Skip Navigation

In-Person Voter Fraud Myth, Justin Levitt Before Senate Committee

  • Justin Levitt
Published: March 13, 2008

>Down­load exec­ut­ive summary
>Down­load PDF of this testi­mony
>Watch archived video of testi­mony

Testi­mony of

Justin Levitt, Coun­sel

Bren­nan Center for Justice at NYU School of Law

Before the

United States Senate Commit­tee on Rules and Admin­is­tra­tion

In-Person Voter Fraud:
Myth and Trig­ger for Disen­fran­chise­ment?

March 12, 2008

On behalf of the Bren­nan Center for Justice at New York Univer­sity School of Law, I thank the Senate Commit­tee on Rules and Admin­is­tra­tion for hold­ing this hear­ing – and for provid­ing the oppor­tun­ity to discuss some of the facts and myths about voter fraud, and the effects that some of these myths are having on eligible Amer­ican citizens. 

My name is Justin Levitt, and I am coun­sel at the Bren­nan Center.  The Bren­nan Center is a non-partisan public policy and legal advocacy organ­iz­a­tion that focuses on funda­mental 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 issues relat­ing to elec­tion admin­is­tra­tion, includ­ing work to remove unne­ces­sary barri­ers to voter regis­tra­tion; to make voting machines more secure, reli­able, usable, and access­ible; and to expand access to the fran­chise and ensure that elec­tions are conduc­ted with integ­rity.  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. 

We have paid partic­u­lar atten­tion in recent years to claims of voter fraud.  We have collec­ted alleg­a­tions of fraud cited by state and federal courts, bipar­tisan federal 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.  Most recently, we published a mono­graph reflect­ing our analysis, entitled “The Truth About Voter Fraud,” which compiles for the first time the recur­ring meth­od­o­lo­gical flaws behind the alleg­a­tions of wide­spread voter fraud that are frequently cited but often unsup­por­ted.[1]  We have simil­arly examined claims of voter fraud in amicus briefs filed with courts around the coun­try, includ­ing cases currently pending at the appel­late level and with the Supreme Court.[2]

We have also reviewed, in detail, the effect of policies and laws that contrib­ute to the disen­fran­chise­ment of eligible citizens.  We attempt to bring reli­able data to bear on the effort to assess the nature and magnitude of the impact of new elec­tion rules, partic­u­larly those with the poten­tial to burden eligible citizens’ efforts to exer­cise their right to vote.  In help­ing to quantify the impact of these rules, we have sponsored surveys and soph­ist­ic­ated stat­ist­ical analyses; we have collec­ted affi­davits and anec­dotes; and we have conduc­ted in-depth review of voter regis­tra­tion forms and voter regis­tra­tion rolls, line by line.

In my testi­mony today, I will share some of our find­ings.  Our research suggests that the incid­ence of fraud by those imper­son­at­ing others at the polls is strik­ingly rare.   Yet we have seen restric­tions proposed to address this perceived or inven­ted threat, often suppor­ted by stor­ies about elec­tion fraud or abnor­mal­it­ies that the restric­tions would not actu­ally prevent.  Further empir­ical research shows that the prob­lems caused by some of these restric­tions are far more seri­ous than the prob­lems they allegedly resolve.  As we stated in The Truth About Voter Fraud, “[t]he voter fraud phantom drives policy that disen­fran­chises actual legit­im­ate voters, without a corres­pond­ing actual bene­fit.”[3]

I.       The Myth of In-Person Voter Fraud

I will focus here on a partic­u­lar type of voter fraud: in-person imper­son­a­tion fraud, or the attempt to imper­son­ate another voter at the polls.  Alleg­a­tions concern­ing the incid­ence of or poten­tial for this sort of fraud have been cited as justi­fic­a­tion for vari­ous restric­tions on the exer­cise of the fran­chise, includ­ing some of the more prom­in­ent elec­tion law contro­ver­sies in the coun­try.  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.  So it is to this in-person imper­son­a­tion fraud, and to our extens­ive research on the topic, that I will direct my testi­mony today.

As explained in more detail below, we conclude that the incid­ence of actual in-person imper­son­a­tion fraud is extraordin­ar­ily rare.  Though it does occur, there are only a hand­ful of recent accounts, even fewer of which have been substan­ti­ated.  During this same period, hundreds of millions of ballots have been cast.  In the past few years in partic­u­lar, the prior­ity placed on the issue should have fostered the discov­ery of any substan­tial quant­ity of in-person imper­son­a­tion fraud; the most notable signi­fic­ance of the incid­ents that have surfaced, however, is how rare they appear to be.

We arrived at our conclu­sion primar­ily through a focus on evid­ence: extens­ive research of reports, cita­tions, and claims of fraud, in popu­lar and schol­arly public­a­tions, and in docu­ments provided to and produced by public and private invest­ig­a­tions.  We have prior­it­ized more recent claims, and partic­u­larly claims purport­ing to reveal in-person imper­son­a­tion fraud.  Our review and analysis spans thou­sands of accounts, includ­ing every single asser­tion of fraud in the most compre­hens­ive collec­tion of claims of in-person imper­son­a­tion fraud to date: the cita­tions presen­ted to the Supreme Court in the Craw­ford v. Marion County Elec­tion Board case. 

Most Alleg­a­tions of Fraud Do Not Involve In-Person Imper­son­a­tion Fraud

We discovered that in order to assess the incid­ence of in-person imper­son­a­tion fraud, it is first neces­sary to cut through a large amount of noise.  Upon survey­ing the news after many, if not most, elec­tions, it is easy to find alleg­a­tions of elec­tion fraud of one sort or another.  The vast major­ity of these alleg­a­tions do not involve in-person imper­son­a­tion fraud.

Many of the appar­ent alleg­a­tions of fraud are not alleg­a­tions of fraud at all.  Some­times “voter fraud” stops at the head­line: the story claims to involve fraud, but the substance of the story does not support the bold­face nine-word summary.[4]  Another set of these reports present, not alleg­a­tions that fraud has occurred, but spec­u­la­tion that fraud might occur in the future.[5]  Still other reports show indi­vidu­als crying foul, but the facts describe straight­for­ward admin­is­trat­ive errors.[6]  Such errors are not to be ignored – but they are also not to be confused with fraud.

Some of these post-elec­tion reports actu­ally do present worri­some alleg­a­tions of fraud – but only rarely do they involve alleg­a­tions of in-person imper­son­a­tion fraud.  Instead, they allege schemes involving fraud­u­lent absentee ballots;[7] or absentee voters who have been coerced;[8] or conspir­acies to buy votes;[9] or efforts to tamper with ballots or machines or count­ing systems.[10]  Occa­sion­ally, and now more rarely, there are reports of fraud­u­lent regis­tra­tion forms – usually involving rogue work­ers hoping to cheat nonprofit organ­iz­a­tions out of an honest effort to register real citizens.[11]  We are aware of no recent substan­ti­ated case in which such regis­tra­tion fraud has resul­ted in an attempt to cast a fraud­u­lent vote.  All of these reports should be invest­ig­ated, and any wrong­do­ing should not be condoned.  Yet they too should not be confused with in-person imper­son­a­tion fraud.

Many Alleg­a­tions of Fraud Are Also Plagued By Recur­ring Meth­od­o­lo­gical Errors

One subset of the reports of alleged fraud is partic­u­larly toxic because, on first blush, these reports appear to entail the trap­pings of science, and there­fore seem more conclus­ive than they actu­ally are.  Alleg­a­tions of this sort rely on efforts to match compu­ter­ized voter rolls either to other compu­ter­ized voter rolls or to compu­ter­ized lists of ineligible indi­vidu­als, such as regis­tries of the deceased; such efforts usually involve an attempt to match names and dates of birth from one list to another.  A vari­ant involves the attempt to screen voter rolls against prop­erty or zoning records.  The results are then usually trum­peted as conclus­ive evid­ence of so many dead voters,[12] or double voters,[13] or voters rendered ineligible because of convic­tion,[14] or voters from vacant lots.[15]

There are common meth­od­o­lo­gical concerns with such match­ing efforts, includ­ing both the qual­ity of the under­ly­ing lists and the partic­u­lar protocol used in match­ing list to list.  We have reviewed these errors at length in several public­a­tions, includ­ing The Truth About Voter Fraud.[16]  Even without an error, however, it is never­the­less a mistake to draw final conclu­sions from these sorts of match­ing exer­cises.  As Professor Michael McDon­ald and I demon­strate in a forth­com­ing article for the peer-reviewed Elec­tion Law Journal, element­ary stat­ist­ics confirms that in any substan­tial pool, it is quite common to find two differ­ent indi­vidu­als who share the same name and date of birth.[17]  When compar­ing one list of millions of voters to another list of millions of ineligible indi­vidu­als, it should not be surpris­ing to find hundreds of perfect “matches” that actu­ally repres­ent differ­ent indi­vidu­als, known to record-link­age experts as “false posit­ives.”  The incid­ence of such matches reveals stat­ist­ics at work, not fraud. 

Moreover, the rare cases when match­ing efforts do actu­ally reveal elect­oral miscon­duct rarely involve in-person imper­son­a­tion fraud.  For double voters or citizens inten­tion­ally voting despite a disen­fran­chising convic­tion, the few wrong­do­ers who may be caught by such match­ing efforts are caught precisely because they vote using their real name.  Those few indi­vidu­als voting in the names of a deceased citizen find it far easier to do so by absentee ballot.  Such instances do not repres­ent in-person imper­son­a­tion fraud.

In sum, we have reviewed many past accounts of fraud, and track contem­por­ary accounts as they arise.  Our research confirms that there are hundreds of reports of alleged fraud, in thou­sands of elec­tions, with millions of ballots cast.  Yet after wading through the false and irrel­ev­ant reports categor­ized above, only a hand­ful of reports remain that even allege, much less substan­ti­ate, instances of in-person imper­son­a­tion fraud.

Notori­ous Alleg­a­tions of In-Person Imper­son­a­tion Fraud Have Been Debunked

Even fewer of these alleg­a­tions stand up to real scru­tiny.  Indeed, care­ful invest­ig­a­tion has more often than not debunked, not confirmed, alleg­a­tions of in-person imper­son­a­tion fraud.  One notori­ous and recur­ring example is a 2000 invest­ig­at­ive report in the Atlanta Journal-Consti­tu­tion, claim­ing that “the actual number of ballots cast by the dead” was “5,412 in the past 20 years.”[18]   The article has been favor­ably cited by an Assist­ant U.S. Attor­ney General,[19] a Governor,[20] a state Secret­ary of State,[21] and several state Attor­neys General,[22] among others. 

This article did not, however, actu­ally reveal 5,412 ballots cast by the dead, much less 5,412 instances of in-person imper­son­a­tion fraud.[23]  Instead, it revealed 5,412 matches of Social Secur­ity death records to voting records.  And it further revealed that these matches are flawed.  The reporter acknow­ledged that death records contain errors, list­ing people as dead who are actu­ally alive, but appar­ently did not invest­ig­ate how many of the 5,412 iden­ti­fied ballots suffered from this error.  The reporter also acknow­ledged that voter records contain errors, reflect­ing data entry mistakes and those who sign the wrong line of a poll­book, but appar­ently could not or did not invest­ig­ate how many of the 5,412 iden­ti­fied ballots suffered from this error.  The reporter neither acknow­ledged nor appar­ently accoun­ted for the stat­ist­ical like­li­hood that a record of John Smith dying and a record of John Smith voting might in fact reflect differ­ent “John Smith”s with the same date of birth.[24]  Finally, the reporter did not indic­ate how many of these 5,412 ballots were cast in person, rather than absentee.

Indeed, the article iden­ti­fied only one indi­vidual concretely alleged to have been the victim of in-person imper­son­a­tion fraud.  It cited the case of “[Alan Jay] Mandel, the tobacco shop owner, whose voter certi­fic­ate was signed at the polls by someone after his death.”[25]  Repeated the reporter, “[S]omebody defin­itely signed his name on a voter certi­fic­ate on Nov. 3, 1998.”[26]

This alleg­a­tion, though amount­ing to only one concrete alleg­a­tion of in-person imper­son­a­tion fraud in approx­im­ately twenty million votes over 20 years,[27] would never­the­less be disturb­ing – if it were true.  Further invest­ig­a­tion, however, proved the alleg­a­tion false.  The signa­ture (and voter certi­fic­ate) in ques­tion belonged to Alan J. Mandell (with two “l”s), who was very much alive and eligible in 1998, but whose vote was mistakenly recor­ded in the name of Alan Jay Mandel (with one “l”).[28]

Invest­ig­a­tion as thor­ough as the invest­ig­a­tion into the vote of Mr. Mandel/Mandell is rare.  Never­the­less, when research­ers do expend the effort to follow through on initial alleg­a­tions of in-person imper­son­a­tion fraud, they often find those alleg­a­tions to be unwar­ran­ted.  A 2007 invest­ig­a­tion of approx­im­ately 100 “dead voters” in Missouri, for example, revealed that every single purpor­ted case was prop­erly attrib­uted either to a match­ing error, a prob­lem in the under­ly­ing data, or a cler­ical error by elec­tions offi­cials or voters.[29]  Like­wise, after compil­ing a list of poten­tial “dead voters” in New York state, a Pough­keep­sie journ­al­ist invest­ig­ated seven local cases – and found that seven out of seven reflec­ted cler­ical errors or other mistakes, not fraud.[30]  An invest­ig­a­tion in Hawaii in 1999, after review­ing precinct poll­books and call­ing allegedly deceased citizens, simil­arly found that not one of 170 poten­tial “dead voters” actu­ally reflec­ted fraud.[31]

Every once in a great while, a report of in-person imper­son­a­tion fraud, or a report of an attempt at in-person imper­son­a­tion fraud, appears to be substan­ti­ated.[32]  It has been known to happen.  What is notable, however, is how rarely it has been known to happen.  Amer­ic­ans are struck and killed by light­ning more often.[33]

Recent Examples Fit the Pattern: Little to No Evid­ence of In-Person Imper­son­a­tion

Two very recent and very prom­in­ent invest­ig­a­tions into voter fraud precisely fit the over­all pattern that I have described above: many claims of wrong­do­ing and irreg­u­lar­ity, but few that even allege in-person imper­son­a­tion fraud, and a tiny portion, if any, that substan­ti­ate the alleg­a­tions.  These examples are: the compil­a­tion of alleged fraud submit­ted to the Supreme Court in Indi­ana’s Craw­ford v. Marion County Elec­tion Board case, and the recent invest­ig­a­tion by the Milwau­kee Police Depart­ment of fraud in Milwau­kee in 2004.

The Craw­ford case now pending before the Supreme Court provided the most prom­in­ent focal event to date for those who believe in-person imper­son­a­tion fraud to be a legit­im­ate concern to present their proof.  In the case, the lower courts cited several media accounts that, the courts claimed, reflec­ted reports of in-person imper­son­a­tion fraud.[34]  In the Supreme Court, respond­ents and amici support­ing respond­ents added cita­tions to more than 250 reports, encom­passing decades of elec­tions.[35]  The Bren­nan Center thor­oughly examined each one of these cita­tions.[36]  The evid­ence of in-person imper­son­a­tion fraud was strik­ingly sparse.  The vast major­ity of cited reports reflec­ted either alleg­a­tions that could not possibly be related to in-person imper­son­a­tion fraud (e.g., absentee ballot prob­lems, or vote-buying schemes, or ballot tamper­ing), or alleg­a­tions that did not mention whether the alleged wrong­do­ing was commit­ted in-person or through more suscept­ible absentee ballots.[37]  In elec­tions encom­passing hundreds of millions of votes over­all, the cited reports mentioned only eleven votes and two attemp­ted but unsuc­cess­ful votes that allegedly involved in-person imper­son­a­tion fraud.  Two of the alleg­a­tions – and the two unsuc­cess­ful attempts – were substan­ti­ated.[38]  The other nine might just have read­ily involved the same sorts of cler­ical errors uncovered in the invest­ig­a­tions discussed above.[39]

A recent invest­ig­a­tion by the Milwau­kee Police Depart­ment high­lights the same pattern.  After the 2004 elec­tion, initial media accounts featured front-page alleg­a­tions of wide­spread fraud in Wiscon­sin.[40]  On Febru­ary 26, 2008, the Milwau­kee Police Depart­ment released a report on that elec­tion, with what appears to be a painstak­ing invest­ig­a­tion of the facts, and policy recom­mend­a­tions offered with less care and disavowed by the Milwau­kee Police Chief.[41]  The depart­ment’s care­ful factual invest­ig­a­tion primar­ily revealed admin­is­trat­ive mistakes and, occa­sion­ally, negli­gence.[42]  It showed that much of what had origin­ally been iden­ti­fied as poten­tial fraud was in fact due to cler­ical error.[43]  It also uncovered several votes by poten­tially ineligible indi­vidu­als, includ­ing some who were allegedly nonres­id­ents, and some who had allegedly been rendered ineligible due to convic­tions.[44]  The report revealed only one poten­tial vote that might have involved in-person imper­son­a­tion fraud, with no docu­ment­a­tion veri­fy­ing that the vote in ques­tion was actu­ally cast.[45]

Exist­ing Safe­guards Would Disclose In-Person Imper­son­a­tion Fraud

Some have claimed that the incid­ence of alleged in-person imper­son­a­tion fraud is extremely low because in-person imper­son­a­tion fraud is diffi­cult to detect.[46]  This is distinct from the issue of whether in-person imper­son­a­tion fraud is diffi­cult to prosec­ute: a topic far more suited to the expert­ise of my fellow panel­ist Mr. Iglesias.  Rather, this claim seeks to explain the fact that reports of poten­tial in-person imper­son­a­tion fraud are rare, by assert­ing that in-person imper­son­a­tion fraud is diffi­cult to discover.

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 in-person 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.[47]  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 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.

As in all law enforce­ment, none of these detec­tion mech­an­isms are perfect.  Yet in hundreds of millions of ballots cast, they have yiel­ded only a hand­ful of poten­tial instances of in-person imper­son­a­tion fraud, precisely during a period when invest­ig­at­ing voter fraud was expressly deemed a federal law enforce­ment prior­ity,[48] and when private entit­ies were equipped and highly motiv­ated  to seek, collect, and dissem­in­ate such reports.[49]  Every year, there are far more reports of UFO sight­ings.[50]  The scarcity of reports of in-person imper­son­a­tion fraud, in this context, is itself mean­ing­ful.

In-Person Imper­son­a­tion Fraud Is Irra­tional

Instead, a more logical explan­a­tion for the extraordin­ary rarity of repor­ted in-person imper­son­a­tion fraud is that in-person imper­son­a­tion fraud is extraordin­ar­ily rare.  In-person imper­son­a­tion fraud is an extremely inef­fi­cient means to influ­ence an elec­tion.  For each act of in-person imper­son­a­tion fraud in a federal elec­tion, the perpet­rator risks 5 years in prison and a $10,000 fine under federal law, in addi­tion to any penal­ties assessed under state law.[51]  In return, the perpet­rator gains at most one incre­mental vote.  It is sens­ible that few indi­vidu­als believe such a trade-off worth­while.    

II.    The Consequences of the Myth

If perpet­rat­ing the myth of in-person imper­son­a­tion fraud had no consequences, it would likely be of little concern to this Commit­tee.  There are consequences, however, and these consequences have been and can be quite seri­ous.

First, “crying wolf” about voter fraud, and partic­u­larly in-person imper­son­a­tion fraud, perpetu­ates the lack of public confid­ence in the integ­rity of the elec­tion process.  Second, it distracts atten­tion from the real and recur­ring prob­lems that do commonly under­mine our elec­tions, includ­ing some of the other forms of fraud or admin­is­trat­ive irreg­u­lar­ity I have discussed above.  Third, it can be used to justify placing undue and improper pres­sure on impar­tial prosec­utors to bring unwar­ran­ted crim­inal prosec­u­tions, a subject that my fellow panel­ist Mr. Iglesias is far more equipped to discuss.  

Claims of Fraud Are Used to Justify Policies that Do Not Correct Real Prob­lems

“Crying wolf” about voter fraud has also been used to justify policies that lead to the disen­fran­chise­ment of eligible citizens, includ­ing the impos­i­tion of overly restrict­ive voter iden­ti­fic­a­tion require­ments, undue purges of legit­im­ate voters from the rolls, and unwar­ran­ted restric­tions on the regis­tra­tion process, impact­ing both indi­vidual voters and organ­iz­a­tions. 

The Milwau­kee Police Depart­ment report provides another timely example.  Though the depart­ment’s chief disavowed the report’s policy recom­mend­a­tions, the report’s authors cited their conclu­sions as a reason to pass legis­la­tion requir­ing photo iden­ti­fic­a­tion docu­ments as a condi­tion of voting.[52]  Over the course of the last week, vari­ous indi­vidu­als have echoed this argu­ment in the press, citing the report’s find­ings as support­ing photo ID legis­la­tion.[53]

The report itself, however, iden­ti­fies no confirmed instance of in-person imper­son­a­tion fraud, despite a thor­ough search for such evid­ence.  That is, the report did identify vari­ous defi­cien­cies with the elec­tion process, none of which could be solved by requir­ing govern­ment-issued photo ID at the polls, and it did not confirm any actual wrong­do­ing that a poll site iden­ti­fic­a­tion require­ment would correct or prevent.  In our exper­i­ence cata­loguing attempts to justify ID require­ments with claims of voter fraud, this logical non-sequitur is a common feature.[54]

Claims of Fraud Are Used to Justify Policies that Cause Real Prob­lems

Moreover, not only are restrict­ive iden­ti­fic­a­tion require­ments poorly tailored to any exist­ing prob­lem of any magnitude, but reli­able empir­ical data demon­strates that they cause prob­lems of their own.

The most prom­in­ent restrict­ive voter iden­ti­fic­a­tion propos­als would require govern­ment-issued photo iden­ti­fic­a­tion as a condi­tion of voting a valid ballot at the polls.[55]  Most eligible voting-age citizens have such iden­ti­fic­a­tion.  Stud­ies have repeatedly shown, however, that many eligible voting-age citizens do not.[56]

Further­more, for those who do not currently have govern­ment-issued photo iden­ti­fic­a­tion, obtain­ing it can be a burden.  In some cases, it takes ID to get ID – for example, a certi­fied birth certi­fic­ate may be required to obtain govern­ment-issued photo iden­ti­fic­a­tion, but govern­ment-issued photo iden­ti­fic­a­tion may be required to get a certi­fied birth certi­fic­ate.[57]  Even when it is possible to get this under­ly­ing docu­ment­a­tion, doing so costs time and money, with fees up to $380 for a replace­ment certi­fic­ate of natur­al­iz­a­tion.[58]  Bring­ing the assembled paper­work to the agency that distrib­utes the photo iden­ti­fic­a­tion costs further time and money, as indi­vidu­als must travel – without a valid driver’s license – to a partic­u­lar office during govern­ment hours.  For indi­vidu­als with disab­il­it­ies or lower-income or more elderly citizens, these burdens may become partic­u­larly acute.

In the two states where govern­ment-issued photo iden­ti­fic­a­tion is required to cast a valid ballot, these laws have already contrib­uted to the disen­fran­chise­ment of eligible citizens.[59]  In one Indi­ana county’s 2007 elec­tions, for example, 32 voters arriv­ing without accept­able photo iden­ti­fic­a­tion cast ballots that were not coun­ted, appar­ently solely because of the voter iden­ti­fic­a­tion law.[60]  Four­teen of these voters had previ­ously voted in at least ten previ­ous elec­tions at these same polling places.[61]  More recently, in Geor­gi­a’s 2008 pres­id­en­tial primary elec­tion, 296 voters arriv­ing without accept­able photo iden­ti­fic­a­tion reportedly cast ballots that were not coun­ted, again appar­ently solely because of the ID law.[62]  It is impossible to know exactly how many addi­tional indi­vidu­als without iden­ti­fic­a­tion arrived at the polls but did not cast the futile provi­sional ballots, or declined to make the futile trip to the polls.  In less than two years, however, there are already reports of far more indi­vidu­als adversely affected by restrict­ive photo iden­ti­fic­a­tion laws than the total cited instances of in-person imper­son­a­tion fraud repor­ted over the last few decades.

There are also disturb­ing indic­a­tions that far more eligible citizens may be affected by such laws in the future.  Reli­able surveys of registered voters conclude that restrict­ive iden­ti­fic­a­tion laws do impact eligible citizens,[63] and that they dispro­por­tion­ately impact minor­ity and elderly popu­la­tions.[64]  Although there is some disagree­ment over the precise magnitude of the effect,[65] even the stud­ies with more modest results estim­ate an impact that would reach more than two million registered voters if applied nation­wide.[66]   Surveys of voting-age citizens – includ­ing eligible citizens who have not been engaged in the elec­tion process, but whom we should hope to engage – are some­what less fully developed, but find an even more substan­tial effect, and similar dispro­por­tion­ate impact.[67]

Some seek to justify these restrict­ive laws, despite their demon­strated impact on many Amer­ican citizens, and despite the fact that they do not correct an exist­ing prob­lem with recur­ring in-person imper­son­a­tion, by claim­ing that they will at least increase public confid­ence in the elec­tion process.[68]  Even if the unfoun­ded fears of the many were suffi­cient justi­fic­a­tion to burden the consti­tu­tional rights of the few,[69] however, a care­ful new study, forth­com­ing in the Harvard Law Review, casts seri­ous doubt on the valid­ity of such asser­tions.  The data show no support for the notion that requir­ing iden­ti­fic­a­tion will increase voter confid­ence; the study found no stat­ist­ic­ally signi­fic­ant correl­a­tion between the rate at which citizens were asked to produce photo ID and their percep­tion that either voter fraud gener­ally, or voter imper­son­a­tion in partic­u­lar, exists.[70]  Photo iden­ti­fic­a­tion laws do not, in short, appear to make citizens feel more secure about their elec­tions.

* * *

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 Commit­tee 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. 

In contrast, there remain seri­ous concerns about aspects of our elec­tion process where exist­ing safe­guards are not suffi­cient, and where remed­ies would not risk harm to eligible voters – includ­ing, for example, threats to the secur­ity of our voting systems, and concerns with the trans­par­ency of purges of the voter rolls.  In this elec­tion season, we still face substan­tial chal­lenges in ensur­ing that all eligible citizens are able to exer­cise the fran­chise effect­ively, and we look forward to assist­ing in the effort to achieve this common goal.

Thank you very much.

[1] Justin Levitt, The Truth About Voter Fraud (2007), avail­able at http://www.bren­nan­cen­­about­voter­fraud/.

[2] Brief of the Bren­nan Center for Justice et al. as Amici Curiae Support­ing Peti­tion­ers, Craw­ford v. Marion County Elec­tion Board, Nos. 07–21, 07–25 (U.S. Nov. 13, 2007) (“Bren­nan Center Brief”); Brief of Bren­nan Center for Justice at NYU School of Law as Amicus Curiae in Support of Plaintiffs-Appel­lants and Reversal, Craw­ford v. Marion County Elec­tion Board, 472 F.3d 949 (7th Cir. 2007); Brief Amicus Curiae of the Bren­nan Center for Justice at NYU School of Law in Support of Plaintiffs/Appel­lants and Reversal, Gonza­lez v. Arizona, 485 F.3d 1041 (9th Cir. 2007); Brief of Bren­nan Center for Justice at NYU School of Law as Amicus Curiae in Support of Plaintiffs/Appellees and Affirm­ance, Case No. 07–02067 (10th Cir. Sept. 17, 2007); Brief of Bren­nan Center for Justice at N.Y.U. Law School as Amicus Curiae in Support of Plaintiffs-Appellees and Affirm­ance, Common Cause/Geor­gia v. Cox, Case No. 05–15784-G (11th Cir. Jan. 14, 2006); Brief of Bren­nan Center for Justice at NYU School of Law as Amicus Curiae in Support of Plaintiffs’ Motion for Summary Judg­ment, ACLU of N.M. v. Santil­lanes, 506 F. Supp. 2d 598 (D.N.M. 2007).

[3] Levitt, supra note 1, at 6.

[4] See, e.g., Megan Matteucci, River­dale Suit Alleges Elec­tion Fraud, Atlanta J.-Const., Nov. 20, 2007; Sandy Cole­man, Randolph Peti­tion Claims Voter Fraud, Boston Globe, May 21, 2006, at 5.

[5] See, e.g., Hoosier Voter Lists Loaded With Bogus Names, South Bend Tribune (Ind.), Nov. 6, 2000, at C4.   Such spec­u­la­tion, for example, may be driven by poor main­ten­ance of the regis­tra­tion rolls, which proper imple­ment­a­tion of the National Voter Regis­tra­tion Act of 1993 and the Help Amer­ica Vote Act of 2002 should correct. See 42 U.S.C. §§ 1973gg-6; 15483(a)(2), (4).

[6] See, e.g., Michelle Hillen, Recount of Runoff Reverses 1st Result, Ark. Demo­crat-Gazette, June 20, 2006.

[7] See, e.g., Pabey v. Pastrick, 816 N.E.2d 1138, 1144–46 (Ind. 2004); In re The Matter of the Protest of Elec­tion Returns and Absentee Ballots in the Novem­ber 4, 1997 Elec­tion for the City of Miami, Flor­ida, 707 So.2d 1170 (Fla. Ct. App. 1998).

[8] See, e.g., Anastasia Hendrix, City Work­ers: We Were Told To Vote, Work for Newsom, S.F. Chron­icle, Jan. 15, 2004; Matthew Purdy, 5 Bronx School Offi­cials Are Indicted in Absentee Ballot Fraud, N.Y. Times, Apr. 25, 1996.

[9] See, e.g., Beth Musgrave, Three Sentenced in Bath Vote Fraud, Lexing­ton Herald-Leader, Sept. 25, 2007; Nick­laus Love­lady, Invest­ig­a­tion Into Vote Fraud in Benton County Nets 14th Arrest, Miss. Clarion-Ledger, Aug. 31, 2007; Tom Searls, Six To Learn Fate in Lincoln Vote Buying Case, Char­le­ston Gazette (W.Va.), May 3, 2006, at 1C; Michael E. Ruane, FBI’s Sham Candid­ate Crawled Under W. Va’s Polit­ical Rock, Wash. Post, Dec. 2, 2005, at A1.

[10] See, e.g., John M. Glionna, S.F., State Wade Into Vote Count Contro­versy, L.A. Times, Nov. 21, 2001.

[11] See, e.g., Todd C. Frankel, 8 Charged in StL Voter Fraud, St. Louis Post-Dispatch, Dec. 21, 2007; Keith Ervin, Felony Charges Filed Against 7 in State’s Biggest Case of Voter-Regis­tra­tion Fraud, Seattle Times, July 26, 2007; Carlos Campos, Bogus Voter Forms Pop Up in Fulton, Atlanta J.-Const., Oct. 21, 2004.

[12] See, e.g., David W. Chen, Among Voters in New Jersey, G.O.P. Sees Dead People, N.Y. Times, Sept. 16, 2005, at B5.

[13] See, e.g., Russ Buettner, Exposed: Scan­dal of Double Voters, Daily News (N.Y.), Aug. 22, 2004.

[14] See, e.g., Scott Hiaasen et al., Felon Purge Sacri­ficed Inno­cent Voters, Palm Beach Post, May 27, 2001, at A1.

[15] See, e.g., Greg J. Borowski, GOP Demands IDs of 37,000 in City, Milwau­kee J. Sentinel, Oct. 30, 2004.

[16] Levitt, supra note 1; see also, e.g., Bren­nan Center for Justice, Invest­ig­at­or’s Guide to “Voter Fraud” (2006), at http://www.bren­nan­cen­­ig­at­ors_guide_to_voter_fraud/

[17] Michael P. McDon­ald & Justin Levitt, Seeing Double Voting: An Exten­sion of the Birth­day Prob­lem, Elect. L.J. (forth­com­ing 2008), avail­able at

[18] Jingle Davis, Even Death Can’t Stop Some Voters, Atlanta J.-Const., Nov. 6, 2000.

[19] Letter from William E. Moschella, U.S. Assist­ant Attor­ney General, to Chris­topher S. Bond, U.S. Senator (Oct. 7, 2005), at

[20] Darryl Fears & Jonathan Weis­man, Geor­gia Law Requir­ing Voters to Show Photo ID Is Thrown Out, Wash. Post, Sept. 20, 2006, at A6.

[21] Brief of State Respond­ents, at 2, Craw­ford v. Marion County Elec­tion Board, Nos. 07–21, 07–25  (U.S. Dec. 3, 2007).

[22] Brief of Texas et al. as Amicus Curiae Support­ing Respond­ents, at 8, Craw­ford v. Marion County Elec­tion Board, Nos. 07–21, 07–25  (U.S. Dec. 2007).

[23] Over twenty years and approx­im­ately 20 million votes cast, even if all 5,412 ballots had in fact been fraud­u­lent, the over­all rate of fraud would have been 0.027%.  Secret­ary of State Cathy Cox, The 2000 Elec­tion: A Wake-Up Call For Reform and Change 11 n.3 (2001), avail­able at­tions/2000_elec­tion_report.pdf.  In real­ity, many of these ballots are likely attrib­ut­able to cler­ical error, data error, or stat­ist­ical coin­cid­ence.

[24] Indeed, if the voter voted in several elec­tions over the twenty-year span reflec­ted in the voter history, each such mismatch would likely account for several “false posit­ive” ballots.  Thus, the 5,412 iden­ti­fied ballots might reflect far fewer matched – or mismatched – voters.

[25] Davis, supra note 18.

[26] Id. (emphasis added).

[27] Cox, supra note 23, at 11 n.3.

[28] Jingle Davis, State Plans to Update Voter Lists, Atlanta J.-Const., Feb. 10, 2001, at 4H; Cox, supra note 23, at 11 n.3.

[29] Steve Cham­raz, “The Dead List” Back­story . . . , News 4 Daily Brief­ing, May 7, 2007, at http://www.belob­­brief­ing/2007/05/the_dead_list_back­story.html

[30] John Ferro, Deceased Resid­ents on Statewide Voter List, Pough­keep­sie J., Oct. 29, 2006.

[31] Waite David, Review Turns Up No Signs of Fraud, Honolulu Advert­iser, Mar. 16, 1999, at A1.

[32] Michael Cass, Poll Worker Indicted in Vote Probe, The Tennessean, Dec. 20, 2007 (discuss­ing one incid­ent in Tennessee in 2007); Madeline Fried­man, Anatomy of Voter Fraud, Hoboken Reporter, July 1, 2007 (discuss­ing one attempt in New Jersey in 2007); Manny Garcia & Tom Dubucq, Unre­gistered Voters Cast Ballots in Dade: Dead Man’s Vote, Scores of Others Were Allowed Illeg­ally, Herald Finds, Miami Herald, Dec. 24, 2000 (discuss­ing one incid­ent in Flor­ida in 2000); Larry J. Sabato & Glenn R. Simpson, Dirty Little Secrets 292 & n.70 (1996) (citing Doug Haaland and Doug Sword­strom, A Report on Elec­tion Law Irreg­u­lar­it­ies: Cali­for­nia 16th Senate District 10 (1995)) (discuss­ing one attempt in Cali­for­nia in 1994); Frank Lynn, Boss Tweed Is Gone, But Not His Vote, N.Y. Times, Sept. 9, 1984 (discuss­ing a scheme in Brook­lyn primar­ily in the 1970s).

Of these reports, several indic­ate that elec­tion offi­cials were compli­cit in the scheme, or that the indi­vidual in ques­tion produced fraud­u­lent photo iden­ti­fic­a­tion. Cass, supra; Garcia & Dubucq, supra; Lynn, supra.

[33] Nat’l Oceanic & Atmo­spheric Admin., Nat’l Weather Service, Light­ning Safety: Medical Aspects of Light­ning, at http://www.light­ning­ (last visited Mar. 7, 2008); Ronald L. Holle, Light­ning Deaths by State, 1997 to 2006 (2007), at http://www.light­ning­–2006_Fatal­it­ies+Rates.pdf

[34] Ind. Demo­cratic Party v. Rokita, 458 F. Supp. 2d 775, 826 (S.D. Ind. 2006); see also id. at 793–94; Craw­ford v. Marion County Elec­tion Board, 472 F.3d 949, 953 (7th Cir. 2007).

[35] Brief of State Respond­ents, Craw­ford v. Marion County Elec­tion Board, Nos. 07–21, 07–25 (U.S. Dec. 3, 2007); Brief for the United States as Amicus Curiae Support­ing Respond­ents, Craw­ford v. Marion County Elec­tion Board, Nos. 07–21, 07–25 (U.S. Dec. 2007); Brief of the Amer­ican Unity Legal Defense Fund as Amicus Curiae Support­ing Affirm­ance, Craw­ford v. Marion County Elec­tion Board, Nos. 07–21, 07–25 (U.S. Dec. 10, 2007); Brief of Demo­crat and Repub­lican Elec­tion Profes­sion­als as Amici Curiae in Support of Affirm­ance, Craw­ford v. Marion County Elec­tion Board, Nos. 07–21, 07–25 (U.S. 2007); Brief of Amicus Curiae Ever­green Free­dom Found­a­tion in Support of Respond­ents, Craw­ford v. Marion County Elec­tion Board, Nos. 07–21, 07–25 (U.S. 2007); Brief for Lawyers Demo­cracy Fund as Amicus Curiae in Support of the Respond­ents, Craw­ford v. Marion County Elec­tion Board, Nos. 07–21, 07–25 (U.S. Dec. 2007); Brief of United States Senat­ors Mitch McCon­nell et al. as Amici Curiae in Support of Respond­ents, Craw­ford v. Marion County Elec­tion Board, Nos. 07–21, 07–25 (U.S. 2007); Brief of the Repub­lican National Commit­tee as Amicus Curiae Support­ing Respond­ents, Craw­ford v. Marion County Elec­tion Board, Nos. 07–21, 07–25 (U.S. Dec. 10, 2007); Brief of Texas et al. as Amicus Curiae Support­ing Respond­ents, Craw­ford v. Marion County Elec­tion Board, Nos. 07–21, 07–25 (U.S. Dec. 2007); Brief of Wash­ing­ton Legal Found­a­tion, as Amicus Curiae in Support of Respond­ents, Craw­ford v. Marion County Elec­tion Board, Nos. 07–21, 07–25 (U.S. Dec. 10, 2007).

[36] Justin Levitt, Analysis of Alleged Fraud in Briefs Support­ing Craw­ford Respond­ents (2007), at http://www.truth­about­­fordAl­leg­a­tions.pdf.

[37] Id. at 2.

[38] Cass, supra note 32(discuss­ing one incid­ent in Tennessee in 2007); Fried­man, supra note 32(discuss­ing one attempt in New Jersey in 2007); Garcia & Dubucq, supra note 32(discuss­ing one incid­ent in Flor­ida in 2000); Sabato & Simpson, supra note 32, at 292 & n.70 (discuss­ing one attempt in Cali­for­nia in 1994).

[39] See Levitt, supra note 36, at 2.

[40] Greg J. Borowski, Inquiry Finds Evid­ence of Fraud in Elec­tion, Milwau­kee J.-Sentinel, May 11, 2005, at A1.

[41] See Special Invest­ig­a­tions Unit, Milwau­kee Police Dept., Report of the Invest­ig­a­tion into the Novem­ber 2, 2004 General Elec­tion in the City of Milwau­kee (2008) (“Milwau­kee P.D. Report”), avail­able at http://graph­ics.json­­ics/news/MPD_2004­voter­fraud­probe_22608.pdf; Greg J. Borowski, Tighter Voting Laws Urged, Milwau­kee J.-Sentinel, Feb. 26, 2008; Daniel Bice, Report’s Stealthy Release Irks Police Chief, Milwau­kee J.-Sentinel, Mar. 1, 2008.

[42] Milwau­kee P.D. Report, supra note 41, at 7–17, 23, 32, 35–36, 41–46, 57–58, 60.

[43] Id. at 7–8, 31, 33–36, 56, 60.

[44] Id. at 15–17, 19, 24, 27, 31–32, 40, 41, 46–47, 49–51.  Some of the invest­ig­at­ors’ analysis in this respect, partic­u­larly with respect to the legit­im­ate addresses of voters, involves legal conclu­sions that are likely beyond the authors’ offi­cial respons­ib­il­it­ies.

[45] Id. at 61.

[46] See, e.g., Craw­ford v. Marion County Elec­tion Board, 472 F.3d 949, 953 (7th Cir. 2007).

[47] It is no answer that the indi­vidual may have submit­ted a fraud­u­lent regis­tra­tion form in a ficti­tious name, presum­ably outside of the pres­ence of an elec­tion offi­cial, before arriv­ing in person to vote in that ficti­tious name.  Federal law already contem­plates this hypo­thet­ical and unlikely possib­il­ity, by provid­ing that any regis­trant new to the juris­dic­tion who submits a regis­tra­tion form by mail must at some point, and through a broad range of means, offer reli­able proof of his iden­tity before voting.  42 U.S.C. § 15483(b).

[48] See Dep’t of Justice, Fact Sheet: Depart­ment of Justice Ballot Access and Voting Integ­rity Initi­at­ive, July 26, 2006, at; Eric Lipton & Ian Urbina, In 5-Year Effort, Scant Evid­ence of Voter Fraud, N.Y. Times, Apr. 12, 2007.

[49] See, e.g., Repub­lican National Commit­tee, You Can’t Make This Up!, at (last visited Mar. 7, 2008).

[50] See, e.g., UFO Case­book, Break­ing UFO News Reports, at http://www.ufocase­

[51] 42 U.S.C. § 1973i(c).

[52] Milwau­kee P.D. Report, supra note 41, at 26.

[53] See, e.g., Press Release: Senator Leibham to Again Ask for Vote on Photo ID Amend­ment (Mar. 5, 2008), at http://www.thewheel­erre­­oid.pdf (last visited Mar. 7, 2008).

[54] See Bren­nan Center for Justice, The Link Between Voter Fraud and Restrict­ive ID, at http://www.truth­about­­ary/the_link_between_voter_fraud_a.html (collect­ing comment­ary and edit­or­i­als on both sides of the issue).

[55] This is currently the law (with limited excep­tions for indi­gency and reli­gious objec­tion) in only two states: Geor­gia and Indi­ana.  Ga. Code § 21–2–417; Ind. Code §§ 3–5–2–40.5, 3–10–1–7.2, 3–11–8–25, 3–11–8–25.1, 3–11.7–5–2.5.  Flor­ida requires voters to produce a photo iden­ti­fic­a­tion card at the polls, but it need not be govern­ment-issued, and voters without such a card may vote a ballot to be coun­ted if the signa­ture on the ballot matches the signa­ture on the regis­tra­tion rolls.  Fl. Stat. §§ 101.043, 101.048.  Every other state offers altern­at­ive, and often more flex­ible, means for a voter to confirm her iden­tity.  See Bren­nan Center Brief, supra note 2, at 33–38.

[56] See infra notes 63–67.

[57] See, e.g., Brief for The Lawyers’ Commit­tee for Civil Rights Under Law et al. as Amici Curiae in Support of Peti­tion­ers at 17–21, Craw­ford v. Marion County Elec­tion Board, Nos. 07–21, 07–25 (U.S. Nov. 13, 2007).

[58] Dept. of Home­land Secur­ity, U.S. Citizen­ship & Immig­ra­tion Servs., Instruc­tions for N-565, Applic­a­tion for Replace­ment Natur­al­iz­a­tion/Citizen­ship Docu­ment (2007), at­str.pdf.

[59] See supra note 55(describ­ing the state laws).

[60] Brief for Respond­ent Marion County Elec­tion Board at 8–9, Craw­ford v. Marion County Elec­tion Board, Nos. 07–21, 07–25 (U.S. Dec. 3, 2007).

[61] Id. at 9–10.

[62] Shan­non McCaf­frey, More Than 400 Voters Lacked Photo IDs in Feb. 5 Primary, The Ledger-Enquirer (Colum­bus, Ga.), Feb. 14, 2008.

[63] See, e.g., Matt A. Barreto et al., The Dispro­por­tion­ate Impact of Indi­ana Voter ID Require­ments on the Elect­or­ate (2007) (“Barreto Indi­ana Study”), avail­able at http://depts.wash­ing­­ments/Indi­ana_voter.pdf; Matt A. Barreto et al., Voter ID Require­ments and the Disen­fran­chise­ment of Latino, Black and Asian Voters (2007), avail­able at http://www.bren­nan­cen­­load_file_50884.pdf; Dr. Robert Pastor et al., Voter IDs Are Not the Prob­lem: A Survey of Three States (2008), avail­able at http://www.amer­­ID­Fi­nalRe­port1–9–08.pdf.  

Another set of empir­ical research seeks to determ­ine the impact of restrict­ive iden­ti­fic­a­tion laws on voter turnout by analyz­ing past voting patterns.  See gener­ally Brief of R. Michael Alvarez et al. as Amici Curiae in Support of Peti­tion­ers, at 10–14, Craw­ford v. Marion County Elec­tion Board, Nos. 07–21, 07–25 (U.S. Nov. 13, 2007) (“Polit­ical Scient­ists’ Brief”); see also Jeffrey Milyo, The Effects of Photo­graphic Inden­ti­fic­a­tion [sic] on Voter Turnout in Indi­ana: A County-Level Analysis (Inst. of Pub. Pol’y, Univ. of Mo., Report 10–2007, 2007).  These stud­ies vary in their meth­od­o­logy; there is substan­tial dispute about the valid­ity of each approach.  In general, each of these turnout stud­ies has neces­sar­ily been constrained by the histor­ical record; because the most restrict­ive voter ID laws are new, the avail­able pool of pertin­ent voting data is relat­ively limited.

[64] See Polit­ical Scient­ists’ Brief, supra note 63, at 7–9.  The research led by Dr. Pastor found a dispro­por­tion­ate effect on minor­it­ies but results appear to have been less conclus­ive for elderly voters.  See Pastor, supra note 63, at 9, 18–19, 23.

[65] See Pastor, supra note 63, at 5–7.

[66] See Pastor, supra note 63, at 8 (conclud­ing that approx­im­ately 1.2% of registered voters in the selec­ted states had no govern­ment-issued photo iden­ti­fic­a­tion); U.S. Elec­tion Assist­ance Comm’n, The Impact of the National Voter Regis­tra­tion Act of 1993 on the Admin­is­tra­tion of Elec­tions for Federal Office 2005–2006, at 8 (2007), avail­able at­ing­house/docs/the-impact-of-the-national-voter-regis­tra­tion-act-on-federal-elec­tions-2005–2006/attach­ment_down­load/file (report­ing at least 172,810,006 registered voters as of the 2006 general elec­tion).

[67] See Bren­nan Center for Justice, Citizens Without Proof: A Survey of Amer­ic­ans’ Posses­sion of Docu­ment­ary Proof of Citizen­ship and Photo Iden­ti­fic­a­tion 3 (2006), avail­able at http://www.bren­nan­cen­­load_file_39242.pdf  (find­ing that approx­im­ately 11% of voting-age Amer­ican citizens do not have current govern­ment-issued photo iden­ti­fic­a­tion); id. (find­ing a dispro­por­tion­ate effect on low-income, minor­ity, and elderly citizens); Barreto Indi­ana Study, supra note 63, at 19 (find­ing that approx­im­ately 16% of voting-age Hoosier citizens do not have current govern­ment-issued photo iden­ti­fic­a­tion).