Of Course Candidates Have to Run in Iowa

Josh Putnam does a terrific job in skewering the latest anti-Iowa diatribe, this time from Nicole Russell in the Atlantic. I'm going to pile on partially because the piece is so bad, and partially because it's an idea that keeps popping up and just doesn't make any sense. This is pretty simple: candidates for the nomination pretty much have to compete in Iowa if they want to win. End of story.

Josh for some reason doesn't quote the top of the article, where Russell runs off the rails right away:

Eleven years ago, George W. Bush campaigned in the Iowa caucuses and won. That was the year Steve Forbes and Alan Keyes trailed behind him, and a virtually unknown John McCain trailed way behind. And that was the last time a Republican who has run (opposed) and triumphed in Iowa has gone on from there to win the presidency. The Iowa caucuses may be first in the nation, but they don't live up to the emphasis placed on them by candidates and the media.

It's true! No Republican since Bush has won a contested Iowa caucuses and gone on to win the presidency! With that kind of observation, an analyst can obviously conclude...

Oh, wait. No Republican since Bush has won a contested nomination and gone on to win the presidency, have they? So I guess perhaps Russell's observation isn't really worth squat, is it? (And it's not just some weird oversight; she later lists Walter Mondale and Bob Dole (1996) as candidates who wasted their time in Iowa because they didn't achieve "eventual electoral success"). Step one: strategic choices made by candidates during the nomination process should not be judged by what happens during the general election.*

The question of whether to compete in Iowa shouldn't be based, at any rate, on how many Iowa winners get the nomination. That's because, as Josh points out, Iowa's function in the process is to winnow the field, not to pick the winner. The correct way to evaluate Iowa as a strategy isn't to see how Iowa winners have done; it's to assess what happens to otherwise viable candidates who skip the state. That's pretty hard to do (was Al Gore in 1988 a viable candidate? Would have have done better had he fought hard and finished, say, third in Iowa?). In my view, the clear evidence is that you can't just skip Iowa and win a nomination anyway.

What I think is more interesting is to think about how much flexibility is allowed. Some candidates have followed an all-in strategy, in which they basically run for president of Iowa for three full years; others split their resources. Of course, it's a lot easier to divide resources if you have more of them! So instead of thinking about whether Mitt Romney is skipping Iowa (he isn't!), it's worth thinking about the advantages and disadvantages of a strategy for a frontrunner that involves downplaying Iowa. Advantages: lowering expectations; more resources to devote elsewhere. Disadvantages: it might yield a lower finish there, which moves everyone else up a spot, and may still, expectations or no, be interpreted as a loss, which could hurt elsewhere. What complicates all this is the dual audiences candidates face. Party elites, including the partisan press, will heavily influence what the next round of primary voters believe both directly and indirectly (by providing resources to favored candidates), so it's important to listen to their expectations; at the same time, the neutral media is important too, once electorates get larger, and so giving them a positive story may matter.

So figured out exactly what Romney is up to and whether or not it's a smart strategy is, I'd say, an interesting exercise. But it has nothing to do with "skipping" Iowa, which is just not a viable nomination path.

*OK, I suppose at the extremes I can imagine an argument that some nomination strategy hurts in November so much that it isn't worth doing pre-nomination. It's hard to picture any such strategy, but at any rate it surely wouldn't be competing in Iowa.