Herding Behavior

AIM 78.3: Explain how the lack of transparency with respect to hedge funds necessitates other strategies for understanding and replicating historical hedge fund performance crises.

AIM 78.4: Discuss how correlations between hedge fund strategies have evolved over time and the implications of those correlations today.

The use of standardized risk measurement and management products by many traders increases the likelihood that these traders create similar portfolios in the market. Over the past ten years, factor risk models, produced by BARRA, Northfield Information Systems, APT, and others produce a market where managers are more likely to have common exposures. Since many traders acted in a similar fashion (with a sudden liquidation) to the August 2007 market, standardized risk measurement likely served as part of the reason for the unwinding of positions as portfolios failed.

The network view of hedge fund performance looks at the degree of "connectedness" in the hedge fund industry over time. This analysis uses a study of the correlation of returns of various hedge funds. Review of a network flow chart shows a clear increase in the connectedness of hedge funds between 1998 and 2007.

One notable result is the fact that the multi-strategy hedge fund, discussed in the previous topic, is more highly correlated with the other hedge fund indices. The increased "connectedness" can be linked to (at least) two key factors.

1. Over the years hedge funds have an increased exposure to traditional market factors (S&P 500 index, government security rates, monetary policy, etc.).

2. The ties due to intricate strategies and multi-strategy funds have increased the correlations.

The increasing connectedness and growing size of the hedge fund industry will require improved risk metrics going forward in order to adequately capture a fund's risk exposure.

It is clear that the hedge fund industry is much more connected than it was just ten years ago. Findings suggest that systematic risk is present in hedge funds and supports the idea of a hedge fund beta. The fact that the universe of long/short funds moved in such a correlated fashion suggests there are particular common factors for those funds. Ultimately, it should be possible to use these common factors in a multifactor model to cost-effectively, passively, replicate these hedge fund's returns despite the lack of transparency in the hedge fund industry.

0 0

Post a comment