Using volatility to measure risk is a big mistake for long-term investors, warns 300 Club member Stefan Dunatov
The investment industry has two great failings: the inability to identify the right risks when considering investment objectives, and then measuring those risks the wrong way.
A long-term investor’s appetite or tolerance for risk should be a direct function of its individual objectives and should not be measured by a short-term metric like volatility. Forecasting returns, not risk, should sit at the heart of long-term investors’ asset allocation strategy.
The emergence of volatility
The increasing focus on forecasting risk, and the emergence of volatility as a popular measure of that risk, is a disappointing response by the investment industry to the recent difficulties it has faced.
Any investment manager would be excused for thinking the last 15 years have been difficult. Since the late 1990s, investors have faced an increase in apparent economic and investment uncertainty, especially in comparison to the preceding two decades.
The rising uncertainty stems from several sources, including significant changes in investment and accounting regulations, not least the transition to marked-to-market pension liabilities under IFRS 13, which crystallises moves in asset prices more frequently on balance sheets. Other sources of uncertainty include the impact on liabilities of a prolonged period of falling interest rates and therefore discount rates, and, perhaps most critically, the single largest market panic since the 1930s.
In investment terms, uncertainty translates into risk. Deep uncertainty, when we are faced with an unlimited set of possible outcomes, creates fear – the sort of fear seen during the Great Depression of the 1930s and again in 2008 when the banking system in the West teetered at the edge of an abyss. In dealing with that fear, investors analyse uncertainty by identifying potential outcomes, using historical events as guides, trying to judge the likelihood of different outcomes occurring and to envisage the circumstances leading to each outcome and their economic consequences. This process effectively reduces uncertainty to mere risk.
In the current period of increased uncertainty, investors are naturally more focused on how to define, measure and manage that risk.
In the search for answers, however, the industry as a whole has failed to develop an appropriate definition and measure of risk. Asset owners have allowed the providers of investment services, including investment managers and consultants, to dictate the approach asset owners should adopt. External advisers naturally prefer to work with an easily definable measure common to many clients. Their solution has been to replace the traditional focus on forecasting returns and all its difficulties with a short-term metric of volatility to forecast risk, which appears to be a more tractable measure. This touches on a separate key issue the industry also has to grapple with – a lack of alignment of interests between the asset owners, consultants and asset managers.
Volatility doesn’t measure real risk
Volatility is a crude, poor measure of risk for a long-term investor. Like the assumption that markets are efficient, the assumption that volatility is a good measure of risk is clearly wrong.
Volatility forecasting is only helpful in the very short term. Volatility measures typically inform us about the state of the world at the current time, and models that forecast volatility tend to only be able to do so with any degree of accuracy over a very short time frame. Correlation assumptions form a key part of volatility forecasting, and these are even more difficult to predict – so much so that most market participants tend to not forecast them at all (correlations are most often static assumptions). The most confident analysts believe six months is a long-time horizon for forecasting volatility, which is of limited use to a professional investor with a medium to long-term investment horizon.
By succumbing to the pressure of providers to focus on short-term measures of risk instead of long-term objectives, investors are giving up their ability to exploit the illiquidity premium. This undermines the premise of being a patient, long-term investor. Long-term investors should be able to absorb short-term fluctuations in risk premiums, yet risk models focus on these short-term fluctuations.
The true measure of risk
The real problem long-term investors face is not short-term volatility. It is how to deal with the uncertainty surrounding the likely paths asset prices might take over their investment time horizon. This is not volatility. Instead, this is about understanding the range of possible macroeconomic outcomes, judging the likelihood of those outcomes occurring and choosing an asset allocation that best fulfils the investment objective within that context.
Investors’ objectives can and should vary widely, based on their particular investment beliefs and characteristics. As such, their measures of risk should also vary widely and, ideally, be bespoke to the individual investor. In identifying their objectives, investors also need to establish the key outcomes they want to avoid and focus on creating the right measure for their risk appetite accordingly.
However, more broadly, almost all investors share a balance-sheet problem, which is not confined to assets alone.
Generally, investors such as pension funds and insurance companies, have a key balance-sheet metric, which is the discount rate applied to liabilities, or, in the case of endowments and sovereign wealth funds, an ‘inflation-plus’ return target. Both establish the rate of growth the assets need to achieve to keep the balance sheet whole or reach the required target.
The true risk for all investors is the extent to which they are prepared to accept sub-optimal outcomes in their efforts to achieve their objective. A sub-optimal outcome is simply one that does not achieve the investment objective. Anything less than the target return is value destructive – anything greater is value creation.
Measuring risk in terms of volatility does not help investors understand or effectively manage this problem. Instead, forecasting returns and economic scenarios are key.
Investors should be focusing on their individual pension, insurance, endowment or sovereign wealth fund objectives, risk appetites and tolerances. The medium to long-term investment horizon available to most of these investors gives them an edge and should not be ignored because of investment managers and consultants’ short-term appeals to solve long-term problems.
Most important, long-term investors should be focusing on forecasting returns, not volatility. They need to better understand the true economic exposure of portfolios and analyse the economic scenarios that are most likely and most worrying in the future.
Stefan Dunatov is CIO at Coal Pension Trustees Investment and a member of the 300 Club
1 Reader's comment