Building a crash proof strategy

The authors of the May 2019 study “The Best of Strategies for the Worst of Times: Can Portfolios be Crisis Proofed?” Campbell Harvey, Edward Hoyle, Sandy Rattray, Matthew Sargaison, Dan Taylor, and Otto Van Hemert looked into the performance of a range of defensive strategies.

Between 1985 and 2018, they looked at both active and passive investments, with a focus on the eight biggest drawdowns (when the S&P 500 Index fell by more than 15%) and three U.S. recessions (8 percent of the full period).

The Findings

  • Continuously holding short-dated S&P 500 put options is the most reliable defensive instrument, but it’s also the most expensive method (-7.4 percent return over all periods). While it performs well during crashes (earning an average of 42.4 percent), it is very expensive during “normal” periods (losing 14.2 percent on average), which account for 86 percent of the sample, and expansionary (non-recession) times, which account for 93 percent of the observations. As a result, passive option protection appears to be too costly to be an effective crisis hedge. They pointed out that trading options is costly, therefore the returns would have been considerably poorer after implementation costs.
  • Futures time-series momentum (which benefits from lengthy equities selloffs) and a quality strategy (which benefits from a “flight-to-quality” impact during crises) are two dynamic strategies that have done well during previous drawdowns. There is a tradeoff between stronger downside protection and higher long-term returns, with shorter-term momentum strategies giving more downside protection but lower long-term returns.
  • In terms of performance during crashes, a beta neutral long-short quality strategy outperforms the rest, with a 32.1 percent return. During normal times, however, it only returns 1.7 percent, for a total return of 5.6 percent. Because time-series momentum and quality have historically uncorrelated returns, they can be used as complimentary crisis-hedge components in a portfolio.
  • Long credit protection (short credit risk) strategies gain during each of the eight equity drawdown periods, albeit in a more unequal fashion, doing particularly well during the credit crisis of 2007-2009. While the credit-protection strategy is less expensive in normal times and non-recessions than the put-buying strategy, it has a negative total return (-3.6%) since it earns 39.6% in a crisis but loses 9.8% in normal times.
  • Holding “safe-haven” U.S. Treasury bonds generates positive carry, but it may be an unreliable crisis-hedging strategy due to the post-2000 negative bond-equity correlation, which is a historical oddity. They noted: “It is only during the drawdowns after 2000 that bonds performed well. During earlier drawdowns, the performance of bonds was mixed, and over the Black Monday period, the bond return was -8.3%.” They added: “The bond performance is consistently positive during the three recessions.” They also suggested: “As we move beyond the extreme monetary easing that has characterized the post-Financial Crisis period, it is possible that the bond-equity correlation may revert to the previous norm, rendering a long bond strategy a potentially unreliable crisis hedge.” 
  • Only those strategies with 100% hit rates and relative hedging efficacy for drawdowns were considered, and the clearest and most compelling winners were L/S quality and long puts, with short credit risk a close second.
  • In terms of returns, 1 and 3 month times series momentum via futures, as well as L/S profitability, were somewhat effective methods. Long bonds and gold performed poorly in comparison to the other strategies.

When they looked at the benefits of combining the two methods of time-series momentum and quality (in equal proportions), they discovered that a 50% allocation was needed to minimize drawdowns, but even a 10% allocation lowered the average drawdown by around 8%, from 44% to 36%.

Average drawdowns would have been decreased to 19% with a 30% allocation. They also discovered that allocating to the two “hedging” strategies enhanced overall performance.

While it was conceivable to hedge crash risk, the authors determined that it was extremely costly in terms of overall return (which should be expected, as insurance comes at a cost). Holding Treasuries is also a risky hedge.

They went on to say that dynamic, rather than systematic, techniques have historically given downside protection as well as increased long-term returns.

The message for investors is that the research shows that, at least historically, allocating to time-series momentum and quality (defensive) stocks not only improves returns but also considerably reduces tail risk, which can lead to financial plan failure.