The economics of Hillary withdrawal

Matthew Yglesias seems a bit puzzled that Democratic party elders haven't stepped in to force Hillary Clinton out of the race. He writes that "insofar as it's really true that [Nancy Pelosi] and 'other leading members of Congress' think [don’t think Hillary can win and want her to give up], they need to communicate it more clearly."

By contrast, Marc Ambinder suggests that party elders have intentionally decided not to step in because "in their minds, the racetrack is open and horses, to beat that metaphor to death, are still trotting around" (via Michael Crowley):

John Edwards, Al Gore, Nancy Pelosi, Harry Reid – if these folks came together and threw their weight behind the nominee, Hillary Clinton would probably drop out by the end of the week. But the party elders have in some cases explicitly abstained from making such a determination because in their minds, the racetrack is open and horses, to beat that metaphor to death, are still trotting around.

But there's actually a third possibility -- that most party elders would prefer that Hillary withdraw but don't want to pay the cost of pushing her out of the race. There are two classic economic problems here. The first is that the collective benefits of pushing Hillary out are much larger than the individual benefit to any one party leader (i.e. there's a positive externality). Why would Pelosi or Reid risk becoming a hated figure to millions of Hillary's supporters? As a result, everyone is likely to sit back and hope that someone else will pay the cost of forcing her out.

The second problem is it's difficult to coordinate a joint effort to push her out. In other words, there's a collective action problem. If all the leaders could magically come together to ask her to withdraw, it might be less costly to them individually to push her out, but any effort to make this happen would inevitably leak, generating untold recriminations and infighting. The incentives to defect from such an agreement would also be strong. As a result, no one is likely to chance it.

For both of these reasons, it's likely that the race will go to the convention unless (a) Hillary decides to withdraw on her own or (b) the accumulation of superdelegate commitments after the primaries drives her out.

Update 3/25 10:01 PM: TNR's Noam Scheiber outlines the possible scenarios in a recent piece -- the key quote to me is in bold:

Democrats have never been known for Spock-like rationality, but even they see the logic of avoiding a convention fiasco. "It's in nobody's interest in the Democratic Party for that to happen," says Mike Feldman, another former Gore aide. "There is a mechanism in place--built into the process--to avoid that." That mechanism, such as it is, involves an en masse movement of uncommitted superdelegates to the perceived winner of the primaries. Almost everything you hear from such people suggests this will happen in time. "I think once we have the elected delegate count, things will move fairly quickly, " says Representative Chris Van Hollen, who oversees the party's House campaign committee. Increasingly, there is even agreement on the metric by which a winner would be named. Just about every superdelegate and party operative I spoke with endorsed Nancy Pelosi's recent suggestion that pledged delegates should matter most.

Assuming Feldman and Van Hollen are right, that means Democrats won't wait much past June 3--currently the last day on the primary calendar--before crowning a nominee. At the same time, it means there's very little chance of ending the contest sooner. Undecided superdelegates on Capitol Hill, along with party elders like Pelosi, Gore, and Harry Reid, "don't want to be seen as elites coming in and overturning the will of the people," says one senior House aide. A Senate staffer says his boss "thinks this give and take is natural, it will be helpful in the end." "That's a view held by a majority of these guys who have been through the cut and thrust of politics," he adds. Which means early June it is.

...The most optimistic scenario I could plausibly construct didn't end the campaign until the second week in May. To make it happen, Obama would have to overtake Hillary among superdelegates--a key psychological barrier. He'd have to limit his margin of defeat in Pennsylvania to ten points, then hold serve two weeks later in North Carolina and Indiana, a pair of states he's slightly favored to win. At that point, Hillary would face nearly impossible odds of overtaking him in the delegate race.

Unfortunately for anyone who wants the race to end soon, there are several problems with this scenario. For one thing, even if all this comes to pass, Hillary would still have to bow out voluntarily--an unlikely twist in any event, but highly implausible if the limbo states of Florida and Michigan still offer her hope. Meanwhile, any one of the aforementioned steps could easily fall through. Polls currently show Obama trailing by double digits in Pennsylvania; the good Reverend Wright could make that tough to change. And, though Obama now leads in North Carolina and Indiana, his advantage is either small or, in the latter case, based on a single, flimsy poll. As for superdelegates, as of this writing, the last two out of the closet opted for Hillary.

So, to review: The most optimistic scenario we have relies on a highly tenuous assumption; it's unlikely to happen even if that assumption holds; and, regardless, it allows the Democratic contest to drag on for six more brutal weeks. The dream may never die, but it's seen some better days.

Update 3/26 8:29 PM -- Note how Harry Reid is also dodging responsibility for settling the issue in this interview (via Kevin Drum):

Q: Do you still think the Democratic race can be resolved before the convention?

Reid: Easy.

Q: How is that?

Reid: It will be done.

Q: It just will?

Reid: Yep.

Q: Magically?

Reid: No, it will be done. I had a conversation with Governor Dean (Democratic National Committee Chairman Howard Dean) today. Things are being done.

In this context, "[t]hings are being done" means "we're all trying to get someone else to do the dirty work." I'm not sure we should believe him -- Howard Dean doesn't want to alienate half the party either.

Update 3/27 9:11 AM -- In comments, Jesse Einhorn questions my claim that "The incentives to defect from such an agreement would also be strong," asking "[w]hat possible incentive would there be to defect back to Hillary's camp" from a possible joint effort to push her out given that she is extremely unlikely to win. However, my claim isn't that the leaders would "defect... to Hillary's camp" but that they would defect to a position of neutrality. Expect lots of mumbling about letting the process go forward, etc etc. The reason to do this that, while Hillary is likely to lose, her supporters (financial, activist, etc.) are and will remain powerful within the party. If you don't believe me, ask Nancy Pelosi, who just got an ominous letter from Hillary's top fundraisers:

We have been strong supporters of the DCCC. We therefore urge you to clarify your position on super-delegates and reflect in your comments a more open view to the optional independent actions of each of the delegates at the National Convention in August. We appreciate your activities in support of the Democratic Party and your leadership role in the Party and hope you will be responsive to some of your major enthusiastic supporters.

Comments

[Mister Snitch]

The Clintons know full well they cannot win. But until Obama cuts a deal with them, they will not quit. They tried to force him into a VP commitment, which he laughed off. They are trying other means to pressure him now. However, it is falling apart for them pretty quickly. They may soon capitulate.

For the Clintons, it is not a matter of merely stepping aside. And Reid et all lack the backbone to stand up to them, as even now they are still too powerful.

[JFred]

The "Elders" probably want to see how the Reverend Wright issue plays out, and may be worried about the vulnerability of Obama to a Republican negative campaign. Somebody pointed out that McCain would rather run against Jeremiah Wright than against Bill Clinton, and those are (sort of) the alternatives. At least, from the viewpoint of campaign strategy, those are the alternatives.

If McCain can launch a classically vicious campaign like the Republicans have been doing since the first Reagan campaign, I think Obama is the more vulnerable of the two.

[Gus]

Is there any merit to the thought that Reid and Pelosi aren't elder enough? Al Gore has nothing to lose politically, but could be the hub for this type of coordination...or is this just wishful thinking?

[silverbird]

The incentives to defect from such an agreement would also be strong.

I don't follow you here. Let's say the collective action game is all set up, the selective incentives are in place (cabinet positions, PAC money, the works) and everyone starts making their move. What possible incentive would there be to defect back to Hilary's camp. What could she possibly offer that would outweigh the inevitable sanction from the inevitable candidate. It seems to me this whole model only works if each side has an EQUAL chance of winning. But we all know that isn't true. If fact, almost everyone is certain that Obama will be the eventual nominee. So what's the problem?

What the Democrats need is Dr. Phil, not Mancur Olson.

[Jade7243]

This is bone-headed. There does not need to be any "coordinated" effort to take Clinton out of the race. The super delegates need to just endorse Obama. Do it one by one or two by two, or in any manner they see fit. They don't have to talk to one another, or have joint announcement. They all just need to get onboard with Obama. They can do it quietly or with fanfare. It just doesn't matter. All of this hand-wringing is unnecessary. This is a numbers game, and when the numbers are no longer in Clinton's favor, she'll drop out.

[Steve]

The numbers are not in Clinton's favor and she's pledging that she'll never drop out. The numbers will have to drive a stake through her heart before she'll drop out. We'll get there eventually, but very unlikely it will happen before June.

[BG]

Of your options above, bnasty, b is by far the most likely.

Right now, Obama needs 400 delegates. He probably can't close things out in pledged delegates, but he can get really close. A 50/50 split of the remaining pledged delegates leaves him short 117. It's hard to see Obama doing much worse than that, especially given that he's likely to pick up a handful of extras from the 127 that Hillary has to hold onto at state conventions in Caucus states.

Next, look at add on delegates. These are almost closer to caucus delegates than they are to the pre-selected superdelegates. It will be hard for a state convention in a heavily Obama leaning state to elect an unpledged superdelegate (look at Alabama, the only Obama state to choose its add-on thus far -- Burkhalter was elected having already announced his support for Obama). Out of the remaining 70 add-ons to be chosen, Obama should take at least 30. That leaves him 87 short.

So that's the real math: Obama needs to find one-third of the remaining uncommitted superdelegates to publicly commit to him. The reality is that while party leaders strategic calculations will probably lead them to stay on the fence, rank and file superdelegates face almost an opposite outlook. They will have increasing pressure from their constituents to choose. Chatter on the ground is getting really bad. The potential for Hillary to punish a State Senator from Missouri is really low. The potential for Obama to reward that same state senator is moderate. The potential for that State Senator's colleagues to blame him or her for allowing the party to go up in flames by sitting on the fence will continue to grow for three months.

Barring any huge changes in the landscape (big wins by either candidate on May 6?), the last possible date for this to be decided is June 21, and there's a good chance it could take exactly that long.

Mark your calendar for the big Democratic Unity rally in Detroit or Miami on June 22.

[Rita]

Super delegates easily end since the hold the key to the outcome. When they turn that key is up to them. Honoring this Clinton induced embargo against delegates committing, except if they're for Hillary of course, is ridiculous. All this delay serves the purpose of one single player. Give her Pennsylvania and cut her loose.

[alan t]

what the pols in washington, the superdelegates are afraid of, is the wrath of the corporate supporters of the DLC, who are unwilling to yield any control over the party to a newcomer and his followers who have a different agenda than the "we don't care if you're a R or a D as long as you work for corporate America!" It is K street that is keeping this nominating process alive, and every CITIZEN democrat should let their elected officials know that if we lose the general, they will be held accountable for their current cowardice. In the age of internet fundraising, there is little need for us to tolerate big moneyed "players" or pols who still carry their water.

[Annabella]

I'm not sure I understand why all the Obama supports, of whom I am one, are so anxious on all these blogs to have HRC drop out. It ain't gonna happen. What should be done is rein in the negativity of her campaign since it is just folder for November... but the more she loses fair and square the more there is hope that at least some of her supporters won't bolt in November to McCain.

There is something to be said for this dragging out... what doesn't kill Obama may be making him stronger. After all no one can say he is a Muslim just because of his name now. As long as he can maintain is grace and his cool, I think it redounds to his benefit.

The trouble is that I get convinced at times that she is in it only to injure him so that she can run again in 2012. I don't think she really still thinks she can pull this off without fracturing the Democratic Party, does she? But then she may want to do that on some unconscious level to show her fiercely Republican Goldwater Daddy up in heaven, that yes she can. As good a reason as I can think of for her playing Ralph Nader in 2008.

You gotta to admit she seems just delusional with an aura of ... what ... I can do or say anything I want...That is coming across loud and clear. So let her continue to self destruct. If Obama can't take it, then he isn't tested enough. Much worse will come his way in November and when he is President. People have to begin having real confidence in him if this is going to work...

[eva]

The total number of pledged delegates (without MI and FL) is 3256.
Whichever candidate gets half of that number (i.e. 1628 pledged delegates) or more, has the absolute majority of pledged delegates and cannot be overtaken by the other candidate. Once this number 1628 is reached, it is beyond spinning.

Obama has right now 1414 (according to Real Clear Politics) pledged delegates.
He needs 214 more to gain the absolute majority.

A plausible calculation gives him 66 delegates in Pennsylvania, 60 in North Carolina, 45 in Indiana, 12 in West Virginia, 30 in Oregon, and 25 in Kentucky. The last two elections take place on the 20th of May.
According to this scenario, Barack Obama will have 1652 pledged delegates, and thus have an absolute majority in pledged delegates on May 20th.

My guess is that this will be the moment all uncommitted superdelegates have been waiting for to move to the candidate that unequivocal math shows cannot be beaten in pledged delegates.

[Leo Regan]

Those superdelegates who support Obama should make their commitment after the North Carolina/Indiana primaries, in a process to resolve this situation. The present conflict does not enhance prospects for the Democrats, and is clearly antagonizing supporters from each camp. The swing voters will dismiss Obama or Clinton and the Democrats who are unable to be decisive. Continuation of the personal attacks does not harden or enable either candidate. The expression, a plague on both your houses, is appropriate. Obama should demonstrate his killer instinct and organise the effective conclusion of this campaign. Clearly he was prepared to antagonize the voters of Florida and Michigan, and with the army of supporters he can offer Democrats at every level a surge in participation and renewal of the party. There really is no contest at grassroots sufficient to justify Clinton's persistence in her delusion. Yet in the enormous political renewal of Obama's campaign, there is hope for sufficient numbers of superdelegates to be decisive, at the moment when Obama has achieved an insurmountable lead of pledged delegates. After North Carolina, the opportunity will have arrived.

[chanel]

I have been saying that all along. No one wants to force Clinton out because they don't want to feel the fire of The Clintons or her supporters, and that makes perfect sense to me because they need to be trying to unite the party. It is Hillary's right to stay in the race but I hope by June if she does not have any logical argument that she will quit and not try to drag things out until August.

[Jeremy]

I project that by the time Oregon and Kentucky have voted (May 20th) that Obama will have a lock on the pledged delegates. By that point, even if Clinton won 100% of delegates in Idaho, Puerto Rico, Montana and South Dakota, Obama would be guaranteed the pledged delegate lead.

So, it's possible that people wanting to force Clinton out may wait for Obama to cross that threshold before acting...

[Rieux]

What eva said, except that there are 3,253 total pledged delegates in play, not 3,256. (See here.) And thus the magic number is 1,627, not 1,628.

Anyway, eva's point stands: there's no need for this to go to June. Once Obama hits 1,627 pledged delegates (which is pretty much inevitable by May 20 at the latest; he needs less than 45% of the delegates between now and then to get there), there can no longer be any doubt that Obama will finish with more pledged delegates than Clinton.

So it's over.

[sue]

yeah, and he will lose to McCain in November.