Party Autonomy in International Property Law - Peace Palace Library

Party Autonomy in 

International Property Law 

© sellier. european law publishers 

www.sellier.de


www.sellier.de


International Property Law 

edited by 

Roel Westrik 

Jeroen van der Weide 


www.sellier.de

ISBN (print) 978-3-86653-173-4 

ISBN (eBook) 978-3-86653-932-7 

The Deutsche Nationalbibliothek lists this publication in the Deutsche National-bibliografie; 

detailed bibliographic data are available on the Internet at 

http://dnb.d-nb.de. 

© 2011 by sellier. european law publishers GmbH, Munich. 

All rights reserved. No part of this publication may be reproduced, translated, 

stored in a retrieval system or transmitted, in any form or by any means, electronic, 

mechanical, photocopying, recording or otherwise, without prior permission 

of the publisher. 

Production: Karina Hack, Munich. Typesetting: fidus Publikations-Service 

GmbH, Nördlingen. Printing and binding: AZ Druck und Datentechnik, 

Kempten. Printed on acid-free, non-ageing paper. Printed in Germany. 


www.sellier.de

Table of Contents 

Introduction 1 

Roel Westrik/Jeroen van der Weide 

A. 

General Aspects of Party Autonomy, as seen from the 

Perspective of Continental Law as well as of Common Law 

1. Choice of Law in International Property Law – 

New Encouragement from Europe 11 

Axel Flessner 

2. Party Autonomy in Dutch International Property Law 41 


3. The Numerus Clausus and Party Autonomy in the 

Law of Property 59 

T.H.D. Struycken 

4. Party Autonomy and Property Rights 83 

Robert Stevens 

B. 

Private International (Property) Law 

5. Party Autonomy in International Property Law: 

A German Perspective 103 

Jan von Hein 

6. Party Autonomy in French and Belgian Law 119 

Vincent Sagaert 

 

v 


www.sellier.de

Table of Contents 

C. 

Developments and Prospects in Europe and in 

European Law Projects 

7. Article 14 Rome I: A Political Perspective 145 

Paulien M. M. van der Grinten 

8. European Conflict Rules for the mutual Recognition of 

Security Rights in Goods 165 

Ulrich Drobnig 

D. 

Assignment in Private International Law; 

Financial Instruments / the Collateral Directive; 

Insolvency Law 

9. Party Autonomy and Assignment 189 

Hendrik Verhagen 

10. Between Articles 14 and 27 of Rome I: How to interpret a 

European Regulation on Conflict of Laws? 207 

Axel Flessner 

11. Financial Collateral Arrangements and Party Autonomy 225 

Reinout M. Wibier 

12. Party Autonomy and Insolvency Law 261 

P. M. Veder 

vi 

 


www.sellier.de

Introduction 

Roel Westrik 

Jeroen van der Weide* 

1. Party Autonomy in (International) Property Law: 

What Is Going On? 

No fewer than twenty-two universities throughout Europe and even Japan 

and Africa were represented at the ‘Party Autonomy in (International) 

Property Law’ conference, held at the Erasmus School of Law on 27-28 

May 2010. Although this number includes speakers at the conference – 

eminent lecturers highly specialised in Civil Law, European and Private 

International Law as well as in International Property Law – it is still a 

remarkable figure. It becomes even more striking when one takes into 

account the main theme of the conference, which seemingly concerned 

an uncomplicated and simple question: Can or should parties to an international 

transfer of property or an international assignment of claims for 

receivables be permitted to make a choice with regard to the applicable 

law? 

It appears the answer is affirmative only with respect to the underlying 

contractual relationship between the parties: for instance, a sales contract 

or a contract to assign. Of course, parties can choose the law applicable 

to their contractual relationship; this party autonomy is in fact the main 

principle, not only in substantive contract law but also in international 

contract law. See, for instance, Article 3 paragraph 1 of the Rome I Regulation 

on the law applicable to contractual obligations (Regulation [EC] 

No. 593 / 2008, OJ L 177 / 6), which clearly states freedom of choice as the 

first and most important ‘uniform rule’ (conflict of laws rule): 

* 

Roel Westrik is Associate Professor in Civil Law, Erasmus School of Law, Rotterdam, 

and senior counsel at Holla Advocaten in Den Bosch, Eindhoven and 

Tilburg; Jeroen van der Weide is Associate Professor in Civil Law, Leiden Law 

School. 

1 


www.sellier.de

Introduction 

‘UNIFORM RULES 

Article 3 

Freedom of choice 

1. A contract shall be governed by the law chosen by the parties. The 

choice shall be made expressly or clearly demonstrated by the terms 

of the contract or the circumstances of the case. By their choice the 

parties can select the law applicable to the whole or to part only of 

the contract.’ 

Then we come to (International) Property Law. Traditionally, within this 

area of law, party autonomy is not allowed, not to say strongly rejected. 

Basically, legal systems have founded their Rules of Property Law and their 

conflict of laws rules within this field on the lex situs rules that is the law 

of the place where the goods are situated. This rule excludes – so it would 

appear – the mere idea of party autonomy. 

Hence, what is going on when parties enter into a contract concerning 

cross-border transportation of movables, an international sale of (im) 

movables, or an assignment of claims for receivables? The following is 

what happens – and is why conference attendance on the part of twentytwo 

universities may not be too astonishing. 

Let us take Rotterdam – one of the world’s largest ports – as an example. 

More than 800 million tons of goods were transferred in 2008 and 2009, 

a period during which the credit crisis caused a severe ‘dip’ in economic 

activity. Goods continue to come into the port of Rotterdam from everywhere, 

and are then transported further all over the world. By the 

time the goods arrive, they have already crossed multiple borders, and 

subsequently will cross even more. The central legal question that arises 

has to do with what happens to the related property rights when borders 

are and will be crossed. Do we recognise a German retention of title or 

an American security interest? Can a French seller and a German buyer 

in a purchase agreement concerning movable or immovable property 

agree that Dutch Law will be applicable in matters of ownership? Is party 

autonomy or, more specifically, a choice of law possibility allowed, or 

should it be allowed, in other areas of (International) Property Law, such 

as the assignment of claims for receivables? Is there a need for harmonised 

European conflict of laws rules on the proprietary rights regime applying 

to goods, claims and securities? 

2 

Roel Westrik / Jeroen van der Weide 


www.sellier.de

Introduction 

At the conference, members approached these questions by concentrating 

on four central themes: 

1. General Aspects of Party Autonomy, as seen from the Perspective of 

Continental Law as well as of Common Law; 

2. Private International (Property) Law; 

3. Developments and Prospects in Europe and in European Law Projects 

(e.g. European conflict of laws rules for Property Law?); 

4. Assignment in Private International Law; Financial Instruments / the 

Collateral Directive; Insolvency Law. 

The scope and aim of the conference was not only to take an inventory of 

the legal controversies in the field, which are many, but also to have the 

academic world put forward its views on the probability or the possibilities 

of allowing party autonomy within (International) Property Law, in 

Continental Law as well as in Common Law. 

Thus, this volume may be approached in at least two ways. The first consists 

in the discussion of legal possibilities and impossibilities for party 

autonomy in national law systems with regard to functioning within the 

frame of – or even being – European and / or (Private) International Law. 

This area is described by making: 

i. a comparison between German Law (von Hein), Belgian / French Law 

(Sagaert), and Dutch Law (van der Weide); 

ii. a comparison between the Continental and the Common Law systems 

(Struycken, Stevens); 

iii. a comparative investigation into more specific legal conceptions such 

as assignment, ‘collateral’ as security, and insolvency (Verhagen, Flessner, 

Wibier, Veder). 

The second approach to this book is to enjoy the diverse authors’ thorough 

investigations into the more fundamental views – questions and 

ideas hidden beneath the legal reality – or to read about the developments 

and prospects in Europe and in European Law Projects concerning the 

subject of party autonomy in (International) Property Law. In particular, 

Professor Drobnig and Ms Van der Grinten have pursued these developments 

and prospects as a specific topic of research; Van der Grinten 

has in fact written on the making of a Regulation within the Brussels’ 

administration. 

Especially with respect to the fundamental questions, this book aims to 

be an invitation for further research, because one other remarkable phe- 


3 


www.sellier.de

Introduction 

nomenon occurs. It emerges that the assumptions and ideas on how legal 

questions should be answered or, even more appropriate in the area of 

(International) Property Law, how these questions should be answered 

imperatively, may often be questionable. One example of an answer that 

is still commonly believed to be highly imperative is the lex situs rule. Traditionally, 

this rule leads to a negative answer to the question of whether 

party autonomy can be allowed in the field of (International) Property 

Law. In each of the contributions to this book, doubt exists as to whether 

this – supposedly – decisive rule of the lex situs still holds. Some of the 

authors even describe legal routes that may replace the lex situs rule in a 

way that will or may better serve what is called the ‘interest of judicial 

matters’ (see for instance Stevens, Verhagen, Wibier). However, even 

this seemingly undisputable variable ‘interest of judicial matters’ looks 

different when the light of research is thrown upon it. In an article published 

in 2010, Flessner analyses – or even deconstructs – extensively the 

assumptions and arguments that so far have seemed to be undisputable 

in the field of party autonomy in International Property Law. 1 Because of 

the general significance of this article, and its being complementary to the 

research on and the study into the underlying arguments that have been 

posited in the contributions to this book, Flessner’s article has been – in 

an English translation – included separately (see Chapter 1). 

Within the Rotterdam School of Law, the topic of the conference is studied 

within the research programme ‘Behavioural aspects of Tort and 

Contract Law: Reasons for Policymaking’, headed by Professor Willem 

van Boom and Professor Michael Faure. The central focus of this programme 

is on the different forces, originating in any academic discipline 

or science, that may or do influence human – legal – behavior. Research 

is being done on the legal norms and legal standards as the underlying 

forces of – existing or preferred – legal rules, and on the influence these 

norms and standards may have on human behaviour or on that of commercial 

enterprises. 

As regards the actual need for party autonomy in the area of (International) 

Property Law, a few observations may be instructive. 

1 

A. Flessner, Rechtswahl im internationalen Sachenrecht – neue Anstösse aus 

Europa, in: Apathy et al. (eds.), Festschrift für Helmut Koziol, Wien 2010. 

4 



www.sellier.de

Introduction 

2. The growing significance of party autonomy in 

(International) Property Law 

Why is it that parties involved in international trade have developed a 

need to choose by themselves the property law that is applicable to their 

transaction? This question of whether party autonomy or, more specifically, 

the possibility of making a choice of law is or should be allowed in 

Property Law is of growing significance in international trade practice, 

and for five reasons. 

Firstly, the importance of international trade practice has grown enormously 

over the last few decades, and new patterns with respect to the 

traditional concept of rights in rem have emerged. The volume of trade, as 

previously mentioned, has grown to hitherto unknown proportions, and 

it is still increasing. Taking into account that legal systems have founded 

their (International) Property Law on the lex situs rule, the boundaries 

between countries currently form barriers in a sense that was previously 

unthought of as well. These are obstacles in a world that has become used 

to the absence of boundaries and that demands flexibility, not only in a 

technical but also in a legal sense. 

Secondly, the questions of (International) Property Law as mentioned 

above are being answered differently in disparate legal systems, even 

though the lex situs rule is deeply established in the conflict of laws rules 

of countries throughout Europe and most parts of the world. The lex situs 

rule is thus becoming increasingly challenged. 

Thirdly, the idea of party autonomy is gaining ground in national legislations 

as well as in European and international rulings. This provides a 

strong reason to reconsider the traditional rejection of party autonomy in 

(International) Property Law. An illustration of this actuality is the fact 

that in the drafting of the European Rome I Regulation, the dissenting 

opinions within the European Commission on the very subject of party 

autonomy (concerning Article 14 Rome I, the assignment of claims) were 

such that a sub-commission had to be installed. This sub-commission was 

expected to provide its report on the subject in June 2010, but the report 

was postponed because further study was necessary. 2 

2 

See the contributions in this book of van der Grinten (Chapter 7), Verhagen 

(Chapter 9) and Flessner (Chapter 10). 


5 


www.sellier.de

Introduction 

In the fourth instance, several and often conflicting interests lie beneath 

the main question as to whether party autonomy in (International) Property 

Law should be allowed. One important example is the idea of sovereignty 

of states. This notion implies that only state law can rule over legal 

issues, such as the question of whether foreign rights have to be acknowledged 

on the home territory. The idea has been the leading principle for 

states and legislators for a considerable time in Private International Law, 

and it still appears to be dominant: for example, in International Litigation 

Law. It needs no explanation that this principle of sovereignty of 

states must be overcome when cross-border trade needs to be facilitated. 

The tension even becomes enormous when (International) Property Law 

has to be made more flexible – see the aforementioned discussion on 

Article 14 Rome I. Another important example of an underlying interest 

is the principle that legal acts in (International) Property Law have an 

effect not only between the contracting parties but also against a third 

party. Many arguments that strongly reject the idea of party autonomy in 

(International) Property Law are based on this very principle, which can 

only be seen in context with – another important example – the numerus 

clauses principle. 

In the fifth instance, it is doubtful whether traditional objections to the 

idea of party autonomy in (International) Property Law still hold. Society 

has changed and continues to do so rapidly; therefore, legal reality should 

change, or should have changed as well. However, the majority of rules 

and regulations have not yet been revised accordingly. 

These five reasons may show why tension mounts rapidly as soon as the 

notion of party autonomy is brought up in matters of Property Law in 

international trade practice. They may explain the growing significance 

of questions of (International) Property Law when parties enter into a 

contract concerning cross-border transportation of movables, an international 

sale of immovables, or an assignment of claims for receivables. 

Nevertheless, it must be emphasised that – even before the globalisation 

of society and trade practice had begun to challenge the classical ideas 

and concepts within Private International Law – insiders in International 

Property Law had acknowledged that the subject of party autonomy was 

controversial. 3 

3 

Constantin Privat, Der Einfluss der Rechtswahl auf die Rechtsgeschäftliche 

Mobiliarübereignung im internationalen Privatrecht, Ludwig Röhrscheid 

Verlag, Bonn, 1964; Axel Flessner, Fakultatives Kollisionsrecht, RabelsZ 34 

6 



www.sellier.de

Introduction 

3. Acknowledgments 

The theme and the first organisational activities relating to the ‘Party Autonomy 

in (International) Property Law’ conference were raised in 2008, 

a year that developed into a period of economical setback. The mere fact 

that the conference – with its captivating lectures and fascinating set of 

contributions as assembled in this book – could be realised in May 2010 

is due to the willingness and the genuine help of a few institutions and 

persons. We, Roel Westrik and Jeroen van der Weide, are deeply indebted 

to all the speakers and authors, to the Erasmus School of Law and its dean, 

Professor Maarten Kroeze, to Professor Willem van Boom and Professor 

Michael Faure, to the Vereniging Trustfonds Erasmus Universiteit Rotterdam, 

to Holla Advocaten, to Clifford Chance, to the Leiden Law School, 

to sellier. european law publishers, and to Jan Sramek Verlag for the kind 

consent to have the aforementioned article of Axel Flessner in the Koziol 

Festschrift translated. A special word of thanks for their significant help 

in several stages of the process is owed to Professor Axel Flessner, Professor 

Rick Verhagen, and Professor Xandra Kramer. The digital existence 

of the conference was made possible – with digital velocity – by Jolanda 

Bloem. The English text was being edited with the utmost care by our 

native speaker, Ms Donna Devine. Finally, without the crucial help of our 

secretaries, Gerda Albers, Maja Lucas, Sandra Klomp, Debbie Kneefel, 

and Margreet den Hond-Schol, there would not have been a conference 

or a book at all. 

 

 

Roel Westrik 


(1970), 547-584; Rolf H. Weber, Parteiautonomie im internationalen Sachenrecht?, 

RabelsZ 44 (1980), 510-530. 


7 


www.sellier.de


www.sellier.de

A. 

General Aspects of Party Autonomy, 

as seen from the Perspective of Continental Law as 

well as of Common Law 


www.sellier.de


www.sellier.de

1. 

Choice of Law in International Property Law – 

New Encouragement from Europe * 

Axel Flessner ** 

International property law invokes the law of the place where the asset 

is situated, the lex rei sitae or lex situs, to control the creation, content, 

exercising and transfer of property rights in the asset (rights in rem) 1 , and 

the prevailing view says that this rule is ‘mandatory’; in principle, the 

parties may not make their own, different choice of law. 2 

In the case of land, the rule invoking the law of the place where the 

property is situated is undisputed, whereas in the case of movable assets 

this is by no means self-evident in view of the great and still increasing 

international freedom of movement of property and people. The Swiss 

International Private Law Act (IPRG) of 1987 tried to take this into 

account by allowing the parties to subject the contractual acquisition of 

property rights in movable tangible assets to the law of the ‘country of 

shipment’, the ‘country of destination’ or the law ‘to which the underlying 

contract is subject’ (Article. 104). It stopped half way, however, 

establi shing that the choice of law agreement ‘cannot be used against 

third parties’ (Article. 104(2)). In addition, there have been extensively 

developed proposals in the international academic literature to allow for 

* 

This article was published earlier as Axel Flessner, Rechtswahl im internationalen 

Sachenrecht – neue Anstösse aus Europa, in: Apathy et al. (eds.), 

Festschrift für Helmut Koziol, Wien 2010, p. 125-146. 

** 

Professor Em. of European Private Law, Commercial Law and Conflict of Laws 

at the Humboldt-Universität, Berlin. 

1 

Section 31 of the Austrian International Private Law Act [IPRG-Gesetz], 

Article 43 of the German Introductory Act to the Civil Code [EGBGB]. 

2 

Verschraegen in Rummel, Kommentar zum Allgemeinen bürgerlichen Gesetzbuch 

[ABGB] II, 3d ed. (2002) § 31 IPRG nr. 19; Heldrich in Palandt, Bürgerliches 

Gesetzbuch [BGB], 67th ed. (2008) Article 43 EGBGB nr. 2. 

11 


www.sellier.de

A. General Aspects of Party Autonomy 

disposals of movable property a choice of the applicable law by the parties 

to the transaction. 3 

In Germany, the further development of these ideas was cut off by the 

Reform Act of 21 May 1999, which, for property law, added Articles 43 to 

46 to the Introductory Act to the Civil Code (EGBGB). These provisions 

codify the reference to the location of the asset as the basic rule, and they 

are regarded as mandatory by their authors and by the view that prevails 

in commentaries and textbooks, allowing no different choice of law by the 

parties themselves. 4 Even the new escape clause in Article 46 EGBGB is 

not regarded by the prevailing view to open a way for an agreed choice of 

a law other than the law where the property is situated. 5 

The national legislature may perhaps silence the academics in its own 

country, but it will not halt academic developments and legislation in 

other countries and in the European Union. The built-in tension between 

going along with the gravitational pull of the place where the property is 

situated, on the one hand, and the fact that property is free to move internationally, 

on the other, has in any case allowed the subject to remain 

virulent in the rest of Europe and has produced academic and legislative 

advances towards greater freedom of the parties in the choice of law. These 

advances will here be introduced (sections A to B), evaluated (section C) 

and developed (sections D to E); the deliberations will conclude with a 

3 

R. Weber, Parteiautonomie im internationalen Sachenrecht? RabelsZ 44 

(1980) 520; Khairallah, Les suretés mobilières en droit international privé 

(1984); Stoll in Staudinger, EGBGB 13th ed., Internationales Sachenrecht 

(1996) nr. 262, 282 et seq.; von Wilmowsky, Europäisches Kreditsicherungsrecht 

(1996) 94-152; Stadler, Gestaltungsfreiheit und Verkehrsschutz durch 

Abstraktion (1996) 673-680; Einsele, Rechtswahlfreiheit im internationalen 

Privatrecht, RabelsZ 60 (1996) 417, 435 et seq.; Ritterhoff, Parteiautonomie 

im internationalen Sachenrecht (1999). 

4 

R. Wagner, Der Regierungsentwurf eines Gesetzes zum Internationalen Privatrecht 

für außervertragliche Schuldverhältnisse und für Sachen, IPRax 

1998, 429, 435; Wendehorst in MünchKomm, BGB X, EGBGB, 4th ed. (2006), 

Vor Article 46 nr. 11, Article 46 nr. 18-20; Heldrich in Palandt, BGB, 67th ed., 

Article 43 EGBGB nr. 2; von Hoffmann / Thorn, Internationales Privatrecht, 

9th ed. (2007) § 12 nr. 10, 11. 

5 

Heldrich in Palandt, BGB 67th ed., Article 46 nr. 3; Wendehorst in MünchKomm, 

BGB X, EGBGB, 4th ed., Article 46 nr. 18-23; von Hoffmann / Thorn, IPR, 9th 

ed., § 12 nr. 12. 

12 

Axel Flessner 


www.sellier.de

1. Choice of Law in International Property Law 

look at the special situation in Europe (section F). It is hoped that these 

reflections will be of interest to Helmut Koziol, who has himself dealt with 

party choice of law for international credit security 6 , a subject on which 

we have occasionally had friendly skirmishes. 

A. Recent academic developments 

I. Paris 

In an unusually extensive Paris thesis or Thèse de doctorat (773 pages), 

which was awarded a faculty prize, the question of which law ought to 

govern property in tangible and intangible assets is fundamentally reassessed. 

7 The author sees a categorical difference between, on the one 

hand, the use of an asset as a physical object, with its physical properties 

and, on the other hand, the creation and transfer of rights in the asset, its 

allotment to a title-holder. The physical use of the property must occur at 

the place where it is situated, and consideration must therefore be given to 

the natural interest of the social environment at that place in regulating 

the use of the property by the legal owner or possessor (enjoyment, use, 

consumption) and the limits to such use (in other words, the defensive 

rights of others). 8 In the legal allotment of an asset to an owner, on the 

other hand, physicality plays no part. The sole consideration is to establish 

rights (in the asset) – a purely intellectual exercise. If what is involved is 

an original acquisition of title (by production, processing, finding of lost 

property, appropriation of res nullius, acquisitive prescription), acquisition 

of title is based on a unilateral action; the law of the place where this 

happens suggests itself for application (it will often be the place where 

the asset is situated). 9 If, on the other hand, a derivative acquisition, that 

is, acquisition from the title-holder by legal transaction is involved, this 

acquisition takes place (among the living) by consensus, and for contractual 

transactions private autonomy and freedom of contract are recognised 

worldwide as well as, in addition, the legal owner’s freedom to dispose. 

6 

Koziol, Probleme der Sicherungszession im grenzüberschreitenden Verkehr 

Deutschland-Österreich, DZWiR 1993, 353. 

7 

D’Avout, Sur les solutions du conflit de lois en droit des biens [with preface 

by Hervé Synvet] (2006); reviewed by Stoll, RabelsZ 73 (2009) 383-395. 

8 

D’Avout, Solutions 15-23. 

9 


Axel Flessner 

13 


www.sellier.de


The legal owner may agree to the sale of the property and its encumbrance 

with a right in rem anywhere, irrespective of the place where the property 

is situated, and with anyone. It is therefore an obvious suggestion to allow 

the parties to determine for the disposal and acquisition also the law that 

is to be applied. 10 Because the physicality of the asset and therefore the 

place where it is situated plays no part in the legal effect of the disposal 

(that is, the acquisition of title), it is also obvious that the disposal over 

corporeal assets (d’Avout: in reality over the rights in such assets) should in 

principle be treated in the same way as the disposal over other assets (receivables, 

company shares, intellectual property) in the conflict of laws, 

unless special features of the object require special provisions. 11 

In summary, the new theory can be formulated as follows: The law of the 

place where the tangible asset is situated should govern in relation to the 

exercising and protection of property rights in the asset (in other words, 

what the legal owner may do with it physically and what protection he 

can expect from the State in respect of this right). In relation to the disposal 

of the property right by legal transaction, on the other hand, it is 

the law elected by the parties themselves or otherwise applicable to the 

contract between them that governs also the property disposal. 

II. 

Amsterdam 

Back in 2001, the Dutch journal Nederlands internationaal Privaatrecht 

(NIPR) published a particularly cogent paper advocating freedom of the 

parties in the choice of law as the basic principle. 12 The author, a young 

Amsterdam practitioner, summarised the Dutch and European opinions 

at the time and came to the same essential conclusion as a few years later 

the French thesis just introduced: In relation to content and exercising, 

the law of the place where the asset is currently situated rightly applies; 

in relation to the acquisition of the right by legal transaction, however, 

freedom in the choice of law produces a better balance between the interests 

of those involved in the transaction and third parties who are not 

involved in the transaction. The main argument as far as the author is 

concerned is the legal owner’s ‘freedom to choose a location’. He may 

determine the place where the asset is situated and by that, according to 

10 

D’Avout, Solutions 589 et seq., 623 et seq. 

11 


12 

Pos, Rechtskeuze in het internationaal zakenrecht, NIPR 2001, 398-405. 

14 

Axel Flessner 


www.sellier.de


the traditional reference to that place, virtually also the law applicable to 

further disposals of the property. 13 It would be practical, then, to let him 

(along with the acquirer) have from the outset also in the conflict of laws 

the choice between the legal systems available. 

An Amsterdam Proefschrift [thesis] which was published later, at the same 

time as the said French Thèse, does not go far afield and theorise about a 

justification of party choice of law, taking that freedom instead as a fundamentally 

accepted possible solution (in the Netherlands at least) for 

international property law. The author designates in the title itself the 

particular issue that makes the application of the law at the asset’s location 

difficult and questionable, namely the ‘mobility’ of movables which, 

under the rule pointing to the location, will necessarily result in a change 

also of the governing law if a tangible asset is moved from the territory of 

one jurisdiction into that of another. 14 If the place where the property is 

situated can be changed at will by the legal owner, even crossing borders, 

would it not be better for that place to be replaced as a principle in the 

conflict of laws by deference to the will of the parties? 

The preliminary draft of the Dutch Staatscommissie (the standing consultative 

body for private international law) proposing an act of parliament on 

international conflicts of laws in property law already existed. It allowed 

party choice of law for a small number of narrowly defined situations in 

which the place where the property is located is actually meaningless – 

thus in the case of goods in international transit (res in transitu) and missing 

property for which the current location is unknown. The author 

agrees with the draft. However, he believes that party freedom in the 

choice of law is limited by the closed number of property types (numerus 

clausus) that each state may establish in its own property law system. 15 

Moreover, the difficulties involved in the change of the governing law as 

a corollary of the asset moving across borders could be overcome in the 

‘receiving country’ through liberal assimilation of the ‘foreign’, ‘imported’ 

right in rem to domestic property law concepts. 16 So, although the author 

13 

Pos, NIPR 2001, 401; the author deals specifically with the arguments put 

forward by Einsele, RabelsZ 60 (1996) 441. 

14 

Van der Weide, Mobiliteit van goederen in het IPR – Tussen situsregel en 

partijautonomie, Deventer (2006). 

15 

Van der Weide, Mobiliteit, 182 et seq., 232, 235. 

16 

Van der Weide, Mobiliteit, 183, 236. 

Axel Flessner 

15 


www.sellier.de


has no fundamental objections to party freedom in the choice of law, he 

does not regard it as urgent in practice. 17 

B. Legislation 

I. Netherlands 

The new Dutch Act of 28 February 2008 on the conflict of laws concerning 

property 18 now essentially grants freedom of choice of law to the parties 

to the extent advocated by the aforementioned draft of the Staatscommissie. 

19 With regard to tangible assets, it basically refers to the law of the 

place where the asset is situated and for registered ships and aircraft to the 

law of the country of registration (Article 2). If goods are to be delivered 

abroad under reservation of title, the parties may, however, choose the 

law of the country of destination from the outset for the effects of the 

reservation, provided the law of the country of destination recognises the 

reservation (Article 3(2)). The same applies to the property effects of a 

leasing agreement (and anyhow for the obligational effects, according to 

Article 3 of the Rome I Regulation), if the property concerned is to be 

used abroad (Article 3(3)). 

The Act also opens up the choice of the applicable property law to the 

parties to a sales contract or other contract that commits them to the 

transfer or creation of a right in rem, if execution of the contract involves 

international transport. The choice of law made by them for the sales or 

other contract then also applies by virtue of the Act to the rights in rem 

to be created under the contract (Article 8(2)). If such a choice of law 

is omitted, the law of the country of destination applies to the property 

rights in the goods transported (Article 8(1)). 

This sort of indirect effect of a choice of law agreement is also allowed 

if goods go missing and their location is unknown. A legal transaction 

on the part of the owner in relation to the goods must then be assessed 

17 

Van der Weide, Mobiliteit, 235. 

18 

Staatsblad 2008 / 70. According to Article 23, the Act is to be referred to as 

‘Wet conflictenrecht goederenrecht’ [Law of Property (Conflict of Laws)], 

which will be generally abbreviated to ‘WCG’. 

19 

See footnote 15. 

16 

Axel Flessner 


www.sellier.de


according to the law of the country in which the goods were located 

before they went missing (Article 7(1)). However, if they were insured 

against loss, it is the law governing the insurance contract (which can in 

principle be chosen by the parties 20 ) that determines whether and, if so, 

how, ownership passes to the insurer on account of the insurance benefit 

(Article 7(2)). 

Finally, the Act also applies to the international assignment of claims, 

since, in accordance with the Dutch Civil Code, it also regards claims 

as assets, which require a ‘proprietary’ attribution to a creditor as the 

‘holder’ of the claim. 21 The assignment of a claim, which is not embodied 

in a negotiable paper, and the creation of a limited right in that claim, 

for instance a pledge, are governed by the law applicable to the contract 

that obliges the holder of the claim to make the transfer or pledge (Article 

10(2)), and the law applicable to that contract can be chosen by the 

parties under Article 3 of the Rome I Regulation. This power of (indirectly) 

choosing the applicable law is excluded only for the transferability 

and chargeability of the claim and for the legal relationship between the 

disposing parties and the debtor under the claim; application of the law 

governing the claim itself, the ‘law of the claim’, is reserved for these issues 

(Article 10(1) and (3)). 

With these provisions, the Dutch Act opens up choice of law by the 

parties for individual cases and objects which have also been discussed 

elsewhere already as exceptions to the lex situs rule, and here and there 

a corresponding conflict has been enacted. But these discussions and enactments 

were aimed only at establishing better objective factors than the 

traditional local one. 22 The fact that the Act recognises choice of law by 

the parties here is therefore an important step, because the Act shows that 

20 

See Article 7 Rome I Regulation. 

21 

Book 6 (‘General Section of the Law of Obligations’) of the Burgerlijk Wetboek 

(BW) [Civil Code] regulates the claim as a relationship between creditor and 

debtor (Article 1 et seq.) and the consequences of transferring the claim to a 

new creditor (Article 142 et seq.) and Book 3 (‘General Section of Property 

Law’) regulates the transfer of the claim as an asset and its charge with a limited 

right in rem (Article 83, 94, 95, 97, 98). 

22 

Thus §§ 32, 33 International Private Law Act (IPRG); Article 45 Introductory 

Act to the Civil Code (EGBGB). Overviews by Verschraegen in Rummel, 

ABGB II, 3d ed., § 31 IPRG, nr. 27-47; Heldrich in Palandt, BGB, 67th ed., 

Article 43 EGBGB nr. 2. 

Axel Flessner 

17 


www.sellier.de


the traditional rule pointing to the place where the property is situated or 

to other objective circumstances is not an insurmountable natural state, 

but rather is based on a deliberate decision on the part of the legislator, 

which can also be different, namely in favour of party freedom in the 

choice of law. 

II. 

Europe 

There are currently no plans in the European Union for legislation on 

international property law. However, Article 14 of the Rome I Regulation 

also regulates the international assignment of claims and in that context it 

necessarily also had to understand the claim as an asset which can go from 

hand to hand or can be encumbered with a limited right in rem, in respect 

of which asset, therefore, in short, a disposal can be made. According to 

Article 14(1) of the Rome I Regulation, the law that is governing this 

disposal is the law that must be applied to the underlying contract between 

the present holder of the claim and the person going to acquire the claim 

(or the limited right in it), and this law applicable to the contract may 

be chosen by the parties themselves in accordance with Article 3 of the 

Rome I Regulation. 

This means that choice of law by the parties is also acknowledged for the 

disposal of the claim as an asset. This should make us seriously reconsider 

the international law of tangible property too. To be sure, tangible 

assets are physical objects and therefore necessarily located in one place, 

while claims are incorporeal and, being an intellectual concept, have no 

‘place’ where they ‘can be found’. Something both items have in common, 

though, is that disposal of them by legal transaction should bring 

about their legal reallocation to a person with effect towards third parties. 

This legal reallocation, the nomination of a person as the new legitimate 

holder of title to the asset, is also a purely intellectual process, which has 

been clearly worked out particularly in the d’Avout thesis. 23 It can be 

asked, therefore, whether party freedom of choice of law, if accepted for 

claims, is not a conceivable solution for tangible goods as well. 

The assumption that is widespread in the literature, however, is that the 

‘third-party effect’ of the assignment of claims is not covered by Article 

23 


18 

Axel Flessner 


www.sellier.de


14 of the Rome I Regulation, but instead has been deliberately set aside. 24 

However, this view is not consistent with the wording and codifying aim 

of Article 14 of the Rome I Regulation. 25 One of the most important 

aspects of the international assignment of claims could not remain unregulated, 

and the wording of the provision concerned in its paragraph 

1 and also the recital therewith 26 suggest the interpretation that the disposal 

effect of the assignment and encumbrance of a claim is completely 

regulated by the provision. It was also possible to show that the provision 

with this understanding is workable in practice. 27 

C. Intermediate findings 

The statements made in the more recent academic and legislative texts 

from other countries in Europe can be summarised as follows: 

Firstly, choice of law by the parties may supplant the ‘objective’ connecting 

factors, where the location of the property becomes indistinct (export 

goods, goods in transit), is unknown (assets gone missing) or does not 

exist (claims). 28 The freedom to choose the legal system has been made 

law here, although in these cases an ‘objective’ substitute reference is 

frequently possible and recommended (country of destination, country 

where the property went missing) 29 or (for the assignment of a claim) 

24 

F. Bauer, Die Forderungsabtretung im IPR (2008) 103 et seq., 167, 301; Kieninger, 

Die Vereinheitlichung des Kollisionsrechts der Abtretung, in: Basedow / 

Remien / Wenckstern (eds.), Europäisches Kreditsicherungsrecht – Symposium 

Drobnig (2010) 149-159. 

25 

For more information: Flessner, Die internationale Forderungsabtretung nach 

der Rom I-Verordnung, IPRax 2009, 35, 38 et seq., and in this book, p. 207-224. 

26 

Recital 38 of the preamble to the Regulation. 

27 

In detail: Flessner, Rechtswahlfreiheit auf Probe – Zur Überprüfung von Art 

14 der Rom I-Verordnung, in: Festschrift Kühne (2009) 687-699. 

28 

According to the Dutch Act and Article 14 Rome I Regulation. 

29 

Overview: Verschraegen in Rummel, ABGB II 3d ed., § 31 IPRG nr. 36-38; 

Schwind, Internationales Privatrecht (1990) nr. 396-398; Wendehorst in 

MünchKomm BGB X, EGBG 4th ed., Article 46 nr. 32-44; Kropholler, Internationales 

Privatrecht, 6th ed. (2006) § 54 IV, V; von Hoffmann / Thorn, IPR 

9th ed. § 12 nr. 36-42; Rauscher, Internationales Privatrecht, 3d ed. (2006) 

nr. 1466. 

Axel Flessner 

19 


www.sellier.de


practised or recommended from the outset by many and in many places 

as the main principle (country of residence of one of the parties involved, 

law of the claim). 30 

Secondly, the rule in the conflict of laws should correlate with the owner’s 

freedom to dispose over the property and to determine its location and 

thereby the law that would be objectively applicable to its use. 31 

Thirdly, a distinction must be made between matters of law that are associated 

with the physical nature and location of the property, and those 

matters of law that relate solely to the legal attribution of a piece of property 

to one person or another. 32 Thus, it seems natural that the owner’s 

right effectively to use and consume the property and the rights of others 

to effectively act upon the property or to seize it are determined according 

to the laws of the place where the property is effectively situated. 

On the other hand, the physicality and location of the property do not 

play a decisive role in the legal, in other words, purely intellectual allotment 

of an asset to a person through a legal transaction. The question of 

where the disposing legal transaction is made (place where the property 

is situated or elsewhere) and between whom (residents or non-residents) 

and the geographical distance between the assets and those who have, or 

are supposed to have, property rights in them is of no consequence to the 

allocation of the right by legal transaction. It is obvious, therefore, that 

here, too, the party freedom in the choice of law generally existing in 

relation to private legal transactions should be granted. 33 It is by the same 

token the continuation of the fundamental freedom of the title-holder to 

dispose in the international context with the means available there under 

the conflict of laws – which is precisely the party choice of the legal system 

that should govern the disposal. 

D. Central ideas and counter-arguments 

The foregoing summary is to some extent at odds with the prevailing 

doctrine in Germany and Austria, which tends to be completely sealed 

30 

More information, with further details Flessner, IPRax 2009, 36 et seq. 

31 

See above in footnotes 12-14. 

32 

D’Avout, Solutions, footnotes 8-11 above. 

33 

D’Avout, Solutions, footnotes 10, 11 above. 

20 

Axel Flessner 


www.sellier.de


against party choice in international property law. Recent European findings, 

instead, encourage fresh deliberation of how international property 

law can be opened up academically to provide greater freedom in the 

choice of law. The following central ideas are based on the findings; they 

are exposed here to the arguments normally presented in support of the 

mandatory nature of international property law, particularly of the rule 

pointing to the law of the place where the property is situated. 

Firstly: Freedom in the choice of law is connected with freedom of disposal 

and acquisition in substantive property law. Insofar as freedom is 

acknowledged in substantive law, it may also be the starting point in the 

conflict of laws. 

Secondly: Freedom in the choice of law is also connected with international 

freedom in the movement of goods and capital. Insofar as this exists, 

the use of the international variety of property laws by the parties 

involved cannot be unwelcome in principle. 

Thirdly: Property rights in tangible assets may be allotted and reallotted by 

transactions far from the place where the assets are situated and without 

any physical contact with or movement of the property. Insofar as this is 

possible, the legal disposal over property rights in tangible assets equates 

to the disposal over non-tangible goods; both are in reality a disposal affecting 

a right and choice of law by the parties can therefore be allowed 

for both with the same authority. 

I. The needs of commercial intercourse 

International intercourse involving property needs a visible connecting 

factor. This alleged need is most commonly cited in opposition to a power 

of the parties to choose the applicable law and in favour of applying the 

law of the place where the assets are situated. It is said that this place can 

be easily ascertained by anyone seeking information on the applicable 

law, whereas the agreement on the choice of law between parties to a 

disposal transaction can not. 34 But ‘needs’ is a blanket argument. Private 

international law, as private law in general, must focus on possible con- 

34 

Wendehorst in MünchKomm BGB X, EGBGB, 4th ed., Vor Art 43 nr. 11, Art 

46 nr. 18; von Hoffmann / Thorn, IPR, 9th ed., § 12 nr. 10, 11. 

Axel Flessner 

21 


www.sellier.de


cerns of conceivable parties interested in this or that solution 35 ; it must 

ask precisely who are the parties possibly involved in property relations, 

what kind of interests they may actually have, to what extent their interests 

are legitimate and how their essentially legitimate interests should 

be measured against essentially legitimate opposing interests, in short – 

in the words of Helmut Koziol: ‘The … conflicts of interest to be resolved 

and the fundamental evaluations … must be worked out and taken into 

account.’. 36 

The parties concerned with property are first and foremost the parties to 

the disposal transaction. In the international setting, they both want to 

know first of all what conditions have to be met for the transfer or creation 

of the right in rem under the applicable legal system. They may determine 

this applicable law by themselves through their agreement – and thereby 

bring into play their knowledge of the chosen legal system or their assessment 

of the place where the acquirer intends to assert his property right. 

By contrast, the objective reference to the law of the place where the asset 

is situated does not give support to such considerations. 

The party acquiring the right in rem will also want to know the law under 

which the disposing party acquired his title – in other words, whether he 

is truly disposing as the entitled party. This law does not have to be the 

law of the present location if the property has come from another country. 

Finding out the location of the property at the time of the earlier acquisition 

is no simpler than asking the disposing party about the (perhaps 

agreed) law covering his earlier acquisition. In the individual case, the 

concerns of the acquiring party are therefore better served overall with 

the power to choose the law than with the mandatory objective reference 

to the location of the property. 37 

The analysis must not be limited, however, to the interests of the parties 

to the individual transaction in question. It must extend to all the potential 

actors in the property market – in other words, to all those who at some 

time may harbour an interest in acquiring property. It is easier for them 

(and for the title-holders) to enter into a property transaction if they 

have a visibly existing standard for the acquisition of title. However, the 

35 

Flessner, Interessenjurisprudenz im internationalen Privatrecht (1990), particularly 

48-66. 

36 

Koziol, Grundlagen und Streitfragen der Gläubigeranfechtung (1991) 2. 

37 

In detail also: Ritterhoff, Parteiautonomie 289-292. 

22 

Axel Flessner 


www.sellier.de


standard provided by the law at the asset’s location will be there regardless 

whether one applies that law from the outset or only subsidiarily, that is, 

in case a choice of law agreement is not made by the parties. Nobody is 

obliged to accept an offered party choice of law. And not having to become 

occupied with the offer of a choice of law should not be an interest 

worthy of protection by the law on conflict of laws. 

It is evident, therefore, that the ‘needs of commercial intercourse’ as an 

argument against party choice of law for the property disposals over tangible 

goods does not withstand a closer analysis of interests. 

II. 

Numerus clausus 

(property rights limited in number and variety) 

According to many authors 38 and comments made in legislation and case 

law 39 , freedom of party choice of law cannot be granted in international 

property law because it would undermine the closed number and variety, 

the numerus clausus of property rights allowed in a legal system. In this 

view, the mandatory reference to the place where the property is situated 

guarantees that legal intercourse within the territory of a legal system must 

only reckon with the types of rights allowed in that system and is thereby 

facilitated. 40 The view brings into play a further ‘interested party’, namely 

the State, which, by a clear and unchanging list of property rights, seeks 

to facilitate economic transactions in its territory with the intention of 

promoting general welfare. 

First of all, the same objection can be raised against invoking the numerus 

clausus principle as against the argument of the ‘needs of commercial 

intercourse’. Invoking the numerus clausus obstructs the freedom of movement 

of actors in the property market far too sweepingly in order to uphold 

a dogmatic principle and without taking account of whether entering 

38 

Verschraegen in Rummel, ABGB) II, 3d ed., § 31 IPRG nr. 19; van der Weide, 

Mobiliteit, 36-39, 182 et seq. with further details. 

39 

The German federal government in its Reform Act Bill of 1999 (see above 

footnote 4), Bundestags-Drucksache 14 / 343, 16; BGHZ 39, 173. 

40 

Thus the general justification of numerus clausus: Koziol / Welser, Grundriss 

des bürgerlichen Rechts I, 13th ed. (2006) 238; Baur / Stürner, Lehrbuch des 

Sachenrechts, 18th ed. (2009) § 1 II 2, 3a. 

Axel Flessner 

23 


www.sellier.de


into property transactions would actually be adversely affected by allowing 

party choice of law generally or by its use in a given case. 41 

But to begin with, the strict doctrine of numerus clausus is already fundamentally 

out of place in today’s international property law. The doctrine 

presumes a legal system that is self-contained and closed off from the 

outside world, such that it can keep its property law system ‘pure’. This 

notion is unrealistic and inadequate in a world where international freedom 

ought to exist in the movement of goods. In this open world, the 

individual country can no longer expect that all property located within 

its territory has been acquired and encumbered according to its closed 

system of rights in rem; rather, the opening of the system to foreign property 

law is quite inevitable. The country where the property is situated 

may wish to keep its own system closed for the exercising of any rights in 

rem within the country and to protect them from attack. However, this is 

not necessary for the recognition of acquisitive transactions, particularly 

if the acquisition took place while the property was abroad. Moreover, 

even in order to exercise and protect the property right which was acquired 

under foreign law but is to be protected according to the law of the 

forum, the conceptual adaptation of the foreign property right to domestic 

notions is necessary, in order to make it compatible with the domestic, 

allegedly ‘closed’ system. 42 This need for adaptation – which requires the 

ability of lawyers to compare legal systems and often to make difficult 

evaluations 43 – undermines the main advantage commonly ascribed to 

the numerus clausus principle, namely the standardisation of the available 

41 

In detail: d’Avout, Solutions 663-690: ‘… disproportion entre le but poursuivi et 

les effets obtenus: l’autonomie de la volonté est abstraitement et généralement 

paralysée pour un objectif de protection des tiers, qui n’est concrètement en 

jeu que de manière exceptionelle’ [‘… disproportion between the aim pursued 

and the effects obtained: the autonomy of will is paralysed abstractly and 

generally for an objective of protecting third parties which is not specifically 

involved other than in an exceptional manner’], 665. 

42 

The literature deals with this adaptation, insofar as it is necessitated by the 

reference to the place where the property is situated, under the title ‘Statutenwechsel’: 

Verschraegen in Rummel, ABGB II 3d ed., § 31 IPRG nr. 20-26; 

Wendehorst in MünchKomm BGB X, EGBGB 4 th ed., Article 43 nr. 125-147, 

both with extensive further details on the literature. 

43 

The ‘change of the governing law’ in text books and commentaries is routinely 

the most important and most difficult subject in international property law. 

24 

Axel Flessner 


www.sellier.de


rights in rem that should be readily accessible for property transactions 

under a given legal system. 

In the internationally open economy, the numerus clausus of property 

rights in a single country can therefore no longer be reasonably invoked 

against the power of the parties to choose the applicable law for the creation 

and transfer of such rights by legal transaction. The principle, in international 

cases, remains sensible only as part of the package of rules 

offered by the legal system that is to be applied under a conflict rule of 

whatever content. Moreover, the notion of a national limit on the number 

and varieties of property rights is on the retreat even for the exercise and 

protection of the foreign property right. Both in academic theory and now 

also expressly embodied in legislation, there is a growing view that the 

foreign right in rem, to be manageable in the domestic forum, need not 

undergo a wholesale reinterpretation beforehand into a corresponding 

legal figure from the domestic list of property rights. Instead, it is sufficient 

according to this view for the foreign right to be ‘equivalent’ to the 

kind of right that the forum would apply in a comparable purely domestic 

case; in other words, the foreign right, in order to find application, must 

only be ‘tolerable’ in the domestic surrounding. 44 The legislature in the 

Netherlands has deliberately adopted this view. 45 The new Act (Article 

5 WCG) reads as follows: 

‘Rights in a tangible asset that have been acquired or created in accordance 

with a law applicable under this Act shall continue to exist, 

even if the asset is taken into another country. These rights may not 

be exercised in a way that is inconsistent with the law of the country 

on whose territory the property is located at the time when the right 

is exercised.’ 46 

44 

This perspective is presented in Verhagen, Conflit mobile bij roerende zaken: 

assimilatie of transformatie? (Deventer – NL, 2007) 16-26; van der Weide, Mobiliteit, 

33-62; Wendehorst in MünchKomm BGB X, EGBGB 4th ed., Article 

43 nr. 147-154; von Hoffmann / Thorn, IPR 9th ed., § 12 nr. 31. 

45 

Evidently influenced by Verhagen, Roerende zaken in het internationaal 

privaatrecht – Kritische beschouwingen over het wetsontwerp voor een ‘Wet 

conflictenrecht goederenrecht’, Weekblad voor Privaatrecht, Notariaat en 

Registratie (WPNR) 2007 / 6711, 468 and and 2007 / 6712, 495. 

46 

On the importance of the provision: Verhagen, Conflit mobile 78 (‘Postscriptum’). 

Axel Flessner 

25 


www.sellier.de


III. 

Third-party effect 

The alienation or encumbrance of a piece of property works against all 

uninvolved ‘third parties’. There seems to be a latent yet widespread notion 

that, for this very reason, those involved in the transaction cannot 

be allowed to determine the applicable property law, to shift aside the 

provisions of the lex situs and thereby to impose at their will legal effects 

on third parties. 47 However, this attitude denies something in international 

property law that is self-evident in domestic property law. The law 

of property allows the disposal with effect for third parties without their 

consent, simply because the holder of the right has the freedom to dispose; 

his disposal must be respected by all ‘third parties’ in the economic and 

social environment – creditors, tax office, prospective acquirers, neighbours. 

The third-party effect is an intrinsic part of the right to dispose in 

property law. This would be contradicted by ‘switching’, in the conflict of 

laws area, to making the untouchability of ‘third parties’ an ideal. Rather, 

it is consistent with the power of disposal of the entitled party to leave 

the decision on the applicable law to that party’s discretion too, that is, 

to include this power under the characteristics of the right in rem with 

the help of which disposal over the property is achieved. Even if the 

third-party effect of a property disposal could be considered a bar to party 

choice of law, then this would be worth considering only for those situations 

where third-party protection would actually be a possible issue in 

substantive property law. This will immediately become clear in the next 

two ‘third party protection’ topics. 

IV. 

Creditors 

The creditors of the disposing party are often portrayed as third parties 

concerned in particular. It is said that they may not be deprived by a 

choice of law agreement of an asset assured to them through the ‘system 

of creditors’ rights’ at the location of the asset. 48 

47 

Thus, for example, Röthel, Internationales Sachenrecht im Binnenmarkt, JZ 

2003, 1027, 1034. The notion also evidently forms the basis of Article 104(2) 

of the Swiss International Private Law Act (see footnote 3 above). Further 

comments of this sort demonstrated in Kieninger, Mobiliarsicherheiten im 

Europäischen Binnenmarkt (1996) 35. 

48 

Stoll, Internationales Sachenrecht, nr. 360, 361; Basedow, Der kollisionsrechtliche 

Gehalt der Produktfreiheiten im europäischen Binnenmarkt: favor of- 

26 

Axel Flessner 


www.sellier.de


This argument ignores essential structures of the law of obligations and 

of liability, which can be encountered everywhere. The mere claim to a 

sum of money by itself does not entitle the creditor to a particular object 

from among the debtor’s assets by way of which the amount owed could be 

settled. The debtor is entirely free to shift the nature and composition of 

his property, to deal with acquisitions and make disposals. If the creditor 

wishes a particular object to be set aside, he must obtain a security right 

in that object. However, the enforcement of the security right is then 

no longer an element of general creditor protection, but is often quite 

opposed to it. 

Protection of the unsecured creditor against unwelcome diminution of his 

debtor’s assets exists only under forced execution and insolvency law and, 

in addition, under the law on the avoidance of legal acts of a debtor that 

is attached to these proceedings, following the Actio pauliana format – in 

Germany the Act on the Avoidance of Legal Acts of a Debtor outside 

Insolvency Proceedings and Sections 129 et seq. of the Insolvency Code. 

Jurisdiction for seizure and sale of the debtor’s property in the course of 

forced execution is necessarily linked to the location of the property, as 

the enforcing country is not allowed to reach beyond its national territory 

with physical power, and this country’s procedural law then determines 

the rights of the creditors and their ranking. 49 Modern insolvency law 

connects its jurisdiction mandatorily to the registered office or place of 

business of the debtor 50 , and possibly, in addition, to the location of the 

property or a branch establishment 51 , and it likewise subjugates the rights 

of all creditors and their ‘ranking’ to the law of the country where the inferentis, 

RabelsZ 59 (1995) 1, 44 (‘Güterordnung’); Kieninger, Mobiliarsicherheiten 

173 et seq. (‘Güter- und Gläubigerordnung’); Kieninger, Das Statut der 

Forderungsabtretung im Verhältnis zu Dritten, RabelsZ 62 (1998) 679, 693 et 

seq. 

49 

Nagel / Gottwald, Internationales Zivilprozessrecht, 6th ed. (2007) § 17 I 2, 

IV 1a; Geimer, Internationales Zivilprozessrecht, 5th ed. (2005) nr. 405, 

3200, 3204; Schack, Internationales Zivilverfahrensrecht, 4th ed. (2006) § 19 

nr. 957. 

50 

Article 3(1) European Insolvency Regulation (EuInsVO); § 63(1) Austrian 

Bankruptcy Act (KO); § 3 German Insolvency Act (InsO). 

51 

Article 3(2) European Insolvency Regulation (EuInsVO); § 63(2) Austrian 

Bankruptcy Act (KO); § 354 German Insolvency Act (InsO). 

Axel Flessner 

27 


www.sellier.de


solvency proceedings take place, the lex fori concursus. 52 Although, on the 

other hand, both these enforcement proceedings focus only on property 

‘belonging’ to the debtor (a debtor is liable only with this property) 53 , the 

issue of where and under which law he acquired the property (or a comparable 

pledgable right) is of no concern for execution and insolvency law; 

both will accept what their forum’s conflict of laws rules together with the 

referred to substantive law of property will tell them about the property 

situation in the case at hand. The same applies to the remedies with which 

owners and other title-holders as third parties can assert their better rights 

in enforcement and insolvency proceedings. For these remedies, the only 

crucial factor is whether the claimed better right, with the substance given 

to it by the applicable law, deserves the protection, exemption or privilege 

provided by the available third party remedy in the forum’s enforcement 

or insolvency proceeding – this, by the way, being the main case for the 

view gaining ground that in a domestic proceeding the foreign property 

right can be honoured directly, without a (intellectually) prior conversion 

into a right recognised in the domestic numerus clausus list. 54 This kind 

of direct application suggests itself also because the law of the forum need 

not phrase its third party protection in terms of property law. Very often, 

rather, the procedural rules by which third party claims can be raised 

against the grip of the enforcement or insolvency proceeding are phrased 

wider than just referring to property law: there may be rights other than 

rights in rem that justify under certain circumstances third party opposition, 

preferred treatment, segregation or separate realization in a given 

forum state’s procedural law. 55 

52 

Article 4(1) and (2) (f-i) EuInsVO; §§ 221, 222 Austrian Bankruptcy Act 

(KO); § 335 German Insolvency Act (InsO). 

53 

Kerameus, Enforcement Proceedings, in International Encyclopedia of Comparative 

Law XVI chapter 10 (2002) sect 82; McBryde / Flessner / Kortmann 

(eds.), Principles of European Insolvency Law (2003) Section 3.1, General 

Commentary 35. 

54 

See footnotes 44, 45 above. 

55 

In Germany: § 771 Code of Civil Procedure (ZPO): ‘a right preventing the 

alienation’; § 805 ZPO: ‘a right of pledge or a preferential right’ that confers a 

‘claim to preferential satisfaction from the proceeds’; § 47 InsO (segregation): 

‘a right in rem or personal right’; § 49-51 (separation): ‘a right to separate satisfaction’. 

In Austria: § 44 KO: ‘right in rem or personal right to segregation’; 

§ 48 KO: ‘claims to separate satisfaction’. 

28 

Axel Flessner 


www.sellier.de


The same applies to the protection offered by the avoidance of acts, i.e. 

the action to set aside transactions causing inacceptable detriment to the 

creditors. For international cases, special conflict rules for this revocatory 

action have been developed in the pertinent statutes, case law and literature; 

they are understandably supposed to be mandatory. 56 Avoi dance in 

the international field does therefore not require assistance from a mandatory 

reference of the detrimental act of transfer to the law of the location 

of the transferred asset. 

In sum, a ‘creditors’ rights system’ ensures its application to cross-border 

cases through its own mandatory standards and rules of jurisdiction, remedies 

at law and choice of law for enforcement, insolvency and avoidability. 

In substantive domestic law, there is no greater creditor protection with 

regard to individual assets than through enforcement, insolvency proceedings 

and avoidance. Hence, there is no reason for international property 

law to make its own further contribution towards a mandatory choice of 

law in the interest of creditors. 

In referring to the ‘creditors’ rights system’, incidentally, the aforementioned 

authors also mean the legal policy pursued by a country to create 

a reasonable overall proportion between secured and unsecured credit, 

that is, to ‘control the volume of available asset-based credit security in 

the economy and thereby protect the class of unsecured creditors’. 57 The 

extent to which this sort of governmental policy is worth enforcing internationally 

against the interest in freedom of choice of law is taken up 

below in the discussion of security rights. 58 

V. Third parties as title-holders 

Freedom to choose the applicable law in property transactions is a serious 

policy problem if it affects the existing rights of other persons in the 

asset. This may occur under substantive law if a non-entitled party alienates 

or encumbers the asset and the other party is acting in good faith 

and therefore, under the applicable law, would acquire the granted right, 

56 

Article 4(2) (m) and Article 13 European Insolvency Regulation (EuInsVO); 

§ 339 InsO; Schwind, Internationales Privatrecht, nr. 508-515. 

57 

Kieninger, Mobiliarsicherheiten, 172. 

58 

Under footnote 80, 81. 

Axel Flessner 

29 


www.sellier.de


while the entitled party would thereby lose it – thus, where an asset is 

alienated by a party who is not the owner, or where it is alienated under 

concealment of an existing encumbrance (such as a lien or pledge), or 

where a limited right in rem is granted by a party who is not the owner or 

where it is granted by the owner, without disclosure of an existing prior 

security right. 

National property laws vary greatly in the way they regulate whether or 

not an indefeasible title, free from encumbrances, can be acquired in such 

cases. Something they have in common, however, is that they require 

actual good faith on the part of the acquirer, but there are a lot of variants 

when it comes to the other prerequisites. 59 Can the parties involved 

in the legal transaction be allowed to choose the legal system that seems 

most beneficial to them to the detriment of the title-holder who is not a 

party to the transaction? 

Even those authors who principally advocate party freedom of choice of 

law can get on a different track here. 60 Sometimes, it is even argued that 

in these cases there is no derivative acquisition at issue, but an ‘original’ 

acquisition by operation of law, just as with manufacturing, specification, 

accession and confusion. 61 What must be crucial, however, for a fitting 

designation of the applicable law is whether the third-party title-holder, 

who is at risk of being expropriated by a good faith purchase, has a legitimate 

interest in this expropriation being determined solely by the law of 

the location of the asset. Such an interest presupposes that the title-holder 

knows or is capable of knowing the country in which the asset is situated. 

This is not usually the case if he has lost the property; where, for instance, 

the stolen car, the stolen jewellery, the suitcase lost in transit is located, 

will mostly be uncertain when the loss is finally discovered, and it will be 

59 

See Thorn, Der Mobiliarerwerb vom Nichtberechtigten (1996); Salomons, 

How to Draft New Rules on the Bona Fide Acquisition of Movables for Europe? 

in Faber / Lurger (eds.), Rules for the Transfer of Movables (2008) 141; 

Armgardt, Der Schutz des gutgläubigen Erwerbers abhanden gekommener 

Sachen in den europäischen Rechtsordnungen und das quotale Lösungsrecht, 

ZEuP 2007, 1006. 

60 

Thus, for example, Ritterhoff, Parteiautonomie, see footnote 62 below; Stoll, 

Internationales Sachenrecht nr. 305: The situs state rules on good faith acquisition. 

A prior, divergent choice of law can be effective only after the asset 

has left the situs state. 

61 

Thus d’Avout, Solutions 604 et seq., 648 et seq. 

30 

Axel Flessner 


www.sellier.de


even more uncertain in which country the lost or stolen asset might be 

offered for sale. 

The situation is different where goods have been entrusted by the titleholder 

to another party (rental, commercial agency, transport, storage, 

security) who then disposes over the property without permission as an 

apparent title-holder. The basic agreement will usually also determine the 

location at which the bailee may have the property and use it. 62 However, 

this sort of contractual specification of location is absent when the contract 

permits the international transportation of the property or other 

international use of the property, such as with car rental or leasing or 

the creation of security rights in such goods. In addition, the contractual 

specification of the location means the contractual use of the property; the 

place of an unlawful disposal is not provided for. The title-holder cannot 

therefore expect under the contract that the property disposed of by the 

bailee in breach of the contract and without title, will in any event still 

be situated at the location provided for in the contract and not in any 

other country. 

It appears, thus, that an interest of the title-holder in being exposed to the 

risk of expropriation through a good faith acquisition precisely under the 

law of the place where the property is located frequently has only a weak 

foundation in law or does not exist at all. Moreover, this interest should be 

weighed against the interests of those involved in the disposal, to being able 

to choose the conditions for an effective legal acquisition independently. 

The latter interests are legitimate when both parties to the transaction are 

acting in good faith, in other words, if the disposing party also considers 

and may consider himself (mistakenly) to be the title-holder – for example, 

where property that has been lost or misappropriated is passed on in the 

selling chain. But even if only the acquiring party seems worthy of protection 

as acting in good faith, his interest deserves to be weighed against any 

existing (and perhaps legitimate) interest on the part of the title-holder 

in the application of the law of the location of the property. There is no 

policy reason to assume that the acquirer should have to succumb from the 

outset in this balancing act. The substantive property law of the different 

legal systems endeavours by means of various solutions to balance out the 

interests of the title-holder with those of the acquirer, in other words, on 

62 

Therefore in this case even Ritterhoff, Parteiautonomie 303, wants to repress 

the power to choose laws in favour of the mandatory reference to the property 

location. 

Axel Flessner 

31 


www.sellier.de


an abstract level: to balance preservation of the status quo and protection 

of bona fide acquisition. 63 This balancing process should also be part of 

the conflict of laws. Respect for the will of the disposing parties need not 

give way automatically to the lex situs idea. On the contrary, the interest 

of the transaction parties expressed in the agreed choice of law is always 

specific, whereas an interest of the title-holder in protection precisely by 

the law at the location of the property is actually possible only in certain 

scenarios and cannot be presumed. In the balancing of interests, therefore, 

the idea of having the parties choose their law by themselves need not be 

suppressed, even for acquisitions from non-entitled parties. 64 

Party freedom of choice of law to support an acquisition from a nonentitled 

party quickly falls under the suspicion of enabling the parties to 

act deliberately to the detriment of the title-holder. However, this way of 

acquisition is impossible from the outset if the prospective acquirer is aware 

of the non-entitlement of the disposing party – because then he will no 

longer be acting in good faith and will not be protected by any legal system. 

If he is not familiar with the non-entitlement, he may have chosen with 

conditional intent that legal system which would be most likely to protect 

him in his (still) good faith if it later should turn out that the other party 

actually had no title. Such conditional intent (intention to acquire, even 

in the event that the disposing party is not entitled to dispose) can never 

be ruled out under substantive property law and it can not automatically 

justify to find the acquirer in bad faith; otherwise an acquisition from a 

non-entitled party – after all a solution provided for by the law – would 

scarcely be possible. Instead, the acquirer may settle for the appearance 

of title (provided, for instance, by possession, registration, publication of 

sale, official involvement), on which the valid acquisition from the nonentitled 

party will be based under the applicable property law. 

There is no reason to be stricter at the conflict of laws level. Whether the 

acquirer should have been suspicious and therefore should have made 

63 

Presentation in Thorn, Mobiliarerwerb; Salomons in Faber / Lurger, Rules for 

the Transfer of Movables 141; Armgardt, ZEuP 2007, 1006. 

64 

Also Stoll, Internationales Sachenrecht, nr. 304-308 (subject to the exception 

given in footnote 60); Einsele, RabelsZ 60 (1996) 442 et seq.; Ritterhoff, 

Parteiautonomie 303-307 (subject to the exception given in footnote 60). 

According to von Wilmowsky, Europäisches Kreditsicherungsrecht 355-360, 

the power of the parties to choose the applicable law also for determining good 

faith acquisition is even required by European law. 

32 

Axel Flessner 


www.sellier.de


enquiries depends largely on the circumstances. 65 If the parties involved 

have chosen an otherwise remote but purchaser-friendly law to apply, 

this may be a suspicious circumstance in itself; if, on the other hand, 

they may have had other good reasons for choosing the law (for example: 

adaptation to the underlying obligation, foreseeable cross-border change 

of the asset’s location, shared country of residence of the parties involved, 

familiarity with the chosen law), the fact that they may also know that the 

chosen law looks favourably upon acquirers will be no reason for imposing 

additional burdens of investigation on the acquirer. 

E. Security rights 

The closer analysis of interests carried out up to this point has shown 

that a mandatory reference to the law of the place where the property is 

located is not necessary for the transfer and creation of property rights 

by a legal transaction and that the power of the parties to choose the 

applicable law is in fact even superior. However, party freedom to choose 

the law must pass the litmus test in relation to security rights – the most 

common application of international property law in judicial practice and 

the hardest, because there is a particularly marked and varied difference 

between national laws in the securing of credit through movable property, 

and countries use to operate clear legal policies in this field. The differences 

relate to the necessary publicity, to protection of the debtor against 

ill-considered moves, against economic constriction of debtors during 

the credit period, against cheating and recklessness in the realization of 

the security, as well as to the protection of unsecured creditors against 

premature and excessive depletion of the debtor’s assets with which the 

debtor is principally liable towards the creditors. Particularly striking are 

the different attitudes towards non-possessory security, towards particular 

types of credit (supplier credit, money lending), towards particular types of 

collateral (current assets, equipment and fixed assets, vehicles, the whole 

enterprise) and towards particular security conditions (security in future 

assets, in bulk assets). 66 Can the parties involved be allowed to pick out 

one of these national credit security systems? 

65 

Thus also Ritterhoff, Parteiautonomie 304. 

66 

See, for instance, the accounts by Kieninger, Mobiliarsicherheiten 23 et seq.; 

Kieninger (ed.), Security Rights in Movable Property in European Private 

Axel Flessner 

33 


www.sellier.de


It is important to keep in mind initially that the physical enforcement 

of the security right in the event of realization (seizure and sale, division 

of the proceeds) and the legal defence against that enforcement are not 

affected by the choice of law agreement, if the realization takes place with 

the help of forced execution and insolvency law. The unalterable application 

of forum law in enforcement and insolvency proceedings is nowhere 

disputed 67 and, as to insolvency, has just been reaffirmed in the European 

Insolvency Regulation (EuInsVO). 68 Some of the reservations about party 

freedom of choice of law that are derived from the idea of a single state’s 

‘creditors’ rights system’ should therefore appear to be settled 69 , because in 

the view of the proponents of that idea the ‘system’ necessarily includes 

the state’s insolvency and enforcement law. 70 Its precedence remains unaffected 

by party autonomy. From the very outset, it is only the provisions 

on creation, transfer and legal content in the security period (until the 

credit is repaid or the security liquidated) that can be reached through 

the choice of law agreement. 

Allowing party autonomy with this limited effect also for security rights 

has a model in Article 14 of the Rome I Regulation. That provision leaves 

the law governing the third party effects of assignment of claims uniformly 

to the will of the parties, regardless of whether the assignment is meant to 

be final (as in purchases and gifts) or to serve as a credit security. The equal 

treatment in the conflict of laws of the property aspects of a sale and of the 

establishing of security rights is therefore already current European law; it 

is consequently obvious in the case of tangible property too. In addition, 

tangible goods and money claims as collateral in certain transactions such 

as extended reservation of title, security transfer of changing stocks of 

goods are treated by the parties as substitutable in any event. 

In the provision of security for credit there is also a particular need for the 

parties to have one and the same law govern the credit agreement and the 

security (which is usually agreed along with the credit). In the case of pure 

transfers of property, such as a sale or gift, delivery performs the contract 

and leaves it behind, so that remaining contractual rights can now be 

Law (2004); Sigman / Kieninger (eds.), Cross-Border Security over Tangibles 

(2007); Plappert, Dingliche Sicherheiten in der Insolvenz (2008). 

67 

See footnotes 49-53 above. 

68 

Article 4, 5, 7 EuInsVO. 

69 

See footnote 48 above. 

70 

Particularly Kieninger, Mobiliarsicherheiten, 174. 

34 

Axel Flessner 


www.sellier.de


viewed in isolation under the conflict of laws. In the case of credit, on the 

other hand, the provided security right must support the credit relationship 

throughout its duration, in other words, until the credit amount is 

settled. Under Article 3 of the Rome I Regulation, the parties can choose 

the law for the credit agreement. In this case, the legal situation for the 

parties is simplified if they are able to regulate the creation and content 

of the security right under the same legal system. 

Finally, on the European level freedom of the parties to choose the applicable 

law shields cross-border secured credit from attacks inspired by European 

Union law. The predominant view now is that freedom of movement 

of goods and capital is restricted if the choice of law reference to the place 

where the property is located results in security rights being invalidated 

or even simply having to be adapted, once the property arrives in another 

Member State whose law does not recognise the security right granted or 

its creation abroad. 71 A divergence in opinions remains only as to whether 

compelling reasons of the common good in a Member State, in the meaning 

developed by the European Court of Justice, can justify this effect. 72 

If the parties are able to select the applicable law for international secured 

credit, security rights then are immune to such challenges, because the 

freedom to choose laws precludes any reproach of an infringement of the 

fundamental freedoms in the European market. 73 

There are, after all, good additional grounds for freedom of the parties 

to choose the applicable law particularly in the case of in rem security 

rights. 74 However, the mandatory application of individual rules in this area 

of law remains to be taken into account, as in many countries a strong 

71 

Basedow, Rabelsz 59 (1995) 43; Kieninger, Mobiliarsicherheiten 152 et seq.; 

von Wilmowsky, Europäisches Kreditsicherungsrecht 97-105, 108-115; Röthel, 

JZ 2003, 1032; W. Roth, Secured Credit and the Internal Market: The Fundamental 

Freedoms and the EU’s Mandate for Legislation, in: Eidenmüller / 

Kieninger, The Future of Secured Credit in Europe, European Company and 

Financial Law Review (ECFR), Special Volume 2 (2008) 36, 50 et seq.; further 

references in Verschraegen in Rummel, ABGB II 3d ed., § 31 IPRG nr. 30. 

72 

For justification: Kieninger, Mobiliarsicherheiten 173 et seq.; Röthel, JZ 2003, 

1032-1035. Against justification: Basedow, RabelsZ 59 (1995) 45 et seq., von 

Wilmowsky, Europäisches Kreditsicherungsrecht 122-151. 

73 

Kropholler, Internationales Privatrecht 75 with further references. 

74 

In agreement: Eidenmüller, Secured Creditors in Insolvency Proceedings, in: 

Eidenmüller / Kieninger (footnote 71 above) 271, 281-283. 

Axel Flessner 

35 


www.sellier.de


policy intention lies behind them. The publicity of security rights, which is 

taken very seriously in many countries 75 , is singled out here as an example. 

Is this publicity a reason for generally rejecting freedom of choice of law 

for security rights (even extending beyond enforcement and insolvency 

law!) or in any case placing limits on it in the form of ‘overriding mandatory 

provisions’ or as part of the forum state’s public policy (ordre public)? 

This question is addressed here using Austria as an example. In Austria, 

transfer of possession is required by the Civil Code (ABGB) for the creation 

of a pledge (§ 452 ABGB). This requirement has been extended by 

case law and doctrine to any kind of security by movable assets. It may not 

be circumvented by a transfer of ownership (which, other than pledging, 

could be effected through constructive possession, in other words, by mere 

agreement, § 428 ABGB). If a transfer of ownership is actually intended by 

the parties to serve as a credit security, the asset must therefore be handed 

over physically (as with a pledge) or another means of publicity must be 

used that is expressly permitted by statute. 76 For conflict of laws, this publicity 

requirement is strengthened, but also limited, by § 31 of the International 

Private Law Act (IPRG), which provides for application of the 

law where the asset is located to the property law transaction. The strict 

publicity requirement therefore applies to the creation of security rights in 

any tangible assets situated in Austria, but not for property located abroad. 

In neither case does it matter where the parties to the credit transaction 

are resident and active (whether that location is in Austria or abroad, in 

the same country or in different countries), where the credit transaction is 

concluded, where and in which currency the credit amount is paid out, received 

and paid back, or by which law the credit agreement is governed. 

For the credit agreement, the parties have a free choice of law (Article 3 of 

Rome I) with which they can take account of such circumstances. In the 

case of credit against security, selecting the security instrument is at the 

heart of the contractual agreement. The parties will therefore also be interested 

in choosing the law for this part of their transaction, rather than 

being legally fixed to the location of the asset – which at the time of the 

agreement may be quite insignificant to them under the circumstances. 

75 

In detail: von Wilmowsky, Europäisches Kreditsicherungsrecht 153 et seq.; 

Kieninger, Mobiliarsicherheiten 23 et seq.; Kieninger (ed.), Security Rights; 

Sigman / Kieninger (eds.), Cross-Border Security. 

76 

Eccher in Koziol / Bydlinski / Bollenberger (KBB), Kurzkommentar zum ABGB 

(2005) § 359 nr. 5; Koch in KBB 2d ed. (2007) § 451 nr. 2. 

36 

Axel Flessner 


www.sellier.de


Publicity of security rights through transfer of possession should serve, to 

be sure above all those people who may extend additional credit to the 

debtor and security provider, in other words, the further and subsequent 

creditors, who would otherwise gain a false impression of the debtor’s 

financial standing. It has already been shown that this creditor interest, 

insofar as it relates to particular pieces of property, has only a weak 

foundation in law and in fact in the time before insolvency and enforcement. 

77 Compared to the interest of the parties to the credit agreement 

in choosing the applicable law, it loses further weight due to the fact that 

Austria focuses its requirement for publicity of the security right not on 

the international credit market or on the protection of creditors of Austrian 

debtors or even on the protection of Austrian creditors of Austrian 

debtors, but rather exclusively on the question whether a piece of property 

intended to secure a loan is located in Austria. Money lending, however, 

which is to be secured by movable assets, is to a large extent internationally 

volatile because the market for lenders and borrowers is wide open 

internationally. An interest, even a public (?) interest, to have publicity 

of credit security rights in tangible movables for the benefit of ‘the 

creditors’ when, and only when, the security collateral is in Austria seems 

scarcely purposeful in this market, rather like an at random policy. In the 

neighbouring field of the assignment of claims for security, the Supreme 

Court has expressly denied a public interest on the part of Austria in the 

international enforcement of the Austrian publicity rule. 78 Such an interest 

should therefore only have a limited bearing when weighing it against 

the contrary interest of parties to the credit agreement, which they have 

expressed in a choice of law clause for the security right. 

The Austrian Supreme Court ruled in 1983, on the other hand, that the 

Austrian publicity requirement should even apply to a tangible asset subject 

to a non-possessory security right validly created under a foreign law 

while the asset was located in that jurisdiction; if the asset were later to 

arrive in Austria, the foreign non-possessory security right would not be 

recognised there. 79 If this precedent were followed nowadays, Austrian law 

would find itself in the sights of the European Union (to which – then the 

European Community – Austria did not belong at the time). Not recog- 

77 

See footnote 49 et seq. 

78 

OGH (Supreme Court) in JBl 1992, 189 = IPRax 1992, 47. 

79 

OGH in SZ 56 / 188 = IPRax 1985, 165. The decision has been welcomed in 

the literature, see Verschraegen in Rummel, ABGB II 3d ed., § 31 IPRG nr. 28, 

29; detailed criticism: Schwind, IPR nr. 393. 

Axel Flessner 

37 


www.sellier.de


nising the security right that has validly been created in another Member 

State in the Member State of the asset’s new location is the most clear-cut 

case of an impediment to the free movement of goods and capital, and it 

is hard to justify such a decision on compelling reasons of Member State 

public policy, since for the enforcement of the ‘imported’ security right, 

precedence is anyhow reserved to the enforcement and insolvency rules 

at the asset’s new location. 80 

The considerations developed here based on the example of the Austrian 

publicity requirement can be generalised. Each publicity requirement on 

the part of a country for security rights, including the need for registration, 

must be questioned as to its international scope because of the actually 

existing and in the European Union even legally guaranteed freedom of 

cross-border movement of money and credit. From a European perspective, 

it seems that even if a Member State prescribes for security rights 

a certain publicity without exception, such requirement may be applied 

only to purely domestic cases in that Member State. For cross-border traffic, 

on the other hand, the different publicity rules of individual states may 

only be regarded as possibilities from which the credit parties should be 

able to choose. 

F. Europe 

The encouragement from Europe taken up here allows a perspective for 

Europe too. Freedom to choose the applicable law for property transactions 

concerning movable assets is required in Europe up to the reserved 

realm of enforcement and insolvency law. This means that the substantive 

property law of the individual jurisdiction in relation to tangible 

goods, hitherto fixed to the soil by the reference to the situs, becomes 

movable just like its subject matter, a title structuring instrument in the 

hands of the parties. The parties have the choice between the property 

laws of the Member States. They may only be limited in their freedom of 

choice by a Member States’ so-called ‘overriding mandatory provisions’ 

and by the reserved national ‘public policy’ (‘ordre public’) – which, in 

turn, must be compatible with Union law. Even mandatory publicity or 

another such provision, which a Member State of the European Union 

only wants to have for its ‘own’ domestic market, to some extent comes 

80 

See footnote 49-54. 

38 

Axel Flessner 


www.sellier.de


into conflict with the guiding idea of the European Single Market. The 

European money and credit market is so highly integrated, even more so 

due to its shared currency, that a mandatory national regulation, even 

if self-restraint is explicitly intended, will have an impact on the entire 

market such that self-restraint would in truth be a contradiction in terms. 

This applies particularly to mandatory requirements for security rights, 

with which a Member State intends to put a squeeze on secured credit 

and give unsecured credit its chance of satisfaction. In the harmonised 

money and credit market, this sort of legal policy directed only at a single 

Member State’s market can merely be referred to as pointless. Against the 

parties’ interest in choosing the applicable law by themselves, it merits 

no European support. 

On the distant horizon is a single ‘European Security Right’. 81 It will be 

offered to the parties as an alternative to Member State security rights. 

It would free legal dealings from the agony of choosing between Member 

State laws, but it would also show through its optional nature that 

party choice of law for property transactions is a practicable solution. 

It is the only realistic way in the world in which the economic players, 

their goods and their money actually move freely – and in Europe with a 

constitutional guarantee – in the international market and across local 

legal boundaries. 

81 

For reflections on this subject, see Kreuzer, Conflict-of-Laws Rules for Security 

Rights in Tangible Assets in the European Union, in: Eidenmüller / Kieninger, 

Future of Secured Credit 297, 314 et seq.; Flessner, Security Interests in 

Receivables – A European Perspective, in: Eidenmüller / Kieninger, Future of 

Secured Credit 336, 340 et seq.; further references in Röthel, JZ 2003, 1034 

footnote 101. 

Axel Flessner 

39 


www.sellier.de


www.sellier.de

2. 


Dutch International Property Law 

Jeroen van der Weide * 

2.1. Introduction 

On 3 February 2010, the French Cour de cassation delivered a judgment on 

the law that was applicable in a case involving seven paintings. 1 In 2000, 

a French-born artist living in New York City had furnished the paintings 

to the American owner of a New York restaurant. In 2005, the artist died, 

and the following year the restaurant was forced to close down because of 

financial problems. The owner of the restaurant then took the paintings to 

France to auction them. In the summer of 2007, the widow of the French 

painter applied to a French court for leave to attach the paintings before 

they were sold. Leave was granted, but the French auction house and the 

American restaurant owner petitioned the court to lift the attachment, 

which the court did in December 2007. The widow appealed to the Paris 

Cour d’appel, which dismissed the appeal. She then appealed to the Cour 

de cassation. 

It will be clear that the central issue in this procedure concerns ownership 

of the paintings. Under French law, a person who possesses movable 

property, and believes he is the rightful owner of the property is presumed 

to be the owner. As was stated in the former Article 2279 (Article 2276 

revised version) of the French Civil Code (Code civil): ‘En fait de meubles, 

la possession vaut titre’, possession in good faith constitutes prima facie evidence 

of ownership. Under American law, however, mere possession does 

* 

Associate Professor in Civil Law, Leiden Law School. 

1 

Cour de cassation, Première chambre civile, 3 February 2010, Arrêt no. 121 

(08-19.293). Commented by Étienne Cornut, La Semaine Juridique (Édition 

Générale) 2010, no. 176, p. 332; Louis d’Avout, L’inexorable territorialité du 

droit français des biens, ou comment la tradition peut aboutir à l’injustice, La 

Semaine Juridique (Édition Générale) 2010, p. 531-534; Caroline Cohen, Rev. 

crit. dr. int. pr. 2010, p. 485-494. The decision of the Cour de cassation may 

be consulted online at www.courdecassation.fr. 

41 


www.sellier.de


not transfer ownership to the possessor. If it had been held in this case that 

American law applied, the beneficiary – the American restaurant owner – 

would have had to prove that he had received the paintings as a gift. 

In its decision of 3 February 2010, the French Cour de cassation confirmed 

its earlier case law 2 and ruled that French law, and French law alone, 

governs issues of proprietary rights in movables situated in France: 

‘la loi française est seule applicable aux droits réels dont sont l’objet des biens 

mobiliers situés en France.’ 

Since the paintings had reached French soil, French law applied, and 

Article 2279 provided sufficient basis to rule that title to the paintings 

had been transferred. 3 

2.2. The lex situs rule 

The decision of the French Cour de cassation illustrates the dominance 

of the lex situs rule in matters of international property law. This conflict 

of laws rule says that international property law issues, regarding both 

movables and immovables, are governed by the lex situs, the law of the 

place where the goods are situated. This rule is found not only in French 

2 

See for example Cour de cassation 8 July 1969, Rev. crit. dr. int. pr. 1971, 

p. 75-86, comm. Philippe Fouchard; Journal du Droit International (Clunet) 

(JDI) 1970, p. 916-923, comm. Jean Derruppé; Juris-Classeur périodique (JCP) 

1970(II), nr. 16182, comm. Hélène Gaudemet-Tallon. See also Cour de cassation 

3 May 1973, Rev. crit. dr. int. pr. 1974, p. 100-109, comm. E. Mezger; JDI 

1975, p. 74-82, comm. Philippe Fouchard. 

3 

Critical on this point: D’Avout 2010, p. 532 (note 1): ‘Si, en France, les 

meubles litigieux avaient été vendus ou bien saisis par un tiers de bonne foi, 

l’application de la loi française aurait été justifiée. Au cas contraire, le recours 

à la loi de situation actuelle constitue une solution seulement formelle des 

conflits de lois, qui manque d’égard pour la particularité du litige et se réduit 

au fond à l’application pure et simple de la lex fori (c’est- à-dire à une transformation 

du litige international à un litige interne ordinaire).’ According 

to Article 2 of the Dutch Property Law (Conflict of Laws) Act, American 

instead of French law would have been decisive, since the paintings had been 

furnished in New York to the American restaurant owner. 

42 



www.sellier.de

2. Party Autonomy in Dutch International Property Law 

law 4 but also, for example, in English, 5 German, 6 Belgian, 7 Swiss, 8 and 

Dutch law. 9 

4 

Article 3 paragraph 2 of the French Civil Code in its official English translation 

reads as follows: ‘Immovables are governed by French law even when 

owned by aliens.’ (‘Les immeubles, même ceux possédés par des étrangers, sont 

régis par la loi française.’). 

5 

For English law, see for example Janeen M. Carruthers, The Transfer of Property 

in the Conflict of Laws. Choice of Law Rules concerning Inter Vivos 

Transfers of Property, Oxford: Oxford University Press 2005, p. 19 et seq. and 

Chapter 8; Dicey, Morris and Collins, On the Conflict of Laws (Volume 2), 

London: Sweet & Maxwell 2006, Chapter 22; Cheshire, North & Fawcett, 

Private International Law, Oxford: Oxford University Press 2008, Chapter 

27 et seq. 

6 

Article 43 paragraph 1 of the German Conflict of Laws Code (Einführungsgesetz 

zum Bürgerlichen Gesetzbuche) in its official English translation reads as 

follows: ‘Interests in property are governed by the law of the country in which 

the property is situated.’ (‘Rechte an einer Sache unterliegen dem Recht des 

Staates, in dem sich die Sache befindet.’). 

7 

Article 87 paragraph 1 of the Belgian Conflict of Laws Code (Wetboek van 

Internationaal privaatrecht) reads as follows: ‘Rights in rem to goods are governed 

by the law of the State in whose territory the goods are situated at the 

time of their enforcement.’ (‘De zakelijke rechten op een goed worden beheerst 

door het recht van de Staat op wiens grondgebied dit goed zich bevindt 

op het tijdstip dat zij worden ingeroepen.’). 

8 

Article 99 paragraph 1 of the Swiss Conflict of Laws Code (Bundesgesetz über 

das Internationale Privatrecht) reads as follows: ‘Rights in rem to immovables 

are governed by the law of the place where the immovable is situated.’ (‘Dingliche 

Rechte an Grundstücken unterstehen dem Recht am Ort der gelegenen 

Sache.‘). Article 100 paragraph 1 of the Swiss Conflict of Laws Code reads 

as follows: ‘Acquisition or loss of rights in rem to movables is governed by the 

law of the State in whose territory the goods are situated at the time of their 

acquisition or loss.’ (‘Erwerb und Verlust dinglicher Rechte an beweglichen 

Sachen unterstehen dem Recht des Staates, in dem die Sache im Zeitpunkt 

des Vorgangs, aus dem der Erwerb oder der Verlust hergeleitet wird, liegt.‘). 

9 

Article 2 paragraph 1 of the Dutch Property Law (Conflict of Laws ) Act (Wet 

conflictenrecht goederenrecht) reads as follows: ‘Unless otherwise provided 

in paragraph 2 or paragraph 3, the proprietary rights regime applying to goods 

is governed by the law of the state in whose territory the goods are situated.’ 

(‘Behoudens voor zover in het tweede en het derde lid anders is bepaald, wordt 


43 


www.sellier.de


There are several reasons that the lex situs rule is undisputed and universally 

accepted. First of all, the place where the goods are situated is a very 

obvious connecting factor for determining which proprietary rights law is 

applicable. In addition, the link with the law of the place where the goods 

are situated serves the interest of facilitating international private law 

transactions, since third parties should be able to rely on the assumption 

that goods situated in a country are governed by the law of that country, 

among other things because proprietary rights law systems typically form 

a closed system (numerus clausus). Furthermore, universal applicability of 

the lex situs rule makes for harmonised decisions – which in German law 

is called Entscheidungseinklang – and thus also for legal certainty. 10 

The principle of party autonomy is at odds with these proprietary rights 

principles. This is expressed very clearly once again, for example, in the 

Outline Edition of the Draft Common Frame of Reference, published in 

2009 by the Study Group on a European Civil Code and the Research 

Group on EC Private Law: 

‘The principle of party autonomy has to be considerably modified in 

property law. Because proprietary rights affect third parties generally, 

the parties to a transaction are not free to create their own basic rules 

as they wish. They cannot, for example, define for themselves basic 

concepts like ‘possession’. Nor are they free to modify the basic rules 

on how ownership can be acquired, transferred or lost.’ 11 

het goederenrechtelijke regime met betrekking tot een zaak beheerst door het 

recht van de staat op welks grondgebied de zaak zich bevindt.’). 

10 

See for example G.C. Venturini, Property, in: Kurt Lipstein & R. David a.o. 

(eds.), International Encyclopedia of Comparative Law (Volume III, Private 

International Law, Chapter 21), Tübingen: J.C.B. Mohr (Paul Siebeck) 1976; 

Kurt Siehr, Internationales Sachenrecht. Rechtsvergleichendes zu seiner Vergangenheit, 

Gegenwart und Zukunft, Zeitschrift für Vergleichende Rechtswissenschaft 

(ZvglRWiss) 2005, p. 145-162. 

11 

Christian von Bar a.o. (eds.), Principles, Definitions and Model Rules of European 

Private Law. Draft Common Frame of Reference (DCFR). Outline 

Edition, München: Sellier. European law publishers 2009, Principle no. 14, 

p. 70. 

44 



www.sellier.de


2.3. The Dutch Property Law 

(Conflict of Laws) Act in general 

In the Netherlands, the Dutch Property Law (Conflict of Laws) Act (Wet 

conflictenrecht goederenrecht) entered into force on 1 May 2008. 12 This Act 

merely lays down conflict rules that determine which law is applicable to 

cross-border proprietary rights issues. 

Conflict of laws is the last large area of Dutch Civil Law still awaiting 

codification. The Dutch legislature has opted for a system of successively 

codifying subareas, with the ultimate aim of bringing them together in 

one Conflict of Laws Code, which will then be Book 10 of the Dutch 

Civil Code. At present, conflict of laws rules have been codified for approximately 

sixteen subareas, ranging from the Registered Partnerships 

(Conflict of Laws) Act to the Corporations (Conflict of Laws) Act. 

The Dutch Property Law (Conflict of Laws) Act was the last area to be 

codified. Its entry into force on 1 May 2008 completed the codification 

process. In September 2009, the Act adopting and implementing Book 10 

of the Civil Code (Vaststellings- en Invoeringswet Boek 10 Burgerlijk Wetboek) 

was sent to the Dutch Parliament. 13 In all probability, Book 10 of the 

Civil Code will be current law by 2011 or 2012. 14 The conflict of laws rules 

12 

Act of 25 February 2008 implying conflict of laws rules on the proprietary 

rights regime applying to goods, claims, and securities (Wet van 25 februari 

2008, houdende regeling van het conflictenrecht betreffende het goederenrechtelijke 

regime met betrekking tot zaken, vorderingsrechten, aandelen 

en giraal overdraagbare effecten), Staatsblad van het Koninkrijk der Nederlanden 

2008, 70. Commented by J.A. van der Weide, in: J.H. Nieuwenhuis 

a.o. (eds.), Tekst & Commentaar Vermogensrecht, Deventer: Kluwer 2009, 

Annex 23 (p. 1689-1734). See also Praxis des Internationalen Privat- und 

Verfahrensrechts (IPRax) 2008, p. 560-562 for a German translation of the 

Dutch Property Law (Conflict of Laws) Act, with an introductory article by 

Teun Struycken & Bartosz Sujecki, Das niederländische Gesetz zur Regelung 

des internationalen Sachenrechts, IPRax 2008, p. 558-559. 

13 

Tweede Kamer der Staten-Generaal, vergaderjaar 2009-2010, 32 137, no. 2 

(text) and no. 3 (commentary). 

14 

On 28 September 2010, the Act adopting and implementing Book 10 of the 

Dutch Civil Code was adopted by the Dutch House of Representatives and 

sent to the Senate. See Eerste Kamer der Staten-Generaal, vergaderjaar 2010- 


45 


www.sellier.de


on the proprietary rights regime applying to goods, claims, and securities 

will be regulated in Chapter 10 (Articles 126-141) of Book 10. 

The Dutch Property Law (Conflict of Laws) Act regulates three categories 

of property: (a) goods, (b) claims, and (c) shares and bonds (to be 

distinguished into bearer shares, registered shares, book-entry shares, and 

bonds). However, it by no means contains an exhaustive set of rules. For 

example, it does not give rules for goods traded by means of nego tiable 

instruments, such as bills of lading. According to the legislature, this subject 

had not yet sufficiently crystallised to be included in legislation. 15 

Other elements of international property law for which the Dutch Act 

does not provide include intellectual property rights, quasi-property law 

rights, factoring, and time-sharing. 16 

2.4. Content of the Dutch Property Law 

(Conflict of Laws) Act 

The Dutch Property Law (Conflict of Laws) Act has six chapters. 17 Following 

an introductory first Chapter, Chapter 2 (proprietary rights regime 

of goods) opens in Article 2 paragraph 1 by laying down the lex situs rule. 

It reads as follows: 

‘Unless otherwise provided in paragraph 2 or paragraph 3, the proprietary 

rights regime applying to goods is governed by the law of the state 

in whose territory the goods are situated.’ 18 

2011, 32 137, A. Recently, on 17 May 2011, Book 10 of the Dutch Civil Code 

has been adopted by the Dutch Senate. 

15 

See Eerste Kamer der Staten-Generaal, vergaderjaar 2007-2008, 30 876, nr. 

C, p. 12. 

16 

See Tweede Kamer der Staten-Generaal, vergaderjaar 2006-2007, 30 876, 

nr. 6, p. 2. 

17 

This Act will be adopted in Book 10, Chapter 10 (Articles 126-141) of the 

Dutch Civil Code without major textual changes. 

18 

In its original Dutch version, Article 2 paragraph 1 of the Dutch Property Law 

(Conflict of Laws) Act (Article 10:127 paragraph 1 Civil Code) reads as follows: 

‘Behoudens voor zover in het tweede en het derde lid anders is bepaald, 

wordt het goederenrechtelijke regime met betrekking tot een zaak beheerst 

door het recht van de staat op welks grondgebied de zaak zich bevindt.’ 

46 



www.sellier.de


Paragraph 4 of this Article contains a non-exhaustive list of issues that 

fall under the proprietary rights regime applying to goods: for example, 

the question of which requirements must be satisfied to effect a transfer 

of goods and which rights can vest in goods. 

Article 2 is followed by a number of special provisions relating – in a 

conflict of laws setting – to retention of title clauses and leasing (Article 

3), the right of lien (Article 4), the doctrine of changes in the connecting 

facts – known in French as conflit mobile and in German as Statutenwechsel 

– (Article 5), 19 protection of third parties (Article 6), 20 transfer of 

title to stolen goods to the insurer (Article 7), and the proprietary rights 

regime applicable to goods in transit (Article 8). 21 

International proprietary rights issues concerning claims and shares are 

regulated in Chapter 3 (Articles 9-11) and Chapter 4 (Articles 12-14). 

19 

Article 5 of the Dutch Property Law (Conflict of Laws) Act (Article 10:130 

Civil Code) reads as follows: ‘Proprietary rights to goods that have been acquired 

or created in accordance with the law that is applicable pursuant to 

this Act will continue to exist, even if the goods are moved to another state. 

These rights may not be enforced in a way that is incompatible with the 

law of the state in whose territory the goods are situated at the time of the 

enforcement.’ (‘Rechten op een zaak, die overeenkomstig het ingevolge deze 

wet toepasselijke recht zijn verkregen of gevestigd, blijven daarop rusten, ook 

wanneer die zaak wordt overgebracht naar een andere staat. Deze rechten 

kunnen niet worden uitgeoefend op een wijze die onverenigbaar is met het 

recht van de staat op welks grondgebied de zaak zich ten tijde van die uitoefening 

bevindt.’). See specifically H.L.E. Verhagen, Conflit mobile bij roerende 

zaken: assimilatie of transformatie? Een bijdrage over roerende zaken in het 

internationaal privaatrecht (Studiekring ‘Prof. mr. J. Offerhaus’, nieuwe reeks 

no. 10), Deventer: Kluwer 2007. 

20 


Civil Code) reads as follows: ‘The legal effects of the acquisition of goods from 

a party who does not have the power to dispose of the goods are governed by 

the law of the state in whose territory the goods are situated at the time of 

acquisition.’ (‘De rechtsgevolgen van de verkrijging van een zaak van een 

beschikkingsonbevoegde worden beheerst door het recht van de staat op welks 

grondgebied de zaak zich ten tijde van die verkrijging bevond.’). 

21 

Articles 3-4 and 7-8 of the Dutch Property Law (Conflict of Laws) Act will 

be discussed in paragraph 6. 


47 


www.sellier.de


Both types of property rights can be either bearer rights or rights of a 

known title holder, hence it is necessary to lay down different conflict of 

laws rules. Nowadays, however, securities transactions are often effected 

by book entry. Chapter 5 (Articles 15-16) is tailored to this new development 

and contains two provisions concerning the proprietary rights regime 

that applies to book-entry shares and bonds. The last chapter, Chapter 

6 (Articles 17-23), contains a number of final provisions. 

2.5. Party autonomy in general 

Because the situs rule is characterised by invariability, foreseeability and 

unambiguousness, it enhances legal certainty and protects the interests of 

smooth private-law transactions. 22 But while the situs rule is static, party 

autonomy offers dynamism. Choice of law gives parties the possibility of 

arranging their mutual legal relationship as they please. In a certain sense, 

party autonomy and the situs rule are thus each other’s opposites. They are 

at the far ends of the spectrum of proprietary rights law. 23 

It is impossible to think of modern private international law without the 

instrument of choice of law. Via international contract law, in which the 

22 

See for example Ann-Christin Ritterhoff, Parteiautonomie im internationalen 

Sachenrecht. Entwicklung eines Vorschlags insbesondere für das deutsche 

Kollisionsrecht unter vergleichender Berücksichtigung des englischen Kollisionsrechts 

(Schriften zum Internationalen Recht, no. 114), Berlin: Duncker 

& Humblot 1999, p. 53 et seq. Cf. also Gerhard Kegel & Klaus Schurig, Internationales 

Privatrecht, München: C.H. Beck 2004, p. 765: ‘Das Verkehrsinteresse 

verlangt daher die Herrschaft der lex rei sitae.’ 

23 

Cf. Louis d’Avout, Sur les solutions du conflit de lois en droit des biens, Paris: 

Éd. Economica 2006, p. 587 et seq.; J.A. van der Weide, Mobiliteit van goederen 

in het IPR. Tussen situsregel en partijautonomie (serie Recht en Praktijk, 

no. 142), Deventer: Kluwer 2006, paragraph 6.2; Axel Flessner, Rechtswahl 

im internationalen Sachenrecht – neue Anstöße aus Europa, in: Peter 

Apathy a.o. (eds.), Festschrift für Helmut Koziol, Wien: Jan Sramek Verlag 

2010, p. 125-146 (in translated version: see Chapter 1 in this book on Party 

Autonomy in International Property Law, Choice of Law in International 

Property Law, New Encouragement from Europe). 

48 



www.sellier.de


choice of law instrument has its origins, 24 party autonomy has found its 

way into numerous subareas of private international law. One finds choice 

of law, for example, in conflict of laws rules regarding liability for unlawful 

acts, 25 matrimonial property law, 26 succession law, 27 and divorce law. 28 

In these branches of law, the relationship between the parties is the central 

element: the legal relationship between offender and offended, between 

spouses, between testator and heirs, and between former spouses. 

Things are different in the field of proprietary rights law, where the legal 

relationship to third parties is at least as significant as the relationship 

between the parties themselves. This is due to a characteristic feature of 

proprietary rights: namely, that they are enforceable against third parties. 

Precisely this aspect is considered to be an important obstacle to the universal 

application of party autonomy in international property law. 29 

2.6. Party autonomy in Dutch international property law 

Nevertheless, there is a clearly perceptible tendency, even in the field 

of international property law, to allow parties greater flexibility in their 

24 

Cf. Article 3 of the Regulation (EC) No 593 / 2008 of the European Parliament 

and of the Council of 17 June 2008 on the law applicable to contractual 

obligations (Rome I), OJ 2008, L 177 / 6. 

25 

Cf. Article 14 of the Regulation (EC) No 864 / 2007 of the European Parliament 

and of the Council of 11 July 2007 on the law applicable to noncontractual 

obligations (Rome II), OJ 2007, L 199 / 40. 

26 

Cf. Articles 3 and 6 of the Convention of 14 March 1978 on the Law Applicable 

to Matrimonial Property Regimes. See also Article 16 of the Proposal for 

a Council Regulation on jurisdiction, applicable law and the recognition and 

enforcement of decisions in matters of matrimonial property regimes, Brussels, 

16 March 2011, LOM (2011) 126 final. 

27 

Cf. Article 5 of the Convention of 1 August 1989 on the Law Applicable to 

Succession to the Estates of Deceased Persons. 

28 

Cf. Article 5 of the Council Regulation (EU) No 1259 / 2010 of 20 December 

2010 implementing enhanced cooperation in the area of the law applicable 

to divorce and legal separation (Rome III), OJ 2010, L 343 / 10. 

29 

Critical on this point: Flessner 2010, p. 137 et seq. (note 23). 


49 


www.sellier.de


mutual relationship. 30 On the one hand, this meets the practical requirements 

of modern legal transactions; on the other hand, it is an attempt to 

overcome the mainly academic problem of conflit mobile. 31 

The Dutch Property Law (Conflict of Laws) Act has followed this development 

and includes six provisions, scattered over the Act, that provide for 

the possibility of choosing which law shall be applicable to the proprietary 

rights regime in the legal relationship between the parties. Such a choice 

of law will usually be based on practical considerations. For example, 

Article 13 paragraph 2 of the Act provides that Dutch companies whose 

stock is traded on the New York Stock Exchange may stipulate that the 

transfer of these ‘American shares’ shall be governed by the law of the 

federal state of New York. It reads as follows: 

‘In derogation from paragraph 1, where, in order to promote the 

marketability of registered shares in a Dutch public limited liability 

company on a regulated foreign stock exchange, such shares are issued 

in a form customary in the state of establishment of that stock 

exchange, the issuing company may provide that the proprietary rights 

regime of the shares will be determined by the law of the state of 

establishment of the relevant stock exchange or the law of the state 

in which, with the consent of the relevant stock exchange, deliveries 

and other proprietary juridical acts regarding the shares may or must 

be performed.’ 32 

30 

See Flessner 2010 (note 23). 

31 

According to Dutch law, most of the foreign proprietary rights will be recognised. 

See Van der Weide 2006, Chapter 4. 

32 

In its original Dutch version, Article 13 paragraph 2 of the Dutch Property 

Law (Conflict of Laws) Act (Article 10:138 paragraph 2 Civil Code) reads as 

follows: ‘In afwijking van het eerste lid kan met betrekking tot aandelen op 

naam in een Nederlandse naamloze vennootschap waaraan, ter bevordering 

van de verhandelbaarheid aan een gereglementeerde buitenlandse effectenbeurs 

een in de staat van vestiging van de beurs gebruikelijke vorm is gegeven, 

door de uitgevende vennootschap worden bepaald dat het goederenrechtelijke 

regime wordt beheerst door het recht van de staat van vestiging van de betrokken 

beurs dan wel het recht van de staat waarin met instemming van de 

betrokken beurs leveringen en andere goederenrechtelijke rechtshandelingen 

betreffende de aandelen kunnen of moeten worden verricht.’ 

50 



www.sellier.de


We may call this type of choice of law a ‘direct choice of law’: the parties 

make a specific juridical act in the field of proprietary rights law (the 

transfer of shares) the subject matter of their choice of law. 

Article 3 paragraph 2 of the Dutch Property Law (Conflict of Laws) Act, 

which lays down conflict rules for retention of title, likewise concerns 

a direct choice of law. It provides that buyer and seller may agree that 

the proprietary consequences of retention of title – regarding of whose 

property the goods form a part and who has the power of disposal of the 

goods – shall be governed by the law of the country of destination: 

‘In derogation from the first sentence of paragraph 1, the contracting 

parties may agree that the proprietary effects of retention of title to 

goods intended for export shall be governed by the law of the country 

of destination provided that the retention of title does not cease to be 

effective under that law until the purchase price has been fully paid. 

Such choice of law will only be effective if the goods are actually imported 

into the designated country of destination.’ 33 

In drafting this latter provision, the legislature specifically had in mind 

Dutch exporters who, by choosing German law as applicable law when exporting 

goods from the Netherlands to Germany, could avail themselves 

of the wider forms of retention of title for which German law provides: 

for example, Eigentumsvorbehalt mit Verarbeitungsklausel (retention of title 

with processing clause). 34 This German form of retention of title extends 

the retention to cover goods (e.g. steel chains) manufactured out of the 

33 



follows: ‘In afwijking van de eerste zin van het eerste lid kunnen partijen overeenkomen 

dat de goederenrechtelijke gevolgen van een eigendomsvoorbehoud 

van een voor uitvoer bestemde zaak worden beheerst door het recht van 

de staat van bestemming indien op grond van dat recht het eigendomsvoorbehoud 

niet zijn werking verliest totdat de prijs volledig is betaald. De aldus 

overeengekomen aanwijzing heeft slechts gevolg indien de zaak daadwerkelijk 

in de aangewezen staat van bestemming wordt ingevoerd.’ 

34 

See Tweede Kamer der Staten-Generaal, vergaderjaar 2001-2002, 28 239, 

nr. 3, p. 8; Tweede Kamer der Staten-Generaal, vergaderjaar 2006-2007, 30 

876, nr. 6, p. 10. 


51 


www.sellier.de


original goods (e.g. coils of steel). 35 Dutch law does not have this form of 

extended retention of title. 36 

In most cases, however, a ‘proprietary’ choice of law under the Dutch Act 

comes into existence indirectly. Pursuant to Article 4, for example, the 

question of whether and when a right of lien comes into existence, and 

the extent of its scope, is governed by the same law that governs the legal 

relationship on which the right is based: for instance, a sales contract 

or a contract of carriage. 37 By choosing the law that is applicable to the 

underlying contractual relationship, the parties can indirectly influence 

which law will govern whether and when a right of lien comes into existence 

and the extent of its scope. 

The legislature has opted for a similar solution in Articles 7 and 8 of the 

Act. Article 7 paragraph 2 provides that in the case of stolen goods whose 

whereabouts are unknown, the transfer of ownership between insured 

and insurer is governed by the law that is applicable to the insurance 

agreement. 38 

35 

Cf. Staudinger / Beckmann, § 449 BGB, Berlin: Sellier de Gruyter 2004, no. 99- 

126; Jürgen F. Baur & Rolf Stürner, Sachenrecht, München: Verlag C.H. Beck 

2009, § 59, no. 49 et seq. 

36 

See for example Jacobien Rutgers, International Reservation of Title Clauses. 

A Study of Dutch, French and German Private International Law in the Light 

of European Law, The Hague: T.M.C. Asser Press 1999. 

37 


Civil Code) reads as follows: ‘(…) the question of whether and when a right of 

lien comes into existence and the extent of its scope is governed by the same 

law that governs the legal relationship in which the right is based. A right of 

lien will only be effective if it is compatible with the law of the state in whose 

territory the goods are situated at the time of the enforcement of the right 

of lien.’ (‘(…) worden het ontstaan en de inhoud van een recht van retentie 

bepaald door het recht dat de daaraan ten grondslag liggende rechtsverhouding 

beheerst. Een recht van retentie kan slechts geldend worden gemaakt 

voor zover het recht van de staat op welks grondgebied de zaak zich bevindt, 

zulks toelaat.’). 

38 

Article 7 paragraph 2 of the Dutch Property Law (Conflict of Laws) Act (Article 

10:132 paragraph 2 Civil Code) reads as follows: ‘If the loss is covered 

by insurance in the circumstances referred to in the preceding paragraph, the 

law that is applicable to the insurance agreement determines whether and, 

if so, how ownership is transferred to the insurer.’ (‘Is in het in het vorige lid 

52 



www.sellier.de


The same applies to goods in transit. Pursuant to Article 8, a choice of 

law laid down in the sales contract is deemed to apply to the proprietary 

rights regime of the goods in transit as well: 

‘If the carriage of goods referred to in paragraph 1 of this provision 

takes place pursuant to a contract of sale or another contract imposing 

an obligation to transfer the ownership of the goods in transit or 

to create rights in rem to these goods, a choice of law laid down in the 

sales contract or the contract to create rights in rem is deemed to apply 

to the proprietary rights regime of the goods in transit as well.’ 39 

For example, if it has been agreed that a Dutch coffee-roasting house will 

transfer title to a shipment of coffee beans to a Belgian coffee-roasting 

house on the high seas, the transfer of ownership is governed by Dutch 

law if the parties have declared Dutch law applicable to the sales contract 

concluded between them. 

Finally, I will not discuss the one remaining possibility of indirect choice 

of law in the Dutch Property Law (Conflict of Laws) Act, because I would 

encroach upon the contributions of Paulien van der Grinten, Rick Verhagen, 

and Axel Flessner to this book. 40 The aforementioned possibility concerns 

Article 10 of the Dutch Property (Conflict of Laws) Act, which regulates 

cross-border assignments. Article 10 paragraph 2 reads as follows: 

bedoelde geval het verlies door een verzekering gedekt, dan bepaalt het recht 

dat de verzekeringsovereenkomst beheerst, of en op welke wijze de eigendom 

op de verzekeraar overgaat.’). 

39 



follows: ‘Indien het in het eerste lid bedoelde vervoer plaatsvindt ter uitvoering 

van een koopovereenkomst of een andere overeenkomst die verplicht tot 

overdracht van de vervoerde zaak, of ter uitvoering van een tot vestiging van 

rechten op die zaak verplichtende overeenkomst, wordt, in afwijking van het 

eerste lid, een aanwijzing van het op de bedoelde overeenkomst toepasselijke 

recht, opgenomen in die overeenkomst, geacht mede betrekking te hebben 

op het goederenrechtelijke regime met betrekking tot de vervoerde zaak.’ 

40 

Paulien M.M. van der Grinten, Article 14 Rome I: A Political Perspective 

(Chapter 7); Hendrik Verhagen, Party Autonomy and Assignment (Chapter 

9); Axel Flessner, Between Articles 14 and 27 of Rome I: How to interpret a 

European Regulation on Conflict of Laws? (Chapter 10). 


53 


www.sellier.de


‘The proprietary regime of a claim of a known creditor is governed 

by the law that is applicable to the contract imposing the obligation 

to transfer the claim or to create rights in rem. That law determines 

in particular: (a) the requirements for the transfer of the claim or the 

creation of rights in rem; (b) the question of who is entitled to exercise 

the rights attached to the claim; (c) the question of what rights can 

exist in the claim and the nature and content of these rights; (d) the 

manner in which these rights are modified, transferred or will extinguish, 

as well as the legal relationship between these rights.’ 41 

Article 10 paragraph 2 of the Dutch Property (Conflict of Laws) Act is 

based on the Bechem / Hansa-judgment of 16 May 1997 of the Supreme 

Court of the Netherlands (Hoge Raad der Nederlanden), in which the 

law applicable to the transfer of a claim of a known creditor is linked to 

the law of the contract to assign. 42 Article 10 has now been superseded 

by recent developments. On 17 December 2009, the Regulation of the 

European Parliament and of the Council of 17 June 2008 on the law applicable 

to contractual obligations (Rome I) entered into effect (OJ 2008, 

L 177 / 6). Article 14 of the Rome I Regulation contains a separate rule for 

cross-border assignments. 

41 


Law (Conflict of Laws) Act (Article 10:135 paragraph 2 Civil Code) reads 

as follows: ‘Voor het overige wordt het goederenrechtelijke regime met betrekking 

tot een vordering op naam beheerst door het recht dat op de tot 

overdracht of vestiging van rechten verplichtende overeenkomst toepasselijk 

is. Dat recht bepaalt in het bijzonder: (a) welke vereisten aan een overdracht 

of vestiging worden gesteld; (b) wie gerechtigd is tot uitoefening van de in de 

vordering besloten rechten; (c) welke rechten op de vordering kunnen rusten 

en welke de aard en de inhoud van deze rechten zijn; (d) op welke wijze die 

rechten zich wijzigen, overgaan en tenietgaan en welke hun onderlinge verhouding 

is.’ 

42 

Hoge Raad der Nederlanden 16 May 1997, Nederlandse Jurisprudentie (NJ) 

1998, 585, comm. Th.M. de Boer. Case notes in the English language by 

M.E. Koppenol-Laforce, Netherlands International Law Review (NILR) 1998, 

p. 129-137; Carla Joustra, Praxis des Internationalen Privat- und Verfahrensrechts 

(IPRax) 1999, p. 280-284. 

54 



www.sellier.de


2.7. Putting party autonomy to the test 

Party autonomy to choose the law governing proprietary rights aspects can 

only be said to exist if the choice of law made with respect to these aspects 

is valid not only between the parties (inter partes) but also vis-a-vis third 

parties, meaning that it can be alleged against third parties. 43 

Foreign legislation in which party autonomy is accepted without limitation 

is – as far as is known – not present. The most far-reaching form of 

such party autonomy is found in the Swiss Conflict of Laws Code (Bundesgesetz 

über das Internationale Privatrecht). Pursuant to Article 104, parties 

may make a choice of law in regard to the acquisition of proprietary rights 

in movables. The choice is limited, however, to the law of the country of 

origin, the law of the country of destination, or the law that governs the 

underlying obligatory legal relationship. 44 However, according to paragraph 

2 of Article 104 of the Swiss Conflict of Laws Code, a choice of 

law made under Article 104 cannot be alleged against third parties and 

therefore has relative effect only: namely, in the relationship between the 

parties themselves: 

‘Die Rechtswahl kann Dritten nicht entgegengehalten werden.‘ 45 

43 

Cf. Kurt Siehr, Internationales Privatrecht. Deutsches und europäisches Kollisionsrecht 

für Studium und Praxis, Heidelberg: C.F. Müller Verlag 2001, 

p. 272-273: ‘Außerdem spielt das Sachenrecht zwischen Vertragsparteien 

keine große Rolle. Sachenrechte entfalten nämlich ihre volle Schönheit gegenüber 

Dritten. Deshalb gibt es auch keine Rechtswahl im Internationalen 

Sachenrecht. Ein Eigentum inter partes ist unsinnig.’ 

44 

In its original German version, Article 104 paragraph 1 of the Swiss Conflict 

of Laws Code reads as follows: ‘Die Parteien können den Erwerb und den Verlust 

dinglicher Rechte an beweglichen Sachen dem Recht des Abgangs- oder 

des Bestimmungsstaates oder dem Recht unterstellen, dem das zugrundeliegende 

Rechtsgeschäft untersteht.‘ 

45 

Critical on this point: Kurt Siehr, Das Internationale Privatrecht der Schweiz, 

Zürich: Schulthess 2002, p. 197: ‘Parteien untereinander brauchen das Sachenrecht 

nicht. Dieses soll sich gerade auch gegenüber Dritten durchsetzen. 

Art. 104 Abs. 2 IPRG verneint dies aber. Die Vorschrift des Art. 104 IPR ist 

verunglückt, solange sie nicht auch gegenüber Dritten, welche die Rechtswahl 

kennen, gilt. Das ist aber nach Art. 104 Abs. 2 IPRG nicht der Fall, der 

zwischen gut- und bösglaubigen Dritten nicht unterscheidet. Solange man 


55 


www.sellier.de


A number of the possibilities of making a choice of law mentioned in the 

Dutch Property Law (Conflict of Laws) Act likewise have only a relative 

effect. Articles 3 (retention of title clauses) and 4 (rights of lien), for 

example, provide that the effect on proprietary rights aspects of crossborder 

retention of title clauses or rights of lien is governed by the law of 

the place where the goods are situated (lex situs). This means that, in the 

strict sense of the word, under Articles 3 and 4 of the Dutch Act there is 

no party autonomy to choose the law governing proprietary rights issues. 

In the case of the retention of title clauses regulated in Article 3, this 

consequence is less visible, because the choice of the law of the country of 

destination will as a rule coincide with the choice of the law of the place 

where the retention of title clause is alleged against third parties – likewise 

being the country of destination. 

It remains to be seen, however, how foreign courts will respond to retention 

of title if its consequences for proprietary rights aspects have been 

determined in advance by means of a choice of law made by the buyer and 

the seller. Would a German court, for example, accept such a choice of law 

in a case involving the export of goods from the Netherlands to Germany? 

Given the circumspect position taken by the German Bundesgerichtshof in 

regard to parties choosing the law that is applicable to proprietary rights 

issues, there is, in my opinion, no clear answer to this question. In its decision 

of 25 September 1996, the German Bundesgerichtshof ruled: 

‘Nach der ständigen Rechtsprechung des BGH (…) gilt im deutschen internationalen 

Sachenrecht für alle Übereignungstatbestände der Grundsatz 

des Rechts des Lageortes (…). Die Zulässigkeit und das Vorliegen einer 

Rechtswahl hat der BGH auch in solchen Fällen nicht in Erwägung gezogen, 

in denen die Anwendbarkeit deutschen Rechts für die Übereignung 

einer im Ausland belegenen Sache möglicherweise dem Willen der Parteien 

entsprochen hätte (…). Mit der uneingeschränkten Anwendung der Situs- 

Regel befindet er sich in Übereinstimmung mit einer breiten Auffassung im 

Schrifttum (…) die eine Rechtswahl im internationalen Sachenrecht mit 

Rücksicht auf das Verkehrsinteresse für unzulässig hält (…).‘ 46 

den Art. 104 Abs. 2 IPRG nicht einschränkend auslegt, ist die Rechtswahl 

überflüssig. In dieser Form ist sie auch nie ernsthaft empfohlen worden.’ 

46 

Bundesgerichtshof (BGH) 25 September 1996 – VIII ZR 76 / 95, Die Deutsche 

Rechtsprechung auf dem Gebiete des Internationalen Privatrechts (IPRspr.) 

1996, no. 56; Neue Juristische Wochenschrift (NJW) 1997, p. 461-464; Praxis 

56 



www.sellier.de


Transfer of ownership is governed by the lex situs, the law of the place 

where the goods are situated. In international property law, party autonomy 

is not allowed. 47 

It is remarkable that the Dutch Property Law (Conflict of Laws) Act 

contains a number of choice of law options that on the contrary do effect 

proprietary rights issues. These are the choice of law provisions of Article 

7 (transfer of stolen goods to the insurer), Article 8 (goods in transit), 

Article 10 (cross-border assignment), and Article 13 (American shares). 48 

Although the legislature is not always clear about the nature of the choice 

of law concerned, it may be assumed that these four choice of law provisions 

have an erga omnes effect and that consequently these are examples 

of party autonomy to choose the law governing proprietary rights aspects 

in its purest form: the choice of law not only has an effect between the 

parties themselves but may also be alleged against third parties. 

Careful scrutiny of these four choice of law provisions shows that the 

goods to which these provisions refer cannot be linked in any way to a 

place where the goods are situated. In the case of stolen goods, this place 

will usually be unknown, while goods in transit are on the move without 

staying in any one place very long. Claims and shares are intangible. A 

place where these goods are situated simply does not exist. It is indeed 

my opinion that party autonomy to choose the law governing proprietary 

des Internationalen Privat- und Verfahrensrechts (IPRax) 1997, p. 411-413 

and p. 422-427, comm. Hans Stoll. 

47 

It should be noted that German law is still supportive of the concept of renvoi 

in the Conflict of Laws. Article 4 paragraph 1 of the German Conflict of Laws 

Code (Einführungsgesetz zum Bürgerlichen Gesetzbuche) reads as follows: 

‘Wird auf das Recht eines anderen Staates verwiesen, so ist auch dessen Internationales 

Privatrecht anzuwenden, sofern dies nicht dem Sinn der Verweisung 

widerspricht. Verweist das Recht des anderen Staates auf deutsches Recht 

zurück, so sind die deutschen Sachvorschriften anzuwenden.’ In the light of 

the current case law of the Bundesgerichtshof in regard to parties choosing the 

law that is applicable to proprietary rights issues, this leads to doctrinal inconsistencies. 

See in this book the contribution of Jan von Hein, Party Autonomy 

in International Property Law: A German Perspective (Chapter 5), paragraph 

5.2.3. Also critical on this point: Ulrich Drobnig, European conflict rules for 

the mutual recognition of security rights in goods (Chapter 8), p. 174 in this 

book. 

48 

See for these provisions paragraph 6 of this Chapter. 


57 


www.sellier.de


rights issues can only function optimally when a link with the situs is 

nonexistent, since the interests of third parties are then not directly at 

stake, and no harm is done to proprietary rights systems. 49 

2.8. In conclusion 

Party autonomy to choose the law governing proprietary rights aspects is 

a last resort, intended for those cases involving proprietary rights aspects 

in which seeking a connection with the place where the goods are situated 

does not bring a solution, and in which only a choice of law will 

lead to results that are acceptable in actual practice. The Dutch Property 

Law (Conflict of Laws) Act regulates four of these cases, of which the 

provisions regarding goods in transit (Article 8) and cross-border assignment 

(Article 10) are the most relevant from the practical perspective. 

As regards the latter legal concept – cross-border assignment – it is hoped 

that the European Commission, in its report to be issued on the scope of 

Article 14 of the Rome I Regulation 50 will clear a path for party autonomy 

to choose the law governing proprietary rights aspects. However, it is to 

be expected that a European-wide consensus at this point will be difficult 

to achieve. 

49 

Cf. Van der Weide 2006, paragraph 6.6 (see note 23). 

50 

See the review clause of Article 27 paragraph 2 of the Rome I Regulation and 

the ‘Open invitation to tender JLS / 2010 / JCIV / PR / 0007 / E4 – Study on the 

question of effectiveness of an assignment or subrogation of a claim against 

third parties and the priority of the assigned or subrogated claim over a right 

of another person’, European Commission, 22 May 2010 (JLS / E4-LK / jbg- 

D(2010) 7418). See also the contribution of Paulien M.M. van der Grinten to 

this book (Chapter 7). 

58 



www.sellier.de

3. 

The Numerus Clausus and 

Party Autonomy in the Law of Property 

T.H.D. Struycken * 


The numerus clausus in the law of property concerns the general principle 

that the number of property rights is limited and these rights cannot be 

expanded or modified by individual parties according to their specific 

wishes and needs. The numerus clausus implies that the law of property 

is mandatory; there is no room for party autonomy. 

The numerus clausus is sometimes called one of the great dogmas of the 

civil law legal systems. Its characterisation as dogma is illustrative of its 

status: emblematic, enigmatic, elusive, fundamental and controversial. In 

an article published in 1953, Vera Bolgár famously stated: 

“The numerus clausus rule, which has no apparent practical value, 

might well be left to the museum of Begriffsjurisprudenz.” 1 

This strong statement as well as the traditional view that the numerus 

clausus is characteristic of civil law legal systems and unknown to the 

common law call for an investigation of the scope, origin and justification 

of the numerus clausus principle. 

The numerus clausus is better known in the Netherlands as the ‘closed 

system’ of property law (het gesloten systeem van het goederenrecht). It is unclear 

where either term comes from. The 1888 explanatory Motive to the 

draft text of the German BGB spoke of ‘die geschlossene Zahl’ of property 

* 

Professor of European Property Law (Utrecht University), finance and insolvency 

lawyer (NautaDutilh N.V., Amsterdam). 

1 

Vera Bolgár, Why No Trusts in the Civil Law?, (1953) 2 Am.J.Comp. Law 

214. 

59 


www.sellier.de


rights. 2 The words numerus clausus seem to have first been used in Germany 

by Philipp Heck in 1930 3 and a few years later by Franz Wieacker 

in 1938. 4 The words reappeared many years later in several places at about 

the same time, in 1952 in Italy 5 , in 1953 in Bolgár’s article and in the title 

of Merryman’s response to Bolgár in 1963, 6 which were both published 

in the United States. A reference to a closed system of property law first 

appeared in the Netherlands in 1961 in the reports of the deliberations 

of the Dutch legislature on the draft of a new civil code, the Burgerlijk 

Wetboek of 1992 (the “1992 Burgerlijk Wetboek” or “BW”). 7 It is only in 

the last two decades that the closed system of property law, or the numerus 

clausus, became the title and subject of voluminous monographs. 8 

2 

Motive zu dem Entwurfe eines Bürgerlichen Gesetzbuches für das Deutsche 

Reich, Band III, Sachenrecht, Amtliche Ausgabe (1888), p. 3. 

3 

Philipp Heck, Grundriß des Sachenrechts, 1930 (reprinted in 1970 and 1994), 

§§ 22 and 23. 

4 

Franz Wieacker, Bodenrecht, 1938, p. 111. 

5 

A. Torrente, Manuale di Diritto Privato, Milano, 1st edition 1952, e no. 167 

on p. 232: “Nell’interesse generale della produzione, l’ordinamento guiridico 

non ritiene opportune gravare la proprietà di limitazioni diverse da quello 

della legge previste e ritenute utili alla generalità: perciò i diritti reali costituiscono 

un numerus clausus.” 

6 

J.H. Merryman, Policy, Autonomy, and the Numerus Clausus in Italian and 

American Property Law, (1963) 12 Am.J.Comp. Law 224-231. 

7 

See VV II (d.d. 25 October 1961), Parl. Gesch. Book 5, p. 248 and 291, 

and furthermore Handelingen II (d.d. 5 October 1965), Parl. Gesch. Book 4, 

p. 227. 

8 

See most notably Bénédict Foëx, Le “numerus clausus” des droits réels en 

matière mobilière, Collection Juridique Romande, Payot, Lausanne, 1987; 

B. Rudden, Economic Theory v. Property Law: The Numerus Clausus Problem, 

in: Oxford Essays in Jurisprudence, Third Series 1987; Vincenzo Giuffrè, 

L’emersione dei “iura in re aliena” ed il dogma del ‘numero chiuso’, Jovene 

Editore, Napoli 1992; T.H.D. Struycken, De numerus clausus in het goederenrecht, 

Serie Onderneming en Recht, vol. 37, Kluwer 2007 (thesis Nijmegen); 

Bram Akkermans, The Principle of Numerus Clausus in European Law 

of Property, Ius Commune Europaeum Series, no. 75, Intersentia 2008 (thesis 

Maastricht), and the contributions to legal periodicals and books mentioned 

infra, and V. Sagaert, Het goederenrecht als open systeem van verbintenissen?, 

Tijdschrift voor Privaatrecht 2005, p. 983-1086. See also Jens Thomas Füller, 

Eigenständiges Sachenrecht?, Jus Privatum 104, Mohr Siebeck Tübingen, 

2006. 

60 



www.sellier.de

3. The Numerus Clausus and Party Autonomy in the Law of Property 

Both terms raise the question of the scope of this principle of Dutch 

property law. Interpreted narrowly, the numerus clausus concerns the 

categories and types of iura in re aliena, limited real rights in the property 

of another (beperkte rechten, hereinafter “limited property rights”). 

Interpreted more widely, the numerus clausus also concerns the content 

of full ownership rights in (volledige rechten) both tangible assets (zaken, 

or “things”), whether movable or immovable, and intangible assets such 

as shares, debts (vorderingen) and intellectual property rights. 9 The term 

‘closed system’ may also be taken to refer to other rules and concepts of 

property law, such as those pertaining to how a property right (goederenrechtelijk 

recht) can be acquired or lost and a power over one’s property can 

be conferred on someone else. 

3.2. The foundations of the numerus clausus 

Several articles in the 1992 Burgerlijk Wetboek enshrine the numerus clausus 

principle more or less explicitly, where they refer generally to types of 

property rights and types of situations that have a statutory basis. Article 

3:81(1) BW, a provision in the part of the code containing general rules 

on the acquisition of property rights, implicitly requires a statutory basis 

for each type. It states: 

‘He who is entitled to an independent and transferable right can, 

within the boundaries of that right, create the limited property rights 

that are mentioned by statute.’ 10 

This article on the types of property rights – sometimes referred to by the 

German word Typenzwang – is thought to reflect most clearly the existence 

of the numerus clausus in the Dutch law of property. 11 The article 

9 

Since the 1992 Burgerlijk Wetboek entered into force on 1 January 1992, 

the term ownership (eigendom) has, strictly speaking, applied only to tangible 

property and not to intangible property, such as debts and intellectual property 

rights. 

10 

Article 3:81(1) BW reads in Dutch: ‘Hij aan wie een zelfstandig en overdraagbaar 

recht toekomt, kan binnen de grenzen van dat recht de in de wet 

genoemde beperkte rechten vestigen.’ 

11 

Two other articles contain rules bearing some resemblance to the principle of 

numerus clausus: Article 3:277(1) BW refers to the equality between creditors 

T.H.D. Struycken 61 


www.sellier.de


further implies that the content of the different types of property rights is 

likewise prescribed by statute (Typenfixierung). 

Under the wider interpretation of the closed system, the rules on transferability 

also appear relevant. Article 3:83(1) BW states that (i) the ownership 

of things, (ii) limited property rights and (iii) debts are transferable 

unless a statutory provision or the nature of the right dictates otherwise. 

Article 3:83(3) BW states that rights other than ownership, limited property 

rights and debts are transferable only if a statute so provides. Similarly, 

Article 3:80(3) and (4) BW states that assets are acquired or lost in 

the manner prescribed for each type of asset by statute. 12 

Finally, the controversial Article 3:84(3) BW enshrines the prohibition 

of fiduciary transfers of ownership. 13 Any such fiduciary transfer, whether 

cum creditore or cum amico, is void. This prohibition is thought to create 

an obstacle to all forms of dual ownership and other ways of splitting 

ownership not laid down by statute. As such, Article 3:84(3) BW may 

be understood as one of the foundations of the numerus clausus in Dutch 

property law. 14 

The Dutch legislature considered the numerus clausus in some detail in the 

1960s and 1970s when it discussed the various drafts of the 1992 Burgerlijk 

Wetboek. The explanatory memorandum to the first draft, both written by 

the great 20 th century Dutch scholar Eduard Meijers, stated explicitly that 

the draft enshrined the numerus clausus principle. Later in the process, 

‘except for grounds of preference that are recognised by statute’ (behoudens 

door de wet erkende redenen van voorrang), and Article 4:42(1) BW defines a 

disposition by last will and testament (uiterste wilsbeschikkingen) as ‘a unilateral 

legal act [..], that is provided for in Book 4 BW or is recognised as such by 

statute’ (Een uiterste wilsbeschikking is een eenzijdige rechtshandeling [..], die in dit 

Boek is geregeld of in de wet als zodanig is aangemerkt.). 

12 

As used in all these articles, the term ‘statute’ refers to a law enacted by parliament. 

Regulations and rules issued or passed by other governmental and local 

authorities fall outside the notion of statute. 

13 

Article 3:84(3) BW reads: ‘A juridical act intended to transfer property for 

purposes of security or which does not have the purpose of bringing the property 

into the patrimonium of the acquirer, after transfer, does not constitute 

valid title for transfer of that property.’ 

14 

See A.H. Scheltema, De goederenrechtelijke werking van de ontbindende voorwaarde, 

2003 (thesis Leiden), p. 371, and Struycken, o.c. 2007, § 8.2.4.2. 

62 



www.sellier.de


the legislature agreed to slightly relax the restrictions on party autonomy 

resulting from this principle by offering parties several statutory possibilities 

to adapt, limit or expand the powers and obligations arising from a 

limited property right they have created. 

In addition, the legislature made it possible to give proprietary effect to 

certain types of obligations in respect of land by registering them in the 

land register. As a general rule, the obligation must be one to allow certain 

action to be taken or to refrain from taking certain action in respect 

of the obligor’s land (dulden of niet doen), and not an affirmative one (in 

faciendo). If registered, such an obligation is binding upon acquirers of that 

land pursuant to Article 6:252 BW. Even though belonging to the law of 

obligations, this type of obligation, which binds the obligor in his capacity 

as owner of the land (kwalitatieve verplichting), can be seen as a statutory 

expansion of the numerus clausus (the Typenzwang). 15 

It is not in the nature of Dutch courts to explicitly embrace a dogma such 

as the numerus clauses or any other dogma. Moreover, in the 19 th century, 

the courts appeared willing to afford parties the freedom to create new 

types of property rights. This changed after the 1905 landmark decision 

of the Hoge Raad in Blaauboer / Berlips. 16 In that case, the court held that 

the obligation of an owner of land to construct a road on adjacent plots 

had not passed to his successor in title to the land because (i) the obligation 

did not take the form of a property interest, (ii) as an obligation in 

faciendo, it could not constitute a property interest and (iii) it had not 

been registered in the public land register. 

Since that decision, and in particular since the 1992 Burgerlijk Wetboek, 

Dutch courts have referred occasionally to the closed system of Dutch 

property law. More importantly, they have seemed inclined to interpret 

the statutory rules of property law rather strictly. 

Nevertheless, Dutch scholars continue to disagree on the scope of the 

numerus clausus. Lurking behind discussions regarding terminology is 

15 

In the common law, such obligations can be compared to covenants running 

with the land. 

16 

HR 3 March 1905, W. 8191. See further S.E. Bartels and J.M. Milo, Contents 

of Real Rights: Personal or Proprietary – a Principled History, in: Contents 

of Real Rights, S.E. Bartels en J.M. Milo (eds.), Wolf Legal Publishers, 2004, 

p. 5-23. 



www.sellier.de


the fundamental debate on whether the numerus clausus is limited to 

the traditional property rights, being ownership and the limited property 

rights, i.e. (i) servitude (erfdienstbaarheid), (ii) usufruct (vruchtgebruik), 

(iii) emphyteusis (erfpacht), (iv) superficies (opstalrecht), (v) mortgage (hypotheekrecht) 

and (vi) pledge (pandrecht), or whether the numerus clausus 

also encompasses ‘modern’ types of property rights such as rights to goodwill 

and internet domain names. 

3.3. Origin of the numerus clausus 

It is generally assumed that the numerus clausus is a German invention 

that was developed by the Pandectists in the 19 th century and incorporated 

in the 1900 draft of the BGB as one of its fundamental principles. 

Without any doubt, the restrictions on party autonomy in the law of 

property formed an essential part of the model for private law that was 

developed by Friedrich Carl von Savigny (1779-1861) and his followers. 17 

Other elements of this model were a new abstract, absolute and total concept 

of ownership, the notion of a thing that excludes intangible assets, 

a strict specificity requirement for property rights, a rejection of transfer 

restrictions with proprietary effect, the autarky of the law of property visà-vis 

the law of obligations and general protection for bona fide acquirers 

against a defect in the transferor’s title. As it is now understood by most 

scholars, the numerus clausus is deeply rooted in 19 th century German 

legal thinking. 

However, an investigation of the Dutch sources shows that the numerus 

clausus of property rights was incorporated in very clear words in early 

drafts for a new Dutch civil code around 1800, most articulately in an 

1816 draft by Joan Melchior Kemper (1776-1824). Kemper’s draft stated 

explicitly that the only rights one could have in the property of another 

17 

See Wolfgang Wiegand, Numerus clausus der dingliche Rechte – Zur Entstehung 

und Bedeutung eines zentralen zivilrechtlichen Dogmas, in: Wege europäischer 

Rechtsgeschichte (Festschrift Karl Kroeschell), Rechtshistorische 

Reihe Band 60, 1987, p. 623-643; see also Wiegand, Die Entwicklung des 

Sachenrechts im Verhältnis zum Schuldrecht, AcP 190 (1990), p. 112-138, 

and Wiegand, Funktion und systematische Stellung des Sachenrechts im BGB, 

in: 100 Jahre BGB – 100 Jahre Staudinger, Beiträge zum Symposion vom 18.- 

20. Juni 1998 in München, p. 107-126. See also Füller, o.c. 2006. 

64 



www.sellier.de


were the rights treated in the code, and that all other rights were personal 

only. The early drafts were superseded by codes modelled on the French 

Code civil. A provision in the old Dutch Burgerlijk Wetboek of 1838 contained 

a non-exhaustive catalogue of real rights, similar to Article 543 of 

the French Code civil and equally inadequate as a statutory foundation for 

a numerus clausus of property rights. Consequently, 19 th century Dutch 

courts were inclined to grant parties considerable liberty to devise new 

types of property rights outside the scope of the old Burgerlijk Wetboek. 

The numerus clausus received a warmer reception in 19 th century Dutch 

doctrine than in the courts. The works of both Diephuis 18 (1817-1892) 

and Opzoomer 19 (1821-1892), two monumental 19 th century legal scholars 

in the Netherlands, show a clear tendency to adhere to a strict interpretation 

of the old Burgerlijk Wetboek; the statutory provisions were their main 

focal point when describing and analysing the limits of the law of property. 

Telling are a number of references in the work of Diephuis to the French 

author Demolombe (1804-1887). 20 Where French courts and doctrine of 

that period predominantly favoured a large role for party autonomy in the 

field of property law, Demolombe argued extensively for adherence to the 

types of property rights chosen and regulated by the French legislature. 

The line taken by Diephuis and Opzoomer became so deeply entrenched 

in Dutch legal thinking that opposition was too little and too late for the 

numerus clausus to be rejected. As mentioned above, in 1905 the Hoge 

Raad cemented the foundations of the numerus clausus in Dutch law. 

It can be concluded that the numerus clausus, as the expression of the will 

of the legislature, has been part of Dutch legal thinking since the era of 

codification in the early decades of the 19 th century. However, the works 

of Savigny and the Pandectists fleshed out this principle, resulting in what 

is now understood as the numerus clausus and the closed system in the law 

18 

G. Diephuis, Het Nederlandsch Burgerlijk Regt naar de volgorde van het 

Burgerlijk Wetboek, volume III, 1846 (1st edition) & 1857 (2nd edition); G. 

Diephuis, Het Nederlandsch Burgerlijk Regt, volume VI, 1880 (1st edition) 

& 1886 (2nd edition). 

19 

C.W. Opzoomer, Het Burgerlijke Wetboek verklaard, Volume III, Art. 555- 

783, 1871; C.W. Opzoomer, Het Burgerlijke Wetboek – Aanteekening op de 

artikelen die thans nog verklaring behoeven, Volume II, 1850. 

20 

C. Demolombe, Cours de Code Napoléon, band IX, De la distinction des biens; 

de la propriété; de l’usufruit, de l’usage et de l’habitation, 1854, nrs. 462-532, 

p. 351-470. 



www.sellier.de


of property. The hallmarks of the Pandectist thinking on party autonomy, 

the concept of ownership, the concept of a thing, free transferability and 

the protection of bona fide purchasers became ever more apparent in 

Dutch legal doctrine. They now characterise Dutch property law, as enacted 

in the 1992 Burgerlijk Wetboek, in very many ways. 21 

3.4. The numerus clausus in other legal systems 

It used to be widely believed that the numerus clausus in the law of property 

was part of the civil law but not of the common law. 22 Indeed, many 

civil legal systems recognise the numerus clausus as a fundamental principle 

of their law of property, although such recognition is mainly in doctrine 

and case law; few codes contain explicit provisions on the numerus 

clausus. Closer comparative analysis shows, however, that the status of the 

numerus clauses in civil law and common law countries is not always as 

supposed and the distinction between them on this point is blurred. 

Strikingly, French law itself is not thought to endorse the numerus clausus. 

On the basis of the classic Caquelard case 23 , decided in 1834, many – but 

not all – authors submit that French law is characterised by party autonomy 

in the field of property law. 24 In practice, however, the French 

courts tend to adhere to the types of property rights laid down in the Code 

civil and other statutes. There appear to be very few examples of cases 

that could be said to recognise some new kind of property right. In the 

field of proprietary security interests, French law is strongly dominated by 

special legislation. It took a special statute in 1980, the Loi Dubanchet, 

21 

See more extensively Struycken, o.c. 2007. 

22 

See René David, Les grands systèmes de droit contemporains, 1 st edition 1964, 

no. 311 (‘Conception anglaise des tenures’), with a reference to F.H. Lawson, 

The Law of Property, 1958, p. 59; F.H. Lawson, in: International Encyclopedia 

of Comparative Law, Volume VI, Property and Trust, F.H. Lawson (ed.), 

Chapter 2, Structural Variations in Property Law, 1975, p. 138, § 2-274 and 

§ 2-35. 

23 

Req. 13 février 1834, D.P. 34.1.218, S. 34.1.205 (Caquelard c. Lemoine); also 

published in: F. Terré / Y. Lequette, Les grands arrêts de la jurisprudence civile, 

11th ed. 2000, tome 1, no. 60 (9th ed. no. 58). 

24 

See for references Struycken, o.c. 2007, p. 250-252; Akkermans, o.c. 2008, 

p. 167. 

66 



www.sellier.de


to give effect to a retention of title clause vis-à-vis other creditors of the 

acquirer. Almost all developments in French property law are the result 

of legislation. 

Another legal system that supposedly does not endorse the numerus clausus 

principle is that of South Africa, one of the few legal systems that are 

often characterised as ‘mixed’ because of their origins in both common 

and civil law. Closer analysis reveals here as well that the South African 

courts have developed strict criteria for judging whether or not to recognise 

a new type of property interest. Property law in South Africa is 

therefore strictly regulated, if not by statute then by the courts. 25 

Common law systems also display a different picture than once believed. 

Many scholars from the common law tradition now think that the numerus 

clausus characterises the legal systems of many countries sharing 

that tradition as much as it does those of the civil law countries. Merrill 

and Smith conclude: 

‘In the final analysis, the idea that property interests may be created 

only in limited numbers of standardized forms has a very odd status 

in the common law. If one observes what lawyers and judges do, it is 

clear that the numerus clausus exerts a powerful hold on the system 

of property rights. (..) from the perspective of the practicing lawyer, 

the entire system presents the picture of a fixed menu of options from 

which deviations will not be permitted. The chances of persuading a 

court to create a new type of property in any particular case are too 

remote to be taken seriously. In this respect, property law has always 

been and continues to be very different from contract law. 

If one examines the official doctrine and the reasoning of the few cases 

that test the validity of the idea, however, the numerus clausus appears 

to have penetrated the consciousness of the common-law lawyers only 

weakly. Perhaps the best characterization of the status of the numerus 

clausus in American common law is that it is simply a fact about the 

way in which the system of property rights operates. The fact is so 

patent and obvious, so deeply entrenched, that it is rarely commented 

25 

See C.G. van der Merwe, Numerus Clausus and the Development of New Real 

Rights in South Africa, in: Contents of Real Rights, o.c. 2004, p. 99-113, and 

M.J. de Waal, Identifying Real Rights in South African Law: the ‘Subtraction 

from Dominium’ Test, in: Contents of Real Rights, o.c. 2004, p. 83-98. 



www.sellier.de


upon. But because it is so rarely commented upon, common-law lawyers 

have little to say in its defense when it is challenged.’ 26 

The absence of a numerus clausus in the common law of property appears 

to be a myth. 27 Furthermore, in the field of the American law of security 

interests, the common law rules have been superseded by Article 9 of the 

Uniform Commercial Code, which creates a uniform legal regime for all 

types of security interest, whatever form they take. 28 

A bone of contention remains the trust. For some, this common law concept 

constitutes the ultimate proof that there is no numerus clausus in 

the common law, because parties can use a trust to devise any type of 

property interest they like, which interest can be invoked by the beneficiaries 

against third parties. For others, the trust creates flexibility without 

burdening third parties, because bona fide acquirers of trust assets tend to 

be well protected against possible rights of the beneficiaries and nowadays 

trustees possess wide powers of disposal over the trust assets. 29 

26 

Thomas W. Merrill and Henry E. Smith, Optimal Standardization in the Law 

of Property: The Numerus Clausus Principle, 110 Yale L.J. (2000), p. 23-24. 

Contrast: Henry Hansmann and Reinier Kraakman, Property, Contract, and 

Verification: The Numerus Clausus Problem and the Divisibility of Rights, 

Journal of Legal Studies 31 (2002) 2, p. 373-420, § XIII, p. 417. 

27 

See also B. Rudden, Economic Theory v. Property Law: The Numerus Clausus 

Problem, in: Oxford Essays in Jurisprudence, Third Series 1987, p. 241 and 

243, and in Lawson / Rudden, The Law of Property, 3 e ed. 2002, p. 194-195, and 

W. Swadling, Property – General principles, in: English Private Law, 2000, vol. 

I, § 4.11-4.14. Cf. Akkermans, o.c. 2008, ch. 6. 

28 

The revolutionary form over substance approach in the field of personal property 

security becomes clearest from the first version of Art. 9 (Former Article 

9), in particular FA § 9-102(1)(a): ‘(..) this Article applies to any transaction 

(regardless of its form) which is intended to create a security interest in 

personal property (..)’. Compare the Official Comment 2 to RA § 9-109(a) 

(1): “When a security interest is created, this Article applies regardless of the 

form of the transaction or the name that parties have given to it.” The notion 

‘security interest’ is defined in the general part of the UCC, § 1-201(37): 

‘“Security interest” means an interest in personal property (..) which secures 

payment or performance of an obligation.’ 

29 

Merrill and Smith, p. 34. See also Merrill en Smith, The Property / Contract 

Interface, 101 Columbia LR (2001) 773, p. 848-849: “The popularity of the 

trust as a means of managing property for the benefit of one or more benefi- 

68 



www.sellier.de


A comparative analysis shows that all legal systems have articulate rules 

on the creation of property rights and a catalogue of specific types of property 

interests. 30 Full party autonomy seems to be anathema to property law. 

The only difference is the role played by statute: in some legal systems 

the numerus clausus is the product of case law rather than of intervention 

by the legislature. 

3.5. Rationale behind the numerus clausus 

The comparatist Rudden was one of the first to point out the enigmatic 

nature of the numerus clausus, not just for common lawyers but also for 

civil lawyers: 

ciaries has rendered the third-party information costs associated with such 

exotic interests far more manageable. The corpus of the trust can be bought 

and sold, invested and reinvested, leased and mortgaged, in the sound discretion 

of the trustee as if the property were an undivided fee simple. The 

complexities of dividing the fruits of these efforts among different generations 

and classes need not trouble the third parties who deal with the trustee as the 

manager of the trust corpus. The division is a concern only of the settlor, the 

trustee, and the beneficiaries, and is dealt with among them as a matter of in 

personam obligation. In effect, the trust is a brilliant device that allows for 

considerable customization of beneficial interests using the building blocks of 

the old common law – thereby reducing the frustration costs associated with 

strict regimentation of property forms – while at the same time consolidating 

the assets used to fund these beneficial interests in a form that minimizes third 

party information costs.”, and Henry E. Smith, The Language of Property: 

Form, Context, and Audience, 55 Stanford LR (2003) 1105, p. 1190: ‘Even 

the trust, that quintessential creation of equity, is designed to hide much 

informational complication behind the trust form where only the settlor, the 

trustee, and the beneficiary – but not potential purchasers and others in the 

“world” at large – have to process the complexity.’ 

30 

For an overview see Sachenrecht in Europa: systematische Einführungen und 

Gesetzetexte, Christian v. Bar (hrsg.), Osnabrück: Univ.-Verl. Rasch, 1999, 

volume 1 (Denmark – England – Finland – Sweden), volume 2 (Polen – Czech 

Republic – Hongary), volume 3 (Greece – Portugal – Netherlands) and volume 

4 (Belgium – France – Italy – Spain). 



www.sellier.de


‘Civilian doctrine has so little doubt that a closed list of real rights is 

necessary that its justifications are perfunctory.’ 31 

In 1987, he and the Swiss scholar Bénédict Foëx each contributed independently 

to a greater understanding of the reasons for having a numerus 

clausus. 32 Further analysis was carried out within the law and economics 

movement; Merrill and Smith, on the one hand, and Hansmann and 

Kraakman, on the other, later deepened and refined the economic analysis 


The reasons for a numerus clausus can be grouped into three clusters. The 

first has an ethical dimension: the numerus clausus serves to protect basic 

freedoms. When drafting the French Code civil, the French legislature 

sought to abolish old feudal rights that violated the fundamental freedoms 

of citizens. In addition, one of the objectives of the French Revolution 

was to obtain better protection of private property against expropriation 

by the sovereign without compensation. Both aims were achieved through 

the enactment of a statutory scheme in respect of property rights which 

scheme excluded property rights that were overly burdensome, banned 

many types of obligations in faciendo arising from property rights and incorporated 

an absolute notion of ownership that was immune from unfettered 

intervention by the state. The French Conseil Constitutionnel made 

clear in 1982 that the link between property and fundamental rights has 

not diminished in modern times: 

‘Si, postérieurement à 1789 et jusqu’à nos jours, les finalités et les conditions 

d’exercice du droit de propriété ont subi une évolution, caractérisée 

à la fois par une notable extension de son champ d’application 

à des domaines individuels nouveaux et par des limitations exigées par 

l’intérêt général, les principes mêmes énoncés par la Déclaration des 

droits de l’homme ont pleine valeur constitutionnelle tant en ce qui 

concerne le caractère fondamental du droit de propriété dont la conservation 

constitue l’un des buts de la société politique et qui est mis 

au même rang que la liberté, la sûreté et la résistance à l’oppression, 

31 

Rudden, o.c. 1987, p. 249. 

32 

Bénédict Foëx, Le “numerus clausus” des droits réels en matière mobilière, 

Collection Juridique Romande, Payot, Lausanne, 1987. See further: Wiegand, 

o.c. 1987, p. 623-643, and Wiegand, o.c. 1990, p. 112-138. 

70 



www.sellier.de


qu’en ce qui concerne les garanties données aux titulaires de ce droit 

et les prérogatives de la puissance publique.’ 33 

The numerus clausus reflects the ethical choices made by the legislature in 

a particular country, not merely by offering some types of property rights 

and abandoning others, but also by incorporating special protection for 

certain parties where deemed necessary. 

The second cluster has an economic dimension: the numerus clausus serves 

to increase trade. The French Code civil and, to an even greater extent, 

the German BGB of 1900 sought to facilitate the free flow of trade and 

commerce by reducing legal restraints and safeguarding the free transferability 

of goods. One of the means used was to deny proprietary effect to 

transfer restrictions and another was to offer protection to bona fide purchasers 

of property from a seller lacking title. More fundamental was the 

standardisation of property rights by providing clear rules on the content 

of each type and the rights and obligations flowing from it. Standardisation 

is a core aspect of the numerus clausus. It reduces the information 

costs and transaction costs in respect of property. 34 In addition, since the 

essence of property rights is that they have effect against third parties, 

the legislature can, by recognising certain rights in respect of property as 

property rights, facilitate and enhance the trade and use of that property 

in the manner it prefers. 35 

The third cluster has an organisational dimension: the numerus clausus 

is part of a legislative technique to increase legal certainty. The numerus 

clausus eliminates the need for general rules on property rights and a general 

definition thereof: adequate rules for specific types suffice. These rules 

concern not only the content of each type of property right, but also the 

specific requirements for its creation, transfer and abandonment. This 

dimension explains the rule set out in Article 3:83(3) BW (cited above), 

a rule that at first seems odd: rights other than ownership of things, li mited 

property rights and debts are transferable only to the extent provided by 

statute. This paradoxical rule need not be interpreted as a restriction on 

the transferability of new types of property rights. Rather, it can be un- 

33 

Conseil Constitutionnel 16 janvier 1982, D. 1983, 169; JCP 1982, II, 19788; 

F. Terré et Y. Lequette, Les grands arrêts de la jurisprudence civile, 11 th ed. 

2000, no. 1. 

34 

Merrill and Smith, o.c. 

35 

Hansmann and Kraakman, o.c. 



www.sellier.de


derstood as a reflection of the organisational dimension of the numerus 

clausus, because it presupposes that each type of right requires adequate 

and appropriate rules for its creation and transfer. This presupposition is 

of course related to the need to provide adequate means of publicity for 

rights that, being property rights, have effect against third parties. 

3.6. Impact of the numerus clausus on the 

creation of property rights 

The numerus clausus restricts party autonomy. In order to measure the 

impact of this restriction in a particular legal system, it is necessary to 

investigate how much room parties have to adapt and modify the type 

of property right they are creating, and what criteria the courts apply to 

determine the validity and effect of the result. An adaptation or modification 

may consist of either supplementing the powers and obligations 

arising from a specific type of property right or, as is more often the case, 

restricting or reducing those powers and obligations. 

Dutch property law does not allow every additional power or obligation 

to become part of the property right. Meijers’ explanatory memorandum 

to his first draft of the 1992 Burgerlijk Wetboek formulates the essence test 

(voldoende verbandvereiste) as follows: if parties agree to additional powers 

or obligations upon creating a property right, those powers and obligations 

may only be treated equally with the powers and obligations that are part 

of the statutory content of that right if the additional powers or obligations 

have a sufficient connection with the essence of that type of property 

right. 36 For example, in respect of emphyteusis, only obligations that 

directly concern the enjoyment and the maintenance of the encumbered 

land can be added to the content of this right and have effect against third 

parties and acquirers of the land. 

Furthermore, Dutch property law is thought to be reluctant to recognise 

obligations in faciendo as proprietary. Although the law on this point is uncertain, 

where a party seeks to add an obligation in faciendo to a property 

36 

TM, General remarks on Book 5, sub 3, third paragraph, Parl. Gesch. Book 

5, p. 3. See further: H.W. Heyman, Contents of the Real Right: Dogmatic 

Rigidity and Pragmatic Flexibility of Dutch Property Law, in: Contents of Real 

Rights, o.c. 2004, p. 71-81. 

72 



www.sellier.de


right, that obligation will only be accepted as part of the property right if 

it resembles an obligation to maintain and service land or buildings, or an 

ancillary obligation to pay a charge for the use of the property of another. 

Maintenance and ancillary payment obligations are the sole examples of 

property obligations in faciendo forming part of the statutory content of 

some types of property rights. 

More often, parties seek to reduce or restrict the scope of a particular 

property right. Three ways of doing this can be distinguished. The first 

is where the grantor and grantee of the property right agree that only 

certain specified powers will accrue to the grantee and not all the powers 

normally arising from that property right. With respect to a right of 

usufruct, this could mean limiting the statutory powers (usus and fructus) 

to, for instance, the power to enjoy the physical fruits and the shade of 

a tree owned by the grantor (but not the dead wood). With respect to a 

servitude of way, this could mean restricting it to the right to enter the 

property on horseback for the purposes of a fox hunt. Modification of a 

property right by specifying and restricting the powers conferred on the 

grantee – the power dimension – implies a deviation from the statutory 

content of such right. 

The second way in which parties attempt to restrict a property right is by 

limiting its scope to a part of the property – the spatial dimension. The 

classic manner of effecting a spatial delimitation of land is by vertically 

splitting the parcel into two smaller ones. The Hoge Raad has held that 

such a delimitation is also possible if there is a building on it, provided 

that the building could, according to common opinion (verkeersopvattingen), 

also be split into functionally and technically separate parts. 37 

Much more doubtful is the validity of splitting the rights to a building into 

horizontally separate parts. Increasingly, complex building projects are 

legally structured by granting separate property rights for different layers, 

such as a right of emphyteusis (erfpacht) for the underground parking and a 

separate right of emphyteusis for the building’s foundations (the former being 

limited to the underground space) plus one or more separate rights of 

superficies (opstal) for one or more levels of the building above the ground. 

It is often difficult to say whether the parties intended to create a property 

right in respect of only part of the property or a property right in respect of 

the whole property but conferring powers merely in respect of a specified 

37 

See in the Netherlands HR 29 May 1985, NJ 1986, 274 (Tandartsenpraktijk). 



www.sellier.de


part thereof. The former legal structure is unlikely to be valid, as it runs 

counter to the principle that a property right can only concern property 

as a whole. The latter structure may be thought to amount to an evasion 

of the statutory regime on apartment rights (appartementsrecht). 

The third way in which parties sometimes attempt to restrict a property 

right is by limiting it in time – the time-dimension. Time restrictions 

come in two forms. Firstly, parties may seek to restrict the duration of a 

property right itself; in Dutch law, only the right of usufruct is characterised 

by mandatory statutory limits on its duration. Secondly, parties 

may restrict the right to exercise the powers conferred by the right to some 

set period each year. Timeshare is a phenomenon that fits rather uneasily 

into the classic notion of a property right. Again, it is often uncertain 

whether, for example, the parties intended to create a property right that 

exists only on Mondays, or a continuous right that confers powers only 

on Mondays. The validity of the former variant of timeshare is even more 

doubtful than the validity of the latter. 

The three dimensions illustrate the ways in which parties may want to 

modify and adapt the types of property rights offered by the law of property. 

The room available for party autonomy in this field depends greatly 

on the boundaries set by the numerus clausus. 

3.7. Dual ownership and the numerus clausus of powers 

The statutory prohibition of fiduciary transfers of ownership in Article 

3:84(3) BW has banned from Dutch property law the internationally 

widely-used technique for securing the repayment of credit by way of a fiduciary 

transfer. Furthermore, it cut short a possible development towards 

ownership that is split into legal ownership and beneficial ownership, 

which is generally regarded as characteristic of the English-style trust. 

Fiduciary ownership, whether cum creditore or cum amico, is thought to 

create dual ownership, which is difficult to reconcile with the civil law 

notion of absolute and exclusive ownership. 

The reasons for the fiducia-verbod are a matter of both principle and 

pragmatism. In the perception of Meijers, a legal system that separates 

legal title (recht) from beneficial interest (belang) is flawed; as a general 

principle, the legislature should opt to confer title on the party with the 

74 



www.sellier.de


prime interest. Conferring full title on a creditor instead of only a security 

interest would give him more rights and powers than he needs and 

would constitute an “overdistribution” (overbedeling) of rights and powers 

to that creditor. The issue here is not so much one of justice as one of 

pragmatism in the organisation of the law: the aim of the fiducia-verbod 

is to provide the debtor with sufficient standing in law (locus standi), by 

way of a bare ownership right, to enable him to safeguard his interest in 

the asset controlled by the creditor. As a result of forcing such parties to 

create a pledge instead of transferring ownership for fiduciary purposes, 

ownership remains with the debtor, who therefore retains his locus standi 

and can safeguard his interests. Furthermore, this enables him to profit 

from his ownership interest by transferring it (subject to the security interest) 

or creating a second right of pledge. 38 

Similar reasons explain the prohibition of fiducia cum amico. If it is necessary 

for one party’s property to be administered by another who is more 

careful or competent, it does not make sense, according to Meijers, to 

deprive the incompetent party of his full ownership and grant it to the 

other party, who requires only powers of administration. In such cases, 

the legislature should opt to confer those powers on the administrator 

and give them effect against both the owner himself and third parties. Instead 

of the English-style trust, Meijers proposed to expand the old Dutch 

concept of bewind. 39 The essence of this concept is that the party whose 

property is made subject to the bewind by a court (the onderbewindgestelde), 

remains the owner, but loses a number of his ownership powers: the powers 

of administration and of disposal are conferred on the administrator 

(bewindvoerder). Thus, the administrator gets the powers he needs, and 

the incompetent party is not deprived of more than is necessary. 

In the light of this, two phenomena in current Dutch property law should 

be mentioned. On the one hand, there is the implicit statutory recognition 

of the retention of title clause as a valid security device that can be 

invoked against other creditors of the buyer. Retention of title is functionally 

so close to a transfer of ownership for fiduciary purposes that the difference 

in treatment is hard to explain. Equally important is the pragmatic 

dimension: the seller does not need more than a security interest, whereas 

38 

See Struycken, o.c. 2007, § 7.4.2. 

39 

E.M. Meijers, De trustee en het burgerlijk recht (recensie preadviezen BCN 

1927), WPNR 2998 (1927), p. 413-416, VPO III, p. 184. See Struycken, o.c. 

2007, § 7.5.3. 



www.sellier.de


the buyer needs more than an expectancy of ownership, not only to protect 

his interest against actions by the seller or third party that affect the 

property, but also to be able to capitalise on it through a transfer or pledge. 

Retention of title therefore runs counter to the underlying philosophy of 

the system of Dutch property law. 

On the other hand, there is the controversial provision in Article 3:85 

BW under which a transfer of ownership for a set period of time is converted 

by operation of the law into the creation of a right of usufruct. 

This rules out the possibility of temporary ownership in Dutch property 

law. This provision, which many find difficult to explain, fits in perfectly 

with Meijers’ views on an efficient and pragmatic manner of organising 

property law. The usufruct provides the intended temporary owner with 

the use and fruits of the property for a limited period of time, while the 

intended subsequent owner already gets a bare ownership right, which not 

only gives him locus standi to protect his future interest but also enables 

him to trade it for value. 

Thus, the system developed by Meijers consists of three categories of 

entitle ments in respect of an asset: personal rights, proprietary rights and 

what can best be described as proprietary powers in respect of an asset. 

Examples of the latter, apart from the powers of a bewindvoerder, are the 

powers of a liquidator (faillissementscurator) in respect of the assets in an 

insolvent’s estate and the powers of a notary and a lawyer in respect of 

client money in a special account (kwaliteitsrekening). 

Such proprietary powers over someone else’s property, which powers 

should be carefully distinguished from rights (subjectieve rechten), are exceptional 

in nature and exist only where prescribed by statute: there is a 

separate numerus clausus of such extraordinary proprietary powers. 40 As a 

consequence, parties cannot split ownership in ways not provided for by 

statute. More specifically, the numerus clausus prevents parties from splitting 

off a power comprised in a property right and conferring that power 

40 

A subjective right (subjectief recht) differs from a ‘proprietary power’ in that a 

subjective right can be transferred, seized by a third party and pass by universal 

succession (vatbaar voor overdracht, beslag en overgang onder algemene titel), 

whereas none of these is possible with a proprietary power. As a consequence, 

what happens if the holder of a special proprietary power dies or ceases his 

activities is determined entirely by statute. See further Struycken, o.c. 2007, 

§ 7.8 and 7.9. 

76 



www.sellier.de


separately on another person. This is possible only if the power either 

takes the form of one of the statutory property rights in re aliena, or is 

designated by statute as a special proprietary power. 

3.8. Scope and consequences of the numerus clausus 

The numerus clausus raises three further questions. 41 First, does the law of 

property allow parties to make use of alternative property rights and structures 

instead of the statutory property right created for such situations? 

For instance, by reshaping and modifying a right of superficies, it seems 

possible to replicate a right of emphyteusis without actually choosing this 

type of right. Another example is a perpetual right of emphyteusis without 

an obligation to pay a charge; such a limited property right mimics 

ownership, except that the holder of the right of emphyteusis may incur 

some obligations arising from the limited property right which obligations 

could not bind him and his successors as owner. In addition, co-ownership 

and a transfer subject to a condition subsequent (ontbindende voorwaarde) 

could in theory make it possible to circumvent statutory restrictions on 

property rights. The extent to which such creativity is allowed under 

Dutch property law is unclear. 

Secondly, where the law offers a type of property right for a specific situation, 

may the parties use the law of obligations to supplement or imitate 

that type of property right? The answer to this question is probably yes, but 

the likelihood decreases if the alternative achieves almost the same effect 

without satisfying the safeguards installed by the law of property. 

The third question is whether the numerus clausus allows for an entirely 

different application of a certain type of property right in new situations. 

A statutory property right can be seen as reflecting a legislature’s choice 

to offer that type of right. The legislature will have had in mind the situations 

in which that right is to be used. Seen in this perspective, it is not 

necessarily self-evident that parties are free to use that type of property 

right in a completely different setting in order to meet a completely new 

and different need. For instance, the right of emphyteusis has evolved from 

a legal framework for the development of raw land for agricultural use to 

41 

A fruitful source of inspiration is Philipp Heck, Grundriß des Sachenrechts, 

1930, §§ 21 and 23. 



www.sellier.de


a device used by municipal authorities to impose restrictions concerning 

the location, use and external appearance of buildings and a vehicle for 

structuring complex building projects. 

These three questions regarding the creativity allowed by the numerus 

clausus principle automatically raise the issue of the consequences of an 

infraction. Nullity is one option; mere contractual effect (whereby the 

right cannot be invoked against third parties or by successors) is another. 

In some cases, the application by analogy of the set of rules governing 

the alternative property right (samenloop) may provide the protection 

envisaged by the numerus clausus. Lastly, a property right modified beyond 

the limits of the numerus clausus could be converted by operation of law 

into another type of property right. The answers to these questions under 

Dutch law are not yet clear. 42 

3.9. The future of the numerus clausus 

The question of whether there is a future for the numerus clausus in the law 

of property should be addressed broadly, because the whole of Savigny’s 

model for private law is at stake. One cannot assess the significance of the 

numerus clausus without also taking into account the notion of ownership, 

the third party effects of powers and obligations that result from property 

rights, the rules on transfer restrictions and the protection of bona fide 

purchasers. The whole system is geared towards enhancing the free flow 

of goods, and also safeguarding legal certainty and foreseeability for those 

involved with property of others in one way or another. The underlying 

reasons for a numerus clausus, be they ethical, economic or organisational 

in nature, have not lost their appeal. 

The Dutch legal system, however, shows a tendency towards shifting control 

from the phase in which rights are created (ex ante) to the phase in 

which they are exercised (ex post). The 1992 Burgerlijk Wetboek contains 

a fairly complete framework of imprévision provisions, which give courts 

discretionary power to, upon request, limit, change or even abrogate property 

rights in re aliena when they have, over the course of time, become 

disproportionately onerous to the parties involved, or run counter to the 

42 

See further Struycken, o.c. 2007, ch. 8. 

78 



www.sellier.de


general interest. 43 More generally, the provisions on abuse of power and 

on objective good faith (redelijkheid en billijkheid) create means for adjusting 

the relationship between parties in a concrete case. 44 It remains to be 

seen, however, whether these new checks and balances provide sufficient 

safeguards. In my view, these new judicial instruments constitute useful 

additional tools but cannot fully replace the numerus clausus as a means 

for enhancing legal certainty and foreseeability – both key elements of any 

law of property – and for facilitating commerce and trade. 

The emergence of ex post control mechanisms opens up a new perspective 

that could give new significance to the numerus clausus and the closed 

system of the law of property. If the doctrines of imprévision, abuse of 

power and redelijkheid en billijkheid together reflect a general principle of 

proportionality, it becomes worthwhile to investigate which other principles 

underlie the law of property. Together with the proportionality principle, 

a total of seven principles can perhaps be extracted from the rules 

of Dutch property law: the causality principle, being the rule that the 

system of transfer of property is causal rather than abstract; the delivery 

principle, being the rule that a transfer of ownership does not occur ex 

solo consensu but requires delivery or some other formality; the specificity 

principle, being the rule that property rights and the creation thereof 

require a specific object; the non-splitting principle, being the basic idea 

that a property right cannot be split in ways not provided for by statute 

and the rule that a property right and an act in respect thereof concern 

the whole of the property to which the right relates; the autonomy or 

transferability principle, being the concept that property is transferable to 

the greatest extent possible and that persons have largely unfettered powers 

of disposal; and the publicity principle, being the imperative that the 

general public is able to ascertain the property rights of each person, their 

content and transfers thereof. The numerus clausus can be understood as 

a mandatory system of fundamental principles that underlie the whole of 

the law of property. 45 

Comparative analysis of other legal systems shows that the numerus clausus 

is not merely a choice for or against party autonomy in the law of property; 

it is also a matter of the allocation of powers between the courts and the 

43 

Articles 5:78, 5:80 and 5:97 and 6:259 BW. 

44 

Article 3:13 BW (abuse of a power) and Articles 3:166(3), 6:2 and 6:248 

BW. 

45 

See Struycken, o.c. 2007, § 9.5.1. Cf. for Germany Füller, o.c. 2006, passim. 



www.sellier.de


legislature. In origin and in essence, the numerus clausus confers exclusive 

jurisdiction for the recognition of new types of property rights on the 

legislature. However, incontestably, the courts in all legal systems in fact 

play an important role in developing and shaping the specific property 

rights and the rules that govern them. A debate on the numerus clausus is 

therefore also a debate on who should take the lead in the development 

and modernisation of a law of property: the courts or the legislature? 

Put in these terms, the future of the numerus clausus touches upon the 

fundamental justifications for codified law. In this debate, I opt for maintaining 

and enhancing the primacy of the legislature. Apart from the 

well-known general reasons for codified law, such as legal certainty, foreseeability, 

and democratic legitimacy, the primacy of the legislature in the 

development of property law is justified by the impact of property rights 

in society. Their effect on third parties requires a higher level of certainty: 

third parties should be able to know what to expect. Also, the long-term 

effects of property rights require policy choices for which the legislature 

is better equipped than a court. 

Maintaining the primacy of the legislature in the field of property law 

should not be a gratuitous gesture. It implies that the legislature has a 

duty to act and enact where and when the rules of property law require 

change or modernisation. The legislature should take a more active and 

leading role in adapting property law to the needs of commerce and society. 

This responsibility, however, is one that the executive and legislative 

branches share with the legal community. Practitioners and scholars 

should proactively support the legislature by clarifying the changes 

necessary to modernise the law of property and drawing up legislative 

proposals. In this way, the numerus clausus in the law of property becomes 

a living dogma, too significant and too vital for Vera Bolgár’s museum of 

Begriffsjurisprudenz. 

3.10. Party autonomy in the law of property – 

the way forward 

The numerus clausus and party autonomy being flip sides of the same coin, 

the question of whether there is a future for the numerus clausus in the law 

of property inevitably entails a discussion of the room that individual parties 

should be given to shape their legal relationships. There is a marked 

80 



www.sellier.de


doctrinal tendency to favour party autonomy in both domestic and international 

property law. 46 Party autonomy is difficult to oppose as a general 

principle. It is consistent both with prevailing political views on the position 

of the individual in relation to his society, and with the practical 

needs imposed by an increasingly complex global economy. Furthermore, 

there is not a single argument against it that is both simple and persuasive. 

However, the discussion among legal commentators on the extent 

to which party autonomy should be given a place in the law of property 

sometimes seems to lack detail and precision. 

In order to understand the impact of party autonomy in the law of property, 

it is important to identify exactly whose interests are at stake. The 

reason why the law of property – in contrast to that of contract – is mandatory 

is generally said to be that property rights are binding on third parties, 

whereas contractual rights and obligations merely bind the parties to 

the contract. Property rights are traditionally presented as absolute rights 

which can be opposed and invoked against third parties. Such parties 

could be prejudiced if, through an agreement, new types of rights in rem 

could be created, existing types could be modified or new methods losing 

or acquiring a right in rem could be devised. If and to the extent such a 

real right imposes an affirmative duty to act (in faciendo), the third party’s 

interest becomes even acuter. 

But who, precisely, are these third parties? To whom, and to what extent, 

does it make a difference whether a right is characterised, or created, 

as either contractual or proprietary? To determine the impact of party 

autonomy in the law of property, it is essential that these questions be 

answered with a sufficient level of detail. Several categories of relevant 

parties should be distinguished, apart from the holder of the right in respect 

of the asset owned by another person: (i) those who succeed to that 

right in respect of the asset of another by universal succession (overgang 

onder algemene titel); (ii) those who succeed to that right by singular succession 

(overgang onder bijzondere titel); (iii) those who have prior rights 

in respect of the same asset; (iv) those who acquire ownership of the asset 

46 

See the outspoken views in favour of more party autonomy expressed by Axel 

Flessner, Rechtswahl im internationalen Sachenrecht – neue Anstöße aus 

Europa, in: Festschrift für Helmut Koziol zum 70. Geburtstag, 2010 Jan Sramek 

Verlag (in translated version: see Chapter 1 in this book on Party Autonomy 

in International Property Law, Choice of Law in International Property Law, 

New Encouragement from Europe). 



www.sellier.de


in respect of which the right has been granted, by succession (whether 

universal or singular); (v) the trustee in bankruptcy of the owner of the 

asset; (vi) those who levy an attachment of the asset; and (vii) generally 

all those who somehow risk to interfere with the asset without asserting 

rights of their own, such as those who cross somebody else’s land or collide 

with somebody else’s car. 

To resolve the conflict between numerus clausus and party autonomy in 

the law of property, it is insufficient to simply state that property rights 

affect third parties and are ‘absolute’; a proper analysis requires pinpointing 

who precisely is affected by the right and in which circumstances. 

Only then will it be possible to argue convincingly either that the law of 

property and the law of contract should be differentiated as far as party 

autonomy is concerned or, alternatively, that party autonomy should be 

extended to the law of property. 

Furthermore, it is important to discuss the safeguards against potential 

excesses of party autonomy, such as the rules on publicity, on the protection 

of a bona fide acquirer of a right from someone who lacks full title to 

the relevant asset, the rules against abuse of power and the courts’ powers 

to modify or terminate a property right in exceptional circumstances. If 

the necessary limits to party autonomy are made explicit, this will help 

to make party autonomy in the law of property more acceptable than it 

used to be. 

82 



www.sellier.de

4. 

Party Autonomy and Property Rights 

Robert Stevens * 

4.1. Numerus clausus 

It is central to the rule of law that we are all readily able to discover the 

duties that we owe, and to whom that we owe them. This basic principle 

has consequences for the class of rights recognised within the law of 

property. 

If we were Romans, what we would mean by a right in rem, in contradistinction 

to a right in personam, would be clear but it would not correspond 

with all modern usages of these concepts. A right in rem would be a right 

in relation to a physical thing, and that right would be exigible against, 

or binding upon, all others. A right in personam is a right against another 

person and is only exigible against that person. 

In England, if I wish to claim for a tort, a wrong, I must show that a duty 

owed to me has been breached. Put identically, I must show that a right I 

had as against the defendant has been violated. If the wrong committed 

is in relation to a thing, I must show that I have a right in relation to the 

thing exigible against the defendant in order to have standing to bring 

a claim. For example, if a railway company leases the use of a bridge to 

enable its trains to cross, and the bridge is negligently damaged by a lorry 

driver causing the company loss because it is unable to operate its service, 

it will be able to claim because a lease of land, in England, involves the 

conveyance to the leaseholder of an interest in the land, a right in rem. 

If, by contrast, the railway company merely has a contractual licence to 

use the bridge, it only has a right as against the owner of the bridge, no 

right in relation to the bridge itself. It cannot, therefore, sue the lorry 

driver for loss it suffers as a result of the damage. 1 The licensee may have a 

claim against the bridge owner, and the bridge owner will have a claim for 

* 

Barrister, Professor of Commercial Law, University College London. 

1 

The Aliakmon [1986] 1 AC 785, HL. 

83 


www.sellier.de


the wrong as against the lorry driver, but the latter claim will only permit 

the recovery of the bridge owner’s losses, not the licensees as such (English 

law does not recognise the German concept of ‘transferred loss’). 

It follows that for any legal system the category of rights which are in rem 

in the Roman sense must be determinate. If parties were free to agree to 

create any rights in rem that they so chose, third parties would be unable 

to discover what the precise content of the duties they were under were, or 

to whom they were owed. Consequently all legal systems have a numerus 

clausus of rights in rem in the full-blooded Roman sense, and English law 

is no exception to this. 

What goes on the list and what does not can, and does, vary from one 

system to another, so the English characterisation of a lease is not one 

which is universally adopted. 2 There is no demonstrably right answer as 

to whether an easement over land should, or should not be accepted 

on to the list. Further, the exact definition of each property right may, 

again, legitimately vary from one jurisdiction to another with none being 

demonstrably superior. However, what constitutes a right in rem in the 

Roman sense must be fixed by law. The parties cannot agree to create a 

right exigible against all others which is not on the list. A common law 

lease, for example, requires the lessee to have exclusive control of the land 

leased. If this is not granted, all that has been created is a licence, which 

as its name implies is simply a privilege as against the right holder to use. 

The parties cannot transform a contractual right into a right in rem by 

choosing to call the right created a right in rem. If the agreement does not 

confer a right to exclusive possession no property right has been created, 

even if the agreement calls itself ‘a lease’. One of the most over-quoted 

judicial statements to this effect is that of Lord Templeman: 

‘The manufacture of a five-pronged instrument for manual digging 

results in a fork, even if the manufacturer, unfamiliar with the English 

language, insists that he intended to make, and has made, a spade.’ 3 

2 

See for example Regulation (EC) No 44 / 2001 on jurisdiction and the recognition 

and enforcement of judgments in civil and commercial matters (Brussels 

I), OJ L 012, Article 22(1): ‘(..) procedings which have as their object rights 

in rem in immovable property or tenancies of immovable property (..)’. As a 

matter of English domestic law, the former includes the latter. 

3 

Street v Mountford [1985] A.C. 809, 819. 

84 



www.sellier.de

4. Party Autonomy and Property Rights 

One feature of the common law which it does not share with all others is 

that it does not have an indivisible unitary concept of ownership, so it is 

perfectly possible to have more than one person with rights good against 

the rest of the world in relation to a thing. Even a thief acquires a title to 

goods which is better than the title of all others to the goods, save the person 

from whom he took them. So, if Y steals a watch from X, and is then 

in turn robbed by Z, Y may sue Z for the wrong of conversion, 4 and obtain 

delivery up of the goods. The fact that there is another party (X) with a 

relatively better title than Y has cannot be relied upon by Z. Y’s title to the 

goods is original and arises by possession, as at root do all rights to physical 

things. X’s right to the goods has not been transferred; it is still vested in 

him. That X and Y both have rights in relation to the goods which they 

can exert as against Z does not offend the numerus clausus principle, as the 

right they are both asserting is simply one of ownership, and not a bespoke 

right falling outside of the categories recognised by the law. The position 

in English law may be contrasted with, for example, the employment 

by the Draft Common Frame of Reference (Book VIII, Article 1:206 et 

secuens) of an ‘owner-possessor’ who is distinct from the straight-forward 

owner of goods. This distinction is unknown to the common law. 

Similarly, some legal systems are not as ready as English law is to allow the 

intentional division of rights in rem in land so as to confer property rights 

in multiple persons in relation to the same thing. Our numerus is not as 

clausus as in some other legal systems, but the list is still very short indeed. 

So, although in England there is no doubt that both a landlord and his 

tenant have a right in rem, so that either could claim against a third party 

who interfered with their interest in the land, a lease of goods by contrast 

does not, by itself, transfer any right in relation to the goods themselves, 

although the lessee will usually have standing to claim if the goods are 

damaged as a result of his rights arising from his possession of them. In 

this regard the lessee of goods does not differ from the thief. Another 

right in rem in relation to land is an easement, whereby someone with a 

freehold or leasehold interest in land has a right to limited use of nearby 

land which he does not own. Again, interference by a third party with this 

right to use is an actionable nuisance, and an easement is consequently a 

right in rem in the full-blown Roman sense. There is no equivalent right 

in relation to goods. 

4 

Armory v Delamirie (1722) 1 Strange 505. 


85 


www.sellier.de


Although it has been argued that the numerus clausus principle is economically 

efficient, 5 this is neither here nor there. Any legal system which 

allowed the creation of idiosyncratic rights which imposed duties on third 

parties which they could not ascertain or predict would offend basic principles 

of morality. 

All modern legal systems have however moved beyond Roman law and 

recognise classes of rights which are neither in rem nor in personam in the 

sense the Romans would have recognised. In modern day usage we confer 

the label “property rights” upon classes of rights which would not have 

been recognised as in rem by our Roman forbears. This has important 

consequences for the scope of the numerus clausus principle. 

4.2. Intellectual property: a brief digression 

Intellectual property rights are not in rem in the Roman sense because 

they do not relate to any physical thing. Indeed, they have no subject matter 

independent of the right itself which is capable of being transferred. If 

I have a right in rem in relation to a computer, if that computer is stolen 

and then given to a third party, I can follow my right in relation to the 

thing into the hands of the donee. With an intellectual property right, 

by contrast, there is no separate thing to follow. If a third party publishes 

my book in breach of copyright, or sells copies of my patented invention, 

or sells a soft drink using the name my rival drink has employed as its 

trademark, a wrong is committed. However, no thing has been transferred 

to or interfered with by the wrongdoer. Although, perhaps astonishingly 

to the outsider, the common law possesses no rei vindicatio in relation to 

personal property, it can be readily seen that no claim of the form “that 

is mine give it back” can be asserted in relation to intellectual property 

rights. Intellectual property rights are exigible against the rest of the world 

but have no independent subject matter. 

It is because of the fact that they are exigible against all others that we 

cannot create any intellectual property right of any form of our choosing. 

To constitute an intellectual property right a right must fall within one 

of the definitions of such a right; if it does not no intellectual property 

5 

H. Smith. and T. Merrill, ‘Optimal Standardization in the Law of Property: 

The Numerus Clausus Principle’, 110 (2000) Yale Law Journal, 1. 

86 



www.sellier.de


right has been created. Information does not, without more, constitute 

intellectual property. At the margin there may be a dispute (e.g. can you 

patent a DNA sequence?) but we have a closed list of intellectual property 

rights for precisely the same reason that we have a closed list of rights in 

rem in the Roman sense. 

4.3. Transfers 

All legal systems have restrictions on the transfer of rights in rem, requiring, 

for example, the registration of title to land or a physical delivery of 

goods before the transfer can take place. One reason for these restrictions 

is a desire to ensure that third parties are able to discover who is the holder 

of these rights, so that they are able to determine to whom they owe their 

duties with respect to the thing. Just as it is important that all of us are able 

to determine the duties we are under, we need to know to whom we owe 

them. A secondary reason for these restrictions is to prevent fraud, fortunate 

earlier ‘transfers’ of property by persons who have become bankrupt 

to relatives are less easy to concoct in a world which requires registration 

for a transfer to be effective. 

In England, in relation to all forms of property, other than goods, an agreement 

in any form cannot alone transfer one party’s rights to another. For 

forms of property other than goods either the transfer, or conveyance, of 

the right is wholly independent of the contract itself under which the 

parties agree to effect the transfer, or the contract must be in a particular 

form to effect transfer. 

Common lawyers are probably most familiar with this proposition in relation 

to land. So, in relation to unregistered land a deed of conveyance 

is required before transfer is made, and in relation to registered land an 

alteration to the land registry is necessary. 6 

Precisely the same applies to the transfer of other rights which are not in 

rem in the Roman sense. So, if debts were freely transferable by creditors, 

debtors would not know to whom their debts were owing and may pay 

the wrong party. It is impossible to transfer receivables by way of contract 

alone: notice to the debtor is required and even this may not suffice where 

6 

Almost all land is nowadays registered. 


87 


www.sellier.de


the receivables are, by their terms, non-assignable. 7 Registrable securities 

such as shares may only be transferred by alteration of the share register. 8 

If securities such as shares could be transferred by agreement alone, how 

could the parties the share imposed duties upon, in particular the company, 

know to whom they were supposed to perform? The disposition of an 

equitable interest, the nature of which we will come to, may be executed 

by contract alone, but this must be in writing signed by the person disposing 

of the same or by his agent. 9 Similarly, the assignment of a trade 

mark 10 , copyright, 11 design, 12 or patent 13 must be in writing and signed by 

or on behalf of the assignor. Those intellectual property rights which are 

registrable also need to be so registered in order to be fully effective as 

against third parties. 14 

Goods are different. The title to goods can, of course, be transferred by 

deed or by delivery (e.g. Christmas presents) but their peculiarity lies in 

the fact that the right to goods can be conveyed by a contract in any form 

without more. One of the phrases English lawyers learn before they can 

walk is that ‘property passes when intended to pass’. 15 This magic spell 

embodies the rule that title to goods can be conveyed without the need 

for any further step or special form to perfect it: the contract alone suffices. 

Here, at least, the Germanic separation between contract and conveyance, 

Trennungsprinzip, is not found at all. The explanation as to why the 

law for the transfer of the right to goods should be different from the law 

for the transfer of all other rights appears to be historical accident. 

7 

Linden Gardens Trust Ltd v Lenesta Sludge Disposals Ltd [1994] 1 AC 85. 

8 

Technically, where what is “transferred” is a share what the ‘transferee’ acquires 

is a new share by way of novation, the ‘transferor’’s old rights being 

cancelled. See M. Ooi, Shares and Other Securities in the Conflict of Laws 

(Oxford University Press, 2003). A share is a proportionate right as against 

the company and other shareholders, the rights of all shareholders of the same 

class are the same. 

9 

Law of Property Act 1925, s 53(1)(c). 

10 

Trade Marks Act 1994, s 24(3). 

11 

Copyright, Designs and Patents Act 1988, s 90(3). 

12 

Copyright, Designs and Patents Act 1988, s 222(3). 

13 

Patents Act 1977, s 30(6). 

14 

Patents Act 1977, s 33; Trade Marks Act 1994, s 25. 

15 

Sale of Goods Act 1979, s 17. 

88 



www.sellier.de


4.4. Party autonomy 

Given then that English law, like all systems, has a numerus clausus in 

relation to all rights which are exigible against all others, including both 

rights in rem in the Roman sense and intellectual property rights, and also 

that English law, like all systems, has several restrictions on the transfer of 

rights by agreement alone, it would at first sight appear to be the case that 

party autonomy is severely restricted. This would be, of course, commercially 

inconvenient. That this is not so is because English law, like other 

legal systems, has transcended the Roman law dichotomy between rights 

in rem and rights in personam. This is also true of legal systems outside of 

the common law world but the pattern is neither uniform nor (unfortunately) 

widely understood. 

The essential concept to grasp is that of rights to rights. 

If A, the owner of Blackacre declares himself a trustee of the right to his 

land in B’s favour, English law does ‘not say that [B is] the owner of the 

land, it says that the trustee was the owner of the land, but adds that he is 

bound to hold the land for the benefit of [B]’. 16 We may illustrate this by 

the fact that B has no standing to sue for violation of the right to the land 

itself: only A may sue a trespasser or the creator of a private nuisance. B 

has no right to the land, and does not have a right exigible against the rest 

of the world but only one as against the person with title to Blackacre from 

time to time. B does not, therefore, acquire in the narrow sense employed 

by the Romans, a right in rem. 

B’s right that A holds the title to the land for his benefit (which is simply 

the converse of A’s duty to hold the title to the land for B’s benefit) is 

not simply a right against A personally, a right in personam in the sense 

employed by Roman lawyers. Rather the subject matter of B’s right is A’s 

right to the land. 17 If, therefore, A gives the (right to) Blackacre to his 

16 

F.W. Maitland, Equity – a course of lectures (A.H. Chaytor and W.J. Whittaker, 

eds., revised by J. Bunyate), Cambridge, 1936, 17-18. 

17 

See also R. Chambers, An Introduction to Property Law in Australia (2 nd ed., 

Lawbook Co, 2008) at [13.90]; L Smith, ‘Unravelling Proprietary Restitution’, 

Canadian Business Law Journal, (2004), 317; L Smith, ‘Philosophical Foundations 

of Proprietary Remedies’ in: R. Chambers et al. (eds.), Philosophical 

Foundations of The Law of Unjust Enrichment (Hart, 2009); L Smith ‘Trust 

and Patrimony’, Revue générale de droit 38 (2008), 379-403; L. Smith, Estates, 


89 


www.sellier.de


son, C, B’s right will bind C because B’s right is in relation to the right 

C has acquired. Similarly, if A goes into bankruptcy, B’s right against A’s 

right will bind A’s trustee in bankruptcy and hence will be protected in 

this event. 18 

Whenever one party is obliged to hold a right for the benefit of another a 

trust arises, because that is, definitionally, what a trust is. So, if A makes a 

binding contract to transfer specific shares, or receivables, or title to land, 

or an intellectual property right, or an equitable interest to B, A becomes 

a trustee of this right upon payment even though the requisites of transfer 

have not been fulfilled. This does not offend the rules relating to the 

transfer of rights because a trust involves no transfer. 19 Once the transfer 

has been made, there is no longer any trust because there is no longer any 

duty owed to anyone. In English law, I have one single indivisible right to 

the computer on which I am currently typing, not two rights one at law 

and another in equity. 

So called ‘equitable assignments’ are also of the above kind. 20 Contractual 

rights were, and are, non-transferable because of privity of contract: 

only someone who has been promised something by the promisor and 

who has provided consideration in exchange for that promise acquires a 

contractual right. Allowing the free transfer of contractual rights would 

offend privity of contract by allowing people who satisfy neither of these 

conditions to sue. An equitable assignment, despite its name, does not 

Trusts and Pensions Journal 28 (2009), 332; W. Swadling in: Burrows (ed), 

English Private Law (2 nd edn, 2007) at 4.140-4.153; B. McFarlane and R. 

Stevens; ‘The Nature of Equitable Property’, Journal of Equity (2010) 1; cf R. 

Nolan, ‘Equitable Property’, LQR 122 (2006), 232. 

18 

For bankrupt individuals this is expressly stated in the Insolvency Act 1986, s 

283(3)(b). It is also true of insolvent companies despite the absence of expression 

in legislation: see e.g. Barclays Bank Ltd v Quistclose Investments Ltd [1970] 

AC 567. 

19 

See e.g. Lysaght v Edwards (1876) 2 Ch D 499; Walsh v Lonsdale (1882) 21 Ch 

D 9. In principle the same concept applies where A’s duty to B with respect 

to a right of A’s arises for reasons other than a contract between them (e.g. 

because of unjust enrichment or an equitable wrong). 

20 

See the careful work by Chee Ho Tham, ‘Notice of Assignment and Discharge 

by Performance’ [2010] LMCLQ 323 and ‘The Nature of Equitable Assignment 

and Anti-Assignment Clauses’ in: J. Neyers & R. Bronaugh (eds.), Exploring 

Contract Law (Hart, 2009), 283. 

90 



www.sellier.de


involve the transfer of a contractual right but rather the creation of a 

new right in the ‘assignee’, the subject matter of which is the ‘assignor’s’ 

contractual right as against the promisor. In this respect it differs from a 

statutory assignment permitted by section 136 of the Law of Property Act 

1925, which does involve a true transfer. 

The terms of trusts can be as many and varied as the wit of human imagination 

allows. There can be any number of beneficiaries, some as yet 

unborn, whose rights are subject to any number of conditions precedent or 

subsequent. The startling variety in the content of the right of a beneficiary 

under a trust does not offend the numerus clausus principle because 

a trust does not involve a right exigible against the whole world, but 

merely a right exigible against whoever is the holder of the right, which 

is the subject matter of the trust, from time to time. There is, therefore, 

no question of imposing undiscoverable duties of an idiosyncratic form 

upon all of rest of the world. 

Similarly, trusts of rights do not require any of the steps or forms for the 

transfer of rights to be satisfied. They do not require the satisfaction of 

any of these steps because they do not involve anything being conveyed 

or transferred at all. 

We find a similar division between the transfer of rights and the creation 

of new rights in relation to pre-existing rights in relation to secured 

transactions. To understand this it is necessary to explain the difference 

between a mortgage and a charge, a distinction which can be lost sight of 

because of their functional similarity. 

A mortgage involves the transfer, or conveyance, of the right which is 

being used for purposes of security. So, just as with an absolute transfer, 

to be effective if what is being conveyed is a right other than a right to 

goods, the relevant notification, registration or formal step must be taken. 

The equity of redemption is the right to have transferred back the right 

conveyed for purposes of security upon the performance of the obligation 

secured. 

A charge differs from a trust because of the different obligations a trustee 

and a chargor are under with respect to the underlying right (or, which 

is identical, because of the different rights the beneficiary or the chargee 

have with respect to the underlying right). If A is obliged to hold a right 

by way of security for B, this is a charge. A is charged (i.e. bound, under 


91 


www.sellier.de


a duty) to hold a particular right by way of security for B. A’s duty to B is 

defeasible upon the performance of the duty secured by the charge. It is 

important to differentiate the obligation to hold a right by way of security 

for another, and the quite separate obligation which this secures. A charge 

involves no equity of redemption because there is nothing to redeem, 

because nothing is transferred. Once the obligation secured is fulfilled, 

the obligation to hold particular rights as security is released. 

Just as an obligation to transfer land creates a trust, if A agrees to mortgage 

his right to his land to B, B acquires a security right immediately before 

transfer takes place because that is what A is obliged to do. 

It is impossible to contract out of the duty of a trustee or chargor to hold 

rights for another because without such a duty there is no trust or charge. 

The difference between a trust and a charge is the content of the duty in 

relation to the right held. 

Although it is impossible to transfer again something which has already 

been transferred away, it is possible to create separate duties to different 

persons in relation to the same right. So, A may declare a trust over his 

title to land in favour of B, C, D and E. Although each trust is equally 

valid, it is not possible for A to fulfil his duty to each of B, C, D and E. An 

issue of priorities then arises. We are more familiar with priority disputes 

of this kind where there are successive charges in relation to the same 

right in favour of successive creditors. By contrast, once I have given 

away my (right to my) car, I cannot validly give it away a second time. 

No question of priority properly so-called therefore arises where multiple 

purchasers argue that they are the party to whom the right to goods has 

been transferred. 

We cannot transfer what we do not yet have, but we can agree to transfer 

what we will acquire in the future. Once those future rights, whether 

to land, goods, shares, debts or anything else, are acquired they cannot 

be, and are not, automatically conveyed to the promisee. 21 Rather there 

arises an obligation to hold the rights acquired for the benefit of someone 

else: a trust. Similarly, whilst a transfer of rights by way of security (a true 

mortgage) is impossible where those rights have not yet been acquired, an 

agreement to hold such rights by way of security when acquired creates an 

obligation so to hold them. A charge therefore arises upon acquisition. 

21 

Holroyd v Marshall (1861-62) 10 HLC 191, 11 ER 99. 

92 



www.sellier.de


Raising our gaze to other legal systems there are two ways in which the 

common law position in relation to security is relatively unusual. In relation 

to non-possessory security within civilian legal systems it is the 

mortgage, not the charge, which is (relatively) unusual. So, although German 

law recognises the transfer of title for security purposes (Sicherungsübereignung) 

this is not the usual form non-possessory security takes. We 

can see therefore that other legal systems have also moved beyond the 

Roman law division between rights against persons (in personam) and 

rights against things (in rem) and have also embraced rights to, or over, 

other rights. 

However, other legal systems have not generally adopted this idea outside 

of the context of non-possessory security. 22 So, trusts of the kind we have 

in the common law world are generally unknown. Instead, the concept of 

a separate fund or patrimony has been employed. The definition in Book 

X of the Draft Common Frame of Reference is instructive because of how 

far it departs from the concept of a trust in the common law world. So, 

the definition commences, unobjectionably (Article 1:201): 

‘A trust is a legal relationship in which a trustee is obliged to administer 

or dispose of one or more assets (the trust fund) in accordance 

with the terms governing the relationship (trust terms) to benefit a 

beneficiary or advance public benefit purposes.’ 

But then continues (Article 1:202(1)): 

‘the trust fund is to be regarded as a patrimony distinct from the personal 

patrimony of the trustee and any other patrimonies vested in or 

managed by the trustee.’ 

Within the common law world, this is flat wrong. The subject matter of 

a separate fund or patrimony is not just rights but duties and liabilities 

as well. So the administrator of the estate of someone who has died does 

not owe the obligations of the estate, the estate does. On the DCFR view 

the trust, and not the trustee, may be the subject of a duty and may itself 

be sued. 23 In English law a trust does not involve a separate fund or patri- 

22 

See the very important work by G. Gretton, ‘Ownership and its Objects’, 

RabelsZ 71 (2007), 802. 

23 

This is expressly contemplated by the Brussels I Regulation (see fn. 2). 


93 


www.sellier.de


mony: trusts do not owe obligations, trustees do. 24 The DCFR definition 

represents, I would suggest, a common misconception amongst civilian 

lawyers attempting to translate the trust into terms with which they are 

familiar from their own legal system. 

So, a charge and a mortgage are alike because they are both security rights; 

just as both lions and crocodiles are alike because they are carnivores. 

A beneficiary’s interest under a trust and a fee simple estate in land are 

alike because they are absolute interests which are non-defeasible upon 

the performance of an obligation; just as rabbits and iguanas are alike 

because they are herbivores. The rights of a chargee and the beneficiary 

under a trust are alike because they relate to the rights of another; just 

as lions and rabbits are alike because they are mammals. 25 An equity of 

redemption is also a mammal. The rights of a mortgagee of land and the 

holder of a fee simple estate in land are alike because their rights relate to 

a physical thing and are exigible against the whole world (i.e. they are in 

rem in the Roman sense); just as crocodiles and iguanas are alike because 

they are both reptiles. 

How English law, indeed how any legal system, permits party autonomy in 

relation to the creation of property rights as between contracting parties 

can only be understood by employing the concept of a right in relation to 

another right (or, put identically, the duty to hold that right for another). 

Once we realise that this right is not effective as against third parties, but 

only as against the right holder from time to time, we can see that it is 

not operating in rem in the Roman sense. It is unsurprising that early legal 

systems, such as Roman law, did not develop in any coherent fashion what 

may be described as a type of supra-law, rights relating to other rights. Restrictions 

on the class of rights we recognise as operating against all others 

and restrictions on our ability to transfer rights simply have no application 

to this category. However it is even more important to realise that this 

category of right does not operate merely in personam. There must be a 

subject matter, another right, and the beneficiary or chargor is protected 

from the counter-parties insolvency just as much as is someone who has 

24 

L. Smith, ‘Trust and Patrimony’, Revue générale de droit 38 (2008), 379- 

403. 

25 

All charges and all trusts are equitable and not legal, but today it is their 

conceptual similarity which is interesting, not their historical origin in one 

corner or another of Westminster Hall. Only equity recognised right to other 

rights, transcending the old Roman law division. 

94 



www.sellier.de


a right in rem in the Roman sense. I have tried to show that. within the 

common law, this conceptual structure is not limited to the trust. Indeed 

within other legal systems which do not, unlike English law, permit the 

divisibility of the ownership of a thing (e.g. a lease of land) the concept 

of a right against another right is of even greater practical utility. 

4.5. Private international law 

We have many reasons to be grateful for our common Roman law inheritance, 

but the sharp and exhaustive division within private international 

law between rights in rem and rights in personam is not one of them. 

Whether someone has created a contractual obligation to another in relation 

to another right is, inevitably, determined by the law governing the 

contract. This statement of the completely obvious is contained in relation 

to claims by Article 14(1) of Rome I: 26 

‘The relationship between assignor and assignee under a voluntary assignment 

or contractual subrogation of a claim against another person 

(the debtor) shall be governed by the law that applies to the contract 

between the assignor and assignee under this Regulation.’ 

This provision is surplusage and all it achieves is confusion. Of course the 

relationship of the parties to a contract is governed by the law applicable 

to that contract: what on earth else would it be governed by? 

Whether I, or somebody else, is the person in whom a particular right is 

vested is determined by the law governing that right. Put identically, to 

whom a particular duty is owed is determined by the law governing that 

duty. This is true of all rights. So, once we have said that the law governing 

the right to land is the lex situs, how the right to land is transferred is 

inevitably governed by the lex situs. Once we have said that the law governing 

shares in a company is that of the place of incorporation, in whom 

that right is vested is determined by the law of the place of incorporation. 

Again, this statement of what is (or should be) completely obvious is 

stated in relation to claims by Article 14(2) of the Rome I Regulation: 

26 

Regulation (EC) No 593 / 2008 on the law applicable to contractual obligations 

(Rome I), OJ L 177 / 6. 


95 


www.sellier.de


‘The law governing the assigned or subrogated claim shall determine 

its assignability, the relationship between the assignee and the debtor, 

the conditions under which the assignment or subrogation can be 

invoked against the debtor and whether the debtor’s obligations have 

been discharged.’ 

Of course the duties the debtor is under, and to whom they are owed, are 

governed by the law governing the debt. How would it be sensible, or even 

meaningful, for this to be determined by anything else? 

If we were Romans we could stop there, and Article 14 would cover all 

that could be or needed to be said. Indeed we could delete Article 14(1) 

and (2) and the law in all Member States would inevitably still be the 

same. What however of a right to another right, which does not involve 

the transfer of the underlying right but rather the creation of a new right, 

and one which will bind transferees of its subject matter and persist upon 

bankruptcy? 

It should be clear that there should be no reason why the validity of a 

trust or charge or any right to another right should be determined by the 

law governing the underlying right. This is because the underlying right 

is unaffected, it has not been transferred in any way. Put identically, the 

correlative duties are not being altered by the declaration of a trust or the 

creation of a charge. In principle the law which determines the validity 

of the tie around the underlying right is the law applicable to the tie. Put 

another way, the law governing the trust or charge should determine not 

just its validity as against the trustee or chargor personally but its ‘proprietary’ 

validity as well. This means that party autonomy should apply, the 

parties may choose for themselves a law to govern their trust or charge and 

that will determine its validity. This deference to party autonomy is embodied 

in the Hague Trusts Convention so that where a trust is expressly 

created it is governed by the law expressly chosen or where there is no 

such choice, by the law with which it has its closest connection. 27 What 

27 

Convention of 1 July 1985 on the Law Applicable to Trusts and on their Recognition 

(Hague Trust Convention), Articles 6, 7 and 8, to be found on www. 

hcch.net / upload / conventions / txt30en.pdf. The Convention was incorporated 

into UK law by Recognition of Trusts Act 1987 (see www.legislation.gov. 

uk / ukpga / 1987 / 14). See also the exemplary decision of the European Court 

of Justice in Webb v Webb Case C-294 / 92 [1994] ECR I-1717; [1994] QB 696 

interpreting what is now Article 22 of the Brussels I regulation (see fn. 2). 

96 



www.sellier.de


law governs the underlying right is neither here nor there because a trust 

or charge has no impact upon the underlying right. 

In relation to charges precisely the same principles apply to that expressed 

in the Hague Trusts Convention. So, in Re Anchor Line (Henderson Brothers) 

Ltd, 28 A Ltd, an English company, owning land and movable property 

in Scotland, granted a floating charge to B, a Scottish bank, executing 

the agreement in Scotland. Under Scottish law at that time this form of 

security did not exist. A Ltd went into liquidation and funds representing 

the Scottish assets subsequently came into the liquidator’s hands. Those 

funds were payable to B in accordance with the floating charge. It made 

no difference that, under the law of the jurisdiction in which the land and 

movables were situated no right in relation to these assets were transferred 

to the chargee, as no charge of any kind involves the transfer of the underlying 

right but rather the creation of a new right. 

Today, a similar dispute would be governed by the European Insolvency 

Regulation. 29 If A Ltd, a company with its centre of main interests in 

England is wound up in England, having given B a charge over assets situated 

in other jurisdictions which do not recognize such a non-possessory 

security, the position remains precisely the same as at the time of Re Anchor 

Line. 30 The assets of the company situated in other Member States 

are to be remitted to England as the lex concursus. Under Article 5(1) of 

the Regulation the opening of the insolvency proceeding shall not affect 

the rights in rem of creditors or third parties situated within the territory 

of another Member State. However by giving effect to the priority of the 

floating charge in the English insolvency proceeding, the lex concursus is 

not in any sense cutting back or diminishing the proprietary rights to the 

assets of creditors or third parties. 

What if the issue of the validity of a trust or charge came before the courts 

of a state which did not recognise rights to other rights such as the trust, 

but instead within its domestic law treated the Roman categories of rights 

in personam and in rem as exhaustive? Private international law in the 

common law world inevitably developed rules governing rights to other 

28 

[1937] 1 Ch 483. See also Re Courtney, es p Pollard (1840) 4 Deac 27. 

29 

Regulation (EC) No 1346 / 2000 on insolvency proceedings, OJ L 160 / 1. 

30 

See also the article by P. Smart, Rights in Rem, Article 5 and the EC Insolvency 

Regulation: An English Perspective, International Insolvency Law 

Review 15 (2006), 17. 


97 


www.sellier.de


rights, such as the trust and the charge, because it had categories of such 

rights within its own domestic law. What if the question of the validity 

of a charge came before the courts of a jurisdiction which did not recognise 

these or analogous classes of rights? In all probability it will have no 

choice of law rules specifically governing such a class of rights unknown 

to its own law, and so they will create no more than personal obligations. 

A state, such as Italy, can incorporate choice of law rules for such rights 

even where they are not accurately reflected in its own domestic law, by 

signing up to the Hague Trusts Convention, but this has proven to be an 

unusual decision for states to take. 

What of the Rome I Regulation? Article 14(3) states: 

‘The concept of assignment in this Article includes outright transfers 

of claims, transfers of claims by way of security and pledges or other 

security rights over claims.’ 

This may be politely described as unhelpful. What is the law which determines 

the proprietary validity of a charge over a claim? We are told that 

Article 14 is intended to cover ‘security rights over claims’ 31 but what does 

this mean? Is the validity of such security determined by Article 14(1) (the 

law governing the relationship between chargor and chargee) as it should 

be or by Article 14(2) (the law governing the underlying unaffected claim 

right) which makes no sense? Article 14(3) is worse than useless as it fails 

to tell us which of 14(1) or 14(2) applies to which issue. 

The fear is that because the conceptual explanation as to why party autonomy 

in relation to the creation of proprietary rights over other rights 

is not only possible but completely unobjectionable has not been understood 

at European level, we will see a disastrous codification of the private 

international law rules in relation to property rights which denies such 

autonomy. In England where commercial law is dependent upon such 

autonomy, the result would be catastrophe. 

However, it is necessary to separate the issue of proprietary validity from 

that of priority. Because, as we have seen, multiple trusts or charges over a 

single asset may be equally valid, a priority competition as between them 

may arise. Say, A Ltd creates a first charge over its premises, its stock in 

trade, and its receivables to B Bank. It then creates a second charge over 

31 

Emphasis added. 

98 



www.sellier.de


the same assets in favour of C Bank. As the agreements between A Ltd 

and the two banks may not be governed by the same law, the issue of the 

priority between the two securities cannot be determined by the same law 

as that which determined each charge’s validity. 

One option would be to have the issue of priorities determined by the law 

applicable to each of the underlying rights secured. Although this is possible, 

it would be extremely inconvenient as the order of priority would 

vary according to the law applicable to the rights used by way of security 

with consequent expense in determining who was paid what. 

The traditional approach of private international (common) law has been 

to apply the lex fori which until recently was probably the best solution 

available, but had the unfortunate consequence that the order of priorities 

could vary according to where the hearing took place. Forum shopping 

and inconsistency of result is therefore a potential problem. 

Today a much more attractive option is to apply the debtor’s centre of main 

interests, or COMI, as defined by the Insolvency Regulation. Criticisms of 

the definition of the COMI can be made, but regardless of whether these 

are justifiable it makes a great deal of sense to have the priority of claims 

in relation to the assets of the debtor determined by the place where the 

main proceedings will take place in case of insolvency. 

Another issue which cannot be left to party choice, and which must be 

separated from it, is whether a third party who acquires a right for value 

in ignorance of the fact that it was held on trust or had been charged 

by way of security takes free of the trust or charge. Again, it would be 

unacceptable if what a third party was taken to know were determined 

by a law of which he had no knowledge or means of knowing. Whether 

the tie around the underlying right can bind such a third party should 

therefore be determined by the law governing the right he acquires. If 

therefore a Ruritanian right, whether to land, goods, payment of a debt, 

shares or anything else is held on trust or by way of security for a bank, 

whether a party acquiring such right from the trustee or chargor takes free 

of it should be determined by Ruritanian law. Party autonomy is vitally 

important, but it is not everything, it cannot govern questions of priority 

or overreaching. 


99 


www.sellier.de


4.6. Conclusion 

Those who favour party autonomy in relation to the creation of property 

rights and those who do not will continue to talk at cross purposes unless 

they define more clearly what they mean by property rights. 

Of course, in relation to the creation of rights exigible against all others or 

the transfer of rights, absolute party autonomy cannot be permitted. The 

law governing a right must determine whether it can be created and the 

conditions under which it can be transferred, if at all. However what is 

frequently overlooked is that some proprietary rights involve no transfer 

at all, but rather the creation of a new right to the underlying rights, a new 

right which does not operate in personam in the Roman sense. 

The conceptual distinction I have drawn underpins many of the modern 

financial market transactions we find today. Intermediated securities, 32 

non-possessory security, securitization, the ability to bring a claim to funds 

misdirected by fraud, and much else depends upon it. Only by recognising 

that none of us are any longer Romans can we understand how party 

autonomy in relation to one class of rights we describe as proprietary is 

possible. 

32 

See B. McFarlane and R. Stevens, ‘Interests in Securities: Practical Problems 

and Conceptual Solutions’, in: L. Gullifer and J. Payne, Intermediated Securities 

(2009). 

100 



www.sellier.de

B. 

Private International (Property) Law 


www.sellier.de


www.sellier.de

5. 

Party Autonomy in International 

Property Law: A German Perspective 

Jan von Hein * 


An essay on party autonomy in international property law from a German 

perspective could be very short. A German lawyer who is asked whether 

black-letter German private international law allows parties to choose 

the law applicable to rights in property would, without hesitation, answer 

‘No’. And this answer would be correct. Yet, on closer inspection, things 

are more complicated than they first appear. In recent decades, German 

private international law has witnessed three stages of codification: In 

1986, the original version of the German Introductory Act to the Civil 

Code (Einführungsgesetz zum Bürgerlichen Gesetzbuch, commonly abbreviated 

as EGBGB), which had been in force since 1900, 1 underwent a 

comprehensive reform. 2 However, this revision for the most part was concerned 

with eliminating violations of modern notions of gender equality 

in international family law and with incorporating into German law the 

Rome convention on the law applicable to contractual obligations of 

1980 3 – today replaced by the Rome I Regulation. 4 Although a draft dealing 

with the law applicable to non-contractual obligations and to rights 

in property had already been elaborated in the Federal Ministry of Justice 

in 1984, 5 those issues were dropped in the course of further legislation. 

* 

Professor of German Civil Law, Private International Law and Comparative 

Law, University of Trier (Germany). 

1 

Reichsgesetzblatt 1896, p. 604. 

2 

Bundesgesetzblatt (BGBl.) 1986 part I p. 1142. 

3 

Convention on the Law Applicable to Contractual Obligations of 19 June 

1980 (Consolidated Version), O.J. 1998 C 27 / 34. 

4 

Regulation (EC) No. 593 / 2008 of the European Parliament and of the Council 

of 17 June 2008 on the law applicable to contractual obligations (Rome I), 

O.J. 2008 L 177 / 6. 

5 

Reprinted as an annex in Kropholler, Internationales Privatrecht, 1 st ed. 

1990. 

103 


www.sellier.de

B. Private International (Property) Law 

It then took the German government nine years to rediscover its own 

well-worn draft 6 and finally initiate legislation in 1998, 7 which, despite 

the considerable period of time that had passed since the presentation of 

the 1984 draft, mainly put the rules proposed therein into German law. 8 

The motivation for this sudden leap into legislative action was rooted in 

fears that Germany would be at a strategic disadvantage in the upcoming 

negotiations on a European Rome II Regulation if it did not have 

a single rule on tort conflicts in its codified law. 9 Although no similar 

European initiatives were to be expected in the field of property conflicts, 

the slightly modified 1984 proposals on this issue made their way into a 

new Chapter 6 to the Introductory Code. After the Rome I and Rome II 

Regulations 10 had been passed, the Introductory Code was changed again: 

Conflicts rules on contracts were abolished, 11 while the rules go verning 

torts were kept on the book because of the Rome II Regulation’s restricted 

6 

A slightly revised draft was presented in 1993, reprinted as an annex in Kropholler, 

Internationales Privatrecht, 2 nd ed. 1994. 

7 

The 1998 draft is reprinted in Praxis des Internationalen Privat- und Verfahrensrechts 

(IPRax) 1998, 513 with an accompanying article by Rolf Wagner, 

Der Regierungsentwurf eines Gesetzes zum Internationalen Privatrecht für 

außervertragliche Schuldverhältnisse und für Sachen, ibid. 429. 

8 

Gesetz zum Internationalen Privatrecht für außervertragliche Schuldverhältnisse 

und für Sachen, BGBl. 1999 part I p. 1026; on the new provisions regarding 

international property law in particular, see Pfeiffer, Der Stand des Internationalen 

Sachenrechts nach seiner Kodifikation, IPRax 2000, 270; Stoll, 

Zur gesetzlichen Regelung des internationalen Sachenrechts in Art. 43-46 

EGBGB, IPRax 2000, 259; on the 1999 reform in general, see Junker, Die 

IPR-Reform von 1999: Auswirkungen auf die Unternehmenspraxis, Recht 

der Internationalen Wirtschaft (RIW) 2000, 241; Kreuzer, Die Vollendung 

der Kodifikation des deutschen Internationalen Privatrechts durch das Gesetz 

zum Internationalen Privatrecht der außervertraglichen Schuldverhältnisse 

und Sachen vom 21.5.1999, Rabels Zeitschrift für internationales und ausländisches 

Privatrecht (RabelsZ) 65 (2001) 383. 

9 

See Rolf Wagner, Ein neuer Anlauf zur Vereinheitlichung des IPR für außervertragliche 

Schuldverhältnisse auf EU-Ebene, Europäische Zeitschrift für Wirtschaftsrecht 

(EuZW) 1999, 709 (710). 

10 

Regulation (EC) No. 864 / 2007 of the European Parliament and of the Council 

of 11 July 2007 on the law applicable to non-contractual obligations (Rome 

II), O.J. 2007 L 199 / 40. 

11 

BGBl. 2009 part I p. 1574; on this reform, see Martiny, Neues deutsches internationales 

Vertragsrecht, RIW 2009, 737. 

104 



www.sellier.de

5. Party Autonomy in International Property Law: A German Perspective 

scope of application. 12 Apart from minor adjustments, Chapter 6 remained 

unchanged. In the discussion preceding the codification of Chapter 6 in 

1999, several voices in academia had advocated party auto nomy in property 

law. 13 Nevertheless, these demands failed to gain traction both with 

the German Federal Supreme Court and with the legislature. Instead, 

both in the case law as well as in the rules codified in Chapter 6, several 

workarounds were developed in order to accommodate the concerns that 

had given rise to claims for more party autonomy in property conflicts. 

Moreover, even the codified German private international law contains 

various points of entry for party autonomy in property issues. In the following 

sections of this paper, I will first describe the reasons for the prevailing 

conservative German attitude towards party autonomy in international 

property law (infra II.1). I will then analyse and critique the workarounds 

that German law has developed as substitutes for the choice of law denied 

to the parties themselves, as well as the points of entry for party autonomy 

already mentioned (infra II.2 and 3). Turning from the German to the 

European perspective, I will then examine which lessons the rules on 

party autonomy in the Rome I and II Regulations have to offer for a possible 

legislation at the EU level (infra III). The paper concludes with an 

evaluation of the current state of German private international law and 

a look at future European steps. 

12 

BGBl. 2008 part I p. 2401; on the preceding draft, see Rolf Wagner, Änderungsbedarf 

im autonomen deutschen internationalen Privatrecht aufgrund der 

Rom II-Verordnung? – Ein Überblick über den Regierungsentwurf eines Gesetzes 

zur Anpassung der Vorschriften des Internationalen Privatrechts an die 

Rom II Verordnung, IPRax 2008, 314. 

13 

See Drobnig, Eigentumsvorbehalte bei Importlieferungen nach Deutschland, 

RabelsZ 32 (1968) 450; id., Entwicklungstendenzen des deutschen internationalen 

Sachenrechts, in: Lüderitz / Jochen Schröder (eds.) Internationales 

Privatrecht und Rechtsvergleichung im Ausgang des 20. Jahrhunderts, Festschrift 

für Gerhard Kegel (1977) 141 (147); Rolf H. Weber, Parteiautonomie 

im Internationalen Sachenrecht, RabelsZ 44 (1980) 510; Einsele, Rechtswahlfreiheit 

im Internationalen Privatrecht, RabelsZ 60 (1996) 417 (435 et seq.); 

Stoll, Internationales Sachenrecht, in: J. von Staudingers Kommentar zum 

Bürgerlichen Gesetzbuch (1996), paras. 282-285; for a recent pleading for 

party autonomy in international property law, see Flessner, Rechtswahl im internationalen 

Sachenrecht – Neue Anstöße aus Europa, in: Festschrift Koziol, 

Wien (2010), 125-146 (in translated version: see Chapter 1 in this book on 

Party Autonomy in International Property Law, Choice of Law in International 

Property Law, New Encouragement from Europe). 


105 


www.sellier.de


5.2. The German codification 

5.2.1. Rejection of party autonomy 

a) Lex rei sitae as the general rule 

Article 43(1) EGBGB contains a clear-cut codification of the familiar 

principle of lex rei sitae (or, as it is frequently put in English legal Latin, the 

lex situs): ‘Rights in property are governed by the law of the state where 

it is situated’. This rule is silent on party autonomy – and deliberately so. 

Shortly before the codification of 1999, the Federal Supreme Court had 

explicitly rejected party autonomy in international property law in the 

‘Polish Cans’ case of 1996. 14 The facts of the case were as follows: Seller 

(S), a German company headquartered in Germany, had sold a number 

of cans containing food to a German Buyer (B), which was also a German 

company domiciled in the Federal Republic. The problem was that 

the cans were not stored on German soil at the time of sale, but were still 

located on the premises of the Polish manufacturer (M) who had canned 

the produce. Under German substantive law, a transfer of ownership basically 

presupposes that the parties agree on the ownership being transferred 

from seller to buyer, and that the seller actually transfers his or her possession 

of the goods to the buyer. For the sake of expedience, however, 

German law in some cases allows parties to substitute the actual transfer 

of possession. One substitute is the assignment of a claim for delivery of 

the goods that the seller has against a third party who actually possesses 

the goods in question. This is what S and B did in the case before the 

Supreme Court: S assigned to B his claim against M for a delivery of the 

cans. Pursuant to German substantive law, ownership would have passed 

from S to B under such circumstances. The tricky question, however, 

was whether German substantive law actually governed the transfer of 

ownership from S to B. Even before the codification of 1999, lex rei sitae 

was accepted by German courts as a rule of customary law. Since the cans 

where stored in Poland at the time that ownership should have passed 

from S to B, Polish law governed in this case as the lex situs. The Court of 

Appeal, however, assumed that Polish law may not allow for a transfer of 

ownership by way of assigning a claim for a delivery of the goods against a 

third person. For reasons of German procedural law, the Court of Appeal’s 

14 

Bundesgerichtshof (BGH) 25.9.1996, Neue Juristische Wochenschrift (NJW) 

1997, 461 = Die deutsche Rechtsprechung auf dem Gebiete des Internationalen 

Privatrechts (IPRspr.) 1996 No. 56. 

106 



www.sellier.de


findings on the content of a foreign law could not be challenged before 

the Federal Supreme Court. Nevertheless, the Court of Appeal did not 

stop at this point. Instead, it argued that S and B had chosen German law 

as the law governing the transfer of ownership between them. Given that 

both parties were German companies, the sale contract was governed 

by German law, and so on. The Federal Supreme Court did not accept 

this derogation from lex rei sitae. The judges rejected party autonomy in 

property law by invoking the following arguments. First of all, the general 

interest in legal certainty is best served by applying the lex rei sitae. After 

all, the transfer of ownership also has implications for third parties, who 

need to ascertain easily which law governs rights in property. Second, the 

Federal Supreme Court argued that accepting choice of law regarding the 

transfer of ownership would violate the principle of a numerus clausus of 

property rights. Who knows what kind of rights in property alien to German 

substantive law the parties might invent? Third, undermining legal 

certainty and stimulating legal creativity in property law would endanger 

the protection of creditors. Finally, the Federal Supreme Court pointed 

out that lex rei sitae was not only a cherished rule of German customary 

law but also an almost universally recognised conflicts principle in other 

countries as well. A strict adherence to the lex situs would thus foster 

one of the principal goals of classic private international law as developed 

by Friedrich Carl von Savigny: namely, the international harmony of 

decisions. 15 Accordingly, the Federal Supreme Court remanded the case, 

leaving the task of determining and applying Polish property law to the 

lower court. The German Government referred to this decision and its 

ratio decidendi in its official explanation for the Act of 1999 (Article 43[1] 

EGBGB). 16 The Court replicated this favour by later referring to the new 

Act as ‘a codification of our settled case law’. 17 

b) Formal validity 

German private international law is not only strict in applying lex rei sitae 

with regard to the substantive requirements to the transfer of ownership; it 

also extends this attitude to the formal validity of a legal transaction crea- 

15 

On the international harmony of decisions as one of the principal goals of the 

conflict of laws, see Friedrich Carl von Savigny, A Treatise on the Conflict of 

Laws (William Guthrie trans., 1880) 69 et seq. 

16 

Bundestags-Drucksache 14 / 343 (1999), p. 16. 

17 

BGH 29.5.2000, IPRspr. 2000 No. 43 (p. 93). 


107 


www.sellier.de


ting or transferring a right in property. Article 11(4) EGBGB provides: ‘A 

legal transaction creating or transferring a right in rem is formally valid 

only if it observes the formal requirements of the law that is applicable to 

the legal relationship forming the subject matter of the legal act‘. This rule 

is an exception to the favor negotii that is generally accepted in German 

private international law. Article 11(1) EGBGB, which mirrors Article 

11(1) of the Rome I Regulation, provides that, generally speaking, ‘a legal 

transaction is formally valid if it satisfies the formal requirements of the 

law which governs it in substance or of the law of the country where it 

is concluded’. A strict adherence to the lex rei sitae with regard to formal 

validity makes sense for real property, but hardly for movable property. 18 In 

the latter group of cases, the reasons that motivate a favor negotii in contract 

law – expediency in international transactions and the protection of 

legitimate expectations of the parties – support a more liberal attitude as 

well. The Federal Government had, in the course of the comprehensive 

reform of private international law in 1986, already announced its intention 

to evaluate and possibly revise this provision in the future. 19 Since 

then, however, two occasions to revisit the question have already been 

missed. The legislature left the rule untouched in the course of the reform 

of 1999, and merely renumbered it in 2009 when the Introductory Code 

was adapted to the Rome I Regulation. 

5.2.2. Workarounds and points of entry for party autonomy 

a) Renvoi 

Nevertheless, German private international law is not completely set on 

denying party autonomy in every case. Within the classic paradigm on 

which German conflicts law is based, the international harmony of decisions 

is of paramount importance, and I have already shown that the 

Federal Supreme Court invoked this argument in order to justify its strict 

adherence to the lex situs (supra II.1.a). Yet the argument of international 

harmony cuts both ways, given that other countries are more liberal in 

their approach to party autonomy in international property law. Gene rally 

speaking, German law is still firmly supportive of renvoi in private international 

law: that is, German rules of conflict usually do not refer directly to 

the substantive laws of another state, but first to its private international 

18 

Kropholler, Internationales Privatrecht, 6 th ed. 2006, p. 316. 

19 

Bundestags-Drucksache 10 / 504 (1983), p. 49. 

108 



www.sellier.de


law. Thus, pursuant to Article 4(1) EGBGB, lex rei sitae (Article 43[1] 

EGBGB) points not to the substantive but to the private international 

law of the situs. If the conflicts rules in force at the situs allow for party 

autonomy, German law accepts this approach. 20 Even before the codification 

of 1999, the Federal Supreme Court had decided accordingly in a case 

involving rights flowing from shared ownership of a Spanish apartment. 21 

We have here an important point of entry for party autonomy into German 

international property law. 

b) A qualified change in the applicable law 

Another important workaround that serves as a functional equivalent to 

party autonomy in some cases is found in Article 43(3) EGBGB. This 

provision reads as follows: ‘If an interest in property brought into this state 

has not yet been perfected, events that occurred in another state are to be 

considered as local events for the determination of the perfection of such 

an interest in this state’. In academic parlance, this phenomenon is known 

as a qualifizierter Statutenwechsel: a qualified change in the applicable law. 22 

The background of this rule is an old chestnut from university conflicts 

courses: the widely discussed ‘Knitting Machine’ case decided by the Federal 

Supreme Court in 1966. 23 The facts of this case were as follows: 

Seller (S) in Italy had sold a knitting machine to Buyer (B) in Germany. 

The parties had agreed on an oral retention of title clause, which was only 

valid inter partes under Italian substantive law. After the machine had 

been delivered in Germany, the question arose as to whether creditors 

of B could seize the machine as B’s property. Under German substantive 

law, an oral retention of title clause was and still is sufficient to trigger an 

erga omnes effect. Not surprisingly, the Federal Supreme Court started its 

legal analysis by professing a strict adherence to the lex situs. This means 

that German law applied only after the machine had entered Germany. 

This would have been too late for S, because the oral retention of title 

clause was concluded when the knitting machine was still on Italian soil. 

Nevertheless, the Federal Supreme Court found an ingenious way to help 

20 

Kropholler (supra note 18), p. 556; Kieninger, in: Ludwig (ed.), juris Praxiskommentar 

BGB, Vol. 6, Internationales Privatrecht (2009) Art. 43 EGBGB 

para. 43. 

21 

BGH 25.9.1997, NJW 1998, 321 = IPRspr. 1997 No. 60. 

22 

Cf. Kropholler (supra note 18), p. 562. 

23 

BGH 2.2.1966, BGHZ 45, 95 = NJW 1966, 879 = IPRspr. 1966-67 No. 54. 


109 


www.sellier.de


S out of his predicament: The court argued that the title clause contained 

a tacit substantive agreement between the parties that full property should 

be transferred back from B to S after B had acquired possession of the 

goods in order to securitise the seller’s claim for payment. Many academics 

criticised the Court for this tongue-in-cheek approach: Denying party 

autonomy in international property law while at the same time inventing 

a tacit agreement under substantive law in order to compensate for 

this deficit was not perceived as methodologically candid and proper. 24 

It would be preferable, the critics argued, simply to honour the parties’ 

tacit choice of German law as the law of the country of destination. 25 The 

compromise between the court’s adherence to lex rei sitae and academic 

calls for party autonomy is found in Article 43(3) EGBGB: An unfinished 

transfer (or retention) of ownership (under Italian law) is ‘perfected’ when 

the knitting machine crosses the (German) border. The retention of title 

clause which was only valid inter partes in Italy is transmogrified into an 

erga omnes rule. Thus, in this specific group of cases, there is no longer 

any practical need for the parties to agree on the law of the country of 

destination as applicable to the retention of title clause, because Article 

43(3) EGBGB ensures the validation of the clause. Formally, this is only 

a unilateral conflicts rule, as it refers simply to ‘an interest in property 

brought into this state [i.e. Germany]’. This approach was deliberately 

chosen because the drafters were reluctant to meddle with the substantive 

property law of other nations. 26 In academia, however, a multilateralisation 

is favoured. 27 Why should the Italian seller be treated differently if 

he had exported the knitting machine not into Germany but into Utopia, 

which also recognises oral retention of title clauses? 

c) Cross-border pollution 

Another point of entry for party autonomy in international property law 

is found in Article 44 EGBGB, which concerns cross-border polluting 

24 

See the references supra note 13. 

25 

See the references supra note 13. 

26 

Bundestags-Drucksache 14 / 343, p. 16; see also Kreuzer (supra note 18), 

p. 450. 

27 

Kegel / Schurig, Internationales Privatrecht, 9 th ed. 2004, p. 773; Wendehorst, 

in: Rixecker / Säcker (eds.), Münchener Kommentar zum Bürgerlichen Gesetzbuch, 

5 th ed. 2009, Art. 43 EGBGB para. 173; this is rejected, however, by 

JurisPK / Kieninger (supra note 20) Art. 43 EGBGB para. 36. 

110 



www.sellier.de


emissions emanating from real property. The rule provides: ‘As to claims 

arising from the adverse effects of emissions from real property, the provisions 

of Regulation (EC) No. 864 / 2007 [Rome II] except for chapter III 

shall apply by way of analogy’. The doctrinal background behind this rule 

is that, under German substantive law, injunctive and similar protective 

claims under neighbour law are divided into tort-related damages claims 

and property-related claims for revindication. 28 In the conflict of laws, 

however, splitting up a neighbour’s remedies between the lex situs and 

the lex loci delicti would lead to practical complications and possibly even 

to conflicting judgments, denying under tort law what is allowed under 

property law, and vice versa. In contrast, a harmonised connection of tort 

law claims and rights in rem is justified by the fact that, at the level of 

substantive law, tort-related and property-related rules are elements of 

a finely balanced, closely interwoven regulatory structure for delimiting 

the respective scope of freedom and action of the polluter and the injured 

party, which must not be divided by a two-pronged approach. This 

was already recognised by the German legislature in 1999. The original 

version of Article 44 EGBGB, however, merely referred to the objective 

conflicts rule for complex torts (Article 40(1) EGBGB) and excluded 

party autonomy (Article 42 EGBGB). After the autonomous conflicts 

rules for torts were superseded by the Rome II Regulation, in particular 

Article 7 Rome II, which relates to environmental damage, 29 the provision 

was adapted to the Regulation. In this respect, it must be noted that 

the reference to Rome II introduced in 2008 is not restricted to Article 7 

of the Rome II Regulation, but includes Article 14 Rome II as well, thus 

allowing for party autonomy even insofar as the pertinent claims are characterised 

as belonging to property law. The practical importance of this 

shift is slightly diminished by the fact that under an autonomous characterisation 

of nuisance claims in international law, such claims are arguably 

within Rome II’s scope of application anyway. The Court of Justice of the 

EU has decided with regard to the Brussels I Regulation that such claims 

must be characterised as delictual within the meaning of Article 5(3) 

28 

For closer analysis, see von Hein / Wolf, Transboundary Environmental Damage 

in the Conflict of Laws, in: Wolfrum / Langenfeld / Minnerop (eds.), Environmental 

Liability in International Law: Towards a Coherent Conception 

(2005) 381 (424 et seq.). 

29 

For closer analysis of this provision, see Bogdan, The Treatment of Environmental 

Damage in Regulation Rome II, in: Ahern / Binchy (eds.), The Rome 

II Regulation on the Law Applicable to Non-Contractual Obligations (2009) 

219. 


111 


www.sellier.de


Brussels I, and not as claims sounding in property law within the ambit 

of the exclusive jurisdiction under Article 22(1) Brussels I. 30 If the Court 

draws the line between property law and tort law accordingly under Rome 

II, the conflicts rules of this Regulation will be directly applicable and the 

transmission effected by Article 44 EGBGB will no longer be needed. 

From a doctrinal point of view, however, one should not underestimate 

the fact that the German legislature has for the first time recognised the 

possibility of party autonomy for claims that are traditionally characterised 

as belonging to property law. 

d) Security rights in means of transportation 

For all practical purposes, a more important point of entry for party autonomy 

into international property law is found in Article 45(2) sentence 

1 EGBGB, which deals with security rights in means of transportation. 

The provision states: ‘The attachment of security interests as a matter of 

law in the above means of transportation [i.e. aircraft, vessels, rail vehicles] 

is governed by the law applicable to the claim that is to be secured’. 

Due to this technique involving an accessory connection, a valid choice 

of law clause for the secured claim (Article 3 Rome I) determines the attachment 

of security interests as well. 31 However, there is an important 

restriction: Pursuant to Article 45(2) sentence 2 EGBGB, lex rei sitae still 

determines the priority as being among several security interests. Thus, 

even the ‘indirect’ party autonomy recognised in Article 45(2) cannot 

lead to adverse effects of the choice of law clause upon third parties. 

e) The escape clause 

Finally, it is controversial as to whether the escape clause found in Article 

46 EGBGB may be used in order to recognise indirectly a choice of 

law clause. Article 46 EGBGB provides: ‘If there is a substantially closer 

connection to the law of a state other than the law that would be appli- 

30 

ECJ 18.5.2006-343 / 04 Land Oberösterreich v. ČEZ as, [2006] E.C.R. I-4557 = 

IPRax 2006, 591, 564 note Thole = RIW 2006, 624, 627 note Knöfel. 

31 


112 



www.sellier.de


cable under Articles 43 and 45, the law of that state shall be applied’. 32 

The provision deliberately omits Article 44 EGBGB on environmental 

damage. The reason for this lacuna is that the Rome II Regulation does 

not allow for recourse to the general escape clause (Article 4[3] Rome 

II) as far as claims for environmental damage (Article 7 Rome II) are 

concerned. 33 Thus, applying the German escape clause to property-related 

claims would disturb the harmonised connection of delictual and 

property-related claims that the legislature has tried to ensure by Article 

44 EGBGB. 

Although the law governing the contract for the sale of goods between 

the parties is a factor that may be taken into account when determining 

a substantially closer connection under Article 46 EGBGB, a mere 

divergence between the law chosen by the parties for their contractual 

obligation and the lex rei sitae is not enough to trigger an application of 

the escape clause. 34 The more difficult question is whether a technically 

invalid direct choice of the applicable property law may serve as a factor 

in determining a substantially closer connection. Some authors have 

advocated such an approach in order to ensure flexible solutions. 35 Most 

writers, however, have rejected this proposal because it would amount to 

a circumvention of the drafters’ clear intention to refuse party autonomy 

in international property law. 36 A mediating approach consists in allo wing 

party autonomy by way of the escape clause, but limiting it to an inter 

partes effect. 37 The Federal Supreme Court has not yet decided on this 

issue, but it would be quite surprising if the judges deviated from their firm 

stance against party autonomy in international property law. 

32 

For an in-depth analysis of Art. 46 EGBGB, see Mansel, Normzweck und 

Tatbestandsstruktur des Art. 46 EGBGB, in: Lorenz et al. (eds.), Festschrift 

für Andreas Heldrich (2005) 899. 

33 

Cf. Bogdan (supra note 29), p. 226 et seq. 

34 


35 

Mansel (supra note 32), p. 905 et seq.; Stoll (supra note 8), p. 264. 

36 

Junker (supra note 8), p. 252; Pfeiffer (supra note 8), p. 274; jurisPK / Kieninger 

(supra note 20) Art. 46 EGBGB para. 6; Thorn, in: Palandt, BGB, 69 th ed. 

2010, Art. 46 EGBGB para. 3 (except in very rare cases). 

37 

This was the intention of the drafters of the 1999 act, see Bundestags-Drucksache 

14 / 343, p. 19; this position is shared by Kropholler (supra note 18), 

p. 567. 


113 


www.sellier.de


5.2.3. Doctrinal inconsistencies 

a) The general interest in legal certainty 

Taking a critical look at the arguments that the Federal Supreme Court 

invoked in the ‘Polish Cans’ case, one finds that a closer inspection 

leads to a more differentiated evaluation. The general interest in legal 

certainty upon which the court sets so much store is already considerably 

undermined by the techniques of renvoi and the escape clause. 

If third parties must be protected from the risks inherent in not being 

able to foresee the law chosen by the parties, it is not self-evident that 

this argument should count for nothing if a foreign law allows for party 

autonomy, or if a court decides after a comprehensive evaluation of the 

specific case at hand that lex situs should be discarded because there is a 

substantially closer connection to the law that the parties have chosen 

for their sales contract. Moreover, the existence of an escape clause is 

actually supportive of allowing parties to choose the applicable law to 

rights in property. In contract and tort conflicts, the parties’ right to 

choose the applicable law (Article 3 Rome I; Article 14 Rome II) is an 

important counterweight to the discretion granted to courts (Articles 

4[3] Rome I and II). Parties may legitimately want to be on the safe side 

by excluding any leeway for a court to second-guess them on the matter 

of determining the applicable law. This interplay between allowing for 

a court’s discretion and the parties’ need for legal certainty is ignored in 

the current German system. 

b) The numerus clausus argument 

The familiar numerus clausus argument is also less impressive than it appears 

at first sight. First of all, it is already considerably restricted by Article 

43(2) EGBGB. This is a kind of particular provision on public policy. 

The rule provides: ‘If the situs of property changes, rights established in 

it cannot be asserted in contradiction to the law of the new situs state’. 

The mere fact that the law of the new situs state does not acknowledge a 

right in property established under the former lex situs does not amount 

to a contradiction to the law of the new situs state. In the past, German 

courts have readily recognised registered securities in movables without 

possession of the item to which the security interest is attached (e.g. a 

lien in a car under the U.S. Uniform Commercial Code [UCC] or under 

114 



www.sellier.de


Italian law). 38 The courts have achieved this result by having recourse to a 

technique called Transposition: 39 The foreign right in property is translated 

into a functional equivalent under German substantive law. Although 

German substantive law demands that a lien in a car may only be established 

if the possession of the vehicle is transferred from one person to 

another, court practice and doctrine have for a long time accepted the 

workaround of what is termed Sicherungseigentum: The owner of a car may 

transfer his property to the party giving him a loan, but the actual transfer 

of possession is substituted by an agreement that allows the former owner 

to keep the car. When the loan is repaid, the creditor transfers ownership 

back to the former debtor. From an economic point of view, the result 

is very similar to a non-possessory lien under American or Italian law. 

Thus, German courts have denied a ‘contradiction’ within the meaning 

of Article 43(2) EGBGB. Therefore, German international property law 

already contains a significant element of flexibility that could also be 

used to accommodate concerns relating to the numerus clausus of property 

rights being endangered by party autonomy. 

c) Protection of creditors 

The protection of creditors is prima facie a more valid argument for restricting 

party autonomy in international property law. Nevertheless, Article 

45(2) EGBGB may serve as a blueprint for a more differentiating 

approach: Why should the parties be refused an option to choose the 

applicable law if it is guaranteed that such a choice will not adversely 

affect the priority as among several security interests? 

d) International harmony of decisions because of a 

universal recognition of lex rei sitae 

Finally, one has to reconsider the argument that an international harmony 

of decisions is fostered by denying party autonomy because of a universal 

recognition of lex rei sitae. This is certainly true for real property, but it 

38 

See on a lien under the UCC Oberlandesgericht (OLG) Karlsruhe 6.7.2000, 

Wertpapier-Mitteilungen (WM) 2003, 584 = IPRspr. 2001 No. 52A; on a 

lien under Italian law, BGH 11.3.1991, NJW 1991, 1415 = IPRspr. 1991 

No. 71. 

39 

On Transposition, see Kropholler (supra note 18), p. 561. 


115 


www.sellier.de


is increasingly doubtful with regard to movables in the light of recent 

codifications: for instance, in the Netherlands 40 or Switzerland. 41 If a more 

liberal attitude towards party autonomy in international property law becomes 

an international trend, at least in Western Europe, the international 

harmony of decisions may be better served by Germany’s jumping 

on the bandwagon. 

5.3. Lessons from Rome I and II 

5.3.1. Third-party effects 

Although there does not exist any European legislative project on international 

property law, the question is legitimate as to which lessons one 

might draw from the existing EU Regulations on the conflict of laws. 42 

The first lesson is that party autonomy must not have an adverse effect 

on third parties. This basic principle is firmly enshrined in Article 3(2) 

sentence 2 Rome I, which provides that rights of third parties must not 

be affected by a change in the applicable law. Likewise, Article 14(1) 

sentence 2 Rome II ensures that rights of third parties (e.g. insurers) are 

not affected by a choice of law clause between tortfeasor and victim. This 

general principle supports an argumentum a fortiori for erga omnes rights: 

namely, party autonomy in international property law will – in principle – 

be restricted to having only an inter partes effect. 

40 

Article 3(2) Wet conflictenrecht Goederenrecht, reprinted in a German 

translation in IPRax 2008, 560 with an introductory article by Struycken / 

Sujecki, Das niederländische Gesetz zur Regelung des internationalen Sachenrechts, 

IPRax 2008, 558. 

41 

Article 104 of the Swiss Bundesgesetz über das Internationale Privatrecht, 

reprinted in Riering (ed.), IPR-Gesetze in Europa (1997) No. 8. 

42 

On party autonomy in the Rome I and II Regulations, see Rühl, Rechtswahlfreiheit 

im europäischen Kollisionsrecht, in: Baetge / von Hein / von Hinden 

(eds.), Die richtige Ordnung – Festchrift für Jan Kropholler (2008) 187; von 

Hein, Of Older Siblings and Distant Cousins: The Contribution of the Rome 

II Regulation to the Communitarization of Private International Law, RabelsZ 

73 (2009) 461 (486-492). 

116 



www.sellier.de


5.3.2. Limitation of eligible laws 

Under both the Rome I and the Rome II Regulation, it is accepted as a 

general principle that no substantial connection between the chosen law 

and the object of choice (contract, tort) is required (Article 3 Rome I; 

Article 14 Rome II). Important exceptions exist, however, for contracts 

for the carriage of persons (Article 5[2] Rome I) and insurance contracts 

covering mass risks (Article 7(3) Rome I). Thus, a limitation of eligible 

legal orders in international property law (e.g. country of destination, 

country of origin) would not be incompatible with a systemic approach. 

5.3.3. Explicit or tacit choice of law 

Both the Rome I and the Rome II Regulation allow for a tacit choice 

of law clause (Article 3 Rome I; Article 14 Rome II). However, legal 

certainty may be a more important consideration in property law than 

in contract / tort law. Thus, it would be legitimate to insist on an explicit 

clause or to define strictly the terms under which a tacit choice of law 

may be honoured. 

5.3.4. Protection of weaker parties 

Under both the Rome I and the Rome II Regulation, it is accepted as 

a general principle that weaker parties (consumers, employees) deserve 

specific protection (Articles 6, 8 Rome I; Article 14[1][a] Rome II). The 

instruments employed to achieve this goal differ considerably, however. 

In the field of consumer and employee protection, the Rome I Regulation 

follows a more favourable law approach. With regard to protecting persons 

who are parties to a contract of transportation and as far as the insurance 

of mass risks is concerned, the Rome I Regulation opts for a limitation of 

eligible laws (Article 5[2] subparagraph 2 Rome I; Article 7(3) Rome I). 

The Rome II Regulation tries to protect weaker parties by differentiating 

the accepted time of choice: ex ante for B2B agreements and ex post for 

B2C clauses (Article 14[1] Rome II). No matter which specific technique 

is adopted, international property law will also have to accommodate the 

need to protect weaker parties. 


117 


www.sellier.de


5.3.5. Exclusion of renvoi 

As a general principle, renvoi should be excluded in European instruments 

(see Article 24 Rome II; in principle, Article 20 Rome I as well). In international 

property law, a possible compromise could consist in allowing for 

an extension of party autonomy under domestic private international law 

(cf. Article 7[3] sentence 2 Rome I for insurance contracts). However, a 

uniform solution would be preferable. 

5.3.6. Public policy and mandatory rules 

Finally, public policy and a special attachment of internationally mandatory 

rules are respected both in Rome I and II (see Article 21 Rome I 

and Article 26 Rome II on public policy; Article 9 Rome I and Article 

16 Rome II on mandatory rules). In international property law, a public 

policy clause will be needed in property law to safeguard essential elements 

of the numerus clausus. In this regard, Article 43(2) EGBGB may 

serve as an inspiration for drafting a clause that ensures a necessary degree 

of flexibility in accommodating foreign property rights. Moreover, 

international property law will have to take into account the existence 

of internationally mandatory rules. In particular, certain formal requirements 

(registration) and principles of publicity may be characterised as 

internationally mandatory. Internally mandatory rules should be applied 

to intra-state and intra-EU cases following the model of Article 3(3) and 

(4) Rome I and Article 14(2) and (3) Rome II. 


The German codification rejects party autonomy as a specific conflicts rule, 

but contains various workarounds and points of entry allowing for flexible 

solutions. This technique leads to doctrinal inconsistencies, however, 

which emphasise the need for a more candid and sophisticated approach. 

In particular, flexibility in international property law by way of an escape 

clause is welcome, but needs to be balanced by allowing parties to choose 

the applicable law. The Rome I and II Regulations contain various general 

principles on party autonomy in European private international law, which 

deserve to be taken into account in drafting conflicts rules on property rights 

as well. In particular, adverse effects on third parties must be prevented. 

118 



www.sellier.de

6. 

Party Autonomy in French and Belgian Law 

The Interconnection between Substantive Property Law 

and Private International Law 

Vincent Sagaert * 


The issue of the scope of party autonomy within the field of property law is 

one of the most challenging questions in property law. As is often the case 

in life, the most challenging issues are also the most complicated ones. 

This contribution is thus limited in its ambitions. It does not aim to be the 

final word as to whether party autonomy is ‘a general principle of modern 

property law’, as is proclaimed by some scholars: 1 this would require the 

scope of a PhD thesis. 2 Nor will we deal with the legal-economic arguments 

that are at stake. 3 

We will stick to certain developments of the subject matter in French 

and Belgian law, both in substantive law and in private international law 

(hereafter: PIL). It is hoped that these developments might lead to further 

research in the field of the interconnection between substantive law and 

private international law in these legal systems. 

* 

Professor of Private Law, University of Leuven; Professor of Property Law, 

University of Antwerp, Barrister at the Brussels Bar. 

1 

T.W. Merrill and H.E. Smith, The Property / Contract Interface, Yale L.J. 2000, 

69. 

2 

See recently B. Akkermans, The principle of numerus clausus in European 

private law, Antwerp, Intersentia, 2008, 657 p; T.H.D. Struycken, De numerus 

clausus in het goederenrecht, Deventer, Kluwer, 2007, 870 p. 

3 

B. Depoorter, Fragmentation of Property: the law and economics of the anticommons, 

Dissertation University of Ghent, Gent, 2003, 136. 

119 


www.sellier.de


6.2. The closed system of property rights in 

French and Belgian law 

6.2.1. Historical introduction: prevailing view 

It is generally thought that French private law provides for the foundations 

of the closed system of property law (numerus clausus) throughout 

Europe. According to general opinion, it would not be possible to create 

property rights that have not been recognised as such by the legislator. 

This principle would, again according to general opinion, be the result of 

a revolutionary reaction against the feudal system at the end of the 18 th 

century. This feudal system was founded in the mixing of obligatory duties 

and property rights. The entitlement to a layer of a ‘property right’ could 

entail numerous personal and positive duties in relation to the feudal 

Lord. The interconnection between personal obligations and property 

rights was the legal basis for a social model that the early 19 th century 

legislator aimed to abolish. In acting against this structure, property law 

would have to be defined in a stricter fashion, and delimited in a clearer 

manner, in order to take away the legal foundations of the feudal system. 

The strict limitation of the number of property rights would therefore be 

the legal counterpart of party autonomy in the field of obligations. 

This view is seen to be hazardous when one looks into the French Civil 

Code itself. This not only emerges from the absence of express recognition 

of the numerus clausus but also from the system of transfer of movable 

property rights, which has been adopted by the French legislator. 

In the field of transfer, obligatory effects and proprietary effects have been 

combined in Article 1138 C.C., which provides that the transferee of 

property rights becomes owner of the transferred goods from the moment 

of delivery, and that moment is situated at the moment of the agreement 

between parties. In other words, the obligatory aspects (rights and obligations) 

and the proprietary effects of the sales agreement are concurring 

and are triggered by the same event (e.g. the consensus). 4 Thus, the 

4 

Article 1138 provides in its original French version: “L‘obligation de livrer 

la chose est parfaite par le seul consentement des parties contractantes. Elle 

rend le créancier propriétaire et met la chose à ses risques dès l‘instant où elle 

a dû être livrée, encore que la tradition n‘en ait point été faite, à moins que le 

débiteur ne soit en demeure de la livrer; auquel cas la chose reste aux risques 

120 



www.sellier.de

6. Party Autonomy in French and Belgian Law 

splitting of the obligatory and proprietary aspects is not reflected in the 

transfer system. 

In the following paragraphs, we shall see in the first instance that the numerus 

clausus principle had been subject to a major development throughout 

the 19 th and 20 th centuries. We will analyze the current status of the 

principle, and determine the relation between this substantive principle 

and the conflict of law rules. 

6.2.2. Legal foundations of the numerus clausus principle 

As we have stated, it appears that the closed system of property rights 

is grounded in a reaction against the feudal system. However, a closer 

look at the statutory provisions of the French Civil Code – which was 

closely inherited from the Revolution – questions the validity of the 

aforementioned opinion. The principle of numerus clausus is not found 

in the Civil Code or in any other statutory provision. Article 543 C.C. 

provides that 

‘one may have a right of ownership, or a mere right of enjoyment, or 

only land services to be claimed on property’. 

However, this provision does not enumerate the property rights in an exhaustive 

manner: first of all, superficies and emphyteusis are not mentioned, 

as these rights were only re-introduced into Belgian law in the Acts of 10 

January 1824. 5 Moreover, real security rights (pledges, mortgages, liens) 

and co-ownership have been omitted in the enumeration. Meanwhile, 

other property rights, such as timesharing, have been recognised by legal 

practice without any amendment to Article 543 C.C. 

The only general foundation of the numerus clausus in the Civil Code 

would be a declaration of Treilhard, who was one of the drafters of the 

Civil Code during preparatory works pertaining to it: 

de ce dernier.” See on the scope and application of this rule V. Sagaert, ‘Consensus 

versus delivery systems. Consensus about tradition?’, in: W. Faber and 

B. Lurger (eds.), Rules for the transfer of movables. A candidate for European 

harmonization or national reforms?, Munich, Sellier, 2008, 9-46. 

5 

Belgian Official Gazette, 1824, nr. 14. 


121 


www.sellier.de


‘(..) l’on ne peut avoir sur les biens que trois espèces de droits: ou un 

droit de propriété, ou une simple jouissance, ou seulement des services 

fonciers. Ainsi notre Code abolit jusqu’au moindre vestige de ce 

domaine de supériorité jadis connu sous les noms de seigneur féodal 

et censuel.’ 6 

However, there existed specific statutory provisions underlying the numerus 

clausus principle in specific cases. For instance, the principle was 

given statutory basis with regard to property security rights. Article 2285 

(old Article 2093) of the Civil Code provides that – here we use the official 

translation 7 

‘the property of a debtor is the common pledge of his creditors; and the 

proceeds of it shall be distributed among them pro rata, unless there 

are legal causes of priority between the creditors.’ 

This article provides that a right of priority can be granted only on legal 

grounds, recognised as such by the legislator. An identical provision applies 

in Belgian law. 8 

6.2.3. Development of the numerus clausus principle 

in the 19 th and 20 th centuries 

The viewpoint that the numerus clausus principle is a reaction against 

the feudal system also becomes questionable when one looks into the 

case law of the early 19 th century. Already in 1834, the French Cour de 

cassation rendered the famous Caquelard judgment, in which it denied any 

restriction on property rights that split up immovable ownership. The case 

dealt with an agreement of co-ownership with regard to bushland, which 

6 

Locre, Législation civile, VI, 32. Free translation: ‘One can have on objects 

only three categories of rights: ownership, right of use, or land burdens. In this 

way, our Code abolishes the foundations of this domain of superiority that was 

previously known as feudal lords’. 

7 

www.legifrance.gouv.fr. The French text of the provision is as follows: ‘Les 

biens du débiteur sont le gage commun de ses créanciers; et le prix s’en distribue 

entre eux par contribution, à moins qu’il n’y ait entre les créanciers des 

causes légitimes de préférence.’ 

8 

Article 8 of the Belgian Mortgage Act 16 December 1851, Belgian Official 

Gazette 22 December 1851. 

122 



www.sellier.de


granted to one co-owner the right of use of the bush, while the other was 

granted the much more valuable rights to the fruits of the trees. The first 

contracting party (Caquelard) claimed part of the fruits also, arguing that 

the powers of parties in co-ownership were mandatorily fixed. However, 

this claim was found to be unjustified. The Court ruled that 

‘en droit, les art. 544, 546 et 552 Cod. civ., sont déclaratifs du droit 

commun relativement à la nature et aux effets de la propriété, mais 

ne sont pas prohibitifs: ni ces articles, ni aucune autre loi, n’excluent 

les diverses modifications et décompositions dont le droit ordinaire de 

propriété est susceptible.’ 9 

This judgment appears to accept an open system of property rights into 

French law. 

This judgment was not a single step. Until the end of the 19 th century, the 

French Cour de cassation had adopted a flexible viewpoint with regard to 

the closed system of property rights, recognising property rights although 

they are difficult to reconcile with the numerus clausus principle. 10 

The numerus clausus principle has been subject to increasing pressure during 

the last decades. The system of property rights seems to open up 

again: For instance, in 1984 the Cour de cassation ruled that a perpetual 

right to fix a placard to another’s wall had proprietary effects. As it was 

impossible to classify this right as a servitude in this specific case, it is hard 

to determine under which legal property right the current right could be 

brought. 11 This view is supported by leading legal scholars, who criticise 

this numerus clausus principle, especially in French law, because it is based 

9 

Cass. fr. 13 februari 1834, Sirey 1834, I, 205. Freely translated, the citation 

states: ‘At law, Articles 544, 546 and 552 C.C. are confirming the general rule 

with regard to the nature and effects of ownership, but are not prohibitive: neither 

these provisions nor another statutory rule preclude diverse modifications 

and fragmentations of the general right of ownership’. See also L. d’Avout, Sur 

les solutions du conflits des lois en droit des biens, Paris, Economica, p. 666, 

n°. 506. 

10 

Cass. fr. 12 December 1899, Sirey 1901, I, 497; Cass. 25 October 1886, Sirey 

1887, I, 373; Cass. 5 November 1866, Dalloz 1867, I, 32; Cass. fr. 20 February 

1851, Dalloz 1851, I, 54. 

11 

Cass. fr. 18 January 1984, Bull. civ., III, No. 16, Dalloz 1985, jur., 504, note F. 

Zenati, J.C.P. 1986, II, n°. 20547, note J.F. Barbieri. 


123 


www.sellier.de


upon historical reasons that are no longer valid at the beginning of the 

21 st century. 12 

It has emerged from the aforementioned developments that the origins 

and foundations of the numerus clausus principle in French law is uncertain 

and hazardous. Most French legal handbooks pay formal lip service 

to the closed nature of property law. 13 However, developments in French 

property law have resulted in case law that seems to abandon a strict application 


More recently, this ambiguous position has been driven to further limits. 

The Association Henri Capitant des Amis de la Culture Juridique Française, 

which also initiated the major reforms of the law of succession and the 

law of security rights, presented a draft text for the reform of property 

law. 14 The proposed Article 516 Civil Code provides that ‘the provisions 

of the new book on « property » are part of the public order, except if they 

provide otherwise’. This means that all contractual deviations from the 

statutory provisions as regards property rights would be absolutely void. 

It has been rightfully observed that such a provision is a turning back 

towards the revolutionary ideas that were abolished two centuries ago, 

and have never been applied in legal practice. 15 The drafters themselves 

were aware of the curtailing of party autonomy, and therefore created an 

open-ended property right, the ‘droit de jouissance spéciale’. It is a property 

right that grants one or more specific powers to use another’s property 

12 

P. Crocq, Propriété et garantie, Paris, L.G.D.J., 1995, p. 201, No. 248-249; M. 

Planiol and G. Ripert, Traité pratique de droit français, Paris, 1964, III, p. 54, 

No. 48; A. Weill, F. Terre and P. Simler, Droit civil. Les biens, Paris, Dalloz, 

1985, p. 47, No. 41. 

13 

C. Beudant, Cours de droit civil français, Paris, Rousseau, 1948, IV, p. 63, 

No. 70; L. Josserand, Cours de droit civil positif français, Paris, Sirey, 1932, I, 

No. 1337; C. Witz, La fiducie en droit privé français, Paris, Economica, 1981, 

p. 76, No. 84; F. Zenati and T. Revet, Les biens, Paris, PUF, 2008, p. 458, n°. 

296 : ‘Les droits personnels peuvent etre créés par engagement volontaire sans 

restriction quant à la nature de la prestation mise à la charge du débiteur. Cette 

liberté est discutée pour les droits réels, car on répugne à admettre que de tels 

droits puissent etre créés en dehors des formes recensées et réglementées par la 

loi. Un droit réel non autorisé par les textes permet à ceux qui le constituent 

de déroger dans une mesure non acceptable au régime des biens.’ 

14 

This text can be found at www.henricapitant.org. 

15 

V. Sagaert, “Aux armes citoyens”, NTBR 2009, 363-364. 

124 



www.sellier.de


(draft Article 608). Its content can be modulated contractually by the 

contracting parties as long as they comply with regard to the destination 

of the property and the rules of public order. 16 In other words, the front 

door of the property house is hermetically sealed, but the back door is 

open and ready to host foreigners. 

In Belgian law, the Supreme Court has, according to the prevailing view, 

confirmed the existence of the numerus clausus principle in the Blieck 

judgment of 16 September 1966, in which the obligation to lay out and 

maintain a park on the part of a parcel of land in favour of the municipality 

was not to be considered as a servitude. The second requirement for 

the valid creation of a servitude – that it is vested in favour of a dominant 

tenement – was not fulfilled. 17 Even if the weight of this judgment as legal 

ground for the closed system is criticisable, 18 the Belgian Supreme Court 

gives a clear opinion on the structure of property law. It seems that the 

Court aims to make clear that the freedom of contract would be opposite 

to the compulsory nature of the statutes with regard to property law. 

6.3. The numerus clausus principle is expressed 

in two major matters 

Despite the weak legal and historical foundations of the principle, the 

general starting point is that freedom of contract of the parties is seriously 

curtailed by the most fundamental characteristic of property law: namely, 

the numerus clausus principle. 19 To repeat the words of Bernard Rudden in 

16 

This seems to be based on circular reasoning: As the property law is part of 

the public order, one cannot create a property law of which the content can 

be freely modulated with the exception for matters of public order. 

17 

Cass. 16 September 1966, Arresten van het Hof van Cassatie 1967, 67, Journal 

des Tribunaux 1967, 59, Revue critique de jurisprudence belge 1968, 166, note 

J. Hansenne. See also H. de Page, Traité élémentaire de droit civil belge, Brussels, 

Bruylant, 1964, I, n° 130; W. van Gerven, Algemeen Deel, in: Beginselen 

Belgisch Privaatrecht, Brussels, Story-Scientia, 1987, n° 34. 

18 

In this judgment, the question of whether new property rights could be created 

was an obiter dictum; thus, the result of the judgment did not depend upon 

it. 

19 

For an extensive analysis of the scope of this principle in Belgian law and comparative 

perspective, see V. Sagaert, Het goederenrecht als open systeem van 


125 


www.sellier.de


his article ‘Things as things and things as wealth’: “fancies” are for contract, 

not for property.’ 20 Unregulated personal rights are allowed as long 

as they are not contrary to public order and bona mores, but unregulated 

property rights are degraded to personal rights. However, this does not 

mean that party autonomy is excluded in the field of property law. Parties 

are free to determine the content of property rights, as long as they respect 

the basic features of these rights. 

With regard to some property rights, the contractual freedom is extensive 

(long lease, building rights). However, this freedom is restricted within 

two major fields of law, in which French-Belgian property law is quite 

restrictive and which belong to the internal public order. It concerns 

(A) the strict requirements for servitudes 21 and (B) the exclusive nature 

of ownership. We will analyze both these issues in the following paragraphs. 

6.3.1. Strict requirements for servitudes 

As regards servitudes, French and Belgian law have remained unchanged 

since the introduction of the Civil Code in 1804. Art. 637 of the French 

and Belgian Civil Code provides a mandatory definition with strict requirements: 

a servitude is ‘a charge imposed on a parcel for the use and 

utility of another parcel belonging to another owner.’ 22 A servitude merely 

allows limited use of the servient tenement, and has to be described expressly 

when the servitude is created. They are different from the general 

rights to use a piece of land, as they are conferred to the holder of a long 

verbintenissen? Poging tot een nieuwe kwalificatie van de vermogensrechten, 

3 Tijdschrift voor Privaatrecht, 2005, 983-1086. 

20 

B. Rudden, Economic Theory v. Property Law: The Numerus Clausus Problem, 

in: J. Eeklaar and J. Bell (eds.), Oxford Essays in Jurisprudence, Third Series, 

Oxford, Clarendon, 1987, 243. 

21 

The word ‘servitudes’ is used here because of the civil law nature of the legal 

systems that are dealt with. The words ‘land burdens’ or ‘easements’ seem to 

be much broader with regard to their scope. 

22 

This translation and all other English translations from the Civil Code that 

are used hereunder are based on the translation on the French governmental 

website: http: // www.legifrance.gouv.fr. Given the inadequacy of some wordings, 

there have been some adaptations. 

126 



www.sellier.de


lease or building right or – to a certain extent – to the usufructuary of a 

parcel of land. 

Three requirements can be set out as essential features of servitudes: (1) 

it is a burden on land (2) in favour of land, and (3) the servient tenement 

and the dominant tenement have a different owner. From a historical 

perspective, it is logical that the legislator was reluctant in relation to 

servitudes, as they were the modern translation of feudal burdens. These 

requirements have gradually been opened up and applied in a more flexible 

manner. 23 Nevertheless, even taking into account the gradually more 

extensive application of these requirements, they restrict the scope of 

servitudes within strict limits. We will not go into further detail with 

regard to these devices, as they are related to real estate and thus do not 

raise any issue of private international law. 

6.3.2. Exclusivity of ownership 

Other subject matter in which the sensitivity and strictness of the closed 

system of property rights is witnessed can be found with regard to this 

ownership feature, and more precisely in the feature that ownership must 

be exclusive. This feature has been at stake in recent debate with regard 

to the French ‘fiducie’ device. The definition of ownership in French and 

Belgian private law is to be found in Article 544 Civil Code, which provides 

that 

‘ownership is the right to enjoy and dispose of things in the most comprehensive 

manner, provided they are not used in a way prohibited by 

statutes or regulations.’ 

As historical research has demonstrated, the wording of ‘in the most comprehensive 

manner’ in Article 544 of the French Civil Code did not refer 

to the absoluteness of ownership. The real meaning of the words was to 

be found at the level of the exclusive nature of ownership. 24 

23 

See on this point V. Sagaert, The fragmented system of land burdens in French 

and Belgian law, in: S. van Erp and B. Akkermans (eds.), Towards a unified 

system of land burdens, Antwerpen, Intersentia, 2006, 31-52. 

24 

See for example R. Derine, De grenzen van het eigendomsrecht in de 19 e eeuw, 

Antwerpen, De Sikkel, 1955. 


127 


www.sellier.de


This feature is not to be found in the internal relation between the owner 

and the object of ownership, but must be looked for in the relation between 

the owner and third parties. It means that civil ownership is – 

traditionally – a unitary ownership. Only one person can have the ‘most 

comprehensive’ (residuary) property rights on a good at a given time. 25 

This is also called the ‘exclusive’ nature of ownership. In other words, 

ownership is the right of its holder to exercise powers with regard to a 

good to the exclusion of others. The powers of a full owner cannot be in 

competition with the powers of another. 26 

The importance of the exclusive nature of ownership can be justified 

and explained in a historical perspective. 27 Before codification, the 

feudal system was grounded in a layered concept of ownership, which 

made the distinction between dominium utile and dominium directum. 

The exclusivity of ownership innovated by creating a non-layered 

owner ship right, which contradicts the legal instrument building a 

feudal society. In doing so, the legislator aimed to distance itself from 

the concept of feudal ownership. As stated, this feature is implicitly 

provided in the French Civil Code through the words ‘in the most 

comprehensive way’, but is more explicitly part of the definitions provided 

by later Codes, such as the German Civil Code (§ 903 BGB) 

and the Dutch Civil Code (Art. 5:1 Dutch Civil Code). Hence, the 

exclusivity of ownership was, at the beginning of the 19 th century, a 

key notion of ownership. Therefore, it is not surprising that all legal 

systems have struggled with legal institutions trying to challenge the 

unitary nature of ownership. 

Contextual property law research has demonstrated that one of the major 

developments in European property law is that the exclusivity of owner- 

25 

H. de Page, Traité élémentaire de droit civil belge, Brussels, Bruylant, 1974, 

V, n°. 893A; J. Hansenne, Les biens, Liège, 1996, I, 584, nr. 632; F. van Neste, 

Eigendom morgen, Tijdschrift voor Privaatrecht 1983, 484, nr. 5. 

26 

Co-ownership is not an exception to the exclusive nature of ownership, as 

the powers of a co-owner are not general powers but are limited to specific 

powers subject to specific requirements, in such a way that the rights of the 

other co-owners are respected. 

27 

As has been explained supra, this historical explanation is questionable for 

the numerus clausus principle in general, but is accountable for some aspects 

of it. 

128 



www.sellier.de


ship has been weakening over the decades. 28 This is principally due to 

European developments that have created layers in ownership protection. 

A fiduciary ownership has been created and recognised in several 

‘civil law’ systems. The French fiducie was an important device, and will 

be dealt with separately (cf. infra). However, the notion of exclusivity 

of ownership has also been at stake at a European level. The Insolvency 

Regulation created a certain exception to this exclusivity with regard to 

title retention. The proprietary protection of the seller in the case of insolvency 

of the buyer is recognised in accordance with European national 

legal systems. However, the position of the buyer is strongly debated if the 

seller becomes insolvent before payment of the full purchase price. In the 

case of a cross-border title retention (where the goods have been transferred 

from one Member State to another), Article 7 par. 2 Insolvency 

Regulation provides for a clear and simple rule: The European legislator 

has directed the national legal systems to introduce a title reservation 

with regard to financial collateral arrangement: 

‘The opening of insolvency proceedings against the seller of an asset, 

after delivery of the asset, shall not constitute grounds for rescinding 

or terminating the sale and shall not prevent the purchaser from acquiring 

title where at the time of the opening of proceedings the asset 

sold is situated within the territory of a Member State other than the 

State of the opening of proceedings.’ 

In other words, the seller under title retention remains owner until the 

moment that the purchaser has paid the full purchase price. However, 

the buyer can claim a proprietary right on the good if the seller becomes 

subject to an insolvency proceeding prior to full payment. If the purchaser 

continues to make payments, he shall acquire title at the end of 

the period set out in the contract. 29 Hence, the buyer also has a virtual 

ownership of the good pendente conditione. The proprietary powers relating 

to the goods are, at different levels, divided between two persons 

(seller and buyer). 

28 

For a more elaborate overview, see V. Sagaert, Nieuwe perspectieven op het 

eigendomsrecht na tweehonderd jaar Burgerlijk Wetboek, in: P. Lecocq, B. 

Tilleman and A. Verbeke, Zakenrecht, Bruges, Die Keure, 2005, 43-85. 

29 

M. Virgos and M. Schmit, Report on the Convention of Insolvency Proceedings, 

3 May 1996, 6500 / 96, n°. 114. 


129 


www.sellier.de


French law also introduced its proper trust-like devices. This development 

began with the Loi-Dailly in 1981, which acknowledged a fiduciary transfer 

for security purposes (‘fiducie-sûreté’) in the financial sector. This Act 

of 2 January 1981 (Loi Dailly), which is now laid down in Article L 313-23 

of the Monetary and Financial Code, provides that a credit institution can 

acquire the fiduciary ownership of professional claims that the debtor is 

entitled to. This transfer is effective in relation to the insolvency of the 

assignor, even if the assigned claim becomes due only after the insolvency 

of the assignor. 30 

A more general trust-like device was introduced in France by an Act of 

2007. After nearly twenty years of attempts being made, the fiducie was 

finally voted by the French legislator, and with most of the features of a 

trust-like device. After 16 years of parliamentary debate – the first project 

dated from 1991 – it was approved by the French legislator on February 

19 th , 2007. 31 As well as the private law provisions, this Act contains a large 

chapter with tax provisions, illustrating the reluctance of the civil law 

relating to the tax effects of the trust. 32 Contrary to the Draft Common 

Frame of Reference, 33 French law defines – in the definition of the fiducie 

being the essentially proprietary nature of the device – the fiducie as 

30 

Cour de cassation, 7 December 2004, Dalloz 2005, jur., 230, note C. Larroumet. 

31 

Act on the Fiducie of 19 February 2007, JORF 21 February 2007 (own emphasis), 

to be consulted at www.legifrance.gouv.fr. This Act has been incorporated 

in Article 2011 et seq. of the French Civil Code. On this Act, see e.g. F. 

Barriere, La fiducie. Commentaire de la loi n° 2007-211 du 19 février 2007, 

Bulletin Joly Sociétés 2007, 440-455; A. Delfosse and J.L. Pelinguel, Premières 

vues sur la fiducie, Defrénois 2007, 581-600; C. Larroumet, La loi du 19 février 

2007 sur la fiducie. Propos critiques, Dalloz 2007, 1350-1353; P. Marini, Enfin 

la fiducie à la française, Dalloz 2007, 1347-1349; C. Witz, La fiducie française 

face aux expériences étrangères et à la convention de La Haye relative au trust, 

Dalloz 2007, 1359; et seq. 

32 

See on these tax provisions P. Fumenier, J. Miller and S. Dorin, Le régime fiscal 

de la fiducie, Dalloz 2007, 1354-1358. 

33 

Article 1:201 DCFR: ‘a legal relationship in which a trustee is obliged to administer 

or dispose of one or more assets (the trust fund) in accordance with 

the terms governing the relationship (trust terms) to benefit a beneficiary or 

advance public benefit purposes’. 

130 



www.sellier.de


‘the operation according to which one or more settlers transfer current 

or future goods, rights or securities, or a universality of current 

or future goods rights or securities, to one or more trustees who act 

with a specified purpose to the benefit of the beneficiary and keep them 

separate from their personal estate.’ 

Moreover, it is expressly stated that the insolvency proceedings opened 

against the ‘fiduciaire’ do not affect the fiduciary property (Art. 2024 C.C.) 

and that personal creditors of the ‘fiduciaire’ cannot seize fiduciary property 

(Art. 2025 C.C.). 

The debate on the proprietary positions of parties came to a head in 

2009, when the French legislator added a paragraph to this provision. 

An amendment of September 17 th , 2009 introduced a second paragraph 

in Article 2011 C.C., stating that the ‘trustee’ (‘fiduciaire’) exercises the 

fiduciary ownership of the goods that are part of the trust estate in favour 

of the beneficiaries, according to the terms of the fiducie. The underlying 

idea behind this reform was not connected to property relations 

but to economic motives, which are to be situated in the recognition 

of foreign legal devices. In reforming Article 2011 C.C., the French 

legislator aimed to facilitate the entrance of Islamic financial institution 

into the French financial market. The Islamic Sukuk requires that the 

person who participates has a property right to the underlying property 

and participates in the risk involved in the use of it. The English financial 

market had already accommodated the Sukuk by agreeing to grant 

equitable ownership to the participators; however, the French legislator 

would not stand behind: he aimed to realise the same goal through this 

amendment. The financial interest of the innovation could not be overestimated. 

Islamic finance transactions amount to 840 billion dollars a 

year and, more importantly, a yearly increase of 15 %. In opening their 

financial market, the French legislator aimed at recovering part of this 

market share. 

However, this provision has not survived the constitutional control exercised 

by the French Constitutional Court. This court ruled in a decision 

of 14 October 2009, in less than a page, that the provision did not have 

a connection, not even indirectly, with the proposal of which they were 

a part. The amendment was included in the Act, aiming at favoring market 

access to small and medium enterprises and at the functioning of the 

financial markets. 


131 


www.sellier.de


Although it had nothing to do with the proprietary implications of the 

amendment, the decision of the French Constitutional Court was highly 

applauded by the most authoritative French scholars. Crocq and Aynès 

called the aforementioned amendment the gate to a restoration of feudalism. 

34 Indeed, as the fiduciaire would not have the full right of ownership, 

but only a fiduciary right of ownership, this would mean that another 

layer of the ownership right would vest on behalf of the beneficiary, 

which would create a splitting up of ownership. Under the 2007 Act, legal 

scholars have indeed justified the obligation of segregation on behalf of 

the trustee by making reference to the absence of property protection on 

behalf of the beneficiary. However, the protection of the fiduciary property 

against the private creditors of the fiduciaire demonstrates that the 

beneficiary is entitled to a proprietary protection. Is the efficiency (‘opposability’) 

of a right in the case of insolvency not the real litmus test for 

the question as to whether somebody has property rights? 

From a theoretical point of view, there are three possibilities to justify this 

proprietary protection: (1) one accepts that this protection is based on a 

layer of a property right – which would endanger a major dogma of exclusivity 

in French property law; (2) one considers this proprietary protection 

as sui generis, which would put at stake the numerus clausus principle; (3) 

one considers these fiduciary assets as part of a separate estate (‘patrimony’), 

in which case the unity and indivisibility of the estate in French 

private law would be at stake. It is clear that the 2009 amendment opted 

for the first of these alternatives. 

6.4. PIL as an instrument for the opening up 

of the legal system? 

As we have seen, there is some debate about the general structure of 

French and Belgian property law. Most scholars assume the existence of 

a numerus clausus principle, which, however, does not exclude party autonomy 

on a large scale in property law. Nevertheless, in some areas of 

property law this party autonomy is scrutinised for historical reasons. In 

the last part of this contribution, we will deal with the issue concerning 

34 

L. Aynes and P. Crocq, La fiducie préservée des audaces du législateur, Dalloz 

2009, (2559) n° 6. See also P. Dupichot, Fiducie et finance islamique, Dalloz 

2010, 1064 et seq. 

132 



www.sellier.de


the interaction with PIL. It could be imagined that private parties use 

PIL to avoid the closed system of property rights. The question is whether 

they can choose a foreign legal system, which has a more open approach 

towards new property rights. 

6.4.1. General starting point: the lex rei sitae 

The general PIL rule in French and Belgian PIL is the same as in other 

legal systems: the law of the place where the goods are situated determines 

the law that applies to these goods. This emerges from Article 3 par. 2 of 

the Civil Code, providing that ‘immovables are governed by French law 

even when owned by foreigners’. 35 In the second half of the 20 th century, 

this rule was extended to movables. 36 

In legal scholarship, this rule is considered to be the logical effect of the 

numerus clausus principle: Since the national legal system acknowledges 

only domestic property rights, only these rights will apply to domestic 

goods. As parties cannot create unregulated property rights, they cannot, 

on the basis of their party autonomy, appeal to foreign legal systems in 

order to introduce them into French or Belgian law. 37 

This rule was recently confirmed by the French Cour de cassation in its 

judgment of 10 February 2010. In this case, an American woman, a widow, 

vindicated a painting that was situated on French territory, and French 

law had to be applied. This also entailed that the woman was confronted 

35 

Oddly enough, the official French translation of this text is: ‘Immovables are 

governed by French law even when owned by aliens’. 

36 

Chambre des Req., 19 March 1872, Recueil Dalloz 1874, I, 475. 

37 

J. Erauw, Internationaal privaatrecht, in: Beginselen van Belgisch privaatrecht, 

Mechelen, Kluwer, 2010, 634, n°. 550. See however: L. d’Avout, o.c., 676, n°. 

513. This author argues that the relationship between the numerus clausus and 

the lex rei sitae is contradicted by the analysis that German law, which would 

stick to the closed system, easily recognises foreign security rights, while the 

French system, which would be open in substantive law, is reluctant. However, 

it can be opposed that German law is – with regard to security rights – more 

flexible than French law. The recognition beyond statutory provisions of the 

Sicherheitsubereignung or Sicherheitsabtretung and the extensive application of 

title retention – including prolongation and extension clauses – demonstrate 

that German substantive law is at least as flexible as French substantive law. 


133 


www.sellier.de


with a rule of possessory protection leading to ownership (‘possession vaut 

titre’ 38 ), which was unknown to her in her own national system. 39 A more 

intricate case was that in which a Court of Appeal was confronted with 

the following case: A famous painting by Pieter Breughel de Oude was 

stolen from the Dresdner Gemäldergalerie in 1945 by Russian soldiers. The 

subsequent fate of the painting remains unclear. Presumably, it was sold by 

a Russian soldier to an American citizen, and afterwards the latter made a 

gift of the painting to his son. The latter brought the painting to Belgium 

and aimed to sell it on the Brussels art market. The police discovered 

the operation, and the painting was seized. The Dresdner Gemäldergalerie 

claimed ownership of the painting, but the question arose as to which law 

should be applied. The Antwerp Court of Appeal, 40 confirmed by the Supreme 

Court (Hof van Cassatie / Cour de cassation), ruled that Belgian law 

applied, as the painting was situated in Belgium. 41 Hence, the vindication 

action was admitted. This judgment is remarkable, as the result obtained is 

radically contrary to a judgment of the Dutch Supreme Court in a quasiidentical 

case, which was decided under the application of Dutch law. 42 

Apparently, an owner must simply be lucky enough to find his property 

38 

This is the previous and famous Article 2279 C. civ., which in French law 

is actually Article 2276 C. civ.: ‘En fait de meubles, la possession vaut titre. 

Néanmoins, celui qui a perdu ou auquel il a été volé une chose peut la revendiquer 

pendant trois ans à compter du jour de la perte ou du vol, contre 

celui dans les mains duquel il la trouve; sauf à celui-ci son recours contre celui 

duquel il la tient.’ 

39 

Cass. fr. 10 February 2010, Receuil Dalloz 2010, 443, note I. Gallmeister. We 

refer for further details on the case to the contribution of our colleague Jeroen 

van der Weide in this book (Chapter 2). 

40 

Court of Appeal Antwerp 3 December 2007, Rechtskundig Weekblad 2008- 

09, 961, Tijdschrift@ipr.be 2008 (found at www.ipr.be), afl. 3, 56. This judgment 

was confirmed by the Belgian Supreme Court in a judgment of 19 June 

2009. See on this judgment J. Erauw, Vier keren een ander situsrecht voor een 

nogal mobiel landschap van Breughel, Tijdschrift@ipr.be 2008, afl. 3, 75-78 

(to be consulted on http: // www.ipr.be); V. Sagaert, Over de lotgevallen van 

een gestolen Breughel in het IPR goederenrecht, Rechtskundig Weekblad 

2008-09, 965-968. 

41 

Substantially, the Court ruled that the current possessor could not be protected 

under Belgian law, as the possession was hidden and thus defective. 

42 

For a comprehensive overview and comparison of both judgments, see J.A. 

van der Weide, Over de grenzen van het recht. De merkwaardige geschiedenis 

van twee schilderijen uit de Gouden Eeuw, in: Ex Libris Hans Nieuwenhuis: 

134 



www.sellier.de


in an accommodating country, or employ the strategy of waiting until his 

property turns up in such country. 

In Belgian law, PIL was codified in the PIL Code of 16 July 2004, which 

became effective in October 2004. 43 Article 87 § 1 PIL-Code provides that 

the rights in rem in respect of an asset are governed by the law of the State 

on the territory of which the asset is situated. The court only makes exceptions 

in the event that a third party had acquired rights on the property 

pursuant to another legal system, but this could not be demonstrated in 

the aforementioned case. 

6.4.2. Mitigations regarding the general starting point: 

party autonomy in conflict rules 

As we have seen, French – more than Belgian – substantive property law 

has already opened up, even if the reluctance of the early-19 th -century legislator 

is still detected in some minds (cf. the Reform Project, supra). The 

opening of the substantive law towards new property rights has been used 

as an argument in order to recognise foreign and domestically unknown 

property rights. For instance, d’Avout argues in his impressive doctoral 

thesis that the flexibility of French property law enables recognition of 

foreign property devices, as long as appropriate measures of publicity are 

provided. 44 This creates a diagnostic relationship between substantive law 

and PIL. 

On the one hand, the receptivity towards foreign legal systems can result 

in an amendment of substantive law. In this manner, the reception and 

recognition of foreign property devices can be facilitated. As an example 

of the amendment facilitating the introduction of the Sukuk in French 

law has also demonstrated (cf. supra), substantive law has been changed 

at times in order to facilitate cross-border relations. 

However, the inverse interaction between PIL and substantive law is more 

regular. It has happened on some occasions in the past that PIL has been 

opstellen aangeboden aan prof. mr. J.H. Nieuwenhuis, Deventer, 2009, Kluwer, 

281 et seq. 

43 

Belgian Official Gazette 27 July 2004. A non-official English translation of 

this Code can be found at www.internationalprivatelaw.be. 

44 

In support of this view, see L. d’Avout, o.c., 666, n°. 506. 


135 


www.sellier.de


amended in order to provide for a remedy to the deficiencies in legal practice 

in French or Belgian law. It is, to a certain extent, an inappropriate 

use of PIL, as PIL is not meant to fill the gaps left by substantive property 

law. We will give two examples below, one with regard to transfer of 

ownership for security purposes and the other with regard to recognition 

of the trust. 

6.4.2.1. Assignment of property for security purposes 

The possible interaction between substantive property law and private 

international law is visible in the field of the fiduciary transfer of ownership 

for security purposes. The developments of commercial transactions, 

which result in gradually increasing cross-border transactions, 

have necessitated the introduction of a more modern security instrument. 

45 This has led the legislator to introduce the fiduciary transfer in 

the framework of banking operations. This was the Loi Dailly, which 

is dealt with supra. Hence, there was no need in French private law 

to accommodate PIL rules with regard to the transfer of ownership for 

security purposes. 

PIL was subject to an analogous development as substantive law. In a 

judgment of 24 May 1933, the French Supreme Court refused to give effect 

to a transfer of goods for security purposes by a French transferor to a 

German transferee. 46 The transferor was dispossessed of the goods and the 

agreement was subject to German law. Nevertheless, the Court ruled that 

this agreement was contrary to French law – as the pactum commissorium is 

prohibited in French law – and thus refused to give effect to the fiduciary 

transfer. Later case law also argued on the basis of the contravention of 

this contractual structure with the requirement of dispossession. However, 

this judgment was criticised by legal scholars. It was observed that PIL was 

not adapted to the developments in substantive law, which acknowledged 

more flexible security interests. L. d’Avout writes significantly that ‘the 

jurisprudential policy consisting in the rejection of foreign non-possessory 

security interests gradually terminates.’ (We will not expand on the desir- 

45 

M. Cabrillac and C. Mouly, Droit des sûretés (Paris: Litec, 2002), n° 530. 

46 

Req. 24 May 1933, Dalloz hebomadaire 1933, 378 and Recueil Sirey 1935, I, 

257, note H. Battifol. This judgment was confirmed in 1969: Cass. fr. 8 July 

1969, Juris-classeurs périodiques 1970, n°. 18.169, note H. Battifol. 

136 



www.sellier.de


ability of this development. 47 ) Therefore, an argument is made by French 

doctrine in favor of the recognition of German Sicherungsübereignung or 

the extended title retentions (verlängerte Eigentumsvorbehalt or erweiterte 

Eigentumsvorbehalt), even if the object is situated in France. 

Belgian law has not followed this development, and does not acknowledge 

a general recognition of a fiduciary transfer for security purposes. 

According to the current state of Belgian law, such a transfer for security 

purposes would – as a general starting point – be ineffective because it 

lacks a third party effect. 48 Belgian law traditionally does not recognise 

the proprietary effects of a fiduciary transfer for security purposes. 49 Even if 

certain recent developments have diminished the strength of this starting 

point, 50 it still remains de lege lata part of Belgian security law. This position 

is criticised by legal scholars, especially when the transfer involves 

claims, as it could disadvantage Belgian financial institutions in relation 

to foreign ones. Belgian financial institutions risk becoming deprived of 

this most perfected property right because of the lack of such a security 

device in Belgian law. 

The conflict of law rules with regard to rights in rem on receivables have 

been the object of heated debate. It is our conviction that the debate 

on the conflict rule is rooted in the insufficiency of the substantive rule. 

Originally, the lex rei sitae rule was applied. Most legal scholars assumed 

this meant that the law of the residence of the debtor of the receivable 

that is the object of the assignment / pledge applied. After the internal 

Belgian law abolished the requirement of notification to the debtor (Act 

47 

L. d’Avout, o.c., 686, n° 521. This development was already anticipated by G. 

Khairallah, Les sûretés mobilières en droit international privé, n° 124 et seq. 

with references. 

48 

For a more extensive analysis: V. Sagaert, ‘Het eigendomsrecht als volwaardig 

zekerheidsinstrument?’, in: XXXIste postuniversitaire Cyclus Willy Delva 

2003-2004, Mechelen, Kluwer, 2006, 189-230. 

49 

Cass. 17 October 1996, Arr.Cass. 1996, 930, Bull. 1996, 992, concl. J. Piret, 

Tijdschrift voor Bank- en Financiewezen.1997, 114, note I. Peeters, R.W. 

1996-97, 1395, note M.E. Storme. 

50 

This rule is subject to exceptions, the most important of which deals with 

financial instruments and cash: Act of 15 December 2004 with regard to securities 

on financial collateral, 1 February 2005. This exception was introduced 

pursuant to European law, and more precisely pursuant to the European Directive 

2002 / 47 / EC on financial collateral arrangements of June 6, 2002. 


137 


www.sellier.de


6 July 1994), the majority of authors applied the law governing the receivable 

that was the object of the assignment / pledge. Pursuant to this criticism, 

the Belgian legislator introduced a new conflict rule in 2002, which 

was quite remarkable and revolutionary in Belgian PIL. Article 145 Act 

of 2 August 2002 amended this conflict of law rule by providing that the 

proprietary aspects of the assignment are governed by the lex contractus 

of the assignment. Party autonomy was introduced in order to determine 

the proprietary effects of the assignment / pledge of receivables in relation 

to third parties. This rule has been the object of critical observations by 

Belgian legal scholars, who have observed that PIL filled the gap left by 

substantive law. 51 Without going into detail concerning these criticisms, 

it can be said that the main observation was that PIL was abused in order 

to provide a remedy for deficiencies in the substantive law. As Belgian 

substantive law did not allow for the same possibilities as the Sicherheitsübereignung 

or Sicherheitsabtretung in German law, PIL had to grant the 

possibility to parties to make use of these possibilities. More technical 

criticisms were addressed to the solution accepted by the legislator, such 

as (1) the uncertainty that follows from contractual freedom if third parties 

are the interested parties (as is always the case in property law) and 

(2) the impossibility of providing a clear-cut solution if different conflicting 

assignments / pledge agreements are governed by another law, which 

provides a different conflict solution. The debate that was raised after 

this amendment had many similarities to the legal debate on the Hansajudgment 

in the Netherlands. 52 

The introduction of the PIL Code has been the occasion to amend, once 

again, the applicable conflict rule. Drafters of the PIL Code were inspired 

by the rule proclaimed by Article 30 of the Uncitral Convention of 12 

December 2001 (Article 87 § 3 PIL-Code). According to this provision, 

the creation of rights in rem with respect to receivables and the effects of 

the transfer of the receivable is governed by the right of the State on the 

51 

V. Sagaert, ‘De zakenrechtelijke werking van de cessie van schuldvorderingen: 

de nieuwe IPR-regeling na de Wet van 2 augustus 2002’, Tijdschrift voor 

Privaatrecht 2003, 561-612; comp. W. van Lembergen, ‘Grensoverschrijdende 

cessie van schuldvorderingen na de Wet van 2 augustus 2002 op het financieel 

toezicht: consensus over consensualisme?’ Bank Fin.R. 2003, 145-159. 

52 

Comp. in Belgian literature: K. Christiaens and I. Peeters, ‘Cessie tot zekerheid 

in een grensoverschrijdende context. Bedenkingen bij het arrest van de 

Hoge Raad (Ned.) van 16 mei 1997’, Algemeen Juridisch Tijdschrift 1998-99, 

277-284. 

138 



www.sellier.de


territory of which the party that created the rights or that transferred the 

rights has its habitual residence at the time of creation or transfer. The 

effects of conventional subrogation on rights in rem is governed by the 

law of the State on the territory of which the party that subrogated has 

its habitual residence at the time of the transfer. 

However, the PIL Code has not been able to terminate the debate. On 

the one hand, discussions about the introduction of specific provisions of 

the proprietary effects of an assignment in the Rome I Regulation have 

given rise to new arguments. 53 Moreover, the rule of the PIL Code has 

been especially criticised by financial institutions (e.g. clearing companies) 

with regard to bank accounts. On the other hand, it is argued that 

bank accounts have a fictitious situs: for instance, the place where the 

payment obligation or obligation to make restitution of the money must 

be complied with. This is the location of the financial institution that is 

debtor with regard to that obligation. 54 

6.4.2.2. Recognition of foreign trusts 

We have already indicated that the French law introduced in 2007 a 

trust-like device called the fiducie. We have also demonstrated that this 

new device has led to a fundamental dogmatic debate about the foundations 

of French property law (cf. supra). Except for the rules of third party 

protection, which are remedied by publicity measures, 55 these rules can 

be implemented without affecting the numerus clausus principle. Several 

proprietary devices can be considered to be functionally equivalent to 

53 

V. Sagaert, H.C. Sigman and E.M. Kieninger, ‘De cessie van schuldvorderingen 

in het voorstel van de Rome I Verordening: een kritische analyse’, Bank en 

Fin. 2006, 336-347. 

54 

G. van Hecke and K. Lenaerts, Internationaal Privaatrecht, in A.P.R., Gent, 

Kluwer, 1989, p. 310. Comp. J. Erauw, o.c., 644-645, n°. 561. 

55 

L. d’Avout, o.c., 675, n°. 512: ‘Le maintien de la sécurité juridique locale 

implique simplement qu’un tiers ne puisse se voir opposer à son détriment un 

droit réel inattendu créé dans le cadre d’une transaction internationale dont 

le contenu lui serait resté inconnu. Il est suffisant de ménager l’ignorance, et 

d’aménager la connaissance du tiers. Les règles locales de publicité y pourvoient 

largement, bien plus que la règle abstraite du numerus clausus des droits 

réels qui ne devrait pas être érigée en obstacle péremptoire à la reconnaissance 

des droits réels constitués en verte de la loi applicable à l’acte juridique.’ 


139 


www.sellier.de


devices that protect the beneficiary of a trust. For instance, the Anglo- 

American tracing rules have a functional equivalent in the concept of real 

subrogation. Hence, French substantive law avoids a larger debate on the 

recognition of foreign trusts. 56 

Belgian law has not made as much progress in substantive law. For a long 

time, Belgium made no efforts to recognise the effects of foreign trusts on 

Belgian territory. In contrast to the Netherlands, Belgium did not ratify 

the Hague Convention. However, this situation proved to be unsustai n- 

able, and Belgian law has taken certain steps in the field of PIL. In the PIL 

Code, Article 124 provides that a trust is governed by the law chosen by 

the settler, and, if he has not made a choice of law, by the law of the State 

on which territory the settler has its habitual residence at the moment of 

the creation of the trust. Although the Belgian legislator emphasised during 

parliamentary works preceding the Act that this Act could not affect 

the numerus clausus principle, 57 it indirectly opens up Belgian property 

law. If, for instance, an American beneficiary makes a claim on financial 

instruments that have been deposited with a Belgian clearing institution, 

the choice of law between parties will be able to have its effect. 

However, the Belgian legislator has – as did the French legislator in 2007 – 

a reluctant attitude: the legislator has excluded from the scope of the Trust 

provisions all matters with regard to the validity of the acts of acquisition 

or transfer of property rights on the trust property, as well as the transfer 

or protection of third party-acquirers of the property (Article 125, § 2 PIL 

Code). It does not emerge from this provision that all property law, and 

especially the proprietary structure of the trust itself, is excluded from the 

scope, since the introduction of the trust chapter in the PIL Code would 

otherwise have been useless and senseless. However, it demonstrates how 

the Belgian legislator – in amending the PIL – followed the same line of 

reasoning and retained the same reluctance as the French legislator in 

amending the substantive law with regard to the fiducie. 

56 

French case law adopted a lex rei sitae approach, applying the law of the situs 

of the trust property: Court of Appeal Paris 10 January 1970, Recueil Dalloz 

1972, 122, note P. Malaurie. However, this position did not prevent French 

case law from accepting the obligatory effects of a trust device; some case law 

even extends this solution to proprietary effects. See the case law cited by G. 

Khairallah, o.c., n° 236 et seq. 

57 

Parl. St. Senaat 2003-04, 3-27 / 7, 236. 

140 



www.sellier.de


6.5. Concluding remarks 

As we pointed out in the beginning of this contribution, our intention 

was not to provide a final outline of the question of party autonomy in 

French and Belgian law. However, we have tried to demonstrate that the 

issue is the subject of a highly ambivalent debate in French and Belgian 

property law. The debate is burdened by the historical background and 

‘sacred’ dogmatic aspects that come into play when this issue is dealt with. 

Moreover, certain developments have also demonstrated that there seems 

to be a flexibilisation of property law, especially in French private law. 

We have also aimed to provide references with regard to the interaction 

between substantive law and national law. It has emerged that certain 

results achieved by the French legislator in substantive law are striven for 

through PIL by the Belgian legislator. However, both means of opening 

up national property law to new devices give rise to the same concerns 

and discussions. 


141 


www.sellier.de


www.sellier.de

C. 

Developments and Prospects in Europe 

and in European Law Projects 


www.sellier.de


www.sellier.de

7. 

Article 14 Rome I: A Political Perspective 

Paulien M. M. van der Grinten * 

‘In order to arrive at a common basis on which to determine questions 

between inhabitants of the countries living under different 

systems of jurisprudence, our courts recognise and act on a division 

otherwise unknown to our law (…).’ 1 


This contribution tells of the ‘making of’ Article 14 on assignment of the 

Rome I Regulation on the applicable law to contractual obligations. 2 I was 

asked to write on this subject because I had negotiated Rome I on behalf 

of the Netherlands. The contribution is a personal account of the negotiations, 

which lasted from 2006 until November 2007. The anecdotes 

may both comfort you, as they show that ‘Brussels’ is capable of having 

in-depth discussions on Rome I, and shock you in their banality. 

The contributions of Flessner and Verhagen also focus on the specific 

subject of assignment in private international law and Article 14 Rome 

I Regulation. 3 Their angle, however, is different from the more political 

approach in this paper. Therefore, I will not go into every detail of the 

rules in Article 14 Rome I, but leave that to the expertise of Flessner and 

Verhagen. 

* 

Coordinating Counselor Department of Law, Ministry of Security and Justice 

of the Netherlands 

1 

Farwell L.J. in Re Hoyles [1911] 1 Ch. 179 at 185. 

2 

Regulation (EC) No 593 / 2008 of the European Parliament and of the Council 

of 17 June 2008, OJ 2008, L 177 / 16 on the law applicable to contractual 

obligations (Rome I). 

3 

H.L.E. Verhagen, Party Autonomy and Assignment, Chapter 9 of this book 

and A. Flessner, Between Articles 14 and 27 of Rome I: How to interpret a 

European Regulation on Conflict of Laws?, Chapter 10 of this book. 

145 


www.sellier.de

C. Developments and Prospects in Europe and in European Law Projects 

What I will do here is describe where we came from with Article 12 of the 

Rome Convention 4 , how we started in 2006 with a Commission proposal 

on Article 14 on assignment, and how we struggled and failed. I shall 

finish with some thoughts about the future of Article 14 Rome I, and a 

short conclusion. 

7.2. Where we came from: the legal chaos under 

Article 12 of the Rome Convention 

In the 1970s, it took the negotiators of the Rome Convention a long time 

to conclude an agreement on a harmonised set of rules on the applicable 

law to contracts. The wording of the Rome Convention has certainly 

not guaranteed a uniform interpretation and application of the rules contained 

therein. Still, the Convention was a big success. 

Therefore, the original idea of the European Commission, at least its JLS 

section – Justice Liberty and Security – was to transform the Convention 

into a Regulation without many changes, simply because Regulations had 

become the suitable instrument under the Treaty of Amsterdam. 5 Some 

amendments might have to be made in areas known to have caused problems 

in practice, but otherwise the rules should remain more or less the 

same. 6 

4 

Rome Convention of 19 June 1980 on the law applicable to contractual obligations 

(80 / 934 / EEC), OJ L 266, 09 / 10 / 1980 p. 1-19, consolidated version 

30 December 2005, OJ 2005, C 334 / 3. 

5 

Treaty of Amsterdam amending the Treaty on the European Union, the Treaties 

establishing the European Communities and related acts, OJ C 340, 10 

November 1997. 

6 

Looking at Article 4 Rome I on the applicable law in absence of a party 

choice-of-law, it is clear that this Article has changed considerably as compared 

with Article 4 of the Rome Convention. The ‘hard and fast’ rules in 

Article 4 of the Rome I Regulation serve mainly to meet the criticism of others 

within the Commission (e.g. Internal Market), who thought that Private International 

Law lacked legal certainty and predictability, and that it should 

better be replaced by mutual recognition based on the law of the country of 

origin. In order to arrive at maximum legal certainty, the scope of the ‘closest 

connection’ rule in Article 4 of the Rome Convention has been narrowed 

down. 

146 



www.sellier.de

7. Article 14 Rome I: A Political Perspective 

This ‘no changes unless’ approach explains why the Commission did not 

make an impact assessment for Rome I, although later in the negotiations 

the Commission came to regret this. Whichever provision we would be 

talking about, any Member State opposed to the proposed change to the 

Rome Convention rule would argue that lack of an impact assessment 

meant the proposed change was not justified. 

Article 12 of the Rome Convention was one of the provisions for which 

the Commission wanted to improve the rules, based on wide consultation. 

Article 12 reads as follows: 

‘1. The mutual obligations of assignor and assignee under a voluntary 

assignment of a right against another person (‘the debtor`) shall be 

governed by the law which under this Convention applies to the contract 

between the assignor and assignee. 

2. The law governing the right to which the assignment relates shall 

determine its assignability, the relationship between the assignee and 

the debtor, the conditions under which the assignment can be invoked 

against the debtor and any question whether the debtor’s obligations 

have been discharged.’ 

In the years after the entry into force of the Rome Convention in the 

different Member States, it became apparent that Article 12 in practice 

lacked a uniform interpretation. This diversity in interpretation applied 

both to paragraph 1 and paragraph 2. There was no uniform answer to 

the question of whether the proprietary aspects were within or outside 

the scope of Article 12, and if they were within the scope, whether they 

should fall under paragraph 1 or paragraph 2. Likewise, if the proprietary 

aspects were outside the scope of Article 12 of the Rome Convention, 

what should the choice-of-law rule be for these aspects? Thus, the lack 

of a clear rule on the proprietary aspects of assignment resulted in considerable 

legal uncertainty. 

As regards Article 12 paragraph 1 on the relationship between the assignor 

and the assignee, some believed this rule should only apply to the 

contract to assign between the assignor and the assignee. 7 Others considered 

that Article 12 paragraph 1 of the Rome Convention should also 

7 

In the Netherlands, this view was widely held in legal literature. See e.g. 

L.F.A. Steffens, Naar een commune Nederlandse goederenrechtelijke verwijzingsregel 

voor de overdracht c.q. de bezwaring van een vordering, NTBR 


147 


www.sellier.de


apply to the proprietary aspects of an assignment, at least as between the 

assignor and assignee. 8 As regards the scope of Article 12 paragraph 2 of 

the Rome Convention, the views differed considerably. Some – mainly 

German – scholars believed that paragraph 2 should also govern the proprietary 

aspects other than those between the assignor and the assignee, 

including the effects against third parties. 9 To support their view, they 

referred to a decision of the German Bundesgerichtshof of 1983 in a case 

in which the Rome Convention did not even apply. 10 Others held the 

view that Article 12 paragraph 2 did not contain any rule to govern the 

proprietary aspects of an assignment. They argued that these aspects were 

completely outside the scope of Article 12 of the Rome Convention and 

were, therefore, left to the national choice-of-law rules of the Member 

States. 11 In France, this led to the application of the law of the debtor’s 

domicile to the effect of the assignment against third parties. 12 

I had my first thoughts on Article 12 of the Rome Convention in 1994, 

shortly after the Rome Convention had entered into force for the Nether- 

1995 / 3, p. 61; M.V. Polak, Vermogensrechtelijke meerpartijenverhoudingen, 

Deventer: Kluwer 1993. 

8 

In favour of application of paragraph 1 of Article 12 to all proprietary aspects 

of an asssigment: the Dutch Hoge Raad in the Hansa-judgement and Dutch 

legal literature, a.o. R.I.V.F. Bertrams, H.L.E. Verhagen, Goederenrechtelijke 

aspecten van de internationale cessie en verpanding van vorderingen op 

naam, WPNR 6088 (1993), p. 261-266; in favour of a more limited scope of 

paragraph 1 only for the proprietary aspects as between the assignor and the 

assignee: A. Senay-Cytermann, Les conflits de lois concernant l’opposabilité 

de créance, Rev. crit. dr. intern. privé 81 (1) 1992, p. 35; H. Gaudemet-Tallon, 

Le nouveau droit privé européen des contrats, Rev. trim. dr. eur. 1981, p. 275; 

P. Lagarde, Le nouveau droit international privé des contrats après l’entrée en 

vigeur de la Convention de Rome du 19 juin 1980, Rev. crit. dr. intern. privé 

80 (1991), p. 335; C. Reithmann & D. Martiny, Internationales Vertragsrecht, 

Köln: Verlag Dr. Otto Schmidt 2004, p. 309. 

9 

Reithmann & Martiny, o.c. footnote 8, p. 310; C. Von Bar, Internationales 

Privatrecht. Zweiter band. Besonderes teil, München 1991, nr. 33. 

10 

BGH 23.2.1983, BGHZ, 87, 19 / 22 f. In 1990 the Bundesgerichtshof decided 

that Article 12, paragraph 2, of the Rome Convention did govern the proprietary 

aspects of an assignment, BGH 20 June 1990, BGHZ, 111, 376. 

11 

E.g. in France, A. Senay-Cytermann, o.c. footnote 8, p. 48 and P. Lagarde, o.c. 

footnote 8, p. 335. 

12 

Idejm footnote 11. 

148 



www.sellier.de


lands in 1992. In my graduation paper, I discussed the question as to 

whether there was a need for a separate choice-of-law rule for the proprietary 

aspects of an assignment of receivables in private international law. 

At the time, I argued that there was no such need. In my view, Article 12 

paragraph 1 of the Rome Convention could be applied by way of analogy 

to the proprietary aspects of an assignment, including the effect of the 

assignment against third parties. Thus, the law governing the contract 

between the assignor and the assignee – which might have been chosen 

by the parties – would also govern all of the proprietary aspects of the 

assignment. 13 

My reasons to argue in favour of this party autonomy were twofold. The 

first reason was purely pragmatic: to limit the number of different laws 

applicable to an international assignment of receivables. The second reason 

was that in my view the validity of the assignment as between assignor 

and assignee was already within the scope of Article 12 paragraph 1 of the 

Rome Convention. At the time, the majority of Dutch authors argued 

that, for Dutch law, Article 12 paragraph 1 was limited in scope to the 

contract to assign between the assignor and the assignee, since the assignment 

itself was a matter of property law and therefore outside the scope 

of the Rome Convention. 

What I argued in my paper was that the distinction made in Dutch substantive 

law between contractual aspects and proprietary aspects of an 

assignment could not be made in the same way when applying Article 

12 of the Rome Convention. The fact that the Rome Convention is an 

international instrument of private international law not only explains its 

different wording but it also requires a different, autonomous interpretation 

of the concepts and the wording contained therein. Hence, the motto 

of this paper is as quoted from a judgment of Farwell L.J. in Re Hoyles: 

‘In order to arrive at a common basis on which to determine questions 

between the inhabitants of two countries living under different 

systems of jurisprudence, our courts recognise and act on a division 

otherwise unknown to our law (…)’. 14 

13 

See also R.I.V.F. Bertrams, H.L.E. Verhagen, o.c. footnote 8, arguing along 

the same line. 

14 



149 


www.sellier.de


In 1997, the Hoge Raad gave its judgement in the famous Hansa case. 15 

It held that Article 12 paragraph 1 of the Rome Convention not only 

governed the obligations between the assignor and the assignee but also 

applied to the proprietary aspects of an assignment. In practice, this meant 

that the parties could choose the law applicable to their contract to assign 

and thereby exercise party autonomy also to the proprietary aspects of the 

assignment. In the Dutch Property Law (Conflict of Laws) Act of 2008, 

this Hansa rule has been codified. 16 

Having the Hansa judgement, the Netherlands did not crave a new specific 

rule on the proprietary aspects of an assignment in Europe. However, 

the total lack of uniformity in the application of Article 12 of the Rome 

Convention in the Member States made us feel the need for European 

codification of the choice-of-law rules regarding the proprietary aspects 

of assignment. Preferably, this would lead to a codification of our Hansa 

rule. 

7.3. How we started: the Commission proposal 

(COM [2005] 650 [final]) 

Let me begin with a point of order. When we negotiated the Rome I 

Regulation, the provision on assignment was Article 13. Since in the 

course of our negotiations a provision on insurance was added, the provision 

on assignment became Article 14 in the final Regulation Rome I as 

it was adopted on 17 June 2008. For the sake of simplicity, I shall refer 

to Article 14 whenever I talk about the assignment provision in Rome I, 

whether in the Commission proposal, in later drafts, or in the Regulation 

as it was adopted. 

15 

Hoge Raad 16 May 1997, Nederlandse Jurisprudentie (NJ) 1998, 585 (Brandsma 

q.q. / Hansa Chemie AG). 

16 

Article 10 Wet conflictenrecht goederenrecht (Act on the Conflict of Law 

Rules Relating to Property Matters); now incorporated as Article 135 of the 

bill codifying all Dutch conflict-of-law rules: Vaststelling en invoering van Boek 

10 (Internationaal privaatrecht) van het Burgerlijk Wetboek (Vaststellings- en Invoeringswet 

Boek 10 Burgerlijk Wetboek) (32137), (Book 10 of the Dutch Civil 

Code). See on this Dutch Property Law (Conflict of Laws) Act of 2008 also 

Van der Weide in this book, Chapter 2. 

150 



www.sellier.de


The Commission launched its proposal for a Rome I Regulation in December 

2005. 17 The proposal for Article 14 Rome I contained a change 

compared to paragraphs 1 and 2 of Article 12 of the Rome Convention: 

namely, to include contractual subrogation in the same rule. This proposal 

was acceptable from the start to all Member States. However, the major 

change was found in the added paragraph 3. This paragraph contained a 

choice-of-law rule for the effects of an assignment against third parties. 

The rule proposed was the law of the assignor’s habitual place of residence. 

The Commission justified the choice of this rule by referring to the UN- 

CITRAL Convention on the Assignment of Receivables in International 

Trade, which contained a similar provision. 18 

Then in 2006, negotiations between the Member States started in the 

Council working group. We had one common goal: a uniform choiceof-law 

rule covering all aspects of an assignment. It may seem an impossible 

task to negotiate a complex and detailed instrument like Rome I 

between 27 Member States with more than twenty different languages 

being spoken and translated simultaneously. However, in negotiations in 

Brussels with regard to future European instruments, it is common that 

only a handful of Member States take part in the discussions on each of 

the specific issues of such an instrument. 

The reason for this limited participation in the discussion is that some 

issues are highly specialised, and Member States cannot always send their 

representative who is an expert with regard to such a specific matter. 

This certainly applies to the choice-of-law rules on assignment, which is 

a highly complex issue. Another reason might be that for some Member 

States the topic under discussion is economically less important than for 

other Member States. 

17 

COM (2005) 650 final, 2005 / 0261 (COD) of 15 December 2010, Proposal 

for a Regulation of the European Parliament and the Council on the law applicable 

to contractual obligations (Rome I). 

18 

Article 30 of the United Nations Convention on the Assignment of Receivables 

in International Trade of 12 December, 2001. To date, only Luxembourg, 

Madagascar, and the United States of America have signed this Convention, 

and only Liberia has ratified it. It has not entered into force yet, see http: // 

www.uncitral.org / uncitral / en / uncitral_texts / payments / 2001Convention_ 

receivables_status.html, last consulted on 22 March 2011. 


151 


www.sellier.de


Member States that did participate in the debates on assignment had 

various reasons for doing so. Some had an important financial industry to 

protect and facilitate. This applied, for instance, to the United Kingdom 

and the Netherlands, whose banks and lawyers were both deeply involved 

in complex structured finance products. Their respective governments 

thought this financial trade deserved protection and support by having a 

choice-of-law rule stimulating rather than complicating their financing 

business. Others happened to have an expert delegate: for example, a legal 

scholar who could contribute dogmatic arguments to the debate. Finally, 

some Member States wanted to protect their existing national choiceof-law 

rule on third-party effects, or to prevent the application of their 

national substantive laws on assignment becoming rarely used if a certain 

choice-of-law rule were to be adopted. 

To determine the Dutch position in the debate, I consulted on a regular 

basis in meetings, by email, and by telephone with financial experts 

from banks, law firms, and universities to learn the key points for the 

Netherlands. During negotiations, a delegation can participate in the 

discussions by making written proposals. In the case of Article 14 Rome 

I, we continuously stressed the needs of the financing industry and legal 

practice, giving examples and demonstrating the practical consequences 

of the proposed rules. 

However, even with few Member States, Brussels discussions sometimes 

are reminiscent of the tower of Babel. It is not so much the language itself 

that causes confusion; rather, the completely different national legal concepts 

we use in substantive law cause us problems in judging the proposed 

international rules of private international law. 

7.4. How we struggled (part I): 

the relationship between the assignor and the 

assignee in Article 14 paragraph 1 Rome I 

As has been determined extensively in legal literature, the legal concepts 

inherent in the substantive laws on assignment of the various Member 

States are varied. Aspects that under one law are a matter of property law 

are under another law part of the law of obligations. 

152 



www.sellier.de


The passing of title between the assignor and the assignee is an example 

of this. Under French substantive law, the conclusion of a contract of sale 

between a seller and a buyer has the immediate effect of passing of title 

(l’effet translatif du contrat). 19 Thus, French substantive law regards the 

passing of title between the parties as a contractual matter, being part 

of the law of obligations. Only the effect of an assignment against third 

parties is treated as a separate matter. 20 When it comes to Private International 

Law, this qualification as a contractual matter is applied in the 

same way. However, Dutch and German substantive law would qualify the 

transfer of title between assignor and assignee as a proprietary aspect of 

an assignment, being part of their respective property laws. Under Dutch 

substantive law, one could not separate from the erga omnes effect of the 

assignment this passing of title under the assignment between the parties. 

Save for a clear provision stating otherwise, the passing of title between 

the assignor and the assignee would be qualified as a proprietary aspect 

of the assignment and not as a contractual matter, also for the purpose of 

Private International Law. If this issue had to be treated differently for the 

purpose of Article 14 Rome I, it would be necessary to clarify the wording 

of this Article. 

Given these differences in legal concepts, how could we arrive at a uniform 

autonomous meaning of Article 14 paragraph 1? The Netherlands 

thought it should be clarified that Article 14 paragraph 1 Rome I governed 

the relationship between assignor assignee as a whole, including the proprietary 

aspects of the assignment between them. This is why, contrary to 

our own substantive law concepts, the Netherlands, Germany, and Au s- 

tria all proposed to amend the wording of Article 14 paragraph 1 Rome I. 

The proposal was to refer to the relationship between assignor and assignee 

rather than the obligations between them. 

The proposed wording of Article 14 paragraph 1 Rome I would lead to 

the same result as Article 12 paragraph 1 of the Rome Convention for 

the Member States whose law regarded the validity between assignor and 

assignee to be a contractual matter. However, it took months of negotiating 

before these Member States accepted the arguments put forward by 

the German and Dutch delegations. We had to explain why the difference 

in legal concept in the various substantive laws on assignment required 

19 

Article 1583 French Civil Code. 

20 

Thus, referred to as ‘third party effects of an assignment’ rather than ‘proprietary 

aspects of an assignment’. 


153 


www.sellier.de


a slightly different wording in paragraph 1 in order to arrive at the same 

interpretation of this paragraph in all Member States, including Germany 

and the Netherlands. Following an initial negative reaction, the negotiations 

moved reluctantly to allow a recital without an amended wording of 

paragraph 1. In the end the Council, the Commission, and the European 

Parliament accepted our proposed amended wording of the text of Article 

14 paragraph 1 Rome I. 

7.5. How we struggled (part II): third party effects 

and priority, Article 14 paragraphs 2 and 3 Rome I 

Most Member States and the Commission knew from the start that the 

choice-of-law rule for third-party effects and for priority of an assignment 

would form a crucial part of the negotiations on Rome I. The Commission’s 

proposal for Article 14 paragraph 3 Rome I, which provided for the 

effect against third parties to be governed by the law of the assignor’s habitual 

place of residence, had received immediate support from countries 

in the South of Europe and from many of the ‘new’ Member States. The 

European Economic and Social Committee (ECOSOC) also supported 

the proposed rule based upon the assignor’s habitual place of residence. 

The UK, however, had problems with the proposed rule in Article 14 

paragraph 3 Rome I. They considered Article 14 a key provision for their 

decision on whether to use their right to opt in to Rome I, and even said 

so explicitly in their written comments of September 2006. The UK believed 

that any rule intended to govern proprietary aspects should focus 

on the entitlement to the proceeds of the assigned claim rather than on 

the effect of the assignment against third parties. Under Article 12 of 

the Rome Convention, the English courts had held that the law governing 

the assigned claim – Article 12 paragraph 2 – was the proper law to 

govern priority and entitlement to the proceeds. 21 One of the arguments 

used in favour of this rule was that the debtor of the claim would often be 

a market participant, such as a bank or a bond issuer, who could not risk 

being mistaken about who owned the claim and thus the proceeds thereof. 

Therefore, there would be a great deal at stake for the UK and its financial 

21 

Re Raiffeisen Zentralbank Österreich AG v. Five Star General Trading LLC 

[2001] EWCA Civ 68 [2001] QB 825. 

154 



www.sellier.de


industry if the Rome I Regulation were to contain a different rule based 

on the assignor’s habitual residence. 

The same applied to the Netherlands, where the financial industry claimed 

it would cost them hundreds of millions of Euros if the rule proposed by 

the Commission were to be adopted. The Dutch financial industry had 

been able to profit from party autonomy in the choice-of-law rules on assignment 

ever since the Hansa case. 22 This had permitted them to conduct 

more transatlantic business than a rule based on the law of the assignor’s 

habitual residence would have permitted. Moreover, financial experts and 

their lawyers claimed that the law of the assignor was counterproductive 

for anything they did other than factoring or ‘bulk assignments’. More 

novel ways of security financing, like cash pooling and syndication of 

loans, were coming into fashion. Besides, the Hansa rule of party autonomy 

for the applicable law to an assignment was included in a Dutch 

bill on choice-of-law rules for property law. 23 In 2006, when we negotiated 

Rome I, this bill had already been submitted to the Dutch Parliament. 

Almost any rule was thought to be preferable to the one proposed by the 

Commission. 

So there we were, the Dutch delegation fighting for a modern, assignment-friendly 

choice-of-law rule for the proprietary aspects of an assignment 

based on the law governing the contractual obligations between the 

assignor and assignee as determined by party autonomy under Article 14 

paragraph 1 Rome I. While we submitted our written proposals to have 

the Hansa rule accepted Europe-wide, the UK argued that paragraph 3 of 

the Commission proposal should simply be deleted. This would allow the 

UK courts to continue the application of the law of the assigned claim in 

paragraph 2 to the proprietary aspects of an assignment. 

However, by the end of 2006, when we had finished our first round of 

negotiations, the assignor’s habitual place of residence tended towards 

majority support. Some Member States had asked to limit its scope to be 

in line with the limitations in the scope of the UNCITRAL Convention 

on the Assignment of Receivables in International Trade. Article 4 paragraph 

2 of that Convention excludes many types of assignments from its 

22 


23 

The bill that led to the Wet conflictenrecht goederenrecht (Dutch Property 

Law (Conflict of Laws) Act), see footnote 16. 


155 


www.sellier.de


scope. 24 For the same reason – keeping in line with UNCITRAL – some 

had argued that we should have a specific rule defining the assignor’s habitual 

residence. The definition of habitual residence in Article 19 Rome 

I deviates from the one in the UNCITRAL Convention, and therefore 

would result in an applicable law different from that designated by the 

UNCITRAL definition. 25 Thus, the envisaged coherence between the 

two instruments would not be realised. 

Neither the UK nor the Netherlands had any chance of a majority in 

favour of their respective proposals for a simple deletion of paragraph 3 

or an application of paragraph 1 of Article 14 Rome I to the proprietary 

aspects of an assignment, let alone a qualified majority as required under 

Article 251 of the EC Treaty. 26 In fact, the Dutch plea for a third-party 

effect rule based on party autonomy was received with considerable hostility, 

and was not supported by any other Member State. It was quite 

clear that the European Union was not ready for any such rule. The UK 

delegates and legal practitioners regarded the Dutch solution to be outrageous 

and quite wrong. 

Since both the UK and the Netherlands regarded the assignor’s habitual 

place of residence as the worst possible choice-of-law rule for proprietary 

aspects of an assignment, and more particularly, the third party effects 

24 

See Article 4, paragraph 2, UNCITRAL Convention of the Assignment of 

Receivables in International Trade: ‘This Convention does not apply to assignments 

of receivables arising under or from: (a) Transactions on a regulated 

exchange; (b) Financial contracts governed by netting agreements, except a 

receivable owed on the termination of all outstanding transactions; (c) Foreign 

exchange transactions; (d) Inter-bank payment systems, inter-bank payment 

agreements or clearance and settlement systems relating to securities 

or other financial assets or instruments; (e) The transfer of security rights in, 

sale, loan or holding of or agreement to repurchase securities or other financial 

assets or instruments held with an intermediary; (f) Bank deposits; (g) A letter 

of credit or independent guarantee.’ 

25 

Cf. Article 5 under (h) UNCITRAL Convention of the Assignment of Receivables 

in International Trade, which uses the place of business rather than 

the place of central administration. 

26 

As of December 2009, replaced by Article 294 of the Treaty on the Functioning 

of the European Union. 

156 



www.sellier.de


and priority thereof, 27 it was time to reconsider our respective positions 

in the negotiations. 

During the first months of 2007, the UK presented a written proposal to 

provide explicitly for a rule on the proprietary aspects of an assignment 

to be included in Article 14 paragraph 2 Rome I, being the law of the assigned 

claim as the preferable rule for matters of ownership and priority. 

They sent in their best experts from the Treasury to substantiate their 

position and to explain the importance of such a rule to the 26 other delegations 

in the meeting room. The proposal was welcomed immediately 

by some Member States, like Austria and Germany, who both wanted to 

regulate these proprietary aspects of an assignment in Rome I. Moreover, 

they were used to applying the law governing the assigned claim to those 

aspects. Some others as well thought the proposal worth considering. 

However, many, including the Commission and Member States from the 

South of Europe, such as France, Spain, and Italy, remained opposed to 

any rule other than the law of the assignor’s habitual place of residence. 

As a result, the UK was still far from a majority. 

The Netherlands went back to consult on the UK proposal with the Dutch 

Standing Committee on Private International Law and with the financial 

experts and lawyers. The question put to them was whether in order to 

prevent the assignor’s habitual place of residence becoming the European 

rule on the proprietary aspects of an assignment, the Netherlands should 

actively support the UK proposal now that the Hansa rule was not going 

to make it Europe-wide. It took intensive debates and a great deal of 

endurance to convince everyone concerned within the Nether lands that 

this was our best option in the negotiations. Many found it hard to let 

go of the Hansa rule. However, once we had sufficient support within the 

Netherlands to go ahead in support of the UK proposal, the Netherlands 

made a written proposal slightly amending the wording of the UK proposal. 

Shortly thereafter, certain Member States that up until then had supported 

the Commission proposal for Article 14 paragraph 3 Rome I were 

prepared to support a different solution. We will never know for certain 

whether they were convinced by the arguments raised by the UK and the 

Netherlands, or by the volumes of financial trade in the UK, or whether 

27 

Other than the validity as between the assignor and assignee to be governed 

by the law designated by Article 14 paragraph 1 Rome I. 


157 


www.sellier.de


they simply felt their change of position could help us reach a compromise. 

The fact is that those Member States could now accept a choiceof-law 

rule for the proprietary aspects of an assignment on the basis of 

Article 14 paragraph 2 Rome I, the law of the assigned claim. In April 

2007, Portugal came with a similar proposal to apply the law governing 

the assigned claim also to third-party effects and priority of an assignment. 

Even Member States that had previously taken a hard line supporting the 

assignor’s habitual residence expressed their wish to reach a compromise 

solution. 

Also in April 2007, the German Presidency asked the Council of Mini 

sters in an orientation debate to agree on some of the Rome I provisions. 

However, the more controversial ones were not on the list. These 

included the provision on overriding mandatory rules (Article 9 Rome I) 

and the choice-of-law rules on consumer contracts (Article 6 Rome I), 

as well as on insurance contracts (Article 7 Rome I) and on assignment 

(Article 14 Rome I). 

At the end of June 2007, two things happened. The first was that the 

German Presidency and the incoming Portuguese Presidency together 

presented a new draft of Rome I. This new draft contained two options 

for a choice-of-law rule on the effectiveness of an assignment and the 

questions regarding priority of an assignment. Option 1 was the law of 

the assigned claim, based on the proposals made by the UK, the Netherlands, 

and Portugal. Option 2 was the law of the assignor’s habitual 

place of residence (paragraph 3 of the Commission proposal), with its 

suggested textual improvements to bring it more in line with the UNCI- 

TRAL Convention on the Assignment of Receivables in International 

Trade. This was a glimpse of hope for those in favour of the law of the 

assigned claim. A Presidency paper in which one’s position in the debate 

is reflected can be – and indeed proved to be – helpful. At the same time, 

it meant that Portugal as Presidency would have to take a neutral position 

in the debate and would no longer be able to actively support one 

solution or another. 

The second thing that happened was that Spain, which had been an 

ardent supporter of the assignor’s habitual place of residence as the preferred 

solution for the proprietary aspects of an assignment, submitted a 

written compromise proposal with two options. Option 1 was to have the 

law of the assignor’s habitual place of residence as the general rule, but 

to provide for a special rule for certain types of assignment, such as bank 

158 



www.sellier.de


deposits. 28 The law governing the assigned claim would apply to the latter 

types of assignment. Option 2 was the other way round. The law governing 

the assigned claim would be the general rule for the proprietary 

aspects of an assignment. 29 However, the law of the assignor’s habitual 

place of residence would apply as a special rule for assignments of future 

receivables and bulk assignments. Spain also suggested including a pledge 

in Article 14, and this suggestion was soon accepted by the rest of the 

Council. 

The Dutch approach in the negotiations had always been that whatever 

the rule might be on third-party effects and priority of an assignment, it 

must not have a negative effect on Dutch financial trade. In our Brussels 

debates, the arguments put forward by the UK and the Netherlands 

were taken seriously, since no one could deny that the UK’s volumes of 

financial trade were much higher than anybody else’s. No one dared to 

take responsibility for the billions of Euros worth of losses that the UK 

finance business might incur if we ignored their arguments and adopted 

the wrong choice-of-law rule. 

Therefore, negotiations continued. Between June and November 2007, 

we strove to reach a compromise solution on Article 14 Rome I. Dozens 

of written proposals were made, based either on option 1 or option 2 of 

the Spanish proposal. We looked into every corner to find ways to accommodate 

both the supporters of the assignor’s law where they needed 

it (factoring and securitisation) and those who wanted the law of the 

assigned claim for the complex varieties in structured finance. However, 

for the UK and the Netherlands, a solution based on the assignor’s law as 

the general rule (option 1) was unacceptable because of its detrimental 

effects on these countries’ financial trade. The Council’s main problem 

was finding proper demarcation lines for the scope of the assignor’s law 

and for the scope of the law of the assigned claim. 

In the meantime, the Presidency negotiated with the European Parliament 

in order to have a first reading agreement on Rome I. The European 

Parliament did not have strong views on assignment, but had very strong 

views on certain other sensitive issues like consumers and insurance. Time 

28 

This would be in line with the scope of the UNCITRAL Convention on the 

Assignment of Receivables in International Trade. 

29 

Other than the validity of the assignment as between the assignor and the 

assignee. 


159 


www.sellier.de


began to run out, as most Member States felt that a second reading by 

Parliament would mean that all the issues on which we had reached such 

a careful compromise in the Council would be re-debated. 

Between September and November 2007, the positions of Member States 

started to shift. Banks and Treasuries in Spain, Italy, and even France no 

longer seemed to be as positive about the assignor’s law as the choice-oflaw 

rule for third-party effects and priority of an assignment. Apart from 

the time-pressure, things looked promising for the UK and the Netherlands. 

7.6. How we failed: no rule on proprietary 

aspects other than the validity between assignor 

and assignee, Article 14 Rome I 

Why is it that when we tried so hard and were so close to having a choiceof-law 

rule on the proprietary aspects of an assignment in Rome I we did 

not succeed? In my view, this is where political banality enters. 

By November 2007, the only outstanding issues in the Council were the 

provisions on consumer contracts (Article 6) and on assignment (Article 

14). The European Parliament and a large part of the Council wanted 

limited party autonomy in consumer cases, while a large minority of the 

Council was very much opposed to the notion. At this point, Rome I 

became a political trade-off. 

In Brussels, in the early morning of Monday, 19 November 2007, we 

made a last attempt to reach a compromise on Article 14 Rome I. The 

Portuguese Presidency gave the interested delegations one hour to find 

a compromise rule. Over the weekend, these delegations – instead of a 

weak compromise – had exchanged many e-mails discussing the pros and 

cons of having no choice-of-law rule on third-party effects and of priority 

of an assignment in Rome I. My hotline communication with the Dutch 

financial experts had led me to the conclusion that the Netherlands could 

not risk having a ‘bad’ choice-of-law rule on assignment. In order to keep 

the end-result of Rome I politically acceptable to them as a whole, some 

Member States could not accept a compromise solution on both Article 

14 on assignment and Article 6 on consumer contracts. The European 

Parliament stood ready to torpedo most of our consensus on the other 

160 



www.sellier.de


Rome I provisions if we did not agree there and then to the first reading 

compromise text of Rome I. 

So that was it. No rule on third-party effects or on the priority of an 

assignment in Article 14 Rome I. To soften the blow, there was only a 

strong review clause in Article 27 Rome I to have a report on the issue 

two years after adoption of the Regulation. 30 This means that for the time 

being our respective national choice-of-law rules on third-party effects 

and on the priority of an assignment remain applicable. However, insofar 

as a court in a Member State would find that these proprietary aspects 

of an assignment were within the scope of Article 14 paragraph 1 or 2 

Rome I, we have the European Court of Justice to give an autonomous 

interpretation of this Article of the Regulation. I very much doubt the 

European Court of Justice would accept such an interpretation of Article 

14 Rome I. National courts that want to apply the rules of Article 14 

paragraph 1 or 2 Rome I to third-party effects or to the priority of an assignment 

should therefore only do this by way of analogy or as part of an 

explicit national rule. 31 

7.7. Future developments: the report on Article 14 Rome I 

The report was due on 17 June 2010. However, the tender for a preliminary 

study was published as late as 22 May 2010. A report is now due no 

sooner than the summer of 2011. 32 In view of our negotiations on Article 

14 Rome I, the tender papers contain a few remarkable things. The study 

is about whether we need to harmonise choice of law rules on third-party 

effects and on priority of assignments. As if this were a question. The 

30 

Which is rather unusual, since normally review clauses take a period of time 

starting from the date when an instrument becomes applicable, not when it 

is adopted. In the case of Regulation Rome I, this means the report had to be 

submitted in June 2010 instead of December 2011. 

31 

See, however, A. Flessner’s contribution to this book, Chapter 10; he seems 

to hold a different view on this point. 

32 

See the Commission’s Invitation to tender JLS / 2010 / JCIV / PR / 0007 / E4, following 

the publication of a contract notice OJS 99 of 22 May 2010. According 

to the Invitation to tender (under 1.7.3 and 1.7.4), the study must be completed 

and approved by the Commission within 11 months after the tender 

contract was signed. 


161 


www.sellier.de


study needs to include an analysis of the legal, financial, and economic 

impacts of solutions presented during the negotiations. The study must 

take into account the relative importance to an economic sector of commercial 

activities, based on assignment and on the fact that a minority 

of Member States accounts for a vast majority of wholesale EU financial 

services. Moreover, the study must look at substantive laws on assignment, 

and especially at recent reforms for deformalisation of assignment. Finally, 

the study must analyse whether we need specific legislation – substantive 

or choice-of-law – on securitisation. 

Other future developments may include discussions on the deformalising 

rules on assignment in the Draft Common Frame of Reference. A totally 

different development concerns of course the effects of the financial crisis. 

Rumour has it that securitisation is already back in business. Will this 

increase the need for specific legislation in this field? Will the financial 

crisis lead to more reluctance to accept party autonomy? We cannot foretell 

this. What we do know is that before a new proposal on Article 14 

Rome I is tabled by the Commission, a thorough impact assessment will 

have to be made. This assessment should contain an in-depth analysis of 

the consequences for financial trade of the various options for a rule on 

third-party effects and on the priority of an assignment. 

7.8. Brief conclusions 

The Rome I negotiations showed me that party autonomy in substantive 

law does not automatically lead to party autonomy in choice-of-law. 33 It 

also struck me that the European Parliament was practically absent in 

the debate on Article 14 Rome I, even though the Article is extremely 

important for the financial trade in many Member States. Moreover, the 

negotiations demonstrated that arguments from financial and legal practice 

are more important in the discussions than is any legal scholar’s authoritative 

view. The emphasis on these aspects in the tender papers for 

the report on Article 14 Rome I as provided for in the review clause of 

Article 27 Rome I underline the importance of these financial and practi- 

33 

See for example the French Loi Dailly nr. 81 of 2 January 1981, which allows 

for bulk assignments with practically no formalities vis-à-vis third parties. 

Thus far, this has not had any effect on the French choice-of-law rules on 

third-party effects or on the priority of an assignment. 

162 



www.sellier.de


cal arguments. However, politics prevail in the end, resulting in a trade-off 

between Member States, the European Parliament, and the Commission. 

For the time being, every Member State is at liberty – within the limits 

of an autonomous interpretation of Article 14 Rome I by the European 

Court of Justice – to use its own choice-of-law rule for third-party effects 

and for the priority of an assignment. We will all continue to do this until 

we have a European choice-of-law rule on third-party effects and on the 

priority of assignments. For the Netherlands, this means that the Hansa 

rule, as codified in our laws, will continue to be applied for some time at 

least. 


163 


www.sellier.de


www.sellier.de

8. 

European Conflict Rules for the 

Mutual Recognition of Security Rights in Goods 

Ulrich Drobnig * 


In its ‘Stockholm Programme’ of 2010 1 , the Council of the EU requests 

the Commission, inter alia, to continue work on the approximation of 

conflict rules in those areas in which it is necessary, such as separation 

and divorce. In addition, it might extend, inter alia, to the field of security 

rights. 2 This contribution will attempt to determine the desirability and 

feasibility of European rules on recognition of security rights in movables 

created in another Member State if the encumbered asset is subsequently 

exported or otherwise brought to another Member State. At the same 

time, the limits of the device of recognition will also be pointed out. 

For several reasons that will become apparent, the two major basic types 

of security in movables – scil. retention of title on the one hand and other 

security rights on the other –will be discussed separately. 

With the exception of recent Swiss and Dutch statutory provisions, the 

relevant conflict rules have been elaborated by the courts; in some countries, 

they have been supported by so many court decisions that one can 

speak of a firm case law. 

Before dealing with the legal issues, it is helpful to specify the economic 

dimension of the underlying export and import transactions. In 2008, the 

last ‘normal’ year of the world economy, the total value of exports from 

* 

Director emeritus at the Max-Planck-Institute for Comparative and International 

Private Law, Hamburg. 

1 

OJ EU 2010 C 115, p. 1. 

2 

Paragraph 3.1.2 sub paragraph 3 sent. 2; cf. Cepelak, Would we like to have a 

European Code of Private International Law?: Eur.Rev.Priv.L. 2010, 705-729, 

712. 

165 


www.sellier.de


Member States of the EU to other areas of the world amounted to 1.310 

billion EURO, and the total value of imports into the European Union 

came to 1.565 billion EURO. However, even greater was the volume of 

merchandise and of semi-finished and finished goods that were moved 

in the internal market across national borders to another Member State: 

the value of imports from other Member States amounted to 2.639 billion 

EURO, and that of exports to other Member States came to 2.714 

billion EURO. 3 These are truly imposing figures, and may give an idea of 

the values that are the main but not even the only objects of European 

conflict rules on movable property. 

The vast majority of these transactions involve payment obligations. 

Many of these monetary obligations are secured by proprietary security 

in favour of the seller. A considerable number of court decisions in all 

European countries have dealt with retentions of title in imported goods. 

However, some major trans-border transactions have also been financed 

by banks and other credit institutions; in these cases, security rights other 

than retention of title have been employed, and again the issue of recognition 

has arisen when the goods involved were shipped to an EU Member 

State. 

Security rights in monetary claims and security rights in aircraft and ships 

are excluded from the present study; the former because they are dealt 

with broadly by A. Flessner 4 and by Verhagen and van Dongen 5 , and the 

latter because of their very special features. 

Before going into technical detail, it is useful to briefly inquire into the 

practical demand for reliable rules in this field. In the framework of an EU 

Study on the interaction of contract and property law, businessmen and 

legal practitioners had been asked about their practical experiences. The 

summary of their replies represents their response: 

‘All the answers, irrespective of trade or country, complained about 

the difficulties in relation to credit securities in the cross-border circulation 

of business transactions … To compound matters, there is a lack 

3 

Eurostat Datenbank Comext. DS-016890-EU27 Trade since 1995, Jan.-Dec. 

2008. 

4 

Infra, Chapter 10. 

5 

Infra, Chapter 9, n. 1. 

166 



www.sellier.de

8. European Conflict Rules for the Mutual Recognition of Security Rights in Goods 

of uniform legislation in this field, and even of mutual recognition of 

national and international securities.’ 6 

In accordance with the enormous practical importance of the retention 

of title as compared to other non-possessory security rights, retention of 

title will be discussed first (infra 8.2), and other non-possessory security 

rights thereafter (infra 8.3). 

8.2. Retention of title 

The following survey restricts itself to the ‘simple’ retention of title. The 

various forms of extension – for instance, into the proceeds from a resale 

of the purchased goods by the buyer 7 or into products of work done on the 

bought goods by the buyer – are excluded here. 

8.2.1. European law 

The importance of the retention of title for export / import transactions 

in the EU internal market is underlined by the fact that the European 

legislator has seen fit to expressly oblige the Member States to recognize 

a retention of ownership in goods imported from another Member State 

and validly agreed upon. Article 4(1) of the Directive on combating late 

payments of 29 June 2000 8 provides: 

‘1. Member States shall provide in conformity with the applicable 

national provisions designated by private international law that the 

seller retains title to goods until they are fully paid for if a retention of 

title clause has been expressly agreed between the buyer and the seller 

before the delivery of the goods.’ 

6 

C. von Bar and U. Drobnig (eds.), The Interaction of Contract Law and Tort 

and Property Law in Europe (Sellier, München 2004), 439-440. 

7 

This aspect is covered in a practical manner in the aforementioned contributions 

of Flessner as well as of Verhagen and van Dongen, supra n. 4 and 5. 

8 

Directive 2000 / 35 / EC of 29 June 2000 on combating late payment in commercial 

transactions (Off.J. EC 2000 L 200, p. 35). See postscript. 


167 


www.sellier.de


Retention of title is defined by Article 2 no. 3 as meaning: 

‘… the contractual agreement according to which the seller retains 

title to the goods in question until the price has been paid in full.’ 

And the purpose of the provision is set out in consid. (21): 

‘It is desirable to ensure that creditors are in a position to exercise a 

retention of title on a non-discriminatory basis throughout the Community 

if the retention of title clause is valid under the applicable 

national provisions designated by private international law.’ 

Remarkably, this provision does not seem to play any role in practice, 

since it has not been mentioned in any specific case. However, in effect 

the courts in most countries act in conformity with that rule, as the following 

brief survey illustrates. 

Instead of on the basis of alphabetically arranged national reports country 

by country, it is more helpful for the conclusions to be drawn from the 

survey to group the reports by the results achieved. A first group comprises 

those countries that in practice achieve full recognition of retentions of 

title created abroad (infra 8.2.2). Into a second group fall those countries 

that, while basically recognizing foreign retentions of title, exclude specific 

types of this security (infra 8.2.2.3). Finally, a third group covers those 

countries that in practice exclude all foreign-created retentions of title, 

unless a specific aspect of these is adapted to the legal system of the state 

of importation (infra 8.2.2.4). 

8.2.2. Foreign retentions of title fully recognized 

a) Austria 

Simple reservations of title that had been created validly in Germany 9 

and in Italy 10 have been recognized by the Austrian Supreme Court. In 

the earliest German-Austrian case, the Supreme Court gave the following 

reasons: 

9 

S.Ct. 19 January 1989, IPRE 3 (1988-1990) no. 97; S.Ct. 31 March 1989, ibid. 

no. 98; S.Ct. 29 May 1990, SZ 63 no. 85 and IPRE 3 (1988-1990) no. 99. 

10 

S.Ct. 30 August 2002, ÖBA 2003, 536, 537 (recognition implied). 

168 



www.sellier.de


‘Since the retention of title is also recognized in Austria, it is not 

subject to form requirements nor to requirements of publicity; the 

change of situs of the sold goods is here irrelevant for purposes of the 

conflict of laws.’ 11 

Two other Supreme Court decisions deserve special mention, since they 

deal with a specific situation that recurs sometimes under comparable 

circumstances. Both cases relate to sales by a Swiss seller to an Austrian 

buyer living in Austria. The validity of the security right was doubtful, 

since in Switzerland a retention of title is effective (vis-à-vis third persons) 

only if the retention of title is entered into a register located at the place 

of business of the purchaser. By definition, however, in cases of exports 

from Switzerland the importer resides abroad, so a registration in a Swiss 

register is not feasible. In two cases, the Austrian Supreme Court decided 

that although no registration of the retention of title had been effected, 

the agreed retention of title is fully effective in Austria as soon as the purchased 

goods are delivered to the buyer in Austria. 12 The Swiss legislator 

has now helpfully clarified the situation: The Law on Conflict of Laws of 1 

January 1989 provides that in the case of exports to foreign countries the 

registration of a retention of title in a Swiss register is not required. 13 

b) France 

The former fundamental hostility of French courts towards foreign nonpossessory 

security rights 14 has today, in view of an increasing number of 

French legislative instruments for non-possessory security, given way to a 

more liberal attitude. Foreign simple retentions of title are today recognized 

implicitly in the French buyer’s bankruptcy if the substantive and procedural 

requirements of French insolvency law are met by the foreign seller. 15 

11 

Cf. S.Ct. 19 January 1989 (supra n. 9). 

12 

S.Ct. 26 April 1961, SZ 34 (1961) no. 67 (p. 183: the reservation of title 

became effective ‘at once’); more precise is S.Ct. 13 May 1987, IPRE 2 (1983- 

1987) no. 114, p. 243 (effective if the object is delivered to Austria). 

13 

Swiss Law on Conflict of Laws of 1989 Article 103. 

14 

Cf. Cass. 24 May 1933, JDI (Clunet) 1935, 380; Cass. 8 July 1969, ibidem 1970, 

916. 

15 

Cf. e.g. Cass. 22 December 1975, Bull.civ. 1975 IV no. 313: the German 

seller had terminated the sales contract before the insolvency proceeding was 

opened. Cf. also Cass. 5 March 1996, Bull.civ. 1996 IV no. 73: successful 


169 


www.sellier.de


c) Germany 

The attitude of the German courts has been strongly influenced by a 

leading case of the Federal Supreme Court of February 1966 published 

in the official collection of decisions. An Italian exporter had retained 

ownership of goods exported to Germany, but had not complied with the 

relevant Italian provisions on making that reservation effective against 

third parties. After the sold goods had arrived at the German buyer’s 

place of business, a creditor of the buyer brought an execution against 

the goods; the Italian seller objected, invoking an earlier retention of 

title. The court took great pains to secure the Italian seller’s preference, 

although under Italian law the seller’s rights had no effects against third 

parties. It ‘construed’ a tacit agreement between seller and buyer about a 

retention of title concluded upon the goods crossing the German border, 

which assured the precedence of the seller over the rights of the German 

execution creditor. 16 German commentators, while approving the 

result, invoked the idea that the Italian seller’s retention of ownership had 

automatically become fully effective upon the merchandise crossing the 

German border. 17 This result of the decision – but not its reasoning – has 

been generally approved. 18 

Many German courts that subsequently had to pass on retentions of title 

agreed by foreign exporters with their German buyers relied on that 

judgment and based their decisions on two issues: first, had the parties 

agreed upon a retention of title? The consequence of an affirmative answer 

reached in all cases was that a valid and fully effective reservation of ownership 

had been achieved upon the affected goods crossing the German 

border. Second, the legal situation in the seller’s country was not invesclaim 

for delivery of German seller’s successor against French buyer’s French 

lessee, who was held to be in bad faith with respect to the buyer’s entitlement 

to the goods, since the lease contract mentioned the reservation of ownership 

(cf. note D. 2000 J 74). Cass. 8 January 2001, Rev.crit.d.i.p. 2002, 328 note 

Bureau, did not recognize an Italian seller’s retention clause on the contractual 

level as not affecting those goods that the seller reclaimed. 

16 

Fed.S.Ct. (BGH) 2 Feb. 1966, BGHZ 45, 95, IPRspr. 1966-1967 no. 54. 

17 

See especially Kropholler, Internationales Privatrecht (ed. 6, Tübingen 2006) 

563 s. 

18 

In addition to Kropholler (preceding n.) Kegel / Schurig, Internationales Privatrecht 

(ed. 9, Munich 2004) 773; von Hoffmann / Thorn, Internationales 

Privatrecht (ed. 9, Munich 2007) 527 ss. 

170 



www.sellier.de


tigated, thus saving time and costs with regard to ascertaining the legal 

situation in the seller’s country. While most of the decisions dealt only 

with the position of the exporter vis-à-vis the importer, 19 in one case a 

third party was affected who had bought the goods from the importer. 20 

The German legislator later adopted the result of this case law in a generalized 

form. Article 43 paragraph 3 of the Introductory Law to the German 

Civil Code, as enacted in 1999, provides: 

‘(3) If an interest in property brought into this state has not yet been 

perfected, events that occurred in another state are to be considered 

like local events for the determination of the perfection of such an 

interest in this state.’ 21 

In effect, the legislator has confirmed, albeit by an involved and vague 

general formula, the result of the Supreme Court’s decision and of the 

succeeding case law. 

d) Netherlands 

In some recent Dutch cases, including one decided by the Supreme Court 

in 1997, the recognition of German reservations of title was taken for 

granted. Only the consequences of events that occurred after importation 

of the goods into the Netherlands were the object of discussion and 

decision. 22 

19 

App.Ct. (OLG) Hamm 13 July 1989, IPRspr. 1989 no. 76 (Italian seller); App. 

Ct. (OLG) Koblenz 16 January 1992, IPRspr. 1992 no. 72 (Dutch seller); App. 

Ct. (OLG) Koblenz 23 December 1988, IPRspr. 1988 no. 35 (Belgian seller); 

Sup. First Inst. Ct. (LG) Köln 24 March 1992, IPRspr. 1992 no. 73 (Belgian 

seller). 

20 

Sup. First Inst. Ct. (LG) Hamburg 28 June 1978, IPRspr. 1978 no. 42 (Spanish 

seller). But see also the leading case of 1966 decided by the Federal Supreme 

Court, supra n. 16. 

21 

Translated by P. Hay, Am.J.Comp.L. 47 (1999) 652. 

22 

S.Ct. (H.R.) 16 May 1997, N.J. 1998 no. 585 (Hansa-case) with note ThMdB.: 

the issue involved a German ‘prolonged’ reservation of title: namely, one extended 

to the proceeds of sale of the imported goods; App.Ct. (Hof) Leeuwarden 

3 March 2009, NIPR 2010, 292: loss of German reservation of title 

by integration of the delivered pieces into a machine; First Inst. Ct. (Rb.) 


171 


www.sellier.de


In the new Dutch Law on Conflict Rules of 2008 23 with respect to the 

proprietary regime of tangible property and so forth, Article 3 deals with 

retention of title. The basic rule of Article 3 paragraph 1 determines the 

proprietary regime of a retention of title as being that of the country in 

which the affected property is located at the time of delivery. This is a 

helpful solution for international sales because, in effect, delivery of goods 

usually will take place in the importer’s country. Thus, restrictive effects 

of the law in the seller’s country will be irrelevant. This result can also be 

achieved by the parties agreeing on application of the law of the country 

of importation, provided that this agreed law upholds that agreement and 

that the goods in fact are imported into that country (Article 3 paragraph 

2). The preceding rules apply also to the proprietary regime of a leasing of 

goods intended for use abroad (paragraph 3). 

e) United Kingdom 

Two decisions of English appellate courts from 1976 and 1991 have 

confirmed that simple retentions of title involving goods imported from 

abroad are recognized in all parts of the United Kingdom. The first case, 

the famous Romalpa case, 24 concerned a Dutch exporter’s extended retention 

of title against an English importer. Some time after the arrival of 

the goods, another creditor of the importer appointed an administrative 

receiver for the importer. The Dutch exporter claimed, first, redelivery 

of the unsold and unpaid-for portion of its merchandise. In the first instance, 

this claim was admitted by the receiver of the defendant importer, 

hence the court ordered accordingly. 25 The Court of Appeal also rejected 

Rotterdam 28 October 2008, NIPR 2010, 121: reservation of title to a car 

moved between the Netherlands and Germany. Cf. also the recent decision 

of the European Court of Justice of 10 September 2009 (C-292/ 08 – German 

Graphics, ECR 2009, 8421; N.J. 2010 no. 541): judicial remedies of the German 

exporter of machines sold under retention of title in the Dutch buyer’s 

insolvency proceedings. 

23 

Dutch Law of 25 February 2008 containing conflict rules for the proprietary 

regime of tangible property and so forth, NIPR 2008, 234 ss. 

24 

G. McCormack, an English expert, evaluated the decision as a ‘seminal case’ 

and describes its effects for English domestic law, cf. G. McCormack, Reservation 

of Title (ed. 2, London 1995) 73 ss. 

25 

Aluminium Industrie Vaassen BV v. Romalpa Aluminium Ltd., All E.R. 1976 [2] 

552 at 555 (Q.B.D. 1975). 

172 



www.sellier.de


the English importer’s defenses. Two of the three Lord Justices expressly 

confirmed the Dutch exporter’s title to the unsold goods still held by the 

English importer. 26 The more doubtful points on the English importer’s 

power to dispose of the plaintiff’s – unpaid for – assets and the rights of 

the Dutch exporter in the proceeds of these sales go beyond the scope of 

the present inquiry. 

While the effect of a simple reservation of title in England was not subject 

to doubts, Scottish courts took a different position. In a comparable 

factual situation, a German exporter of steel had to appeal to the House 

of Lords in order to achieve a comparable result. In this case, the German 

exporter also had retained ownership in the goods (in this case steel) sold 

to the Scottish importer for whom receivers were also appointed later. In 

keeping with earlier decisions of Scottish courts, the German exporter’s 

claim for return of the unpaid-for steel had been refused by the Scottish 

courts in the first and the second instance. This was based on the grounds 

that the exporter’s clause amounted to the creation of a non-possessory 

security, which is invalid in Scotland. However, in the House of Lords, 

the two Lord Justices who gave substantive reasons for their opinions 

agreed in clearly distinguishing between a retention of ownership by a seller 

and the creation of a non-possessory security by a debtor. 27 A retention of 

owner ship is not subject to the formalities of creating a security right. 28 

It is significant that in both cases the reasons for recognition are based 

exclusively upon English substantive law; the law of the country of exportation 

is nowhere mentioned. 

f) Conclusion 

The five countries dealt with in this section – Austria, France, Germany, 

the Netherlands, and the United Kingdom – recognize reservations of 

title created in another Member State if the goods concerned are brought 

26 

Aluminium Industrie Vaassen BV v. Romalpa Aluminium Ltd., All E.R. 1976 [2] 

552 at 557 ss., 561, 563 (per Roskill, L.J.) and 567 (per Megaw, L.J.). 

27 

Armour v. Thyssen Edelstahlwerke AG, 1991 (2) A.C. 339 (H.L.), at 351-353 

and 354 s. 

28 

Cf. the opinions of Lord Keith of Kinkel and Lord Jauncey of Tullichettle cited in 

preceding footnote. 


173 


www.sellier.de


into the importer’s Member State. Hence, these five Member States fully 

comply with the aim of the European Directive on mutual recognition of 

reservations of title, though without invoking it. 

It is significant that the emphasis of recognition has shifted from the 

proprietary to the contractual level. Except in the German leading case, 

the judges – including all other German judges – do not inquire into the 

proprietary status of the retention of title in the country of exportation. 

Relevant are two other factors: first, had the parties agreed on a retention 

of title? If so, is the retention of title valid and effective according to the 

law of the country of importation? In the light of this broad judicial practice, 

one may have some doubt about the new Dutch rule that allows the 

parties to agree on the application of the law of the country of importation 

but subject to the acceptance of this choice of law by the law of the country 

of importation. 29 This provision may risk encouraging the argument e 

contrario that without such express choice of law the desirable application 

of the new lex rei sitae is doubtful. In addition, is it really legitimate for the 

old lex rei sitae to determine the application of the future new lex rei sitae, 

although the latter has to agree? I do not know of any legal system with a 

rule dealing with such a choice by the parties. 

The aforementioned general practice may explain why the recognition 

approach used by the European legislator for retentions of title created 

in another Member State is not used and invoked in judicial practice. 30 

More relevant may be the German legislator’s approach in its new conflict 

rules on property of 1999: ‘events’ (Vorgänge) that occurred abroad 

before things were moved to Germany are regarded as having occurred 

in Germany. 31 However, this clause has also not been invoked in the 

relevant cases – if consciously so, probably because the matter has been 

taken for granted. 

29 

Cf. supra II 2 d). 

30 

Cf. supra II 1 in fine. 

31 

Introductory Law to the Civil Code Article 43 paragraph 3, as enacted in 1999 

and quoted supra at p. 171 lit. c). 

174 



www.sellier.de


8.2.3. Foreign retentions of title recognized – 

unless resale is allowed 

a) Principle and exception 

Denmark and Sweden generally recognize foreign retentions of title. 32 

However, an exception to this rule is made if the Danish or Swedish buyer 

is a merchant and is entitled, according to the terms of the contract of 

sale, to dispose of the goods bought in favour of a sub-buyer. 33 Economically, 

the exception alluded to in the title of this section is very important, 

since it has a broad field of application. In effect, it voids retention of title 

clauses in sales to merchants reselling to other merchants. The effect of 

these clauses is limited to sales to producers and to merchants reselling 

to consumers. 

b) Conclusion 

Is the Scandinavian exception compatible with the European obligation 

of mutual recognition of retentions of title? In my view it is. Recognition 

of security rights in the field of corporeal movables does not mean 

that retentions of title must be accorded the full effects of the security 

right that they may enjoy in the country of exportation. The effect of 

recognizing retentions of title must be limited to the effect that a cor- 

32 

Denmark: English seller’s retention of title in machine to be used in Danish 

buyer’s enterprise recognized: S.Ct. 3 November 1983, UfR 1984 A 8. In 

Sweden, no directly relevant case has been found. However, already in 1932 

a German seller’s retention of title was recognized in the Swedish buyer’s 

bankruptcy: S.Ct. 27 April 1932, NJA 1932 I no. 84. 

33 

Denmark: S.Ct. 8 February 1983, UfR 1983 A 311; Vestre Landsret 31 January 

1992, UfR 1992 A 373; Østre Landsret (ØL) 6 April 1998, UfR 1998 

A 1073. If the asset sold by a German seller but not paid for by the Danish 

buyer is retaken by the German seller with the consent of the Danish buyer 

shortly before the latter’s bankruptcy, the buyer is entitled to reclaim the purchase 

price from the German seller: S.Ct. 21 August 1987, UfR 1987 A 766. 

Sweden: In the bankruptcy of the Swedish buyer, the German seller, despite 

a reservation of ownership clause, was not allowed to reclaim the sold but 

unpaid-for machines from the buyer’s bankruptcy estate: S.Ct. 31 October 

1978, NJA 1978 no. 114. 


175 


www.sellier.de


responding retention of title enjoys in the country of importation. This 

result is supported by the observation made in the preceding section 

(8.2.2 f) that recognition of security rights, in practice, boils down to 

conferring those rights to the foreign secured creditor that holders of 

the corresponding security rights generally enjoy in the Member State 

of importation. 

8.2.4. Foreign retentions of title only recognized if formal 

requirements of country of importation observed 

a) Court decisions 

The domestic law of important southern European countries, such as Italy 

and Spain, establishes formal requirements unknown in other European 

countries. In Italy, a ‘data certa’ of the agreement creating the security 

is required, which must be furnished in a public document. 34 In Spain, 

a contract of sale providing for deferred payment of the purchase price, 

whether with or without a reservation of ownership, must currently be 

entered into a register of deferred sales of movable goods. 35 

As could be anticipated, these provisions have been applied to imported 

goods, and even to foreign automobiles passing through the two countries. 

36 

b) Compatibility with European Law? 

Here again the question arises as to whether this insistence on invoking 

domestic law is compatible with the European obligation to recognize 

34 

CC artt. 1524, 2704. 

35 

Ordinance for the register of sales with deferred payment of 19 July 1999 (B.O. 

no. 172 of 20 July 1999). 

36 

Cass. 21 June 1974 no. 1860, Riv.dir.int.priv. 1975, 335 (Austrian lorry in 

transit); Corte d’appello Milano 6 April 1956, Foro it. 1957 I 1856 (German 

lorry, despite dated notarial document); Trib. Latina 19 February 1973, Giur. 

it. 1974 I 2, 421 (Greek retention of title in lorry; however, the vehicle documents 

designated the buyer as owner). Spain: Audiencia Provincial de Barcelona 

13 September 1989, Rev.Esp.Der.Int. 42 (1990) 644 with note Sanchez 

Lorenzo (German bus). 

176 



www.sellier.de


retentions of title created validly in another Member State. The issue here 

is even more aggravating, since it is concerned not only with the refusal 

of recognition in a specific – although an important – sector of business 

(supra 8.2.3 a); rather, it involves the general disregard of any retention of 

title created in any other Member State without a ‘data certa’ contract. 

In 2005, the EU Commission brought a proceeding against Italy, claiming 

that Article 1524 paragraph 1 of the Italian CC violated the Directive; 

according to this provision, a retention of title can be upheld only against 

the buyer’s creditors if the retention clause is contained in a document 

that has a data certa that precedes the date of an execution brought by another 

creditor. A later decree added two further conditions. However, the 

action failed. 37 The European Court of Justice distinguished between the 

effects of a reservation of ownership as between the parties and as against 

third persons; it maintained that third parties should not be affected by 

the Directive. This reasoning is not convincing. 

Nothing in the terms of Article 4 or in the definition in Article 2 of the 

Directive 38 justifies the aforementioned distinction. Moreover, restricting 

the effect of a retention of title to the internal relationship between 

seller and buyer runs counter to the very purpose of the clause, which is 

intended to protect the seller against the execution and the insolvency 

creditors of the buyer. 

However, the result of the decision may be justified on other grounds. In 

the field of property law, which is so strongly dominated by the fundamental 

principle of lex rei sitae, the duty to recognize retentions of title created 

in another Member State may have to be defined differently than in other 

fields. A retention of title created validly in another Member State cannot 

by virtue of recognition have broader proprietary effects than in the 

country of origin. This thesis is to some degree supported by the case law 

on imports of merchandise from Italy and Spain: especially the German 

courts disregard, consciously or unconsciously, the limited effect that retentions 

of title have in the country of exportation. They only apply the 

stronger effect that the law of the country of importation attaches. The 

ratio is convincing: commercial relations as well as legal certainty demand 

within each country one regime of proprietary rights. 

37 

European Court of Justice 26 Oct. 2006 (C 302 / 05), ECR 2006, 10 597. 

38 

Cf. texts supra 8.2.1. 


177 


www.sellier.de


Seen from this angle, it is justified to attach to retentions of title in imported 

goods the proprietary effects of the new lex rei sitae. It is therefore 

irrelevant whether the retention of title that is recognized had narrower 

or broader effects in the exporter’s country. In the result, therefore, the 

decision of the European Court of Justice of 2006 is correct. 

8.3. Non-possessory security rights other 

than retention of title 

8.3.1. Uniform function, but lack of uniform 

terminology and regulation 

The somewhat vague heading of this section reflects the very considerable 

differences not only of terminology but also of substance existing in this 

field in the Member States of the EU. 

While retention of title is a concept and a type of security in movables 

that is used broadly in Europe and understood everywhere in the same 

sense – even though some details differ – such uniform understanding is 

absent for other types of non-possessory security in movables. The multitude 

of terms is already indicative of the substantive differences. English 

equivalents of the national institutions and names are terms like chattel 

or movable mortgage (England and Italy), non-possessory pledge, and 

security transfer of ownership. 

The broad variety of names for the various national institutions indicates 

considerable variations of the substantive rules. Consequently, the risk of 

non-recognition of these security rights upon transfer of the encumbered 

asset to another Member State is probably greater than for retentions of 

title. The only common denominator of the differing national institutions 

is their common economic objective. Non-possessory security rights other 

than retention of title typically serve as security for creditors other than 

sellers, especially of banks and other financial institutions. 

178 



www.sellier.de


8.3.2. Recognition in other Member States 

a) Denmark 

In what is apparently the only relevant case, a Danish appellate court 

recognized a German security transfer of ownership to a German bank. 39 

b) Germany 

Germany applies the liberality that it exercises in domestic law also to 

foreign non-possessory security devices that are unknown in Germany 

itself. In one case, the Federal Supreme Court recognized a French ‘gage 

automobile’, although this institution is unknown in Germany: A French 

lorry had been charged in France by such a non-possessory pledge as security 

for a bank-credit. When circulating in Germany, it was seized by 

a German creditor of the owner of the vehicle. In litigation about the 

validity of this seizure, the French creditor won. The Supreme Court 

equiparated the French security with a German security transfer of ownership 

and therefore permitted the French creditor to prevail over the 

subsequent German execution creditor. 40 The same result on the same 

grounds was reached in a comparable case: an Italian luxury car that had 

been encumbered in Italy by its owner with an ‘automobile mortgage’ was 

stolen there; the thief brought the car to Germany, where he sold it. The 

Italian bank’s action against the German buyer was successful; the Federal 

Supreme Court again used the analogy to a German security transfer of 

ownership. 41 It may be added that a German appellate court has also recognized 

a security interest in a motor car, created under Article 9 of the 

American UCC and duly noted on a certificate of registration. When the 

39 

A Dane living in Germany had financed the acquisition of a car by a security 

transfer of ownership of the car to the financing German bank. After his 

return to Denmark, the holder of the car sold it to a Danish firm that bought 

it in good faith. Nevertheless, the German bank succeeded in reclaiming ‘its’ 

car (ØL [appellate court] 27 March 1963, UfR 1963 A 704; extract in JDI 

(Clunet) 1965, 691). 

40 

Fed.S.Ct. 20 March 1963, BGHZ 39, 173, IPRspr. 1962 / 63 no. 60. 

41 

Fed.S.Ct. 11 March 1991, IPRspr. 1991 no. 71 (sub 1 d). The court denied the 

defendant’s defense of good faith, since buyers of foreign cars are expected to 

exercise particular circumspection. 


179 


www.sellier.de


car was later imported to Germany, the security interest was converted to 

a German security transfer of ownership. 42 

c) Netherlands 

The most interesting and important Dutch case decided in 2001 dealt 

with a Tanzanian floating charge. It is not as exotic as it may seem, because 

Tanzanian law in this respect, as in many others, is derived from 

English law, and this appears to be the only instance in which a typically 

English type of security has been brought before a Continental jurisdiction. 

The case is also remarkable because the amount in issue was over 2 

million US$ and it came twice before the Dutch Supreme Court; hence, 

it received a very thorough treatment. 43 

The somewhat complicated facts may be summarized as follows: Cardago, 

a Tanzanian exporter of sisal, had a line of credit of US$ 2 million with 

NBC, a Tanzanian bank, for the financing of its exports. As security, the 

exporter granted a floating charge over its present and future receivables. 

In 1994, Sisal, an Italian firm, arrested assets of Cardago amounting to 

US$ 2.5 million held by a Dutch bank in the Netherlands as security for 

claims that it asserted against Cardago; the Dutch bank held assets of 

some US$ 2.350 million for Cardago derived from documentary credits. 

When Cardago went into receivership in Tanzania in March 1995, all 

claims of its financer NBC fell due, and the floating charge ‘crystallized’ 

to a fixed charge, encumbering all existing receivables. In the procedure 

for distribution of the assets into which Sisal had brought execution in 

Rotterdam, the Tanzanian bank NBC applied for preferential satisfaction. 

The appellate court (Hof Amsterdam) granted that application on the 

grounds that a foreign security right must be recognized in the Netherlands, 

provided it pursues the same purpose as Dutch law and that in 

comparable circumstances it leads to the same result as Dutch law. The 

42 

Higher Appellate Court (OLG) Karlsruhe 6 July 2000, IPRspr. 2001 no. 52 

A. 

43 

S.Ct. 23 April 1999, N.J. 2000 no. 30 and S.Ct. 14 December 2001, N.J. 2002 

no. 241; cf. also Hof Amsterdam 21 December 2000, NIPR 2001 no. 109 (also 

reprinted in the second decision of the Supreme Court). 

180 



www.sellier.de


comparable Dutch institution is now – since the security transfer of 

owner ship is no longer allowed – the ‘stil pandrecht’: namely, the pledge 

of monetary claims without notice to the debtor. 44 The Supreme Court 

adopted the aforementioned formula of the appellate court in general, 

but specified that it must be applied with respect to the specific purpose 

that is in issue. 45 

In view of the long procedure and the intensive review, this decision carries 

a high degree of persuasion for the treatment of English-style floating 

charges by Continental courts. 

d) Sweden 

Two cases involving cars registered in Germany came before the Supreme 

Court. Both cars had served as security for loans to the holders of the cars; 

these had been transferred under a transfer of ownership for security to the 

financing banks. After the automobiles had been brought temporarily to 

Sweden, public authorities laid hands on them. In one case, the secured 

creditor successfully claimed its asset. 46 By contrast, the owner of the other 

car did not succeed, because the car had been misused for smuggling and 

selling drugs illegally in Sweden, and was confiscated. 47 

e) United Kingdom 

It can hardly come as a surprise that an Irish floating charge was recognized 

in England. 48 

44 

Hof Amsterdam (preceding n.) sub 4.20 of the court’s reasons. 

45 

No. 3.3 of the reasons. 

46 

S.Ct. 1 October 1984, NJA 1984 I no. 125. 

47 

S.Ct. 18 October 1968, NJA 1968 I no. 63 in the criminal proceedings, confirmed 

after sale of the car by S.Ct. 17 May 1972, NJA 1972 I no. 34. 

48 

Cretanor Maritime Co. Ltd. v. Irish Marine Management, Ltd. [1978] 1 W.L.R. 

966, 977 s. per Lord Buckley (C.A.). 


181 


www.sellier.de


8.3.3. Non-recognition in other Member States 

a) Austria 

Since Austrian legislation does not provide for any legal instrument allowing 

the creation of non-possessory security rights in movable assets, 

the courts originally used the same instrument as German court practice: 

namely, the security transfer of ownership without transfer of possession to 

the secured creditor. 49 However, later court practice in Austria retracted 

from this position and now also requires transfer of possession to the creditor. 

One consequence of this departure from the German model is regrettable. 

A security transfer of ownership of assets validly effected in Germany according 

to German law is not recognized in Austria if the ‘encumbered’ 

asset later passes to Austria. 50 

In one recent case, a German bank had accepted a security transfer of 

ownership in a lorry from the debtor living in Germany, who later moved 

to Austria, also moving the lorry there. When the debtor stopped payments, 

an insolvency proceeding was opened in Austria. The German 

bank attempted in vain to satisfy its claims involving the lorry. The Austrian 

debtor brought a claim for damages against the German bank, based 

upon the cost of proceedings and expenses spent on enforcing his rights 

in the car. The Austrian Supreme Court held the German bank liable in 

damages, since the bank had insisted on its ownership rather than seeking 

legal advice on Austrian law, and had thereby caused considerable 

expense to its Austrian customer. If the bank had investigated whether 

it was still security owner of the lorry under Austrian law, it would have 

learned its legal position and avoided considerable expenses involving 

competing Austrian creditors. The German bank was held liable for a 

high amount of damages. 51 

49 

Cf. supra 8.3.2 b). 

50 

S.Ct. 14 December 1983, J.Bl. 1984, 550; First Inst. Appellate Court (LG) 

Linz 27 May 1986, IPRE 2 (1983-1987) no. 115. 

51 

S.Ct. 28 March 2002, ÖBA 2002, 937 (approving note Koziol). 

182 



www.sellier.de


b) France 

One French case dealt with an involved arrangement between two Dutch 

companies and a French company: The basic transaction was, in essence, 

a purely Dutch security transfer of ownership – hidden under a contract 

of sale of equipment to a bank combined with a rehire-purchase contract 

to the seller – between two Dutch companies. The Dutch hire-purchaser 

brought the material to France, where a French creditor of that company 

sought to bring execution into the material. The Dutch owner’s attempt 

to oppose that execution failed: the Dutch contract was analysed as being 

in reality a non-possessory security of a type unknown to French law and 

therefore void. 52 

8.3.4. Conclusion 

Despite the broad variety of security rights other than retention of title, 

to be found in the Member States, the courts are willing to recognize 

and adapt these security rights. This is remarkable for two reasons: first, 

the European legislator does not apply any pressure in this field; second, 

the transformation of foreign security rights of a nature unknown to the 

respective lex fori and their translation into and adaptation to the latter 

appear to be difficult. However, this burden of translation into and 

adaptation to the lex fori is facilitated by the fact that most legal systems 

dispose of only one major and basic non-possessory security device other 

than retention of title. 

8.4. Results 

A. An exporter’s retention of title in goods for exportation is, as a rule, 

subject to the law of the country of importation. 

B. For all other cases of retention of title – except with regard to automobiles, 

cf. infra D – the two-step procedure of recognition is necessary: 

first, it has to be ascertained whether an effective retention of title had 

been created in the Member State of exportation; second, if there is a 

52 

Cass. 3 May 1973, Rev.crit.d.i.p. 1974, 100 with note Mezger, J.D.I. (Clunet) 

1975, 74 with note Fouchard. 


183 


www.sellier.de


valid retention of title according to A, the Member State of importation 

is obliged to recognize the retention of title, but may and usually will apply 

its law to the effects of the retention of title. 

C. In essence, the same rules as under A and B apply to security rights 

in movable goods other than retention of title. This holds true, although 

there is no binding obligation under Community law. However, existing 

practice may be based upon the generally recognized basic principle of lex 

rei sitae and the legal consequences of a change of the situs. 

D. In my view, the preceding three principles should not apply to security 

rights of any kind created in a motor vehicle registered in another Member 

State. Looking at the factual objects of the decisions reviewed above, 

one sees that in almost a dozen cases security rights created in the state of 

registration of a vehicle were involved. However, this fact was disregarded; 

instead, the law of the present physical location of the car was applied: 

namely, the law of a station in a mere transit. 53 Nevertheless, this exception 

should be construed narrowly. It would not apply, for instance, to the 

importation of a car 54 or to the frequent instances of theft if the thieves 

moved the stolen car to another country. 55 


In summary, the accessible judicial practice in the ‘old’ Member States in 

Central, Northern, and Southern Europe does not disclose the necessity 

of an intervention by the European legislator – except for one narrow 

fact pattern: security rights in motor vehicles in transit should as a rule be 

subject to the law of the country in which the vehicle is registered. 

Whether the preceding conclusion also holds true for the new Member 

States in Central, Eastern, and South-Eastern Europe would require separate 

investigation. Since one cannot yet expect legal certainty from court 

practice in these countries, specific legislation in the field of the conflict 

of laws may be necessary. 

53 

Cf. supra n. 36 (four cases), 39-42, 46, 47, and 51. 

54 

Cf. supra at n. 42. 

55 

Cf. supra at n. 41. 

184 



www.sellier.de


8.6. Postscript 

Directive 2000 / 35 / EC of 29 June 2000 has just been replaced by a new 

version: Directive 2011 / 7 / EU of 16 February 2011 on combating late 

payment in commercial transactions (O.J.EU. 2011 L 48, p. 1) As far as 

the recognition of retentions of title created in another Member State is 

concerned, all relevant texts of the former Directive have been retained. 

Merely the numbering of the provisions has been changed: The definition 

of retention of title is now contained in art. 2 no. 9; the rule on recognition 

is now contained in art. 9 para. (1); and the explanation is to be 

found in no. 31 of the considerations. 


185 


www.sellier.de


www.sellier.de

D. 

Assignment in Private International Law; 

Financial Instruments / the 

Collateral Directive; Insolvency Law 


www.sellier.de


www.sellier.de

9. 

Party Autonomy and Assignment 

Hendrik Verhagen * 

9.1. Advantages of party autonomy 

The principle of party autonomy is referred to in the Rome I Regulation as 

‘one of the cornerstones’ of the system of European private international 

law in contractual matters. 1 However, it is not always remembered that, 

until the second half of the 20 th century, recognising party autonomy 

in the field of contracts was not such a self-evident concept. For how 

could the parties, merely by their mutual consent, set aside the mandatory 

rules of the law that would otherwise have been applicable? 2 Moreover, 

would not allowing parties to choose the law governing their contractual 

relationship certainly lead to abuse of this freedom? In the last decades 

these objections have been overcome. It is now accepted wisdom that 

* 

Rick Verhagen is professor of private international law, comparative law and 

civil law at Radboud University, Nijmegen, and attorney at Clifford Chance, 

Amsterdam. 

1 

Regulation (EC) No 593 / 2008 on the law applicable to contractual obligations 

(Rome I), OJ 2008, L 177 / 6, recital (12). I have already expressed my 

views in several other publications. See in particular A. Flessner & H. Verhagen, 

Assigment in European Private International Law, München: Sellier. 

European Law Publishers 2006; H.L.E. Verhagen & S. van Dongen, Cross-Border 

Assignments under Rome I, Journal of Private International Law (JPIL) 6 

(2010), p. 1-21. 

2 

To mention but one example from Dutch case law: in the 1960‘s the Hague 

Court of Appeal ruled that the parties‘ choice of the applicable law could only 

derogate from the non-mandatory rules of the law that would have been applicable 

in the absence of the choice and that therefore the mandatory rules 

of that law continued to be applicable to the contract in question. In the 

Alnati-decision the judgment of the court of appeal was annulled by the Hoge 

Raad, which ruled that the chosen law exclusively governed the contract, to 

the exclusion of both the non-mandatory and the mandatory rules of the law 

that would otherwise have been applicable. HR 13 May 1966, Nederlandse 

Jurisprudentie (NJ) 1967, 3 (Alnati), Rev. Crit. 1967, p. 522. 

189 


www.sellier.de

D. Assignment; Financial Instruments; Insolvency Law 

allowing party autonomy is beneficial for cross-border transactions, that 

cases of abuse of this freedom are extremely rare and, even if they do occur, 

that they can be adequately remedied by using existing techniques 

(ordre public, overriding mandatory provisions, consumer protection etc.). 

Fairly recently, the principle of party autonomy has been extended by the 

European legislator to tort, unjust enrichment and other non-contractual 

obligations (Article 14 of the Rome II Regulation) 3 and even to family law 

(alimony). 4 But once we step outside the realm of the law of obligations, 

party autonomy is thought to remain problematic. The same objections 

formerly forwarded against party autonomy for contracts again raise their 

heads. In the case of property law in particular, party autonomy is still 

regarded by many as anathema, due to the belief that it would enable the 

parties to choose a particular law in order to adversely affect the rights 

of third parties. 5 At the conference, one of the speakers called this the 

‘knockout argument’ (Totschlagargument): and many believe rightly so. 

But only after the relevant third parties have been identified and it has 

been carefully examined how their interests are precisely affected, is it 

possible to assess the true value of this criticism. 6 

3 

Regulation (EC) No 864 / 2007 on the law applicable to non-contractual obligations 

(Rome II), OJ 2007, L 199 / 40. 

4 

Article 15 of Regulation (EC) No 4 / 2009 on jurisdiction, applicable law, 

recognition and enforcement of decisions and cooperation in matters relating 

to maintenance obligations OJ 2009, L 7 / 1, referring to Article 7 and 8 of the 

Hague Protocol of 2007. 

5 

E-M. Kieninger, Brussels I, Rome I and Questions Relating to Assignment and 

Subrogation, in: J. Meeusen et al (eds.), Enforcement of international contracts 

in the European Union, Convergence and divergence between Brussels 

I and Rome I, Antwerpen: Intersentia 2004, p. 378-379; Max-Planck-Institut 

für ausländisches und internationales Privatrecht, Comments on the European 

Commission’s Green Paper on the conversion of the Rome Convention of 

1980 on the law applicable to contractual obligations into a Community instrument 

and its modernization, RabelsZ 2004, para. 10.1.3; T.H.D. Struycken, 

The proprietary aspects of international assignment of debts and the Rome 

Convention, Article 12, Lloyd’s Maritime and Commercial Law Quarterly 

1998, p. 354-356; P. Lagarde, Retour sur la loi applicable à l’opposabilité des 

transferts conventionnels de créances, Mélanges Jacques Béguin, Paris: Litec 

2005, p. 425. 

6 

See also Flessner‘s contribution to this book (Chapter 10) and Axel Flessner, 

Rechtswahlfreiheit auf Probe – Zur Überprüfung von Article 14 der Rom I- 

190 



www.sellier.de

9. Party Autonomy and Assignment 

Not everyone is of the view that party autonomy and assignment are irreconcilable 

notions. There is even a decision from an EU Member State 

supreme court concerning assignment in which party autonomy played a 

leading role. In the famous – according to some, infamous – Hansa-case, 7 

party autonomy was used by the Dutch Hoge Raad as a powerful argument 

for finding that the governing law of the assignment in that case should be 

referred to Article 12(1) of the Rome Convention (being the law governing 

the contract between assignor and assignee) rather than Article 12(2) 

of the Rome Convention (being the law governing the claim assigned). 8 

The Hoge Raad indicated that to hold otherwise would be to go against 

a trend in favour of the principle of party autonomy, as can be discerned 

in Article 3 of the Rome Convention and in other areas of private international 

law, such as various Hague Conventions and, for assignment in 

particular, Article 145(1) of the Swiss IPRG. 9 

A solution based upon party autonomy has considerable practical advantages, 

which are well-known and need not be repeated here. 10 What does 

need to be repeated is that in practice there are always good and honest 

reasons for the parties choosing a particular law and cases of manipulation 

are extremely rare, if they exist at all. The fear that party autonomy 

would be abused by the contracting parties, which still seems to pervade 

the views of several contemporary authors, in the past also existed in 

respect of contracts, but proved to be unfounded there. The same can be 

established in respect of party autonomy and assignment: in the Nether- 

Verordnung, in: J.F. Baur et al. (eds.), Festschrift für Gunther Kühne, Frankfurt 

am Main: Verlag Recht und Wirtschaft 2009, p. 703-715. 

7 

HR 16 May 1997, Nederlandse Jurisprudentie (NJ) 1998, 585 (Brandsma/ 

Hansa Chemie AG). Case notes in English by: M.E. Koppenol-Laforce, 

Nether lands International Law Review (NILR) 1998 129-137; C. Joustra, 

IPRax 1999, p. 280-284. 

8 

Rome Convention of 19 June 1980 on the law applicable to contractual obligations 

(80/ 934 / EEC), OJ L 266 , 09 / 10 / 1980 p. 1-19, consolidated version 

30 December 2005, OJ 2005, C 334 / 3. Article 12 of the Rome Convention is 

the predecessor of Article 14 of the Rome I Regulation. 

9 

Bundesgesetz über das Internationale Privatrecht vom 18. Dezember 1987 

(IPRG). 

10 

See A. Flessner & H. Verhagen, Assignment in European Private International 

Law, 2006, p. 67-70; H.L.E. Verhagen & S. van Dongen, Cross-Border Assignments 

under Rome I, Journal of Private International Law (JPIL) 6 (2010), 

p. 17-19. 


191 


www.sellier.de


lands we have more than 13 years of experience with the Hansa-case and 

there are no indications at all that this decision has led to the frustration 

of creditors or other unfair results. 

9.2. Party autonomy and property law 

Rules of private international law (conflict rules) often reflect values of 

substantive law. This has never been more powerfully expressed than in 

Von Savigny’s famous statement that private international law’s primary 

task is to ensure that a legal relationship will be referred to that legal 

system to which it – given its distinctive legal nature – belongs: 

‘dasjenige Rechtsgebiet …, welchem dieses Rechtsverhältniss seiner eigenthümlichen 

Natur nach angehört’. 11 

We can see this principle reflected clearly, in the Rome I Regulation. Freedom 

of contract in substantive law here finds its parallel in the freedom 

to choose the applicable law: Article 3 Rome I. Where in substantive law 

freedom of contract is curtailed in order to protect structurally weaker 

parties, this finds its parallel in the specific conflict rules for consumer contracts 

and employment relationships, pursuant to which party autonomy 

often cannot deprive consumers or employees of the protection offered by 

the law of their socio-economic environment (Articles 6 and 8 Rome I). 

What are the issues in the substantive laws on assignment, that could 

have an impact on a conflict rule for assignment? Broadly, if we confine 

ourselves to the laws of the Member States, we can see that there are three 

important issues on which the laws of the Member States diverge. Firstly, 

in some Member States notification to the debtor of the assigned receivable 

not only serves to inform that debtor that assignment has taken place, 

but importantly is also an essential requirement for the assignment itself. 

Secondly, in some jurisdictions the protection of bona fide assignees against 

an earlier assignment by the same assignor is based on notification to the 

debtor. Thirdly, the laws of some Member States do not allow security 

assignments: security over claims must be created in the form of a charge 

(pledge). In addition there is the requirement, existing in several Member 

11 

F.C. von Savigny, System des heutigen Römischen Rechts, Vol. 8, 1849, 

p. 28. 

192 



www.sellier.de


States (e.g. United Kingdom), that assignments, or certain (security) assignments, 

must be filed in a public register. This issue of public filing will 

be discussed later. In this part of my contribution I shall examine whether 

the first three substantive issues need to have an impact on the treatment 

of assignment at the level of private international law, more in particular 

on the desirability of allowing party autonomy in this area. 

9.2.1. Notification 

At the outset, I would like to emphasise that as far as notification serves to 

inform the debtor of the assigned claim or has an impact on his ability to 

invoke set-off and other defenses against the assignee, it is widely accepted 

that the debtor should be able to rely on the proper law of the assigned 

claim. This is reflected in Article 14(2) Rome I, which provides that 

the law governing the assigned claim determines the conditions under 

which the assignment can be invoked against the debtor and whether the 

debtor’s obligations have been discharged. The assigned debtor is therefore 

adequately protected under the present version of Article 14 Rome I 

and need not further concern us here. 

Let us now turn to the function of notification in the law of property. 

While in many jurisdictions (e.g. Germany: § 398 BGB) notification to 

the debtor is not required in order to transfer a claim by way of assignment, 

this is not the case for every jurisdiction. Thus Article 1690 of the 

French Code civil (Cc) requires that the assignor or the assignee formally 

notify the debtor of the assignment via a bailiff (huissier). 12 Unless this 

condition is fulfilled, the assignment cannot be invoked against third parties 

(tiers). 13 The word ‘tiers’ in Article 1690 Cc includes not only the 

debtor, but also any interested third party, such as other creditors of the 

assignor or subsequent assignees. However, the substantive laws of many 

European jurisdictions (including France) have been changed in the last 

decades, in order to adapt to the new commercial reality that claims are 

assigned on a large scale, within the framework of transactions such as factoring 

and securitisation. In jurisdictions where notification to the debtor 

12 

Alternatively the assignment is acknowledged by the assigned debtor by way 

of an ‘authentic’ deed. 

13 

P. Malaurie, L. Aynès & P. Stoffel-Munck, Les obligations, Paris: Lextenso 

éditions 2007, p. 778-779. 


193 


www.sellier.de


is (or was) still an essential requirement for the transfer of receivables, this 

requirement has been abolished for certain types of transaction. 14 

In the ‘Low Countries’ the rules on assignment of the Civil Code have 

even been fundamentally revised, in order to facilitate factoring and securitisation 

transactions and other transactions involving ‘bulk assignments’. 

Thus Article 1690 of the Belgian Civil Code has been amended 

to the effect that the conclusion of the agreement to assign makes the 

assignment enforceable against third parties (including the assignor’s 

creditors). After its amendment, the Dutch Civil Code now recognises 

two forms of assignment: disclosed (being an assignment notified to the 

debtor) and undisclosed. An undisclosed assignment requires either an 

authentic deed or a document to be registered (in a non-public register) 

with the tax authorities. But even in the Code civil things may be different 

in the future. In the Avant-projet Catala the rule embodied in Article 1690 

has also been abandoned: the proposed Article 1254 Cc provides that the 

execution of a deed suffices in order for the assignment to be enforceable 

against third parties. 15 In current Belgian and Dutch law and, if the projet 

Catala were to be enacted, also in French law, notification only serves to 

inform the debtor of the assigned claim. 

It follows that at the level of substantive law the assignment is ‘consensualised’: 

the assignor and the assignee can among themselves decide when 

a claim transfers. The logical consequence of this for private international 

law would be that here also the assignment would be ‘consensualised’. 

How? By allowing the assignor and assignee to designate the governing 

law themselves. 

Although there are some jurisdictions where notification is an essential 

requirement for the transfer of the claim, this does not mean that party 

14 

See Article L214-43, L313-27 and L515-21 of the French Monetary and Financial 

Code (Code monétaire et financier). See also P. Malaurie, L. Aynès & 

P. Stoffel-Munck, Les obligations, 2007, p. 786-787; A. Flessner & H. Verhagen, 

Assignment in European Private International Law, 2006, p. 14-15; H.L.E. 

Verhagen & S. van Dongen, Cross-Border Assignments under Rome I, Journal 

of Private International Law (JPIL) 6 (2010), p. 7. 

15 

L‘avant-projet de réforme du droit des obligations (articles 1101 à 1386 du 

Code civil) et du droit de la prescription (articles 2234 à 2281 du Code civil), 

http: // www.lexisnexis.fr / pdf / DO / RAPPORTCATALA.pdf (accessed 7 October 

2010). 

194 



www.sellier.de


autonomy is a defunct concept in such situations. The idea behind making 

notification a requirement for the transfer of a claim is that this creates 

an element of publicity, so that interested persons can be aware of the 

assignment. Notification, however, is an extremely ineffective device for 

creating publicity, not only in purely domestic cases, but even more so 

in cross-border assignments, in particular in circumstances where the assigned 

debtors reside in different countries to the assignor. For, how can, 

other than in rare cases, 16 notification to assigned debtors resident in different 

countries to the assignor create adequate publicity for the creditors 

of the assignor? If anything, the notification requirement would constitute 

an argument in favour of the law of the assigned debtor’s residence as the 

governing law of the assignment, but this conflict rule must be rejected 

as being unsuited for many assignment-based cross-border transactions. 

It would entail that a bulk assignment of a company’s international receivables 

portfolio would have to comply with as many substantive laws 

as there are countries to which the company distributes its products. 17 

9.2.2. Protection of bona fide assignees 

Although notification is not required for a valid and effective assignment 

in some European legal systems (e.g. Belgian law, English law) , it does fulfill 

a function for the protection of assignees against earlier or subsequent 

assignments: the assignee who is the first to give notice has priority. 18 This 

rule can also be found in Article 11:401(1) PECL and Article III.– 5:121 

of the Draft Common Frame of Reference. 19 

16 

In particular cases in which the majority of the assignor’s debtors reside in the 

same country as the assignor’s creditors, this country not being the assignor’s 

country of residence. 

17 

In the same vein, Mance L.J. in Raiffeisen Zentralbank Österreich AG v. Five 

Star General Trading LLC [2001] 3 All ER 257, para. 38. This criticism is also 

endorsed in recent French legal writing (where this conflict rule originally 

attracted much support), see e.g. P. Lagarde, Retour sur la loi applicable à 

l’opposabilité des transferts conventionnels de créances, Mélanges Jacques 

Béguin, 2005, p. 425. 

18 

Dearle v Hall (1828) Russ 1, 38 ER 475; Article 1690 Belgian Cc. 

19 

Cf. C. von Bar, E. Clive & H. Schulte-Nölke (eds.), Principles, Definitions and 

Model Rules of European Private Law: Draft Common Frame of Reference 

(DCFR), Outline edition, München: Sellier. European law publishers 2009, 

p. 265. 


195 


www.sellier.de


Under a conflict rule based upon party autonomy, subsequent assignees 

are adequately protected when assignment is referred to the law chosen 

by the parties. They can determine in accordance with the law they have 

selected themselves whether they would be protected against an earlier assignment 

and whether such protection requires notification of the assignment. 

In transactions where there are multiple assignments of the same 

claim and the subsequent assignment is cross-border in nature, it must be 

admitted, that the cross-border character of the second assignment and 

the resulting application of some other law may sometimes come as a disappointment 

to the first assignee, particularly where the law governing the 

second assignment would give priority to a subsequent assignment (e.g. 

English law), while the law under which he acquired the claim would not 

protect a second assignee (e.g. German law). This is, however, part of the 

intrinsic risk of conducting business in a world of different laws, and does 

not really increase the basic risk of there being competing assignments, 

particularly as the law governing the subsequent assignment has to start 

with the results achieved under the previous assignment and its law (see 

below). Moreover, if the second assignee, aware of an earlier assignment, 

deliberately chooses a legal system protecting a second assignee, his lack 

of good faith may be fatal. 20 

The use of a conflict rule based upon party autonomy has been said to 

lead to a deadlock situation in the case where a receivable has been assigned 

multiple times and each assignment is governed by a different law. 

As a result of this a priority conflict is said to arise between the assignees, 

for which a conflict rule referring assignment to the proper law of the 

underlying contract between the assignor and the assignee would provide 

no solution. 21 As has been stressed before, this point of criticism is 

clearly wrong. 22 The proper law of the subsequent assignment would not 

20 

Article 1690 Belgian Cc. 

21 

See e.g., V. Sagaert, De zakenrechtelijke werking van de cessie: de nieuwe iprregeling 

na de wet van 2 augustus 2002, Tijdschrift voor Privaatrecht (TPR) 

40 (2003), p. 561, 580; H.C. Sigman & E-M. Kieninger, The Law of Assignment 

of Receivables: In Flux, Still Uncertain, Still Non-Uniform, in: H.C. 

Sigman & E-M. Kieninger (eds.), Cross-border Security over Receivables, 

München: Sellier. European law publishers 2009, p. 61. 

22 

See A. Flessner & H. Verhagen, Assignment in European Private International 

Law, 2006, p. 32-36; H.L.E. Verhagen & S. van Dongen, Cross-Border Assignments 

under Rome I, Journal of Private International Law (JPIL) 6 (2010), 

p. 17-18. See further A. Flessner, Rechtswahlfreiheit auf Probe – zur Überprü- 

196 



www.sellier.de


only govern the intrinsic validity of that assignment, but also determine 

whether the subsequent assignee would be protected from the earlier assignment. 

The principle is exactly the same as with tangible property, 

where ownership acquired by one person under the original lex situs may 

be defeated by the rights acquired by another person under the rules on 

bona fide purchases of the new lex situs. 

9.2.3. Prohibition of security assignments 

Some legal systems allow security over claims to take the form of a security 

assignment (e.g. English law and German law), while other legal systems 

force the parties to use a pledge, by outlawing security assignments (e.g. 

Dutch law). 

One of the characteristics of European systems of private international 

law is that, by operating with multilateral conflict rules, they demonstrate 

a large openness towards foreign laws. A fundamental principle is 

that legal systems are equivalent and that in international transactions 

there may be good reasons to apply the laws of other jurisdictions, even 

if these laws are different from the laws of the forum state. I have not 

been able to identify a compelling reason why this should be different 

here. The assigned claims are not the assignor’s creditors’ property. In a 

purely domestic transaction it is up to the assignor and the assignee (and 

not their creditors) whether and under what conditions security will be 

created. In an international transaction the parties should be able to create 

security over intangible assets in accordance with the law which 

they consider to fit for their needs, whether by way of assignment or by 

way of charge. 

According to Kieninger, the Hansa-case is a good example of allowing 

a contract to the detriment of a third party, in this case the assignor’s 

unsecured creditors. For, by referring the assignment to the proper law of 

the underlying contract to assign (German law) the Hoge Raad allowed 

receivables to be encumbered in a manner (i.e. by a security assignment) 

fung von Article 14 der Rom I-Verordnung, in: J.F. Baur et al (eds.), Festschrift 

für Gunther Kühne zum 70. Geburtstag, 2009, p. 706-715, who explains various 

cases of multiple assignments. See also Einsele, Die Forderungsabtretung 

nach der Rom I-Verordnung – Sind ergänzende Regelungen zur Drittwirksamkeit 

und Priorität zu empfehlen?, RabelsZ 74 (2010), p. 114-115. 


197 


www.sellier.de


which would not have been possible under Dutch law (which prescribes a 

pledge). 23 At first sight this might seem a convincing argument. However, 

if we look more closely at the Hansa-case, a different picture emerges. 

In the transaction at issue in the Hansa-case, virtually the same result 

as reached under German law could have been reached under Dutch 

law, by creating an undisclosed pledge over the debtor’s claims against 

his customers. The only condition would be that such a pledge needs 

to be registered (in a non-public register) and can only be created over 

claims arising from existing contracts. However, the so-called ‘abridged 

registration procedure’ commonly used in Dutch practice makes it very 

easy to ensure that the secured creditor also acquires a security interest 

on claims arising under contracts which have been entered into after 

the registration of the original deed of pledge. The parties periodically 

register additional short form deeds of pledge containing a generic description 

of the recently pledged receivables (e.g. “all receivables arising 

under contracts entered into by the pledgor in Month [x]”), with a 

cross-reference to the terms and conditions of the original ‘Master Deed 

of Pledge’. Another interesting case in this respect is the Sisal-case. 24 In 

this case the Dutch Hoge Raad upheld a decision by the court of appeal 

in which the holder of a floating charge created in 1967(!) under the 

law of Tanzania, over all present and future claims of the chargor and 

extending to a claim arising under a contract entered into in 1994, was 

granted the same rights of enforcement as the holder of a Dutch right of 

pledge. In fact, one of the reasons why the court of appeal in this procedure 

allowed assimilation of this floating charge, was that virtually the 

same results could be achieved under Dutch law, by using the abridged 

registration procedure. 

The Sisal-case raises another issue. Of course it is true that where in a 

particular jurisdiction security assignments are not recognised, it is also 

23 

E-M. Kieninger, Brussels I, Rome I and Questions Relating to Assignment 

and Subrogation, in: J. Meeusen et al (eds.), Enforcement of international 

contracts in the European Union, Convergence and divergence between Brussels 

I and Rome I, 2004, p. 378-379; See also E-M. Kieninger, Das Statut der 

Forderungsabtretung im Verhältnis zu Dritten, RabelsZ 62 (1998), p. 693-695; 

G. Mäsch, Abtretung und Legalzession im Europäischen Kollisionsrecht, in: 

S. Leible (ed.), Das Grünbuch zum internationalen Vertragsrecht, München: 

Sellier. European law publishers 2004, p. 199. 

24 

HR 14 December 2001, Nederlandse Jurisprudentie (NJ) 2002, 241 (Societa 

Italiana e Affini Lavorata S.p.a. / NBC Holding Corporation). 

198 



www.sellier.de


possible that the insolvency legislation of that jurisdiction has no provisions 

relating to this type of security. Does this mean that a creditor to 

whom claims have been assigned by way of security would not be bound 

by insolvency rules curtailing the enforcement of security over claims 

(e.g. a statutory freeze), because the only type of security contemplated 

by these rules is a pledge or charge over receivables? It does not. This is a 

classic case of assimilation: (insolvency) rules designed for domestic legal 

issues are applied by way of analogy to foreign legal institutions which 

are functionally equivalent. This is precisely what the Hoge Raad ruled in 

the Sisal-case. In the event that a creditor seeks to enforce its rights as a 

security assignee in the Netherlands, the relevant rules designed for pledges 

of the Dutch enforcement and insolvency laws will be applied, provided 

that for the purposes of the Dutch provisions in question, the foreign 

security right can as regards its content and purpose be treated equally as 

the related Dutch security right. 25 On this basis, the holder of a common 

law floating charge was allowed to request a so-called ranking order from 

the court in the same manner as a pledgee under Dutch law would be 

able to do. In this case, the assimilation process worked for the benefit of 

the holder of a foreign security interest. However, it can also restrict the 

enforcement of foreign security. For instance, article 63a of the Dutch 

Bankruptcy Act, which imposes a statutory freeze on the enforcement 

of pledges, should also be applied to foreign security assignments. In other 

words, there are sufficient safeguards in the law of insolvency in order to 

prevent the parties from evading its restrictions by choosing a different 

property regime. 26 

9.3. The effectiveness of an assignment against 

third parties and priority 

The conclusion which can be drawn from the foregoing observations 

is that from the perspective of property law the principle of party autonomy 

can be recognised. This is not to deny, however, that there are 

persons other than the assignor and the assignee which have legitimate 

interests in respect of claims which the law needs to take into account. 

Sometimes it is the law of property itself which does so, as when bona 

25 

A. Flessner & H. Verhagen, Assignment in European Private International 

Law, 2006, p. 40-41. 

26 

See Michael Veder‘s contribution to this book (Chapter 12). 


199 


www.sellier.de


fide assignees are protected against an earlier assignment. In most cases, 

however, the interests of other persons are addressed outside the law of 

property, namely by laws on insolvency, voidable preference (pauliana), 

attachment by individual creditors and sometimes tort. At the level of 

private international law these laws have their counterpart in special 

conflict rules for insolvency, voidable preference, attachment and tort. 

Perhaps the main reason why so much disagreement and confusion exists 

on assignment in private international law is that the relevance of these 

special conflict rules is not always fully recognised. This in turn may 

be caused by the fact that expressions like ‘third party effects’, ‘opposability’, 

‘the effectiveness of an assignment against third parties’ and 

‘priority’ are too general to provide clear guidance for finding the appropriate 

conflict rules. These expressions should be broken down to more 

precise categories: it then becomes apparent that many if not all of the 

objections against allowing party autonomy for the proprietary aspects 

of assignment disappear. I will illustrate this with reference to Article 

27(2) Rome I. 

9.3.1. The effectiveness of an assignment of a 

claim against third parties 

In Article 14(1) Rome I the words ‘mutual obligations’ of Article 12(1) 

Rome Convention have been replaced by the word ‘relationship’. According 

to recital 38 of the preamble to Rome I, this should clarify that 

‘article 14(1) also applies to the property aspects of an assignment, as between 

assignor and assignee (…)’ (emphasis added). Article 27(2) Rome 

I provides that the European Commission shall submit ‘a report on the 

question of the effectiveness of an assignment or subrogation of a claim 

against third parties (…)’. In an article, Sanne van Dongen and I have 

submitted an approach to the issue which Article 27(2) Rome I seeks to 

address, which would remove the apparent incongruity of Article 14(1) 

and Article 27(2) Rome I. In our view, the distinction that should be 

made is not between the inter partes property effects (whatever that may 

mean) and the erga omnes property effects. The appropriate distinction 

is rather between the ‘property aspects’ of an assignment and the ‘other 

third party effects’ thereof. The property aspects of assignment essentially 

concern the requirements for a valid transfer of the assigned claim from 

the patrimony of the assignor to that of the assignee and the ranking of 

competing property interests in respect of the same claim. If Article 14(1) 

Rome I were to be interpreted in this way, it would entirely apply to these 

200 



www.sellier.de


property aspects of assignment. 27 The effectiveness of an assignment against 

third parties concerns rules pursuant to which a transfer, which – under the 

property law designated by Article 14(1) – is intrinsically valid, cannot 

be objected to by certain (or all) creditors. 28 

In many jurisdictions rules on voidable preference (pauliana) or the attachment 

of claims by individual creditors have this effect. For instance, 

as a consequence of Article 475h of the Dutch Code of Civil Procedure, 

an assignment of a claim cannot be invoked against a creditor who has 

previously levied an attachment on this claim. Although as a matter of 

property law a transfer of the claim from the assignor to the assignee has 

taken place, a creditor of the assignor who has levied the attachment can 

treat the claim as if it still belongs to the assignor. A similar effect takes 

place where an individual creditor is able to attack an assignment as constituting 

a voidable preference (actio pauliana). As a matter of property 

law the claim would still belong to the assignee. However, this creditor 

can take recourse against the claim, as if it was still owned by his debtor 

(the assignor). 

These rules concerning attachments and voidable preferences, although 

directly affecting the legal consequences of an assignment, are clearly not 

within the scope of Article 14(1) Rome I. As a consequence, although a 

claim may have transferred under the chosen (property) law applicable 

pursuant to Article 14(1) Rome I, the question of whether certain creditors 

can ignore this transfer, may be governed by a different law. 29 

27 

See also A. Flessner, Die internationale Forderungsabtretung nach der Rom I 

Verordnung, IPRax 29 (2009), p. 38-43; A. Flessner, Rechtswahlfreiheit auf 

Probe – zur Überprüfung von Art 14 der Rom I-Verordnung, in: J.F. Baur et 

al (eds.), Festschrift für Gunther Kühne zum 70. Geburtstag, 2009, p. 703- 

704; D. Einsele, Die Forderungsabtretung nach der Rom I-Verordnung – Sind 

ergänzende Regelungen zur Drittwirksamkeit und Priorität zu empfehlen?, 

RabelsZ 74 (2010), p. 96; C. Reithmann & D. Martiny, Internationales Vertragsrecht, 

Köln: Verlag Dr. Otto Schmidt 2010, p. 295. 

28 

See also H.L.E. Verhagen & S. van Dongen, Cross-Border Assignments under 

Rome I, Journal of Private International Law (JPIL) 6 (2010), p. 6-13. 

29 

See also H.L.E. Verhagen & S. van Dongen, Cross-Border Assignments under 

Rome I, Journal of Private International Law (JPIL) 6 (2010), p. 10-11. On 

voidable preference under the EU Insolvency Regulation, see Michael Veder’s 

contribution to this book (Chapter 12). 


201 


www.sellier.de


This also applies to public filing. In England, certain security assignments 

have to be filed in the appropriate public registers in order for the security 

to be effective against subsequent transferees (including assignees), 

chargees and garnishees, and in insolvency proceedings commenced in 

respect of the company granting the security. Public filing is subject to its 

own conflict rules. 30 As far as public filing affects the ability of a secured 

creditor to invoke a security assignment against the insolvent estate of 

an assignor having its centre of main interests in an EU Member State, it 

is submitted that this is a matter within the scope of the EU Insolvency 

Regulation. 31 It is also submitted that, as a consequence of Article 2(g) 

and 5 thereof, the effects of (lack of) public filing are limited to assigned 

claims against debtors having their centre of main interests (COMI) in 

the same EU Member State as the insolvent assignor. A fortiori, the effects 

of (lack of) public filing in respect of subsequent assignees, chargees and 

garnishees should also be limited to ‘domestic’ claims. 32 

9.3.2. Priority 

Article 27(2) Rome I provides that the European Commission shall submit 

‘a report on (…) the priority of the assigned or subrogated claim over 

a right of another person’. Firstly, this provision demonstrates how sloppy 

the drafting of EU legislation can be. All of the different language versions 

that I have reviewed refer to the priority attached to the assigned claim 

itself. 33 If this is really what the report should advise upon, then this part of 

the report would be rather brief, certainly where priority in insolvency is 

concerned. For, Article 4(2)(i) of the Insolvency Regulation already pro- 

30 

See also Einsele, Die Forderungsabtretung nach der Rom I-Verordnung – Sind 

ergänzende Regelungen zur Drittwirksamkeit und Priorität zu empfehlen?, 

RabelsZ 74 (2010), p. 109-113, who argues that such registration requirements 

could be regarded as ‘overriding mandatory provisions’ (Article 9 Rome I). 

31 

Regulation (EC) No 1346 / 2000 on insolvency proceedings, OJ 2000, L 

160 / 1. 

32 

See for public filing also Flessner’s contribution to this book (Chapter 10). 

33 

‘(…) rang de la créance faisant l‘objet de ladite cession ou subrogation par 

rapport aux droits détenus par d‘autres personnes’; ‘(…) privilegio del credito 

ceduto o surrogato rispetto al diritto di un’altra persona’; ‘(…) den Rang dieser 

Forderung gegenüber einem Recht einer anderen Person’; ‘(…) la prioridad 

del derecho cedido o subrogado sobre un derecho de otra persona’; ‘(…) prioridade 

do crédito cedido ou sub-rogado sobre um direito de outra pessoa’. 

202 



www.sellier.de


vides that the law of the State of opening of insolvency proceedings shall 

govern ‘the ranking of claims’. However, the ranking of a claim which has 

been assigned is usually not different from other claims: the transfer of a 

claim normally does not change its substance and identity. In other words, 

we are not dealing with an issue which is specific to assignment, so it must 

be assumed that ‘priority’ in Article 27(2) has an alternate meaning. 

In respect of the assignment and charge of claims, there are two types of 

‘priority’: (i) the question whose right in rem prevails in case of multiple 

dispositions (assignments, charges) of a claim and (ii) the distribution of 

the proceeds of a claim where several creditors have an interest in that 

claim. In my view, the first type of priority, which one could call “proprietary 

priority”, is currently governed by Article 14(1) Rome I. Only 

the second type of priority, which one could call “distributive priority” is 

outside the scope of Article 14 Rome I. Thus questions such as whether 

and how a claim can be attached by an individual creditor, to whom the 

debtor of the claim must pay in case of attachment and how the proceeds 

of this claim should be distributed among competing claimants, are outside 

the scope of a proprietary conflict rule such as Article 14(1) Rome I. 

This is not to say that proprietary priority and distributive priority do not 

interact. In particular, a creditor’s entitlement to the proceeds of a claim 

may depend on his proprietary priority. In other words, to the extent that 

under the law governing the distribution of proceeds a creditor’s proprietary 

interest is relevant, this proprietary interest must be determined 

in accordance with the law(s) designated by Article 14 Rome I. As to 

the law governing distributive priority: in insolvency proceedings this is 

the lex (fori) concursus, as follows from Article 4(2)(i) of the Insolvency 

Regulation. The law governing distributive priority outside insolvency is 

currently not regulated in EU law. One would expect that most Member 

States would apply the law of the country where enforcement takes place 

(lex loci executionis). 

9.4. Limitations on party autonomy? 

As I hope to have demonstrated, a solution based upon party autonomy 

for determining the law applicable to the property aspects of assignment 

would be reconcilable with the ‘distinctive legal nature’ of assignment. 

If one would nevertheless prefer to reduce the ‘risk’ that a manipulative 

choice-of-law is made, one could provide that the assignor and the as- 


203 


www.sellier.de


signee shall only be able to choose a law that is connected with the parties 

or the transaction. This is not without precedent in EU law: see Article 

5(2) of Rome I concerning the contract of carriage of persons. 34 However, 

some Member States seem to be hyper allergic to party autonomy and 

even this relatively low dose of party autonomy runs the risk of causing 

anaphylactic reactions. In our recent paper, Sanne van Dongen and I have 

therefore suggested the following compromise solution: the applicability 

of the proper law of the assigned claim, in combination with a limited 

choice of law possibility for the law of the assignor’s residence. We believe 

that this solution could offer the ‘best of both worlds’. Most importantly, 

the applicability of the law governing the assigned claim would mean that 

the assignor, assignee and debtor would only have to look at one legal 

system in order to assess their rights in respect of the relevant claims. 35 In 

situations where this conflict rule may be difficult to operate, in particular 

a bulk assignment of a portfolio of receivables governed by different laws 

and / or future receivables, whose applicable law cannot be determined 

or predicted, the parties to the assignment may choose instead the applicability 

of the law of the assignor’s residence. Since the range of laws 

that can be chosen from is thus extremely limited and these laws are both 

closely connected with the assignment, even the most ardent adversaries 

of party autonomy should not have a problem with this compromise 

solution. 36 


The ‘knockout argument’ that party autonomy must be rejected in respect 

of assignment, because this would unjustifiably harm the interests of third 

parties, proves to be too unbalanced to be true. Once the relevant third 

parties (e.g. bona fide assignees, creditors of the assignor, creditors of the 

assignee) are identified and their respective interests are assessed against 

34 

Examples in the Hague Conventions are: Article 3 of the Convention of 14 

March 1978 on the Law Applicable to Matrimonial Property Regimes and 

Article 4 of the Convention of 5 July 2006 on the Law Applicable to Certain 

Rights in Respect of Securities held with an Intermediary. 

35 

A. Flessner & H. Verhagen, Assignment in European Private International 

Law, 2006, p. 48. 

36 

H.L.E. Verhagen & S. van Dongen, Cross-Border Assignments under Rome I, 

Journal of Private International Law (JPIL) 6 (2010), p. 19-20. 

204 



www.sellier.de


the background of the substantive law issues at stake, it appears that there 

is no reason to be afraid of party autonomy. From the perspective of property 

law, it has been established that all of the characteristic issues of 

assignment in substantive law can be adequately dealt with by a conflict 

rule allowing the assignor and the assignee to choose the governing law. 

In so far as other ‘third party effects’ of assignment (public filing, voidable 

preference, attachment) and (distributive) ‘priority’ are concerned: 

they are outside the scope of a conflict rule for the proprietary aspects of 

assignment and accordingly do not need to be a matter of concern when 

allowing party autonomy for the proprietary aspects of assignment. 


205 


www.sellier.de


www.sellier.de

10. 

Between Articles 14 and 27 of Rome I: 

How to interpret a European Regulation 

on Conflict of Laws? 

Axel Flessner * 

10.1. Introduction: Property issues in Rome I 

Of all the types of assets to which international property law may be applicable, 

only claims against another person (the debtor), as understood to mean 

‘a right (of a creditor) to performance of an obligation (by the debtor)’, 1 and 

book entry securities serving as collateral for financial market obligations 

have thus far been addressed by European legislation. For property issues 

of ‘financial collateral arrangements’ relating to book entry securities, the 

Financial Collateral Directive of 2002 has a special conflict of laws provision 

(Article 9). For the ‘assignment’ of claims, Article 14 of the Rome I 

Regulation designates the applicable law. 

The reason that assignment of claims figures in a regulation ‘on the law 

applicable to contractual obligations’ has to do with the ‘hybrid nature’ 

of a claim, it being a ‘personal right, whether of a contractual or of a noncontractual 

nature (e.g. a tort claim), whose substance is determined by 

the law of obligations’, and at the same time forming ‘an intangible asset, 

…, which is itself capable of being transferred, charged, or otherwise 

encumbered under the law of property’. 2 If we had already had in Europe 

a regulation on the applicable law of property, assignment of claims could 

* 

Professor Em. of European Private Law, Commercial Law and Conflict of 

Laws at the Humboldt-Universität, Berlin. 

1 

This definition of a ‘claim’ can be assembled from the relevant provisions in 

the Draft Common Frame of Reference (DCFR), Book III, Chapters 1 and 

5, Articles 1:102 and 5:101 (von Bar / Clive / Schulte-Nölke (eds.), Principles, 

Definitions and Model Rules of European Private Law: Draft Common Frame 

of Reference (DCFR), Outline edition, Munich 2009). 

2 

Verhagen / van Dongen, Cross-Border Assignments under Rome I, Journal of 

Private International Law 6 (2010) 1-21, at p. 2. 

207 


www.sellier.de


equally have been placed into such an instrument, as has been demonstrated 

by the recently enacted Dutch statute on international property 

law. 3 However there was already a rule like Article 14 in the predecessor 

of Rome I: the Rome convention of 1980 on the law applicable to contractual 

obligations (Article 12). Hence, given the economic importance 

of assignments, it seemed natural to continue with the placement of the 

subject in the regulation on contract law conflicts, and not to wait until 

some possible regulation on conflict of property laws, which is not yet on 

the European legislative horizon. 

In the meantime, Article 14 Rome I has been amply explained. 4 However, 

trouble emerges from Article 27 paragraph 2, which charges the European 

Commission to reassess the matter as to ‘the question of the effectiveness 

of an assignment or subrogation of a claim against third parties and the 

priority of the assigned or subrogated claim over a right of another person’. 

This third-party effectiveness has been the hottest issue in the making of 

Article 14. What precisely is that issue, and who are the third parties? 

First, they are the creditors of the assignor and those of the assignee. 

When an assignment has been made, do the creditors of the assignor have 

to accept that the claim no longer belongs to their debtor, and can the 

creditors of the assignee claim it to belong to their debtor? Second, when 

successive assignments of the same claim have been made, either by the 

same assignor or by successive assignors, which of the various assignees 

can regard the claim to be his against the other assignees? 

Many commentators think that Article 27(2) completely excludes thirdparty 

effectiveness from the scope of Article 14. In their opinion, the 

legislator has postponed the decision on that issue to the special review 

procedure required by Article 27(2). 5 The view is surprising, because the 

3 

Articles 9 and 10. See in this book van der Weide, Party Autonomy in Dutch 

International Property Law (Chapter 2), and the text of the statute. 

4 

Martiny, in: Münchener Kommentar zum Bürgerlichen Gesetzbuch, vol. 10: 

Internationales Privatrecht, 5. Aufl., München 2010, Art. 14 Rom I-VO; Martiny, 

in: Reithmann / Martiny (eds.), Internationales Vertragsrecht, 7. Aufl., 

Köln 2010, p. 289-308; Plender / Wilderspin, The European Private International 

Law of Obligations, 3d ed., London 2009, p. 367-387. 

5 

Sigman / Kieninger, The Law of Assignment of Receivables: in Flux, Still uncertain, 

Still non-Uniform, in: Sigman / Kieninger (eds.), Cross-Border Security 

over Receivables, Munich 2009, p. 1-74, at p. 54; Plender / Wilderspin (note 

4) 386; Garcimartín Alférez, Assignment of claims in the Rome I Regulation: 

208 

Axel Flessner 


www.sellier.de

10. Between Articles 14 and 27 of Rome I 

text of the provision requires, first, only a report and, only ‘if appropriate’, 

a proposal by the Commission to amend the Regulation; second, it does 

not and cannot promise a corresponding vote by Council and Parliament. 

In its report, the Commission may find an additional rule unnecessary, 

or Council and Parliament may be unwilling to vote on or to accept a 

submitted proposal. Hence, if the exclusionary view should be correct, we 

would have to do without a European rule on the issue, not only temporarily 

but also perhaps for a considerable period, and the present version 

of Article 14 may be there to stay for quite some time. In this situation, 

its interpretation becomes important: In the light of Article 27(2), does 

Article 14 cover third-party effects or not? 

I have earlier submitted that it does, 6 and also that third-party effects are 

within the scope of Article 14 paragraph 1 7 – which means that Article 14 

embodies the solution found by the Dutch Hoge Raad: namely, to apply 

the law applicable to the contractual relationship between the parties to 

the assignment: for example, a sale of the claim, a loan against security, 

or a factoring arrangement. 8 This brings into play Article 3 of Rome I, 

and thus also opens the third-party effects to choice of law by the parties 

to the assignment. 

We must be aware, however, that whatever view we hold on the scope of 

Article 14, it may be influenced by our assessment of the solution that can 

be achieved by applying the provision. Hence, all the authors who have 

been pleading to apply the law of the assignor’s location are among those 

Art. 14, in: Ferrari / Leible (eds.), Rome I Regulation – The Law Applicable 

to Contractual Obligations in Europe, Munich 2009, p. 217-249. at p. 226; 

Bauer, Die Forderungsabtretung im IPR – Schuld- und zuordnungsrechtliche 

Anknüpfungen, Frankfurt am Main 2008, p. 103, 167, 301. 

6 

Flessner, Die internationale Forderungsabtretung nach der Rom I-Verordnung, 

Praxis des Internationalen Privat- und Verfahrensrechts (IPRax) 2009, p. 35- 

43, at p. 38-39. This view is shared by Martiny (note 4), in: Münchener 

Kommentar, Art. 14 nr. 16, and in: Reithmann / Martiny, p. 295; Verhagen / 

van Dongen (note 2) 5; Verhagen, in this book (Chapter 9); Einsele, Die 

Forderungsabtretung nach der Rom I-Verordnung, Rabels Zeitschrift für ausländisches 

und internationales Privatrecht (RabelsZ) 74 (2010), p. 91-117, at 

p. 99-101, 112, 114. 

7 

Flessner (note 6) 39-40. 

8 

On that solution, see Verhagen, in this book (Chapter 9); Flessner / Verhagen, 

Assignment in European Private International Law, Munich 2006, p. 8-12. 

Axel Flessner 

209 


www.sellier.de


who think that Article 14 in its present version does not apply to third 

parties in the first place. 9 Indeed, there is no hope of them finding their 

preference codified in the present version, whereas other solutions – the 

law of the underlying transaction between assignor and assignee or the law 

of the assigned claim – can be accommodated in the present text. 10 Therefore, 

under cover of a discussion on the scope of Article 14, there may in 

fact rage a dispute concerning whether this or that conflict rule would be 

the appropriate substantive policy for the third-party issue. It seems helpful 

in this situation to take a step back by asking how the opposing views as 

to the scope of Article 14 fit in with accepted standards of interpretation 

of European Community and Union law. 

10.2. Statutory interpretation of European Law 

The legislative output under the European treaties has the character of 

statutes. It is therefore basically recognised that the usual methods of 

statutory interpretation can be employed here, particularly to regulations 

on matters of private law, as these regulate directly the rights and duties 

of the citizens and often replace existing Member State law, such as the 

Rome I Regulation does. The four classical roads on which to establish 

the meaning of a statutory rule are: first, to ask what the text, the wording 

of the rule, actually says; second, to ask what the legislator meant to say, 

judging from the legislative history of the rule; third, to determine how the 

rule best fits with existing or even prospective neighbouring law: in other 

words, whether it fosters or disturbs the coherence of the legal system; and 

fourth, to assess what the rule is to achieve: namely, its purpose. 11 The 

European Court of Justice has repeatedly, and quite recently, expressly af- 

9 

Sigman / Kieninger (note 5) 51; Bauer (note 5) 283-293. 

10 

See Flessner (note 6) 39. 

11 

See in detail Arnull, The European Union and its Court of Justice, 2d ed., 

Oxford 2006, p. 607-620; Weatherill / Beaumont, EU Law, 3d ed., London 1999, 

p. 184-192; Waelbroeck, in: Commentaire Mégret, Le droit de la communauté 

économique, 2ème éd., Bruxelles 1993, vol. 10, Article 164 nr. 10-23; 

Riesenhuber, Europäische Methodenlehre, 2. Aufl., Berlin 2010, §§ 1 and 11; 

Henninger, Europäisches Privatrecht und Methode, Tübingen, p. 264-322; 

Langenbucher, Europarechtliche Methodenlehre, in: Langenbucher (ed.), 

Europarechtliche Bezüge des Privatrechts, 2. Aufl., Baden-Baden 2008, § 1; 

Vogenauer, Die Auslegung von Gesetzen in England und auf dem Kontinent, 

Tübingen 2002, p. 344-429; Kropholler, Die Auslegung von EG-Verordnungen 

210 

Axel Flessner 


www.sellier.de


firmed and used these methods of interpretation when applying the Brussels 

I Regulation on jurisdiction and recognition of judgments (which is 

complementary to the Union’s conflict of laws regulations). 12 

Each one of these four classical roads – textual, historical, systematic, and 

purposive – can, when directed to a European regulation, take on some 

special European Union colour, in comparison with a statutory interpretation 

of Member State law, with an interpretation of European treaty 

law, and with an interpretation of European directives and regulations in 

fields other than conflict of laws. This will be illustrated here briefly for 

each of them. 

As to the wording of rules, it has often been noted that it is less compelling 

than in Member State law because the text of a regulation exists in more 

than twenty official languages of equal weight, which makes for less precision 

of content and for more options to choose from. 13 This is certainly 

true, but once we have got a rule in its twenty or so versions, the rule will 

and must carry more weight by its sheer existence, since European legislation 

is much slower and much more cumbersome to make and to change 

than are Member State statutes. This means that once we have the rule, 

we must learn to love it. 

As to the legislative history, it is conventionally said that it serves to determine 

the real intention of the legislator. But we all know that ‘the 

legislator’ is an artificial notion, which in a democratic and parliamentary 

government designates a great number of often invisible people having 

been involved in the making of the statute – those who drafted it, those 

who deliberated on it, and those who eventually passed it in the legislative 

assembly. To say that there is a will of the legislator is wholly fictional. The 

fiction allows us to create a shorthand summary of the many ideas that 

went into the legislative text we are reading and to use them again for a 

legitimate – now ex post – argument. 

zum Internationalen Privat- und Verfahrensrecht, in: Basedow et al. (eds.), 

Aufbruch nach Europa, Tübingen 2001, p. 583-594, at p. 590-594. 

12 

See European Court of Justice, judgment of 23 April 2009, C 533 / 07 – Falco 

Privatstiftung, with further references, nr. 20. 

13 

Weatherill / Beaumont (note 11) 185; Waelbroeck (note 11) nr. 12; Riesenhuber 

(note 11) § 11 nr. 19. But see recently, and critically, Weiler, Grammatikalische 

Auslegung des vielsprachigen Unionsrechts, Zeitschrift für Europäisches Privatrecht 

(ZEuP) 2010, 863-882. 

Axel Flessner 

211 


www.sellier.de


The European legislator is, moreover, much less tangible and much less 

reachable than a state legislator, even less than in the larger Member 

States. To begin with, we have three institutional legislators, or at least 

two and a half, with the European Commission having the exclusive right 

to bring bills, followed by the Council and the Parliament. But the involvement 

in European law-making in fact reaches far into the Member 

States’ governments and parliaments, and at the same time and at all 

levels the people actually involved are much less visible and audible – 

union-wide and in each Member State. The social and communicative 

distances between legislators and the concerned public are much greater 

at the European than at the national level. A will to be relied on when it 

later comes to apply the legislative product is more difficult to establish 

and therefore carries less weight. It would seem that, compared with national 

law, there is in Europe a different correlation between the text of 

the statute and its legislative history; the history counts less and the text 

counts more than in the national environment. 14 

The systematic approach 15 is to ask what the position of the rule is in 

relation to other existing or prospective statutory law. This aspect has 

particular significance for the Regulations that have been and will be 

enacted on conflict of laws and international civil procedure. The Community 

has made, and the Union is now making, great strides in building 

up a system of enactments on conflict of laws, from the law of obligations 

to family law and succession law, and it seems clear that in interpreting 

these enactments we must be very careful to give the applicable rule its 

proper place in this existing and emerging system. 

Finally, the European colour of the purposive interpretation: 16 This way 

of looking at a text has much more legitimacy in the European Union 

than in Member States, because the Union is still a political body that 

14 

Accord: Hartley, The Foundations of European Community Law, 6th ed., Oxford 

2007, p. 74; Weatherill / Beaumont (note 11) 188-190; Stotz, Die Rechtsprechung 

des EuGH, in: Riesenhuber (ed.) (note 11) § 22 nr. 15; Langenbucher 

(note 11) nr. 11; Henninger (note 11) 296; Vogenauer (note 11) 320-321, 

420-423. 

15 

Also called ‘contextual’, see Arnull (note 11) 626; Weatherill / Beaumont (note 

11) 184, 190. 

16 

Also called ‘teleological’; see Plender / Wilderspin (note 4) 35-38; Arnull (note 

11) 607; Weatherill / Beaumont (note 11) 190 fn. 198; and usually so in the 

German language literature, see Riesenhuber (ed.) (note 11) § 11 nr. 40-46, 

212 

Axel Flessner 


www.sellier.de


declares itself to strive at aims, at ‘integration’, and, particularly, to build 

up and equip with legislation the Area of Freedom, Security, and Law. 17 

The new provisions of the Lisbon treaty on this particular project will 

constantly provide legal practice with strategic documents driving forward 

and explaining further legislation, such as the Stockholm Programme of 

the European Council now does. 18 This is an environment quite different 

from the more staid and settled legal and political systems of the Member 

States. It is clear that the purposes of the rules created for that process 

must play a very prominent role in their interpretation. 

10.3. Application to Articles 14 and 27 

What can we gain from these European interpretation rules with relation 

to our problem? We have to assume for this investigation the text of Article 

14 as it stands today, either before or after the review process under 

Article 27(2) has been completed, but without a modification of Article 

14 resulting. Does Article 14 then provide a conflict rule for third-party 

effects? 

10.3.1. The wording 

The wording is ambiguous. Paragraph 1 says that the conflict rule relates 

to the relationship between assignor and assignee; it does not mention 

third persons. But Recital 38 of the Preamble explains that this wording 

was to make sure that ‘property aspects’ was also covered. 19 What 

are property aspects that do not include third-party effects? Even Profes- 

§ 22 nr. 17; Langenbucher (note 11) nr. 12, 15; Heiderhoff , Gemeinschaftsprivatrecht, 

2. Aufl., München 2007, p. 45-49. 

17 

Articles 67-89 of the Treaty on the Functioning of the European Union. 

18 

On the Stockholm Programme, see R. Wagner, Die politischen Leitlinien zur 

justiziellen Zusammenarbeit in Zivilsachen im Stockholmer Programm, IPRax 

2010, 97-100. The European Commission has meanwhile produced its ‘Action 

Plan’ implementing the Stockholm programme, COM(2010) 171 final, see 

www.ec.europa.eu / justice ‘policies’ , under ‘Latest Developments ‘. 

19 

The relevant text of the Recital is: ‘In the context of voluntary assignment, 

the term ‘relationship’ should make it clear that Article 14 (1) also applies to 

the property aspects of an assignment, as between assignor and assignee, in 

Axel Flessner 

213 


www.sellier.de


sor Paul Lagarde, the great authority on conflict of laws, arguing from 

a French perspective, has doubted whether one can split the property 

aspects into internal ones – those between the parties to the assignment – 

and external ones – those towards third parties. 20 But in fact, with regard 

to assignment of claims, there is a line in French legal thinking that divides 

the problem in just that way: namely, the mere consensus of the 

assignment parties transferring the claim between them, and only an additional 

step – notification to the debtor or his express assent – making 

the assignment effective towards others. 21 In providing exactly for this 

external aspect, Article 27 shows that the European legislator finds just 

such a split not as incongruous as it may appear to observers outside of 

France. 22 

In contrast, Article 27(2) is very clear in calling for a report that possibly 

proposes a ‘modification’ of the Regulation with regard to the 

topic of third-party effects. 23 The text obviously presupposes that the 

Regulation does contain a rule on the topic, which, however, should be 

reconsidered and, if it appears necessary, be modified; it also seems to 

allow Article 14 to remain unaltered if the Commission or the Council 

and the Parliament so decide. Thus, all in all, the textual interpretation 

ends with a somewhat mixed result. Taken alone, the text of Article 

14 neither compels nor excludes its application to third-party effectiveness. 

legal orders where such aspects are treated separately from the aspects under 

the law of obligations.’ 

20 

Lagarde, Retour sur la loi applicable à l’opposabilité des transferts conventionnels 

de créances, in: Droit et actualité: Études offertes à Jacques Béguin, Paris 

2005, p. 419, 424. 

21 

The system has recently been explained to foreign readers by Verhagen / van 

Dongen (note 2) 6-7; Bauer (note 5) 64-77. 

22 

For the latter view, see Flessner (note 6) 40; Verhagen / van Dongen (note 2) 

5-8. 

23 

The English version speaks of the envisaged proposal ‘to amend’ the Regulation. 

This might evoke the idea of an addition to the existing text, but all the 

other versions in Germanic and Romanic languages use the term ‘modification’: 

modificación (Sp), modification (Fr), modifica (It), alteracao (Port), 

wijziging (Du), aendring (Dan), ändring (Swed), Änderung (Ger). 

214 

Axel Flessner 


www.sellier.de


10.3.2. Legislative intent 

We know the following from the accessible documents: first, the Commission 

had proposed a special rule for third party effects, which was to 

apply the law of the assignor’s place; second, long negotiations ensued; 

third, in the final stages in the Council, the presidency proposed, as a 

compromise, either to follow the Commission proposal or to declare the 

law of the assigned claim to be applicable; and fourth, neither proposal 

has been made law. 24 There are people, some of whom participated in 

the deliberations, who tell us that the decision on that special provision 

was to be postponed until the review process. 25 However, there is 

no formal decision as to this effect; Article 27, as we have seen, does 

not say so, and there is no documentary evidence of such a decision 

in Council or Parliament, although majority voting would have been 

possible. Therefore, all we can see in the legislative history is that the 

Commission made a proposal for a Regulation with a special provision 

for third-party effects, and that the Regulation has been eventually enacted, 

but without such a provision. To this extent, the proposal of the 

Commission to have a special provision on third-party effects appears to 

have been rejected without a substitute, thus leaving Article 14 as the 

general, unqualified rule. 

As mentioned earlier, in European Union law the enacted text must 

weigh more heavily than a possibly diverging but not visibly expressed 

legislative intent. The inherent weakness of invoking a legislator’s alleged 

real intent can be overcome only by official documentary evidence 

of such intent. The European Court of Justice has at times been 

extremely outspoken in this respect. 26 We have no such evidence here 

for an alleged decision of the legislator to carve out from the original 

legislative project an important topic to be dealt with only at some later 

time, if at all. 

24 

For details of the process, see van der Grinten, Art. 14 Rome I: A Political Perspective, 

in this book (Chapter 7); Sigman / Kieninger (note 5) 50-53; Plender / 

Wilderspin (note 4) 370-373; Garcimartín Alférez (note 5) 246-247. 

25 

Sigman / Kieninger (note 5) 54; Garcimartín Alférez (note 5) 234-235. 

26 

See cases noted by Arnull (note 11) 619, 620; Weatherill / Beaumont (note 11) 

189, 190. 

Axel Flessner 

215 


www.sellier.de


10.3.3. System 

In the emerging European system of conflict rules, considerable emphasis 

is laid on avoiding overly fine distinctions, and, in particular, on avoiding 

distinctions that are familiar only in some Member States but not in 

others. See Recital 38 of the Preamble (above, note 19). It says that Article 

14(1) has been drafted so as to make clear that in the relationship of assignor 

and assignee no distinction should have to be made between the 

property aspects and the contractual aspects of assignments, even when 

a given Member State’s legal system separates them. This separation is 

thought to be basic and necessary, for instance, in German legal thinking 

but less so in other countries. Thus, what the Preamble here says 

is that this peculiar division of contractual and property aspects should 

not bother us in the application of Article 14. The distinction between 

property effects between assignor and assignee and property effects against 

third parties is another such peculiarity, this time from the French orbit. 27 

As long as this distinction has not been clearly adopted in the envisaged 

review process, disregarding it for the time being would keep Article 14 

in line with the overall endeavour for simplicity in the European conflict 

rules. It would save us from a split between the property effects (those 

between assignor and assignee and those towards third parties), and from 

working with different conflict rules for either side of the split. 

The distinction between the internal and external property aspects is also 

missing in two provisions that are very close to Article 14, viz. Article 

15 of Rome I and Article 19 of Rome II. These provisions apply in the 

case of subrogation by operation of law when debts are paid by a person 

other than the debtor: for instance, by an insurer or by a guarantor or 

surety. According to them, the subrogation is governed by the same law 

that, as between creditor and payer, caused or justified the latter to pay. 

They do not provide for a special treatment of third-party effects, which 

means that these effects as well are governed by the law designated by 

the overall rule. Contractual subrogation or transfer by assignment and, 

on the other hand, subrogation by operation of law may be very close 

together, even interchangeable. Payments by non-debtors can lead either 

to an obligation of the paid creditor to assign his claim against the debtor 

to the payer, or to an assignment to the payer by operation of law; it is a 

matter of legislative discretion and legal technique. However, as to the 

position of third parties other than the payer (being the recipient of the 

27 

See above at note 21. 

216 

Axel Flessner 


www.sellier.de


assignment), it should be better not to make a difference. Not creating 

a separate and different rule for the third-party effects in either of these 

situations would keep Article 14 sensibly in line with Article 15 of Rome 

I and Article 19 of Rome II. 28 

It would also keep Article 14 Rome I in line with the Financial Collateral 

Directive, which, in its Article 9, shows in the clearest manner 

how the proprietary effects of a transaction can be dealt with by a unitary 

conflicts rule for the parties to the transaction and third parties alike. 29 

The provision subjects the proprietary effects of a financial collateral arrangement 

involving book entry securities to the law of the country where 

the ‘account’ is maintained: that is, the register in which title entries 

concerning the securities are made (Article 9[1]). That country’s law is 

(in paragraph 2) declared to apply to: 

a) the legal nature and proprietary effects of book entry securities collateral; 

b) the perfection of such collateral and more generally to the completion 

of the steps necessary to render such an arrangement effective against 

third parties; and 

c) whether a person’s title in such book entry securities is overridden by 

or subordinated to a competing title, or whether a good faith acquisition 

has occurred. 

A legislator could not have demonstrated more clearly that it is perfectly 

all right to cover by one and the same conflicts rule the property effects 

between the parties to a transaction and the property effects towards third 

parties. Why should Article14 Rome I not embody the same approach? 

Furthermore, to carve out the third-party effectiveness from Article 14 

would raise the question of where to put this topic in the overall European 

system of conflict rules. Imagine that we had a European Regulation 

of international property law. Should we place third-party effects of assignments 

into that regulation and keep in Rome I the property effects 

between assignor and assignee? Certainly it would be better to have both 

28 

See in greater detail Einsele (note 6) 98-99, and, as to Article 13 of the Rome 

convention, the predecessor of Article 15 Rome I, Flessner / Verhagen (note 8) 

27-28. 

29 

Directive 2002 / 47 / EC on financial collateral arrangements, Official Journal 

2002, L 168, p. 43. In the same sense, but less elaborately, Article 9(2) of the 

Settlement Finality Directive 1998 / 26 / EC of 19 May 1998. 

Axel Flessner 

217 


www.sellier.de


issues together in either one of them. As yet, there is no property Regulation. 

Thus, the place for third-party effectiveness is only in Rome I. We 

have no other place at present and for the foreseeable future. 

Finally, one needs to consider how Article 14 – if deprived of application 

to third-party effectiveness – would fit in with the European Treaties. 

The Preamble in Recital 43 expressly claims Rome I to accord with the 

principle of subsidiarity, laid down in Article 5 of the Union Treaty (formerly 

Article 5 of the Community Treaty), because the legislative aim, it 

says, could be better achieved at the Community level than at the level 

of Member States. The overall legislative aim is, and must be, to build 

up and equip with rules on conflict of laws the European Area of Justice; 

the Regulation is expressly based on former Article 65 of the Community 

Treaty, now Article 81 of the Treaty on EU Functioning. However, not to 

include the property effects with regard to third parties in Rome I would, 

under present conditions, mean leaving this issue to the legislative competence 

of the Member States. 30 As a consequence, it could no longer be 

submitted to the European Court of Justice. The issue has been termed by 

all the experts as one of utmost importance and intricacy. To exclude it 

from Rome I would therefore run counter to the very constitutional idea 

on which Rome I was based by the legislator. 

10.3.4. Purpose 

The overall objective of rules on conflict of laws is to provide certainty 

by designating the applicable law where an abundance of competing and 

possibly applicable laws produces insecurity. In the European Union, this 

central idea takes on a special flavour. The Union is set up to be an area 

of freedom, security, and justice, and an internal market ‘without internal 

borders’, 31 although it is not able, constitutionally and politically, to provide 

unity of substantive law. Union-wide unified rules on the inevitable 

conflict of laws are eagerly sought as a way out of this dilemma, and they 

30 

There is agreement on this result among most authors, although not whether 

the result is, or would be, welcome; Flessner (note 6) 38, 39; Sigman / Kieninger 

(note 5) 54, 55; Garcimartín Alférez (note 5) 235; Bauer (note 5) 301; Plender / 

Wilderspin (note 4) 379; Martiny (note 4) Article 14 nr. 16; Verhagen / van 

Dongen (note 2) 12; van der Grinten (note 24). 

31 

Article 3 (formerly Article 2) of the Union Treaty; Article 26 of the Treaty on 

the Functioning of the EU (formerly Article 14 of the Community Treaty). 

218 

Axel Flessner 


www.sellier.de


must make legal diversity manageable for a single market still to be fully 

achieved. Hence, they are strongly purposive. 

With the purposive character of Rome I, the well-known principle of 

practical effectiveness (effet utile), often used by the Court of Justice, 

comes into play. It says that whenever the text of a European legal rule 

allows for a choice to be made, the interpretation that better promotes 

the rule’s objective should be preferred. A clear rule on the third-party 

effectiveness of assignments is regarded by all the experts – practitioners 

and academics alike – to be of utmost importance to the assignment 

business. 32 In the purposive view, it follows that its inclusion rather than 

its exclusion from the present Article 14 is the required interpretation. 

Without such inclusion, Article 14 would be incomplete, a lex imperfecta, 

for the assignment of claims economy. 

In addition, there are special grounds in this case for looking into purpose 

and practical effectiveness. Under the predecessor of Article 14 Rome I, 

which was Article 12 of the Rome Convention of 1980, the third-party 

effects indeed had been regarded as being included in the provision. Certainly 

some doubts about that had been expressed in the literature, but 

in the case law of Member States the inclusion was practised. The Dutch 

Hoge Raad regarded it as a matter of Article 12 paragraph 1; the German 

Bundesgerichtshof as a matter of paragraph 2; and the English Court 

of Appeals relied on both paragraphs. 33 It follows that not to include the 

third-party effects in Article 14 Rome I would actually reduce an at least 

partial unity that previously had been achieved and that had the chance 

of being perfected some day by a ruling of the Court of Justice. Such an 

interpretation would not only neglect but would directly run counter to 

the purpose of European conflicts law. It would also frustrate, maybe forever, 

the hope of having the European Court of Justice rule one day on the 

aforementioned differences in the case law and in the literature as to the 

proper treatment of third-party effects. If Article 14 in its present version 

32 

See recently, Sigman / Kieninger (note 5) 54. 

33 

See references and analysis Flessner / Verhagen (note 8) 8-14; Plender / Wilderspin 

(note 4) 380-386. Most authors thought, and still think, the Court of Appeal 

to have belonged to the paragraph 2 faction, but a full reading of Lord Justice 

Mance’s opinion, written for the court, shows that he relied on both the Dutch 

and the German interpretations, which had been brought to his attention; see 

the judgment (2003) AllER 257, at p. 273 et seq, 275. 

Axel Flessner 

219 


www.sellier.de


did not cover them, the issue would have been given back to Member 

State law and could no longer be submitted to the European Court. 

10.4. Summary 

With the new and the forthcoming regulations on conflicts of law, and 

on the basis of successive ‘programmes’ (The Hague, Stockholm), the European 

Union is gradually building up the legal equipment for interstate 

activity of people and enterprises in the European area of freedom, security, 

and justice. For civil proceedings and conflicts of law, the European 

legislator is moving away from what has often been criticised as selective 

and uncoordinated action to building a statutory system. This movement 

entails a new challenge to disciplined interpretation, by which statutes are 

seen as durable public assets rather than as governmental policy measures. 

This approach makes recourse to the accepted standards of interpretation 

necessary. To be sure, these standards are maxims rather than dictates; 

they are not mandatory law, but certainly they are legitimate avenues to 

reasoned judgment and legal persuasion. 

As to Article 14 of Rome I, its interpretation as a directly applicable and 

existing law may fall under the foreshadow of the forthcoming review 

procedure under Article 27(2), which, judging from the debates before 

and in the making of the Regulation, may again be dogged and futile. 

Recourse to the accepted standards of interpretation could result in the 

withdrawal of the current Article 14 from these unresolved legislative 

policy debates. The foregoing application of these methods has produced 

a strong preponderance of arguments supporting the view that Article 14 

in its present form is applicable also to third-party effectiveness. Only the 

textual interpretation rendered a mixed result; the other ways (legislative 

history, system, and, above all, purpose) clearly favoured the inclusive 

interpretation. With this result, the present text can stand fully on its 

own, and a recent case-by-case study has shown that the provision in its 

present version is workable in every respect. 34 

34 

Flessner, Rechtswahlfreiheit auf Probe – Zur Überprüfung von Art. 14 der Rom 

I-Verordnung, in: Baur et al. (eds.), Festschrift für Gunther Kühne, Frankfurt 

am Main 2009, p. 703-715. 

220 

Axel Flessner 


www.sellier.de


10.5. Consequences 

If Article 14 applies also to third-party effects, the conflict solution must 

be taken either from its first or its second paragraph – there is no other 

choice. Third-party effects are therefore governed by the law applicable to 

the contract between the assignment parties (paragraph 1) or by the law 

governing the assigned claim (paragraph 2). In the competition between 

the two, the law of the assignment contract has a clear advantage by a 

majority of supporting arguments. 35 Applying Article 14(1) means that 

together with the assignment contract, third-party effects are open to 

choice of law by the assignment parties according to Article 3 of Rome I. 

This is an added contextual argument for the inclusionary interpretation. 

In the overall European system of conflict rules, choice of law by the parties 

is on the march forward. 36 It has been officially called a ‘cornerstone’ 

for contractual obligations; 37 it has been extended onto non-contractual 

obligations; 38 it is part already of family law; 39 and it has been proposed 

even for succession law 40 (which, as well as property law, is the prototype 

of law providing for effects on third parties regardless of their consent). 

To include third-party effects in the present Article 14 would also bring 

this part of Rome I into line with the general, party-friendly system. It 

would indeed seem strange for the European Union to break away just 

at this point – assignments being such an important instrument on the 

international financial market – from party autonomy and to insist here 

on mandatory law. 

People are very often uneasy with the notion that the assignment parties 

should be able to choose the applicable law with effect to third parties 

without those parties consenting. The second paragraph of Article 14 

expressly shields the debtor under the claim from detrimental effects of 

35 

See in detail Flessner (note 6) 39-42. 

36 

Overview by Verschraegen, Kritische Bestandsaufnahme zur Rechtswahl, in: 

Verschraegen (ed.), Rechtswahl – Grenzen und Chancen –, Wien 2010, 

p. 111-131, at p. 120-128. 

37 

Recital 11 of the Preamble to Rome I. 

38 

Article 14 of Rome II and pertinent Recital 31in the Preamble. 

39 

See Articles 7 and 8 of the Hague Protocol on maintenance obligations, which 

have been incorporated by the Regulation of 18 December 2008 on maintenance 

obligations (Article 15), Official Journal of the EU 2009, L 7, p. 1. 

40 

Proposal of the European Commission for a Regulation on matters of succession, 

Article 17, COM (2009) 154 final. 

Axel Flessner 

221 


www.sellier.de


the assignment, by reserving to him the continued applicability of the law 

governing the assigned claim. Hence, the misgivings can realistically refer 

only to possible disadvantages for the creditors of assignor and assignee 

and for other, subsequent assignees of the same claim. 

Such qualms are unfounded. The ‘external’ effect on third parties – creditors, 

neighbours, tax authorities, potential buyers – is the very essence of 

property law insofar as property enables title holders to dispose freely of 

their assets. Why should we suddenly turn away from this basic principle 

once we enter the field of conflict of laws? 41 Creditors are sufficiently protected 

by the applicable local rules on the enforcement of judgments and 

on insolvency. 42 Potential buyers can inquire under which law the offered 

asset has been acquired by their seller no less than they can inquire about 

the validity of the seller’s title under whatever applicable law. They are 

further protected against surprise by Article 3(3) Rome I, which excludes 

party choice of law in purely domestic transactions. Moreover, the general 

public is protected by the reservation in Article 9 Rome I of overriding 

mandatory provisions of the forum state and of the state where the contract 

has to be performed. 

It is often thought that party choice of law could not be allowed where we 

find for security rights in assets (including rights on claims) the requirement 

of registration or some other form of publication of such rights. 

Authors (including myself) have been prepared to take this situation into 

account: for instance, with a special conflict rule – which might then 

become a further paragraph of Article 14 43 – or by seeing registration 

requirements as overriding mandatory rules, thus putting them under Article 

9 Rome I, 44 or by sharply restricting Article 14(1) to the obligational 

aspects between assignor and assignee, and referring all third-party effects 

41 

See in greater detail Flessner / Verhagen (note 8) 22-36, and Verhagen, in this 

book (Chapter 9), in footnotes 11-17; for movable property in general, Flessner, 

Rechtswahl im internationalen Sachenrecht – neue Anstösse aus Europa, 

in: Apathy et al. (eds.), Festschrift für Helmut Koziol, Wien 2010, p. 125-146, 

at p. 137-138 (in translated version: see Chapter 1 in this book on Party 

Autonomy in International Property Law: Choice of Law in International 

Property Law, New Encouragement from Europe, par. D, section IV). 

42 

See in detail Flessner (note 41) 137-138 (German text) / par. D, section IV 

(English text); Verhagen (note 41). 

43 

Flessner / Verhagen (note 8) 71-76; Verhagen / van Dongen (note 2) 9-10. 

44 

Einsele (note 6) 107-113. 

222 

Axel Flessner 


www.sellier.de


(not only public registration requirements) to the law at the assignor’s 

location. 45 

Upon further reflection on the needs in the European market and the European 

area of justice, we should be wary of such deference to publication 

requirements in single Member States. A national rule that invalidates 

or denies full effect to an otherwise irreproachable cross-border security 

assignment in the internal – the common – market is an impediment to the 

free movement of capital. How can it be justified in the European Court of 

Justice? A national preference for registration seems not enough to justify 

a Member State’s reaching out with invalidating rules into cross-border 

business in the common market. The purely local affairs are in any case 

taken care of by Article 3(3) Rome I, and the enforcement and ranking 

of security rights is reserved to the state where the encumbered asset 

has been seized for a judgment claim enforcement or for an insolvency 

procedure. 46 

Moreover, in single Member States there is a particular aspect of registration 

requirements that should raise European concern. Registration 

statutes for movable property usually open, and restrict, the register to 

parties located in the register state. However, mandatory legal ties of market 

actors to a single Member State are questionable in the European 

market, where everybody should be able to move freely and make use of 

local opportunities without discrimination. Registration of security rights 

may be regarded as a burden as well as an advantage. To impose them 

only on locals and their assets and to deny their benefits to non-locals 

may be perceived as discrimination by either side. European Union law is 

principally opposed to legal requirements that tie down market actors to 

the rules of their home state and prevent other market participants from 

making use of those rules. 

This dilemma would be avoided from the outset if everybody, by an agreement 

with the other party in the transaction, would be able to choose the 

applicable law and thus opt in or out of whatever publication requirement 

or opportunity exists anywhere in the Union. The aforementioned case 

45 

Sigman / Kieninger (note 5) 60-69. 

46 

I have developed this in more detail elsewhere: Flessner (note 41) 141-146 

(German text), par. E-F (English text); likewise Verhagen, in this book (Chapter 

9), in footnotes 25-28. 

Axel Flessner 

223 


www.sellier.de


study 47 has shown that such a rule can do full justice to assignments for 

security and to existing publication requirements. Until a possible modification 

after review, Article 14 turns out to be a full-fledged conflicts rule 

with no particular need of early revision or completion. The European 

Commission decided in May 2010 to postpone its report under Article 

27(2) by at least one year until no sooner than the end of 2011. 48 To 

obtain the benefit of more years of experience, the review could just as 

well be postponed further until it becomes part of the review of the whole 

regulation, which is due in three years under Article 27 paragraph 1. 

47 

See above note 34. 

48 

A preliminary study has been called for on the basis of a contract notice (Official 

Journal, S 99, of 22 May 2010) by an invitation to tender (JLS / 2010 / 

JCIV / PR / 0007 / E4). According to the invitation, the study is due for delivery 

to the Commission by August / September of 2011, and it must be evaluated 

by the Commission by November / December of 2011. 

224 

Axel Flessner 


www.sellier.de

11. 

Financial Collateral Arrangements and 

Party Autonomy 

Reinout M. Wibier * 

11.1. What are financial collateral arrangements? 

If the credit crisis has shown anything, it is the interconnectedness of the 

various market participants in the financial world. Not only did a fall in 

real estate prices in the United States trigger a ripple that developed into 

a hurricane tearing through the entire financial system, the fall of one investment 

bank, Lehman Brothers, in September 2008 almost brought the 

entire financial system to its knees. One of the reasons for this interconnectedness 

is the large number of contracts that financial intermediaries, 

in particular pension funds, insurance undertakings, credit institutions 

(i.e. banks), hedge funds, investment banks and mutual funds, enter into 

with each other on a daily basis. 

Inevitably, these contracts give rise to payment obligations which, considering 

that the parties we are talking about have a tendency to ‘think 

big’, may prove quite substantial. As a significant element of these obligations, 

the exposures that parties have against each other may therefore 

also prove quite large. An important method of reducing the risk of nonpayment 

is the provision of financial collateral (e.g. cash, debt or equity 

securities) for the obligations that arise out of these types of contracts. 

These are arrangements whereby the party that has a payment obligation 

to another party provides collateral to that other party, either by way of 

an outright transfer of collateral or by way of a charge or pledge or other 

security interest over the collateral. Providing collateral in this way enables 

the parties to minimize their net exposure to each other because, in a 

sense, the payment obligation has already been partially met by the provision 

of collateral. 1 If either party does not meet its payment obligations, 

the other, non-defaulting party will be able to obtain payment through 

1 

Alastair Hudson, The Law of Finance, Sweet & Maxwell, 2009 (Hudson 2009), 

§§ 45-06 and 45-07. 

225 


www.sellier.de


the collateral that has previously been provided. Collateral usually takes 

the form of the most liquid assets, i.e. cash (or cash equivalents such as 

United States Treasury bills) or other highly rated debt securities. 

This chapter looks at the provision of collateral and at the freedom that 

the parties to collateral arrangements have in this regard. The central 

issues are the extent to which parties have the freedom to shape these 

arrangements and the extent to which the provisions of mandatory law 

limit such freedom. Before I address these questions, I would first like to 

give a brief overview of the underlying reasons for the provision of collateral 

(§ II). Then, in § III I will deal with party autonomy in traditional 

financial collateral arrangements, whilst § IV will discuss some novel 

ways of providing financial collateral. Here, I will examine whether there 

are contractual alternatives available for the provision of collateral that 

would work at least as well as traditional solutions based on property law. 

Finally, a brief conclusion can be found in § V. 

11.2. The reasons for providing financial collateral 

Market participants enter into various types of financial contracts and 

transactions, with derivatives contracts being particularly common in 

the context of financial collateral arrangements. Derivatives contracts 

include futures or forwards, swaps and similar contracts. What these have 

in common is that their value is determined by some kind of underlying 

value, for instance currencies, interest rates or commodities. 2 An example 

would be an option to purchase a number of shares in Company X exactly 

one year from the date of entering into the option contract at a price 

of EUR 10 per share. The value of this contract would then obviously 

depend on the price of shares in Company X at the exercise date, a year 

from conclusion of the contract. 3 In order to make things less abstract I 

will use the specific example that will serve as our point of reference in 

this chapter: a credit default swap – which have become infamous as a 

result of the credit crisis – in relation to Issuer N.V. (Issuer) entered into 

by Alpha Bank N.V. (Alpha Bank) as protection buyer and Rock Solid 

2 

For a more comprehensive description of the various types of derivatives 

please see Hudson 2009, chapter 43, with further references. 

3 

If each share were worth less than EUR 10 the option would be worthless, but 

if the share price were higher, the option would still have value. 

226 



www.sellier.de

11. Financial Collateral Arrangements and Party Autonomy 

Bank N.V. (Rock Solid Bank) as protection seller. The background to the 

contract could be the following. 

Imagine that Alpha Bank holds a position in bonds issued by a company, 

Issuer. It could try and hedge its risk of a default of Issuer (i.e. the risk 

of not being paid) by entering into a credit default swap with another 

financial institution. In a credit default swap the protection seller (Rock 

Solid Bank in our example) agrees to pay a certain amount to the protection 

buyer (Alpha Bank in our example) if certain pre-determined credit 

events occur in relation to the reference entity (Issuer in our example). 4 If 

Alpha Bank (for instance) holds corporate bonds in Issuer in an amount of 

EUR 20,000,000, it could then try to find a protection seller that would be 

willing to pay this EUR 20,000,000 upon any of a number of specific credit 

events taking place such as a rating downgrade of Issuer, a default under 

one of the loan agreements that is in place in relation to this entity or a 

straightforward default under its payment obligations under the bonds. 

Upon the occurrence of a credit event, Alpha Bank may be under an obligation 

to deliver the relevant bonds to the protection seller in order to 

be entitled to payment under the credit default swap. The credit default 

swap can also be cash-settled, meaning that no delivery of the bonds from 

Alpha Bank to the protection seller takes place. Instead the protection 

seller pays the difference in value only, i.e. the difference between the 

amount received by Alpha Bank from Issuer and the amount due under 

the bonds issued by Issuer Another form of contract, which is often used, 

would involve Alpha Bank paying all amounts actually received by it from 

Issuer under the bonds to Rock Solid Bank In return, Rock Solid Bank 

pays a fixed amount equal to the amount that Issuer should have paid 

under the bonds to Alpha Bank, minus a fee. Actual payments are netted, 

i.e. only differences in these two payment obligations are actually paid. 

As a result of this arrangement, Alpha Bank hedges its exposure under 

the bonds issued by Issuer Alpha Bank, as the protection buyer, pays a 

fee to Rock Solid Bank to assume this risk. If no credit event occurs, the 

protection seller can pocket this fee which will leave it with a nice profit. 

Selling credit protection may be a lucrative business if premiums exceed 

4 

Credit default swaps can be structured in many different ways and consequently 

there are a number of possible contracts, each with a different name. 

An overview can be found in Philip Wood, Set-off and Netting, Derivatives, 

Clearing Systems, 2 nd edition, Sweet & Maxwell, London 2007 (Wood 2007a), 

§§ 10-024-10-036. 


227 


www.sellier.de


the amounts paid out under the credit protection sold by the protection 

seller. Moreover, credit default swaps enable protection sellers to take on 

credit risks against entities without having to invest directly in securities 

issued by those entities. In our example, there may be regulatory reasons 

that make it difficult for Rock Solid Bank to purchase debt instruments 

issued by Issuer directly, but by selling credit protection it has assumed 

the risk without needing to purchase debt instruments. This is also one of 

the reasons why selling credit protection may be quite dangerous; without 

credit default swaps, taking on the credit risk of EUR 20,000,000 

against Issuer in our example would require an initial investment of EUR 

20,000,000. If the risk is taken on through a credit default swap, no such 

initial investment by the protection seller is required. Consequently, 

credit default swaps increase leverage. 5 

Moreover, as was shown during the credit crisis, they increase the overall 

strains on the financial system because total losses of all market participants 

may well increase due to a single issuer of debt defaulting on its 

obligations. There is no need to actually have a credit exposure against 

the reference entity (Issuer) to purchase credit protection. Many parties 

may have bought (and sold credit) protection and consequently, a single 

default by Issuer may give rise to a great number of payment obligations 

that in aggregate may be many times larger than the total amount of debt 

issued by Issuer. 

The above shows that a credit default swap in its most basic form could 

be seen as an insurance against a credit risk. It provides protection against 

non-performance of a debtor usually by a third party (the protection 

seller) undertaking to pay an amount to the creditor if its debtor does 

not perform its obligations. 6 From a legal perspective, however, credit 

default swaps are not (and should not be) qualified as insurance contracts. 

Qualification as an insurance contract would lead to regulatory issues 

which might lead to the contracts being void because insurance may only 

be provided by licensed insurance undertakings. The key point here as a 

matter of English law is that under a credit default swap, payments must be 

5 

See EC Commission Communication, ‘Ensuring efficient, safe and sound 

derivatives markets: Future policy actions’, 20 October 2009, COM(2009) 

563 / 4, available at ec.europa.eu / internal_market / financial-markets / docs / 

derivatives / 20091020_563_en.pdf. 

6 

See Hudson 2009, § 43-42 et seq.; and Eric Dickinson, ‘Credit Default Swaps: 

So Dear To Us, So Dangerous’, available at ssrn.com / abstract=1315535. 

228 



www.sellier.de


made regardless of whether the payee has suffered a loss, which of course 

is not true for insurance contracts. 7 

Credit default swaps are used for many purposes. The protection buyer 

might want to hedge an actual credit risk (because it is holding bonds in 

a company but is not willing to bear this risk). It may also enter into the 

credit default swap for purely speculative reasons. We have already seen 

that there is no legal requirement for the protection buyer to have an 

actual insurable risk; anyone can purchase credit protection in relation to 

a specified reference entity. This is another key difference between credit 

default swaps and insurance contracts. It is not possible for me to take out 

insurance on my neighbour’s house; I need to have an insurable interest 

in order to be able to obtain insurance. However, by entering into a credit 

default swap, I can insure against a risk that I did not run in the first place, 

which might give rise to perverse incentives because I may now have an 

interest in the reference entity going bust. This is one of the failings of 

credit default swaps which became increasingly apparent following the 

credit crisis. 8 On the other hand, as we have just observed, payments must 

be made under the credit default swap irrespective of whether there was 

an actual loss. This then is the main argument for not classifying these 

types of products as insurance contracts. 9 

The amounts involved in the credit default swap markets are truly mindboggling. 

The Bank for International Settlements publishes statistics on 

a regular basis and these show that the notional amounts outstanding 10 

7 

Hudson 2009, § 43-54. 

8 

Hershey H. Friedman and Linda W. Friedman, ‘The Global Financial Crisis of 

2008: What Went Wrong?’, available at ssrn.com / abstract=1356193, pp. 14 

& 15. 

9 

Hudson 2009, § 43-54. 

10 

The following description of this measure was taken from p. 3 of the Bank of 

International Settlements’ report on OTC Derivatives markets of November 

2009, available at http: // www.bis.org / publ / otc_hy0911.pdf?noframes=1: 

‘Nominal or notional amounts outstanding provide a measure of market size 

and a reference from which contractual payments are determined in derivatives 

markets. However, such amounts are generally not those truly at risk. 

The amounts at risk in derivatives contracts are a function of the price level 

and / or volatility of the financial reference index used in the determination of 

contract payments, the duration and liquidity of contracts, and the creditworthiness 

of counterparties. They are also a function of whether an exchange of 


229 


www.sellier.de


varied between USD 42,581 billion in June 2007 and USD 36,046 billion 

in June 2009. But even if the gross market value is taken as a measure, 

which might be a better indicator given that this is much closer to the 

actual exposures of market participants in part because notional amounts 

are not exchanged under these types of contracts, 11 the numbers are still 

very big, running into the hundreds of billions of dollars. 12 

There are many issues that are currently being discussed in relation to 

credit default swaps, which are partly blamed for the credit crunch because 

they might have increased, rather than decreased, risks in the global 

financial system. 13 A case in point is AIG, an insurance company, which 

had entered in so many of these transactions that authorities felt that 

it could not be allowed to fail. 14 These types of instruments are understandably 

under great scrutiny at this point in time and there will most 

certainly be a regulatory response. Most proposals seem to concentrate on 

the clearing and settlement of these types of instruments and increased 

transparency. 15 

notional principal takes place between counterparties. Gross market values 

provide a more accurate measure of the scale of financial risk transfer taking 

place in derivatives markets.’ 

11 

In our example the underlying amount of bonds issued by Issuer was EUR 

20,000,000. It is, however, unlikely that such an amount will actually be paid 

under the contract. 

12 

www.bis.org / statistics / otcder / dt1920a.pdf. 

13 

An overview can be found in the joint report issued by the Financial Services 

Authority and HM Treasury of December 2009, ‘Reforming OTC Derivative 

Markets A UK Perspective’, available at www.fsa.gov.uk / pubs / other / reform_ 

otc_derivatives.pdf; and Eric Dickinson, ‘Credit Default Swaps: So Dear To Us, 

So Dangerous’, available at ssrn.com / abstract=1315535. 

14 

Andrew Ross Sorkin, ‘The Case for Saving A.I.G., by A.I.G.’, New York Times, 

2 March 2009, available at www.nytimes.com / 2009 / 03 / 03 / business / economy 

/ 03sorkin.html; and William K. Sjostrom Jr., The AIG Bailout, Washington 

and Lee Law Review, 2009 Vol. 66, p. 943, available at papers.ssrn.com / sol3 / 

papers.cfm?abstract_id=1346552. 

15 

In addition to the EC Communication (COM(2009) 563 / 4) and the joint 

report of the Financial Services Authority and HM Treasury mentioned above 

(at footnotes 6 and 14 respectively), the following references might be useful: 

Christian Johnson, ‘The Enigma of Clearing Buy Side OTC Derivatives’, Futures 

& Derivatives Law Report, December 2009, Volume 29(11); Christopher 

L. Culp, ‘The Treasury Department’s Proposed Regulation of OTC Derivatives 

230 



www.sellier.de


I am not going to discuss these issues, although they are of course of great 

importance. But one thing it is worth noting is that the credit crisis has 

also shown is that having the protection of a credit default swap is only 

helpful if your counterparty, the protection seller, is able to make good on 

her promise to pay up if a credit event occurs. Of the various mechanisms 

that are used in order to mitigate this counterparty risk, a number of which 

are listed below, the provision of financial collateral – the subject matter 

of this chapter – is only one. These risk reduction instruments include: 

– 

– 

– 

Payment netting and close-out netting: reducing exposures by agreeing 

that there will only be one net payment due between parties that 

have entered into numerous contracts with differing payment obligations. 

Netting involves the cancellation of all outstanding contracts 

followed by the setting off of all amounts due so that one net payment 

remains. Netting thus reduces the number of payment obligations and 

amounts involved; 

Further hedging: entering into a matching transaction (e.g. a credit 

default swap) with a different party in order to cover the risk of the 

counterparty’s non-performance in the original transaction. 

Providing financial collateral for the amounts due under a financial 

contract: transferring collateral or providing a security interest over 

collateral in order to cover the outstanding payment obligations. 

The last of these risk reduction instruments – the providing of financial 

collateral – is the subject matter of the remainder of this chapter. 

That the provision of financial collateral will continue to play an important 

role, notwithstanding the fact that credit default swaps are subject 

to a lot of scrutiny right now and currently seem unlikely to be able to 

avoid more stringent regulation, is clear from the recent initiatives of the 

EU Commission aimed at improving the derivatives markets. The initiatives 

indicate that the Commission is planning to introduce legislation 

Clearing and Settlement’, available at ssrn.com / abstract=1430576; Stephen 

G Cecchetti, Jacob Gyntelberg and Marc Hollanders, ‘Central Counterparties 

for over-the-counter derivatives’, available at ssrn.com / abstract=1472960; 

and Craig Pirrong, The Clearinghouse Cure’, Regulation, Winter 2008-2009, 

pp. 44-51. 


231 


www.sellier.de


requiring financial firms to post collateral when entering into derivatives 

transactions. 16 

When a party has an exposure to its counterparty, providing collateral 

may of course reduce risk and this in turn may lead to less onerous 

capital adequacy requirements for an exposure against a counterparty. 

The advantage of having collateral is that the exposure against a counterparty 

is reduced by the value of the collateral. In fact, these types of 

risk-reducing arrangements are deemed so important that the EU legislator 

has adopted a Directive aimed at facilitating them, the Collateral 

Directive, 17 which has been amended recently in order to widen the 

scope of acceptable collateral from cash and tradable securities to bank 

loans (which are however ignored as potential collateral for purposes of 

this chapter). 

In this chapter I will argue that, from a legal perspective, the parties to 

financial collateral arrangements have almost unlimited freedom in establishing 

the terms and conditions of their arrangements in such a way as 

to achieve their precise objectives, at least as a matter of Dutch law. This 

freedom includes the freedom to have almost any legal system govern the 

arrangements. 

11.3. Party autonomy in relation to financial collateral 

arrangements in the Netherlands 

11.3.1. Introduction 

Let us now go back to our example. Alpha Bank, based in the Netherlands, 

has an exposure of EUR 20,000,000 against Issuer, a Dutch cor- 

16 

EC Commission Communication, ‘Ensuring efficient, safe and sound derivatives 

markets: Future policy actions’, 20 October 2009, COM(2009) 563 / 4, 

pp. 5 & 6. Other measures proposed include central clearing of derivatives 

and / or additional capital charges depending on the type of product. 

17 

EU Directive 2002 / 47 / EC (the Collateral Directive) available at eurlex.europa.eu 

/ LexUriServ / LexUriServ.do?uri=OJ:L:2002:168:0043: 

0050:EN:PDF, as amended by EU Directive 2009 / 44 / EC, available at eurlex.europa.eu 

/ LexUriServ / LexUriServ.do?uri=OJ:L:2009:146:0037:0043: 

EN:PDF. 

232 



www.sellier.de


poration specializing in telecommunications services and the issuer of a 

large number of bonds. Alpha Bank has hedged this exposure by entering 

into a credit default swap with Rock Solid Bank, also based in the Netherlands. 

Under this credit default swap, in the case of a default of Issuer, 

Alpha Bank will receive payments from Rock Solid Bank in order to 

make good its loss. The maximum exposure of Rock Solid Bank under 

this arrangement is EUR 20,000,000, i.e. the total amount due from Issuer 

to Alpha Bank. The parties have agreed that Rock Solid Bank is to 

provide financial collateral to Alpha Bank in order to limit this exposure. 

At a certain point in time this means that collateral with a value of EUR 

2,000,000 needs to be provided. 18 How could this be effected? There are 

at least four options: 

(i) Rock Solid Bank could post 19 cash collateral with Alpha Bank 

through a title transfer arrangement; 

(ii) Rock Solid Bank could charge or pledge 20 a cash balance that it holds 

with either Alpha Bank or a third party; 21 

18 

The value of the collateral to be provided depends on the contract between 

the protection seller and the protection buyer and is often arrived at through 

the use of complicated mathematical models of credit risk. Moreover, since the 

collateral itself may consist of debt securities and since these types of securities 

are constantly subject to changes in value, there may be various moments during 

the lifetime of the credit default swap where additional collateral needs to 

be provided or where collateral that has previously been provided may have to 

be returned to the collateral provider. I will leave all these complications to a 

side and just assume that there is a need to provide collateral in the amount 

(i.e. with a market value) of EUR 2,000,000. 

19 

‘Posting’ collateral is the industry jargon for collateral arrangements that involve 

an outright transfer of collateral. 

20 

The term ‘pledge’ is used here as understood under Dutch law: the standard 

form security interest over an asset (generally) entitling the holder to sell the 

asset or (if it is a claim) collect it from the debtor and keep the proceeds. In 

an English law context, the term may have a more restrictive meaning (see 

Hudson 2009, § 45-25) and the standard form security interest would be a 

charge. In the remainder of this chapter, I only use the term pledge because 

my reference point is Dutch law security interests. 

21 

Since the credit crunch, market practice has been that a third party custodian 

is no longer acceptable in light of such custodian’s additional insolvency 

risk. An exception is Euroclear Bank, the Brussels (Belgium) based financial 

services company that specializes in the settlement of securities transactions 

as well as the safekeeping and asset servicing of these securities. 


233 


www.sellier.de


(iii) Rock Solid Bank could post securities collateral with Alpha Bank 

through a title transfer arrangement; or 

(iv) Rock Solid Bank could charge or pledge securities it holds with Alpha 

Bank or in an investment account with a different intermediary 

for the benefit of Alpha Bank. 

Once EU Directive 2009 / 44 / EC, amending the Collateral Directive, 22 has 

been implemented, collateral may also take the form of credit claims, i.e. 

‘pecuniary claims arising out of an agreement whereby a credit institution 

(…) grants credit in the form of a loan’. This type of collateral will be 

ignored for the purpose of this chapter because cash and securities would 

seem to be more appropriate for collateralizing credit default swaps owing 

to their liquidity. 

In the following paragraphs, I will be considering the level of party autonomy 

available for each of the four types of arrangements listed above. 

11.3.2. Cash collateral transfer arrangements 

(posting cash collateral) 

What happens from a legal perspective if Rock Solid Bank posts cash collateral 

with Alpha Bank? First, the parties will probably enter into some 

sort of contractual arrangement in addition to the credit default swap, 

specifying when the cash must be posted, whether Alpha Bank is under 

an obligation to keep the cash in a separate account, the circumstances 

under which Alpha Bank will be allowed to appropriate the cash, when 

it should be returned (if it is to be returned), etc. The banks, being professional 

parties active in the financial markets, should be able to agree 

almost everything in this respect without the provisions of mandatory 

law standing in the way of their party autonomy. They can (and are likely 

to) use one of the standard form contracts provided by the International 

Swaps and Derivatives Association (ISDA). 23 In our example, the parties 

22 

The Collateral Directive. 

23 

ISDA is an important organization in the field of derivatives transactions, 

amongst other things providing standard documentation and arranging for 

legal advice as to the enforceability of this documentation in different jurisdictions. 

For further information see www.isda.org. A general description of 

the ISDA documentation can be found in Hudson 2009, chapters 44 and 45, 

234 



www.sellier.de


are likely to use the ISDA Master Agreement to document their credit 

default swap and the Credit Support Annex that goes with it. 24 

There may be some doubt as to whether the parties would be free to 

choose the law applicable to their arrangement. The most obvious standard 

form precedent for the provision of the collateral would be the ISDA 

Credit Support Annex (which is used to create a title transfer arrangement) 

and this is usually governed by English law. Since in our example 

both contracting parties are Dutch, the EU Regulation on the law applicable 

to contractual obligations (Rome I) 25 might not allow a deviation 

from the provisions of mandatory Dutch law. Although the parties to 

contractual arrangements are generally free to choose the applicable law 

(article 3(1) Rome I), article 3(3) Rome I restricts the ability to deviate 

from the provisions of mandatory law if all elements relevant ‘to the situation’ 

are located in another country. 26 This may restrict the parties in 

our example from deviating from the Dutch provisions of mandatory law 

since all elements relevant to the situation are Dutch – even the reference 

entity is Dutch. However, the fact that the parties have entered into a 

number of international contracts in this situation (a credit default swap 

perhaps entered into on the basis of an English law governed ISDA Master 

Agreement and a Credit Support Annex, that is also likely to be governed 

by English law) would seem to be sufficient to allow them to also choose 

English law in respect of mandatory provisions. These contracts may be 

seen as inherently international even if, in a specific case, the parties 

happen to be Dutch. 

The parties could choose Dutch law to govern the documentation but a 

major drawback would be that these documents were simply not designed 

to be governed by Dutch law. Terminology, legal concepts and references 

where chapter 44 deals with the ISDA Master Agreement and chapter 45 

with collateralization. 

24 

A general discussion of this documentation can be found in Hudson 2009, 

chapters 44-46. 

25 

EU Regulation (EC) No 593 / 2008, available at eur-lex.europa.eu / LexUri- 

Serv / LexUriServ.do?uri=OJ:L:2008:177:0006:0016:EN:PDF. 

26 

The full provision reads as follows: ‘Where all other elements relevant to 

the situation at the time of the choice are located in a country other than 

the country whose law has been chosen, the choice of the parties shall not 

prejudice the application of provisions of the law of that other country which 

cannot be derogated from by agreement.’ 


235 


www.sellier.de


have all been inspired by the English legal system. This is not to say that 

they cannot be amended fairly easily so as to be suitable to be governed 

by Dutch law, but the parties may still not be willing to do so preferring to 

use the documents as they are. Moreover, applying the mandatory provisions 

of Dutch law instead of the standard English law provisions would 

seem to serve no purpose at all in our example, where the professional 

parties do not need the protection of mandatory Dutch law and so there 

should be no recourse to Dutch law / and so Dutch law provisions should 

be irrelevant if this reflects the parties’ preferences. 27 If the parties had 

instead chosen New York law as applicable to their arrangement, the 

provisions of article 3(4) Rome I, restricting the ability to deviate from 

mandatory community law if all elements relevant to the situation are 

located in a Member State, could still be relevant. 28 However, for much 

the same reasons as mentioned immediately above, professional parties 

in my view should have the freedom to apply New York law to their collateral 

arrangement without being subject to the restrictive provisions of 

article 3(4) Rome I. Restricting party autonomy in this situation would 

not seem to serve any purpose. 

These matters are of course only relevant if there are in fact mandatory 

provisions under Dutch or community law that would apply to the transaction, 

which does not appear to be the case. Indeed, I cannot think of any 

rule that might be relevant in this respect as long as the analysis is limited 

to the contractual and contract law aspects of the arrangement. 29 Having 

said that, articles 3(3) and 3(4) of Rome I might in principle be viewed 

as a restriction on party autonomy. The question then could be whether 

there are ways around these restrictions. Could the parties, by posting 

the collateral in an account outside the Netherlands, for instance, make 

the transaction sufficiently international in order to avoid application of 

articles 3(3) and (4) of Rome I? In my view this would in fact solve the 

27 

For the Netherlands: R.I.V.F. Bertrams and S.A. Kruisinga, Overeenkomsten 

in het internationaal privaatrecht en het Weens Koopverdrag, Kluwer, December 

2007, p. 25, with further references. 

28 

The full provision reads as follows: ‘Where all other elements relevant to the 

situation at the time of the choice are located in one or more Member States, 

parties’ choice of applicable law other than that of a Member State shall not 

prejudice the application of provisions of Community law, where appropriate 

as implemented in the Member State of the forum, which cannot be derogated 

from by agreement.’ 

29 

The property law aspects will be discussed presently. 

236 



www.sellier.de


problem because then the balance of the relevant elements of the transaction 

could well shift in favour of either England or New York. Perhaps, if 

the reference entity or the protection seller were a non-Dutch entity, the 

balance would shift even more. But this sort of scheming, which would 

only be possible if the transaction is entered into for speculative reasons 

because only then might it be possible to choose a different reference 

entity, should not be necessary. Because the documentation that is used 

for these types of transactions has been developed in an international context, 

and these types of contracts are therefore inherently international, 

the mere fact that there happen to be two Dutch parties that have entered 

into the contract should not deprive them from the option of choosing 

an applicable law other than Dutch law. The provisions of articles 3(3) 

and 3(4) of Rome I were developed to prevent parties from abusing the 

possibility to choose the applicable law to avoid certain requirements of 

mandatory law, but that is clearly not what happens in our scenario. The 

parties may simply want to use standard form documentation and avoid 

the uncertainty that might result from applying Dutch law to documentation 

that was designed for English law or the costs involved in amending 

the documents for Dutch law. 30 

Another restriction on the freedom to choose the law applicable to the 

contractual elements of the collateral arrangement can be found in article 

9 Rome I, dealing with mandatory provisions crucial for safeguarding 

public law interests, which can never be deviated from by a choice of 

law. These types of provisions are usually not relevant for collateral arrangements 

due to the narrow scope of their application. There would 

seem to be no mandatory provisions of such a grave nature that could 

interfere with the type[s] of contracts that we are currently dealing with. 

The restriction to party autonomy imposed by article 9 Rome I is therefore 

unlikely to be relevant to the types of transactions described above. 

30 

This is not to say that choosing Dutch law as the law applicable to the arrangements 

will not also offer benefits, depending on the exact circumstances of the 

case. It is not very difficult to adapt the ISDA documentation in such a way 

that it can be governed by Dutch law. If this is combined with submission to 

the jurisdiction of the Dutch courts, the cost benefits may be substantial. The 

main topic of this chapter is parties’ autonomy in arranging their collateral 

structures and evidently there would be no problems in this respect if two 

Dutch entities were to choose Dutch law and submit to the jurisdiction of the 

courts of the Netherlands. 


237 


www.sellier.de


In conclusion then, parties to a credit default swap that want to enter into 

a collateral arrangement would be unlikely to run into too much trouble in 

terms of restrictions to their party autonomy. They can choose English or 

New York law to govern their arrangement and standard documentation is 

readily available. The contracts can be signed and the collateral can subsequently 

be provided in accordance with the terms of these contracts. 

There is, however, a separate problem in this scenario. Is the actual provision 

of collateral not a matter of property law rather than contract law, 

contrary to what has been assumed up until now in this chapter? Should 

there not be a proprietary transfer of collateral which is mandatorily governed 

by Dutch law? And does this not form a serious restriction of the 

parties’ autonomy? 

In order to actually post the collateral, the collateral provider (Rock Solid 

Bank) will have to transfer an amount of cash to Alpha Bank to cover 

the risk of non-payment under the credit default swap. This will take 

the form of a bank transfer. There is no property law involved because 

bank transfers are a matter of contract law. Rock Solid Bank will simply 

instruct its bank to transfer an amount of funds to Alpha Bank However, 

the word ‘transfer’ is in fact misleading here. An ordinary bank transfer 

takes the form of a reduction of the balance held by the ‘transferor’ with 

its own bank followed by an increase in the balance of the ‘transferee’. 

For example, if I transfer 100 EUR to my friend Paul, the balance with 

my bank will be reduced by EUR 100 and Paul’s balance with his bank 

will be increased by the same amount. There is no actual transfer or other 

property law aspect involved in the entire transaction. 31 The parties (Alpha 

Bank and Rock Solid Bank in our example) will maintain accounts 

with one another in which they record the amount of collateral that has 

been provided in accordance with the credit support annex. 32 The actual 

transfers are made in accordance with the instructions of the transferee 

or transferor as applicable by transfer into one or more bank accounts 

specified by the recipient. 33 

31 

See, for English law, Charles Proctor and Roy Goode, Goode on payment obligations 

in commercial and financial transactions, 2 nd edition, Sweet & Maxwell, 

London 2009, § 4-20, and Hudson 2009, § 2-10; and for Dutch law, R.J. 

Abendroth and R.M. Wibier, ‘Giraal betalingsverkeer en het faillissement van 

de rekeninghouder’, WPNR 6752, 2008, pp. 324-332. 

32 

Hudson 2009, § 45-43. 

33 

ISDA Credit Support Annex, § 3(a). 

238 



www.sellier.de


The analysis given above means, in my view, that the parties are not restricted 

in their freedom to give their collateral arrangement the contents 

that they prefer. Party autonomy in cash collateral arrangements extends 

very widely indeed due to the fact that the entire arrangement is based 

on contract law, which generally offers far wider freedoms than property 

law and, moreover, appears to contain no formal limitations, in terms of 

substantive law, on these types of arrangements. 

If there are no actual mandatory provisions that limit party autonomy in 

the legal system with which the situation has ‘a close connection’ 34 the 

question of whether it is possible to deviate from such (non-existing) 

provisions automatically becomes moot. It may be useful to note that this 

would also be true if there was no such thing as the Collateral Directive, at 

least as a matter of Dutch law and probably also as a matter of English law. 

In other words, the Collateral Directive was not needed to enable cash 

collateral arrangements where the cash is ‘transferred’ to the collateral 

taker because Dutch substantive law did not contain any rules limiting 

party autonomy even before the Collateral Directive was implemented. 

Even the provision prohibiting the transfer of assets for security reasons 

(article 3:84(3) of the Dutch Civil Code) has never been relevant, because 

cash ‘transfers’ by way of book entries are not actually transfers in 

a legal sense. 

11.3.3. Pledging a cash balance 

If the cash collateral takes the form of a security right, the most common 

approach, from a Dutch legal perspective, would be to pledge a cash balance. 

Much of the same issues that were just discussed would also arise 

in this scenario. The main question would again be whether Alpha Bank 

and Rock Solid Bank would be able to choose English law (or New York 

law) as the law applicable to their (contractual and proprietary) pledge 

arrangement. ISDA has also developed a standard form document for 

pledging cash, the Credit Support Deed, which is part of the ISDA Master 

Agreement and is itself governed by English law. 

The main difference from the posting cash collateral scenario is that when 

the parties choose to create a security interest over a cash balance, it may 

not be entirely possible to avoid the application of the mandatory provi- 

34 

Article 4(3) Rome I. 


239 


www.sellier.de


sions of Dutch property law. In particular, if the security interest takes the 

form of the pledge of a cash balance in a bank account held with a Dutch 

bank, it is unlikely that Dutch provisions on pledging cash balances can 

be avoided altogether. 

There are basically two forms that a pledge might take. The first would be 

for the collateral provider (Rock Solid Bank) to open an account with the 

collateral taker (Alpha Bank) and post the relevant cash amount into this 

account. The account would be in the name of Rock Solid Bank, which 

would consequently be the owner of the balance held in this account. 

Subsequently, the account should be pledged to Alpha Bank to effect the 

collateral arrangement. As a matter of Dutch law, this would require a 

deed of pledge, followed by giving notice to the debtor of the claim, which 

in this example would be Alpha Bank itself. 35 These requirements have 

not been amended or simplified by the implementation of the Collateral 

Directive in the Netherlands. Instead, the Dutch legislature has taken the 

view that requiring a deed of pledge and giving notice to the debtor of 

the claim is not inconsistent with the Collateral Directive. I have argued 

elsewhere 36 that this view might not be correct, because the Collateral 

Directive does not allow for any formalities other than placing collateral 

within the control of the collateral taker, which might be effected through 

a mere transfer of the funds to Alpha Bank. Requiring the execution of 

a deed of pledge, which must be signed by the pledgor, may be seen as 

an additional formality inconsistent with the Collateral Directive. From 

a practical perspective this may not be too significant since in any event 

the parties are likely to execute a document – probably the standard form 

credit support deed issued by ISDA. 37 The execution of this document 

would almost certainly be sufficient to meet the Dutch law requirements 

for pledging an account balance, provided that it has been signed by the 

collateral provider (Rock Solid Bank in our example). The only argument 

35 

These requirements follow from Article 3:94(1) of the Dutch Civil Code 

which deals with the assignment of contractual rights. It is applicable to 

pledges pursuant to Article 3:236(2) of the Dutch Civil Code. 

36 

R.M. Wibier, ‘Financial Collateral in the Netherlands, England and under 

the EU Collateral Directive – Traditional and novel ways of taking security 

over cash and securities’, Journal of International Banking Law and Regulation, 

Volume 24, 8, 2009 (Wibier 2009), pp. 388-411, available at ssrn.com / 

abstract=1287095. 

37 

Which it might then be useful to amend for the avoidance of doubt so as to 

indicate that a pledge is being created; see Hudson 2009, § 45-52. 

240 



www.sellier.de


why this might not be the case would be that the document is governed 

by English law and not by Dutch law and that to qualify as a Dutch law 

deed of pledge, the document purporting to be a deed of pledge should be 

governed by Dutch law. In my view, this is a very weak ground indeed for 

invalidating the pledge. For one thing, this requirement is nowhere to be 

found in the relevant provisions of the Dutch Civil Code and the Code 

on Civil Proceedings. Any signed document that is intended to prove 

something qualifies as a ‘deed’ 38 as a matter of Dutch law on the basis of 

the explicit provisions of article 156 of the Code on Civil Proceedings. 39 

Also, a more restrictive interpretation would be completely inconsistent 

with the Collateral Directive and its requirements. 

The pledge would take the form of a charge-back in our example. 40 The 

cash balance is a claim owed by Alpha Bank to Rock Solid Bank and by 

pledging this claim to Alpha Bank, Rock Solid Bank would have a pledge 

over a claim against itself. Although the practice of charge-backs is controversial 

under English law, there seems to be little doubt that these types 

of arrangements are valid. As regards Dutch law, the validity of chargebacks 

as such has never been subject to serious doubt. 41 

38 

Note that the sense of ‘deed’ is different under English law, where it means a 

document in a specific form. 

39 

In my own translation, Article 156 of the Code on Civil Proceedings reads as 

follows: ‘Deeds are signed documents, intended to be used as proof.’ 

40 

For a discussion of the validity of charge-backs as a matter of English law see: 

Roy Goode, Commercial Law in the Next Millennium, Sweet & Maxwell, 

London 1998 (Goode 1998), pp. 69-71. See also Roy Goode and Louise Gullifer, 

Goode on Legal Problems of Credit and Security, 4 th edition, Sweet & 

Maxwell, London 2008 (Goode-Gullifer 2008), § 3-12; [[re] Bank of Credit 

and Commerce International SA (No 8) [1998] AC 214; [1997] 4 All ER 

568; Philip Wood, Comparative Law of Security Interests and Title Finance, 

2 nd edition, Sweet & Maxwell, London 2007 (Wood 2007b), § 29-025; F. Oditah, 

‘Financial Trade Credit: Welsh Development Agency v. Export Finance 

Corporation’ [1992] J.B.L. 557; Roy Goode, ‘Charge-backs and legal fictions’, 

Law Quarterly Review, 1998, 114 (Apr), pp. 178-181; and Edwin C. Mujih, 

‘Legitimising Charge-Backs’, Insolvency Lawyer, 2001, 1 pp. 13-16 for an overview 

of the arguments for and against charge-backs. The general view would 

seem to be that charge-backs are in fact valid as a matter of English law. 

41 

See Wibier 2009, pp. 388-411 for a further discussion of this issue. 


241 


www.sellier.de


The second form the pledge of cash collateral might take would be for 

the collateral provider (Rock Solid Bank) to pledge a cash balance that it 

holds with a different bank to Alpha Bank This might be the Dutch Central 

Bank, but in principle it could just as easily be another commercial 

bank. In practice, only the former option might be available, the reason 

being that posting the collateral with another commercial bank would 

introduce the risk that this bank would go bankrupt and that the collateral 

would (at least temporarily) become stuck in the bank’s bankruptcy 

estate. This risk might not be acceptable in a transaction where the aim 

is to reduce risks, even if the commercial bank were of good standing and 

had a high credit rating. 42 

Pledging an account balance with another bank would equally require a 

deed of pledge signed by the collateral provider (Rock Solid Bank) followed 

by notice to the account bank. In addition, the possibility that the 

account bank may have a right of set off or its own right of pledge over 

the balance should be taken into account. If this is the case, the account 

bank should waive these rights to prevent them from interfering with 

Alpha Bank’s pledge. 43 

The above shows that using a right of pledge over cash does in fact mean 

that there are some restrictions on the parties’ freedom because they will 

need to comply with the requirements of Dutch property law in relation to 

the creation of pledges. This restriction of party autonomy may, however, 

be very limited. It would probably still allow the use of the ISDA standard 

form precedent governed by English law and since the parties were probably 

going to execute this documentation in any event, it might not be 

seen as a true restriction of their autonomy. 

42 

As noted before, Euroclear Bank, the Brussels (Belgium) based financial services 

company that specializes in the settlement of securities transactions as 

well as the safekeeping and asset servicing of these securities, might be the 

exception. 

43 

In theory there would also be the possibility of pledging the account balance 

without giving notice to the account bank pursuant to Article 3:239 of the 

Dutch Civil Code. 

242 



www.sellier.de


11.3.4. Securities collateral transfer arrangements 

(posting securities collateral) 

Securities, and in particular debt securities, are also a very useful form 

of collateral. Indeed, they are often regarded as (almost) equivalent to 

cash, especially if the debt has been issued by the central government of 

the United States, Germany, the Netherlands or a comparable country 

with a good credit rating and solid past performance in terms of debt 

repayment. If the collateral arrangement takes the form of an outright 

transfer, the collateral provider (Rock Solid Bank) will transfer the relevant 

securities to the collateral taker (Alpha Bank). Alpha Bank has 

an obligation to return equivalent collateral, i.e. securities of the same 

sort but not the exact same securities transferred to it by Rock Solid 

Bank, once the collateral provider has performed its obligations under 

the credit default swap or the contract has otherwise been terminated 

(for instance by Issuer having performed all its obligations in full under 

the bonds). 

The parties can use the ISDA Credit Support Annex in order to effect 

their collateral arrangement. As to whether they have the freedom to 

choose English law, the same considerations apply as those discussed in 

§ III.2. My conclusion there was that the parties do have this freedom and 

that even if they did not, this would in practice probably not be relevant. 

Party autonomy would include the freedom to determine that the collateral 

will be subject to an outright transfer for the benefit of the collateral 

taker (as opposed to a mere security interest), which would include the 

freedom to use the collateral as the collateral taker sees fit. The parties 

would also be free to set the terms and conditions under which equivalent 

collateral should be returned. Also, the arrangements are likely to include 

an obligation to return part of the collateral if it has gone up in value so 

as to exceed the amount of collateral to be provided. If the collateral has 

decreased in value, Rock Solid Bank is likely to have an obligation to post 

additional collateral. 44 

This form of security (i.e. an outright transfer by way of security) is not 

available under the general rules of creating security in the Dutch Civil 

Code. Article 84(3) of Book 3 of the Dutch Civil Code explicitly states 

that a transfer by way of security is invalid. However, if the arrangement 

qualifies as a financial collateral arrangement under the legislation im- 

44 

Hudson 2009, § 45-12. 


243 


www.sellier.de


plementing the Collateral Directive in the Netherlands (which would be 

the case in our example) an outright transfer would be possible. Article 

3:84(3) of the Dutch Civil Code has been explicitly disapplied in the 

legislation implementing the Collateral Directive in the Netherlands, 

meaning that it does not impose restrictions on party autonomy. 45 

Contrary to the cash collateral scenario, the question of whether mandatory 

rules of property law apply to the actual provision of the collateral 

is very relevant if securities are to be transferred. Securities are generally 

held through chains of intermediaries and there are specific property law 

provisions that apply to transferring these securities. Although bearer 

certificates (i.e. pieces of paper representing the interest of the investor 

against the issuer) were once an innovative way of transferring rights 

against the issuers of securities, in today’s world bearer certificates have 

become almost irrelevant. Instead, the most important aspect of securities 

transfers has become the account the investor has with its intermediary 

where the investor is credited with the securities. 46 A securities account 

is much like a cash account held with a bank, allowing for the transfer 

of fungible securities held in the account by way of book entries. This 

means that for a valid collateral arrangement there must be an entry in 

the books of the intermediary where the collateral provider (Rock Solid 

Bank) holds its securities, debiting the amount of securities specified, and 

a corresponding booking in the account of the collateral taker (Alpha 

Bank), crediting its account. This booking must comply with the provisions 

of property law applicable to the account, which, pursuant to the 

private international law requirements of the Collateral Directive as implemented 

into Dutch law, will be the property law for the jurisdiction 

in which the account is maintained. 47 This means that if the account is 

maintained in the Netherlands, there must be a booking that complies 

with the Dutch Securities Transfer Act. These bookings are very similar 

45 

Article 7:55 of the Dutch Civil Code. 

46 

See Roy Goode, Legal problems of credit and security, 3 rd edition, Sweet & 

Maxwell, London 2003 (Goode 2003), § 6-02 for a description of this development. 

See also Wood 2007b, §§ 29-052 and 29-053; Matthias Haentjens, Harmonisation 

of Securities Law – Custody and Transfer of Securities in European 

Private Law, Kluwer Law International, 2007 (Haentjens 2007), pp. 29-41; and 

Joanna Benjamin, Interests in Securities, Oxford University Press, Oxford 2007 

(Benjamin 2007), §§ 1.78-1.92. 

47 

Article 16 of the Dutch Property (Conflicts of Law) Act (Wet Conflictenrecht 

Goederenrecht), which implements Article 9 of the Collateral Directive. 

244 



www.sellier.de


to bookings made for cash ‘transfers’ and therefore should not be too 

much of a problem, but nevertheless, in this case the parties would, it 

seems, not have the freedom to deviate from the relevant (Dutch) provisions. 

If the securities are held in an account outside the Netherlands, the 

laws of that jurisdiction in relation to transfers should be complied with 

in accordance with the provisions of private international law mentioned 

immediately above. 

Other than this, there seem to be no limitations on the freedom of the 

parties to shape the collateral arrangement as they like. Specifically there 

are no documentary requirements in the Netherlands for outright transfer 

arrangements in relation to fungible securities. 

11.3.5. Pledging securities collateral 

Perhaps counter-intuitively, there are more restrictions on party autonomy 

if the parties choose to use a security right instead of an outright transfer 

as their preferred collateral arrangement. We have already seen this 

with the previous example of cash collateral. Equally counter-intuitively, 

however, using securities collateral may actually reduce the (mandatory) 

property law requirements, if compared to cash collateral arrangements, 

because there would be no need for a formal deed of pledge on the basis 

of Dutch substantive law. This is so because the Dutch Securities Transfer 

Act enables the pledging of securities through a (mere) book entry. Under 

the Dutch Securities Transfer Act, a right of pledge in favour of the 

intermediary where the securities are held (in our example if Rock Solid 

Bank holds its securities with Alpha Bank) is simply created by a pledge 

agreement. There are no reasons why the ISDA credit support deed would 

not qualify for these purposes. 48 If the right of pledge is in favour of a third 

party (i.e. not the intermediary where the securities account is held), 

there is not even a need for a pledge agreement: a book entry indicating 

the pledge will suffice. This may be a booking into the pledgee’s (pledge) 

account or into the securities account of the pledgor. 49 This may also be 

effected by flagging the specific securities account as having been pledged 

to a third party. In this scenario, there are no formal requirements for a 

48 

Other than the rather weak argument, dismissed by me above in § 3.2, that 

the document should be governed by Dutch law in order to qualify as a deed 

of pledge. 

49 

Article 20 of the Dutch Securities Transfer Act. 


245 


www.sellier.de


deed of pledge in any specific form, leaving the parties with a great deal 

of party autonomy to shape their arrangement. 

Consequently, Rock Solid Bank can pledge securities held by it with either 

Alpha Bank or with another intermediary on the basis of the ISDA 

standard form documentation, possibly followed by a book entry which 

must be in accordance with the laws of the jurisdiction where the securities 

account is maintained. Other than that, there would seem to be no 

significant restrictions to party autonomy that might restrict the parties’ 

freedom. 

11.3.6. Enforcement issues 

The final question then becomes one of whether the security taker would 

be restricted in its enforcement rights. This is important because enforcement 

will have to be effected in the Netherlands, possibly within the 

context of Dutch law governed insolvency proceedings in our example. 50 

Because enforcement is outside the scope of this chapter, I will limit myself 

to a few remarks of a more general nature. 

One of the most straightforward enforcement scenarios would be for Alpha 

Bank to sell the collateral and to apply the proceeds of the sale towards 

the secured obligations. This would not lead to any problems as a 

matter of Dutch law, since the sale of collateral in the relevant markets 

is an accepted method of enforcement under Dutch law in these types of 

arrangements. 51 

As a matter of English law (which we have assumed to be the law of 

choice of Rock Solid Bank and Alpha Bank for their credit default swap 

and collateral documentation), upon certain events (primarily events of 

default under the documentation) the security taker is authorized ‘to deal 

50 

A Dutch Bank will be subject to Dutch insolvency proceedings pursuant to 

Article 3(1) of EU Directive 2001 / 24 / EC on the reorganization and winding 

up of credit institutions, as implemented in Article 212h of the Dutch Bankruptcy 

Code and Article 3:202 of the Dutch Financial Supervision Act. 

51 

See Article 3:250(2) of the Dutch Civil Code which allows for the sale of assets 

tradable in an exchange in the markets even if the relevant assets do not 

qualify as financial collateral. 

246 



www.sellier.de


with the collateral as though its absolute owner’. 52 This might be problematic 

under Dutch law in cases where a security right has been created 

(as opposed to transfer arrangements) because a pledgee may generally 

only sell the relevant assets and keep the proceeds of the sale up to the 

amount of the secured liabilities. However, if the arrangement qualifies 

as a financial collateral arrangement, the Dutch legislation implementing 

the Collateral Directive would enable the secured party to appropriate 

the collateral or to set off the secured obligations against the value of the 

collateral. The most important Dutch law provision that has been disapplied 

for this purpose is the prohibition on appropriation 53 of the pledged 

securities which would normally apply pursuant to article 3:235 of the 

Dutch Civil Code. 54 Consequently, there appear to be no restrictions in 

comparison with the possibilities that the secured party has under English 

law in this respect. Even in terms of enforcement, there would seem to be 

party autonomy and very few restrictions indeed (if any). 

11.3.7. Final remarks 

There are three further issues that have not yet been discussed and that 

will only be briefly touched upon here. The first is that a choice of jurisdiction 

for applicable law should preferably be accompanied by a submission 

to the jurisdiction of the relevant courts. An English court applying English 

law is probably preferable to a Dutch court having to apply English 

law (simply because it will be more difficult for the Dutch court to fully 

grasp what English law entails). This generally does not cause any prob- 

52 

Hudson 2009, § 45-61. 

53 

The term ‘appropriation’ is used here to indicate that the collateral taker takes 

the relevant collateral as his or her own upon enforcement and sets off the 

secured obligations against the value of the collateral, without being under 

the obligation to sell or otherwise realize the collateral. 

54 

Nor is there a general right of appropriation under English law. Appropriation 

as a new remedy for the secured creditor has been introduced in England pursuant 

to the Collateral Directive. See Joanna Benjamin, ‘Financial Collateral 

Arrangements: Lessons from Cukurova’, Insolvency Intelligence, 2008, 21(5), 

pp. 65-72. The article also offers a view on what is required for the appropriation 

of financial collateral. 


247 


www.sellier.de


lems since the parties have the freedom to submit disputes to the courts 

of their jurisdiction of choice. 55 

Secondly, if only Dutch parties are involved in the collateral arrangement, 

it may be beneficial not to choose English or New York law but to 

adapt the ISDA documentation for Dutch law. If this is combined with 

a submission to the jurisdiction of the Dutch courts, this may sometimes 

lead to substantial cost benefits, depending on the precise circumstances 

of the case. Dutch law offers sufficient flexibility to have the ISDA documentation 

governed by Dutch law, although it would need some minor 

amendments. However, no problems would arise either in terms of contract 

law or in terms of proprietary issues. Ultimately, whether English, 

New York law or Dutch law is most suitable for the parties would be a 

matter of preference. I have of course focused on English (and to a lesser 

extend New York law) because that is where problems might be expected 

in terms of party autonomy (although, as I have argued, such problems are 

largely absent). As regards Dutch law, there would be no issues in terms 

of party autonomy and therefore Dutch law in this respect has not been 

dealt with in any detail in this chapter. 

Finally it may be of interest to know that Euroclear Bank, which is located 

in Brussels, Belgium, offers various services to market participants 

in terms of both cash and securities collateral arrangements. Euroclear 

Bank is often seen by market participants as a safe entity to use when 

placing the relevant collateral. Euroclear Bank’s services can help reduce 

administrative tasks for the collateral provider and collateral taker. It also 

means that parties using these services may prefer Belgian law as the law 

applicable to their collateral arrangement and submit to the jurisdiction 

of the courts of Belgium. 56 

55 

See Article 23 of Council Regulation (EC) No 44 / 2001 of 22 December 2000 

on jurisdiction and the recognition and enforcement of judgments in civil and 

commercial matters. See Hudson 2009, § 6-27. 

56 

Euroclear Bank’s website: https: // www.euroclear.com / site / publishedFile?D 

ocumentName=MA0060 Collateral management broch_tcm86-132206. 

pdf&action=dload. 

248 



www.sellier.de


11.4. Contractual alternatives for collateral arrangements 

11.4.1. Intermediate conclusion: 

very few restrictions on party autonomy 

The above has shown that the parties to financial collateral arrangements 

are not likely to run into too much trouble in terms of limitations to their 

party autonomy. They will be able to choose the law applicable to their arrangements 

and they will thus be able to use the standard form documentation 

that is more or less commonplace for these types of transactions. 

Also, since the provisions of mandatory property law are very limited in 

this respect, the most that may be required is a deed of pledge that has 

been signed by the pledgor or compliance with book-entry requirements 

for cash and fungible securities. Implementation of the Collateral Directive 

has had a further positive effect on party autonomy since it has disapplied 

a number of restrictive provisions. 57 The main reason why the parties 

enjoy so much freedom is, it seems, that their arrangements are largely of a 

contractual nature and that restrictive and mandatory property law seems 

to be of very limited relevance indeed. 

In this section 4 I will briefly set out my argument that there may 

be even greater opportunities of increasing party autonomy. My main 

point is that collateral arrangements may be given a form that is even 

more contractual than has been so far suggested in this chapter. For this 

purpose, I will first briefly discuss what a bank balance is from a legal 

perspective (§ 11.4.2). I will argue that it is no more than a contractual 

claim of the deposit holder against the bank. Subsequently, I will take 

the position that this allows for alternative approaches to taking security 

over bank balances because the bank and the account-holder have the 

freedom to manipulate this contractual claim (§ 11.4.3.). Finally, I will 

discuss what this means for collateral arrangements involving fungible 

securities (§ 11.4.4). 

57 

The most important restriction that has been lifted is the prohibition on 

transfers for security reasons which applies under general Dutch law (Article 

3:84 of the Dutch Civil Code). Also, the enforcement of security entails fewer 

formalities and rules if the security arrangement qualifies as a financial collateral 

arrangement. 


249 


www.sellier.de


11.4.2. What is a bank balance? 58 

As a matter of Dutch law, if a person has a deposit with a bank, such 

person has a contractual claim against the bank. The view that a positive 

account balance or a deposit constitutes a claim against the bank seems 

to be widely accepted internationally. 59 This is shown in the use of the 

words ‘or similar claims for repayment of money’ in the very definition of 

cash in the Collateral Directive, which strongly supports the conclusion 

that money held in an account also gives rise to a claim for repayment. 60 

Another way of looking at this is to consider a deposit with a bank as the 

right to obtain judgment against the bank for the value of the deposit, 61 a 

view that is often accepted in an American context. 62 Another interpretation, 

which is often applied in an English context, is that a bank deposit 

is a chose in action: 

‘The depositor’s right to claim payment of his deposit is a chose in action 

which the law has recognized as property.’ 63 

These interpretations all are in line with the Dutch view that a deposit 

constitutes a contractual right of the depositor against the bank. 

Generally, the freedom to determine the contents of a contract rests with 

the parties to the contract. This enables the parties to a claim that re- 

58 

Parts of this § 4.2 and the following § 4.3 were previously published in Wibier 

2009, pp. 388-411. 

59 

See Hudson 2009, § 30-01 for English law. 

60 

The full definition reads as follows: ‘‘cash’ means money credited to an account 

in any currency, or similar claims for repayment of money, such as money 

market deposits’. 

61 

Taken from Joseph H. Sommer, ‘Where is a bank account?’, Maryland Law 

Review, 1998, 57, 1. 

62 

Also see Brian M. McCall, ‘Money everywhere but not a drop to secure: a 

proposal for amending the perfection rules for security interests in money and 

deposit accounts’, Tennessee Law Review, 2007, Vol. 74:669. 

63 

Lord Hoffmann in: House of Lords, In re Bank of Credit and Commerce International 

SA (No 8) [1998] AC 214; [1997] 4 All ER 568. There has been 

consensus on this point for some time. See Edwin C. Mujih, ‘Legitimising 

Charge-Backs’, Insolvency Lawyer, 2001, 1 pp. 3-16 with further references. 

See also E.P. Ellinger, E. Lomnicka and R.J.A. Hooley, Ellinger’s Modern Banking 

Law, 4 th edition, Oxford University Press, Oxford 2006, p. 122. 

250 



www.sellier.de


sults from a contract to determine the exact contents of the claim. For 

instance, the parties to a claim resulting from a contract for the sale of 

goods can determine that the claim for payment of the purchase price will 

be payable only upon delivery of the goods. As a result of such a provision, 

the seller would have a claim against the buyer that is contingent upon 

delivery of the goods. Likewise, the parties to the sale contract can determine 

that this claim will be non-transferable, resulting in the contract 

generating a non-transferable claim. They could also agree that the claim 

for payment will diminish by 15 % for each day after e.g. September 1 that 

the relevant goods remain undelivered by the seller. They could even 

agree that if the goods are not delivered by September 1, there will be no 

amount due at all, resulting in a claim that would extinguish if the goods 

are not delivered by such date. Contingencies not relating to delivery of 

the goods are also possible. An example would be a provision in the contract 

stating that payment does not need to be made if one dollar is worth 

at least 95 eurocents at a particular date. Put more generally, the parties 

to a contract are able to give claims arising from a contract (almost) any 

contents they like. 

Third parties are bound by the contents that the parties to a contract 

have given to the claims that arise from it. For instance, if the claim for 

payment from the sale and purchase contract in the example above is 

subsequently assigned to a third party or if the seller goes bankrupt, neither 

the assignee nor the bankruptcy trustee would be able to claim more 

from the purchaser than the seller could have claimed if the claim had 

not been assigned or if the seller had not become insolvent. In the case 

of a transfer of the claim to a third party, this is simply an application of 

the nemo dat rule: no one can transfer more than he or she actually owns. 

In the case of a bankruptcy trustee being appointed, the effect can be explained 

by the trustee finding the assets of the insolvent party as they are, 

even if these assets are defective, faulty or flawed. So if, as in the example 

above, the parties to the sale and purchase contract have agreed that the 

claim for payment of the purchase price will diminish by 15 % for each 

day after September 1 that the relevant goods remain undelivered by the 

seller, neither the assignee of the claim for payment of the purchase price 

nor the seller’s bankruptcy trustee will be able to claim the full amount 

from the purchaser if the goods are delivered on or after September 2. It 

might be possible to use the principle that allows the parties to a contract 

to determine the contents of the claims that arise under that contract as 

a method of granting security over bank deposits. This would offer almost 


251 


www.sellier.de


unlimited party autonomy to professional market participants because the 

collateral arrangement would become entirely contractual in nature. 

11.4.3. Alternative security structures allowing even 

more party autonomy 

From a legal perspective, the rights of an account-holder towards its 

bank are contractual. These contractual rights may be, and in practice 

often are, given certain contents by agreement between the bank and 

its account-holder, for instance if they agree that the deposit will not be 

payable before January 2, 2012. This agreement leads to a claim of the 

account-holder against the bank, but this has an inherent flaw: if the 

account-holder (or its bankruptcy trustee or assignee) were to claim payment 

of the balance on April 12, 2011, the bank would refuse payment on 

the basis that the claim of the account-holder does not allow for release of 

the deposit before January 2, 2012. This ability of the parties to shape the 

claims that result from their contractual relationship offers possibilities in 

terms of creating security over bank accounts / deposits. 

The first situation that comes to mind is the charge-back scenario. Here, 

it is the account bank that takes security over the balance it holds. In our 

example, this would be the situation if the collateral provider (Rock Solid 

Bank) provided a right of pledge or charge over the cash balance held by 

it with the collateral taker (Alpha Bank). 

In my view, it would be possible and in fact conceptually a lot cleaner for 

the parties to simply contract that the balance held by Rock Solid Bank 

with Alpha Bank will only be payable to Rock Solid Bank if and to the 

extent it has first paid all the amounts due to Alpha Bank under the credit 

default swap (or for this contract to terminate in such a way that no future 

payment obligations can be incurred). This would seem to be an attractive 

alternative to the rather weak structure of charging the account balance in 

favour of Alpha Bank and the result would be quite similar to that where 

a right of pledge or a charge is created over the balance in favour of Alpha 

Bank. The account balance serves as security for the account bank and if 

the loan is not repaid, the account bank would be under no obligation to 

repay the balance – and all this is achieved under a contractual arrangement, 

providing the parties with maximum party autonomy. 

252 



www.sellier.de


As explained above, Rock Solid Bank’s bankruptcy trustee should be 

bound by the limitations that have been contractually agreed by the parties 

and the same would be true for the assignee of the account balance 

and, for that matter, for third parties in general. Effectively the result 

would be similar to a pledge of the account balance to Alpha Bank as the 

account bank (a charge-back) without triggering the conceptual problems 

that may arise in the case of a charge-back under English law. 64 

As a matter of Dutch law, it should also be possible to use this method 

of granting security over a cash deposit if the secured creditor is not the 

account bank but a third party. Going back to our example, this would be 

the situation if the parties have arranged for Rock Solid Bank to pledge 

or charge a cash balance held by it with some other bank (e.g. Euroclear 

Bank) that is not otherwise involved in the credit default swap and its collateralization. 

Rock Solid Bank, the third party account bank (Euroclear 

Bank) and Alpha Bank would then agree that the balance will be released 

to the account-holder only if and to the extent that Rock Solid Bank has 

paid its obligations under the credit default swap to Alpha Bank in full (or 

if this contract has terminated in such a way that no future payment obligations 

can be incurred). Moreover, the account bank would undertake to 

release the balance to Alpha Bank if it has not been presented with proof 

that such payment has taken place at a certain date. Rock Solid Bank’s 

bankruptcy trustee or the assignee of the claim against the bank would 

be bound by these arrangements for the reasons discussed above. This 

method of using a bank deposit for security could be called a ‘contractual 

approach’ because it relies entirely on contractual arrangement between 

the account-holder, the account bank and the secured party. 

This contractual approach to security is, it seems, just as robust as proprietary 

security because third parties would be bound by these arrangements. 

The most relevant third parties would of course be Rock Solid Bank’s 

creditors, assignees and (potential) bankruptcy trustee. We have already 

seen that these parties would simply step into the shoes of Rock Solid 

Bank and thus would only be entitled to the contingent claim against the 

account bank, as this has been shaped by the contractual arrangement 

entered into by Rock Solid Bank. 

The contractual approach to security shows strong similarities to the English 

law concept of a ‘flawed asset’. This is a contractual structure whereby 

64 

Goode 1998, pp. 69-71. See also Goode-Gullifer 2008, § 3-12. 


253 


www.sellier.de


the debtor of a claim agrees with its creditor that the debt will not be payable 

unless a specific event happens. 65 This precise type of arrangement 

is often used in relation to bank deposits. An example is a deposit placed 

with a bank by a company which will only be repayable if and to the extent 

that the company first pays its indebtedness to the bank under a loan. 

In this approach, parties enjoy full autonomy to arrange their affairs as 

they like. It would be possible to attach any number of conditions to the 

account bank’s obligation to repay the deposit, for instance performance 

of a group company’s obligations or even the obligations of an entirely 

unrelated party towards the bank. Whether there are limits to the actual 

conditions that could be used goes well beyond the scope of this chapter 

and will not be discussed in any detail. Obvious limitations might result 

from the ultra vires doctrine and from fraudulent preference rules. These 

limitations are of little relevance here, however, because the central theme 

of this chapter is the more fundamental question of whether contractual 

arrangements could be used to give the claim against the bank specific 

contents in order to create a (quasi-) security interest allowing the parties 

to enjoy complete freedom in structuring their collateral arrangement. 

The concept of a flawed asset providing effective security as an alternative 

to charge-backs seems to have been explicitly accepted under English 

law for instance by the House of Lords in Bank of Credit and Commerce 

International SA (No 8) although, as with the conceptual acceptance of 

the charge-back, it was in an obiter dictum. Lord Hoffmann: 

‘The Court of Appeal said that the bank could obtain effective security 

in other ways. If the deposit was made by the principal debtor, it could 

rely upon contractual rights of set-off or combining accounts or rules 

of bankruptcy set-off under provisions such as rule 4.90. If the deposit 

was made by a third party, it could enter into contractual arrangements 

such as the limitation on the right to withdraw the deposit in 

this case, thereby making the deposit a ‘flawed asset.’ All this is true. 

It may well be that the security provided in these ways will in most 

cases be just as good as that provided by a proprietary interest. But 

that seems to me no reason for preventing banks and their customers 

from creating charges over deposits if, for reasons of their own, they 

want to do so. (…)’ 

65 

See Wood 2007b, § 29-021 and Goode-Gullifer 2008, § 1-21. 

254 



www.sellier.de


Of course the contractual approach discussed here may also have drawbacks. 

The most important drawback is that it requires the account bank 

to sign up to the security arrangement. This would not be a problem if 

the account bank is a party to the underlying contract (the credit default 

swap in our example) but if Rock Solid Bank wishes to use a balance held 

with an entirely different bank, this other account bank, which would 

otherwise be uninvolved in the transaction, may not be keen on getting 

involved at least not without being paid a (hefty) fee. 66 If an account 

balance is pledged or charged to a third party, the account bank is more 

or less an innocent bystander in relation to this pledge. Its involvement 

(and its risk of being held liable) would be much greater if, under certain 

specific circumstances, it was asked to release the balance held in the 

name of Rock Solid Bank to Alpha Bank, a bank with which the account 

bank did not have any prior legal relationship. The reason of course is that 

the account bank would then have to check whether the requirements 

for release of the deposit to either its account-holder, Rock Solid Bank, 

or the collateral taker, Alpha Bank, have been fulfilled. However, this 

does not take away the fact that the alternative approach suggested here 

may be helpful in specific circumstances, particularly if the account bank 

itself is the collateral taker. In our example of Rock Solid Bank providing 

collateral for the benefit of Alpha Bank to secure its obligations under the 

credit default swap, an obvious scenario would be for Rock Solid Bank to 

post the relevant cash collateral with Alpha Bank The parties could then 

enter into detailed arrangements as to if and when the cash collateral will 

be repayable, etc. 

The question of whether there are legal reasons under Dutch law why 

the contractual approach proposed here would not work was researched 

in the author’s doctoral thesis. No legal objections were found and it was 

concluded that there are no reasons why this alternative approach would 

not work in principle both in relation to security taken by the account 

bank and security taken by a third party. Of course, all this remains true 

provided that the account bank is willing to cooperate. This is not to 

say that there are no limitations to the freedom of contract, which is the 

principle that allows the parties to use the contractual approach described 

here in the first place. Freedom of contract is only a starting point which 

has limitations that would also apply to contractual security arrangements. 

For example, an arrangement whereby the account balance needs 

to be paid to a subsidiary of the account-holder upon the account-holder’s 

66 

See Wood 2007a, § 14-020. 


255 


www.sellier.de


insolvency would probably be invalid for reasons of fraud, especially if 

its main purpose were to prevent creditors from taking recourse against 

the account-holder by having her assets disappear into thin air upon her 

insolvency. However, if the security arrangement is part of an arm’s length 

transaction and is entered into on sound commercial terms, these limitations 

will be the exception to the rule that, generally, taking contractual 

security over bank deposits is possible as a matter of Dutch law. 

Although the reasoning for taking this view was specific to Dutch law, a 

similar line of reasoning might lead to similar results in other jurisdictions 

because a lot of the principles (bank deposits constitute a claim against 

the bank, freedom of contract is an essential part of the legal system and 

claims resulting from contracts can be given any preferred contents by 

the parties to the relevant contract) are likely to apply in many Western 

jurisdictions. 

11.4.4. Collateral arrangements involving securities 

The final question then is whether this alternative contractual approach 

could also be feasible in a scenario where securities collateral is to be provided. 

Apart from the points mentioned in paragraphs 11.4.2 and 11.4.3 

above, the main difficulty appears to be whether an investor’s interest in 

securities can be qualified as a contractual right against the intermediary 

where he or she holds here securities account. 

This in some ways is probably a rather large step. The starting point is, or 

rather used to be, that securities were issued in bearer form to make their 

transfer easier. The transfer of securities took the form of handing over the 

piece of paper representing the relevant security. This way of transfer has 

strong property law implications because the holder of the piece of paper is 

seen as the owner of the rights that are embodied in the piece of paper. 

In modern securities transfer law, the importance of bearer certificates has 

diminished dramatically. Bearer certificates are safely tucked away in a 

vault of a central securities depository and securities are often in registered 

form only, removing the need for bearer certificates altogether. The investor 

holds his or her rights through a chain of intermediaries that record 

securities holdings in securities accounts. Transfers take the form of book 

256 



www.sellier.de


entries and there is no transfer whatsoever of paper. 67 On the other hand, 

under Dutch law, investor protection against an intermediary insolvency 

takes the form of a proprietary right in the pool of securities held by that 

intermediary, which is granted to the investor by statute. The situation 

under English law is somewhat similar: the right of the investor against 

an intermediary is deemed to be of a proprietary nature precisely because 

it is protected against the intermediary’s general creditors in the event of 

its bankruptcy. 68 

In view of this proprietary nature of the rights of an investor, it is difficult 

to abandon altogether the connection between physical bearer securities 

(and the corresponding property law implications) and simply summarize 

the investors’ rights as a contractual claim against the intermediary that 

can be manipulated by the account-holder and the intermediary as they 

please. 

There are, however, various arguments to support the view that the rights 

of an investor are ultimately of a contractual nature only and that, consequently, 

security over these rights can take the same form as was argued 

for cash balances, i.e. manipulation of the contractual rights of the holder 

of the securities account against its intermediary. I will only list my arguments 

here 69 – for a more comprehensive overview reference will be made 

to other publications: 

i. the connection with physical pieces of paper (where it exists at all) is 

only symbolic in modern securities law; the most important right of 

an investor is the investor’s rights against his or her securities intermediary, 

whose records determine who will be entitled to exercise the 

rights under the relevant securities with there being no way of proving 

entitlement to these rights other than through an account statement 

with a securities intermediary; 70 

67 

See Goode 2003, § 6-02 for a description of this development. See also Wood 

2007b, §§ 29-052 and 29-053; Haentjens 2007, pp. 29-41; and Benjamin 2007, 

§§ 1.78-1.92. 

68 

See Benjamin 2007, chapter 2 (specifically, §§ 2.33-2.34), chapter 13 and chapter 

14 (specifically, §§ 14.06-14.27). 

69 

Wibier 2009, pp. 388-411, specifically, § 3.4. 

70 

Hudson 2009, § 35-09: ‘The proprietary right of the investor is therefore in the 

chose in action against the registrar and not in any individual bond.’ 


257 


www.sellier.de


ii. proprietary rights are only relevant in protecting an investor’s rights 

in relation to the intermediary in an intermediary insolvency, and so 

for all other intents and purposes can be safely ignored; 

iii. the system of chains of intermediaries through which securities are 

held may be viewed as a sophisticated clerical system which records 

to whom the issuer should pay principal and interest; it does not, or at 

least should not, have any independent proprietary significance; and 

iv. investors are generally unable to bypass their intermediaries and directly 

exercise their rights against either (a) intermediaries higher up 

the chain, (b) physical pieces of paper representing the securities or 

(c) the issuer of the securities, meaning that the main asset is the investors’ 

right against the intermediary, which is a right of a contractual 

nature only. 

Realistically however, it would be a bit optimistic to state that using the 

contractual approach to security, which was described for cash balances in 

this section, over securities held in investment accounts is already widely 

possible, let alone practiced. In principle though, there is no reason at 

all why manipulation of the contractual rights of an investor against the 

intermediary where the securities are held should not offer rock solid security 

without any rules of property law preventing this from happening. 

Obviously, this would be a plus for party autonomy because the parties 

would enjoy the freedom to attach any number of conditions to the release 

of the collateral to either one of them, again as explained for cash 

balances in this section. 


In this chapter I have argued that the parties to financial collateral arrangements 

enjoy a good deal of freedom (party autonomy). There seem 

to be only minor limitations to their freedom to give their arrangements 

any contents they like. There might be a small number of formalities 

that need to be adhered to, such as the execution of a deed of pledge and 

compliance with the rules applicable to securities transfer, but these would 

not prevent the parties from using the standard documentation provided 

by ISDA and having English law apply to their arrangements even if all 

the parties involved are Dutch, notwithstanding that parties that prefer 

the application of Dutch law would of course also have the freedom to 

choose Dutch law. 

258 



www.sellier.de


I have also argued that security might take the form of a contractual 

arrangement between the security provider, the security taker and the 

account bank (or financial intermediary if the collateral consists of fungible 

securities). These contractual arrangements would offer the parties 

ultimate freedom and thus the highest level of party autonomy, but there 

are remaining doubts as to the validity of these arrangements when the 

collateral is to consist of fungible securities. 


259 


www.sellier.de


www.sellier.de

12. 

Party Autonomy and Insolvency Law 

P. M. Veder * 


The theme and title of this contribution is ‘Party Autonomy and Insolvency 

Law’. Why this theme? Does insolvency law in any way influence 

the debate on whether to accept party autonomy in international property 

law? 

Of course, the question as to whether a person has merely a personal 

claim against an insolvent debtor or instead a proprietary interest in the 

debtor’s property can mean the difference between recovering nothing at 

all or achieving a substantial, if not full, recovery. However, the role of 

insolvency law as such in the debate on whether to accept party autonomy 

in international property law is, I think, somewhat limited. 

Insolvency of one of the parties to a transaction does not alter the law 

applicable to their relationship. Nor does the mere fact that insolvency 

proceedings have been opened in respect of one of the parties to a transaction 

in principle alter the rights and obligations emanating from their 

relationship. However, insolvency law may influence the extent to which 

and the conditions under which these rights and obligations can be enforced 

during or after the debtor’s insolvency. 

As regards proprietary (security) rights in assets, insolvency law takes 

such rights as it finds them. Whether a proprietary (security) right exists, 

and what its contents are, is not a matter of insolvency law but one of 

(international) property law. What insolvency law does is regulate how 

such rights can be enforced within the framework of an insolvency proceeding. 

Of course, in regulating the position of, in particular, security 

rights in insolvency, legislators make policy choices: for instance, with 

respect to safeguarding or strengthening the interests of the general body 

* 

Michael Veder is professor of law at the Molengraaff Institute of Private Law, 

Utrecht School of Law, and attorney at RESOR N.V. in Amsterdam. 

261 


www.sellier.de


of unsecured creditors. However, I agree with Professor Flessner – as he has 

recently set forth in his contribution to the Koziol Festschrift 1 – that such 

policy choices in insolvency legislation cannot exert a decisive influence 

on the debate as to whether to accept party autonomy in international 

property law. These policy choices, as they are for example reflected in the 

ranking of claims, remain unaffected and will be given the desired effect, 

since the enforcement of security rights remains governed by the relevant 

enforcement statute, which cannot be freely chosen by the parties. As to 

the enforcement of security rights in movables outside of insolvency, this 

will be the law of the place where the movables are located. With respect 

to the enforcement of security rights in insolvency, this will be the law 

of the state where insolvency proceedings have been opened (with the 

proviso, of course, that pursuant to Articles 5 and 7 of the EC Insolvency 

Regulation [EIR] rights in assets situated in other Member States will 

remain unaffected by the proceeding). Party autonomy in international 

property law can only be relevant for the creation, transfer, and determination 

of the content of security rights and not in respect of their enforcement. 

When it comes to applying the rules of the enforcement statute, 

it is in principle irrelevant whether a particular right was created under 

foreign law. It is decisive whether the right in question can be given the 

same effect as ‘local’ proprietary (security) rights within the rules of the 

enforcement statute. 

Nevertheless, it must be noted that on an international level, the position 

of parties in insolvency proceedings may be ‘manipulated’ in a number 

of ways. 

Notwithstanding the intention of the EC Insolvency Regulation to avoid 

incentives for parties to seek to obtain a more favourable position by 

forum shopping, 2 practice shows that the applicable insolvency regime 

can to a certain extent be manipulated by the debtor shifting his centre 

of main interests to another – more favourable – jurisdiction. Pursuant 

to Article 4 EIR, the law applicable to insolvency proceedings and their 

1 

A. Flessner, Rechtswahl im internationalen Sachenrecht – neue Anstösse 

aus Europa, in P. Apathy et al. (eds.), Festschrift für Helmut Koziol zum 70. 

Geburtstag, Jan Sramek Verlag, 2010, p. 137 / 138, (in translated version: see 

Chapter 1 in this book on Party Autonomy in International Property Law, 

Choice of Law in International Property Law, New Encouragement from Europe, 

par. D, section IV). 

2 

Preamble EIR, par. 4. 

262 

P. M. Veder 


www.sellier.de

12. Party Autonomy and Insolvency Law 

effects shall be that of the Member State within the territory of which 

such proceedings are opened: the lex concursus. This general and exclusive 

applicability of the lex concursus is subject to a number of exceptions, 

however, which in some cases allow for a choice of law made by the parties 

to determine their position in insolvency. These exceptions are laid down 

in Articles 5-15 EIR and are generally intended to protect legitimate expectations 

and the certainty of transactions in Member States other than 

that in which proceedings are opened. 3 

In some respects, the special position awarded to a particular creditor is 

based on the location of assets, as is for example the case with Articles 

5 and 7 EIR, where the location of the assets in which a right in rem is 

invoked is decisive for the effects that the opening of insolvency proceedings 

will have on such rights. In other cases, a choice of law may influence 

a creditor’s position. This is, for example, the case with regard to set-off. 

Pursuant to Article 6 EIR, in cases where set-off is not allowed under the 

lex concursus, a creditor may nevertheless invoke a right of set-off if set-off 

is permitted under the law applicable to the insolvent debtor’s claim (including 

its insolvency law). The lex causae of the insolvent debtor’s claim 

is something that parties are generally free to choose and, consequently, 

parties may select the law that is most favourable to set-off in the case of 

insolvency. 

Another important exception to the general and exclusive applicability 

of the lex concursus is provided for in respect of avoidance actions. That 

is what I would like to focus on in this paper. Pursuant to Article 4 EIR, 

the lex concursus determines the rules relating to the voidness, voidability, 

or unenforceability of legal acts detrimental to all the creditors (Article 4 

[2] [m] EIR). However, it follows from Article 13 EIR that a transaction is 

safe, even if it is subject to reversal under the lex concursus, if it cannot be 

challenged under the law governing the transaction, including its insolvency 

law. In my view, the protection thus offered by the EC Insolvency 

Regulation, and in fact also by rules of customary private international 

law in a number of jurisdictions in cases that do not fall within the scope 

of the EC Insolvency Regulation, 4 by a law that parties are free to choose, 

is a factor, albeit not a decisive one, that should be taken into considera- 

3 

Preamble EIR, par. 24. 

4 

For Netherlands law, see for example Hoge Raad 24 October 1997, NJ 1999, 

316, comm. Th.M. de Boer (Gustafsen q.q. / Mosk), JOR 1997 / 146, comm. 

H.L.E. Verhagen. 

P. M. Veder 

263 


www.sellier.de


tion in the debate on whether to accept party autonomy in international 

property law. 

12.2. Avoidance actions 

12.2.1. Introduction 

Juridical acts that the debtor has performed in the period (immediately) 

prior to the opening of an insolvency proceeding, including the transfer 

of and creation of security rights in assets, are generally subject to special 

scrutiny. Insolvency law generally provides for mechanisms to render ineffective 

juridical acts performed by the debtor prior to the opening of the 

insolvency proceedings and that have prejudiced the creditors. Insolvency 

law under certain circumstances allows for the reversal 5 of juridical acts 

that the debtor has performed to the unfair detriment of the (other) creditors. 

Such actions are intended to protect the integrity of the estate and 

to counter infringements on the paritas creditorum (or, rather, the statutory 

ranking of creditors). This is a recognition of the fact that a debtor who 

is insolvent, or on the brink of insolvency but not yet subject to formal 

insolvency proceedings, may be tempted to transfer assets gratuitously or 

at an undervalue, or to benefit certain creditors to the detriment of others. 

The result of such actions may for example be that assets or proceeds 

thereof are returned to the estate, to the extent that they would otherwise 

have been available for the satisfaction of creditors’ claims, or to allow 

the administrator to realise assets unencumbered: for instance, following 

reversal of the creation of a security right immediately prior to the opening 

of the proceeding to secure a previously unsecured claim. 

The conditions under which juridical acts may be reversed differ widely 

among jurisdictions: for example, with respect to requirements of proof 

of collusion or awareness of the prejudice to other creditors or fraudulent 

intent by the debtor. 6 Some jurisdictions in principle only allow for 

5 

‘Reversal’ is a term derived from the Principles of European Insolvency Law, 

see W.W. McBryde et al. (eds.), Principles of European Insolvency Law, Deventer: 

Kluwer, 2003. 

6 

Comparative surveys can be found in J.J. Forner Delaygua (ed.), La proteccion 

del crédito en Europa: La acción pauliana, Barcelona: Bosch, 2000; P.R. 

Wood, Principles of International Insolvency, 2 nd edition, London: Sweet & 

264 

P. M. Veder 


www.sellier.de


the reversal of juridical acts that have been performed in a particular 

period prior to the opening of insolvency proceedings (or the moment 

determined by the court that the debtor was in fact insolvent): a ‘période 

suspecte’. Timeframes and conditions may differ depending on the nature 

of the juridical act concerned. Jurisdictions also differ with respect to the 

question of whether action must be taken by the administrator to obtain 

the invalidation of the prejudicial acts or whether the decision to open 

insolvency proceedings automatically entails invalidation. 

Given these differences in national substantive law, it is therefore important 

to establish which law governs such avoidance actions in an international 

context. 

12.2.2. Lex concursus applies; ‘veto’ of the lex causae 

The law applicable to avoidance actions is determined by Articles 4 (2) 

(m) and 13 EIR. This applies both to main and territorial (secondary) 

proceedings. 7 Pursuant to Article 4 (2) (m) EIR, the lex concursus determines 

the rules relating to the voidness, voidability, or unenforceability of 

juridical acts detrimental to all the creditors. The lex concursus determines 

the conditions for reversal of a juridical act, the manner in which such 

reversal operates – e.g. by operation of law or requiring action from the 

liquidator (either in court or outside of court proceedings) – and the legal 

consequences thereof. 

If under the lex concursus a juridical act is not subject to reversal, the 

liquidator cannot challenge the act successfully even if it is subject to 

reversal under the law governing the act concerned. If for example the act 

concerns a transfer or encumbrance of assets situated in a Member State 

where the debtor has an establishment within the terms of Article 2 (h) 

Maxwell, 2007, p. 458 et seq. See also the UNCITRAL Legislative Guide on 

Insolvency Law, part II, par. 148-203. 

7 

The division of powers between the liquidators in main and territorial proceedings 

to institute avoidance actions will not be further dealt with in this 

paper. On this issue, see for example P.M. Veder, Cross-Border Insolvency 

Proceedings and Security Rights, Kluwer Law International, 2004, p. 309- 

311. The issue of jurisdiction in respect of avoidance actions will also not be 

addressed in this paper. In this respect, reference is made to ECJ 12 February 

2009, Case C-339 / 07 (Seagon q.q. / Deko Marty). 

P. M. Veder 

265 


www.sellier.de


EIR, the opening of secondary proceedings may be helpful if under the insolvency 

law of that State, unlike the law governing the main proceeding, 

the transfer or encumbrance can be reversed and claw back is possible. 

On the other hand, the fact that a juridical act is subject to reversal under 

the lex concursus does not ipso facto entail that it can indeed be challenged 

successfully by the liquidator. Pursuant to Article 13 EIR: 

‘Article 4 (2) (m) EIR shall not apply where the person who benefited 

from an act detrimental to all the creditors provides proof that: 

– the said act is subject to the law of a Member State other than that 

of the State of the opening of proceedings, and 

– that law does not allow any means of challenging that act in the 

relevant case.’ 

Article 13 EIR thus provides a possible defence to the person who has 

benefited from a juridical act that is challenged by the liquidator. The 

burden of proof is placed on him, however. He must first prove that the 

challenged act is governed by the law of a Member State other than the 

State where the proceeding has been opened. He must then prove that, 

under the circumstances of this particular case, that law – including its 

insolvency law – does not allow any means of challenging the juridical 

act in question. The phrase ‘in the relevant case’ is used to express the 

notion that in the determination of the vulnerability of the act under the 

lex causae all relevant circumstances of the case must be taken into consideration. 

It is not sufficient to determine whether it can be challenged 

in the abstract. 8 

Article 13 EIR acts as a ‘veto’ against the invalidity of the act decreed by 

the lex concursus. It does not prescribe a cumulative application of both 

the lex concursus and the lex causae. The conditions for and the consequences 

of the voidability are governed by the lex concursus. The only 

purpose of Article 13 EIR is to reject the application of the lex concursus 

in a given case. The aim of Article 13 EIR is 

‘to uphold legitimate expectations of creditors or third parties of the 

validity of the act in accordance with the normally applicable national 

law, against interference from a different “lex concursus”.’ 9 

8 

Virgós / Schmit Report, Nr. 137. 

9 

Virgós / Schmit Report, Nr. 138. 

266 

P. M. Veder 


www.sellier.de


12.2.3. Protection of legitimate expectations or is insolvency a 

foreseeable risk? 

This ‘double test’ has been incorporated in the EC Insolvency Regulation 

notwithstanding that in legal writing it was and is often rejected. 10 It unnecessarily 

restricts the possibilities of re-establishing the estate for the 

benefit of the creditors. The question really is whether creditors or third 

parties entering into a transaction with the (later insolvent) debtor should 

not reckon with the possibility that the transaction could be subject to reversal 

in accordance with the insolvency law of the Member State where 

the debtor has his centre of main interests. It is submitted that if the insolvency 

law of the Member State where the centre of the debtor’s main 

interests is located provides that the act concerned is subject to reversal, 

there can be no legitimate expectation that the act would be valid only 

because the lex causae of that act, including its insolvency law, does not 

allow for its reversal. The opening of an insolvency proceeding in respect 

of one of the parties to a transaction in the State where that party’s centre 

of main interests is located is a foreseeable risk that should be taken into 

account when a transaction is entered into 11 . To this it may be added that, 

in my view, even the opening of insolvency proceedings in the Member 

10 

Cf. Veder 2004 (footnote 7), p. 315, H.L.E. Verhagen and P.M. Veder, De 

‘Pauliana’ in het Nederlandse internationaal privaatrecht, NIPR 2000 / 1, p. 5; 

R.J. van Galen en J.C. van Apeldoorn, Grensoverschrijdende aspecten van 

insolventieprocedures buiten verdrag, in: R.J. van Galen, J.C. van Apeldoorn 

and A.J. Berends, Grensoverschrijdende insolventieprocedures, Preadviezen 

voor de Nederlandse Vereniging voor Internationaal Recht, Deventer: Kluwer, 

1998, p. 130; A. Trunk, Internationales Insolvenzrecht, Systematische 

Darstellung des deutschen Rechts mit rechtsvergleichenden Bezügen, Tübingen: 

MohrSiebeck, 1998, p. 190; A. Flessner, Internationales Insolvenzrecht in 

Deutschland nach der Reform, IPRax 1997, p. 9; H. Hanisch, Extraterritoriale 

Wirkung eines allgemeinen Veräusserungsverbots im Konkurseröffnungsverfahren 

– Revisibilität ermessenshafter Ermittlung ausländischen Rechts – 

Durchgriff – Auf die Insolvenzanfechtung anwendbares Recht, IPRax 1993, 

p. 72. 

11 

The practice of COMI-shifts must be taken into consideration. Relocating the 

debtor’s centre of main interests to another jurisdiction will lead ta a charge of 

the applicable lex concursus and, consequently, the applicable rules relating to 

the avoidance of pre-insolvency transactions. In such cases the protection of 

the other parties’ legitimate expectations regarding the validity of the transaction 

is justified. 

P. M. Veder 

267 


www.sellier.de


State where the debtor has an establishment as defined in Article 2 (h) 

EIR is a foreseeable risk; hence, also in such cases the protection of legitimate 

expectations may not be at stake where the transaction involved the 

transfer or encumbrance of assets situated in that State. 

12.2.4. Articles 5 (4) and 7 (3) EIR 

The system thus laid down in Articles 4 and 13 EIR also applies with 

respect to the reversal of the transfer or encumbrance of assets. With 

respect to proprietary rights (including reservation of ownership), this 

has been laid down explicitly in Articles 5 (4) and 7 (3) EIR. Pursuant to 

these provisions, the fact that rights in assets located in a Member State 

other than the State where the insolvency proceeding has been opened 

are ‘isolated’ from the effects of the main insolvency proceeding ‘shall not 

preclude actions for voidness, voidability or unenforceability as referred 

to in Article 4 (2) (m)’. 

Even though not explicitly referred to in Articles 5 (4) and 7 (3) EIR, the 

defence provided for in Article 13 EIR can also be invoked by the creditor 

who has obtained a proprietary right within the terms of Articles 5 and 7 

EIR. 12 If, for example, a Spanish liquidator, appointed in main proceedings 

opened in Spain, invokes the voidness or voidability of a right of pledge 

created in assets situated in the Netherlands, the pledgee cannot claim 

that, pursuant to Article 5 (1) EIR, the right of pledge is not affected by 

the claw back provisions that are in place under Spanish insolvency law. 

This is clearly expressed by Article 5 (4) IR. The pledgee can, however, 

on the basis of Article 13 EIR, claim that the right of pledge is valid and 

enforceable against the insolvent estate if it cannot be challenged under 

Dutch (insolvency) law. 

12.2.5. Lex causae: the law governing the proprietary 

aspects of the transfer or encumbrance of assets 

The EC Insolvency Regulation has adopted a system whereby a juridical 

act can only be challenged successfully if it is subject to reversal under 

the lex concursus as well as under the lex causae, including its insolvency 

12 

Cf. Veder 2004 (footnote 7), p. 316. 

268 

P. M. Veder 


www.sellier.de


rules. In this respect, it is important to determine which is the relevant lex 

causae in the case of the transfer or encumbrance of assets. 

The question arises as to whether in determining the lex causae relevant 

for the application of Article 13 EIR with respect to the encumbrance or 

transfer of assets effect must be given to the distinction that in some legal 

systems – on the level of substantive national law as well as that of private 

international law – is made between aspects pertaining to the law of obligations 

and those pertaining to the law of property. The law governing 

the contract in which the transfer or encumbrance of assets is agreed upon 

is not necessarily the same as the law governing the proprietary aspects of 

the transfer or encumbrance itself. A contract between a Dutch company 

and a Spanish bank – pursuant to which the bank will extend credit and as 

security for the repayment of the loans extended the debtor company will 

provide a right of pledge in inventory and stock situated in the Netherlands 

– may be governed by Spanish law. On the other hand, the right of 

pledge in the inventory and stock will – under prevailing conflict of laws 

rules – be governed by Dutch law as lex rei sitae. Which is the lex causae 

against which the vulnerability of the creation of the right of pledge must 

be tested? Under Dutch law, in the event there was an enforceable legal 

obligation for the debtor to provide security (e.g. pursuant to a credit 

agreement), the liquidator can only challenge the right of pledge if at the 

time of its creation the bank knew that a petition for the opening of an 

insolvency proceeding in respect of the debtor had been filed for, or there 

was collusion between the bank and the debtor with the intent of giving 

preference to the bank over other creditors. The liquidator will therefore 

generally only be successful in challenging the provision of security if 

he can successfully challenge the contract from which the obligation to 

provide the security emanates: namely, the credit agreement, which, if reversed, 

under Dutch law automatically leads to the reversal of the security 

provided on the basis of that agreement. If, however, Spanish law, as the 

law governing the credit agreement, were to provide that the agreement 

itself cannot be challenged, the avoidance action would not be successful, 

even if under Spanish law the right of pledge based on that agreement 

itself were subject to reversal. Consequently, when, for purposes of Article 

13 EIR, a distinction should be made between the contractual (obligatory) 

and the proprietary aspects of the transfer or encumbrance of assets, the 

liquidator would not be able to successfully challenge the right of pledge, 

even in cases where under both legal systems the end result – that is, the 

preferential position of the bank resulting from the security – were to be 

subject to reversal. 

P. M. Veder 

269 


www.sellier.de


It is submitted that the vulnerability of detrimental acts by which the 

debtor has disposed of or encumbered assets is a matter that is most closely 

connected to the law that governs the proprietary aspects of the transfer or 

encumbrance, irrespective of the fact that the technical legal cause of the 

reversal may lie in the reversal of the contractual obligation underlying 

the proprietary right created. I consider that, for the purpose of Article 

13 EIR, only the law governing the proprietary aspects of the transaction 

should be taken into consideration. 13 This law should govern not only the 

vulnerability of the (proprietary) act of transfer or encumbrance of the 

asset(s) concerned but also the vulnerability of the (contractual) obligation 

to transfer or encumber the asset, irrespective of the law applicable to 

that obligation under the normal conflict of laws rules. The law governing 

the conditions under which a right has effect against third parties, such as 

the liquidator and the other creditors of the insolvent debtor, should also 

govern the protection of such creditors where the creation or transfer of 

such a proprietary right has prejudiced their rights of recourse in respect 

of the debtor’s assets. 

To the extent that party autonomy is not accepted in international property 

law, this approach would eliminate some of the perceived adverse 

effects of party autonomy. The protection of the interests of creditors in 

whose interest avoidance actions are included in insolvency laws cannot 

be circumvented by a choice of law. Unlike the law governing the 

contractual obligations between parties, the law governing proprietary 

aspects of the transfer or encumbrance of assets is – under current private 

international law rules – generally not determined by a choice of law by 

the parties. But even in cases where a choice for the applicable proprietary 

regime is accepted in international property law (as, for example, 

is the case under rules of Dutch private international law [and, as some 

authors have argued, under Article 14 of the Rome I Regulation] 14 regarding 

the assignment and charging of receivables), this approach would lead 

to all relevant aspects of the validity and enforceability of the transfer 

and encumbrance of assets to be governed by the law with the closest 

connection. Furthermore, this approach takes into account that not in 

all legal systems a clear distinction is made in respect of the juridical act 

13 

Cf. Veder 2004 (footnote 7), p. 319; Verhagen / Veder, NIPR 2000 / 1 (footnote 

10), p. 14. 

14 

Cf. A. Flessner, Die internationale Forderungsabtretung nach der Rom-I 

Verordnung, IPRax 2009 / 1, p. 38; S. van Dongen and A.P. Wenting, Europa en 

internationale overeenkomsten. EVO wordt Rome I, NTBR 2009 / 3, p. 89. 

270 

P. M. Veder 


www.sellier.de


that is challenged by the administrator: the contractual obligation or the 

subsequent proprietary act(s) by which the transfer or encumbrance is 

effected. Referring the issue of the vulnerability exclusively to the law 

governing the proprietary aspects of the transfer or encumbrance makes 

this distinction between substantive national laws less important and the 

rule easier to apply. 

12.2.6. Scope of Article 13 EIR; lex causae is not 

the law of a Member State 

Article 13 EIR explicitly confers a ‘veto’ on the lex causae only if that 

is the law of another Member State. Judging by its verbatim text, the 

protection offered by Article 13 EIR only applies in cases where the challenged 

act is governed by the law of another Member State. On the face 

of it, Article 13 EIR does not lead to the inapplicability of Article 4 EIR 

in cases where the lex causae of the challenged act is the law of a non- 

Member State; hence, in such cases the reversal of the juridical act would 

be governed by the lex concursus exclusively. 

However, the question is whether such an a contrario reasoning is justified. 

The need, if any, for the protection of the legitimate expectations of 

parties dealing with the debtor that an act will not be subject to reversal 

under the lex causae is not greater or lesser depending on whether the 

law of a Member State or the law of a non-Member State applies to the 

challenged act. 

The Virgós / Schmit Report rejects such an a contrario reasoning. 15 In this 

report it is observed that Articles 4 and 13 EIR aim to regulate the issue 

of avoidance actions in line with the general scope of the EC Insolvency 

Regulation, which only deals with the intra-Community effects of insolvency 

proceedings. Following the Virgós / Schmit Report on this point, 

this would mean that the EC Insolvency Regulation does not uniformly 

prescribe that in the event the challenged act is governed by the law 

of a non-Member State, the other party should be allowed to invoke 

protection under the lex causae. But it also does not prescribe that in 

such cases the vulnerability of the act only depends on the rules of the 

lex concursus. In cases that go beyond the scope of the Regulation, the 

Member States are free to decide which rules they deem most appropri- 

15 

Cf. Report Virgós / Schmit, Nr. 93. 

P. M. Veder 

271 


www.sellier.de


ate. Whether, in analogy to Article13 EIR, a ‘veto’ against reversal of the 

act can be derived from the law governing that act in the event that that 

law is the law of a non-Member State is a matter of customary private 

international law of the Member States. Consequently, the matter may 

be dealt with diffe rently depending on the rules of private international 

law of the forum. 16 

A question that arises with regard to the scope of the EC Insolvency 

Regulation concerns in which cases a claw back action initiated by a 

liquidator goes beyond the scope of the Regulation, i.e. no longer concerns 

the intra-Community effects of the proceeding. The Virgós / Schmit 

Report suggests that this is already the case if the challenged act is governed 

by the law of a non-Member State. In my view, it is doubtful 

whether this is a correct assumption. Would there not be more reason 

to conclude that only in cases concerning transactions performed with 

a counterparty situated outside the EU or transactions concerning assets 

situated outside the EU, intra-Community aspects of the proceeding are 

no longer at stake, even if such transactions are governed by the law of 

a Member State? Can a transaction performed with a counterparty situated 

in a Member State or a transaction concerning assets situated in a 

Member State really be considered not to concern the intra-Community 

effects of the proceeding only because they are governed by the law of 

a non-Member State? In any event, starting from the assumption that it 

is necessary to protect counterparties based on their legitimate reliance 

on the validity and enforceability of the acts under the lex causae, I see 

no compelling argument for the protection provided by (a rule similar 

to) Article 13 EIR to be restricted to transactions with counterparties 

situated in the EU that are governed by the law of a Member State, and 

not to apply (by analogy) to other cases. There is certainly no reason 

to withhold the protection offered by Article 13 EIR to counterparties 

situated in a Member State only because the challenged act is governed 

by the law of a non-Member State. 

Several jurisdictions apply rules similar to the EC Insolvency Regulation 

in cases that fall outside the scope of the Regulation, such as the Netherlands, 

Germany, and Belgium. 

16 

The courts of the Member State where the insolvency proceeding is opened in 

which the avoidance action is initiated have jurisdiction, cf. ECJ 12 February 

2009, Case C-339 / 07 (Seagon q.q. / Deko Marty). 

272 

P. M. Veder 


www.sellier.de


Under current Dutch law, an action brought by a foreign liquidator to 

reverse a juridical act that is governed by a law different from the lex 

concursus will only be successful if and to the extent that the challenged 

act is also subject to reversal pursuant to the law governing the act. This 

follows from a decision of the Hoge Raad of 24 October 1997 in a case 

concerning an action brought by a German liquidator to reverse payments 

made to a Dutch creditor. The Hoge Raad observed that: 

‘In the opinion of the Hoge Raad present Dutch private international 

law refers the action brought in the Netherlands by a foreign liquidator 

to reverse a juridical act that has prejudiced the creditors to the 

law governing the insolvency proceeding (the ‘lex concursus’), which 

determines the existence and content of the powers of the liquidator. 

However, the protection of legal certainty requires that due attention 

is given to the circumstance that a Dutch counterparty of the insolvent 

party – the party that has benefited from the challenged juridical 

act performed by the insolvent debtor – does not have to be prepared 

for an action aimed at reversal of the juridical act under foreign law, 

to the extent that the juridical act itself is not governed by that same 

law and such law imposes less strict requirements for the reversibility 

of the juridical act than the law governing the juridical act itself (the 

‘lex causae’). In the event the lex causae is different from the lex 

concursus, the success of the action brought by the liquidator must 

therefore be assessed not only in accordance with the rules of the latter 

but also the rules of the lex causae, and will only lead to reversal 

of the juridical act if the requirements of the lex concursus as well as 

the lex causae have been met. Support for this approach can be found 

in international developments, in particular the Convention on insolvency 

proceedings that was concluded on 23 November 1995 and 

has meanwhile been signed by the Netherlands.’ 17 

The conflict rule thus presented by the Hoge Raad, which in my view 

applies equally in the case of insolvency proceedings opened in the 

Netherlands, 18 differs from the approach adopted in the EC Insolvency 

Regulation in that the court must always examine the vulnerability of the 

juridical act in accordance with its lex causae, and it places the burden 

of proof on the liquidator. It is not a mere defence that may be invoked 

17 

HR 24 October 1997, NJ 1999, 316, comm. Th.M. de Boer (Gustafsen q.q. / 

Mosk), JOR 1997 / 146, comm. H.L.E. Verhagen. Translation by the author. 

18 

Cf. Veder 2004 (footnote 7), p. 327. 

P. M. Veder 

273 


www.sellier.de


by the beneficiary of the challenged act. In the proposals for a reform of 

Dutch insolvency law that were presented to the Dutch government in 

November 2007, a conflict rule that is more in tune with Article 13 EIR 

has been suggested. 19 These proposals have not as yet led to any preparation 

for legislation. 

Article 339 of the German Insolvenzordnung contains a rule similar to that 

of Article 13 EIR: 

‘A juridical act can be reversed if the conditions for reversal of the 

insolvency law of the State where the insolvency proceeding is opened 

have been fulfilled, unless the other party provides proof that the juridical 

act is governed by the law of another State and the juridical 

act is in no way subject to reversal under that law.’ 20 

Belgium has also incorporated a rule similar to Article 13 EIR in Articles 

119 (4) and 121 (2) of its Act on private international law of 2004: 21 

‘Art. 119. § 1. The insolvency proceeding opened in accordance with 

article 118, § 1, second paragraph, and its effects are governed by Belgian 

law. Belgian law determines the conditions for opening the proceeding, 

its conduct and closure. It determines in particular the issues 

referred to in article 4, § 2, (a)-(m) of the insolvency regulation. 

(…) 

§ 4. In deviation from § 1: 

1° if the person who benefited from an act detrimental to all the creditors 

proves that that act is governed by the law of another State and 

that that law does not allow any means of challenging that act in the 

relevant case, the voidness, voidability or unenforceability of legal acts 

detrimental to all the creditors is governed by the law of that State; 

(…) 

Art. 121. § 1. A foreign judicial decision concerning the opening, conduct 

or closure of an insolvency proceeding, and that has not been 

19 

Cf. Art. 10.4.9 Voorontwerp Insolventiewet, published in S.C.J.J. Kortmann 

and N.E.D. Faber, Geschiedenis van de Faillissementswet, Voorontwerp Insolventiewet, 

Serie Onderneming en Recht, deel 2-IV, Deventer: Kluwer, 

2007. 

20 

Translation by the author. 

21 

Wet van 16 juli 2004 houdende het Wetboek van internationaal privaatrecht, 

Staatsblad 27 juli 2004, p. 57344 et seq. 

274 

P. M. Veder 


www.sellier.de


handed down on the basis of the insolvency regulation, is recognized or 

declared enforceable in Belgium in accordance with article 22 (…). 

§ 2. A foreign judicial decision as referred to in § 1 may not be given 

effects in Belgium that are contrary to the rights of parties pursuant to 

the rules established in article 119, § 2-§ 4.’ 22 

Consequently, the protection of reliance on the irreversibility of a particular 

transaction under the lex causae in many cases may extend to transactions 

with counterparties in any jurisdiction and governed by any law. 

This means that the number of legal systems available to parties that 

wish to make their transaction safe against avoidance in insolvency is in 

theory unlimited. 

12.3. Concluding observations 

Accepting party autonomy in international property law has consequences 

for the protection offered to other creditors under insolvency law by avoidance 

rules. The point of departure is that the reversal of pre-insolvency 

transactions is governed by the lex concursus. However, it is a common 

feature of the EC Insolvency Regulation and rules of customary private 

international law in a number of jurisdictions that the ‘legitimate’ expectations 

that a transaction was valid and not subject to reversal under the 

law governing that transaction are protected. One of the consequences 

of allowing party autonomy in international property law would be that 

parties could choose the law that is most favourable to them in terms of 

avoidance of the rights they have acquired. Parties can avoid avoidance 

by way of a choice of law. This in itself is nothing new, but allowing party 

autonomy in the field of international property law would add to the 

choices available for parties to make their transaction safe against subsequent 

interference by a liquidator in insolvency proceedings. 

This in itself, however, should not be considered a decisive argument 

against the possibility of party autonomy in international property law. In 

most cases, the choice for a particular law will be based on a number of factors, 

such as the parties’ knowledge of and familiarity with the chosen law, 

its useful purpose for the transaction (other than making it safe against 

avoidance actions), and so on. It is also not an unknown feature that legal 

22 

Translation by the author. 

P. M. Veder 

275 


www.sellier.de


systems accept a choice of law with third-party effects. In this respect, 

reference can be made to, for example, the rules regarding the assignment 

and charging of receivables under Dutch private international law 

(and, as some authors argue, Rome I in its current form) and the Hague 

Trust Convention. In the absence of uniform rules for a great number of 

issues that are of direct relevance to international (financial) practice, 

accepting party autonomy in the field of applicable law, also with effect 

against third parties, allows parties to choose a legal system that best caters 

for their needs. However, the interests of third parties such as the other 

creditors should not be left out of the equation. This can be achieved by 

imposing certain restrictions on the choice made for a particular law in 

view of the application of rules like Article 13 EIR. In the exceptional 

case that parties use their freedom to choose the law applicable to (the 

proprietary aspects of) their transaction with the sole objective of making 

the transaction immune against avoidance actions in insolvency, such 

a choice of law should be disregarded for purposes of applying rules like 

Article 13 EIR, and the lex causae should be determined in accordance 

with the objective conflict of laws rules. With respect to such ‘fraudulent’ 

choices of law, a presumption could apply that a choice for a legal system 

that in the circumstances of the particular case renders the transaction 

immune for avoidance actions is presumed to have been effected with 

the sole purpose of avoiding such insolvency avoidance if that law has no 

reasonable connection with the transaction at hand (e.g. the location of 

the parties or assets concerned). 

276 

P. M. Veder 


www.sellier.de

Party Autonomy in International Property Law - Peace Palace Library

Create successful ePaper yourself

Delete template?

Save as template?