Determinants and effects of Venture Capital and Private Equity ...

Determinants and effects of Venture Capital and Private Equity
investments on Italian SMEs
Vincenzo Capizzi, Ph.D.,
Associate Professor of Banking and Finance
Department of Business Studies and Environment, Vice Director,
Eastern Piedmont State University, Novara, Italy
Telephone: +39 0321 375 438
Fax: +39 0321 375 405
E‐mail: vincenzo.capizzi@eco.unipmn.it
Renato Giovannini, Ph.D.,
Associate Professor of Banking and Finance
Department of Economics and Management
Guglielmo Marconi University Rome, Italy
SDA BOCCONI School of Management Milan, Italy
Telephone: +39 02 5836 5853
Mobile: +39 348 330 5002
Fax: +39 02 5836 5920
E‐mail: renato.giovannini@unibocconi.it
Valerio Pesic, Ph.D.,
Assistant Professor of Banking and Finance
Department of Banking, Insurance and Markets
La Sapienza University Rome, Italy
Telephone: + 39 06 4976 6260
Fax: + 39 17 8275 2912
E‐mail: valerio.pesic@uniroma1.it
Abstract
Numerous studies have discussed that small and medium enterprises (SMEs) are financially
more constrained than large firms: therefore, venture capitalists (VC) are often the only
available source of financing to small and young companies, especially in those cases where
intangible assets are at the core of the business. By the analysis of the interaction occurred
between a sample of italian SMEs and VC and PE operators (from a dataset of 730 deals
resulting during period 1997‐2007, a final sample of 160 VC/PE‐backed companies was
obtained), we research for an empirical evidences on the determinants and effects that VC and
PE investments played in Italy for small and medium enterprises. We find that, as in the United
States, VCs and PEs more likely finance younger, smaller and thus riskier firms; moreover,
coupling these results with sustained investments in intangible assets both ex‐ante and ex‐post
the date of the deal, it supports the theory which sees VC and PE firms as a solution to
problems of asymmetric information.
JEL code: G24, G30, G31 G32
Key words: venture capital, firm value, firm financing, SMEs
- 1 -

1. Introduction
In the few last decades there has been an increasing emphasis on the dynamism of small
medium enterprises (SMEs), as a major source for speeding up the economic path which
coupled with a greater focus placed upon the efficiency of financial markets in meeting the
needs of these businesses. In the specific case of Italy, whose economy has always taken
advantage of the wealth produced by the industrial districts, this interest should be particularly
binding.
On the one hand, several studies have shown that small‐medium enterprises count for a
product innovation rate more than proportional in regard of R&D formal activities, what
explains their high development rates. In particular this is true for high‐tech and industrial
enterprises (Audretsch, Santarelli and Vivarelli 1999). Moreover, Thurik et al. (2002) show that,
as for 18 OECD countries, higher level of entrepreneurship comes with higher rate of growth
and lower unemployment. Furthermore, studies on regional development are in favour of
young and small firms devoted to innovative business as the leading components in the
economic growth of those areas 1 .
On the other hand, the availability and cost of finance is one of the factors which affect the
ability of a business to grow. Carpeter and Petersen (2002a) showed that in a panel of more
than 1,600 US small firms, the growth of most of them is constrained by internal finance,
together with a small leverage effect. In contrast, the small fraction of firms that make heavy
use of new share issues exhibits growth rates far above what can be supported by internal
finance. In particular, the phase which separates the identification of the innovative project
from the marketing of the product appears to be crucial. In part because they do not have the
financial and other resources to withstand a sustained period of poor performance, the rate of
disbandment among small organizations is quite high, so that a potential investor might see the
prospected returns reset to zero as a likely outcome. Accordingly, when looking for external
capitals a small business may be exposed to the rationing of funds because it lacks of a long
track record, a fact that makes the evaluation more difficult.
From the perspective of a new entrepreneur, venture capitalists (VC), acting as professional
investors with a deep knowledge of the market based on former managerial experience, are
often the only available source of financing to start up a company, especially in those cases
where intangible assets are at the core of the business. Additionally, their involvement usually
implies sharp changes to both corporate governance models and relations with stakeholders,
factors which are perceived as basic starting points for better future performance (Hellmann
and Puri, 2002); the latter are common features shared with private equity firms (PE) which
1 As for United Kingdom refer to Hart, Hanvey, (1995). Callejon and Segarra (1999), point out a positive relation
between start‐up and the growth of Spanish regions and industries.
- 2 -

serve larger established companies in need of either consolidating the results or solving
ownership succession’s problems.
Empirical evidence for the United States highlights some common factors of the VC and PE
industry: the financing is mainly directed to smaller firms operating in high‐tech sectors whose
performance is significantly different from that of similar firms which did not receive this form
of financing. Differences in performance pertain to many aspects, such as R&D intensity, firm
sales growth and investment which have been found to be generally higher for backed firms
than others.
However, when looking at the European market it is clear that it differs from the North
American one for some substantial aspects. A much larger market size, a more developed
capital market, different composition of the investments – generally more prone to start‐up
and high‐tech firms – and of financial sources – pension funds constitute a larger share in
United States while the main contribution in Europe comes from banks – are the most
important differences to enumerate.
Given the differences in the industry characteristics it might be assumed that determinants
and effects of VC and PE may differ in the European countries.
The first aim of this study is to advance knowledge of the Italian market of SMEs financing
and to contrast the results with other experiences. In order to obtain this result we use a
database in which information on venture capital and private equity deals has been matched
with balance‐sheet data for a representative sample of backed Italian firms (data are picked‐up
from the Private Equity Monitor yearly newsletters, AIDA and ZEPHIR – Bureau Van Dijk). More
specifically, the empirical exercises use probit regression analysis to test the relation between
the probability of VC and PE deals and a group of variables (such as size, age, level of collateral,
etc.) found to be relevant in the US. The empirical analysis also compares the performance – in
terms of various balance‐sheet indicators – of backed firms with that of non backed ones. We
applied fixed effect estimation, which controls for unobservable heterogeneity, in order to
explore these relations. The ex‐post analysis of the performance is also useful to discriminate
among different theories.
One of these theories consists in the prediction of the so‐called “certification effect”, in
terms of ability of third‐parties to certify the quality of information issued by relatively
unknown firms (Megginson and Weiss, 1991). The latter notion has been tested and evaluated
in many different bodies of research. Borisova (2007) applied it to the privatization process
which has characterized the European countries during the three last decades, finding that a
decrease in government ownership by one percentage point leads to an increase in the credit
spread, used as a proxy for the cost of debt, by one‐half of a basis point. Sufi (2006) analyzed
another context, namely the introduction of syndicated bank loan ratings by Moody’s and
Standard & Poor’s showing that borrowers that obtain a loan rating gain increased access to the
capital of less informed investors such as foreign banks and non‐banks institutional investors.
- 3 -

Many researchers also applied it to the context of the VC and PE industry. However, most of
them aimed at examining how financial institutions help to resolve the asymmetric information
inherent in the Initial Public Offering (IPO) process. On the contrary, there is scant literature
which interprets the VCs and PEs as agents able to produce information about the qualities of
SMEs, and above all we could not find any evaluation of the Italian case. Indeed, the second
contribution relates to the provision of firm evidence supporting this view.
This study adds to the existing literature by testing the “certification effect” through a
combination of variables that have been singularly suggested by different studies at
confirmation of the presence of such effect (Beatty, Ritter, 1986; Del Colle, et al., 2006;
Borisova, 2007; Hyytinen and Pajarinen, 2007). In particular, we apply the same econometric
procedure as for the effects of VC and PE to proxy variables of access to credit granted by banks
and trade credit, defined as the average length of purchases over the fiscal year.
Our results confirm that, as in the United States, VCs and PEs more likely finance younger,
smaller and thus riskier firms. Coupling these results with sustained investments in intangible
assets both ex‐ante and ex‐post the date of the deal we can support the theory which sees VC
and PE as a solution to problems of asymmetric information. Moreover, looking at the patterns
of growth, rate of investment and sales recorded after the deal we find evidence that is
consistent with the role of the external investor as a consultant. As another interpretation of
these facts we reject, as far as the Italian market is concerned, the theory of venture capital
spurring the innovation. As a matter of fact, most of deals show a slowdown of growth and
investment in fixed assets which follows an higher than average period of investment and
growth, which we interpret as an implicit aim to contribute to the consolidation of a firm’s
result. Finally, the results bring forth the presence of certification effect as confirmed by the
broadening of access to bank credit at better conditions and the consequent reduction of trade
credit, consistent with the theory of Petersen and Rajan (1997).
The paper proceeds as follows: Section 2 draws on the main features of SMEs and financial
industry to set the potential positive interaction between VC and PE. Section 3 describes the
theoretical background and the major contributes of the existing literature. In Section 4 we
briefly recall various corporate finance theories and empirical evidence that are useful to
highlight the likely determinant and effects of VC and PE financing. In addition, it presents the
econometric models in use. Section 5 describes the sources of our data, its main features and
the results for the econometric analysis on the determinants and effects of VC and PE financing.
Section 6 concludes.
- 4 -

2. SMEs challenging the financial industry.
2.1 About SMEs
During the post war period there has been an almost unanimous chorus by academics and
practitioners supporting the clear drawbacks of small‐medium enterprises when compared to
larger ones. Empirical evidence pervasively indicated that SMEs were less efficient, were
remunerating less their employees, were less prone to innovate and thus bound to a decline as
for economic growth. Generally, small size was perceived as a preparatory step in view of a
consolidation of a tiny number of efficient large firms (Audretsch 2002).
Conversely, since 80’s this position has been reconsidered owing to an upwarding trend in
terms of magnitude and relevance of SMEs in the most developed industrial areas. According to
the recent economic literature there are several key elements which can explain the new
setting.
First of all, radical changes have occurred to production processes by means of softening
the well‐established relation size‐efficiency for the benefit of those ones able to recognize the
technological gap and set up flexible organisations. As a second factor, which still makes
reference to the production factors, we should highlight the changes that have occurred to the
labour force that started to appraise positively the above mentioned needs for flexibility. On
the demand side, the enlargement of the markets followed the expansion of customers’ taste,
highlighting, once again, the urge for specialised production and better skills in allocating the
resources. Under these different circumstances, new theories based on the Knowledge
economy appraise the role taken by SMEs as fundamental components of new patterns in
economic growth. As a matter of fact, the tiny margins stemming from continuous
improvements of products, materials, processes, can be of a sufficient magnitude for the
profitability of a small business; while for large firms this could not be always the case.
Along with the review of SMEs as entities capable to drive upward the general economy, a
large body of literature paid much attention to the issue of the financial constraints faced by
new business which may hinder their birth and their development. Acute financial risks add to
the normal operating risk due to the lack of internal resources. The kind of financial risk
attached to young SMEs arise from market imperfections which make reference to information
asymmetry factors, to transaction and agency costs and more generally to the inadequacy of
the financial systems meeting the needs of these businesses.
Firstly, smaller and younger firms are usually in short supply of managerial skills and of the
ability of conveying structured information to investors (Caselli, 2004). Secondly, a higher
degree of opaqueness applies to the case of SMEs, due to the fact that the identity of owners
and managers often coincide: as a result, backers, while running their evaluations, will privilege
real guarantees to the future returns indicated by the firm. Finally, the innovative entrepreneur
- 5 -

might be less keen on revealing details of his business due to the risk of losing his ideas and the
related competitive advantages (Ueda, 2004).
Traditionally, banks have always played an important role among all the financial
intermediaries devoted to the support of SMEs. However, under the circumstances outlined
above, the business model of the traditional bank appears to better meet the needs of medium
size firms which have already an established track record and operate in traditional sectors.
In fact, the provision of debt by a bank to a small business is essentially an agency problem
in which the bank (as principal) is using the firm (as agent) to generate a return on money
advanced. This process occurs under conditions of imperfect and asymmetric information
(Berger and Udell, 1995; Keasey and Watson, 1993) which relate both to the ex ante evaluation
of the project and the entrepreneur (adverse selection) and to the ex post monitoring of
performance (moral hazard). Such information problems are not unique to the small firms
sector, but are considerably more prevalent there because of the anticipated higher costs of
information collection.
Traditionally, there is an agreement on the fact that the degree of information asymmetry
may be reduced through two mechanisms: the provision of collateral as part of the debt
contract and the development of a close working relationship between the lender and the
borrower (Binks and Ennew, 1996). Specifically, the low‐risk borrowers who leave the market in
the Stiglitz‐Weiss model (1981) can signal their status by a willingness to offer appropriate
levels of collateral; a close relationship has the potential to provide the bank with a better
understanding of the operating environment facing a particular business, a clearer picture of
the managerial attributes of the owner and a more accurate overview of the prospects for the
business. Stein (2002) stresses this issue showing that local regional banks have superior skills
in acquiring the soft information stemming from the strict contact with small firms active in the
neighbourhood. The less hierarchical and rigid modus operandi of local banks are the key
elements which allow for the acquisition of non‐computable information which is the typical
outcome of the relationship lending business model.
Nevertheless, there are intrinsic characteristic of SMEs which can hinder the process: on the
one hand because of the technology‐intensive nature of their activity and their lack of a track
record, they face severe adverse selection and moral hazard problems; on the other hand, most
of their assets are firm‐specific or intangible and hence cannot be pledged as collateral. By this
meaning we confirm that the credit rationing is especially acute for some clusters of customers:
smaller, younger and independent firms report more difficulties than other firms when asking
for bank credit. Moreover, Del Colle, Finaldi Russo and Generale (2006) show that small
business are usually affected by multiple lending relationships with banks which can imply a
lower information disclosure. The last outcome points out the weaknesses of the relationship
lending model, which sees the unique long‐term nature interaction between firm and bank as
the way information asymmetries can be sort out. Further, the study of Panetta, Schivardi,
Shum (2004) on the effects of the concentration of the Italian banking industry, suggests that,
- 6 -

after a merger operation, the portion of credit allocated to small business decreases in the long
run after mergers, due to more pronounced size change and more complex organizational
structure.
2.2 Venture capital as a solution
In line with the above, several authors have suggested that a central distinction between
venture capitalists and private equity firms, together, and other financial intermediaries is that
the VCs and PEs operate in situations where asymmetric information is particularly significant.
Indeed, this form of financing has been very successful in the United States and has spurred the
growth of many high‐technology firms. The well known fortunes of such ventures as Yahoo!,
eBay, Microsoft, Apple, brought many policy makers and entrepreneurship scholars to regard
start ups and venture capital as driving forces for economic growth, job creation and structural
change.
As for the Italian case, data gathered by the Private Equity Monitor (PEM © ) on the
transactions occurred in 2006 and 2005, portrayed in figure 1, show that more than half have
dealt with firms recording less than € 60 mil. in terms of sales, confirming the strict relation
between SMEs and PE.
There are several different reasons which tilt in favour of the involvement of VCs and PEs.
First of all, they hold a stake in the firm, but the control rights are proportionately greater when
the entrepreneur must be induced to put more effort into ensuring the success of the project.
In other words, their presence is enforced by an optimal mix of debt securities and equity
securities which, together, ensures the possibility of the backer to become a creditor or a
partner according to what it is needed for the full engagement of the entrepreneur. Kaplan and
Stromberg (2004) refer to this feature as a separation between control and flow rights.
Specifically, control rights allow the venture capitalist to participate to the main decisions of the
entrepreneur. As an additional tool, Gompers (1995) recalls the practice of staging of capital
infusions, for which prospects for the firm are periodically revaluated: the shorter the duration
of an individual round of financing, the more frequently the VC monitors the entrepreneur’s
progress and the greater the need to gather information. Finally, they fully embody the
characteristics of the subjects able to provide the “certification effect”. We have already argued
that the quality of small companies often cannot be observed directly; thus, evaluators must
appraise the company based on observable attributes that are thought to co‐vary with its
underlying but unknown quality. Resource holders therefore assess value by estimating the
conditional probability that a firm will succeed, given a set of observable characteristics of the
organization.
There are two qualitatively distinct categories of information that influence perceptions of
the probability that a young company will succeed. First, important constituencies such as
potential investors and customers make quality judgments through careful consideration of the
- 7 -

previous accomplishments of the organization. Secondly, the identity of exchange partners
becomes a primary consideration when potential investors, customers, employees, suppliers,
and other exchange partners decide whether to commit their resources to a new enterprise.
Since SMEs are mostly affected by information opaqueness and they usually lack of long
track record, we are particularly interested on the second category. The starting point of much
of this work is the observation that social or industrial structures can be represented as a set of
positions that are arranged hierarchically according to the prominence of their occupants.
Baum and Oliver (1991) demonstrated that organization‐to‐institution ties signal conformance
to institutional prescriptions and thereby facilitate young organizations in their attempts to
acquire legitimacy and other resources (see also Aldrich and Auster, 1986). Summing up, it is
possible to identify three eligible social mechanisms that may lead would‐be investors,
customers and other potential exchange partners to take into account the characteristics of a
focal new venture's affiliates as they strive to assess its unobserved and uncertain quality
(Stuart, Hoang and Hybels, 1999): (1) relationships have reciprocal effects on the reputations of
those involved, (2) the evaluative capabilities of well‐known organizations are perceived to be
strong, and (3) relationships with prominent organizations signal a new venture's reliability,
and, thus, its high likelihood of survival.
Together, these three social processes suggest that, for a small venture, gaining a
prominent affiliate serves to enhance perceptions of its quality.
3. Literature Review and Hypothesis
3.1 Literature related to the determinants of Venture Capital and Private Equity financing
The theoretical literature regarding the financing of small firms commonly shares the view
on the additional difficulties faced in order to obtain external finance owing to greater
information opaqueness. Asymmetric information problems between firms and financiers
strongly affect their relationship and shape the nature of the contract between them, such as
the choice of debt versus equity, and for debt the presence of collateral, covenants and the
maturity of the loan. First, there are agency problems. Debt increases moral hazard problems:
following Jensen and Meckling (1976) firms can substitute high‐risk projects for low risk
investments; high‐risk projects increase the probability of bankruptcy, but offer no offsetting
gain to debt‐holders if success is achieved. Second, as Carpenter and Petersen suggest (2002b),
the marginal cost of financial debt could increase very quickly for small firms because fewer
tangible assets can be used to secure loans. Thus, bank finance may not be viable.
Unlike debt, equity finance does not increase the probability of bankruptcy. Moreover,
upside returns are not bounded for investors who buy equity. Aghion and Bolton (1992) and
Aghion et al. (2004) propose a model, based on control rights, which comes to the conclusion
- 8 -

that when size of projects becomes sufficiently large or when assets are increasingly intangible,
firms will give more control rights to outside investors by issuing new equity.
On the empirical side, there is increasing evidence confirming the theoretical prediction
that innovative firms rely more on internal finance and less on leverage. It also seems clear that
outside equity is a valuable source of funds for innovative firms, at least in some countries. The
results of Carpenter and Petersen (2002a) on a panel of publicly‐traded US high‐tech companies
can give useful hints also for our purposes. Although large innovative companies have at their
disposal more collateral to pledge against bank debt, their growth appears to be constrained in
any case. Following the reasoning, small firms can have more trouble in financing their
innovative activity. When they need external finance, they are more likely to obtain debt from
lenders at worse conditions (higher interest rates or shorter maturity) due to their asymmetric
information problems, whereas outside equity seems to be the most suitable financial source.
Unclear growth potential, large investments in intangibles assets (as it may the case for
high‐tech companies) are other variables which might hinder the raising of additional capital.
Myers (1977) argues that firms whose value is largely dependent upon investment in future
growth options would make less use of debt because the owner/manager can undertake
investment strategies that are particular detrimental to bondholders. Testing for the
relationship between market‐to‐book ratios 2 and leverage, Rajan and Zingales (1995) find it to
be negative, giving empirical support to that prediction. Similarly, Barclay and Smith (1995) find
that debt maturity declines with firms’ market‐to‐book ratio.
The nature of firms’ assets may also have important implications for expected agency costs
leaving open the specific possibility to venture capitalists to step in. Williamson (1988) argues
that leverage should be positively related to the liquidation value of assets. As a matter of fact
tangible assets are on average easier to sell and receive a higher fraction of repayment than do
intangibles assets like patents or copyrights. Thus, higher liquidation value implies that default
is less costly. Subsequent tests, as in Rajan and Zingales (1995), proved the relationship
between liquidation value (measured as the ratio of tangible assets to total assets) and
leverage to be positive.
The framework presented gives factual hints towards the role of an external financer.
Where conditions as the ones just outlined are in place, there is the scope for VCs and PE funds
to add value more than other intermediaries. Screening and monitoring activities, although
being far from perfect 3 , and participation in the company’s board can potentially overcome
most of the problems outlined. This may not apply to banks: because regulations limit banks’
ability to hold shares directly, they cannot use equity to fund projects. Moreover, as far as the
Italian market for SMEs funding is concerned, Panetta, Schivardi, Shum (2004) find that the
bank’s specialization in terms of credit policy seems to be affected by M&As: the portion of
2 Myers suggests that a firm’s market‐to‐book ratio may be related to the fraction of firm value that is comprised
of future growth opportunities.
3 Cfr. Gorma and Sahlman (1989); Manigart et al. (2002).
- 9 -

credit allocated to small business decreases in the long run after mergers, due to more
pronounced size change and more complex organizational structure.
In light of the above discussion hypothesis on the determinants of venture capital financing
can be test.
Hypothesis 1:
Given the specific setup in which VCs and PEs operate it is possible to identify a cluster of
variables which can drive their investment decisions.
3.2 Literature on the economic impact of VCs and PE funds
The interest on economic impact of VCs and PE funds can be grouped in two: (1) specific
studies addressing the outcomes for the general economy; (2) researches focusing on the
performance of the small business after new capital has been acquired.
In relation to the first group of studies, Hellman and Puri (2000) and Kortum and Lerner
(2000) show that there is a strong positive correlation between venture capital and innovation.
More precisely, the formers argue that VC backed firms appear to be faster in implementing
new patents; Kortum and Lerner conclude that a dollar of VC is three times more effective in
promoting patent creation than a dollar from a corporation. Yet, the direction of causality
between Venture Capital and the degree of innovation is an open issue. Some empirical studies
have found that more VC financing fosters innovation (“venture capital first hypothesis”, e.g.
Kortum and Lerner, 1998), while others document the opposite, meaning that the entrance of
the external financer follows the discovery of a new technology and meets the need to market
such innovations (“innovation first hypothesis”). Hirukawa and Ueda (2003) find that venture
capital financing is more frequent in industries that have had an increase in total factor
productivity, which the Authors interpret as a proxy for innovation, while after the deal, this
relation turns to be negative.
In terms of job creation, research has focused on understanding the relationship between
employment growth and VC/PE funding in macroeconomic terms. For example, Wasmer and
Weil (2000) find evidence of the impact on employment of an increase in the ration VC
investment/GDP in a panel of 20 OECD countries. Belke et al. (2003) extend the scope of the
key question testing a virtuous circle between entrepreneurial dynamism, innovative start ups,
dynamic venture capital industry and job creation. The paper delivered pioneering empirical
evidence of such a link at the macroeconomic level, showing that venture capital is able to
significantly raise employment growth and job creation.
Focusing on the second field, recent studies examined empirically the relationship between
receiving venture capital and firm performance. Sapienza (1992) found that the provided
services are positively related to the performance of venture‐backed firms. Most notably, the
stylized facts are: (1) the greater the innovation pursued by the venture, the more frequent the
- 10 -

contact between the lead investor and the CEO; (2) the more open the communication, and the
less conflict of perspective in the venture capitalist‐CEO pair, the greater the value of the
involvement. Lerner (1999) evaluates the long run success of firms participating in the Small
Business Innovation Research (SBIR) program, a major public assistance initiative in the United
States for high‐tech firms. Those receiving assistance from SBIR achieve significantly higher
employment and sales growth rates than similar No‐SBIR assisted firms. These differences are
even more pronounced in ZIP codes with high venture capital activity. Jain and Kini (1995) add
that VCs advice can include marketing services and the upgrading of the commercial network,
fostering an increase in sales. Partially in contrast with the above results stands the work of
Manigart and Hyfte’s (1999) whose findings for 187 Belgian venture‐backed firms are quite
different. Belgian venture‐backed firms do not achieve a significant higher employment growth
compared to non‐venture backed firms of the same industries, of similar size, and similar age.
However, higher growth rates in total assets and cash flow are obvious.
Summarising, many studies have argued and showed that the presence of professional
investors can strengthen the performance of a company owing to many different actions they
can undertake. Thus, as for the case of Italian small business we can test the following
hypothesis:
Hypothesis 2:
It is possible to determine the enhancing role played by VCs and PE funds when
evaluating the post‐investment performances of Italian SMEs.
3.3 Literature on the “Certification Effect”
There is a shared acknowledgment of the fact that financial intermediaries can bring
positive contributions as agents able to produce information about the qualities of firms.
The seminal work of Akerlof (1970) brought the attention to the plausible failures of a
market featured by imperfect information. The model predicts that, in the lack of both defined
guarantees and distinguishable quality, the market may fail: only the average quality of the
goods will be considered, which in turn will have the side effect that goods that are above
average in terms of quality will be driven out of the market, causing the “Lemon market”.
Using the same framework Chan (1983) shows that when all investors have positive search
costs, i.e. they are uninformed investors, the entrepreneurs will find it in their interests to offer
less desirable projects leading to the degeneration of the projects undertaken. Thus, only
“lemons” are offered and investors will not enter the market. Conversely, when some investors
have zero search costs, there is an improved allocation of resources in terms of the
entrepreneurs’ efforts spurring projects with higher investor returns.
- 11 -

Although both of the above researches mentioned the lack of guarantees as a feature of the
imperfect market none of them went further. In the attempt of doing so Booth and Smith
(1986) found important evidence of the so called “Certification effect”. The underlying theory
derives from the expanding literature on the use of reputational signaling to guarantee product
quality, most notable effort is the work by Klein and Leffler (1981). They demonstrate the
conditions under which a non‐salvageable capital expenditure can serve as an effective bond to
guarantee the quality of a firm’s products. As a matter of fact the non‐salvageable investment is
perceived by customers as a commitment to product quality which will be rewarded as long as
the firm does not cheat.
Booth and Smith extend the reputational capital paradigm to explain the role of the
investment banker in certifying the pricing of equity and risky debt issues. In a market where
insiders have an information advantage which might bring the opportunity of a wealth transfer
from outsiders, issuing firms can have the option of “leasing” the brand name of an investment
banker to certify that the issue price reflects available inside information. Otherwise, absent
the ability of insiders to credibly communicate their beliefs or the ability of outsiders to buy
information, a potential market failure of the type identified by Akerlof results: other things
being equal, the proportion of over‐valued firms seeking new outside equity will be greater
than the proportion in the population, leading the outsiders to raise their expected probability
that a firm is over‐valued, finally turning into a decline in firm market value.
To further explain the role of an investment banker as a certifying agent and the above
mentioned “lease” of the brand name, it is worth to illustrate the following example. Consider a
firm that has only limited investment opportunities such that, given the scale economies
associated with new issues, it will seek new equity infrequently (say, every t years). The bond
provided by such a firm is non‐productive except in those infrequent periods when the firm
elects to seek new capital. If instead, it can lease the use of a bond from an underwriter for the
period necessary for inside information to become public, then a perpetuity of rental payments
with frequency t can be substituted for a non‐salvageable investment in the determination of
the firm value, as the one suggested by Klein and Leffler. Thus, n such firms could successively
employ the underwriter’s bonding investment over a single issue interval of length t. In this
manner an underwriter can be perceived as a firm specializing in leasing the bonding
investment to other firms seeking to raise capital.
In order to test the predictions of the model, Booth and Smith developed testable
hypothesis regarding, among the others, the decision to use an underwriter and the amount
charged for the certification. The analysis provides evidence supporting the certification
hypothesis by means of increasing firm value if bonding investment are made to certify the new
issue price and above all, greater net benefit when issuing firms employ a specialist (say,
investment banker) who has made the requisite bonding investment.
Most notably, the mechanism works because the financial institution exploits its own
established reputation and charges the issuing firm accordingly to the magnitude of the
- 12 -

asymmetric information. On this topic we can generalize the contributions of Stuart, Hoang and
Hybels (1999) who document how the performance of young biotechnology firms is affected by
an inter‐organizational certification, or endorsement process, as it operates in the industry's
strategic alliance and equity ownership networks, as well as through the connections between
new ventures and the investment banks that underwrite their securities offerings.
Several different bodies of work stemmed from the above implications: studies were
conducted on new models based at least in part on the formal certification hypothesis 4 ; others
have examined more specifically how financial institutions help to resolve the asymmetric
information inherent in the Initial Public Offering (IPO) process 5 .
Meggison and Weiss (1991) brought consistent evidence finally linking the certifying
hypothesis and the role played by venture capitalists as better informed agents in the
competing market of IPOs. They found that the involvement of a private equity fund at the IPOs
leads to less underpricing. Their substantial contribution is proven by the broad literature
aimed at testing further their finding under different hypothesis, on different markets. Brav and
Gompers (1997) investigated the long‐run return for private equity‐backed (PEB) and non‐PEB
initial public offerings. They proved that the PEB public firms perform better than the non‐PEB
ones, providing evidence that the book‐to‐market ratio at the offering date has a significant
influence on the aftermarket performance. Dai (2007) found that stock performance of VC‐
invested firms is significantly better than Hedge Fund‐invested firms both in the short run and
in the long run. He concludes that coupling the positive role of VCs with substantial ownership,
request board seats and long term investment strives towards the presence of a certification
effect.
Worth a mention, however, are contributions of other authors who found different results.
Arikawa and Imad’Eddine (2006) find out that only the largest top four VCs have a significant
negative impact on the underpricing, and that the top three underwriters have a significant
positive impact. Munsters and Tourani Rad (1994) have been unable to determine the
certification effect for IPOs in the Netherlands.
Despite the fact that this field of study has captured the attention of many researches, as
just shown, there has not been much work on providing evidence of the “Certification effect”
applying in the case of VCs and PE funds backing SMEs not involved in IPOs processes.
In our view this brings about the opportunity to improve previous researches by means of
evaluating eligible variables which can untangle the effects of having such an financial
institution certifying the reliability of a private company, absent other public information.
Therefore the hypothesis to be tested is:
4 Cfr. James (1990) and Blackwell, Marr and Spivey (1990).
5 Cfr. Johnson and Miller (1988), Carter (1990).
- 13 -

Hypothesis 3:
There is a positive relationship between the VC funding and the overall perception of
stakeholders, captured by such economic variables as interest costs to total debt, trade
credit and access to institutional credit.
4. Theory and evidence on the role of VC and PE
We have seen from section 3 that there is not a single theory that by itself is able to explain
the rationale of venture capital and private equity contracts. Hence, in this part we try to draw
on corporate finance theories and previous empirical evidence hints for identifying a list of
controls which will help addressing the hypothesis moved and their relations.
4.1 The Determinants
When testing for the determinants of VC and PE financing specifically addressed to SMEs
firms, a first step implies understanding what are their main characteristics and which of them
are eligible to play a relevant role in the investment decision. Therefore, the purpose of this
section, along with the theoretical predictions found in the literature, is to present a set a
variables which might be perceived suitable for our analysis.
In the field of economic research, it’s a common practice to use firms’ youth and size as
proxies for informational opaqueness 6 , making a link with the two major outcomes of the well‐
known asymmetric information theory: “adverse selection” and “moral hazard”. Both of them
may arise in any investment environment, but they seem particularly acute in the
entrepreneurial finance. With large established firms, investments are made safer by the use of
existing assets as collateral, and the development of reputation. Collateral and reputation
effects can mitigate the negative effects of both adverse selection and moral hazard. Because
entrepreneurial firms lack assets to provide as collateral, and because they lack the “track
record” necessary to establish their reputation, the effects of informational market failures are
more severe in entrepreneurial finance than in financing established firms. Moreover, the
degree of asymmetric information is also likely to be high for firms whose assets are difficult to
evaluate, such as those whose main asset is a new product yet to be launched on the market or
those with a large share of intangible assets in their balance sheets. Thus, not surprisingly, the
financial literature contends that because of their superior scouting and monitoring capabilities,
VC and PE investors are able to deal effectively with the adverse selection and moral hazards
problems.
6 Cfr. Bertoni, Colombo and Croce (2008) for a review.
- 14 -

These first arguments already provide some indications on which variables can be chosen
for our purpose. In the empirical analysis of the determinants of venture capital, we use the
logarithm of firm sales (Size) as a proxy for company size. Age (in logarithm) is calculated using
the date of incorporation of the firm. Finally, we use Intangibles, defined as the share of
intangible over the sum of intangible and tangible assets, to proxy for the difficulty of external
investors in evaluating the activity of the firm. In particular, the variables as Size and Age have
an expected negative sign on the probability of VC and PE finance, meaning the lower the age
and the size, the higher the probability. On the contrary, we expect higher rate of Intangibles
associated to backed firms. Information asymmetry also involves the decision upon the choice
between internal and external capital: negating the hypothesis of frictionless financial markets,
as set by Modigliani and Miller (1958), inefficient equilibria arise. As a consequence, firms
adhere to a ‘pecking order’ in financing their investments (Myers and Majluf, 1984): they first
rely on internal capital, which is the source of financing with the lowest opportunity cost; then,
when internal capital is exhausted, they turn to the external capital source with the lowest cost,
which is usually debt (at least for firms with low leverage). However, as we have already
discussed, SMEs and, in particular, the innovative ones have peculiar characteristics which bring
Sau (2008) to renew the “pecking order” theory allowing venture capital to precede debt
financing. On this basis we introduce the variables Leverage and Short debt (in logarithm)
respectively defined as the share of debt over the sum of debt and equity and overall amount
of short‐term debt granted to the company both in terms of commercial and financial debt. As
in the case for Age and Size, for the same reasons, we expect a negative sign.
Profitability and performance of the firm are other figures an investor may be interested in,
although the evaluation of such deals is usually take into consideration the perspectives of
future earnings. In consideration of the latter, we elect to plug in ROE as fundamental measure
of profitability. In this case, making any predictions on the sign is a difficult task, since
conflicting interpretations coexist a priori: on the one hand, an high value may convey
reassuring information on future returns to investors; on the other hand, it might be perceived
as abundance of internal resources which is negatively related with the probability of venture
financing. Furthermore, Ebitda proxies for performance. Provided that it gives insights on the
ability of the firm to generate cash flows from the core activity, higher values are expected to
rise the probability of financing.
Finally, from the strand of research which tries to find connections between VC and PE
financing and the degree of innovation, both at aggregate and individual level, we identify
Growth, defined as the rate of change of sales for each company, Capex, defined as the rate of
change of fixed assets which proxies for firm investment, and the High‐Tech dummy, which
takes the value 1 in industries with a high “innovative” content 7 , as additional variables of
7 Using the four‐digit industry codes, we classify a firm to be a high‐technology one if it belongs to one of the
following industries: chemical and pharmaceutical products, aerospace, electronic equipment, media,
telecommunications, software and hardware.
- 15 -

interest. As a matter of fact one can argue that especially for small firms, VCs and PEs bring
along additional consultancy which include marketing services and the upgrading of the
commercial network; both aimed at fostering the increase in sales. This acts as a response to
the enterprises’ need to enhance their results, after a period of an intensive investment
campaign and/or growth. In this case the expected sign will be positive. Intangibles, once
considered as close estimates for the effort put in innovation, should, again, have a positive
sign.
4.2 The effects
In the previous paragraph we showed that theories on asymmetric information have a
relevant impact on the characteristics of SMEs enterprises, which in turn can shape the sign of
the relation between them and the probability to be backed by VCs and PE funds. On the other
side we will find that similar intuitions will drive the expectations on the eligible outcomes once
the investment took place. In the first instance, Ueda (2004) presents a model in which venture
capitalist and the entrepreneur are equally informed about the projects; this fact facilitates
financing to firms with low collateral but exposes the entrepreneur to the risk of project
expropriation by the venture capitalist. For this reason the model also predicts that, after the
deal, profits should increase to compensate the entrepreneur for the risk of being expropriated
by the venture capitalist. Hence, some measure of profitability (such as ROE – return on equity
– or ROA – return on assets) should be comparatively higher than that of other firms.
Nevertheless, these relations can be undone by other factors such as, respectively, a striking
increase in the figures of equity as recorded in the balance sheet and an ongoing campaign of
assets acquisition after the involvement of the external investor.
At this stage it is possible to further develop the issue concerning the presumed impact of
the venture capital industry on degree of innovation. Indeed, gathering evidence on the
consequences of VC should help to shed some light on the direction of causality; if venture
capital first hypothesis dominates in the Italian case, then we should expect an increase in
Capex, Growth, Intangibles. In fact, the injection of capital will stimulate the acquisition of the
essential inputs enabling the production and the marketing of the product. If, on the contrary,
is the innovation first hypothesis that dominates, we do not expect either accumulation of
tangibles assets (measured by Capex) nor Growth to continue after the deal, but rather figures
indicating the consolidation of firm’s results. The organization should have already the assets
required and it’s mainly seeking a push for its sales. As a consequence, VCs and PEs can also be
perceived as consultants, owing to their ability to transfer to firms in which they are involved
new values, as the above mentioned efficiency, and welcome them in a wider network of
relationships, a fundamental source for future outstanding results. In accordance to the latter
Sales of the backed firms should prove better results than non‐backed firms. Moreover, in the
- 16 -

case of turn‐around or buy‐out operations (presumably for larger firms) we expect an increase
in efficiency, which we proxy with the (log of) Value added per employee.
Finally, in this paragraph we can make some remarks on Hypothesis 3 concerning outcomes
of VC and PE financing which can allow for the presence of certification effect. Practice tells us
that inter‐organizational exchange relationships can act as endorsements that influence
perceptions of the quality of young organizations when unambiguous measures of quality do
not exist or cannot be observed. As a result, the valuations of young firms are at times
influenced by the characteristics of the affiliates of the companies under scrutiny. Because
strong relationships with prominent organizations convey the fact that young companies have
earned a positive evaluation from experienced and influential actors, associations with high‐
status organizations elevate the reputations of new ventures. Therefore, endorsement by a VC
or PE investor makes it easier for portfolio firms to obtain access to other external financial
resources and to tangible and intangible assets (e.g. distribution channels, manufacturing
facilities, sales force) possessed by other firms through the establishment of alliances (Colombo
et al., 2006; Hsu, 2006). This latter effect further relaxes the financial constraints these firms
would otherwise experience, as these resources do not need to be built internally.
As a first consequence of a relationship with prominent organizations the expected
likelihood of survival would be higher than ex‐ante. Thus, we could trace beneficial effects in
the perceptions of suppliers: according to this Trade Credit, a measure in days which represents
the average length over the fiscal year of the purchases, would likely increase. Nevertheless
Petersen and Rajan (1997) provide striking evidence for different relations: they found Trade
credit to be linked to access to credit granted by banks in a way that suppliers could substitute
for financial institutions owing to an advantage over traditional lenders in investigating the
credit worthiness of their clients, as well as a better ability to monitor and force repayment of
credit. As a consequence, under this setting, the expected result from our analysis would have a
negative sign.
Similar consideration could also apply to the measures related to debt, which we consider
here as another component of the certification effect. Variables Debt short and Debt, calculated
in logarithm terms, are used for this purpose. We have already encountered a series of
contributions devoted to address the financial constraint borne by SMEs. According to these
studies, VCs and PEs are able to exploit their established reputation, and hence, can meliorate
the rating of firms in which they decide to invest, resulting with a broader and better access to
the funds of a bank. However, we can reach opposite findings when interpreting VCs and PEs in
their role of helping to re‐balance the financial structure. if this is a valid theory, then we would
expect to see lower values for both Debt and Debt short. To test whether backed firms improve
their credit standing we resort to a definition of cost of debt (Financial Expenses) as the one
identified by Hyytinen and Pajarinen (2007) expressed by the ratio of interest costs to total
- 17 -

debt 8 . A negative sign, in particular coupled with an increase of debt, would give clear evidence
of better conditions applied to backed firms, in turns allowing for the presence of the
certification effect.
4.3 The econometric set‐up
In order to test the hypothesis outlined we rely on econometric techniques applicable to
panel data, allowing us: (1) to control for unobservable individual heterogeneity; (2) to use a
large amount of information, including many companies and several years for each company,
thus increasing the degrees of freedom and reducing colinearity between the explanatory
variables; (3) to analyse the evolution over time of the variables in a group of companies.
A logarithmic transformation has been applied to most of the variables. Applying this
procedure we obtain beneficial effects such as: steadying the variance, reducing multiplicative
effects into additive ones and normalizing the distributions. Since the transformation it is not
feasible when there are null and negative values we first add 100 to each value and then
calculate their logarithms. In particular, this was the case for variables obtained as variation
between two subsequent periods such as Growth, Capex and ROE.
4.3.1 Econometric set‐up for the determinants
In this section we present a multivariate analysis to test Hypothesis 1 which will allow us to
quantify the importance of the different determinants for the financing through Venture
Capitalists and Private Equity funds.
Based on the theoretical predictions on the variables that should affect the likelihood of an
external funding, we estimate various versions of the following probit model:
Pr(Backedi,t=1)= F(β1Agei, t‐1 + β2Sizei,t‐1 + β3Size 2 i,t‐1 + β4Intangibles i,t‐1 + β5Ebitdai,t‐1 +
β6ROEi,t‐1 + β7Leveragei,t‐1 + β8High‐Techi,t‐1 +β 9Capexi,t‐1 +
β10Growthi,t‐1 + β11Short Debti.t‐1 + κiArea +ηiYear)
The multivariate Probit model uses Backed as a discrete variable representing a choice from a
set of mutually exclusive choices: it equals 1 when firms are backed, 0 otherwise. Yet to be
described are controls Area and Year. The former focuses on geographical characteristics that
may be involved in the investment decision. The firms are divided into three groups according
to the location of their registered offices: north, centre and south are the different macro‐
regions identified. This strategy joins the procedure for the selection of the control group
8
The cost of debt calculated in this way will underestimate the real one since the scaling variables (total debt)
include items that are not interest bearing.
- 18 -
(1)

described in section 4.1. Finally, Year includes the years when the deal took place and controls
for specific common effects.
4.3.2 The econometric set‐up for the effects
In this section we present the procedure adopted to analyze the performance – in terms of
various balance sheet indicators – of backed firms relative to the companies that did not
receive this form of financing.
For the main accounting and financial variables (denoted yi,t) we estimate the following
fixed‐effect regression:
yi,t = α +β1Deal0 + β2Deal13 + ut + dt + εi,t
Where Deal0 is a dummy variable which takes value 1 in the year of the deal. It should be
pointed out that if the firm is financed more than once in our sample period, the dummy takes
value 1 more than once, specifically in the year of each operation. Deal 13 is a dummy equal to
1 in the three years after the deals, which is considered the average holding period for the
industry.
Regarding the estimation method, there is a discussion as to whether the individual effects
should be treated as fixed or random variables. However, this is not an important distinction
because we can always treat the individual effects as random variables without loss in
generality (Wooldrige 2002). However, it is really important to determine whether or not these
individual effects are correlated with the variables observed. To test for the existence of this
correlation, the Hausman test (1978) is usually used. If this test does not reject the null
hypothesis that the individual effects are not correlated with the explanatory variables, the
most suitable estimation would then be the random‐effects model and the best estimator
would be Balestra‐Nerlove’s (1966) generalised least squares estimator. If, on the contrary, the
null hypothesis is rejected, the within groups ordinary least square estimator would then be the
most suitable one. More intuitively, implementing a fixed effect regression allows us to control
for firm‐specific characteristics that are time‐invariant but that could be correlated with the
deals, such as industry or managerial quality.
5. Methodology and data
5.1 Methodology
The key objective of the research is to outline a general framework for the investment deals
realized by VCs and PE firms with respect to Italian SMEs. Being this our aim, foreigner investors
operating in Italy are also included.
- 19 -
(2)

The first step required was the construction of a solid database of firms which received
some sort of external financing and for which separate financial accounts exist. Resorting to the
Italian Private Equity and Venture Capital Association newsletter, we were able to identify
names of most of the actors involved in the Italian market. Only privately held VCs and PE funds
were considered, thus excluding all the deals carried out by publicly‐controlled investors such
as agencies for regional development. This is consistent with the aim of the research of
providing evidence for the PE industry alone. We also excluded all the deals which were not
characterized by a direct involvement of the investor, such as acquiring shares of fund of funds
given a general priority of portfolio diversification. In case the information on the deals was not
published on the investors’ web sites we resort to ah‐hoc databases such as Zephir (© Bureau
Van Dijk Electronic Publishing) specialized in reporting information on M&A activity, joint
ventures and private equity deals. For the period 1997‐2007 a dataset of 730 deals resulted,
including also firms with more than one stage of financing.
As for identifying a threshold of what to consider as SMEs we looked at the criterion
established by the European Commission which states 9 :
“A medium‐sized enterprise is defined as an enterprise which employs fewer than 250
persons and whose annual turnover does not exceed EUR 50 million or whose annual balance‐
sheet total does not exceed EUR 43 million. A small enterprise is defined as an enterprise which
employs fewer than 50 persons and whose annual turnover and/or annual balance sheet total
does not exceed EUR 10 million. A micro‐enterprise is defined as an enterprise which employs
fewer than 10 persons and whose annual turnover and/or annual balance sheet total does not
exceed EUR 2 million.”
At this stage we decided to focus only on industrial firms. As suggested by Booth and Smith
(1986) public utilities and banking firms operate in regulated sector and thus, should show
different patterns in terms of firm‐specific risk. Moreover, due to the above mentioned
regulations, the latter industries tend to have limited certification costs which is, instead, one of
the issue this research tries to untangle.
The presence of leverage buy‐out operations (LBOs) implied a special attention to the
correct identification of the legal entity entitled to issue the representative balance‐sheet data.
If a reverse merger scheme occurred the target company was the entity of interest; in case of a
forward merger scheme we looked for the balance‐sheets of both NewCo and target company.
In the light of the above concerns, we filtered the group of firms just constructed obtaining
a final sample of 160 VC/PE‐backed companies. This number also reflects the availability of
9
Cfr. Commission Recommendation 2003/361/EC of 6 May 2003 concerning the definition of micro, small and
medium‐sized enterprises [Official Journal L 124 of 20.05.2003].
- 20 -

sufficient financial data in the database AIDA 10 . Common problems such as the lack of a
sufficient number of observations for each firm and the exclusion of some of them from the
database itself prevented the sample to have a bigger size. Presumably, this can be a direct
consequence of the fact that we are dealing with SMEs firms, in the sense that looser
regulations might apply in terms of disclosure policy. Manigart et al. (2005) bolsters this
intuition finding clear evidence that generally firms switch to a higher disclosure policy just one
year before the stake acquisition by a PE investor takes place.
Further, the analysis implied a second step devoted to identify another set of firms with
similar characteristics to the first ones but did not receive any external funding, so that we were
able to match each company in the sample group with a non backed one. The method for the
selection of the control group is as follows: (1) we picked out all the companies active in the
same region to control for the economic growth in that particular context, (2) we selected
those belonging to the same sector of activity using the ATECO 2002 11 codes in four or six digits,
upon what was the case, (3) we filtered the companies that were within the same range of
sales in the year of the funding event, (4) we finally selected the company that was closer in age
to that of the sample. In many cases we found a company founded in the same year.
Provided that growth patterns differ in companies at various stages of development, we
classify the VC‐PE backed firms and their respective comparables into three groups: start up,
growth and late stage financing. Firms that receive the first round of VC funding from the start‐
up point to the moment they reach break‐even are included in the start‐up group. Those firms
with a track record of earnings, that receive their first round in order to finance the expansion
of the business through a capital increase, are grouped into the growth stage. Buyouts,
turnaround and replacement capital deals, which generally do not involve an entry of fresh
money into the firm, are classed as late stage investments 12 . The same intuition may also apply
when comparing smaller firms and bigger firms; accordingly, within the sample and the control
groups, we divided the firms with a value of total assets of less than 10 million Euro in the first
year of presence in the database from those with a value above the threshold set.
10 © Bureau Van Dijk Electronic Publishing. The financial data are provided by Honeyvem (www.honeyvem.it)
which acquires and revise the balance sheets deposited in the Italian Chambers of Commerce. For each company
AIDA merges in one document the figures of the previous 10 years, or less upon the availability, and adds
information on shareholdings and management for the first 20.000 Italian firms.
11 ATECO 2002 is the classification proposed by the Italian Institute for Statistics (ISTAT) based on the lines set by
the European counterpart, NACE. Under this classification the different economic activities are grouped in
sections, sub sections, branches, groups, classes and categories. For most of the sample group we were able to
identify a control company active in the same region and in the same category of activity. From 1 January 2008 it’s
available another classification, ATECO 2008, which substitute the different activity codes for statistics purposes
and fiscal purposes with one code.
12 In the case that there was not public information available we use the firm’s age as a proxy for the stage of
investment. In particular, we define early stage investments as investments of a VC in firms no more than three
years old. As an approximation it may fail to state the real nature of the deal, however, a longer distance between
foundation date and begin of involvement make it more difficult to take into account the initial founding
characteristics as crucial determinants for firm growth.
- 21 -

The methodology adopted has some drawbacks mainly related to the usage of sole public
data. Many others studies take advantage of questionnaires sent directly to VCs and PE funds
to acquire more in depth information as for financing terms of the investment, firms’ equity
ownership, contingencies to future financing. However, the validity of the samples obtained
might be affected by: (1) survivorship bias, in the way that the survey is delivered to investors
who are still in business, (2) positive bias, because it is likely VCs and PE funds report
performance of only those firms doing well. Conversely, our dataset potentially addresses these
issues since it includes information for firms which are either undertaking winding‐up
procedures or did not perform well. Nevertheless, as already mentioned, we could not track
down the financial data of each and every company due to two main reasons: (1) the source at
our disposal was not comprehensive of all the activities, (2) especially for small and medium‐
size companies the amount of information provided has a positive relation with costs to
produce and publish its financial data and eventually rises the awareness among its own
competitors.
5.2 Data
Figure 2 and table 1 aim to provide an overview of the Italian market for Venture Capital
and Private Equity, as the one captured by the firms included in our dataset over the survey
period 1997‐2008. It gives some indications on how the industry behaved as for geographical
distribution, typology of investment and preferred sectors of activity. Not surprisingly the North
of Italy, with 101 deals, shows a very high grade of entrepreneurship with Lombardia being the
region leader in Italy in terms of investments attracted. The Centre Italy accounts for 32,5% of
the market where Emilia‐Romagna and Toscana played a relevant role. The South Italy is
characterized by a low level of interest: Campania, Puglia and Basilicata saw one investment
each while Calabria three. However, since we did not include publicly‐controlled investors in
our survey the flow of investments for Southern regions should be considered underestimated.
Not surprisingly either, from figure 2 we see that sectors with a high potential of innovation
were preferred: as a matter of fact, DL sector, by its own definition, it’s involved in activities
more prone to be technology driven, while companies belonging to sector K work in fields such
as IT solutions for firms and consultancy services where intellectual capital is fundamental; at
the same time within the more traditional DK sector there is a broad range of activities which
imply an intensive usage of technology.
Panel A of Table 2 reports the summary statistics on the control sample. Data are averages
over the period before the involvement of the external financer. The time span considered is
not constant over the firms but it’s calculated upon the first year in which the firm is present in
the database and the year the deal takes place. The median firm of this group records a value of
sales of € 12.2 million, total assets of 10.4 millions, 50 employees and is 20 years old. Intangible
assets represent less than 5% of intangibles and fixed assets. As for profitability, the return on
- 22 -

equity is 9 per cent; the median added value per employee is 51 thousands of Euros. In terms of
financing the sample shows a high grade of leverage (defined as the ratio of debt over the sum
of debt and equity) mainly related to short‐term debt (defined as the sum of commercial and
bank debt due within 12 months).
Panel B reports statistics for backed firms. For each variable a star indicates whether the
difference between the mean of the control sample and that of the backed firms is significant
at the level of 5%. The first thing to note is that backed firms are generally younger (18 years).
As regards size, in this case firms tend to be bigger. The median firm exhibits higher sales, total
assets and number of employees, respectively 13,2 millions 11,9 millions and 61 employees.
This is in contrast with the suggestions of theory and empirical evidence relative to the United
States. Coupling this results with the ones reported in the second column, which reports the
mean, it can be argued there are few big firms which strive upwards the averages. Indeed,
worth nothing is the fact that in the backed‐sample there is a wider dispersion – as measured
by the difference between the 99 th and 1 th percentile – of the variables that proxy for size. As
for profitability ROE and Value Added per employee show better performance though for the
former the difference is not statistically significant. Intangibles are sensibly higher; assuming
intangibles can proxy for innovation this result may strengthen the intuition that backed firms
tend to be more innovative than others. To notice, also, the lower level of leverage which is
statistically significant: both in terms of overall level and short‐term, the median firm relies less
on debt. Trade credit, which is a variable of interest when investigating for the certification
effect, lasts longer for backed firms. This is consistent with the intuition pursued by Petersen
and Rajan (1997) who find that small firms which don’t have broad access to credit from
financial institutions exploit much more trade credit, meanwhile firms with better access to
credit offer more trade credit. In confirmation of this aspect, debt service applied to backed
firms is sensibly higher than the one applied to others, indicating increasing interests for the
formers.
Panel C and D switch the attention to the large firms sub‐group of the sample. The
comparison indicates that backed‐companies are younger and bigger in size, following the
results showed in previous panels. Moreover, they show better results in terms of
performance, though ROE is again not significant; intangibles are still higher giving hints for a
stronger innovative attitude. Contrary to the evidence for smaller firms, large backed‐firms are
more indebted than those in the control group. Short‐term debt represents the most important
source of financing for both the groups since the difference is not significant. Worth a mention
is the fact that both backed‐subgroups grow less in terms of variation of sales, while the better
level of marginality is reassuring
Summing up the descriptive analysis shows that backed firms are younger, grow less, have a
larger share of intangibles and lower profits. For larger firms another difference with respect to
the control sample is the higher level of indebtness.
- 23 -

5.3 The determinants of venture capital financing
Table 4 presents the results obtained estimating equation (1). In column (a) attention is
confined to the variables with the coefficients from β1 to β8 (controlling for geographical area
and time dummies). This is done in order to maximize the number of observations on which
estimation is performed; in fact, employing variables such as Capex and Growth would imply a
loss of observations given that they are calculated over year t‐1.
Consistently with the theories of asymmetric information, the signs of Size and Age turn out
to be negative; the existence of a non‐linear relation between the probability of receiving
external capital and size also emerges. In the same direction goes the positive sign and the
statistical significance of the variable Intangibles. Forthcoming considerations will conclude on
the intuition by which banks are more keen on granting credit to firms with higher liquidation
value of assets. In terms of profitability Ebitda appears to be an important determinant; we can
anticipate here that its sign and its magnitude will be constant in all the different specifications
of the model. Provided that Ebitda is a measure widely used in the financial industry for the
assessment of the value of a company, e.g. the “multiples approach”, thus, this result can be
considered coherent with the state of art in use. Conversely, investors seem not to rely much
on the figures for ROE and see other variables as important factors for predicting the future
performances of a firm. Looking at the negative sign we can assume that the internal finance is
not sufficient and is curbing firm’s investment decisions, hence its growth opportunities. In this
way, we would strengthen the theory of Carpenter and Petersen (2002a). Finally, the negative
sign of Leverage backs the intuition that predicts a higher demand for venture capital finance
by firms which encounter more difficulties in accessing debt financing.
We check the robustness of these results in various ways. We re‐estimated specification (a)
using different lags of the variables. In particular, if the variables are entered with a lag of one
year (results not reported), the basic results are confirmed, with the exception of Size and
Leverage which become not significant.
Column (b) of the table reports the results obtained estimating the richer model for the
whole set of variables reported in equation (1). In particular we introduce Capex and Growth to
measure the expansion of the firm and Short Debt to capture the access to credit. Results
confirm that firms that are younger, smaller and have a higher share of intangibles assets in
their balance sheet are more likely to be financed. Leverage is still significant though at 10% of
significance, the dummy High‐Tech and Roe are still not significant. As regards the adding
variables of specification (b), we find contrasting evidence between the two determinants
chosen to proxy for firm expansion: while Capex is positive and not significant, Growth is
positive but significant. On this basis it is possible to argue that the firms that have as their
most valuable asset their knowledge are more likely to be financed by venture capitalists.
- 24 -

Finally, we find Short Debt being negative and statistically significant. This constitute a solid
ground for stating that firm with less access to the channels of traditional funding will seek
more likely for capital through issuing equity. Column (c) shows the results when data for early
stage financing are withdrawn. The estimates are consistent with the previous models, except
for Leverage which becomes not significant and changes its sign.
5.3.1 Differences according to firm size
As already outlined, attitude of smaller companies may substantially differ from the attitude
of bigger ones. Thus, this part provides estimations of the same kind as in the previous section,
while applied to a different sample of data, namely, firms with total assets above 10 million
Euros. We run estimations only for specification (a) and (b) since in this sub group there are no
firms eligible for early stage financing deals. Moreover, given the new framework that we are
considering, we see private equity funds as the main players.
Column (a) presents the estimates for the first block of variables. There are striking
differences such as Age being no longer significant and High tech positive and significant. The
first outcome reasonably highlights the positive relation between size and age, where older
firms are more likely bigger. The second outcome gives a new intuition according to which
Private Equity funds invest mainly in consolidated High tech firms, that have already a feasible
market for their products; this comes as support of the “innovation first hypothesis”. Applying
the same reasoning, we can interpret the lack of significance of High‐tech in the case of smaller
firms
Adding the second block of variables does not change the results of the previous paragraph.
Conversely, a final comparison between the two richer models can be drawn. It shows
important changes like the loss of significance for both Growth and Short Debt, when compared
to column (b) of panel B. For the first case a conceivable explanation is that firms could operate
in mature markets and the investment of a PE fund is driven by the need of renewal of board of
directors or a restructuring of the company foreseeing a new phase of expansion; as for Short
Debt, since bigger firms have a grater access to credit granted by banks the second variable can
not be a key driver for attracting external capital.
5.4 The effects of VC and PE financing
Table 6 separately reports the estimates for equation (2) for smaller enterprises against
larger firms of the sample. We start with addressing the profitability of SMEs. ROE drops with
respect to the other firms in the interim period (i.e. from t+1 to t+3). This is confirmed when
using another measure for profitability, ROA. In this case we find negative and statistically
significant sign, too. At a first sight this may convey bad news. Instead, if we couple the
estimate of ROA with Total assets more reassuring evidence arise: once the deal has been
- 25 -

sealed, backed firms went through an important period of investments which shrank the index.
As Intangibles shows, the investments pointed towards booth kind of assets.
We tried to untangle the effect on performance of backed firms using such variables as
Sales and Value added per employee. Both have positive signs. The presence of the venture
capitalist or private equity fund seems to be effective in terms of sales during the interim
period. As we have already argued, we find explanation for this owing to the additional
consultancy services provided by the external professional investors. Instead, the Value added
is statistically significant only in the year of the deal. We can try to explain this taking advantage
of the outcome of the variable Employees. In fact, the latter points towards a positive and
significant difference in the first year, while in the subsequent time frame it does not. Hence,
assuming that this can be reflected in its relative proxy for performance, we find a sound
justification for the change in significance, namely the increased number of employees.
Unfortunately, if we assume this relation to be valid, we can not outline any interpretation as
for efficiency.
The block of the three subsequent variables allows us to add some comments on the
question whether the presence of an external financial institution plays also an important role
determining what we called a “certification effect”. A first standpoint is that backed firms get
greater access to credit as shown by the variable Debt, which is positive. This is an important
difference since we saw in the ex‐ante analysis that this was not the case. Moreover, the proxy
we use for cost of debt (Financial expenses), has a significant and negative coefficient, leading
us to argue that with the entrance of a professional investor the firm’s credit standing improves
and the interest rate charged lowers. Finally, Trade credit is negative and significant. As an
explanation of this we can borrow again the theory and evidence brought by Petersen and
Rajan (1997) establishing a negative relation between amount of credit granted and the
exploitation of trade credit as a form of funding.
When looking at larger firms it is possible to draw similar conclusions for most of the
variables. In our view variable Debt has different interpretations which leave some open
questions. When compared to smaller firms we can see that’s not significant in the first period
while it does so in the following period with a smaller magnitude, though. One could argue that
this is a hint for the intervention of the external financer in re‐balancing the financial structure.
At the same time it is possible to state that large firms which successfully dealt with private
equity firms often undertake structured operation such as MBOs or LBOs which always imply a
high usage of leverage, which usually merges into the company balance sheet. Thus, explaining
the positive coefficient. Unfortunately, our database doesn’t provide us with sufficient
information to untangle this issue.
As a final remark the evidence on the ex‐post performance, coupled with the results of the
ex‐ante, that showed a significant correlation between growth and the probability of venture
capital, seems to indicate that the “innovation first hypothesis”, rather than then the “venture
capital first hypothesis” is validated by our data. The funding occurs after a higher than average
- 26 -

period of investment and growth. On the other hand, the rate of change of fixed assets (Capex)
decreases while for sales its growth can not be assumed higher than the one of the control
group. These arguments contribute to the intuition of the consolidation of a firm’s result, rather
than spurring further innovation and growth. Thus, VCs and PE funding, though not directly
affecting the measures of growth, seem to facilitate consolidation in firms’ results.
6. Conclusions
In this paper, we study the characteristics of the relationship between SMEs, venture capital
and private equity investors by means of a database including 160 deals signed in Italy.
The empirical analysis has shown that venture capitalists are more likely to step in for firms
which are younger, smaller and more endowed with intangible assets than the average.
Additionally, this brings factual hints on the their positive role when asymmetric information
problems are of utmost importance, and there is broader scope for adding value.
At the same time, the ex‐post analysis indicates that smaller firms benefit from venture
capitalists achieving better results, in comparison with the control group, in terms of sales,
employment and expenditures on innovation. Moreover, considering that the capital infusions
appear to be more frequent after periods of higher than average growth and investment, we
can argue that larger firms resort to private equity with the aim of consolidating their
performance. Accordingly, firms appraise the additional consultancy services granted.
We also tested the hypothesis of VCs and PEs as certifying parties. Departing from the
traditional background of IPO underpricing, which has been already widely investigated, we
outlined a new framework based on a cluster of balance sheet indexes. Thus, the original result
of this study is, in our view, the confirmation of the presence of the “certification effect” under
new circumstances and applying to SMEs, which are seldom considered. However, for a better
reliability of our results it would be advisable to spread the analysis to a wider geographical
area and verify whether there are similar patterns across different countries.
As a final remark, the empirical evidence supports the thesis of venture capital backing
innovative businesses rather than supporting new entrepreneurial ideas from scratch. From a
practical viewpoint, it highlights the limits of the sole private initiative in encouraging
innovative companies, embodied here by VCs and PEs, and leaves to policy‐makers the task to
bridge the gap. Not surprisingly, this is also the position expressed in several official documents
by the European Commission.
In line with the above, we can suggest, as a future line of research, to further investigate
the characteristics of the Italian financial system which hinder venture capital to fully expand its
range: the experience of Anglo‐Saxon countries states that, under proper regulations, this could
be attained. A branch in this field could be the assessment of the performance of state‐owned
- 27 -

egional agencies and the role of Universities as incubators. Although we did not include them
in our survey, we perceive their contribution as a fundamental support to start‐up a new
business.
- 28 -

An overview of the Sample
50
40
30
20
10
0
Source: Rapporto PEM© 2006
2006 vs 2005: % Distribution by target company sales (€ Mln)
300
Others
28%
DG
5%
2005 2006
% Distribution by Sector of activity
DJ
7%
DA
8%
- 29 -
G
9%
K
18%
DK
12%
DL
13%
Figure 1
Figure 2
Following ATECO 2002 classification K : Hiring services, IT solutions, Research e Services for firms; DL : Electric, Electronic and Optic Machineries
and Equipments; DK : Mechanical Machineries and Equipments; G: Wholesale and retail trade; DA: Food and Tobacco; DJ: Metallurgy; DG:
Chemistry.
Source: Our analysis
_____________________________________________________________________________
Geographical
distribution
Source: Our analysis
Characteristics of backed companies
North
Center
South
101
52
7
63%
33%
4%
Stage of
financing
Early stage
Expansion
Late stage
13
62
85
8%
39%
53%
tot 160 100% tot 160 100%
Table 1

Descriptive statistics Table 2
in Panel A, the summary statistics refer to the control sample of SMEs. In Panel B data refer to SMEs that received financing. In Panel C, they
refer to the control sample ot large enterprises (defined as those with at least 10 million of total assets). In Panel D data refer to large firms
that were financed. A * indicates that a test of the equality of means between the control sample and the backed sample is rejected (at least
at 5 per cent).
Variable
Number of
Obs.
Median Mean Std. Dev 1° petile 99° petille
Panel A: Control sample ‐ SMEs
Age 160 20 23,8 14,9 6 94
Sales (€ mill.) 673 12,1 16,2 14,1 0,6 59,8
Ebitda (€ mill) 673 0,9 1,55 21,9 ‐3,3 9,3
Total Assets (€ mill) 673 10,4 14,7 149,0 0,5 71
Equity (€ mill) 673 1,9 3,4 0,0 43,1 19,3
ROE 658 9 11,8 25,9 25,9 91
Intangibles (€ mill) 672 0,1 0,4 10,7 0,0 5,8
Tangibles (€ mill) 673 1,1 2,7 43,3 0,1 23,2
Leverage 673 0,7 0,8 0,17 0,3 0,99
Debt (€ mill) 673 6,3 10,2 12,0 0,3 55,9
Debt short (€ mill) 673 5,5 9,0 115,0 0,0 55,9
Gross margin (€ mill) 673 2,7 4,3 52,5 0,03 33,3
Debt Service (€ mill) 673 0,1 0,3 4,1 0,01 1,87
Corporate Taxes (€ mill) 673 0,2 0,47 7,3 ‐0,02 3,6
Growth 574 7,6 17 62,2 ‐80,3 207,9
Number of employees 648 50,0 85,1 113,2 2 601
Value added x
employee 661 51,0 59 37,5 2 263
Capex 527 ‐0,2 39,3 806,0 ‐77 110
Trade Credit 671 97,5 99,0 98,3 0 434
Net profit (€ mill) 654 0,3 0,6 11,1 ‐0,2 4,9
Panel B: backed‐SMEs
Age 160 18 21,4* 15,5 2,0 71,0
Sales (€ mill.) 573 13,2 18,9* 180,1 0,0 83,9
Ebitda (€ mill) 572 1,7 2,6* 28,5 ‐1,5 12,8
Total Assets (€ mill) 568 11,9 19,7* 27,0 0,3 141,0
Equity (€ mill) 572 3,1 5,2* 72,7 0,0 31,5
ROE 565 12,9 11,6 59,2 ‐102.7 81,5
Intangibles (€ mill) 567 0,21 1,3* 30,3 0,0 16,4
Tangibles (€ mill) 574 1,4 3,7* 62,8 0,0 25,8
Leverage 571 0,7 0,7* 0,2 0,1 1,0
Debt (€ mill) 572 6,1 9,4* 228,0 0,1 99,0
Debt short (€ mill) 567 5,5 8,9* 131,0 0,0 59,2
Gross margin (€ mill) 572 3,8 6,05* 7,2 63,0 44,7
Debt Service (€ mill) 573 0,2 4,4* 7,2 0,0 3,2
Corporate Taxes (€ mill) 574 0,4 0,66* 8,7 ‐0,4 4,4
Growth 471 10,5 15,6 42,2 ‐92,5 191,2
Number of employees 571 61 107* 13,1 4,0 560,0
Value added x
employee 529 56 67* 47,8 0,0 251,0
Capex 449 26 36 71,1 ‐7,9 146,0
Trade Credit 577 109,4 121,1* 150,0 0,0 93,6
Net profit (€ mill) 564 0,3 0,6 13,9 ‐2,0 4,9
- 30 -

Variable
Number of
Obs.
- 31 -
Table 2/ctd
Median Mean Std. Dev 1° petile 99° petille
Panel C: Control sample ‐ Large Firms
Age 160 21 24,8 14,9 6 94
Sales (€ mill.) 815 25,8 27,5 15,0 2,2 0,3
Ebitda (€ mill) 815 2,1 29,7 37,0 ‐3,3 18,2
Total Assets (€ mill) 815 23,2 27,5 191,2 6,2 104,0
Equity (€ mill) 815 5,1 7,1 62,8 0,1 34,2
ROE 801 8,0 9,3 21,3 ‐63,0 62,0
Intangibles (€ mill) 814 0,2 0,9 24,0 0,0 13,9
Tangibles (€ mill) 814 2,8 5,5 87,6 0,6 47,8
Leverage 673 0,7 0,7 0,17 0,3 0,99
Debt (€ mill) 815 13,7 18,5 17,3 1,6 86,3
Debt short (€ mill) 815 11,4 16,0 164,1 1,1 84,3
Gross margin (€ mill) 815 5,9 7,4 5,6 0,3 27,7
Debt Service (€ mill) 815 0,3 0,5 5,5 0,1 2,6
Corporate Taxes (€ mill) 815 0,5 0,8 9,2 ‐0,6 2,9
Growth 815 18,7 30,5 351,3 1,2 64,5
Number of employees 786 110,5 134,3 119,1 14,0 783,0
Value added x employee 801 57,0 65,3 37,1 11,0 229,0
Capex 727 ‐1,1 32,8 28,7 ‐84,5 445,5
Trade Credit 815 107,0 120,4 91,0 0,0 463,9
Net profit (€ mill) 796 0,4 0,7 16,8 ‐4,3 6,1
Panel D: backed‐Large Firms
Age 160 19 22,4* 15,5 2 71
Sales (€ mill.) 1627 26,1 29,8* 23,8 0,5 99,3
Ebitda (€ mill) 1625 25,7 3,5* 43,1 ‐3,9 20,4
Total Assets (€ mill) 1626 25,0 33,3* 297,2 5,7 146,0
Equity (€ mill) 1626 6,4 9,1* 10,0 ‐0,1 50,4
ROE 1600 7,0 7,1 44,6 ‐0,1 84,0
Intangibles (€ mill) 1625 0,4 3,5* 90,2 0,0 47,8
Tangibles (€ mill) 1624 3,2 6,3* 93,5 0,06 43,6
Leverage 571 0,6 0,7* 0,2 0,1 1,0
Debt (€ mill) 1626 14,7 21,9* 24,2 2,1 129,0
Debt short (€ mill) 1626 11,9 17,2 17,6 1,6 92,8
Gross margin (€ mill) 1625 6,5 8,6* 7,5 ‐0,3 39,9
Debt Service (€ mill) 1625 0,4 0,6* 9,2 0,03 4,0
Corporate Taxes (€ mill) 1627 0,6 0,9 1,1 ‐0,9 4,6
Growth 1627 11,4 13,9 1,8 9,9 17,3
Number of employees 1594 113,0 145* 3,6 9,0 809,0
Value added x employee 1544 60,0 68,13* 4,1 0,0 234,0
Capex 1432 ‐0,1 37,6 38,8 ‐0,8 0,4
Trade Credit 1634 109,5 125,3* 106,6 0,0 466,0
Net profit (€ mill) 1594 0,4 0,6 261,3 ‐6,8 6,2

Variable definition
Age age of the firm (in logarithm)
Size log of total firm sales
Intangibles ratio of intangible over the sum of intangible and fixed assets
Ebitda log of Ebitda
Leverage debt over the sum of debt and equity
ROE profit over equity
High‐tech dummy equal to 1 for companies in high‐tech sectors
Capex rate of change of fixed assets
Growth rate of change of sales
Short debt log of total short‐term debt
Total assets log of total assets of the firm
Employees log of total number of employees
Value added x employee ratio value added over number of employees
Trade credit (average of commercial debt over purchases)*360
Financial expenses ratio interest costs over total debt
- 32 -
Table 3

- 33 -

Determinants of Investment Decisions
Probit regression results for the probablity of venture capital and private equity finance. The dependent varibale is 0 if the company is not financed, and 1 in the year of the deal
(firms are dropped after the first deal). The regressors are lagged one year. Column (a) reports the regression for coefficients from 1 to 8. Column (b) reports the whole model.
Column (c) reports the whole once early stage deals are dropped. Large firms reports only first two specifications. *** indicates significance level of 1% or less; ** between 1 and
5%; * between 5 and 10%.
SMEs
- 34 -
Large firms
Variable (a) (b) (c) Variable (a) (b)
Age ‐0.311 *** ‐0.382 *** ‐0.382 *** Age ‐0.129 ‐0.181
(0.112) (0.121) (0.121) (0.148) (0.152)
Size ‐2.315 * ‐3.128 *** ‐3.12 *** Size ‐12.581 ** ‐10.544 *
(1.258) (1.394) (1.394) (5.422) (5.614)
Size 2 0.061 0.082 ** 0.082 ** Size 2 0.358 ** 0.300 *
(0.039) (0.042) (0.042) (0.160) (0.165)
Intangibles 0.572 *** 0 .773 *** 0 .773 *** Intangibles 0.629 *** 0.810 **
(0.242) (0.259) (0.259) (0.327) (0.340)
Ebitda 0,429 *** 0 .582 *** 0.582 *** Ebitda 0.478 *** 0.642 ***
(0.101) (0.104) (0.104) (0.157) (0.160)
Leverage ‐0.544 * ‐0.424 * 0.11 Leverage ‐0.375 0.056
(0.329) (0.309) (0.605) (0.469) (0.809)
ROE ‐0.002 ‐0.001 ‐0.001 ROE ‐0.001 ‐0.001
(0.002) (0.001) (0.001) (0.001) (0.001)
High‐tech 0.113 0.143 0.143 High‐tech 0.365 ** 0.323 *
(0.131) (0.138) (0.138) (0.187) (0.193)
Capex 0.0001 0.000 Capex 0.001
(0.0003) (0.000) (0.000)
Growth 0.002 * 0.002 ** Growth ‐0.001
(0.001) (0.001) (0.001)
Short debt ‐0.470 ** ‐0.470 * Short debt ‐0.255
(0.171) (0..171) (0.380)
Number of Obs. 492 453 448 Number of Obs. 268 251
Pseudo R 2 0.1519 0.1625 0.1883 Pseudo R 2 0.1501 0.1296
Observed probability 0.000 0.000 0.000 Observed probability 0.000 0.000
Table 4

Variable Age Size
Age 1.000
Size 0.195 * 1.000
Total
assets
Total assets 0.226 * 0.877 * 1.000
Intangibles ‐0.224 * ‐0.075 * 0.024 * 1.000
Correlation coefficients
Ebitda 0.204 * 0.758 * 0.739 * ‐0.047 * 1.000
Intangibles Ebitda High‐tech Leverage
High‐tech ‐0.157 * ‐0.073 ‐0.063 0.206 * ‐0.061 1.000
Leverage 0.058 ‐0.072 ‐0.053 0.021 ‐0.230 * 0.024 1.000
Debt short 0.215 * 0.825 * 0.897 * ‐0.003 0.610 * ‐0.021 0.319 * 1.000
Capex ‐0.015 0.022 0.044 ‐0.083 * 0.012 0.028 ‐0.018 0.030 1.000
- 35 -
Debt
short
Table 5
Capex Growth
Growth ‐0.149 * ‐0.087 * ‐0.145 * 0.009 ‐0.102 * ‐0.001 0.117 * ‐0.070 0.080 * 1.000

Effects on backed firms
For each variable listed, we estimated the equation: yi,t = α +β1Deal0 + β2Deal13 +ut +dt +εi,t. where Deal0is a dummy equal to 1 in every first year of the deal; Deal13 takes value 1 in the 3 subsequent years. ut is
the firm‐specific effect, dt is the calendar‐year effect,εi,t is a random error with zero mean. The specification is estimatd with a fixed effect method by using each company as control for itself after the deal, which
enables to control for firm‐specific characteristics that are time‐invariant. *** indicates significance level of 1% or less; ** between 1 and 5%; * between 5 and 10%.
SME LARGE FIRMS
Variable Number of obs Year 0 Years 1‐3 F‐test Number of obs Year 0 Years 1‐3 F‐test
ROE 2817 ‐2.252 ‐4.820 ** 3.07 ** 1600 ‐5.034 ‐4.181 1.66
(2.272) (1.993) (3.529) (3.016)
ROA 2791 ‐0.009 ** ‐0.011 *** 7.01 *** 1584 ‐0.018 *** ‐0.017 *** 12.01 ***
(0.003) (0.003) (0.005) (0.004)
Total assets (log) 2851 0.198 *** 0.292 *** 38.82 *** 1626 0.142 *** 0.148 *** 13.12 ***
(0.041) (0.035) (0.040) (0.034)
Intangibles 2847 0.044 *** 0.044 *** 16.94 *** 1622 0.028 ** 0.031 *** 6.20 ***
(0.010) (0.009) (0.012) (0.010)
Ebitda 1574 0.122 ** 0.155 *** 6.22 ** 1508 0.065 0.011 0.59
(0.047) (0.055) (0.06) (0.051)
Sales 2884 0 .060 0.236 *** 18.37 *** 1619 0.041 0.128 *** 4.42 **
(0.044) (0.031) (0.051) (0043)
Employees 2809 0.024 0.221 *** 25.07 *** 1593 ‐0.074 * 0.074 *** 10.08 ***
(0.036) (0.031) (0.041) (0.034)
Debt short 2783 0.027 0.172 *** 9.03 *** 1623 0.015 0.091 ** 1.95
(0.046) (0.040) (0.054) (0.046)
Value added x employee (log) 2768 0.092 *** 0.032 3.88 ** 1562 0.102 *** 0.035 4.5 ***
(0.034) (0.029) (0.0409 (0.033)
Trade credit 2431 ‐0.007 ‐0.066 *** 4.74 *** 1590 0.023 ‐0.086 *** 7.15 ***
(0.025) (0.021) (0.029) (0.024)
Financial Expenses 2858 ‐0.002 ** ‐0.001 ** 3.40 ** 1624 ‐0.005 *** ‐0.004 *** 8.91 ***
(0.001) (0.001) (0.001) (0.001)
Debt 2860 0.113 ** 0.268 *** 21.36 *** 1626 0.063 0.123 *** 3.7 **
(0.047) (0.041) (0.055) (0.046)
Capex 2506 ‐0.029 ‐0.061 ** 1.39 1432 ‐0.002 ‐0.045 0.61
(0.043) (0.031) (0.049) (0.041)
Growth 2544 ‐0.031 0.005 0.41 1453 ‐0.033 0.017 0.32
(0.037) (0.032) (0.054) (0.045)
- 36 -
Table 6

References
Aghion P., Bolton P., (1992). An incomplete contracts approach to financial contracting.
Review of Economic Studies, 59: 473‐494.
Aghion P., Bond S., Klemm A., Marinescu I. (2004). Technology and financial structure: are
innovative firms different? Journal of the European Economic Association, 4: 277‐288.
Akerlof G., (1970). The market for “lemons”: Quality and the market mechanism. Quarterly
Journal of Economics, 84: 488‐500.
Aldrich H., Auster E.R., (1986). Even dwarfs started small: Liabilities of age and size and their
strategic implications. In L. L. Cummings and B. M. Staw (eds.), Research in Organization
Behaviour. 8: 165‐198. Greenwich, CT:JAI Press.
Arikawa Y., Imad’Eddine G. (2006). Who provides certification effect in Japan? Evidence
from IPOs on the Jasdaq, Mothers and Hercules. Working paper. Waseda University.
Audretsch D.B., Santarelli E., Vivarelli M. (1999). Start‐up size and industrial dynamics: some
evidence from Italian manufacturing. International Journal of Industrial Organization, 17:
965‐983.
Audretsch D.B., Thurik A.R., Verheul I., Wennekers S., (2002). Entrepreneurship:
Determinants and Policy in a European‐U.S. Comparison, Kluwer Academic Publishers,
Boston/Dordrecht, 2002.
Balestra P., Nerlove M., (1966). Pooling Cross Section and Times Series Data in the
Estimation of a Dynamic Model: The Demand for Natural Gas. Econometrica, 34(3): 585‐
612.
Barclay M. J., Smith C.W.Jr. (1995). The maturity structure of corporate debt. Journal of
finance, 50: 609‐631.
Baum J.A.C., Oliver C., (1991). Institutional linkages and organizational mortality.
Administrative Science Quarterly, 36: 187‐218.
Beatty R., Ritter J. (1986). Investment banking, reputation and the underpricing of initial
public offerings. Journal of Financial Economics, 15: 213‐232.
Belke A., Fehn R., Foster N. (2003). Does venture capital investment spur employment
growth? Working Paper. University of Wien.
Berger A.N., Udell G.F. (1995). Relationship lending and lines of credit in small firm finance.
The Journal of Business, 68(3): 351‐381.
Bertoni F., Colombo M.G., Croce A., (2008). The Effect of Venture Capital Financing on the
Sensitivity to Cash Flow of Firm's Investments. European Financial Management.
- 37 -

Binks M.R., Ennew C.T., (1996). Growing firms and the credit constraint. Small Business
Economics, 8: 17‐25.
Blackwell D., Wayne M., Spivey M., (1990). Shelf registration and the reduced due diligence
argument: Implications of the underwriter certification and the implicit insurance
hypothesis. Journal of Financial and Quantitative Analysis, 25: 245‐259.
Booth J.R., Smith R.L., (1986). Capital raising, underwriting and the certification hypothesis.
Journalof Financial Economics, 15: 261‐281.
Borisova G., (2007). Does privatization affect the cost of debt? Evidence from the European
Union. Working paper, University of Oklahoma.
Brav A., Gompers P., (1997). Myth or reality? The long‐run performance of IPOs:
evidence from venture capital‐backed companies. Journal of Finance, 52: 1791‐1821.
Callejón M., Segarra A., (1999).Business Dynamics and Efficiency in Industries and Regions:
The Case of Spain. Small Business Economics, 13: 253‐271.
Carpenter R., Petersen B., (2002a). Is the growth of small firms constrained by internal
finance? The Review of Economics and Statistics, 84(2): 298‐309.
Carpenter R., Petersen B., (2002b). Capital market imperfections, high‐tech investments and
new equity financing. The Economic Journal, 112(1): 54‐72.
Carter R., (1990). Underwriter reputation and repetitive public offerings, Working paper,
Iowa State University.
Caselli S., (2004). Corporate Banking Strategies: Products, Markets and Channels, in De
Laurentis G., Strategy and Organization of Corporate Banking. Springer Verlag, Berlin, New
York.
Chan Y., (1983). On the positive role of financial intermediation in allocation of venture
capital in a market with imperfect information. The Journal of Finance, 38(5): 1543‐1568.
Colombo M.G., Grilli L., Piva E., (2006). In search for complementary assets: The
determinants of alliance formation of high‐tech start‐ups. Research Policy, 35: 1166–1199.
Dai N., (2007). Does investor identity matter? An empirical examination of investments by
venture capital funds and hedge funds in PIPEs. Journal of Corporate Finance, 13:538–563.
Del Colle D.M., Finaldi Russo P., Generale A., (2006). The causes and consequences of
venture capital financing. An analysis based on a sample of Italian firms. Bank of Italy.
Discussion Paper n. 584.
European Commission, (2003). Definition of micro, small and medium‐size enterprises.
Official Journal, Recommendation 2003/361/EC of 6 May 2003.
- 38 -

Gompers P.A., (1995). Optimal staging, monitoring, and the staging of venture capital. The
Journal of Finance, 1(5): 1461‐1489.
Gorman M., Sahlman W.A., (1989). What do venture capitalists do? Journal of Business
Venturing, (4): 231‐248.
Hart M., Hanvey E., (1995). Job Generation and New and Small Firms: Some Evidence from
the Late 1980s. Small Business Economics, (7): 97‐109.
Hausman J., (1978). Specification tests in econometrics. Econometrica, 46(6): 1251‐1271.
Hellmann T., Puri M., (2000). The interaction between product market and financing
strategy: the role of venture capital. The Review of Financial Studies, 13(4): 959‐984.
Hellmann T., Puri M., (2002). Venture capital and the professionalization of start‐ups:
empirical evidence. Journal of Finance. 57(2).
Hirukawa M., Ueda M., (2003). Venture capital and productivity. University of Wisconsin,
mimeo.
Hsu D.H., (2006). Venture capitalists and cooperative start‐up commercialization strategy.
Management Science, 52: 204–219.
Hyytinen A., Pajarinen M., (2007). Is the cost of debt capital higher for younger firms?
Scottish Journal of Political Economy, 54(1): 55‐71.
Jain B.A., Kini O., (1995). Venture capitalist participation and the Post‐issue operating
performance of IPO firms, Managerial and Decision Economics, 6: 593‐606.
James C., (1990). Relationship specific assets and the pricing of underwriter services.
Working paper, University of Florida.
Jensen M.C., Meckling W.H., (1976). Theory of the firm: managerial behaviour. Agency costs
and ownership structure, Journal of Financial Economics, 3(4): 305‐260.
Johnson J., Miller R., (1988). Investment banking prestige and the underpricing of initial
public offerings. Financial Management, 17: 19‐29.
Kaplan S.N., Strömberg P. (2004). Characteristics, Contracts, and Actions: Evidence from
Venture Capitalist Analyses. Journal of Finance, 59(5): 2177‐2210.
Keasy K., Watson R., (1993), Small Firm Management, Blackwell, Oxford.
Klein B., Leffler K., (1981). The role of market forces in assuring contractual performance.
Journal of political Economy, 89: 615‐641.
Kortum S., Lerner J., (1998). Does venture capital spur innovation? NBER Working paper, n.
6846.
- 39 -

Kortum S., Lerner J., (2000). Assessing the contribution of venture capital to innovation.
RAND Journal of Economics, 31(4): 674‐692.
Lerner J., (1999). The government as venture capitalist. Journal of Business, 72: 285‐318.
Manigart S., Baeyens K., Van Hyfte W. (2002). The survival of venture capital backed
companies. Venture Capital, 4(2):103‐124.
Manigart S., Beuselinck C., Deloof M., (2005). Private equity investment and disclosure
policy. Vlerick Leuven Gent Management University, Working Paper Series 2005/1.
Manigart S., Hyfte W.V., (1999). Post‐investment evolution of Belgian venture capital
backed companies: an empirical study. Paper presented at the Babson Entrepreneurship
Conference.
Megginson W.L., Weiss K.A., (1991). Venture Capitalist Certification in Initial Public
Offerings. The Journal of Finance, 46(3): 879‐903.
Modigliani F., Miller M., (1958). The Cost of Capital, Corporate Finance, and the Theory of
Investment. American Economic Review, 48(3): 261‐297.
Munsters J., Tourani Rad A., (1994). Lange termijn prestaties van venture pitalondersteunde
ondernemingen na introductie op de Amsterdamse Effectenbeurs. Working paper, Erasmus
Universiteit Rotterdam.
Myers S., (1977). Determinants of corporate borrowing. Journal of Financial Economics, 5:
147‐157.
Myers S., Majluf N., (1984). Corporate financing and investments decisions when firms have
information that investors do not have. Journal of Financial Economics, 13: 187–221.
Panetta F., Schivardi F., Shum M., (2004). Do mergers improve information? Evidence from
the loan market. Bank of Italy, Discussion Paper n. 521
Petersen M.A., Rajan R.G., (1997). Trade credit: theories and evidence. The Review of
Financial Studies, 10(3): 661‐691.
Rajan R.G., Zingales L., (1995). What to we know about capital structure? Some evidence
from international data. Journal of Finance, 50: 1421‐1460.
Sapienza H.J., (1992). When do venture capilists add value. Journal of Business Venturing, 7:
9‐27.
Sau L., (2008). Venture Capital and New Pecking Order Financing for Innovative Firms.
International Economics and Finance Journal, 3(2): 2‐20.
- 40 -

Stein J.C., (2002). Information production and capital allocation: decentralized versus
hierarchical firms. The Journal of Finance, 2.
Stiglitz J., Weiss A., (1981). Credit Rationing in Markets with Imperfect Information.
American Economic Review, June, 391‐410.
Stuart T.E., Hoang H., Hybels R., (1999). Interorganizational endorsements and the
performance of entrepreneurial ventures’, Administrative Science Quarterly, 44: 315–349.
Sufi A., (2006). The real effects of debt dertification: evidence from the introduction of bank
loan ratings. Working paper, University of Chicago: Booth School of Business.
Ueda M., (2004). Banks versus Venture Capital: Project Evaluation, Screening and
Expropriation. Journal of Finance. 59(2).
Wasmer E., Weil P., (2000). The macroeconomics of labour and credit market imperfections.
IZA Discussion Papers, n. 179.
Williamson O., (1988). Corporate finance and corporate governance. Journal of Finance, 43:
567‐591.
Woolridge J.M., (2002). Econometric analysis of cross section and panel data. Cambridge:
MIT Press.
- 41 -