One way to manage the type of uncertainty found in financial markets is through redundancy. Since the 1950s and the introduction of modern portfolio theory, investment managers have worked to optimise their portfolios. What they inadvertently produce is fragility. Portfolios that are supposed to be insulated from fluctuations are actually at greater risk of failure.
Portfolios are optimised for either return or volatility. Portfolio managers focus on adjusting the portfolio mix between stocks, bonds and cash in order to maximise the return for a given level of volatility or minimise the volatility for a given level of return. If you imagine a curve rising toward the right and leveling off as it moves right, this is called the efficient portfolio frontier. In theory, t is not possible to move above the frontier, as it relates return to volatility. Portfolios below the frontier are unoptimised. The suggestion is made that moving a portfolio onto the frontier will increase the return for the same volatility, which is seen as gaining return at no “cost”.
It is easy to see that this view of optimisation is not always appropriate. Mutual funds, for example, need to maintain a cash balance in order to fund redemptions. Cash has no volatility, but also almost no return. It cannot be used to optimise a portfolio, because it doesn’t have a low or negative correlation with other asset classes. From the perspective of optimisation, it is seen as having no benefit, but mutual funds (as well as any opportunistic investor) has a need to hold cash. Further, optimisation has not worked. Correlations are not static, proving that the statistics that underly the modern portfolio theory are an inappropriate model of reality. In practice, all portfolios, optimised or not, present similar performance during market crashes: they all fall in value.
While optimisation focuses on reducing the fluctuations of a portfolio, robustness focuses on protecting value. Using redundancy, a portfolio has several parts, each of which has a different function and can profit in a different environment. The environment remains unpredictable, but the portfolio is prepared to profit. Investments can be classified into as many asset classes as you like. Let’s take stocks, bonds, cash and real estate as the most common examples. During some multi-year periods, stocks perform very well, far better than other asset classes. This was the case during the period from 1982 to 2000, with falling interest rates, low inflation and increasing leverage. During the same period, bonds performed quite well. But bonds shone during the period of 1999 to 2002, while stocks crashed and interest rates fell to rock bottom. Following that, real estate soared, especially in the United States, boosted by low long-term financing rates and increasing leverage. When the market fell apart, all types of equity and debt, except some forms of government debt, crashed at the same time and cash provided the best safe haven, outperforming almost all other investments.
It’s easy to look at a narrative like the one above and think: “If only we had known which asset class would perform well at which time. Aren’t there clues that make it obvious to the initiated?” The answer is no, because if there were, those people would have struck it rich, cashed in, and would either be well-known or no longer giving advice. A more realistic approach is to own each asset class. Own some stocks, some bonds, some real estate and keep some cash, just in case. We don’t do this to optimise the portfolio, but to be ready for any environment.
Further, a single investment may have multiple purposes in a portfolio. As an example, let’s look at a hypothetical portfolio for a retiree. Stocks provide dividends, which provide necessary income, as well as growth, which addresses inflation and longevity. Bonds also play multiple roles in the portfolio, providing guarantees for the value of the principal, providing some income, and moving opposite to stocks during a market correction. Cash provides a source to draw from when there is a need for a lump sum, as well as “dry powder” with which to buy during a market correction. Real estate is also useful to provide some income, while providing an effective hedge against inflation. As the economic environment shifts, the portfolio is able to provide for different needs.
This raises the question of proportions. The proportion is related to the purpose of the investor. In the example above, the retiree may be able to produce most of the income he needs from dividends. In this case, there is only a small need for bonds. The bonds should provide enough guaranteed capital to cover the likely length of a market correction, probably 3 to 5 years. Assuming a withdrawal rate of 5% (or less), having 15% to 25% of the portfolio in bonds would be adequate. This is less than the traditional 40% – 50% recommended at retirement, because the purpose is conceived differently.
This way of building a robust portfolio using redundancy isn’t very different from traditional portfolios. A robust portfolio holds a variety of assets, but the purpose is different from modern portfolio theory. Instead of seeking to optimise return and minimise volatility, the assets are chosen and proportions determined based on the purpose of the investor. A retiree will have various sources of income, where a growth-oriented investor will have various sources of growth. The portfolio will always maintain a portion in cash to take advantage of opportunities. The portfolio will never be optimal and may leave some potential returns untapped, but it should provide the most effective protection against total failure.