Paradoxes in the proposed Colorado Amendment 36
The scheme on Colorado's November 2 ballot to amend the state constitution to do away with the winner-take-all awarding of the state's electoral votes in favor of a proportional system tries to do something that is arguably laudible, but unfortunately it turns out that there is no "fair" scheme to do this. Much like the subtle mathematical problem of the decennial apportionment of the House of Representatives (see either a brief history or for a more detailed history, the book Fair Representation by Balinski and Young) this seemingly simple task of splitting 9 things based on a proportion in a fair manner can be quite tricky.
The Colorado scheme proposed by Make Your Vote Count works like this. Take the percentage of the state's popular vote received by each candidate, multiply it by the number of electoral votes, and then round to the nearest integer. This makes sense, but this does not guarantee that the number of electoral votes will sum to the number that the state has, which for Colorado is 9. If the sum is more than 9, then electoral votes are subtracted from the candidate getting the fewest popular votes but still receiving an electoral vote. If the total is less than 9, then the candidate receiving the most electoral votes gets the unawarded electoral vote. This seems reasonable.
Suppose in the November election results look like this:
However, does the following seem fair? Consider what happens if Colorada actually had 10 electoral votes. For the same popular vote it would look like this:
But this totals to 11 eleven electoral votes, and so Nader, the candidate receiving the fewest popular votes, loses his. Thus giving Colorado one more electoral vote to award takes away an electoral vote for Nader to award one more to both Bush and Kerry!
Now consider a second example. Return to 9 electoral votes but modify the popular results taking votes from Kerry and giving them Bush to give this:
Again there are more than the alloted electoral votes and Nader's is taken away. Therefore Nader's getting an electoral vote is not only a function of how many people vote for him, but it also depends upon the margin of victory of the first place finisher over the second place finisher! That does not seem fair at all.
I have no idea whether this Amendment will pass, and then whether it can constitutionally be applied to this year's election, which will of course be over before we know whether or not this will pass. But we could get into another mess in the courts. And it will certainly make it more difficult for the networks to call the results with only a mere 99% of precincts reporting in. It's possible for Nader to have an electoral vote based on the 99% report, but lose it with the results of that last precinct, despite increasing his total percentage of the popular vote.
The Colorado scheme proposed by Make Your Vote Count works like this. Take the percentage of the state's popular vote received by each candidate, multiply it by the number of electoral votes, and then round to the nearest integer. This makes sense, but this does not guarantee that the number of electoral votes will sum to the number that the state has, which for Colorado is 9. If the sum is more than 9, then electoral votes are subtracted from the candidate getting the fewest popular votes but still receiving an electoral vote. If the total is less than 9, then the candidate receiving the most electoral votes gets the unawarded electoral vote. This seems reasonable.
Suppose in the November election results look like this:
Candidate | Popular Vote % | Electoral Vote (Not Rounded) | Electoral Vote (Rounded) |
George Bush | 48% | 4.32 | 4 |
John Kerry | 46% | 4.14 | 4 |
Ralph Nader | 6% | 0.54 | 1 |
However, does the following seem fair? Consider what happens if Colorada actually had 10 electoral votes. For the same popular vote it would look like this:
Candidate | Popular Vote % | Electoral Vote (Not Rounded) | Electoral Vote (Rounded) |
George Bush | 48% | 4.8 | 5 |
John Kerry | 46% | 4.6 | 5 |
Ralph Nader | 6% | 0.6 | 1 |
But this totals to 11 eleven electoral votes, and so Nader, the candidate receiving the fewest popular votes, loses his. Thus giving Colorado one more electoral vote to award takes away an electoral vote for Nader to award one more to both Bush and Kerry!
Now consider a second example. Return to 9 electoral votes but modify the popular results taking votes from Kerry and giving them Bush to give this:
Candidate | Popular Vote % | Electoral Vote (Not Rounded) | Electoral Vote (Rounded) |
George Bush | 50% | 4.50 | 5 |
John Kerry | 44% | 3.96 | 4 |
Ralph Nader | 6% | 0.54 | 1 |
Again there are more than the alloted electoral votes and Nader's is taken away. Therefore Nader's getting an electoral vote is not only a function of how many people vote for him, but it also depends upon the margin of victory of the first place finisher over the second place finisher! That does not seem fair at all.
I have no idea whether this Amendment will pass, and then whether it can constitutionally be applied to this year's election, which will of course be over before we know whether or not this will pass. But we could get into another mess in the courts. And it will certainly make it more difficult for the networks to call the results with only a mere 99% of precincts reporting in. It's possible for Nader to have an electoral vote based on the 99% report, but lose it with the results of that last precinct, despite increasing his total percentage of the popular vote.