Census

Over the weekend, I tried out a new regular feature in which I ask everyone to step back and think about what actually matters, instead of just focusing on the hype and all.  Speaking of that topic, here's one: new projections about which states will gain and lose in the reapportionment following the 2010 census. 

A few things to keep in mind as you look at the linked Richard E. Cohen article, which I thought was pretty good.  First, Cohen is focused on the effects for the House, but of course reapportionment matters for the presidency, perhaps even more.  Texas adding four seats is a big deal, with New York losing two.  That's a nice gain for the GOP.  The other two multiseat states are a wash, with swing state Ohio losing two and swing state Florida gaining two.  The other gainers are GOP states UT, SC, and GA; Dem state WA; and swingers NV and AZ.  Losing a seat each are Democratic states IL, IA, MA, MI, and NJ; Republican LA; and swingers PA and MO.  Add 'em up, and it's clearly a net gain for Republicans...although that is balanced, perhaps, to the extent that increased Latino voting in Florida and Arizona push those states away from the GOP.  Indeed, my guess would be that Arizona may be on its way to moving into the top tier of key states in the battle for the presidency -- or perhaps a better way to think of it is that Florida (projected EV 29) will increasingly be in a tier by itself, with Ohio (projected EV 18) and Pennsylvania (20) dropping back into a pack that includes Arizona (11), Colorado (10), Virginia (13), and a few others. 

As far as effects on the House, there are a couple of things to keep in mind.  One is to remember that we're mostly interested in changes from new redistricting, which involves knowing just how the old districts played.  So remember that Texas is beginning from the DeLay gerrymander that helped Republicans, not from the 2001 Democratic gerrymander, when speculating about how many seats will be won or lost for each party.  Another is that most gerrymanders tend to be bipartisan gerrymanders, which protect all incumbents.  Parties that try to maximize seat gain by keeping their winning margins down (and thus allowing their voters to have the majority in more districts) risk losing a lot very bad years, and at any rate incumbent Members of the House are usually all too happy to exchange larger party goals of maximizing seats for the very different goal of maximizing their own safety.  Bipartisan incumbent protection gerrymanders such as California's ten years ago don't result in very many close elections, but because politicians tend to like them they are often enacted, meaning that redistricting usually isn't nearly as big a deal as some believe.  So it matters, but don't get carried away.

By the way, for an argument for the democratic advantages bipartisan gerrymanders, see Justin Buchler's "Case Against Competitive Congressional Districts."