How professionals manage portfolio risk

Thomas – we are good at assessing risk in our everyday lives. That’s why we, for example, look left and right before crossing the road. But what risks do investors on the capital markets run – and how can they be managed?

For investors, risk arises in part from the extent to which an investment’s value fluctuates, i.e. volatility. However, drawdown, the maximum loss in value that a share, bond, commodity or fund can suffer, is also a crucial risk measure. It is possible to measure both metrics historically and estimate their future trajectory. I can say that a fund has fluctuated in value by 34 percent over a certain time – say, a week or a year. And I can also say that the investment’s value dropped by a maximum of 21 percent during the last market crash. Fund managers are usually assessed on their portfolios’ historical volatility and drawdown as well as the returns they achieve. This means they face the task of estimating and managing future volatility and drawdown in addition to the return. Estimates can be achieved using appropriate risk models to calculate these two risk metrics.

However, alongside the metrics calculated using models, there is a need for a risk management process that helps the fund manager to interpret the numbers and translate them into decisions. It is therefore essential to establish a risk management process before defining and estimating the risk metrics. Ideally, a fund’s risk objectives should be defined first, and these should fit the potential client profile. Once the fund's risk objectives are known, catalogues of measures can be drawn up and risk limits defined. A suitable investment approach can then be developed and implemented within these limits.

And how can diversification be used in risk management?

The aim with diversification is to fill the portfolio with components that exhibit low correlation with each other. Ideally, this means that losses in one part of the portfolio are at least offset by gains in other parts. The trick with this strategy is to identify elements that do not behave in diametrical opposition, because then gains in one part of a portfolio are always eaten up by losses in another part. Ideally, a portfolio should therefore combine investments that generate positive returns over the long term but that overcompensate for each other in periods of weakness. Personally, I feel that diversification strategies are the supreme risk management discipline in a fund management context. The most important factor here is the uncorrelated nature of the individual risks.

So, multi-asset and mixed funds, for example, could diversify across asset classes.

Correct. Multi-asset funds invest in different asset classes, for example in shares, bonds, precious metals, and real estate. If these individual assets or components are negatively correlated, at least in stress phases, meaning they move in opposite directions in certain market phases, then the individual risks balance each other out. Sifting such diversification effects out from the investment universe and making them usable is another component of risk management. The combination of equities and gold is well known as a classic pair of opposites with an often negative correlation in times of stress. However, during the coronavirus crash of March 2020, we were also able to detect differences within the equity asset class. While most stocks suffered sharp price falls, manufacturers of computer chips, for example, performed much less badly. And technology companies were the driving force in the recovery, which had a compensatory effect with strong price gains. So, during that market crash, it paid to remain invested in those companies. We also use certain factor criteria to search the equity universe, for example, whether a certain company size ensures that a sector behaves particularly defensively or offensively in the market. This is known as the size factor. If you integrate these findings into a strategy, it is possible to control volatility in a targeted way.

So, risk managers really need to be good statisticians first and foremost to make sense of this quantitative output?

Yes, in a way. Quantitative risk management also includes scenario analysis, which simulates how a portfolio would behave if a certain event were to occur, for example if the dollar were to drop sharply. This can also be expressed in figures. However, in practice there is much more to risk management than just numerical calculations. We always combine our quantitative calculations with qualitative assessments. During last year’s coronavirus stock-market crash, for example, we made an ad hoc decision not to move too heavily into cash holdings that preserve capital or to hedge our equity positions completely, despite a strong drawdown on the stock market of well over ten percent. At that time, we had assessed the probability of the markets falling significantly further as lower than that of a rapid market recovery. We would not have been able to take advantage of the recovery on behalf of our investors if our cash position had been too large.