Changes in the factor exposures of hedge funds

Brealey, R A and Kaplanis, E C (2001) Changes in the factor exposures of hedge funds. Working Paper. London Business School nan.

Abstract

Recent years have seen considerable interest in the activities of hedge funds, commodity trading advisors (CTAs), and the proprietary trading desks of commercial and investment banks. Part of this interest represents a need to understand the investment policies and performance of an important sector of the asset management industry, but the activities of hedge funds have also raised issues of public policy. For example, the decision by George Soros's Quantum Fund to sell sterling short in the fall of 1992 is widely believed to have brought significant pressure on the currency and to have hastened its departure from the ERM. In 1997 hedge funds again attracted adverse publicity when the prime minister of Malaysia protested that sales by hedge funds of Asian currencies were responsible for the depreciation of the ringitt. Some critics of hedge fund activity argue that not only may the funds deal in thin markets in substantial volume, but they are also guilty of herding, or even of collusion and manipulation. Other concerns about the activities of hedge funds were prompted by the nearfailure of LTCM in the fall of 1998. Many of these worries centered on the degree of leverage employed by hedge funds. For example, the President's Working Party report on hedge funds concluded "The central public policy issue raised by the LTCM episode is how to constrain excessive leverage more effectively. As events in the summer and fall of 1998 demonstrated, the amount of leverage in the ficial system combined with aggressive risk taking can greatly magnify the negative effects of any event or series of events. By increasing the chance that problems at one ficial institution could be transmitted to other institutions, leverage can increase the likelihood of a general breakdown in the functioning of ficial markets." The potentially disruptive effect of leverage on markets may be accentuated by the feedback trading that may result from the need to liquidate positions to meet the need for additional margin or collateral. One response to these concerns has been to argue for greater transparency for hedge fund exposures. For example, at the IMF/World Bank meetings in September 2000 the G24 group of emerging market countries complained: "Standards in the area of transparency are being pressed upon developing countries without a commensurate application of corresponding obligations for disclosure by ficial institutions, including currently unregulated highlyleveraged institutions." However, not only may it be difficult to obtain agreement to increase disclosure, but the multiplicity of factors to which hedge funds are potentially exposed makes it difficult to specify what information would be useful or how it could be produced in a digestible and timely form. The concerns about the activities of hedge funds have not been universally accepted. For example, there is some evidence that hedge funds did not in fact take large short positions in Asian currencies ahead of the devaluations (with the possible exception of the Thai baht) and did not make profits from these devaluations. Other apologists draw on Milton Friedman's argument that, rather than being destabilizing, profitable speculation drives prices towards their equilibrium value. On this view the price disturbances that followed the near collapse of LTCM were a consequence of a shortage of speculative capital rather than the reverse. This paper seeks to shed light on the factor exposure of hedge funds by examining the loading of the returns of a sample of hedge funds on a number of explanatory factors. We look both at a fund's average exposure to different factors and we then examine how the exposure varies across time. We look at these changing exposures for individual funds and we examine how far funds make parallel changes in their positions. The paper is organized as follows. Section II reviews the literature and Section III provides a description of our methodology and a discussion of our empirical results. Section IV summarizes and concludes the paper.