Executive Summary
Sequence of return risk is different from most other forms of risk borne by investors. It is not a risk inherent in an asset class, nor can it be seen in any portfolio analysis that looks at a single period of returns. As a result, it is not much of a surprise that sequence of return risk is entirely ignored in much of academic finance. But it is a meaningful risk for the vast majority of investment portfolios and there are useful tools that can mitigate its effects. We believe a portfolio construction framework that takes into account the lifespan of the portfolio and its expected cashflows can account for sequence of return risk better than any standard single-period optimization. And by dynamically reallocating portfolios in light of the returns assets have actually delivered and the consequent impact on their expected returns, portfolio managers can substantially further improve outcomes for their clients. Adopting these tools can be difficult, as both require combating natural human behavioral tendencies. But investors and their advisors would be well-served by at least understanding the ways those tendencies lead to sub-optimal long-term outcomes and determining whether it is worth some of the emotional pain involved in taking a different path.
As one of us points out relentlessly, risk isn’t a number, rather it is a notion or a concept. In the past, we have talked about the permanent impairment of capital as being the true risk you want to avoid. We have also highlighted the three paths that can lead to this outcome:
- Valuation risk. Buying an asset that is expensive means that you are reliant upon all the good news about that asset being delivered (and then some). There is no margin of safety in such investments.
- Fundamental risk. As Ben Graham put it. “Real investment risk is measured…by the danger of a loss of quality and earnings power through economic changes or deterioration in management.” We often talk of depression risk as a “deep” risk to equities because a depression can absolutely destroy the cash flows of stocks.
- Financing Risk. This is essentially anything that can force you to exit a position before you would otherwise choose to. For example, if you are using leverage, you can be forced to sell an asset because of, say, a margin call. One of the key features of financing risk is that it introduces path dependence.
In financial planning circles, one prominent “risk” discussion focuses on the sequence of returns. We will argue that sequence risk is actually a form of financing risk. The order in which returns occur will matter if (and only if) money is being added to, or withdrawn from, a portfolio. If no money is being added or withdrawn, there is no sequence risk. In this sense, sequence risk is not a risk inherent to a portfolio (as are valuation risk and fundamental risk) but depends on the external cash flows of the investor and is thus a financing risk. Hence the specifics of the risk – i.e., whether it’s better to have good returns followed by bad or bad returns followed by good – depends on whether you are withdrawing or contributing to your account.
But simply because sequence risk is a financing risk doesn’t make it any less real from an investor’s perspective. The evidence that financing risk/sequence risk matters is all too clear. Studies such as Morningstar’s “Mind the Gap” regularly show that investors achieve somewhere between 1 to 2 percentage points per year less in terms of return than their investment funds achieve – a gap generated by unfortunately-timed buy and sell decisions. Compound that over the kind of horizons involved in saving for retirement and it is little wonder that sequence risk is oft discussed by practitioners.
However, financing risk tends to get short shrift in academic finance circles. 1 It is generally assumed to not be an issue for anyone who is not investing with borrowed money. And indeed, for a “steady state” unlevered portfolio where money is neither being added nor withdrawn, there is no financing risk and no path dependence. But in the real world almost no portfolios fit that steady state model: as soon as money enters or exits a portfolio, financing risk and its consequent path dependence become an important issue.
Practitioners have long realized the salience of this path dependence. But understanding the existence of a problem does not imply success in mitigating it. Unfortunately, the tools of academic finance weren’t generally designed to handle this problem, leaving practitioners largely on their own to deal with sequence risk through heuristics and oversimplified simulations.
So, the question remains: how do we deal with it? This missive lays out two potential ways to mitigate sequence risk. The first is to start the investment process by asking the right question.
Framing Risk
When building a portfolio, it is important to clearly define the objective: what are you, the investor, trying to accomplish? In this sense, the portfolio is the answer. In standard finance theory, optimal portfolios are constructed to efficiently trade off return and volatility. This theoretical underpinning forms the basis for interactions with clients whereby questions are framed in terms of a client’s risk tolerance and/or risk capacity. This “risk tolerance” is then generally translated into a volatility limit (at odds of course, with the idea expressed above that risk isn’t a number). In this sense, the answer is driving the question, rather than the other way around.
As Inker and Tarlie (“Investing for Retirement: The Defined Contribution Challenge”) discuss, falling short of your financial needs is a much more intuitive framing of risk. In this framing, the right question is, “What do you need, and when do you need it?” It is not that volatility is inconsequential, as it is an important characteristic of the portfolio. However, volatility is not the primary driver of the portfolio. Instead, the primary driver is minimizing shortfall relative to suitably chosen targets that reflect the client’s wants, needs, and circumstances.
[Before moving on, it is worth pointing out that defining shortfall relative to suitably chosen targets does not mean that minimizing shortfall applies to problems only where the objective is to avoid falling below a basic level of subsistence. Because shortfall can be defined relative to any future set of targets spanning the investor’s horizon, the framework is useful, for example, for wealthy investors looking to compound their wealth for future bequests. Circumstances such as this are often framed in terms of risk capacity. But the concept of risk capacity is really just the other side of the “what you need/want” coin. One advantage of framing the problem in terms of needs or wants is that it is much more intuitive: asking someone what they need or want is a much easier conversation than asking them what their risk capacity is. Another advantage is that the needs/wants framing systematizes the portfolio construction process, directly aligning the financial plan with the portfolio in a way that the risk tolerance/risk capacity framing cannot.]
Framing risk as “not having what you need when you need it” and then constructing portfolios that minimize this risk does two things. First, the “what you need” question is the flip side of the risk capacity coin, framing the issue in terms of return rather than risk. Second, the “when you need it” question brings horizon into the problem. And bringing horizon into the problem can naturally address sequence risk. This is because when you frame risk in terms of shortfall, horizon enters the picture in two ways. The first is the total portfolio horizon, e.g., how long you expect to live. However, there is another horizon, and that is when you start caring about shortfall. And it is this second horizon that is the key to addressing shortfall risk.
If you have a long horizon until you care about shortfall, e.g., a 45-year-old saving for anticipated retirement at age 65 has 20 years until shortfall becomes a concern, sequence risk is not as big a deal as it is when your shortfall horizon is shorter. 2 As the 45-year-old ages and approaches retirement, concern about total portfolio horizon shrinks naturally. And as this horizon shrinks, a portfolio that minimizes shortfall reduces its weight in stocks in a way that still accounts for how much longer the 45-year-old expects to live.
This pattern is evident in the glidepath in Exhibit 1, which illustrates how the optimal shortfall portfolios change with age, assuming: 1) all assets are fairly priced; and 2) in the pre-retirement phase the asset owner desires to compound at 5% over inflation, and in the post-retirement phase at 3% over inflation. 3 In this chart we see that the weight in stocks falls at an accelerating rate as retirement age approaches, which, from the perspective of sequence risk, is an intuitive pattern. Framing risk as “not having what you need when you need it” and constructing portfolios that minimize this risk naturally accounts for the increased saliency of shortfall risk at retirement. In this way, framing risk as falling short of what you need and when you need it addresses, at least partially, sequence risk. 4
Continue reading here.
