Hedge fund leverage: 2002–2017

• Author: Bing Liang,Liping Qiu
• Published date: 01 September 2019
• DOI: http://doi.org/10.1111/eufm.12202
• Date: 01 September 2019

DOI: 10.1111/eufm.12202

ORIGINAL ARTICLE

Hedge fund leverage: 2002–2017

Bing Liang

1

|

Liping Qiu

2

1

Isenberg School of Management,

University of Massachusetts Amherst,

121 Presidents Dr., Amherst,

Massachusetts

Email: bliang@isenberg.umass.edu

2

School of Business, University of

Connecticut, 2100 Hillside Road, Storrs,

Connecticut

Email: liping.qiu@uconn.edu

Abstract

Using a large panel data set, we investigate the dynamics of

hedge fund leverage from 2002 to 2017 and find consider-

able variations in both time series and cross section. More

than 70% of hedge funds use leverage and almosthalf of the

leveraged funds are levered through margin borrowing. On

average, hedge funds decreased leverage prior to the

beginning of the financial crisis, with leverage remaining

below the pre-crisislevels. We find that the level of leverage

and its changesare related to fund characteristicssuch as age,

governance, performance,risk, fees, liquidity, survival, and

the birth of a new fund.

KEYWORDS

financial crisis, hedge funds, leverage, newborn funds, survival

JEL CLASSIFICATION

G23, G28

1

|

INTRODUCTION

Excessive leverage is said to be the root cause of the global financial crisis in 2008 and the European

sovereign debt crisis.

1

The excessive use of leverage in the hedge fund industry has created concerns

for the risk of the industry and its impact on market stability. The debacle of Long Term Capital

Management (LTCM) in 1998 and the blowup of two Bear Stearns credit strategy funds in 2007, all

related to exorbitant leverage levels, heightened concerns over the risks associated with high leverage,

We thank participants at the 2013 FMA meetings and seminar participants at the University of Connecticut for helpful

comments.

1

Speech by Mr Norman T. L. Chan, Chief Executive of the Hong Kong Monetary Authority, at the Economic Summit 2012

‘Roadmap to Hong Kong Success,’Hong Kong, 9 December 2011.

Eur Financ Manag. 2018;1–34. wileyonlinelibrary.com/journal/eufm © 2018 John Wiley & Sons, Ltd.

|

1

a

t

l

t

l

i

c

l

m

L

l

a

i

F

o

c

l

f

S

l

h

f

p

s

l

(

t

m

s

t

h

o

r

c

J

l

a

t

W

a

t

d

2

L

908 © 2018 John Wiley & Sons, Ltd. wileyonlinelibrary.com/journal/eufm Eur Financ Manag. 2019;25:908–941.

n

s

l

l

,

2

1

and the possible impact on systemic risk. In a recent paper, Cao, Liang, Lo, and Petrasek (2018) found

that although hedge funds play a positive role in making the stock market more efficient, during a

liquidity crisis leveraged hedge funds are forced to deleverage and unwinding their positions can cause

the prices to move further away from the equilibrium. Hence, monitoring the dynamic usage of

leverage in hedge funds and investigating the implications of leverage are important issues not only for

investors, but also for the regulatory authorities.

The hedge fund industry is one of the fastest growing sectors in finance. Hedge funds are generally

characterized as highly leveraged and sophisticated investment vehicles. The widespread use of

leverage is one of the dimensions in which hedge funds differ from other managed portfolios. Unlike

mutual funds, hedge funds are not explicitly governed by regulations that limit their leverage level.

Leverage plays an essential role in hedge funds’investment strategies. Hedge funds not only use

leverage to take advantage of investment opportunities, but also vary leverage dynamically in order to

adjust market risk exposure, aiming at amplifying returns while attaining a level of volatility desired by

investors, which are otherwise unachievable.

Previous literature about hedge fund leverage focuses on four areas: (1) The estimation of leverage.

For example, McGuire and Tsatsaronis (2008) employ factor regressions to estimate hedge funds’use

of leverage. Eichengreen and Park (2002) show that hedge funds reduced their use of leverage and

credit following the Russia–LTCM crisis. Ang, Gorovyy, and Inwegen (2011) track hedge fund

leverage in time series from a data set provided by a fund of hedge funds. (2) The determinants of hedge

fund leverage. Eichengreen and Park (2002) find that larger funds tend to use more leverage.

Schneeweis, Martin, Kazemi, and Karavas (2004) document that hedge fund strategies with relatively

lower risk generally report higher leverage. Brown, Goetzmann, Liang, and Schwarz (2008) show that

hedge funds with higher operational risk tend to have lower leverage ratio. (3) The impact of hedge

fund leverage. Agarwal and Naik (2004) find no empirical evidence of significant under- or over-

performance between levered funds and unlevered funds. Similarly, Schneeweis et al. (2004) find no

statistically significant difference between the risk-adjusted performance of funds with below median

leverage usage and those with above median usage. (4) The dynamics of hedge funds. Fung and Hsieh

(1997) first point out that hedge fund strategies are highly dynamic. Chen and Liang (2007) illustrate

that the market exposure of self-claimed market-timing hedge funds varies with future changes in

market return and volatility. Bollen and Whaley (2009) show that a sizable percentage of hedge funds

shift their risk exposures significantly over time. Patton and Ramadorai (2013) propose a new method

to capture hedge fund dynamics by using high-frequency instruments.

Although the research on hedge fund leverage is not new, the nature of voluntary reporting for

hedge funds to data vendors complicates the monitoring of their use of leverage. In addition, research

on hedge fund leverage and the implication for hedge fund operations as well as market stability is

relatively scarce in the academic literature.

In this paper, we explore hedge fund leverage on a cross-sectional and time-series basis. Our data

consist of all hedge funds covered by the Lipper TASS Hedge Fund database (hereafter TASS) from

January 2002 to December 2017. Based on the monthly average leverage ratio and indicators of

leverage usage and leverage methods over time, we examine the following research questions: What

are the dynamics of hedge fund leverage? What drives the changes in leverage level over time? How do

these changes relate to each other and to a fund's past performance, flows, and other characteristics?

What are the effects of the changes in hedge fund leverage on future performance, risk-taking behavior,

and capital flows from investors? What determines the leverage level of newborn hedge funds? What is

the relation between hedge fund leverage and their survivorship?

Our empirical findings provide important insights on several theoretical predictions relating to the

dynamics of hedge fund leverage. Overall, more than 70% of hedge funds use leverage and almost half

2

|

LIANG AND QIULIANG AND QIU

|

3

909

1.

of the leveragedfunds are levered using marginborrowing. Among different leveragestrategies based on

the four major leveragemethods, margin borrowing alone is themost widely used strategy. But after the

financial crisis, thepercentages of leveraged hedge funds using only marginborrowing are much lower

than the percentages priorto the financial crisis. More and more hedge funds are levered using options

and/or forwards or combiningmargin borrowing with other leverage methods. On average,hedge funds

decreased their leveragein mid-2007 prior to the financial crisis. In fact, we find thatfunds newly added

to TASS over our sampleperiod, which is from January 2002 to December 2017, deleveragedan average

of 7 months earlier. We also find that when comparedto their peers, funds tend to enhance leverageusage

when the current fund returnis high, return volatility is low, and the fundis young. Funds with stronger

governance are generallyable to obtain credit more easily and thus are more likely to increase leverage

(see Brown et al., 2008; Brown,Goetzmann, Liang, & Schwarz, 2009, 2012). This finding is consistent

with established funds having good performance choosing a higher leverage level in order to augment

their returns furtherwhen both the fund and the lender feel safe to do so. On theother hand, hedge funds

tend to deleveragewhen the current leverage level is high. Deleveragedfunds also tend to charge a higher

management fee. This finding is in line with the general belief that fund managers decrease leverage

usage in order to reduce risk when the fund's leverage is already high and when the interests of the

managers and the investorsare better aligned. In the cross sectionof newborn hedge funds, we find that

from the macroeconomic perspective, higher Volatility Index (VIX) and Treasury over Eurodollar

(TED) spread lead to lower leverage. In addition to the economy-wide factors, fund-specific variables

have a profound impact on the leverage level of newborn funds. Specifically, shorter notice periods,

higher minimum investmentamounts, manager co-investments, higher incentivefees, high-water mark

provisions, and stronger fund governance are all associated with higher fund leverage.

We also examine the impact of leverage change on fund performance, risk, and investor flows. We

find that increases in leverage do not result in improved performance. On the contrary, the performance

for leverage increased funds is not only worse than their performance prior to the increases, but also

worse than their peers. Fund flows indeed increase after funds increase their leverage. Funds

deleverage when the current leverage ratio is high. We find that the difference between the reductions

in volatilities of deleveraged funds and of the unchanged funds is statistically significant. Deleveraging

lowers return volatility in deleveraged funds. We also find that funds employing leverage generally are

more likely to survive longer, and a higher leverage usage is associated with a longer lifetime,

consistent with the operational risk literature (see Brown et al., 2008, 2009, 2012).

Our study contributes to the research on hedge fund leverage in several ways. Firstly, our

exploration on the dynamics of hedge fund leverage is related to estimating the leverage. Estimating

hedge fund leverage is a difficult task as these funds have significant flexibility as to the types of assets

they can invest in. Weak reporting requirements for hedge funds complicate the monitoring of their use

of leverage. McGuire and Tsatsaronis (2008) use factor regressions with time-varying betas to estimate

hedge funds’use of leverage on the basis of public data. Employing the data drawn from two annual

surveys of 647 managers, Eichengreen and Park (2002) provide evidence that hedge funds reduced

their use of leverage and credit following the Russia–LTCM crisis. Ang et al. (2011) track hedge fund

leverage in time series from a data set provided by a fund of hedge funds over the period from

December 2004 to October 2009. They find that hedge fund leverage decreased prior to the start of the

financial crisis in mid-2007. In this paper, we investigate the dynamics of hedge fund leverage using the

monthly average leverage data from TASS over the period of January 2002–December 2017. In

addition to the leverage ratio, we also estimate leverage through the usage of margin borrowing,

futures, options and/or forwards, and FX-credit. Consistent with Ang et al. (2011), we find that on

average, hedge funds decreased their leverage in mid-2007. Secondly, our study on hedge fund

leverage is also related to the previous work on the determinants of hedge fund leverage. Eichengreen

LIANG AND QIU

|

3

a

h

d

i

f

e

i

r

i

l

l

l

c

A

b

i

f

m

s

l

t

B

m

f

r

s

h

a

u

m

r

t

c

w

c

h

h

R

i

p

u

d

a

S

F

4

L

2

|

LIANG AND QIU

910