Playing for Both Teams

Senator Arlen Specter arrives at the US Capitol today (Alex Wong/Getty)

The politics of why Sen. Arlen Specter (formerly R-PA, now D-PA) has switched parties today is pretty simple: Facing a tough, and probably politically fatal, primary from a conservative Republican, the moderate/liberal Republican chose to live — be it as a Democrat.

What’s more complex, I think, is the psychology of party switching.

While party switches get a lot of attention when they happen, what’s really remarkable is that there aren’t far more of them. When a new party takes control of a legislative body, there often could be fairly huge benefits available to a politician willing to switch sides to the new majority — committee assignments, getting desired legislation through, getting pork for one’s district. And when a particular potential party switcher could tip a major milestone — such as, say, a 60-vote majority in a Senate where 60 votes are necessary to do just about anything — the new party has the incentive to offer any number of inducements.

So, why don’t we see a lot more of this?

There is, of course, the problem that some senators and representatives are from strongly conservative (or liberal) states or districts. Their constituents have strong party alliances, and so a party switch would mean some short term gains followed by certain political death.

But plenty of politicians, like Specter before this switch, live in “purple” or even opposite-party-leaning states and districts. Still, very few of these folks switch.

The reasons they don’t switch are obviously heavily social and psychological.

Here’s one paper taking a stab at some of the barriers:

Since, for elected leaders partisan loyalty is more important than for other citizens, a change in partisanship for these leaders can be presumed to be more consequential as well. Such a change has personal as well as professional implications. It affects not only the politician’s current and future career but, at least in the short and middle term, the entire network of friendships and relationships built up, both within and outside the legislature, over the course of that career. As a result of a party switch, the leader is perceived as less predictable, less constrained by the “normal rules of the game.” The natural question that arises is, “what other untoward behavior may be expected from this person.” And from the switchers perspective, the reactions of others become less certain, especially in the near term, when the partisan thread is severed.

I would guess that this social effect is pretty big. But I’d also guess a major change in self-image is pretty jarring as well.

In his statement, Specter says:

While I have been comfortable being a Republican, my Party has not defined who I am. I have taken each issue one at a time and have exercised independent judgment to do what I thought was best for Pennsylvania and the nation.

Specter may believe that being a Republican hasn’t defined him. Or it may be something he feels he needs to say, now. But it’s quite unlikely to be true. Group identities, even identities that should be very weak (such as in a lab setting), have been proven to be much stronger than we’re capable of realizing.

In fact, adopting this new group identity is likely to change Specter’s ideological and voting behavior, even though he’ll go to any length to tell himself and the public that it hasn’t:

My change in party affiliation does not mean that I will be a party-line voter any more for the Democrats that I have been for the Republicans. Unlike Senator Jeffords’ switch, which changed party control, I will not be an automatic 60th vote for cloture. For example, my position on Employees Free Choice (card check) will not change.

Whatever my party affiliation, I will continue to be guided by President Kennedy’s statement that sometimes party asks too much. When it does, I will continue my independent voting and follow my conscience on what I think is best for Pennsylvania and America.

Maybe not. But Specter’s voting behavior is likely to change significantly, as this paper (referenced in this paper) from 2000 found.

Humans can’t help it. They’re social animals. Change the group you affiliate with, change your behavior.