Subscribe to DSC Newsletter

Elections and the markets, avoid silly data-mining charts edition (from the Financial Time)

Every time there’s an election in the US, analysts and pundits start data-mining through previous political cycles to predict what could happen next to the markets.

This kind of thing is more prevalent before presidential elections, but congressional elections aren’t completely immune.

So we’ve always been sceptical about the utility of charts like this new one from S&P (hat tip Reuters), and this time is no different:

If all goes as expected, the category you’re looking for is Split Congress, Democratic President — a combination that has produced an average annual loss of 3.7 per cent.

But so what? That combination has only happened in two prior years — not exactly a hefty sample size, and we won’t even get into variance or other issues. Reuters goes on to note in the same article:

Yet other historical indicators point to a good 2011 for stock market investors. The best year in the political cycle for investors is typically the third year of a president’s term: Prices have moved up in 94 percent of those years, averaging a 17 percent annual return. And “we’re in the sweet spot right now,” Stovall told Reuters. The best quarters in the stock market typically come in the fourth quarter of a mid-term election and the first two quarters of that post-election year. That all points to nice returns for the end of 2010 and 2011.

In other words, whatever happens next will break from one tradition or another, and charts like this don’t help you know which — or tell you much of anything, really. They’re just something interesting if pointless for market historians to keep track of.

We last railed against this kind of data-mining just before Barack Obama was elected, right after John Tammy and Rob Arnott had written this:

Are Mr Obama’s soaring prospects for victory causing global markets to crash? Or is the global market crash driving voters into Mr Obama’s camp? Is the global recession and liquidity crisis both creating the market crash and setting the stage for an Obama victory? Statistics alone cannot answer these questions. But many readers may be surprised to learn that they support all three.

There is more than a 20 per cent correlation between today’s market action and the changes in Mr Obama’s prospects, with markets falling when these prospects improve.

We’re not endorsing one side or the other here, just pointing out how wrongheaded this kind of exercise can make people look.

And that said, here is how the S&P 500 has performed since Obama was sworn in:

Related links:
Obama, the markets and the perils of presidential data mining – FT Alphaville
Elections and the markets – FT Alphaville

Views: 170

Comment

You need to be a member of AnalyticBridge to add comments!

Join AnalyticBridge

On Data Science Central

© 2020   TechTarget, Inc.   Powered by

Badges  |  Report an Issue  |  Privacy Policy  |  Terms of Service