I wonder what the outcome would be for Election 2000 if all the electoral votes were allocated in a proportional manner like in Maine and Nebraska. As I recall, the rules in those states are two electoral votes to the state-wide popular vote winner and one electoral vote to the popular vote winner of each congressional district.
Has anyone seen an analysis that determines the results using those rules?
Good question, but VERY difficult to answer.
The problem is that most states only report results down to the county level, like this for New Jersey. But congressional districts don’t correspond to county maps, as this set of House results by Congressional district shows. Notice all those counties listed as partially included in the total for each district. So unless you can convince the states to give you the Presidential results in the House race form, you can’t even do the analysis. Sorry…
Somebody with a computer and access to precinct level data could probably figure it out. I don’t think any precincts are ever split between House districts.
It would probably be a nice project for some think tank.
This was covered on NPR this evening. It turns out that if all fifty states had the Maine/Nebraska system, the electoral vote would have been 271 for Bush and 267 for Gore.
I too wonder what the electoral result of this system would be. However, I’m very glad that we don’t have the Nebraska-Maine system nationwide. It was bad enough having to fight over Florida, but can you imagine what the legal fight would be like in a race this close if you could pick up electoral votes in districts anywhere in the country? <shudder>
Good find GKittridge! Here’s a link to a real-audio clip of the NPR piece: http://www.npr.org/ramfiles/atc/20001218.atc.14.rmm. The relevant part starts at about 5:30 in.
But for their source, they actually reference a USA Today story. I dug it up, here’s a link: http://www.usatoday.com/news/vote2000/bush25.htm. Interesting article. Here’s what they say about their methodology:
FWIW, I did a “back-of-an-envelope” calculation that came up with the same number on 29/11/00. I posted the results in a this thread http://boards.straightdope.com/sdmb/showthread.php?threadid=48473 that turned out to be a thread killer!
Certainly not claiming any expertise in this matter. What this simply shows is that overall, the voter distribution between districts did not differ markedly from the state averages. This would not be unexpected in a first-past-the-post system. However, if proportional voting was used the result is a little different because Nader would start picking up Electoral College votes.
A person on a Washington Post chat session said today that if the proportional system had been in place in 1960, there would have been a 269-269 tie in the electoral vote. However, in the end Kennedy won by a sizeable margin in the electoral vote (303-219 I believe with a few scattered votes by faithless electors).
Now wouldn’t that have been fun?
How about:
Adding a third factor to the electoral vote total for a state: area. Rhode Island (or whatever state in the future is the smallest) gets one more. Every other state get one more vote for every RI-area that will fit in it. This way, Alaska would get 545 more votes, for a total of 548. Maybe then someone would campaign up there. :D:D
Instead of every state getting getting two at-large votes for their Senate seats, let’s base the number of Electoral Votes solely on the number of congressional districts, e.g. Montana gets 1 instead 3, California gets 52 instead of 54, etc.
This would help minimize the disparity of every state getting at least three votes, even if a state had just (for argument) ten residents. With their focus on equal protection, I’m sure the Supreme Court would support this idea.
Woolly, does your back-of-the-envelope show who would have won under this system?
(The ten residents would be quite busy - two US Senators, one US Representative, one Governor, one State House Rep, one State Senator, one State Chief Justice and three private citizens.)
Well I am just guessing but there is no possible way either of them could get 270 EC votes that way:)
Rillian, I think they talked about that possibility in the NPR piece, right after they gave the results for OP’s scenario. Unfortunately, I stopped listening and didn’t hear the results. 
But you might want to listen to it and see what they came up with.
And of the three private citizens, one’s a stay-at-home Mom, one’s a welfare bum, and the last has to pay enough in taxes to support the entire state! 
“Instead of every state getting getting two at-large votes for their Senate seats, let’s base the number of Electoral Votes solely on the number of congressional districts, e.g. Montana gets 1 instead 3, California gets 52 instead of 54, etc.”
According to an item I heard on the radio a few weeks ago, had this system been used, Gore would have one. (Bush won in a lot of the dinky states.)
“With their focus on equal protection, I’m sure the Supreme Court would support this idea.”
Indeed, the present system gives much more weight (something like 40% more IIRC) to the votes of voters in states like N. Dakota than to the voters in states like California.
Actually, the most screwed states in the Electoral College system are the mid-sized states. Here’s a summary from Slate’s Chatterbox column http://slate.msn.com/code/chatterbox/chatterbox.asp?Show=12/13/2000&idMessage=6680
“As Chatterbox has noted before, the distortions of the Electoral College aren’t as simple as widely believed–that is, they don’t just favor little states over big ones. Because of the winner-take-all feature, they actually favor big states more than they do little ones (it’s the medium-sized states that tend to get screwed). Longley and Peirce have actually calculated the relative voting power in each of the 50 states. Montanans have the least relative voting power. Californians have the most. A Californian, in fact, enjoys two-and-a-half times the voting power of a Montanan and 72 percent more voting power than the average voter nationwide. After California, the states with the most Electoral College clout are: Texas, New York (they both have not-quite-twice as much clout as Montana), Florida, Pennsylvania, and Illinois (which has a little more than one-and-a-half times as much clout as Montana). Chatterbox has just named all the states where voters enjoy more-than-average clout in presidential elections. After Montana, the states that have the least voting power in the Electoral College are Kansas, West Virginia, Maine, Arkansas, Utah, and Nevada.”
“Actually, the most screwed states in the Electoral College system are the mid-sized states.”
That’s rich. Longley and Peirce must be statisticians.
Evidently they are making some assumptions about the distribution of votes.
Here is an assumption-free analysis:
Montana: 3 electoral votes/633,666 people = 0.0000047 electoral votes per person
Montana: 54 electoral votes/33,145,121 people = 0.0000016 electoral votes per person
In my book, 0.0000047 is more than 0.0000016.
I do admit there are circumstances in which big state voters could have an advantage (for example, if people voting for president always voted only for candidates resident in their state, only Californians could ever be elected).
Oops! The first Montana above should be North Dakota and the second should be California.
I believe the idea of “voting power” Longley and Peirce are using is “the chance your vote could make the difference in who becomes president”. A single voter in California can throw all 54 of the state’s electoral votes to one side or the other. Sure this isn’t likely (Yeah gave the odds), but if it does happen, those 54 votes have a very high chance of changing the winner of the entire election.
Contrast that to Montana, where it is obviously more likely that a single voter could change the direction of the 3 votes. But it’s much less likely that those 3 votes would make the difference in choosing the president.
Or something like that. How they get actual numbers out of this idea is for someone else to answer.
Another problem with the Electoral College is that there are several possible ways of assigning representatives in Congress. We currently use something called the “method of equal proportions,” which I do not pretend to understand very well. But we haven’t always used this system–only since WWII or so. There’s no obvious reason why this method is more fair than any other: each method has its paradoxes and quirks, and none can claim to be totally free of bias in one direction or another.
Other methods would produce about the same result as the equal proportions one, with the size of most states’ delegations (and number of electoral votes) being unchanged. But in an election this close, “about” doesn’t cut it. One early–and easy to understand–method of determining representation was Alexander Hamilton’s, which was approved by Congress in 1792 and then was vetoed by Washington–the first veto ever.
According to my own back-of-the-envelope calculations (actually, I used an index card, but what the hey) the use of Hamilton’s method would have transferred one legislative seat each from Oklahoma and a second Bush state (don’t have my notes here) and given them to Massachusetts and New Jersey. Thus, minus two electoral votes for Bush, plus two for Gore, and we have…
You fill in the rest.
One problem with using the Maine/Nebraska system for the entire country would be that the time-honored practice of Gerrymandering could be used to give an advantage to one party’s candidate for ten years. So whoever controls the House in years ending in zero would have a much better chance not only of continueing to control the House, but also of electing a president.
Perhaps a system of neutrally-devised “Elector Zones” equal to the number of House seats could be devised. But that’s a whole new can of worms. What does “neutrally-devised” mean? Why would the House ever pass a system that diminished their own power?
At first I wanted to be in favor of electoral college reform instead of electoral college abolition, but the more I think about it, the more inclined I am to just chuck the whole thing and go to a straight popular vote.