Guide to COST-BENEFIT ANALYSIS of investment projects - Ramiri

VALUE TRANSFERValue transfer encompasses the adjustment of WTP in order to take into account the differences between the study and policy sites. The mostcommonly used adjustment is based on income, because it is thought that it is the most important factor resulting in changes in WTP. Thus ifthe WTP for an environmental good is X in a region when income per capita is Y, it may be X*f(Z/Y) in a different region where per capitaincome is Z.Other determinants might systematically differ between study and project sites, the main ones include:- the socio-economic and demographic characteristics of the population- the specific physical characteristics of the area- the extent of the change involved (values derived for small improvements may not apply to large changes)- the market conditions (availability of substitutes)- the changes of valuation over timeFor all types of adjustments the quality of the original study is of paramount importance for the validity of themethod.Some databases have been set up to facilitate benefit transfer. This is the case with the EVRI database 120 developedby Environment Canada and the US Environment Protection Agency. More than 700 studies are currently availablein the database, but only a minority are of European origin and this fact reduces the usability of the database in aEuropean context. GEVAD is an online European database, which was co-funded by the European Social Fund andGreek government resources. The aim of the project was to create a free online environmental valuation database, bygathering a critical mass of European valuation studies. About 1,400 studies were reviewed, focusing on the ones thatwere spatially more relevant to Europe. Emphasis was also placed on the most recent research results. So far, morethan 310 studies have been included in the GEVAD database. These studies are classified according to theenvironmental asset, good or service, which is valued (e.g. amenities, water and air quality, land contamination, etc.),the valuation method used, the main author and the country of the ‘study site’. 121Recent Estimate of the VOSL (Value of Statistical Life) in the UKWTP for mortality risk reductions is normally expressed in terms of the value of statistical life (VOSL). This entails dividing theWTP for a given risk reduction by that risk reduction in order to obtain the VOSL. The following table has a variety of estimatesof the VOSL, mostly for the UK. There is some unease about using the value of statistical life in contexts where remaining yearsmay be few for the affected individuals and this has led to the use of ‘life year’ valuations derived from VOSL. For example, theconcern is that estimates of VOSL from studies of workplace accidents (which tend to affect healthy, middle-aged adults), androad accidents (which tend to affect median age individuals) are ‘too high’ when transferred to environmental contexts wheremortality-related air pollution impacts tend to mostly affect the very elderly or those with serious respiratory problems.VOSL $MillionStudy Type of study Risk Context(year prices)Markandya et al. 2004 Contingent valuation Context-free reduction in mortality riskbetween ages of 70 and 801.2 - 2.80.7 – 0.80.9 – 1.9(2002) 3Chilton et al. 2004 Contingent valuation Mortality impacts from air pollution 0.3 – 1.5(2002) 3,4Chilton et al. 2002 Contingent valuation Roads (R), Rail (Ra) Ratios:Ra/R=1.003 6Beattie et al.1998 Contingent valuation Roads (R) and domestic fires (F) 5.7 3Carthy et al. 1999 Contingent valuation/standard gamble Roads 1.4 – 2.3(2002) 3,5Siebert and Wie 1994 Wage risk Occupational risk 13.5(2002) 3Elliott and Sandy 1996 Wage risk Occupational risk 1996: 1.2(2000) 3Arabsheibani and Marin 2000 Wage risk Occupational risk 1994: 10.7(2000) 3Source: Adapted from Pearce, D.W., G. Atkinson and S. Mourato (2006)Note: 1: median of the studies reviewed; 2: range varies with risk reduction level, lower VOSLs for larger risk reductions. 3: UK £ converted to US$ using PPP GNPper capita ratio between UK and US. Range reflects different risk reductions. 4: based on WTP to extend life by one month assuming 40 years of remaining life. 5: basedon trimmed means. 6: this study sought respondents’ relative valuations of a risk relative to a risk of death from a road accident. Numbers reported here are for the 2000sample rather than the 1998 sample. Between the two sample periods there was a major rail crash in London120121The database is accessible through the following link: The database is accessible through the following link: 227