Sean Trende at Real Clear Politics is one of the rare pundits who actually takes time with data. His posts are regularly full of polling, demographic and election data. However, they’re also typically rife with assumptions that when questioned show his entire argument to be specious. Usually that process requires taking apart several assumptions, which means showing how he’s wrong is often tedious.
But today, showing his claims are dubious is quite easy. Trende claims Obama’s recent immigration decision will be a net negative. He has three points. One is that Latinos aren’t a monolith, and they don’t really care about immigration as much as everyone thinks. But white working class voters, they really do care about immigration, and Obama already has problems with them so this means they’ll flock to Romney. [Never mind that Trende, as Larry Bartels and others have shown ad nauseum, is dead wrong in thinking "white working class" voters are a monolith, and that contra Trende, white working class voters in the North--which overlaps with the states won by Gore and Kerry--have actually moved to the Democrats in recent years; it's only in the South that white working class voters have become more Republican.]
My favorite part of his piece, though, is this:
1) Latinos are underrepresented in swing states. While the Latino vote is frequently portrayed as a critical voting bloc, in truth it is concentrated in only a few swing states with just a handful of electoral votes. The only states where Latinos make up more than 10 percent of the electorate are: Arizona (16 percent of the electorate in 2008), California (18 percent), Colorado (13 percent), Florida (14 percent), Nevada (15 percent), New Mexico (41 percent), and Texas (20 percent).
Of these, only Colorado, Florida, and Nevada are swing states; New Mexico and Arizona are at best borderline swing states. In Florida, the Latino vote largely (though decreasingly) comprises voters of Cuban descent and is therefore atypical of other Latino electorates.
So in the end, we’re talking about Colorado and Nevada as the states where this is likely to produce dividends of any size, for a total of 15 electoral votes.
Yes, why should we give much importance to those two states with their paltry 15 electoral votes? Let’s say Obama won only the Gore states, plus New Hampshire (which Gore lost by less than a percentage point), Colorado and Nevada. That would mean he’d lose the electoral college, right? Oh, uh, hold on…
I don’t think Obama adopted the new immigration policy purely for electoral gain. But I do think it will be a net gain. Trende wants us to believe it won’t, and to believe him, we’d have to believe that Obama will lose New Hampshire or at least one of the states won by Al Gore and not win any other states except possibly Nevada and Colorado. Well, that’s a plausible scenario, but if so, it would almost certainly be because of factors other than immigration. But if Obama runs about like Kerry and Gore, but adds Nevada and Colorado, he won’t be 19 electoral votes short like Kerry, or 3 electoral votes short like Gore. Instead, Colorado and Nevada would give him the presidency with a couple electoral votes to spare.
From an electoral standpoint, Obama’s immigration decision might seem puzzling, but only if you’re puzzled by the potential decisiveness of Obama locking down the electoral votes of Nevada and Colorado.