The Electoral College: When Losers Win - Island of Sanity

Island of Sanity

Campaigns & Elections

The Electoral College: When Losers Win

In 2000, George W Bush was declared the winner of the presidential election ... even though he got fewer total votes than his opponent, Al Gore. This happened because in the United States, we do not elect our presidents by a direct, popular vote, but rather by a somewhat complex system called the "electoral college". Every time we have a presidential election, people talk about the significance and implications of the electoral college, and there is always some discussion that it should be abolished and replaced with a direct election. But a case like this has prompted more discussion that usual.

A quick summary of how the electoral college works

Under the electoral college system, voting is done by states. Each state gets a number of "electors" equal to the number of votes it gets in the House of Representatives plus the number of votes it gets in the Senate. (The number of votes in the House is proportional to the state's population, at present about one vote for every half-million people. The number of votes in the Senate is always two.) The electors then vote for president. The Constitution says that the legislature of each state shall decide how its electors are chosen. In practice, every state chooses them by a popular vote, in which the electors are listed on the ballot by who they intend to vote for for president. In many states the name of the elector is not even listed -- all that is shown is the name of the presidential candidate that that elector said he would vote for.

This "two step" process of election would make little difference were it not for two important facts:

First, in 48 of the 50 states, the election of electors is "winner take all". If a state gets, say, 10 electoral votes, and 60% of the people vote for Jones while 40% vote for Smith, that does not mean that Jones gets 6 electors and Smith gets 4. Rather, Jones gets all 10 and Smith gets zero. (The two exceptions are Maine and Nebraska.)

Second, because of the "Representatives plus Senators" rule, the number of votes is not strictly proportional to population. More on this in a moment.

Previous minority winners

Mr Bush is just the third president in our history to be elected with less than a majority, and two of the previous "minority winners" were special cases.

In 1824, John Quincy Adams beat Andrew Jackson, even though Adams had only 105,321 votes while Jackson had 155,872. But this wasn't really because of the electoral college: Jackson had more electoral votes than Adams, too, 99 to 84. The issue here was that there were four candidates in the race, and no one got a majority: Henry Clay got 46,587 popular votes and 37 electoral votes, and William Crawford got 44,282 popular votes and 41 electoral votes. Under another clause in the Constitution, if no candidate for president gets a majority, then the House of Representatives chooses from among the front-runners. Even if it had gone strictly by the popular vote, then assuming this "run off" rule was still in place, the outcome would have been the same. (We could debate having the House decide these run-offs, but that's a whole different question.)

In 1876 Rutherford B Hayes got 4,033,950 popular votes and 185 electoral votes, while Samuel Tilden got 4,284,757 popular votes and 184 electoral votes. Thus Hayes was declared the winner. In that election there were sharp disputes about who the real winner was in Florida, Louisiana, and Oregon. A commission was appointed consisting of 3 Republicans, 3 Democrats, and 1 independent to decide the winners in those states, and in the end they decided that all three disputed states rightly belong to Hayes, thus giving him the election. This result was, perhaps, similar to 2000, where Bush was declared the winner only after a contentious fight over who really won Florida. (Funny how poor Florida keeps popping up here.)

In 1888, Benjamin Harrison got 5,444,337 popular votes and 233 electoral votes, while Grover Cleveland got 5,540,050 popular votes but only 168 electoral votes. This was the only truly "clean" win for a candidate gained solely because we use the electoral college rather than a popular vote. And by the way, Cleveland went on to defeat Harrison in the next election.

Some implications

The electoral college system has a number of interesting, sometimes subtle effects on presidential elections.

Winning with a minority

Surely the most commonly noted oddity about the electoral college is that it makes it possible to win with less than a majority of the vote. Indeed, it is possible, at least in theory, to win with a rather small percentage of the total vote.

How small? What is the fewest votes you could get and still win? Think of it this way: To win you need a majority of the electoral votes. That is, you need to win in states totaling 50% of the population plus 1. But to win these, you only need 50% of the vote in each state. So in theory, you could win with just 25% of the vote. That is, if you got just 50% plus 1 in half the states while your opponent got the remaining 50% minus 1, and in the other half of the states your opponent got 100% of the vote, you would still win. If this isn't clear, let's take a simple example: suppose there were just four states, A, B, C, and D, as follows: (and let's assume each state gets 1 electoral vote for each 1 million voters)

State Voters Electoral Votes
A 15 million 15
B 11 million 11
C 6 million 6
D 8 million 8

Now suppose Smith and Jones run, and the election turns out like this:

Popular Vote Electoral Votes
State Smith Jones Smith Jones
A 7.6 7.4 15 0
B 0.0 11.0 0 11
C 3.1 2.9 6 0
D 0.0 8.0 0 8
Total 10.7 29.3 21 19
Percent 27% 73% 53% 47%

Smith wins the electoral vote with a comfortable edge, even though he got a mere 27% of the popular vote.

Indeed, the real situation is slightly more extreme than this. Recall that the number of electoral votes that each state gets is not directly proportional to its population. Rather, it is proportional to the states population plus two. The number of electoral votes a state gets is equal to the number of representatives plus the number of senators. The number of representatives is proportional to the population, but the number of senators is always two. On top of that, each state is guaranteed a minimum of at least one representative. This gives something of a bias toward small states. For example, California, the most populous state, has about 29.7 million people and gets 54 electoral votes, or 1 electoral vote for every 550,000 people. Wyoming, the least populous state, has 450,000 people and gets 3 electoral votes, or 1 electoral vote for every 150,000 people.

So what's the real minimum? In Chart 1 I list the number of electoral and popular votes for each state. If you win all the smallest states, then you get the highest ratio of electoral votes to popular votes. So in Chart 2 I list a set of states, starting with the smallest and working up, that are just enough to get a majority of the electoral votes. The result: it is possible to win the presidency of the United States with just 23% of the total popular votes.

When I discussed this article with my wife she pointed out that I am making an important assumption: that voter turn-out will be the same as it was in real life. The results could be even more dramatic if voter turn-out in the states that you win is very low, while in the states that you lose it is very high. After all, the number of electoral votes a state gets depends on its population as determined by the census, not by the number of people who actually vote. To take the far-out extreme here, suppose you won just the 11 biggest states: CA, FL, IL, MI, NJ, NY, NC, OH, PA, TX, and VA. This would get you 270 votes, just enough to win. You lose all the other states. In the other states, turnout is the same as it was in 1996, for a total of over 43 million votes. But in these 11 states, just one person in each state bothers to vote, and he votes for you. You then win the election with just 11 votes out of over 43 million cast.

Okay, that's a pretty unlikely scenario. But it makes the point that, if we take the possibility of variations in voter turn-out into account, it is possible to win with almost any arbitrarily small percentage of the vote.

Some red herrings

The "real" winner

After the 2000 election, many people pointed out that Al Gore had, in fact, received more popular votes than George Bush, and thus, he "should have been the winner", or he was the "real winner". After all, if we had a pure popular vote, then Al Gore would have won. At first this statement sounds like it is just obviously true, but a little thought will show that it is not. Yes, if we had a direct popular vote for president, and both candidates had campaigned exactly where, when, and how they did in real life, then Al Gore would have won. But that's a pretty meaningless fact. Suppose that two teams play a football game, and Team A wins. The coach of Team B says that if only field goals were worth 4 points each instead of 3, his team would have won. The only logical response would be, So what? If the rules were different, Team A would not have used the same game plan, and it is not at all obvious that they would have lost under the alternative set of rules. In any contest, I'm sure one could come up with some hypothetical alternative scoring system that would result in the other team winning ... if you assume that both teams play just as they would under the real rules. But no serious contender plays a game without considering what is worth points.

In the case of an election, the candidates surely know that we have this electoral college system. And given the millions of dollars that it takes to run a presidential campaign, I'm quite sure they take the time to carefully analyze just how it affects their campaigns.

For example: In this election, George Bush did very little campaigning in New York, even though it is the second most populous state in the country. Why? Because he figured out very early on that his chances of winning New York were extremely slim. Perhaps by working very hard and devoting lots of time and money to the state, he could have gotten 30 or even 40% of the vote. So what? He would still be very far from that magic 50% plus one vote that he would need to get anything of value from the effort. Thus, he wrote off New York very early. Both candidates made similar decisions about numerous states. Both devoted almost all of their efforts to states that were seen as reasonably likely to go either way.

But if we didn't have an electoral college, they almost surely would have behaved differently. Instead of concentrating on toss-up states, they would have concentrated on toss-up cities, toss-up ethnic groups, toss-up occupational groups, etc etc. The campaign would have been very different.

The founders didn't trust the people

A statement I have heard made many times by reporters and news analysts is that the framers of our Constitution created the electoral college because they didn't believe the people were smart enough to make a wise choice of president. Instead, they created this special elite who would make the choice in the name of the people.

There are two big problems with this theory.

The first is that it doesn't make sense. Even if it was true that the founders didn't trust the people to make a wise choice in this matter, how would this two-step system help? If they were afraid that, if allowed to vote for president directly, the people might vote for, say, a Nazi, then doesn't it follow that they would be just as likely to vote for electors who were Nazis? And then the Nazi electors would elect a Nazi president anyway. (Okay, that's a modern bugaboo, the Nazi party wasn't formed until well after 1789, but you get the idea.)

The second problem is that the founders wrote at great length about their reasons for structuring the Constitution as they did. They gave a number of reasons which made excellent sense given the political realities of the time. I am not aware of any surviving writing that mentions distrust of the people as a reason. Quite the contrary: One of the alternative suggestions offerred was to have Congress choose the president. But many of the delegates to the Constitutional convention worried that Congress might be too corrupt, or too mired in partisan differences, to make a good choice, and eventually the idea was voted down. (Of course, no one today worries about Congress being less than perfect ...)

Okay, one could speculate that if that was the reason they might have tried to keep it quiet. And we could speculate that they didn't think out the problem of who the voters would choose as electors. But for this theory to hold up we have to pile speculation on top of speculation. We end up with a theory where the only evidence to support it is guesses by people who weren't there about what the people who were there might have been thinking. We have to assume that the people who were there not only lied about why they did it, that all the hundreds of people involved all carried this conspiracy of lies to their deaths without anyone spilling the beans, and that it was all to protect a plot that didn't make any sense to begin with. This is just getting too far fetched to be taken seriously.

The real reason why we have an electoral college

So why did they create the electoral college? There were a lot of factors, but it basically came down to this: At the time the Constitution was written, there were serious concerns by the small states that if voting was purely by popular vote, their interests would routinely be over-ruled by the big states. Remember there were only 13 states back then, so each state was pretty much a "region" by itself, with distinct regional interests. The framers were worried that, in the extreme case, voters in each state might all vote for someone from their own state, and so whichever state had the most people would routinely succeed in electing someone from their own state. After all, the people of each state would naturally expect "one of their own" to best represent their interests and beliefs. And it would be difficult for people to fairly judge someone from another state: how much could they know about someone who lived hundreds or thousands of miles away? A local candidate would probably have had to work his way up through numerous minor offices, and so he would have a record that people could know and evaluate. (Even if he was not a career politician, to be a serious candidate he would have to be well-known as a businessman, military leader, or something.) All they would know about a candidate from a far-off state was what he said about himself when he came by giving campaign speeches, or what his opponent said about him, neither of which would be particularly reliable. This was, perhaps, more true in an era before television and the internet than it is today. But surely even today one would have to be incredibly naive to suppose that we really know much of anything about a candidate from some 30-second sound bytes on television.

So the framers idea was that the people would choose electors, and then the electors could meet, interview candidates, study their records, and so on, and make a wise choice. The electors would be local, so the people could know their records and choose people who would represent them well. But the electors would then have the leisure to travel around the country if necessary to learn about the candidates.

This solved the "knowledge" problem, but it still left the problem of big states versus small states. The big states wanted voting to be in proportion to population. The little states wanted each state to get the same number of votes. So they came up with a compromise: the system would be slightly skewed toward the small states, to give them slightly more votes than they would have gotten if it was purely proportional, but still not as many as the big states got.

In the context of the times, it made a lot of sense. Of course, half of the original idea has now been abandoned: the electors no longer make any decision, but are chosen purely on the basis of what candidate they pledge to vote for.

If the Constitution were being written today, similar challenges would have to be faced. How do we deal with the problem that all people know about a candidate is what he says in TV commercials or what a few reporters say about him? How do we protect the interests of minorities? Maybe today the issue wouldn't be different regions, but some other division: racial minorities, union members, farmers, religious groups, etc. Or it might be regional, for example, people in the South might worry about being out-voted by people in the Northeast. In the end, similar compromises would have to be reached.

Pros and cons

Is the electoral college obsolete? Should it be abolished?

The obvious argument against keeping the electoral college is that it allows a candidate to win even though his opponent got more votes. In a democracy, the goal is supposed to be to carry out the will of the majority.

On the other hand, this is what the electoral college was supposed to do. It might well be said that in the 2000 election, the electoral college functioned exactly as it was intended to. There was a sharp division between people who lived in big cities, like New York, Los Angeles, Detroit, and Chicago, who overwhelmingly voted for Gore, versus people who lived in suburbs and rural areas, who overwhelmingly voted for Bush. So Gore won in a relatively small number of states with big populations, like New York and California, while Bush won in a lot of states with small populations, like Wyoming and Mississippi. The electoral college system biased the vote slightly toward the rural people, and thus gave Bush the election. The original goal of protecting the interests of small states was achieved.

A second argument in favor of the electoral college is that it reduces the opportunity to challenge the outcome of an election. In 2000 we saw an incredible fight over the vote count in Florida. This was surely an undesirable thing. But suppose there was no electoral college, and the vote had been this close. Then we could have seen arguments like this over the whole country. That is, under the electoral college, there would have been little point in, say, Bush partisans fighting over a few thousand votes in New York, or Gore partisans fighting over a few thousand votes in Texas. In order to change enough votes to change the winners in those states, they would have had to claim errors or vote fraud on an absolutely massive scale, and somehow convince the elections officials, courts, and the general public that these charges were true. In most of the states that were close enough that claims of errors or fraud would be sufficient to change the outcome, like New Mexico and Oregon, it just didn't matter, they didn't have enough votes to change the outcome anyway. That was why there was such a fight about Florida: it was the only state that was very close, and that would change the outcome if it could be moved into the opposite column. But suppose there was no electoral college; suppose there was a straight popular vote. Then this reasoning would go away. A candidate would have every reason to fight tooth and nail in every state: by combining a few thousand votes in Boston, a few thousand in Chicago, a couple of hundred in Fulton, Missouri and Mechanicsburg, Pennsylvania, etc etc he might be able to put together enough to change the outcome.

Another argument against the electoral college is that it distorts the race: it's complex effects make the race more of a game that can be won by cleverly manipulating the rules, then a true measure of the candidate's popularity with the people. A candidate who devotes his resources to just barely winning in a sufficient number of states will defeat a candidate who continues to campaign in a state after he has already won enough voters to take it while neglecting another state where he is just a few short. Instead of concentrating on getting his message to all the people, a candidate is forced to carefully target his appeal.

On the other hand, a defender of the electoral college could reply that every election is just such a game anyway. This one just has slightly different rules from most. No candidate for a major office truly just goes out and tells people where he stands on the issue, without regard for who is listening. Candidates always carefully target their message: They know that some people already strongly agree with their agenda, and so their votes can be counted on, and they can generally be ignored. (Except for some occasional assurances that, yes, I haven't forgotten you, and if I sound like I'm "moderating" or appeasing the other side, don't worry, that's just campaign rhetoric, you can count on me to toe the line once elected.) They know that other people strongly disagree with their agenda, and so they must simply be written off. If you're black, you know you're not going to get the Klan vote, for example, and you just don't worry about it. So candidates concentrate their efforts on people in the middle, the moderates and undecideds. And they adjust their message to the interests of particular groups. Even assuming a candidate isn't going to lie and pander to the audience, assuming he's completely honest, still, if he's campaigning in front of a group that generally agrees with his position on A but disagrees with his position on B, he'd be a fool to give an uncompromising talk on B while never bringing up A. He will either emphasize how important A is and how they should support him to get these policies on A enacted, and/or he will try to explain that he's not really that far from these people on B and they shouldn't worry about this unimportant issue. Etc etc. Every campaign is a game in this sense.

In any case, one could argue that any changes to how elections are run is an uphill struggle. By definition, the group that is in power at any given moment presumably got there through the current system. If it was a winner for them, why should they want to change it? And if the people who are in power don't want to change it, how is it going to get changed?

Chart 1: Electoral Votes by State

This chart shows the number of electoral votes each state received following the (soon to be obsolete) 1990 census. The number of "popular votes" is the total number of votes cast for president (for all candidates) in the 1996 election. (I chose that year simply because it's the latest year for which I have a handy list of vote totals.)

State Electoral Votes Popular Votes (thousands)
AL 9 1,523
AK 3 229
AZ 8 1,387
AR 6 870
CA 54 9,646
CO 8 1,462
CT 8 1,359
DC 3 179
DE 3 268
FL 25 5,273
GA 13 2,281
HI 4 346
ID 4 485
IL 22 4,275
IN 12 2,118
IA 7 1,218
KS 6 1,063
KY 8 1,380
LA 9 1,764
ME 4 585
MD 10 1,764
MA 12 2,517
MI 18 3,808
MN 10 2,144
MS 7 886
MO 11 2,133
MT 3 403
NE 5 671
NV 4 447
NH 4 490
NJ 15 3,017
NM 5 537
NY 33 6,192
NC 14 2,502
ND 3 265
OH 21 4,491
OK 8 1,201
OR 7 1,309
PA 23 4,448
RI 4 382
SC 8 1,143
SD 3 321
TN 11 1,879
TX 32 5,575
UT 5 650
VT 3 249
VA 13 2,389
WA 11 2,165
WV 5 634
WI 11 2,144
WY 3 209
Total 538 94,676

Chart 2: A winning set of states

One could win the presidency of the United States by winning the following states. Beside each I list the approximate minimum number of votes you would need to win each state. Without doing an intensive calculation, I believe this is the minimum, or at least close to the minimum, percentage of the popular vote that one could take and still win the presidency.

State Electoral Votes Popular Votes (thousands)
AL 9 762
AK 3 115
AZ 8 694
AR 6 435
CO 8 731
CT 8 680
DC 3 80
DE 3 134
GA 13 1,141
HI 4 173
ID 4 243
IN 12 1,059
IA 7 609
KS 6 532
KY 8 690
LA 9 882
ME 4 293
MD 10 882
MA 12 1,259
MN 10 1,072
MS 7 443
MO 11 1,067
MT 3 202
NE 5 336
NV 4 224
NH 4 245
NM 5 269
ND 3 133
OK 8 601
OR 7 655
RI 4 191
SC 8 572
SD 3 161
TN 11 940
UT 5 325
VT 3 125
VA 13 1,195
WA 11 1,083
WV 5 317
WY 3 105
Total 270 21,655
Percent 50.5% 23%


Famighetti, Robert, ed. 1998 World Almanac and Book of Facts. Mahwah, NJ. 1997.

Madison, James. Notes of Debates in the Federal Convention of 1787. Norton edition pub 1966.

© 2001 by Jay Johansen


Hafida Jul 23, 2014

Lorem ipsum dolor sit amet, consectetuer adspiicing elit, sed diam nonummy nibh euismod tincidunt ut laoreet dolore magna aliquam erat volutpat. Ut wisi enim ad minim veniam, quis nostrud exerci tation ullamcorper suscipit lobortis nisl ut aliquip ex ea commodo consequat. Duis autem vel eum iriure dolor in hendrerit in vulputate velit esse molestie consequat, vel illum dolore eu feugiat nulla facilisis at vero eros et accumsan et iusto odio dignissim qui blandit praesent luptatum zzril delenit augue duis dolore te feugait nulla facilisi. Nam liber tempor cum soluta nobis eleifend option congue nihil imperdiet doming id quod mazim placerat facer possim assum. Typi non habent claritatem insitam; est usus legentis in iis qui facit eorum claritatem. Investigationes demonstraverunt lectores legere me lius quod ii legunt saepius. Claritas est etiam processus dynamicus, qui sequitur mutationem consuetudium lectorum. Mirum est notare quam littera gothica, quam nunc putamus parum claram, anteposuerit litterarum formas humanitatis per seacula quarta decima et quinta decima. Eodem modo typi, qui nunc nobis videntur parum clari, fiant sollemnes in futurum.

Add Comment