Election reform, the Voting Rights Act, the Help America Vote Act, and related topics -- with special attention to the voting rights of people of color, non-English proficient citizens, and people with disabilities
Dan Tokaji's Blog
- Election Law Blog (Rick Hasen)
- Election Updates (Michael Alvarez & Thad Hall)
- Votelaw Blog (Ed Still)
- Leave it to the Lower Courts: On Judicial Intervention in Election Administration, 68 Ohio State Law Journal 1065 (2007)
- The New Vote Denial: Where Election Reform Meets the Voting Rights Act, 57 South Carolina Law Review 689 (2006)
- Early Returns on Election Reform: Discretion, Disenfranchisement, and the Help America Vote Act, 73 George Washington Law Review 1206 (2005)
Friday, June 2
Back to Ohio: The Rolling Stone Piece
Robert F. Kennedy, Jr. has published this article in Rolling Stone, entitled "Was the 2004 Election Stolen?" It's a long article but, for those anxious to get to the bottom line, his answer to the title's question is an emphatic "yes."
Kennedy's treatment of his subject differs from most "stolen election" arguments we've heard since 2004, in that it's much more comprehensively researched -- including over 200 footnotes that would make a law professor proud. While there's not a whole lot of new information here, Kennedy does a nice job of explaining and cataloguing the numerous problems that did in fact occur in Ohio's 2004 election. For reasons explained below, I don't think he makes a persuasive case that the election was "stolen" (i.e., that Kerry really won). The article is nevertheless useful in exposing how shoddy election administration practices can result in lost votes, and how some recently enacted laws will make things worse rather than better.
Let me start with the aspects of Kennedy's article that I liked. The article extensively describes the various impediments to voting that occurred in Ohio's 2004 election. And there were a lot of problems in that election, some probably driven by the desire to suppress votes and others by plain incompetence. To summarize some of the most significant problems that Kennedy describes:
- As in most other states, Ohio's chief election official is elected on a partisan basis. This inevitably creates the risk that the official will discharge his or her duties in a partisan fashion, and there's evidence to support the conclusion that Secretary of State Ken Blackwell did just that in 2004. Foremost among the decisions that might reasonably be questioned are his order not to count provisional ballots cast in the wrong precinct, and the order (later rescinded) to require that registration forms be on 80-lb. paper weight.
- In the weeks preceding the 2004 election, the Republican Party mounted an extensive effort to challenge voters' eligibility. According to Kennedy, the GOP sent letters to over 200,000 newly registered voters, and later sought to have more than 35,000 voters taken off registration lists when the letters were returned, a practice known as "caging." These pre-election challenges were ultimately stopped by a federal district court, properly so in my opinion.
- The punch-card voting systems that were used by over 70% of voters in Ohio's 2004 election resulted in the loss of tens of thousands of votes. Overall, there were around 95,000 votes on which no vote for President was recorded, the majority of those cast by punch card. Other voters used non-notice optical scan systems, which are also prone to error. Kennedy estimates (based partly on my work) that 66,000 votes were lost due to unreliable voting equipment. This seems to me a reasonable estimate if we include the votes lost due to non-notice punch-card and optical scan systems.
- There were inexcusably lines at many polling places in Ohio, including those in Franklin County and Knox County. Kennedy reports that some voters near Kenyon College had to wait for eleven hours before voting. He also finds that the reported problems with long lines were concentrated in urban areas with high minority populations, and the mostly anecdotal evidence that I've seen suggests that this is probably accurate. A DNC-sponsored survey found that 3% of voters, which would be over 174,000, left the polls without casting ballots due to long lines. (Interestingly, the percentage who left and didn't come back was about the same 3% for both black and white voters.) While it's difficult to measure with any degree of precision how many votes were lost due to long lines, this undoubtedly was a serious problem that discouraged some people from voting.
A Grain of Salt
There are other aspects of Kennedy's report that would be very disturbing if true, but appear to rest on somewhat weaker evidence. For example:
- Kennedy describes a group of Republican operatives known as the "Mighty Texas Strike Force" which allegedly "us[ed] pay phones to make intimidating calls to likely voters." Kennedy's source for this allegation is a report produced Democrats on the House Judiciary Committee in January 2005, entitled "Preserving Democracy: What Went Wrong in Ohio," also known as the Conyers Report, which quotes a statement made by an unidentified hotel worker. While this allegation is what we lawyers would call hearsay -- actually, it's triple-hearsay, since the Conyers Report was relying on a statement made at a hearing by someone other than the hotel worker -- if true it's obviously very troubling.
- The Greater Cleveland Voter Registration Coalition (GVREC) found that about 16,000 voters in that area were disenfranchised by data-entry errors on the part of election officials in Cuyahoga County, and another 15,000 votes were denied due to trivial omissions on registration cards. If extrapolated to the whole state, this would mean 72,000 or so votes denied. There undoubtedly were data-entry errors, but as GVREC candidly acknowledged, this was a "ballpark" estimate. I think it's on the high side, since those voters wrongly omitted from registration lists should still have been permitted to cast provisional ballots. And as summarized here, there were only 158,642 provisional ballots cast, of which 123,548 (78%) were counted -- meaning that just over 35,000 people cast provisional ballots that didn't end up being counted. Now, it's possible that some voters were wrongly denied a provisional ballot, but those should be captured in the 3% survey figure noted above.
To summarize: There's no doubt that some votes were lost due to faulty voting technology (let's say 66,000). There were also likely some voters whose provisional ballots weren't counted due to registration errors (the most that this could possibly be is 35,000). Add to that voters who report leaving the polls without casting a ballot (174,000, generously estimated), and you've still only got a total of 275,000 lost votes-- and keep in mind that this is indulging very optimistic assumptions in Kerry's favor. If my algebra is right, Kerry would still have had to pick up about 196,500 of those lost votes, or over 71%, in order to tie Bush, given the 118,000 margin of victory. Perhaps not beyond the realm of possibility, but not very likely.
Hard to Swallow
This brings me to the aspects of Kennedy's report of which I'm most dubious. To conclude that the election was probably "stolen," Kennedy has to rely on the discrepancy between exit polls and the reported results. I've previously expressed my doubts about this line of reasoning here, but Kennedy's report relies on something new: a forthcoming book by Steve Freeman & Joel Bleifuss, entitled Was the 2004 Presidential Election Stolen? Exit Polls, Election Fraud, and the Official Count. (According to Kennedy's footnote, this book is coming out next month and it's listed on Amazon.com.)
While it's hard to evaluate a claim based on a source that isn't yet available, I remain very skeptical of the argument that exit poll discrepancy prove widespread election fraud. As an initial matter, it's worth noting that Mystery Pollster Mark Blumenthal, a reliable source on matters of exit polling, has argued that Freeman's prior work exaggerated the significance of the "errors" within Ohio and the other states in which such discrepancies existed. Blumenthal has collected his prior posts on the exit polls surrounding the 2004 election, all of which are enlightening for those interested in learning more about the subject, here.
The biggest problem with relying on exit polls to "prove" widespread election fraud is that it there's no plausible explanation of how such fraud could actually occur. The discrepancies existed not just in one but in many counties, each of which is administered by its own bipartisan board of elections, as Kennedy notes. These counties used a variety of different types of voting equipment, including punch card, optical scan, and direct record electronic systems manufactured and sold by different vendors. To believe that the exit polls prove election fraud, you have to believe that a group of people somehow managed to orchestrate the manipulation of results in not just one but in many counties, all of which run their own elections. Moreover, because the exit poll discrepancy existed in 30 states, according to Kennedy, you have to believe that there was widespread ballot stuffing and/or ballot snuffing across state lines, in hundreds of counties using various different types of equipment, many of which have chief election officials who are either nonpartisan, bipartisan, or Democrats. No plausible explanation has been offered as to how one could pull off such an extensive conspiracy -- and it is a conspiracy theory -- without detection.
Even if we take Kennedy's characterization of Freeman & Bleifuss' finding at face value, it doesn't eliminate the most plausible explanation for the reported result/exit poll discrepancy, which Blumenthal has discussed here and here, and which I noted here. That's the possibility that Bush voters actually avoided those conducting the exit polls, or even lied about whom they voted for, given that those voters might have assumed that a "college student approaching with a clipboard" was a Kerry supporter, as Blumenthal puts it. Kennedy's basis for rejecting this possibility is that: "In Bush strongholds, Freeman and the other researchers found that fifty-six percent of voters completed the exit survey -- compared to only fifty-three percent in Kerry strongholds" (citing Freeman & Bleifuss' forthcoming book). Even if this is true, however, it doesn't tell us who those voters completing the exit survey in Bush strongholds were -- or more precisely, whom they supported. More specifically, it doesn't eliminate the possibility that Kerry voters in Bush strongholds were more likely to speak with interviewers than were Bush voters in those strongholds.
Kennedy also rests on the fact that in twelve sparsely populated counties, a Democratic candidate for chief justice of the state supreme court, Ellen Connally, drew more votes than Kerry. Following the Conyers report, he questions how a "gay-friendly black judge roundly rejected in Democratic precincts" could have gotten more votes than Kerry in those counties. Sounds like a hard question ... until you consider that many voters probably didn't know that she was gay-friendly, black, or even a Democrat. Although I'm an Ohio voter, and was following the 2004 election pretty closely, I must confess that I didn't know she was "gay-friendly" or black until reading Kennedy's article. These down-ballot contests simply don't receive a lot of media attention (which is one of the big problems with electing judges in the first place, but that's a different subject). I did know that Connally was a Democrat, but other voters might not even have known that. Although judicial candidates in Ohio are typically endorsed by the parties, the offices are nominally nonpartisan and, under Ohio law, "[n]o name or designation of any political party" is supposed to appear by judicial candidates' names on the ballot. So inferring election fraud in 12 counties based on Connally's vote total is, in my view, quite a stretch.
What to Make of All This?
After reading Kennedy's account, my overall reaction is similar to that which he quotes Senator John Kerry as having:
Can I draw a conclusion that [the Republicans] played tough games and clearly had an intent to reduce the level of our vote? Yes, absolutely. Can I tell you to a certainty that it made the difference in the election? I can't. There's no way to do that. If I could have done that, then obviously I would have found some legal recourse.As I said shortly after the November 2004 election, and still believe today, I don't think that Kerry won Ohio's election. And as I've said in a law review article that Kennedy cites at several points in his story, I think that Kerry made the right call by not contesting the result in court. The 118,601 votes by which Bush took Ohio in 2004 is several orders of magnitude greater than the 537 votes by which he took Florida in 2000. It's perfectly reasonable to believe that the latter result was erroneous, but it strains credulity to believe the same about the former result.
The most important question we now face, however, is not whether Kerry really won. It is instead what ought to be done about the very real and serious problems that emerged in Ohio and other states in 2004, which Kennedy exhaustively documents, for the most part quite accurately.
Unfortunately, some of the measures being implemented in the states would do little or nothing to improve elections. One of them is the so-called "voter verifiable paper trail" (VVPAT) that's part of the "Count Every Vote Act" that Kennedy mentions. If there's anything that should be clear from Ohio's 2004 election, in which the substantial majority voted with punch card and optical scan ballots, it's that paper doesn't guarantee election integrity or public confidence in the results. Whatever type of system is used for voting, the key to election integrity is trustworthy procedures -- maybe less sexy that the VVPAT, but likely to be much more effective. Unfortunately, there were many aspects of Ohio's 2004 election, including the recount, in which such procedures don't appear to have been observed. That can only damage public confidence in the results.
Even more damaging are the measures being proposed in some states to impede access to the vote. Kennedy properly notes that the restrictive voter ID laws recently enacted in Georgia and Indiana are tailored to suppress voter turnout. Unfortunately, there's a well-orchestrated campaign to pass these vote suppression measures in other states, which I've previously described here. The State of Missouri's legislature recently enacted a similar law, and others seem likely to follow, absent a groundswell of opposition.
It's perfectly reasonable to look back at the 2004 election and try to learn from the mistakes that were made. I'm therefore not one who believes that those distressed by the 2004 result should simply "get over it." But arguing over who really won shouldn't distract us from issues like voter ID and felon disenfranchisement, where the present fight over vote suppression really lies.