Close Communications: Hedge Funds, Brokers and the Emergence of Herding
| Published date | 01 January 2017 |
| DOI | http://doi.org/10.1111/1467-8551.12158 |
| Date | 01 January 2017 |
British Journal of Management, Vol. 28, 84–101 (2017)
DOI: 10.1111/1467-8551.12158
Close Communications: Hedge Funds,
Brokers and the Emergence of Herding
Neil Kellard, Yuval Millo,1Jan Simon2and Ofer Engel3
Essex Business School, University of Essex, Wivenhoe Park,Colchester CO4 3SQ, UK, 1School of
Management, University of Leicester, University Road, Leicester LE1 7RH, UK, 2Beedie School of Business,
Simon Fraser University, 500 Granville Street,Vancouver, BC, Canada V6C 1W6, and 3Centre for the Analysis
of Time Series, Tower 1, 11th Floor, London School of Economics and Political Science, Houghton Street,
London WC2A 2AE, UK
Corresponding author email: nkellard@essex.ac.uk
We examinehow communication, evaluation and decision-making practices among com-
peting market actors contribute to the establishment of herding and whether this has im-
pact on market-wide phenomena such as prices and risk. Data are collected from inter-
views and observations with hedge fund industry participants in Europe, the USA and
Asia. We examine both contemporaneous and biographical data, finding that decision-
making relies on an elaborate two-tiered structure of connections among hedge fund
managers and between them and brokers. This structure is underpinned by idea shar-
ing and development between competing hedge funds leading to ‘expertise-based’herding
and an increased probability of over-embeddedness.We subsequently present a case study
demonstrating the role that communication between competing hedge funds plays in the
creation of herding and show that such trades aect prices by introducing an additional
risk: the disregarding of information from sources outside the trusted connections.
Introduction
As seen most notably in the global financial
crisis of 2008, unstable financial markets can
have a detrimental impact on today’s economy
and society. Amongst a variety of explanations
for such instability (Stein, 2015), a contributing
factor is often thought to be herding among
market participants. In the context of markets,
herding commonly refers to several actors making
the same investment decision either at the same
time or in close succession, leading to a high
concentration of similar market orders and higher
risks. Although there is much empirical evidence
of the existence of herding in the finance literature
(Jiao and Ye, 2014), there are few explorations of
the possible underlying social and organizational
practices through which herding in financial
markets may come about. In particular, the rich
conceptualization that evolved in the manage-
ment and economic sociology literatures around
the notion of embeddedness (e.g. Granovetter,
1985; Ingram and Roberts, 2000; Uzzi, 1996,
1997; Uzzi and Lancaster, 2003) has not been
utilized eectively. Motivated by this apparent
gap in empirical knowledge and conceptual-
ization, in this paper we analyse communica-
tion and information-sharing practices among
financial organizations and use this qualitative
empirical examination to contribute to a theory
on the conditions that enable and frame collabo-
rative decision-making between managers in these
competing organizations and, under certain con-
ditions, contribute to the emergence of herding.
Our data are drawn from hedge funds, a sec-
tor that is often considered emblematic of the
financial industry. Hedge funds, although often
managing billions of dollars in assets, are typically
‘boutique’, small in terms of employees. However,
like other larger financial market participants
such as investment banks or mutual funds, these
small firms are faced with an almost limitless
© 2016 British Academy of Management. Published by John Wiley & Sons Ltd, 9600 Garsington Road, Oxford OX4
2DQ, UK and 350 Main Street, Malden, MA, 02148, USA.
Close Communications 85
opportunity set of assets to trade. How do
they gain information, interrogate possibilities
and make decisions? Social and organizational
research in other contexts suggests that these
competitors may communicate closely (Ingram
and Roberts, 2000; Uzzi and Lancaster, 2003).
If they do, does this contribute to herding and,
subsequently, eects on market-wide measures
such as prices and risk?
The qualitative empirical material was collected
through interviews with a large number of hedge
fund professionals and field visits to hedge funds
and brokerage firms. We corroborate the quali-
tative empirical findings by constructing a map
of the interactions-based connections, analysing
them in the light of the institutional, biographical-
historical and geographical datacollected, and cal-
culating the relevant social network measures. Fi-
nally, to illustrate the impact of communication
practices, over-embeddedness and its outcomes,
we study a single trading position held by a num-
ber of connected firms, over a period of several
months. By combining these three modes of inves-
tigation we aim to capture more comprehensively
than previous research the multifaceted nature of
the phenomenon of investment-related communi-
cation among professional financial investors.
To understand the communication practices
among competing hedge funds we first appeal to
Podolny(2001) and, in particular, the uncertainties
around both market opportunities and the qual-
ity of other actors that drive the use of selective,
close-knit ties. The shared and repeatedanalysis of
information that takesplace between such ties mo-
tivated us to regard this conceptuallyas ‘expertise-
based’ herding, where actors who consider each
other ‘smart’ adopt similar trades. Additionally,
the seminal work initiated byUzzi (1996) and Uzzi
and Lancaster (2003) suggests that the structure
of connections among economic actors can also
become over-embedded, whereby a limited set of
ideas is circulated among close ties. This can ad-
versely aect the actors through their exposure
to a limited information set leading to insulated
and potentially risky decision-making. We argue
that the eects of the sociological concept of over-
embeddedness can be extended into financial mar-
ket theory, as tightly grouped hedge funds under-
weight relevant information about prices and risk
from sources outside the trusted connections.
The remaining parts of the paper are divided
into seven sections. In the nextsection we consider
the literature on communication and herding in fi-
nancial markets, whilst the following section pro-
vides an overview of the hedge fund industry
and likely social connections. Methods and data
are then presented. Interviews and field observa-
tions are used to explore communication practices
whilst the followingsection provides the historical-
biographical origins of hedge funds’ communica-
tion practices. The penultimate section examines
the emergence of herding and wider market
risks and finally a discussion and conclusion are
provided.
Communication and herding in financial
markets
To begin our explanation of herding in financial
markets we turn first to the organizational liter-
ature that has explored the conditions that af-
fect communicationand decision-making amongst
participants in markets per se. An important no-
tion is that economic activity is embedded in
pre-existing networks of social ties (Baker, 1984;
Granovetter, 1985; Uzzi, 1996, 1997), where ex-
change of information (Ingram and Roberts, 2000)
and organizational learning (Uzzi and Lancaster,
2003) take place.In addition to social connections,
the widespread adoption of similar trading soft-
ware and financial models (Callon and Muniesa,
2005; Zaloom, 2003) may serve as a conduit for
imitation between dierent investors, thus increas-
ing the homogeneity of views in financial markets
(Beunza and Stark, 2012; MacKenzie, 2003).
By contrast with its organizational counterpart,
the mainstream finance literature has examined
investor behaviour such as herding by focusing on
prices or trades and then, via some quantitative
approach, noted whether these are consistent with
investor rationality. For example, a body of work
has observed correlated trades and subsequently
suggested that investors herd because they have
access to the same public information or they
infer useful information from each other’s trading
patterns or they mimic others’ trades for reputa-
tional and benchmarking concerns (Boyson, 2010;
Graham, 1999; Grinblatt, Titman and Wermers,
1995; Lakonishock, Shleifer and Vishny, 1992;
Sias, 2004). However, the finance literature pays
little attention to the possible social mechanisms
behind herding. Exceptions include Hong, Kubik
and Stein (2005), Cohen and Frazzini (2008) and
© 2016 British Academy of Management.
To continue reading
Request your trial