The United States electoral system continues to grind on in its creaky nineteenth century fashion, and will eventually produce final results from last week’s election. (Three Senate races are still undecided, one of them having a runoff next month, and most news organisations have still not called the presidential race in Missouri.) But the figures are now complete enough to allow some interesting analysis. (The New York Times has user-friendly numbers.)
To some extent observers have been misled by the previous two presidential elections, which were both extraordinarily close. This year just returned to normal: a decisive victory, but certainly not a landslide. Indeed, leaving out 2000 and 2004, it was the closest election since 1976.
Nonetheless, a closer look at the result discloses some worrying signs for the Republican party.
The GOP carried 22 states, nine fewer than in 2004. Eight of them are spread across the Rocky Mountains and the plains (including Alaska, which is usually categorised with the Rockies for convenience), but the rest constitute a solid block across Appalachia and the deep South: West Virginia, Kentucky, Tennessee, Missouri, Kansas, Arkansas, Oklahoma, Texas, Louisiana, Mississippi, Alabama, Georgia and South Carolina.
This area is the only place where the Republicans are gaining ground. In terms of swing from 2004, they outperformed their national average by more than 1.5% in eleven states: three of them due to personal factors (Arizona, Alaska and Massachusetts, the respective homes of John McCain, Sarah Palin and John Kerry), and the other eight all in the southern and Appalachian heartland.
In five states — Arkansas, Louisiana, Tennessee, West Virginia and Oklahoma — McCain actually improved on the 2004 results. No prizes for guessing what feature of Obama’s candidacy was alienating voters there.
Dave Leip’s atlas shows the trend graphically (Leip’s figures are slightly different to mine because his are raw rather than two-party, but the pattern is the same).
At county level, the picture is even more striking: the New York Times has a map showing the strength of swing by county (based on exit polls, so there will be small discrepancies). It makes for sobering viewing, as does the accompanying story on attitudes in the south.
Even in the new Republican heartland, they are losing ground at the edges, where the influx of Hispanic and/or northern immigrants is changing demographics. Georgia and Texas both swung to Obama by more than the national average, and South Carolina came close. If the trend continues, those states will go the same way as Virginia and North Carolina. In the Rockies and the plains, that trend is even clearer: six of Obama’s nine biggest swings occurred in the region.
An obvious comparison is with Bill Clinton’s victory in 1992, which was of very similar magnitude (just five more votes in the electoral college, and 0.3% more of the two-party vote). Clinton won in the more traditional way, picking off a string of southern states whose Democrat heritage could still be tapped by a southern candidate. Obama, however, is riding a demographic wave; the solid swings among younger, better educated and more cosmopolitan voters bode well for the Democrats’ future.
Only time well tell whether this is a true realignment: as Nate Silver says, “ask me in eight years”. But it looks enough like it to give the Republicans plenty to worry about.