Elections 2012: Suppressing Fraud or Suppressing the Vote? : Statistics, Politics and Policy Jump to ContentJump to Main Navigation
Show Summary Details

Statistics, Politics and Policy

Editor-in-Chief: Middleton, Joel

Access brought to you by:

provisional account

See all formats and pricing

Select Volume and Issue
Loading journal volume and issue information...

Elections 2012: Suppressing Fraud or Suppressing the Vote?

1U. Massachusetts Medical School, Quantitative Health Sciences, Worcester, MA, 01655 USA

2Dartmouth College, Mathematics, Hanover, NH, 03755, USA

Corresponding author: John W. Lamperti, Dartmouth College, Mathematics, Hanover, NH, 03755, USA, Tel.: +1 603 646 2866, Fax: +1 603 646 1312

Citation Information: Statistics, Politics and Policy. Volume 4, Issue 1, Pages 14–28, ISSN (Online) 2151-7509, ISSN (Print) 2194-6299, DOI: 10.1515/spp-2012-0002, January 2013

Publication History

Published Online:


Recent years have seen a wave of laws making it more difficult to vote in the United States. Their ostensible purpose is to prevent voting by persons not legally qualified, and thus to improve public confidence in electoral integrity. Are such laws in fact needed to address serious fraud problems? On the other hand, how many and what kinds of legitimate voters will be, or have been, disenfranchised by them? Is voter confidence positively, or negatively, affected by voter ID laws? This article surveys what is known about these issues and offers suggestions for how statisticians can contribute.

1 Introduction

The United States has a history of promoting democracy – and of restricting it. The Constitution of 1789 did not define who could vote, implicitly leaving that to the various states; originally, only White male property owners could vote. Since then Constitutional Amendments and enabling legislation have hugely extended citizen suffrage.

The 15th Amendment to the US Constitution (ratified in 1870) granted voting rights to “citizens of the United States” regardless of their skin color, but not their sex. In 1920 the 19th Amendment took care of that omission. The 23rd Amendment allowed District of Colombia residents to vote for president, the 24th outlawed poll taxes, and the 26th lowered the voting age to 18. The Voting Rights Act of 1965 provided federal enforcement to ensure that the rights guaranteed by the 15th Amendment became more nearly the reality. The National Voter Registration Act of 1993, also known as “Motor Voter,” made registering to vote easier. Many states have taken further steps, such as election-day voter registration, with the same goal of increasing citizen participation in elections.

During the last few years, however, and at an accelerated pace since the 2010 elections, the tide has turned. New York University’s Brennan Center for Justice calls it “the biggest rollback in voting rights since the Jim Crow era.” At the state level – where most voting rules and regulations are defined – many recent laws make registering and voting more difficult, especially affecting the ease, or even the possibility, of voting for the poor and minorities.1 Much of this movement has been facilitated by the American Legislative Exchange Council (ALEC), a corporate-funded think tank that supplies model legislation to conservative state legislators.

What sorts of measures are we talking about? They include:

  • Requiring voters to have specific types of photo ID documents;

  • Instituting onerous regulations and high penalties for minor infractions by third parties (such as the League of Women Voters) that conduct registration drives;

  • Eliminating same-day voter registration;

  • Instituting broad purges of registration lists, requiring many legitimate voters to jump through hoops to re-establish their right to vote;

  • Making it more difficult to establish “place of residence” to be able to vote (affects especially students and those who move frequently);

  • Reducing the number of polling places and/or their open hours;

  • Curtailing absentee and early voting.

Laws and regulations making it harder for people to vote have been adopted in more than 30 states, many in just the last 2 years. Estimating how many potential voters will be affected by these measures, and how many will actually be deterred from voting, is challenging. The Brennan Center report cited above finds that the restrictions “fall most heavily on young, minority, and low-income voters, as well as voters with disabilities.” The numbers? “These new laws could make it significantly harder for more than five million eligible voters to cast ballots in 2012.” Of course putting roadblocks in the way of that many people does not translate into 5 million fewer voters, since some will successfully negotiate the roadblocks and others would not have voted anyway. Statistics should provide the tools to estimate the actual effect of the laws: which people are seriously burdened and which (and how many) are actually prevented from voting.

2 To Prevent Fraud?

The sponsors of these measures say the new laws are needed to “prevent fraud.” In fact, electoral error and fraud have been serious problems throughout American history.

Electoral error is all too common. Grilled by a Congressional committee about problems with network coverage of the 2000 election, NBC News president Andrew Lack said: “Millions of votes are thrown out in election after election in this country. Now there’s a story, there’s a screw-up.”2 The Caltech/MIT Voting Project, a study of U.S. voting techniques and problems begun immediately after that election, estimated that 4–6 million votes of legitimate citizens were “lost” in the 2000 presidential election due to a range of electoral process issues.3 Astonishingly, these numbers represent about 5% of the officially tallied 105 million votes. The largest sources of these errors included problems with registration, long lines and other accessibility issues at polling places, as well as difficulties with voting machines or other aspects of usability, such as Palm Beach County’s confusing “butterfly ballot.” Irregularities in counting the votes actually cast are also of concern; since 2000, the American Statistical Association has focused on improving electoral auditing.4

Electoral fraud is a small part of electoral error, although with a process as decentralized and error-prone as voting in America, it is sometimes hard to tell innocent errors from malicious ones. But innocent or malicious, meaningful manipulation of the system requires the involvement of partisans who are, or collude with, election officials – those in the best position to place a heavy thumb on the scales. (Remember: “It’s not who votes that counts, it’s who counts the votes!”) In contrast, the new restrictions are mainly aimed at potential fraud by individual voters; that is, at voter fraud. The distinction is critical.

Photo ID requirements, for example, do not prevent stuffed or lost ballot boxes, rigged voting machines, gerrymandered district boundaries, vote buying, or other time-honored and far more effective ways to steal elections. And they don’t address any of the problems identified by the Caltech/MIT project. The only problem ID laws address is that of an individual impersonating a legitimate voter in order to cast a fraudulent ballot. Such impersonation already carries draconian penalties (in Florida, for example, it is a third degree felony), for miniscule partisan gain. Does anybody ever intentionally do this? What is the evidence?

3 Voter Fraud is Rare

Indeed, evidence of voting by individuals who are not legally entitled to vote or who vote in another’s name (and especially intentional – i.e., fraudulent – voting) is extremely scarce. But the ID restrictions will differentially suppress voting by certain population groups. The issue is highly politicized, since those disenfranchised are disproportionately likely to vote Democratic. And in fact nearly all the push for the new restrictive measures is from Republicans; the pushback is from Democrats.

Doug Chapin from “Electionline.org,” a project of the University of Minnesota, tried to frame the issue impartially during a recent discussion on NPR:5

… I really think that the hardest core partisans on both sides of this argument would agree that there is not zero instance of fraud, nor is there zero instance of otherwise eligible people being turned away. The problem we face, quite frankly, is that we don’t know how big the problem is either way… And in the absence of that, it’s really hard for the policy process to make policy judgments, and I think that explains a lot of the heat we’re seeing on both sides of these issues.

If a lack of reliable data is fundamental, statisticians should be able to help!

In fact, our ignorance is nowhere near as total as Chapin’s comment suggests. (Yes, sometimes “man bites dog” but the problem is not on the same scale as “man” being bitten by his “best friend”.) Wendy Weiser of the Brennan Center replied:

…despite the huge … resources that have been put into trying to identify them [the number of cases of illegal voting] are infinitesimal.

There is little evidence of the kind of voter fraud that ID laws could help prevent – one person, not entitled to vote, impersonating a legitimate voter. This has been amply demonstrated.6 One particularly meticulous study concludes:

The data … are consistent with a logic that works against the fear that individual voters are corrupting elections. The best facts we can gather to assess the magnitude of the alleged problem of voter fraud show that, although millions of people cast ballots every year, almost no one knowingly and willfully casts an illegal vote in the United States today.7

A striking confirmation of that was provided when the State of Indiana defended its 2005 strict voter ID law before the US Supreme Court in 2008. The state claimed that the photo ID requirement would reduce instances of voter fraud. However, even as the Court upheld the law on a 6–3 vote it acknowledged that there was no record of any fraud in Indiana’s history that such a requirement would have prevented! The few cases of voter fraud that had been documented all occurred with absentee ballots. In his dissenting opinion, Justice David Souter argued that Indiana needed to show the actual existence of fraud, as opposed to merely claiming that fraud was possible, before imposing “an unreasonable and irrelevant burden on voters who are poor and old.”8 Souter’s dissent goes to the heart of the issues at stake today.

4 But Fear of Fraud is Widespread

Some proponents of strict voter ID and other requirements also argue that even if voter fraud is actually rare, there is widespread public belief that it is common. This belief, they say, will then reduce voter turnout. Finally, they claim that such (erroneous) beliefs and their undesirable consequences can be eliminated, and public faith in the system restored, by passing voter ID laws. This was precisely the stated basis for the Supreme Court majority’s decision in the Indiana case, where Justice John Paul Stevens wrote, “Indiana’s interest in protecting public confidence in elections, while closely related to its interest in preventing voter fraud, has independent significance, because such confidence encourages citizen participation in the democratic process.”9

Justice Stevens’ reasoning was bizarre; it hardly seems plausible that adding an obstacle to voting would increase turnout. The Court had no facts before it to support those speculations; indeed, there is now evidence that they are false. An opinion study reported in the Harvard Law Review in 2008 investigated the connections between belief in several types of possible fraud and voting behavior. Indeed, belief in voter fraud is widespread: 35% of Republicans and 15% of Democrats agreed that “voting more than once or voting if not a US citizen” was “very common.” However, a person’s belief in the prevalence of voter fraud was unrelated to either the presence of strict ID requirements, or to his or her intention to vote. The study concludes: “The use of photo identification requirements bears little correlation to the public’s belief about the incidence of fraud. The possible relation of such beliefs to participation appears even more tenuous… [We] conclude that … public perceptions do not provide a firm justification for voter identification laws.”10

5 The Cost of Requiring Specific ID

In contrast to the scarcity of voter fraud, Weiser continued on NPR, “We do know on the ID front … of people who have been denied the right to vote because they didn’t have ID.” Hans von Spakovsky of the Heritage Foundation, an advocate for restrictive laws including voter ID requirements, responded that “There is no evidence whatsoever” that large numbers of people had “been turned away from the polls because of voter ID laws.” But even though the restrictive ID legislation is recent, numerous cases have already been reported of would-be voters who do not have and cannot easily get the required documentation; some of them, such as a group of elderly nuns in Indiana, have quite literally been “turned away from the polls.” Reliable overall figures are, however, elusive.

In addition, the costs of implementing voter ID laws may be substantial. Such costs are in part financial, if government at various levels accepts the responsibility to provide acceptable IDs at no cost to the voter. Charging a fee for a required voting document would be a kind of “poll tax” and therefore subject to legal challenge. Moreover, acquiring even a nominally free ID imposes a high barrier for citizens without a driver’s license because of the need to pay for and assemble acceptable proof and get to a place that will issue the ID during the limited hours made available for this.11

The greater cost of the ID requirements and other restrictive measures, however, is their potential damage to the democratic process, since such measures affect some groups far more than others. A study by the NAACP found that the restrictive measures in general, and voter ID requirements in particular, will “exact a disproportionate price on Afro-American and other voters of color.”12 And, while the number of legally qualified people who will be prevented from voting by the new ID laws can only be estimated, it certainly exceeds the instances of voter fraud.

Still, some plausible estimates are available. The Brennan Center’s calculation that more than five million will find it “significantly harder” to vote aggregated results from several states, aiming to, presumably, provide a lower bound of those affected.13 A 2006 Brennan Center paper reported results from a national telephone survey of 987 voting-age citizens; it found that “Twenty-five percent of African-American voting-age citizens have no current government-issued photo ID, compared to eight percent of white voting-age citizens.”14 Respondents were asked about possession of documentary proof of citizenship and of photo ID. The sample answers indicated that as many as 11% of US citizens, over 21 million individuals, did not have government-issued photo IDs. The proportion was higher for the elderly, for minorities, and for low-income citizens. In addition, the IDs that people did have often did not contain current information.

Five years later, Hans von Spakovsky and Alex Ingram of the Heritage Foundation contested these estimates. They argued, among other things, that the wrong population was surveyed, that alternative forms of ID, acceptable in some states, were not covered, and that other surveys have found “only a small percentage of voters without ID.”15 Three Brennan Center authors, Wendy Weiser, Keesha Gaskins and Sundeep Iyer, quickly responded, defending the study’s methods and results.16

What are the issues, and who is right? Clearly statisticians could perform a national service by evaluating the competing claims and augmenting the available evidence on these issues.

6 New Hampshire

While the proposal that voters must have government ID may seem reasonable, it involves complications and costs for both individuals and the state. Not everyone who is entitled to vote has these IDs. We consider, for example, the state of New Hampshire, where the Republican-controlled legislature recently passed a voter ID law over the (Democratic) governor’s veto.

Until 2012 New Hampshire was one of 20 states with no voter ID requirements; it also apparently never had problems with voter impersonation.17 However, in 2010 the state elected strong Republican majorities in both House and Senate, and the new legislature soon passed a law requiring voters to show a valid photo ID; that attempt was vetoed by Governor John Lynch in June 2011. “There is no voter fraud problem in New Hampshire. We already have strong elections laws that are effective in regulating our elections,” Lynch wrote.18 Backers of the ID requirement did not give up, and despite opposition from the Secretary of State and many town and city clerks, a new, complicated ID law passed in June 2012 over the governor’s second veto. The law is supposed to be implemented in two stages with the stricter ID requirements going into effect only in 2013; it also requires clearance from either the Department of Justice or a federal district court.

Students at Dartmouth College sought to assess the impact of new ID requirements.19 The Department of Motor Vehicles supplied the numbers of current license holders over the age of 18. (A New Hampshire driver’s license will be acceptable ID under the new law.) Estimating the number of eligible voters (citizens) among this group, they compared the result to the total number of potential voters obtained from the 2010 census. Their conclusion is that roughly 3% of eligible voters, some 28,600 persons, do not have a driver’s license. Although some of these may possess other acceptable ID such as a passport, most people without driver’s licenses do not have passports. It seems likely that at least 25,000 citizens, eligible to vote in New Hampshire, will be inconvenienced or worse when trying to exercise that right.

Similar studies have been conducted in other states,20 and many of the estimates (of the percentage without “acceptable” ID) are considerably higher. (In rural New Hampshire a driver’s license is more important than it is for many city dwellers.) At least 10 additional states have passed new voter ID laws since 2010. Similar laws have been proposed (not necessarily passed) in at least 34 states.

The disparity between the probable cost of voter ID proposals and the lack of any serious voter fraud in New Hampshire calls into question the motive for stricter measures. (What happened to the Republican ideal of “limited government”?) Yet proponents usually insist that eliminating fraud is their only concern. New Hampshire House Speaker William O’Brien, a vocal supporter of voting restrictions, added fuel to the debate in a speech to a “tea party” group in Rochester, N.H. Advocating an end to election-day registration as well as stricter ID and residence requirements, O’Brien deplored the effects of students voting since students are “basically doing what I did when I was a kid, which is voting as a liberal. That’s what kids do – they don’t have life experience, and they just vote their feelings.”21 As opposed, presumably, to rational analysis of the issues by mature citizens such as himself. But voting one’s feelings, at any age, is not fraud.

7 Florida

Once again, as in the controversial presidential election of 2000, the state of Florida has been in the eye of the storm. The controversy there is not primarily about voter ID requirements; Florida has had an ID law for many years, although not of the most restrictive kind. But voter ID requirements are only one way to limit the franchise, and several other tactics have been more prominent in the “Sunshine State.” A brief summary:

1) Restricting Voter Registration Drives.

In the 2008 presidential election, the nation saw an impressive increase over 2004 in voting by minorities: votes cast by African Americans were up 15%, by Asian Americans 21%, and by Hispanic or Latino citizens about 28%. These increases were facilitated in many areas, including Florida, by voter registration drives carried out by organizations such as the League of Women Voters (LWV), Rock the Vote, NAACP, and others. These drives are important: “It’s a documented fact that minority and low-income citizens, in particular, rely on community voter registration drives to register to vote,” says Michael Slater, executive director of Project Vote.22 In 2008 almost 20% of African-American voters had registered in such drives, compared to only 6% of whites.

Florida has a history dating back to 2005 of obstructing voter registration drives, but the LWV and others have successfully challenged the restrictions in court. Between 2009 and 2011 the issues seemed to have been settled and an agreement reached for the future. In May 2011, however, the legislature passed, and newly-elected Governor Rick Scott signed, a bill (HB 1355) creating serious new obstacles to registering and voting in Florida. The stated rationale was preventing fraud, even though the Florida Department of State itself noted “just 31 cases of alleged voter fraud” between 2008 and 2011, only two of which resulted in arrests.

The new law included regulations hindering voter registration drives by third parties, such as a strict 48-hour time period for turning in all names – with heavy penalties for missing that limit. These regulations were so threatening that even the LWV could not realistically comply and stopped its registration activities. The impact has already been substantial: a recent study showed registrations down by over 80,000 compared to the corresponding period of the 2008 election cycle.23

In late May 2012, a Federal District Court in Florida blocked implementation of the most restrictive of the new rules obstructing registration drives. Calling parts of HB 1355 “onerous,” Judge Robert Hinkle wrote: “If the goal is to discourage voter-registration drives and thus also to make it harder for new voters to register, the 48-hour deadline may succeed.”24 The League of Women Voters and other organizations quickly announced they would resume their efforts to register as many voters as possible, although they still face more obstacles than in the past. “Registering citizens to vote is part of our core mission, and we’re excited to get back to work,” said St. Petersburg LWV President Darden Rice.25 But the time is short, and they are unlikely to be able to make up all the lost ground.

2) Purging the Registration List.

The state of Florida does not have a good history with voter lists. In 2000 a purge from voter rolls of alleged former felons (who are ineligible to vote under Florida law) was based on highly flawed data and erased the registrations of some 12,000 legitimate voters.26 African Americans, some 11% of the state’s electorate, made up over 40% of those purged, and many were prevented from casting ballots in an election where George W. Bush’s official margin of victory was 537. A similar wrongful purge was narrowly averted in 2004.

In 2002, in the aftermath of the year 2000 electoral problems, Congress passed the Help Americans Vote Act (HAVA). This law requires, among other things, that each state maintain a computerized, statewide list of registered voters, and that the list be updated periodically to keep it accurate. But doing that job well is not simple27 and can open the door to abuses similar to Florida’s previous purge debacles. In fact, similar controversy is alive and well in the same state in 2012.

Late in 2011, Governor Scott ordered the Secretary of State to identify and remove noncitizens from the voter registration list. Lacking accurate citizenship data, the Secretary used motor vehicle department records to come up with over 180,000 names of “potential non-citizen registered voters.” The list was not made public. Instead, the state released a subset of 2625 (with no information about how the subset had been selected) to county election supervisors, requesting that they investigate and, if justified, remove the names from their voter rolls. Among others, Florida Congressman Ted Deutch objected, asking that the purge be suspended. Deutch pointed out that although Florida has “no history of mass voter fraud,” it does have a history of “mass voter disenfranchisement” dating from the elections of November 2000.

According to The Miami Herald (May 12, 2012), only 13% of the “potential noncitizen” list is non-Hispanic White, even though this demographic is fully two thirds of Florida’s voter rolls. Blacks make up 14% of the list, about their voting proportion, but Hispanics are vastly overrepresented.

The supervisors found that the lists contained many errors. Miami-Dade County accounted for a majority – 1638 – of the subset. Of those, 385 quickly proved to be US citizens, while most of the remainder did not immediately respond to a letter sent by the county Supervisor of Elections. Many county supervisors received complaints from angry citizens, typically from people recently naturalized and proud of it. But there was also Bill Internicola, born in Brooklyn in 1922, who had earned a Bronze Star fighting in World War II. Internicola had to show his Army discharge papers to remain registered.

At the end of May, the US Justice Department called for a halt to the purge. Florida’s procedures had not been “pre-cleared” as required under the Voting Rights Act, and there was also an issue about the late timing of the purge, which appeared to conflict with the 1993 “Motor Voter” Act. Governor Scott ordered the purge to continue, but other state officials, and almost all county election supervisors, were not eager to go on. The original list of “potential non-citizens” was eventually released to the public although the state now called it “obsolete” and insisted that it would not be used to purge any voters this fall. In late July, one commentator judged that “the purge is in limbo” and thinks “In the end, the voter purge battle in Florida may have been fought to a draw.”28

“It clearly is partisan and political, but actually, there is no purge going on at the moment,” said Ion Sancho, the elections supervisor in Leon County and a voting-integrity advocate. Sancho predicted that he and many of his fellow clerks would simply refuse to implement voter removals out of concern the evidence wasn’t sufficient.29 Earlier this year 64 of 67 county supervisors didn’t implement the first purge effort, so Sancho may well be right.

The state has also announced that it will remove some 53,000 dead persons from the voter rolls. This should not be controversial – provided they are really dead.

How to control both “type I and type II” errors in maintaining accurate voting lists is a problems that statisticians could help to get right.

3) Cutting Back on Early Voting.

Getting to the polls on Election Day is hard for many would-be voters, especially low-income workers who can’t afford to miss part or all of a day’s wages. Early voting can be the answer. Moreover, in 2008 the last Sunday before Election Day (Tuesday) made a big contribution to that year’s high voter turnout. “Souls to the Polls” campaigns, encouraging churchgoers to cast their votes as a group, were especially effective with minority voters; African Americans made up about one third and Hispanics almost a quarter of Florida’s last-Sunday voters in that year.

Florida’s new voting law (HB 1355) would change that, by cutting early-voting days from 14 to 8, explicitly eliminating voting on that last Sunday.30 Since this part of the law disproportionately affects low-income and minority voting, its implementation has been blocked by a federal court in Washington, at least for those counties covered by Section 5 of the Voting Rights Act.

4) Disenfranchising Those Convicted of a Felony.

The United States is one of the world’s most restrictive democracies with respect to denying the vote to people convicted of crimes even after they have completed their sentences. Nearly 6 million Americans, about 2.5% of the total voting-age population, cannot now vote due to “felon disenfranchisement.” That number has risen rapidly; as recently as 1980 it was less than 1.2 million. Like US prisoners as a group, those affected are disproportionately African American; their disenfranchisement rate is four times that of whites.31

States differ widely on this issue, and Florida’s policy is particularly harsh. As of January 2011 both Florida’s disenfranchisement rate and the total number of people thus banned from voting were the highest in the nation, about 10.4% and 1.54 million, respectively. Of these, some 520,000 were African Americans, amounting to 23.3% of their voting age population.

These laws and numbers have been changing on a state-by-state basis, often in the direction of greater leniency. The reverse is the case in Florida, which is one of six states that disenfranchise ex-felons for life. However, Florida law does provide for applying to have one’s voting rights restored. In 2007, then-Governor Charlie Crist streamlined and liberalized the restoration procedure, allowing some 150,000 Floridians to regain their franchise. That policy was quickly reversed in March 2011 by Governor Scott. There is now a 5–7 year waiting period before a former felon can even apply for restoration. Furthermore, the process is complicated with no certainty of success. Of course this, like the other measures discussed above, will significantly reduce voting by minority and low-income populations.

8 Why Make it Harder to Vote?

Of course New Hampshire and Florida are only examples; since the beginning of 2011 at least 41 states have introduced measures restricting access to the ballot. The number of legitimate voters who will be seriously handicapped in casting their ballots by the new voter ID requirements, difficult as that is to measure, certainly vastly exceeds the few cases of voter fraud those IDs might prevent. In view of that disproportion, it is hard to believe that preventing fraud is the real motive for the new restrictive laws. Sometimes their proponents have publicly admitted other reasons for favoring them. New Hampshire’s O’Brien was one example: students “just vote their [liberal] feelings.” Another case recently surfaced in Pennsylvania, when House Majority Leader Mike Turzai, in listing his party’s accomplishments for the Republican State Committee, called out: “Voter ID, which is going to allow Governor [Mitt] Romney to win the state of Pennsylvania, done!”32 A Democratic legislator commented on the same partisan advantage: “Voter impersonation, which is the only form of voter fraud that voter ID would address, never occurs in Pennsylvania … This was always about suppressing the votes of people who vote the wrong way.” The law is being challenged in court.

In Florida too, the sponsor of the restrictive HB 1355 openly admitted that his bill was meant to make voting more difficult. “I don’t have a problem with making it harder,” said State Senator Mike Bennett. “[Voting] should not be easy.”33 But – much harder for some than for others.

In Ohio as well, where early voting on weekends has been eliminated, GOP partisan Doug Preisse made a surprisingly frank admission to the press: “I guess I really actually feel we shouldn’t contort the voting process to accommodate the urban – read African-American – voter-turnout machine,” he said. Preisse is a county party chair and adviser to Governor John Kasich. What he said publicly is what many believe privately – keeping turnout down among Obama supporters is the best way for the GOP to win the 2012 election.34

Estimating just how many eligible citizens will be deterred from voting by the ID laws and other new restrictions is a big task, requiring separate studies by state. Of course, we cannot directly measure the effect until such laws are implemented. Once those numbers are available, they can be compared to the number of voter fraud cases, if any, that may have been prevented. But even when the facts are known, policy decisions will still depend on value judgments. How many fraudulent votes avoided or detected would justify disenfranchising even one or two – or several hundred thousand – legitimate voters?

9 Concluding Thoughts

Half a century ago, broad participation in politics and elections seemed almost synonymous with democracy itself. For example, on election eve in 1956 President Eisenhower urged participation, saying: “If we don’t vote, then we are forfeiting one of the greatest privileges we have, of participating in the decisions of this country. I am not speaking now of how you vote. I am talking about the act itself… that is the thing that we must do.” And Congress largely agreed; the Senate vote for the 1965 Voting Rights Act was 77 to 19, with 30 out of 32 Republicans in favor! (Almost all negative votes were from Southern Democrats.) This ideal is still alive, as in this recent editorial:

Any attempt to narrow the voting population is suspect; increasing voter participation should be a national goal. To reverse this [restrictive] trend, voters need to elect representatives who value voting rights.35

For decades, however, Americans have tolerated electoral errors that regularly disenfranchise millions of voters. Reforms are badly needed, but the recent spate of restrictive laws and regulations do not address these problems.

Whether we make voting harder or easier, and whether we seriously invest in improving our elections, reflect our values as a nation. Many realistic steps could make our elections more trustworthy. The Brennan Center for Justice, the Caltech/MIT Voting Project, the Sentencing Project, and the American Statistical Association, among others, have laid a good foundation. We must continue to identify, quantify, and work to eliminate the major causes of electoral error, including voter disenfranchisement. Serious attempts at electoral reform should address the nation’s many real electoral problems, not the mythical ones that photo ID laws purport to solve.


  • 1

    See, for example, Wendy Weiser and Lawrence Norden, “Voting Law Changes in 2012,” Brennan Center for Justice (NYU School of Law), 2011.

  • 2

    Los Angeles Times (online), February 15, 2001.

  • 3

    Caltech/MIT Voting Technology Project, Voting: What Is, What Could Be, July 2001. See also MIT News, July 16, 2001.

  • 4

    http://www.amstat.org/outreach/electionauditingresources.cfm, accessed August 27, 2012.

  • 5

    “New Voter ID Laws and the 2012 Elections,” National Public Radio, The Diane Rehm Show, June 11, 2012.

  • 6

    For example see Justin Levitt, “The Truth About Voter Fraud,” Brennan Center for Justice, 2007; also Lorainne C. Minnite, The Myth of Voter Fraud, Cornell University Press, 2010.

  • 7

    The Myth of Voter Fraud, pp. 5–6.

  • 8

    Supreme Court of the United States, Crawford et al. v. Marion County Election Board et al., opinions of Justice Stevens (for the majority, to uphold the Indiana law) and Justice Souter (dissenting).

  • 9


  • 10

    Stephen Ansolabehare and Nathaniel Persily, “Vote Fraud in the Eye of the Beholder: The role of public opinion in the challenge to Voter identification Requirements,” Harvard Law Review 2008: 1737–1773.

  • 11

    See among others Keesha Hawkins and Sundeep Iyer, “The Challenge of Obtaining Voter Identification,” Brennan Center for Justice, July 17, 2012. Their paper contains this extreme example:The Mississippi Catch-22Although Mississippi’s restrictive law is not yet in force, citizens there without ID face particularly perverse rules. To secure government-issued photo ID, many voters will need a birth certificate. Yet the state requires a government-issued photo ID to obtain a certified copy of a birth certificate. These rules jointly represent another major hurdle to voting placed in the path of those with the least means to surmount them.

  • 12

    “Defending Democracy: Confronting Modern Barriers to Voting Rights in America,” A Report by the NAACP Legal Defense and Education Fund and the NAACP, 2011.

  • 13

    See note 1, “Voting Law Changes in 2012,” p. 37.

  • 14

    “Citizens Without Proof: A survey of Americans’ possession of documentary proof of citizenship and photo identification,” Brennan Center for Justice, November 2006.

  • 15

    “Without Proof: The Unpersuasive Case Against Voter Identification,” Heritage Foundation Legal Memorandum no. 72, August 24, 2011.

  • 16

    “‘Citizens Without Proof’ Stands Strong,” Brennan Center for Justice, September 8, 2011.

  • 17

    Never, that is, until January 2012 when James O’Keefe and the ill-named “Project Veritas” sent several individuals to New Hampshire who requested and received ballots under false names, presumably to show that committing voter fraud was possible. These activities were crimes, but the individuals immediately left the state and so far have not been identified or prosecuted.

  • 18

    Governor Lynch’s Veto Message (June 27, 2011) regarding SB 129.

  • 19

    Stephen Prager, Manav Raj and Joseph Singh, PRS Policy Brief 1112–05, Rockefeller Center at Dartmouth College, March 6, 2012.

  • 20

    See for example Justin Leavitt, “Fast Facts on the Impact of Photo ID: The Data,” Brennan Center for Justice, 2008.

  • 21

    See for example Peter Wallsten, “In States, Parties Clash Over Voting Laws that Would Affect College Students, others,” Washington Post (online), March 8, 2011. See also New York Times editorial “Keeping Students From the Polls,” December 26, 2011.

  • 22

    Project Vote Newsletter, March 29, 2012.

  • 23

    New York Times study, March 27, 2012.

  • 24

    CBS News Political Hotsheet (on line), May 31, 2012.

  • 25

    The Miami Herald, August 9, 2012.

  • 26

    “United States v. State of Florida: Court Cases,” Brennan Center for Justice, June 26, 2012.

  • 27

    See for example Myrna Pérez, “Voter Purges,” Brennan Center for Justice, September 30, 2008.

  • 28

    Molly Ball, “Will Florida’s Voter Purge Cost Obama the Election?” The Atlantic (online), July 24, 2012.

  • 29


  • 30

    See testimony of Michael Herron and Daniel Smith before the U.S. Senate Judiciary Committee, Subcommittee on the Constitution, Civil Rights and Human Rights, January 27, 2012.

  • 31

    Figures in this and the following paragraph are from Uggen, Shannon and Manza, “State-Level Estimates of Felon Disenfranchisement in the United States, 2010,” The Sentencing Project, July 2012.

  • 32

    Pittsburgh Post-Gazette, June 28, 2012.

  • 33

    Quoted in Brentin Mock, “Florida to People of Color: Don’t Vote Here,” The Nation, July 16–23, 2012.

  • 34

    Columbus Dispatch, August 19, 2012.

  • 35

    San Jose Mercury News, June 20, 2012.

Comments (0)

Please log in or register to comment.