In Defense of South Carolina: Institutions Matter

We all know the story of the 2000 Republican presidential primary in South Carolina.  John McCain won New Hampshire by double digits, leading a massive increase in campaign donations, campaign volunteers and press.  In response, the Bush campaign went negative in South Carolina–using push polls and other means to spread some nasty rumors, including one that McCain had fathered a black child out of wedlock.  Messy stuff for sure.
Christopher Lamb, a colleague of mine at the College of Charleston, writing in yesterday’s Huffington Post, warns Mitt Romney to avoid “going naked to a knife fight.”  This is certainly prescient advice.  In explaining why Mitt Romney should be especially prepared in South Carolina Lamb writes:

To understand politics in South Carolina, one needs to be aware of the quote from the Unionist James Louis Petigru who responded to the state’s decision to secede from the United States in December 1860 by saying, “South Carolina is too small for a republic and too large for an insane asylum.”

As a social scientist, in particular a political scientist, I have an issue with Lamb’s claim. In my mind this quote implies that South Carolina’s politics (particularly it’s primaries) is unusually negative and hyperbolic because it’s people are somehow atypical.  In simple terms, the state’s politics is a function of its 4.7 million citizens.  But while South Carolina is more conservative than most states (and no doubt differs in other ways), I highly doubt its citizenry alone leads to vicious primaries.  In fact, I’m skeptical that the people of South Carolina are even a critical factor in this case.

A political scientist (Lamb is a communications professor)–in particular one approaching this topic from an institutional perspective–might look to other factors.  Specifically, while I think Lamb’s overall point is accurate, what he overlooks in my view is that the primary system itself, and the rules making South Carolina “first in the South,” greatly affect the nature of the state’s primaries.  Because the primary system is an iterated process (rather than a one-shot, 50 state election), political “momentum” is critically important (see this paper by John Aldrich for a formal proof of this dynamic).  Simply put, candidates who win early primaries like Iowa and New Hampshire are likely to receive greater support in subsequent states because of sophisticated or “front runner” voting (see this paper) as well as generate greater campaign donations and support.  This, in turn, improves their chances of winning subsequent primaries.  Because South Carolina is third in this sequence, there is an incentive for opposition candidates to go negative independent of the state’s demographics.

In sum: I don’t think it’s that South Carolina is atypical per se, it’s just the way the primary system is designed and run.