Left or Right, and Up or Down
August 8, 2006 • Posted in Political Indicators
Should investors lean toward governments at one end of the country political spectrum to find outperforming equity markets? In their October 2003 paper entitled “The Presidential Puzzle: Political Cycles and the Stock Market”, Pedro Santa-Clara and Rossen Valkanov examine monthly U.S. stock market performance versus executive branch party across 18 Presidential elections (1927-1998, 864 months) encompassing 10 Democratic and 8 Republican Presidencies. In their July 2006 paper entitled “Political Orientation of Government and Stock Market Returns”, Jedrzej Bialkowski, Katrin Gottschalk and Tomasz Wisniewski investigate whether the political orientation of 173 different governments systematically affects the performance of 24 international (mostly European) stock markets. Findings are:
“The Presidential Puzzle: Political Cycles and the Stock Market” concludes that:
- The broad value-weighted stock market has beaten the 3-month Treasury bill (T-bill) rate by an average of about 11% under Democratic Presidents and just 2% under Republican Presidents — an economically and statistically significant 9% difference. Equal-weighting of stock returns produces a difference of 16%, with small-capitalization stocks strongly outperforming large-capitalization stocks. (See the chart below.)
- The difference in returns holds for sub-periods 1927-1962 and 1963-1998, and is generally independent of business cycle explanations.
- The difference in returns is not election-centered. It accumulates gradually across Presidential terms, suggesting it is due to systematic surprises in economic policies.
- Volatility of returns is somewhat higher during Republican than Democratic Presidencies, indicating that stock market outperformance during the latter is not reward for risk.
- The party in control of Congress does not indicate significant differences in excess stock market returns.
- Given the limitations of the data, the impact of party-in-Presidency on the stock market in this study could be a statistical fluke.
The following chart, taken from this paper, shows the average annualized excess (versus the T-bill) value-weighted returns for each Presidential term during 1927-1998. Darker shading denotes Republican Presidencies. The dash-dotted line indicates the the average excess return for the entire period. Democratic (Republican) Presidencies tend toward higher than average (significantly lower than average) excess returns.
In summary, Democratic Presidencies are probably good for the the U.S. stock market.
The authors conjecture that party-in-Presidency affects the stock market through fiscal and regulatory policies.
The approach of this paper assumes two reasonably consistent economic policy regimes, one Democrat and the other Republican, alternately in power across 864 months of stock market returns. However, the sample spans only seven changes in party-in-Presidency. It may be more appropriate simply to test returns from change to change, a very small sample, rather than month to month or term to term. Also, the periods from election to economic policy implementation to economic policy efficacy can be many months, even years, suggesting that economic policy effects substantially lag party-in-Presidency. It is arguable that investors could be eliminating the lag by substantially anticipating policy implementation and effects.
The following chart shows two perspectives of the change in the Dow Jones Industrial Average (DJIA) across the seven regimes covered by the paper, plus the balance of the Democrat’s last regime and the to-date part of the current Republican regime. One perspective uses DJIA performance roughly coincident with regimes. The other lags DJIA performance by 12 months, arbitrarily assuming that actionable evidence of policy changes from regime to regime take one year to emerge. Neither the results nor the sample size justify assignment of stock market outperformance to a political party.
“Political Orientation of Government and Stock Market Returns” extends the study of regimes versus stock markets to an international setting, concluding that:
- There are no statistically significant differences in returns between left-wing and right-wing regimes.
- Investors do not adjust their discount rates based on election outcomes.
- International investment strategies based on the political orientation of countries’ leadership are likely to be futile.
- The anomaly observed in the U.S. appears to be country-specific.
In summary, economic activity generally transcends politics in politically free societies.
This bottom-line conclusion further suggests that the U.S. result may be a fluke.