Decision Making on Behalf of Others
John R. Hibbing Department of Political Science University of Nebraska-Lincoln
[email protected]
John R. Alford Department of Political Science Rice University
[email protected]
Paper presented at the annual meeting of the American Political Science Association, Washington, DC, September 1-4, 2005. Financial support for this project was provided by the Decision, Risk, Management Sciences as well as the Political Science Programs of the National Science Foundation (SES 0350387). −1−
Abstract In politics specifically and society generally people are often in a position where either they make decisions on behalf of someone else or someone else is making decisions on their behalf. But empirical social science research has provided little information on the extent to which decisions are different when they are made on behalf of others. In this paper, we draw on evolutionary theory to formulate a set of expectations regarding the differences between decisions people make for themselves and decisions they make for others (representation). Using laboratory experiments, we then provide preliminary tests of these hypotheses. The finding of most interest is that representatives are at least as careful of constituent resources as constituents are of their own resources. Though this result would seem to go against standard principal-agent theory, it makes perfect sense when seen in an evolutionary context. People care about what others think of them and this caring will lead representatives to pander to their constituents even if their position as representative is not hanging in the balance. Representatives’ solicitude vis-à-vis constituents may be due more to basic human psychology than to institutional structures such as elections.
−2−
Decision making is at the core of human behavior and therefore of the social sciences, for if we know the basis on which people make choices we will understand much about the human condition. Accordingly, research on decision making has been intense, with scholars in experimental economics, social psychology, organizational behavior, and neuroscience all providing substantial insights into the nature of decisions individuals make. The decisions these researchers typically study, however, are those people make on their own behalf. Perhaps an experimental subject is asked to divide with another player a small sum of money or perhaps the subject is asked to decide how much money to invest in a public goods communal account. But in the real world many decisions are made on behalf of others; in other words, people often are given responsibility to make decisions for someone else. In fact, decisions made on behalf of others occur in an incredible range of contexts and are often among the most interesting and important of all human decisions. CEOs make decisions on behalf of employees; pastors on behalf of their congregants; generals on behalf of their troops; teachers on behalf of their students; gang leaders on behalf of their followers; parents on behalf of their children; members of corporate boards on behalf of shareholders; lawyers on behalf of their clients; and one spouse on behalf of the other. To be sure, these situations differ from each other in important ways but they are united by the fact that a decision made by one individual implicates others for whom the decision maker feels some responsibility (as such, decisions made on behalf of someone else are different than simple third-party decisions).. Perhaps the clearest example of decision making on behalf of others occurs in politics where representational democracy calls for a representative to make decisions on behalf of constituents. In most political arenas, decisions are not made by directly by ordinary people but rather are made indirectly by representatives who have acquired responsibility for a group of −3−
constituents. The goal of our research project is to investigate the manner in which giving authority to another individual changes the perspective of both the representative and the represented.
Theory and Previous Research “A boy found two ripe apples as he was walking home from school with a friend. He kept the larger one for himself and gave the smaller one to his friend. ‘It wasn’t fair to keep the larger one for yourself,’ the friend replied. ‘What would you have done?’ the first boy asked. ‘I’d have given you the larger one and kept the smaller one for myself,’ said the friend. To which the first boy responded, ‘Well, we each got what you wanted, so what are you complaining about?’” Robert Frank, 1994: 21
Though this scenario does not entail a pure representational decision, it makes the important point that decisions we would make ourselves may not be the decisions we want others to make. The friend would have been quite pleased with the smaller apple if he had made the decision to give the large one to the other boy, but when he was given the smaller apple by someone who kept the larger one, his satisfaction plummeted. If this account is generalizable, representatives concerned for the happiness of their constituents should not make decisions by asking themselves, “What would my constituents do if this were up to them?” The very fact that it is not “up to them” means that they want something different than they would if they were in a position to decide for themselves. Rather, to please constituents, representatives must be more attentive to constituents than constituents would be to themselves. More importantly, assuming political representatives are aware of this fact, they will respond by being more protective of constituents’ interests than those constituents would be of their own. This means that the parochialism engaged in so frequently by representatives may not be attributable to a selfish desire to maintain their positions so much as to innate human psychology. Just to provide one −4−
example, national surveys in the United States in the early 2000s consistently reported that the public did not want tax cuts yet politicians rushed to embrace, enact, and take credit for tax cuts that they claimed the hard-working, long-deprived American people so richly deserved. Why would people acting as representatives be more solicitous of the interests of others than those “others” would be of their own interests? Much current thinking about representation is based on principal-agent theory, an extension of rational actor theory into representation. Principal-agent theory assumes the central problem of representation is how to bend agents acting in their own self-interest toward actions that reflect the interest of the principal (Kalt and Zupan, 1984; Goff and Grier, 1993; Rothenberg and Sanders, 2000). As such, the problem typically anticipated in principal-agent relationships is that agents will fail to act with sufficient concern for the interests of principals. Institutional structures, it is believed, must be constructed in order to prevent shirking of this sort. But our expectation is just the opposite of this wellknown theory and we believe the greater danger with agency comes not from shirking but from pandering. Principals do not want agents to make the same decisions the principals would make if they were calling the shots. They want more. Agents realize this and, unlike the boy who found the apples, make decisions accordingly. What theory could account for such behavior? The theory of evolutionary biology. As increasingly applied to social and political science (see Orbell et al., 2004; Alford and Hibbing, 2004), this theory stresses the centrality of group life for human provisioning, procreation, recreation, and protection from predators. Given the usefulness of groups, it is not surprising that humans have developed genetic, neurological, and behavioral machinery to both nurture groups and to protect their own status within groups (see Tooby and Cosmides, 1992; Sober and Wilson, 1998; Ostrom, 1998). Part of this machinery is an acute individual sensitivity −5−
to status and reputation within the group and this sensitivity makes people eager to avoid both being played for a sucker (a sign that status within the group is not what it should be) and being perceived as a leech (a reputation that is certain to harm status within the group). As such, this theory accounts for why the boy would have been pleased with the smaller apple if the decision had been his but was displeased with the smaller apple when the decision was made by his friend. The size of the apple received is not the real issue; the way other people view us and treat us is—and there are perfectly sound evolutionary reasons for this relentless concern for our place within the group as well as with the sociability of fellow group members (especially those with resources). People want to be known as thoughtful toward others but they do not want others to take advantage of them (Hibbing and Alford, 2004). So, evolutionary theory when applied to a principal-agent scenario such as we see in the relationship between elected representatives and their constituents predicts 1.) that principals will want an agent to dote on them more than the principals would dote on themselves if they had the power directly and 2.) that agents will in turn dote on principals more than agents would dote on themselves. If these predictions should turn out to be accurate, they would constitute an interesting and important inversion of the traditional maximizing concepts of decision making (see Buchanan and Tullock, 1962: 18) that predict self-interested behavior will be most evident when people are making decisions on their own behalf. Given the potentially important implications of evolutionary theory for principal-agency specifically and for representative democracy generally, we intend to analyze people’s behavior in circumstances where they are deciding on someone else’s behalf. Certainly there is overwhelming evidence that individuals playing economic games are incredibly sensitive to “others” involved. If the others are computers, people play (and brains −6−
operate) differently than if the others are human beings (McCabe et al., 2001). If the others are anonymous abstractions, people play differently than if they are personified, even if this personification consists of nothing more than cartoonish eye-spots placed at the top of a computer screen (Haley and Fessler, 2005; more generally, see Sullivan and Masters, 1988; Masters and Sullivan, 1989; Dawes, van de Kragt, and Orbell, 1990; Blount, 1995). If subjects know that others will learn of their decisions, people play differently (Hoffman, McCabe, Shachat, and Smith, 2000; van Dijk and Vermunt, 2000; Larimer, 2003). If the others are believed to have intentionally rather than accidentally made an ungenerous decision, people behave differently (Hibbing and Alford, 2004). And if the others are believed to be ambitious, people behave differently toward them even when the objective decisions are the same (Smith, Larimer, Littvay, and Hibbing, 2004). What we do not know is if people behave differently when deciding on their own behalf as opposed to when deciding on behalf of others and if evolutionary theory accurately predicts any differences that do exist. People seem sensitive to others but this does not mean they are more sensitive to those in their charge than they are to themselves. To test this possibility we need results from parallel decision environments, one documenting traditional personal decisions and the other documenting agents’ decisions on behalf of principals, and this is where existing research comes up short. Students of the decision sciences have conducted useful analyses comparing self decisions with the advice people tend to give others (Kray, 2000; Jonas and Frey, 2003), comparing self decisions to decisions made in groups (Robert and Carvenale, 1997; Messick, Moore, and Bazerman, 1997; Morgan and Tindale, 2003), and comparing self decisions to neutral third-party decisions (Sniezek and Van Swol, 2001; Fehr and Gachter, 2002). But −7−
representational decision making is not the same as advisor-based decision making, group-based decision making, or judge-based decision making. Representatives are not merely members of a group faced with a collective decision; unlike judges, representatives have an attachment to a particular principal or group of principals; and unlike advisors, representatives actually make the final decision. So the empirical differences attending typical “self” decisions and representational decisions are still undetermined. Political theorists have devoted great attention to issues surrounding the representational relationship. Much of the early attention was directed at the normative issue of the manner in which representatives should approach their assignment. Theorists such as John Stuart Mill (1975 [1861]) and Jean-Jacques Rousseau (1946 [1762]) thought that, if people could not act on their on behalf, then representatives should at least try to do exactly what the represented would have done (the so-called ‘instructed delegate’ model). Edmund Burke (1949 [1775-7]), on the other hand, thought representatives should behave as ‘trustees,’ a role in which they act in the interest of the represented even as these constituents may have expressed a different preference. Pitkin (1972) provides a masterful listing and analysis of the many variations on the concept of representation. And modern empirical political scientists have conducted extensive research on the extent to which the decisions of representatives are congruent with the preferences of their constituents or principals. The classic work in this vein was undertaken by Warren Miller and Donald Stokes (1963) and they found that policy congruence was present for some salient issues such as race and, to a lesser extent, social welfare, but not for less salient issues such as foreign policy. Informative empirical studies of political representation have been conducted by Matthews, 1960; Froman, 1963; Sullivan and O’Connor, 1972; Clausen, 1973; Kingdon, 1973; −8−
Fiorina, 1974; Kuklinski and Elling, 1977; Erikson, 1978; Monroe, 1979; McCrone and Kuklinski, 1979; Erikson and Wright, 1980; Bianco, 1984; Page, Shapiro, Gronke, and Rosenberg, 1984; Wright, 1989; Bartels, 1991; Poole and Romer, 1993; Oppenheimer, 1996; Alvarez and Gronke, 1996; Bailey and Brady, 1998; Lee and Oppenheimer, 1999; Hill and Hurley, 1999; Bishin, 2000; Erikson and Wright, 2001; and Hurley and Hill, 2003, just to name a few. Thanks to this research, scholars now possess a deep understanding of the conditions under which policy congruence is more or less likely to be present in representational democracy. Policy congruence has not been the sole focus of this empirical work. Innovative and oftcited research by Kingdon (1973), Mayhew (1974), Fiorina (1977), and Fenno (1978) provides evidence of the importance people attach to non-policy matters. Taking positions, explaining actions, servicing constituents, and generally projecting a style of operation with which constituents are comfortable all are vital components of political representation, at least in the case of the United States Congress where most of the research has been concentrated. Acting on behalf of others apparently includes not just the act (roll call vote) itself but also the process of decision making, the style of the decision maker, and the manner in which that decision is justified. As helpful as they may be, these normative observations and empirical findings on representation deal only with decisions made on behalf of others. The question we propose is not do (or should) those acting on behalf of others actually follow the dictates of those to whom they are responsible, but rather do people acting on behalf of others behave differently than they would if their actions only affected themselves and, if so, how. As specified above, our theoretical expectation is that representatives will be more careful of their constituents’ resources than constituents are of their own. Eventually, we also want to know about the desires of constituents and we expect that, like the boy and the apple, people −9−
have different preference structures and reactions when others are deciding for them than when they are deciding for themselves. An outcome deemed acceptable when constituents decided it themselves will frequently be deemed unacceptable when it is the result of a representative’s decision. Finally, we hope to determine whether various institutional structures push decision making on behalf of others farther from or closer to decisions people make on behalf of themselves. Consistent with evolutionary theory but in contrast to most political science assumptions about the importance of institutions, we believe securing constituent-sensitive behavior is not contingent upon giving representatives institutional (usually electoral) incentives. In other words, we believe the effects of incorporating elections into the representational relationship will be surprisingly minimal since whenever representatives know they are deciding on behalf of a real person (or real people) they will be protective of constituent resources.
Research Design In order to measure the behavioral differences created when decisions are given to representatives, we need to construct decision situations that are as similar as possible except for the fact that in one the decision maker is deciding on his/her own behalf and in the other the decision is being made for others. Such requirements demand an experimental setting and that is what we employ here. Our long-term goal is to compare people’s behavior when they make decisions... 1. 2. 3. 4.
on their own behalf on behalf of an anonymous “other” on behalf of an “other” they have met personally on behalf of an “other” they have met personally and who has selected the decision maker to act on their behalf
−10−
5. and on behalf of an “other” they have met personally, who has selected the decision maker to act on their behalf, and who will have the opportunity to reelect them (or not), in light of the decisions that have been made. And we would also like to have data on the reaction of constituents to representational decisions in each of these situations (except for #1 where there is no constituent). It takes two to tango and at least two for there to be a representational relationship so we would like to obtain information on both the behavior of the representative and the reaction of the constituents. At this early stage of the project, however, we have data on only the first three experimental cells depicted above and only for representational behavior. Information on cells 4 and 5 as well as on constituent reactions in cells 2-5 will be obtained through additional experiments over the course of the next few months. In the late spring and summer of 2005, 92 experimental subjects were recruited and brought to Rice University’s Behavioral Research Lab, an NSF-funded facility with approximately 30 computer terminals, a waiting room, a debriefing room, and adequate parking. Most of the subjects were Rice students, meaning their average age is somewhat lower than would be the case in a national sample just as the average family income is substantially higher (basic demographic information is provided in Table 1). But since subjects were randomly assigned to one of the three cells and since our interest is in behavioral differences across the three cells, the typicality of subjects is not a pressing concern. Once at the lab, subjects were assigned either to make decisions on their own behalf, to make decisions on behalf of an unidentified other individual or to make decisions on behalf of an individual they had just met after signing the experimental consent form but before beginning the experiment. Each subject played a version of the dictator game. In this game, subjects are given a sum of money, in this case $10, and told to divide it in any fashion desired with an always −11−
unspecified other player (for the sake of simplicity only whole dollar divisions were permitted). The dividing player is fully aware that the receiving player has no recourse and this is what makes the experimental subjects dictators of a sort. The central variable in our study, then, is the extent to which subjects in cell 1 keep a significant portion of the $10 for themselves and the extent to which subjects in cells 2 and 3 keep a significant portion of the $10 for their constituent. Subjects in cells 2 and 3 were told that they would receive a set fee for serving as the representative of another person. Thus, instead of needing to decide how much of the $10 they would keep for themselves, they needed to decide on behalf of another person how much of the $10 that “constituent” would keep and how much would be passed along to the unspecified other player. Respondents in both cells 2 and 3 were acting as representatives. The difference is that those in cell 2 were representing an individual they had never met and who was referred to in the experiment only by initials while those in cell 3 were representing an individual they had met just a few moments before the beginning of the experiment and who they believed to be occupying an adjacent computer terminal during the experiment. Whereas principal-agent theory, representational theory, and conventional wisdom all predict that, lacking an institutional incentive to behave differently, subjects will keep more money when they are playing for themselves, our expectation derived from evolutionary theory is that the subject as representative will be more protective of constituent interests than people are protective of their own interests and that representatives will become more protective as social contact with the constituent increases.
−12−
Preliminary Results In most divide-the-dollar games of this sort, subjects display a strong attraction to equal divisions and our results are no different in this regard. 47 of the 92 respondents opted for a $5$5 division. In light of this fact, perhaps the most appropriate way of presenting the results is to compare the percent of subjects keeping more than half (as opposed to half) of the $10 (one subject actually gave away more than half and will be combined with those subjects who split the money equally). When this is done we obtain the results presented in Figure 1 and Table 2. [Figure 1 and Table 2 about here] Those subjects assigned to play the standard dictator game - that is, those assigned to make decisions for themselves - were reasonably generous, with nearly 54 percent dividing the money equally and just 46 percent keeping more than half for themselves (see the bars on the left of Figure 1). Consistent with countless other experiments employing economic games (see Ostrom, 1998; Camerer and Loewenstein, 2004, for a good summaries), people tend to be less maximizing, even in anonymous one-shot games, than standard microeconomic theory predicts. But our central interest is in comparing the decisions people make on their own behalf with those they make on behalf of others and here we see some fascinating contrasts. When the constituent is not made tangible or given any social reality (as is true of the situations producing the results for the middle bars of Figure 1), subjects serving as representatives are much less likely to keep more than half of the money for their constituent than was the case for subjects deciding on their own behalf. Only 26 percent of respondents kept more than half for an anonymous constituent. This is not what we had expected and is in fact the reverse of our expectations. Rather than being less likely to offer a selfless $5-$5 split, subjects are actually more likely to offer an −13−
equal division (53.8% in the ‘self’ condition versus 73.3% in the “anonymous constituent” condition. Likewise, the proportion of subjects keeping more than two-thirds of the money for themselves or their constituent drops from 42.3% in the ‘self’ condition to only 20.1% in the “anonymous constituent” condition (see Table 2). In effect, the pattern of a rough balance between equality and selfishness when subjects decide for themselves shifts to a pattern that decidedly favors equality when subjects decide on behalf of an anonymous constituent. But any inclination for representatives to be less generous to constituents than people are to themselves reverses if the representative and constituent have met. When the representative has a face to attach to the constituent, representatives, as we expected, become much more likely to keep more than half of the money for their constituents. Over 60 percent of the subjects in this category awarded more than half of the $10 to their constituents (see the bars on the right-hand side of Figure 1), and the proportion of subjects keeping more than two-thirds of the money for their constituent jumps from 20.1% in the ‘anonymous constituent’ condition to 43.7% in the ‘known constituent’ condition (nearly identical to the 42.3% for the ‘self’ condition). Interestingly, even though subjects deciding on behalf of a known constituent are more likely to keep over half of the money than are subjects deciding for themselves, if we calculate the mean amount kept, this difference vanishes (both keep an average of about $7 while the mean amount of money kept for an anonymous other is only $5.70). This is due to the fact that several of those subjects deciding on behalf of a known constituent awarded more than half to that constituent but barely more than half ($6 or $7). Conversely, when those respondents deciding for themselves kept more than half they almost always (just two exceptions) went the entire way and kept all $10. As a result, the averages are very similar even as the percent keeping more than half is different. Taken together this suggests that people are at least as −14−
generous to known constituents as they are to themselves and much more generous to known constituents than they are to anonymous constituents. How can we can we reconcile our original expectations with these results that appear at least to partly contradict them? The key is to understand that the self-interest of the representative is not the only threat to the interests of the constituent. As we mentioned above, the results in the ‘self’ condition are clearly incompatible with a strict self-interest explanation. What they suggest instead is competing concerns for self-interest and for the interest of others. While the balance of these concerns undoubtedly varies substantially across individuals, the general result is that people are far less self-interested than rational choice assumptions would lead us to predict. When we shift to a subject deciding for an anonymous constituent, the pressure of direct self-interest has been removed by explicitly separating their compensation from their division decision. This has been replaced by an instructed obligation to a constituent, but the concern for others that was working in competition to self-interest is still present. When the link to the constituent is as weak as it is in this anonymous condition, the result is something similar to anonymous third-party decision making, with concerns for all the affected parties largely overpowering the obligation to the constituent. Shifting to a known constituent, however, while only a relatively modest change in institutional terms, dramatically alters the balance of concerns. Concern for others is still a force in evidence, but for the first time an equal division of the funds is no longer a majority result. Instead, obligations felt for the constituent appear to trump general concern for others.
−15−
Discussion and Conclusion The results presented here are admittedly preliminary. However, if they should hold up with larger numbers of subjects and improved procedures, as we think they will, the implications could be important. The case of representation when both the representative and the constituents are anonymous does not have many real world applications and is valuable mostly as a point of comparison with other, more realistic scenarios. Situations in which a representative makes decisions on behalf of constituents who are completely unknown to them and, more to the point, in which a representative is completely unknown to constituents are rarely found. Certainly in the context of legislative institutions, the most salient political example of individuals deciding for other individuals, representatives have always met some of their constituents. Representatives can connect real faces to the likely impact of decisions they must make. Obviously, they have not met every single constituent but that is not necessary for there to be an effect of social contact on decisions. Outside of experimental games, it is extremely rare for decisions to be made in a double-blind context. Typically, people know or can find out who their representative is and representatives know some constituents. The more interesting comparison is between the level of self-interest on display in situations that form a rough parallel with, first, direct democracy (the bars on the left in Figure 1) and, second, indirect democracy where the representative has met a constituent (the bars on the right in Figure 1). Certainly, the situation we have created in the lab, with just one constituent and just one decision to be made, is a simplification of political reality, but if we hope to compare decisions made for a constituent with decisions made for oneself, these simplifications are necessary. We can think of no reason why a representative would be concerned with the interest of one constituent but not two. −16−
Our key finding, provisional as it may be, is that constituents have to be concerned about more than controlling the self-interest of their representatives. They must also contend with the equal strong altruism of their representatives. As the results for subjects representing anonymous constituents show, removing self-interest alone does not insure adherence to the interests of the constituent. Concern for the broader good is at least as potent a contender against constituent interest as it is against self-interest. Making the constituent a human reality through even minimal contact restores the balance of forces and may even give constituents a leg up on the self-interest of the representative. What we expect to see as we add more institutional features to the representational connection is that the balance will tip more decidedly against concern for others and toward concern for the self-interest of constituents. If we prove correct, people’s self-interest may be better served when they allow representatives to make decisions for them than when they make decisions themselves. That is, representatives may be more protective of their constituents’ interests than the constituents would be if they were making decisions directly. At one level, this sounds completely counterintuitive. Why wouldn’t people themselves be in the best position to look out for their own interests? Much of the governmental reform movement is motivated by the belief that people should make decisions themselves rather than allow them to be made by elected officials, and those people who defend representative as opposed to direct democracy usually fall back to the claim that ordinary people are incapable of making informed political decisions probably because they do not have the motivation or ability needed to collect and synthesize information on political issues of the day. Our anticipated results would suggest there is another important defense of representative democracy that is substantially less elitist: people are more willing to protect the interest of −17−
someone they represent than they are to protect the interest of themselves. If this is true, there are advantages to having someone else decide political issues not because that “someone else” is smarter or more politically informed but simply because that person is someone else. The extreme social sensitivities of most individuals lead them to want to treat others well. When making decisions for themselves, this works in opposition to their self-interest. If they have been assigned to make decisions on behalf of another, this can be recruited to form the basis of a bond to the constituent’s interest, but only if care is taken to recognize that it is a two-edged sword when other similarly situated individuals are potential beneficiaries. We have not yet conducted our planned experiments on the effects of election and reelection in stimulating representatives to protect the interests of constituents, but it is notable that the tendency to protect appears to be in evidence even in the absence of electoral and other institutional mechanisms. All it takes is for the representative to realize that the constituent is a real person. Representatives protect the interests of constituents at least as much as constituents protect their own interests. With this kind of social concern operating, it is quite likely that representatives would act with great sensitivity to the interests of their constituents even without electoral attachments. Long ago, Joseph Schlesinger stated that “the desire for election and, more important, for re-election becomes the electorate’s restraint upon its public officials. No more irresponsible government is imaginable than one of high-minded men unconcerned for their political fortunes” (1966: 2). Our preliminary experimental findings—as well as the absence of evidence consistent with Schlesinger’s contention, lead us to question it. Politicians, like virtually all people, are eager to please, ambition or not—elections or not. This may be why even after representatives have decided to retire from political office, their behavior changes only minimally (Hibbing, 1982: Chap. 4). −18−
Our findings should not be taken to mean indirect democracy is preferable. Do we want government to be so eager to accommodate the self-interest of constituents? Just because politicians act in the interest of the people does not mean these actions are beneficial to the larger polity. Pandering could easily result when representatives make decisions on behalf of constituents. Our research is not going to identify the policies that the government should enact or the individual interests that should be advanced. But it does allow us to say that people who make decisions on behalf of someone else have a natural and empirically demonstrable inclination to be solicitous of the interests of others—in fact, at least as solicitous of constituents’ interests as people are of their own interests when deciding for themselves. Conventional assumptions and mid-level theory may be off-base in taking it as given that representatives are motivated largely by, and best controlled by, concerns for their own selfinterest. We see evidence from our experiments that the interests of others are an equally powerful force. Thanks to humans’ remarkably social nature, even without elections representative democracy protects the interests of the represented. The only question is whether it elevates the self-interest of constituents to a status it does not warrant—a status that, if direct democracy were in place, the people themselves would not grant their own self-interest.
−19−
Percentage
FIGURE 1
100 90 80 70 60 50 40 30 20 10 0
Equal Division More for Self or Constituent
Self
Anonymous Known Constituent Constituent
Equal Division
53.8
73.3
37.5
More for Self or Constituent
46.2
26.7
62.5
Game Type
Difference of Means Tests:
sig. (2-tailed)
Self v. Anonymous Constituent (53.8 v. 73.3)
.134
Self v. Known Constituent (53.8 v. 37.53)
.220
Self v. Both ‘Other’ Conditions (53.8 v. 54.8)
.993
Anonymous v. Known Constituent (73.3 v. 37.5)
.004
−20−
TABLE 2
Division $10-$0 $9-$1 $8-$2 $7-$3 $6-$4 $5-$5 $2-$8 n
Self 38.5% 0.0% 3.8% 0.0% 3.8% 53.8% 0.0% 26
Anonymous Constituent 6.7% 6.7% 0.0% 6.7% 6.7% 70.0% 3.3% 30
Known Constituent 25.0% 3.1% 3.1% 12.5% 18.8% 37.5% 0.0% 32
REFERENCES Alford, John R., and John R. Hibbing. 2004. “The Origin of Politics: An Evolutionary Theory of Behavior.” Perspectives on Politics 2 (December): 707-724. Alvarez, R. Michael, and Paul Gronke. 1996. AConstituents and Legislators: Learning about the Persian Gulf Resolution.@ Legislative Studies Quarterly 21: 105-27. Bailey, Michael, and David W. Brady. 1998. AHeterogeneity and Representation: The Senate and Free Trade.@ American Journal of Political Science 42: 524-44, Bartels, Larry M. 1991. AConstituency Opinion and Congressional Policy Making.@ American Political Science Review 85: 457-74. Bianco, William T. 1994. Trust: Representatives and Constituents. Ann Arbor, Mich.: University of Michigan Press. Bishin, Benjamin G. 2000. AConstituency Influence in Congress.@ Legislative Studies Quarterly 25: 389-416. Blount, Sally. 1995. AWhen Social Outcomes Aren=t Fair: The Effects of Causal Attribution on Preferences.@ Organizational Behavior and Human Decision Processes 63: 131-44. Buchanan, James M., and Gordon Tullock. 1962. The Calculus of Consent: Logical Foundation of Constitutional Democracy. Ann Arbor, Mich.: University of Michigan Press. Burke, Edmund. 1949 [1755-77]. Burke=s Politics. Ross J.S. Hoffman and Paul Levack, eds. New York: Alfred Knopf. Camerer, Colin F., and George Loewenstein. 2004. “Behavioral Economics: Past, Present, Future.” In Colin F. Camerer, George Loewenstein, and Matthew Rabin, eds. Advances in Behavioral Economics. New York: Russell Sage Foundation. Clausen, Aage. 1973. How Congressmen Decide. New York: St. Martins. Dawes, Robyn M.; Alphons J.C. van de Kraagt; and John M. Orbell. 1990. ACooperation for the Benefit of UsBNot Me, or My Conscience.@ In Jane J. Mansbridge, ed. Beyond Self-Interest. Chicago: University of Chicago Press. Erikson, Robert S. 1978. AConstituency Opinion and Congressional Behavior.@ American Journal of Political Science 22: 511-35. Erikson, Robert S., and Gerald C. Wright. 1980. APolicy Representation of Constituent Interests.@ Political Behavior 2: 91-106.
Erikson, Robert S., and Gerald C. Wright. 2001. AVoters, Candidates, and Issues in Congressional Elections.@ In Congress Reconsidered (Seventh Edition). Lawrence Dodd and Bruce Oppenheimer, eds. Washington: CQ Press. Fehr, Ernst, and Simon Gachter. 2002. AAltruistic Punishment in Humans.@ Nature 415: 137-40. Fenno, Richard F., Jr. 1978. Home Style. Boston: Little Brown. Fiorina, Morris P. 1974. Representatives, Roll Calls and Constituencies. Lexington, Mass.: Lexington Books. Fiorina, Morris P. 1977. Congress: Keystone of the Washington Establishment. New Haven, Conn.: Yale University Press. Frank, Robert H. 1994. “Local Status, Fairness, and Wage Compression Revisited,” unpublished manuscript, Department of Economics, Cornell University. Froman, Lewis A., Jr. 1963. Congressmen and their Constituencies. Chicago: Rand McNally. Goff, Brian L., and Kevin B. Grier. 1993. AOn the (Mis)Measurement of Legisltor Ideology and Shirking.@ Public Choice 76: 5-20. Haley, Kevin J., and Daniel M.T. Fessler. 2005. “Nobody’s Watching: Subtle Cues Affect Generosity in Anonymous Economic Games.” Evolution and Human Behavior 26 (3): 245-56.. Hibbing, John R. 1982. Choosing to Leave: Voluntary Retirement from the US House of Representatives. Washington DC: University Press of America. Hibbing, John R., and John R. Alford. 2004. AAccepting Authoritative Decisions: Humans as Wary Cooperators.@ American Journal of Political Science 48: forthcoming. Hill, Kim Quaile, and Patricia A. Hurley. 1999. ADyadic Representation Reappraised.@ American Journal of Political Science 43: 109-37. Hoffman, Elizabeth; Kevin McCabe; K. Shachat; and Vernon L. Smith. 1994. APreferences, Property Rights, and Anonymity in Bargaining Games. Games and Economic Behavior 7: 346-80. Hurley, Patricia A., and Kim Quaile Hill. 2003. ABeyond the Demand-Input Model: A Theory of Representational Linages.@ Journal of Politics 65: 304-26.
Jonas, Eva, and Dieter Frey. 2003. “Information Search and Presentation in AdvisorClient Interactions.” Organizational Behavior and Human Decision Processes 91: 15468. Kalt, Joseph P., and Mark A. Zupan. 1984. ACapture and Ideology in the Economic Theory of Politics.@ American Economic Review 74: 279-300. Kingdon, John W. 1973. Congressmen=s Voting Decisions. New York: Harper and Row. Kray, Laura J. 2000. “Contingent Weighting in Self-Other Decision Making.” Organizational Behavior and Human Decision Processes 83 (September): 82-106. Kuklinski, James H., and Richard Elling. 1977. ARepresentational Role, Constituency Opinion, and Legislative Roll Call Behavior.@ American Journal of Political Science 21: 135-47. Larimer, Christopher W. 2003. ABehavior as a Function of Publicity.@ Paper presented at the annual meeting of the Human Behavior and Evolution Society, Lincoln, Neb., June 2003. Lee, Francis E., and Bruce I. Oppenheimer. 1999. Sizing Up the Senate. Chicago: University of Chicago Press. Masters, Roger D., and Dennis G. Sullivan. 1989. ANonverbal Displays and Political Leadership.@ Political Behavior 11: 121-53. Matthews, Donald R. 1960. U.S. Senators and their World. Chapel Hill, NC: University of North Carolina Press. Mayhew, David R. 1974. Congress: The Electoral Connection. New Haven, Conn.: Yale University Press. McCabe, Kevin, Daniel Houser, Lee Ryan, Vernon Smith, and Theodore Trouard. 2001. “A Functional Imaging Study of Cooperation in Two-Person Reciprocal Exchange.” Proceedings of the National Academy of Science 98 (25 September): 11832-35. McCrone, Donald R., and James H. Kuklinski. 1979. AThe Delegate Theory of Representation.@ American Journal of Political Science 23: 278-300. Messick, David M., Don A. Moore, and Max H. Bazerman. 1997. “Ultimatum Bargaining with a Group: Underestimating the Importance of the Decision Rule.” Organizational Behavior and Human Decision Processes 69 (February): 87-107. Mill, John Stuart. 1975 [1861]. Three Essays. Richard Wolheim, ed. London: Oxford University Press.
Miller, Warren E., and Donald E. Stokes. 1963. AConstituency Influence in Congress.@ American Political Science Review 57: 45-56. Monroe, Alan D. 1979. AConsistency between Public Preferences and national Policy Decisions.@ American Politics Quarterly 7: 3-19. Morgan, Pauline M., and R. Scott Tindale. 2002. “Group vs. Individual Performance in Mixed-Motive Situations: Exploring an Inconsistency.” Organizational Behavior and Human Decision Processes 87 (January): 44-65. Oppenheimer, Bruce I. 1996. AThe Representational Experience.@ American Journal of Political Science 40: 1280-99. Orbell, John M., Tomonori Morikawa, Jason Hartwig, James Hanley, and Nicholas Allen. 2004. “’Machiavellian’ Intelligence as a Basis for the Evolution of Cooperative Dispositions.” American Political Science Review 98 (February): 1-16. Ostrom, Elinor. 1998. “A Behavioral Approach to the Rational Choice Theory of Collective Action.” American Political Science Review 92 (1): 1-22. Page, Benjamin; Robert Y. Shapiro; Paul W. Gronke; and Robert M. Rosenberg. 1984. AConstituency, Party, and Representation in Congress.@ Public Opinion Quarterly 48: 741-56. Pitkin, Hannah. 1972. The Concept of Representation. Berkeley, Cal.: University of California Press. Poole, Keith T., and Thomas Romer. 1993. AIdeology >Shirking= and Representation.@ Public Choice 77: 185-96. Robert, Christopher, and Peter J. Carnevale. 1997. “Group Choice in Ultimatum Bargaining.” Organizational Behavior and Human Decision Processes 72 (November): 256-79. Rothenberg, Lawrence S., and Mitchell S. Sanders. 2000. ASevering the Electoral Connection: Shirking in the Contemporary Congress.@ American Journal of Political Science 44:316-25. Rousseau, Jean Jacques. 1946 [1762]. The Social Contract and Discourses. G.D.H. Cole, ed. and translator. New York: E.P. Dutton. Schlesinger, Joseph A. 1966. Ambition and Politics. Chicago: Rand McNally.
Smith, Kevin B., Christopher W. Larimer, Levent Littvay, and John R. Hibbing. 2004. “‘Big Men’ Behaving Badly: Getting People to Tolerate Unfavorable Outcomes.” Paper presented at the annual meeting of the Midwest Political Science Association, Chicago, April 2004. Sniezek, Janet A., and Lyn M. Van Swol. 2001. “Trust, Confidence, and Expertise in a Judge-Advisor System.” Organizational Behavior and Human Decision Processes 84 (March): 288-307. Sober, Elliott, and David Sloan Wilson. 1998. Unto Others: The Evolution and Psychology of Unselfish Behavior. Cambridge, Mass.: Harvard University Press. Sullivan, Dennis G., and Roger D. Masters. 1988. AHappy Warriors: Leaders= Facial Displays, Viewers= Emotions and Political Support.@ American Journal of Political Science 32: 345-68. Sullivan, John L., and Robert E. O=Connor. 1972. AElectoral Choice and Popular Control of Public Policy.@ American Political Science Review 66: 1256-68. Tooby, John, and Leda Cosmides. 1992. “The Psychological Foundations of Culture.” In Jerome H. Barkow, Leda Cosmides, and John Tooby, eds. The Adapted Mind: Evolutionary Psychology and the Development of Culture. New York: Oxford University Press. Van Dijk, Eric, and Riel Vermunt. 2000. AStrategy and Fairness in Social Decision Making: Sometimes It Pays to be Powerless.@ Journal of Experimental Social Psychology 36: 1-25. Wright, Gerald C. 1989. APolicy Positions in the Senate: Who Is Represented?@ Legislative Studies Quarterly 14: 465-86.
