Wall Street’s predictive power

Every presidential election cycle we are invited to ignore the regular polls and look instead for the likely winner in the sales of dumb things, like bobble-head dolls, candidate T-shirts, colas and so forth. While these are stupid fun, there is one type of prediction market that is serious business: Wall Street stocks. 

Our supposed understanding of the predictive power of equity markets is a little more established than fortune-telling predicated on bobble-head sales. Or so I thought. But the more I read, the less sure I am that we know anything at all about the exact nature of the relationship between the stock market and politics.


It seems that the longest-standing notion is that the market’s trends, particularly in the final few months before the election, have been a good predictor of the outcome. Generally speaking, if there is a bullish market, the party in power is rewarded. But if the bears are bloodying Wall Street, the challengers prevail.

This pattern has occurred almost 90 percent of the time, say those who have studied the numbers. While some analysts don’t really speculate how this happens — they just accept the predictive power for what it is — the assumption of most who do explicate is that voters feel confident about the economy, so they buy stocks and also vote to reelect incumbents. That would make some sense. But it collides with other notions we have about how the market works, especially for individual investors, and it cannot possibly square with data about the limited involvement that most voters have in the market today.

Gallup does an annual poll asking Americans about their stock ownership. Lately, in the wake of the economic bust, the percentage of Americans in the market has declined dramatically, from 65 percent in 2007 to only 54 percent in the latest survey, taken last April. Because Gallup’s survey is of all adult Americans, not voters, it might underestimate the electorate’s involvement in Wall Street. But, at the very least, it suggests that a sizable chunk of voters is not likely to be calibrating their politics with their portfolios this November.

Furthermore, we know that most of those who own stocks are not really traders who follow the markets regularly. They don’t know bull from bear. They are just workers with a 401(k) that someone else manages. So, again, can we say that their investments are actively influencing their votes, or vice versa? Certainly in some cases that would be true, but realistically it seems dubious for most workers.

And even if some voters are active investors with a savvy take on markets, can we determine how that will affect their voting? The American Association for Individual Investors, with members who are just such people, are regularly polled about their perceptions of the market. In the last poll, taken Feb. 22, 29 percent said they are neutral about the market over the next six months, neither bearish nor bullish. That’s a lot of serious investors who won’t know how to vote in November unless something gets clearer.

The other problem I have with all this is the direction of the causality. Are voter investors going to react to the market with their votes, or is their reaction to the market presaging their perceptions of the election outcome? President Obama’s supporters will say that a bull market over the next few months will demonstrate his success, thereby inducing voters to reward him with reelection. Republicans, on the other hand, might say that a bull market is the result of investors sensing that Obama is going to lose and that his Republican successor will create a more business-friendly climate. A bull market cannot favor both parties, so someone is going to be wrong.

David Hill is a pollster that has worked for Republican candidates and causes since 1984.