Actually, Andy writes that “close elections really have become more common” and then counts how many elections are close using a couple different thresholds. This raises an interesting distinction, as having more close elections may not necessarily mean that elections have become closer on average, as long as the uncompetitive elections tend to be landslides. In his piece, Brinkley wrote:
bq. But for the past 40 years, close and unpredictable elections have increasingly become the norm.
On its face, this seems closer to Andy’s interpretation than to an interpretation about the average closeness of elections, which was the focus of my quickie analysis in the earlier post.
Here’s some more data, this time charting the winning presidential candidate’s margin of victory from 1856-2004. (1856 was the first election in which the Republican Party competed.) I have put a dashed line at the average for this period (10 points).
To me, it’s tough to say categorically that elections have become closer, on average. 2000 and 2004 were close, obviously, but 1988-96 were not too far below the historical average and were much less competitive than 1960, 1968, or 1976. And then there were three landslides in 1964, 1972, and 1984. At the same time, as Andy notes, nailbiter elections have become more numerous since 1960. (Although, as I said before, I’m not quite sure why we should be starting 40 years ago, as Brinkley does. This seems arbitrary.)
Since I don’t really buy Brinkley’s account of partisan dealignment, I think the closeness of the most recent races may derive from greater parity between the two parties in terms of party identification and registration (due largely to the rise of the Republican Party in the South) and to various idiosyncratic factors that influenced individual races.