In this blog post, we explore some of the questions that clients invested in low-beta strategies have asked us: What should I do? Have the benefits of low beta been arbitraged away, or was 2020 driven by idiosyncratic reasons rather than structural changes? Are technology stocks the new downside protection?

The low-beta promise of better risk-adjusted returns is being challenged by recent performance.

Academics believe a myriad of theories for understanding why beta risk is not compensated in the stock market. One explanation for low beta outperformance is driven by the benchmarking of active management. Since active managers are typically anchored to cap-weighted benchmarks, they are less likely to subject their portfolios to a moderate-to-strong negative exposure to beta.¹ In addition, some individual investors tend to have lottery-type preferences, and in seeking high payoffs, they shun low-risk stocks.² These effects result in an opportunity for investors who do not face the same structural impediments or are not subject to the same biases. Given these explanations, investors in low-risk strategies expect higher risk-adjusted returns over long periods.

However, recent performance of low beta has lagged the market significantly.³ This is due to the strong market performance after the COVID-19-related drawdown in early 2020. From April 2020 to the end of June 2021, the S&P 500 returned 50.9%, compared to 29% for the iShares MSCI USA Min Vol ETF. Since, by construction, low-volatility strategies have low market exposures (the beta of the minimum volatility index to the S&P 500 is 0.75), some of this underperformance is expected. The expected returns to USMV over this period on a beta-adjusted basis were 38.1%. In a very strong recovery where risk-on strategies have dominated, it is not surprising that risk-off strategies like low risk have lagged.

Historically, low-risk stocks have had higher Sharpe ratios.

We take performance data aggregated by beta quintiles constructed by Kenneth French.⁴ As seen in Table 1, the annualized return to each of the beta buckets was relatively constant, with the lowest-beta stocks returning 12.28% per month and the highest-beta stocks returning 10.35%. While the long-run returns were approximately the same, the low-beta stock returns had approximately half the volatility of the high-beta stocks. The result was a Sharpe ratio for the low-beta stocks of 0.86 vs. 0.39 for high-beta stocks. This is the key finding that is unchanged since the key academic studies on the low-volatility effect: Higher-risk stocks have approximately the same returns as lower-risk stocks but have much higher risk—the exception is that for the very highest-risk stocks, returns are “abysmally” low.⁵ Thus, low-beta stocks are a more attractive investment from a Sharpe ratio perspective. Table 1 also reveals that the same pattern held for only large-cap stocks. Even over the last 10 years, despite the strong performance of the mega caps like FAANG (Facebook, Amazon, Apple, Netflix, and Google) stocks, the Sharpe ratio of low-beta, large stocks was still superior to that of the high-beta, large stocks.

Table 1: Sourced from Kenneth R. French–Data Library, 2021. All calculations use data from July 1963 to June 2021. For all stocks, calculations are first done within each size quintile. The size quintile summary statistics are then averaged to generate “All” stock statistics. For “Large Cap” stocks, only the calculations over the largest size quintile are reported.

But since 2020, things have been different.

Building on Table 1, Figure 1 looks at the difference between the average return to low-beta stocks and high-beta stocks over rolling 10-year periods. From this chart, it is evident that the current trailing 10-year spread in these returns is low but not a historic low. In February 2000, the spread in Figure 1 reached –11.04%. Today’s trailing 10-year spread is approximately 92 bps for all stocks, but even for large-caps stocks, the spread is –1.8%. Although the current differences are unusual, low-beta returns have climbed back from a deeper trough than today’s.

Figure 1: Sourced from Kenneth R. French–Data Library, 2021. All calculations use data from July 1963 to June 2021. Low-beta returns are taken from Table 1 then summarized over time. Similarly, high-beta returns are from Table 1. These high-beta returns are then subtracted from the low-beta returns to yield monthly low-beta returns relative to high-beta returns. Rolling 10-year geometric averages are then calculated for the above lines.

We’ve seen that the returns to low- and high-beta securities are similar over time, but they can diverge over specific periods. Low-risk strategies typically outperform in down markets and underperform in up markets (see Figure 2). The period since April 2020 has been a very strong up market, and thus it is not unusual for low-risk strategies to underperform. Given this difference in performance, low-beta investing involves taking on significant tracking error relative to a cap-weighted benchmark. To benefit from a low-total-risk strategy, investors should be patient: When markets enter a down period, low-risk strategies can outperform and help protect the portfolio. For investors that evaluate their active mandates using the information ratio statistic, which penalizes strategies with high active risk, low-beta strategies will likely be disappointing. Although they provide low total risk, they tend to have elevated active risk.

Figure 2: Sourced from Kenneth R. French–Data Library, 2021. All calculations use data from July 1963 to June 2021. S&P 500 data is from Morningstar direct. Index performance is for illustrative purposes only and does not reflect any management fees, transaction costs, or expenses. Indexes are unmanaged, and one cannot invest directly in an index. Past performance does not guarantee future results. The monthly returns to the S&P 500 from July 1963 to June 2021 are divided into quintiles, which define the buckets above. From there, the returns in Table 1 are averaged over each of the quintiles. Compounded and average performance across each of the quintiles is then reported. These averages do not match Table 1 because the averaging over quintile buckets yields different results from directly averaging over all months.

The moderate downside protection in 2020 was somewhat surprising.

Although low-risk strategies tend to outperform when the market is falling, there was almost no outperformance during the 2020 pandemic drawdown, where USMV returned –19.5% from January to March 2020, slightly outperforming the S&P 500 return of –19.9%. This result was well below historic norms of significant outperformance during similar periods. During the COVID-19 surprise drawdown, the types of stocks that were low risk shifted—tech stocks held up better than such historically low-beta stocks as Consumer Staples.

Idiosyncratic reasons or structural changes to the strategy?

Have the benefits of low-beta strategies been arbitraged away? Or was there something unusual about the market upturn from the second quarter of 2020 to today that makes the recent underperformance for low-risk strategies a one-off event?

We believe that the structural impediments of investment constraints that many institutions and active managers have traditionally faced are still binding. In today’s environment, individuals have gravitated to ever-more risky areas of the market. Thus, it is likely the economic causes for the low-risk premiums in historical data are still valid.

The market reactions to COVID-19 over 2020 and 2021 were unique, and they were also unique for low-risk strategies. When pandemic-related lockdowns hit, nearly all brick and mortar shops closed, and the world became virtual. Not surprisingly, volatile growth stocks, especially those providing tech to connect us in a locked-down world, dramatically outperformed. This “surprise” led to a boost in high-risk stocks as the crisis set in. As the world reopens, this shift is likely to prove temporary.

Investors should keep the following points in mind when evaluating their low-risk strategies:

• The recent underperformance is not without precedent.
• Investors have always needed to understand that these strategies are risky from a tracking error perspective. The strategies are most appropriate from a long-term perspective.
• The underperformance in the market rebound since the COVID-19 crisis onset is concerning but may be due to idiosyncratic reasons rather than structural changes to the strategy.

What should you do if you own a low-risk strategy?

If you believe in historic precedence and market cycles, and desire some protection against market downturns, maintaining exposure to low-risk strategies could be valuable. Past drawdowns of these strategies have been worse than the current underperformance. However, if you view the downturn in 2020 as evidence of a structural shift in the stocks that offer downside protection, then moving to a cap-weighted index makes sense. The key question is, will the new norm for downturns be that tech stocks outperform low-risk stocks such as Utilities and Consumer Staples? Possibly, but inferring that from one observation during a period that is nearly defined by the word abnormal is hard to do with certainty.

Conclusion:  No free lunch.

Higher risk-adjusted returns are attractive, but they are not a free lunch. In the short run, especially with strongly upward-trending markets, low-beta strategies can underperform. If investors can tolerate short-term tracking error, low-beta strategies can outperform over the long run. That protection can help investors stay the course and not divest during market crashes; that downside protection hasn’t been needed in a risk-on bull market. But over the long run, we may be glad to have some protection in our portfolios.

Send questions or comments to aperio.blog@blackrock.com.

¹ Malcolm Baker, Brendan Bradley, and Jeffrey Wurgler, “Benchmarks as Limits to Arbitrage: Understanding the Low-Volatility Anomaly,” Financial Analysts Journal 67, no. 1 (January/February 2011): 40–54.

² Kewei Hou and Roger K. Loh, “Have We Solved the Idiosyncratic Volatility Puzzle?” Journal of Financial Economics 121, no. 1 (July 2016): 167–94.

³ For example, from 07/31/2020 to 07/31/2021, the iShares MSCI USA Min Vol ETF was down 15.47% and the Invesco S&P 500 Minimum Variance ETF was down 10.92%.

⁴ Kenneth R. French–Data Library, 2021.

⁵ Andrew Ang, Robert J. Hodrick, Yuhang Xing, and Xiaoyan Zhang, “The Cross‐Section of Volatility and Expected Returns,” Journal of Finance 61, no. 1 (February 2006): 259–99.

MKTGH0921U/S-1847210