National Civic Summit - Lorraine C. Minnite
Published on: Mar 3, 2016
Transcripts - National Civic Summit - Lorraine C. Minnite
An Analysis of
Voter Fraud in
The United States
By LorrAine C. MinniTe,
Senior FeLLow, DeMoS
(Adapted from the 2003 report Securing The Vote,
by L. Minnite and D. Callahan, with updates.)
Dēmos: A Network for Ideas & Action
I. Voter Fraud in the United States: An Overview 1
What Is Fraud and Why Does It Matter? 2
Research Methodology 4
A Framework for Understanding Fraud 5
Election Fraud Today 6
II. Voter Fraud and the Law 7
State Laws and Enforcement 7
Federal Laws and Enforcement 8
III. Major Recent Cases of Alleged Election Fraud 9
The 1997 Mayoral Primary Election, Miami, Florida 9
The 1996 Dornan/Sanchez Contest, Orange County, California 11
The 2000 Election, St. Louis, Missouri 12
I. VoteR FRAUD In tHe UnIteD stAtes: An oVeRVIeW
Since the 2000 election, a historic effort has been underway in the United States to strengthen voting
systems across all 50 states and to address obstacles to broader electoral participation. At both the federal
and state levels, however, efforts to advance a reform agenda have been frequently complicated by heated
debates over the integrity of voting systems—and by allegations of widespread election fraud, and its co-
hort, voter fraud.
In Congress, and elsewhere in the national arena, fear of fraud in elections has been revived in many
policy discussions about our voting system. In 2002, for example, allegations of widespread voter fraud
fueled disagreement over voter identification provisions in federal election reform legislation, resulting
in an acrimonious legislative process that delayed passage of the Help America Vote Act of 2002 (HAVA).
Oddly enough, fraud and election integrity have been implicated in debates over terrorism and security.
The September 11, 2001, terrorist attacks spurred a major legislative initiative, the Real ID Act of 2005,
which requires states to meet costly minimum security standards for drivers’ licenses that many fear will
move the country toward a system of national ID—with negative consequences for electoral participation.1
Voter ID debates currently raging across the states join the disparate issues of voter fraud and election
integrity, security against terrorist attacks, and identification requirements. Proponents argue that more
stringent ID requirements like those embodied in a “Real ID” driver’s license are needed to protect against
voter fraud at the polls, and opponents argue that voter ID is a solution in search of a problem.2
Consequently, the specter of fraud has had a major role in the delay or defeat of some election reforms
enacted at the state level, and has drawn federal agencies into political controversy. In 2002, ballot initia-
tives in California and Colorado that would have enacted Election Day Registration (EDR) into law were
defeated, with opponents of the initiatives arguing that EDR would increase the potential for fraud.3 Since
then, fraud allegations have fueled stringent and contested voter ID proposals in states like Indiana, Geor-
gia and Missouri. In Arizona, where allegations of illegal non-citizen voting have flourished, voters passed
a ballot initiative to make voter registration more difficult by requiring proof of citizenship.
Moreover, controversy over research on voter ID, voter fraud and voter intimidation conducted for the
Election Assistance Commission, a federal agency created by HAVA, which suggested little evidence of
voter fraud and a potentially negative impact of restrictive ID rules on voter turnout, has undermined the
agency’s credibility, further politicizing the issues.4 Finally, as has been uncovered this year, unfounded
fraud allegations surrounding the 2004 and 2006 elections are at the center of both a new Justice Depart-
ment program to increase federal fraud prosecutions, and the dismissal of several U.S. Attorneys who
allegedly failed to prosecute voter fraud, embroiling the Bush Administration in scandal.
Debates over election fraud are not new. They have been a staple of discussions about elections and
democracy in the United States for more than a century. But in recent years, issues of fraud and voting
integrity have increasingly come to the forefront of public policy discussions over the health of America’s
democracy. Even before the 2000 election, consistently low voter turnout rates and obstacles to participa-
tion motivated various efforts to increase voter registration and turnout—efforts that in turn raised ques-
tions about voting integrity. Critics of some reforms—such as the institution of mail-in voting in Oregon,
the loosening of guidelines for absentee ballot use, and, most notably, the National Voter Registration Act
of 1993 (also known as NVRA or the “motor voter” law)—have charged that these reforms increase the po-
tential for voter fraud. Other frequently proposed reforms, such as Election Day Registration, continue to
be criticized on the same grounds.
An Analysis of Voter Fraud in The United States
As federal and state officials consider future reform efforts, as well as the merits of existing reforms, there
is an acute need for better information and analysis about election and voter fraud issues. While the issue
of fraud is raised continually in discussions of election reform, to date there have been few major studies of
voter fraud in the United States.5 Too often in this area, hearsay and anecdotal information are put forth
as fact in important public policy debates.6 Many key questions about fraud remain unanswered, includ-
ing: How often does voter fraud occur? How serious a problem is fraud compared to other problems with
the election process, such as those that occurred in Florida in the 2000 election or in Ohio in 2004? What
kinds of voting methods are most vulnerable to corruption? What administrative, technological and legal
steps can be taken to reduce the chances of voter fraud while also expanding opportunities to register and
What Is Fraud and Why does It Matter?
Elections are the mechanisms by which people choose their representatives. Given that the integrity of
this process is central to democracy, there can be no compromise on the need for fair elections determined
without the taint of fraud—whether on the part of voters, political parties, election administrators or others.
Generally speaking, election fraud is a crime involving conduct that corrupts the process of “obtaining and
marking ballots, the counting and certification of election results, or the registration of voters.”7 Elec-
tion fraud may involve wrongdoing by either individual voters or, as is more often the case, by organized
groups such as campaigns or political parties.
Beyond these traditional conceptions of fraud, many people are concerned about official efforts to corrupt
the election process or erect barriers to participation. For example, election officials could manipulate
registration databases to remove the names of people likely to vote in a certain way so that these people
are unable to cast regular ballots when they arrive at polling places. Corruption of this kind was widely
alleged to have taken place in Florida and other states during the 2000 election.9
Deliberate disfranchisement of voters may also occur because of other kinds of official misconduct: turn-
ing away voters already in line when polls close; intimidating or misinforming voters when they arrive at
the polls; producing misleading or poorly designed ballots; failing to provide bilingual voting materials,
as required by law; failing to upgrade or repair voting systems in specific election districts; and by other
Overall, the disfranchisement of voters through antiquated voting systems, errors, mismanagement of
registration bases, and intimidation or harassment is a far bigger problem today than traditional forms of
election fraud. The problems in Florida in 2000, which determined the outcome of a presidential election,
are dramatic evidence of this point.10 No number of allegations of voter fraud in 2000, let alone substanti-
ated cases of such fraud in that election, came close to rivaling the number of eligible voters who wanted to
vote but were unable to do so. This fact is due to a combination of deliberate actions on the part of partisan
election officials and gross administrative error or incompetence (and that’s in just one state).
These problems have been analyzed and highlighted in a number of studies and reports over the last
several years.11 Civil rights advocates have been particularly active in challenging official forms of election
malfeasance as violating various provisions of the Voting Rights Act. This report, however, does not focus
on these issues. Rather, it looks exclusively at voter fraud as the problem has commonly been discussed
over the past century.
Dēmos: A Network for Ideas Action
Organized Fraud. Fraud is easier to commit for Individual Fraud. Unlike most other Western
organized groups than it is for individual voters democracies, voting in the United States is a two-
because such groups have direct access to election stage process. In nearly all states, an eligible citizen
machinery and the resources necessary to com- who wants to vote must first register using his or
mit the significant amount of fraud it takes to flip her permanent home address. After successfully
election results. Average voters usually lack these completing a voter registration application, the
things. In all but the most extraordinary of cas- voter goes to the polls or receives voting materials
es—for instance, when an election victory depends through the mail, and then casts his or her ballot.
on a handful of votes—fraud must be committed
through a conspiracy to have an impact on the out- Individual vote fraud would happen, conceivably, in
come of an election. a couple of ways:
Such fraud can take several forms. For example: • Voters may violate laws governing the registra-
tion process by misrepresenting themselves
• Political parties, campaign organizations or as eligible when they are not, or submitting
other groups can perpetrate organized fraud registration applications for fictitious people,
by obtaining and voting fraudulent absentee or dead people, or real people who may be either
mail-in ballots. eligible or ineligible to vote and who may or
may not know of or consent to the fraud.
• Interested groups can organize large-scale
“commercialization of the vote” or vote buy- • Voters may commit fraud at the point of vot-
ing—for example, by providing incentives for ing. A voter may vote multiple times using the
otherwise uninterested voters to go to the polls name or names of another voter. In the case of
and vote in a certain way or coordinate efforts a vote cast using the name of a real person, that
to help enough voters vote more than once. person may or may not be eligible to vote and
may or may not consent to the fraud. Voters
• Local election administrators or poll work- consenting to the appropriation of their vote
ers can commit clear-cut fraud through vote by another may do so because they do not plan
rigging, by not counting or destroying ballots, to vote, have little interest in voting, or receive
allowing votes that should have been barred, some kind of material benefit.
and tampering with ballots.
This last example is one of the reasons why partisan
control of election administration has historically
created the greatest potential for election fraud. In
fact, the most comprehensive studies of election
fraud conducted when local political machines were
strong and controlled the machinery of election ad-
ministration in the early decades of the 20th century
found that, “most election frauds of the period were
committed by the precinct officers themselves.”8
An Analysis of Voter Fraud in The United States
Methodology For deMos’ 2003 report, Securing the Vote
The administration of elections for all public offices in the United States, from county dogcatcher to the
President of the United States, is controlled by state and local election officials. This causes election pro-
cedures to vary from state to state and in many places, from county to county. Given the complexity, it is
difficult to assess the overall integrity of U.S. election systems. No other aspect of American politics has re-
ceived as much scrutiny over the last 50 years as the behavior of the American electorate, and yet, election
fraud has received very little attention by scholars. Remarkably, there are no definitive academic studies of
election fraud in the contemporary period, nor are there focused studies of fraud by government agencies
concerned with the administration of elections in this country.12
The difficulty of gathering data on fraud explains much of this vacuum in analysis. Like many of the
rules governing American elections, the rules dealing with election fraud and the state and local agencies
assigned the responsibility of handling fraud claims vary from state to state. In most states the secretary
of state is the chief elections officer, and his or her office has the primary responsibility for maintaining
statewide election records. In some states, complaints of election fraud are first received and investigated
by the state attorney general. In a few states, neither the secretary of state nor the attorney general main-
tains voting and elections records or handles any matters related to fraud at all. Instead, those responsibili-
ties are assigned to a state board of elections or other elections agency. Since so few fraud claims evidence
criminal intent, state law enforcement agencies are only occasionally involved in prosecuting cases. In
general, the formal responsibility for receiving and investigating complaints of election fraud is allocated
to local or county boards of elections or district attorneys, with little to no responsibility or accountability
vested in any state agency.
While the analysis of this report is limited by the lack of comprehensive and accessible statistical data on
election fraud, the authors of Securing the Vote were able to develop an in-depth analysis of voter fraud in
the United States today by drawing on a wide range of sources.
• First, we conducted an analysis of the incidence of voter fraud from 1992 to 2002 in 12 states that
collectively represent about half of the electorate and are drawn from all of the major regions of the
country. These states include: Alabama, California, Florida, Georgia, Illinois, Minnesota, Missis-
sippi, New York, Oregon, Pennsylvania, Texas and Wisconsin. For each of these states, we conducted
Lexis-Nexis searches of news databases, as well as the statutory and case law for evidence of a record of
prosecution of voter fraud. We also contacted selected state officials, including attorneys general and
secretaries of state.
• Second, we conducted a complete Nexis search on voter fraud throughout the United States from 2000
to 2002, supplemented by searches related to several high-profile cases of election fraud that occurred
before 2000. The search produced thousands of articles and references, each of which was thoroughly
• Third, we surveyed the academic literature, a wide variety of government documents, congressional
testimony and research reports, law journal articles, and other sources on election reform from profes-
sional, research and reform organizations.
• Finally, we analyzed in considerable depth some of the highest-profile cases of real or alleged fraud in
the United States over the past decade, including notable cases in Missouri, California and Florida.
Dēmos: A Network for Ideas Action
a FraMeWork For understandIng Fraud
While heated debates over election fraud have arisen episodically for more than a century, the circum-
stances that surround voting and elections have changed dramatically over that time and continue to
evolve rapidly today. Elections can be as contested as ever, but the conditions conducive to election fraud
have steadily declined, and the trend is likely to continue in the foreseeable future. Three factors account
for it: the declining strength of local political parties and machines; strengthened election administration;
and improved voting technology. While some voter fraud may occur with any electoral system, current
trends suggest that it is more possible than ever to further open up the electoral process and facilitate vot-
ing without bringing about greater fraud. Exaggerated fears of fraud should not stand as an obstacle to
reforms aimed at expanding participation.
Declining Local Political Machines. Historically, local political party organizations played an important
role in perpetrating election fraud. During the late 19th century and well into the 20th century, a key motive
for fraud was the immense local patronage benefits afforded to winning parties. Under these conditions,
parties, patronage and fraud were intertwined. Election fraud was perpetrated by partisans acting together
to steal elections. Local party organizations competed for voters and controlled votes through patronage.
When elections were fully controlled by local party organizations, ballots were easily destroyed, miscount-
ed, or falsely multiplied, and voters could be strongly influenced by bosses or local elites to vote in specific
ways. Typically, cases of election fraud involved organized efforts by partisan election officials, party lead-
ers and politicians rather than by the voters themselves.13 Today, local party organizations are relatively
weak to nonexistent, in part because their access to patronage has all but disappeared. They no longer
control lucrative franchises, run police and fire departments, set utility rates, or build large-scale public
works. However, in many states key election officials are openly partisan and may also play an active role
in partisan political campaigns, a conflict of interest that increases the potential for organized fraud.
Strengthened Election Administration. At the same time that local political party organizations have
weakened, modern election administration has become more sophisticated. The reforms put in place in
the late 19th century and early 20th century required voters to register in advance of elections and election
authorities to keep registration records. While some reforms such as voter registration and the secret bal-
lot reduced opportunities for organized fraud, they also had a negative impact on electoral participation,
making voting especially more difficult for poor and working-class people.14 The passage of the National
Voter Registration Act of 1993 as well as the advent of Election Day Registration in a number of states have
helped to reduce the obstacles to voting that accompanied the imposition of voter registration require-
ments over a century ago. In the wake of the 2000 election, considerable attention has been focused on
ways to improve election administration to strengthen the integrity of the election process; a number of
reform measures have already been passed in the states. Particular attention has been given to the need
to create accurate statewide computerized registration systems in all states, as now required by the Help
America Vote Act of 2002. To the extent that further reforms upgrade accuracy, record-keeping and ac-
countability, professionalize service delivery, and simplify the process for voters, election administration
can continue to be strengthened as a bulwark against fraud.
Improved Voting Technology. Mechanization and the application of vote counting technology in the
1960s loosened the control local party workers could exercise over vote counting and significantly reduced
opportunities for organized election fraud. In fact, in his research on election administration in the 1920s,
the early election reformer Joseph Harris came to believe that most electoral corruption was committed
by precinct workers during the vote count. Then, like now in many places around the country, the polls
were staffed on Election Day by people chosen by the major political parties which dominated many local
An Analysis of Voter Fraud in The United States
boards of elections. Then, unlike now, when the voting was finished, poll workers opened up the ballot
boxes and counted the paper ballots at the polling site in back rooms out of the glare of public scrutiny
or official supervision. Harris believed that one of the best ways to clean up American elections was to
take the vote counting responsibility away from the control of party machines and their poll workers. In
the early 1960s, his thinking about the sources of election fraud led him to develop the concept for what
became the punch card ballot. Harris reasoned that if poll workers were corrupting the vote count there
were at least two ways to address the problem: either get rid of partisan poll workers, or take away their
responsibility for counting the votes. He eventually sold his concept for machine-readable ballots to IBM
which designed the punch card technology to do just that. Vote counting was physically moved out of the
polling site, out of the hands of thousands of unskilled poll workers, and into the high tech world of com-
puters programmed and operated by a smaller cadre of better trained, better educated, supposedly politi-
cally neutral bureaucrats.15 Today, we don’t worry so much about poll workers corrupting the vote count
because poll workers don’t count the votes.
On the other hand, technology itself provides no simple guarantee or solution to fraud, as the current
controversy over the reliability and security of electronic voting equipment makes clear. An ideal voting
system should accurately record and tally the votes of all eligible voters who desire to vote and who cast
legal votes, and should provide evidence that the election results reflect their will. Technology can help
us achieve these goals if it is part of a larger effort to secure voting rights and a fair count. Since the 2000
election, states have moved to implement major technology upgrades that help prevent voter fraud, com-
puterizing and centralizing their voter files, for example, though much more work in this area remains to
electIon Fraud today
Based on the research and analysis conducted for Securing the Vote, we offer several conclusions about
election fraud in the United States today:
• Voter fraud appears to be very rare in the 12 states examined in that report. Legal and news records
turned up little evidence of significant fraud in these states or any indication that fraud is more than a
minor problem. Interviews with state officials further confirmed this impression.
• Notable election reforms of the past 10 to 15 years—such as the NVRA, more permissive absentee
balloting rules, all mail-in voting in Oregon, and the enactment of Election Day Registration in several
more states—have not facilitated voter fraud.
• Analysis of several cases of election fraud that have received significant attention in recent years sug-
gests that some of the most notable allegations of fraud have proved to be baseless. While the 1997
mayoral primary election in Miami, Florida, was one of the most egregious election fraud cases in re-
cent memory, there are other noted cases where charges of significant vote fraud have been disproved,
such as the 1996 Dornan/Sanchez contest for the U.S. House of Representatives in Orange County,
California. There are yet other cases, such as the 2000 election in St. Louis, Missouri, in which politi-
cians have made great hay, but charges of widespread fraud have not been substantiated. A new Dem-
os report on voter fraud in states offering Election Day Registration finds that despite the hundreds
of news stories reporting on allegations of voter fraud in Wisconsin in the 2004 presidential election,
practically no fraud has ever been proven. An intensive effort on the part of the federal government to
uncover and prosecute voter fraud in Wisconsin resulted in only 14 indictments and five convictions
or guilty pleas for illegal voting in an election in which over 3 million ballots were cast.16
Dēmos: A Network for Ideas Action
The low level of voter fraud in the United States today does not preclude the need for continued vigilance
to ensure the integrity of election systems. It does, however, suggest that reforms aimed at simplifying
registration and voting can be implemented without risking corruption of elections by fraud. Even if only
partly implemented, some of the many technological and administrative reforms recommended by nation-
al and state commissions since the 2000 election, as well as other “best practices,” can go a long way toward
enhancing election integrity.17 These are reforms such as continued upgrading of the accuracy of voter
registration lists, once-in-a-lifetime registration, better notice and opportunity to voters to correct their
records, improved use of communications technology to better inform voters and resolve administrative
problems on Election Day, automatic notification to election officials upon the restoration of voting rights
of convicted felons, improved efficiency in the transmittal to elections officials of voter registration forms
collected at NVRA-designated state agencies, and increased recruitment, pay and training of poll workers.
These same reforms can facilitate programs, such as Election Day Registration, that are intended to make
voting easier and more convenient for eligible voters.
II. VoteR FRAUD AnD tHe LAW
The opportunity to commit voter fraud is constrained by a matrix of state and federal laws. Voter fraud is
a serious crime that can be prosecuted at the federal and state levels, where penalties carry fines, lengthy
prison terms, and, in the case of illegal voting by non-citizens, deportation.18 However, the effectiveness of
laws depends on their enforcement and implementation.
state laWs and enForceMent
The U.S. Constitution grants states broad jurisdiction over the election process, though the authority of the
states in these matters is not absolute.19 The Fifteenth, Nineteenth and Twenty-sixth Amendments prohibit
states from restricting the franchise based on race or color, gender or minimum age (18 years) of the voter,
respectively. The Supreme Court has found that Congress is within its constitutional authority to pass
laws governing the timing of federal elections, voter registration, access to the ballot for the elderly and
disabled, and, perhaps most important, in the area of prohibitions against racially discriminatory voting
practices.20 However, states are granted wide powers to qualify voters and establish rules for conducting
federal, state and local elections.
Within this complex framework for regulating the electoral process, states have exhibited a full flowering
of differences in the manner in which they administer elections. State election laws governing voting vary
in their level of specificity, with many states granting localities considerable discretion in the way they run
elections. For example, Oklahoma has standard Election Day procedures and a single voter registration
and election management system, and it uses only one type of voting machine. In contrast, before the
enactment of HAVA, Pennsylvania’s election law provided few statewide guidelines and near-autonomy
to the state’s 67 counties in the matter of Election Day procedures. Pennsylvania had 67 different election
systems using a variety of voting machines.21
Federalism, and the authority over election procedures granted to states, also explains why laws criminal-
izing fraud differ across the states. All states have laws governing election crime. However, because of
the historically parochial manner in which states administer elections, they vary in how they handle the
problem of criminal election fraud. All states prohibit voting by non-citizens (although some localities
permit such voting in local elections); most states have various restrictions that bar voting by individuals
convicted of felonies,22 as well those who have been ruled mentally incompetent by a court. Most states
An Analysis of Voter Fraud in The United States
have prohibitions against falsifying voter registration information, voting more than once in an election,
impersonating another voter, intimidating or coercing voters, and bribing voters or buying votes. Most of
these crimes are classified as felonies and carry fines and prison sentences. In some states, a person con-
victed of certain voter fraud crimes can permanently lose his or her right to vote.
State election laws allocate the responsibility for ensuring fair elections to various agencies and officials,
and it is their responsibility to administer and monitor the electoral process to ensure that it is free of cor-
ruption. Local election and law enforcement officials also play a role in enforcing election laws, although
the familiarity of these officials with the ins and outs of election laws and the lines of enforcement author-
ity varies considerably within states. While it is incumbent upon government officials to bring criminal
charges where appropriate, all states also empower private citizens and organizations to bring civil suits to
contest election results.23 Likewise, the NVRA provides a private right of action to any person aggrieved
by a violation of the Act.24
Federal laWs and enForceMent
Despite state jurisdiction over election administration, there is a role for the federal government in pros-
ecuting voter fraud when federal interests are at stake.
Historically, the federal role has extended to ensuring elections that are free of corruption and in eliminat-
ing discrimination against minority voters protected by the Voting Rights Act of 1965, as amended. The
enforcement of the Voting Rights Act concerns civil offenses and is handled by the Justice Department’s
Civil Rights Division. Election-related crimes are handled by the Public Integrity Section of the Justice
Department’s Criminal Division.
Federal election law is an amalgamation of statutes. Some of them expressly apply to elections and vot-
ing, and others, such as statutes prohibiting mail fraud, have been used to prevent and punish voter fraud.
Most federal statutes apply only to federal or mixed federal/state and local elections. In order for election
crime to rise to the level of federal prosecution, there must be some “substantive irregularity relating to the
voting act ... which has the potential to taint the election itself.”25 The Supreme Court has found a constitu-
tionally guaranteed right to vote and Congress has passed legislation to protect this most fundamental of
all rights. There remains debate, however, over whether or not the Constitution guarantees a right to vote
in purely state and local contests—here the judicial record is inconsistent. Federal prosecutors, therefore,
avoid investigating fraud allegedly committed in these elections.
Federal election law can be divided into two categories: anti-intimidation laws and anti-trafficking laws.
Anti-intimidation laws make it a felony to conspire to “injure, oppress, threaten, or intimidate any inhab-
itant of any State, Territory, or District in the free exercise or enjoyment of any right or privilege secured
to him by the Constitution or laws of the United States or because of his having exercised the same.”26
They also provide for criminal punishment of anyone who deprives another of federally secured rights to
vote. Anti-trafficking laws, on the other hand, restrict a citizen’s right to vote by prohibiting the offering,
making, soliciting or receiving of payments in return for voting or withholding a vote. Penalties include
a fine of up to $10,000 and five years imprisonment. Until a policy change instituted in 2002 by the Bush
Administration’s first Attorney General, John Ashcroft, the Justice Department, as a matter of practice,
did not prosecute voters whose only involvement in voter fraud was in compromising their votes. Nor did
it advocate the prosecution of isolated instances of illegal registration, vote buying, or fraudulent voting
because “isolated incidents do not implicate federal interests sufficiently” to warrant federal interference in
what is traditionally a state function.27 It appears that the majority of vote buying schemes that are pros-
ecuted involve small amounts of money and occur in low-income neighborhoods.28
Dēmos: A Network for Ideas Action
On October 1, 2002, U.S. Attorney General John Ashcroft announced the Ballot Access and Voting Integ-
rity Initiative, aimed at enhancing the Department of Justice’s “ability to deter discrimination and election
fraud, and ... to prosecute violators vigorously whenever and wherever these offenses occur.”29 The initia-
tive involved the creation of task forces of district election officers who are assistant U.S. attorneys ap-
pointed by each of the U.S. attorneys to serve in this new capacity for two-year intervals; and FBI officials,
whose job it is to participate in “on-the-ground investigative and prosecutorial coordination” with state
and local elections and law enforcement personnel to “deter and detect discrimination, prevent electoral
corruption, and bring violators to justice.”30
Federal monitoring of elections has been around since the Reconstruction period, but most often it has
been directed toward protecting the voting rights of minority groups at the polls. What is significant about
the Justice Department’s involvement in recent elections is the linking of voting rights with protection
from corruption of the electoral process by voter fraud, reflecting a new view that voter fraud deserves the
same level of scrutiny from federal law enforcement officials as was historically required to guard against
racial discrimination in voting. And yet, the Justice Department’s new project has still to yield any evi-
dence that contradicts the basic findings of this report on the low incidence of voter fraud in contempo-
rary U.S. elections. Despite the elevated law enforcement priority, the federal government indicted only
40 voters for election crimes related to illegal voting between 2002 and 2005, obtaining 26 convictions or
guilty pleas, for a nationwide average of eight to nine illegal voters a year. These were all isolated instances,
in some cases, of confusion or voter error, having no impact on electoral outcomes. In Milwaukee where
a third of the indictments were handed down, an assistant U.S. attorney who prosecuted a number of the
cases was forced to conclude that, “There was nothing that we uncovered that suggested some sort of con-
certed effort to tilt the  election.”31
III. MAJoR ReCent CAses oF ALLeGeD eLeCtIon FRAUD
The cases described below illustrate the following conclusions drawn from the analysis presented in this
report: 1) the potential for abuse of absentee ballot laws remains a concern; 2) for fraud to threaten elector-
al integrity, it must be organized; 3) organized fraud that does affect electoral outcomes is not inordinately
difficult to detect and punish because it largely occurs in public, involves many players, and leaves a paper
trail;32 4) mistakes by voters and election officials happen, perhaps more regularly than we know, but they
are not fraud; nevertheless, under certain conditions partisans will exploit mistakes, errors and incompe-
tence in the administration of elections to create the appearance of fraud and gain competitive electoral
the 1997 prIMary Mayoral electIon, MIaMI, FlorIda
Perhaps the best-known contemporary case of uncontroverted absentee ballot fraud is the disputed 1997
primary mayoral election in Miami, Florida.33 Running for reelection as mayor, Joe Carollo received 51.4
percent of the ballots cast at the polls, while his opponent, former Mayor Xavier Suarez, received 61.5 per-
cent of the absentee ballots, giving Suarez a slim lead of 155 votes over Carollo in total balloting. Because
neither candidate received more than 50 percent of the vote, a run-off election was held, and Suarez nar-
rowly won both the precinct and absentee ballots.34
Carollo immediately challenged the initial November 4th election results, claiming fraud in the absentee
ballot vote that had swung that election to Suarez, denying Carollo the majority support he received at the
polls and forcing him into a run-off. A week after the first election, the Florida Department of Law En-
An Analysis of Voter Fraud in The United States
forcement had arrested two Suarez supporters for buying absentee ballots and falsely witnessing absentee
ballots. Once Carollo lost the run-off to Suarez, he petitioned the Circuit Court for the Eleventh Judicial
Circuit of Florida to overturn the results of the November 4th primary on grounds of fraud.
The trial was held in February 1998. For two and a half weeks, the court heard evidence and read de-
positions from 87 witnesses and examined 195 exhibits.35 Its March 3rd decision noted “a pattern of
fraudulent, intentional and criminal conduct” in the extensive abuse of absentee ballot laws.36 An expert
documents examiner testified that 225 absentee ballots cast had forged signatures; there was evidence of
14 stolen ballots and 140 improperly witnessed ballots. Another 480 ballots were procured or witnessed
by 29 “ballot brokers,” 27 of whom invoked their Fifth Amendment privilege against self-incrimination
instead of testifying at trial. One such ballot broker was 92-year-old Alberto Russi, a campaign volunteer
for Humberto Hernandez, a Suarez ally on the five-member City Commission. Within days of the primary
election, Russi was arrested and charged with three counts of election fraud. Police traced Russi to the
absentee ballot of a dead man whose ballot he witnessed. When police searched Russi’s home they seized
75 absentee ballots already filled out and intended for the November 13 run-off, many of which were ad-
dressed to Russi’s home in the names of other voters. A separate grand jury, convened to investigate the
fraud allegations and make recommendations for improvements in the absentee ballot process, found
that absentee ballots were stolen from mailboxes, that “unscrupulous individuals” had secured ballots for
people under the guise of “helping the voter,” and that voters had been coerced into voting for particular
candidates in return for past favors done for them.37
At the center of what the trial court subsequently found to be “a massive, well-conceived and well-orches-
trated absentee ballot voter fraud scheme” were a large number of absentee ballots—nearly 70 percent of
the total—cast from Little Havana. Little Havana voters reinstalled Commissioner Hernandez, the em-
battled Suarez ally who won reelection to the City Commission by a large majority after being removed
from office by the governor following a 23-count indictment for bank fraud and money laundering.38 An
expert in statistical analysis testified at trial that the large number of absentee ballots from Little Havana
was a statistical “outlier,” the Little Havana absentee ballot rate was an “aberrant case” so unlikely that it
was “literally off the [statistical probability] charts.”
The trial judge, Thomas S. Wilson Jr., concluded that “the evidence shows a pattern of fraudulent, inten-
tional and criminal conduct that resulted in such an extensive abuse of the absentee ballot laws that it can
fairly be said that the intent of these laws was totally frustrated.... This scheme to defraud, literally and
figuratively stole the ballot from the hands of every honest voter in the city of Miami.”39 Judge Wilson over-
turned the results of the November 4 election and ordered a new election, but his remedy was overturned
on appeal. The appellate court affirmed the finding of fraud but voided the remedy of a new election and
remanded the case to the lower court with instructions to enter a final judgment that voided all of the
absentee ballots, determining the outcome of the election by the machine total alone. This decision took
victory out of Xavier Suarez’s hands and gave it to Miami’s new mayor, Joe Carollo.
The 1997 Miami mayor’s race presents one of the more egregious cases of election fraud in recent memory.
As an outlier, it is nevertheless instructive. To affect the outcome of an election, fraud needs to be “mas-
sive,” as the trial judge found. This requires organization and coordination among multiple actors willing
to break the law and risk significant punishment; in other words, a conspiracy. The need for organization
to orchestrate enough fraud to flip the results of an election is the hook to detection. Here, Florida’s civil
election contest procedures worked as they should have, providing the candidate—and by extension, the
voters—harmed by the fraud a means of challenging the distorted results. The fraud was immediately
discovered through conventional criminal investigation.
0 Dēmos: A Network for Ideas Action
News coverage of the fraud scheme and trial was extensive and national and local leaders and residents
loudly bemoaned the further tarnishing of the city’s image as one steeped in political corruption. The state
legislature acted quickly to pass a $4 million election law reform package to root out voter fraud. Unfor-
tunately, the law did much more than that and had a perverse effect on election administration. In its zeal
to address the embarrassing behavior of politicians in Miami, the Republican-controlled legislature passed
a law that paved the way for one of the scandals of the 2000 election in Florida: the massive disfranchise-
ment of Florida voters—most of them African American—whose names erroneously appeared on felony
In May 1998, the legislature added Section 98.0975 to Title IX, Chapter 98 of Florida’s statutes. Section
98.0975 required the Division of Elections in the secretary of state’s office to contract with a private com-
pany to compare the central voter file with databases of persons deceased, those with felony convictions,
and those adjudicated mentally incompetent and to provide lists of matching names to the division. The
rules for matching names were developed by state officials, against the advice of the company hired to
produce the lists, requiring a “match” based on only the first four letters of the first name, 90 percent of the
last name and an approximate date of birth.41 The elections division was required to provide the informa-
tion to the county supervisors of elections who were to undertake their own verification process on local
voter registration databases. The lists were highly inaccurate, for example, one list contained the names
of 8,000 Florida residents who had committed misdemeanors, not felonies, in Texas.42 There is little doubt
that thousands of legal Florida voters were erroneously purged from the rolls and prevented from voting in
Florida in 2000. Florida was the only state in the United States to require its local election officials to verify
their voter rolls using data processed by a private firm.43 Those arrangements were easily manipulated by
partisan election officials to shape electoral outcomes.
the 1996 dornan/sanchez contest For the u.s. house oF representatIves,
orange county, calIFornIa
Orange County, California, is the fourth largest county in the United States, with 2.8 million people count-
ed in the 2000 census, more than one-quarter of them Latino. The 46th Congressional District is nestled in
the heart of Orange County and includes centers of Latino concentration, Santa Ana, the county seat, and
most of Garden Grove and Anaheim, giving the 46th district a population that is nearly two-thirds Latino.
Vast growth and demographic change, along with careful redistricting by Democrats in California’s state
legislature, have facilitated political change in Orange County. Orange County was once a Republican
stronghold, a core constituency for the Republican party in presidential elections because it could swing
California to the party.44 As late as 1988, voters in the 46th district gave 62 percent of their votes to George
H.W. Bush. By 2000, however, a 24 percent Republican margin in presidential elections had been replaced
by a 12 percent Democratic margin when Al Gore won the 46th with 54 percent of the vote to 42 percent
for George W. Bush. The increasing ability of new immigrant and Latino voters to define Orange County
politics and the transformation in party dominance toward the Democrats set the stage for an explosive
case of alleged voter fraud in 1996.
The contested election between the nine-term Republican incumbent Robert K. Dornan and a little-known
businesswoman named Loretta Sanchez involved a blizzard of allegations of registration fraud, non-citizen
and illegal immigrant voting, double voting, voting from nonresidential addresses, illegal inducements to
register and vote, voter intimidation, ballot box tampering, and absentee ballot fraud, all under the canopy
of a bitter and protracted partisan battle that quickly bled into national politics.
An Analysis of Voter Fraud in The United States
One day after the November 5, 1996, election, Dornan led Sanchez by 233 votes, but 12,000 absentee and
provisional ballots had yet to be counted. A week later, when about 3,000 ballots were still left to tally, the
Associated Press called the election for Sanchez, who had moved into the lead with a 929-vote margin. As
the count proceeded, Dornan repeatedly raised the issue of “non-citizen” voter fraud and vowed to take his
re-election fight to the floor of the House of Representatives if he lost. He added that his Republican col-
leagues were looking for a case to use in challenging the recently implemented National Voter Registration
Act, signaling the likely entry of national political forces into the fray.45 Dornan specifically charged that
a well-known Latino rights group and the Democratic Party signed up illegal voters in a drive he argued
may have led to “the first case in history where a congressional election was decided by non-citizens.”46 His
lawyer later called the case “what we think is the single largest example of voter fraud in a federal election
in the last 50 years, and, yes, maybe in this century.”47
On November 22, 1996, the Orange County Registrar of Voters certified Loretta Sanchez the winner by
984 votes,48 and a 14-month battle to deny Sanchez a seat in the House was joined. State electoral and law
enforcement agencies were the first to open investigations into the alleged election irregularities. Then, on
December 26, 1996, Dornan filed a three-page Notice of Electoral Contest in the House of Representatives
requesting an investigation of the election. This was within keeping of his prerogative and the consti-
tutional authority of the House under Article 1, Section 5, Clause 1, which provides that each House of
Congress shall be the judge of the “elections, returns and qualifications” of its members. Under the rules
of the federal Electoral Count Act, the contest is first heard by the Committee on House Oversight, which
conducts its own investigation, and then by the whole House, which disposes of the contest by resolution
or majority vote. In the 105th Congress, the eight-member committee was chaired by Rep. William M.
Thomas, a Republican from Bakersfield, California, and dominated five-to-three by Republican members.
Thomas created a three-person task force comprised of Rep. Vernon Ehlers (R-Mich.) and Rep. Robert Ney
(R-Ohio), and, later, Rep. Steny Hoyer (D-Md.) to conduct the investigation and recommend a course of
action to the full committee. Along with the Orange County D.A. and secretary of state investigations, the
House committee’s investigation took a year to complete and produced, in the end, a disputed finding of
fraud that was too insubstantial to convince the Republican dominated House to upset or reverse Sanchez’s
victory.49 On February 12, 1998, the House voted 378-33 to dismiss Dornan’s contest.
The Dornan-Sanchez electoral dispute fits squarely with what political scientists Benjamin Ginsberg and
Martin Shefter call “politics by other means.” Politics by other means involves the use of legal strategies
and the courts, revelation, prosecution and investigation, and the media to win.50 The fraud allegations and
subsequent 14-month investigations by state, county and federal government agencies cost American tax-
payers well over $1.4 million.51 In the end, very little voter fraud was convincingly substantiated. On April
29, 1998, California’s Republican secretary of state announced that the people identified by the task force
as illegal, non-citizen voters in the 46th Congressional District election of 1996 would not be prosecuted for
voter fraud, the secretary deciding that they had registered in error and not from criminal intent.52
the 2000 electIon, st. louIs, MIssourI
Like most big cities, St. Louis has had its share of election fraud investigations in the past.53 In the wake
of the 2000 election, allegations of voter fraud in St. Louis were raised that included illegal registration;
voting by deceased people, felons, and people whose addresses appeared to be vacant lots; multiple voting;
and unqualified election judges permitting unqualified voters to cast illegal ballots. Before the facts were
known, wild accusations of a vast conspiracy on the part of the Democrats to undertake “a major criminal
enterprise designed to defraud voters” captured national media attention.54
Dēmos: A Network for Ideas Action
The St. Louis case gained national notoriety because the partisan conflict between a senior Missouri Re-
publican senator and a newly elected St. Louis Democratic representative underlying it erupted in congres-
sional hearings and other public venues, giving the story a wider national audience than it would have had
otherwise. It also received attention in the wake of the Florida 2000 debacle and then again going forward
as Congress struggled with election reform over the next year and a half. As will be explained, the contem-
porary St. Louis case is a good example of how the forces promoting expanded access to the franchise are
forced to compromise under the cloud of fraud—sometimes real, but more often conjured up—to achieve
In the months leading up to the election, African-American leaders in St. Louis became concerned that
the recent removal of more than 30,000 names from the registration rolls to an “inactive” list would create
problems at the polls on Election Day. State Senator William Lacy Clay Jr., a candidate for a seat in the
U.S. House of Representatives, gave a speech the day before the election in which he warned that if eligible
voters were prohibited from voting at the polls because of inaccurate registration records, lawsuits would
be brought to keep the polls open past their legal closing time of 7 p.m.55 In fact, that is exactly what hap-
pened. Hundreds of eligible voters were unable to vote because their names had been put on “inactive”
lists which were not distributed to St. Louis’s more than 250 polling places. Poll workers were told to call
headquarters to verify the eligibility of voters whose names were not on their lists, but the problem was
so extensive, the phone lines were jammed for most of the day. When poll workers were unable to get
through, they told voters to go down to the Board’s main office to plead their case. Hundreds of people
tried to cram into the Board’s office at 300 North Tucker Boulevard. Many were still standing in line at
10 p.m., demanding their right to vote.
In the late afternoon of Election Day, Lacy Clay’s campaign, the Gore-Lieberman campaign, and the Mis-
souri State Democratic Committee filed suit in St. Louis City Circuit Court to keep the polls open late so
that registered voters whose records had been mishandled by the Board could vote. A sympathetic judge
issued an order to extend voting hours, but the Missouri Court of Appeals quickly overruled her. The polls
in St. Louis shut down at 7:45 p.m., with only about a hundred additional votes cast after the official 7 p.m.
poll closing time.56
As expected, the Democrats did very well in St. Louis, a heavily Democratic city, but they also did well
statewide where Republicans had long held power, electing a Democrat to the U.S. Senate and as governor.
These were troubling signs to Missouri Republican leaders. Within two days of the election, the state’s
senior U.S. Senator, Republican Christopher “Kit” Bond, called for a federal investigation of voting in St.
Louis, hinting at a conspiracy behind the Democrats’ efforts to extend polling place hours. “What I saw
and heard on Tuesday night is an outrage,” he screamed on national television as he pounded a lectern at a
local Republican party event, adding that the St. Louis Election Board and the Democratic Party should be
investigated for “orchestrat[ing] a concerted scheme to deny all Missouri voters a valid count by keeping
the polls open.”57
Post-election investigations conducted within weeks of the election by the newly elected Republican
secretary of state, Matt Blunt, and the St. Louis Post-Dispatch suggested only a marginal amount of voter
fraud may have been committed in 2000. Voter and election worker error which later would prove to be
the source of the irregularities did not receive the same amount of attention, at least at first. The extent
of St. Louis’ election administration meltdown unfolded slowly, even as the partisan rhetoric heated up.
Most of the initial charges about criminal conspiracies and the defrauding of Missouri voters were even-
tually shown to be overblown. For example, hundreds if not thousands of “ghost voters” were allegedly
An Analysis of Voter Fraud in The United States
registered at vacant lot properties. Bond and others jumped on the allegations to further fuel their fraud
charges. The secretary of state’s probe significantly reduced the number of vacant lot addresses to 79 vot-
ers, and subsequent investigations a year later by reporters at the St. Louis Post-Dispatch discovered that
“dozens of St. Louis voters are being wrongly accused of casting ballots from fraudulent addresses” in the
2000 election. The Post-Dispatch surveyed 1,000 supposedly vacant lots and found that 704 of them had
buildings on them, some of them more than 50 years old.58 Errors in the city’s property records and meth-
ods for classifying vacant a multi-parcel address if only one of the parcels at the address is vacant account
for the mistakes in the voter records.
With no indictments for fraudulent voter registration and the problem of vacant lot addresses largely
solved, Bond and Blunt focused on court orders permitting 1,233 people to vote. Bond again suggested a
criminal conspiracy to create chaos on Election Day so that the Democrats could stuff ballot boxes with
fraudulent votes procured through illegally obtained court orders. The court orders were issued by St.
Louis City and St. Louis County election judges for reasons Blunt argued did not conform to Missouri law.
Some of the court orders appeared to be granted to people who acknowledged that they had failed to reg-
ister by the October 11 deadline, although judges interviewed by the St. Louis Post-Dispatch said that they
believed their court orders complied with state law.59 St. Louis County judge Robert S. Cohen said that
election officials first screened voters who believed they were eligible to vote but who were not on voter
registration lists; voters then had to wait in long lines to have their cases reviewed by an election judge.
Cohen explained to a reporter:
“This process had taken them hours and hours. Some had babies with them; some had wheel-
chairs; some had taken off work. We were trying to accommodate people in a long line and get
them in and out. We were erring on the side of allowing people to vote. Rejecting an American
citizen at the poll who appears to have engaged in no fraud ... it’s a difficult thing to turn that
person away and say you cannot vote, you cannot participate in the democracy today.”60
It would take an FBI investigation in which all of the registration and voting records from the St. Louis
City Elections Board for the month preceding the election were subpoenaed before the facts could
emerge.61 Once that investigation was completed, to the surprise of many, Bond included, the Justice
Department threatened the Board with a lawsuit for abusing the voting rights of thousands of eligible St.
Louis voters by illegally purging their registration records in violation of the NVRA. It was these illegal
purges that created much of the chaos on Election Day leading to more record-keeping errors and mis-
takes, and the appearance of election irregularities. The resulting consent decree stipulated how the Board
would change its policies and procedures for maintaining accurate registration records and how it would
comply with federal requirements for notifying voters of their registration status and handling voter lists
on Election Day. Thus, the alleged voter fraud conspiracy in St. Louis was nothing more than a case of
managerial ineptitude, administrative under-funding, and poor implementation of the NVRA on the part
of St. Louis and Missouri election officials—made worse by partisan rancor and racial politics in a high
interest, competitive election.
The politics of voter fraud in St. Louis spilled over into the national debate on election reform that gripped
the country after the Florida election debacle of 2000. Bond came to play a key role in the legislative
battles over the Help America Vote Act of 2002. At his insistence, the act included a plank that requires
voters to cast only provisional ballots if voting occurs after polls have been ordered to stay open beyond
their legal closing time by state or federal courts.62 Bond was also a leading proponent of federal identifica-
tion requirements for registration and voting. During the debates over the Senate’s initial version of the
Dēmos: A Network for Ideas Action
bill, intense squabbling over voter ID provisions which were demanded by Bond and other Republicans
nearly derailed legislative passage. The House bill included no such provisions and civil rights and elec-
toral reform advocates vowed to work hard against any compromise that included new ID requirements.
But during the House and Senate conference on the bill, Bond and Mitch McConnell of Kentucky, the
Senate’s Republican bill managers, let it be known that Republicans would not back down and that the
voter ID issue was a deal-breaker.63 The final bill included ID requirements but limited them to first-time
voters registering by mail. The retreat by congressional Democrats on the ID issue split the liberal advo-
cacy community which had waged a long and hard-fought effort to win minimum standards and funding
to fix the nation’s electoral machinery. They split, with some condemning the final bill and others offering
tepid support.64 In hindsight, the HAVA ID requirement, limited as it may be, nevertheless paved the way
for a partisan movement in the states that uses allegations of voter fraud to impose more and more restric-
tive identification requirements on voting. The lesson is that the politics of voter fraud matter more for the
election rules we get than the actual evidence of voter fraud itself.
An Analysis of Voter Fraud in The United States
1. Real ID Act of 2005 (P.L. No. 109-13), Div. B, Title II, Sec. 202(c) requires that a state may not issue a driver’s license or non-driver’s ID card unless the
applicant presents documentary proof of his or her full legal name and date of birth, Social Security number or the fact that the applicant is not eligible
for one, address of principal residence and U.S. citizenship.
2. Commission on Federal Election Reform, Building Confidence in U.S. Elections (September 2005), http://www.american.edu/ia/cfer/report/full_report.
pdf; for a dissent see, Brennan Center for Justice and Spencer Overton, Response to the Report of the 2005 Commission on Federal Election Reform, 2005,
3. Kevin Fagan and Mark Simon, “Election-Day Registration Loses; Opponents Said Prop. 52 Would Have Opened Door to Fraud,” San Francisco Chronicle,
November 6, 2002, A5.
4. Matthew Murray, “EAC Blasted Again for Burying Study,” Roll Call, April 9, 2007; Ian Urbina and Eric Lipton, “U.S. Panel Is Said to Alter Finding on
Voter Fraud,” The New York Times, April 11, 2007, A1; Eliza Newlin Carney, “Election Assistance Commission Under Fire,” The National Journal, April
23, 2007; David Nather, “Election Board Facing Votes of No Confidence,” CQ Weekly, April 23, 2007. At the time of this report (August 2007), the Elec-
tion Assistance Commission’s research contracting procedures were under investigation by the Commission’s Inspector General, a review requested by
the Commission’s Chair. See U.S. Election Assistance Commission, “EAC Requests Review of Voter ID, Vote Fraud and Voter Intimidation Research
Projects,”Press Release, April 16, 2007.
5. Dēmos’ 2003 report, Securing the Vote, addresses some of these questions. New briefing papers and a book on fraud to be published in 2008 will expand
on those findings. The U.S. Election Assistance Commission’s controversial report on voter fraud and intimidation, Election Crimes: An Initial Review
and Recommendations for Future Study identifies some of the major gaps in our understanding of what constitutes fraud and intimidation, but neither
this report nor the draft report submitted by the agency’s consultants present new empirical findings or even a coherent definition of voter fraud. Both
the final report and the consultants’ draft report can be found at the EAC’s website, http://www.eac.gov.
6. See note 5.
7. Craig C. Donsanto and Nancy L. Simmons, Federal Prosecution of Election Offenses, 7th Ed. (U.S. Department of Justice, May 2007), pg. 2, www.usdoj.
8. Joseph P. Harris, “Election Reform: The Perspective of Forty Years,” in Richard J. Carlson, ed., Issues of Electoral Reform (New York: National Municipal
League, 1974), pg. 102.
9. See for example “Second Amended Complaint,” N.A.A.C.P., et al. v. Harris, et al., U.S. District Court for the Southern District of Florida, Miami Divi-
sion, Case No. 01-0120-CIV-GOLD/SIMONTON, (March 12, 2002); Greg Palast, The Best Democracy Money Can Buy (Virginia: Pluto Press, 2002), pg.
6-43; Andrew Gumbel, Steal This Vote (New York: Nation Books, 2005); Lance deHaven-Smith, ed., The Battle for Florida: An Annotated Compendium of
Materials from the 2000 Presidential Election (Gainesville: University Press of Florida, 2005).
10. U.S. Commission on Civil Rights, Voting Irregularities in Florida During the 2000 Presidential Election (June 2001).
11. See for example Caltech/MIT Voting Technology Project, Voting: What Is, What Could Be (July 2001); Democratic Investigative Staff, How to Make a
Million Votes Disappear: Electoral Sleight of Hand in the 2000 Presidential Election (U.S. House Committee on the Judiciary, August 2001); the excellent
multi-volume series of studies conducted for Congress since 2000 by the U.S. General Accounting Office (now, The Government Accountability Office),
especially Elections: Perspectives on Activities and Challenges Across the Nation (October 2001), GAO-02-3; and The Nation’s Evolving Election System As
Reflected in the November 2004 General Election (June 2006), GAO-06-450; see note 10; and Anita Miller, ed., What Went Wrong in Ohio: The Conyers
Report on the 2004 Presidential Election (Chicago: Academy Chicago Publishers, 2005).
12. Spencer Overton, “Voter Identification,” University of Michigan Law Review 105(1) (2007), pg. 631-682.
13. Dayna Cunningham, “Who Are to Be Electors? A Reflection on the History of Voter Registration in the U.S.,” Yale Law and Policy Review 9(2) (1991), pg.
14. Ibid. pg. 384, citing Joseph P. Harris, Election Administration in the United States (Washington, D.C.: The Brookings Institution, 1934); Frances Fox Piven
and Richard A. Cloward, Why Americans Still Don’t Vote and Why Politicians Want It That Way (Boston: Beacon Press, 2000), pg. 25–6 91–2.
15. In the 1920s, Harris conducted detailed studies of election administration for the Brookings Institution. His research led him to conclude that elections
were more poorly managed than just about any other area of public administration at the time. “Our elections,” he lamented, “have been marked by
irregularities, slipshod work, antiquated procedure, obsolete records, inaccuracies, and many varieties of downright fraud.” See Joseph P. Harris, The
Registration of Voters in the U.S. (Baltimore: The Lord Baltimore Press, 1929), pg. 3. He nurtured ideas about how to use technology to reduce election
fraud for decades before inventing the Votomatic Vote Recorder (punch card voting machine) toward the end of his academic career. See Interview with
Joseph P. Harris by Harriet Nathan, Oral History (University of California, 1980), http://bancroft.berkeley.edu/ROHO/Vote/. For an enlightening look
at the little studied subject of voting technology, see Roy G. Saltman, The History and Politics of Voting Technology: In Quest of Public Integrity and Public
Confidence (New York: Palgrave, 2006).
16. Dēmos, Election Day Registration: A Study of Voter Fraud Allegations and Findings on Voter Roll Security (August 2007), www.demos.org.
17. The National Commission on Federal Election Reform, To Assure Pride and Confidence in the Electoral Process (August 2001), www.tcf.org/Publications/
18. The Illegal Immigration Reform and Immigrant Responsibility Act of 1996 (P.L. 104-208), perhaps the most radical reform of U.S. immigration laws
ever, makes it much easier to deport otherwise lawful non-citizens for illegal voting in federal elections. Title II, Sec. 216 amended Title 18 of the U.S.
Criminal Code by adding Sec. 611, “Voting by aliens,” which states: “(a) It shall be unlawful for any alien to vote in any election held solely or in part
for the purpose of electing a candidate for the office of President, Vice President, Presidential elector, Member of the Senate, Member of the House of
Representatives, Delegate from the District of Columbia, or Resident Commissioner, unless — (1) the election is held partly for some other purpose;
(2) aliens are authorized to vote for such other purpose under a State constitution or statute or a local ordinance; and (3) voting for such other purpose
is conducted independently of voting for a candidate for such Federal offices, in such a manner that an alien has the opportunity to vote for such other
purpose, but not an opportunity to vote for a candidate for any one or more of such Federal offices. (b) Any person who violates this section shall be
fined under this title, imprisoned not more than one year, or both.”
19. U.S. General Accounting Office, Elections: The Scope of Congressional Authority in Election Administration, GAO-01-470 (March 2001).
Dēmos: A Network for Ideas Action
20. Ibid., 4–7.
21. Ibid., 11.
22. Elizabeth Simson, Justice Denied: How Felony Disenfranchisement Laws Undermine American Democracy, (Americans for Democratic Action Education
Fund, March 2002); Jeff Manza and Christopher Uggen, Locked Out: Felon Disenfranchisement and American Democracy (New York: Oxford University
Press, 2006); Elizabeth A. Hull, The Disenfranchisement of Ex-Felons (Philadelphia: Temple University Press, 2006); see also the many reports and publi-
cations of The Sentencing Project, www.sentencingproject.org.
23. One critic of the NVRA has suggested in recent testimony before the Senate Committee on Governmental Affairs that the lack of an evidentiary record
of voter fraud prosecutions should not be taken as indicative of a lack of voter fraud. This assumes that only the states can pursue fraud claims in court,
which is not the case. See the testimony of Deborah Phillips, founder and chair of the Voting Integrity Project, who asserts without evidence that
“Prosecutors do not like election fraud cases because they take precious resources from strained budgets needed for more serious crimes.” U.S. Congress,
Senate Committee on Governmental Affairs, Hearing on Election Administration Reform, 107th Cong., 2nd sess. (May 3, 2001).
24. 42 U.S.C. Section 1973gg-9(b).
25. Donsanto and Simmons, 24.
26. 18 U.S.C., Section 241 (1988); see also Richard Craswell, “Comments: Federal Prosecution for Local Vote Fraud Under Section 241 of the Federal Crimi-
nal Code,” University of Chicago Law Review 43 (1976), 542–572.
27. The 1995 6th Edition of the Justice Department’s Federal Prosecution of Election Offenses manual included the following instruction (p. 30): “As a general
rule, section 1973i(c) [a criminal statute incorporated into the Voting Rights Act] should not be used to prosecute isolated instances of illegal registra-
tion, vote buying, or fraudulent voting, because such isolated instances do not implicate federal interests sufficiently to warrant federalization of matters
otherwise better left to state election administration and law enforcement…Prosecution of individual and uncoordinated acts should be considered only
where they evidence a widespread systemic abuse which jeopardizes the integrity of the voting process in a particular locale.” This paragraph was wholly
excised in the Department’s most recent (May 2007) revision of the manual.
28. See United States v. Daugherty, 952 F.2d 969, 971 (8th Cir.1991); United States v. Saenz, 747 F.2d 930, 935 (5th Cir. 1984); United States v. Canales, 744
F.2d 413, 416 (5th Cir. 1984).
29. Voting Integrity Symposium, “Prepared Remarks of Attorney General John Ashcroft” (U.S. Department of Justice, October 8, 2002), http://www.usdoj.
31. Eric Lipton and Ian Urbina, “In 5-Year Effort, Scant Evidence of Voter Fraud,” The New York Times, April 12, 2007, A1.
32. Donsanto and Simmons, 2.
33. The City of Miami is not to be confused with Miami-Dade County government. Approximately 365,000 people live in the City of Miami, one of 30
municipal jurisdictions within Miami-Dade County where consolidated government represents the larger “Miami” community and performs most of
the functions of local government.
34. Suarez received 23,598 votes or 53.2 percent to Carollo’s 20,739 votes (46.8 percent). See “Former Mayor Wins an Upset in Miami Ballot,” The New York
Times, November 14, 1997, A30.
35. William T. McCauley, “Florida Absentee Voter Fraud: Fashioning an Appropriate Judicial Remedy,” University of Miami Law Review 54 (2000), 627.
36. In re Matter of Protest of Election Returns and Absentee Ballots in the November 4, 1997, Elections for the City of Miami, Dade County, Fla., No. 97-
25596 CA 09 (Fla. Dade County Ct. March 4, 1998).
37. “‘Outright Fraud’ Found in Miami Mayoral Elections,” The Washington Post, February 3, 1998, A4.
38. Hernandez was eventually convicted of attempting to cover-up the election fraud scheme and sentenced to one year in jail. Governor Lawton Chiles
again removed him from office.
39. Mike Clary, “Miami Without Mayor as Judge Voids Election: Voting: ‘Massive ... Fraud’ in Absentee Forms Cited; New Balloting to Take Place Within 60
Days,” Los Angeles Times, March 5, 1998, A11; Donald P. Baker, “New Mayoral Election is Ordered for Miami; ‘Fraud and Abuse’ Prevalent, Judge Says,”
The Washington Post, March 5, 1998, A2.
40. At the time, Florida was one of only seven states that permanently disenfranchise persons convicted of felony crimes. Its felony disenfranchisement laws
are still one of the harshest in the country; with about one third of all disenfranchised ex-felons in the U.S. residing in Florida. Human Rights Watch and
the Sentencing Project, Losing the Vote: The Impact of Felony Disenfranchisement Laws in the United States (1998); http://www.hrw.org/reports98/vote.
41. Greg Palast, “Florida’s ‘Disappeared Voters’: Disenfranchised by the GOP,” The Nation February 5, 2001; and Palast, The Best Democracy Money Can Buy,
42. Laleh Ispahani and Nick Williams, Purged! How a Patchwork of Flawed and Inconsistent Voting Systems Could Deprive Millions of Americans of the Right
to Vote (New York: American Civil Liberties Union, Demos, Right to Vote, 2004), pg. 1.
43. Palast, The Best Democracy Money Can Buy, pg.16.
44. Lisa McGirr, Suburban Warriors: The Origins of the New American Right (Princeton: Princeton University Press, 2001).
45. Peter M. Warren, “Dornan Vows Appeal to House if Lead is Lost,” Los Angeles Times, November 12, 1996, A1.
47. Dexter Filkins, Peter M. Warren, and Jean O. Pasco, “Dornan, Sanchez Square Off Before House Task Force,” Los Angeles Times, April 20, 1997, A1.
48. A recount requested by Dornan reduced the final margin by five votes.
49. U.S. Congress, Committee on House Oversight, Task Force for the Contested Election in the 46th Congressional District of California, Dismissing the
Election Contest Against Loretta Sanchez: Report of the Committee on House Oversight on H.R. 355, Together with Minority Views (February 12, 1998), H.
50. Benjamin Ginsberg and Martin Shefter, Politics by Other Means: Politicians, Prosecutors and the Press from Watergate to Whitewater, Revised and Updated
An Analysis of Voter Fraud in The United States
Edition (New York: W.W. Norton Company, 1999), pg. 44.
51. Cost estimates are as follows: investigation by the House Oversight Committee ($300,000); expenditures by the INS to conduct data analysis ($500,000);
reimbursement to Dornan for costs associated with his contest ($320,000); reimbursement to Sanchez for her defense ($250,000). These costs do not in-
clude the expenditures by the Orange County Registrar of Voters to conduct an internal review and assist the Committee in its investigation, the Orange
County District Attorney’s Office for its criminal investigation of Hermandad Mexicana Nacionale, or the California Secretary of State’s Office for its
investigation of non-citizen voting in Orange County.
52. “California Won’t Prosecute Non-citizen Voters,” The Washington Post, March 1, 1998, A19.
53. Tracy Campbell, Deliver the Vote: A History of Election Fraud, an American Political Tradition – 1742-2004 (New York: Carroll Graf, 2005), pgs. 160-
54. Carolyn Tuft, “Bond Wants Federal Investigation of Problems at City Polls; He Accuses Democrats of ‘Criminal Enterprise’ in Keeping Polls Open Late;
Democrats Criticize Election Board,” St. Louis Post-Dispatch, November 10, 2000, A1.
55. Jo Mannies, “Vote Fraud Charges are Hogwash, Clay Says, But Bond Stands Firm,” St. Louis Post-Dispatch, February 21, 2001, C2.
56. Safir Ahmed, “Slimin’ the City,” Riverfront Times, November 15, 2000.
57. See Tuft; and Jo Mannies, “City Aimed to Repair Its Image,” St. Louis Post-Dispatch, August 18, 2002. Bond’s charge of criminal intent behind the effort to
clear out what all parties agree was a chaotic situation inside many St. Louis polling places, may have had a personal dimension. Speaking of his success-
ful run for governor in Missouri in 1972, Bond said, “They [St. Louis Democrats] tried the same stunt on me. This time was one too many” — referring
to his belief that St. Louis Democrats intention - ally kept the polls open until midnight in 1972 to prevent his election as the youngest governor in the
state’s history. See Mannies, February 11, 2001; see also, Bond’s statement on the floor of the U.S. Senate, February 13, 2002, in which he linked his own
experience in 1972 to the Democrats’ efforts to keep the polls open late in the 2000 election: “Same state, same city, same play called from the same fraud
play-book.” One of the lawyers for the Democrats in 2000 was Douglas Dowd, the son of Ed Dowd Sr., the man Bond defeated in 1972.
58. Jo Mannies and Jennifer LaFleur, “City Mislabeled Dozens as Voting From Vacant Lots; Property Records Appear to be in Error, Survey Finds; Just 14
Ballots Are Found Suspect,” St. Louis Post-Dispatch, November 5, 2001, A1. Further investigation by the Post-Dispatch fully debunked the vacant lot
claims. According to reporter Jo Mannies, “Basically, we checked every one of the 2,000-plus props [properties] listed as vacant lots with voters, and
found virtually all had houses on them — had been misclassified by the assessor’s office.” (E-mail correspondence with the author, February 25, 2002.)
59. Even Blunt had to admit that communications between local polls and the St. Louis Board of Elections on Election Day was “grossly inadequate.” See
Making Every Vote Count: A Report of Secretary of State Matt Blunt to the People of Missouri (Office of the Secretary of State of Missouri, January 29,
2001). One of his recommendations for improving election administration on Election Day was providing working cell phones to local poll workers and
elections judges and the installation of more telephone lines at Board of Elections headquarters so that poll workers could more easily access the inactive
voter files by calling in their inquiries. In 2000, many cell phones at the local polls had no batteries and a new telephone system at the Board of Elections
malfunctioned, preventing judges from checking whether voters were listed on the inactive file. Under such circumstances, Missouri voters must get a
court order to vote, at least part of the explanation for why so many court orders were issued.
60. Jo Mannies, “Secretary of State Says Local Judges Erred in Election; 1,233 People Were Improperly Allowed to Vote, Report Says,” St. Louis Post-Dispatch,
July 25, 2001, A1.
61. Jo Mannies, “FBI Subpoenas Records From Election Board; Action Follows Charges of Vote Fraud in Recent Elections; Federal Grand Jury Will Get
Documents,” St. Louis Post-Dispatch, April 17, 2001, A1.
62. Section 302(c) of the Help America Vote Act of 2002 (P.L. 107-252) states: “(c) VOTERS WHO VOTE AFTER THE POLLS CLOSE.—any individual
who votes in an election for Federal office as a result of a Federal or State court order or any other order extending the time established for closing the
polls by a State law in effect 10 days before the date of that election may only vote in that election by casting a provisional ballot under subsection (a).
Any such ballot cast under the preceding sentence shall be separated and held apart from other provisional ballots cast by those not affected by the
63. The Act requires that applicants for voter registration must supply their state driver’s license number, or, if they do not possess a driver’s license, the last
four digits of their Social Security number on their registration form. This provision was added to gain support of senators from Oregon and Washington
who were worried that the bill would inhibit their vote-by-mail systems. For all mail registrants, applications must include a copy of documentary proof
of their identity and address. If mail-in registrants fail to provide the documentation, they are required to do so when they vote for the first time in their
new jurisdiction. See, generally, Help America Vote Act of 2002 (P.L. 107-252), Sec. 303(a)(5). On voter ID as a deal-breaker for congressional support
from Senate Republicans, see Karen Foerstel and Emily Pierce, “Hopes for Quick Accord on Election Standards Bill Face Liberals’ Objections,” CQ
Weekly, April 13, 2002; and Derek Willis and David Nather, “Conferees Strike Deal on Election Overhaul,” CQ Monitor News, October 4, 2002.
64. The civil rights and liberal advocacy groups that came together to press for reform after the 2000 election were split over whether the ID requirements
included in the final legislation were reason enough to withdraw support of the bill. The American Civil Liberties Union, Mexican American Legal
Defense and Educational Fund, National Council of La Raza, National Association of Latino Elected and Appointed Officials Education Fund, Lawyer’s
Committee for Civil Rights Under Law all sent letters to Congress opposing the Help America Vote Act in its entirety. The National Association for the
Advancement of Colored People and disability groups endorsed the bill. See, Brian Kim, “Recent Developments: Help America Vote Act,” 40 (Summer
Dēmos: A Network for Ideas Action
Related Resources from Demos
challenges to Fair elections series national voter registration act
▷ Voter Intimidation and Vote Suppression ▷ Implementing the National Voter
▷ Poll Worker Training Registration Act in Public Assistance
Agencies: A Guide for Election Officials and
▷ The Case Against Felony Human Services Professionals
▷ A Promise Still Unfulfilled: The National
▷ Ballot Access for Language Minority Voters Voter Registration Act Twelve Years Later
and People with Disabilities
▷ Expanding Voter Registration for Low-
▷ Voter ID/Proof of Citizenship Requirements Income Citizens: How North Carolina is
Realizing the Promise of the NVRA
▷ A Fallible ‘Fail-Safe’: An Analysis of
Provisional Balloting Problems in the 2006
▷ Continuing Failures in ‘Fail-Safe’ Voting: A
election day registration Preliminary Analysis of Provisional Voting
▷ Election Day Registration in 2007: State Problems (Election 2004)
Legislative Activity ▷ Placebo Ballots: Will ‘Fail-Safe’ Provisional
▷ Anatomy of a Successful Campaign for EDR Voting Fail? (Election 2004)
▷ Election Day Registration: A Ground-Level Candidate Briefing Book
View (A Survey of Election Clerks) ▷ Fulfilling America’s Promise: Ideas to
▷ Election Day Registration: A Study of Voter Expand Opportunity and Revitalize
Fraud Allegations and Findings on Voter Democracy
Roll Security A Dēmos briefing book, with state- and
▷ Same Day Voter Registration in North federal-level application, to help elected
officials advance new policies that
▷ Election Day Voter Registration in Iowa promote electoral participation.
▷ Voters Win with Election Day Registration
election Integrity ▷ Momentum: Igniting Social Change in the
▷ An Analysis of Voter Fraud in the U.S. Connected Age
▷ Fusion Voting: An Analysis ▷ Stealing Democracy: The New Politics of
▷ Improving Access to Voting: A Report
on the Technology for Accessible Voting ▷ Conned: How Millions of Americans Went
Systems to Prison and Lost the Vote
▷ Re-Drawing Lines: A Public Interest
Analysis of California’s 2006 Redistricting
▷ One to a Customer: The Democratic
Downsides of Dual Office Holding
AboUt tHe AUtHoR
Lorraine C. Minnite teaches political science at Barnard College, Columbia University. She is also a Senior
Fellow at Demos, and the co-author of the 2003 Demos report, Securing the Vote: An Analysis of Election
Fraud. Some of the research and analysis presented here are derived from her forthcoming book on voter
fraud in contemporary American elections.
AboUt tHe DeMoCRACy PRoGRAM
The Democracy Program works to strengthen democracy in the United States by reducing barriers to voter
participation and encouraging civic engagement. Dēmos supports state and national reform efforts by con-
ducting research on current and long-range issues; advancing a broad agenda for election reform; provid-
ing advocates and policymakers with technical support; and strengthening reform networks. Through our
recent alliance with the National Voting Rights Institute, we are now able to utilize complementary chan-
nels of policy, advocacy and litigation to achieve our goals.
A NET WOR K FOR IDE AS ACTION
Dēmos: A Network for Ideas Action is a non-partisan public policy research and advocacy organization
committed to building an America that achieves its highest democratic ideals. We believe this requires a
democracy that is robust and inclusive, with high levels of electoral participation and civic engagement; an
economy where prosperity and opportunity are broadly shared and disparity is reduced; and a strong and
effective public sector with the capacity to plan for the future and provide for the common good. Founded
in 2000, Dēmos’ work combines research with advocacy—melding the commitment to ideas of a think
tank with the organizing strategies of an advocacy group.
As with all Dēmos publications, the views expressed in this briefing paper do not necessarily reflect the views
of the Dēmos Board of Trustees.
Visit www.demos.org or contact:
Stuart Comstock-Gay, Director, Democracy Program
firstname.lastname@example.org | (617) 624-3900
Media inquiries: Timothy Rusch, Communications Director
email@example.com | (212) 389-1407
220 Fifth Avenue, 5th fl., New York, NY 10001
T. (212) 633.1405 F. (212) 633.2015
firstname.lastname@example.org | www.demos.org