A risk approach to running portfolios
It is May 1998. The euro’s birth has just been announced for the end of the year. A sea change will sweep continental Europe. As currencies become one, interest rates and bond yields become one across the Euro-zone. Equity markets will explode, losing their national idiosyncrasies. Equities will be best analysed by sectors. Europe would become a less risky place. Or so we were told by brilliant strategists who spell a bear market as a ‘bare’ market. pension managers in Europe mostly followed the advice, shifting from bonds into equities, making their portfolios more international and demanding obsolete restrictions to be eased, even lifted altogether.
At Fonditel, we did not see it exactly that way. We saw opportunities everywhere. We took the view that sectors would become more important and countries less important, but that size would be more important than both of them. We took the view that risk would still be rewarded wherever it came from, and that equities would still be one risk source, not the only return oasis. We took the view that correlations between stocks and bonds would also shift, providing new opportunities. We thought that the vanishing of currencies would make Europe a riskier, not a safer, place to invest, as diversification opportunities would shrink.
Caught in the middle of a process to increase our equity exposure and our geographical reference area, we went for the major asset classes, not for the hottest stocks. We kept in mind our main driver: if we keep risk at 6% we should get where we want to go to, eventually.
We think of ourselves as risk managers, not return hunters. We rely on the time-honoured observation that risk exposure provides over the long term capital preservation and a few points of real growth, which should be enough to meet our goals. Risk management is like balloon flying, not like rocket launching. You can have an influence on distance to ground and some maneuverability while on the air, but direction is dictated by the winds, not by you. We try and keep risk exposure stable, and hope the winds will be favourable: returns will be commensurate with risk. But how do we go about managing risk?
Many people confuse risk measuring with risk management. You can’t manage what you can’t measure, but building indicators of risk and not acting on their flashing lights is no better than not having any at all. In our view, the value at risk (VAR) is a good tool for forensic investing, but not so good a gadget for risk management.
Risk management is about knowing where your risk comes from (sources), what makes it tick (causes), and what you can do to shift up or down your exposure to it (uses) so that you keep it pretty much stable and can feel confident the rewards will eventually come.
Risk is spelled as volatility most of the time, and that is a correct spelling. However, other dimensions of risk, like credit, currency, duration and inflation are not negligible. And then there is the risk of implementation. It is a humbling experience to realise how often you end up short of whatever benchmark you are following.
Keeping it to volatility for this time, we approach risk management by setting a benchmark, measuring volatilities for each of its components and calculating correlations among them. We then compute the aggregate portfolio risk (volatility). This is what we try to measure on a daily basis and control for on a weekly basis. If portfolio risk shifts it must be because benchmark risk is shifting or because our wedge from the benchmark is shifting (tracking error). If benchmark risk is shifting, it may be an underlying asset risk that is shifting or correlation that is changing. Then we try to understand why that particular asset class risk is changing and whether the change is short lived or permanent. If we believe the change is not short lived, we adjust risk by decreasing or increasing our exposure to it, directly or via derivatives, if they are available. If correlation is the source of change (say it goes up) then we try to adjust the combined exposure of the now more correlated asset classes by reducing our exposure to both, which can also be done by increasing our exposure to all other asset classes. If the source of change is tracking error, we follow a similar process, identifying the sources, estimating whether they are permanent or short lived and acting consequently.
Notice I have not said a word about positive or negative returns. Some managers identify increasing volatility with negative returns. Unfortunately, risk is pretty much both symmetric around its mean and mean reverting. Mean reversion means that risk tends to go back to its long run average, symmetry that at a given point in time you are as likely to experience higher risk than lower risk going forward. We try to keep risk stable, not returns. Returns are not stable, they do bounce around a lot. But there is no sure way of anticipating turning points, although most of the investments industry makes its living trying to do just that. Some argue that returns are mean reverting, which is a stronger statement than saying that risk mean reverts. The problem is the time frame. It is of little use to wait for mean reversion if it is 2000 and you have been invested for 10 years in Japanese stocks. Japanese returns may eventually mean-revert, but you will not be there to enjoy the ride.
Many investors associate risk with particular asset classes like equities, or emerging market bonds. We see risk everywhere. Inflation being the largest drain on capital over long periods of time, it is a natural feeling to think of equities as inflation hedges and bonds as an efficient way to die slowly, but surely. We think the focus on equities is historically justified, but largely misleading. Diversification keeps you alive in bad times and stabilises your trip in good times. But diversification works better among asset classes than within asset classes. Thus if diversifying within equities is good, diversifying among asset classes is sheer heaven. Investors that focus on equities exclusively on the ground that they provide powerful long term returns are taking more risk than they need. We take the view that risk taking is eventually rewarded. But risk is a many faceted gem. Looking at just one of its faces, equities, means giving up on other wonderful opportunities. Risk may be rewarded, but equity risk may not be rewarded for long periods of time and vast areas of land.
Remember the managers from the beginning of this article? They were told to go international, to shift into equities, to look through the sector prism. Many did. Tactically that was the right thing to do. But confusing risky tactics with sound strategy can be costly. Should we enter a colder period in equity markets, a lot of new money that found a home in the blossoming equity spring of the last years may have second thoughts about equity risk taking being a one-way street.
Santiago Fernández is chief executive officer of Fonditel, the asset manager of the Telefonica pensions funds, in Madrid