Printer Friendly

Voter turnout: from cost to cooperation.

I.   Introduction
II.  State Action and Voter Turnout
III. The First and Second Generation of Reforms
         A. The First Generation
         B. The Second Generation
IV. Voter Motivations
        A. Self-Interest
        B. Social Identity
        C. Altruistic Cooperation
        D. Community Norms
V. Community Vote Drives
VI. Conclusion


Political scientists have generally concluded that state efforts to increase voter turnout will continue to flounder so long as those efforts remain focused on lowering the already low cost of voting. (2) Accordingly, this Article argues that future efforts to achieve consistently higher and widespread voter turnout among all demographics must consider other determinants of voter behavior. The primary goal of this Article is to craft a framework based on a thorough understanding of voter motivation and behavior that helps conceptualize and analyze public efforts to increase voter turnout. The framework fills a gap in the literature by drawing from a range of fields--including election law, social psychology, sociology, and political science--that have not been previously synthesized and applied to voter turnout law and policy. This Article identifies and analyzes four core voter motivations: self-interest, social identity, altruistic cooperation, and community norms. A more complete understanding of the motivations that emerge in this Article leads to new insights into the promise and limits of specific efforts to increase voter turnout. Moreover, in a section that introduces a new concept called community vote drives, this Article ties together the analysis of the four motivations and demonstrates how they can inform future state efforts to improve voting rates.

The first and second generation reforms to increase voter turnout focused on lowering the cost of voting. (3) The first generation removed discriminatory barriers to voting, successfully erasing turnout differences between African Americans in the South and those living in the rest of the country. (4) The second generation reforms were designed to reduce structural and administrative costs of voting, primarily by reforming voter registration laws and expanding vote-by-mail options. (5) However, these second generation reforms have not led to significant changes in turnout. In the U.S., turnout remains low--by historical standards and in comparison to other countries--and lower voting rates for minority and socio-economically disadvantaged groups persist. (6) Political scientists attribute the lack of success of these more recent reforms to the fact that the cost of voting--the time to register and show up at the polls--is already so low that further reductions in cost will effect little change in voting behavior. (7) Indeed, the consensus among political scientists is now that structural reforms that make voting easier will yield little improvement in turnout rates. (8) Accordingly, future efforts to achieve consistently higher and widespread turnout among all demographics must consider other determinants of voters' behavior.

As a point of clarification, when this Article discusses voter behavior and voter motivation, it refers to decisions about whether to vote, not decisions about which political party to vote for. This Article proceeds as follows: Part II states the case for affirmative state action to increase turnout, beyond removing state-imposed barriers to voting; Part III(A) and (B) describe the first and second generations of reforms aimed to increase voter turnout, and conclude with the claim that these reforms focused on lowering the cost of voting, a strategy that will no longer produce higher turnout; Part IV is the central focus of the paper because it devotes a subpart to each of the four core voter motivations that describe the theoretical and empirical support for the existence of the motivation, and then analyzes how schemes to increase turnout based on that motivation have worked or could work in practice; Part V ties the discussion of the four motivations together by introducing and discussing the concept of community vote drives; and Part VI concludes.


It is now an established part of the constitutional canon that states may not maintain discriminatory barriers to voting, such as poll taxes. (9) Legal and political science scholarship has also recognized that states should minimize the state-imposed structural costs of voting, such as registration, to avoid electoral meltdown (10) and suppressed voting rates, even if the aggregate effect on voter turnout is small. (11) But once discriminatory and onerous state-imposed barriers to voting are removed or properly diminished, should the state do more? This Part introduces an account for why low voter turnout is a cause for concern, why consistent and widespread high turnout is good, and why local and federal state actors should work to boost turnout levels and ensure they remain high. Doing so enriches the previous justifications for state action to increase turnout by supplementing the democratic theory and political science behind these arguments with findings from social psychology.

First, I will present the facts. Voter turnout has fallen significantly during the past several decades. Most estimates show a decline of ten to fifteen points for turnout in both presidential and non-presidential elections from the 1960s to today. (12) The official numbers from the U.S. Census report turnout at 69.3% in 1964 and at 58.3% in 2004, which was up from the low of 54.2% in 1996. (13) The drop in turnout looks better or worse depending on how one crunches the numbers, (14) but the bottom line is that turnout in the past several decades has declined in most advanced democracies with the United States at the bottom of the pack among this group in terms of overall turnout. (15) Some may be satisfied with the fact that turnout is up so far this decade in comparison to the last decade. However, it is imprudent to depend on once-in-a-lifetime campaigns or political events to mobilize voters. If the goal is sustained, high turnout nationally among all demographic groups should increase.

The essence of the argument for why high voter turnout matters starts with the premise that democracy depends on some level of self-determination and governmental legitimacy. High turnout is one legitimating factor. (16) From a standpoint of democratic theory, legitimacy is threatened when the state unduly restricts access to voting booths--even if the political system would remain reasonably effective and stable. However, even after the state has removed improper or onerous barriers to voting, situational forces remain that depress turnout. These negative forces are particularly acute among socio-economically disadvantaged groups. (17) Consistently lower voter participation among these groups has two effects: their preferences are not fully aggregated in elections and they have less influence after elections, as politicians tend to neglect the interests of non-voters. (18) Higher turnout generally helps counteract these effects. Consistently, the greater number of citizens who vote nationally, the more the electorate reflects the entire population, including socio-economically disadvantaged groups. (19)

Political scientists have long known the inequalities in political representation and influence that stem from unequal rates of political participation are not randomly distributed, but "systematically biased in favor of more privileged citizens--those with higher incomes, greater wealth, and better education--and against less advantaged citizens." (20) Social psychology helps explain the situational forces that produce the differences in voting rates. Professors Jon Hanson and David Yosifon explain that, in terms of social psychology, situation "includes anything that influences our attitudes, memories, cognitions, emotions, behaviors, and the like in ways that we tend not fully to appreciate or control." (21) Many situational factors are clearly at work in explaining voting behavior. For example, community attachment, social connectedness, high levels of education, and high rates of newspaper readership are all significantly and positively correlated with high turnout. (22) Again, legitimacy is particularly threatened to the extent that these predictive factors mean racial minorities and socio-economically disadvantaged groups are less likely to vote.

Another key situational determinant of voter behavior is community voting norms. (23) Past state-imposed barriers to voting that hit disadvantaged groups hardest--as poll taxes and registration requirements did--may have not just prevented voting at the time, but also stunted voting norms in communities. Therefore, even after the state removes these barriers, an underdeveloped community voting norm remains. States that attempt to correct this situation are certainly justified.

Overall, situational factors can have just as real an impact on voter turnout as do more readily observable barriers like poll taxes and literacy tests. Indeed, even decades after states removed many discriminatory barriers to voting, the negative relationship between voting rates and socioeconomic status has persisted. (24) Ignoring the situation behind these voting numbers, which "likely advantages those people, groups, and interests with the greatest autonomy to start with" and, thus, "helps to create--and then legitimate and maintain--power relationships." (25) With democratic legitimacy at stake, legal and state actors are well justified in their attempts to overcome situational barriers to voter turnout-particularly when those situational barriers produce unequal turnout among racial minorities and socio-economically disadvantaged groups.

In response to concerns about low turnout, some claim that not voting is an affirmative decision--often based on a lack of political interest or satisfaction with the status quo. (26) Indeed, "there is a venerable literature claiming that low participation may testify to satisfaction, and that high turnout is undesirable." (27) The gist of these claims is that non-voters do not pose a threat to the health of democracy, so long as the non-voters are not denied political access when they feel moved to participate.

These scholars are right that indifference and political satisfaction are not necessarily harmful to democratic legitimacy. However, granting indifference and satisfaction too much explanatory force risks discounting situational factors that may better explain voting behavior. A long line of social psychology research has shown that "humans tend to overstate the role of individual disposition and under-appreciate the role of situation in accounting for human behavior." (28) With this in mind, the dispositional explanations of low turnout--individuals' political disinterest and satisfaction--are likely overstated, while the situational factors are likely understated in the literature. So long as situational factors largely determine voting rates, then low turnout threatens legitimacy.

Aside from arguments based on the situation of turnout, state actors have an interest in high turnout because voting helps sustain a peaceful democratic government. When voting norms atrophy in democratic countries, their citizens may cease to view voting as an expedient form of participation and political expression. With citizens less conscious of voting as a desirable form of participation, they are more likely to resort to protests, violence, and unrest. (29) A society "in which a large proportion of the population is outside the political arena is potentially more explosive than one in which most citizens are regularly involved in activities which give them some sense of participation in decisions which affect their lives." (30) Higher turnout then gives more citizens a sense of effective participation, and this safeguards stability over the long-term. Again, to the extent that the voting norm is lower in poor communities to begin with, not acting to boost turnout leads to higher probabilities of unrest in these already disadvantaged communities.

Professor Lani Guinier offers another potential benefit of higher voter turnout. She maintains that increased participation and turnout may lead to a "sense of shared fate [that] can lead individuals to join with others to challenge unfairness." (31) In this way, increasing turnout is a step, albeit an insufficient one, on the way to transforming voters into "collectively efficacious citizens" who "actively help to make, rather than who merely consume, democracy." (32) Under this aspirational view, citizens who vote are more likely to be engaged and to communicate with their representatives, thus helping to shift the metric of success in a representational democracy from "the quality of (representative) services to the quality of (representative/citizen) relationships." (33) Similarly, sociologist Robert Putnam observes that voting is intertwined with other important forms of civic engagement and social capital. (34)

All told, widespread and high voter turnout is a public good. High turnout across racial and socioeconomic lines is an important legitimating force. High turnout also helps sustain democracies and generate other positive forms of civic engagement. Ultimately, how much weight to put on getting each additional voter to the polls is a normative question. Although some may give it little weight in comparison to other activities, most would assign it a positive value. (35)

Crucially, state actors cannot leave turnout efforts to private actors and political parties. Some have criticized partisan mobilization efforts because they increase the need for money in politics and campaigns. (36) This Article takes no position on the desirability of partisan get-out-the-vote drives generally. However, it is clear that the objectives of parties' mobilization efforts do not align with the public interest in high voter turnout. Parties aim to win near-term contests; they are not primarily concerned with democratic legitimacy and stability. (37) In the U.S. electoral system, this means that parties' mobilization strategies ignore most states and pump the bulk of their limited resources into select battleground states, and within those states they target only those select voters likely receptive to their message. (38) As a result, turnout spikes in the dozen or so battleground states, and stagnates in all the others. In 2004, turnout was five points higher in battleground states than in non-battleground states that ordinarily have comparable turnout levels in mid-term elections. (39) If high and widespread voter turnout is the goal--as this Article argues it should be--then leaving get-out-the-vote efforts to the parties and their affiliates is an inadequate solution.


This Part describes the history and assesses the impact and continued efficacy of the first and second generation of reforms on voter turnout. While mostly descriptive, this Part concludes that these reforms are aimed at lowering the cost of voting, a tactic that will no longer work to increase turnout. Therefore, future efforts must consider other determinants of voter behavior.


In the late nineteenth and early twentieth centuries, voter turnout was high and was not a major public issue. (40) Voter fraud, however, was a substantial problem. (41) State lawmakers combated fraud by passing a slew of new regulations. Most states enacted registration requirements for the first time, (42) they switched to secret ballots, (43) and they constrained election officials' discretion with time, place, and manner restrictions. (44) The reforms succeeded in curtailing fraud. (45) In 1934, election administration expert Joseph Harris wrote: "Since 1900 the general tone of election administration had greatly improved throughout the country, and frauds, formerly so widespread, have tended to disappear in all but a few communities." (46)

But the tighter regulations came at the expense of voter turnout, which fell heavily in the two decades after 1900. (47) The states' move to a secret ballot hurt political parties' mobilization efforts, as they could not count on the votes from the people that they delivered to the polls. (48) Registration laws added onerous bureaucratic requirements to the act of voting. (49) The new regulatory landscape was particularly harsh for African-Americans in the South. After Reconstruction, black male turnout was high, with a majority voting in all but two Southern states. However, after the turn of the century, black turnout in the South dropped steeply. (50) Southern officials abused the trend of tighter regulations to disenfranchise minority voters with arbitrary registration practices, poll taxes, and literacy tests, among other racist devices. (51) Southern politicians fought hard against federal efforts to end these Jim Crow practices, with Southern representatives in Congress killing attempts to outlaw poll taxes. (52) Moreover, when the Department of Justice began investigating and prosecuting local Southern officials, they "engaged in ingenious forms of resistance--losing records, or purging black voters from the rolls shortly after their enfranchisement." (53)

Eventually though, the racist barriers began to fall. In the 1960s, the Supreme Court held poll taxes unconstitutional. (54) More importantly, reformers mustered enough votes to pass the Voting Rights Act of 1965 (VRA). (55) Section 2 of the VRA prohibited literacy tests nationwide. (56) Section 5, the most sweeping and ambitious part of the law, applied only to certain jurisdictions (mostly in Southern states) and required these jurisdictions to receive preclearance from the Attorney General, or from a panel on the D.C. District Court for any changes to their voting laws or election structures. (57)

Voter turnout was a key variable in the formula to determine which jurisdictions section 5 would cover. (58) Congress decided that all jurisdictions with a voter turnout rate below fifty percent and that in the previous presidential election had used a test or device--such as a poll tax or literacy test--would fall under section 5. (59) The formula was derived to avoid "the politically fraught task of merely picking and choosing jurisdictions based on subjective judgments about their relative lack of protection for minority voting rights." (60) Predictably, covered states bristled at the unprecedented federal intrusion. However, the Supreme Court upheld a challenge to the constitutionality of the VRA, finding that the voter turnout part of section 5's coverage formula was valid and rational because "a low voting rate is pertinent for the obvious reason that widespread disenfranchisement must inevitably affect the number of actual voters." (61)

Despite huge progress under the VRA in the past forty-plus years, the law remains necessary to combat persistent forms of voting rights deprivations. Accordingly, Congress voted to reauthorize the law in 2006. (62) Nonetheless, voter turnout is no longer a handy measuring stick for VRA progress and violations. In the 2006 congressional debates over which jurisdictions section 5 should now cover, voter turnout was useless as a metric for assessing the jurisdictions that most egregiously thwarted minority political participation. Although nationwide turnout among African Americans is typically five to ten points lower than the national average, (63) there is now no difference in turnout between section 5 covered and non-covered jurisdictions. (64) In fact, African Americans vote at higher rates than do whites in some covered jurisdictions. (65) Some states may still have electoral structures that have particularly negative impacts on minority voters, but overall, turnout rates no longer help us separate these states. (66)

After digging through the legislative history of the 2006 reauthorization of the VRA, Professor Nathaniel Persily concluded that "[f]or the 2006 reauthorization, voter registration and voter turnout statistics--either aggregate registration and turnout by state or differential registration by racial group--did not help the cause much." (67) Ultimately, even as the VRA continues to combat racial discrimination in voting, it will likely lead to little discernable increase in voter turnout.


In the 1980s, lawmakers began a second generation of reforms intended to increase turnout. (68) They designed these reforms to lower the structural costs of voting by either making the act of voting easier--e.g., mail-in voting--or by making the act of registering to vote easier--e.g., motor voter registration. (69)

Since then, many states have been expanding their vote-by-mail systems, generally loosening their absentee voting laws and allowing voters to cast ballots as early as a few weeks before Election Day. (70) Oregon is leading the way in this trend. (71) In 2000, it became the first state to conduct a statewide vote-by-mail system in a presidential election (72) Unlike absentee voting, where the voter must register to receive an absentee ballot, Oregon simply mailed ballots to all voters on the rolls. (73) The voters had three weeks to complete and mail back ballots. (74) Several other states are following suit; including Washington, which now has about ninety percent of its voters casting ballots by mail. (75)

Mail-in voting has two goals: increase turnout by making voting convenient and eliminate the administrative costs of setting up and operating polling places. (76) With similar motivations in mind, some states have considered internet voting; however, it is more vulnerable to fraud and hacking than mail-in-voting and there is an inequity due to the wealth-based digital divide in internet access. (77) Before Oregon's first vote-by-mail election, some estimated the mail-in system would increase turnout by ten percent. (78) However, vote-by-mail has not proved a boon to voter turnout. One study found that vote-by-mail increased turnout substantially for low salience elections--such as local government races--but had only a limited effect on high stimulus elections in which interest and campaign activity were high. (79) Lending support to this finding was an analysis of elections in several California counties where some voters had been assigned by the government to vote by mail because they lived in less populous precincts. (80) Controlling for demographics, the researchers compared turnout in vote-by-mail precincts to turnout in in-person precincts at the polls. They found that in salient elections such as presidential and gubernatorial races, turnout for vote-by-mailers was almost three percentage points lower than in-person turnout at the polls, while in less salient local elections, turnout for vote-by-mailers was 7.6% higher. (81)

One hypothesis is that in high salience elections, the incessant campaigning and media coverage make the cost of gathering information about candidates low. Therefore, any further decrease in costs through voting by mail has little effect. Conversely, voting by mail helps offset the increased costs of gathering information for low salience races, in which the limited campaigning and media coverage make it more burdensome to gather relevant information. (82) Whatever the reason, the conclusion is the same: vote-by-mail is unlikely to boost turnout in important elections.

Analyses of reforms of registration laws lead to similar lackluster conclusions. In the past few decades, voter registration has been the "single factor most often cited as responsible for low turnout in the U.S." (83) In most states, voters must register twenty to thirty days before Election Day. (84) Until the mid-1990s, many residents could register only at select state offices during restricted hours. (85) These registration requirements may seem like small costs, but they are relatively significant when compared with the low cost of showing up at a precinct on Election Day and filling in a ballot. A rational choice line of reasoning suggests that because voting is a low-benefit activity, "small changes in the costs of voting might have sizable effects on overall turnout rates." (86) Thus, making registration easier should increase turnout.

Along these lines, in 1993, Congress enacted the National Voter Registration Act (NVRA) (87) to "increase the number of eligible citizens who register to vote" and to "enhance[] the participation of eligible citizens as voters in elections for Federal office." (88) Most importantly, the NVRA mandated that states offer residents the chance to register through their regular dealings with state motor vehicle departments and made states offer registration at all state offices providing public assistance to the poor. (89) The public assistance agency provision was added to the NVRA to compensate for the fact that minorities and economically disadvantaged groups were less likely to own cars and, thus, reap the benefits of the motor voter provision. (90)

Before its implementation, many researchers were optimistic about the NVRA's projected impact on voter turnout. Several states had already begun experimenting with motor voter laws, (91) allowing researchers to extrapolate on the effects of nationwide motor voter registration. One study, for example, estimated that the motor voter registration provision would increase turnout at least 4.7%. (92) Indeed, the NVRA did lead to millions of new registered voters, although no significant change in voter turnout appears to have resulted. In the 1996 election, the first presidential election after states implemented the NVRA's provisions, turnout was 6% lower than in the 1992 election. (93) In 1995 and 1996, more than 40 million people had registered at DMVs, but significantly fewer of them voted than new registrants generally. (94) Moreover, unknown millions of these DMV registrants would have registered by some other, albeit less convenient means, if the NVRA had not been law. Numbers from one election in 1996 are hardly dispositive, particularly given that the race involved a popular incumbent president. However, these lackluster returns have been backed up by more recent turnout numbers and studies. (95) Thus, it has been common for political scientists to conclude that lowering the cost of registering to vote, as the NVRA did, mostly leads to an increase in registered voters who do not vote. (96)

The NVRA was successful in one respect. Under NVRA mandates, the poorest and least educated families were just as likely or even more likely to register to vote. (97) However, racial disparities in registration persisted, with whites somewhat more likely than African-Americans or Latinos to register. (98)

What explains the failure of the NVRA to significantly improve turnout? The NVRA lowered the cost of voting for past non-voters, but it did not do enough to increase their desire to vote. Perhaps more importantly, it did nothing to remove registration barriers that prevent likely voters from having their ballots counted. Election Day Registration (EDR) or a universal registration law would solve this problem. EDR, which is currently applied in six states, allows unregistered voters to register when they show up at the polls. (99) It is superior to motor voter registration because it reaches cohorts of voters the NVRA misses: late-deciders, young voters, and mobile voters. (100) Many Americans do not pay attention to elections until the end of October. (101) By that point, many states' registration deadlines have passed, thus, barring late-deciders from deciding to vote, just when interest in the election is peaking. EDR removes this deadline problem. Moreover, people who have moved recently and mistakenly believe their registration transfers with them to their new address routinely show up at the polls only to find they are not registered and cannot vote. (102) EDR also solves this re-registration problem. Finally, EDR also helps many young, first-time voters who may not be aware of registration requirements. (103) Consider someone who gets her driver's license when she is sixteen. She cannot register to vote at that time because she is too young, and the first election in which she is eligible to vote may occur before she has to renew her license or have any other reason to deal with the department of motor vehicles again. Thus, the NVRA's motor voter provision does nothing to inform her about registration and make it easy for her.

States have not adopted EDR because they must bear the added administrative expenses. (104) Professor Richard Hasen proposes a universal voter registration system run by the federal government in conjunction with a voter identification program. (105) Although the primary purpose of Professor Hasen's proposal deals with fraud and electoral meltdown, it should improve turnout as much as universal EDR would. (106) Studies, though, show even an overhaul of registration schemes through EDR or universal registration would not increase turnout significantly--just a few points. (107) Ultimately, further structural reforms such as EDR or keeping polling places open longer will have only a limited impact on turnout.

The findings on the vote-by-mail systems and motor voter laws add to "the growing body of research that suggests that relaxing administrative requirements is not likely to be the panacea for low turnout among the disenfranchised." (108) If the goal is consistently high turnout across the country, initiatives must focus beyond state-imposed costs of voting. They should focus on more complete and complex determinants of voter behavior.


This Part identifies and analyzes four key voter motivations: self-interest, social identity, altruistic cooperation, and norms. It presents the theoretical and empirical support for each motivation, and then assesses how schemes to increase turnout based on each motivation would hold up in practice. By drawing from a range of scholarly fields that previously were not in complete conversation with each other on this subject, the analysis below offers a comprehensive understanding of voter behavior and motivation as well as new insights into how those motivations can inform efforts to increase turnout.


In the 1960s, rational choice theorists began to frame voting as a public good, collective action problem in which everyone benefits from having an election producing a democratically legitimate government but not everyone contributes to the effort. (109) Anthony Downs, and later William Riker and Peter Ordeshook, developed the rational actor's calculus of voting. (110) The model estimates the expected potential benefit a self-interested voter would receive from having her preferred candidate win and multiplies that by the chance that the voter will cast the decisive vote in the election. (111) The benefit and probability are then offset by the costs--the time and energy needed to acquire information on the candidates and then cast a ballot. (112) Their conclusion was that although the costs are small, the chance of casting the deciding vote is even smaller. (113) Therefore, the rational voter will choose to stay home and take a free ride on other voters' efforts. This produces the paradox of rational choice theory and voting: if all voters were rationally self-interested, then none would vote. The rational choice model holds limited predictive value for voter turnout. In this country, more than half of eligible voters typically cast a ballot for the president; they do not stay home as rational choice theory predicts. (114) The rational choice calculus overlooks essential determinants of voting behavior--such as partisan identity, altruistic cooperation, and norms. However, rational choice is not completely irrelevant for turnout.

It has value on the margins and when used to predict changes to turnout levels, but not baseline turnout rates. For example, when a race is close, turnout increases. (115) Rational choice would predict a small increase for close races because the probability of casting the deciding vote is higher--even if any given vote is still unlikely to be decisive. Moreover, rational choice explains how structural barriers to voting, like poll taxes, may drive down voting rates because the tax operates as an added cost. (116)

A basic rational actor understanding of voting appeared in political literature long before Downs. More than a century ago, some analysts began predicting that, as suffrage rights expanded, the more educated citizens would make the rational choice and stop voting. (117) Even then, the rational actor model of voting was off the mark. For decades, turnout consistently has been lower among poorer, less-educated citizens. The analysts lacked an understanding of the situational factors behind decisions of whether to vote.

Despite its limits, a rational actor view of voting underlies many efforts to increase turnout for the simple reason that the variables in the rational actor calculus are often relatively easy to manipulate. In terms of sheer turnout numbers, the most successful rational actor-based interventions are compulsory voting laws that sanction non-voters. (118)

Voting is compulsory in several countries, with Australia often discussed as the example most relevant to Western democracies. Voting has been compulsory in federal elections in Australia since 1924. (119) Turnout there hovers around 95%. (120) Australia's mandatory voting law contemplates a fine of up to $50 for non-voters. (121) However, it is hard to tell how much of that turnout is due to the external force of the threatened sanction--a factor that would clearly count as a cost in the rational choice calculus--and how much is due to voting norms that have been cemented over the decades--a factor not fully incorporated in the rational choice calculus. In practice, non-voters often avoid fines by providing the election commission with written excuses for why they were absent from the polls, almost all of which are accepted. (122) Therefore, some argue the ease of avoiding actual sanctions suggests norms, created in part by the compulsory law's expressive effect, the real force behind the high turnout. (123) However, for evidence that compulsory voting does not lead to enduring voting norms, others point to dramatic drops in turnout in the Netherlands, after it abandoned compulsory voting; and also to the relatively measly 47% turnout when Australians voted without the threat of a fine for delegates to their 1998 Constitutional Convention. (124)

Regardless of whether costs or norms are doing most of the work, compulsory voting undeniably increases turnout while external sanctions are in place. Accordingly, calls for compulsory voting in the United States are fairly constant. (125) However, as Professor Richard Hasen points out, it is not politically feasible in this country; indeed, "hackles rise when compulsory voting is mentioned," perhaps because of Americans' libertarianism and individualism. (126)

Instead of sanctioning non-voters, states may entice voters with financial incentives--a less coercive alternative that should not offend libertarian sensibilities the same way. Along these lines, some have proposed the state run a voters' lottery in which a jackpot is awarded to one voter whose name is randomly drawn from the list of all those who went to the polls in the past election. The idea received some support from Professor Pamela Karlan several years ago: "The popularity of lotteries, particularly among the lower income groups who are least likely to register and to vote might make this a particularly effective method of increasing voter turnout because many potential voters might be more motivated by the chance of a significant payout for voting than by the certainty of having their costs reimbursed." (127)

In 2006, Arizona citizens voted on whether to implement a voter lottery scheme in their state. (128) Tucson ophthalmologist Mark Osterloh put the Arizona Voter Reward Initiative on the ballot. (129) It would reward one random voter a $1 million prize in every presidential primary and general election. (130) To entice voters, Osterloh made the lottery retroactive, so that all the two million Arizona voters in 2004 were eligible to win the prize in 2006 if the initiative passed. (131) In an interview with the The Arizona Republic, he defended the initiative:
   I've seen get-out-the vote campaigns that spend a lot of time and
   an incredible amount of money, and they don't work. So I was
   thinking, and it came to me: Motivate people to do it. If you
   reward people for doing their patriotic duty, if you make them want
   to go to the polls, it would be so much easier. Why not have a
   system where if you vote, you have a chance to get rich? (132)

The initiative lost by a sound margin--66.6% to 33.4%, or 991,280 votes against to 496,636 votes for. (133)

Appealing to voters' self-interest through compulsory voting, a lottery, or other external monetary incentives is problematic for several reasons--reasons that do not exist or are much less significant when states instead appeal to motivations that are intrinsic or capable of internalization, such as altruistic cooperation, social identity, and norms. First, paying voters with lottery tickets or cash may simply be a waste of money. The low cost of a lottery ticket--often just $1--may not be enough to draw to the polls people who would otherwise not come. It may be that monetary incentives high enough to significantly increase turnout would simply drain government coffers too much. Moreover, if a monetary scheme worked and increased turnout significantly, it could have detrimental effects on electoral outcomes and campaign tactics. When people vote primarily because of an external incentive like money, they have little incentive to become informed about the race and the issues that matter to them; the external incentive only motivates them to have their name checked off by a poll worker.

Whether uninformed voters distort outcomes is a current debate in the political science literature. (134) However, there is convincing evidence from Australia, where compulsory voting draws to the polls blocs of apathetic voters, that large numbers of uninformed voters tilt outcomes toward incumbents and lead elected officials to rely on rent-seeking campaign strategies. (135) Incumbents have advantages in most election systems, but Australian law Professor Graeme Orr observes that the incumbency advantage is particularly acute in Australia because officials peddle public money and services for votes from disengaged voters: "[T]he incumbency advantage is now tied to the ability of the executive government, and its political advisers in its tight-knit party machine, to engage in a constant soft-soaping of the electorate through the manipulation of public resources." (136)

Second, in places where a strong community norm encourages voting, schemes like lotteries that introduce money into voting are publicly perceived as cheapening a core civic duty. For example, The Arizona Republic blasted the lottery initiative as "pure silliness," (137) and The Yuma Sun likened it to "[b]ribing people to vote" and concluded it would rather have low turnout. (138) The head of the non-partisan think tank, the Center for the Study of the American Electorate in Washington, captured the idea that reinvigorating voting norms is superior to relying on rational choice-inspired monetary incentives: "We need to rekindle the religion of civic duty, and that is a hard job, but we should not make voting crassly commercial." (139)

Finally, overlooked in the voting literature, these responses point toward another problem with enticing voters with money--the risk of a crowding out effect. Crowding out occurs when increasing external or monetary incentives actually reduce the desired outcome because they reduce the pre-existing intrinsic motivations. (140) The intrinsic motivation suffers when individuals view the external incentives as controlling. (141) The effect has been documented in studies on people's willingness to absorb costs for the public good. (142) If the effect applies to voting, monetary incentives would not only invite uniformed voters, it would actually crowd out the intrinsic motivation of the voter--that is, it would crowd out the motivations from social identity, altruism, and community norms--and could lead to a reduction in turnout overall.

It is difficult to know whether voters would perceive a lottery ticket reward or any cash payment as controlling and, thereby, reduce the intrinsic motivation. Arizona's lottery initiative, if it had passed, could have served as a valuable natural experiment. However, there is probably good reason democracies have avoided paying voters. As Professor Cass Sunstein observes, if votes were treated as commodities, "we would have a different conception of what voting is for--about the values that it embodies--and this changed conception would have corrosive effects on politics." (143) Ultimately, efforts to increase turnout that rely less on money and more on social identity, altruistic cooperation, and norms will avoid distorting and reducing community norms that favor voting.

Nevertheless, the rational calculus of voting is not only about paying or fining voters. Some proposed structural changes to our election system may increase turnout by changing a different variable in the calculus on self-interest and voting: the probability of affecting the outcome of the election. The probability variable positively correlates with the competitiveness of the election, a factor that has been shown to affect voter turnout. (144)

There are two ways to increase the probability variable through structural changes to elections. One is to switch from our winner-takes-all system to proportional representation (PR). Countries that have a PR system consistently have higher turnout rates than countries that do not. (145) This higher turnout almost certainly happens because votes matter more in PR system elections. In the U.S.'s winner-take-all system, the candidate who gets fifty percent of the vote plus one wins everything. A vote is entirely meaningless--from an instrumental, outcome determinative perspective--if it is cast for a candidate who will fall short of that mark or if it is cast for a candidate who will exceed that mark without the help of that vote. In a PR system, however, a vote is more likely to count because seats are awarded proportionally to the parties based on the percentage of the vote they win. A party that receives twenty percent of the vote will hold twenty percent of the seats in the legislature. Votes are not wasted because they have a much higher chance of determining whether a party will pick up an extra seat. Professor Lani Guinier, and others, have advocated for some form of PR system in the U.S. for years. (146) Based on voting rates in PR system countries, if a PR system were adopted in the U.S., it would likely increase turnout here. Regardless of a PR system's relative merits, however, a dramatic shift away from our traditional winner-takes-all system would involve severe transition costs and is not politically feasible on the federal level.

More feasible are proposals that state actors should draw congressional districts that consist of equal parts of Democrats and Republicans. Professor Samuel Issacharoff is the leading proponent of drawing competitive districts, although not because it increases turnout. (147) His main argument is that competitive districts promote accountability by ensuring "competitive processes by which voters can express choice." (148) Professor Nathaniel Persily has responded that there is no a priori reason to prefer competitive districts; indeed, "competitive districts may lead both to greater power for the ideologically centrist, median voter and to almost half of the voters' being unhappy with their representative...." (149) The purpose here is not to settle the debate over the desirability of competitive districts but to inject into it the factor of voter turnout, which has not yet been discussed in the debate. The bottom line here is that more competitive districts may lead to higher turnout. (150)

In sum, schemes to increase turnout based on rational actor models and voters' self-interest inherently involve several problems--possibly huge government expenditures, distorted electoral outcomes, rent seeking, and crowding out effects--that more or less do not exist for schemes based on the more intrinsic motivations. A rational actor understanding of voter turnout can, however, contribute to the debates about proportional representation and districting.


Social identity theory concerns how an individual's awareness that he or she belongs to a social group affects their behavior and motivation. (151) The theory makes two claims relevant to this Article. First, after people identify as members of a group, they will "seek to achieve positive self-esteem by positively differentiating their ingroup from a comparison outgroup on some valued dimension." (152) Second, they will exhibit "a willingness to exert effort on [the group's] behalf." (153) Group members will exhibit this behavior, even if it is not in their narrow self-interest, because their motivation "derives not from the unique qualities of individuals but from their collective sense of who they are and what they feel compelled to do in order to maintain and promote that identity." (154) However, self-identification with a group is not simply binary, but a continuum. The greater extent to which individuals internalize their membership in a group, the more likely they are to act collectively on its behalf. (155)

Dictator game experiments have shown that social identity is in fact a determinant of voter turnout. In one experiment, subjects were given lottery tickets and then told to act like dictators by either keeping all the tickets for themselves or dividing them up as they saw fit with an anonymous individual. (156) The identity of the recipient was never revealed, but some subjects were told whether the recipient was a registered Democrat or Republican. (157) Predictably, Democrats gave significantly less tickets to Republicans than they did to either other Democrats or anonymous recipients whose party-affiliation was unknown. (158) Republicans exhibited the same ingroup loyalty. (159) Meanwhile, independents gave more to entirely anonymous recipients than to those who they knew were identified with either party. (160) After grafting these findings onto the subjects' reported turnout rates, the researchers found that those who shared more with fellow partisans were also more likely to have voted in past elections. (161) This finding shows that those who identify strongly with a political party, as measured by their willingness to share more with fellow members of their party than with others, are more likely to vote. The study's results are also a rebuke to a limited rational choice model of voting, which predicts that self-interested subjects would not share any of the lottery tickets.

Empirical data confirms the finding from the dictator game experiment. Election after election, those who self-identify as strong partisans vote at higher rates than those who consider themselves weak partisans or nonpartisans. (162) However, the important factor here is not party registration, but strength of self-identification with a party. About nine out of ten independents already habitually lean toward one of the two major parties, and they vote accordingly. (163) Self-identified independents who lean toward the Democrats or Republicans in fact vote at about the same rate as voters who self-identify as weak Democrats and weak Republicans. (164) Strong partisans, by contrast, vote at significantly higher rates. These numbers suggest that simply substituting the label independent for the label democrat is not enough to improve partisan identity, and therefore, not enough to increase turnout. What matters is whether the voter has strongly internalized a partisan identity, and thereby achieves esteem by contributing to his or her partisan group by voting for its candidates. On this point, social identity theory goes beyond the traditional rational choice theory of voting by adding the benefit of esteem or value to decisions on whether to go to the polls.

Recent results from a series of social psychology experiments known as minimal group experiments illuminate the point that merely assigning a label to someone is not enough to engender ingroup loyalty. (165) Early minimal group experiments by psychologists found that simply assigning people a social category led them to behave in ways that favored their assigned group. (166) Based on this finding, voter turnout should increase if, even for the most superficial and random reasons, more independents nominally join political parties. The numbers on voting rates for partisans and independents, however, suggest that this is not the case. Indeed, later minimal group experiments have called into question the earlier studies' findings. These recent experiments found that people only behave favorably toward group members with whom they are randomly grouped and from whom they expect benefits in return. (167) If these more recent studies are right, then enticing independents to join political parties in name only will not increase turnout. New Republicans and Democrats must either receive some benefit for voting for their new parties or internalize their partisan identity to a greater extent, thus, receiving esteem when they vote for their group's candidates. To some extent then, until partisan identity becomes more internalized, the voter will behave as predicted for a rational actor--namely, he only absorbs the cost of contributing to the ingroup if he expects a direct benefit in return.

Can state actors draw from these understandings of social and partisan identity to increase turnout? In practice, state regulation of political parties is one potential tool to boost partisan identity and thereby impact turnout. (168) Regulations of political parties are rarely, if ever, directly aimed at increasing turnout by boosting partisan identity. However, regulations of parties often impact turnout, although usually inadvertently and sometimes negatively. For example, turnout decreased after state reforms designed to combat graft diminished the power of urban party machines. (169) When urban bosses could no longer deliver jobs and services to community members, voters had less incentive to identify with the party and the parties themselves had less power to mobilize voters. (170)

Nevertheless, despite recent findings on partisan identity and voter turnout, this Article argues that the current understanding of how voting and partisan identity interact is still too imprecise to form a reliable basis from which political party regulations could produce higher turnout. Indeed, such attempts may prove ineffective and counterproductive. Consider the effects of primary election laws designed to reward affiliated party members and punish registered independents. Many states have closed primaries in which only Democrats and Republicans can participate in their respective primaries, which exclude independents from the system. Other states have open primaries in which independents, and in some cases any registered voter regardless of the party, can vote in either the Democratic or Republican primaries. States with closed primaries give party members an advantage--i.e., they get to vote--and put independents at a disadvantage--i.e., they do not get to vote. In contrast, open primaries treat registered party members and independents equally.

Do primary systems that put independents at a disadvantage entice them to join political parties and thereby increase partisanship and turnout? Assuming the primary laws only increase formal party affiliation but not internalized partisanship, the findings of the minimal group experiments and the data on partisanship's link to turnout suggest that the answer should be no. In fact, analyses by economists that isolate primary type as a variable for affecting turnout have shown that open primary systems actually produce slightly higher turnouts in general elections than do closed primary systems. (171)

There are a few possible explanations for why turnout is higher in open primary systems. First, open primaries produce more centrist candidates leading some to conclude that these candidates attract a greater number of voters in the general election. (172) Alternatively, allowing more voters into the process at the primary stage may help cement a voting habit or community voting norm that spills over into the general election. Another explanation, a form of cognitive dissonance based on psychologists' findings, suggests that after voting for a particular candidate, voters immediately view their candidate of choice more favorably than they did before, particularly in comparison to other candidates. (173) In other words, because open primaries draw more voters to the polls than do closed primaries, (174) more voters will walk away from the primary election with an inflated view of the candidate they voted for, and thereafter will want to support that candidate again in the general election. Regardless of the explanation, the fact remains that closed primaries do not increase turnout. (175)

This discussion indicates that state regulations to increase party affiliation will not necessarily also increase turnout. Indeed, incentives to join political parties without added measures to increase partisan feelings may actually hurt turnout--witness the effects of closed primary systems. (176) Ultimately, it remains uncertain whether there are methods that state actors can rely on to boost partisan identity, not just partisan affiliation, and thereby increase turnout. Moreover, attempts to engineer partisanship run the risk of generating undesirable hyper-partisanship. Until more is known, state actors should instead look to other factors that motivate people to vote--such as altruistic cooperation and community norms.


Altruistic cooperation explains, in part, why humans contribute to public goods like participating in choosing representatives in government. For this Article's purposes, altruistic cooperation refers to pro-social, cooperative behavior in which humans contribute to the public good beyond what a rational actor model would predict. (177) Altruistic behavior of this sort has been witnessed in countless laboratory experiments and in the real world. (178) Individuals have different predispositions for altruistic cooperation; (179) however, such behavior is dramatically affected by situations. (180) This last point suggests that ambitious system designers can try to stimulate altruistic behavior.

Like social identity theory, applying altruism theory to voting helps correct the paradox that results from Down's rational calculus of voting. A rational actor understanding of voting holds that a purely self-interested person will not vote because the cost of voting outweighs the benefit received from their preferred electoral outcome combined with the small probability of affecting the election with one vote. (181) Under an altruistic view, voting makes more sense because the value of the benefit is larger than it is under a rational actor model. (182) An altruistic voter is motivated by a desire to contribute to the public good, not just a desire to help himself. (183) His one vote works for the benefit of a large number of others. (184) Therefore, "people who care about benefits to others and who think one of the alternatives makes others better off are more likely to vote." (185)

Some of the dictator game experiments that found that partisan identity is a determinant of voting behavior also showed altruism at work. In one such experiment, subjects were told that they were eligible to receive a prize of $100 and were asked how much of that prize they would choose to share with an anonymous individual. (186) The game tested subjects' altruistic behavior because they were under no obligation to share the prizes--that is, they were dictators who could distribute the money as they saw fit or keep it all for themselves. The results revealed that "variation in concern for the well-being of others in conjunction with strength of party identification is a significant factor in individual turnout decisions." (187) The most likely subjects to have voted in the past election, a California primary, were those who exhibited altruistic behavior by giving away half of their prize. (188) Specifically, an altruist who shared half of his prize was nearly twice as likely to have voted as an egoist who kept the entire prize for himself. (189) The study concludes that an altruistic "citizen who cares about the well-being of others may have a much larger incentive to go to the polls." (190)

In practice, two crucial factors for increasing altruistic behavior are face-to-face communication and empathy. (191) Talking to others demonstrates a "genuine willingness to cooperate and to determine others' willingness to do so" while helping others to "see the connection between their individual interest and that of the group," with "more talk leading to more cooperation." (192) Meanwhile, people are more likely to behave cooperatively with, and generously to, each other if they have an opportunity to empathize with one another or humanize themselves. (193) For example, experiments have documented that, after seeing the face of someone else, a subject is more likely to treat him or her generously. (194) Additionally, when a person reveals personal information, such as his/her hobbies, generosity towards that person increases further. (195)

These understandings of communication and empathy have a couple of implications for political participation and voting. First, bringing community members together into small groups where they can see each others' faces and learn about each other should increase the level of regard they have for each other. If these meetings occur before elections and political issues are discussed, they could increase turnout by the other-regarding behavior generated in the context of political participation. Second, small meetings with community members before an election may also increase turnout because participants can communicate with each other and pick up on the cues or explicit statements that signal cooperation and intent to vote. When people sense that others are likely to contribute to a public good--that is, they sense cooperative norms in place--they are more likely to contribute themselves. (196)

These views of cooperation and political participation also fit with Robert Putnam's work on social capital. Social capital refers to the "features of social organization, such as trust, norms, and networks that can improve the efficiency of society by facilitating coordinated action." (197) For Putnam, the decline in voter turnout is part of a broader decline in political participation as social capital has atrophied in communities over the past several decades. (198) It follows that, if the sense of civic engagement and trust among neighbors is rebuilt, they will more likely contribute to public goods like voting. While this is an ambitious task, fostering engagement with particular regard to election activities is perhaps more manageable and more likely to generate immediate returns at the polls.

Along these lines, Yale law Professor Bruce Ackerman and Stanford communications Professor James Fishkin teamed up in 2004 to propose Deliberation Day--a national holiday held two weeks before each presidential election. (199) On Deliberation Day, groups of voters would gather in their neighborhoods, watch a live presidential debate, and then discuss central issues in the campaigns. (200) The event would last all day with attendees alternating between discussions in small groups and deliberation in a larger assembly of 500. (201) The citizens would collectively set the agenda of what issues to discuss. (202) At the end of the day though, they would not vote--"[t]he point is not to come to final decisions, but to think about what is at stake." (203) Everyone in attendance would receive $150, and they would have the right to take a day off work. (204) Ackerman and Fishkin plan the holiday to last two days so that only half the work force would be off on one day and essential services could continue running. (205) The goal is for deliberators to come away more informed, civically engaged, and with a greater sense of the public good. (206)

The idea of Deliberation Day stems in part of Fishkin's years of taking deliberative polls. (207) These polls gather citizens to discuss issues--asking them their views both before and after deliberation. (208) Instead of trying to capture a snapshot of public opinion, the polls answer the question: "What would the public think about an issue under good conditions for considering it?" (209) The public deliberation fostered by the polls motivated the participants to become informed beforehand and to continue their education after they arrived, so they left with more information and new opinions that better match their true voting intentions. (210) More importantly, the polling discussion led respondents to consider broad concerns, "look[ing] beyond the most narrow and immediate constructions of self-interest to support the provision of public goods." (211) Ackerman and Fishkin conclude that, "if the first few DDays succeed, tens of millions of Americans will begin to cultivate civic habits of attendance and preparation." (212) All of this will signal a healthier populace.

Deliberation Day has been attacked on several fronts. Ackerman and Fishkin themselves flag the price tag as an issue, estimating that in a four-year cycle the costs of Deliberation Day will range from $1.2 billion to $2.3 billion, depending on how many people participate. (213) Others have ridiculed the holiday's idealism and questioned its viability. (214) Some have questioned whether deliberation would actually produce the benefits the authors' claim. Professor Cass Sunstein, along with two other researchers, documented how, when liberal-minded neighbors met to discuss hot button issues like climate change and same-sex marriage, they returned home even more liberal on those issues. (215) The same polarizing effect was observed for a group of conservatives. (216) In short, "deliberation among like-minded people produced ideological amplification--an amplification of preexisting ideological tendencies, in which group discussion leads to greater extremism." (217) Indeed, these findings on polarization point to the problem of hyper-partisanship discussed in the section on social identity. Turnout may increase after people's adherence to and identity with ingroup doctrine spikes, but perhaps at the cost of tolerance and measured dialogue.

Others have documented this amplification effect in deliberation but have also found that under certain conditions, extremism can be avoided and "minority opinion can lead majorities to consider new alternatives and perspectives." (218) Ackerman and Fishkin maintain that Deliberation Day would avoid amplification in part because deliberators would rotate between large and small groups. (219) However, at least one commentator has, in turn, argued that Ackerman and Fishkin do not adequately respond to Sunstein's criticisms. (220)

The point here is not to take sides in this back and forth on deliberation. Rather, it is to discuss an overlooked possible benefit of Deliberation Day--increased voter tumout. Ackerman and Fishkin briefly reference turnout, predicting in passing that "[o]nce those citizens are enmeshed within the conversational net [on Deliberation Day], they may well decide to vote on Election Day as a result." (221) The discussion here on theories of altruism, cooperation, and social identity lends credence to this prediction. As discussed in the beginning of this subsection, schemes like Deliberation Day that promote face-to-face communication and the humanization of fellow citizens should boost turnout by building a sense of altruistic cooperation as well as eventually helping to build a community voting norm. However, it may also boost turnout by generating unhealthy levels of partisan identity. Whatever the case, the effect on voter turnout should be included in the dialogue on the proposed holiday and other work on political deliberation in communities.


For decades, some scholars have hypothesized that voting norms are a major determinant of voter turnout. (222) The basic idea is simple--you are more likely to vote if your neighbors vote. (223) Norms can motivate behavior either intrinsically or extrinsically. (224) Voters may internalize the voting norm as part of their own identity as a community member or vote because neighbors enforce the norm externally--perhaps through shaming and public disapproval, which may also initially operate as an external cost in rational actor terms. However, a norm that initially operates extrinsically can become internalized over time as voting becomes a habit. The benefit of an internalized voting norm is that the voter may take unseen steps in furtherance of the norm--such as becoming informed about the election--and also may contribute to the enforcement and spreading of the norm.

Stephen Knack, the leading scholar examining how norms affect turnout, tested the norms hypothesis by looking at how social sanctions affect voting. (225) He found a 21% increase in the likelihood of a married person voting if the spouse had indicated in a survey that he or she was the kind of person who would express disapproval to a friend who did not vote. (226) He similarly showed that people were more likely to vote if they reported having "friends, neighbors, or relatives who would be disappointed or angry" if they did not vote. (227) Knack's work was criticized for being inconclusive. Professor Richard Hasen, for example, questioned whether Knack's findings demonstrate "norm internalization as the product of social sanctioning or ... merely a randomly (but prevalently) distributed taste for voting[.]" (228)

Knack has since found other support for his community norms thesis. He and Martha Kropf examined how county-level census response rates correspond with voter turnout. (229) Census responses are informative here because, when a household mails in its census, it saves taxpayers from having to resort to more expensive population counting efforts; it is, thus, a public good to respond to the Census. Americans, by and large, are aware of the pro-social nature of filling out census forms. Surveys show that most Americans believe that responding to the Census is an indication of "good citizenship" and "the patriotic thing to do" in part because the Census "helps people in my community." (230) Brochures accompanying the Census reinforce the message by asking residents to "help make sure your community gets its fair share of federal and state funding" and determine "how many members of Congress your state sends" to Washington. (231) In short, both participating in the Census and voting are examples of cooperative norms--that is, norms that concern people's shared beliefs about expectations for contributing to a common public good. (232)

Indeed, after cross-referencing county-level census response rates with other reported levels of cooperation in communities, Knack and Kropf determined that census response is a valid measure of a community's cooperative norms. (233) Knack and Kropf ultimately found that counties with high census response rates had significantly higher voter turnout than counties with low response rates. (234) They concluded that the cooperative norms and "the social context of a community--as proxied by county-level census response--influences whether an individual will choose to vote." (235)

Knack and Kropf's conclusion on norms and voting is consistent with a National Election Studies survey of voting behavior, which found that 25% of non-voters misrepresented their past voting behavior, while only 1% of those who were regular voters misrepresented their behavior. (236) "Fear of a stranger's disapproval is the only reasonable hypothesis to explain why non-voters were much more likely to lie and say that they voted than voters were to lie and say that they did not vote," explains Hasen. (237)

All told, empirical evidence is now fairly conclusive that individuals' voting behavior is dictated in significant part by community norms. (238) Historical evidence also points to the importance of voting norms. In general, public information is crucial to establishing norms. It signals a norm exists and who is abiding by that norm, allowing norm enforcers to reward contributors and punish violators of the norms. Historically, it appears voting was once governed by norm management in communities when the public nature of voting enabled the norm to take hold. More than a century ago, voters socialized throughout the day with other citizens and party officials. (239) On the way to the polls, community members gleaned information about their fellow voters. (240) Moreover, poll workers observed all voters, and gossip spread the word about who had voted. (241) With voting as such a public and social affair, it was easy for a norm to spread, motivating people to continue to cast ballots. (242)

Can norm management reinvigorate voting habits today? Writing about norm management generally, Professor Ann Carlson presents a framework for when and how it can succeed. (243) Her central conclusion is that "norm management efforts that involve face-to-face communication or individual feedback can have some success in inducing behavioral change." (244) Aside from personal interaction, Carlson shows that convenience and low effort are also key factors. (245) Both factors overlap with factors from other determinants of voter behavior. Face-to-face communication is essential to creating altruistic cooperation, while convenience and low effort fit into the cost variable in rational choice. (246)

In passing, Carlson argues that voting rates could respond well to improvements in the convenience of voting. (247) Carlson, though, is wrong on this point. Voting is already a convenient activity, and efforts to make it more convenient--for example, by expanding vote-by-mail systems--have produced little increase in turnout, as discussed earlier. Nonetheless, Carlson's conclusion that voting should prove amenable to norm management remains valid. (248) Voting is a pro-social, public, and local activity, which make norm management easier. (249) Voting is also a low cost activity and requires periodic, not routine, effort--also pluses in Carlson's framework on norms. (250)

Some localities have already tried to reinvigorate the voting norm, relying in part on factors identified in Carlson's framework. In 2005, two New Hampshire towns agreed to take part in a pilot study of Election Day festivals designed to attract voters by holding local fairs near polling places. (251) One town served as the control; meanwhile, advertisements for an Election Day festival spread through the other town. (252) On Election Day, a tent outside the polling place attracted residents with food, cotton candy machines, music, dancing, raffles--the quintessential American fair ambience. (253) To avoid any accusations of vote buying, the festivals had to be advertised and held in a way so as not to suggest that the free food was contingent on casting a vote. Nonetheless, it does seem that shame and publicity at least in part would motivate fair attendees to vote. How could you go to the precinct for the free food and then leave without voting, while your neighbors watched?

The experiment of Election Day festivals has been replicated dozens of times by other precincts, providing researchers with enough data to conclude that first-time festivals increase turnout on average 2% at a cost of about $28 per vote (254)--making it one of the most cost-effective methods to mobilize voters. The leading political scientists studying get-out-the-vote methods concluded that turnout from Election Day festivals should increase further after the initial festival in a town because "Americans do not associate festivals with elections," therefore, "[i]t might take a festival or two for locals to see social gatherings as an expected and attractive aspect of coming to vote." (255)

In general, norm activation strategies are not one-shot efforts. A two percent increase in turnout after one Election Day festival may appear small; but that number will likely increase in subsequent elections as more non-voters perceive an improved voting norm in their communities. Increased cooperative behavior "tend[s] to generate patterns of collective behavior characterized by multiple equilibria punctuated by tipping points." (256) Therefore, each election cycle may produce only slight increases in turnout before reaching a point where turnout rises sharply. (257)


This Article has presented four key voter motivations: self-interest, social identity, altruistic cooperation, and norms. It has analyzed how these motivations work in theory and in practice. This Part ties together understandings of all four motivations by introducing and analyzing community vote drives, a promising idea for reinvigorating voter turnout.

Community vote drives are entirely public efforts. The specifics, however, are based on scholarship on get-out-the-vote (GOTV) efforts by private organizations like parties and advocacy groups, most notably from work by political scientists, Alan Gerber and Donald Green. (258) Although their studies are aimed at advancing the private "market for campaign services," (259) there is no reason why public efforts to increase turnout cannot appropriate cost-effective GOTV tactics and use them to advance the public good.

Gerber and Green's comprehensive work has shown that some private GOTV strategies are remarkably cost-effective. Although robo-calls urging people to vote are ineffective, phone banks operated by people generate about one vote per thirty-five calls, at a cost of about $38 or $53 per vote, depending on whether the callers are volunteers or paid workers. (260) The most reliably effective tactic though, is door-to-canvassing. On average, canvassers convince one in fourteen people to vote, for a cost of $29 per vote generated--if the canvassers are paid the going rate. (261) Even better results are likely if the canvassers live in the same neighborhood where they are canvassing. (262) It is unsurprising that mobilization results improve as the degree of interpersonal interaction and community ties increase. Consider that face-to-face contact is a key factor in generating norms and in establishing empathy necessary for altruistic cooperation. (263)

The question becomes whether public actors can adopt the parties' GOTV techniques for the public good. If so, instead of Republican or Democratic vote drives, there could be public community vote drives operated by volunteer community members in cooperation with local officials. As discussed earlier, private GOTV efforts are necessarily limited to a dozen or so swing states during presidential elections. If the goal is high voter turnout nationally and across demographic groups, then the parties' selective GOTV strategies are inadequate. Community vote drives are one solution to this shortcoming.

Community vote drives could be even less expensive and more successful than private mobilization efforts. First, communities could rely on volunteers to perform nonpartisan door-to-door canvassing and phone banks, instead of paying for this work. More importantly, local volunteers for community vote drives would better trigger the norms, altruistic cooperation, and social identity that motivate voters. The interpersonal communication between neighbors would establish empathy and signal a community commitment to a voting norm. Furthermore, while parties' GOTV efforts seek to parlay voters' partisan identities into votes, community vote drives could appeal to residents' social identity as part of their local community. Local communities often engender feelings of ingroup loyalty. (264)

Community vote drives could also adopt tactics that move beyond the limits of private GOTV tactics. For example, some communities could embrace a vote drive model based on community chest drives. In the early twentieth century, community chest drives successfully elicited contributions to local social welfare agencies across the country. (265) Writing in 1932, one sociologist observed that the community chest movement tended "to unite a whole community in the bonds of common philanthropy." (266) Towns and cities demonstrated how much community members had donated by displaying a "giant thermometer, clock, or other device located in the center of the city, on which the daily progress of the campaign can be registered." (267) Community vote drives could borrow from this idea and prominently display a voter turnout thermometer or clock with the turnout goal and the turnout rate for the last election. A looming turnout thermometer would signal a strong voting norm and alert altruists to a public good in need of contribution. Moreover, the public drive would help a community social identity cohere and also exert significant social pressure to vote.

Even more ambitiously, communities could exploit the power of public shaming by publicizing lists of those who did not vote in the last election. In a recent study, Green and Gerber, with another researcher, documented the massive potential of using shaming to increase turnout. (268) Before the 2006 primary election in Michigan, they sent one of four different mailings to 80,000 households. (269) One mailing reminded the households: "DO YOUR CIVIC DUTY--VOTE!" and exhorted: "Remember your rights and responsibilities as a citizen. Remember to vote." (270) The other three mailings began with that message, but also included more. The second mailing added that researchers were going to study their turnout based on public records. (271) The third mailing actually displayed the voter turnout records for each member of the household. (272) And the fourth mailing displayed both the household's turnout and their neighbors' turnout. (273) This last mailing, "[b]y threatening to 'publicize who does and does not vote,'" was designed to exert the greatest costs through "social pressure." (274) Moreover, by perhaps learning that their neighbors were voters, residents were alerted to the possibility of a strong voting norm in their community.

Using a control group as a baseline, each mailing successively increased turnout, with the fourth treatment bringing 8.1% more voters to the polls. (275) The mailings cost only thirty cents each, translating into a cost of $1.93 per vote generated by the fourth mailing, the most cost-effective voter mobilization effort studied to date. (276) Moreover, after running a regression analysis of the data, the authors found the social pressure to vote created by the mailings did not crowd out any intrinsic motivations to vote. (277) Therefore, in this regard, these sorts of mailings have an advantage over external sanctions or rewards for voting that risk crowding out intrinsic motivation from norms or altruistic cooperation.

Nonetheless, this form of public shaming may prove too severe for local officials and community members. Indeed, in the experiment, many recipients called a number provided on the "mailing and demanded to be removed from future mailing lists." (278) Less drastically, communities could approvingly publicize the names of those people who did vote and assented to the public disclosure of their names. They could list the names and turnout from the last election alongside a target turnout rate for this election, while exhorting other residents to join the list of voters and help meet the turnout goal. Voters with privacy concerns could opt to remove their name from the publicized lists. Nonprofits have used the positive publicizing of cooperators in similar ways to generate great effects in fundraisers. (279) However, people may view voting differently and still bristle at what could be perceived as a governmental attempt to manipulate emotions by inspiring either pride or shame. Ultimately, it will be up to each individual community to create the community vote drive that works best for them and does not offend people's sense of privacy. The tradeoff is, the more privacy awarded, the more difficult it is for a voting norm to take hold.

However, the appearance of impropriety, not a privacy issue, is likely the greatest problem for community vote drives. To the extent local officials are involved in running the drives or finding volunteers, perceptions of improper partisan motives may dog their efforts. For example, some officials who helped coordinate the Election Day festivals expressed concern that the nonpartisan festivals may still "appear to serve some partisan purpose." (280) However, the problem here is more about perception than a real threat of partisan manipulation that skews outcomes. Most American localities have a tradition of politics that are less divisive along party lines than are national politics, (281) due in part to the Progressive Era reforms that led most cities to hold nonpartisan local elections in which the candidates are not affiliated with any party on the ballot. (282) Moreover, it is not clear that any party would gain from increased turnout. A long list of studies from political scientists has shown that increasing voter turnout by bringing non-voters to the polls favors neither major party. (283) Nonetheless, a misperception that increased turnout helps Democrats persists, (284) and this belief could affect support for vote drives in some communities. Community leaders and local officials responsible for vote drives will have to work together to combat perceptions of partisan impropriety.

Ultimately, community vote drives have the potential to increase turnout significantly; perhaps returning voting rates to numbers not seen in half a century. However, they will not reach these numbers overnight. It could require multiple election cycles before norms cement, and the situation is such that other-regarding altruistic behavior in the form of voting, blossoms. Individual communities may have to experiment before finding the right mix of GOTV techniques that cause turnout to jump in their precincts. The purpose here is not to provide a detailed playbook for community vote drives. Rather, it is to provide a general outline of a promising idea in a way that also illuminates how the key motivations behind voter turnout may interact in future state efforts to increase turnout.


Widespread, high voter turnout legitimizes government and helps correct for lower levels of political participation and influence among socio-economically disadvantaged groups. However, state efforts to increase turnout are unlikely to succeed if they merely chip away at the already low cost of voting. To reach consistently high turnout, state actors must look to richer understandings of voter behavior. Drawing from a range of scholarly fields, this Article introduces a comprehensive framework for how state actors can conceive of and contemplate efforts to increase turnout. An understanding of how to engage core voter motivations, such as self-interest, social identity, altruistic cooperation, and community norms, must inform these efforts. Specific attempts to increase turnout based on these motivations may often prove infeasible or too unmanageable or costly. However, new ideas, such as community vote drives, can succeed by borrowing from tried and true cost-effective strategies honed by private parties. Voting in America was once driven by strong community voting norms. Committed, cost-effective efforts in communities can still produce sustained high voter turnout today.


(1.) Post-Graduate Research Fellow, J.D., Harvard Law School, 2008. I would like to thank Professor Yochai Benkler for his insights and multiple rounds of comments and Professor Terry Fisher and the members of the Spring 2008 Motivation Seminar at Harvard Law School for their input.


(3.) Id. at 23. The first generation of voter turnout reforms tracks what is commonly known as the first generation of voting rights. Heather K. Gerken, Understanding the Right to an Undiluted Vote, 114 HARV. L. REV. 1663, 1671 (2003). However, the second generation of voting rights--dilution claims--is entirely separate from what this Article calls the second generation of reforms to increase turnout. Id. at 1671; see Samuel Issacharoff, Polarized Voting and the Political Process: The Transformation of Voting Rights Jurisprudence, 90 MICH. L. REV. 1833, 1838-39 (1992).

(4.) Issacharoff, supra note 3, at 1838-39.

(5.) See NIEMI & WEISBERG, supra note 2, at 31-32.

(6.) Id. at 24-25.

(7.) Id. at 23.

(8.) Id. at 33.

(9.) United States Department of Justice, Introduction to Federal Voting Rights Laws, (last visited Nov. 10, 2008).

(10.) Richard L. Hasen, Beyond the Margin of Litigation: Reforming U.S. Election Administration to Avoid Electoral Meltdown, 62 WASH. & LEE L. REV. 937, 945-46 (2005).

(11.) See infra text accompanying notes 60-66.

(12.) U.S. Census Bureau, Voting and Registration, Historical Time Series Tables A7 & A9, available at (last visited Oct. 25, 2008).

(13.) Id. The Census numbers show a drop from 55.4% in 1966 to 42.3% in 2002 for non-presidential elections. Id. Niemi and Weisberg estimate that turnout was 65.4% in the 1960 presidential election and 50.8% in the 1996 election; meanwhile, turnout dropped from 47.5% in 1962 to 36.0% in 1998 for non-presidential elections. See NIEMI & WEISBERG, supra note 2, at 23 tbl.2-1.

(14.) Michael P. McDonald & Samuel L. Popkin, The Myth of the Vanishing Voter, 95 AM. POL. SCI. REV. 963 (2001). For example, Michael McDonald and Samuel Poplin have argued that turnout dropped sharply between 1966 and 1972 but then leveled off; however, Thomas Patterson has attacked their methodology and argued that the drop in turnout could be as severe as twenty points if one screens for the racist barriers to voting in place in the South in the 1960s. See id. at 966 thl.1; THOMAS E. PATTERSON, THE VANISHING VOTER 8-9 (2002).


(16.) Richard L. Hasen, Voting Without Law?, 144 U. PA. L. REV. 2135, 2137 (1996).

(17.) See Arend Lijphart, Unequal Participation: Democracy's Unresolved Dilemma, 91 AM. POE. SCI. REV. 1, 1-2 (1997).

(18.) See id. at 4.


(20.) Lijphart, supra note 17, at 1.

(21.) Jon Hanson & David Yosifon, The Situational Character: A Critical Realist Perspective on the Human Animal, 93 GEO. L. J. 1, 33 (2004).

(22.) See NIEMI & WEISBERG, supra note 2, at 27-30.

(23.) Id. at 31-32.

(24.) Id.

(25.) See Hanson & Yosifon, supra note 21, at 138.

(26.) See Henry E. Brady et al., Beyond SES: A Resource Model of Political Participation, 89 AM. POL. SCI. REV. 271, 271 (1995).

(27.) See, e.g., Robert W. Jackman, Political Institutions and Voter Turnout in the Industrial Democracies, 81 AM. POL. SCI. REV. 405, 406 (1987).

(28.) Hanson & Yosifon, supra note 21, at 6.

(29.) See G. Bingham Powell, Jr., American Voter Turnout in Comparative Perspective, 80 AM. POL. SCI. REV. 17, 37 (1986) (noting that higher levels of turnout in democratic nations are "associated with less frequent citizen riots and protests").


(31.) Lani Guinier, Beyond Electocracy: Rethinking the Political Representative as Powerful Stranger, 71 MOD. L. REV. 1, 17 (2008).

(32.) Id. at 4.

(33.) Id.


(35.) For a brief discussion of why some argue that low voter turnout is good, consider Jackman, supra note 27, at 418.

(36.) One study found that state parties allocate 30-40% of their budgets to direct voter contact and get-out-the-vote efforts. See Stephen Ansolabehere & James M. Snyder, Jr., Soft Money, Hard Money, Strong Parties, 100 COLUM. L. REV. 598, 616 (2000).

(37.) See Frontline, Interview Matthew Dowd, Jan. 4, 2005, (last visited Dec. 31, 2008).

(38.) Id. (showing specifically that both parties ignore purely independent voters that do not lean toward either party).

(39.) Daniel E. Bergan et al., Grassroots Mobilization and Voter Turnout in 2004, 69 PUB. OPINION Q. 760, 772 (2005).


(41.) Id. at 123.

(42.) Id. at 152.

(43.) See Jac C. Heckelman, The Effect of the Secret Ballot on Voter Turnout Rates, 82 PUB. CHOICE 107 (1995).


(45.) Id. at 20.

(46.) Id. at 19.

(47.) See Powell, supra note 29, at 20-21.

(48.) See Heckelman, supra note 43, at 119-20.

(49.) See PAUL KLEPPNER, WHO VOTED? THE DYNAMICS OF ELECTORAL TURNOUT, 18701980 220 (Gerald M. Pomper ed., 1982).


(51.) See id. at 76-77.

(52.) See id. at 459.

(53.) Id. (internal quotation marks omitted).

(54.) See Harper v. Va. State Bd. of Elections, 383 U.S. 663, 670 (1966).

(55.) See Voting Rights Act of 1965, Pub. L. No. 89-110, Title I, [section] 2, 79 Star. 437 (current version at 42 U.S.C. [section] 1973 (2006)).

(56.) See 42 U.S.C. [section] 1973(a).

(57.) See [section] 1973; see Nathaniel Persily, The Promise and Pitfalls of the New Voting Rights Act, 117 YALE L. J. 174, 195 (2007).

(58.) See Persily, supra note 57, at 195.

(59.) Id.

(60.) Id. at 196.

(61.) South Carolina v. Katzenbach, 383 U.S. 301, 330 (1966).

(62.) See The Fannie Lou Hamer, Rosa Parks, Coretta Scott King, Cesar E. Chavez, Barbara C. Jordan, William C. Velasquez, and Dr. Hector P. Garcia Voting Rights Act Reauthorization and Amendments Act of 2006, 42 U.S.C. [section] 1973 (2006).

(63.) Jan E. Leighley & Arnold Vedlitz, Race, Ethnicity and Political Participation: Competing Models and Contrasting Explanations, 61 J. POL. 1092, 1109 (1999).

(64.) See Charles S. Bullock, III & Ronald Keith Gaddie, Focus on the Voting Rights Act: Good Intentions and Bad Social Science Meet in the Renewal of the Voting Rights Act, 5 GEO. J.L. & PUB. POL'Y 1, 7 (2007).

(65.) See Persily, supra note 57, at 197.

(66.) See id. at 197-98.

(67.) Id. at 196.

(68.) See ISSACHAROFF ET AL., supra note 50, at 106-07.

(69.) Id. at 106.

(70.) See id.

(71.) See id.

(72.) See id.

(73.) See id.

(74.) See ISSACHAROFF ET AL., supra note 50, at 106. A challenge to the legality of Oregon's system under federal law was rejected in Voting Integrity Project v. Keisling. See Voting Integrity Project v. Keisling, 259 F.3d 1169 (9th Cir. 2001), cert. denied, 535 U.S. 986 (2002).

(75.) See Wash. State Grange v. Wash. State Republican Party, 128 S. Ct. 1184, 1194 n.8 (2008).

(76.) See ISSACHAROFF ET AL., supra note 50, at 107.

(77.) See generally Eben Moglen & Pamela S. Karlan, The Soul of a New Political Machine: The Online, the Color Line, and Electronic Democracy, 34 LOY. L.A.L. REV. 1089 (2001).

(78.) ISSACHAROFF ET AL., supra note 50, at 106-07.

(79.) See Jeffrey A. Karp & Susan A. Banducci, Going Postal: How All-Mail Elections Influence Turnout, 22 POL. BEHAV. 223, 227-28 (2000).


(81.) Id.

(82.) See Karp & Banducci, supra note 79, at 234.

(83.) VOTING BEHAVIOR, supra note 2, at 31.

(84.) See Miles Rapaport & Jason Tarricone, Election Reform's Next Phase: A Broad Democracy Agenda and the Need for a Movement, 9 GEO. J. ON POVERTY L. & POL'Y 379, 386 (2002).

(85.) See Rebekah Evenson, Motor Voter in the United States: Approaching the Goal of Universal Enfranchisement, in VOTING AND DEMOCRACY REPORT: 1995 (1995),

(86.) Benjamin Highton, Voter Registration and Turnout in the United States, 2 PERSP. ON POE 507, 508 (2004).

(87.) 42 U.S.C. [section] 1973gg (2006).

(88.) Id. at [section] 1973gg(b)(1), (2).

(89.) Id. at [section] 1973gg-2.

(90.) See Raymond E. Wolfinger & Jonathan Hoffman, Registering and Voting with Motor Voter, 34 PS: POL. SCI. & POL. 85, 87 (2001).

(91.) See Highton, supra note 86, at 510.

(92.) Benjamin Highton & Raymond E. Wolfinger, Estimating the Effects of the National Voter Registration Act of 1993, 20 POL. BEHAV. 79, 79 (1998).

(93.) Highton, supra note 86, at 511.

(94.) See Wolfinger & Hoffman, supra note 90, at 89.

(95.) See Highton, supra note 86, at 511.

(96.) Wolfinger & Hoffman, supra note 90, at 90.

(97.) Id. at 88.

(98.) Id.

(99.) See Rapaport & Tarricone, supra note 84, at 386-91.

(100.) Id.

(101.) Id.; see Hasen, supra note 10, at 965.

(102.) See Highton, supra note 86, at 509.

(103.) Id.

(104.) See id.

(105.) See Hasen, supra note 10, at 945.

(106.) Id.

(107.) See Craig Leonard Brians & Bernard Grofman, Election Day Registration's Effect on U.S. Voter Turnout, 82 SOC. SCI. Q. 170 (2001) (predicting a 7% increase in the average state if EDR were adopted). But see Mary Fitzgerald, Greater Convenience But Not Greater Turnout: The Impact of Alternative Voting Methods on Electoral Participation in the United States, 33 AM. POE. RES. 842, 856 (2005) (predicting an increase of only 1% in presidential elections and 3% in congressional elections if EDR were adopted).

(108.) Karp & Banducci, supra note 79, at 223.


(110.) Id. at 260-76; William H. Riker & Peter C. Ordeshook, A Theory of the Calculus of Voting, 62 AM. POL. SCI. REV. 25 (1968).

(111.) Id.

(112.) Id.

(113.) Id.

(114.) See Franklin, supra note 15, at 91-118.

(115.) Id.

(116.) See Hasen, supranote 16, at 2143.

(117.) See Lijphart, supra note 17, at 1.

(118.) Maria Gratschew, Compulsory Voting, in VOTER TURNOUT SINCE 1945: A GLOBAL REPORT 105, 110 (2002), available at

(119.) Id. at 105.

(120.) International Institute for Democracy and Electoral Assistance, Voter Turnout: Australia,

(121.) See Commonwealth Electoral Legislation Amendment Act of 1983, sec. 99, [section] 128A, 1983 AUSTL. ACTSS No. 143.

(122.) See David Bamford, Current Issues in Australian Electoral Law, 1 ELECTION L.J. 253, 258 n.29 (2002).

(123.) See Patricia Funk, Is There An Expressive Function of Law? An Empirical Analysis of Voting Laws with Symbolic Fines, 9 AM. L. & ECON. REV. 135, 155 (2007).

(124.) See Eric Fleisig-Greene, Law's War with Conscience: The Psychological Limits of Enforcement, 2007 B.Y.U.L. REV. 1203, 1231, 1231 n.71.

(125.) See, e.g., Note, The Case for Compulsory Voting in the United States, 121 HARV. L. REV. 591 (2007); Sean Matsler, Note, Compulsory Voting in America, 76 S. CAL. L. REV. 953 (2003); Christopher W. Carmichael, Proposals for Reforming the American Electoral System After the 2000 Presidential Election: Universal Voter Registration, Mandatory Voting, and Negative Balloting, 23 HAMLINE J. PUB. L. & POL'Y 255 (2002).

(126.) Hasen, supra note 16, at 2176.

(127.) Pamela S. Karlan, Not By Money But By Virtue Won? Vote Trafficking and the Voting Rights System, 80 VA. L. REV. 1455, 1472 n.54 (1994).

(128.) See Robbie Sherwood, Lottery Could Reward Voters, ARIZ. REPUBLIC, Aug. 10, 2003 at 9B.

(129.) Id.

(130.) Id.

(131.) Id.

(132.) Id.

(133.) AZ Central, Election Results,

(134.) See VOTING BEHAVIOR, supra note 2, at 100-79; see also Daniel R. Ortiz, The Democratic Paradox of Campaign Finance Reform, 50 STAN. L. REV. 893, 913 (1998) (arguing that uninformed voters are likely to vote for "the candidate with the better advertising campaign").

(135.) See Graeme Orr, Book Review, The Nitty-Gritty of Electoral Integrity, 6 ELECTION L.J. 299, 300 (2007).

(136.) Id.

(137.) Editorial, Put on Your Waders; Propositions Fill Ballot; Our Stand: Some Are Worthy, Others Not, ARIZ. REPUBLIC, Sept. 18, 2006, at 4.

(138.) Randal C. Archibold, Arizona Ballot Could Become Lottery Ticket, N.Y. TIMES, July 17, 2006, at A1.

(139.) Id.

(140.) Bruno Frey & Reto Jegen, Motivation Crowding Out Theory: A Survey of Empirical Evidence, 15 J. ECON. SURVEYS 589 (2001).

(141.) Id.

(142.) Id. at 597.

(143.) Cass R. Sunstein, Incommensurability and Valuation in Law, 92 MICH. L. REV. 779, 849 (1994).

(144.) See Franklin, supra note 7, at 91-118.

(145.) See id.


(147.) See Samuel Issacharoff, Gerrymandering and Political Cartels, 116 HARV. L. REV. 593, 627-28 (2002).

(148.) Id. at 600.

(149.) Nathaniel Persily, Reply, In Defense of Foxes Guarding Henhouses: The Case for Judicial Acquiescence to Incumbent-Protecting Gerrymanders, 116 HARV. L. REV. 649, 650 (2002).

(150.) However, a complicating factor is that drawing districts with bizarre shapes that ignore traditional districting factors like compactness may hurt turnout. See Gerken, supra note 1, at 1715.


(152.) Id. at 21.

(153.) Id. at 75.

(154.) Id. at 60.

(155.) Id. at 21-22.

(156.) James H. Fowler & Cindy D. Kam, Beyond the Self: Social Identity, Altruism and Political Participation, 69 J. POL. 813, 814 (2007).

(157.) Id. at 814.

(158.) Id. at 821.

(159.) Id.

(160.) See id.

(161.) See id. at 822.


(163.) See id. at 51. True independents are extremely rare. See id.

(164.) See id. at 50 tbl.3.5.

(165.) See Toshio Yamagishi & Toko Kiyonari, The Group as the Container of Generalized Reciprocity, 63 SOC. PSYCHOL. Q. 116, 117 (2000).

(166.) See id. at 116-18 (summarizing the history of minimal group experiments).

(167.) Id. at 118.

(168.) Although parties are private associations, historically, states have extensively regulated political parties. See Richard H. Pildes, The Constitutionalization of Democratic Politics, 118 HARV. L. REV. 29, 50-52 (2004).

(169.) See NIEMI & WEISBERG, supra note 8, at 378, 379.

(170.) Id. at 378.

(171.) Timothy Besley & Anne Case, Political Institutions and Policy Choices: Evidence from the United States, 41 J. ECON. LITERATURE 7, 30 (2003); Peter T. Calcagno & Christopher Westley, An Institutional Analysis of Voter Turnout: The Role of Primary Type and the Expressive and Instrumental Voting Hypotheses 16 (Dep't of Econ. and Fin., Coll. of Charleston, Working Paper No. 1, 2008), available at wpaper/1.html; see Kristin Kanthak & Rebecca Morton, The Effects of Electoral Rules on Congressional Primaries, in CONGRESSIONAL PRIMARIES AND THE POLITICS OF REPRESENTATION 116, 117 (Peter F. Galderisi, et. al eds., 2001).

(172.) Calcagno & Westley, supra note 171, at 18; see Bruce E. Cain, Party Autonomy and Two-Party Electoral Competition, 149 U. PA. L. REV. 793, 800 (2001).

(173.) Ryan K. Beasley & Mark R. Joslyn, Cognitive Dissonance and Post-Decision Attitude Change in Six Presidential Elections, 22 POL. PSYCHOL. 521, 523 (2001); Sendil Mullainathan & Ebonya Washington, Sticking with Your Vote: Cognitive Dissonance and Voting 22 (Nat'l Bureau of Econ. Research, Working Paper No. W 11910, 2007).

(174.) Trevor Potter & Marianne H. Viray, Barriers to Participation, 36 U. MICH. J.L. REFORM 547, 558 (2003).

(175.) Id. at 558.

(176.) See id. at 558.

(177.) See Ernst Fehr & Simon Gachter, Altruistic Punishment in Humans, 415 NATURE 137 (2002) (explaining theories of cooperative behavior by analyzing a public goods experiment); Ernst Fehr & Urs Fischbacher, The Nature of Human Altruism, 425 NATURE 785, 785 (2003) (explaining that human altruism is a powerful and unique force).

(178.) See Yochai Benkler, Law, Policy, and Cooperation, in GOVERNMENT AND MARKETS: TOWARD A NEW THEORY OF REGULATION (E. Balleisen & D. Moss eds., forthcoming 2009) (manuscript at 6, on file with author).

(179.) See Dan Kahan, The Logic of Reciprocity: Trust, Collective Action, and Law, 102 MICH. L. REV. 71, 79 (2003) (describing that there is a "distribution of cooperative dispositions across the population").

(180.) See ZIVA KUNDA, SOCIAL COGNITION: MAKING SENSE OF PEOPLE 449 (1999) ("Even slight variations in the features of a situation can

lead to dramatic shifts in people's behavior.").

(181.) See ANTHONY DOWNS, AN ECONOMIC THEORY OF DEMOCRACY 260, 271 (1957) (explaining that every rational man decides whether to vote by determining if the returns of voting outweigh the costs).

(182.) See id. at 262. When the cost of voting is zero, everyone who is not indifferent will vote because the return from doing so, though small, is larger than zero. Id. at 262.

(183.) See id. at 267 (explaining that rational men in a democracy are motivated to some extent by a sense of social responsibility relatively independent of their own short run gains and losses).

(184.) Id. at 274 (explaining that one of the returns from voting stems from the realization that democracy cannot function unless many people vote).

(185.) James H. Fowler, Altruism and Turnout, 68 J. POL. 674, 674 (2006).

(186.) Id. at 677.

(187.) Id. at 674.

(188.) See id. at 680 (revealing that those who gave away half of their prize had a 20% higher probability of voting).

(189.) Id. at 680.

(190.) Id. at 681.

(191.) See Michael X. Delli Carpini et al., Public Deliberation, Discursive Participation, and Citizen Engagement.. A Review of the Empirical Literature, 44 AMER. REV. POL. SCI. 315, 324 (2004) (explaining that face-to-face communication is the single greatest factor in increasing the likelihood of cooperation). Face-to-face deliberation is expected to lead to empathy and a broadened sense of people's interests through open-minded argumentation. Id. at 320.

(192.) Id. at 324.

(193.) See id. at 325 (pointing out that minority opinion in the deliberative process can lead to consideration of new perspectives by building empathy).

(194.) Benkler, supra note 178 (manuscript at 9, on file with author).

(195.) Id.

(196.) See Kahan, supra note, 179, at 72.


(198.) PUTNAM, supra note 34, at 41 (explaining that the frequency of virtually every form of community civic involvement has decreased in the last two decades).


(200.) Id. at 24-25.

(201.) See generally id. at 24-36 (explaining the details of the holiday).

(202.) Id. at 26.

(203.) Id. at 36.

(204.) Id. at 17.

(205.) See ACKERMAN & FISHKIN, supra note 199, at 17.

(206.) See id. at 17, 22-23.

(207.) See id. at 44.

(208.) See id.

(209.) Id. at 44 (emphasis in original).

(210.) See id. at 52-54.

(211.) ACKERMAN & FISHKIN, supra note 199, at 55.

(212.) Id. at 73.

(213.) See id. at 227 tbl.A.

(214.) See Alan Ryan, Time Out?, 51 N.Y. REV. OF BOOKS 16 (2004) (reviewing and calling Deliberation Day an "awkward rhetorical marriage between a utopian novel and contemporary political science," and arguing that "Ackerman and Fishkin are nothing if not optimistic").

(215.) David Schkade et al., What Happened on Deliberation Day?, 95 CAL. L. REV. 915 passim (2007).

(216.) See id.

(217.) Id. at 917.

(218.) See Delli Carpini et al., supra note 191, at 325.

(219.) ACKERMAN & FISHKIN, supra note 199, at 64.

(220.) See Chad W. Flanders, Deliberative Dilemmas: A Critique of Deliberation Day from the Perspective of Election Law, 18-20 (2007), student.

(221.) ACKERMAN & FISHKIN, supra note 199, at 23.

(222.) See, e.g., Hasen, supra note 16, at 2135.

(223.) Seeid. at 2153-54,2159.

(224.) Benkler, supra note 178 (manuscript at 11, on file with author).

(225.) See Stephen Knack, Civic Norms, Social Sanctions, and Voter Turnout, 4 RATIONALITY & SOC'Y 133, 137-42 (1992).

(226.) See id. at 139.

(227.) Id. at 142.

(228.) Hasen, supra note 16, at 2160.

(229.) See Stephen Knack & Martha E. Kropf, For Shame! The Effect of Community Cooperative Context on the Probability of Voting, 19 POL.. PSYCHOL. 585 (1998).

(230.) Id. at 589.

(231.) Id.

(232.) Id. at 595.

(233.) Id. at 591 ("The index comprises four activities: volunteer work, charitable giving, working on a community problem, and willingness to serve on a jury.").

(234.) Id. ("Each increase of 5 percentage points in census response corresponds to an increase in the likelihood of voting of about 1 percentage point, on average.").

(235.) Knack & Kropf, supra note 229, at 595.

(236.) Brian D. Silver et al., Who Overreports Voting?, 80 AM. POL. SCI. REV. 613, 613-14 (1986).

(237.) Hasen, supra note 16, at 2160-61.

(238.) Knack & Kropf, supra note 229, at 591-95.


(240.) Hasen, supra note 16, at 2154.

(241.) Id.; see also Michael Schudson, Voting Rites': Why We Need a New Concept of Citizenship, AM. PROSPECT, Fall 1994, at 59-62.

(242.) See Hasen, supra note 16, at 2154 ("With social connectedness usually assured, the norm arose among those groups demanding a norm of voting.").

(243.) See generally Ann E. Carlson, Recycling Norms, 89 CAL. L. REV. 1231 (2001).

(244.) Id. at 1232.

(245.) Id. at 1236, 1296, 1298.

(246.) See id. at 1298-99.

(247.) Id.

(248.) See id. at 1298, 1300.

(249.) Alluding to its public, civic nature, Professors Michael Vandenbergh and Anne Steinemann, in their article on norm activation and carbon emissions, properly identify voting as an activity susceptible to norm management. Michael P. Vandenbergh & Anne C. Steinemann, The Carbon-Neutral Individual, 82 N.Y.U.L. REV. 1673, 1711 (2008) ("In addition to changing carbon-emitting behaviors, norm activation can influence civic behaviors, such as voting and joining advocacy groups.").

(250.) See Carlson, supra note 243, at 1299.


(252.) Id.

(253.) Id. at 112,

(254.) Id.

(255.) Id. at 113.

(256.) Kahan, supra note 179, at 75.

(257.) Cf id. at 75 (pointing out that people who see others contributing in a certain way are more likely to adopt the behavior themselves and thus contribute to the overall increase in the numbers of those who contribute).

(258.) See Donald P. Green et al., Getting out the Vote in Local Elections: Results from Six Door to Door Canvassing Experiments, 65 J. POL. 1083, 1084 (2003).

(259.) GREEN & GERBER, supra note 251, at 157.

(260.) Id. at 139 tbl.10-1.

(261.) Id. at 43.

(262.) Id. at 40.

(263.) See id. at 45.

(264.) See Richard Briffault, Localism and Regionalism, 48 BUFF. L. REV. 1, 22 (2000).

(265.) See Arthur J. Todd, Some Sociological Principles Underlying the Community Chest Movement, 10 SOC. FORCES 476, 477 (1932).

(266.) Id. at 479.

(267.) Id. at 480.

(268.) Alan S. Gerber et al., Social Pressure and Voter Turnout: Evidence from a Large-Scale Field Experiment, 102 AM. POL. SCI. REV. 33, 35 (2008).

(269.) Id. at 33.

(270.) Id. at 37-38.

(271.) Id. at 38.

(272.) Id.

(273.) Id.

(274.) Gerber et al., supra note 268, at 38.

(275.) Id.

(276.) ld.

(277.) See id. at 39-40.

(278.) Id. at 40.

(279.) See James Andreoni & Ragan Petrie, Public Goods Experiments Without Confidentiality: A Glimpse Into Fundraising, 88 J. PUB. ECON. 1605, 1605 (2004).

(280.) GREEN & GERBER, supra note 251, at 113.

(281.) See Fernando Ferreira & Joseph Gyourko, Do Political Parties Matter? Evidence from U.S. Cities, 3 (October 11, 2007) (showing that local politics is less divisive along party lines).

(282.) See id. at 6, 18; see also David Schleicher, Why Is There No Partisan Competition in City Council Elections? The Role of Election Law, 23 J.L. & POL. 419, 465 (2007).

(283.) NIEMI & WEISBERG, supra note 2, at 33.

(284.) See Highton & Wolfinger, supra note 92, at 94.
COPYRIGHT 2009 St. Thomas Law Review
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2009 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Marisam, Jason
Publication:St. Thomas Law Review
Date:Jan 1, 2009
Previous Article:Why the United States Supreme Court got some (but not a lot) of the Sixth Amendment right to counsel analysis right.
Next Article:Potato, potahto: whether ineffective assistance or due process, an effective rule is overdue in termination of parental rights cases in Florida.

Terms of use | Privacy policy | Copyright © 2021 Farlex, Inc. | Feedback | For webmasters |