Different Unfair System

CQ Politics reports that if every state divvied up its electoral votes by congressional district (as Maine and Nebraska, but no other state, do), then Obama would still have won the election, but by a smaller margin. Instead of getting all 55 of California's electoral votes, Obama would have won only 44, with McCain getting 11; on the other hand, Obama would have picked up 11 EVs in Texas instead of 0, so that's a wash. The biggest change would have been in Florida, where Obama would have won only 12 EVs instead of all 27, with 15 switching to McCain. Overall, with all the pluses and minuses considered, Obama's victory would have been 301-237 instead of 365-173.

It's kind of interesting, I suppose, but really, why do we care what would have happened under some different unfair system? The current system, under which most states allocate their electoral votes on a winner-take-all basis, is obviously unfair, in that a candidate who gets 51% of a state's popular vote wins 100% of its electoral vote, while a candidate who gets 49% of a state popular vote gets 0 electoral votes in that state -- all the votes for that candidate are effectively discarded. On a nationwide basis, this means that the candidate with fewer popular votes can win the election, as, of course, occurred in 2000. This will happen if the "winning" candidate candidate wins enough states by a slim margin, while the "losing" candidate wins states by big margins. CQ's hypothetical system (and the actual Maine and Nebraska systems) just reproduce this unfairness at a different level.

If we're going to think about different systems of allocating electoral votes, what would be interesting would be to consider what would happen if states allocated their electoral votes proportionally to the statewide popular vote (as was proposed, but not adopted, in Colorado in 2004). This would be a lot fairer than the current system, although there would still be some issues about accounting for fractions. If a state has, say 5 electoral votes, and one candidate gets 70% of the popular vote while another candidate gets 30%, it's necessary to say whether the 70% candidate should get 3 of the state's 5 EVs (60%) or 4 (80%). Obviously, there's no perfect way to make this decision. Also, third-party candidates who get, say, 5-10% in a particular state would have to be accounted for. But such a system would be fairer than what we have now, or the CQ/Maine/Nebraska system of reproducing the current unfairness at the congressional district level. Fairest of all would be to eliminate the electoral college and elect the President by popular vote.

Of course, in politics, things don't get decided by what's fairest, but by what best promotes the interests of those in control of the system. A popular-vote system would have far-reaching, perhaps difficult-to-predict effects. It is predictable that a candidate would have much more incentive to run up popular vote margins in big states. Currently, a candidate who's ahead 60%-40% in California has little incentive to spend any money there, but under a popular vote system the leading candidate would have every incentive to try to run the vote total up to 70%, which would be 3 million more votes. And candidates would probably spend less in small states -- why work hard to push your vote up an extra 10% in Wyoming when that's only another 60,000 votes?

So small states would strenuously resist a change from the current system. But why should small states have a disproportionately large impact on our presidential elections?