The Willingness to pay and Willingness to accept disparity:
the incentive to protect against a feeling of regret
A literature review
University of Amsterdam
BSc. In Economie en Bedrijfskunde Track: Economie
Name: Margot Akerboom Student number: 10591044 Date of submission: 14-7-‘17 Supervisor: A.R.S. Woerner
2 Statement of originality
This document is written by Margot Akerboom who declares to take full responsibility for the contents of this document. I declare that the text and the work presented in this document is original and that no sources other than those mentioned in the text and its references have been used in creating it. The Faculty of Economics and Business is responsible solely for the supervision of completion of the work, not for the contents.
3 Table of Content
Abstract ... 4
1 Introduction ... 5
2 Literature ... 7
2.1 Economic assumptions and expectations... 7
2.2 Theory and experiments about the disparity ... 7
2.2.1. Endowment effect and Coase theorem ... 8
2.2.2. Substitution and income effect ... 9
2.2.3. Prospect theory and loss aversion ...10
3 Recent studies ...12
3.1 Additional insights and current trends ...12
4 Challenges ...13
4.1 Misconceptions and ambiguity in the literature ...13
5 Suggestion & discussion ...14
5.1 Optional measurements WTP and WTA ...14
5.2 Limitations and further research ...14
6 Conclusion ...15
4 Abstract
This study examines the willingness to pay (WTP) and willingness to accept (WTA) disparity. Previous literature has shown that a considerable fraction of people have a higher WTA than WTP, which is in contrast to standard economic theory wherein the two variables are suggested to be equivalent. From economic theory about neoclassical preferences follows that by relating the income effect with substitution effect, the preferences of individuals depend on those impacts. The WTP/WTA gap is related to theories such as the endowment effect, benefit-cost analysis and prospect theory which contribute to the reason why the gap occurs and what the consequences are on several sectors and fields of study such as behaviour economics and psychology.
Keywords: Willingness To Accept (WTA), Willingness To Pay (WTP), disparity, endowment effect, substitutability and income effect
5 1 Introduction
For many years, researchers have studied the relation between willingness to pay and willingness to accept. The literature finds a disparity between the two variables, which is widely discussed in studies through the use of theory to run experiments for obtaining further knowledge about the gap between WTP and WTA.
One of the relevant variables is the willingness to pay (WTP), which describes the maximum amount an individual would pay to obtain a certain object. This is also described in an experimental setup as if the subject buys an object (Horowitz and McConnell, 2002). The other variable involved in the disparity is the willingness to accept (WTA), which is the minimum amount an individual want to receive from the experimenter in exchange for giving the object back. The experimental setup in the article of Horowitz and McConnell (2002) shows that the WTA is related to the event that the individual sells the object.
The gap between WTP and WTA is an interesting topic for research, especially in the fields of economics and psychology, because there are several contradictions, uncertainties and different perspectives resulting from several fields by reviewing literature. However, there is also motivation for research that comes from daily life activities, such as implications for the management of natural resources and environmental decision making.
Relevant with respect to the economic field of research, is the evidence from theory about the endowment effect and loss aversion that individuals value an object more when they have received it and have to give it back, compared to the case when an individual has to pay a certain amount to obtain the object (Kahneman et al., 1990). Besides this theory, there are many other theoretical explanations about the asymmetric valuation between gains and losses, such as the substitution and income effect, which are related to the gap between the WTP and WTA that will be explained in this paper.
The results from literature are that the disparity occurs because of difference in valuation techniques and experimental design used in the studies. Furthermore, economic theories give different explanations about the matter of the difference between WTP and WTA. In chapter two is the theory of the endowment effect explained combined with the substitution and income effect used in the neoclassical theory of preferences.
In recent studies they focus on the effect of factors related to experimental design to search for reasons of the occurring disparity. Also, in studies of the twenty-first century are theories combined to interpret the overall effect of the disparity. Primary contribution of recent studies is that the experimental setup is based on a combination of effects found in earlier studies which correct for misconceptions.
This paper proceeds in five parts. The next chapter of this paper provides the theoretical framework based on the literature, the available theories that play a role in the disparity and
6 experiments that are done in several studies so far related to the WTP and WTA. Chapter 3 provides recent research about the WTP/WTA gap and shows current trends in the findings of articles from recent years. Furthermore, in chapter 4 are the overall results reviewed next to the challenges and implications of the experiments about the gap through comparing the contribution of recent studies with earlier studies. Suggestions for possible options to decrease the WTP/WTA gap and a discussion about the available literature are given in chapter 5. In the final chapter, the results and implications of the literature are discussed and directions for further research are given.
7 2 Literature
2.1 Economic assumptions and expectations
Before reviewing the literature about the discrepancy between the willingness to pay (WTP) and willingness to accept (WTA), the theoretical assumptions and expectations that lead to the conclusion of equality in the variables have to be explained. The micro-economic concept about the concave shaped utility functions explains that there is a relation between the amount of good ‘x’ that can be exchanged to obtain a certain amount of good ‘y’ and vice versa (Snyder & Nicholson, 2012). Furthermore, the utility functions are assumed to be not influenced by a reference of initial endowment and are also supposed to be not related to loss aversion, according to Willig (1976). Resulting from his study in which he focussed on the bounds of price, he concludes that when the income effects are small, the difference between WTA and WTP is too small to be taken into account. Willig (1976) states in his article that the income effect is negligible, but in the literature of Randall and Stoll (in Hanemann, 1991) they have controversial results.
2.2 Theory and experiments about the disparity
It is a fact that from the perspective of economists the variables willingness to pay and willingness to accept are assumed to give equivalent estimates of value, based on ignorance of the impact of the income effect (Willig 1976). Knetsch and Sinden (1984) mention in their article that the two variables of the disparity are used to valuate assets and the use of resources. This valuation is usually measured by experiments, which are done in several studies to see the difference in valuation outcomes of the two variables and the underlying conditions of the experiments. The theories and studies explained and defined in this paragraph will give an overview of the aspects relevant for determining the WTP and WTA.
In the article of Knetsch and Sinden (1984) the experiments are done with real transactions and lottery tickets in which the design consists of several lay-outs with respect to the variation in monetary value assigned to the tickets and the groups of individuals used. In the experiments monetary values are assigned to the lottery tickets, which propose the value for which the individuals can purchase or sell the tickets and which correct for the wealth effect. The results Knetsch and Sinden (1984) obtain are comparable to the outcomes of studies with hypothetical questions in the experimental design. In the experiments of Knetsch and Sinden (1984) it is assumed that there is a random distribution of tastes between the ticket or monetary value. This presumption however is not sustained.
Following from the occurring difference between WTP and WTA, the valuation is probably affected by theories as the endowment effect and loss aversion.
8 2.2.1. Endowment effect and Coase theorem
In the theory defined by Morrison (1997) about the endowment effect is explained that there is no rejection of the relation with the WTP/WTA gap, since Morrison (1997) has done further research related to the findings of Kahneman et al. (1990).
Kahneman, Knetsch and Thaler (1990) suggest theory based on the endowment effect and Coase theorem, which theorem states that in the case there are no transaction costs, subject to the income effects, the allocation of resources is expected to be autonomous of property rights. Also, the matter of valuation of goods related to the measured initial endowment is discussed in the article of Kahneman et al. (1990). This statement has led to the experimental setup including market-based goods and tokens, which both are supposed to be unrelated to original endowment. In other experiments “induced- value” is stated to have effect on the valuation of goods by individuals. Based on the design of the experiment Kahneman et al. (1990) expect to detect a genuine effect of reference positions on preferences. Resulting from their experiments, ordering of preferences is not independent of endowments and the endowment effect reduces gains from trade. In the findings is also referred to loss aversion which will be further reviewed in the next part of this chapter.
Related to the endowment effect Shogren et al. (in Kolstad and Guzman, 1999) are doubtful about the effect of the effect on the disparity since the result from their experiments is that the gap becomes smaller after four repetitions of the experiment. Initially the WTP and WTA were divergent, but after repeating the setup which can be seen as learning by the individuals, the disparity resulted to be less. The findings of Shorgen et al. (in Kolstad and Guzman, 1999) are against the evidence of Kahneman et al. (1990), about the existence of an endowment effect and the effect on the WTP/WTA disparity.
In the research of Morisson (1997) is looked at the combined effect of the substitution and endowment effect by using indifference curves to see if there is a relation between those effects and the WTP/WTA disparity. Conclusions following from the results of the pilot experiment prescribed in the article are not striking. From the first condition in the experiment results that both the endowment effect and the degree of substitution have effect on the disparity, but in the second condition the endowment effect itself is studied which leads to a contradiction between the substitutability and valuation based on the endowment effect. Morrison (1997) mentions that there is not enough evidence to state that the endowment effect has no relevant impact on the disparity. Furthermore, Morrison’s findings are in favour of Knetsch’s research about the endowment effect. The results from the research of Hanemann (1991) are not corresponding with the findings in the study of Morrison (1997) about the substitution effect.
9 2.2.2. Substitution and income effect
First of all, in this paragraph the relation between the substitution effect and the WTP/WTA disparity will be explained, based on available theory from literature. In the research done by Hanemann (1991) the compensating and equivalent variations, which are welfare measures for prices related to WTP and WTA, are used to re-examine the work done by Randall and Stoll. The result of the research done by Randall and Stoll (in Hanemann, 1991) implies that WTP and WTA should not differ, unless there are unexpected income effects. This outcome is related to the economic theoretical findings of Willig (1976) about the bounds of price explained in the beginning of this chapter. Following from this, Hanemann (1991) is triggered by the fact how the empirical evidence concludes that there is a disparity and at the other side theoretical analysis claim that there are no differences between WTP and WTA. Hanemann (1991) derives an expression of elasticity about substitution in his study from the formulas obtained from Willig combined with the Hicks-Slutsky decomposition. From this expression follows that for the changes in quantity the extent of difference between C and E depend, besides the income effect, also on the substitution effect; which is the degree of changeability of public goods into private goods. He concludes also that if the demand function is highly inelastic or if there are just a few substitutes for the good, this has a divergent effect on C and E which is evidence for the WTP/WTA disparity. To reflect this findings to the conclusion in the article of Morrison (1997), it is hard to find out why the derivation by formulas of Hanemann would not be significant relevant. Between research of Hanemann (1991) and Morrison (1997) there is difference in theoretical deriving of formula’s and interpretation that leads to variation in experimental results. It is also not evident that the experimental design of the study of Morrison leads to significant results, such as the result she obtains that the substitutability argument in Hanemann is not significant. This will be evaluated in chapter 4 about the challenges and misconceptions of studies.
Secondly, the gap between WTP and WTA is studied in the article of Horowitz and McConnell (2003) to find out if there is a significant relation with the income effect which is stated in economic theory as a factor of impact on the disparity. In most studies about the disparity, the research is done the other way around compared to this article. In many other studies theories are used to find the influence on the disparity but in this case the gap is the start of research. Horowitz and McConnell (2003) argue in their article that Hanemann (1991) has provided the most convincing neoclassical explanation of the WTP/WTA disparity. He states that the disparity depends on the ratio of income elasticity of demand for the good to the Allen-Uzawa elasticity of substitution between two comparable goods. By using this ratio, when the substitution elasticity raises the WTA/WTP disparity is supposed to be larger. The only problem that influences the ratio is that the Allen-Uzawa elasticity of substitution is not
10 often measured, contrary to the amount of available statistics about the income effect obtained from survey evidence and market development. Sugden (in Horowitz and McConnell, 2003) made use of the WTA/WTP ratio and the WTP-income effect, which is further studied by Horowitz and McConnell based on the following equation derived by Sugden: 𝜕𝑊𝑇𝑃
𝜕𝑦 ≈ 1 − 𝑊𝑇𝑃
𝑊𝑇𝐴 . In other words, from the equation follows that the income effect is more or less equal to one minus the disparity between WTP and WTA.
The study in the article of Horowitz and McConnell (2003) is designed on looking at the observed WTP/WTA values, the income effect and observing η, which is the income elasticity of willingness to pay with the incentive to prove if the observations are consistent with the equation of Sugden. The variable they start the analysis from is the mean WTA/ mean WTP observed from several earlier studies. Furthermore they do three types of research: based on intuition, general valuation studies and studies on the WTA/WTP gap.
In the results section of the article it is explained that the WTA/WTP disparity is too high to be consistent with the neoclassical theory of preferences. The income elasticity of WTP and income effect based on the findings of the WTA/WTP ratio are very high following from experimental studies compared to the theoretical literature. The different results obtained in the study of Horowitz and McConnell (2003) in comparison with the neoclassical theory of the WTA/WTP disparity, can be due to weakness of used survey methods as for example contingent valuation. This means that there can be a problem in the valuation technique of the surveys, which in the case of non-market resources can lead to problems. Another conclusion about difference with the neoclassical theory is that the experimented individuals do capture true values in their responses of WTP and WTA (Horowitz and McConnell, 2003).
2.2.3. Prospect theory and loss aversion
Related to the disparity between WTP and WTA the prospect theory is analysed in the article of Kahneman and Tversky (1979), which is about individual decision making under risk. In the experimental studies that try to find evidence about the WTP/WTA gap is also decision making by the individuals involved. The possible outcomes that depend on choice making are not often highlighted or discussed in studies.
Besides the economic theories that try to find evidence for the difference in payment incentives and valuation of objects by individuals, the effect of the possible options and outcomes, such as rejecting or accepting, is also related to the amount of WTP and WTA. The prospect theory is divided in two parts of the choice making process: the first part is the analysis that consists of the optional prospects and the second step is the evaluation wherein the prospect with the highest valuation is chosen (Kahneman and Tversky, 1979). Individuals
11 normally perceive outcomes as gains and losses, which are compared to the value of a reference point, as in the theory of loss aversion. The location of the reference point is stated to be affected by the proposed options and by the expectations of the individuals in experimental setups (Kahneman and Tversky, 1979). In the prospect theory also the isolation of options is mentioned, which is not totally relevant in the research of the difference between the WTP and WTA of individuals. For sure the individuals are triggered by taking an outside option in their mind. By studying the theory mentioned in the article of Kahneman and Tversy (1979) it follows that ,next to the economic theories about income elasticity and preferences based on the endowment effect, also human behaviour and the way of thinking about how individuals select a valuable option or price influence the outcomes of WTP and WTA. All in all, prospect theory as well as loss aversion give relevant evidence for the disparity.
12 3 Recent studies
In previous research, performed since 1970, literature has been based on theories and economic effects that provide evidence for in-equivalence between WTP and WTA. In the meta-analysis done by Sayman and Öncüler (2005) they focus on the effects of factors related to study design on the WTA/WTP disparity. They study the following effects: explicit statement of the price, iterative bidding, within/between-subjects nature of design and out-of-pocket payments. The outcomes of the study verify that unless the statement of price, all three effects have impact on the disparity.
3.1 Additional insights and current trends
Earlier studies have in general focussed on the effect of just one factor on the disparity contrary to the study of Sayman and Öncüler (2005) wherein context variables are reviewed such as rational-economic effects as well as strategic misrepresentation, mental reference/framing and emotional effects. Those variables have indeed effect on the disparity and they provide evidence for incentive compatibility and learning, which are in prior research not studied in detail.
The research of Plott and Zeiler (2005) consists of a replication of the experiment related to the endowment effect done by Kahneman et al. (in Plott and Zeiler, 2005). Consistent with the findings in Kahneman’s experiment, Plott and Zeiler (2005) obtain from the experiment a mean WTP of $1.74 and a mean WTA of $4.72 which support that the median WTA is significantly greater than the median WTP. Furthermore, Plott and Zeiler (2005) are describing experimental procedures to study whether subject misconceptions give a satisfactory record of the observed disparity. The conclusion that follows from the reported data is that WTP/WTA disparities do not reflect a fundamental base of human preferences.
From the studies of Sayman & Öncüler (2005) and Plot & Zeiler (2005) follows that in recent studies about the disparity researchers focus on the overall significance of variables that influence the WTP/WTA gap. Searching for evidence of the disparity by studying one factor, such as loss aversion, is typical performed in earlier studies. Contributing insights which are obtained from recent studies are the combined factors relevance and the correction for misconceptions.
Current trends in recent studies are not directly visible from the articles of Sayman & Öncüler (2005) and Plot & Zeiler (2005), but overall the most recent studies are designed to combine and replicate theoretical literature and experiments to come up with more evidence about the WTP/WTA gap.
13 4 Challenges
In the research about the difference between WTP and WTA there are some challenges. Reflecting to the amount of years that the disparity has been studied, there are several complications shown in the literature. In chapter two the experiments and theories that contribute to the gap are analysed, from which becomes clear that there are opposite findings by reviewing the substitution effect by Hanemann and Morrison. Besides this, evidence about the endowment effect follows from the experimental setting of Kahnemann but is analysed as well by Shogren et al. mentioned in the article of Kolstad and Guzman, who pretend to have not enough evidence for impact on the gap.
4.1 Misconceptions and ambiguity in the literature
In literature there is not only research done from the perspective of the occurring effects and stated theories, to find out what the effects are on the disparity, but also vice versa. Horowitz and McConnell (2003) begin their research with an experimental setup to obtain WTP and WTA valuations, while other researchers are studying the economic theories to find a relation with the disparity.
An ambiguity in the literature is furthermore that the economic theories do not complement each other in direction of evidence for the gap, but lead to several directions in research. Sometimes it is possible to combine effects, such as relation of the income and substitution effect and the endowment effect related to theories as loss aversion and prospect theory.
Following from the reviewed articles also problems occurring in survey settings are related to the experimental designs. By the different experimental setups such as making use of hypothetical questions, open or closed questions, use of tokens or private goods, the outcomes are affected. This results in difficulties by comparing the outcomes of the studies and the effect on the WTP/WTA disparity.
14 5 Suggestion & discussion
When the different economic theories and experiments are reviewed to obtain evidence about the WTP/WTA gap, it is observed that there are contradictions in supporting evidence following from literature. This leads to the discussion which other experimental designs can be done and which theories are relevant to the gap.
5.1 Optional measurements WTP and WTA
Following from the theoretical review of evidence about the gap between the WTA and WTP, it is not easy to state which optional measurements of the valuations are not used in
research done so far but are for sure relevant to explain the difference.
Besides the mean and median values derived from experiments, are there possibly other measures that give a more consistent outcome used for the research about the difference between WTA and WTP. This has to be defined and studied more intensively.
5.2 Limitations and further research
Research related to the disparity which might be relevant for further research is the combination of human preferences and economic theory, to interpret the total effect that the disparity bring forward. In current literature those theories are reviewed separately, which induces the effect of contradictions by relating the experimental outcomes to find evidence for the gap.
15 6 Conclusion
From studies about the WTP/WTA disparity related to the endowment effect is found that individuals valuate a good different when they obtain it by giving money in return, compared to ‘selling’ the good by obtaining their willingness to accept. Several related experiments are done to provide evidence for the gap.
In economic theory the substitution and income effect are also relevant for the valuation of goods by individuals, which effects are studied in the articles of Hahnemann, Morrison and Horowitz & McConnell. They have different findings in their research. The neoclassical explanation is rejected by the findings of Horowitz and McConnell (2003) which was first studied by Hanemann (1991). These contradictions in evidence can be due to experimental misspecifications or difference in assumptions.
Besides those aspects the WTP/WTA disparity is affected by prospect theory and loss aversion. From the study of Kahneman and Tversky (1979) results that individuals have different valuations of WTA and WTP. By the theory of loss aversion individuals have a significant higher WTA compared to WTP, since they value gains positive while they interpret the WTA as a loss, which is corresponding to evidence from research. To obtain more consistent and relevant evidence about the disparity, there have to be more research done on combinations of influences on the difference between WTP and WTA.
16 7 References
Hanemann, W. M. (1991). Willingness to pay and willingness to accept: how much can they differ?. The American Economic Review, 81(3), 635-647.
Horowitz, J. K., & McConnell, K. E. (2002). A review of WTA/WTP studies. Journal of
environmental economics and Management, 44(3), 426-447.
Horowitz, J. K., & McConnell, K. E. (2003). Willingness to accept, willingness to pay and the income effect. Journal of Economic Behavior & Organization, 51(4), 537-545.
Kahneman, D., Knetsch, J. L., & Thaler, R. H. (1990). Experimental tests of the endowment effect and the Coase theorem. Journal of political Economy, 98(6), 1325-1348.
Kahneman, D., & Tversky, A. (1979). Prospect theory: An analysis of decision under risk. Econometrica: Journal of the econometric society, 263-291.
Kolstad, C. D., & Guzman, R. M. (1999). Information and the divergence between willingness to accept and willingness to pay. Journal of Environmental Economics and
Management, 38(1), 66-80.
Knetsch, J. L., & Sinden, J. A. (1984). Willingness to pay and compensation demanded: Experimental evidence of an unexpected disparity in measures of value. The
Quarterly Journal of Economics, 99(3), 507-521.
Morrison, G. C. (1997). Willingness to pay and willingness to accept: some evidence of an endowment effect. Applied Economics, 29(4), 411-417.
Plott, C. R., & Zeiler, K. (2005). The Willingness to Pay–Willingness to Accept Gap, the. The
American Economic Review, 95(3), 530-545.
Sayman, S., & Öncüler, A. (2005). Effects of study design characteristics on the WTA–WTP disparity: A meta analytical framework. Journal of economic psychology, 26(2), 289-312.
Snyder, C., & Nicholson, W. (2012). Microeconomic Theory Basic Principles and Extensions. (11th international ed., pp. 133-168). Canada: South-Western.
Willig, R. D. (1976). Consumer's surplus without apology. The American Economic
