International Choice Modelling Conference, International Choice Modelling Conference 2017

Font Size: 
Anna Malgorzata Bartczak, Susan Chilton, Mikolaj Czajkowski, Jurgen Meyerhoff

Last modified: 28 March 2017


In this paper we examine the impact of financial risk preferences and loss aversion on willingness to pay (WTP) for avoiding externalities from the renewable energy development in Poland. Additionally, we investigate the impact of individual’s increase and decrease of income on their stated willingness to pay. In our survey, we use both a discrete choice experiment (DCE) concerning externalities of renewable energy production in the proximity of respondent’s place of residence and a multiple price list lottery (MPL) choice task to elicit financial risk preferences and loss aversion. As far as we know we are our study is the first that investigates the effects of financial loss aversion risk preferences on the direction of change in the price vector in the DCE.

In the DCE we use both increases and decreases of the electricity bill to depict the uncertainty regarding the effect of new renewable energy production sites on the current price level. Both increases and decreases can occur on the same choice tasks. Attributes and alternative labels presented in the DCE describe the source and the location of energy production sites and transition lines. Although in the DCE uncertainty over the utility gain or loss from avoiding externalities from the production sites is not addressed directly, i.e. through an attribute that presents risk levels, we assume that the respondent may be uncertain regarding the outcome. Given the nature of the externalities associated with renewable energy production, we do not think that this is too strong an assumption to make a priori.

The choice sets were created using a Bayesian efficient design using the NGene software using the c-error optimization criterion. The final design comprised 24 choice sets that were blocked into four subsets with each six choice sets. The order of choice sets appearance was randomized as was the order of the first three labelled alternatives. We took steps to ensure that each choice sets and each alternative was presented on every position a comparable number of times. To capture financial risk preferences and loss aversion, we apply the MPL with paired lotteries introduced by Tanaka et al. (2010). This design is one of the most often used experimental risk elicitation methods (see: Binswanger 1980 or Holt and Laury, 2002). Tanaka et al. (2010) extended the standard MPL to allow for the calculation not only risk preferences parameters and but also a loss aversion parameter.

Our survey took place in January 2016. The sample was representative of the Polish population in terms of age, gender, agglomeration size and geographical location. In total, 800 interviews were collected. Interviews were conducted face-to-face using the computer-assisted personal interviewing (CAPI) system by a professional polling agency.

Data were analyzed by employing a mixed logit model (MXL). The monetary attribute enters the model as two variables allowing for a different parameter of the marginal utility of money when one chooses an alternative with positive values of monetary attribute i.e. surcharge or an alternative with negative values as respondents would receive a rebate on current electricity bill. Consumers’ preference heterogeneity is incorporated to the model my making the utility function parameters random according to some a priori selected parametric distribution – for this reason each attribute is associated with the estimate of the mean and standard deviation of its distribution in the population. Although the coefficients do not have direct interpretation their signs reflect whether more of an attribute is perceived as good or bad while their relative values indicate their relative importance.

We find that marginal utility of money seems to be lower with a rebate on the energy bill than with a surcharge to avoid. This result adds to the literature suggesting that people care about the mechanisms by which the funds for public projects are raised (see e.g. Ozdemir et al., 2016, Aravena et, al. 2014). Our findings also indicate that financial risk preferences impact peoples’ choices in a case of a surcharge, while loss aversion for money impacts them in the case of a rebate. These results are in line with Novemsky and Kahneman (2005) and Weber et al. (2007) who argue that loss aversion in money is not present during a buying process. Surprisingly, however, we find that financial loss aversion seems to play a role in a selling process. In the context of our study this observation can be connected with potential uncertainty concerning all effects of renewable energy development.

We find that the more risk seeking people are in a financial domain, they are less cost sensitive and are willing to pay more for proposed changes in renewable energy development, whereas the more loss averse people are with respect to money, they require more compensation before they accept externalities from renewable electricity production. Intuitively, this makes sense. People are assumed risk averse with respect to small decrements form current wealth, which would occur in the context of a surcharge and this would be expected to increase the more uncertain they were over the impact on their utility, resulting in a lower WTP the more risk averse an individual. Likewise, the more loss averse an individual the higher the rebate required for them to prefer the alternative over the future status quo.


Aravena, C., Martinsson, P., Scarpa, R. 2014. Does money talk? – The effect of a monetary attribute on the marginal values in a choice experiment. Energy Economics 44, 483-491.

Binswanger, H.P., (1980). Attitudes toward risk: experimental measurement in rural India. American Journal of Agriculture Economics 62 (3), 395–407.

Holt, C.A. and Laury, S.K. (2002). Risk Aversion and Incentive Effects, American Economic Review, 92(5), 1644-1655.

Novemsky, N., Kahneman, D. 2005. The boundaries of loss aversion. Journal of Marketing Research 42, 119-128

Ozdemir, S, Johnson, F.R., Whittington, D. 2016. Ideology, public goods and welfare valuation: An experiment on allocating government budgets. Journal of Choice Modelling

Tanaka, T., Camerer, C., Nguyen, Q. 2010. Risk and time preferences: Linking experimental and household survey data from Vietnam. American Economic Review, 100(1): 557-571.

Weber, B., Andreas A., Neuhaus, C. Trautner, P., Elger, Ch. E., Teichert, T. 2007. Neural evidence for Reference-dependence in real-market-transactions. NeuroImage 35, 441–447.

Conference registration is required in order to view papers.