Until recently, hedge funds were considered as a ‘black box’ which aimed to deliver absolute returns and stipulated cash as a benchmark. Current research conducted at the London Business School’s hedge fund research centre focuses on two issues. The first pertains to the systematic risks of hedge fund strategies while the second issue focuses on hedge fund investor behaviour. This article summarises the findings of two recent research papers.
The ‘Risk and Portfolio Decisions involving Hedge Funds’ working paper (no. HF-009) argues that hedge funds, compared with mutual funds, can follow more dynamic trading strategies, can take long as well as short positions and therefore can bet on spreads like the small-cap/large-cap spread or the value-growth spread.
As a result, hedge funds can offer exposure to risk factors that traditional ‘long-only’ strategies cannot. The key issue is to understand the kinds and nature of the risk factors associated with different hedge fund strategies. It shows that a wide range of equity-oriented hedge fund strategies exhibit a non-linear payoff structure resembling a short position in the put option on the equity market. They also show a significant exposure to the small-cap/large-cap spread. Having documented the presence of non-linearities, the paper assesses its implications for portfolio construction. It argues that the traditional mean-variance framework ignores the presence of non-linearities and shows how an alternative framework - Conditional Value-at-risk (CVaR) can be applied to construct hedge fund portfolios.
Finally, the paper investigates how the limitations of short history of hedge fund returns can be overcome by working with underlying risk factors estimated using a multi-factor model. Since the underlying assets have longer history, this approach provides insights into the long-term risk return trade offs of hedge funds.
The paper starts by developing the theoretical framework for modelling non-linear payoffs in assets pricing and assigning a value to the skill of the manager generating non-linear payoffs. It illustrates the applicability of this approach by implementing it for equity-oriented hedge fund strategies. It uses hedge fund returns data from HFR and CSFB/Tremont databases.
The paper estimates systematic risk factors of different hedge fund strategies and demonstrates its applicability using out of sample analysis. Next, the paper analyses portfolio decisions with hedge funds. After building the theoretical framework for Conditional Value at Risk (CVaR), it compares the tail-risk of mean-variance and mean-CvaR efficient portfolios. Finally, using the systematic risk factor model, it compares the long run performance of hedge funds with the recent performance.
This paper presents three main findings. Firstly, the option-like payoffs are not limited to ‘trend followers’ and ‘merger arbitragers’, but an integral feature of the payoffs on a wide range of hedge fund strategies. In particular, the payoffs on a large number of equity oriented hedge fund strategies resemble those from writing a put option on the equity index. Secondly, the analysis shows that the expected tail losses of mean-variance optimal portfolios could be underestimated by as high as 54% compared with mean-CVaR portfolios.
This suggests that ignoring tail risk can result in significantly higher losses during large market downturns.
Finally, the extrapolation of hedge fund returns using underlying risk factors suggests that the hedge funds’ performance during last decade is not representative of their long-term performance. In particular, the mean returns during the 1927-1989 period are significantly lower and their standard deviations are significantly higher compared with those of their recent performance.
The results of this paper demonstrate the importance of allowing non-linear risk return relation while analysing hedge funds. The findings of this paper are useful to investors and regulators as they can now better understand the risk-return characteristics of hedge funds. In addition, the analysis presents tools that can be used to measure net and gross risk exposures of hedge funds and also assist with hedge funds portfolio construction The ‘Flows, Performance, and Managerial Incentives in Hedge Funds’ paper (working paper no. HF-016) points out that hedge funds differ from mutual funds in variety of aspects. They are less regulated, less transparent, charge performance based compensation; offer limited liquidity (lock-up, notice and redemption periods) as compared with mutual funds. These differences have important implications for incentives of the hedge fund managers to deliver superior performance and the investment behaviour of investors. Motivated by these issues, this paper investigates the complex relationship between flows, performance and managerial incentives in hedge funds.
This paper has two key objectives. First is to analyse the determinants of money flows into hedge funds. In particular, the focus is on the relationship between money flows and past performance, managerial incentives, impediments to capital withdrawals, and past money flows. The second objective is to analyse the relationship between fund size, past flows, managerial incentives, impediments to withdrawals and future performance.
The paper models the incentive-fee contract as a call option written by investors on assets under management. The net asset value (at which the money flows enter the fund), hurdle rate and high watermark provisions determine the strike price of the call option. Since capital invested at different times are subjected to different high watermarks, the incentive-fee contract amounts to the manager holding a portfolio of call options with different strike prices.
The paper uses a comprehensive hedge funds database created by merging three large databases, namely HFR, TASS and ZCM/MAR. The database includes net returns, monthly assets under management, and other fund characteristics such as lock-up and restriction periods, management and incentives fees, inception data and fund strategy.
Using a multivariate set up, it examines the relation between money flows, and fund characteristics and past performance. For estimating the determinants of future performance, it regresses hedge fund returns against fund size, past flows, managerial incentives, and impediments to capital withdrawal (lock-up and restriction periods).
The paper has four main findings relating to flows, past performance and managerial incentives. First, the money-flows chase recent returns. Second, the money-flows are significantly higher for consistent winners, ie, funds with above median returns two years in a row, and vice-versa for losers. Third, there is a positive relation between flows and managerial incentives; funds where the call option granted to the manager via the performance fee is near-the-money receive significant capital inflows. Funds with high watermark provision also enjoy more inflows. Finally, the investors do not prefer impediments to capital withdrawals. If there exist two funds that are identical in every respect, but one imposes longer lock-up and redemption/notice periods, then that fund receives lower inflows, all else being equal.
This work is currently being examined at a more macro level, which investigates whether money flows chase total returns or value added by the manager (the alpha). Preliminary analysis suggests hedge fund investors do not distinguish between the alpha and beta related components of total return. This may be because the standard two-and-twenty fee contract does not distinguish between alpha and total return.