Electoral votes and the popular vote
In this morning’s WP, George Will reminds us that the 2000 and 2004 elections were, in historical terms, remarkably close. Unlike many of the elections since 1900, the winning party in 2000 and 2004 (the Republicans) received roughly the same percentage of the electoral vote that it won in the popular vote. In 2000, Pres. Bush received 0.7% more of the electoral vote than he did the popular vote. In 2004, he received 2.0% more.
The following table (Table 1) shows the historical relationship between the percent of the two-party electoral vote and the percent of the two-party popular vote the winning candidate received. The four elections with the smallest differences are 1916, 1976, 2000, and 2004.
One additional thing to note from Table 1 is that the Electoral College exaggerates any popular vote majority a candidate may have. Winning candidates, on average, get 19% more of the electoral vote than they do the popular vote. The effect is particularly striking in blow-out elections. In 1984, for example, Ronald Reagan won 59.2% of the two-party vote while receiving 97.6% of the electoral vote. Even in close elections, however, the Electoral College distorts the popular vote. Bill Clinton won “just” 53.5% of the two-party vote in 1992 while receiving 69.8% of the electoral vote.
You may ask, what about the impact of third party candidates like Ross Perot? Table 2 gives us some information about their impact on the election outcomes. The red bars indicate election in which a third party candidate received at least five percent of the popular vote but did not win any electoral votes. The yellow bars indicate elections in which a third party candidate received at least one electoral vote without receiving at least five percent of the popular vote. The green bars represent elections in which a third party candidate received more than five percent of the popular vote and at least one electoral vote. There doesn’t seem to be any consistent effect.
What about 2008? How different will it be? Right now, we have no way of knowing for sure what the final outcomes will be. There are, however, lots of predictions out there. (See especially the current edition of PS, which has this year’s political science prediction models.) If the predictions are right, 2008 will be different from 2000 and 2004, but it will also be below the historical average. To illustrate: FiveThirtyEight.com has one of the more generous predictions currently available. They estimate Sen. Obama will win 64.5% of the electoral vote and 51.9% of the popular vote, for a difference of 12.6%.