By Daniel P. Tokaji
Prof. Tokaji served as an expert consultant to the Carter-Baker Commission, but had no role in preparing its report and was not privy to the Commission’s deliberations.
Yesterday, the Commission on Federal Election Reform, better known as the Carter-Baker Commission released its report “Building Confidence in U.S. Elections.” For those who had hoped for an evidence-driven inquiry into the best way to improve our elections system, the report is largely a disappointment. The report is long on pronouncements about what ought to be changed, but short on evidence that supports its chosen recommendations. Two of its most significant recommendations embrace simplistic solutions, one urged by those on the left and the other by those on the right. These recommendations seem designed to broker a political compromise, but would actually make our election system worse rather than better.
Because advisory commissions lack the power to pass laws, their power rests largely on credibility. At their best, bipartisan commissions can deploy systematic empirical research to cut through partisan self-interest and provide a basis for consensus among those of diverse ideological perspectives. The immediate predecessor of the Carter-Baker Commission, the National Commission on Federal Election Reform (more commonly known as the “Carter-Ford Commission”) was a fine example. The Carter-Ford Commission’s 2001 report was based not just on the opinions of a diverse group of election experts, but also on meticulous research. This included a series of task force reports that rigorously analyzed the problems that our election system faced. The Carter-Ford Commission made recommendations based upon this careful research, that eventually led to enactment of the Help America Vote Act.
At their worst, bipartisan commissions simply paper over difficult issues, substituting superficially appealing solutions for rigorous inquiry into the problems at hand. Such reports may produce a report that can be marketed to politicians and the public, but are unlikely to present a workable and effective solution to the issues they purport to address. Worse still, the ultimate recommendations may play more upon rhetoric than on a careful examination of the evidence. If a commission report lacks a sound evidentiary basis, then its credibility – and thus power to persuade – is diminished.
Regrettably, the newly published report from the Carter-Baker Commission comes much closer to the latter type of report than the former. The two recommendations likely to generate the most attention are: 1) to require that voters show photo identification in order to have their votes counted, and 2) to require that electronic voting machines generate a contemporaneous paper record of the electronic vote, or “voter verifiable paper audit trail” (VVPAT). Advocates on the right have been pressing for strict ID requirements, while support for the VVPAT has largely come from those on the left. The Carter-Baker Commission’s report marshals precious little evidence to support either of these key recommendations – and what little evidence that the Commission identifies cuts directly against both of them.
The “REAL ID”
The question whether to require voters to show photo identification has been the most contentious issue to emerge since the 2004 elections. In Indiana and Georgia, Republican legislators have succeeded in enacting bills to require that voters show photo identification when they appear to vote. These laws are ostensibly motivated by concerns about voting fraud. When one actually looks at the evidence, however, there’s little of it to support the conclusion that fraud is widespread – much less than requiring photo identification is an effective way to curb what little fraud does exist. Voting rights advocates have also pointed out that elderly voters, people with disabilities, the poor, and racial minorities are much less likely to drive, and therefore more likely to lack photo identification. A recent Wisconsin study, for example, found that only 22 percent of black men between the ages of 18 and 24 have a driver’s license.
Instead of taking serious account of the evidence, the Carter-Baker Commission recommends that all voters be required to show so-called “REAL ID” cards if they wish to have their votes counted. Though the Commission portrays itself as adopting a middle-ground approach, the proposal that is actually more extreme than that which has been taken by any state to date. Effective 2010, a government-issued “REAL ID” would be required of all voters who wish to have their votes counted.
The Carter-Baker Commission makes this recommendation against the available evidence. As its report admits: “There is no evidence of extensive fraud in U.S. elections or of multiple voting.” The report brushes to the side the evidence uncovered by the Carter-Ford Commission’s task force, that between six and ten percent of voting-age Americans lack a driver’s license. Nor does the Carter-Baker Commission attempt even a cursory cost-benefit analysis – it does not include any assessment of whether or not the supposed benefits of requiring photo identification outweigh the considerable burdens it imposes on our most vulnerable citizens.
The Carter-Baker Commission can at best be characterized as paying lip-service to the grave voting rights concerns surrounding strict voter ID laws that have led civil rights advocates to label such proposals the “new poll tax.” Already, the Brennan Center for Justice and Professor Spencer Overton, a member of the commission, have released a lengthy rebuttal to the Carter-Baker Commission’s REAL ID recommendation. This response references empirical research that the Carter-Baker Commission avoided, making a strong case that requiring photo ID would impose substantial barriers on voting rights while having little tangible benefits.
The Paper Trail
The Commission’s recommendations regarding voting technology are similarly lacking in evidentiary support. The Commission recommends that Congress require that voting equipment generate a “voter verifiable paper audit trail.” Here again, the purported justification for the recommended reform is to deal with worries about fraud – in this case, the concern that an insider might tinker with voting machine software to rig an election. On this issue, however, concerns about fraud have mostly come from those on the left rather than those on the right.
As is the case with the concerns of voter fraud that drive ID proposals, one cannot say that the prospect of machine fraud is beyond the realm of possibility. But here again, the Commission exaggerates the magnitude of the problem and proposes a solution that is likely to make things worse rather than better. While there’s a long history of election officials tampering with paper ballots, there’s little evidence that this is occurred with electronic voting software, or that it could occur if appropriate procedural checks are in place. More important, the Commission’s proposed fix is unlikely to be a workable and effective solution to the legitimate security concerns it raises.
The Commission would have been much better off had it carefully examined the experience of the states. While a handful of states have adopted laws to require that electronic voting machines produce a VVPAT, only one state – Nevada – has experimented with this device on any significant scale. A logical starting point for analysis would be to engage in a careful study of Nevada’s experience in the 2004 election. But the most that the Carter-Baker Commission can say about Nevada’s VVPAT experiment is that it “appear[s] to have worked well.” Instead of relying on research on what actually happened, the Commission makes policy recommendations – for federal legislation no less – based on appearances.
Even worse, the only authority that the Carter-Baker Commission cites for the proposition that VVPAT machines “appear to have worked well” is a study by Professor Ted Selker, co-director of the Caltech-MIT Voting Technology Project. But as discussed here, Professor Selker’s real views are, in his words, “exactly the opposite” of those that the Carter-Baker Commission attributes to him. Professor Selker’s research has raised serious questions about whether voters will actually check paper copies of their electronic ballots and thus whether they will be an effective antidote to concerns of ballot security.
Such errors on the Commission’s part would be easier to excuse, if there were more evidence to support its conclusion that the VVPAT is a workable and effective solution. But in fact, the evidence points in precisely the opposite direction. The Commission does not examine the experience of California and Ohio, states that have adopted VVPAT requirements. In California’s recent test of one VVPAT system showed that 10 percent of machines experienced paper jams. The problems in implementing the VVPAT have led Los Angeles County – the nation’s largest and most diverse voting jurisdiction – to stay with a subpar central-count optical scan voting system. The Commission also overlooks the experience of Ohio, where requiring the VVPAT led counties to stay with unreliable punch card systems in the 2004 elections. Ohio’s VVPAT law will also result in one (or at most two) voting machine companies being given a monopoly on the state’s electronic voting business, despite the fact that neither VVPAT device has been used in real-life elections on any significant scale. The Carter-Baker Commission’s failure to examine the evidence – and, even worse, its mischaracterization of the slender evidence upon which it purports to rely – raises serious questions of credibility.
Credibility
The credibility of the Carter-Baker Commission report is further called into question by the way in which it chose to handle dissent. Buried at the end of the report (pp. 88-91) are no less than ten separate statements by members of the commission, including dissents on the issues of voter ID and the VVPAT. The report says that “[a]ll of the Commission Members are signatories” to the report and that some commissioners were “asked to limit [separate statements] to 250 words.” In fact, some of the dissenting commissioners were not merely asked but required to limit their separate statements to this length. And it is abundantly clear that there are several recommendations – including those concerning ID and the VVPAT – on which there was dissent. This forced Professor Overton to create a separate website (http://www.carterbakerdissent.com/) in which to explain his reasons for dissenting from the REAL ID recommendation. Professor Overton also highlights some other problems that he believed to exist with the processes the Commission followed.
Given the importance and complexity of the issues addressed, limiting dissenting statements to such a short length is inexplicable. One is only left to conclude that the Commission’s leadership sought to create an appearance of consensus on issues as to which none existed. For a body whose ability to influence policy depends entirely on its credibility, this is especially distressing.
In sum, the Carter-Baker Commission’s discussion of the voter ID and paper trails – in addition to its handling of dissenting viewpoints – leaves much to be desired. Rather than adopting an evidence-based inquiry, the Commission has provided a sop to the left and a sop to the right. That’s a shame, given that there are other proposals in the report that have much more to recommend them, foremost among them its proposal for nonpartisan election administration. It would be unfortunate if such worthy recommendations were lost, due to the absence of rigorous analysis evident in the Commission’s treatment of hot-button issues like voter ID and paper trails, as well as its handling of dissenting views. These aspects of the Commission’s report can only diminish the credibility – and thus the ultimate impact – of its more considered recommendations.