The Duration of Government Shutdowns

I only have a few minutes, so my contribution to our shutdown discussion will be limited. Will a government shutdown occur?  I don’t know.  My intuition tells me no, as I think the political costs are too high for both sides.  But after hearing Reid say on the Senate floor that the differences are wholly ideological, I’m starting to think a shutdown is a real possibility.
In his last post, Nate looked back to 1995 to compare and contrast those government shutdowns and the looming one.  I wanted to take a different approach and look cross-sectionally at all recent government shutdowns.  Indeed, there have been three state government shutdowns in the past decade to go along with the two federal government shutdowns in the 104th Congress.  What I quickly realized is all government shutdowns have one thing in common–they are short in duration.  Despite the gloom and doom that most people attribute to a government shutdown, they don’t last very long.  Here is the informative portion of this post:

• Federal Government Shutdown Redux–12/16/1995 to 1/6/1996 (21 days)
• Minnesota Shutdown–7/1/2005 to 7/9/2005 (8 days)
• New Jersey Shutdown–7/1/2006 to 7/8/2006 (7 days)
• Federal Government Shutdown no. 1–11/14/1995 to 11/19/1995 (5 days)
• Pennsylvania Shutdown–7/11/2007 to 7/12/2007 (1 day)

Both Nate and Josh in their previous posts mentioned that Boehner is in a precarious position.  The conventional wisdom is that the divisions within the Republican Party (between the Tea Partiers and more moderate representatives) have limited Boehner’s ability to negotiate a moderated spending plan with Reid and Obama.  I don’t dispute this point.  Thus, I think a short shutdown (one that lasts over the weekend) could give Boehner some political cover on his right (lets call this “Tea Party street cred”) and result in a moderated bill as the political pressures rise on both parties.  Indeed, I think the ephemeral nature of government shutdowns suggests that they are highly political creatures.

Bonus: my semi-parametric Bayesian hierarchical generalized estimating equation (with robust standard errors, of course) says there will in fact be a government shutdown and it will last until Tuesday, April 12.