Social Mood Conference | Socionomics Foundation
By Matt Lampert
Originally published in the October 2012 Socionomist


According to a historical review conducted by the Socionomics Institute, the presidential party in power has little bearing on the stock market. On the contrary, as our recent paper shows, social mood has substantial bearing on both the stock market and who is elected president.

Nevertheless, every U.S. election cycle brings new discussions about which party’s presidential candidate will benefit or hurt the stock market. Recent articles have claimed that since 1926, the stock market has gained much more per year on average when the president was a Democrat than when he was a Republican. Does this claim stand scrutiny?

It is true that from 1926 to 2011, the Dow’s average annual rate of return under Democratic presidents was 9.77%, while under Republicans it was 4.57%, less than half as much. But what if you look at all the relevant data?

The Republican Party was founded in 1854 and ran its first candidate for president in the election of 1856. It therefore makes sense to analyze the data forward from 1857, the year that the winner of the election of 1856 took office, in order to incorporate the results of every presidential election in which the Democrats and Republicans competed against each other.

We did so and found that from 1857 to 2011, there was a 6.92% average annual rate of return in the stock market under presidents who were Democrats, while under Republicans the rate was 5.90%.

From a socionomic perspective, it is not particularly surprising to find that when all the data are considered, the market has done about equally well under both parties. Still, the Democrats appear to have a narrow lead. Is this lead robust? No. When we take a closer look, even that small difference disappears. Anytime you look at averages, a few extreme values can have a substantial impact on the results. Once you remove them, you get a clearer picture of whether a relationship truly exists. Such is the case here.

From its high in 1929 to its low in 1932, the Dow fell a whopping 89%. No other four-year period in U.S. history comes close to such a move. Thus, the average annual rate of return under Republicans jumps to 7.68%, slightly higher than the 6.92% performance under Democrats, when we omit 1929-32 from our analysis.

All of the advantage in the market’s average performance under the Democrats over a period of 154 years is due to the change during just four years, a time when Herbert Hoover, a Republican, happened to be president. In other words, removing just four years out of 154 swings the benefit in the Republicans’ favor, indicating that the Democrats’ advantage in the overall data series is not robust.

If we also omit the next four years, 1933-1936, when Democrat Franklin Roosevelt was president, the data favor the Republicans even further by 7.68% vs. 5.25%.

This article is a further reminder that when it comes to data on any topic, including presidential elections and the markets, it is important to remember that the conclusions analysts draw may be largely due to the time periods they’ve studied. When you read qualifiers such as “Since 1926…” or “Over the past 50 years…” ask yourself why the claim utilizes a cut-off point not required by the data. At least when analysts look at all the data, you can be more confident that their results are not an epiphenomenon of a particular sub-period and data-mined to produce a certain result.

To read the Institute’s paper on U.S. presidential elections and social mood, click here and then select the “Download This Paper” button at the top of the page.


An Interesting Aside
There is some discrepancy about to which party one of the presidents belonged. Andrew Johnson, who was elected as Republican Abraham Lincoln’s vice president on the National Union Party ticket and who took over the presidency in April 1865 after Lincoln was assassinated, spent nearly his entire life as a Democrat. However, some historians note that during his presidency Johnson acted as an independent who unsuccessfully attempted to turn the National Union Party into a major political brand. Listing Johnson as a Republican instead of a Democrat leaves the gap between party results from 1857-2011 essentially unchanged (Democrats, 7.02%; Republicans, 5.87%).

Matt Lampert is the Socionomics Institute's Research Fellow at the University of Cambridge.

Access 6 years of still-relevant Socionomist archives — in just one easy step >>

(Socionomist subscribers: Log in for your complete, exclusive archive.)