Risk. An important but misunderstood word in finance. Important, because investors should try to avoid it unless they are well paid for taking it, so we need to be clear about what exactly ‘it’ is. Misunderstood, because this clarity is seldom the case.
Residents of the United States, please read this important information before proceeding
Please read this important information before proceeding.
Volatility is merely a turbulent journey
Frequently, we are told, risk is volatility, a measure of the amount that investments fluctuate along the journey at short horizons. But if this is what we expend effort avoiding, we’re focussing on the wrong thing. A good analogy might be an urgent sea voyage where every minute counts. Going full engine into waves creates turbulence along the way that is unpleasant, but is not risky unless this turbulence increases the risk of a slow journey.3 If your aim is a pleasant journey, then by all means take it slow; but if your chief aim is a getting there on time, then you are better advised to invest in tablets for seasickness.
Investment volatility is distressing. But it is not risk. Unless we need the money along the journey4 then the fluttering of sentiment along the way is not what matters at all. What matters is the chance our portfolio is not worth much when we need it. The potential for low values in the long-term increases the risk, while the potential for high values decreases risk… regardless of the volatility of the journey.
The three dotted lines in Figure 1 each show possible paths of investment returns. The bottom two are perfectly smooth, with no volatility, but end up with average, or negative returns respectively. The higher path offers high returns at the end of the investor’s time horizon, but is extremely volatile. Investors who represent ‘risk’ as volatility often weed out portfolios with good returns to avoid the temporary discomfort of turbulence. To get the best risk adjusted returns we need to focus on the outcome, not a smooth ride.
Risk is the chance of a poor final outcome
Imagine all possible future paths a portfolio may take over the next five years. Risk is about how many of all those possible future paths end up with low values. If we mistake volatility for risk and seek to avoid it, we will weed out portfolios with rough journeys regardless of the outcome, good or bad. We have mistaken comfort for success.5
We should not take on risk unless it increases the average portfolio return sufficiently to compensate us for the chance of bad outcomes. To do this we need a precise way to measure the likelihood of bad outcomes. Armed with this, we can choose the portfolio that offers the best returns, after compensating us for the risk we are prepared to take.
The traditional way of measuring volatility is to calculate the standard deviation of fluctuations along the path. This looks at the return each period and compares it to the average returns. Deviations away from the average count as risky. The further away from the average, the more this adds to the standard deviation.
Calculating risk of bad final outcomes frequently uses the same computation except that, instead of looking at deviations from the average path over time, the standard deviation is computed by looking at where the various paths can end up. Paths that end up at the average outcome, like the mid path in Figure 1, do not add to risk, but paths, like the upper and lower paths in the figure, that end away from the average outcome do. The further the deviation from the average, the more this possible path adds to the ‘risk’ of the portfolio.
So by choosing a portfolio with a low standard deviation of outcomes, we’re weeding out portfolios that have a high dispersion of possible future values, in favour of those that are likely to end up with a value close to the average. At first glance, this seems sensible.
Figure 1: Distinguishing between volatility and risk
“Traditional risk measurement filters out portfolios with high variation in outcome, penalizing the good, along with the bad.”
Don’t weed out the flowers
However, when looking at the possible paths that we exclude by minimising standard deviation, we notice something less reasonable. We’re filtering out portfolios that offer a chance of ending up on a really good path as well as those that have bad outcomes. Standard deviation doesn’t distinguish between good and bad outcomes, it just penalises variation. According to this view, both the top and bottom path in Figure 1 add to risk, and should be avoided! This is like telling an investor that you’re “really sorry, but there may a terrible chance you’ll earn 5% more than expected next year. But not to worry, using our risk measure, we are minimising the chance of this undesirable outcome for you.”
The possibility of good outcomes is simply not risk: and we certainly shouldn’t be minimising it by filtering out portfolios with good upside outcomes along with those which have high potential for bad outcomes. Instead we need a risk measure that increases when there are lots of bad outcomes, but ensures that good outcomes have the effect of reducing risk. The possibility of encountering the lower path increases the risk of a portfolio, but the existence of the upper path should actually decrease risk.
Minimising such a measure means we weed out portfolios with lots of worse than expected outcomes, but doesn’t have the counterintuitive implication that we throw away portfolios just because they have variability in outcomes. At Barclays we have developed just such a risk measure, Behavioural Risk, which is grounded in the extensive evidence from the field of Behavioural Finance about how investors should actually think about risk when making risk-return trade-offs in their portfolio.
It enables us to build portfolios that focus on what really matters to investors: not volatility along the journey, but the value of our wealth at the destination; and not mitigating the dispersion of possible outcomes, but reducing the chance of bad ones.
3 Or, of course the ship sinks altogether, which could happen if your ship is badly constructed and unsuited to the voyage. In a portfolio context this can happen when your portfolio is insufficiently diversified or highly concentrated: any one investment can be sunk by events along the journey. A diversified portfolio merely experiences turbulence.
4 In which case this is the destination, not the journey…and we should have minimised the chance of such unexpected needs through insurance and careful planning.
5 This is not to say that comfort is unimportant: a crossing could be so unpleasant that we despair and turn back or, worse, jump overboard. But this is about controlling our behavioural responses to the journey: volatility only matters insofar as we respond to it.