Posted: September 14, 2010
Top-10 List and Top-Two Candidates: Some Thoughts on This Election Season
As our nation heads into the final stretches of this campaign season - today ends the primaries for this year - Election Law @ Moritz has refurbished its web site. The goal is to provide the same type of substantive content that we have in the past, but to do so in a more user-friendly format. We hope that you will poke around and let us know what you think.
We have also used the occasion to reflect on what we as a team think likely will be the biggest legal issues of this election cycle. Here’s our “top ten” list. Look forward also to Weekly Comments (of which this is the season’s opener), as well as a new Weekly Summary (found as one of the links at the top of the home page), between now and Election Day.
It is not surprising that campaign finance and redistricting are the first two items on this “top ten” list. It is widely believed that campaign finance will be the big story of this year’s election in the wake of the U.S. Supreme Court’s decision in Citizens United v. Federal Election Commission. And hardly a day goes by without a new story, or blog post, on the effect that this year’s elections will have on the legislative districts drawn after the completion of the new census. (New maps, of course, will be needed in each state for both the state’s own legislature as well as its delegation to the U.S. House of Representatives.)
But as I think more about what this year’s elections will tell us about the health of the procedures we use to elect government officials, I begin to wonder whether our biggest concern should focus on a different element of the electoral process: the means by which we winnow the candidates down from a field to a single winner.
To be sure, I still believe we need to pay close attention to how we cast and count our ballots. Indeed, many items of our “top ten” list reflect this continuing concern. This year is likely to see even greater use of early and absentee voting, and this trend may cause problems similar to those that Minnesota confronted in 2008. And although it is true that many issues arising with the implementation of the Help America Vote Act (HAVA) have been sorted out by now, there are still potential pitfalls lurking within the registration database and provisional voting requirements of this adolescent (and thus not yet fully mature) statute.
Even so, one can be hopeful that this November’s general election will avoid any major dispute over the casting and counting of ballots. The recounts that have occurred during the primaries, like those in Alaska and Vermont, have gone smoothly—at least from the perspective of not triggering litigation. (While Vermont’s recount among Democratic gubernatorial candidates was a friendly affair, Alaska’s recount among Republican candidates for U.S. Senate was politically testy.) Recounts in general elections tend to be uglier than recounts in primaries, for the obvious reason that candidates from different parties are usually more antagonistic than candidates in intramural competition, and thus we cannot assume that there will be no major recount battle in this year’s general election. But it takes the unusual—and unfortunate—combination of both an exceedingly narrow apparent margin of victory (less than 1000 votes) and a problem in the administration of the voting process susceptible of litigation (which in turn requires some uncertainty in the relevant legal rules) in order for a recount to become a full-scale confrontation of the kind that occurred in Minnesota. If the nation is lucky, it will escape that kind of “perfect storm” in this electoral season.
The health of an electoral system, however, is not measured solely by its ability to accurately tally the votes cast for each candidate. Instead, it is also necessary that the winning candidates represent the electoral preferences of the voters who cast ballots. This essential condition cannot be assumed to occur if, for example, the winning candidate in a three-way race received less than forty percent of the votes. If the runner-up would have prevailed in a head-to-head runoff with this plurality winner, then democracy is arguably disserved by the failure of the electoral system to take account of this preference within the electorate. Imagine a race in which the candidates split the vote 35-33-32, with a politically moderate candidate in the middle of this split. It should not matter whether the plurality winner is a left-wing or right-wing “extremist” to make the same point: if moderates together with the plurality winner’s ideological opponents would have combined to defeat this “extremist” candidate, then the electoral system should give voters the opportunity to have this preference prevail.
I readily acknowledge that there is no perfect system for translating voter preferences into electoral outcomes. Kenneth Arrow proved as much in the theorem that bears his name. Philosophers of democracy can endlessly debate the merits of proportional representation against the winner-take-all system that we have in the U.S. for legislative elections. There is growing interest in Instant Runoff Voting (IRV), as evidenced most recently by Rick Pildes’s advocacy on behalf of it. But there are different versions of IRV, as well as some competitors to (and criticisms of) it.
This piece is not the place to argue for any particular electoral system. Instead, I will simply observe that this year may end up exposing flaws with our conventional method of moving from primary to general elections, without the opportunity for runoffs in most places. The rise of the Tea Party movement and the ferment within Republican Party primaries show that it is not easy to design a sensible system for moving from many candidates to a single winner. Should separate Tea Party candidates appear on the general election ballot, potentially draining significant support for Republicans, who thereby might lose even though they would win a head-to-head matchup against their Democratic counterparts? Witness the allegations we are seeing concerning “fraudulent” Tea Party (and Green Party) candidates, who purportedly are placed on the ballot solely for this siphoning purpose.
This ferment has also caused the phenomenon of “establishment Republicans” pursuing independent candidacies in general elections. Charlie Crist, Florida’s governor elected as a Republican, is running for U.S. Senate as an independent because it became clear that he would lose the Republican primary to the conservative insurgent Mark Rubio. In Alaska’s U.S. Senate election, incumbent Lisa Murkowski did lose in the Republican primary to another conservative insurgent, Joe Miller, and apparently she is contemplating the possibility of a write-in candidacy as an independent.
It is easy to imagine that Florida’s U.S. Senate race might end up like the hypothetical three-way split described above. Rubio might get less than 40%, perhaps as low as 35%, of the vote and still be the plurality winner. Crist might trail Rubio by just a percentage point—35% to 34%, for example—with the Democratic candidate (Kendrick Meek) in third place. If Crist would beat Rubio in a head-to-head runoff, because Democratic voters would hand Crist their support in a two-man contest, should not Floridians be given the opportunity to express this collective preference?
Likewise, it is possible that Murkowski would beat Miller in a two-person race in which all Alaskans voted, rather than just those Alaskans participating in the Republican primary. Thus, it is intriguing to consider what the U.S. Senate election in Alaska would have looked like if Alaska had adopted the same kind of “top two primary” system that they have in Washington and now California. A “top two primary” essentially converts the general election into a head-to-head runoff between the top two vote-getters in the primary, regardless of party. In the primary elections that Alaska held this year, Miller and Murkowski each received over 50,000 votes, whereas the winner of the Democratic primary received only 18,000 votes. Indeed, the total of votes cast for all four candidates in the Democratic primary was only 36,000. Had these numbers been the same in a “top two primary” system, or even if the three other candidates in the Democratic primary had bowed out in order not to drain votes from the leading Democrat, the result would have been that both Miller and Murkowski would have gone on to the general election. Arguably, that situation would have permitted Alaska’s electorate collectively to express its preference among Miller, Murkowski, and the leading Democrat much better than the current situation in which Murkowski must fight for status as a write-in candidate.
California’s new “top two” system does not take effect until the 2012 elections. It will be interesting to observe whether that new system, in contrast to the state’s former conventional system, causes winning candidates to better reflect California’s electorate as a whole. Nor will California’s experience, even if positive, prove that its “top two” system is better than some version of Instant Runoff Voting—or even, what is possible, a combination of an IRV primary (to identify the “top two” vote-getters) and then a head-to-head general election. But because IRV has not yet been adopted on a statewide basis in any state, whereas “top two” now has, those who are interested in improving the health of democracy in the U.S. should consider focusing on “top two” as a politically feasible measure of reform.
Even if the “top two primary” is worthy of attention in the future, is the issue of how to winnow candidates from many to one really more important this year than either campaign finance or redistricting? I certainly do not want to discount the importance of these two topics. Yet there are significant distinctions between the how-to-winnow issue, on the one hand, and either campaign finance or redistricting, on the other. The how-to-winnow issue concerns the voting process itself: is democracy functioning to produce electoral results that properly aggregate voter preferences among the pool of competing candidates?
The campaign finance issue is different, at least as it is being debated this year. The fear of increased corporate and union money after Citizens United is not that it will prevent voters from casting ballots in favor of the candidates whom they most prefer; rather it is a fear that these candidates, once in office, will become beholden to their financial supporters. The concern, then, is that governance, rather than electoral preference-identification, will be distorted.
The concern about redistricting is related to electoral-preference identification. The fear is that gerrymandered districts will cause the election of candidates who do not best reflect the preferences of the electorate, considered collectively. But this fear concerns the candidates who will be elected pursuant to the new legislative districts drawn next year after the completion of this year’s census. Even though redistricting is on the brain of all political professionals, there is no current claim that the redistricting issue is causing a distortion of electoral preference-identification this year—beyond whatever distortion occurred as a result of gerrymandering after the 2000 census. In other words, the fact that this year’s election results will determine who gets to draw the new maps next year is not, by itself, a fact that is distorting the operation of the electoral process this year. The voters will vote their preferences this year, and the electoral system to the extent that it is able will aggregate those preferences into electoral results, and this process will occur without being affected by the fact that this year’s winners will engage in redistricting next year.
By contrast, the absence of a “top two” system—or IRV, or some other better method of aggregating voter preferences—in this year’s election necessarily will have an effect of the ability of democracy to do its basic job this year. Because of the conventional system in place, we may get plurality winners who would not win a head-to-head runoff with the runners-up they beat. This risk is not limited to Florida and Alaska. Indeed, some of the dynamics I have been discussing may play out in today’s primaries, particularly Delaware’s or New Hampshire’s. Moreover, the frequency with which this problem occurs may increase as a result of the political turmoil the nation is experiencing. Even if there is no perfect solution to this problem, it is one about which we should be concerned—and increasingly so.
Thus, let’s keep a sharp eye on the winnowing issue this year, and see if it bears out that we need to put in place new ways - whether “top two,” IRV, a combination, or something else - to pick a winner from a fractured field of candidates.
Edward B. Foley is Director of the Election Law @ Moritz program. His primary area of current research concerns the resolution of disputed elections. Having published several law journal articles on this topic, he is currently writing a book on the history of disputed elections in the United States. He is also serving as Reporter for the American Law Institute's new Election Law project. Professor Foley's "Free & Fair" is a collection of his writings that he has penned for Election Law @ Moritz. View Complete Profile