Andrew's Angle

Style Factors and Elections

Oct 19, 2020

The capacity of factor strategies appears large 

Since 1788, every four years Americans have voted to choose their federal government—through wars, depressions, disease, and natural disasters. Millions of Americans march to the polls or mail in ballots, and many stay up through the night to see who will become the next President of the United States. The 2020 US election will be unique. Never has a Presidential election been held during a global pandemic, where election results may take more than a day, week, or month to be settled. However, since we’ve had over 50 Presidential elections through American history, we can say some things about how the market factor and style factors—like value, quality, momentum, low volatility, and small size—have fared over election periods and under different political regimes.

Periods around presidential elections are volatile

In Figure 1, we show how the US market has had lower returns and higher risk from September through November in election years, going back as far as 1926. Over long periods, US stock market volatility has been 21% in election months compared to 17% in non-election months. The higher volatility in election months has also corresponded to lower returns: 3% in election months vs. 10% in non-election months.

This higher volatility around elections is not surprising. There’s uncertainty about whether a party will stay in power, or whether there will be a change of government. Even an incumbent victory doesn’t guarantee a President won’t pursue different policies during a second term in office.

Figure 1: Returns and volatility of the market

The style box, circa 1992

Source: Fama and French (https://mba.tuck.dartmouth.edu/pages/faculty/ken.french/data_library.html). U.S. Market represented by the market return of all NYSE, AMEX and NASDAQ Firms, value weighted. Data period from July 1926–August 2020. For election month return, periods from September-November in election years were isolated and annualized.. The same process was done with the remaining months for non-election months. Performance does not reflect any management fees, transaction costs or expenses. Past performance does not guarantee future results.

If investors want to mitigate some of that potential volatility, defensively oriented factors such as quality and low volatility can be potential risk-reducing factors. Both these defensive factors have meaningfully outperformed during election months when market volatility was heightened. The style factors shown here increased their average volatility in election months by just 16% compared to market volatility increasing by 27%. Interestingly, the momentum factor has also fared well around election months.

Figure 2: Returns of style factors

The Factor Box shows exposures across six factors

Source: Fama and French (https://mba.tuck.dartmouth.edu/pages/faculty/ken.french/data_library.html) and AQR (https://www.aqr.com/Insights/Datasets) data libraries. Size represented by SMB, Value by HML, Quality by RMW and Momentum by MOM in the Fama and French data set. Low Volatility represented by BAB in the AQR data set. SMB represents small minus big companies. HML represents high book-to-market minus low book-to-market companies, RMW represents robust operating profitability minus weak operating profitability companies. MOM represents high price momentum minus low price momentum companies. BAB represents Low beta minus high beta companies. Data periods for Size and Value are from July 1926–August 2020. Quality from July 1963–August 2020. Momentum from November 1926–August 2020. Low volatility from December 1930–June 2020. For election month return, periods from September-November in election years were isolated and annualized.. The same process was done with the remaining months s for non-election months. Performance does not reflect any management fees, transaction costs or expenses. Past performance does not guarantee future results.

Insights in red and blue markets

While this may be no indication of future returns, our results show that historically markets have done better under democratic presidents—but the market posts positive returns for both parties. Timing and business cycles have likely played at least some role in these results, but academics have found that the democratic presidential premium persists even when accounting for the business cycle, the stance of monetary policy, and other control variables.1

Figure 3: Equity market return by presidential party

The Factor Box shows exposures across six factors

Source: Fama and French (https://mba.tuck.dartmouth.edu/pages/faculty/ken.french/data_library.html). U.S. Market represented by the market return of all NYSE, AMEX and NASDAQ Firms, value weighted. Data period from July 1926–August 2020. Days among republican and democrat presidents were isolated and annualized.

Performance does not reflect any management fees, transaction costs or expenses. Past performance does not guarantee future results.

How have style factors performed when different parties have held the presidency? Figure 4 shows that except for size, all factors have been able to post positive premiums whether under a democratic or republican president. Interesting to note, conventional wisdom may suggest that republican regimes benefit smaller-cap securities through a potential focus on lower taxes and deregulation. However, historically the size premium has been negative under republican presidents. On the other hand, the momentum factor has performed significantly better under republican administrations.

Figure 4: Style factor returns by presidential party

style factor returns by presidential party

Source: Fama and French (https://mba.tuck.dartmouth.edu/pages/faculty/ken.french/data_library.html) and AQR (https://www.aqr.com/Insights/Datasets) data libraries. Size represented by SMB, Value by HML, Quality by RMW and Momentum by MOM in the Fama and French data set. Low Volatility represented by BAB in the AQR data set. SMB represents small minus big companies. HML represents high book-to-market minus low book-to-market companies, RMW represents robust operating profitability minus weak operating profitability companies. MOM represents high price momentum minus low price momentum companies. BAB represents Low beta minus high beta companies. Days among republican and democrat presidents were isolated and annualized. Performance does not reflect any management fees, transaction costs or expenses. Past performance does not guarantee future results.

The long-term vote is for your portfolio

While the election approaches, prepare to buckle in for a period of elevated volatility. Investors may consider allocating to quality and minimum volatility to potentially mitigate some of the potentially higher volatility during election months. Investors should consider staying the course, however, as volatility has historically declined after the election.

Republican or democrat – the most important thing is for long-term investors to stay diligent. However, as you can see in Figure 5 below, equity markets have wavered and rallied under both political regimes while consistently marching higher. This serves to remind investors that markets tend to go up, and that its about time in the market, rather than timing the market, that may be the most prudent way to meet long-term financial goals.

Figure 5: Time in the market, rather than timing the market

Time in the market, rather than timing the market

Source: Fama and French (https://mba.tuck.dartmouth.edu/pages/faculty/ken.french/data_library.html). U.S. Market represented by Mkt-Rf with the risk free rate added back. Data period from July 1926–August 2020. Days among different presidents were separated and annualized using 252 days and compounding.

Andrew Ang
Andrew Ang
Head of Factor Investing Strategies
Andrew Ang, PhD, Managing Director, coordinates BlackRock’s efforts in factor investing. He leads BlackRock’s Factor-Based Strategies Group which manages macro and style ...
Factor videos
Factor videos
Tools
Tools
Factor commentary
Factor commentary