The Best Strategies for the Worst Crises

Overview

It is notoriously hard to find an affordable yet effective hedge against large equity market selloffs.

• Holding, and continuously rolling, at-the-money S&P 500 put options is costly, but the most reliable of the candidate hedge strategies considered.
• Holding ‘safe-haven’ US Treasury bonds is an unreliable hedge generally, despite providing a positive and predictable long-term yield, since the post-2000 negative bond-equity correlation is a historical rarity.
• We argue that long gold and long credit protection sit between long puts and long bonds strategies in terms of both cost and reliability.

We investigate two dynamic strategies that, in contrast to these passive investments, appear to have generated positive performance both in the long-run and particularly so during historical crises:

• Futures momentum has parallels with long option straddle strategies, allowing it to benefit during extended equity sell-offs.
• The quality stock strategy takes long positions in highest-quality and short positions in lowest-quality company stocks, benefitting from a ‘flight-to-quality’ effect during crises.

These two dynamic strategies historically have uncorrelated return profiles, making them complementary crisis risk hedges. We examine both strategies and discuss how different variations may have performed in crises, as well as normal times, over the years 1985 to 2016.

Introduction

Investors typically hold a large allocation to equities, which makes strategies that have the potential to do well during equity market sell-offs valuable diversifiers. In this paper, we examine active and passive strategies that hold potential to perform positively during the worst crises. We focus on liquid strategies with intuitively-sound features, such as long volatility or an overweight to less risky securities. The performance of the strategies is analyzed from a historical perspective, looking at seven instances between 1985 and 2016 where the S&P 500 fell by more than 15%.

Section 2 begins with an examination of two types of passive strategies. First, approaches that benefit directly from a falling market. A strategy that buys, and then rolls, one-month, at-the-money S&P 500 put options performs well in each of the seven crises. However, it is very costly during the ‘normal’ times, which constitute 86% of our sample, and as such seems too expensive to be a viable crisis hedge. A strategy that goes long credit protection (short credit risk) also benefits during each of the seven crises, but in a more uneven manner, benefiting particularly during the 2007-2009 Financial Crisis, which was a credit crisis. On the other hand, the short credit risk strategy is less costly during normal times than the put strategy.

Next, we consider so-called ‘safe-haven’ investments. A strategy that holds long positions in 10-year US Treasuries performed well in the post-2000 equity crises, but was less effective during previous crises. This is consistent with the negative bond-equity correlation witnessed post-2000, which on further analysis appears atypical from the longer historical perspective. 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. A long gold strategy generally performs better during crisis periods than at normal times, consistent with its reputation as a safehaven security. However, its appeal as a crisis hedge is potentially diminished by the fact that its long-run return, measured over the 1985-2016 period, is close to zero and that it carries substantial idiosyncratic risk unrelated to equity markets.

We then turn our attention to dynamic strategies. Certain dynamic strategies – such as shorting currency carry or taking long positions in on-the-run Treasury bonds against short positions in off-the-run bonds – may perform well during crisis periods, but are expensive in the long-term. Because of the costs involved in managing any active strategy, we choose to focus only on those that are, at the least, positive in expectation before costs.

In Section 3, we consider futures time-series momentum strategies, which historically have been shown to provide “crisis alpha” potential.1 Hamill, Rattray, and Van Hemert (2016) argued that this is because trend followers add to winning positions (ride winners) and reduce losing positions (cut losers), much like a dynamic replication of an option straddle strategy. We show that such strategies performed well over the seven equity crises. We also explore various methods to limit the exposure to equities, with a view to enhancing the crisis performance potential. We find that simply not allowing long equity positions worked as well as more complicated approaches, such as capping the beta of the strategy to the equity market at zero. We argue that this is because estimation is challenging when imposing beta restrictions, given that betas vary through time, and possibly abruptly so when entering a crisis.

In Section 4, we consider long-short US equity strategies. A review of the factors proposed in the academic literature suggests that those that take long positions in high-quality and short positions in low-quality companies are most promising as crisis hedges, since they can potentially benefit from flights to quality when panic hits markets. The definition of a quality business is, of course, open to debate. However, broadly speaking, such companies will be profitable, growing, have safer balance sheets, and run investor-friendly policies in areas such as payout ratios. We examine a host of quality metrics, and examine improvements that may be made to strategies from employing more sophisticated portfolio construction techniques than are typically found in the academic literature.

Finally, in Section 5 we show that futures time-series momentum strategies and quality long-short equity strategies are not only conceptually different, but also have historically uncorrelated returns, meaning that they can act as complementary crisis-hedge components within a portfolio. We demonstrate the efficacy of the dynamic hedges through some portfolio simulations.