2008

The Need for a Structurally Nonpartisan Tribunal

by

It is difficult to imagine a better advertisement for the experimental McCain v. Obama simulation than yesterday’s en banc ruling in Ohio Republican Party v. Brunner. The experiment tests whether election litigation might be better handled by a court specially structured to be inherently bipartisan: one Democrat, one Republican, and a third judge mutually chosen by the first two. As others have already noted about yesterday’s en banc ruling, its 9-6 split was virtually (although not entirely) along party lines. All the judges in the majority but one were appointed by Republican presidents; all the judges in dissent but one were appointed by Democratic presidents (and the lone Republican-appointed dissenter was originally nominated by a Democrat).

One need not doubt the conscientiousness or integrity of any of these Article III federal judges to be troubled by this 9-6 divided ruling in an “election eve” case. I am sure that all of these judges were striving to find the true “right answer” to the legal issues before them, in accordance with the model of the ideal judge (Justice Hercules) famously put forward by the legal philosopher Ronald Dworkin. The problem is that it is not so easy to identify objective “right answers” in lightning-speed emergency litigation involving the propriety of a Temporary Restraining Order (TRO) under the so-called “balance of the equities” test. When the topic to be addressed in this inauspicious circumstance concerns the procedures for the casting and counting of ballots in a presidential election—and Republicans and Democrats are lined up on opposite sides of the case (the Republican Party as the lead plaintiff and the Democratic Party having intervened to support the Secretary of State, who is an elected Democrat)—one cannot help but wonder whether the party background of these Article III judges inadvertently affected how they weighed the equities.

To be sure, the majority opinion commendably strives to find “common ground” and puts its reasoning forth in simple, straightforward language that suggests tentativeness about much of its analysis. But ultimately the majority’s decision depends upon its best guess, in the absence of hard evidence, of the feasibility of implementing a change in administrative procedure this close to the start of the ballot-counting process—and its assessment of how necessary a procedural change is in order to make the ballot-counting process “fair” to all voters. The main dissent accepts the Secretary of State’s contention that a court-ordered administrative adjustment might trigger unintended adverse consequences on the voting process; and, in any event, the dissent sees the matter of electoral fairness very differently from the majority. In the face of this kind of disagreement, it is a conjecture that the public (or at least the segment of the public that pays attention to “election eve” judicial decrees) might have more confidence that a court responsible for supervising the administration of the voting process tilts neither Democratic nor Republican, but instead decides issues straight down the middle, if the composition of the court were structured so that it always has an equal number of Democratic and Republican judges, with a tie-breaking member chosen mutually by both sides.

The problem of partisanship appearing to affect tribunals authorized to adjudicate disputes over the voting process is nothing new in our nation’s history. Yesterday, while the Sixth Circuit was releasing its en banc opinion, I was delivering a lecture on how partisanship affected the adjudication of the dispute over the counting of ballots in New York’s gubernatorial election of 1792. The thesis of the lecture was that understanding how partisanship affected that dispute, and the consequences of its doing so, remains important to us today. In brief, the lecture showed that the Founding Fathers had failed to anticipate how disputes over the counting of ballots for Chief Executive could become mired in polarized two-party politics and therefore failed to include in the constitutional system they designed an institution capable of handling this kind of dispute in a way consistent with their own constitutional values. The Founding Fathers themselves suffered from their oversight in the election of 1792—John Jay and his supporters thought he was robbed the governorship by a partisan ruling, and they almost resorted to their revolutionary “first principles” to take their victory back—and, as a nation, we continue to suffer from this oversight, since we have never added the structurally nonpartisan institution necessary to fulfill the Founding Fathers’ vision.

The simultaneous 9-6 split of the en banc appeals court is further confirmation of the point the lecture was then making: our nation continues to lack an institution capable of resolving disputes over the casting and counting of presidential ballots in a way that the institution appears to be nonpartisan and thus fair to both campaigns that so energetically seek a presidential victory. As the lecture observed, James Kent (who participated in the 1792 dispute) and Joseph Story warned that the Framers’ oversight on this issue left the nation vulnerable. Let us hope that the experimental McCain v. Obama simulation begins a process of identifying a way to redress this institutional deficiency.

Purges & Provisional Ballots

by

This morning’s front-page New York Times story on purges is, I believe, somewhat inaccurate in one potentially significant respect.  After describing apparently unlawful purging practices in several states, the article says: “Some states allow such voters to cast provisional ballots.”

My understanding of HAVA–and I believe the prevailing understanding in the legal community (of those who focus on HAVA-related issues)–is that HAVA requires every state to let all individuals cast a provisional ballot if they are willing to affirm that they believe themselves to be registered voters, as presumably would occur if there an unlawful or otherwise erroneous purge.  The relevant provision is 42 U.S.C. 15482(a).  (This requirement is inapplicable to states with Election Day Registration or without any form of voter registration, but the purging problem discussed in the article does not arise in those states.)  Thus, any voter wrongly purged should be able to cast a provisional ballot, and it would be a qualitatively different problem on Nov. 4 if any voter who thinks he or she has been wrongly purged is denied a provisional ballot.  That’s what occurred in 2000, and that’s what HAVA’s provisional voting mandate was designed to avoid.

Of course, there is a separate question of whether the provisional ballots get counted, as the New York Times article notes.  But given the predicate assumption that a voter was purged from a state’s voter registration rolls in a manner that violates federal law, either HAVA or NVRA, it would seem that there is a very strong argument HAVA would require the counting of that provisional ballot (unless, perhaps, the review of the provisional ballot turned up some different basis for concluding that the voter was ineligible to vote–for example, being under age 18).  HAVA specifically requires states to count provisional ballots if a review of those ballots shows the provisional voters to be “eligible under State law”.  42 U.S.C. 15482(a)(4).  I’ve written previously on whether or not “eligible” in this sentence means something different from “registered.”  See Foley, The Promise and Problems of Provisional Voting, 73 Geo. Wash. L. Rev. 1193 (2005).  But regardless of the resolution of that issue, it would seem that if the basis for purging a voter from the state’s registration list (and thus deeming the voter not “eligible” to vote) is a basis that violates federal law, then the states would be preempted in this instance from considering the voter not registered under state law–and thus would be required by HAVA to count the provisional ballot.

Contrary to the implication of the New York Times article (“But [provisional ballots] are often not counted because they require added verification.”), I don’t think it would be tenable under HAVA in this unlawful purging situation to disqualify a provisional ballot for failure to comply with some extra identification requirement applicable to provisional voters (who cast provisional ballots for reasons unrelated to the unlawful purges).  If the voter should have been deemed registered (since, by hypothesis, the purge was unlawful), the only ID requirements that could be imposed at the polling place were the ones applicable to all registered voters, who would then cast regular ballots.  The unlawfully purged voter should have been considered a registered voter, and thus should be treated as such for purposes of counting the provisional ballot if that voter complied with the ID requirements of a registered voter.  If a review of the provisional ballot shows that the voter should have been deemed registered (because the purge was unlawful), and the voter should have received a regular ballot (but for the unlawful purge), then the provisional ballot should be counted as if it were the regular ballot that the voter should have received.  Perhaps a more simple way to put the point is this: if what took the voter off the roll was a violation of federal law, then the voter should be treated as if he or she were on the rolls–and HAVA does not permit disqualifying a provisional ballot of a registered voter unless the individual was not entitled to register in the first place, being under 18 for example.

To be sure, it would be better not to have unlawful purges in the first place.  (I have no independent knowledge of what the New York Times article describes as unlawful purging is occurring, or is a violation of federal law.  I’m assuming the truth of those points for purposes of this analysis.)  For one thing, as I and others have observed, a large percentage of provisional ballots complicates the identification of the winner of a close election.  And if it does happen that a large number of provisional ballots are cast because of what the New York Times describes, then close attention will need to be paid on how officials in those states conduct their process of reviewing those provisional ballots.  But the assumption should not be that unlawful purges will lead to voter disenfranchisement.  In fact, if HAVA works as intended, the assumption should be just the opposite: provisional voting will protect voters from disenfranchisement despite unlawful purges.

Thus, whatever they read or hear about problems that state officials may be having in maintaining their voter registration lists, citizens who think themselves registered should go to the polls.  If poll workers tell them they are not on the list of registered voters, then they should cast provisional ballots–they should not leave without doing so, and no poll worker should deny them this right.  Then, if it turns out they were wrongly removed from the list of registered voters, then HAVA should do its job of making sure that the erroneous purge does not prevent their ballots from being counted as valid votes.

The First Bush v. Gore — and the Next: New York’s Election of 1792 & Its Continuing Relevance

by

Over the next two weeks, I will be engaged in two related endeavors that examine disputed elections from opposite temporal perspectives. One week from today (Tuesday, October 14), I will deliver a lecture on the history of disputed elections. Then, the following Monday, I will be in D.C. to observe the oral argument in McCain v. Obama, which involves a hypothetical dispute over ballots cast in the very near future.

Bush v. Gore in an historical mirror

The historical lecture will go all the way back to our nation’s very beginning, focusing on the 1792 gubernatorial election in New York between George Clinton, the incumbent Democratic-Republican, and John Jay, the Federalist challenger (who was serving at the time as the first Chief Justice of the United States). The dispute over this election truly was our nation’s first Bush v. Gore. The parallels are striking. The two candidates “lawyered up” then, as they did in 2000. On both sides, there were organized efforts to recruit legal luminaries to press their partisan position. As one Jay biographer has written, “every lawyer” in New York City “rummaged in his books for legal arguments.”

So if someone says that the battle over ballots in Florida eight years ago was unprecedented for the army of lawyers each side employed, don’t you believe it. While the lawyers then had no CNN for pitching their legal positions to the public, one of Jay’s legal lieutenants reported that the other side was “stuffing news papers with dissertations” on the dispute, but not to worry because a “refutation” from their “own side” would “appear in tomorrow’s paper.” Indeed, a multitude of pamphlets and broadsides was written to support each candidate, and the news printers were happy to circulate them to an aroused citizenry.

Aaron Burr, in defending Clinton’s position, was the James Baker of his day, while Rufus King (like Burr, a U.S. Senator from New York at the time) was Jay’s Warren Christopher, or, perhaps more apt, Ron Klain (based on the prominence of his role for Gore, as recently portrayed in HBO’s Recount movie). Moreover, Burr and King made essentially the same arguments as their modern-day counterparts. Burr, along with the other lawyers he recruited to Clinton’s side, including Edmund Randolph (then serving as the first U.S. Attorney General), made the argument that election statutes must be strictly followed in order to protect the integrity of elections. King and the rest of Jay’s legal team, which included a prominent former judge in Philadelphia, argued that technical breaches of election statutes should not be permitted to disenfranchise innocent voters. When King wrote that “election law is intended to render effectual the constitutional right of suffrage” and “should therefore be construed liberally” so that “the means should be in subordination to that end,” Gore would have been happy to have those words in his own legal briefs.

It is not merely an historical curiosity that 1792 foreshadows 2000. Rather, I suggest that every high-stakes disputed election has the potential for sparking the basic debate between the Burr-Baker strict constructionist position and the King-Klain voter-protection view. Each side, regardless of its prior ideological commitments, will take whichever of these jurisprudential positions will support a victory in the particular election at hand.

In the 200-plus years since 1792, our constitutional democracy has not advanced beyond the simple repetition of this basic jurisprudential debate. The reason is that we lack an institution capable of developing an impartial body of precedent on how to resolve disputed elections. The specific contest in 1792 ended when the state’s Canvassing Committee, whose decisions by statute were final and unreviewable, ruled in favor of Clinton by a partisan 7-4 vote. The result was unsatisfactory then, precisely because it was so partisan.

In 2000, the Supreme Court, perhaps fearing the partisanship that would beset a congressional resolution of the presidential election, took it upon itself to attempt a resolution according to its interpretation of the Constitution. But the 5-4 decision caused the Court, rightly or wrongly, to be accused of its own partisanship. It wasn’t able to maintain itself as institutionally “above the fray,” because it wasn’t institutionally designed to handle the uniquely polarizing case of two presidential candidates each claiming a legal entitlement to a declaration of electoral victory.

I believe there are valuable historical lessons in the New York election of 1792 on why we still lack the institution necessary for this category of cases. Those lessons, I conjecture, concern the almost immediate development of two-party politics despite the desire of our Constitution’s Framers to avoid that development. But I’ll leave those points for Tuesday’s lecture.

Testing the Potential Future of Bush v. Gore

What is most important, looking forward, is how ready our system of constitutional democracy is to handle the next Bush v. Gore, whenever it may come. It is unlikely to happen this year, for the simple reason that the odds are against it happening in any given year. But come again it will, at some point in nation’s future. There is a story to tell about how our system progressed (if one can call it that) from 1792 to 1876, when the dispute over the Hayes-Tilden presidential election necessitated a special commission—just as the story progressed (or at least moved through time) from 1876 to 2000. My Moritz colleague Steve Huefner and I are in the process of writing a book-length chronicle of major disputed elections in U.S. history—a chronicle that has some interesting tales along the way, of which 1792 is the only the starting point. But whenever that book goes to press, there will be new chapters to be written in subsequent editions, and what will they say about the improvement of our society’s ability to resolve disputed elections?

One specific question for the future is, what effect will the legal precedent of Bush v. Gore have in the next dispute? The McCain v. Obama simulated adjudication is designed, in part, to shed some light on that question. (In my last comment I recounted the genesis and goals of McCain v. Obama, in which hypothetically the outcome of this year’s election turns on whether or not to count provisional ballots cast in Colorado as a result of a severe storm that triggers an extension of polling hours in Denver.) Even now, I can tell that the indeterminacy of the Equal Protection principle articulated in Bush v. Gore poses a challenge to any court confronted with a future disputed election that requires consideration of that precedent. In my role as “Acting Clerk of Court” for the McCain v. Obama simulation, I have prepared a “Neutral Memorandum of Law” for the benefit of the participating attorneys and jurists. Undertaking that task has caused me to realize, even more than I did when I responded to Dan Lowenstein’s critique of my initial effort to classify potential Equal Protection claims based on Bush v. Gore, just how fluid that classification must be until it gets pinned down by future precedents. (It’s hard to know conceptually how to treat the storm-induced provisional ballots under Bush v. Gore for reasons I endeavor to explain, much less to decide definitely which side should win under that precedent.)

But if McCain v. Obama truly has powerful arguments on both sides, as my analysis in the Neutral Memorandum of Law proffers for the consideration of the counsel and Court, then the openness of the law places all the more of a premium on the need for an impartial tribunal. In McCain v. Obama, the attorneys can appeal to the same “first principles” of election law—strict enforcement to protect the integrity of elections, or flexible interpretation to avoid voter disenfranchisement—as their predecessors did in Bush v. Gore, and before them their predecessors in the disputed election of 1792. Without a structurally impartial tribunal, the decision of which side’s “first principles” should prevail is likely to seem just as arbitrary, and thus unsatisfactory, as the previous decisions did to many contemporaneous observers. What is more, there can be no advancement from one dispute to the next; each time, the two sides are condemned to repeat the same warring first principles, with another apparently arbitrary ruling by a body perceived by many to be partial to one side.

The avoidance of this dispiriting repetition is the hope of the institutional experiment underlying the McCain v. Obama simulation. If the institution that adjudicates the dispute is structurally impartial, and perceived as such, then its decision has a chance of being accepted as principled. Moreover, that principled decision can serve as precedent in the next dispute. It might be that even a structurally impartial tribunal will not always side with the same “first principle”: strict enforcement to protect the integrity of elections, or flexible interpretation to avoid voter disenfranchisement. The particular facts of each dispute might cause the impartial tribunal to lean towards one of these two competing “first principles” in some cases, while tipping towards the other when the facts are different. But if the tribunal remains truly impartial over time, it will be able to explain why it leans one way sometimes but not others. By this process, a principled corpus of precedent concerning disputed elections potentially could develop, and ultimately the future of disputed elections might look very different from the past.

But McCain v. Obama is an experiment, and as with any genuine experiment we cannot predict how it will turn out. So, over the next few weeks, I hope that we can learn from the future as well as the past.

McCain v. Obama, a U.S. Supreme Court hypothetical (one hopes)

by

Election Law @ Moritz, with co-sponsors AEI-Brookings Election Reform Project and the Supreme Court Institute at Georgetown’s law school, will conduct simulated adjudication of a hypothetical case involving a dispute over the outcome of this year’s presidential election. This experiment will test an idea proposed in recent scholarship:  that the public more likely would perceive that courts are fair and impartial in resolving lawsuits over the outcome of high-stakes elections if the courts that adjudicate these cases are specifically structured to be strictly bipartisan.

In May, I posted on our Election Law @ Moritz website a hypothetical I gave my Election Law students as part of their final exam. Now, thanks to the co-sponsorship of the AEI-Brookings Election Reform Project and the Supreme Court Institute at Georgetown’s law school, we are going to have an actual adjudication of this hypothetical case (or, more precisely, a revised version of the hypothetical). We are conducting this simulated adjudication to test an idea that I have proposed in my recent scholarship: that the public more likely would perceive that courts are fair and impartial in resolving lawsuits over the outcome of high-stakes elections if the courts that adjudicate these cases are specifically structured to be strictly bipartisan.

The panel of retired jurists that has agreed to adjudicate the hypothetical is truly stellar and follows the model that I set forth in my proposal: two judges, each identifiable as affiliated with one of the two major political parties while simultaneously having an impeccable reputation for judicial integrity, with a third judge selected by the mutual agreement of the first two. (This method of selecting the third judge makes the overall panel intrinsically neutral.) Our three-judge panel consists of Thomas Phillips, the former Chief Justice of the Texas Supreme Court, and Patricia Wald, the former Chief Judge of the U.S. Court of Appeals for the D.C. Circuit, with David Levi, former Chief Judge of the United States District Court for the Eastern District of California (and now Dean of Duke Law School), chosen by Justice Phillips and Judge Wald to be the third and presiding judge. I think all would agree that one could not ask for a better three-judge panel to simulate how a highly respected, inherently bipartisan court would handle a high-stakes lawsuit over the outcome of a major election.

We are equally blessed that two advocates of the highest reputation, with extensive experience in U.S. Supreme Court litigation, have agreed to argue this case before our panel. Glen Nager, who chairs the Issues & Appeals group at Jones Day, will present McCain’s side of this hypothetical case. Walter Dellinger, who similarly chairs O’Melveny’s appellate practice, will present Obama’s side of the case. As we explain in the stipulated Statement of the Case, neither advocate has been retained by the actual presidential campaigns for purposes of participating in this simulated adjudication, but both will zealously advocate their side of the case in accordance with professional norms, as if they really were representing their hypothetical clients.

The oral argument of this case will occur at 10:00 a.m., on Monday, October 20, at Georgetown’s Hart Auditorium in McDougal Hall (600 New Jersey Ave., N.W., Washington, D.C. 20001). It’s open to the public (no admission fee), and anyone who would like to attend is welcome to do so. (All information about this project, including a webcast of the oral argument for those unable to attend, will be collected at this web page.) It should be interesting “political theater,” as well as an informative “social science experiment” of the proposed inherently bipartisan election court. We ask only that individuals planning to attend please RSVP to Tina Drake Zimmerman, draket@law.georgetown.edu, so that we can estimate the size of the audience.

For all those not yet familiar with the facts of this hypothetical (or who have understandably forgotten about it since May), the essence is that the outcome of this year’s presidential election will turn on disputed provisional ballots from Colorado. These ballots were cast following an extension of polling hours in Denver due to an unusually severe winter storm on the afternoon of Election Day. The dispute concerns whether or not to count these ballots under applicable provisions of state and federal law.

It is not farfetched to think that Colorado will be “the swing state” this year, as pollster Stuart Rothenberg predicted last week. Nor is it farfetched that bad weather might cause an extension in polling hours, as happened during the primaries this year in Ohio and Maryland. Likewise, the partisan affiliations of Denver election officials, the Colorado Secretary of State, the justices of the Colorado Supreme Court, and the Governor, as well as the current partisan makeup of the state legislature’s two houses — not to mention the content of the relevant provisions of the Colorado election code — all collectively conspire to make realistic a partisan battle over whether to count provisional ballots, were a storm-related extension of polling hours to occur. As I mentioned in May, however, this scenario is only one of many that one could concoct for various potential swing states, including Ohio, if it turns out that somewhere besides Colorado were to become the “Florida” of 2008.

One noteworthy feature of our McCain v. Obama hypothetical case is that it contains Equal Protection and Article II issues parallel to those in Bush v. Gore. As election experts well know, and most other members of the public will remember, the Equal Protection and Article II issues in Bush v. Gore arose in the context of the rules for determining whether a “dimpled” or “punctured” or “hanging” chad would count as a vote. In the Colorado hypothetical, there are no chads of any kind, but eliminating them does not eliminate the possibility of Equal Protection and Article II issues comparable to those in — and left unresolved by — Bush v. Gore. On the contrary, provisional ballots cast as a result of extending polling hours in storm-suffering Denver, but not elsewhere in Colorado, would trigger new and vexing Equal Protection and Article II issues.

Thus, for anyone with lingering doubts about whether the U.S. Supreme Court handled Bush v. Gore as best as it possibly could have (and apparently Justice O’Connor herself has expressed such doubts), our McCain v. Obama hypothetical offers a way to test the idea of a bipartisan election court, not just generally, but also specifically in the context of a case exactly parallel to Bush v. Gore: another disputed outcome in a presidential election, where the federal constitutional questions presented to the U.S. Supreme Court are essentially equivalent.

Will our distinguished bipartisan three-judge panel be unanimous or divided on how to handle these federal constitutional questions? If divided, will their disagreement be less rhetorically divisive than the impression left by the totality of opinion-writing in Bush v. Gore? More fundamentally, even if the panel splits 2-1, will the fact that the third and presiding judge was chosen by the other two cause the public to perceive that the resolution of the dispute was as fair as possible under the circumstances — and fairer than either if a conventional court would have decided it, or if the dispute had been left to Congress rather than a court to resolve? What is wonderful about this project’s being a genuine experiment is that we don’t know the answers to these questions in advance, and thus we truly expect to learn something valuable as a result of our three-judge panel’s actual deliberations over this hypothetical case.

Our three-judge panel is planning to issue its decision in McCain v. Obama within 10 days of the oral argument, so that we all will have it before Election Day itself. An additional benefit of this pre-Election Day release relates to another aspect of the scholarship that underlies this project. Although I have suggested the enactment of legislation to give special bipartisan courts jurisdiction over lawsuits challenging the results of high-profile elections, I have also recognized that any such legislation is not likely soon, and certainly not before this year’s election. Consequently, I have proposed that the private sector create a “shadow court,” or what I now prefer to call an “Amicus Court” (since it sounds friendlier), to deliberate in advance of a real court’s decision in any high-stakes election dispute that would benefit from an inherently bipartisan tribunal. The Amicus Court’s deliberations then could be submitted to the real court in the form of an amicus brief, with all the persuasive force that its reasoning would have by virtue of the Amicus Court’s distinguished and inherently bipartisan membership.

Our McCain v. Obama simulated adjudication can be viewed as an early-stage potential “test drive” of the Amicus Court concept. To be clear, our three-judge panel for the McCain v. Obama hypothetical is adjudicating the case as if it has the full jurisdiction and authority of the U.S. Supreme Court in that case. (In other words, it is the U.S. Supreme Court for purposes of that single case.) It is not thinking of its role as a private-sector advisory body that would submit its deliberations to the actual nine-member Supreme Court that sits on First Street, N.E., in Washington, D.C. Still, the decision that it releases before Election Day may prove relevant to the Amicus Court concept in the following way. Suppose that, despite all hope to the contrary, after Election Day litigation really does occur over the outcome of this year’s presidential election, either in Colorado or elsewhere. Then, our three-judge panel’s decision in the hypothetical McCain v. Obama may serve as a form of “non-binding, but persuasive authority” for the real court confronted with the real case. Even if the facts and issues in the real post-Election Day McCain v. Obama (or Obama v. McCain) don’t exactly match the hypothetical that we have devised, our distinguished three-judge panel’s treatment of the issues in the hypothetical McCain v. Obama may be analogous enough, and sufficiently powerful in its deliberative reasoning, that the real court will wish to take advantage of its wisdom. After all, a precedent from one state supreme court (or one federal appeals court) need not be “on all fours” with the case pending before another court for that precedent to be relevant and valuable, even though it is obviously not binding. In the same way, the actual decision in our hypothetical McCain v. Obama may be influential in the resolution of any real dispute that arises after Election Day. If so, and insofar as this influence comes from the fact that our three-judge panel in McCain v. Obama is inherently bipartisan in its composition, comprising retired jurists with exceptionally high reputations of judicial integrity, then this influence will show how the Amicus Court similarly might work.

In short, I must say that I find it incredibly exciting and gratifying that scholarly ideas can be “field tested” in this way. Whether as a model of an official election court enacted into law as a government body, or as a prototype of the private-sector Amicus Court, the simulated adjudication of McCain v. Obama will be an experiment of the core idea that an inherently bipartisan tribunal can help improve public perception concerning the resolution of election disputes. It is not often that a scholar’s work has the opportunity to be put in practice in this way.

Of course, I’m indebted to many other individuals for helping to develop this scholarship. I will be thanking them publicly in connection with the materials and program we are developing for the October 20 oral argument. Here, it is worth reflecting that the idea of bipartisan election courts originated in the research that my Moritz colleagues and I undertook for From Registration to Recounts, was initially presented in a paper presented at a Stanford Law School symposium, and then crystallized in its “shadow” or “Amicus” court form in another paper presented to a Tobin Project workshop. The idea has been refined and “sharpened” considerably as it has progressed to the current stage of preparing for our three-judge panel’s adjudication of McCain v. Obama. It will be interesting to see what happens to the idea in the aftermath of this adjudication.

McCain Likely to Win Ballot Request Fight

by

By Nathan Cemenska

Web Editor, Election Law @ Moritz
Moritz College of Law

On September 17, Republicans sued Ohio Secretary of State Jennifer Brunner after she issued a memo to local boards of election instructing them to reject allegedly flawed absentee ballot applications that had been designed, printed, and issued to voters by the McCain-Palin campaign (Ohio law does not require such applications to be in any particular form, as long as they contain certain enumerated statements). Two days later, Republicans sued Brunner again, in an action that appears, thus far, substantially identical to the first. According to Brunner, the applications were flawed because voters failed to place a check mark in a square that was printed beside the following statement on the application form: “I am a qualified elector and would like to receive an Absentee Ballot for the November 4, 2008 General Election.” Brunner argued that the applications of voters who had forgotten to place a check in this square could not be honored because Ohio law requires applicants to indicate on the application form that they are qualified electors. Because it is pretty clear from the form that voters are supposed to check the square, Brunner has a point.

For their part, Republicans in their complaints cull together various bits of the Ohio election code and claim that Brunner’s actions are contrary to law. They point out that there is no law that explicitly says that a check box on a ballot application has to be checked to make the application effective (however, it is worth noting that there also is no law that explicitly says that failure to check such a box will not make such an application ineffective). Republicans also point out in their complaint that Brunner has ordered local boards of election to overlook the failure of voters to place a check in boxes that appear on Brunner’s recommended absentee ballot application form. That is true, but it is a different form, and is susceptible to being interpreted differently. Particularly, the check boxes on Brunner’s form do not appear next to the qualified voter statement, but instead appear next to “blanks” where voters are supposed to fill in their social security numbers or other identification information. If the voter fills in these numbers, the voter has complied with the relevant statute by providing the required information– checking or not checking the square next to these statements does not really add or subtract anything. In contrast, there are no “blanks” for the voter to fill in information on the disputed portion of the McCain form, only a square next to the qualified voter statement. Voters who check the square have clearly indicated they are qualified, but it is less clear with voters who fail to check the box and nevertheless send in the form.

Despite efforts on the part of both parties to make this case seem very complicated, the basic standard for deciding it is really quite simple: To be effective, absentee ballot applications must contain a “statement that the person requesting the ballots is a qualified elector.” R.C. 3509.03. Do the allegedly flawed applications contain such a statement? As with a Rorschach test, the answer to this question depends on whom you ask. On the one hand, the forms do have printed on them the words “I am a qualified elector….” On the other hand, there is a square next to those words, one that the voter was clearly supposed to mark. Some voters placed a mark in that square, and others did not. Is the emptiness in the square enough to take the words “I am a qualified elector” and change their meaning so that they now mean the voter is not a qualified elector, or that the voter is not taking a position on this issue either way? Does anyone actually think there is a “right” answer to this question?

Despite the ambiguity, the equities fall so heavily in favor of the Republicans that they alone are enough to decide the case. Nobody is seriously contending that the voters who failed to check the rectangles are not qualified. Furthermore, by filling out and mailing in application forms that had printed on them the words “I am a qualified elector,” the voters clearly intended to communicate that they were qualified. And for Brunner to claim that she did not get the message makes her look more than a little bit obtuse. This cannot do anything except generate prejudice against her in the hearts of the Supreme Court justices.

There are other issues in the case, including the precarious question of whether mandamus relief is even available, an area in which the Court has seemed to draw some pretty arbitrary distinctions in elections cases as of late (see State ex rel. Summit Cty. Republican Party Executive v. Brunner, 118 Ohio St.3d 515 (Ohio, 2008); State ex. rel Mackey v. Blackwell, 106 Ohio St.3d 261 (Ohio, 2005)). However, people who get too caught up in the “legal analysis” of a case like this run the risk of analyzing mere shadows on the wall. The real issue is that, to a lot of people, Brunner’s order is going to seem unfair. Furthermore, chances are, at least four of those people sit on the Ohio Supreme Court.

EL@M Publishes Major New Research

by

By Nathan Cemenska

Web Editor, Election Law @ Moritz
Moritz College of Law

The 2000 presidential contest was a disaster and, since then, disturbing election problems have occurred in Ohio, Colorado, Florida and elsewhere. Will the 2008 presidential election be similarly flawed, or will it go smoothly? To help answer this question, Election Law @ Moritz researched dozens of election law issues over 17 states, categorized the approach of each state, and plugged them into a series of interactive maps and charts.

The project features an interactive web chart and database that allows the user to perform side-by-side comparisons of 44 critical legal questions across 17 key states. The project also allows the user to access the same information in a more visual way be viewing 44 interactive US maps that are color coded to faciliate analysis and discussion. Finally, the project includes an interactive map that allows the user to see our analysis of the states most likely to experience election-related litigation prior to November. All of these maps and charts allow the user to see more detailed factual and legal research, including citations, by clicking through to the heart of our database. The executive summary of the project may be found here.

The states were chosen based on our analysis of whether they might be critical to the result of the presidential election. We divided the issues up into ten groups: Institutional arrangements, voter registration/statewide database, challenges to voter eligibility, provisional voting, early and absentee voting, voting technology, polling place operations, ballot security, emergency preparedness, and post-election processes (counting, recounting and post-election lawsuits).

Here are the key conclusions of our research in each of these areas:

  • Institutional arrangements. Most states could do more to improve the quality and extensiveness of the training required of local election administrators. Many states should also look at the entity that is responsible for counting provisional ballots and consider whether that entity should not include some more representation from the minority political party. The party affiliations of local election officials are easy to determine in some states, and difficult to determine in others. On average, more election administration problems seem to have occurred in states where it is easy to identify political affiliation.
  • Voter registration/statewide database. The federal Help America Vote Act asks that states make some attempt to match the personal information contained on incoming voter registration applications against information contained in external government databases. However, the language of HAVA is vague, and in most cases there is no black-letter state law to clarify the rules. In the absence of any clear guidance, administrations have simply created their own rules, most of which are not contained in any publicly available document. Fortunately, the current practice in many states allows most voters to cast a regular ballot at the polls, even if officials were unable to match their information.
  • Provisional voting. There is no one standard for determining whether a provisional ballot counts, but a host of standards, each of which only applies in certain situations. The most generic standard applies when the voter’s name is not on the precinct roster, in which case the ballot will generally count if the voter is registered and eligible. This standard does not vary much across states. However, there is a great degree of interstate variance on the issue of whether officials should count provisional ballots that are cast in the wrong precinct, and also on the issue of whether officials should count provisional ballots when the voter failed to present ID at the polls.
  • Voting technology. Most of the 17 states that will be important to the 2008 presidential result use optical scan technology predominately or exclusively. However, Colorado, Indiana, North Carolina, Nevada, Ohio, Pennsylvania and Virginia use significant numbers of touchscreen voting machines. Colorado, Indiana, Pennsylvania and Virginia do not require touchscreens to be equipped with a paper trail.
  • Polling place operations. The required amount of poll worker training varies widely and, in some states, it is not required at all. States could do more to make their training programs more frequent and more robust.
  • Ballot security. Various states require different forms of ID at the polls, but that is not the real issue. The real issue is the legal consequence of failure to present ID. We found that only Colorado, Missouri and Indiana require physical ID of all voters, force voters without ID to cast a provisional ballot, and refuse to count provisional ballots unless the voter later returns with some form of ID. In the rest of the states, most voters without ID should nevertheless be able to cast a ballot that will count without post-voting follow-up from the voter.
  • Emergency preparedness. Most states have little or no law telling officials what to do if there is a violent natural disaster or other emergency that interferes with the conduct of an election. While this does not mean that officials are not prepared for such occurrences– they are free to develop their own voluntary preparations– it does mean that courts would have a very difficult time trying to decide a lawsuit by a candidate who claimed he or she would have won the election but for a disaster that interfered with voting. States should do more to develop clearer rules in this area, to avoid putting a court in the position of having to make up its own rules on an ad hoc basis. Many states also have little or no policy telling officials or the courts what to do in case of a ballot shortage or touchscreen malfunction that interrupts voting.
  • Post-election processes. The deadlines for approving official results and other post-election processes in many states are late enough that there is a risk of bumping up against the federal safeharbor deadline that protects states from having their choice of president disregarded. In 2000, the US Supreme Court used this risk as a justification for ending further attempts at obtaining an accurate vote count in Florida. However, there is not necessarily any reason for concern in this area, because officials do not generally take all of the time they are legally permitted to approve official results. The true danger is that a few states are still relying on a vague “intent of the voter” standard to determine whether a ballot with nonstandard markings should count. This is similar to the standard that caused the delays and confusion in 2000.

The executive summary includes a more detailed discussion of these and other issues. We hope that you will find it interesting, accurate and, most of all, useful. We also intend to provide further updates to the maps and charts in the future, to ensure that this project remains useful.

We would like to thank the dozens of state and local election administrators and other elections experts across the country who provided important information that was necessary to complete this project. Please let us know what you think of our project or whether there is any way you see that we can improve upon our work (you can send an email to me at cemenska.1@osu.edu). We would also like to thank the JEHT Foundation, which funded this project. Finally, we would like to thank the numerous people from within Election Law @ Moritz and the Moritz community who helped conceive of, obtain funding for, and execute this project.

Social Initiatives on State Ballots

by

By Dale A. Oesterle

It is time to assess the effect of social initiatives on the November Presidential election. In 2004, eleven states had same-sex marriage bans on the ballot and many observers believed that they helped turn out the vote for incumbent President Bush in critical states.

The situation is cloudier this year.

First, the number of ballot initiatives is down significantly. There were 204 initiatives in 2006, a non-presidential year, and there are currently only 108 measures certified for state ballots. At most another ten or twenty that are in the certification process will be added. Moreover, a number of ballot issues are sponsored by Christian groups to affect John McCain, not Barak Obama. The groups want to force McCain to declare on social agenda questions on which McCain has otherwise been “weak.”

But a deeper look into the issues shows the same old story. The critical swing states are Indiana, Ohio, Michigan, Virginia, Colorado, New Mexico, Nevada, Missouri, Florida, and New Hampshire. The other states show clear leads for one candidate or the other. So what is happening in the swing states?

In several of the biggest swing states, well-publicized ballot issues would appear to energize people to vote who will also vote for McCain. To the extent that ballot issues get people to vote who otherwise might not vote, the higher turnout would seem to favor McCain.

Florida voters will face a marriage definition amendment, defining marriage as solely between one man and one woman. McCain is a clear supporter of the amendment. Colorado voters will face an amendment banning affirmative action, a “right to work” (anti-union) provision, and an amendment defining life as “beginning from the moment of fertilization.” McCain will support the first two but has yet to declare on the latter. Ohio has a measure to require companies with more than fifty workers to provide seven paid days of sick leave to employees. McCain will oppose the measure as will the Ohio business community. Missouri votes will decide on whether to declare English an official state language. McCain, despite, his ambivalence on the issue, may still benefit here.

In short, in the swing states, social ballot issues may still be very important in bringing out voters that will tend to support one candidate or the other. At present, I suspect that McCain will be the beneficiary. Of course, given Obama’s dramatic appeal to new voters, the young, and minorities, the ballot-issue related turnout advantage for McCain may just be a drop in the bucket.

Do We Need a Box for “Abstain”?

by

By Dale A. Oesterle

I offer another “no chance” academic recommendation: Add an “abstain” box to the ballot in all elections, for all candidates, and on all ballot issues.  The number of “abstain” votes in any given election would provide a real choice for voters and also provide real information for political parties, election statisticians, and pundits.  Furthermore, I would relish the chance to vote formally for “abstain” on the Presidential line on the ballot. I suspect that I am not alone. I would vote. I could be heard. And it would be the vote I want recorded. I do not want the absence of a recorded mark to be my vote. 

The just-completed election for seats on the Yahoo board of directors featured a contest between “yes” votes and “abstain” votes on the Yahoo proxy forms. There was no alternative candidate on the proxy or any other proxy and there was no box for a “no” vote. Notably, the CEO of Yahoo, who had successfully blocked an acquisition by Microsoft, collected over a thirty percent abstain vote from all shareholders voting. This is a remarkably high number and a statement by a substantial number of shareholders that they were displeased with the CEO’s performance. In uncontested elections the normal percentage of abstain votes is less than three percent.

The abstain vote was all the more startling because it came on the heels of a settlement of a dissident shareholder proxy contest. Carl Icahn, one of our longest-engaged and best known activist hedge fund operators, had fielded a slate of candidates for the board and was soliciting votes. A few weeks before the election, Icahn withdrew his challenge and accepted the management’s agreement to nominate him and two delegates to Yahoo board seats. Shareholders who would have voted for Icahn’s dissident slate were left only with “abstain” votes for incumbent insiders who had blocked the Microsoft deal.

The Yahoo dissident shareholders were upset that the Yahoo CEO and board had not accepted a Microsoft acquisition offer that was, in value, considerably higher than the trading price of Yahoo stock, unaffected by rumors of the merger.

Those who ridicule a “just say no” campaign with only an “abstain” vote note that one yes vote outweighs any number of “abstain” votes. One check in the yes box on one proxy and 999 checks in abstain boxes on all other proxies puts an incumbent back on the board for a year. They overlook the power of the poll on stock prices (all abstain voters are potential sellers of the stock), and on the next annual election (when a hedge fund may be encouraged to run a dissident slate).

Faced with an election for President of the United States in November I am in the uncomfortable position of not feeling able to vote for any candidate running when I show up to vote. And I will show up; voting is important to me. I can simply leave the President line (or box) blank and vote for others. But this is an admission of defeat; I want to vote. Or I can attempt a write-in, which will test my patience and the patience of the poll workers who must handle my ballot specially.

I would relish the chance to vote formally for “abstain” on the Presidential line on the ballot. I suspect that I am not alone. I would vote. I could be heard. And it would be the vote I want recorded. I do not want the absence of a recorded mark to be my vote.

The number of “abstain” votes in any given election would provide a real choice for voters and also provide real information for political parties, election statisticians, and pundits. So I offer another “no chance” academic recommendation: Add an “abstain” box to the ballot in all elections, for all candidates, and on all ballot issues.

Bipartisanship in Election Administration: Is There Enough?

by

By Nathan Cemenska

Web Editor, Election Law @ Moritz
Moritz College of Law

A fundamental assumption underlying the setup of many of our elections institutions is that we cannot necessarily trust administrators to do a good job in one-party environments. Rather, it is better to require that at least one important player in each entity be from each of the largest political parties. This reduces the risk of hanky-panky, and also reduces the risk of mistakes made by people who try to do their best, but who might fail because they do not appreciate the viewpoint of the other side. I do not think we should automatically conclude that every level of election administration must be permeated with bipartisanship, and even think that requiring this could in many cases lead to deadlock, partisan grandstanding, and endless, fruitless debate. Nevertheless, requiring bipartisanship in our elections institutions is a good rule of thumb, and one that we should not deviate from without having specific reasons. Moritz has completed some research on 16 states that suggests that America is following this rule of thumb in some areas, but not in others.

One area where operations are sufficiently bipartisan is the composition of poll workers. Fourteen of the 16 states we researched required at least one member of the minority party to serve in each polling place (one of the states, Oregon, does not use poll workers because it votes entirely by mail). The outlier, Pennsylvania, actually elects its poll workers, and does nothing to ensure that there is representation from both parties (except in the case of filling vacancies by appointment). Pennsylvania should consider reforming its system, but even the other states might want to examine what they are doing to see whether there is a way to improve. One question to consider is whether it is important that poll workers be evenly bipartisan (as is required in New Hampshire), or whether it is sufficient to have just one or two representatives from the minority party. Another question to consider is whether officials who appoint poll workers are actually complying with the requirements of bipartisanship. This can be a problem in jurisdictions where members of the minority party are rare, or in jurisdictions where the requirement simply is not taken seriously. Sometimes officials need to be freed up to recruit younger workers or workers from out of the jurisdiction in order to meet these important requirements. Reforms that made it easier to comply with requirements of poll worker bipartisanship played an important role in ending the massive ballot-box stuffing operations that used to occur in Chicago prior to the early 1980’s.

One area that is potentially, though not necessarily, concerning is local administrators. Only four of the states we researched (Indiana, Ohio, Pennsylvania, and North Carolina) had bipartisan boards administering elections in every local jurisdiction, although two states had this type of board in some jurisdictions but not others (Wisconsin and Missouri). In the rest of the states—Washington, Oregon, Nevada, Colorado, New Mexico, Iowa, Minnesota, Michigan, Florida, and New Hampshire—there is a single individual, elected or appointed, with chief authority over elections at the local level. While it is reckless to suggest that local administrations cannot be trusted just because they are not inherently bipartisan, making them inherently bipartisan would do a lot to quiet those who like to make such reckless suggestions (it also might do some real good). On the other hand, reformers only have so much political capital and this might not be the most cost-effective area in which to spend it. That is particularly true because local administration is an area where it actually might be beneficial to have a single individual in charge who can get things done without having to persuade everyone and reach a consensus. It may turn out that, rather than this being an area where there is a “right” answer, it is one in which there is a trade-off and the decision depends on what one values more: Insurance against iniquity, or insurance against inaction.

An area for real concern is counting of ballots. Unlike the fast-paced decision-making that is sometimes necessary in the actual planning and execution of elections, counting should be a purely ministerial task in which the participants have no real discretion. It is also an area where, despite our eagerness for quick election results, we can afford to go more slowly without offending any substantive rights. For these reasons, there really is no downside to requiring some bipartisanship in the counting process. Despite this, most states leave the counting up to an entity—either the local election official or a counting board—that does not have to have even a single member of the minority party. Washington, Nevada, New Mexico, Minnesota, Iowa, Michigan and Florida count ballots using a canvassing board of unconstrained partisanship, while Oregon and New Hampshire allow the counting to occur under the direction of a single individual. Election officials have admitted to Moritz in the past that it would be possible to manipulate the result of an election by making “mistakes” in aggregating precinct results, something that is more likely when no one from the minority party is required to be present during the process. The situation is even worse in the case of provisional ballots, which are often counted by an entity other than the one that counts regular ballots. Washington, Oregon, Nevada, Colorado, New Mexico, Wisconsin and Michigan count provisionals under the direction of a local election authority that is generally of unconstrained partisanship, while Florida counts them under the direction of a separate, but unconstrained, county canvassing board. This is a problem because the standards for counting provisional ballots in many states are not exactly clear, and it is less likely that they will be applied fairly when representatives of only one party are present during counting. The concern is mitigated by the fact that the laws of most of these states explicitly say that counting must be public or that party observers have a right to be present, but the truth is that in many cases this gives false hope because nobody can be trusted to reliably show up and observe.

With the possible exception of the counting of precinct results and provisional ballots, I would not expect reform in any of these areas any time soon. The required composition of poll workers generally seems fair enough (provided the requirement is observed) and local election administrators have a fairly powerful lobby that can probably prevent overhauling local institutions in most states. However, the case against entities of unconstrained partisanship counting ballots is strong, and it would not take much authority away from local election administrators to require that they share some of the responsibility in this area with members of the opposite party. For that reason, those who are interested in reform should focus their efforts in this area.

Regulation of voter registration drives: Will there be further litigation?

by

By Nathan Cemenska

Web Editor, Election Law @ Moritz
Moritz College of Law

The EL@M team is currently determining likely areas for litigation going into the 2008 Presidential election. One potential area appears to be litigation surrounding the regulations that some states apply to third-party voter registration drives performed by entities such as ACORN and the League of Women Voters. Lawsuits attacking such regulations have been filed in the past in Ohio, Florida, Georgia, and Pennsylvania, and we may see more pre-election suits of this kind. Litigation is more likely in swing states with a large number of electoral votes, and in states that have the most restrictive kinds of regulations.

One state to watch is Colorado. Many analysts think it will be key to the 2008 Presidential outcome, both because it is a swing state and because it carries a relatively large number of electoral votes (9). This alone puts it at risk of litigation, and that risk is compounded by Colorado’s somewhat restrictive regulations on voter registration drives. Organizations cannot engage in voter registration drives in Colorado unless the leader of the organization first files with the state and undergoes special training. § 1-2-701. Completed applications generally must be submitted within 15 business days of the date they are completed (within 5 business days if the application was completed within thirty days of the registration deadline). § 1-2-702. The rules prohibit tying a ground-level registration worker’s pay to the number of applications submitted. Violations of these rules can lead to fines of $50 to $1,000.

Although these regulations may be restrictive enough to anger voter registration groups and trigger suit, they are by no means draconian compared to regulations that are currently being litigated in other states. The now-enjoined Ohio provisions, for instance, required all compensated voter registration workers—not just those at the top—to register with the state and complete online training. § 3503.29. They also required that ground-level registration workers submit completed forms directly to the Secretary of State, rather than through the workers’ supervisors. § 3599.11. Like Colorado and many other states, Ohio made it illegal to tie the pay of ground-level workers to the number of registration forms submitted. § 3599.111.

The Colorado regulations are more similar to the regulations in Florida, which spurred an ongoing suit. As in Colorado, each Florida registration drive must register a representative with the state. § 97.0575. However, ground-level workers do not have to register and receive training. Also as in Colorado, the code levies fines against registration groups for failing to return completed applications promptly (Florida requires they be turned in within 10 days of completion). However, the Florida regulations are actually milder than the Colorado ones in some ways. For instance, they limit the total amount of fines levied to $1,000 per year. But the Florida law may be tougher in other ways, as it arguably permits the imposition of fines not only against voter registration drive organizations, but also against the individual workers in those drives.

Fortunately from a litigation standpoint, most of the other states we researched had no regulations or only mild regulations that apply to voter registration drives. The most common requirements are some kind of registration with the state (ex: Missouri) and a prohibition on tying the pay of registration workers to the number of applications submitted (ex: Iowa). Other states, such as Nevada, require those requesting large numbers of applications to submit a plan that describes the activities of their drive. Still others, such as Indiana, have no significant regulations at all. Still, we researched only 16 states, and it is possible that there are regulations in the remaining 34 that are strict enough to create controversy.

If litigation is filed, it will not be concluded prior to the election, but there will be time for courts to issue preliminary injunctions. Unfortunately, prior decisions (I, II) in these types of cases have not produced much guidance, but instead have rested on conclusory statements that the challenged rules either did or did not violate the First Amendment or the constitutional “right to vote” (but see here). For that reason, it is difficult to predict how such cases will go. Suffice it to say that any decisions that come down are not likely to derive from pure legal reasoning, but instead are more likely to derive from “common sense,” whatever that is.