Book Review: Reducing the Risk of Black Swans
Book Review: Reducing the Risk of Black Swans: Using the Science of Investing to capture returns with less volatility
by Larry Swedroe and Kevin Grogan
The book is thought provoking from start to finish. For context, readers should understand the concept of a “Black Swan”, introduced in the brilliant Nassim Nicholas Taleb’s book, “Fooled by Randomness” and expanded on in “The Black Swan.” The idea is that we tend to assess risk based on what has been observed in the recent past, rather than on what could or could have happened. Some events may be so rare, that there is no record of them occurring… yet. These could be extremely impactful. The term has been expanded (inappropriately in my opinion) to mean very rare events generally. Swedroe and Grogan examine how to protect a portfolio against large declines in value.
In the Foreword, Ross Stevens lists exceptionally low interest rates as a major factor in how people approach risk. Not only do low rates push investors into riskier assets seeking a return on investment, but low rates themselves create additional risk in fixed income, which is generally considered a safe asset class. He notes that the duration (a measure of how long your money is committed) has increased from 3.7 years to 6.0 years in the aggregate bond index, effectively levering up the portfolio by 62%. In simple terms, the average interest rate risk in the bond market is much higher, at the same time interest rates got to extremely low levels.
The authors start by explaining why historical stock returns are unlikely to continue. The price paid per dollar of earnings has increased tremendously over time, particularly since the start of the 1990’s. This is not only unlikely to continue, but it makes future returns harder as investors get less earnings per dollar invested.
The authors look at ninety years of data, Over that period, the total stock market earned 8.4% per year more than government bonds. The smaller half of companies returned 3.3% per year more than the larger half, and the cheapest 30% of stocks made 5.1% per year more than the most expensive 30%. There is more risk in smaller and cheaper stocks, but the point is that by taking different risks uncorrelated to market risk[1], as long as the risks are compensated, investors can reduce exposure to just market risk. Importantly, for risk diversification to work, the risks can’t be correlated to each other. In this case, the risk factors have shown to be independent over time. However, in order to stick with a portfolio exposed to multiple risk factors, investors must be willing to endure periods of lagging the market return.[2] The authors then demonstrate using historical data, how a portfolio with size and value exposure can take less market risk and attain a similar return with less total risk. This is important because in a large market crash the investor will have less exposure to stocks. This is a different way to think about diversification. Rather than diversifying across all asset classes and styles, this approach diversifies across risk factors, seeking exposure to factors that have historically generated returns with low or no correlation to the market.
In part two of the book, the authors examine alternative investments to seek returns that are uncorrelated to the equity market. The first of these is alternative lending. This is peer-to-peer lending, or lending that bypasses banks and all the associated costs of operating a bank. Over its short history, the authors found that alternative lending experienced returns comparable to stocks, with much lower volatility. Loans tend to be short-term, so interest-rate risk is low. Because these loans would tend to underperform during periods of economic stress due to credit risk, the authors suggest taking from the equity allocation. There are a few funds available that invest in alternative loans. Because the earnings are taxed as ordinary income, this goes best in retirement accounts. The second alternative category is reinsurance. This is selling insurance to insurance companies. Reinsurance pays out in extreme catastrophes such as wildfires, earthquakes and hurricanes. Catastrophe bonds are bonds that pay a high interest rate, which is reduced if catastrophes occur. These have averaged a higher return than the stock market (though the authors only expect a comparable return going forward) with lower volatility and low correlation to the market. The correlation did rise in the ’08 financial crisis, as these bonds use collateral, and thus have risk to a properly functioning financial market. Investors can buy individual “cat” bonds, or use a fund that buys these bonds. Because the funds are illiquid, they cannot be held in retirement accounts (unless the investor has a “self-directed” account.) The third alternative investment is more complicated. It is called the Variance Risk Premium (VRP). This is based on the phenomenon that options sellers historically have had a positive return (and options buyers a negative return). Because people are risk-averse, they will pay more than the expected value for someone else to take risk from them. Since VRP investing involves accepting the downside risk in an asset class (e.g. taking risk of a really big loss), it is important to diversify across many asset classes. The authors suggest one interval fund for diversified VRP investing. The fourth alternative investment is also very technical. It invests according to styles or factors (e.g. value, momentum, carry) that have earned a positive return over time, but doing so through holding the favorable investments and being short the unfavorable investments. This allows investors to benefit from both overpriced and underpriced securities, and to earn the various risk premia without taking market risk. The authors recommend a specific fund from AQR that is invested across premia, asset classes and geographies. They expect this fund to continue to achieve market-like returns with about half of market risk, and no correlation to the equity market. This means the fund can make money in up or down markets. The fifth alternative investment is time-series momentum. This holds investments (e.g. asset classes, sectors, etc.) while they are going up and shorts them as they are going down. It is pure trend-following. This can be done through futures contracts, so this is often called “managed futures.” Again, because the strategy has the ability to go long or short, it can make money in up or down markets. The strategy can diversify across a lot of different markets and asset classes. It can also adjust leverage to hit a desired risk target. The authors suggest an AQR fund that is expected to continue to earn a return comparable to stocks, with about half of the volatility and little to no correlation.
Finally, the book has a short chapter on implementation. The authors prefer using the higher expected returns from risk factor exposure to reduce the overall equity allocation. They also suggest 10-30% allocation to alternative assets classes, equally weighted across all the alternatives. Whether this exposure is taken from the equity or fixed income exposure depends on how much exposure the investor has to each. Proportionally reducing exposure between both to deploy in alternatives makes sense.
Conclusion: Reducing The Risk of Black Swans looks at how to go beyond just a stock/bond portfolio to reduce risk of major loss by taking exposure to other, uncorrelated risks. Investors would do well to gain exposure to time-tested risk-factors that are relatively uncorrelated to stocks such as value, small companies and momentum. Using alternative asset categories that have historically earned equity-like returns with below equity risk and no equity correlation can further diversify the risk exposure an investor takes. Investors should be careful, however. While some of the recommended alternative investments have done very well, their history is very short, and they have not been widely held. As investment classes go from fringe to mainstream, they often get revalued upward, boosting returns. Investors who get in after an investment is widely held will earn a lower yield than those who got in during the boutique stage, and will also not enjoy the multiple expansion the first investors got. Of note, most of these investments have not done well since the book was published – this may or may not be predictive of future results. For instance, the authors note that the Swiss Re CAT Bond index averaged 8% per year from 1/2002 to 6/2016. They use the Stone Ridge fund (SRRIX) with clients. This fund has an annual fee of a whopping 2.43%. Over its eight-year life, it has averaged just 0.12% annual return from 10/2013-10/2021. Since 10/16, it has averaged a loss of about 4.5% per year. The alternative lending fund recommended by the book (LENDX) has an annual expense ratio of 4.03%. Part of this is interest expense on borrowed funds to lever up the portfolio, but even without that the annual expense is well over 3%, making it hard for an investor to get a good return. Even so, it has averaged over 10% per year in its first six years, through 2/2022. Data is not yet available to see how this leveraged loan fund did over the last year when interest rates and credit spreads rose. I am skeptical that management can take 3-4% per year out of the interest payments and still continue to make an attractive return for investors without taking significantly more risk. During a strategy of low interest rates (cheap leverage) and a calm economy, this type of strategy should work great, but when the cost of leverage increases significantly, economic conditions worsen, and struggling small businesses can neither pay off loans nor refinance, defaults could spike. Since these are mostly short-term loans that were offered when rates were low, the risk of borrowers not being able to refinance could be acute.
Importantly, correlation is not static. Some asset classes may have low correlation to stocks during normal or moderately stressful times, but may be highly correlated if the financial system breaks. Alternative lending in particular, has not been around long enough to see how it will handle a period of extreme economic stress. The authors note that Cat Bonds did poorly in the Great Financial Crisis due to counterparty risk, which wouldn’t be observed when looking at correlation during normal times. VRP would also likely increase in correlation during times of extreme stress.
Investors in alternative asset class must be willing to accept high fees, illiquidity, and potentially a decade of poor relative returns while stocks are doing very well. This is hard to do. Liquidity is most valued in market crashes, so even if these alternative funds do protect value, it may be inaccessible when it is most needed. The alternatives are interesting, and should diversify risk most of the time, but may fail to provide the required protection during a Black Swan event. Very high fees could eat up most of the return above short-term government bonds. Investors should be cautious on alternatives.
Overall, the book is a great exploration of how to reduce risk in a portfolio without reducing expected return through an insightful and effective way of viewing diversification. Investors should carefully consider the recommendations offered in this book.
[1] Market risk, also called equity risk premium, market beta, or simply beta is a measure of how much a stock or portfolio will change in value in relation to the total market changes.
[2] This is hard because people measure success based on what happened, not based on what might have happened. For instance, if Portfolio A has a 50% chance of earning 8% and a 50% chance of losing 8% and Portfolio B has a 50% chance of earning 10% and a 50% chance of losing losing 30%, investors in Portfolio A will tend to feel like they made a bad choice if the market goes up and they only made 8%, ignoring the risk they avoided, since the bad outcome didn’t materialize this time.