Hedge funds can cut portfolio risk in half
A new study by EDHEC shows that an allocation of 20 per cent to hedge funds can reduce a fund's probability of extreme loss by 50 per cent.
The results of the study -- 'The Benefits of Hedge Funds in Asset Liability Management' by Lionel Martellini and Volker Ziemann of the Edhec Risk and Asset Management Research Centre -- are due to be presented in full at the Edhec Hedge Fund Days in London on 14-16 February.
The study shows that it is possible to construct diversification benchmarks that allow the risk related to holding stock or bond portfolios to be reduced in a very significant and robust way by appropriately selecting the alternative strategies and optimising these with proven techniques (minimising the extreme risks, as measured by the Value-at-Risk of the overall portfolio).
The main points of the study are as follows:
• Hedge funds are not a strategic asset class because they are heterogeneous and cannot be modelled
• The studies that attempt to evaluate the role that hedge funds should play as an allocation class are naïve and subject to sample risks.
Edhec has set up a new approach to asset-liability management in the presence of hedge funds. This study allows for the identification of hedge fund styles that enable the risk parameters of the stock and bond classes to be improved over the long-term.
Rather than identifying a hedge fund class, this approach therefore involves including hedge funds in fixed-income and equity management. The results show that, with an approach of this kind, an allocation to hedge funds of 20 per cent allows the probability of extreme loss of a pension fund to be reduced by 50 per cent.
Moreover, the introduction of 10 per cent of hedge funds into the overall allocation can lead to a reduction of more than 25 per cent in the probability of the value of the assets falling below 75 per cent of that of the liabilities (asset-liability deficit greater than 25 per cent).
- By Category
- News from other sites
- Special Reports
- Partner events