A Time to Scatter Stones, and a Time to Gather Them: The Annual Cycle in Hedge Fund Risk Taking

Published date01 November 2018
Date01 November 2018
AuthorAchim Mattes,Olga Kolokolova
DOIhttp://doi.org/10.1111/fire.12169
The Financial Review 53 (2018) 669–704
A Time to Scatter Stones, and a Time to
Gather Them: The Annual Cycle in Hedge
Fund Risk Taking
Olga Kolokolova
Alliance Manchester Business School, University of Manchester
Achim Mattes
The University of Konstanz
Abstract
Analyzing a sample of hedge fund daily returns from Bloomberg, we find a seasonal
pattern in their risk taking. During earlier months of a year, poorly performing funds reduce
risk. The reduction is stronger for funds with higher management fees, shorter redemption
periods, and recently deteriorating performance, consistent with a managerial aversion to
early fund liquidation. Toward the end of a year, poorly performing funds gamble for res-
urrection by increasing risk. It is largely achieved by increasing exposure to market factors,
and can be linked to stronger indirect managerial incentives during the second half of a
year.
Corresponding author: Alliance Manchester Business School, University of Manchester, Booth
street East, Crawford House 2.6, Manchester M15 6PB, UK; Phone: +44 (0)161 3062081; E-mail:
olga.kolokolova@manchester.ac.uk.
We thank M. Amman, G. Barone-Adesi, M. Bowe, M. Braun, M. Brennan, J. Carpenter, S. Darolles,
G. Franke, J. Jackwerth, A. Kostakis, A. Plazzi, S.-H. Poon, N. Wang, and the participants of the 13th
International Research Conference on Finance, Risk and Accounting Perspectives in Cambridge 2013, the
6th Annual Hedge Fund Conference in Paris2014, the American Finance Association annual meeting 2015,
and the seminars at the Swiss Finance Institute, Universityof Lugano 2015, the EMLYON Business School
2015, and the University of Chile 2017 for helpful comments and discussions. Wethank DAAD for partly
financing this research. Earlier versions of this paper have been circulated under the title “Recovering
managerial incentives from daily hedge fund returns: incentives at work?” All remaining errors are
our own.
C2018 The Eastern Finance Association 669
670 O. Kolokolova and A. Mattes/The Financial Review 53 (2018) 669–704
Keywords: hedge funds, risk taking, incentives, seasonality
JEL Classifications: G11, G23
1. Introduction
The optimal managerial action in response to incentives is a timeless topic in
management science. Hedge funds have gained a lot of attention in the recent years,
as they provide a natural playground for such analysis. On the one hand, typical
compensation contracts of hedge fund managers create complex incentive schemes.1
On the other hand, being loosely regulated, hedge fund managers have direct control
over the fund risk.
There is ongoing debate in the theoretical literature on the optimal response of
hedge fund managers to performance. The literature generally predicts two alternative
reactions to poor performance. Offensive (increasing) risk taking is expected for fund
managers with a finite optimization horizon (Hodder and Jackwerth, 2007; Buraschi,
Kosowski and Sritrakul, 2014). Defensive (decreasing) risk taking is expected for
managers with an infinite optimization horizon (Lan, Wang and Yang, 2013).
Besides making contradictory theoretical predictions, recent literature also high-
lights the importance of indirect managerial incentives. Lim, Sensoy and Weisbach
(2016) analyze managerial incentives resulting from investor inflows in response to
good performance. They find that the indirect link between compensation and per-
formance via future fees on inflows creates stronger incentives than the direct link
from the incentive contract.
The empirical evidence on the magnitude and the pattern of risk taking is,
however, still scarce. Some papers do not find any significant performance-risk
relation at all (Brown, Goetzmann and Park, 2001), while others find offensive risk
taking (Aragon and Nanda, 2012; Buraschi, Kosowski and Sritrakul, 2014). We aim
at closing this gap in the literature and recovering the full pattern of managerial risk
taking empirically.
We use a previously unexamined sample of daily hedge fund returns from
Bloomberg. While the hedge funds in our sample are very similar to the majority of
funds reporting monthly returns with respect to their risk taking, the higher reporting
frequency allows us to estimate fund risk on a monthly basis. A semiparametric
panel regression approach allows us to capture nonlinearities in the performance-risk
1A typical compensation contract includes a management fee, which is a constant share of the fund’sassets
paid out on a pro rata temporis basis, and a performance fee, calculated as a share of the fund’s profits
in excess of a high-water mark (HWM; previously achieved end-of-year maximumnet asset value), often
paid at the end of a calendar year. Theoretically,such a compensation structure induces highly nonlinear
managerial risk taking (e.g., Hodder and Jackwerth, 2007; Panageas and Westerfield, 2009; Lan, Wang
and Yang,2013).
O. Kolokolova and A. Mattes/The Financial Review 53 (2018) 669–704 671
relation. This setup further enables us to analyze potential mechanisms behind the
observed risk shifts.
Thereby, we make several contributions to the literature. We reveal a highly
nonlinear performance-risk relation with a strong seasonal pattern over the course of
a calendar year. Conditional on fund underperformance relative to the HWM, hedge
fund managers reduce the risk during earlier months of a year, but strongly increase
it toward the end of a year.2The risk alterations are economically significant and
range from an average 14% decline to a 20% increase relative to the expected level
of risk. The observed nonlinearity offers a potential explanation for the absence of
significant results in some previous papers that use linear specifications. And the
detected seasonality allows reconciling theoretical predictions on offensive versus
defensive risk taking.
Our additional tests suggest that during earlier months of a year, hedge fund
managers are mainly concerned with fund survival. Consistent with the theoretical
predictions in Lan, Wang and Yang (2013), funds with a shorter notice period prior to
redemption, recently deteriorating performance, or younger age, corresponding to a
higher liquidation probability, exhibit a stronger risk reduction. During later months
of a year, the focus of poorly performing fund managers shifts toward delivering
high returns. We find that the flow-performance relation becomes stronger toward
the end of a year, compared to the beginning of a year.Such seasonal variation in the
flow-performance sensitivity (FPS) impacts the tradeoffbetween implicit and explicit
incentives of fund managers. Accordingly, the end-of-year risk increase is driven by
funds that are not capacity constrained and are expected to have a higher FPS (Lim,
Sensoy and Weisbach, 2016). It is also more pronounced during times when the
market performs poorly and the competition for investor flowsbecomes more severe.
Remarkably,the end-of-year gamble by poorly performing funds is independent from
the actual compensation structure and strongly pronounced for funds not charging
incentive fees, too. Our results confirm a material impact of indirect flow-related
incentives (Lim, Sensoy and Weisbach, 2016), which are especially pronounced
toward the end of the year.3
Our analysis also offers insights in the operational implementation of risk shifts.
De-risking during the first half of a year is achieved by proportionally reducing
exposure to market risk factors and idiosyncratic risk. Increasing the risk toward the
end of a year, however, is disproportional. The increase in market risk is almost twice
the idiosyncratic risk increase. Thus, hedge fund investors should not only be aware
of seasonal variations in the risk taking, but also of a changing risk composition
underling the returns over the course of a calendar year.
2We also detect a similar seasonal pattern using a larger group of hedge funds reporting conventionally
on a monthly basis (see Section 6.3). The lower frequency of reporting, however, does not allow for an
analysis as detailed as for the group of daily reporting funds.
3Chevalier and Ellison (1997) are one of the firstto suggest that convexity of fund flows implicitly creates
convexity of managerial compensation evenin the absence of an HWM provision.

To continue reading

Request your trial

VLEX uses login cookies to provide you with a better browsing experience. If you click on 'Accept' or continue browsing this site we consider that you accept our cookie policy. ACCEPT