Hostname: page-component-74656f9884-c52kr Total loading time: 0 Render date: 2026-11-09T01:59:07.562Z Has data issue: false hasContentIssue false
  • English
  • Français

Voter Registration and Turnout in the United States

Published online by Cambridge University Press:  01 September 2004

Benjamin Highton
Benjamin Highton
Affiliation:
Benjamin Highton (bhighton@ucdavis.edu) is assistant professor of political science at the University of California, Davis
Rights & Permissions [Opens in a new window]

Abstract

In a democracy, voting is the most fundamental act of political participation and therefore holds a central location in the study of political behavior. One significant research tradition focuses on the relationship between registration and turnout and raises important and related questions. How do registration laws influence turnout levels? What types of people are most affected by them? What are the partisan implications of registration requirements? Spanning the late nineteenth through early twenty-first centuries, this essay places these questions in the appropriate theoretical context and then answers them.Previously Benjamin Highton was an APSA Congressional Fellow and worked for Senator Paul Wellstone on health and welfare policy. He appreciates the advice and criticism from Henry Brady, Jennifer Hochschild, Randolph Siverson, Raymond Wolfinger, and the anonymous referees for Perspectives on Politics.

Information

Type
Research Article
Information
Perspectives on Politics , Volume 2 , Issue 3 , September 2004 , pp. 507 - 515
Copyright
© 2004 American Political Science Association

In November of 2000, about 106 million Americans voted for their president. What if a registration “fee” of $27.72 had been due the previous May? Undoubtedly, far fewer people would have voted, and the effects would have been especially pronounced among the less well off. This simple logic underlies a substantial body of research that investigates the relationship between registration laws and voter turnout in America.

The figure of $27.72 is the value of the Virginia poll tax, adjusted for inflation, that was adopted in 1904.1

Virginia residents were required to pay an annual poll tax of $1.50 in order to register to vote. Adjusting for inflation, $1.50 in 1904 is equivalent to $27.72 in 2000. See Friedman 2003. The Virginia poll tax could accumulate over three years. Thus, by 1907 a new registrant would have to pay $4.50 or $83.16 in 2000 dollars.

Although they were the most straightforward example of registration “costs,” poll taxes are only one of the various costs that registration requirements have imposed on Americans during the last century. Measuring and explaining how these laws have influenced turnout has been a priority for political scientists and politicians alike. The questions are theoretically engaging and their answers are politically significant.

In this essay, I argue that there are minimal partisan implications of contemporary registration laws and that registration reform has probably reached its limits of enhancing turnout. To reach these conclusions, I review the costs and benefits of voting and explain how registration laws might lower turnout by making the costs of voting greater than the benefits. Then, I assess evidence on the impact of registration laws on overall turnout and on those citizens who are most and least affected by registration provisions. This analysis supports the conclusion that those who hope to increase voter turnout substantially in the United States must look beyond registration.

The Benefits and Costs of Voting

Why vote? People vote when the benefits of doing so outweigh the costs. This means that there are really two questions to address: What are the benefits of voting, and what are the costs?

Voting benefits

Any act of political participation might provide people with instrumental, expressive, or intrinsic benefits. In terms of voting, coherent explanations should not depend on the instrumental rewards. The reason for this is straightforward: Any single vote “is lost in a sea of other votes,”2

Downs 1957, 246.

so that “my chances of determining who becomes President [or influencing just about any election outcome] are of about the same order of magnitude as my chances of being killed driving to the polls—hardly a profitable venture…. [Voting] is irrational as an instrumental action toward an egocentric end.”3

Meehl 1977, 11. The instrumental reward of voting depends on the product of the probability of influencing the outcome (P) and the “differential benefit” (B) one would gain from one candidate prevailing over another, P*B. See Riker and Ordeshook 1968. Theoretically, larger ideological differences between candidates and higher political offices (because they come with more power to influence policy outcomes) increase the differential benefit. But, the minute chance of influencing an election outcome (P) means that even with a high differential benefit, the costs of voting outweigh the instrumental rewards for almost everyone.

Because “there are so few concrete payoffs to voting,”4

people must receive expressive or intrinsic benefits; as John Aldrich puts it, “most of the action is, in fact, in the intrinsic values of voting per se.”5

Aldrich 1993, 266. Or, as Wolfinger and Rosenstone explain, the “more important benefits of voting, however, are expressive rather than instrumental.” Wolfinger and Rosenstone 1980, 7.

Voting may bring the “feeling that one has done one's duty to society, to a reference group (Democrats, blacks, bankers, liberals, feminists, conservationists, and so forth), and to oneself; or the feeling that one has affirmed one's allegiance to or efficacy in the political system.”6 In addition, just as some people like the opera or sports events, some like voting. As William Riker and Peter Ordeshook observe,7 “for those who enjoy the act of informing themselves for the decision, who get social satisfactions out of going to the polling booth … [some] supposed costs are actually benefits.”8

For a more extensive discussion of how, for some, “bearing the cost becomes part of the benefit,” see Verba, Schlozman, and Brady 1995, 103.

Although these expressive and intrinsic benefits are easy to identify, measuring their magnitudes precisely (How good does it feel to vote?) is probably impossible. Still, there is little doubt that for nearly everyone, voting is a low benefit activity.9

Aldrich provides the clearest articulation of this observation along with its implications. See Aldrich 1993. Verba, Schlozman, and Brady (1995) analyze a variety of forms of political participation (campaign work, financial contributions, protesting, et cetera) and explain why voting provides the fewest benefits of them all. The fact that voting is the most common form of political participation among Americans is explained by its low costs (and low resource demands), not its high benefits.

Politics, in general, is a peripheral concern to most Americans.10 Even among the politically engaged and interested, the positive feelings of fulfilling one's civic duty (or the guilt for shirking it) are not very strong and do not last much past election day. Moreover, the opportunities for expressive and intrinsic voting benefits are neither rare nor unpredictable because American elections occur at regular and frequent intervals. In addition, because of the “big brute fact” that no single vote is decisive (even in a case like the 2000 presidential election), the electoral implications for not having voted in a particular election are largely inconsequential. All of this means that for most people, voting or not voting in any given election does not affect one's subjective or objective well being, in either a positive or negative way, very much.

Voting costs

The observation that voting is a low-benefit activity leads to three propositions. First, even small voting costs may deter people from voting. Second, given that many people do cast ballots, the costs of voting for nearly everyone must not be very high. Third, as a “low-benefit, low-cost activity,” in general, small changes in the costs of voting might have sizable effects on overall turnout rates and influence the turnout of some groups more than others.11

Since its near-universal adoption, voter registration has been perhaps the main cost of voting in America.12

North Dakota is the only state that currently does not require registration. It abolished registration requirements in 1951. Instead, precincts maintain lists of voters from previous elections. Those not on the lists who show up to vote “may be asked to sign a sworn affidavit swearing to the fact that he or she is a qualified elector of the precinct and therefore qualified to vote.” See Jaeger 2002, 1. Despite the obvious potential for mischief, the North Dakota Secretary of State reports that there have been no incidents of “widespread voter fraud” in North Dakota. See Jaeger 2002, 2.

While in many other countries, the government is responsible for registering voters, individuals in the United States have always had to “pay” the costs themselves. Writing more than seventy years ago, Harold Gosnell linked this difference to lower voter turnout.

[T]here are certain common elements running through the election systems studied which make voting an easier task than it is in this country…. In the European countries studied, a citizen who is entitled to vote does not, as a rule, have to make any effort to see that his name is on the list of eligible voters. The inconvenience of registering in this country has caused many citizens to become non-voters…. In France, Germany, Belgium, and Switzerland the local officials check up at least once a year in order to see that the voters' lists are accurate and up to date.13

Because registration in America has been the responsibility of potential voters, both the magnitude and timing of registration costs influence turnout. Consider the poll tax. Clearly, higher poll taxes represent greater costs and consequently cause lower turnout. In addition, if two poll taxes are equal in amount, but one is payable on election day while the other is due months in advance, the latter will more greatly affect turnout. As V. O. Key wrote,

[M]any persons to whom the dollar or two is no burden do not pay the tax because of conditions surrounding its collection…. When the tax is due six months or more in advance of a campaign, before the candidates have announced and before political interest is aroused, the natural result is a smaller degree of payment than would occur if collection continued until a shorter time before the voting.14

Key 1949, 585.

Whatever the costs associated with registration, people will be more interested and therefore more likely to deal with them closer to election day.

Registration is significant not only because it influences overall turnout levels, but also because registration laws may affect some people more than others. If this is the case, registration may account for some of the “distortion” between the voting population and the general population (i.e., voters are older, better educated, wealthier, whiter, etc.).15

For instance, Powell concludes that “it seems very likely, although we cannot demonstrate it directly, that the difficulty of registration in America is … responsible for this remarkable distinctiveness of American voting processes.” See Powell 1986, 31.

Similarly, if the types of people most affected by registration laws tend to share party affiliations, registration laws may have important partisan implications as well.

Registration Laws and Voter Turnout

Beginning with Robert Erikson, scholars have noted that among those who register, turnout rates are extremely high and “even unlikely registrants are relatively frequent voters when they do register.”16

For example, in the last three presidential elections, fully 85 percent of those who report being registered voted.17

This figure is based on my own analysis of the Current Population Survey (CPS) Voter Supplements for 1992, 1996, and 2000, which are available through the Inter-university Consortium for Political and Social Research.

It is thus tempting to infer that registration is the chief barrier to voting.18

Piven and Cloward argue that “people vote if they are registered. Nonvoting is almost entirely concentrated among those who are not registered. This is prima facie evidence of the deterrent impact of registration procedures on voting.” See Piven and Cloward 1988, 260. In their more recent work, Piven and Cloward back away from their strong claims on the impact of registration laws on turnout. See Piven and Cloward 2000.

This, however, is an important instance of a selection problem. Research that models registration separately from voting shows that registered individuals are better educated, older, politically interested, and more socially connected than non-registered citizens.19

Erikson 1981; Uhlaner 1989; Mitchell and Wlezien 1995; Jackson 1996; Timpone 1998. By separating the registration decision from the turnout one, scholars are better able to understand why some factors influence turnout. For example, jury aversion may lower turnout through lower registration rates because jurors are sometimes selected from registration lists. See Oliver and Wolfinger 1999. In contrast, campaign mobilization affects turnout by increasing the probability of voting among those who are already registered. See Jackson 1996, 2002. Brown, Jackson, and Wright 1999.

Because these factors are associated with lower voting costs and increased benefits, it is a mistake to argue that if registration barriers were removed, non-registered citizens would vote at the same high rate as those who are registered.20

Consider the analogy of Ivy League admissions and graduation rates. Few, if any, would argue that high Ivy League graduation rates (upwards of ninety percent) imply that those who were not admitted would graduate at the same high rate if they had been admitted. The reason is simple: admissions committees select students who they think will graduate. If they admitted a class that was representative of the overall pool of high school graduates (i.e., those who are eligible to attend Ivy League schools), graduation rates would certainly fall. Similarly, those who register have attributes that make them likely voters. Just as one would not infer that those who are not accepted to Ivy League schools would graduate if they were accepted, one should not infer that those who are not registered would vote if they were registered.

This point may be illustrated by examining turnout in North Dakota, which does not have voter registration. If registration is the primary factor that keeps people from voting, turnout among North Dakotans should approach one hundred percent, or at least the turnout rate among those who are registered in other states. But, turnout in North Dakota lags the turnout of the registered in other states.21

An alternative approach to assessing the effects of registration laws on turnout takes advantage of the substantial variation in registration requirements across states and over time. Analyses of this sort reveal that of all the myriad registration laws, four stand out for their impact on turnout: poll taxes, literacy tests, registration closing dates, and “motor voter” laws.

Poll taxes and literacy tests

Key, Morgan Kousser, and Jerrold Rusk and John Stucker tell the history of the poll taxes and literacy tests that were enacted in southern states beginning in 1890.22

Each of the eleven former Confederate states enacted poll taxes between 1890 and 1904, and seven adopted literacy tests.23

Although the South receives most of the attention, literacy tests were used elsewhere as well. As Riker put it: “Eighteen states have adopted the test, seven to disfranchise Negroes, five to disfranchise Indians and Mexicans and Orientals, and six to disfranchise European immigrants.” See Riker 1965, 60.

All poll taxes required some payment in order to register, and the tax often accumulated over time. Literacy tests required potential voters to demonstrate either reading or writing skills, though many states had some form of exemption, such as a “grandfather clause.” Both poll taxes and literacy tests impose costs on potential voters in advance, often well in advance, of election day.

Distinguishing the impact of poll taxes and literacy tests on turnout is difficult because both measures were often adopted at or near the same time, amid already declining southern turnout. In addition, people in the 1890s paid less attention (by today's standards) to collecting and maintaining the sort of data that are useful to social scientists. Nevertheless, there is little doubt that the adoption of such laws contributed to a significant decline in turnout. Rusk and Stucker find, for example, that the rate of turnout decline increased by an average of 15 percentage points after states adopted poll taxes and 9 points after they adopted literacy tests.24

I calculate these figures based on Rusk and Stucker's tables 6.13 and 6.14. They represent averages across all states that adopted poll taxes and literacy tests. Because of the often overlapping adoptions, the figures should not be considered as the independent effects of each provision (i.e., one should not conclude that the combined effect of a poll tax and literacy test would be to reduce turnout 24 percentage points). In addition, no two poll taxes or literacy tests were identical, so one would not want to say that a poll tax or literacy test reduces turnout by X percent. Higher poll taxes and more difficult literacy tests will lower turnout more than lower taxes and easier tests. See Rusk and Stucker 1978.

Blacks and poor whites were the main groups disenfranchised by poll taxes and literacy tests. Although no law explicitly mentioned blacks, targeting them was easy. Writing about the Mississippi state constitution, Key observes that “all the suffrage provisions … were phrased to exclude from the franchise not Negroes, as such, but persons with certain characteristics most of whom would be Negroes.”25

Key 1949, 538.

Indeed, most blacks in the South were poor and illiterate. Moreover, they could not expect to benefit from the discretion local (white) officials often had regarding literacy tests and their various “exemptions” and “alternatives.”26

Key's analysis led him to conclude that “no matter from what direction one looks at it, the southern literacy test is a fraud and nothing more. The simple fact seems to be that the constitutionally prescribed test of ability to read and write a section of the constitution is rarely administered to whites.” See Key 1949, 576.

Steven Rosenstone and John Mark Hansen, and James Alt confirm the disparate racial impact of these laws.27 The poll tax kept many poor whites from voting as well. In fact, given the many other methods used to disenfranchise blacks, Key argues that the poll tax likely had a larger effect on poor whites than on blacks.28

The significance lies in a sometimes overlooked aspect of southern politics. Maintaining one-party Democratic dominance meant it was necessary to exclude as much as possible those who might hold or develop sympathies for Republican or Populist candidates. Blacks were obviously one such constituency. Poor whites were another.

Registration closing dates and election day registration

With the adoption of registration laws, states required citizens to register in advance of election day. Early registration closing dates provide less time to register and may limit the mobilizing effects of campaigns, which grow more intense as election day nears. In addition, requiring registration before election day usually entails a separate trip to register that could “require a longer journey, at a less convenient hour, to complete a more complicated procedure—and at a time when interest in the campaign is far from its peak.”29

Stanley Kelley, Richard Ayres, and William Bowen conducted the first systematic study of closing dates.30

They analyzed registration and turnout rates across 89 cities in 1960 and found that a closing date more than three months before election day lowered registration levels by as much as 15 percentage points. Steven Rosenstone and Raymond Wolfinger analyzed the direct connection between closing dates and turnout among individuals.31 By the time the data they analyzed were collected in 1972, nearly all the states had closing dates no longer than a month before election day. Even with these shortened periods, Rosenstone and Wolfinger found that the effect of closing dates on turnout could be as much as nine percentage points.32

The relationship between closing dates and turnout is quite robust. It has been replicated in analyses of turnout in different election years using different methodologies and a variety of data. See Squire et al. 1987; Nagler 1991, 1994; Teixeira 1992; Rosenstone and Hansen 1993; Mitchell and Wlezien 1995; Rhine 1995; Rhine 1996; Highton and Wolfinger 1998; Timpone 1998; Brians and Grofman 2001.

Researchers could directly investigate the effect of having no closing date after Maine, Minnesota, and Wisconsin adopted election day registration in the 1970s. Four more states followed. Oregon allowed election day registration in the early 1980s before later repealing it. New Hampshire, Wyoming, and Idaho adopted election day registration in response to the 1993 National Voter Registration Act. Through election day registration, these states allowed campaigns to realize their mobilizing potential fully and at the same time virtually eliminated the costs of registration. The evidence reveals a noticeable boost in turnout of about five percentage points.33

What types of people are likely to take advantage of the opportunity to register closer to election day or on election day itself? One “type” is an individual who in the past has had fewer opportunities to register. Consider two people, alike in nearly every way except that one is twenty years old and the other is sixty. With the approach of the 2004 presidential election, the young person has the opportunity to vote in her first election. To do so, of course, she will need to register. Having had few opportunities to register in the past, it seems more likely that her desire to register would manifest during the fall campaign, perhaps near the end when media coverage is greatest and party mobilization efforts reach their peak. For her, a short closing date might be the difference between registering and not. In contrast, the sixty-year-old has lived through many more election seasons and has had many more opportunities to register. One more opportunity will likely have less impact on whether this person registers and votes. Consequently, early closing dates are likely to have their largest effects among the young.

In a related vein, one would predict larger effects for the residentially mobile. Moving requires people to register with their new address in order to vote. Movers, then, are like twenty year-olds: an upcoming election is the first for which they are eligible to vote (at their new address). Because reregistering at a new address is often low on recent movers' priority lists, a political campaign can provide the needed motivation.34

“People in a new home have all sorts of arranging and adjusting to do, from redecorating to mastering innumerable details of domestic administration…. Registering to vote is drab and boring, a weak claimant for attention.” See Squire, Wolfinger, and Glass 1987, 50.

Short closing dates or election day registration enable these individuals to translate campaign interest into voting.

Political interest also matters for considering the turnout effects of closing dates. People who are most interested in politics are very likely to make sure they are registered. Only rarely will they fail to register by the waning weeks of a national campaign. As a result, closing dates influence the turnout of these highly motivated people very little. Those least interested in politics are also unlikely to be influenced by closing dates. These citizens have virtually no motivation to vote; their voting benefits are nearly zero. They pay little, if any, attention to political campaigns and are therefore unlikely to be activated by them. Late closing dates, or even election day registration will not bring these people to register and vote.

Between these extremes are individuals who take some interest in politics, and who may be spurred to register and vote by the increased campaign interest that attends the approach of election day. A late closing date allows for this possibility. If the deadline for registration is well before election day, however, it is unlikely that campaign interest will be translated into turnout. For this group of people, registration closing dates ought to matter more.

Evidence confirms that closing dates have disparate turnout effects. Both the young and the residentially mobile are more responsive to registration closing dates and election day registration.35

With regard to political interest, I am aware of no study that has directly investigated its interaction with registration closing dates. However, a number of studies look at the interaction between education—a good proxy for interest—and closing dates. The evidence suggests that those with either a high school degree or some college education are most affected. Those without a high school degree—the least interested—and those who completed college—the most interested—are influenced less.36

Nagler 1991; Nagler 1994; Highton and Wolfinger 1998; Brians and Grofman 2001. This finding is at odds with those of Rosenstone and Wolfinger who found the largest effects among the least educated. See Rosenstone and Wolfinger 1978. Nagler shows how this finding was an artifact of the methodology they employed. See Nagler 1991; Nagler 1994.

“Motor voter” registration

“Motor voter” laws are the last type of registration provisions that research consistently shows to affect turnout levels. Beginning in the 1970s, some states started adopting laws that allowed voter registration at motor vehicle agencies. By the beginning of 1992, two dozen states had some form of motor voter program.37

Because most people have driver's licenses or state identification cards, which they periodically have to renew, linking registration to the application and especially the renewal process lowers registration costs. Registration becomes attached to an activity that people will engage in anyway; the need for a separate trip and effort to register are thereby removed. As a result, motor voter laws can come “close to eliminating registration as a separate activity with its own costs.”38

As states formulated and implemented motor voter laws, an important distinction emerged. Some states either used a combined registration/licensing form or had DMV personnel explicitly ask people if they wanted to register. These “active” motor voter states were distinguished from “passive” ones where registration was not effectively integrated into the ordinary course of DMV transactions; forms were available but no prompting was provided.39

Not surprisingly, “active” programs had more substantial effects on turnout—about five percentage points—than “passive” ones, the effects of which were not always possible to distinguish from zero.40

Like election day registration, an active motor voter program removes the costs associated with a separate trip to register.41

Moreover, by offering an easy opportunity to register in advance of election day, motor voter laws can further enhance turnout by facilitating political mobilization because parties and other groups often rely on lists of registrants. The same is not true for election day registrants.

As a consequence, it should particularly benefit the young, the residentially mobile, and those with middling levels of education. Highton and Wolfinger tested two of these hypotheses by analyzing turnout in Colorado before and after it implemented an active motor voter program.42 The results generally confirmed the age and education predictions. No measure of residential mobility was available, so it was impossible to test the hypothesis about movers.

Other registration provisions

In addition to poll taxes, literacy tests, registration closing dates, and motor voter programs, states have implemented a variety of other registration laws. For example, some allowed mail registration; some required registration offices to be open regularly for at least 40 hours a week or on evenings and weekends; some required registered voters to reregister at periodic intervals; some purged nonvoters from registration lists after a specified amount of time. All of these provisions relate to registration costs. None, however, exhibits clear and consistent effects across the wide range of empirical analyses of the turnout effects of registration laws.43

The minimal effects, if any are discernable at all, suggest that the costs associated with these laws are too small to influence turnout very much.

Another kind of registration law prevents a whole class of individuals from registering. For example, many states had residency requirements until the latter part of the twentieth century. Rosenstone and Wolfinger report that in “the early 1960s, 38 states required at least a year's residence in the state before one could register.”44

For those who had not lived in a state long enough, the residency requirement was an absolute barrier to voting. At the same time, though, the negative effect of residency requirements on overall turnout levels was limited because nearly everyone (even most of the residentially mobile) met the requirements.45

For instance, the Census Bureau reports that over a one-year period in the 1960s, 96 percent of people remained in the same state. See U.S. Census Bureau 1964. Only four percent of the population moved from one state to another. Among those who moved (about 20 percent of the population), slightly more than 80 percent moved within their state of residence.

Nevertheless, these restrictions effectively kept recent immigrants from voting. Similarly, felon disenfranchisement laws keep convicted felons from registering and voting, in some cases even after release from prison. The aggregate turnout effects are small; only about one in fifty adults is unable to vote, either currently or permanently, because of a felony conviction.46 Nonetheless, the effects are far from uniform; for instance, black men are disenfranchised at a rate seven times the national average.47

Ibid. Uggen and Manza conduct an extensive analysis of the political implications of felon disenfranchisement laws. See Uggen and Manza 2002.

Partisan Consequences of Registration Laws

The history of the Jim Crow South illustrates the great possibilities for partisan effects of registration laws. Once the Union Army left, the South maintained one-party Democratic rule in part by keeping those least likely to support the Democrats (poor whites and especially blacks) from registering to vote.48

Key documents a variety of other ways by which one-party dominance was maintained. See Key 1949.

Against this historical backdrop, it is not surprising that scholars have investigated the partisan consequences of other registration laws.

The partisan impact of a registration law depends on two factors: the number of people affected by a law and their political distinctiveness. If the effect of a registration law is limited to either a small number of people or to a group that resembles the voting population in terms of partisanship, the partisan consequences will be minimal. Poll taxes and literacy tests in the South illustrate the importance of these factors. Those who were kept from voting by these laws (blacks and poor whites) formed a sizable proportion of the population, and their political inclinations differed greatly from those among voters.

In contrast, today's registration laws appear to have much smaller partisan effects. Scholars have investigated the partisan implications of changing registration laws and consistently found little, if any, noticeable effects.49

Although these findings are generally at odds with the views of many journalists, politicians, and even some political scientists, they are understandable.

Several factors explain the minimal partisan effects. First, consider the differences between voters and nonvoters in general and what might happen if the United States saw universal turnout. On the whole, there are small differences between voters and nonvoters in terms of their partisan and policy preferences. The earliest estimate of universal turnout showed that if everyone had voted in 1972, the partisan balance would have barely changed and that on a range of issues “voters are a virtual carbon copy of the citizen population.”50

Subsequent studies report similar results.51 Given that “as a whole, nonvoters appear well represented by those who vote” and that universal turnout would represent a substantially larger turnout increase than would changes in registration laws, one would expect the partisan effects of registration laws to be even less significant than the small estimated effects of universal turnout.52

There are two caveats about the partisan implications of registration laws, but they do not change the prediction of minimal partisan effects. First, whereas the work cited above relates to the nation as a whole, there might be larger effects in state or local elections. This, however, is unlikely to be the case. Studies of the relationship between turnout and election outcomes for a variety of offices generally find little, if any, partisan effects.53

Higher turnout does not consistently benefit either party significantly; for example, full turnout in Senate elections would have benefited Democrats by just about a single percentage point, on average, in the elections between 1994 and 1998.54

The second caveat relates to the characteristics of those disproportionately affected by registration laws. If they are politically distinct, the partisan effects would be larger. As I have discussed, registration closing dates and motor voter laws have greater impact on turnout among the young and residentially mobile, the largest blocs of nonvoters.55

Current Population Survey Voter Supplement data for the 2000 election indicates that 33 percent of nonvoters were under the age of 30, 41 percent had lived at their current address for two years or less, and 55 percent were either residentially mobile or young.

Neither group's partisan preferences differ much from those of the rest of the population, however, which means that adding a disproportionate number of them to the ranks of voters through registration reform would barely alter the partisan balance of the electorate.56 A similar lack of political distinctiveness is evident for those with middling education levels. In short, the partisan preferences of marginal voters, those for whom variations in registration laws matter most, closely mirror those of voters.

Contemporary Registration Laws and Voter Turnout

Registration was widely adopted as a requirement for voting between 1876 and 1912.57

Since then, the long-term trend has been toward making registration a less imposing and onerous task. The 24th Amendment, ratified in 1964, outlawed poll taxes for federal elections. In 1966 the Supreme Court extended the prohibition to all elections when it ruled that a poll tax violated the Equal Protection Clause of the 14th Amendment.58 Literacy tests have also been left in the past, eliminated by the Voting Rights Act of 1965 and Voting Rights Act Amendments of 1970. Registration closing dates, although still in existence, have been substantially curtailed through the 1970 Voting Rights Act Amendments and a Supreme Court decision.59

Dunn v. Blumstein.

Whereas Kelley, Ayres, and Bowen's analysis of turnout in 1960 included people living in states with closing dates of one hundred or more days before election day, Rosenstone and Wolfinger's analysis of turnout in 1972 included just four states with closing dates greater than 31 days.60

Kelley, Ayres, and Bowen 1967; Rosenstone and Wolfinger 1978. The 1970 Voting Rights Act Amendments and Dunn v. Blumstein also restricted residency requirements.

At present, no state has a closing date greater than 31 days, and the most populous state in the country has recently cut its closing date in half, from 30 to 15 days before an election. In addition, as I mentioned, motor voter laws started to proliferate in the 1970s.61

Commenting on all of the changes, especially the congressional actions and Supreme Court decisions, Keyssar writes:

By 1975, the nation had witnessed a legal revolution…. What occurred in the course of a decade was not only the re-enfranchisement of African Americans but the abolition of nearly all remaining limits on the right to vote. Poll taxes, literacy tests, understanding clauses, pauper exclusions, and good character provisions had been swept away. Property and tax requirements for voting in special elections had been all but eliminated; durational residency qualifications had been drastically cut and the definition of residency broadened; the voting age had been lowered to eighteen, and language barriers had been dropped. The total number of new voters added to the electorate cannot be counted with precision, but the figure was surely in excess of twenty million. See Keyssar 2000, 281–82.

At the same time, Keyssar identifies “two uneasy pieces,” ongoing questions relating to the rights of felons (and ex-felons) and immigrants.

The most significant recent changes to registration laws began with passage of the National Voter Registration Act (NVRA) of 1993. The law included four registration provisions: “active” motor voter, public agency registration, universal mail registration, and prohibition of “purging for not voting.”62

The law requires that a “motor vehicle driver's license application (including any renewal application) … shall serve as an application for voter registration. See 107 Stat. 68, 1993.

The only way a state could be exempt from these requirements is if it allowed election day registration. Many groups hailed the legislation as a key step toward removing barriers to voting. Alexander Keyssar writes that:

The registration measure was the final act of the drama that had begun in the 1960s: it completed a lurching yet immensely important forty-year process of nationalizing the voting laws and removing obstacles to the ballot box. As such, the Motor Voter bill was also a critical step in dismantling the multiple impediments to voting that had been erected between the 1850s and World War I. By the end of the twentieth century, what had been a long historical swing toward contraction of the franchise had been decisively reversed.63

In terms of enhancing turnout, the key provision was the requirement that every state implement an active motor voter program or permit election day registration. In essence, the NVRA eliminated the costs of a special trip to register, and thereby virtually eliminated the costs of registration for nearly all eligible voters.

Ironically, in the first presidential election after the adoption of the NVRA, turnout dropped by about six percentage points from its 1992 level, hardly an encouraging sign. This probably was due more to unusually high turnout in 1992 than to any negative effects of the NVRA on turnout, especially given that the law took effect in 1995 and license renewal cycles are typically four or five years. Still, turnout in the 1998 and 2002 midterm and the 2000 presidential elections was not substantially higher than pre-NVRA levels (even if we set aside the 1992 election and make comparisons to presidential turnout in the 1980s).

What can be inferred from the post-NVRA turnout rates? The individual costs of registration are now lower than they have been at any time since the widespread adoption of state registration requirements. And, given that all states are now required to have motor voter programs or election day registration, registration costs cannot go much lower.64

To be sure, there is still some room for improvement. In its most recent report to Congress, the Federal Election Commission (2001) notes that some states have been having implementation problems with their motor voter laws.

Thus, it is no longer reasonable to sustain the proposition, first articulated by Gosnell (1930), that American registration laws are responsible for lower turnout in the United States compared to other countries. Nor is it reasonable to contend that registration requirements are the primary cause of the demographic skew of the voting population. If either of these propositions were correct, American turnout would be substantially higher and the demographic correlates of turnout would have weakened considerably. Neither of these outcomes has been observed, which suggests that the search for causes must extend beyond registration laws.

Conclusion

In the United States, registration has been and remains an individual responsibility. People are not automatically registered when they first become eligible to vote or take up a new residence. However, the government can make registration easier (or more difficult) with predictable consequences for turnout. At present, due to earlier reforms and the National Voter Registration Act of 1993, registration laws are more uniform and registration costs are lower than at any point since registration requirements became widely adopted. There is now little room for enhancing turnout further by making registration easier. It is still possible to lower voting costs somewhat by extending polling place hours and mailing sampling ballots and polling place information to registrants.65

Wolfinger, Highton, and Mullin, forthcoming. Although unlikely to be widely adopted, conducting elections by mail is another means to reduce voting costs and increase turnout. See Berinsky, Burns, and Traugott 2001.

Furthermore, efforts to mobilize voters, especially through personal canvassing, can be effective.66 Nevertheless, continued nonvoting by substantial numbers of citizens suggests that for many people, voting remains an activity from which there is virtually no gratification—instrumental, expressive, or otherwise. Consequently, for those whose goal is a democracy where most people engage in the fundamental act of political participation, a pessimistic conclusion cannot be avoided.

References

REFERENCES

Aldrich, John H. 1993. Rational choice and turnout. American Journal of Political Science 37 (1): 246–78.Google Scholar
Alt, James E. 1994. The impact of the Voting Rights Act on black and white voter registration in the South. In Quiet revolution in the South: The impact of the Voting Rights Act, 1965–1990, ed. Chandler Davidson and Bernard Grofman, 351–77. Princeton: Princeton University Press.
Bennett, Stephen Earl, and David Resnick. 1989. The implications of nonvoting for democracy in the United States. American Journal of Political Science 34 (3): 771–802.Google Scholar
Berinsky, Adam J., Nancy Burns, and Michael W. Traugott. 2001. Who votes by mail: A dynamic model of the individual-level consequences of voting-by-mail systems. Public Opinion Quarterly 65 (2): 178–97.Google Scholar
Brady, Henry E., Sidney Verba, and Kay Lehman Schlozman. 1995. Beyond SES: A resource model of political participation. American Political Science Review 89 (2): 271–94.Google Scholar
Brians, Craig Leonard, and Bernard Grofman. 2001. Election day registration's effect on U.S. voter turnout. Social Science Quarterly 82 (1): 170–83.Google Scholar
Brown, Robert D., Robert A. Jackson, and Gerald C. Wright. 1999. Registration, turnout, and state party systems. Political Research Quarterly 52 (3): 463–79.Google Scholar
Citrin, Jack, Eric Schickler, and John Sides. 2003. What if everyone voted? Simulating the impact of increased turnout in senate elections. American Journal of Political Science 47 (1): 75–90.Google Scholar
Delli Carpini, Michael X., and Scott Keeter. 1996. What Americans know about politics and why it matters. New Haven: Yale University Press.
DeNardo, James. 1980. Turnout and the vote: The joke's on the Democrats. American Political Science Review 74 (2): 406–20.Google Scholar
Downs, Anthony. 1957. An economic theory of democracy. New York: HarperCollins.
Dunn v. Blumstein, 405 U.S. 330 (1972).
Erikson, Robert S. 1981. Why do people vote? Because they are registered. American Politics Quarterly 9 (3): 259–76.Google Scholar
Federal Election Commission. 2001. The impact of the National Voter Registration Act of 1993 on the administration of elections for federal office. Report to the 107th Congress.
Fenster, Mark J. 1994. The impact of allowing day of registration voting on turnout in U.S. elections from 1960 to 1992. American Politics Quarterly 22 (1) 74–87.Google Scholar
Franklin, Daniel P., and Eric E. Grier. 1997. Effects of motor voter legislation: Voter turnout, registration, and partisan advantage in the 1992 presidential election. American Politics Quarterly 25 (1): 104–17.Google Scholar
Friedman, S. Morgan. 2003. The inflation calculator. www.westegg.com/inflation/.
Gerber, Alan S., and Donald P. Green. 2000. The effects of canvassing, direct mail, and telephone contact on voter turnout: A field experiment. American Political Science Review 94 (3): 653–63.Google Scholar
Gosnell, Harold F. 1930. Why Europe votes. Chicago: University of Chicago Press.
Harper v. Virginia Board of Elections, 383 U.S. 663 (1966).
Highton, Benjamin. 1997. Easy registration and voter turnout. Journal of Politics 59 (2): 565–75.Google Scholar
Highton, Benjamin, and Raymond E. Wolfinger. 1998. Estimating the effects of the National Voter Registration Act of 1993. Political Behavior 20 (2): 79–104.Google Scholar
Highton, Benjamin, and Raymond E. Wolfinger 2001. The political implications of higher turnout. British Journal of Political Science 31 (1): 179–92.Google Scholar
Jackson, Robert A. 1996. A reassessment of voter mobilization. Political Research Quarterly 49 (2): 331–49.Google Scholar
Jackson, Robert A. 2002. Gubernatorial and senatorial campaign mobilization of voters. Political Research Quarterly 55 (4): 825–44.Google Scholar
Jaeger, Alvin A. 2002. North Dakota…. The only state without voter registration. North Dakota Secretary of State, Elections Division. http://www.state.nd.us/sec/forms/pdf/votereg.pdf.
Kelley, Stanley, Jr., Richard E. Ayres, and William G. Bowen. 1967. Registration and voting: Putting first things first. American Political Science Review 61 (2): 359–79.Google Scholar
Key, V. O., Jr. 1949. Southern politics in state and nation. New York: Alfred A. Knopf.
Keyssar, Alexander. 2000. The right to vote: The contested history of democracy in the United States. New York: Basic Books.
Knack, Stephen. 1995. Does “motor voter” work? Evidence from state-level data. Journal of Politics 57 (3): 796–811.Google Scholar
Knack, Stephen 1999. Drivers wanted: Motor voter and the election of 1996. PS: Political Science and Politics 32 (2): 237–43.Google Scholar
Knack, Stephen 2001. Election-day registration: The second wave. American Politics Quarterly 29 (1): 65–78.Google Scholar
Knack, Stephen, and James White. 2000. Election-day registration and turnout inequality. Political Behavior 22 (1): 29–44.Google Scholar
Kousser, J. Morgan. 1974. The shaping of southern politics: Suffrage restriction and the establishment of the one-party South, 1880–1910. New Haven: Yale University Press.
Meehl, Paul E. 1977. The selfish voter paradox and the thrown-away vote argument. American Political Science Review 71 (1): 11–30.Google Scholar
Mitchell, Glenn E., and Christopher Wlezien. 1995. The impact of legal constraints on voter registration, turnout, and the composition of the American electorate. Political Behavior 17 (2): 179–202.Google Scholar
Montjoy, Robert S. 1992. Innovations in election administration: Motor voter registration programs. Washington, DC: National Clearinghouse on Election Administration, Federal Election Commission.
Nagle, Jack H., and John E. McNulty. 1996. Partisan effects of voter turnout in senatorial and gubernatorial elections. American Political Science Review 90 (4): 780–93.Google Scholar
Nagler, Jonathan. 1991. The effect of registration laws and education on U.S. voter turnout. American Political Science Review 85 (4): 1393–1405.Google Scholar
Nagler, Jonathan 1994. Scobit: An alternative estimator to Logit and Probit. American Journal of Political Science 38 (1): 230–55.Google Scholar
National Clearinghouse on Election Administration. 1992. Fast facts on state registration and election procedures. Technical report 2. Washington, DC: Federal Election Commission.
Oliver, J. Eric, and Raymond E. Wolfinger. 1999. Jury aversion and voter registration. American Political Science Review 93 (1): 147–52.Google Scholar
Piven, Frances Fox, and Richard A. Cloward. 1988. Why Americans don't vote. New York: Pantheon Books.
Piven, Frances Fox, and Richard A. Cloward 2000. Why Americans still don't vote: And why politicians want it that way. Boston: Beacon Press.
Powell, G. Bingham, Jr. 1986. American voter turnout in comparative perspective. American Political Science Review 80 (1): 17–43.Google Scholar
Rhine, Staci L. 1995. Registration reform and turnout change in the American states. American Politics Quarterly 23 (4): 409–26.Google Scholar
Rhine, Staci L. 1996. An analysis of the impact of registration factors on turnout in 1992. Political Behavior 18 (2): 171–85.Google Scholar
Riker, William H. 1965. Democracy in the United States. 2nd ed. London: Macmillan.
Riker, William H., and Peter C. Ordeshook. 1968. A theory of the calculus of voting. American Political Science Review 62 (1): 25–43.Google Scholar
Rosenstone, Steven J., and John Mark Hansen. 1993. Mobilization, participation, and democracy in America. New York: Macmillan.
Rosenstone, Steven J., and Raymond E. Wolfinger. 1978. The effect of registration laws on voter turnout. American Political Science Review 72 (1): 22–45.Google Scholar
Rusk, Jerrold G., and John J. Stucker. 1978. The effect of the southern system of election laws on voting participation. In The history of American electoral behavior, ed. Joel H. Silbey, Allan G. Bogue, and William H. Flanigan, 198–250. Princeton: Princeton University Press.
Sentencing Project. 2003. Felony disenfranchisement laws in the United States. www.sentencingproject.org/pdfs/1046.pdf.
Squire, Peverill, Raymond E. Wolfinger, and David P. Glass. 1987. Residential mobility and voter turnout. American Political Science Review 81 (1): 45–65.Google Scholar
Teixeira, Ruy A. 1992. The disappearing American voter. Washington, DC: Brookings Institution.
Timpone, Richard J. 1998. Structure, behavior, and voter turnout in the United States. American Political Science Review 92 (1): 145–58.Google Scholar
Uggen, Christopher, and Jeff Manza. 2002. Democratic contraction? Political consequences of felon disenfranchisement in the United States. American Sociological Review 67 (6): 777–803.Google Scholar
Uhlaner, Carole Jean. 1989. Turnout in recent American presidential elections. Political Behavior 11 (1): 57–79.Google Scholar
U.S. Census Bureau. 1964. Population characteristics: Mobility of the population of the United States, April 1961 to April 1962. Series P-20, No. 127. Washington, DC: Government Printing Office.
Verba, Sidney, Kay Lehman Schlozman, and Henry E. Brady. 1995. Voice and equality: Civic voluntarism in American politics. Cambridge: Harvard University Press.
Wolfinger, Raymond E., Benjamin Highton, and Megan Mullin. Forthcoming. How postregistration laws affect the turnout of registrants. State Politics and Policy Quarterly.
Wolfinger, Raymond E., and Steven J. Rosenstone. 1980. Who votes? New Haven: Yale University Press.
Zaller, John R. 1992. The nature and origins of mass opinion. New York: Cambridge University Press.