Expressive behavior in economics and politics

European Journal of Political Economy 26 (2010) 403–418

Contents lists available at ScienceDirect

European Journal of Political Economy

j ourna l homepage: www.e lsev ie r.com/ locate /e jpe

Expressive behavior in economics and politics

Arye L. HillmanDepartment of Economics, Bar-Ilan University, Ramat Gan 52900, IsraelBESA Center for Strategic Studies, Bar-Ilan University, Ramat Gan 52900, IsraelCESifo Munich, GermanyCEPR, London, United Kingdom

a r t i c l e i n f o

E-mail address: [email protected].

0176-2680/$ – see front matter © 2010 Elsevier B.V.doi:10.1016/j.ejpoleco.2010.06.004

a b s t r a c t

Article history:Received 9 March 2010Received in revised form 11 June 2010Accepted 16 June 2010Available online 23 June 2010

Expressive behavior is the self-interested quest for utility through acts and declarations thatconfirm a person's identity. Expressive voting is an example of expressive behavior. I introduceexpressive behavior in the forms of expressive rhetoric and expressive generosity. Thequestions for society and for public policy arewhether expressive behavior affects others, and ifso whether beneficially or disadvantageously. In experiments, expressive behavior oftenbenefits others. There are adverse social consequences when, in real-life decisions, expressivebehavior results in unwanted public policies of expressive-policy traps. I consider the prospectsfor avoiding or exiting expressive-policy traps.

© 2010 Elsevier B.V. All rights reserved.

JEL classification:D6H53

Keywords:IdentityVotingRhetoricGenerosityPolicy trap

1. Introduction

Hypocrisy: The practice of professing standards, beliefs, etc., contrary to one's real character or actual behavior, esp. the pretenceof virtue and piety. Collins English Dictionary

Neo-classical economic models describe people rationally maximizing utility derived from material aspects of life —

consumption or leisure and income or wealth. Behavioral economics explains perceived “non-rational” behavior inconsistent withthe traditional rationality axioms. Expressive behavior also departs from the traditional neo-classical view of behavior but inpositing that people behave rationally in seeking expressive utility from acts or decisions that substantiate or confirm personalidentity.

Attributes of identity can be chosen to reflect a view that people have of themselves in terms of who they are and what theystand for, or support or oppose. People obtain expressive utility, for example, by confirming pleasing attributes of being generous,cooperative, trusting and trustworthy, or ethical and moral. In conflict situations, confirming a pleasing identity can be confirmedor signaled by being conciliatory and open to compromise and being opposed to violence.

All rights reserved.

http://dx.doi.org/10.1016/j.ejpoleco.2010.06.004

mailto:[email protected]

http://dx.doi.org/10.1016/j.ejpoleco.2010.06.004

http://www.sciencedirect.com/science/journal/01762680

404 A.L. Hillman / European Journal of Political Economy 26 (2010) 403–418

An identity can be chosen as self-pleasing or pleasing to others, or if possible simultaneously both. The expressive utility frombeing pleasing to oneself is founded in psychological aspects of self-approval. Pleasing others has a basis in wishing to be liked orpopular, which can be achieved by expressively signaling conformity with group-defining norms.1

Insofar as all people have an identity that influences their behavior, all people behave expressively. The questions for societyand for public policy are whether expressive behavior affects others, and if so whether beneficially or disadvantageously.

Benefits for others are revealed in experiments when people choose behavior that confirms a pleasing identity to themselves orto others. Outside of experiments, in the domain of real-life decisions, the same pleasing behavior has socially adverseconsequences when public policies are supported only for reasons of personal expressive utility. As a result of the quest forexpressive utility, societies can find themselves in disadvantageous expressive-policy traps with majority-supported policies thatthe majority does not actually want.

Section 2 summarizes elements of expressive behavior. Appendix A provides a more extended formal presentation. InSection 3, expressive behavior is illustrated through expressive voting. Section 4 introduces behavior that I call expressive rhetoric.Section 5 introduces behavior that I call expressive generosity. Section 6 reviews evidence on expressive behavior. Section 7addresses remaining issues: I consider the relation between expressive and other types of behavior, the effects of institutions onexpressive behavior, and solutions for social costs. Section 8 provides brief concluding remarks.

2. Expressive behavior

Expressive behavior is self-interested and rational. If the behavior affects others, the consequences for others are notinternalized in individuals' decisions. Expressive behavior is therefore distinct from, and devoid of, altruism and malice.2

The attributes of identity confirmed by expressive behavior can be either predetermined or discretionary. Predeterminedidentity, for example, affects behavior through gender and ethnicity.3 I shall be concerned with attributes of identity that are achoice.4

For exposition, I shall view total utility as the sum of material and expressive utilities, or

1 Pauthereby

2 I co3 Ros

centurywomanidentity

4 Brewhich t

5 Kirc6 See

when r7 Exp

For con(1984),literatu

Totalutility = Materialutility + Expressiveutility:

There need be no conflict betweenmaterial and expressive utility. Because ofmoral hazard and the undermining of thework ethic,a high-income person may object to a policy of high taxation that finances generous welfare payments. The high-income person'spolicy position is consistent with personal material self-interest. At the same time, the high-income person can obtain expressiveutility from substantiation of identity as a person who believes that incentives of public policy should encourage self-reliance.

In other cases, expressive utility is inconsistent with material self-interest. A person may, for example, forego income bychoosing to live in a location for expressive reasons. From the observed locational preference, we infer that, if material losses fromthe locational decision are high, expressive utility from the locational choice is higher.

Material losses incurred because of expressive behavior can be insignificantly low or zero. In such cases, people disregardmaterial utility and make decisions based on expressive utility alone. Such low-cost expressive behavior has been extensivelystudied in the form of expressive voting.5

3. Expressive voting

The expressive-voting hypothesis contrasts with an instrumental view of rational voting (Downs, 1957) in which voters aredescribed as believing that their vote is with some significant likelihood decisive in determining the outcome of an election.Whether the institutions are those of representative or direct democracy, and whether voting is optional or compulsory, a singlevote in general does not affect a voting outcome. The “paradox of voting” is that instrumental voters, if rational, are predicted notto vote because the cost of voting in terms of time exceeds the expected benefit of voting based on the insignificant likelihood ofone vote being decisive.6 The addition of expressive utility from voting can however change the rational-voting cost–benefitcalculation.7

l Rubin (2002) describes the small-group hunter–gatherer origins of much human behavior that ostensibly include the wish to appear likeable, andbe liked by others. Chen and Xin Li (2009) and Klor and Shayo (2010) describe experiments that reveal benefits from shared group identity.nsider coexistence of expressive behavior with altruism or malice in Section 7.e (1988) proposed that gender-based identity explained the emergence of specialized “breadwinner–homemaker” households in England in the 19th(see also De Vries, 2008, on preferences, technology, and incomes). Akerlof and Kranton (2000) proposed that gender-based identity explains why aearning market income may contribute disproportionately to housekeeping (the woman does not wish to emasculate the man) and observed that ethniccan affect behavior through attitudes to educational achievement (see also Austen-Smith and Fryer, 2006, on “acting white”).

nnan and Hamlin (2000) described choice of identity as people choosing “dispositions”; Lewisch (2004) described people as choosing “windows” througho view the world. On people choosing identity, see also Glaeser and Ward (2006).hgässner (1992) described “low-cost” decisions.Tullock (1967). Pivotal-voter instrumental-voting models (see the overview of such models by Feddersen, 2004) are of limited real-world relevanceational instrumental voters are aware that their vote is not decisive.ressive voting was suggested in Buchanan (1954) and has origins in Tullock (1971). Fiorina (1976) distinguished expressive from instrumental voting.tinuations and elaborations, including descriptions of expressive behavior that are more general than expressive voting, see Brennan and BuchananBrennan and Lomasky (1984, 1993), Glazer (1987), Brennan and Hamlin (1998, 2000), and Schuessler (2000). Hamlin and Jennings (2010) survey there. Expressive voting is central to the public-choice perspective on political economy (Mueller, 2003; Hillman, 2009).

8 Jones and Hudson (2000).9 Kliemt (1986) described decisions made behind a “veil of insignificance”. The comparative reference is to the veil of ignorance of Rawls (1971). The

metaphor of the veil of ignorance describes people confronting uncertainty about their future selves. The veil of insignificance involves no uncertainty. Peopleknow who they are but they know that they are insignificant in determining actual outcomes or policy decisions.10 The example is from Hillman (2009, chapter 7).11 On the “volunteer” public good, see Hirshleifer (1983) and Diekmann (1985); for a textbook exposition, see Hillman (2009, chapter 3).12 For the values in Table 1, in the mixed-strategy equilibrium, a voter vetoes the income transfer with probability 0.5. The likelihood of the outcome (−1,−1is therefore 0.25.13 On social traps and the volunteer public good, see Diekmann (1986).14 The outcome is the “inefficient unanimity” of Brennan and Lomasky (1984). A voter cannot influence the material loss of −2 from paying to finance thetransfer but obtains expressive utility of 1 by voting non-decisively in favor of the transfer, resulting in utility of −1.

Table 1Expressive voting.

Person 2 votes against the income transfer Person 2 votes in favor of the income transfer

Person 1 votes against the income transfer 0, 0 0, 1Person 1 votes in favor of the income transfer 1, 0 −1, −1

405A.L. Hillman / European Journal of Political Economy 26 (2010) 403–418

Voting involves two related decisions: whether to vote, and for whom to vote.8 The decision whether to vote can involveconceptions of civic duty that underlie expressive confirmation of identity as a socially responsible person. Having decided to vote,individuals also make an expressive decision regarding for whom to vote. They can obtain expressive utility by voting for policiesthat they truly wish to see implemented. However, rationally recognizing that their individual vote will not affect the votingoutcome, voters can choose to vote expressively for candidates and policies that they would oppose if they believed that their votewere decisive.9

In Table 1 two expressive taxpayers choose how to vote on financing an income transfer to a third person.10 The two voters willbe taxed, and the income transfer will take place, only if there is a consensus vote in favor. Abstention by one voter is therefore aveto and so is sufficient to block the income transfer.

With expressive utility of voter j denoted by UjEXP and material utility by Uj

MAT, and total utility determined as the sum ofexpressive and material utility so that Uj=Uj

EXP+UjMAT, for each person in Table 1:

UEXPj fromvoting for thetransfer = 1

UMATj frompaying for thetransfer = −2

Uj = UEXPj + UMAT

j fromvotingagainst thetransfer = 0:

Voting against the income transfer provides no expressive utility and results in no material loss. A voter is best off withexpressive utility 1 from voting for the income transfer that is not made because the other person has vetoed the transfer. Themaximum personal benefit is achieved in the efficient Nash equilibria at (1, 0) or (0, 1) in which the other voter has vetoed theincome transfer. The circumstances are those of the volunteer-public-good game. The public good that the “volunteer” provides isavoidance of the common loss that is incurred if the income transfer is made.11

If voting is sequential (and observed), the order of voting determines which of the Nash-equilibria (1, 0) or (0, 1) is attained.The person voting first has expressive utility from voting to be generous and the second voter maximizes utility (for both voters)by vetoing the income transfer.

In a mixed-strategy equilibrium, the outcome can be (−1, −1), in which case utility of the taxpayers is minimized.12 Thetaxpayers are then in a policy trap that has arisen because of expressive voting.13 They have both expressively voted in favor of apolicy that neither wants.

In majority voting in actual elections, there are many voters, each of whom is aware that a single vote is not decisive indetermining whether the income transfer will take place. Each voter maximizes expressive utility by voting in favor of the incometransfer, which therefore takes place, resulting in the expressive-policy trap in which all voters have utility −1.14

Externalities are present in voting because voters do not internalize the consequences of their voting decisions for others(Tullock, 1959; Hillman, 2009, chapter 6). With expressive voting, there are mutually disadvantageous externalities amongexpressive voters if voting results in an expressive-voting trap.

Expressive voting can result in expressive-voting traps in which there is majority support for public policies that each individualmember of the supportive majority would veto if he or she could.

To summarize, other than the time taken to vote, voting does not affect a voter's material utility because of the effectively zeroprobability of one vote being decisive in determining policy outcomes in usual elections. Voters can therefore rationally disregardmaterial utility and vote to maximize expressive utility, and the result can be an expressive-policy trap.

)


4. Expressive rhetoric

Expressive rhetoric, like expressive voting, is low-cost behavior that can provide expressive utility. The rhetoric can be anaccompaniment of expressive voting on income redistribution. The rhetoric then takes the form of declarations such as “we shouldhave a social conscience” or “people less well off than we are should be helped”. For diversity, I shall describe expressive rhetoricagainst the background of the policy issue of public safety when an adversary has supreme-value objectives.

4.1. Public safety in the face of supreme values

Supreme values allow no trade-offs in objectives.

Definition. Supreme values (Bernholz, 1993)

15 Ber16 OnExamplwas boterroriseducateDanielNoneth17 ProappearaFrey (2Frey's p

A supreme-value ideology or belief system ranks objectives lexicographically.

The supreme-value objective may be annihilation of peoples who are declared inferior (as in national-socialism), change in theconsciousness of people (as in communism), or submission of all people to a designated belief system (as in the case of radicalIslam).15 Although compromise with a supreme-value adversary is impossible, expressive utility can be obtained from conciliatoryexpressive rhetoric. The rhetoric can use the narrative of “strong” and “weak” conjoined with “rich” and “poor”:

“When one side in a conflict is strong and rich and the other side is poor and weak, the strong side should be conciliatoryand generous, and make offers of compromise”.

The source of expressive utility is in the continuation, implicit or explicit:

“I would be conciliatory and generous toward the weak if I were strong”.

The conjoining of “weak” and “poor” in the expressive rhetoric parallels expressively voting for policies that would requirehigh-income people (the strong) to be generous in sharing income or wealth with low-income people (the weak). Evidence doesnot substantiate the association between being a threat to public safety and being poor. However, to let the rhetoric proceed, weallow the evidence to be disregarded.16 Expressive utility from conciliatory rhetoric then nonetheless requires that supreme valuesbe called into question:

“Supreme values should not be taken at face value: all people are reasonable (as I am), and compromise with reasonablepeople is always possible.”

The next step in the rhetoric after the calling into question of supreme values is the rhetoric of “engagement”:

“We should engage with adversaries to achieve compromise and mutual understanding”.17

Table 2 shows the volunteer public-good game implicit in the above expressive rhetoric. The public good “public safety” isprovided if at least one of the two persons acknowledges that an adversary exists and poses a threat to public safety. The followingvalues are used in Table 2.

UMATj =

(10 if policiesensurepublicsafety

−20 if publicsafety is threatened

UEXPj fromfavoringpublicsafety =

(2whenbothadmit the threat topublicsafety

−5whentheotherdeniesthere isa threat topublicsafety

UEXPj fromdenyingpublicsafety is threatened =

(5whenbothdenythethreat topublicsafety

10whentheother favors policies toprovide public safety

nholz (1993) describes different supreme-value systems. On radical Islam, see also Bernholz (2004), Hillman (2007), and Frisch and Inbar (2008).the evidence that terror is not due to low income or low wealth, see Krueger and Maleckova (2003), Abadie (2006), Piazza (2006), and Krueger (2007).es indicate on the contrary a positive relation between wealth and being a terror threat. Particular cases confirm the general evidence. Osama bin Ladenrn into a high-wealth Saudi family. September 11 terrorists and other terrorists in the US, the UK, and elsewhere have been university-educated. Thet from Nigeria who sought to down a plane on December 25, 2009 on the approach to Detroit airport was the son of high-wealth parents: he had beend at a college of London University, and had lived in one of the more exclusive areas of London. The convicted killer of the Wall Street Journal reporterPearl, whose beheading was videotaped, had attended an English private high school and had been a student at the London School of Economics.eless, the conjoining of “weak” and “poor” allows the rhetoric: “Ending poverty is the means of ending threats to public safety”.minent examples of this rhetoric include U.K. Prime Minister Arthur Neville Chamberlain's 1938 “engagement” with German chancellor Hitler. Innces of the rhetoric in recent literature, Cowen (2004) reflects on the Coase theorem as a guide for achieving compromise with supreme-value terror and004) weighs incentives for compromise (a carrot) against defense (a stick). See Plaut (2004) for a response to Cowen and Inbar (2006) for a review ofroposals and a response.

Table 2Expressive rhetoric and public safety.

Person 2 expressively declarespublic safety is not threatened

Person 2 admits that publicsafety is threatened

Person 1 expressively declares public safety is not threatened −15, −15 20, 5Person 1 admits that public safety is threatened 5, 20 12, 12

18 The rhetoric in alternative circumstances in which both persons admit the existence of the threat to public safety is: “if you were peace-loving andconciliatory as I am, we would be liked by everybody.”19 Franck et al. (2005) describe the moral dilemma when the population from which terrorists come is known but not the specific identity of terrorists.20 See Mueller (2004).21 On the multi-person volunteer game when one person is decisive in providing the public good, see Xu (2001).


The benefits of a policy that provides public safety, and the losses if public safety is not provided, are in terms of material utility.Both persons have utility of 10 if at least one of them admits the threat to public safety (in which case policies to ensure publicsafety are implemented) and they both lose 20 when neither admits that public safety is threatened (in which case there is nopublic safety).

There are externalities because the expressive utility from rhetoric depends on the rhetoric of the other person. In thesymmetric cases, common expressive utility is 2 when both persons admit and 5 when both deny the existence of the threat topublic safety. In asymmetric cases, expressive utility of 5 is lost by admitting the existence of the threat to public safety when theother person denies that public safety is threatened, and expressive utility of 10 is gained by denying the existence of the threat topublic safety when the other person acknowledges that the threat is present.

When decisions are sequential, public safety is provided in the efficient Nash equilibria at (20, 5) and (5, 20), with the firstperson to decide having high utility and the other person low utility. The expressive externality can be through the rhetoricalquestion of the high-utility person:

“Why are you not peace-loving and conciliatory as I am?”18

With no means of pre-commitment, we have a game of chicken. In the mixed-strategy equilibrium, both persons can havedeclared themselves to be “peace-loving” and can find themselves in the expressive-policy trap at (−15,−15) where both incur amaterial loss (danger to their lives) but benefit from their expressive behavior.

Themoral dilemma in providing public safety also provides a means of seeking expressive utility. An apprehended terrorist hasinformation that, if divulged in time, will save innocent lives, but compelling means are required to elicit the information.19 Inthese circumstances, the case in support of public safety is:

“Public policy should ensure public safety, even if the means are inconsistent with usual standards of legal and civilrights.”20

The expressive counter case is:

“Saving innocent lives can never justify forceful means of eliciting information.”

The hypocrisy is in the continuation, which may remain unspoken:

“Unless it is my life that is saved.”

In the 2-person game in Table 2, one person can “volunteer” to acknowledge the need for policies that provide public safety. Asin the case of voting, we proceed to a multi-person game in which a sufficiently large number of “volunteers” (for example amajority) is required to admit the need for public safety.21With each individual recognizing that he or she is personally ineffectualin ensuring public safety, material utility is disregarded and all people maximize their individual expressive utility. The outcome isthe expressive-policy trap (−15, −15).

The conclusions regarding expressive rhetoric parallel those of expressive voting:

People can maximize utility by expressively proclaiming the merits of policies that they know would be personally harmful ifactually implemented.

People may expressively proclaim the merits of policies that they would oppose if they were decisive.

Expressive rhetoric that influences policy can result in expressive-policy traps.


4.2. Soft power

Soft power (Nye, 2004) is a concept proposing that persuasion is effective in achieving conciliation with adversaries throughappeal to the rationality of avoiding conflict. There is no qualification for supreme values. Advocates of soft power in the face of asupreme-value adversary obtain expressive utility from their expressive rhetoric. Soft power is often expressed as a “carrot” setbefore the adversary.22

4.3. “Useful idiots”

Lenin described theWestern supporters of communism as “useful idiots”. The behavior of “useful idiots” is expressive. They arenot decisive and obtain utility from the rhetoric that displays their peaceful non-violent identity.

4.4. Advice for others

“Useful idiots” obtain expressive utility by denying the need for their own defense. Denial of the existence of a threat to safetyand thereby of the need for self-defense can be advice for others. For example, following the terror attacks of September 11, 2001,large majorities in western European countries declared that they were of the opinion that the U.S. should not act to preemptfurther terror attacks.23 The responses may have been an indication of the presence of malice (see Section 7). However, thedeclarations are also consistent with expressive rhetoric that confirms an identity of being opposed to violence.24 Expressiveconciliatory rhetoric in face of threats that others face imposes costs on those others who are placed in the unnatural circumstancesof being told not to defend themselves. The externalities are in the diminished feeling of self-worth and indicated low value of lifefor those to whom self-defense is denied. Advocates of soft power and “useful idiots” deny, for their expressive utility, thatsomeone wishes to do them harm but would change their rhetoric and would acknowledge the need for self-defense if decisiveand actually personally threatened. There are no such self-limiting bounds on people obtaining expressive utility for themselvesfrom expressive rhetoric that advises against self-defense for others.

4.5. Self-defamation to please others

The intention of expressive rhetoric may be to please others. The personal benefits from the rhetoric can take the form ofexpressive utility (from being liked) or material reward (monetary payments or privileged invitations). Pleasing others mayrequire self-defamation. Thus, for example, someU.S. commentators joined Europeans and others in rhetoric that declared that theUnited States had brought the September 11 terror attacks upon itself.25 The rhetoric of self-defamation takes the form:

22 As i23 OppGerman24 On25 See26 Expdefamin“improv27 In ahigh incsocial s

“We deserve what others do to us because we are condescending in supposing the primacy of our culture”.

The rhetoric can be conjoined with the familiar:

“We are strong and the populations from which the terrorists come are weak and we should be more respectful”.

Self-defamation is predicted in particular from persons with foreign peer groups.26

5. Expressive generosity

Expressive behavior can take a form that I call expressive generosity. For some people, voting for or expressively declaring thevirtue of generosity while not actually personally giving may not provide expressive utility. Rather, expressive utility fromconfirming generous identity may require the act of actual giving.27

I have defined expressive utility from voting and rhetoric without interdependent utility. Likewise:

Expressive generosity is motivated solely by own expressive utility and not by altruism in caring about consequences of givingfor others.

f the supreme-value adversaries were rabbits.osition to a U.S. defensive response against countries harboring terrorists was voiced by 71% of French respondents, 79% of Italians, 82% of British, 83% ofs, and 88% of Spaniards. Source: Pew Global Attitudes Project, reported in Kirchick (2009).the relation between survey responses and true preferences, see for example List and Gallett (2001).Miller (2008) who reviews the themes of post 9/11 U.S. novelists.ressive benefits from group approval and material benefits compensate for loss of expressive utility from self-defamation. Frey (2003) observed self-g utility-maximizing behavior in academic publishing when authors acquiesce to idiosyncratic requirements of reviewers whose proposedements” are a requisite for a favorable publication decision.counter to the usual perspective, Corneo and Grüner (2000) suggested that generosity can decrease expressive utility. If utility from social status requiresome or wealth, people who give money away can find themselves lacking the wealth or income that allows them to express themselves as having hightatus.


In the absence of altruism andwith no expressive utility, the decision is to give nothing. As with voting and rhetoric, expressiveutility can change behavior, in this case to result in actual giving.

We again consider low-cost decisions. Expressive generosity can be low-cost because it is the act of giving that providesexpressive utility rather than necessarily the amount given.

Beneficial externalities from expressive generosity are indicated in behavior in experiments (which I consider in the nextsection). There are negative externalities from expressive generosity in the following cases.

5.1. Expressive generosity in a natural experiment

Two high-income visitors to a low-income country encounter a school-age child offering trinkets for sale at a time of day whenthe child confirms that she should (and could) be in school. One of the visitors (an economist) points out that the purchase oftrinkets will discourage the child's parents from sending her to school. The other visitor responds by withdrawing from thepurchase, but later is seen furtively buying trinkets from the child. The purchase of the trinkets is inconsequential for the materialutility of the buyer: the amount of money spent on the trinkets is insignificant, and the buyer obtains no material utility from thetrinkets. The purpose of the purchase of the trinkets is only to provide expressive utility for the buyer by confirming a generousidentity. There is a negative externality from the expressive behavior through the adverse incentives for the child's education. Thebuyer's decision was however not based on outcomes for the child. Only the buyer's own expressive utility mattered.

5.2. Overfeeding children

In another example, parents obtain expressive utility through expressive generosity by overfeeding children. Utility is obtainedby the parent, at the expense of the overfed child.

5.3. Ineffective aid as expressive generosity

Donors in the forms of international agencies, governments, and non-government organizations have provided developmentaid in face of conclusive evidence that their aid is ineffective in promoting economic growth and in benefiting the poor in the low-income countries (Easterly, 2001, 2006; Hillman, 2002; Doucouliagos and Paldam, 2008). The aid is ineffective because of thepredatory nature of government in the aid-receiving countries and absence of the accountability of democracy (Hillman, 2004;Borooah and Paldam, 2007). Beyond being ineffective, the aid is harmful because of the rent-seeking incentives associated withcontestability of the aid (Svensson, 2000).28 Donors would prefer that the aid that they give be effective. However, aid beingineffective and harmful does not deter the giving of the aid. The giving of the ineffective aid allows the donors expressively tosignal:

28 Theindustriindustrialso thethis parnot to bneed. A29 Wereciproc30 Becthroughdisparitrecipien

We care about poor people in poor countries.

Expressive utility thus explains why ineffective and harmful aid continues.29 One donor among many might wish to ceasegiving the ineffective harmful aid. Ceasing to give is expressively costly when a donor who continues giving ineffective aid asks:

Why are you not generous in giving aid as I am?

An expressive-policy trap sustains the ineffective or harmful aid. Each donor continues expressively to give aid, although eachdonor if given the opportunity would veto all aid.30

6. Is behavior expressive?

I have described expressive-policy traps that are the consequence of expressive voting, expressive rhetoric, and expressivegenerosity. The substantiation of the policy traps requires evidence of prevalence of expressive behavior. I turn now to theevidence. I begin again with expressive voting.

following is from an interview (Spiegel On-line International, accessed April 20, 2010) with James Shikwati, an economist in Kenya. Spiegel: Thealized nations of the West want to eliminate hunger and poverty. Shikwati: Such intentions have been damaging our continent for the past 40 years. If theal nations really want to help the Africans, they should finally terminate this awful aid. The countries that have collected the most development aid areones that are in the worst shape. Despite the billions that have poured into Africa, the continent remains poor. Spiegel: Do you have an explanation foradox? Shikwati: Huge bureaucracies are financed (with the aid money), corruption and complacency are promoted, Africans are taught to be beggars ande independent. In addition, development aid weakens the local markets everywhere and dampens the spirit of entrepreneurship that we so desperatelys absurd as it may sound: Development aid is one of the reasons for Africa's problems.are concerned here with aid predicated on expressive utility from generosity. Aid can be based on reciprocal benefit. See Younas (2008) on trade asating aid and Dreher et al. (2008) on reciprocation of US aid in exchange for UN General Assembly voting.ause aid-dispensing bureaucracies benefit by being the conduit for the expressive generosity, the ineffective aid may provide more than expressive utility,the material benefits to donor-countries’ citizens employed in donor bureaucracies (Vaubel, 1996). See Weaver (2008) on the World Bank and the

y between aid rhetoric and development outcomes. Welzman (2010) investigates the consequences of asymmetric information between donors andt governments.


6.1. Is voting expressive?

A possible alternative to expressive utility as the explanation for voting is avoidance of regret. If no one voted because everyonebelieved that a single vote is not decisive, every voter would regret not having voted. No one voting is therefore not a Nashequilibrium. Because in actual elections no individual voter can reasonably expect to be decisive, regret in having missed theopportunity to be decisive by not voting cannot explain why people vote. Regret can, however, confirm expressive voting. Newinformation or changed sentiments can lead people to express regret about how they voted. Such regret is clear evidence ofexpressive voting because the opportunity to change a voter's decision would not change the electoral outcome.31

Instrumental voters would maximize utility by voting for the policy or candidate closest to their ideal from among availablealternatives and abstain from voting only if indifferent between alternatives. Expressive voters are, in contrast, influenced by thedistance between their ideal and political parties' policy positions or candidates' attributes (see Appendix A). They abstain whenthere is too great a distance between their ideal and the alternatives offered (Brennan and Hamlin, 1998; Hillman, 2009,chapter 6). People who do not vote because they declare that they have “no one to vote for” confirm expressive voting.

Guttman et al. (1994) investigated whether voter abstentions are due to “indifference” as predicted by instrumental voting or“alienation” as predicted by expressive voting. The evidence was consistent with expressive voting: distance of availablecandidates from voters' ideal policies and not indifference determined whether people voted or abstained.32 The same study alsofound that the propensity of an individual to vote decreased with the number of “politically eligible adults” in the household. Suchbehavior is inconsistent with instrumental voting but consistent with expressive voting: the number of household members thatvotes does not affect the voting outcome but expressing identity can be delegated within the household.

An instrumental view of voting predicts greater participation in voting by low- than high-income people because low-incomepeople have a lower value of time. The evidence indicates to the contrary a greater likelihood of voting by high-income people,which is consistent with expressive voting. We expect low-income people in general to be more concerned with the materialrequisites of life than with seeking expressive utility through expressive voting.33

6.2. Other cases of expressive voting

Expressive voting is revealed in circumstances other than election of political representatives. Feigenbaum et al. (1988)hypothesized that if capital owners voted in their self-interest, they would support nuclear power, but found that they votedexpressively to oppose nuclear power. Fort and Bunn (1998) also found expressive voting on the issue of nuclear power: theyconcluded that people who incurred high costs of participation in voting voted against nuclear power because their voteprovided sufficiently high expressive utility to compensate for their high costs of participation in voting. Glazer (1992)proposed that majority decisions of workers to strike were instances of expressive behavior. Voting on a flag is expressive(Karahan and Shughart, 2004), as is voting on the choice between retaining a non-resident monarch and becoming a republic(Davidson et al., 2006), voting in the Eurovision song contest (Ginsburgh and Noury, 2008),34 and voting on the officiallanguages of multinational bodies (Fidrmuc et al., 2009).35 Voting in the United Nations General Assembly is expressive(Potrafke, 2009).36

Behavior is expressive in ways other than voting. People expressively confirm identity, for example, through the clothesthat they wear, the tone of voice that they adopt, the neighborhood in which they choose to live, and the cars that theydrive.37

6.3. Behavior in experiments

If behavior is expressive, this should be revealed in experiments. Indeed, expressive utility appears to explain much posited“non-rational” behavior in experiments.38

31 As Brennan and Hamlin (2000, p. 31) point out: “If you made a mistake in the polling booth and voted for the ‘wrong’ candidate, that mistake would almostcertainly not alter the electoral outcome — though, presumably, it would remain a mistake from your point of view.”32 The study used panel survey data from the 1976 U.S. presidential election.33 The relation between incomes and voting was studied by Wolfinger and Rosenstone (1980). Greater focus on expressive behavior when incomes are higheris consistent with observed greater public support for generous distributional policies in prosperous times: see for example Markussen (2008). Frey (1971)linked the greater likelihood of high-income people voting to better access to political information. The information can be used instrumentally (and irrationally)or expressively (and rationally). Voter turnout among low-income people was higher than usual in the 2008 U.S. presidential election, suggesting expressivevoting “in having someone to vote for”.34 People cannot vote for their own resident country's song, but voting can be influenced by common elements of identity through geographical proximity andlanguage, and immigrants can identify with their cultural home.35 Governments and people want their language included as an affirmation of identity.36 There is a low probability that a vote will be decisive and governments vote to confirm their identity through alignments.37 For examples of non-voting expressive behavior, see Copeland and Laband (2002), Laband et al. (2009), and Johansson-Stenman and Peter Martinsson(2006). The conspicuous consumption described by Veblen (1899) was expressive: the servants lined up to greet the master and mistress confirmed wealthrather than indicating the need for servants.38 The literature proposing irrational behavior in experiments is vast: for an example, see DellaVigna (2009).


6.3.1. Experiments on expressive votingAlthough individual voters are not decisive in real-life voting in usual elections, in experiments voters can be informed that

they will be decisive with indicated probabilities. Expressive voters are predicted by incentives paralleling those of table 1 to voteto givemoney to charity when the probability of being decisive is small, and to keepmoney for themselves when the probability ofbeing decisive is high. Voters who are expressively generous are predicted to vote to give away small sums of money when theprobability of being decisive is high. Experiments designed to investigate expressive voting reveal expressive generosity.

Carter and Guerette (1992) presented students from economics and accounting classes at the College of the Holy Cross inWorcester Massachusetts with the choice between keeping $6 or $9 (in different experiments) for themselves and giving $2 tocharity. Giving to charity was thus relatively costly. Only “weak support” was indicated for expressive voting: students tended tovote to donate to charity when decisive. The students' behavior exhibited expressive generosity. The students did not know— anddid not seem to care— to whom they were giving charity.39 They chose to confirm generous identity. In conversations with friendsand family, they would have incurred a loss of expressive utility when describing their participation in the experiment andreporting that “I took the money where possible for myself and did not care about donating to charity”.

Tyran (2004) conducted a similar experiment: students were given the equivalent of $6 and were informed of their probabilityof being decisive in a majority vote on alternatives of keeping the money and giving the money “to charity”. In a first type ofexperiment, if a majority voted in favor of charity, all students were obliged to donate their money without regard for how theyhad personally voted. In a second type of experiment, if a majority favored giving to charity, students could keep their money forthemselves, provided that they had personally voted against donating the money. Expressive voters would vote in favor of givingto charity in the first type of experiment but not the second. There was, however, no significant difference in behavior in the twotypes of experiments.40 Students in either case ostensibly voted to confirm a generous identity without regard for how theconditions of the experiment affected the likelihood of giving up the small sums of money that were at stake.41

In an experiment reported by Fischer (1996), students were offered the possibility of keeping $200 for themselves. The sum ofmoney was sufficiently high to evoke expressive behavior without expressive generosity. Students voted to keep the money whentheir likelihood of being decisive was high and voted in favor of giving to “charity”when their likelihood of being decisive was low.

6.3.2. Experiments involving cooperation and trustIn the single-interaction prisoners' dilemma or in public-good games, it is rational behavior not to cooperate, yet often in

experiments students achieve the efficient cooperative outcome. In repeated games with a predetermined number of rounds, theyoften cooperate until before the final round, and sometimes in the final round as well. Cooperation is predicted if the payoffsperceived by the students include expressive utility from confirmation of a cooperative identity. The attempt to free ride to exploitthe goodwill or kindness of others is inconsistent with “being a nice person”.42 In the trust game, expressive behavior by the donorallows confirmation of being a trusting person and expressive behavior of the recipient allows confirmation of being a trustworthyperson. In the predicted Nash equilibrium based on material gain alone, no money is transferred. Expressive utility explains whymoney is transferred and also returned.43

6.3.3. Experiments involving distributionIn ultimatum games, there is often affront and rejection of offers when recipients perceive unfairness. Recipients who feel that

they have been treated unfairly are prepared to incur a personal material cost in response to the perception that the donor is not agenerous person. Rejection of offers perceived as unfair confirms the identity predicated on reciprocating behavior that “I am areasonable person provided that others are reasonable in their behavior to me”. Expressive utility also explains otherwiseseemingly anomalous outcomes in which very high offers are rejected: the high offers are regarded as an affront to pride and thusas condescending (see for example Chen and Tang, 2009). The intended recipient would lose expressive utility if the “excessively”generous offer of the donor was accepted.

The dictator game is the purest test of expressive generosity. Because the decision whether to share is unilateral, people giveonly to confirm a generous identity. Donors do not in general know towhom they are giving, andmaywell be giving to people whoare better off than themselves. The purpose of their giving is not to help people who are less fortunate but to obtain expressiveutility from their expressive generosity.44

39 The students could have voted to take the larger sum for themselves and independently donate more than $2 to charity, thereby effecting a Paretoimprovement. Carter and Guerette noted that the presence of true charitable intentions would have had a “confounding” effect on their experiments. Theyviewed a claimed intention to donate part of retained money to charity as “rationalizing” a vote for personal money.40 That is, whether a student voted to be charitable did not depend on whether the outcome of majority voting would oblige students to be charitable(experiment 1) or being charitable was a personal decision (experiment 2).41 Tyran reported that he attempted to eliminate ethical considerations by not informing the students of the ethical implications of their decisions (p. 1652).Yet, surely students understood that ethics and ethical identity were involved in the choice between taking money for themselves and donating to charity. InTyran's experiments, students were given $3 for correctly predicting how others would vote. Tyran interpreted the experimental results as indicating thepresence of bandwagon effects and suggested that “it may be more rewarding to vote for a morally worthy cause if others are expected to do so, too”. Ashworthet al. (2006) found “bandwagon effects” in data on actual voting in Belgium. People may vote for sure winners to express themselves as “winners” in theirpersonal lives.42 On the prisoners' dilemma amended for utility from having a cooperative identity, see Hillman (2009, chapter 1); on public-good games, see chapter 3.43 Identity confirmed through expressive behavior also explains why money is not returned. An individual who does not return money in the trust game has anidentity for which utility is maximized through material success.44 On donor behavior when attributes of recipients are known, see Fong (2007). Donors then care about whether recipients are deserving of charity.


Eichenberger and Oberholzer-Gee (1998) reported outcomes of “bandit” games in which a student could take money fromanother student. The behavior was consistent with expressive generosity, with “bandit” students leaving some money for theother student. The behavior was not charitable or generous in leaving money for people in need. The behavior was expressivelygenerous because there could be no presumption that the participant with whom a student had been paired deserved charity.

List (2007) and Bardsley (2008) reported that behavior changeswhen the “dictator” can takemoney aswell as give. Less is thenon average given. The less generous behavior is expressive. When taking as well as giving is an option, a generous identity can beconfirmed by not taking and giving less.

Fischer (1996) reported that expressively voting for charity increased when behavior was observed by others, indicating thepresence of a social approval motive. Expressive utility obtained from approval from others is also indicated by outcomes in whichnothing is given when the decision is completely anonymous, including with respect to the researcher overseeing the experiment,and by evidence (Cason and Mui, 1997) that donors are more generous in their proposals for giving when decisions are madecollectively.

6.3.4. Behavior of economics studentsThe experimental evidence indicates that knowledge of economics is correlated with being more self-interested.45 A question

regarding behavior of economics students is whether the students have chosen their identity in the sense of having preselectedthemselves through predetermined attributes to study economics— or whether their behavior is a consequence of an identity thatis formulated through their education in studying economics. Perhaps there is pre-selection through the decisionwhether to studyeconomics. However, when economics students are introduced to the prisoners' dilemma, they not only learn the concepts ofdominant strategy and Nash equilibrium; they are also told that payoffs as material rewards are the sole source of utility.Expressive utility is not included in the valuation of outcomes. Non-economics students tend to show greater awareness ofexpressive utility and seem to be more inclined to seek to confirm an identity of being generous and cooperative. Economicsstudents can be expressive in not cooperating. Nonetheless, even economics students, provided that the material costs aresufficiently small, may continue to cooperate in a repeated prisoners' dilemma because of the expressive utility from confirming anidentity as a nice person, and indeed the identity of a nicer person than people who do not cooperate.

6.3.5. Gender differencesThere are gender differences in behavior.46 Although gender is predetermined, there are aspects of gender-related identity that

are nonetheless a choice. Croson and Gneezy (2009) summarize the evidence as indicating that men and women often behavedifferently but the “exceptions to the rule” are “managers and professional populations”. Men and women, thus, are indicated topredicate behavior on similar identities in professional roles, but they differ in the identities confirmed inmore personal aspects ofbehavior. In dictator games, for example, when giving is subject to a penalty, women tend to be less deterred by the cost of givingwhen deciding whether or how much to give to others (Andreoni and Vesterlund, 2001). Women therefore exhibit behaviorconsistent with a more sympathetic and less calculating identity than men, and are more expressive than men in confirming agenerous identity without regard for the personal cost of giving.

6.3.6. Intrinsic motivationExpressive behavior explains “intrinsic motivation”. People are often unwilling to create the markets predicted by the Coase

theorem but are willing to do for no material reward that for which they reject payment.47 The offer of money is an affront to theiridentity. The decline in expressive utility from accepting payment outweighs the material benefit.

6.4. Expressive behavior as a unifying explanation in experiments

In various experimental situations, being nice or pleasant is not predicted to yield best personal outcomes based on rationalbehavior motivated by material reward.48 Yet, some cases of economics students aside, people are often, or even in general, inexperiments. The niceness is expressive behavior. Expressive utility makes the observed nice behavior in experiments rational.People respond in their behavior not only to material benefit, but also to the expressive utility from decisions about whether tocooperate, be reciprocally kind, trust others, be trustworthy, and be fair or ethical in sharing. As is clear from the behaviordescribed as due to intrinsic motivation, material gain is willingly forgone when offsetting expressive utility is higher.

7. Final issues

I now address some final issues concerning (1) other related types of behavior, (2) institutions, and (3) resolution of problemswhen there are social costs of expressive behavior.

45 See for example Frank et al. (1993) and Frank and Schulze (2000). For an overview of the literature comparing behavior of economists and non-economists,see Kirchgässner (2005).46 For summaries, see Eckel (2008) and Eckel and Grossman (2008).47 On the evidence, see Hillman (2009, chapter 5).48 In programming experiments in repeated games based on rational behavior with solely material benefits, nice behavior does not ensure efficient outcomes(Donninger, 1986).


7.1. Related behavior

Various types of behavior are related to or can coexist with expressive behavior.

7.1.1. Morality and ethicsMorality and ethics, of course, affect behavior.49 Moral or ethical behavior is subsumed into expressive behavior through choice

of identity as being a moral or ethical person.50 Some people might choose to behave morally only if material benefit foregone issufficiently low. However, people may also refuse to compromise their identity as an ethical person irrespective of the materialincentive.51

7.1.2. Altruism and maliceWhat are the consequences when expressive behavior coexists with altruism or malice?Voting is expressive when a person votes for the altruistic reason of supporting democratic principles because one vote is

inconsequential for sustaining democracy. Malice is present if the decision for whom to vote is made to “punish” a candidate orpolitical party.52 Because an individual vote is inconsequential in “punishing”, voting to “punish” (or to protest) is expressive.

In the case of expressive rhetoric, malice can underlie the declaration that other people should not defend themselves. Thedeclaration can also be expressive and intended to signal “your defense would entail additional violence and I am not a violentperson”. In terms of outcomes if the advice is followed, the reason for denial of self-defense does not of course matter.53

Malice has no role in expressive generosity because disadvantageous consequences for recipients are not intended. Altruism is,on the other hand, a natural complement of expressive generosity: people may give to confirm a generous identity and alsobecause they obtain utility from better outcomes for others.54

7.1.3. DelusionThe literature describes behavior based on delusion. Akerlof (1989) described willful delusionary behavior in which people

choose bias in perception of information that trades off their “desires to feel good about themselves” and reality. Cowen (2005)proposed that voters disregard information that is inconsistent with their self-image or that would imply that they had madeincorrect decisions in the past; their delusionary behavior includes the belief in being the decisive voter. Caplan (2007) proposedthat people willingly delude themselves by believing what they want to believe and choose the beliefs that give them highestutility. Expressive behavior differs from these descriptions of delusionary behavior, because behavior can change according towhether people's decisions are expected to affect their material utility.55

7.2. Institutions

The media and political parties can reflect or internalize expressive behavior.

7.2.1. Expressive mediaThe media have incentives to base reporting on consistency with the requisites of the expressive utility of readers and

listeners.56 Themedia can profit by catering to identity-confirming interpretations and perspectives of expressive populations. Themedia can, for example, expressively prescribe soft power. When populations to whom the media cater obtain expressive utilitythrough expressive support for one side in a conflict, the media can provide substantiation of biased expressive perceptions byselective prejudicial reporting. True information that contradicts the predispositions of the media's expressive audience woulddiminish the expressive utility that the media provides — and would diminish media profits.57

49 See for example Holler (1986), who returns to Adam Smith's Moral Sentiments, and Holler and Kanniainen (2010).50 Thus for example Feddersen et al. (2009) described generosity in expressive-voting experiments as “moral bias” but “moral bias” is expressive behavior.51 The saying “every man has his price” presupposes expressive utility. The sufficiently high “price” compensates for expressive utility lost when principles ofmoral behavior are compromised. The saying is most famously attributed to the first British Prime Minister Robert Walpole (1676–1745) but, as could beanticipated, predates him. Walpole made the statement in the context of contemporary politics and as a politician of his time was concerned only with the priceof men.52 See Glazer (2008), who describes voting to anger (or please) others.53 If there is malice, we are led to seek the reasons. See Glaeser (2005).54 Andreoni (1990) distinguished “impure” from “pure” altruism in personal contributions to a public good. Altruism was “impure” when the source of utilitywas increased availability of the public good and “pure”when utility increased only because of the act of giving. In the context of expressive behavior, utility from“pure altruism” is expressive through confirmation of generous identity, and is not altruistic if the donor is motivated only by own expressive utility.55 People can be delusional when voting expressively. Sobel and Wagner (2004) observed greater redistribution of income by government in more populous U.S.states, which they attributed to voters in these states perceiving themselves as less likely to be decisive than voters in stateswith small populations. There is in that casedelusion because the probability of being a decisive voter is negligible when the size of the electorate reaches that of the U.S. states with the smallest populations.56 The financial media in general refrains from being expressive: accurate information and not expressive rhetoric is sought when decisions are made aboutpersonal finance and wealth.57 Mullainathan and Shleifer (2005) describe manipulation of reporting of news. Iyengar and Hahn (2009) present experimental evidence showing that peoplematch personal preferences with choice of media ideology.


7.2.2. Expressive political partiesThere would be disappointment if, after a majority of voters has expressively voted for an outcome that the voters do not want,

elected representatives set out to implement the policies. Disappointment is avoided when political representatives are alsoexpressive. Expressive voters then achieve their preferred outcome of obtaining expressive utility without incurring the personalmaterial costs of the policies for which they voted. Before an election, the political party supported by the expressive majority usesexpressive rhetoric to proclaim the need for generous sharing of income and wealth, or proclaims the merits of compromise andconciliation in face of supreme-value adversaries. Then, having won the election, elected representatives may be aware thatproceeding to implement the expressive policies will result in defeat in a future election by another political party that isexpressive in rhetoric and understands that voters do not actually want the policies for which they expressively vote. The policiesproclaimed before the election are therefore not implemented. Expressive rhetoric can continue, and increases when the rhetoricsubstitutes for the unimplemented expressively supported policies.58

7.3. Solutions

When expressive behavior imposes the social costs of expressive-policy traps, a society might seek means of avoiding orescaping the traps.

7.3.1. Usual remedies for social costsUsual remedies cannot be applied to the social costs of expressive behavior. Coase bargaining and Pigovian tax-subsidy

solutions cannot be used. Direct regulation of expressive behavior is also not feasible.59

7.3.2. Resistance to remediesThere are societal incentives to resist solutions to social costs of expressive behavior. The resistance is through the conventions

of political correctness, which impose the restriction: “You are not allowed to say (or write) that — even if it is true”.60 Peoplebehaving expressively do not wish to be informed that their expressiveness is in disregard of facts and data. They do not wish to betold what they may already know.61 As an example, prosperous people in rich countries who seek to expiate feelings of guilt byexpressively supporting a domestic welfare state and open immigration, lose expressive utility when it is pointed out that thepolicies are mutually inconsistent.62 In another example, expressive utility of donors is protected when political correctnessrequires ineffectiveness of aid to be attributed to geographical disadvantage rather than avarice of rulers and governments inrecipient countries. The claim that “the reason for development failure is geography” allows the self-blaming expressive rhetoric(see, for example, Sachs, 2006) that “we in rich countries should give more, because aid would be effective if only more aid wereprovided”.

7.3.3. Salient eventsExpressive behavior has been changed by salient events. Thomas Moore (1551) and then, more influentially, Karl Marx and

Frederick Engels in the 19th century, described “utopia” as a state devoid of the inequalities of private property. Expressive supportfor the utopian vision persisted in the west into the latter part of the 20th century, undeterred by professional analysis (Hayek,1944) and personal description (Solzhenitsyn, 1974) of an elitist enslaving system. The expressive support subsided whencommunism ended.63 Salient change may similarly diminish the expressive behavior that sustains other expressive-policy traps.In welfare states, demographic change and adverse selection through immigration may diminish expressive support for generouswelfare entitlements and for immigration independently of the human capital and productiveness of immigrants. For issues ofpublic safety, expressive soft-power rhetoricmay declinewhen threats that previously confronted only thosewho are preached to,begin to make an appearance in the preachers' own personal lives. In foreign aid, on the other hand, expressive generosity may bedifficult to curb because of the symbiotic benefits to the expressive donors and to the recipient governments engaged in thepretense of benevolence.

58 An expressive political party such as “new Labor” in the United Kingdom allowed the middle class to vote expressively — as similarly has been the case withthe Labor Party in Australia and “social democratic” parties elsewhere.59 Paternalistic solutions that can be effective for children (prompting by parents can lead the children to shed their expressive identities) cannot be readilyapplied to adults.60 In contrast, cognitive dissonance (for example, Akerlof and Dickens, 1973) is unwillingness to accept verified information as objective truth.61 There may be public knowledge (everyone knows); political correctness attempts to protect expressive utility by preventing information from becomingcommon knowledge (everyone knows that everyone knows). Loury (1994) pointed out the reputational concerns that underlie incentives for voluntaryadherence to restraints of political correctness. Morris (2001) showed more formally that a person with known true information has a reputational incentive tolie in order to avoid being regarded as politically incorrect. In Hillman (1998), I considered political correctness as a pedagogic devise to protect an ideology.Kuran (1995) described the inhibitions on truth when people are obliged to live with “private truths” and “public lies”. Stigler (1982, p. 13) noted the restraintsof political correctness in describing economists as preachers and proposing that “the main lesson that I draw from our experience as preachers is that we arewell received in the measure that we preach what the society wants to hear”.62 On the mutual inconsistency, see for example Sinn (1997) and Nannestad (2004, 2007, 2009). A generous welfare state cannot be sustained in face of non-replacement domestic fertility and ongoing arrival of immigrants who disproportionately draw on welfare benefits.63 In academia, in the year of the beginning of the end of Soviet communism, a popular textbook informed students that: “the Soviet economy is proof that,contrary to what many skeptics had earlier believed, a socialist command economy can function and even thrive” (Samuelson and Nordhaus, 1989, p. 837).Norwood (2009) documents prior support in U.S. academia for German national-socialism. I cite these examples in a context of expressive support forcommunism and national-socialism even in the U.S., compared to Europe where the support has been more pervasive.


8. Concluding remarks

I have proposed expressive behavior as a replacement for the explanations of behavioral economics for why people (accordingto behavioral economics) depart from rationality. It is not irrational to choose to forego a small sum of money in order to confirm apleasing identity or to decide that the expressive utility from being accommodating without monetary compensation exceeds thematerial gain from participating in Coase negotiations. Public policy is not at issue in the personal decisions of experiments. Insocietal outcomes, the expressive low-cost behavior that is beneficial in experiments is the reason for the unwanted public policiesof expressive-policy traps. This has left us with the question of how expressive-policy traps might be escaped.

Acknowledgements

This paper is a development of a plenary address to the European Public Choice Society in Athens, April 2009. Versions of thepaper were presented at seminars in Israel at Bar-Ilan University, in Australia at Australian National University, Deakin University,Monash University, the University of Tasmania, the University of New South Wales, and the University of Wollongong, and inEurope at a PCRC workshop on Power, Games, and Institutions in Åland Finland in August 2008, as a plenary lecture at the CESifoConference on Political Economy in Dresden in December 2009, and at the University of Freiburg, the University of Göttingen, theUniversity of Hamburg, and the University of Paris 1— Panthéon Sorbonne. I thank for helpful discussionsMichael Brooks, ManfredHoller, Oliver Landmann, Vai-LamMui, Yew Kwang Ng, Günther Schultze, and Heinrich Ursprung. Helpful observations were alsoprovided by Nick Baigent, Peter Bernholz, Frank Bohn, Wilfred Ethier, Raphael Franck, Louis Lévy-Garboua, Amihai Glazer, DrorGoldberg, Robert Gregory, Joel Guttman, Alan Hamlin, Carsten Hefeker, Colin Jennings, Bryan McCannon, Dennis Mueller, NiklasPotrafke, Joan Rogers, Jacob Saciuk, William Shughart II, Thierry Verdier, Xin Meng, and Mor Zahavi.

Appendix A

The appendix sets out with more formality the description of expressive behavior, using the characterization of expressivevoting. An individual j in a population of size n obtains expressive utility by voting to confirm own identity, or perhaps does notvote. There arem candidates for office. We denote individual j's decision to vote for candidate v by xj

v and the decision to abstain byxjA. Personal benefits and costs are defined with reference to the consequences of the voting or abstention decision. Individual j isrationally aware that his or her vote is with certainty not decisive. Material utility from voting is

64 Rath

UMAT xkj� �

= Bj vð Þ−Cj xkj� �

; k = A;1;…m; j = 1;…n: ð1Þ

Bj(v) is personal material benefit for individual j if candidate v wins the election, which in a general setting can depend on theindividual's decision xj

k. Cj(xjk) is the material cost personally incurred from the decision xjk. The cost of abstaining is zero for all

individuals:

CAj = 0; j = 1;…n: ð2Þ

The material cost in terms of time of individual j voting for one of the m candidates is positive:

Cvj = Cj N 0; v = 1;2;…m; j = 1;…;n ð3Þ

When individual j is not decisive, the voting decision does not affect material benefit from voting. The individual benefitsmaterially when the outcome of voting is election of a candidate with favorable policies; however, the benefit from the election ofthe favorable candidate is not attributable to the individual's voting decision when the individual's vote is not decisive. GivenEqs. (2) and (3), individual j if not decisive maximizes material utility (1) by abstaining from voting.64

If individual j is decisive (or pivotal) with respect to his or her preferred candidate vj⁎, then:

vj⁎ = argmax Bj vð Þ−Cj

h i; v = 1;…;m; j = 1;…n: ð4Þ

The decision in Eq. (4) by a decisive voter to vote for candidate vj⁎ is based onmaterial utility (1). However, when individual j isnot decisive, material utility does not influence the voting decision. If only material utility mattered, the individual wouldrationally abstain. Expressive utility can lead the non-decisive individual to vote.

Expressive voters care about closeness of candidates' attributes to their expressive ideal. A Euclidean measure of distancebetween candidates is therefore required as perceived by individual voters with respect to their expressive ideal. With vj

ideal

indicating individual j's ideal expressive candidate, expressive utility of individual j from voting for a candidate vk is

UEXPj xkj

� �= Ij−Lj jxidealj −xkj j

� �: ð5Þ

er than zero probability of being decisive, the voter may perceive some small probability of being decisive.


An expressive individual may decide to abstain, which can provide expressive utility, or there may be loss of expressive utilityfrom being unable to express oneself through voting. We denote expressive utility from not voting by:

UEXPj xAj

� �= QA

j : ð6Þ

When the individual chooses to vote for one of the candidates, the choice that maximizes expressive utility is:

xvE*j = argmax UEXPj xvEj

� �: ð7Þ

If xjideal is feasible, that is, if xjideal=xz for some z=1, …, m,

xvE*j = xideal: ð8Þ

If xjideal is not feasible, expressive utility declines with the distance of xjideal from xjvE⁎.

The decision whether to vote depends on whether there is a candidate sufficiently close to the individual's expressive ideal.Total utility is:

Uj xkj� �

= UMATj xkj

� �+ UEXP

j xkj� �

: ð9Þ

An individual expressively votes for one of the candidates if:

Lj jxvE*j −xideal j� �

≤ Ij−Qj−Cj: ð10Þ

Define the voting decision that maximizes Eq. (9) as:

xk*j = argmax Uj xkj� �

: ð11Þ

For an individual who is not decisive, possibilities are

xk*j =

xAj

xv*j = xvE*j

xvE*j ≠ xv*j

8>>>><>>>>:

ð12Þ

An individual who abstains for material reasons may also not vote for expressive reasons: the presence of expressive utilitythus does not ensure that a non-decisive individual votes. When, based on expected material utility alone, an individual woulddecide not to vote, expressive utility changes the decision to voting for the preferred feasible expressive alternative, if thecondition (10) is satisfied.

All individuals with the same material preferences need not have the same expressive preferences. Also, some people mightmaximize expressive utility by voting contrary to material self-interest whereas others do not. However, in the absence ofexpressive utility, voting would never be contrary to material self-interest and expressive-policy traps could not arise.

References

Abadie, A., 2006. Poverty, political freedom, and the roots of terrorism. American Economic Review 96, 50–56.Akerlof, G.A., 1989. The economics of illusion. Economics and Politics 1, 1–15.Akerlof, G.A., Dickens, W.T., 1973. The economic consequences of cognitive dissonance. American Economic Review 63, 307–319.Akerlof, G.A., Kranton, R.E., 2000. Economics and identity. Quarterly Journal of Economics 115, 715–753.Andreoni, J., 1990. Impure altruism and donations to public goods: a theory of warm-glow giving. Economic Journal 100, 464–477.Andreoni, J., Vesterlund, L., 2001. Which is the fairer sex? Gender differences in altruism. Quarterly Journal of Economics 116, 293–312.Ashworth, J., Geys, B., Heyndels, B., 2006. Everyone likes a winner: an empirical test of the effect of electoral closeness on turnout in a context of expressive voting.

Public Choice 128, 383–405.Austen-Smith, D., Fryer Jr., R.G., 2006. An economic analysis of “acting white”. Quarterly Journal of Economics 120, 551–583.Bardsley, N., 2008. Dictator game giving: altruism or artefact? Experimental Economics 11, 122–133.Bernholz, P., 1993. Necessary conditions for totalitarianism: supreme values, power, and personal interest. In: Radnitzsky, G., Boullion, H. (Eds.), Government:

Servant or Master. Rodopi, Amsterdam and Atlanta, pp. 267–312.Bernholz, P., 2004. Supreme values as the basis for terror. European Journal of Political Economy 20, 317–333.Borooah, V.K., Paldam, M., 2007. Why is the world short of democracy? A cross-country analysis of barriers to representative government. European Journal of

Political Economy 23, 582–604.Brennan, G., Buchanan, J.M., 1984. Voter choice: evaluating political alternatives. American Behavioral Scientist 28, 185–201.Brennan, G., Hamlin, A., 1998. Expressive voting and electoral equilibrium. Public Choice 95, 149–175.


Brennan, G., Hamlin, A., 2000. Democratic Devices and Desires. Cambridge University Press, Cambridge, U.K.Brennan, G., Lomasky, L., 1984. Inefficient unanimity. Journal of Applied Philosophy 1, 151–163.Brennan, G., Lomasky, L., 1993. Democracy and Decision. Cambridge University Press, Cambridge, U.K.Buchanan, J.M., 1954. Individual choice in voting and the market. Journal of Political Economy 62, 334–343.Caplan, B., 2007. The Myth of the Rational Voter: Why Democracies Choose Bad Policies. Princeton University Press, Princeton NJ.Carter and Guerette, 1992. An experimental study of expressive voting. Public Choice 73, 251–260.Cason, T.N., Mui, V.-L., 1997. A laboratory study of group polarization in the team dictatorship game. Economic Journal 107, 1465–1483.Chen, K., Tang, F.-F., 2009. Cultural differences between Tibetans and ethnic Han Chinese in ultimatum bargaining experiments. European Journal of Political

Economy 25, 78–84.Chen, Y., Xin Li, S., 2009. Group identity and social preferences. American Economic Review 99, 431–457.Copeland, C., Laband, D.N., 2002. Expressiveness and voting. Public Choice 110, 351–363.Corneo, G., Grüner, H.P., 2000. Social limits to redistribution. American Economic Review 90, 1491–1507.Cowen, T., 2004. Road map to Middle Eastern peace? A public choice perspective. Public Choice 118, 1–10.Cowen, T., 2005. Self-deception as the root of political failure. Public Choice 124, 437–451.Croson, R., Gneezy, U., 2009. Gender differences in preferences. Journal of Economic Literature 47, 448–474.Davidson, S., Fry, T.R.L., Jarvis, K., 2006. Direct democracy in Australia: voter behavior in the choice between a constitutional monarchy and a republic. European

Journal of Political Economy 22, 862–873.de Vries, J., 2008. The Industrious Revolution. Cambridge University Press, Cambridge, U.K.DellaVigna, S., 2009. Psychology and economics: evidence from the field. Journal of Economic Literature 47, 315–372.Diekmann, A., 1985. Volunteer's dilemma. Journal of Conflict Resolution 29, 605–610.Diekmann, A., 1986. Volunteer's dilemma. A social trap without a dominant strategy and some empirical results. In: Diekmann, A., Mitter, P. (Eds.), Paradoxical

Effects of Social Behavior: Essays in Honor of Anatol Rapoport. Physica-Verlag, Heidelberg Wien, pp. 187–197.Donninger, C., 1986. Is it always efficient to be nice? A computer simulation of Axelrod's computer tournament. In: Diekmann, A., Mitter, P. (Eds.), Paradoxical

Effects of Social Behavior: Essays in Honor of Anatol Rapoport. Physica-Verlag, Heidelberg Wien, pp. 123–134.Doucouliagos, H., Paldam, M., 2008. Aid effectiveness on growth: a meta study. European Journal of Political Economy 24, 1–24.Downs, A., 1957. An Economic Theory of Democracy. Harper and Row, New York NY.Dreher, A., Thiele, R., Nunnenkamp, P., 2008. Does US aid buy UN General Assembly votes? A disaggregated analysis. Public Choice 136, 139–164.Easterly, W., 2001. The Elusive Quest for Growth: Economists' Adventures and Misadventures in the Tropics. MIT Press, Cambridge MA.Easterly, W., 2006. The White Man's Burden: Why the West's Efforts to Aid the Rest Have Done So Much Ill and So Little Good. The Penguin Press, New York NY.Eckel, C.C., 2008. The gender gap: using the lab as a window on the market. In: Bettio, F., Verashchagina, A. (Eds.), Frontiers in the Economics of Gender. Siena

Studies in Political Economy, Routledge, London, pp. 223–242.Eckel, C.C., Grossman, P.J., 2008. Differences in the economic decisions of men and women: experimental evidence. In: Plott, C., Smith, V. (Eds.), Handbook of

Experimental Economics Results, Volume 1. Elsevier, Amsterdam, pp. 509–519.Eichenberger, R., Oberholzer-Gee, F., 1998. Rational moralists: the role of fairness in democratic economic politics. Public Choice 94, 191–210.Feddersen, T.J., 2004. Rational choice theory and the paradox of not voting. Journal of Economic Perspectives 18, 99–112.Feddersen, T., Gailmard, S., Sandroni, A., 2009. Moral bias in large elections: theory and experimental evidence. The American Political Science Review 103,

175–192.Feigenbaum, S., Karoly, L., Levy, D., 1988. When votes are words, not needs: some evidence from the Nuclear Freeze Referendum. Public Choice 58, 201–216.Fidrmuc, J., Ginsburgh, V., Weber, S., 2009. Voting on the choice of core languages in the European Union. European Journal of Political Economy 25, 56–62.Fiorina, M.P., 1976. The voting decision: instrumental and expressive aspects. Journal of Politics 38, 390–413.Fischer, A.J., 1996. A further experimental study of expressive voting. Public Choice 88, 171–184.Fong, C., 2007. Evidence from an experiment on charity to welfare recipients: reciprocity, altruism, and the empathic responsiveness hypothesis. Economic Journal

117, 1008–1024.Fort, R., Bunn, D.N., 1998. Whether one votes and how one votes. Public Choice 95, 51–62.Franck, R., Hillman, A.L., Krausz, M., 2005. Public safety and the moral dilemma in the defense against terror. Defense and Peace Economics 16, 347–364.Frank, B., Schulze, G.G., 2000. Does economics make citizens corrupt? Journal of Economic Behavior & Organization 43, 101–113.Frank, R.H., Gilovich, T., Regan, D.T., 1993. Does studying economics inhibit cooperation? Journal of Economic Perspectives 7, 159–171.Frey, B.S., 1971. Why do high income people participate more in politics? Public Choice 11, 101–105.Frey, B.S., 2003. Publishing as prostitution? — Choosing between one's own ideas and academic success. Public Choice 116, 205–223. Reprinted in Congleton, R.D.,

Hillman, A.L., Konrad, K.A. (Eds.), 2008. 40 Years of Research on Rent Seeking 2 — Applications: Rent Seeking in Practice. Springer, Heidelberg, pp. 749–767.Frey, B.S., 2004. Dealing with Terrorism: Stick or Carrot? Edward Elgar, Cheltenham, U.K.Frisch, H., Inbar, E. (Eds.), 2008. Radical Islam and International Security: Challenges and Responses. Routledge, London.Ginsburgh, V., Noury, A.G., 2008. The Eurovision song contest. Is voting political or cultural? European Journal of Political Economy 24, 41–52.Glaeser, E.L., 2005. The political economy of hatred. Quarterly Journal of Economics 120, 45–86.Glaeser, E.L., Ward, B.A., 2006. Myths and realities of American political geography. Journal of Economic Perspectives 20, 119–144.Glazer, A., 1987. A new theory of voting: why vote when millions of others do. Theory and Decision 22, 257–270.Glazer, A., 1992. An expressive-voting theory of strikes. Economic Inquiry 30, 733–741.Glazer, A., 2008. Voting to anger and to please others. Public Choice 134, 247–254.Guttman, J.M., Hilger, N., Shachmurove, Y., 1994. Voting as investment vs. voting as consumption: new evidence. Kyklos 47, 197–207.Hamlin, A., Jennings, C., 2010. Expressive political behaviour: foundations, scope and implications. Working Paper 08-19, Politics, University of Manchester. http://

www.socialsciences.manchester.ac.uk/disciplines/politics/about/themes/mancept/workingpapers/.Hayek, 1944/1972. The Road to Serfdom. University of Chicago Press, Chicago IL.Hillman, A.L., 1998. Political economy and political correctness. Public Choice 96, 219–239. Presidential Address, European Public Choice Society, Prague, April

1997. Reprinted in: Congleton, R.D., Hillman, A.L., Konrad, K.A. (Eds.), 2008. Forty Years of Research on Rent Seeking 2— Applications: Rent Seeking in Practice.Springer, Heidelberg, pp. 791–811.

Hillman, A.L., 2002. The World Bank and the persistence of poverty in poor countries. European Journal of Political Economy 18, 783–795.Hillman, A.L., 2004. Nietzschean development failures. Public Choice 119, 263–280.Hillman, A.L., 2007. Economic and security consequences of supreme values. Public Choice 30, 259–280.Hillman, A.L., 2009. Public Finance and Public Policy: Responsibilities and Limitations of Government, 2nd Ed. Cambridge University Press, New York NY.Hirshleifer, J., 1983. From weakest-link to best-shot: the voluntary provision of public goods. Public Choice 41, 371–386.Holler, M.J., 1986. Moral sentiments and self-interest reconsidered. In: Diekmann, A., Mitter, P. (Eds.), Paradoxical Effects of Social Behavior: Essays in Honor of

Anatol Rapoport. Physica-Verlag, Heidelberg Wien, pp. 223–234.Holler, M.J., Kanniainen, V. (Eds.), 2010. Special issue on ethics and economics. European Journal of Political Economy 26, 293–402.Inbar, E., 2006. Review of B.S Frey, Dealing with terrorism: stick or carrot. European Journal of Political Economy 22, 343–344.Iyengar, S., Hahn, K.S., 2009. Red media, blue media: evidence of ideological selectivity in media use. Journal of Communication 59, 19–39.Johansson-Stenman, O., Peter Martinsson, P., 2006. Honestly, why are you driving a BMW? Journal of Economic Behavior & Organization 60, 129–146.Jones, P., Hudson, J., 2000. Civic duty and expressive voting: is virtue its own reward? Kyklos 53, 3–16.Karahan, G.R., Shughart II, W.F., 2004. Under two flags: symbolic voting in the state of Mississippi. Public Choice 118, 105–124.Kirchgässner, G., 1992. Toward a theory of low-cost decisions. European Journal of Political Economy 8, 305–320.Kirchgässner, G., 2005. (Why) are economists different? European Journal of Political Economy 21, 543–562.

http://www.socialsciences.manchester.ac.uk/disciplines/politics/about/themes/mancept/workingpapers/

http://www.socialsciences.manchester.ac.uk/disciplines/politics/about/themes/mancept/workingpapers/


Kirchick, J., 2009. What price popularity. Commentary 128, 22–27.Kliemt, H., 1986. The veil of insignificance. European Journal of Political economy 2, 233–244.Klor, E.S., Shayo, M., 2010. Social identity and preferences over redistribution. Journal of Public Economics 94, 269–278.Krueger, A., 2007. What Makes a Terrorist? Economics and the Roots of Terrorism. Princeton University Press, Princeton NJ.Krueger, A., Maleckova, J., 2003. Education, poverty, and terrorism. Journal of Economic Perspectives 17, 119–144.Kuran, T., 1995. Private Truths, Public Lies: The Social Consequences of Preference Falsification. Harvard University Press, Cambridge MA.Laband, D.M., Pandit, R., Sophocleus, J.S., Laband, A.M., 2009. Patriotism, pigskins, and politics: an empirical examination of expressive behavior and voting. Public

Choice 138, 97–108.Lewisch, P., 2004. A theory of identification. International Review of Law and Economics 23, 439–451.List, J.A., 2007. On the interpretation of giving in dictator games. Journal of Political Economy 115, 482–493.List, J.A., Gallett, C., 2001. What experimental protocol influence disparities between actual and hypothetically stated values? Evidence from a meta-analysis.

Environmental and Resource Economics 20, 241–254.Loury, G.C., 1994. Self-censorship in public discourse: a theory of ‘political correctness’ and related phenomena. Rationality and Society 6, 428–461.Markussen, S., 2008. How the left prospers from prosperity. European Journal of Political Economy 24, 329–342.Miller, C., 2008. 9/11 and the novelists. Commentary 126, 32–35.Moore, T., 1551/1949. Utopia. (Translated and edited by H.V.S. Ogden). Appleton-Century-Crofts, New York NY.Morris, S., 2001. Political correctness. Journal of Political Economy 109, 231–265.Mueller, D.C., 2003. Public Choice III. Cambridge University Press, Cambridge, U.K.Mueller, D.C., 2004. Rights and citizenship in a world of global terrorism. European Journal of Political Economy 20, 335–348.Mullainathan, S., Shleifer, A., 2005. The market for news. American Economic Review 95, 1031–1053.Nannestad, P., 2004. Immigration as a challenge to the Danish welfare state? European Journal of Political Economy 20, 755–767. Reprinted in: Congleton, R.D.,

Hillman, A.L., Konrad, K.A. (Eds.), 2008. Forty Years of Research on Rent Seeking 2— Applications: Rent Seeking in Practice. Springer, Heidelberg, pp. 281–293.Nannestad, P., 2007. Immigration and welfare states: a survey of 15 years of research. European Journal of Political Economy 23, 512–532.Nannestad, P., 2009. Unproductive immigrants: a socially optimal policy for rational egalitarians. European Journal of Political Economy 25, 562–566.Norwood, S.H., 2009. The Third Reich in the Ivory Tower. Cambridge University Press, Cambridge, U.K.Nye, J.S., 2004. Soft Power: The Means to Success in World Politics. Public affairs, New York NY.Piazza, J.A., 2006. Rooted in poverty? Terrorism, poor economic development, and social cleavages. Terrorism and Political Violence 18, 159–177.Plaut, S., 2004. Misplaced applications of economic theory to the Middle East. Public Choice 118, 11–24.Potrafke, N., 2009. Does government ideology influence political alignment with the US? An empirical analysis of voting in the UN General Assembly. Review of

International Organizations 4, 245–268.Rawls, J.A., 1971. A Theory of Justice. Belknap Press, Cambridge MA.Rose, S., 1988. Gender antagonism and class conflict: exclusionary strategies of male trade unionists in nineteenth century Great Britain. Social History 13,

191–208.Rubin, P.H., 2002. Darwinian Politics: The Evolutionary Origin of Freedom. Rutgers University Press, New Brunswick NJ.Sachs, J., 2006. The End of Poverty: Economic Possibilities for our Time. The Penguin Press, New York NY.Samuelson, P.A., Nordhaus, W.D., 1989. Economics, 13th Ed. McGraw Hill, New York NY.Schuessler, A.A., 2000. A Logic of Expressive Choice. Princeton University Press, Princeton NJ.Sinn, H.-W., 1997. The selection principle and market failure in systems competition. Journal of Public Economics 66, 247–274.Sobel, R.S., Wagner, G.A., 2004. Expressive voting and government redistribution: testing Tullock's charity of the uncharitable. Public Choice 119, 143–159.Solzhenitsyn, A.I., 1974. Gulag Archipelago 1918–1956: An Experiment in Library Investigation. Parts I–II. Harper and Row, New York. (Completed in 1968.

Unbanned in Russia, 1989).Spiegel On-line International, 2010. Interview with James Shikwati (7 April 2010).Stigler, G.J., 1982. The Economist as Preacher and Other Essays. The University of Chicago Press, Chicago IL.Svensson, J., 2000. Foreign aid and rent seeking. Journal of International Economics 51, 437–461. Reprinted in: Congleton, R.D., Hillman, A.L., Konrad, K.A. (Eds.),

2008. 40 Years of Research on Rent Seeking 2. Applications: Rent Seeking in Practice. Springer, Heidelberg, pp. 165–189.Tullock, G., 1959. Problems of majority voting. Journal of Political Economy 67: 571–9. Reprinted in Arrow, K.J., Scitovsky, T. (Eds.), 1969. Readings in Welfare

Economics. Richard D. Irwin, Homewood IL, pp. 169–178.Tullock, G., 1967. Toward a Mathematics of Politics. University of Michigan Press, Ann Arbor MI.Tullock, G., 1971. The charity of the uncharitable. Western Economic Journal 9, 379–392.Tyran, 2004. Voting when money and morals conflict: an experimental test of expressive voting. Journal of Public Economics 88, 1645–1664.Vaubel, R., 1996. Bureaucracy at the IMF and the World Bank: a comparison of the evidence. World Economy 19, 195–210.Veblen, T., 1899/1934. The Theory of the Leisure Class. Modern Library, New York NY.Weaver, C., 2008. The Hypocrisy Trap: The World Bank and the Poverty of Reform. Princeton University Press, Princeton NJ.Welzman, Y., 2010. The Persistence of Ineffective Aid. PhD dissertation, Bar-Ilan University.Wolfinger, R.E., Rosenstone, S.J., 1980. Who Votes? Yale University Press, New Haven CT.Xu, X., 2001. Group size and the private supply of a best-shot public good. European Journal of Political Economy 17, 897–904.Younas, J., 2008. Motivation for bilateral aid allocation: altruism or trade benefits. European Journal of Political Economy 24, 661–674.

Documents

Expressive behavior in economics and politics