Access denied: The problem with voter ID laws

June 12, 2007

In Wisconsin, voter fraud is rampant. Or so thought U.S. Attorney Steven Biskupic, who began a hunt for fraudulent voters after John Kerry won Wisconsin by just 11,000 votes over George W. Bush in 2004, in an election that Republicans claimed was tainted by widespread voter fraud. But by the time he completed his work, Biskupic reported that he had uncovered no conspiracy to commit fraud. His prosecutors ended up charging only 14 people with voting illegally—and only four of them, all felons ineligible to vote, were convicted.

Lawmakers in many states are saying that there’s only one way to stop this epidemic of fraud: have every voter show ID at the polls—ideally a state-issued photo ID. But experts on elections say that voter fraud of the kind that could be countered by ID requirements is rare. What’s more, requiring photo IDs would disenfranchise millions of voters. The supposed remedy, these experts say, would turn out to be far worse than the actual problem.

Since 2002, the Justice Department has made an all-out effort to track down and convict fraudulent voters. By 2006, those efforts had yielded just 86 convictions nationwide, and many of those incidents, like the four Wisconsin cases, would not have been prevented by a voter ID requirement.

Meanwhile, a study by the Brennan Center for Justice (see truthaboutfraud.org) found that some 21 million citizens—including a disproportionately large number of African Americans and elderly people—do not have government-issued photo ID. As many as 4.5 million people have a photo ID that lacks their current address or current legal name; many of these are young adults and people with lower income who move frequently. The proof-of-citizenship requirements that some ID advocates propose are especially onerous for married women: 32 million voting-age women do not have documents to prove citizenship that reflect their current legal name.

Proponents of strict voter ID press their case by magnifying the size of the fraud problem while minimizing the impact of voter ID laws. Ohio is one state that strengthened its ID laws after the 2004 election. The League of Women Voters teamed up with a housing advocacy group there to find out how many cases of individual voter fraud had been pursued in relation to the 2004 presidential election. They came up with a statewide total of four, or 0.00004 percent of the nearly 10 million votes cast.

But to hear voter ID proponents tell the story, fraudulent voters were everywhere in Ohio. One master of magnification is Mark “Thor” Hearne of the American Center for Voting Rights. If Web presence is any indicator of an organization’s legitimacy, the ACVR should raise eyebrows: its Web site didn’t appear until March 2005, and it disappeared exactly two years later. But after the 2004 election, the ACVR was everywhere, testifying at hearings and filing lawsuits that claimed voter fraud.

Hearne, who was national election counsel for Bush-Cheney ’04, was called to testify on behalf of the ACVR before the House Administration Committee—chaired at the time by Bob Ney (R., Ohio), who is now serving prison time for corruption—when that committee was looking into irregularities in the 2004 election. Hearne claimed that fraud was reported “in every corner” of Ohio, and that “the fraudulent voter registrations totaled in the thousands.”

Hearne offered the committee this rhetorical flourish: “Ohio citizens deserve the confidence that they the voters—not trial lawyers, activist judges and special-interest groups soliciting fraudulent votes with crack cocaine—determine the result of Ohio elections.” He was inspired to include the last example by the case of a hapless addict who did indeed confess to accepting cocaine in lieu of cash as payment for turning in completed registration forms. (The fraudulent forms were spotted because officials wondered why so many people with names like Mary Poppins and Michael Jackson lived on a single block and had the same handwriting. Presumably this is one of the four cases the League of Women Voters discovered. The addict was charged with a felony, and Mary Poppins didn’t get to vote.)

Proponents of strict voter ID laws make their recommendations sound like a matter of common sense. “Every day millions of Americans show a picture ID to pay by check, board a plane or buy alcohol or tobacco,” argued Vernon J. Ehlers (R., Mich.) when the U.S. House was preparing its own voter ID bill. “Surely the sanctity of the ballot warrants as much protection as these other activities. Our voting rights are too important to rely on an ‘honor system.’” Sound bites like these echoed from coast to coast as the bill skated through the House on a near-perfect party-line vote.

But the analogy is not accurate. First of all, it’s not true that you have to have an ID to get on a plane. As George Washington University law professor Spencer Overton pointed out in an interview, “If you don’t have an ID [at the airport] there’s a different process, more of a search, but you don’t need ID. Even in the context of terrorism there are exceptions for people flying without photo ID.”

Many of the new voter ID laws, on the other hand, offer no exceptions: some states allow you to vote provisionally if you don’t bring your ID to the polls, but require you to bring the ID to the board of elections within a couple of days if you want your vote counted; other states don’t allow even this fallback option.

Ehlers’s point about the “honor system” is also disingenuous. Most states do require voters to prove their identity in some way, and those who come to the polls without documentation can sign an affidavit attesting to their identity. If suspicions of fraud surface, the affidavit becomes a tool for investigators. Penalties for voter fraud are so high that it’s unlikely many individuals would be willing to risk imprisonment just to add a single vote to their favorite candidate’s column.

“If there are so many fraudulent votes out there, more than the legitimate votes that would be excluded, then we can consider voter ID,” Overton said. “But let’s get the facts on the table to make an assessment instead of using anecdote and inappropriate analogies to analyze the problem.”

Overton came to the voter ID conversation by way of his membership on the Commission on Federal Election Reform—called the Carter-Baker Commission because it was chaired by Jimmy Carter and former secretary of state James Baker. The Carter-Baker Commission gave an enormous boost to voter ID proponents in 2005 when it recommended that voters be required to present a REAL ID, a proposed government-issued photo ID that indicates citizenship status. Overton believes that the commission got some bad advice.

“The top experts in the field weren’t brought in,” Overton said. “People who were brought in included John Fund, a journalist and Wall Street Journal political commentator with a particular perspective.” (Fund is author of Stealing Elections: How Voter Fraud Threatens Our Democracy.) “They didn’t bring in academics.” Another of the panelists was Colleen McAndrews, an attorney who had fought challenges to the Gray Davis recall in California and served as campaign treasurer for Arnold Schwarzenegger.

The commission held only two hearings, allocating under three hours to voter ID and related issues, and allowed no public comment. The Lawyers’ Committee for Civil Rights Under Law issued a press release echoing Overton’s concerns: the commission “gave little attention to detail,” it said. “There were no separate task forces devoted to any particular aspects of our election system. There was little attempt to gather rigorous empirical data to support any conclusions. The result is a report based on anecdote and supposition, rather than rigorous analysis of real-world facts.”

The Carter-Baker deliberations were also shaped by the post-9/11 political context: Congress had just passed the REAL ID Act, which requires states to issue a uniform ID card that indicates citizenship status as well as identity. Overton suggests that in the absence of a careful “cost-benefit analysis,” Carter and the commission’s Democrats operated on a “hunch” that because “everyone will have photo ID for security purposes, it’s not a big deal to ask for ID.”

Some of the commissioners, said Overton, “would make the argument that the objective is national security, and therefore encouraging an ID for voting will encourage everyone to get the ID.” (Lee Hamilton, chair of the 9/11 Commission, was also a Democratic member of the Carter-Baker Commission.)

Now, though, many states are balking at implementing REAL ID. Arkansas, Maine and Idaho have passed resolutions rejecting the law, and legislation is pending in Congress to repeal the REAL ID Act.

Overton and two other commissioners dissented from the voter ID recommendation (you can read Overton’s reasons at carterbakerdissent.com). Commissioner Shirley Malcolm did not, but she said in an interview that her support had been “based on the idea that REAL ID was going to happen.” Asked if she would have supported the recommendation if REAL ID hadn’t been in the picture, she answered firmly: “No, I would not have.”

The dissenters had hoped to make their case in formal statements that would be added to the commission’s report. But the heads of the commission announced that the dissents would be limited to 250 words. Overton said that he “felt compelled to write an academic article because the whole debate was reduced to sound bites.”

That article, which appeared in the Michigan Law Review in January 2007, begins the sort of cost-benefit analysis that was missing from the commission’s deliberations. “Policymakers,” he challenges in the article, “should await better empirical studies before imposing potentially antidemocratic measures.”

Voter ID proponents haven’t been waiting for better studies. With the Carter-Baker wind at their back, they continue to push for stricter voter ID laws in state after state while keeping up the steady complaint about voter fraud.

Meanwhile, voting-rights activists have been waiting for the release of a pair of studies requested by the U.S. Election Assistance Commission—one on voter fraud, the other on the impact of voter ID laws.

The voter ID report, completed in June 2006 by researchers at the Eagleton Institute of Politics and the Moritz College of Law, was finally released (but not adopted) by the EAC in March 2007. That was too late for its finding—that strict voter ID requirements correlate with lower voter turnout—to affect the debate over voter ID that preceded the 2006 congressional elections.

The report’s authors acknowledge an important limitation of their study: if experts are to assess “the effectiveness of voter ID as a way to protect the integrity of the ballot,” then their research “should logically include an estimate of the nature and frequency of vote fraud.” Their research didn’t do that. Nor did it measure how effective various voter ID rules are for countering fraud.

The Eagleton/Moritz researchers weren’t looking at voter fraud because the EAC had hired someone else to cover that topic: Tova Wang, a progressive election reform expert with the Century Foundation, and Job Serebrov, a conservative Arkansas attorney. Wang and Serebrov reviewed literature and legal cases related to voter fraud, and they interviewed elections officials and election law experts from across the country and across the political spectrum. But their final report has never seen the light of day because the EAC has refused to release it.

A summary of Wang and Serebrov’s findings submitted in May 2006 and a draft of their final report obtained by the New York Times this year provide some clues about the content of the unreleased document. Among the experts the researchers talked to, agreement was all but unanimous that polling-place fraud of the sort that would be prevented by strict voter ID rules is very rare.

Paul DeGregorio, chair of the EAC at the time, told USA Today last fall that the agency was sitting on the report because “there was a division of opinion” within the agency. “We’ve seen places where fraud does occur,” he said. By December, EAC staff had written their own report on election fraud, titled “Election Crimes: An Initial Review and Recommendations for Future Study.” It is ostensibly based on Wang and Serebrov’s research but presents a conclusion opposite theirs: EAC staff write that there is no consensus on individual voter fraud. In support of this reversal, they point to the allegations of massive fraud in Wisconsin that were debunked by U.S. Attorney Biskupic in 2005.

Freedom of Information Act requests haven’t succeeded in getting the EAC to cough up the real Wang/Serebrov study; nor have requests from Congress or the press. When I spotted an article in April saying (incorrectly, as it turns out) that the EAC was now making the Wang/Serebrov report available to journalists, I called to ask for a copy. The EAC press liaison sent me the “Election Crimes” document. I think I was supposed to believe that this was the real thing. A colleague who made the same request a month later met with the same result: twice she requested the report authored by Wang and Serebrov and twice a press liaison sent her “Election Crimes,” giving no indication that these were two different documents.

Because of contractual obligations, neither Wang nor Serebrov can respond to press inquiries about their findings. But in an April press release protesting the continuing censorship, Wang does discuss her communications with the EAC. She writes that from July 2006, when she and Serebrov submitted their final report, to December 2006, when the “Elections Crimes” document was released, “no member of the EAC Commission or staff contacted me or my coauthor to raise any concerns about the substance of our research.”

In an e-mail obtained by the New York Times, Serebrov gives his own hint about what was happening behind the scenes; he wrote in the e-mail to an EAC staffer: “I could care less that the results are not what the more conservative members of my party wanted. Neither one of us was willing to conform results for political expediency.”

While Wang and Serebrov’s conclusions remain buried, the voter ID movement presses on. One house of the Texas legislature has just passed a voter ID measure, and Mississippi lawmakers are poised to do the same.