Our Approach to Risk

It has become more common to apply quite complex risk models in investment management. To us, it is important to challenge this development and simplify the process to construct a portfolio. Most complex theories and modern risk measures are based on historical data. That includes standard measures like volatility, value-at-risk and tracking error. But a lot of the disasters in the financial world have resulted from a reliance of quantitative models that are based on historical data. For instance, in a position, where the fund manager might go long a utility and short another utility, with the belief that the correlation between the two stocks are very high, could over time generate substantial losses due to structured change in the nature of the businesses.

We focus on long-term economic relationships between individual positions in the portfolio that lead us to a portfolio structure with a desired risk/reward profile. A desired risk/reward basically means that the potential downside on any investment should be disproportionately low to the upside of the investment.

Exposure management, family style

As we see exposure management differently than others, we have invested in an in-house risk manager. We concluded that we would be best served by building our own system for portfolio and risk management rather than buying a 3rd party system. This way we can continuously grow together with the system instead of purchasing a 3rd party system that we have to adapt to our needs, being dependent on their release cycle and the requirements that their customer base has.

To make our exposure-model more dynamic, we define the most important factors for each investment, and add or reduce factors continuously as they crop up or disappear to make the model real-time. We monitor how much exposure each factor adds to the portfolio according to the size of the position. The portfolio managers may reduce certain factors by adjusting positions or hedge out undesired exposures. For example, if the portfolio has an extensive exposure in oil-related investments, the portfolio manager could reduce the exposure by reducing, eliminating or hedging to allow a proper level of exposure.

In some cases an investment has exposure factors that are not hedge able, but only observable. For instance, a classic exposure factor for a pharmaceutical company is the approval of a new product. In most cases, the only way to come around this exposure is to adjust the size of the position.

The exposure model also has a subsystem to carefully monitor liquidity in our positions.