We’re approaching an election year. That means lots of arguments and name calling, and that just describes my family gatherings! Which political party is the best for the stock market? It’s a simple question with a not-so-simple answer.
Everyone has probably seen bar charts showing things like S&P 500 Index returns during Democratic and Republican administrations. These charts are useless for making investing decisions. Why? Most simply, the president doesn’t determine the markets’ returns. Markets are complex systems driven by many forces. Who is president may matter, but maybe only a miniscule amount. Policies matter more than party labels. Which policies get passed depends also on the composition of Congress.
Here’s an illustration of some of the complexities around looking at charts that purport to prove one party is superior to another.
First, how do dividends and inflation factor into the charts? Most comparisons look at the price return of the S&P 500 Index under various administrations. The problem is that inflation was very low in the 1950s but very high in the 1970s. Therefore, a percentage-point return in the 1970s wasn’t worth as much as a percentage-point return in the 1950s. Dividends are also an important source of returns, which is why I looked at the total return (dividends included). Note: This does not adjust for taxes. If you were to look at these returns on an after-tax basis, that opens up a whole new can of worms. So, what does the inflation-adjusted total return look like under various presidents?
We’ll address that question soon, but first, what start date should you use when trying to attribute equity market returns to a president? Should the clock start on Inauguration Day? Or should it start from the day a president is elected? Most analyses just look at calendar-year returns, which would mean, for example, that all of 2009 belonged to President Obama, even though he didn’t take office until January 20, 2009. Do those first 20 days of 2009 belong to President Bush or President Obama? Markets are forward-looking, so shouldn’t stock prices move in anticipation of what the president might do? Arguably, I think it makes more sense to start the clock with the election, which is the first Tuesday after the second Monday of November of years divisible by four. Using that approach, here’s a comparison chart: The inflation-adjusted average annual return of the S&P 500 Index by party of the president from 1857 (or, 1856, when looking at the election-year data) to 2014.
When results flip-flop that easily, you should really question the robustness of the results. Part of the problem with such data is the variability in returns, which means the averages aren’t as significant as they appear.
Earlier, I mentioned the importance of Congress’ composition in influencing which policies get passed while a president is in office. Let’s go back to that. When the executive and legislative branches are controlled by one party, you are likely going to see less gridlock than when the executive branch is controlled by a different party than Congress. Here are the average annual inflation-adjusted total returns, depending on the party composition of the executive and legislative branches.
The highest returns are when congress is split (one house is Republican-controlled and the other is Democrat-controlled) and the president is Democratic. A unified Republican government is a close second. Don’t read too much into these results, though. The differences aren’t that significant, especially when you consider that only 12 years out of the 157 years’ worth of data we’ve measured fall into the Democratic/split category, and 20 years fall into the Republican/split category.
So what does this election year hold in store, in terms of how the market will perform as one presidential administration transitions into another? If you’re looking for a clear forecast, history isn’t a very good guide, as either another Democrat will succeed President Obama or a Republican will. Only four times since 1920 has a Republican followed a Democrat and the average real return for the election year was 7%. Having another Democrat follow a Democrat is rarer. In fact, you could argue that it hasn’t really happened. Truman took over from FDR after FDR died in office. Lyndon Johnson took over from Kennedy after Kennedy was assassinated. Both Truman and Johnson were reelected with real S&P 500 Index returns of 17% and 15.2%, respectively, their first election years.
|Presidential transitions||Average annual real total return||Number of times since 1856|
|Democratic to Republican||6.6%||9|
|Republican to Democratic||-1.5%||8|
|Democratic to Democratic||NA||0|
Source: Global Financial Data and author’s calculations
No matter how you slice it, and there’s a lot of slicing that goes into looking at this data, an election year isn’t something that worries me too much as an investor. Fundamentally, I tend to believe that policies are more important than party labels. Party labels tend to be broad and shifting umbrellas: Calvin Coolidge was a very different Republican than Herbert Hoover or Richard Nixon. Bill Clinton was a very different Democrat than Franklin Delano Roosevelt or Lyndon Johnson. When it comes to the markets, elections matter, but the relationship between who wins and whether investors win isn’t always neatly summarized by the election results. Vote, but don’t let your politics taint your portfolio.