The Electoral College as a Weighted Market

Barack Obama led the popular vote by 7% on November 4. A hefty margin for sure in a presidential race, but nothing at all compared to the electoral vote turnout: a good 37% margin. And except with rare occasions where the granularity of the state system falls on the other side of a close race (i.e., the 2000 election), electoral vote margins are almost always starkly amplified over the popular vote margin. Why is this, and what does it tell us about markets?
Earlier I wrote about Efficient Market Theory and its shortcomings in failing to describe overweighted information in stock prices. The example used there was horse betting, but the electoral college is also a great, and much more accessible example, whose design highlights the way information may become overweighted.
The key to this overweightedness hinges on the fact that the popular vote is not a probability. A 53%-46% spread does not mean that John McCain has a 46% chance of winning. As a matter of fact, if we are to look at the electoral college as a measure of the probability of either candidate winning, John McCain was severely overrepresented in the college with respect to the poll numbers. If supporters of McCain and Obama were homogeneously distributed throughout the country, even a 1% edge in the popular vote would give Obama the entire college. In a winner-take-all system, a few points (the margin of error) away from 50% in the polls and probability of a win drops precipitously.
The exact same effect can be observed in the House and Senate with regard to party control and vote outcomes. Each additional Democratic congressman shifts the aggregate passed legislature 0.45% in the House and 0.63% in the Senate in the liberal direction, according to one measure of such things (Stimson, 2008) - until the halfway point. The single congressman that gives the Democrats control of Congress carries with him a 48.9% shift in policy in the House, and a 34.8% shift in the Senate (conversely, the same is true for Republican control in the opposite direction). Legislation, like the presidency, is winner-take-all, and the only probability that matters is at the margin when the proportion is 50%.
Predictions in the stock market are similarly boolean, in that investors decide whether a particular stock will return a profit or a loss over the course of their investment. They are not interested in how it meanders up and down along the way except insofar as that represents a trend upwards or downwards - at least no more than the popular vote matters to the election. A stock price carries no more information about the future performance of that stock than the electoral college contains about the popular vote: it too is not a probability. And also like the electoral college, these predictions influence the outcome. Buying a stock on the expectation of a positive return will cause its price to rise, just as casting a vote for a candidate will increase his popular vote proportion. Finally, a stock in price/earnings equilibrium (that is, when all information is present and weighted optimally) can be thought of as at the 50% margin with regard to gains vs. losses, in that, according to Efficient Market Theory, its price is equally likely to go up or down depending on news - basically, any news while the stock is at equilibrium is that last senator that pushes the stock into positive or negative territory.
Stock movement is thus decided at the margin, where the expectation boolean is determined. If enough stock is sold at once in a particular company because of a bad piece of news, the stock price drops and the boolean turns negative, which tells people that the expected earnings no longer justify the (original) price - it will generate a loss with respect to the overall market if bought at the current price. But this triggers more people sell while they have the chance, and before long, the negative data has driven the price down to where it is justified by the new expected earnings. Just like the electoral college, a company need only move a few points down from price/earnings equilibrium to initiate a precipitous drop in stock price.
SHARE{ DiggRedditNewsVineStumbleUponSlashdotFacebookTechnoratidel.icio.usMa.gnoliaLive }



