Investment portfolios will typically contain a number of risk factors, including equity market beta, liquidity, interest rates, credit spreads, currency risk, inflation and commodity risks. Tail risk hedging should seek to address all of these factor risks.
For a tail risk hedge to be effective, it should possess several important characteristics — a tail risk hedge must be negatively correlated to asset returns and exhibit strongly convex behavior to the upside during periods of market stress. In effect, its characteristics must mirror those of the tail risks inherent in a portfolio. Typically, implementation of tail risk hedging will involve the use of derivatives which exhibit convex behavior. For equity tail risk, these hedges may take the form of equity options. Similarly, for interest rate and credit risks, swaptions and credit default swaps may be used.
To develop an understanding of the intuition behind tail risk hedging, consider “regime-switching” as a model for financial market behavior. This model states that markets are well-behaved most of the time, when asset correlations are stable and asset return distributional assumptions are approximately normally distributed. This is termed the “good” regime. However, on rare instances, markets switch to a highly stressed environment marked by illiquidity, falling asset prices and significantly higher asset correlations. This is the “bad” regime, and the fall of 2008 is a perfect example. After a period of time, the model suggests that markets switch back to the “good” regime, and the cycle continues forever. Forecasting an imminent regime switch is great, but acknowledging and preparing for it is equally important.
There are three methods by which investors with a modern mean-variance asset allocation can deal with a regime-switching model for financial markets. The first is simply to ignore the “bad” regime; that is, realize that diversification and allocation benefits will protect an investor most of the time but losses are inevitable. This is essentially the passive approach that most pensions and endowments knowingly or unknowingly take toward market declines like what was experienced over the past year. To borrow a term, this is the “fair-weather” approach to asset allocation and portfolio management.
Another option taken by some is an “all-weather” approach that seeks to optimize performance over both “good” and “bad” regime scenarios. Separate distributional assumptions and correlations in the allocation are used to determine optimal portfolios within each regime, which are then combined together based on probability estimates of the likelihood of each regime. The resulting portfolio of this middle-of-the-road approach is actually suboptimal most of the time given that the markets are normally behaved. In exchange for underperforming during the good times, an investor will outperform during the bad times. While this approach has its merits, there are serious issues to consider (e.g., will boards and investment teams have the conviction to accept lagging performance vs. peers during the good times in exchange for the outperformance during tail events).
The third approach seeks to implement a “fair-weather” asset allocation while integrating tail risk hedging into the portfolio. This is achieved by setting aside a portion of expected return for the purpose of purchasing hedges that will protect the portfolio from the adverse conditions during a tail event. Intuitively, an investor gives up a small part of the portfolio expected return in exchange for lower portfolio expected volatility and truncated market losses. This “barbell” approach to asset allocation integrates easily into existing asset allocations and does not conflict with objectives of investment teams. As discussed below, it will be shown that integration into current investment policies is fairly straightforward and traditional investment team incentives need not be affected by explicitly separating the cost and benefits of tail risk hedging.