Determining the degree of risk that is appropriate and tolerable in clients’ financial plans is central to an advisor’s role. I will show how advisors should deconstruct risk into six components and then integrate them using a framework to provide the best recommendations for clients.
I'll focus this analysis around the time of retirement, with clients needing to plan how to spend down savings and how much investment risk to take. Clients need to develop retirement plans that will achieve their financial goals, but they also need to be able to psychologically withstand the market volatility that such plans may entail. A retirement plan that meets the financial plans on paper but results in the client bailing out when markets are down is not viable.
Plans should first be developed with a focus on the financial goals. The psychological issues can then be considered, perhaps calling for cutbacks in risky investments and setting more realistic financial goals. Ideally the advisor and client would handle the psychological issues in a way that minimizes their disruption on the financial plan.
This idea of separating the financial and psychological issues was recently proposed by Michael Kitces in a blog post. He raised the concern that many of the tools used to assess risk tolerance actually attempt to measure the combination of risk tolerance and risk capacity. He demonstrated how these combined measures can send the wrong signals and lead to inappropriate investment advice. The methodology I propose builds on Kitces' recommendation of separating risk tolerance and risk capacity.
Risk components
A client's overall risk profile consists of six components. A brief description of each component follows:
Risk tolerance – the psychological capacity to tolerate turmoil in investment markets without undue anxiety.
Risk perception – the client's view of the riskiness of financial markets, which may be influenced by current market volatility.
Loss aversion – a behavioral economics concept that describes the tendency to strongly prefer avoiding losses to acquiring gains.
Risk aversion – a classical economics concept that describes the reluctance to accept uncertain payoffs in favor of certain payoffs with lower expected values.
Required risk – the amount of investment risk a client needs to assume in order to build a viable financial plan.
Risk capacity – the financial capability to withstand losses without undue disruption to financial plans.
The six different risk components are a recipe for confusion. I'll attempt to sort them out.
Building a financial plan
I'll first demonstrate how the approach that I favor for building retirement plans effectively deals with risk aversion, required risk and risk capacity. I recommend a two-step process of building a plan that first deals with these three components before addressing the other components.
There are a variety of different strategies for generating retirement income. I favor the essential/discretionary approach. The particular version I prefer uses guaranteed lifetime income sources including Social Security, pensions and annuities to cover basic living expenses. Remaining funds are used for discretionary spending, which adjusts each year based on investment performance. The choice of the stock/bond mix for discretionary funds depends on how comfortable the client will be making year-to-year changes in discretionary spending. For example, a high stock allocation for the discretionary funds would suit an individual able to tolerate significant volatility in annual spending to achieve higher average spending.
With this retirement income approach in mind, we can evaluate how well it addresses the three risk components noted above. We will start with risk aversion, an esoteric economics concept tied to utility theory. In the context of retirement planning, it can be thought of as spending flexibility. A highly risk averse retiree would prefer a spending pattern that doesn't vary a lot from year to year. Low risk aversion describes a retiree comfortable with a more variable pattern of spending. The essential/discretionary approach outlined above directly addresses risk aversion.
The essential/discretionary approach also directly addresses the required risk component. If an individual does not have enough savings to pay for essential retirement expenses with low-risk assets like bonds or annuities, the individual needs to consider other options, including working longer and cutting expenses. Some might argue that such a predicament could be resolved by funding essential expenses with riskier assets. However, as I demonstrated in this Advisor Perspectives article, taking more risk in this situation will likely lead to an intolerably high probability of running out of money.
Risk capacity refers to the financial capability to withstand losses. It is affected by the amount of available savings in relation to financial needs (which, in turn, determine the required withdrawal rate) and the relationship between basic living expenses and guaranteed lifetime income sources. The essential/discretionary approach deals with basic living expenses directly by utilizing guaranteed lifetime income sources to meet essential needs.
The essential/discretionary approach also addresses the discretionary element, but in a more complicated way. It might seem logical that someone with a lot of savings in relation to desired discretionary spending (and therefore a low required withdrawal rate) could afford to take risk and invest heavily in stocks. However, if we consider an individual with high risk aversion, a heavy stock allocation might be inappropriate since he or she would not gain significantly in utility or well-being from a higher potential level of spending. This latter consideration gives an indication of the overlapping nature of the three risk components discussed in this section. In building a financial plan for a client, an advisor could test the overall impact by going through the steps to build a retirement plan based on the essential/discretionary approach and then checking how the plan addresses the three components.
Dealing with the psychological components
We have dealt with the risk components related to building a plan to meet financial goals, and now we need to focus on the psychological components – risk tolerance, risk perception and loss aversion. Unlike the other components, which we could analyze separately, these three are intertwined and are best addressed together. For example, loss aversion is the flip side of risk tolerance. The key question we are trying to answer is whether the client will be psychologically capable of sticking with the financial plan the advisor has developed – a single behavioral risk that encompasses the three psychological components. To put it less elegantly, we are trying to assess the "freak out" potential.
The accepted practice in financial planning is to have the client fill out a risk-tolerance questionnaire. However, despite their widespread use, risk-tolerance assessments may miss the mark in addressing the freak out issue.
Per Kitces’ blog post, some risk-tolerance questionnaires go off subject and also include questions about risk capacity. For example, the introduction to Fidelity's questionnaire states: "Your risk tolerance measures how comfortably you can handle declines in the value of your investments both emotionally and financially."
There are also questions about whether investor behavior in down markets is more a function of risk perception or risk tolerance. Geoff Davey, the co-developer of the FinaMetrica risk tolerance assessment tool, made this case in an Advisor Perspectives article. He writes, "With major events such as a stock market correction, typically it is an investor's perception of risk that changes, which results in a change in their behavior." His research has shown risk tolerance is a much more stable personal trait.
Texas Tech Professor Michael Finke, in research with colleague Michael Guillemette found that questions on risk-tolerance assessments measuring loss aversion were good predictors of cashing out of stocks in down markets, but many of the other questions were poor predictors. This finding raises questions about the predictive power of overall risk-tolerance scores.
A fourth issue is the one I focused on in this Advisor Perspectives article, arguing that risk-tolerance assessments are likely to be poor predictors of how clients will react under stress. This failure stems from the "two mind" concept, popular in psychology1. Different parts of the brain are engaged in filling out a questionnaire versus reacting to market turmoil, so risk tolerance scores may turn out to be poor predictors of behavior.
Psychologist Jonathan Haidt came up with perhaps the most descriptive metaphor to draw attention to the two-mind phenomenon, describing this mental division as akin to a rider on the back of an elephant. Translating the metaphor to investor behavior, the rider fills out a risk-tolerance questionnaire but has little control over the elephant's reaction when spooked by market turmoil.
Although this two-mind concept has been popular with psychologists for many years, it is only beginning to influence the way advisors deal with clients on matters of risk. Michael Finke, Russell James and Michael Guillemette are members of the academic community who have been involved in financial planning research that recognizes the two-mind concept. System 1 versus System 2 has been popularized by Daniel Kahneman's best-selling book, Thinking, Fast and Slow. However, despite the increased attention, there is still a lot to be done to integrate what psychologists have learned with advisor practice.
A better way forward
We definitely need a better way to assess client freak out potential and broaden the focus beyond tolerance questionnaires. Although the questionnaires produce numerical risk scores and even asset allocation guidelines, the precision is an illusion and we miss the most important predictors.
Perhaps the best indicator of how a particular client will behave in the next financial crisis is how he or she behaved in the last one. Those who freaked out and sold in the turmoil of 2008 will have a strong tendency to react the same way the next time the stock market crashes. Another consideration is the degree to which particular clients have been willing to follow advice even when it has conflicted with their gut instincts. A related consideration is the client's view of how much control an advisor can effectively exercise in investing – whether it's feasible to move in an out of the stock market to avoid the downdrafts. Clients who believe in market timing will be more prone to making mistimed market moves.
There is also a feedback loop in how the financial plan is set up. A client who has the assurance that guaranteed lifetime income sources will pay for basic living expenses will remain calmer in a market crisis than one hoping that a mix of stock and bond investments will last a lifetime.
Unfortunately, these indicators cannot be precisely scored like risk-tolerance assessments; there is no way to translate the indicators into specific portfolio recommendations. The best an advisor can do is make a subjective judgment about how much to dial back the riskiness of plans to anticipate client behavior. Educating clients may also help, but education has a greater impact on rational thinking than on influencing on emotions.
Final word
I've presented a framework to help advisors deal with the variety of risk components that make up their clients' overall risk profiles. This framework relies heavily on the essential/discretionary retirement income strategy that I favor. It would be worthwhile for others who utilize different retirement income strategies to evaluate how their approaches deal with the various risk components.
Joe Tomlinson, an actuary and financial planner, is managing director of Tomlinson Financial Planning, LLC in Greenville, Maine. His practice focuses on retirement planning. He also does research and writing on financial planning and investment topics.
Read more articles by Joe Tomlinson
