Are you wondering how the upcoming U.S. presidential election might affect your investments? This is not a new kind of concern. Elections are stressful times, and the heightened rhetoric from across the political spectrum doesn’t help. With so much at stake, it may seem obvious the outcome could impact the stock market.

I cannot turn down the volume, nor am I going to take a political side. But I can tell you what the data say about the relationship between U.S. presidential elections and U.S. stock markets. Bottom line, what may seem so “obvious,” may not be. Not even when the election could be a tight one.

Post-Election Views

Let’s start with a presidential election’s short-term effects on stock market returns. I have compared the returns for the 12 months starting in November, for both election and non-election years from 1926–2019.

Source: Benjamin Felix; Data Source: Dimensional Returns Web

On average, the 12 months following a U.S. presidential election (the green bars) have delivered slightly lower returns, at 10.6% versus 11.9% for all the other November–October 12-month periods (the gold bars). Of the 23 post-election 12-month periods, seven of them had negative returns, while the rest were positive. The negative post-election periods were in 1936, 1940, 1956, 1968, 1972, 1976, and 2000.

How does the market decide whether an election is good or bad news, at least in the short run? Each investor has their own political biases, which affect their view of how the outcome should impact the economy and the stock market. A person with one set of views will tend to look favourably on their party’s win. Others might see it just the opposite.

In other words, a win for either party will not be universally viewed as a good or bad thing. The end result? The range of opposing views can have a muting effect on asset prices. This was studied in detail in a 2012 paper, “Political Climate, Optimism, and Investment Decisions. The authors looked at a large sample of data from Gallup surveys, the National Longitudinal Survey of Youth, and portfolio holdings and trading data from a large U.S. discount broker.

They found individuals became more optimistic and perceived the markets to be less risky when their political party was in power. This is important to keep in mind before worrying that asset prices are going to collapse following an election result. The market aggregates the expectations of all participants, not just those aligned with a given political view. Prices are set at the equilibrium of all expectations. Based on this, we wouldn’t generally expect major stock market events to stem directly from election results, which is exactly what the data show.

Stock market volatility, on the other hand, may increase around elections, especially tight ones. In a 2019 Journal of Index Investing paper, “With Greater Uncertainty Comes Greater Volatility, the authors found that the U.S. Economic Policy Uncertainty Index seemed to spike around tight presidential elections, such as in 2000, 2004, and 2016; and that stock market volatility seemed to be linked to both economic policy uncertainty and the business cycle.

Source: With Greater Uncertainty Comes Greater Volatility; Inna Zorina, Jamie Khatri, Carol Zhu, James J. Rowley; The Journal of Index Investing Nov 2019, 10 (3) 6-14; DOI: 10.3905/jii.2019.1.077

Market Politics

Forget about short-term expectations. A more interesting question might be how the political party in power affects stock market returns throughout a presidential term. Do politics make a longer-term difference in U.S. stock market outcomes?

In fact, they seem to, but with a twist. In a 2003 Journal of Finance paper, “The Presidential Puzzle: Political Cycles and the Stock Market, authors Pedro Santa-Clara and Rossen Valkanov examine the stock market through election cycles from 1927–1998. Again, they found no significant evidence of stock price changes immediately before, during, or after elections. This finding is consistent with a 1989 paper, “What Moves Stock Prices? which found important news does not tend to be related to large stock market returns.

However, in their samples, Santa-Clara and Valkanov did find that stock markets delivered much higher returns on average when Democrats were in power. Under Democratic presidents, the excess return of the U.S. market over three-month Treasury bills was on average 9% higher per year than under Republican presidents. They observed a consistent increase in this effect related to company size. The largest firms had an annual excess return of 7%, increasing to 22% per year for the smallest firms.

This was a puzzle, because the difference in returns was not explained by business-cycle variables, and was not concentrated around election dates. So how do we explain the results?

 

Piecing Together the Presidential Puzzle

Before concluding there is a causal relationship between positive stock market outcomes and Democratic leadership, there is a more theoretically consistent explanation to consider. In a 2017 paper, “Political Cycles and Stock Returns, authors Lubos Pastor and Pietro Veronesi update and respond to the conundrum.

First, they updated the stock return analysis under Democratic and Republican leadership to cover a longer timeframe, from 1927–2015. They found the excess return under Democrats was 11% higher per year than it was under Republicans over the full period. In fact, they found all the equity premium over the 89 years had been earned under a Democratic president. This statistically significant relationship in their out-of-sample test seems to confirm Santa-Clara and Valkanov’s original work.

Source: Pastor, Lubos and Veronesi, Pietro, Political Cycles and Stock Returns (May 26, 2019). Chicago Booth Research Paper No. 17-01, Fama-Miller Working Paper , Available at SSRN: https://ssrn.com/abstract=2909281 or http://dx.doi.org/10.2139/ssrn.2909281

A persistent stock market performance difference based on political leadership poses a theoretical problem. For larger stock returns to persist based on Democratic policy initiatives, the market would need to be consistently underestimating their positive economic benefits. Pastor and Veronesi instead suggest an explanation that aligns with a rational stock market:

 

It is not Democratic policy that results in positive excess returns, but the timing of when Democrats have been elected.

To test this explanation, Pastor and Veronesi developed a model of political cycles driven by time-varying risk aversion. In their model:

  • When risk aversion is high, such as during economic crises, voters are more likely to elect a Democratic president because they demand more social insurance.
  • When risk aversion is low, voters are more likely to elect a Republican because they want to take more business risk.
  • If risk aversion is higher under Democrats, a higher equity risk premium exists when Democrats are elected, which explains the higher average returns.
  • But the higher risk premium is not caused by the Democratic president. Both the higher risk premium and the Democratic presidency are caused by higher risk aversion leading up to the election.

 

A Model in Action: Additional Evidence

Pastor and Veronesi went on to show their model’s findings align with the empirical literature on when Democrats are more likely to get elected.

  • In a 2016 Journal of Financial Economics paper, “Time varying risk aversion”, the authors used survey data to show that risk aversion surged after the 2008 financial crisis, even among investors who did not experience losses.
  • In a 2010 paper, “Partisan Financial Cycles”, Lawrence Broz examined bank crises in developed countries and found that left-wing governments are more likely to be elected after financial crashes.
  • In a 2012 American Political Science Review paper, “Unemployment and the Democratic Electoral Advantage”, John R. Wright showed that U.S. voters tend to elect Democrats when unemployment is high.

Anecdotally, we see similar results:

  • During the Great Depression, incumbent Republican President Herbert Hoover lost the election to Democrat President Franklin D. Roosevelt in November 1932.
  • During the Great Recession, incumbent Republican President George W. Bush lost the election to Democrat President Barack Obama in November 2008.
  • President John F. Kennedy was elected in 1960, during the 1960–1961 recession.
  • President Jimmy Carter was elected in 1976, shortly after the 1973–1975 recession.
  • President Bill Clinton was elected in 1992, shortly after the 1990–91 recession.

In all of these cases, elections took place when voters were likely to be more risk-averse than usual. Pastor and Veronesi argue this is not a coincidence. They suggest it is explained by their model relating risk aversion to election outcomes.

 

Near and Far: Only Time Will Tell

Everything I just covered was a long way of saying, there have been statistically reliable differences in stock returns under different political party leadership. But it is reasonable and theoretically consistent to believe these differences do not result from the political leadership. Rather, the political leadership results from the same conditions that have led to historically higher stock returns.

In other words, the stock market is going to do what it’s going to do. The economic conditions that lead to higher risk aversion, and therefore higher expected stock returns, have also tended to favour Democratic platforms.

Beyond that, it’s anybody’s guess whether the relationships will persist in the future – even the near future of the November election. I’m not predicting election or stock market outcomes based on this information. I’m merely stating an empirical fact: Stock market outperformance under Democratic leadership can be reasonably explained by the level of risk aversion at the times when Democrats have tended to be elected.

So, as we approach this year’s election, it is important to remember:

In the short term: Election outcomes have not had any meaningful relationship with stock market returns. Tight presidential elections like those from 2016 and 2000 resulted in increased volatility as the market priced in economic policy uncertainty. But it all came out in the wash pretty quickly.

In the longer-term: While there seems to be a positive correlation between Democratic leadership and stock market returns, the causal relationship is much more likely between risk aversion and Democrats being elected, which also happens to result in a higher equity risk premium over those periods.

At the end of the day: The equity risk premium has been persistent through time for Democrats, Republicans, and any other investor who has stayed invested long enough to capture it.

 

The main takeaway is this: None of this is useful for trying to time the market. Risk premiums show up quickly and unexpectedly regardless of the political cycle. You have to be there to capture them, which means staying invested. U.S. and Canadian citizens alike are better off staying invested over time, and dedicating their time and energy instead to voting sensibly for the candidates of their choice.