About this Blog
About The Author
Email Me

RealClearPolitics HorseRaceBlog

By Jay Cost

« The Puerto Rico Wild Card | HorseRaceBlog Home Page | On Obama's Speech »

What Will Happen with Florida?

It looks like the Florida re-vote plan is not going to happen. This is from the Miami Herald.

Florida law prohibits election officials from authenticating votes cast in the Democratic Party's proposed do-over primary by mail, state officials said Thursday, a potentially fatal blow to the increasingly embattled plan.

''There's no authority under Florida law that would allow county supervisors of election or the state to verify signatures in an election of a state party,'' said Sterling Ivey, a spokesman for Florida's secretary of state and Division of Elections.

Verifying the identity of anyone who votes by mail -- either through a conventional absentee ballot or in the Florida Democrats' proposed and unprecedented statewide mailed election -- is considered a key bulwark against electoral fraud.

In addition, the plan floated Wednesday by state party chief Karen Thurman lacked a key requirement to protect the anonymity of voters -- an inner ''secrecy envelope,'' though aides said Thursday that the envelope would be included if the proposal gains momentum.

So, if there will be no vote-by-mail, what will there be? My sense is that the most likely result is no action at all.

The question at hand is whether the status quo - which is no re-vote - will be retained, or whether there will be a switch to some alternative. There are, by my count, four actors who have a say in whether there will be a switch: the Florida Democratic Party, the Democratic National Committee, and (by virtue of DNC mandate) the Obama campaign and the Clinton campaign. Of course, the lines between these four groups can get blurry. For instance, members of Congress could be affiliated with a campaign as well as a party organization. Also, other entities are involved in setting the rules of the game - e.g. the governor decided that a re-vote will not come on the government's nickel. But you get the idea. These four entities are the major players.

Each of them has different interests or goals and, in true Madisonian fashion, all of them are entitled to a veto. Nobody is going to get railroaded here. Either these four entities agree to a solution, or the status quo wins.

This complicates a re-vote.

When different entities have closely related goals, a compromise can often be found. For instance, if this were simply up to the DNC and the Florida Democratic Party, a compromise would probably happen. After all, both groups are principally interested in maximizing participation.

It is also possible to find a compromise when actors have different goals that do not necessarily conflict. This is akin to killing two birds with one stone. For instance, if this were up to the DNC, the Florida Democratic Party, and the Clinton campaign - a compromise might be found. The Clinton campaign wants to maximize its share of votes/delegates, which is quite different than what the DNC or the Florida Democratic Party wants. However, neither side really opposes the other. So, there might be some proposal that accomplishes both goals at once.

However, it is hard to reach a compromise in which actors have mutually exclusive goals and both are empowered to check the other. This is like giving the Road Runner a veto over Wile E. Coyote, and vice-versa. The cartoon would consist of the two of them just sitting there.

That is pretty much where the Democrats are. Consider the following.

Let's assume that the goal of the Clinton campaign is to maximize its share of the delegates without diminishing its share of the popular vote. It anticipates that the Florida delegation will not be seated as it is. It also anticipates that it will have to make an argument to the super delegates, and that these delegates will be at least somewhat persuaded by her popular vote victory in January. So, while it wants to net delegates coming out of Florida, it does not want to do so at the expense of the 295,000-vote margin it won in January.*

Accordingly, the Clinton campaign's decision rule for any proposed re-vote would be:

SUPPORT the re-vote if:
(Expected Total Votes Cast in Re-vote) X [(Clinton's Expected Share of Vote) - (Obama's Expected Share of the Vote)] >= 295,000 votes,

OPPOSE the re-vote if:

(Expected Total Votes Cast in Re-vote) X [(Clinton's Expected Share of Vote) - (Obama's Expected Share of the Vote)] < 295,000 votes,

An example will enliven this. Suppose that 4 million Democrats receive mail-in ballots, and the Clinton campaign expects that 3 million voters will mail ballots back. It also expects to beat Obama 55 to 45 in the vote count. Thus, it expects to net: (3 million votes) X (55% - 45%) = 3 million votes X 10% = 300,000 votes. Accordingly, the Clinton campaign would support the re-vote.

However, Obama would surely oppose it. Why? It helps Clinton! She nets 5,000 votes and at least a few pledged delegates. Plus, the re-vote eliminates the vote count that favors him the most (i.e. excluding FL and MI). It's a loser for him. Generally, any plan that is good for her is probably bad for him, and vice-versa. The two are engaged in a zero sum game. If the probability of Clinton winning the nomination increases, the probability of Obama winning necessarily decreases. This makes a compromise unlikely.

Unlikely - but not impossible. Remember that to delineate this basic logic, we had to make some assumptions. They might not hold. As they fail to do so - it becomes possible for a re-vote to happen. I see three potential ways that we might see a re-vote.

First, it might happen if both sides think they will come out better. Obviously, both of them cannot. Obama's gain is Clinton's loss. Clinton's gain is Obama's loss. However, our above logic assumes that Clinton and Obama have the same expectations. They might not. They might have different estimates of expected turnout as well as different estimates of how each will fare. So, Clinton might think it will favor her. Obama might think it will favor him.

This is not implausible. Predicting election results is tricky - especially primary elections where you do not have the preferences of 85% of the electorate "anchored" by partisanship. Each side might honestly expect that it will benefit from a re-vote. In that situation, both could support it.*

Second, it might happen if one side or the other becomes desperate. Our model assumes that both candidates see themselves as having some chance of victory without the re-vote. If they see themselves as standing no chance - they might be willing to embrace the re-vote even if they expect to do badly.

Remember that expectations play an important role in this kind of interaction. Both sides have to make assessments about what will happen in the re-vote. So, they will have an expectation for turnout and for each candidate's share, but they will also have a range of values in which each number could fall (from best-case to worst-case scenarios). If one side were to assess that it would assuredly lose the nomination under the status quo (or at least that a loss is very likely) - it might agree to a plan under which it expects to be harmed, but under which there is a small chance of great success. In football, they call this the Hail Mary.

Third, social pressure might be a factor. Our above model assumes a kind of "closed system." That is, nobody pays a price for acting in their own interests. This might not be the case - especially in a sensitive situation like this.

There are two types of interests at stake. The party organizations have an interest in openness and/or fairness. The candidates have an interest in maximizing their own electoral advantage. However, they cannot be honest about their interests. Neither candidate can be seen to be "disenfranchising" voters for the sake of electoral victory. That would be a public relations nightmare. So, each must somehow argue that a proposal is not free and fair, even if all that matters to it is victory.

This means that a candidate might be forced to accept a proposal that would leave it less well off. If the proposal seems to everybody to be an open and fair re-vote, how could the candidate squash it without seeming ruthlessly ambitious? That might be what is moving things forward in Michigan. One candidate or the other is going to be better or worse off from a Michigan re-vote. The one that will be worse off (assuming that he or she expects to be worse off) will not want the re-vote. However, if the deal seems to create a free and fair election - the expected loser will have little wiggle-room.

I think this last exception is the most plausible of the three. Maybe somebody will come up with a proposal that creates an obviously open and fair vote. Of course, that has not yet happened - and the fact that it hasn't happened despite the attention of national and state Democrats to the situation does not augur well. Perhaps a successful compromise in Michigan will put sufficient pressure on Florida to hammer out an agreement.


[*] Matters are may be more complicated than this. We have taken these assumptions to illustrate a simple decision rule, which we shall do presently. The assumptions might not hold in "minor" ways, i.e. ways that make the decision rule more complex, but essentially the same. For instance, the Clinton campaign might accept a lower vote margin in Florida if it knew it could net some pledged delegates via a re-vote. This would make her computation more complicated than what we have outlined - but the same basic pattern would hold (her win is his loss; so, whatever she will support he will oppose). The assumptions might not hold in "major" ways, i.e. ways that might actually foster a compromise solution. These are discussed below.

[*] Note that the disagreement might not involve the actual primary results, but what those results imply down the line. Both candidates might predict exactly the same result in the primary itself - but that the primary will have different implications on the broader nomination battle. For instance, Obama might expect to lose a Florida re-vote, but will ultimately have his hand strengthened because it will take the issue from Clinton, who also expects to be stronger by virtue of the primary. In that situation, there is a disagreement about broader expectations that makes a compromise possible.

-Jay Cost