A lot of people have pointed to Dan Balz's piece about how Democrats are registering record numbers of voters as evidence that the extended primary isn't so bad for the party. Believe me, I'm all for bucking the C.W. But I don't find this very persuasive. In order for this to be much of a benefit, you'd have to show that the registration numbers wouldn't be up if one candidate or the other had already locked up the nomination. Actually, you'd have to show more than that. You'd have to show that the registration numbers wouldn't have ended up in the same place by November under either scenario. (For most states the general-election registration deadline is sometime in October.) I don't think you can really make that case.

For one thing, if we had a nominee, that person would already be organizing a lot of states for the general. And all the evidence suggests he or she would be having a lot of success registering people as Democrats. As Balz himself points out:

The Pew Research Center offered fresh evidence of this last week with a report that aggregated interviews with 5,566 voters during the first two months of the year. It found that 36 percent of respondents identified themselves as Democrats and 27 percent called themselves Republicans, a drop of 6 percentage points since the 2004 election. The report noted that, on an annualized basis, this is the lowest GOP identification in 16 years of surveys.

This has little to do with the fact that there's still a primary contest going on. It has to do with the fact that a lot more people want to be Democrats these days than Republicans.

I'll concede that, from a registration perspective, there's some advantage to having a candidate actually appear in a state--as opposed to just having staff organize it--and that competitive primaries have some positive effect on registration. But, again, for Balz's point to be relevant, you'd have to argue that the registration increases we'd forgo already by having a nominee wouldn't be made up by the fall. I have a hard time believing they wouldn't be.

The only relevant consideration here is down-ballot races in states that might otherwise get neglected but which are getting attention because the candidates are competing in primaries there. But you have to weigh that against that fact that these states won't help a Democrat get elected president. If Clinton or Obama had already clinched the nomination, they could be spending time in states they actually have a chance of carrying in November (or which at least count in the electoral college), as opposed to Montana, South Dakota, Kentucky, Mississippi, Wyoming, Indiana, and Puerto Rico.

Of course, even if you buy Balz's logic (and how could you after all that!), no one's suggesting the primary contest should have ended after, say, South Carolina. They're suggesting it should have ended after March 4, by which point it was clear that Hillary had very little chance of overtaking Obama in the pledged-delegate count, and that most superdelegates were reluctant to overturn Obama's delegate lead. By that point, the vast majority of states had already held their primaries and caucuses.

--Noam Scheiber