Article published in Le Jeudi (Luxembourg) on 6 March 2014

Ninety years ago, an US economist, Frank Knight, first drew the distinction between risk and uncertainty. Risk is a statistical concept. It can be calculated on the basis of historical performances. Uncertainty is another matter entirely. Uncertainty is our awareness that things can happen, combined with our inability to assess their probability of doing so, simply because we are unable to back-test our hypotheses. In this environment, whenever we use the word “probability”, it is based on intuition and not statistical analysis. Uncertainty is quite frustrating: we are aware that we lack some of the information needed for a clear response. The former US defence secretary, Donald Rumsfeld, called these “known unknowns”. This frustration is caused by the feeling that we have lost control. While “risk” means “control” (it is possible to calculate, with a high degree of precision, that in a given year, the equity market will rise by 20% or that bond yields will fall by 1 per cent), “uncertainty” rhymes with “lack of control”.

An increase in uncertainty affects the financial markets through its effects on the economy. It can undermine growth and the currency (which would exacerbate inflation), and trigger an interest rate hike (to defend the currency). It also has a psychological impact on the economy (by reducing risk appetite). In spite of all this, uncertainty can be managed. The simplest way to do so consists in diversifying. Depending on developments, some asset classes (government bonds) do well, while others (equities) do less well. Diversification provides broad exposure and smooths out portfolio performance over time. A second way of managing uncertainty consists in setting limits: if a completely unexpected event sends the equity markets into a downward trend, positions are reduced beyond a certain threshold. A third way consists in exploiting decorrelation between asset classes. This isn’t “naive” diversification; on the contrary, instruments are carefully chosen to produce negative correlation. This is where we find safe havens, such as high-quality government bonds or gold. If uncertainty rises, for example, because of political tensions, higher prices of bond (which move in the opposite direction of long bond yields) and gold can cushion the portfolio impact of falling equity prices. Academic research has studied many episodes of geopolitical uncertainty and found that they often cause oil prices to rise. A fourth way of managing uncertainty is through a scenario-based analysis, which has the advantage of requiring a thorough process based on defining a baseline scenario and alternative scenarios, along with a description of how they might unfold (i.e., catalyst and cause-and-effect relationship) and an assessment of their impact on various assets classes. The resulting transparency justifiably creates a feeling of control and mastery, although we don’t know what scenario will come to pass. At the very least, it lets us compare the portfolio’s behaviour through various scenarios and take action accordingly, while an inventory of catalysts can be used to draw up a check-list of points to monitor closely.

So, all in all, uncertainty can be managed. Reacting by fleeing and avoiding any investment in risky assets (equities, corporate bonds, etc.) entails an opportunity cost, which, over time, will be very significant. Most of all, the investor must ensure that he has decorrelated assets in his portfolio.


William De Vijlder

Vice – Chairman of BNP Paribas Investment Partners