Download testimony [pdf]
The Brennan Center for Justice at NYU School of Law
United States Senate Committee on the Judiciary
Subcommittee on the Constitution, Civil Rights and Human Rights
New State Voting Laws: Barriers to the Ballot?
September 8, 2011
The Brennan Center for Justice at NYU School of Law thanks the Senate Judiciary Subcommittee on the Constitution, Civil Rights, and Human Rights for providing the opportunity to present testimony at this important hearing, and in particular, discuss the consequences of strict voter identification requirements.
The Brennan Center is a nonpartisan think tank and legal advocacy organization that focuses on issues of democracy and justice. Among other things, we seek to ensure fair and accurate voting procedures and systems, and to promote policies that maximize citizen enfranchisement and participation in elections. We have done extensive work on a range of voting issues of concern to Americans, including voter identification, voter registration, and voting system security. Our work on these topics has included the publication of studies and reports; assistance to federal and state administrative and legislative bodies with responsibility over elections; and, when necessary, participation in litigation to compel states to comply with their obligations under federal law and the Constitution.
In the past year, state governments across the country have enacted a record number of laws restricting access to the franchise. These laws have taken many forms—from eliminating election-day registration, to restricting third-party voter registration activities, to reducing the number of days for early voting and limiting the number of days for voter registration. By far the most common legislation of this type introduced in the 2011 legislative session was legislation requiring photo ID for in-person voting. Thirty-four states introduced laws requiring voters to produce photo IDs for in-person voting. Of the states that do not have a voter ID laws, only three—Oregon, Vermont and Wyoming—did not consider voter ID legislation this year. Of the states considering new legislation this year, twenty-two defeated photo ID legislation in the legislature. Six states passed strict “no-photo, no-vote” voter ID laws; and three extended the new photo ID requirements to absentee voters. In five states, governors’ vetoes prevented photo ID legislation from becoming law. And, as a result of activity this session, in November 2011 in Mississippi and November 2012 in Missouri, voters will consider ballot measures to amend their state constitutions to require photo IDs for all voters.
The Brennan Center’s testimony will focus on these new voter ID laws. Following a brief discussion of the constitutionality of the new photo ID laws, the Brennan Center will provide an assessment of the evidence related to two issues that are critical to understanding the impact of these new laws, specifically: (a) available empirical evidence related to rate of possession of photo ID by voting-age American citizens, including the effect of voter ID laws on voter turnout, and; (b) the true incidence of voter fraud—particularly with respect to voter impersonation, including an evaluation of the most oft-cited and recent allegations of voter fraud in the photo ID debate.
I. Voter ID Laws Are Not Per Se Constitutional
Advocates for photo ID laws often subscribe to the inaccurate belief that the U.S. Supreme Court’s divided decision upholding Indiana’s photo ID law in Crawford v. Marion County Election Board rendered all state photo ID laws immune to constitutional challenge. In Crawford, the Court upheld Indiana’s photo ID law against a broad “facial” attack to its constitutionality. In doing so, the Court made clear that the photo ID law remained subject to challenge as a matter of law by particular groups or individuals who were unconstitutionally burdened by the law. The Court expressly singled out as groups who might bring a successful challenge “elderly persons born out of state,” “persons who because of economic or other personal limitations may find it difficult to secure a copy of their birth certificate” or other documents needed for photo ID, homeless people, and people with a religious objection to being photographed. In addition to leaving the door open to challenges by affected voters, the Court also left the door open to challenges to other photo ID laws that burden voters more than Indiana’s.
The factual record in Crawford largely compelled the Court’s decision refusing to invalidate Indiana’s photo ID law. The Court held that the scant record put before it did not present sufficient evidence of the burdens the law would impose on voters to justify striking down the law. Consequently, a future lawsuit with a more developed factual record may lead to a different result than Crawford.
Finally, photo ID laws may also be vulnerable to lawsuits based on state constitutional rights. For example, Missouri’s photo ID law was struck down by the state’s Supreme Court, which found that the Missouri Constitution had stronger voter protections than the federal constitution.
For all of these reasons, the Supreme Court’s divided decision in Crawford v. Marion County Elections Board should not be read as a universal constitutional endorsement of Indiana’s photo ID law, let alone all photo ID laws.
II. Photo ID Requirements for Voters Negatively Impact Voting
The best available empirical evidence shows that significant percentages of voting-age American citizens do not possess valid, government-issued photo identification.
Regardless of whether new photo ID laws are constitutional, they are certainly bad policy. Statutes obliging American citizens to obtain and produce government-issued photo ID before being permitted to vote threaten to disenfranchise millions of Americans. The seminal study on this issue, Citizens Without Proof—a report based upon a nationwide survey by the National Opinion Research Corporation sponsored by the Brennan Center in late 2006—found that 11% of voting-age American citizens do not have current, government-issued photo IDs. That represents more than 21 million individuals. Any policy that would operate to completely bar or limit access to a population of this size from the polls must be viewed with scrutiny and compels demand for a well‑supported rationale for its implementation.
These findings in Citizens Without Proof are consistent with other independent studies. For instance, a 2008 Collaborative Multi-ethnic Post-election Survey of registered voters in eighteen states found that 8% lack a valid, state-issued photo ID with their current address. And a 2007 Indiana survey found that over 13% of registered Indiana voters lack a valid Indiana driver’s license or an alternate Indiana-issued photo ID.
Also of significant concern is the substantial evidence that poor, elderly, and non-white citizens are less likely than the general population to possess the requisite current government-issued photo ID. In Citizens Without Proof, we found:
- Citizens earning less than $35,000 per year are more than twice as likely to lack current government-issued photo identification as those earning more than $35,000. Specifically, at least 15% of lower-income voting-age American citizens do not have this type of documentation.
- Of citizens age sixty-five and above, 18% lack current government-issued photo ID.
- Up to 25% of voting-age African-American citizens lack valid, government-issued photo ID, compared to 8% of voting-age white citizens.
Again, other empirical studies have come to the same conclusions. For example:
- A 2005 telephone survey of Indiana voters aged sixty and older found that 10% lack a valid state driver’s license. Within this population, the survey concluded that 30% of non-white voters do not have a valid license.
- A 2006 national survey sponsored by the Center on Budget and Policy Priorities found that 8.9% of African-Americans born in the U.S. do not have a passport or birth certificate available.
- A 2007 statewide telephone survey of Indiana residents concluded that state residents with only a high-school degree are 9.5% less likely to have access to valid photo ID than college graduates.
- A 2007 regression analysis of data from the Georgia Secretary of State and the Georgia Department of Driver Services determined that, compared to white voters, black voters were over three times more likely to lack valid forms of identification. Women and the elderly were also significantly less likely to possess a valid driver’s license or state ID card.
- The 2008 Collaborative Multi-ethnic Post-election Survey found that 14% of blacks and 7.3% of Latinos lack a valid, state-issued photo ID with their current address, compared with 5.8% of whites. 
Attempts to Discredit Studies That Show a Significant Number of Americans do not have Government Issued ID are Unpersuasive.
As the seminal national study on the question of how many Americans eligible to vote possess current government issued photo ID, Citizens Without Proof is regularly cited by academics, politicians and advocates as conclusive evidence of how many Americans will be impacted by laws imposing harsh and restrictive photo ID requirements on voters. It has also, consequently, been the focus of criticism of studies on the subject from proponents of strict photo ID laws.
The results reported in Citizens Without Proof are based on a national survey conducted by the nationally well-respected research firm Opinion Research Corporation (ORC). ORC helped determine neutral question wordings, conducted the questioning of survey participants, and then corrected for biases. The Brennan Center correctly and ethically disclosed all results, along with their statistical significance. The Brennan Center and ORC acted under proper methodological and ethical protocols for gathering data and reporting conclusions.
The most recent attack on Citizens Without Proof comes from Hans von Spakovsky and Alex Ingram, who recently published a Heritage Foundation memo claiming that Citizens Without Proof is flawed in both its conclusions and methodology. Von Spakovsky and Ingram make a number of false and misleading about Citizens Without Proof. A detailed response to those criticisms can be found in a document (“Citizens Without Proof Stands Strong”) that is annexed as an exhibit to this testimony.
The three main responses in Citizens Without Proof Stands Strong are as follows: first, von Spakovsky and Ingram wrongly criticize the survey because it “could have included illegal and legal aliens.” This speculation baseless. As Citizens Without Proof clearly reports, ORC specifically questioned survey participants as to whether they were U.S. citizens, using questions generally accepted in the industry. The survey results were limited to U.S. citizens of voting age and did not include illegal or legal aliens. Von Spakovsky and Ingram also try to criticize the survey on the ground that it was not limited to “actual or likely voters, registered voters, or even eligible to vote at all [sic].” This too is baseless. Citizens Without Proof does not purport to present findings of how many actual, likely, or registered voters do not have the documents. Actual or likely voters are not the only citizens who have the right to vote. It is certainly relevant to assess how many of those who are entitled to vote would be prevented from doing so if they tried because of a photo ID or proof of citizenship requirement.
Second, von Spakovsky and Ingram entirely misrepresent the survey questions asked by ORC for the survey used by Citizens Without Proof. For instance, they claim that the survey “did not ask respondents whether they had government-issued IDs”, even though question one in the survey asked respondents whether they have a “current, unexpired government-issued ID.”
Third, von Spakovsky and Ingram cite two questionable studies and their own haphazard data analyses to show that Citizens Without Proof is wrong. Von Spakovsky and Ingram entirely ignore the overwhelming weight of the academic research documented above, all of which supports the Brennan Center’s conclusions. In their own analyses, they repeatedly cite the flawed list-matching comparisons documented below to bolster their flawed allegations.
In short, von Spakovsky and Ingram fail to raise any legitimate criticism of Citizens Without Proof. Their baseless assertions should not be treated with the same seriousness as a national study conducted by one of the nation’s best and most respected research firms, or all of the independent and empirically rigorous studies conducted before and since.
Comparisons of voter registration lists with driver’s license and state identification card lists are not sufficient to demonstrate conclusive rates of ID ownership in a particular state.
Proponents of strict voter ID laws have also attempted to discredit studies showing many Americans do not have ID required under these laws by comparing the numbers of state issued IDs with the numbers of registered voters, or voting age citizens, in a given state. Some advocates claim, using this flawed list-matching methodology, that the number of government-issued photo IDs actually exceeds the number of possible voters. For instance, in an attempt to refute the fact that thousands of voting-age Kansans lack state-issued photo IDs, Kansas Secretary of State Kris Kobach made a simple list comparison between the number of US citizens in his state (2,126,179 persons), and the number of Kansas driver’s licenses or non-driver ID in circulation (2,156,446). Based upon his comparison, Secretary Kobach concluded that there are more photo IDs in circulation than there are eligible voters in Kansas.
A review of the list-management practices of the Kansas Department of Motor Vehicles shows Secretary Kobach’s conclusion is flawed. The Kansas Department of Motor Vehicles (“DMV”) only purges people who have moved outside of the state if they apply for a license in another state, and then that state notifies Kansas. If Kansas is not notified, the license remains on the DMV list until it has been inactive for five years after the license’s expiration date. Between 2005 and 2009, 351,462 Kansas residents over 18 left the state, raising the possibility that many of these movers remain on the current DMV list. Likewise, unless a family member submits a death certificate, deceased persons remain on the DMV list for five years after expiration. Assuming that all Kansans over 18 who died in the past five years had a license that expired in 2006 or later, there could be as many as 118,000 people on the 2010 license lists who are no longer alive. Moreover, Kansas’ driver’s license list and voting-age population statistics both include permanent residents and other non-citizens who cannot vote. According to the American Community Survey, in 2009 there were 104,731 residents of Kansas who over 18 but not citizens. Finally, Kansas’ driver’s license list includes expired licenses, which are not sufficient identification under the state’s voter ID law. For all of these reasons, one cannot draw any meaningful conclusions from the fact that there are a greater number of Kansas driver’s licenses and state ID cards in circulation than there are voting-age citizens.
Comparing state’s driver’s license lists with voter registration lists is inadequate to determine whether or not significant numbers of voting-age citizens lack proper ID for voting purposes. Such lists should not serve as a substitute for empirically rigorous surveys that show large numbers of Americans citizens do not have the kind of government issued photo ID required in the most restrictive state laws.
The best available evidence shows that strict voter ID laws depress turnout.
Given the myriad factors that influence voter turnout and the very short period of time that any photo ID laws have been in effect, it is extremely difficult to isolate the precise affect of voter ID laws on voter turnout. However, the most rigorous empirical study to date, recently described in the leading journal of political science methodology, Political Analysis, concludes that the strictest forms of voter ID requirements reduce turnout among registered voters. This study also found that less educated and less wealthy voters are particularly likely to be deterred. Notably, that study relied upon data from 2000 to 2006, and thus does not examine the impact of the more recent, stricter voting ID laws—which would likely have an even greater negative effect. In their study, Alvarez, et al. documented the effect of voter identification requirements on registered voters as they were imposed in states in the 2000 and 2004 presidential elections, and in the 2002 and 2006 midterm elections. By using four election cycles and individual responses from the Current Population Survey, the authors could isolate the effect of voter identification requirements on voter turnout. The state-level panel data allows them to control for changes in the electoral environment both across states and across time, which they could not do with only one year of data. The individual-level data allowed them to answer questions about whether certain subpopulations are disproportionately affected by these regulations, something that is not possible using aggregate data. Ultimately they found that aggregate data showed no evidence that non-strict voter identification requirements had any effect on voter participation. They did find, however, that the strictest forms of voter identification requirements — combination requirements of presenting an identification card and positively matching one’s signature with a signature either on file or on the identification card, as well as requirements to show picture identification — had a negative impact on the participation of registered voters.
Despite these credible empirical findings, supporters of photo ID laws claim that such laws do nothing to depress turnout. A common survey relied upon by advocates for strict voter requirements is a 2009 study by Jason D. Mycoff, Michael W. Wagner, and David C. Wilson finding that photo ID requirements do not have an effect on voter turnout. Many of Mycoff’s conclusions are based upon aggregate data, which has limited applicability in this context because it fails to consider how voter ID requirements affect individuals. Consequently, as a statistical matter, Mycoff et al.’s conclusion that photo ID does not affect voter turnout is suspect.
While this makes sense as a statistical matter, it is also a matter of common sense. Group or aggregate data can tell you how a large group behaves but it cannot give any meaningful information about the impact of specific inputs on individuals within a large group. For example, say 10 people per week visit a grocery store in their neighborhood for four weeks to buy milk and eggs. At the end of the four weeks the grocery store suddenly changes to a cash-only policy. That cash-only policy may deter shoppers who only use credit cards for grocery purchases. After another four weeks that grocery store may still have 10 customers per week, but there is no way of knowing, based upon the aggregate numbers whether any of the original 10 customers were deterred by the new policy. Similarly, if the same or similar aggregate numbers of voters turnout for elections – there is no way to know, without additional research, whether individual voters within that group are deterred by strict voter ID requirements.
In another example, Georgia’s Secretary of State Brian Kemp wrote a letter to the Washington Post citing increased turnout among black voters between 2006 and 2010 as proof that the state’s voter ID laws had no negative impact on voting. This fact was subsequently repeated to argue that voter ID laws do not, in fact, deter voting. Secretary Kemp drew his conclusions based upon Georgia’s turnout numbers without controlling for other influencing factors. Consequently, after careful evaluation his assertion about increased black voters is not persuasive. A more appropriate way to examine these numbers is to compare Georgia’s turnout with that in other similar states that did not have voter ID requirements during the relevant time period. For instance, in nearby North Carolina, the black voter turnout rate jumped by 40% between 2006 and 2010. This increase rate dwarfs Georgia’s. In 2006, when there was no voter ID requirement, 42.9% of registered black voters turned out in Georgia; in 2010, after the state’s restrictive ID requirement became law, turnout was just 50.4%—representing an increase of 17.5%. In other words, the increase in black voter turnout in North Carolina, a state without voter ID laws, was more than twice the size it was in Georgia, a state with stringent voter ID laws. When appropriately contextualized, Georgia’s voter ID law no longer looks quite so harmless.
III. The Myth of Widespread Voter Fraud
The Brennan Center has paid particular attention in recent years to claims of voter fraud. We have collected allegations of fraud cited by state and federal courts, commissions, political parties, state and local election officials, authors, journalists, and bloggers. We have analyzed these allegations at length, to distinguish those which are supported from those which have been debunked; furthermore, we have created and published a methodology for investigating future claims, to separate the legitimate from the mistaken or overblown. In 2007, we published a monograph reflecting our analysis, entitled “The Truth About Voter Fraud,” which compiled for the first time, the recurring methodological flaws behind the allegations of widespread voter fraud that are frequently cited but often unsupported. Allegations concerning the incidence of or potential for voter fraud have been cited as justification for various restrictions on the exercise of the franchise, specifically photo ID laws. There has been much assertion concerning the appropriate degree of concern regarding such fraud, but relatively little attention paid to the facts that we know.
In-person impersonation fraud is the only type that voter ID laws have the potential to address but study after study confirms that such fraud is extremely rare.
The Brennan Center has analyzed claims of rampant voter fraud in order to distinguish unfounded and exaggerated tales of fraud from reliable, verified claims of election misconduct. We published the results of our analysis in a monograph entitled “The Truth about Voter Fraud,” which compiles the methodological flaws that lead to allegations of voter fraud, and debunks baseless — though often repeated — reports of voter fraud. In our research we have found virtually no fraud of the type that a photo ID requirement could fix. As to allegations of other types of voter fraud, our research indicates that these claims often prove baseless upon inspection. The Brennan Center’s exhaustive research revealed that there is little to no reliable evidence of in-person impersonation fraud in the country. And, of course, this form of fraud is the only misconduct that photo ID laws address.
All available studies show that the incidence of in-person voter impersonation is extraordinarily rare.
- Of the 9,078,728 votes cast in Ohio’s 2002 and 2004 general elections, a total of four were deemed as ineligible or “fraudulent” and found by the Board of Elections and County Prosecutors to have merit and pursued legal action. With no evidence of any of the four convictions being preventable by a photo ID law.
- Between October 2002 and September 2005, the Department of Justice brought just 38 cases were brought nationally, and of those, 14 ended in dismissals or acquittals, 11 in guilty pleas, and 13 in convictions, with only one conviction potentially being preventable by imposition of photo ID laws.
- Analysis of Analysis of 2004 Washington gubernatorial election revealed 6 cases of possible double voting and 19 cases of alleged voting in the name of deceased individuals out of a total 2,812,675 ballots cast; the rate of ineligible voting that thus might have been remedied by ID requirements was 0.0009%
- In a comprehensive survey of election fraud, Professor Lorraine Minnite and David Callahan conducted a review of news and legal databases and interviewed attorneys general and secretaries of state in Alabama, California, Florida, Georgia, Illinois, Minnesota, Mississippi, New York, Oregon, Pennsylvania, Texas and Wisconsin, collectively representing about half of the national electorate. The study found that voter fraud of any kind is “very rare,” is not more than a “minor problem” and “rarely affects election outcomes.” Notably absent from the study are any confirmed cases of in-person impersonation fraud.
Some claim the low incidence of voter impersonation fraud demonstrated above is because it is so difficult to detect. In truth, there are multiple means to discover in-person impersonation fraud, all of which might be expected to yield more reports of such fraud, if it actually occurred with any frequency. An individual seeking to commit impersonation fraud must, at a minimum, present himself at a polling place, sign a pollbook, and swear to his identity and eligibility. There will be eyewitnesses: pollworkers and members of the community, any one of whom may personally know the individual impersonated, and recognize that the would-be voter is someone else. There will be documentary evidence: the pollbook signature can be compared, either at the time of an election or after an election, to the signature of the real voter on a registration form, and the real voter can be contacted to confirm or disavow a signature in the event of a question. There may be a victim: if the voter impersonated is alive but later arrives to vote, the impersonator’s attempt will be discovered by the voter. (If the voter impersonated is alive and has already voted, the impersonator’s attempt will likely be discovered by the pollworker; if the voter impersonated is deceased, it will be possible to cross-reference death records with voting records, as described above, and review the actual pollbooks to distinguish error from foul play.) If the impersonation is conducted in an attempt to influence the results of an election, it will have to be orchestrated many times over, increasing the likelihood of detection.
During a period when investigating voter fraud was expressly deemed a federal law enforcement priority, and when private entities were equipped and highly motivated to seek, collect, and disseminate such reports there have been hundreds of millions of ballots cast, it is telling that during that time, with so many ways to identify voter impersonation fraud there have been only a handful of potential instances of impersonation fraud. Every year, there are far more reports of UFO sightings. The scarcity of reports of in-person impersonation fraud, in this context, is itself meaningful.
Many allegations of fraud are plagued by recurring methodological errors, which are widely discredited.
A common source of fraud claims are list-matching exercises gone wrong. In many cases, the failure of disclosure for the methodology for matching lists makes it impossible to determine the accuracy of the final numbers giving rise to the claims of voter fraud. Comparing lists means comparing multiple fields for accuracy. An abundance of caution is required in any attempt to accurately match information in one data system with information in another. Nevertheless, there are consistent problems when advocates for restrictive voting laws make list comparisons in attempts to prove voter fraud.
Incomplete Matches: “False Positives”
Attempts to match data on one list to data on another list will often yield “false positives”: two records that at first appear to be a match but do not actually represent the same person. If middle initials, prefixes or suffixes are ignored, it can generate thousands of false matches. For example In 2008, the Elections Director for Muscogee County, Georgia, sent out 700 letters to local residents informing them that they were ineligible to vote because they were convicted felons. More than one-third of the voters called to report that there had been a mistake. The purged voters included an octogenarian who insisted she had never even received a parking ticket. According to media reports, the list that went to Muscogee County was generated by a new computer program, and included voters whose names, but not necessarily other information, corresponded or “matched” the names of those with felony convictions.
Complete but Deceptive Matches: “The Birthday Problem”
Advocates for greater restrictions on voting have compared non-citizen lists with voter registration lists, or lists of people voting in a particular election, matching only name and birthday  Too often, the unsupported conclusion is that non-citizens are illegally voting. There is a problem, however, with presuming that two records with the same name and date of birth must represent the same person. This presumption is not consistent with basic statistical principles. The statistical coincidence of matching birthdates is a proven phenomenon, so much so it has a name – “The Birthday Problem.” For example, in a modestly sized group, the probability that two people have the same birthday – day and month – is, for many observers, surprisingly high. In a group of just 23 people, it is more likely than not that two will share the same birthday. For 40 people, the probability is 90%. Applying the “Birthday Problem” to voter registration lists is fairly straightforward. By including the year (and thus the full birth date), the statistics change somewhat, but the threshold is still surprisingly small to many: given some reasonable assumptions about the average lifespan, the probability that at least two of 150 people have the same exact birth date – day, month, and year – is 50%.  And in a group of 300 people, the probability that two share a birth date match is approximately 90%. When one begins to compare lists of hundreds of thousands of voters, it is expected that there will be potentially thousands of name and birthdate matches by simple statistical probability and do not reflect any evidence of voter fraud.
Improper addresses: “You have the wrong house”
There are often allegations that voters have illegally registered from a fake address. Allegedly improper addresses may not show fraud. Apartment numbers may be incorrectly read as part of a street address by computerized matching systems, causing the system to reject an accurate address. A typographical or other data entry error may make a legitimate address appear fictitious or appear to be located outside of the relevant precinct. A voter may be registered as living in a building that has been demolished since the registration was processed; if the voter moved within the same precinct, she may not be required to re-register. In addition, an individual (e.g., a site manager of a business) may actually live at what appears to be an invalid business address. An example of this was in the 2000 elections in Missouri. Then-Secretary of State Matt Blunt saw widespread chaos in the hotly contested 2000 election and made claims of widespread fraud. He claimed, as did others, that 79 votes were cast by voters with allegedly invalid addresses. But a subsequent Post-Dispatch survey of every one of those suspect properties turned up something else: hundreds of bona fide houses and apartment buildings that were wrongly classified by the city assessor’s office as vacant lots. Because of those inaccurate records, many of those properties’ occupants have been wrongly tagged as registering to vote from fake addresses. At that time, city records indicated that 2,214 residents appear to be registered to vote from 1,000 vacant lots. Consequently, state and local elections officials targeted 79 of them for casting possibly illegal ballots. The Post-Dispatch did in fact find 432 city residents registered from 296 truly vacant lots. But most of those residents haven’t voted in years, an indication they had likely moved.
The most notorious allegations of in-person voter fraud are baseless.
There are a few alleged cases of voter fraud that are cited repeatedly, and many are explained by the matching and database errors identified above. Serious review of these exaggerated claims demonstrate them to be little more than hyperbole and distorted facts. The first such example is drawn from the Special Investigations Unit of the Milwaukee Police Department report into the November 2, 2004 general election in the City of Milwaukee. This report is widely cited as evidence that voter fraud is a problem, and frequently bootstrapped onto claims that photo ID laws are needed to prevent such fraud. Yet while the report revealed administrative mistakes and, occasionally, negligence on the part of election officials, it identified only one potential vote that might have involved impersonation fraud, and no documentation verifying that the vote in question was actually cast. The report expressly disavowed any claim that “thousands of fraudulent ballots were cast in Wisconsin.” Indeed, the report showed that much of what had originally been identified as potential fraud was in fact due to clerical error. The report uncovered several votes by potentially ineligible individuals, including some who were allegedly nonresidents of the City of Milwaukee and some who had allegedly been rendered ineligible due to convictions; however, none of these problems would have been prevented by the imposition of photo ID laws. When questioned about the report, Milwaukee Police Chief Edward Flynn stated that the report itself was never properly vetted within the chain of command and the recommendations contained therein were ill-advised and should have never been included.
Another frequently cited event involved a Brooklyn voter fraud ring, which conspired to manufacture votes decades ago, and consisted of coordinated “ballot stuffing” fraud by local officials and election workers. The crime in question involved forgery of voter registration cards, filing of the cards with the New York Board of Elections and recruitment of people to cast votes in the names of the fictitious voters. Proponents of photo ID laws argue that this fraud ring could have been stopped by imposition of photo ID laws,  but this is a dubious conclusion. That is because election workers themselves were instrumental to the fraud. Photo ID laws, of course, depend on honest enforcement of the law by election officials. If they are part of a conspiracy to commit fraud, requiring them to check photo ID would not prevent fraud.
Another claim commonly used to support imposition of photo ID laws is the 2010 state representative race in Kansas City, Missouri that was allegedly “stolen” when one candidate, J.J. Rizzo, allegedly received more than 50 votes illegally cast non-citizens of Somali descent. Due to the closeness of the race, that event was fully investigated and litigated. There no indictments or no prosecutions after a full review of the incident in question and no finding of illegal voting by the Missouri Court of Appeals after full and fair litigation of all the allegations of election irregularities. The court, in reviewing the reports of election irregularities found credible the election judges who testified, “without contradiction, that all persons who were given a ballot were registered voters who verified their identity by showing proper identification in the check in process,” including the Somali voters who were allegedly non-citizens at the time of the election. Nonetheless, proponents of photo ID laws continue to use this incident in articles, op-eds and blogs as an example of non‑citizen voting.
The most recent allegations of voter fraud are similarly unsupported and unreliable.
More recent allegations of voter fraud that fit the pattern of relying upon allegations that seem particularly egregious on the surface, but when closely examined contain little evidence of voter fraud, and utterly no evidence of impersonation fraud that would be solved by enactment of photo ID laws.
In an attempt to prove voter fraud is a serious problem in New Mexico, Secretary of State Dianna Duran turned over 64,000 cases to the state police for investigation of whether voter fraud was committed. In June 2011, Secretary Duran turned over a list of names equal to approximately 5% of New Mexico’s registered voters to the New Mexico State Police for voter fraud investigation. Secretary Duran took this ill-advised action during the consideration of a proposed photo ID law by the New Mexico legislature. To identify the names, Secretary Duran stated that her office used names and birthdates to affect the matches between the voter registration lists and the lists of foreign nationals. She further stated that 117 registrants from the voter registration list had social security numbers that did not match their name. It is important to note that New Mexico has over 900,000 registered voters, most of whom fill out their voter registration card by hand, from which the data must be entered into a centralized data system.
Ultimately, there is no indication that Secretary Duran’s analysis included any evaluation or follow‑up to determine if any or all of the alleged incidents of voter fraud were the result of data entry errors, unreadable voter registration forms or some other accidental source for the confusion. When interviewed, Secretary Duran’s office admitted that the 64,000 names turned over for investigation could not be considered evidence of voter fraud and may have simply been a result of administrative errors. In addition, in reviewing the hundreds of thousands of names on the list of registered voters and the hundreds of thousands of names on the foreign national license holders lists in New Mexico over an 8‑year period, one should expect to find people on both lists with matching names and birthdates. Consequently, any conclusion of fraud, based solely upon name and birth date matches in a population of hundreds of thousands of people should be viewed with suspicion. Once again, there is no evidence that any of the types of voter fraud alleged here could be prevented by the introduction of photo ID laws.
In a report issued on March 8, 2011, Colorado, the Secretary of State Scott Gessler’s office identified 11,805 non-citizens allegedly illegally registered to vote in the state, of whom 4,947 allegedly cast a ballot in the 2010 elections. Secretary Gessler’s office, based upon the report was, “nearly certain” that 106 American immigrants were improperly registered to vote in Colorado. The report’s conclusion that there are over 11,805 improperly registered voters and of those 4,000 people improperly voted in the 2010 elections are called into question by the qualifying statements and equivocal recommendations contained in the report.
Secretary Gessler’s report admits that inconclusive voter registration data does not prove that all 11, 805 persons it identified were registered improperly. It concludes that even where there are improper registrations, they could have been due to unintentional registration, clerical or other administrative failure without any intention of the registrants to vote or commit voter fraud. The report is utterly silent on how it arrived at the conclusion that over 4000 of the “improper registrants” voted in the 2010 election. There is simply a barely-supported, conclusory statement that “it is likely” that many of the 4,214 registrants in question were not citizens when they cast their vote in 2010. Compare the 106 registered voters that the report alleges are “virtually certain” that they are not citizens, with no attempt to suggest that any of those 106 persons actually voted in 2010 or intended to commit fraud.
Most important, the analysis itself was flawed. The study used a non-citizen resident list dating from 2006 to identify over 4900 non-citizens who allegedly voted in 2010. While his process removed duplicates created when a person used two different non-citizen sources of identification to apply for or renew their driver’s license or identification card, there is no indication that Gessler’s process removed people from the list once they had become citizens. The Secretary did not cross-check those names against the names of over 30,000 Americans that became citizens in Colorado between 2006 and 2009 before making unsupported allegations that 4900 non-citizens voted in the 2010 election in Colorado. Like many of the other common allegations of potential voter fraud, Secretary Gessler’s report is insufficient to support any real claim of voter fraud. Moreover, it is worth noting that because Secretary Gessler’s methodology was to compare non-citizen lists with motor vehicle license or state identification card lists, each one of the allegedly “illegal” voters had photo ID to prove their identity.
A May 23, 2011 letter to the Wall Street Journal, Kansas Secretary of State Kris Kobach used 221 incidents of reported voter fraud in the 13-year period between 1997 and 2010 to allege that voter fraud is a concern in Kansas. This allegation of voter fraud relies upon data about “reported” events and “allegations” of problems with no reference to actual prosecutions, arrests or actual findings of voter malfeasance. A review of the 221 incidents of reported “voter fraud” that he cites revealed that the categories of violations included: electioneering too close to a polling location, failure to deliver voter registration cards, improper ballot challenges, registration cards containing improper zip codes, non-citizen registration (no allegation of non-citizen voting), intimidation of poll workers, double‑voting and voter impersonation. Of the seven convictions arising out of the incidents of “voter fraud” there were two for electioneering and the remainder for double-voting between states or counties. None of the seven convictions based upon the 221 allegations over 13 years would have been prevented by the introduction of photo ID laws.
Recently, student voters in Maine have been targeted for criminal investigation based on their student status. In July, Maine Republican Party Chairman Charlie Webster claimed that a list of 206 University of Maine students who both paid out-of-state tuition and voted in Maine elections was evidence of potential voter fraud. He turned this list over to the newly-elected Maine Secretary of State, who immediately announced that his office would add the student list to an inquiry about voter fraud and called upon the Attorney General to assist in a criminal investigation.
But under Maine law, as in other states, the residency rules for tuition are very different from those for voting; many students meet the legal voting residency requirements while still being ineligible for in-state tuition. Maine students are eligible to vote in local elections so long as they meet state voting residency requirements, which require an individual to affirm that his or her residence “is that place where the person has established a fixed and principal home to which the person, whenever temporarily absent, intends to return.” As long as a student considers a campus address to be a fixed residence and has no immediate intention to leave, he or she may lawfully register to vote at school. The receipt of in-state tuition, on the other hand, is governed by a completely different and more restrictive set of rules. The University of Maine System’s residency guidelines require a student to have lived in Maine for a full year to be considered for in-state tuition, among other requirements. It is worth noting that this rule, if applied as voting residency requirements, would be plainly unconstitutional.
A student paying out-of-state tuition who voted in a Maine election is not evidence of voter fraud. Like many “anecdotal” claims of voter fraud, this is an example of alleging fraud where no illegal conduct exists. The practice of using a population that may be particularly vulnerable to charges of voter fraud and alleging illegal conduct to raise the suspicions of citizens and politicians is particularly abhorrent. Once again, there are no allegations here that would be prevented by the introduction of a photo ID law.
* * *
Given the amount of speculation and misinformation in the public sphere concerning in-person impersonation fraud, and restrictions ostensibly intended to address such fraud, we thank the Subcommittee for sponsoring this hearing. This represents a welcome effort to ensure that the serious policy debate around election reform remains grounded in the facts.
The available empirical research shows that although in-person impersonation fraud is an occurrence of extraordinary rarity, it has been used to justify policies that appear to offer little benefit and impose substantial cost. The existing safeguards and deterrents have been successful in preventing in-person impersonation fraud to any significant degree; further measures are not only unnecessary, but risk compromising the integrity of our elections to the extent that they shut out eligible citizens.
 For an overview of these new laws, see Wendy Weiser & Nhu-Y Ngo, Voting Rights in 2011: A Legislative Roundup, at http://www.brennancenter.org/content/resource/voting_rights_in_2011_a_legislative_round-up/ (July 15, 2011). A comprehensive report on these laws will be available next week at the same page.
 Alaska, Arkansas, California, Colorado, Connecticut, Delaware, Hawaii, Illinois, Iowa, Maine, Maryland, Massachusetts, Mississippi, Nebraska, Nevada, New Mexico, New York, Ohio, Virginia and West Virginia. There is pending photo ID legislation in New Jersey and Pennsylvania.
 Alabama, Kansas, South Carolina, Tennessee, Texas, and Wisconsin.
 Kansas, Texas, and Wisconsin.
 Minnesota, Missouri, Montana, New Hampshire and North Carolina. The New Hampshire State Senate will take up the measure to override the gubernatorial veto in September 2011.
 Those challenging photo ID laws to date have argued that they violate federal or state constitutional protections under a variety of theories: they impose constitutionally unjustified burdens on voters without photo ID; the differential treatment of voters with and without photo ID violates the Equal Protection Clause of the Fourteenth Amendment; requiring a voter without ID to pay to obtain one violates the Twenty-fourth Amendment’s prohibition on poll taxes; and photo ID laws run afoul of various state constitutional protections of the right to vote, which may be stronger than their counterparts in the U.S. Constitution.
 See generally Vishal Agraharkar, Wendy Weiser & Adam Skaggs, The Cost of Voter ID: What the Courts Say, Brennan Center for Justice 3–4 (2011), available at http://brennan.3cdn.net/2f0860fb73fd559359_zzm6bhnld.pdf.
 Crawford v. Marion County Election Bd., 553 U.S. 181, 199 (2008).
 Id. at 200–03.
 Id. at 201.
Brennan Center for Justice, Citizens Without Proof (2006), available at http://www.brennancenter.org/page/-/d/download_file_39242.pdf.
 Lorrie Frasure et al., 2008 Collaborative Multi-racial Post-election Survey: Comparative Multi-racial Survey Toplines 24 (2008), available at http://cmpstudy.com/assets/CMPS-toplines.pdf.
 Matt A. Barreto, Stephen A. Nuno, & Gabriel R. Sanchez, The Disproportionate Impact of Indiana Voter ID Requirements on the Electorate (2007), available at http://depts.washington.edu/uwiser/documents/Indiana_voter.pdf
 Brennan Center for Justice, supra note 11.
 Voter Identification in Indiana: A Demographic Analysis of Impact on Older Indiana Citizens (2005) (survey of 843 Indiana registered voters aged 60 and older). This survey was attached to the Plaintiffs’ Motion for Summary Judgment at a district court proceeding in Crawford v. Marion County Election Board. See Motion for Summary Judgment by Plaintiffs, attach. 8, Ind. Democratic Party v. Rokita, 458 F. Supp. 2d 775 (2007).
 Robert Greenstein, Leighton Ku & Stacey Dean, Survey Indicates House Bill Could Deny Voting Rights to Millions of U.S. Citizens 1 (2006), available at http://www.cbpp.org/files/9–22–06id.pdf..
 Matt A. Barreto, Stephen A. Nuno, & Gabriel R. Sanchez, Voter ID Requirements and the Disenfranchisement of Latino, Black and Asian Voters (2007), available at http://www.brennancenter.org/dynamic/subpages/download_file_50884.pdf.
 M.V. Hood III & Charles S. Bullock, III, Worth a Thousand Words?: An Analysis of Georgia’s Voter Identification Statute, 36 Am. Politics Research, no. 4, July 2008 at 555–579, available at http://apr.sagepub.com/content/36/4/555.abstract.
 Frasure et al., supra note 12.
 See Hans A. von Spakovsky & Alex Ingram, Without Proof: The Unpersuasive Case against Voter Identification (2011), available at http://origin.heritage.org/research/reports/2011/08/without-proof-the-unpersuasive-case-against-voter-identification.
 See Wendy Weiser, Keesha Gaskins & Sundeep Iyer, “Citizens Without Proof” Stands Strong, Sept. 8, 2011, http://www.brennancenter.org/content/resource/citizens_without_proof_stands_strong/. .
 von Spakovsky & Alex Ingram, supra note 20.
 IDs exceed voter-age residents, The Columbus Dispatch (July 25, 2011) http://thevotingnews.com/state/ohio/ohio-ids-exceed-voter-age-residents-the-columbus-dispatch/.
 Telephone Interview by Maria da Silva with Division of MotorVehicles clerk, Kansas Department of Revenue (May 27, 2011).
 It is not unreasonable to assume that most of these residents still had current (or very recently expired) licenses at the time they moved. Because Kansas purges its license lists after five years of inactivity following expiration, most of the people who moved between 2005 and 2009 could still have been on the state’s DMV rolls in 2010. 350,000 is 16.2% of the number of licenses that Secretary Kobach cites.
 Da Silva Interview, supra note 27.
 According to the Centers for Disease Control, 122,059 Kansans died between 2005 and 2009. Only about 3% of deaths are among those below 24. Accordingly, if we conservatively estimate that 3% of deaths were among those below 18, then about 118,400 Kansans above 18 died between 2005 and 2009. See Centers for Disease Control and Prevention, National Vital Statistics Report, Deaths by 10-year Age Groups: United States and Each State, 1999–2007 (2010), http://www.cdc.gov/nchs/nvss/mortality/gmwk23f.htm; Centers for Disease Control and Prevention, National Vital Statistics Report, Births, Marriage, Divorces, and Death: Provisional Data for 2009 (2010), http://www.cdc.gov/nchs/data/nvsr/nvsr58/nvsr58_25.htm (including estimates from 2008 and 2009).
 U.S. Census Bureau, 2009 American Community Survey Estimates (2010) (custom table identifying citizenship status of Kansas residents by age).
 These concerns apply to similar comparisons made in other states. A July 2011 article in the Columbus Dispatch reported that Ohio has 28,000 more driver’s licenses than voting age residents. But in 2009, 165,954 people over the age of 18 left Ohio. Approximately 105,000 individuals over the 18 die in Ohio annually (National Vital Statistics System, Centers for Disease Control and Prevention, Deaths by 10-year age groups: United States and each state, 2007 (2010), http://www.cdc.gov/nchs/data/dvs/MortFinal2007_Worktable23f.pdf.). And according to the American Community Survey, in 2009 there were 191,439 residents of Ohio who were over 18 but not citizens. As in Kansas, many of these people are likely to be on the State’s driver’s license lists, but none of them are eligible to vote in the state.
 R. Michael Alvarez, Delia Bailey, & Jonathan N. Katz, the Effect of Voter Identification Laws on Turnout (2007), available at http://brennan.3cdn.net/c267529e2bb704e85d_u0m6ib08s.pdf.
 R. Michael Alvarez, Delia Bailey, & Jonathan N. Katz, An Empirical Bayes Approach to Estimating Ordinal Treatment Effects 26–30 (2010), available at http://brennan.3cdn.net/a5782740e4185414a8_snm6bhfwg.pdf.
 Alvarez et al. supra note 34, at 2.
 Id. at 2 – 3.
 Id. at 3.
 See David Muhlhausen, Photo ID Laws Do Not Reduce Voter Turnout, Heritage Foundation, May 5, 2009. Available at: http://www.heritage.org/research/testimony/photo-id-laws-do-not-reduce-voter-turnout#_edn15 (citing Jason D. Mycoff, Michael W. Wagner, and David C. Wilson, The Empirical Effects of Voter-ID Laws: Present or Absent" PS: Political Science & Politics, 42 (2009)).
 Mycoff et al. do apply a regression model to individual-level data in order to determine how photo ID affects voter turnout. In statistics, regression analysis includes any techniques for modeling and analyzing several variables, when the focus is on the relationship between a dependent variable and one or more independent variables. More specifically, regression analysis helps one understand how the typical value of the dependent variable changes when any one of the independent variables is varied, while the other independent variables are held fixed. But instead of assessing the effect of photo ID on turnout among groups most likely not to have ID, the authors group together all voters when examining how photo ID affects turnout. This could “drown out” the effect of photo ID on turnout among minorities and elderly, groups less likely to possess ID.
 Hans Von Spakovsky, Op-Ed, Voter ID Was a Success in November, Wall St. J., Jan. 30, 2009, http://www.washingtonpost.com/opinions/how-voter-id-laws-keep-elections-honest/2011/06/22/AGS6UkjH_story.html.
 Brian Kemp, Letter to the Editor, How Voter ID Laws Keep Elections Honest, Wash. Post, June 24, 2011, http://www.washingtonpost.com/opinions/how-voter-id-laws-keep-elections-honest/2011/06/22/AGS6UkjH_story.html.
 Justin Levitt, The Truth About Voter Fraud (2007), available at http://www.brenna