|
COPYRIGHT 1996 University of Pennsylvania, Law School
As now, when here's the fixed Assembly Day, And morning come, and no one in the Pnyx. There in the Agora chattering, up and down Scurrying to dodge the cord dripping red
--Aristophanes, The Acharnians(1)
INTRODUCTION
In ancient Athens, election officials corralled voters with a red-dyed rope, herding them from the marketplace to the Assembly's voting area at the nearby Pnyx.(2) Athenian officials in later years simply paid voters to attend the Assembly's In contemporary Italy, those who fail to vote face the prospect of having their names posted by the mayor on the communal notice wall and of being branded a nonvoter in official papers.(4) In (where else but?) California, a voting stub obtained after casting a ballot has entitled voters to a free half-dozen "Yum-Yum" doughnuts or a discounted spinal adjustment by a chiropractor.(5)
Carrots and sticks have been employed to increase voter turnout since the birth of democracy, in reaction to what rational choice theorists have termed "the paradox of voting":(6) given the infinitesimal chance that one's own vote could affect the outcome of most elections or the stability of the electoral system, it often appears rational to abstain from voting. The paradox facing rational choice scholars is that many people do vote in the absence of visible carrots or sticks, although not in the same numbers in comparable elections or across state or national boundaries.
Social norms may solve the paradox of voting in a meaningful way. A social norm could induce voting through (1) social sanctions that raise the costs of nonvoting; (2) social rewards that raise the benefits of voting; and (3) internalization of a norm of voting. The norm of voting overcomes at least two collective action problems: First, society is better off if all (or at least many) people vote because a large turnout legitimates democratic government, which is itself a public good. Second, a group is better off if all of its members vote in a bloc for a particular candidate or party because politicians offer groups rents or other advantages in exchange for bloc votes.(7) Despite these benefits, voting has (at least opportunity) costs and each person's individual vote alone has a negligible effect on the provision of these public goods. The norm of voting overcomes the apparent irrationality of voting, thereby facilitating the provision of these public goods.
This Article examines the plausibility and implications of a norm-based explanation for voting. Part I reviews rational choice models for voting, contrasts the rational choice models with the social norms hypothesis, and examines the empirical evidence regarding the existence of a norm of voting. As Part I demonstrates, rational choice explanations have offered only a tautological explanation of why people vote: People vote when the psychic benefits of voting exceed its costs.(8) Unlike the rational choice explanation, a norm-based explanation of voting can explain plausibly why some people vote, as well as explain aggregate changes in voter turnout over time. Although the norms hypothesis is plausible, evidence supporting the hypothesis is sketchy and may be consistent with alternative explanations for voting. The analysis in Part I illustrates a general proposition that norm-based explanations are about as easy to conjure up as they are difficult to prove.
Assuming that a norm of voting has served to overcome collective action problems for only certain groups in the United States, and assuming that the norm has eroded over time even among these groups because of a decrease in social connectedness, arguably the state should take on the role of social sanctioner of last resort through a compulsory voting law. Part II of this Article, building upon Ellickson's pathbreaking work, Order Without Law,(9) examines the substitutability of state- and societal-based mechanisms for social control in the voting context. In particular, this Part considers whether compulsory voting laws could serve as a good substitute for a norm of voting.
Part II demonstrates that state- and societal-based methods of control are not always substitutable. Enactment of a compulsory voting law in the United States, even if desirable as a method of overcoming collective action problems, and even if proven effective as a means of increasing turnout in other states, is unlikely to occur because of a widely held libertarian belief against government interference in the decision to vote.
I. USING SOCIAL NORMS TO SOLVE THE PARADOX OF VOTING
A. Like Voting for Chocolate: The Vacuity of the Rational Choice Voting Model
The paradox of voting has proven to be something of a "roach motel" for rational choice theorists: entering the debate has proven much easier than emerging from it unscathed. Indeed, opponents of rational choice theory have pointed to the paradox as indisputable proof of the theory's failure to explain political phenomena generally.(10) This is not the place to examine whether the paradox is symptomatic of general "pathologies" of rational choice theory;(11) rather, I focus here on various rational choice attempts to explain the paradox of voting.
Although he never used the phrase, Anthony Downs is credited with first recognizing the paradox in his classic work, An Economic Theory of Democracy.(12) Downs made the following assumptions necessary to the paradox: (1) there are costs associated with acquiring information about whom to vote for(13), and, more importantly, with voting itself;(14). (2) each citizen derives benefits from living in a democracy, and those benefits depend upon a sufficient number of people voting;(15) and (3) each voter "will actually get this reward even if he himself does not vote as long as a sufficient number of other citizens do."(16)
Downs's solution to the paradox rested on what is now widely recognized as faulty rational choice logic: He argued that the rational voter "is willing to bear certain short-run costs he could avoid in order to do his share in providing long-run benefits."(17) As Brian Barry has explained, Downs's reasoning
requires our citizen to reason that since the benefits he gets
depends on the efforts of others, he should contribute too. This
may be good ethics, but it is not consistent with the assumptions
of the model, which require the citizen to compute the advantage
that accrues to him from his doing x rather than y; not the
advantage that would accrue to him from himself and others doing
x rather than y, unless, of course, his doing it is a necessary and
sufficient condition of the others doing it.(18)
On this reasoning, even a pure altruist in the conventional rational choice model would conclude that it is irrational to vote.(19)
Riker and Ordeshook addressed the paradox using a model consistent with rational choice theory, but virtually devoid of explanatory power. In their model, a citizen chooses to vote rather than abstain whenever:
pB- C+D > 0,(20)
where B is "the differential benefit, in utiles, that an individual voter receives from the success of his more preferred candidate over his less preferred one,"(21) p is the probability that a citizen will bring about B by voting, where [less than or equal to] p [less than or equal to] 1,(22) C is "the cost to the individual of the act of voting,"(23) and D is the utility of other psychic benefits of voting, including "the satisfaction from compliance with the ethic of voting."(24)
Recognizing that in most elections p is close to zero,(25) the equation reduces to the expression that people vote when
D >C,
or when the psychic benefits of voting exceed the costs.(26) As evidence supporting their hypothesis, Riker and Ordeshook pointed to the high correlation between voting and a sense of "citizen duty," as measured by Election Day polling conducted by the Survey Research Center at the University of Michigan.(27)
Gordon Tullock and others have concluded from the body of work following Riker and Ordeshook that rational choice theory has very little to say about why people vote. Tullock has characterized voting as an act of consumption, driven by a taste for voting exogenous to the economic model, much like the taste for chocolate.(28) Although a consumption explanation can tell us about voting on the margin (for example, the amount of voting goes down, like the demand for chocolate, when its price relative to other goods goes up),(29) it tells us nothing about who has this taste, where it is acquired, or whether the taste for voting changes over time.(30) As Barry has remarked, Riker and Ordeshook's analysis "does not leave any scope for an economic model to come between the premises and the phenomenon to be explained. Instead, the question shifts back to: `Why do some people have this kind of motivation more than others?'"(31)
Rational choice theorists have had greater success explaining voting on the margin, given a taste for voting distributed somehow in the population. This work has focused primarily upon how changes in costs, such as elimination of the poll tax or increased availability of political information, affect turnout.(32)
A few brave theorists have ventured to explain why people vote without relying on a "taste for voting" argument. These efforts have largely failed, however. One line of research has examined circumstances under which p could be high enough to make pB > C. Palfrey and Rosenthal noted that the value of p depends upon strategic interaction among all voters-after all, if a voter knew everyone else was planning to abstain, p = 1 for that voter and her vote is decisive.(33) The authors concluded that in a society in which every voter had complete information about the preferences and voting costs of every other voter, p could be very close to 1 in equilibrium.(34) They later conceded, however, that the full information assumption is unrealistic for large electorates, and that without that assumption, the only citizens who vote are those "whose sense of duty outweighs any cost in voting. We have come full circle and are once again beset by the paradox of not voting."(35)
Schwartz also has presented a model with a nontrivial value for p, positing that although one person's vote has no chance of determining an election winner, "it has, at least by comparison, a non-negligible chance of determining which candidate carries her precinct, and that might well determine whether and at what levels her precinct receives distributive benefits."(36) Unfortunately, Schwartz failed to provide any evidence demonstrating that p is truly non-negligible in what he terms the voter's "sub-electorate," or that voters go to the polls because they believe their vote affects the provision of distributive benefits in their precinct.
Another line of research has looked for tangible economic benefits from voting, thereby injecting narrow self-interest into the D term.(37) Schwartz, arguing from the unproven premise that "often it is obvious to others" how one voted, posits that voting can earn the voter "access, influence and credit toward political favors."(38) Although Schwartz's proposition does not appear to be true generally,(39) there is evidence that narrow self-interest motivates at least some voters. Wolfinger and Rosenstone found that government employees in patronage states tend to have a higher turnout as a group than government employees in states using a merit system for government employment,(40) presumably because politicians can monitor patronage employees' trips to the voting booths and dole out punitive sanctions for nonvoting. Schwartz's thesis has nothing to say, however, about why government employees in nonpatronage states and federal government employees (whose voting behavior is not monitored by politicians) tend to have a higher voter turnout than private employees possessing similar education and income levels.(41)
In a more plausible line of analysis, Uhlaner has argued that group leaders obtain tangible benefits from politicians in exchange for promises to deliver votes from group members. Group leaders then use some of the benefits supplied by politicians to provide selective incentives to group members to vote. As Uhlaner argues:
Sometimes the benefits are economic selective incentives (e.g., dollar
bills in envelopes handed to known supporters at the polls or repair of the
sidewalk in front of the homes of those who voted). Some of the surplus may be
used to decrease the costs of voting (e.g., by providing information or
transportation to the polls). Many times the surplus may go in to
increasing the normative benefits of voting, such as enhancing a sense of
fulfilling a group-specific duty to participate.(42)
Unfortunately, so far no one has produced compelling evidence supporting Uhlaner's thesis.(43) None of the people I know personally (an undoubtedly unscientific sample) admit to receiving tangible economic benefits from any group leaders in exchange for their vote. Given the recent furor over allegations that African-American ministers in New Jersey received money to suppress voter turnout in a recent governor's race,(44) I suspect we would see more than occasional public reports that voters received tangible selective incentives for their votes if the practice was widespread enough to account for much voting.
In sum, the standard rational choice model of why people vote has not moved beyond positing a taste for voting distributed randomly in the population. Although creative rational choice theorists have set forth novel explanations for voting without relying on the taste argument, these theorists have provided no evidence thus far sustaining their claim. Rational choice theory seems best suited to explaining voting on the margin only, by examining how changes in costs (the C term) affect turnout.
B. The Origin, Stability, and Possible Decline of a Norm of Voting
If proven, a norm solves the paradox of voting in a meaningful way; that is, by doing more than merely positing a taste for voting distributed randomly in the population. This subpart sets forth a theoretical account of the origin, stability, and possible decline of a voting norm. Part I.C then examines whether the empirical evidence supports the norms hypothesis.
1. The Origin, Stability, and Possible Decline of Norms
A social norm is a behavioral pattern characterized by three practices: (1) nearly everyone in a group conforms to it; (2) nearly everyone in the group approves of this conformity and disapproves of deviation from it; and (3) the approval and disapproval helps to ensure that nearly everyone in the group conforms.(45) Although sanctions and rewards induce some norm-based behavior, a central component of social norms is internalization, whereby "an individual comes to have an internal sanctioning system which provides punishment when he carries out an action proscribed by the norm or fails to carry out an action prescribed by the norm."(46) Thus, many group members follow the norm even when others cannot observe their behavior.
Early work on theories of social norms had a decidedly functionalist orientation, "explaining the emergence of a norm by the functions it serves for the set of actors who hold it."(47) Functionalist explanations of norms violate the principle of methodological individualism, constituting "an unacceptable deus ex machine--a concept brought in at the macrosocial level to explain social behavior, yet itself left unexplained."(48) In functionalist accounts, norms present a typical free-rider paradox: "Everyone is better off if everyone enforces a norm, but because enforcement is costly each is motivated not to bother enforcing it himself."(49)
Fortunately, scholars committed to methodological individualism--most notably Ullmann-Margalit,(50) Sugden,(51) Elster,(52) and Coleman(53)--have moved the theory of norms beyond its functionalist orientation. Coleman has set forth two requirements for the emergence of a social norm. First, a demand for the norm must arise, such as in the presence of a collective action problem.(54) Second, those who would benefit from the norm, acting rationally, can share appropriately either the costs of sanctioning or the costs of inducing others to sanction.(55) The second condition, which Coleman terms the "second order free-rider problem," is met when there are sufficient social relationships among the beneficiaries of the norm.(56)
Coleman uses the example of team sports, whereby a norm emerges for a team member to work harder than rational choice theory would predict, given the free-rider problem. The encouragement of others, that is, positive sanctions, motivates the team member to act with "zeal;"(57) "[t]he shouts of encouragement to an athlete from his teammates may cost them little but provide him with rewards that lead him to work even harder."(58) The team member rationally increases the effort because the payoff for acting with zeal is greater than the payoff for free-riding.
Although Coleman explains the rationality of the teammate's zeal in a straightforward manner, he is less than clear in explaining why other team members provide positive sanctions to the zealous teammate. Why does each individual member not free-ride rather than shout encouragement?(59)
Social relationships are the key to sanctioning.(60) Positive (and sometimes negative) sanctioning often is either costless or provides a benefit to the sanctioners. Using an important example, Braithwaite explains that people enjoy gossip, an activity that presupposes existing social relationships.(61) Talking about others' activities--positively or negatively--is a consumption activity. People playing team sports similarly enjoy shouting encouragement to the zealous team member; the activity increases, not decreases, utility.
Moreover, people in social relationships care about others' opinions of them--what McAdams has called "esteem"(62)--and Pettit has explained how people valuing esteem may be rewarded or punished, even if others expend no costs on sanctioning:
[It is commonly assumed] that the enforcement of norms must
involve intentional action and since action always generates at least
time costs that it must therefore be potentially costly for those
who conduct it. The surprising thing however is that this is
false.... [P]eople do not have to identify violators intentionally;
they just have to be around in sufficient numbers to make it likely
that violators will be noticed. And equally, people do not have to
discipline violators intentionally, going out of their way for
example to rebuke them or report them to others; they just have
to disapprove of them--or at least be assumed to disapprove of
them--whether that attitude ever issues in intentional activity.
....
We care not just about the rebukes and commendations we
receive from others but also about whether they take a negative or
positive view of what we do .... How can we know about other
people's dispositions if they are not exercised? Easy. We know
what they know of us and, ascribing similar standards to them, we
know whether they are likely to think well or badly, to take a
favorable or unfavorable attitude.(63)
To clarify and summarize the argument thus far, for a social norm to arise, group members must demand a norm, as they might in the presence of a collective action problem. The norm will emerge, however, only when it is rational for group members to provide positive or negative sanctions. In the presence of social relationships, sanctioning may be a beneficial (or at least a costless) activity. The greater the intensity of social relationships (or "social connectedness"), the greater the opportunity for positive and negative sanctioning. Social connectedness means that there will be (1) unintentional observation of group members' behavior; (2) opportunities for gossip; and (3) concern about esteem.
Once the norm emerges, it has a certain "stickiness." Group members internalize the norm, and social sanctions reinforce that internalization. Norms have whet don Elster terms "a grip on the mind."(64) Individuals follow the norm and sanction others without going through an outcome-oriented rational calculus.(65) The "stickiness" ensures that norms will remain stable over time even if demand for the norm ceases. An often-used example of this phenomenon is the norm of dueling. Some claim that the norm arose as a product of the aristocracy's effort to raise their relative social status. But it is impossible to understand the rationality of a particular aristocrat participating in a duel--even if the...
Read the full article for free courtesy of your local library.
|
Bookmark this article
Print this article
Link to this article
Email this article
Digg It!
Add to del.icio.us
RSS
