Battle for Alphas: Hedge Funds versus Long-Only Portfolios

Abstract

The study reported here empirically examined whether the alphas of hedge funds and those of long-only portfolios present different distributions and are derived from different risk factors. Adjusted for return volatility differences, hedge funds seem to offer more consistent alphas for potential transfer to either equity or bond asset classes than do long-only portfolios—even under extreme market conditions. Potential explanations for the findings include lack of data reliability and differences between hedge funds and actively managed long-only funds in compensation, investment constraints, and structures. Factors related to market index returns do not adequately detect hedge funds' risk postures beyond a fund's exposure to the market-directional risk of standard asset classes. Risk factors derived from asset prices in financial markets do provide timely and systematic descriptions of the risks underlying trading strategies used by hedge funds. The multifactor style-risk analysis presented here can effectively monitor a hedge fund's exposure to systematic versus idiosyncratic risks and volatility-risk factors over time.

The technology investing debacle, diminishing equity return expectations, and the inability of long-only portfolios to generate positive alphas have greatly contributed to institutional investors' recent interest in hedge-fund investments. The optionlike return pattern of hedge funds, however, is a challenge for investors in analyzing risk exposures. Single measures of risk and return always have the potential to be misleading, but they are especially inadequate in analyzing hedge-fund risk. Investors need to carefully examine the return patterns of a fund at times of various market conditions and consider the fund's exposure to various forms of systematic risk factors.

This article addresses two questions relevant to investor analysis of hedge-fund performance: First, do alphas from hedge funds and long-only portfolios have different return attributes? And second, do the sources of alpha for the two types of investments derive from different risk factors? Because investors can transfer alpha obtained from hedge funds to their traditional asset-class portfolios, a more appropriate way to evaluate hedge funds and long-only portfolios than simply comparing their returns is to compare their returns in excess of their respective benchmarks. The empirical findings presented here indicate that hedge funds, especially equity market-neutral strategies, provide more consistent transferable alpha than do long-only portfolios for both equity and bond asset classes and in various market environments. I assess possible explanations of this performance difference from the standpoints of data reliability and differences between hedge funds and long-only funds in fees, manager incentives, investment constraints versus flexibility, and other structural factors.

Risk factors derived from asset prices in financial markets are timely and useful for hedge-fund risk analysis. Most relevant to hedge funds' risk profiles are exposure to the market-directional risk of standard asset classes and various volatility risks. I show that the way in which a hedge fund manages its exposures to implied volatilities in extreme market conditions may be the key to consistent performance. The results presented in the article highlight the importance of diversifying one's hedge-fund investments among different strategies—between funds and within a fund—for achieving consistent performance.

An analytical framework that incorporates multiple risk factors (rather than a single factor) gives investors a more complete picture of hedge-fund risk taking. Therefore, in the spirit of equity style analysis, which is so popular among practitioners, I evaluate common risk factors driving the performance of hedge funds and long-only portfolios. In this style-risk approach, various financial market risk indicators are categorized into (1) exposure to market risks (market-directional risk) and (2) exposure to volatility risk, which provides a concise assessment of risk exposures over time.

I also review another approach to analyzing hedge-fund risk—direct replication of the hedge fund's optionlike payoff profile, trading strategies, or arbitrage opportunities. Return series derived from this “mimicking” approach are particularly useful for studying the risk factors and performance attributes of hedge-fund investing. It also provides a promising direction for future research into hedge-fund asset pricing.

The author would like to thank Pengfei Xie and Kam Chang for their insightful research assistance. The author is also grateful for many useful discussions with colleagues in the Global Fixed Income Group and constructive comments from Stan Kon, Ron Leisching, Eric Tang, and participants at the Spring 2001 Q Group Conference.