Is Volatility A Better Measure Of Risk Than We Give It Credit To Be?

It has been popular in recent years to bash volatility, and standard deviation as its most common way of being quantified, as a terrible measure of risk. Not just because of the criticisms associated with standard deviation itself and whether market returns are normally distributed, but at a more basic level: is the up-and-down volatility of an investment what a client really cares about? Shouldn’t risk be more focused on loss, the impact of losses on goals, and the probability of achieving goals, than just the raw choppy volatility itself? Yes, perhaps, but on the other hand maybe we don’t give volatility itself enough credit for the risk that it does create: volatility in investment returns leads to volatility and uncertainty about the timing of retirement and other goals and the risk that they cannot be achieved in the time anticipated.

The inspiration for today’s blog post comes from a series of email conversations I’ve had as a follow-up to a blog post I made a few months ago about how typical retirement projections create a reliance on doubling your money in the investment markets in a very short period of time. which got a lot more visibility last weekend when it was picked up by the New York Times. The essence of the conversations were that perhaps volatility is actually more relevant than we give it credit for; we just don’t translate it well into real world consequences.

For instance, most clients cannot relate in any tangible way to the idea of portfolio A has a 10% expected return and an 18% standard deviation, while portfolio B has a 6% expected return but only an 8% standard deviation. We sort of understand return; we have difficulty understanding compounding returns; and we have no intuitive capacity to understand the real difference between higher return with higher standard deviation and lower return with lower standard deviation. How are we supposed to modify our return expectations in light of the difference standard deviation assumption? The typical translation has been pretty typical: it all evens out in the long run, so pick the higher returning investment and just make sure you stay the course.

Indeed, we perhaps make the point even more salient in the direction of always encouraging the higher return investment by translating it back to savings. If you pick the higher return investment, you can save $300/month over your working years and generate enough wealth to retire. If you pick the lower return investment, you’ll have to spend less now, do fewer things you enjoy, and save $600/month to get the same retirement outcome. No wonder there’s so much pressure to always pick the higher return investment; the standard approach implies that everyone should always take the highest amount of risk they can possibly tolerate without falling off the wagon. Taking less risk only translates to one outcome: if you take less risk, you have to spend less now and save more. In other words, under the traditional framework, lower risk has a big cost (I give up spending on things I want to enjoy now), and higher risk has little cost (just stay invested for the long run).

However, part of what my discussion about compounding in the final retirement years reveals is that higher volatility can actually have a very big cost: it makes the final date of the accumulation goal more uncertain. Since a higher volatility portfolio has a greater potential of underperforming the anticipated return by a significant magnitude and/or for a long period of time, the impact of a higher return/higher volatility portfolio is that the end-point of the goal has a material risk of being pushed back. So in the context of the earlier example, the trade-off is not just that high-return/high-volatility means saving $300/month, and lower-return/lower-volatility means cutting spending and saving $600/month. Instead, we can and should also discuss the uncertainty of the timing of the outcome. You can save $300/month and, depending on market returns, retire not at age 62, but anywhere between age 52 and 72 with your early 60s as a probable but uncertain outcome. Or you can choose to save $600/month and be certain to retire between age 59 and 65. The lower return and higher savings scenario doesn’t just have a cost in the form of reduced current spending; it also has a benefit, in the form of certainty about when you get to retire, and a dramatic reduction in how much longer you’ll have to work if things don’t go well (pushing back to only age 65 with bad low-volatility returns, instead of pushing back to age 72 with bad high-volatility returns that come at the wrong time).

So maybe volatility is actually a decent way to look at risk, once we translate it into terms that are relevant for a client and show the benefits and the costs. What do you think? Is this a reasonable way to translate the impact of volatility into more tangible outcomes that illustrate the trade-off? Would you use this as a way to communicate the concept to your clients?