Pdf finance and economics discussion series divisions of research

Pdf File 256.23 KByte, 45 Pages

Title: Corporate Hedging, Investment and Value
Subject: FEDS Working Paper
Keywords: Currency derivatives, firm value, underinvestment, debt capacity, cost of capital
Author: Jose M. Berrospide, Amiyatosh Purnanandam, and Uday Rajan
Creator: TeX
Producer: MiKTeX pdfTeX-1.40.3
CreationDate: Mon Mar 24 22:11:44 2008
ModDate: Mon Mar 24 22:11:44 2008
Tagged: no
Form: none
Pages: 45
Encrypted: no
Page size: 612 x 792 pts (letter) (rotated 0 degrees)
File size: 262383 bytes
Optimized: no
PDF version: 1.4

  • Finance and Economics Discussion Series Divisions of Research ...
  • Finance and Economics Discussion Series Divisions of ...
  • PDF Finance and Economics Discussion Series Divisions of Research ...

PDF Finance and Economics Discussion Series Divisions of Research ...

Finance and Economics Discussion Series
Divisions of Research & Statistics and Monetary Affairs
Federal Reserve Board, Washington, D.C.
Corporate Hedging, Investment and Value
Jose M. Berrospide, Amiyatosh Purnanandam, and Uday Rajan
2008-16
NOTE: Staff working papers in the Finance and Economics Discussion Series (FEDS) are preliminary
materials circulated to stimulate discussion and critical comment. The analysis and conclusions set forth
are those of the authors and do not indicate concurrence by other members of the research staff or the
Board of Governors. References in publications to the Finance and Economics Discussion Series (other than
acknowledgement) should be cleared with the author(s) to protect the tentative character of these papers.
Corporate Hedging, Investment and Value
Jose M. Berrospide Amiyatosh Purnanandam Uday Rajan?
February 2008
Abstract
We consider the effect of hedging with foreign currency derivatives on Brazilian firms in the
period 1997 through 2004, a period that includes the Brazilian currency crisis of 1999. We find
that, derivative users have valuations that are 6.7-7.8% higher than non-user firms. Hedging
with currency derivatives allows firms to sustain larger capital investments, and also removes the
sensitivity of investment to internally generated funds. Thus, it mitigates the underinvestment
friction of Froot, Scharfstein, and Stein (1993), at a time when capital in the economy as a
whole is scarce. We further show that hedging increases the foreign currency debt capacity of a
firm, and that foreign debt is a cheaper source of capital than domestic debt during our period
of study.
JEL Codes: G30, G32.
Keywords: Currency Derivatives, Firm Value, Underinvestment, Debt Capacity, Cost of Capital.
We would like to thank Tim Adam, Sudheer Chava, Kathryn Dominguez, Jie Gan, Han Kim, Peter Mackay,
Paige Parker, Amit Seru, Rene Stulz, Linda Tesar, Jing Zhang, and seminar participants at CSSS (Calcutta), Hong
Kong UST, Michigan, National University of Singapore, Notre Dame, and Singapore Management University for
helpful comments. All errors remain our own. The views expressed are those of the authors and do not necessarily
reflect those of the Board of Governors of the Federal Reserve System or its staff.
Federal Reserve Board; Tel: (202) 452-3590, E-mail: jose.m.berrospide@
Ross School of Business, University of Michigan; Tel: (734) 764-6886, E-mail: amiyatos@umich.edu
?Ross School of Business, University of Michigan; Tel: (734) 764-2310, E-mail: urajan@umich.edu
1 Introduction
Despite the widespread use of derivatives by firms, and the impressive growth in the derivatives
market in the last two decades, empirical evidence on the effect of derivative usage on firm value
is limited and often mixed. There are several reasons for this. First, there is a lack of data on
the extent and direction of a firm's exposure to risk. It is often non-trivial to even identify the
frictions that hedging may help overcome. Second, data on the kinds of derivatives used by a firm,
and indeed on whether firms are hedging or speculating, may be unavailable. Finally, hidden or
unobservable firm characteristics introduce additional complications that affect the interpretation
of empirical results.
For example, suppose we find that a firm using derivatives has a higher value than an
observationally identical non-user firm. It is possible that derivative usage is correlated with an
omitted variable that has a positive effect on firm value, such as managerial skill. We may then
incorrectly conclude that derivative usage has a positive effect on firm value. Therefore, if one finds
a link between firm value and derivative usage, it is an equally important empirical task to identify
the specific channels via which derivatives might add value to a firm.
We study the effect of derivative usage on firm value and financial policy in a setting that
is less likely to suffer from some of these concerns. We consider a panel of Brazilian firms in a
period surrounding the 1999 currency crisis. Until January 1999, Brazil had a managed exchange
rate policy (a "crawling peg") for its currency, the real. Following a severe currency crisis, the
policy was abandoned in favor of a fully floating exchange rate in January, 1999. The economy
suffered through a large recession as a result of the crisis. In this period, many firms in Brazil had
liabilities denominated in foreign currency, primarily US dollars. During the managed exchange
rate regime these firms faced minimal exposure to foreign currency risk. However, this exposure
increased dramatically in the floating exchange rate regime. While only a handful of firms hedged
their currency exposure prior to 1999, the usage of currency derivatives increased steadily from
1997 to 2004, our period of study, with over 40% of firms in the sample using currency derivatives
1
by the end of the period.1
Our setting has three noticeable advantages. First, following the radical change in exchange
rate policy and the upheaval in the economy, Brazilian firms were in a state of disequilibrium. It
would require some time for firms to determine what their optimal policies should be in the new
environment. If all firms are behaving optimally, the effect of hedging on value may be hard to
detect in equilibrium.2 That is, if hedging does appear to be of value, firms that do not hedge must
either be behaving sub-optimally, or perhaps have some hidden cost to hedging relative to hedged
firms that is unobserved by the econometrician.
Second, the nature of the risk faced by firms in our study is clear. A large number of firms
in our sample had some liabilities denominated in foreign currency over the period of study. In
the managed exchange rate period, such positions were implicitly hedged via government policy.
Once the exchange rate was floated, firms were required to hedge on their own to protect against
a further fall in the Brazilian real. Thus, both the source of risk and the direction of the hedge are
immediate to deduce.
Finally, the crisis itself was exogenous to the actions of any particular firm, though of course it
came about as a result of a weakness in the overall economy. A typical problem in any regression
model of firm value is the issue of endogeneity between firm value and the decision variable of
interest, such as the hedging decision in our case. A natural experiment that randomly assigns
firms to be hedgers or non-hedgers would be an ideal research design to get around this problem.
Since the decision to hedge is endogenous, our setting does not represent a pure natural experiment.
Nevertheless, it is a setting in which the exogenous change in the exchange rate regime led to a
corresponding exogenous change in the marginal value of currency hedging, which was close to zero
in the managed exchange rate period. If hedging has any value, its value should be demonstrable
when such a large shock occurs. By comparing hedgers and non-hedgers as well as by comparing
a firm's own value before and after it started hedging, our experiment design is less likely to be
affected by endogeneity issues.
1We use the terms "hedging" and "using derivatives" synonymously in this paper.
2For example, in the model of Adam, Dasgupta, and Titman (2007), in equilibrium some firms hedge and others
do not, but each firm is indifferent between hedging and not and has the same value in either case.
2
We find in pooled regressions that hedgers have significantly higher market valuations (as
measured by the market-to-book ratio or Tobin's q) than their non-hedger counterparts, after
controlling for key drivers of firm value such as firm size, profitability and growth potential. To
reduce the likelihood that the result is due to unobservable factors correlated with derivative usage,
we estimate a firm fixed effects model, which by design controls for any omitted variables as long
as they remain constant over time. We continue to find a large valuation premium (of 6.7-7.8%)
for currency derivative usage. This suggests that a substantial portion of the valuation effect that
we document in the pooled regressions is causal in nature.
We then explore the reasons behind the valuation premium we obtain. The theoretical literature
posits that hedging allows firms to overcome frictions due to bankruptcy costs (Smith and Stulz
(1985)). Stulz (1996) suggests that the primary reason to hedge is to reduce the likelihood of costly
"left-tail" events such as financial distress or an inability to carry out an investment policy. In
a dynamic setting, Leland (1998) shows that the reduction in the probability of financial distress
leads to increased debt capacity and thus to a higher firm value via increased interest tax shields. A
different approach is offered by Froot, Scharfstein and Stein (1993), who show that, when external
financing is costly, hedging can remove inefficiencies from a firm's investment decisions by providing
capital in low cash flow states. At the empirical level, the extant literature has provided significant
insights into the debt capacity channel of hedging using data from the US (see Graham and Rogers
(2002)), but the investment channel has largely remained unexplored.
Our study contributes to the literature by showing that hedging helps firms smooth their
investment policies. We find that hedgers invest considerably more than non-hedgers, controlling
for the investment opportunity set at the industry level. Just as importantly, we find that a hedged
firm's investment is not sensitive to its operating cash flow. A non-hedger, on the other hand,
is significantly dependent on its internal funds for investment. This provides direct evidence in
support of the Froot, Scharfstein, and Stein (1993) model. A basic friction faced by Brazilian firms
in our period of study is a shortage of capital, which results in underinvestment by non-hedged
firms.3 Hedging enables firms to maintain capital expenditure at a time of overall economic stress,
3Over a different period of study, Bonomo, Martins, and Pinto (2003) find a significant drop in corporate investment
3
and thus mitigates the underinvestment friction.
To examine why the level of investment is higher for hedgers in our sample, we consider the
effects of derivative usage on leverage. In our setting, the debt channel has an additional effect
on firm value as compared to the Leland model. As we show in Section 2.2, the intermediation
capability of the domestic banking sector was severely hampered during the period of our study,
with low bank credit availability and a high spread between lending and deposit rates. In such an
environment, foreign currency debt can be a cheaper source of capital than domestic debt. Thus, in
addition to the extra tax-shields generated by additional debt, currency derivatives may add value
through their effect on the cost of capital to the firm. While we do not have direct data on the cost
of domestic and foreign debt, we can infer their relative costs from the financial expenses and net
incomes reported by firms. We show that a real of foreign currency debt is cheaper for a hedger
firm than a real of domestic debt. We further demonstrate, both in univariate analysis and via
regressions, that hedged firms have significantly greater foreign leverage, and a greater proportion
of foreign currency debt.4 The level of domestic debt is the same across hedged and unhedged
firms. Thus, overall, hedged firms do have higher leverage, as in the Leland (1998) model.
Finally, we account for the likely endogeneity between the decision to hedge and the extent
of foreign currency debt taken on by a firm. We first model the hedging decision of a firm via
a hazard rate process. We then use the key determinants of hedging in an instrumental variable
estimation of derivative usage and foreign currency debt. Even after accounting for the endogeneity
of these decisions, we find that hedged firms have significantly higher foreign currency debt than
non-hedgers. Overall, therefore, our results show that in a segmented capital market, derivative
usage can add value by increasing the firm's access to a cheaper source of capital.
There has been much empirical work on the effect of derivative usage on firm value in recent
years. Using a sample of US firms over the period 1990-95, Allayanis and Weston (2002) estimate
that users of foreign currency derivatives command an average premium of 4.87% in firm value.
However, Guay and Kothari (2003) suggest that the extent of derivative usage by US firms is too
after a 1991 devaluation in Brazil.
4Berrospide (2007) extends the Holmstrom and Tirole (1997) model of external financing to an open economy,
and demonstrates theoretically that hedging can increase foreign debt capacity.
4

Download Pdf File Online Preview