How Many Non-Citizen Voters? Enough to make a difference

By David Simcox
Volume 18, Number 4 (Summer 2008)
Issue theme: "Illegal alien voters"

Can U.S. residents who are prohibited from voting, but vote anyway, affect the political future of the country or its political subdivisions?

If you believe the word of open suffrage nonprofits1 and think tanks,2 such as those whose challenge to Indiana’s voter ID laws was rejected by the Supreme Court in 2008,3 the number of non-citizens voting is negligible. Dismissing the lax voter registration process ushered in by the 1993 Motor-Voter law, those advocates argue that non-citizens have nothing to gain and a lot to lose by illegal voting, such as loss or delay of naturalization or, if illegal aliens, detection and deportation.4

Those claiming the number of proven cases of fraud is inconsequential often cite a Department of Justice (DOJ) five-year campaign begun in 2002 in which only 120 people have been charged and 86 convicted. These are cases in which DOJ was involved. The scoffers often ignore the sizable number of cases investigated at the state level, including those handled quietly and without criminal penalties in administrative processes.

Statements of various U.S. Attorneys involved invoke a permissive legal doctrine that discourages prosecution of non-citizen voting fraud, such as absence of “concerted effort to tilt elections,” little evidence of “widespread, organized fraud,” “mistakes or misunderstandings by immigrants, not fraud,” and no indications of “conspiracy.”5 Apparently, citizens must tolerate the spreading access of ineligible voters to the ballot boxes as long as it is “disorganized, not concerted, lacking criminal intent, and non-conspiratorial.”

Some cases since 1995 in which community and ethnic non-profit groups have been caught registering non-citizens, such as Hermandad Nacional Mexicana, DemocraciaUSA, or the Association of Community Organizations for Reform (ACORN), are, according to these arguments, just the result of honest mistakes or confusion among the non-citizen registrants themselves about the citizenship requirement or their own status. Some of these non-compliant groups have been beneficiaries of federal grants.

But Americans favoring more, not less, ballot security remain convinced that non-citizen voters in 1996 provided the narrow winning margin in Democrat Loretta Sanchez’s upset of long-time incumbent Republican Bob Dornan in California’s 47th Congressional district (Orange County). They believe that the incident was not isolated and that it presaged a growing threat to good government in general and the Republican party in particular. Few accept the official 1997 finding of a California grand jury that the 624 proven votes by ineligible aliens—out of more than 4000 claimed by Dornan’s attorneys—would not have altered the outcome.6

Those arguing for added safeguards against ineligible voters have provided abundant anecdotal cases of non-citizen registration and voting, but few macro-statistics showing national or regional dimensions of the practice. Much of the evidence tends to be circumstantial. For example, consider census data showing that 41 percent of Hispanics and 33 percent of Asians are non-citizens. Yet a national survey of reasons for not voting showed 13.8 percent of Hispanics and 13.1 percent of Asian gave “ineligibility” as their reason for not registering.7 How much of the gap can be attributed to the unfamiliarity of newcomers with the terms and regulations?

Cases numbering in the hundreds have surfaced in which non-citizen registered voters have admitted non-citizenship to escape a summons to jury duty. But such scattered cases are little help in projecting overall numbers for the nation or its major political subdivisions.

The lack of data is not surprising. It’s not something busy voter registrars in high immigration states want to examine carefully, risking the hostility of open suffrage and ethnic political advocates who impute racism or oppression of the poor to rigorous voting rules. The former Immigration and Naturalization Service (INS)—now the Citizenship and Immigration Service (CIS)—has shown distaste for being drawn into investigations, alleging lack of useful data to prove non-citizenship.8

In another example of reliance on the honor system, the current naturalization application form used by CIS, the N-400, asks applicants if they have registered or voted in a U.S. election. CIS so far has declined to disclose the number answering “yes.” Any such a number would have questionable validity, given the possible complications for the applicant who admits having voted.

But indicators of significant registration of non-citizens continue to pop up, and the current vast voter registration campaigns of Hispanic and Asian ethnic interests since Congress’s 2006 rejection of mass amnesty may force local registrars to increase their rubber-stamping of applications, producing new legions of non-citizens to try to vote in 2008.

South Florida :

A “Welcome the World” Electorate

While exact figures of non-citizen registrants are impossible to determine, a number of indicators suggest at least the order of magnitude of non-citizen voting. In 2001 an article in Insight on the News, “Motor-Voter Law Responsible for Increases in Voter Fraud,” claimed that two to four percent of the votes in the 2000 presidential election were cast by non-citizens, and in some Florida counties 10 percent to 15 percent of votes were cast by non-citizens.9 (In Florida’s Miami-Dade County, over half the population is foreign born and nearly a third of the voting age population is non-citizen.)

The Insight report is apparently based on research on the effects of the 1993 Motor-Voter law by U.S. Border Control, an immigration reform non-profit. USBC found that between 1994 and 1998, Hispanic voter registration in Florida soared by 557 percent, from less than 100,000 to 655,000. In Miami-Dade County in the same period the number of Hispanic registered voters increased 20-fold.10

A look at Census and voter registration data for Miami-Dade indicates a disproportionately high voter registration among Hispanics if the more than half-million non-citizen Latino adults are removed from the county’s pool of eligible Hispanic voters. The percentage of Hispanics registered out of the reduced pool of eligible citizen Hispanic adults is a remarkably high 88.4 percent. The comparable figure for registration of non Hispanic whites, with their usually higher propensity to naturalize and to register and vote is 79.9 percent.

The 88.4 percent registration percentage also dwarfs the national percentages projected for the Hispanic vote in 2008 by the Pew Hispanic Center in December 2007. Pew projects that only 58.2 percent of 18.2 million eligible Hispanic adult citizens will be registered in 2008.11 If Miami-Dade’s eligible adult citizen population registered at the 58.2 percent rate projected by Pew, there would be 354,000 registered Hispanic voters, not the present 536,000.

If the Miami-Dade Hispanic registration figures were generalized to the entire national Hispanic population, there would be 16.1 million registered Hispanic voters instead of the 10.6 million projected by Pew.

The strikingly high Hispanic non-citizen registration in Miami-Dade is probably not fully representative of large Hispanic populations elsewhere. Cubans are 52 percent of the county’s Hispanic population and are the most mobilized and politically aggressive of Latino voting blocks, with a strong proprietary attitude toward local governance. The close identification of most Cubans with the Republican Party intensifies partisan and intra-ethnic political competition in the region—including in registration and mobilization of voters.

California : Tenuous

Eligibility Standards

Another indicator of the order of magnitude of the non-citizen vote is in a 2007 study of California voting trends by a respected think tank, the Public Policy Institute of California (PPIC)12. Its statewide survey of voter registration found 31 percent of the state’s immigrant population to be registered. If registration eligibility laws were fully effective, the entire 31 percent should consist of naturalized citizens. But if a registration rate of naturalized citizens nationally found by Census in 200413 applies here (61.2 percent, according to 2004 Census figures), non-citizens probably account for 300,000 of those registrations (Table 1).

The Census finding of 61.2 percent naturalized citizen registration rate had a margin of error of 3.5 percent. Thus the residual could range from a high of 443,000 non-citizen registrations at a naturalized registration rate of 57.7 percent to a low of 156,000 registrations of ineligibles at 64.7 percent (see Table 1). The Pew Hispanic Center projects for 2008 a naturalized citizen national registration rate of 60.8 percent. Substituting that figure for the 2004 Census figure of 61.2 percent registration rate for naturalized citizens would imply 316,000 non-citizen registrations.

Estimated registration of non-citizen voters in 2007 in California,

If these figures for non-citizen registration in California, home to more than one-quarter of the country’s non-citizens, are projected to the entire country they imply national registration of non-citizens between 600,000 and 1.71 million, with a mid-point of 1.2 million.

Reliance on Census data on the non-citizen population may skew these estimates toward the low side. If the total illegal alien population of the U.S. is 16 to 20 million, rather than the 11 to 12 million currently used by Census, California’s population of non-citizens (including illegal aliens) would be at least a million higher than shown above – or about 6.6 million – and the state’s total immigrant population would approach 11 million. PPIC’s figure of 31 percent registration would imply in column 5 below a non-citizen registration twice as high, at 612,000, or about 2.5 million if applied to the entire nation.

Another indicator so far overlooked or dismissed in the debate is a 2007 research survey by an obscure southern California think tank that shows significant non-citizen voter registration. The study, done by the immigrant-friendly Leavey Center for the Study of Los Angeles of Loyola Marymount University (LMU), and titled 2007 Los Angeles Riots 15th Year Anniversary Resident Follow-up Survey Report,14 surveyed 1,651 L.A. residents, which it broke down into nine sub-populations such as Latino, Anglo, Korean, African-American, U.S.-born, naturalized citizen, and non-citizen.

Question 27 of this poll asked respondents, including non-citizens: “ At your current address are you registered to vote as a Democrat, a Republican, an independent, with another political party, or are you not registered to vote at your current address?

The number of non-citizen respondents in the sample is not stated. If their share of all respondents corresponds to their share of the city’s population (24.6 percent), the sample size would be a statistically acceptable 400. According to Table 2, the non-citizens (along with other major subpopulations) replied as follows to that question:



The structure of the question and the responses suggest these percentages probably understate the numbers actually registered. The 12 percent of respondents are those imprudent or clueless enough to admit to a pollster for publication that they had broken a 1996 federal law (PL 104-208, Sec. 215-216). More likely than not, the 3 percent that played it safe and refused to answer also represents registered voters.

Responses to LMU Poll Question No. 27

The 86 percent declaring themselves not registered is highly likely to include some registered voters who felt it safer to deny it. Another fraction of the deniers would include those respondents who adhered strictly to the question’s limiting terms to answer “not registered,” but either had been registered in the past or at that time were registered outside of the voting jurisdiction for their present address. With these considerations, a reasonable conservative estimate would be somewhere between 15 percent and 20 percent of the non-citizens polled were or had been registered.

Another striking feature of the responses is the party preferences, with registrants favoring the Democrats over the Republicans two to one. Recent exit polls generally show that over half of independents also vote Democratic. Little wonder the Republicans have consistently pushed for more proofs of voter eligibility, while Democrats dismiss them as unnecessary and repressive.

Table 3 on the following page posits from Census data that 88 percent of the non-citizen population are of voting age and 12 percent are registered, as suggested by the LMU poll, and that 80 percent of those registered will actually vote, the percentage projected for registered Hispanic voters in the 2008 presidential elections.15 The table (below) further generalizes the percentages of non-citizen registrants shown in Los Angeles to the national level.

Projected number of non-citizen registrants and voters in 2006 — Low estimate

These projections are based on census estimates of the non-citizen population, including an estimated nearly 12 million resident illegal aliens. But an increasing number of estimates now place the illegal alien population between 16 million and 20 million. The following estimates assume the presence of 16 million illegal aliens, the mid-point between the census estimate of 12 million and the estimates of Bear Stearns and others of 20 million.16 The percentage of non-citizens registered is raised to 15 percent as suggested above in assessing the LMU poll.

Ineligible Voters

and Political Outcomes

The foregoing projections suggest a non-citizen electorate in 2008 ranging from a low estimate of 1.2 million to a high estimate of 2.7 million, with a mid-point just under 2.0 million . How significant are 2.0 million unlawful votes in an overall voter turnout that reached 126 million in the 2004 presidential race? Could 1.6 percent of the total electorate significantly affect the distribution of political power?

However modest the percentage of ineligible registrants seems, in areas heavily populated with non-citizens, such as Los Angeles City and County, Orange County, and the State of California or other states with large foreign-born populations, such as Texas, Florida, New York, and Arizona, those numbers could add up to a balance-tipping bloc of voters. The distortion of political outcomes will become more prevalent in the future if illegal immigration continues to add a half-million new non-citizens a year.

Projected number of non-citizen registrants and voters in 2006 — High estimate

In three presidential elections since 1960, the number of popular votes separating the two top contenders has been far less than the 2.0 million ineligible voters projected here for the country. In the 2008 California primary, Hillary Clinton edged out Barack Obama by less than 400,000 votes, slightly more than the 343,000 non-citizen votes conservatively estimated for that state.

In several California congressional districts, the ineligible alien vote may well decide the election. In the 47th district, Loretta Sanchez, who ousted Republican Bob Dornan in 1996 in a disputed election, eked out a 14,000 vote victory over Asian immigrant Tan Nguyen in 2006. Sanchez’s victory margin was little more than the 13,600 non-citizen voters projected here.

In California, Texas, and Florida, a number of Congressional districts are nearing ethnic tipping points where the extra edge of non-citizen votes could hasten the transition from African-American or Anglo to Latino incumbency. Perversely, African-Americans may be the victims of the abuse of the lenient suffrage rules they have long championed. Particularly vulnerable in rapidly Latinizing districts are such sitting Texas Congress members as Al Green (9th), Immigration subcommittee Chair Sheila Jackson-Lee (18th), and Eddie B. Johnson (30th).

African-American Congresswomen in California whose once heavily Black districts are rapidly Latinizing include Maxine Waters (35th) and Laura Richardson (37th). And in the 51st district (Imperial County), which would have upwards of 13.2 non-citizen voters, increasingly lonely Anglo Congressman Bob Filner may be among the next to succumb. A notable irony is that entrenched Cuban-American Republican Congress members in three south Florida districts are now imperiled by the rapid rise of mostly Democratic non-Cuban Latino numbers in their districts.

Narrow election victories are not rare at the state and local level, even in the more populous jurisdictions. In the following recent cases, the margin of victory was well within the non-citizen voter numbers estimated here:

* In the 2004 Virginia Senate race, Democrat Jim Webb shaded out George Allen by 7,231 votes. This study estimates 42,000 votes cast by non-citizens.

* In Washington State, Christine Gregoire won the Governor’s race by a tiny 261 vote margin – a trifle within the 47,000 ineligible registered voters in that state.

* In 2004, Democratic presidential candidate John Kerry lost Nevada’s electoral votes by a margin of only 22,000 popular votes – a little under the projected 25.600 votes cast by non-citizens.

If 58 percent of the non-citizens registered are either Democrats or Democratic-leaning independents, as these LMU poll numbers imply, it could seriously imperil the Republican presidential candidate in close races in states such as Texas, Arizona, and Florida.

Motor-Voter: An Honor

System Short on Honor

Worth asking is not how did these numbers get so large, but why aren’t they even larger? The 1993 Motor-Voter Act, as many conservatives warned, has made voter registration of applicants for drivers’ licenses and social services virtually automatic, putting applicants’ responses to such key questions as U.S. citizenship on an honor system. For illegal aliens, now at least 55 percent of the non-citizen population, drivers’ licenses and, until recently, voter registration cards were documents of choice to satisfy prospective employers on identity and eligibility to work in the U.S.

The Federation for American Immigration Reform (FAIR), for example, in 2004 claimed that a large portion of the 250,000 persons issued New York State drivers’ licenses while providing erroneous social security numbers were illegal aliens who were then quasi-automatically registered to vote under the procedures of the Motor-Voter law.17 The New York experience was echoed in a number of states, until a public backlash after 9/11 forced tighter strictures on issuing licenses to illegal alien license seekers. Motor-Voter, however, still eases voter registration for the 45 percent of the immigrant population who are legal permanent residents, but won’t pay the price of naturalization.

Presumption of citizenship is increasingly the policy of voting registrars. Simply signing voter registration forms and mailing them is a satisfactory affirmation of citizenship — no further questions asked. In some states such as Iowa and South Dakota, eligibility to vote is presumed even if the applicant fails to affirm U.S. citizenship.

In some immigrant “sanctuary” cities, asking about citizenship status even in such state-controlled transactions as voter registration runs counter to an emerging “don’t ask — don’t tell” ethos. Qualifications for voter registration are being eroded by immigration advocates’ efforts to define an individual’s immigration status as an “intimate personal matter” protected by the right to privacy.18 Incredibly, your citizenship is presumably no one’s business in exercising a fundamental right of U.S. citizenship.

Non-citizens signing up to vote quite correctly don’t see much risk of penalties in it. The federal government has shown no interest in enforcing its 1996 ban on non-citizen registration and voting, and the federal congress has shot down all initiatives to require proof of citizenship. And U.S. Attorneys, notwithstanding White House pressures, have shown little interest in giving the issue priority attention.

Only Arizona has legislatively required proof of citizenship in registering, a provision approved in 2004 but held up by challenges in federal court. Texas, California, Missouri and 16 other states have considered similar legislation. None of the pending state measures, if adopted at all, would go into effect until after the 2008 elections. Most of the draft bills, including the one rejected soundly in California, would require proof of U.S. citizenship from first-time applicants for registration, and in some states each succeeding re-registration. Appropriate penalties would be imposed along with better verification systems for state registrars.

Following the Democratic Congressional sweep in 2006, any political momentum in Congress for keeping non-citizens out of the polls has slowed or in some cases reversed. An indication is Senator Barack Obama’s 2007 bill, the Voter Intimidation and Deceptive Practices Prevention Act (S. 453)19.

Obama and other authors of this bill effectively played on the calculated outrage and indignation mounted by Latino and other populist organizations at a letter reportedly disseminated by a California immigration reform organization during the 2006 race in California’s 47th Congressional district between incumbent Loretta Sanchez and unsuccessful Republican challenger Tan Nguyen.

The letter, cited in the findings of Obama’s Senate bill, warned (correctly) that “illegal aliens” and (incorrectly) “immigrants” were ineligible to vote.20 Obama’s legislation, as now written, is more likely to do the “intimidating,” mostly of those citizens who would make an issue of non-citizen voting. The bill would seriously chill political free speech. While the House version passed in 2007, Obama’s bill is still tied up in the Senate — a condition likely to change in an Obama presidency.

Absentee and Mail-In Ballots:

Invitations to Fraud

Some states’ encouragement of the use of absentee ballots has reduced the risk to non-citizens of facing challenges at the polling place. Also facilitating ineligible voters is the greatly increased use of absentee ballots in immigrant-rich states. In California, 31 percent of all votes were cast by mail; in Arizona and Colorado 32 percent and 29 percent, respectively. In those jurisdictions, non-citizens can register on-line or by mail and regularly cast ballots without risking face-to-face challenges or close scrutiny at the registration office or at the polls. A possibly revealing statistic for the 2004 election period is that 23 percent of naturalized citizens registered by mail compared to only 12 percent of native-born citizens.21

An extensive report by the Florida Department of Law Enforcement (FDLE) on voter fraud, much of it in Miami-Dade, saw absentee voting as the greatest facilitator of fraud.

The desire to facilitate the opportunity for each person to vote has resulted in increased opportunity to use absentee ballots improperly. (Once one has registered fraudulently, he or she can obtain an absentee ballot for every election thereafter if he or she wishes. The lack of “in-person, at-the-polls” accountability makes absentee ballots the “tool of choice” for those inclined to commit voter fraud)22

Three factors probably boosting current estimates heading into the November presidential election would be the surge of young Latinos now reaching voting age, the massive ethnic voter registration drives during the past two years, and the increased militancy of ethnic political groups over Congress’s repeated rejections of amnesty legislation and support for tightened enforcement since 2005.

Voter registration drives by ethnic lobbies or their spin-offs have proliferated in the past two years. They are appealing to liberal foundation grant-givers. Such organizations as DemocraciaUSA, VotoLatino, and Project Vote work closely with major Latino organizations, such as La Raza, League of United Latin American Citizens—(LULAC), the National Association of Latino Elected officials, the League of Women Voters, the heavily Hispanic Service Employees International Union (SEIU), and Univision Spanish-language media network. Funding comes from such foundations as People for the American Way, and Pew.

Searching for new voters among Asian immigrants are the Asian-American Legal Defense and Education League and Asian and Pacific Islander Americans ( Many of the Hispanic and Asian groups boast slick high-tech websites featuring on-line registration (raising further questions about the integrity of the registration process.)

In Los Angeles, even local college chapters of the Aztlan Chicano Student Movement (MEChA) in Pasadena and East Los Angeles have joined the Latino get-out-the-vote effort. It’s hard to imagine that MECHA ideologues could be too conscientious about following the voter registration rules of a government they hold to be an illegitimate occupying power in their homeland.

Voter registration can be profitable as well as fulfilling. Registration agents contract with political or interest groups to bring in newly signed registration forms for a fee—now from two dollars to five dollars a head. How many agents are going to turn down a registrant—and a fee—over a minor technical matter of citizenship?

Ethnic political entrepreneurs count on lots of votes, ineligible or otherwise, to swell their own political leverage. For the non-citizens, registering and voting is a way of ensuring friendly representation in the local halls of power and repaying the patronage they receive from their local “ patrones.” Ineligible registration and voting fits easily within the prevailing civic culture of quiet, contagious mendacity in immigrant enclaves, where most aspects of life itself are lived “off the books,” and one’s identity and antecedents depend on the circumstances of the moment.

It’s Pay Back Time at the Ballot Box

The perception of alien entitlement to vote is further nourished by the rising campaigns in a number of states and cities to make non-citizens eligible to vote on local issues, further confusing non-citizens about the current restrictions and increasing the atmosphere of permissiveness and illegitimacy in the existing system. Adding to this is a “cut the red tape” frustration among legal immigrants toward naturalization. Washington’s delays in reducing the huge backlog of naturalizations has left many feeling that a premature exercise of the suffrage is justifiable.

The failure of ethnic interests in Congress in 2005–2006 to win amnesty injected a new militancy in mobilizing immigrant voters—a sense of gain power or else! Washington’s 2007 crackdown on illegal entry and employment and the spread of state restrictive laws toward illegal aliens have heightened the urgency.

A common placard at Hispanic pro-amnesty demonstrations in 2006 was a warning to Anglo America: “Hoy Marchamos, Mañana Votamos!” (“Today we march, tomorrow we vote”). The message implies a determination to reward or punish at the ballot box—with or without the cachet of citizenship.

A measure of the urgency for immigrants to enroll for their survival is the exhortation of a popular Univision comedian, Latino folk hero and former illegal alien, Eddie Sotelo23 (popularly known as “ Piolin”). Sotelo’s exhortation echoes the current back-to-the-wall rhetoric among illegal immigrants: “Si no votan, nos botan” (“If you don’t vote they’ll throw us out”)

There is no less urgency for citizens, naturalized and native. And they must decide whether voting should remain an exclusive attribute of citizenship and demand the safeguards to make it so. ■


1. For example, People for the American Way;

2. Brennan and


4. See Brennan Center for Justice, “The Truth about Fraud: Alleged Voting by Non-Citizens”

5. Ibid.

6. Jim Miller, “The Sour Lesson of Bob Dornan’s Defeat” (


8., pp. 7–8.

9. “Motor-Voter Law Responsible for Increase in Voter Fraud.” Insight on the News, August 20, 2002. Accessed at

10. U.S. Border Control, PR Newswire release of November 9, 2000. “Will illegal aliens decide who will be America’s next president?”

11. Pew Hispanic Center. “Hispanics and the 2008 Election: Hispanic Population, Voting and Registration Trends,” December 8, 2007, page 16, Table 4.


13. Op. cit., endnote 7.


15. Pew Hispanic Center, op. cit.

16. Robert Justich and Betty Ng, CFA, “The Underground Labor Force Is Rising to the Surface,” Bear Sterns, January 3, 2005. See also Wayne Lutton, “Assessing the Illegal Alien Population – Key Findings from an Exclusive Report.” The Social Contract, Summer 2007,





21. Op. cit., endnote 7.

22. www.fdle.state.fl.usl Op. Cit p 1.


About the author

David Simcox, former chair of the Policy Board of the Center for Immigration Studies, is the director of Migration Demographics, a research organization.