John Harwood of the Wall Street Journal correctly pointed out in a Nov. 10 column that the GOP almost got seriously screwed by the Electoral College:
[I]f 75,000 voters in Ohio had swung toward Sen. John Kerry, all 20 of the state's electoral votes would have gone to him, too—even though the Republican incumbent still would have received his impressive 51% majority of the national popular vote, a margin of three percentage points over his opponent. Mr. Kerry would have won the election with a 272-266 Electoral College victory.
Bush got 3 and a half million more votes than Kerry, but Ohio could have killed it for him. That should be reason enough for conservatives to want to abolish the Electoral College, and a few do. It should be an especially easy sell for conservative populists. And yet the Conservative Caucus, chaired by the veteran right-wing activist Howard Phillips, is sending every presidential elector a free copy of The Importance of the Electoral College, by George Grant, a humanities professor at Franklin Classical School, an evangelical high school in Tennessee.
Grant's book, which is elegantly written and tightly argued—kindly note this blue-stater's respect for red-state evangelist scholarship—is preoccupied with states' rights, which the Electoral College does indeed enhance (at the expense of the voters). Its most arresting argument, though, is the following nightmare scenario:
If the federal hedge of the Electoral College were not in place then it would be perfectly possible for a candidate to lose as many as 49 states and still win the Presidency. Imagine the chaos if George Bush had won every state but lost Massachusetts by a popular vote margin sufficient to cost him the Presidency. If he had won each of those states by fairly small margins—akin, say, to his actual margin of victory in Florida—his lead might amount to less than half a million overall. If Gore had won Massachusetts by a million votes—mostly from the city of Boston and its suburbs—then Gore would have won the overall popular vote total, despite having lost 49 of the 50 states. His popular vote would be 500,000 votes more than Bush's. With the Electoral College, Bush would have won 523-13. In a direct election, Gore would have prevailed.
Popular-vote margins this thin aren't unheard-of. In 2000, Gore won the popular vote by a margin of 540,000. In 1960, John F. Kennedy won the popular vote by a margin of only 113,000. Theoretically, one could pick up the votes necessary to win the popular vote in a single state while losing all the other states.
In practice, though, it seems unlikely. What Grant is positing is essentially 49 more or less tie votes that all resolve themselves in Candidate A's favor. It's like imagining 49 consecutive coin tosses that all come up heads. I'm sure that's happened sometime, somewhere, but you'd be foolish to plan your life around the possibility. Plus, one state would have to give Candidate B a popular-vote victory margin about 20,000 times as great as Candidate A received in any other state. We live in a diverse nation, but it isn't that diverse. If any one state showed results so dramatically different from the results in each of the other 50 states, the likeliest explanation would be that someone had tampered with the polls.
But let's imagine that, in spite of these obstacles, Grant's nightmare scenario came about without any stuffing of ballot boxes. Why should we regard the outcome as a terrible thing? None of Candidate A's 49 states would have rejected Candidate B by a statistically significant margin. In truth, Grant's example describes a popular vote that was tied in 49 states, with the 50th breaking the tie.
Grant's argument is a pretty good example of one key distortion in our thinking brought about by the Electoral College. It makes imaginary victories look real. Who cares who won in a given state if the popular-vote difference was statistically insignificant? In reality, nobody won. But under the Electoral College system, somebody always has to win at the state level, or else you can't award state electors. Under a popular-vote system, we wouldn't have to play that game. We'd just count up the ballots and see who got the most votes.
America's Worst College Archive:
Oct. 19: "Why the 'bye, bye, Nevada' argument is wrong"
Sept. 9: "Why the 'no more majorities' argument is wrong."
Aug. 17: "Why the '50 Floridas' argument is wrong."
Aug. 11: "Let's choose our president by popular vote."