Health & Medicine

Hedge fund stock trading in the financial crisis of 2007-2008

Description
Hedge fund stock trading in the financial crisis of 2007-2008
Published
of 56
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Share
Transcript
   Hedge Fund Stock Tradingin the Financial Crisis of 2007-2008 Itzhak Ben-David Fisher College of Business, The Ohio State University Francesco Franzoni Swiss Finance Institute and University of Lugano Rabih Moussawi Wharton Research Data Services, The Wharton School, University of Pennsylvania December 2010 Abstract We document a drastic reduction in hedge fund stock ownership during the recent financial crisis. In thetwo quarters around the Lehman collapse (2008Q3-Q4), hedge funds cut their equity holdings by about29% and nearly every fourth fund dumped more than 40% of its equity portfolio in each quarter. Wedirectly establish that investor redemptions were a primary driver of these selloffs and provide suggestiveevidence that pressure from lenders was also an important determinant of stock sales. These channelswere more relevant for funds with low restrictions on investors’ withdrawals and low bargaining power vis-à-vis their brokers. Also consistent with fire sales, hedge funds were more likely to sell high-volatilitystocks and liquid stocks. Finally, we show indirect evidence suggesting that part of the stock selloffsoccurred because hedge funds reallocated capital to other assets in a flight to quality or in pursuit of profitopportunities.   _____________________ * We thank Viral Acharya, Giovanni Barone-Adesi, Alexander Eisele, Vyacheslav Fos, Craig Furfine, YeeJin Jang,Pete Kyle, José-Miguel Gaspar, Massimo Massa, Loriana Pelizzon, Alberto Plazzi, Steven Ongena, TarunRamadorai, Ronnie Sadka, René Stulz, Dimitri Vayanos, and seminar and conference participants at the Ohio StateUniversity, the 2 nd Annual Conference on Hedge Funds in Paris, the 3 rd Erasmus Liquidity Conference, theWharton/FIRS pre-conference, the FIRS conference (Florence), LUISS University (Rome), and the C.R.E.D.I.T.Conference (Venice), for helpful comments.  1 1. Introduction  It is widely believed that in normal times hedge funds provide liquidity to financialmarkets. 1 Because the recent financial crisis was characterized by a substantial deterioration inliquidity across asset classes, a natural question is whether and how hedge funds altered their  behavior in this critical period. This issue relates to the increasing body of theoretical andempirical literature studying the mechanics of liquidity at times of market stress. 2  We address this question by exploiting a new data set srcinating from the match of institutional ownership of U.S. stocks with a proprietary list of hedge funds. These data aremanually matched to TASS to draw information on hedge fund characteristics and returns.Overall, we are in the position to provide a detailed analysis of the changes in hedge fund stock holdings during the recent crisis and their trading motives.The main message of the paper is that hedge funds exited en masse the U.S. stock marketas the crisis evolved primarily due to more stringent financial constraints. Figure 1 provides agraphic impression of the drop in hedge funds’ market participation. It plots the fraction of U.S.market capitalization held by the hedge funds in our data set. This is a lower bound on the truefigure because we exclude investment houses for which the hedge fund ownership data areconsolidated with other asset management activities (e.g., Goldman Sachs). The shaded areasdenote the quarters around two events of special market stress: the Quant meltdown (2007Q3)and the Lehman Brothers’ bankruptcy (2008Q3). The figure shows drastic declines in hedge fundholdings around these events. In more detail, we find that hedge funds reduced their equityholdings by about 5% in each of the third and fourth quarters of 2007, and by about 15% in each 1 Khandani and Lo (2009) document that the returns of hedge funds are correlated with the returns of illiquid assets,and Aragon (2007) and Sadka (2009) find that hedge funds earn premia related to liquidity level and risk,respectively.   Brophy, Ouimet, and Sialm (2009) and Agarwal, Fung, Loon, and Naik (2009) present evidence thathedge funds provide liquidity in niche assets. 2 Gromb and Vayanos (2002) and Brunnermeier and Pedersen (2009) provide a theoretical framework to analyze thethe link between funding liquidity and market liquidity. Some recent empirical studies focus on the constraints toliquidity provision (Aragon and Strahan 2010, Comerton-Forde, Hendershott, Jones, Moulton, and Seasholes 2010,Hameed, Kang, and Viswanathan 2010, Nagel 2010, among others). Khandani and Lo (2007) study the behavior of hedge funds in the summer of 2007. Boyson, Stahel, and Stulz (2010) study contagion across hedge funds in thecontext of liquidity dry-ups.  2 of the third and fourth quarters of 2008, on average. 3 Moreover, we show that other institutionalinvestors (excluding mutual funds) and retail investors took the other side of hedge fund trades. 4  After establishing this novel evidence on hedge fund stock selloffs during the lastfinancial crisis, the goal of the paper is to understand the economic forces that drove hedge fundsout of the equity market. Theoretical guidance for our analysis comes from the observation that,at times of crisis, there is a shift in hedge funds’ incentives. On the one hand, in bad times,limits-to-arbitrage can emerge that constrain hedge funds’ ability to provide liquidity (Shleifer and Vishny 1997, Gromb and Vayanos 2002, Vayanos 2004, and Brunnermeier and Pedersen2009, among others). These forces can manifest themselves as investors’ redemptions, margincalls, and the risk limits which are in place to preempt future capital calls. We label these driversas the  financial constraints channel . On the other hand, the occurrence of liquidity dry-upssimultaneously across asset types (Chordia,   Sarkar, and Subrahmanyam 2005, Goyenko 2006,Goyenko and Ukhov 2009, Baele, Bekaert, and Inghelbrecht 2010) gives unconstrainedarbitrageurs an incentive to reallocate capital towards more attractive investment opportunities.In this case, we talk about the asset reallocation channel , which is consistent with continuationin liquidity provision.We provide direct evidence on the primary role played by redemptions in causing theselloffs. Redemptions account for roughly 6% of the 12% decline in average hedge fund equityholdings during the selloff quarters. Because we can directly measure redemptions, this findingis the soundest evidence on the motives behind the equity sales. Moreover, some indirect butvery robust results suggest that lender pressure was also a major driver of hedge funds’ sales.Using hedge fund average leverage as a proxy for lender pressure, we conclude that the financialconstraints channel (redemptions plus lender pressure) explains about 9.5% of the 12% decline in 3 These figures translate to an exit of 0.2% on average of total market capitalization in each of the third and fourthquarters of 2007, and to 0.4% on average of total market capitalization in each of the third and fourth quarters of 2008. 4 We complement the analysis with an examination of short side trades. Because hedge fund short equity positionsare not disclosed, we rely on the conjecture that most short selling is performed by hedge funds (see, e.g., GoldmanSachs 2010). We document a decline in the overall short interest reported by the exchanges in these four quarters,which is partly driven by the short selling ban on financial stocks in September and October 2008. While closingshort positions may inject liquidity into the market, we show that there is only a small overlap between the stocksthat were sold by hedge funds and those bought when short interest declined. Thus, the selloffs by hedge funds andthe reduction in short interest did not cancel out in terms of liquidity provision.  3 average hedge fund equity holdings. We view this first set of findings as consistent with thelimits-of-arbitrage literature cited above.To gain more intuition on the sources of heterogeneity in hedge fund behavior, weexplore the characteristics of hedge funds that sold off during the crisis. As expected, crisisselloffs caused by redemptions were more likely for funds considered liquid by investors (i.e.,those with short lockup period, short redemption notice, and short redemption frequency).Consistent with the margin call interpretation, crisis selloffs related to high leverage were morelikely for funds with poor past performance.To bring further evidence on the financial constraints motive, we study the characteristicsof the stocks that hedge funds traded during the crisis. We report that hedge funds were morelikely to close positions in high- rather than low-volatility stocks. Symmetrically, short interestdecreased more strongly for high-volatility stocks. These results reveal a risk managementmotive. As the limits-to-arbitrage literature predicts, a reduction in exposure to high-volatilityassets can derive from margin calls (Brunnermeier and Pedersen 2009), or from internal risk management practices, such as Value at Risk (VaR) models (Vayanos 2004, Brunnermeier andPedersen 2009). Interestingly, hedge funds were more likely to sell liquid stocks during thecrisis, consistent with Scholes’ (2000) observation that during a crisis, investors deleverage their  portfolios by liquidating their most liquid securities first. This finding suggests that hedge fundswere trying to manage the price impact during the fire sales.In addition to financial constraints, we provide indirect evidence that the assetreallocation motive was a significant driver for some hedge funds. In particular, we conjectureand test that non-equity-focused hedge funds were more likely to sell stocks due to the assetreallocation motive as they had better access to other markets. This channel appears to explainabout 2% of the 12% decline in average hedge fund equity quarterly holdings during the crisis.Further evidence on the asset reallocation channel comes from hedge fund return correlationswith market indexes. We infer that top sellers during the crisis significantly increased their exposure to the fixed income markets (both government and corporate bonds). While the  4 investment in government bonds may reflect a flight to quality, 5 the investment in corporate bonds could reveal an attempt to capture profit opportunities in an illiquid market.Finally, we relate hedge fund returns to their trades during the crisis. We find that hedgefunds that sold stocks during the selloff quarters achieved higher returns in other markets in thequarter after the selloff quarter. We view this finding as indirect evidence in favor of the assetreallocation channel: some of the hedge funds that sold equities invested the proceeds in other, potentially less liquid, markets and captured the rebound.Our results provide some perspective on the findings of some recent work. By studyingthe correlation of hedge fund returns with the stock market, Cao, Chen, Liang, and Lo (2009) andBillio, Getmansky, and Pelizzon (2010) conclude that hedge funds are able to time the marketand avoid liquidity dry ups. Our evidence suggests that much of this “timing” behavior is theresult of hedge fund capital evaporation during crises. Sadka (2010) shows that hedge fundreturns contain a premium related to aggregate liquidity risk. Our evidence can explain this premium in terms of the financial constraints that prevent hedge funds from providing liquidityin times of crisis. Our finding that redemptions are a major constraint to hedge fund ability tocapture the illiquidity premium resonates with the results in Hombert and Thesmar (2009) andTeo (2010). After the first draft of our paper, Boyson, Helwege, and Jindra (2010) have analyzedsimilar data, but using a smaller set of hedge funds, and found that during the crisis hedge fundssold more equity holdings than required to just face redemptions. We provide some perspectiveon the other channels, besides redemptions, that were likely the cause of the selloffs. Also later than our paper, Brown, Green, and Hand (2010), have argued that fire sales were not awidespread phenomenon during the crisis because many funds did not experience negativealphas. In our view, the absence of negative alphas is not by itself evidence of a lack of fire sales,especially if the distress condition is reflected in the risk factors themselves. Rather, we take thefact that in a depressed equity market hedge funds were not able to capture underpriced securities(which is testified to by the lack of positive alphas) as consistent with our finding of severefinancial constraints for the hedge fund sector.The paper proceeds as follows. Section 2 describes the data sources we use. Section 3explores the aggregate behavior of hedge funds during the crisis and studies the distribution of  5 A flight-to-quality, although not directly related to the need to return money to investors and lenders, can still beascribed to the financial constraints explanation, as it can be motivated by the need to preempt future capital calls.
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks
SAVE OUR EARTH

We need your sign to support Project to invent "SMART AND CONTROLLABLE REFLECTIVE BALLOONS" to cover the Sun and Save Our Earth.

More details...

Sign Now!

We are very appreciated for your Prompt Action!

x