Loss aversion for the value of voting rights: WTA/WTP ratios for a ballot

https://doi.org/10.1016/j.irle.2021.106041Get rights and content

Highlights

  • This article presents an empirical evidence of people’s loss aversion for the value of voting rights.

  • The willingness-to-accept compensation (WTA) is much higher than the willingness-to-pay price (WTP) for a ballot(s).

  • Observed WTA/WTP ratios for a ballot(s) are 5.00–27.36, which are much larger than the ratios for ordinary market goods.

  • WTA for a mere 10% dilution of voting rights is still higher than WTP for full voting rights.

Abstract

What is the subjective value of voting rights? This article presents an empirical demonstration of people’s loss aversion for the value of voting rights. By way of vignette experiments using scenarios of the 2020 U.S. presidential election (Studies 1 and 2, with U.S. citizens) and a fictitious direct premier election in Japan (Study 3, with Japanese citizens), the present research measured the willingness-to-accept compensation (WTA) and the willingness-to-pay price (WTP) for a ballot(s) in political elections. A great disparity between WTA and WTP was found regardless of the electoral setting, the proportion and width of ballot alteration, and the initial status. Specifically, the observed WTA/WTP ratios for a ballot(s) were within the range of 5.00–27.36, which is much larger than the ratios for ordinary market goods (about 1.5–2.5) and which is comparable to the ratios for other non-market goods. In addition, WTA for a mere 10% dilution of voting rights is still higher than WTP for full voting rights; the powerful effect of loss aversion might help explain universal barriers against suffrage extension.

Introduction

This article explores people’s subjective evaluation of the value of voting rights. It empirically examines people’s loss aversion for voting rights. Loss aversion is a core element of the prospect theory, a theory of behavioral economics developed by Kahneman and Tversky (1979). The prospect theory models that subjective satisfaction of people is determined by relative changes of their holding values rather than the absolute values they hold. The relative changes can be categorized into two domains, gains and losses, from an initial status (which is called a “reference point”). In a nutshell, people tend to be more sensitive to losses than gains even if the alteration width is objectively the same.

Implications of loss aversion for political decision making have been discussed from early times in various topics, especially in the area of international relations (Levy, 2003, McDermott, 2004). There has also been a growing body of research about voters’ loss aversion in their vote choice. For instance, economic performance reportedly has asymmetric influences on support for government and the incumbent—people react more strongly toward negative economic shocks than toward positive economic growth (Bloom and Price, 1975, Garz and Martin, 2020, Kappe, 2018). In a similar vein, negative attitudes of voters toward candidates have been found to provoke a higher voter turnout than do positive attitudes (Kernell, 1977, Lau, 1985, Patty, 2006). Furthermore, voters seem to have a general tendency to support the currently existing policy; in other words, people have a status quo bias from loss aversion (Alesina and Passarelli, 2019, Cohen and Knetsch, 1992). At the same time, it has been found that voters in the domain of economic loss tend to vote for a risky change, whereas those in the domain of economic gains prefer the status quo, a less risky choice (Carreras, 2019, for Brexit; Morisi, 2016, for Scottish independence; see also Quattrone and Tversky, 1988). However, all of these previous studies of voters’ loss aversion addressed vote choice and not the value of voting rights.

A method commonly used for measuring people’s loss aversion for a certain good is to observe the disparity between the compensation offered for obtaining it, the maximum willingness to pay (WTP) and the compensation required for losing it, the minimum willingness to accept (WTA). There is a phenomenon, known as the endowment effect, that WTA is higher than WTP. The prevailing theory for the endowment effect is loss aversion for the good (Thaler, 1980); accordingly, a loss aversion ratio is calculable as WTA/WTP (recent criticisms of this theory are noted in Section 5). For instance, to measure WTA/WTP for a mug, an experimenter gives a mug to seller-participants to ask them how much they would sell it for and to ask buyer-participants how much they would pay for it (Kahneman et al., 1990). A large body of research has revealed that WTA/WTP for ordinary market goods such as mugs is usually 1.5–2.5 (Booij et al., 2010, Novemsky and Kahneman, 2005). In other words, subjectively, people are inclined to value losses as 1.5–2.5 times greater than the equivalent gains in the case of ordinary goods.

The gap between WTA and WTP has been observed not only for ordinary goods. It has been observed for public goods and non-market goods as well. For example, goods such as hunting permits (Bishop and Heberlein, 1979), safety (Gerking et al., 1988, McDaniels, 1992), visibility (Rowe et al., 1980), and (to avoid drinking) distasteful drink (Coursey et al., 1987) have been studied empirically. Meta-analysis studies have shown that the ratios of WTA/WTP for public or non-market goods are higher than those for ordinary market goods (Horowitz and McConnell, 2002; Tunçel and Hammitt, 2014). This is probably because there is no guidance on prices. Evaluation of market goods can fall within a reasonable range around the commonly shared understanding of their prices; in contrast, people cannot rely on preconceived notions about their prices in the case of public goods or non-market goods. WTA/WTP ratios have varied largely by good and by study, but the average ratio for public or non-market goods has been around 10 (Horowitz and McConnell, 2002). Because no guidance exists for pricing, WTA/WTP ratios for voting rights can be predicted to be somewhat comparable to 10, as is the case with other public or non-market goods. It is predicted to be at least larger than 1.5–2.5, a range for ordinary market goods. Nonetheless, no research thus far has examined WTA/WTP ratios for the value of voting rights.

Actually, the value of voting rights is related to a long-standing controversial puzzle in political science, known as the paradox of voter participation. An act of voting has no material benefit for the voter, considering that an individual vote has no direct effect on the outcome of large-scale elections (as a practical matter, one vote cannot change an election outcome except for the extremely-low-probability event of an exact tie). Therefore, according to the rational choice theory (Downs, 1957, Riker and Ordeshook, 1968), voting is an irrational act for an economically rational person. From this perspective, voting rights could be worthless. However, the rational choice theory has trouble explaining empirical observations of high turnout rates for elections in reality (Green and Shapiro, 1994, Matsusaka and Palda, 1993). Consequently, studies have addressed voters’ subjective factors to explain the reality: there have been two lines of literature. One approach has been to expand the rational choice model by incorporating misconceptions of voters (e.g., Dittmann et al., 2014; Duffy and Tavits, 2008). This approach suggests that voters overestimate the effects of their own individual vote (i.e., the probability of being pivotal) on the electoral outcome. Another approach has highlighted the effect of voters’ subjective benefits for voting, such as expressive motives (e.g., Brennan and Hamlin, 1998; Kan and Yang, 2001) and moral duties (e.g., Blais et al., 2000; Güth and Weck-Hannemann, 1997). Such approach suggests that voters get satisfaction by expressing a “cheer” or “boo” for a candidate (even if the vote does not change the winner) or by fulfilling their civil duty of political participation to maintain democracy.

To date, however, very few studies have measured the value of voting rights. While previous research on the paradox of voter participation has addressed variables influencing voter turnout, the value of voting rights itself has not received attention. One exception is a study by Güth and Weck-Hannemann, 1997. They conducted an intriguing field experiment to measure the WTA value of a ballot for the German Bundestag election in 1994. In their experiment, participants (students) were offered a real incentive, payment in compensation, for destroying their actual voting card. But, because of budget constraints, they measured WTA of only a portion of participants who assigned low values on the ballot. In their experiment, 90 (63.38%) out of 142 participants refused to abandon their voting card even at the maximum possible bidding price, which was DEM 200 (≈ USD 130).1 In any case, previous research has not measured both WTA and WTP for the same voting rights.

Thus, the present research contributes to the field of law and economics by application of loss aversion in an as-yet unexplored subject: the value of voting rights. It is also the first attempt to measure both sides (WTA and WTP) of the subjective value of voting rights. Furthermore, it offers implications for the issue of enfranchisement. Loss aversion might help explain why suffrage extensions have been difficult to occur in many societies during their histories.

The present research measures WTA/WTP ratios for the value of voting rights through three vignette experiments. In all of the experiments, participants were instructed to consider a hypothetical situation in which ballot-dealing became completely legalized and they would be able to buy and sell their ballots freely in the election. Then, they were asked to name the selling or buying price of a ballot (a between-subjects design). In Study 1, which was conducted just before the 2020 U.S. presidential election, participants in the U.S. answered their minimum selling price or maximum buying price of a ballot for the upcoming election. In Study 2, which was administered one month after the presidential election, participants in the U.S. considered their minimum selling price or maximum buying price of a ballot for the hypothetical rerun election. The structural difference between Studies 1 and 2 was the initial status (reference point) upon measuring WTP. Study 1 asked the price for buying an additional ballot from the status of holding one ballot, the same status as reality; study 2 asked the price for buying a ballot from the artificial initial status of holding no ballot. In Studies 1 and 2, participants also considered the selling or buying price of a ballot when they held ten ballots—creating a situation of 10% dilution or condensation of their voting right. The purpose of covering a situation of dilution was to provide implications for the issue of suffrage extension, where voteless people are granted a ballot at the sacrifice of current voters to accept dilution of their voting rights. Study 2 included an additional set of conditions to ask the selling or buying price of aggregated ten ballots. In Study 3, participants in Japan answered their minimum selling price or maximum buying price for a ballot for a hypothetical direct election of their prime minister. The experimental structure was similar to Study 2—WTP was measured from the initial status of holding no ballot. Table 4, shown in Section 5, summarizes the features and structures of each study.

Section snippets

Study 1

Study 1 measured WTA and WTP for a ballot for the U.S. presidential election in 2020. It also measured WTA and WTP for one ballot when participants held ten ballots—representing WTA for 10% dilution and WTP for 10% condensation of one’s voting right.

Study 2

Study 2 measured WTA and WTP for voting rights one month after the U.S. presidential election in 2020, using a fictitious vignette of rerun election. Similar to Study 1, Study 2 measured WTA and WTP for a ballot. But, the initial status upon measuring WTP in Study 2 was holding no ballot, whereas it was holding one ballot in Study 1. WTA and WTP for one ballot when participants held ten ballots were also measured. In addition, Study 2 measured WTA and WTP for the aggregated ten ballots.

Study 3

Study 3 measured WTA and WTP for a ballot for a fictitious direct premier election in Japan, presuming that direct voting by the public was introduced to elect the prime minister. Studies 1 and 2 found a large disparity between WTA and WTP for voting rights. But, are the large WTA/WTP ratios unique to the particular 2020 presidential election in the U.S.? Or are they found in other elections as well? Study 3 attempts to explore another electoral setting in a different country.

General discussion

What is the subjective value of voting rights? What is the people’s loss aversion ratio for voting rights? Although the paradox of voter participation has been a well-known puzzle in political science, research into the value of voting rights itself has been scarce. In addition, attempts to apply the concept of loss aversion have been seen widely in political science studies (including the issue of voters’ vote choice). Nonetheless, the loss aversion ratios for voting rights have not been

Declarations of interest

None.

Acknowledgements

I would like to thank Ayako Hatano, Ayako Hirata, Yoko Kawada, Daisuke Mori, Kanji Muramatsu, Shozo Ota, Takashi Shimizu, Gakuto Takamura, and participants at the Conference of Japan Association of Sociology of Law, the Conference of Japan Law and Economics Association, and the Fundamental Jurisprudence Workshop at the University of Tokyo for helpful comments. This work was financially supported by a grant from Japan Society for the Promotion of Science, Japan (KAKEN 18K1261).

References (51)

  • R.C. Bishop et al.

    Measuring values of extramarket goods: are indirect measures biased?

    Am. J. Agric. Econ.

    (1979)
  • A. Blais et al.

    The calculus of voting: an empirical test

    Eur. J. Polit. Res.

    (2000)
  • H.S. Bloom et al.

    Voter response to short-run economic conditions: the asymmetric effect of prosperity and recession

    Am. Polit. Sci. Rev.

    (1975)
  • A.S. Booij et al.

    A parametric analysis of prospect theory’s functionals for the general population

    Theory Decis.

    (2010)
  • G. Brennan et al.

    Expressive voting and electoral equilibrium

    Public Choice

    (1998)
  • Chapman, J., Dean, M., Ortoleva, P., Snowberg, E., Camerer, C., 2021. On the relation between willingness to accept and...
  • D. Cohen et al.

    Judicial choice and disparities between measures of economic values

    Osgoode Hall Law J.

    (1992)
  • D.L. Coursey et al.

    The disparity between willingness to accept and willingness to pay measures of value

    Q. J. Econ.

    (1987)
  • A. Downs

    An Economic Theory of Democracy

    (1957)
  • J. Duffy et al.

    Beliefs and voting decisions: a test of the pivotal voter model

    Am. J. Polit. Sci.

    (2008)
  • Durkee, A., 2020. Trump and the GOP have now lost more than 50 post-election lawsuits. Forbes....
  • J. Eichhorn

    Votes at 16: new insights from Scotland on enfranchisement

    Parliam. Aff.

    (2018)
  • M. Frondel et al.

    WTA-WTP disparity: the role of perceived realism of the valuation setting

    Land Econ.

    (2021)
  • D. Gal et al.

    The loss of loss aversion: will it loom larger than its gain?

    J. Consum. Psychol.

    (2018)
  • M. Garz et al.

    Media influence on vote choices: unemployment news and incumbents’ electoral prospects

    Am. J. Polit. Sci. Adv. Online Publ.

    (2020)
  • Cited by (0)

    View full text