J. of Multi. Fin. Manag.

16 (2006) 142159

Empirical evidence concerning incentives to hedge

transaction and translation exposures
Niclas Hagelin, Bengt Pramborg
School of Business, Stockholm University, Stockholm SE-10691, Sweden
Received 30 March 2005; accepted 15 May 2005
Available online 7 July 2005

Abstract
We investigate Swedish firms use of financial hedges against foreign exchange exposure. Our
survey data lets us distinguish between translation exposure and transaction exposure hedging. Survey
responses indicate that over 50% of the sampled firms employ financial hedges, and that transaction
exposure is more frequently hedged than is translation exposure. The likelihood of using financial
hedges increases with firm size and exposure, and liquidity constraints are important in explaining
transaction exposure hedging. Importantly, the existence of loan covenants accounts for translation
exposure hedging, suggesting that firms hedge translation exposure to avoid violating loan covenants.
2005 Elsevier B.V. All rights reserved.
JEL classication: F23; F31
Keywords: Hedging; Foreign exchange exposure; Translation exposure; Loan covenants; Bond covenants

1. Introduction
Why widely held firms, whose owners have the ability to hold diversified portfolios of
securities, elect to hedge their exposures is the subject of debate. Theoretical research has
shown that risk-reduction efforts may enhance value if they reduce cash-flow variability,
thus mitigating costs associated with market imperfections (see Smith and Stulz, 1985;
Bessembinder, 1991; Froot et al., 1993).

Corresponding author. Tel.: +46 8 674 7427; fax: +46 8 674 7440.
E-mail address: bpg@fek.su.se (B. Pramborg).

1042-444X/$ see front matter 2005 Elsevier B.V. All rights reserved.
doi:10.1016/j.mulfin.2005.05.003

N. Hagelin, B. Pramborg / J. of Multi. Fin. Manag. 16 (2006) 142159

143

One important source of risk that is widely hedged is foreign exchange (FX) exposure
(see, e.g. Bodnar et al., 1996, 1998; Berkman et al., 1997; Alkeback and Hagelin, 1999;
Bodnar et al., 2003). Numerous studies have investigated whether firms use of currency
derivatives conforms to the predictions of theoretical research.1 Mian (1996) examines the
currency derivatives usage of a sample of US firms, finding that hedging exhibits economies
of scale; however, Mian finds only weak support for theoretical models that seek to explain
hedging activities. Like Mian (1996), Geczy et al. (1997) apply a dichotomous measure
of currency derivatives use to a sample of US firms, and find that firms with extensive FX
exposure and economies of scale are more likely to use currency derivatives. They also
find that firms with less liquidity and greater growth opportunities (as measured by R&D
expenses) are more likely to use currency derivatives. They interpret this as evidence that
firms use currency derivatives to reduce cash-flow variation that might otherwise preclude
their investing in valuable growth opportunities.
Graham and Rogers (2000) investigate 161 US firms and report that they hedge in
response to expected financial distress costs, firm size, and investment opportunities. They
also show that firms hedge to increase debt capacity, though not in response to convexities
in tax schedules. Allayannis and Ofek (2001) analyze a sample of 378 US firms to uncover
the characteristics associated with the adoption of particular levels of currency derivatives
use. They corroborate the findings of Geczy et al. (1997) concerning the decision to use
currency derivatives, and show that the degree of FX exposure is the only significant factor
explaining the notional value of currency derivatives used. Bartram et al. (2004) examine a
large sample of firms from 48 countries and find that firm-specific factors associated with
derivatives use vary little from country to country. Lel (2004), like Bartram et al. (2004),
examines a large sample of firms from various countries, finding that economies of scale,
growth opportunities, and financial distress costs are the firm-specific determinants of FX
exposure hedging. Brown (2001) studies an American manufacturing firm in depth, and
somewhat surprisingly finds that management tends to focus on the impact of hedging on
reported earnings rather than directly on cash flow.
Firms can also manage FX exposures through specific operational strategies, in addition to using financial hedges. Martin et al. (1999) find that changes in FX exposure are
influenced by operational adjustments. Importantly, Allayannis et al. (2003) find that operational hedging is not an effective substitute for financial hedging, but that it may benefit
shareholders when used in combination with financial hedging.
In addition to choosing among various hedging techniques, firms must also decide which
type of FX exposure to hedge. FX exposure can be categorized as either economic exposure,2
1 Nance et al. (1993), Dolde (1995), Berkman and Bradbury (1996), Gay and Nam (1998), and Guay (1999)
analyze why firms use derivatives in hedging, but do not focus on any particular type of exposure. Tufano (1996)
and Haushalter (2000) examine why firms use derivatives to hedge gold and oil price exposure, respectively.
2 Economic exposure is usually broken into two components: the exposure of identifiable anticipated transactions
and competitive exposure (the exposure of unidentifiable future cash flows). Our survey does not include hedging
of competitive exposure using financial instruments, which could possibly bias our results. The reason for not
investigating competitive exposure is that few firms tend to hedge this type of exposure with currency derivatives
(see, e.g. Bodnar et al., 1996). Instead, the longer-termed competitive exposure is typically managed by operational
techniques. From an examination of annual reports, it is suggested that this is also the case for Swedish firms.
Firms may however use foreign-denominated debt for competitive exposure hedging, and for those firms that use

144

N. Hagelin, B. Pramborg / J. of Multi. Fin. Manag. 16 (2006) 142159

transaction exposure, or translation (TL) exposure. TL exposure, in contrast to economic

exposure and transaction exposure, has no direct cash-flow effects. For this reason, standard
hedging theory does not apply to TL exposure hedging as it does to transaction exposure,
and the finance literature generally recommends not hedging TL exposure (see, e.g. Butler,
2004; Eiteman et al., 2001).3 Nonetheless, some firms do hedge their TL exposure (see, e.g.
Bodnar et al., 1996, 1998; Berkman et al., 1997; Alkeback and Hagelin, 1999; Bodnar et al.,
2003), and empirical evidence suggests that managers and investors care about TL exposure
(see, e.g. Aggarwal, 1991; Martin et al., 1998). Importantly, Martin et al. (1998) note that
TL exposure may have indirect cash-flow consequences as it affects reported balance sheet
accounts. For example, changes in net worth may affect borrowing capacity and cost of
capital.
Focusing on Swedish firms exposures, Hagelin (2003) examines the use of currency
derivatives for transaction and TL exposure hedging. Hagelins (2003) results support the
conjecture that firms hedge their transaction exposure so as to increase firm value by reducing
costs associated with market imperfections. However, no evidence is found to support the
notion that TL exposure hedges are used to increase firm value.
This study analyzes the association between characteristics of firms and their hedging of
transaction and translation exposures. Our major contribution, however, is that we examine
the relationship between TL exposure hedging and the existence of loan covenants. Firms
may well hedge TL exposure because of loan covenants requiring them to uphold certain
standards of financial performance as a condition for retaining access to funds, and such
performance measures are affected by translation gains and losses (see, e.g. Butler, 2004;
Eiteman et al., 2001). Smith and Warner (1979) suggest that firms enter into covenants
that may be affected by TL gains and losses. They show that covenants based, implicitly
or explicitly, on net worth are common. To the best of our knowledge, the present study
is the first attempt to document empirically the relationship between the existence of loan
covenants and TL exposure hedging.4

foreign debt for this purpose only, we would record them as non-hedgers although they use financial instruments
for hedging purposes. We expect this problem to be relatively small. Butler (2004) provides a discussion on the
problems associated with using financial hedges to manage longer-termed competitive exposure (for more on this
topic see also Chow et al., 1997).
3 TL exposure depends on the translation method used. The International Accounting Standard 21 (IAS 21)
suggests using the current rate method for self-contained foreign affiliates and the temporal method for integrated
foreign affiliates and for foreign affiliates in countries with high inflation. Under the current rate method, exposure
is the net amount of assets and liabilities translated at the current exchange rate. Changes in exchange rates as they
affect foreign operations thus always cause changes in group equity; under the temporal method, these changes
also affect group net income. The Swedish firms examined in this study follow the Swedish Institute of Authorized
Public Accountants proposal to recommendation, which builds on IAS 21. In practice, the current rate method
dominates among Swedish firms.
4 Other studies do examine the effect of loan covenants. For example, Core and Schrand (1999) use an option
pricing framework to model equity valuation for cases in which firms face costs associated with violating
accounting-based debt covenants. Their model shows that the value of equity depends on two factors: the economic value of the firm and the probability of the firm violating the covenant. Other examples include Roberts
and Viscione (1984), who examine the effect of covenants on bond yields, and Leland (1994) who investigates the
effect of covenants on optimal capital structure.

N. Hagelin, B. Pramborg / J. of Multi. Fin. Manag. 16 (2006) 142159

145

Questionnaire data covering a sample period from 1998 to 2001 is used to study whether
recorded relationships hold over time. We define hedging as including not only hedging
with currency derivatives, but also hedging with foreign-denominated debt. This, together
with the use of data covering four years, should improve the reliability of the results.
Use of financial hedges to reduce FX exposure is widespread, and over 50% of the sampled firms hedge FX exposure. We find that transaction exposure is more frequently hedged
than TL exposure, although up to 20% of the firms hedge their TL exposure. The survey
responses indicate that about one-tenth of the sample firms changed their hedging polices
concerning TL, committed transaction (CT), and anticipated transaction (AT) exposures
each year in favor of adopting hedging; these policy changes seldom meant that firms quit
hedging.
We estimate logit regressions to analyze firm characteristics associated with the likelihood of using financial hedges. In line with prior studies of US firms (Mian, 1996; Geczy
et al., 1997; Graham and Rogers, 2000; Allayannis and Ofek, 2001), we find that larger
firms are more likely to employ financial hedges. We also find that firms with greater FX
exposure are more likely to use financial hedges. Our findings regarding firm size and FX
exposure are consistent with the conjecture that fixed costs act as a barrier to hedging for
small firms and firms with less FX exposure. The logit results also show that the likelihood
of using financial hedges for CT exposure increases with decreasing liquidity, supporting
the view that firms with low liquidity hedge to reduce the probability of encountering financial distress, as suggested by Nance et al. (1993). Interestingly, we document a significant
positive association between TL exposure hedging and the existence of loan covenants; this
suggests that firms hedge TL exposure to avoid violating loan covenants.
There is still the question of to what extent the results of this study may be generalized
to non-Swedish firms. The results of Bartram et al. (2004) and Lel (2004) indicate that
firm-specific factors associated with derivatives use vary little from country to country
(including Sweden); this suggests that our results could well be valid for non-Swedish
firms. This perception is also supported by the across-the-board similarity in terms of
hedging practices, indicated in recent survey studies (see Bodnar et al., 1996; Berkman
et al., 1997; Alkeback and Hagelin, 1999; Bodnar and Gebhardt, 1999; De Ceuster et al.,
2000; Mallin et al., 2001; Bodnar et al., 2003), among firms in different countries.5
The study is organized as follows: Section 2 describes the sample selection procedure
and presents our findings regarding firms FX exposure and hedging practices. Section
3 describes the research design and defines the variables; the cross-sectional results are
presented in Section 4, which is followed by a conclusion.

5 However, survey evidence suggests that US firms and firms in other countries may differ in their focus on
accounting numbers. For example, Berkman et al. (1997) found that New Zealand firms were more likely to focus
on accounting earnings than were US firms. The relatively high importance of accounting earnings has also been
documented in surveys on European firms; see, e.g. Bodnar and Gebhardt (1999) for evidence on German firms,
De Ceuster et al. (2000) for Belgian firms, Mallin et al. (2001) for UK firms, and Alkeback et al. (2006) for Swedish
firms. This difference in focus may not result in differences in hedging practice (see, e.g. Bartram et al., 2004;
Lel, 2004; Pramborg, 2005), and our belief is that the results of this study may be generalized to non-Swedish
firms. However, because of the differences suggested by the above mentioned surveys, possibly due to institutional
differences, research covering other markets is warranted.

146

N. Hagelin, B. Pramborg / J. of Multi. Fin. Manag. 16 (2006) 142159

2. Sample selection, FX exposure, and hedging practices

2.1. Sample selection
Because detailed data concerning firms FX exposures and hedging practices are not
publicly available, we used two questionnaires to determine firms exposures and use of
financial hedges.6 The first questionnaire was sent to 275 firms in March 2000, and contained
questions concerning the respondents inherent exposures and hedging policies. The second
questionnaire was sent to 261 firms in September 2001. The two questionnaires are nearly
identical; however, the second asked firms whether they were subject to loan covenants
or not. The questionnaires of 2000 and 2001 asked about the 19981999 and 20002001
periods, respectively. The complete questionnaires are presented in Hagelin and Pramborg
(2004).
The questionnaires were sent to firms meeting the following three criteria: (i) the firm
was listed on the Stockholm Stock Exchange, (ii) the firm was non-financial, and (iii) the
firm was headquartered in Sweden. Financial firms were excluded because the focus of
the study is on end users rather than producers of financial services. Foreign firms those
headquartered outside Sweden were excluded to eliminate inter-firm differences arising
from differences in accounting standards between countries.
One hundred thirty and 128 usable responses were obtained from the first and second
questionnaires, representing response rates of 47% and 49%, respectively. The two surveys
produced a total of 516 firm-year observations. To check for non-response bias, responding
firms were compared with non-responding firms. Specifically, we compared whether the
two groups differed in terms of industry classification and the accounting variables that we
use to explain hedging (see Section 3). No significant differences were found, indicating
that no response bias exists.7
2.2. FX exposure and hedging practices
The questionnaires asked what proportions of total revenues (FR), total costs (FC), and
total equity (FE), measured in terms of book value, were denominated in foreign currencies
each year. Respondents were also asked whether they hedged FX exposure with currency
derivatives or foreign-denominated debt, and, if so, whether their hedging concerned TL,
CT, and/or AT exposures. One type of foreign exchange exposure that is not included in
the questionnaire is long-term competitive exposure. This has the potential to bias our
results for FX exposure hedging (but not for TL, CT, and AT hedging). Since few firms
6 The accounting practices of Swedish firms regarding exposures and hedging did not follow consistent rules
throughout the sample period. New rules and regulations based on IAS 39 were not in effect during the sample
period. Recommendations from the Swedish Accounting Standards Board (BFN R9) stipulated that firms should
report net revenues, investments, and number of employees for each geographical market, but allowed considerable
freedom in determining what the geographical market was. Of the two accounting methods allowed for hedging,
deferral accounting and mark-to-market, almost all firms used deferral accounting. Firms were required to disclose
derivatives positions in footnotes, and most firms reported their positions on a net basis, rather than per currency,
type of exposure hedged, or exposure partitioned over time. Notably, practices varied widely among firms.
7 Results are available from the authors by request.

N. Hagelin, B. Pramborg / J. of Multi. Fin. Manag. 16 (2006) 142159

147

Table 1
Data on foreign exchange exposure and frequency of hedging
Panel A: FX exposure (%)

FR
FC
FE
ANE

Panel B: Frequency of hedging (%)

All

Small

Medium

Large

38.7
29.4
17.1
16.9

32.0
23.8
8.3
21.9

35.6
26.4
16.0
14.7

56.5
44.9
31.5
17.4

FX
TL
CT
AT

All

Small

Medium

Large

54.2
23.1
42.6
31.8

32.2
9.5
25.5
17.6

54.5
22.5
38.9
30.5

86.2
43.1
76.6
56.0

Panel A presents foreign revenues (FR), foreign costs (FC), foreign equity (FE), and the absolute value of the
difference between FR and FC (ANE) for the pooled sample of firm-year observations. Panel B presents the
percentage of firms that hedged foreign exchange (FX) exposure, translation (TL) exposure, committed transactions
(CT), and anticipated transactions (AT) for the pooled sample of firm-year observations. The sample consists of
non-financial firms listed on the Stockholm Stock Exchange that responded to the questionnaires, and is divided
into groups based on firm size. A firm with a book value of total assets less than SEK 500 million is considered
small, between SEK 500 million and SEK 5000 million is considered medium, and greater than SEK 5000 million
is considered large.

use derivatives for this type of exposure (see Bodnar et al., 1996), and we expect that few
firms use foreign-denominated debt for this purpose only, we expect this potential problem
to be relatively small.
Table 1, Panel A, presents FR, FC, FE, and the absolute net exposure (ANE), defined as
the absolute value of the difference between FR and FC. We use FE as our proxy for translation exposure, which is natural since the exposure is defined in terms of accounting values.
In addition, loan covenants are typically expressed in terms of accounting ratios (respondents were specifically asked whether they had entered into loan covenants expressed in
accounting terms). The variable ANE, our proxy for transaction exposure, draws on a
suggestion of Marstons (2001) and is similar to the imbalance measure that Martin et al.
(1999) successfully use to explain cross-sectional differences in FX exposure between firms.
Table 1 presents statistics concerning the pooled sample of firm-year observations, in
which the sample firms are divided into groups according to size. A firm with book value
of total assets less than SEK 500 million is considered small, between SEK 500 million and
SEK 5000 million is considered medium, and greater than SEK 5000 million is considered
large.
From Panel A in Table 1 it can be seen that the average FR of the pooled sample is
38.7% while the average FC is 29.4%, suggesting that our sampled firms are on average
more involved in exporting than importing. From the table it is evident that large firms, on
average, are characterized by higher FR and FC than are small firms. Even though large
firms have higher levels of FR and FC than do small firms, there is a tendency for small firms
to have higher levels of ANE than do large firms. It is likely that large firms have greater
opportunity to employ operational hedges than do small firms. For the pooled sample of
small firms, the average ANE is 21.9%, showing that small firms have considerable FX
exposure. Panel A also shows that the average FE is 17.1% for the pooled sample, and that
FE increases with firm size.
Panel B in Table 1 shows, from the pooled sample of firm-year observations, the percentage of firms that hedged FX, TL, CT, and AT exposures. Like Panel A, Panel B presents the
results broken down by size group. From Panel B it can be seen that 54.2% of the pooled

148

N. Hagelin, B. Pramborg / J. of Multi. Fin. Manag. 16 (2006) 142159

Fig. 1. Comparison of variability in hedging policy. This figure reports the percentage of firms that began hedging,
quit hedging, or did not change their hedging practices with reference to foreign exchange (FX), translation (TL),
committed transaction (CT), and/or anticipated transaction (AT) exposures. The total number of observations is
319.

sample used currency derivatives or foreign-denominated debt to hedge FX exposure, 23.1%

hedged TL exposure, and 42.6% and 31.8% hedged CT and AT exposures, respectively.
From Table 1 it can be seen that larger firms are more likely to hedge than small firms are.
Given the evidence that small firms have considerable FX exposure, the higher propensity
for hedging among larger firms is likely to be a result of economies of scale in hedging (see,
e.g. Mian, 1996; Chow et al., 1997; Geczy et al., 1997; Martin and Mauer, 2004). We do
not report the figures for each sample year, but note that the findings in Panel A hold for
each year and size group.
In addition, we investigated whether firms changed their hedging policies concerning
FX, TL, CT, and AT exposure, finding, as is evident from Fig. 1, that most firms did not
change these policies. However, of the firms that did change their policy, there are more
observations of firms that began rather than quit hedging.

3. Research design and variable denitions

This section presents the research design and variables used to distinguish among the
possible explanations for using financial hedges for FX, TL, CT, and AT exposures. To do
this, we estimate logit regressions.8 For example, one of the logit specification we use to

8 The questionnaires yielded data concerning how much of their TL, CT, and AT exposures firms hedged with
currency derivatives. On average, firms hedged a substantial part of each exposure. This is consistent with the
survey evidence of Hakkarainen et al. (1998), and suggests that the decision whether or not to employ hedges is
of more interest to us than is the decision regarding how much exposure to hedge. Our results using degree of
exposure being hedged as the dependent variable are in line with those of Allayannis and Ofek (2001), and are
available from the authors on request.

N. Hagelin, B. Pramborg / J. of Multi. Fin. Manag. 16 (2006) 142159

149

investigate firms hedging for FX exposure can be written as:


PFX,i
ln
= + 1 (loan covenants)i + 2 (firm size)i + 3 (liquidity)i
1 PFX,i
+4 (leverage)i + 5 (market-to-book)i + 6 (ANE)i
+7 (FE)i + 8 (manufacturing)i + 9 (primary products)i + i
(1)
where PFX ,i is the probability that firm i hedges FX exposure, and the right-hand-side
explanatory variables are presented below. We use separate regressions for each particular
exposure investigated (FX exposure, TL exposure, CT exposure, and AT exposure) where
the binary variable is assigned a value of one if a firm uses financial instruments, i.e.
currency derivatives or foreign-denominated debt, to hedge the investigated exposure, and
zero otherwise.
The explanatory variables for these regressions include proxies for economies of scale,
expected costs of financial distress, underinvestment associated with costly external financing, and firms FX exposure.9 The variable definitions and hypothesized relationships are
as follows:
(a) Loan covenants. Butler (2004) and Eiteman et al. (2001) are among those arguing
that the existence of loan covenants can partly explain why firms engage in hedging
activities. The rationale is that loan covenants require that a firm maintain certain levels
of financial performance, and that violating a loan covenant can lead to a reduction
in borrowing capacity. Martin et al. (1998) argue that changes in net worth induced
by FX exposure may affect borrowing capacity. In these circumstances, hedging can
ensure that the firm retains its access to funds. Although a positive relationship could be
expected in the cases of both transaction and TL exposure hedging, we are particularly
interested in the results for the latter due to the scarcity of other explanations. This is of
special interest as this study, to the best of our knowledge, is the first attempt to analyze
empirically the relationship between hedging and loan covenants. We use a dummy
variable assigned a value of 1 if loan covenants exist, and 0 otherwise. (Note: Data are
only available for the years 2000 and 2001.)

9 In addition to the rationales for hedging we investigate, theorists provide another two: hedging to reduce tax
costs due tax schedule convexity, and hedging to maximize managers private utility. We chose not to examine
tax-based explanations because of the failure of earlier studies to document a relationship between use of currency
derivatives and tax convexity (see Mian, 1996; Geczy et al., 1997; Graham and Rogers, 2000; Allayannis and
Ofek, 2001). Graham and Rogers (2000) conclude that firms do not hedge in response to tax convexity because
the incentive is small compared to other hedging incentives. In a later study, however, Graham and Rogers (2002)
show that although hedging cannot be explained by tax convexity, hedging may reduce costs associated with
taxes since it allows for more debt in the capital structure. Similarly, we chose not to examine motives related
to managers private utility because of the relatively weak support for this class of explanations found in earlier
empirical studies of currency derivatives use (see Geczy et al., 1997; Graham and Rogers, 2000; Hagelin, 2003).
A notable exception is a recent study by Knopf et al. (2002); it shows that as the sensitivity of managers stock
option portfolios to stock return volatility increases, firms tend to hedge less.

150

N. Hagelin, B. Pramborg / J. of Multi. Fin. Manag. 16 (2006) 142159

(b) Firm size. Empirical research supports the perception that starting and managing a
hedging program is associated with significant fixed costs that may be justified when it is
possible to achieve economies of scale (see, e.g. Mian, 1996; Geczy et al., 1997; Martin
and Mauer, 2004).10 Because fixed costs arise from, for example, training employees
and developing hedging strategies economies of scale should be associated with all
types of hedging. Hence, we would expect to find a positive relationship between firm
size and all types of hedging. We use the log of the book value of total assets as a proxy
for size.
(c) Liquidity. Hedging can increase the value of the firm by lowering the expected costs
of financial distress (see Smith and Stulz, 1985). Nance et al. (1993) argue that the
probability of encountering financial distress can be reduced by maintaining more liquid
assets, thereby reducing the need for hedging. Thus, high liquidity can be considered
as a substitute for transaction (i.e. CT and AT) exposure hedging, suggesting a negative
relationship between liquidity and propensity to hedge. We use the ratio of current
assets to current liabilities as a proxy for liquidity.
(d) Leverage. Hedging can reduce the variance of firm value and thereby the expected cost
of financial distress (see Smith and Stulz, 1985). For this reason, we would expect
leverage to be positively related to transaction (i.e. CT and AT) exposure hedging. We
use the book value of debt divided by the book value of equity as a proxy for leverage.
(e) Market-to-book ratio. Bessembinder (1991) demonstrates that hedging reduces the
incentive to underinvest, as hedging shifts individual future states from default to
non-default outcomes. Because firms with more valuable growth opportunities are
more likely to be affected by underinvestment, these firms may be more likely to
hedge. We therefore predict that there should be a positive correlation between transaction (i.e. CT and AT) exposure hedging and growth opportunities. As a measure of
future growth opportunities, we use the ratio of the market to the book value of total
assets.
(f) Foreign exchange exposure. Given the fixed costs associated with a derivatives hedging
program, the decision to initiate one is not determined solely by whether FX exposure
exists or not, but also by the extent of exposure. As a proxy for transaction exposure,
we use the absolute value of the difference between FR and FC (i.e. ANE). As a proxy
for translation exposure, we use the percentage of total equity (measured in terms of
book value) denominated in foreign currency (i.e. FE).
(g) Industry dummies. We divide the sample into various industry sectors because typical levels of exposure may vary between industry sectors. We use the same industry sector categories as Bodnar et al. (1996): primary products, manufacturing, and
services.
Data for creating the explanatory variables ANE, FE, and loan covenants are taken from
the questionnaires. The accounting data required to calculate the explanatory variables
firm size, liquidity, leverage, and market-to-book ratio are collected from the Nordbanken
10 While most of the previous evidence on economies of scale concern transaction exposure hedging or FX
exposure hedging in general studies that examine scale economies and TL exposure do exist. For example, Hagelin
(2003) shows that larger firms are more likely to hedge TL exposure than small firms are.

N. Hagelin, B. Pramborg / J. of Multi. Fin. Manag. 16 (2006) 142159

151

Aktieguide Sommar 2001 stock market guide and from annual reports. We use the industry
classification system devised by Statistics Sweden and the SNI 92 standard to classify firms
into our three distinct industry groups.
Table 2 presents yearly median values for the firm size (in million SEK), liquidity,
leverage, and market-to-book ratio variables, and mean values for the loan covenant, industry
dummy, ANE, and FE variables for the pooled sample of firm-year observations. It is evident
from the table that there are no apparent trends in the values that fluctuate from year to year.
Table 2 also shows that about 60% of the firms are in the service industry category, while
under 10% are categorized as primary producers. Interestingly, up to one quarter of the firms
are subject to loan covenants; such a widespread incidence of covenants is in accordance
with the findings of Smith and Warner (1979).
Table 3 displays pair-wise correlations between the variables representing firm characteristics. In general, the correlations are not high in absolute terms, and only a few correlations
are higher than 0.20. Excluding the industry dummies, the highest correlation in absolute
terms is between FE and firm size (0.38), which is expected from the results displayed
in Table 1. In the next section, we use cross-sectional regressions in order to investigate
relationships between different types of hedging and the variables in Table 3. The relatively
low correlations suggest that multicollinearity should not materially affect our results.

4. Cross-sectional results
Table 4 presents the logit regressions we estimate to distinguish between the potential
reasons for financial hedge use; and contains the coefficient estimates, P-values, classification accuracy, and McFadden R2 values. In Panel A, the dependent variable is set to one for
firms using financial hedges to hedge FX exposure and to zero for those that do not. Panels
B, C, and D present results broken down by type of FX exposure hedged. More specifically,
Panel B uses TL exposure hedging as the dependent variable, Panel C uses CT exposure
hedging as the dependent variable, and Panel D uses AT exposure hedging as the dependent
variable in the logit regression estimation.
The data include observations of an unbalanced panel of 89115 firms per year for
four years. In addition to the results for each year, we also present regression results for
the pooled sample. Pooling firm-year observations treats each observation as independent,
which tends to underestimate standard errors and overstate reported P-values to the extent
that firm values are correlated from year to year. Therefore, as a robustness test, we also
performed panel data regressions using population average models; the results are similar
to those for the pooled sample in Table 4 and are therefore not presented.11
The logit regression results in Table 4 show that hedging is significantly associated with
the existence of loan covenants, at least for TL exposure hedging. We interpret this as evidence that firms hedge TL exposure in order to avoid violating loan covenants, thereby
maintaining access to funds, as Butler (2004) among others suggests. This finding is important given that the theoretical models of Smith and Stulz (1985), Bessembinder (1991), and
Froot et al. (1993) do not attempt to explain TL exposure hedging.
11

The results are available from the authors by request.

152

1998
1999
2000
2001
Pooled

Firm size

Liquidity

Leverage

Market-to-book

Manufacturing

Services

Primary products

ANE (%)

FE (%)

Loan covenants

885
1245
1202
853
1026

1.72
1.78
2.12
1.91
1.89

1.24
1.24
0.99
1.11
1.14

1.25
1.65
1.50
1.24
1.33

0.32
0.32
0.32
0.32
0.33

0.63
0.63
0.63
0.63
0.62

0.05
0.05
0.05
0.05
0.05

13.9
14.1
19.4
20.1
16.9

17.7
18.5
16.1
16.2
17.1

0.24
0.28

Summary statistics regarding explanatory variables for each year and for the pooled sample of firm-year observations. The sample consists of non-financial firms listed on
the Stockholm Stock Exchange that responded to the questionnaires. All variables are as defined in Section 3. The table reports the median values for firm size, liquidity,
and market-to-book ratio. For the variables manufacturing, services, primary products, foreign equity (FE), and the absolute value of the difference between FR and
FC (ANE), and loan covenants mean values (i.e. the proportion of firms) are reported.

N. Hagelin, B. Pramborg / J. of Multi. Fin. Manag. 16 (2006) 142159

Table 2
Summary statistics regarding explanatory variables

19982001
Liquidity
Leverage
Market-to-book
ANE
FE
Manufacturing
Primary products
Services
20002001
Loan covenants

Firm size

Liquidity

Leverage

Market-to-book

ANE

FE

Manufacturing

Primary products

0.11
0.04
0.16
0.04
0.38
0.21
0.02
0.22

0.20
0.15
0.16
0.10
0.13
0.04
0.15

0.13
0.16
0.00
0.24
0.07
0.20

0.13
0.08
0.18
0.10
0.22

0.02
0.09
0.24
0.02

0.13
0.03
0.14

0.18
0.89

0.30

0.26

0.18

0.19

0.18

0.17

0.26

0.17

0.07

Services

0.20

The table displays Pearson correlations. The variables are: rm size, the book value of total assets; liquidity, the current ratio; leverage, the debt-to-equity ratio; marketto-book, the ratio of the market value and book value of equity; industry dummies (manufacturing, primary products, and services); foreign equity (FE), the proportion of
equity that is denominated in foreign currency; absolute net exposure (ANE), the absolute value of the difference between the proportion of foreign revenues (FR) and
foreign costs (FC); industry dummies (Manufacturing, Primary products, Services); and loan covenants, a dummy that is set to one if a firm has loan covenants attached
to its loans, and zero otherwise.

N. Hagelin, B. Pramborg / J. of Multi. Fin. Manag. 16 (2006) 142159

Table 3
Pearson correlations

153

154

N. Hagelin, B. Pramborg / J. of Multi. Fin. Manag. 16 (2006) 142159

Table 4
Logit regression estimates of the likelihood of using financial hedges
Panel A: Hedging FX exposure
1998
Loan covenants
Firm size
Liquidity
Leverage
Market-to-book
ANE
FE
Manufacturing
Primary products
Log likelihood
McFadden R2
% Correct (gain)
Observations

1999

2000

2001

Pooled

0.464 (0.533)
0.688 (0.351)

0.772 (0.001)
0.889 (0.000)
0.658 (0.000)
0.368 (0.098)
0.688 (0.000)
0.182 (0.171)
0.075 (0.465) 0.143 (0.036) 0.282 (0.157)
0.141 (0.000)
0.304 (0.203) 0.096 (0.439)
0.255 (0.382)
0.108 (0.507)
0.000 (0.994)
0.671 (0.000) 0.142 (0.195) 0.046 (0.565)
0.728 (0.035)
0.101 (0.064)
0.029 (0.096)
0.017 (0.138)
0.034 (0.003)
0.031 (0.036)
0.029 (0.000)
0.088 (0.000)
0.066 (0.003)
0.029 (0.155)
0.026 (0.230)
0.044 (0.000)
1.107 (0.182)
0.908 (0.124)
0.465 (0.499)
1.370 (0.049)
0.755 (0.008)
2.615 (0.098)
3.234 (0.046)
41.196 (0.000)
36.119 (0.000)
2.889 (0.004)
36.785
45.904
48.671
36.281
186.087
50.8%
31.3%
38.4%
38.7%
35.3%
86.1 (71.2)
79.8 (59.3)
78.9 (54.0)
79.8 (47.1)
80.3 (57.1)
108
109
109
89
417
Panel B: Hedging TL exposure
1998

Loan covenants
Firm size
Liquidity
Leverage
Market-to-book
FE
Manufacturing
Primary products
Log likelihood
McFadden R2
% Correct (gain)
Observations

1999

2000

2001

Pooled

1.451 (0.022)
1.513 (0.025)

0.320 (0.080)
0.039 (0.848)
0.437 (0.028)
0.260 (0.121)
0.339 (0.000)
0.499 (0.051) 0.232 (0.407) 0.197 (0.069) 0.690 (0.067)
0.221 (0.034)
0.330 (0.213) 0.320 (0.172)
0.129 (0.696) 0.037 (0.765)
0.062 (0.412)
1.491 (0.060) 0.411 (0.355) 0.501 (0.023)
0.212 (0.559)
0.387 (0.036)
0.024 (0.027)
0.030 (0.010)
0.030 (0.011)
0.022 (0.096)
0.020 (0.000)
0.184 (0.812) 0.460 (0.521)
0.981 (0.184)
1.574 (0.035)
0.373 (0.237)
0.247 (0.772)
0.243 (0.781)
3.049 (0.000)
4.045 (0.001)
1.451 (0.000)
39.780
42.278
31.931
33.830
170.443
28.9%
20.9%
48.0%
36.4%
24.4%
79.6 (4.4)
83.5(14.3)
90.0(59.3)
81.1(32.0)
78.5 (6.25)
108
109
110
90
419
Panel C: Hedging CT exposure
1998

Loan covenants
Firm size
Liquidity
Leverage
Market-to-book
ANE
Manufacturing
Primary products
Log likelihood
McFadden R2
% Correct (gain)
Observations

1999

2000

2001

Pooled

0.924 (0.098)
0.331 (0.537)

0.810 (0.000)
0.722 (0.000)
0.397 (0.009)
0.319 (0.025)
0.513 (0.000)
0.192 (0.165)
0.107 (0.234) 0.244 (0.030) 0.218 (0.149)
0.161 (0.000)
0.374 (0.065) 0.037 (0.619)
0.074 (0.721)
0.077 (0.441)
0.020 (0.716)
0.435 (0.039) 0.142 (0.148)
0.013 (0.861)
0.146 (0.542)
0.088 (0.065)
0.015 (0.204)
0.020 (0.062)
0.024 (0.020)
0.025 (0.017)
0.020 (0.000)
1.436 (0.038)
1.397 (0.009)
0.455 (0.424)
0.395 (0.470)
0.784 (0.002)
2.802 (0.037)
2.890 (0.037)
0.233 (0.886) 0.220 (0.895)
1.331 (0.082)
50.640
57.772
59.915
54.514
240.397
35.2%
26.4%
22.6%
14.6%
19.9%
79.6 (56.6)
80.1 (55.1)
76.5 (41.3)
68.8 (29.3)
72.2 (36.1)
113
115
115
93
438

N. Hagelin, B. Pramborg / J. of Multi. Fin. Manag. 16 (2006) 142159

155

Table 4 (Continued )
Panel D: Hedging AT exposure
1998
Loan covenants
Firm size
Liquidity
Leverage
Market-to-book
ANE
Manufacturing
Primary products
Log likelihood
McFadden R2
% Correct (gain)
Observations

1999

2000

2001

Pooled

1.075 (0.039)
0.539 (0.305)

0.527 (0.001)
0.411 (0.022)
0.387 (0.002)
0.251 (0.076)
0.427 (0.000)
0.086 (0.427)
0.009 (0.911) 0.035 (0.489) 0.128 (0.204)
0.054 (0.131)
0.177 (0.390) 0.097 (0.611) 0.053 (0.845)
0.125 (0.241)
0.003 (0.950)
0.247 (0.127) 0.218 (0.044) 0.090 (0.242)
0.330 (0.136)
0.135 (0.003)
0.013 (0.258)
0.020 (0.010)
0.023 (0.010)
0.019 (0.077)
0.019 (0.000)
0.609 (0.319)
0.443 (0.415)
0.402 (0.527)
0.387 (0.512)
0.345 (0.166)
2.777 (0.026)
1.030 (0.431)
0.309 (0.747)
0.186 (0.864)
1.034 (0.103)
57.982
62.484
62.282
54.082
248.809
20.6%
16.7%
17.5%
14.9%
14.1%
74.1 (27.5)
73.7 (28.6)
74.8 (30.1)
66.7 (22.5)
69.7 (20.5)
112
114
115
93
436

Logit regression estimates of the relationship between the likelihood of a firm using financial hedges and proxies
for incentives to hedge and proxies for foreign exchange exposure. The sample consists of non-financial firms
listed on the Stockholm Stock Exchange that responded to the questionnaires. In Panel A the dependent variable
is set to one if the firm uses financial hedges to hedge foreign exchange (FX) exposure. Panel B uses translation
(TL) exposure hedging with financial hedges as the dependent variable. Panel C uses committed transaction
(CT) exposure hedging with financial hedges as the dependent variable. Panel D uses anticipated transaction (AT)
exposure hedging with financial hedges as the dependent variable. The explanatory variables are: firm size, liquidity,
leverage, market-to-book ratio, industry dummies (manufacturing, primary products, and services), foreign equity
(FE), and the absolute value of the difference between the proportion of FR and FC (ANE). P-values are presented
in parentheses and based on a two-sided test. Values less than 0.10 are shown in boldface type. The significance
tests are based on QML (Huber/White) standard errors and covariance. The last row of each panel shows the
percentage of observations that are correctly predicted by the model (using a cut-off rate of 0.5) and the percentage
gain, in terms of correct classification, of including the explanatory variables to an intercept.

All regressions presented in Table 4, except for two using TL exposure as the dependent
variable, show that greater firm size is significantly associated with a greater probability
of financial hedging at least at the 10% level. This strongly supports the hypothesis that a
firms decision as to whether or not to use financial hedges is influenced by the existence
of economies of scale, consistent with the findings of earlier studies of currency derivatives
use (see Mian, 1996; Geczy et al., 1997; Graham and Rogers, 2000; Allayannis and Ofek,
2001; Hagelin, 2003).
From the table it is evident that the results for the liquidity variable vary across
time period and exposure category investigated. Significance is indicated for the pooled
regression models using FX, TL, and CT exposures as dependent variables and for some of
the individual years for these exposure categories. The result for CT exposure is consistent
with our expectations, and we interpret this as evidence that firms hedge CT exposure in
response to their liquidity and to reduce the likelihood of financial distress. The result for TL
exposure, however, is unexpected because liquidity issues emphasize cash-flow constraints,
while TL gains (losses) tend to be unrealized and have few direct cash-flow consequences.12
12

In general, since TL hedging mainly affects accounting numbers (net worth), there should be no relation
between liquidity and TL exposure hedging. However, loan covenants may in some cases set restrictions on
book liquidity. Thus, firms with loan covenants may have higher liquidity, in which case these variables could be
positively correlated. The results in Table 3 does not support this conjecture.

156

N. Hagelin, B. Pramborg / J. of Multi. Fin. Manag. 16 (2006) 142159

This could be because over 95% of the firms that hedged TL exposure also hedged transaction exposure. To investigate this possibility we undertook two robustness tests. First, we
re-ran the regression presented in Table 4 for the pooled sample of firms, treating observations of firms that hedged TL exposure only as TL hedgers and all other observations as
non-hedgers. This treatment produces only 12 observations (of 419) classified as representing TL hedgers, and this small number of data points calls for a cautious interpretation of
the results.13 The results of the regression (not presented) indicate that liquidity is unrelated
to TL hedging. As a second test, we re-ran the regressions on TL exposure, allowing only
firms that did not hedge any other type of FX exposure to remain in the sample as nonhedgers. With this specification the significance of liquidity is indicated less frequently than
is reported in Table 4. These robustness tests suggest that the relationship between liquidity
and TL exposure indicated in Table 4 is uncertain.
As for leverage, only one coefficient is found to be marginally significant, and contrary to
the prediction, this coefficient is found to be negative. The finding of no significant positive
association between leverage and use of financial hedges is consistent with the findings of
Geczy et al. (1997), Allayannis and Ofek (2001), and Hagelin (2003), but runs counter to
those of Graham and Rogers (2000).
In contrast to our prediction, for the pooled regressions the coefficients of the marketto-book ratio are negative and significant at the 10% level. We can gain insight into
this result by studying the regressions for separate years. Specifically, note that the sign
of the coefficient changed during the sample period, from negative over the 19982000
period to positive in 2001. A possible explanation for the negative coefficients over the
19982000 period is the high valuation of firms belonging to the so-called new economy until early 2000. Because many firms comprising this group typically had small cash
flows, they had no reason to employ hedges. In general, the risks these firms faced were
not related to changes in exchange rates but rather to project risk and access to external
financing. Our failure to document a positive relationship between the market-to-book ratio
and hedging is in line with that of earlier studies of currency derivatives use (see Mian,
1996; Geczy et al., 1997; Graham and Rogers, 2000; Allayannis and Ofek, 2001; Hagelin,
2003).
As predicted, firms with greater absolute differences between revenues and costs denominated in foreign currency (ANE) are more likely to use financial hedges to hedge FX, CT,
and AT exposures. The coefficients of ANE are positive in all regressions and significance
is indicated for each of the three pooled sample regressions and for 9 of the 12 regressions for individual years. The results for foreign equity (FE) are also as predicted, i.e. the
likelihood of TL exposure hedging increases with FE. This finding runs counter to that of
Hagelin (2003), who argues that his failure to document the predicted relationship may be
because many firms classified as non-hedgers do in fact hedge TL exposure using foreigndenominated debt. Our results may differ because we classify both currency derivatives
hedging and foreign-denominated debt as hedging activities.
Finally, the coefficients for the industry dummies suggest that the likelihood of hedging
is related to industrial sector. Briefly stated, firms in the primary products and manufacturing
categories are more likely to hedge FX, TL, and CT exposures than are firms in the service
13

The sample size did not allow for the inclusion of any industry dummies in this specification.

N. Hagelin, B. Pramborg / J. of Multi. Fin. Manag. 16 (2006) 142159

157

category. This is consistent with the survey evidence of Alkeback and Hagelin (1999) for
Sweden and of Bodnar et al. (1996) for the US.
The percentage of correctly classified observations for the regressions in Table 4 ranges
from 66.7% to 90.0%, which is similar to Geczy et al.s (1997) results. The percentage gain,
in terms of number of correctly classified observations, from including the explanatory
variables (in addition to an intercept) ranges from 4.4% to 59.3%. As evident from Panel B,
the worst performance in terms of gain is recorded for the models explaining TL hedging
for which the variable Loan Covenants is not included. In untabulated regressions for the
years 2000 and 2001, we re-ran the regressions excluding Loan Covenants. For the year
2000 (2001), the percentage gain fell from 59.3% to 22.2% (32.0% to 12.0%). This supports
the importance of loan covenants as an explanation for TL exposure hedging.
To examine the robustness of the results presented in Table 4, we use logit regressions to
compare those firms that began to hedge TL, CT, and AT exposures with those that did not.
The potential benefit of analyzing changes in hedging policies, according to Skinner (1996)
and Guay (1999), derives from the difficulty in controlling for cross-sectional differences
in core business risk. We did not investigate firms that began to hedge FX exposure or firms
that quit hedging because these groups contain too few observations to allow for meaningful
analysis (see Fig. 1). The use of explanatory variables differs in that we use both the levels
of the explanatory variables as well as their change, in percent, from the previous year.
However, we are unable to include the dummy indicating the existence of loan covenants
(because of an insufficient number of observations). Because this is the variable of primary
focus for this study, we do not present the results for the change regressions in any detail,
but we note that the results corroborate those in Table 4.14

5. Conclusions
In this study, we examine Swedish firms use of financial hedges to hedge foreign
exchange exposure for the 19982001 period. We use survey data that lets us differentiate
between hedging aimed at translation exposure and hedging aimed at transaction exposure.
The survey responses show that use of financial hedges is widespread: over 50% of
the responding firms employ financial hedges, and transaction exposure is more frequently
hedged than is translation exposure. About 20% of the sampled firms hedge their translation
exposure, an interesting finding given that the finance literature generally recommends not
hedging translation exposure. Butler (2004), among others, however, argues that translation
exposure hedging may be rational in the presence of loan covenants that require a firms
financial performance to remain at certain levels, and given that violating a loan covenant can
lead to reduced borrowing capacity. We document a positive relationship between existence
of loan covenants and translation exposure hedging; this relationship supports the conjecture
that firms hedge translation exposure in order to secure their access to funds.
As theoretically predicted, liquidity is negatively related to transaction exposure hedging,
supporting the assumption that firms hedge in response to expected financial distress costs.

14

The results are available from the authors by request.

158

N. Hagelin, B. Pramborg / J. of Multi. Fin. Manag. 16 (2006) 142159

We also find that the likelihood of hedging foreign exchange exposure increases with firm
size and exposure, suggesting that the existence of economies of scale influence a firms
hedging decisions. Thus, the evidence suggests that Swedish firms hedge in way that is
consistent with shareholder value maximization.
Acknowledgement
The content of this article expresses the opinions of the authors alone and may not be
representative of their respective institutions. The authors would like to thank an anonymous
referee, as well as Harvey Arbelaez, Animesh Ghoshal, and Art Stonehill. Comments from
seminar participants at the 46th annual meeting of the Academy of International Business
in Stockholm, 2004, are acknowledged.
References
Aggarwal, R., 1991. Management of accounting exposure to currency changes: role and evidence of agency costs.
Managerial Finance 17, 1022.
Alkeback, P., Hagelin, N., 1999. Derivative usage by nonfinancial firms in Sweden with an international comparison. Journal of International Financial Management and Accounting 10, 105120.
Alkeback, P., Hagelin, N., Pramborg, B., 2006. Derivatives usage by non-financial firms in Sweden 1996 and 2003:
what has changed? Managerial Finance, in press.
Allayannis, G., Ofek, E., 2001. Exchange rate exposure, hedging, and the use of currency derivatives. Journal of
International Money and Finance 20, 273296.
Allayannis, G., Ihrig, J., Weston, J.P., 2003. Exchange-rate hedging: financial vs. operational strategies. American
Economic Review Papers and Proceedings 91, 391395.
Bartram, S., Brown, G., Fehle, F., 2004. International evidence on financial derivatives use. Working Paper,
Lancaster University, University of North Carolina, and University of South Carolina.
Berkman, H., Bradbury, M.E., 1996. Empirical evidence on the corporate use of derivatives. Financial Management
25, 513.
Berkman, H., Bradbury, M.E., Magan, S., 1997. An international comparison of derivatives use. Financial Management 26, 6973.
Bessembinder, H., 1991. Forward contracts and firm value: investment incentive and contracting effects. Journal
of Financial and Quantitative Analysis 26, 519532.
Bodnar, G.M., Gebhardt, G., 1999. Derivatives usage in risk management by US and German non-financial firms:
a comparative study. Journal of International Financial Management and Accounting 10, 153187.
Bodnar, G.M., Hayt, G.S., Marston, R.C., 1996. 1995 Wharton survey of derivatives usage by US non-financial
firms. Financial Management 25, 113133.
Bodnar, G.M., Hayt, G.S., Marston, R.C., 1998. 1998 Survey of financial risk management by US non-financial
firms. Financial Management 27, 7091.
Bodnar, G.M., de Jong, A., Macrae, V., 2003. The impact of institutional differences on derivatives usage: a
comparative study of US and Dutch firms. European Financial Management 9, 271297.
Brown, G.W., 2001. Managing foreign exchange risk with derivatives. Journal of Financial Economics 60, 401448.
Butler, K.C., 2004. Multinational Finance, third ed. South-Western, Mason, Ohio.
Chow, E.H., Lee, W.Y., Solt, M.E., 1997. The economic exposure of US multinational firms. Journal of Financial
Research 20, 191210.
Core, J.E., Schrand, C.M., 1999. The effect of accounting-based debt covenants on equity valuation. Journal of
Accounting and Economics 27, 134.
De Ceuster, M.J.K., Durinck, E., Laveren, E., Lodewyckx, J., 2000. A survey into the use of derivatives by large
non-financial firms operating in Belgium. European Financial Management 6, 301318.

N. Hagelin, B. Pramborg / J. of Multi. Fin. Manag. 16 (2006) 142159

159

Dolde, W., 1995. Hedging, leverage, and primitive risk. Journal of Financial Engineering 4, 187216.
Eiteman, D.K., Stonehill, A.I., Moffett, M.H., 2001. Multinational Business Finance, ninth ed. Addison-Wesley,
Boston, MA.
Froot, K.A., Scharfstein, D.S., Stein, J.C., 1993. Risk management: coordinating corporate investment and financing policies. Journal of Finance 48, 16291658.
Gay, G.D., Nam, J., 1998. The underinvestment problem and corporate derivatives use. Financial Management
27, 5369.
Geczy, C., Minton, B.A., Schrand, C., 1997. Why firms use currency derivatives. Journal of Finance 52, 13231354.
Graham, J.R., Rogers, D.A., 2000. Does corporate hedging increase firm value? An empirical analysis. Working
Paper, Fuqua School of Business, Duke University, Durham, NC.
Graham, J.R., Rogers, D.A., 2002. Do firms hedge in response to tax incentives? Journal of Finance 57, 815839.
Guay, W.R., 1999. The impact of derivatives on firm risk: an empirical examination of new derivative users. Journal
of Accounting and Economics 26, 319351.
Hagelin, N., 2003. Why firms hedge with currency derivatives: an examination of transaction and translation
exposure. Applied Financial Economics 13, 5569.
Hagelin, N., Pramborg, B., 2004. Hedging foreign exchange exposure: risk reduction from transaction and translation hedging. Journal of International Financial Management and Accounting 15, 120.
Hakkarainen, A., Joseph, N., Kasanen, E., Puttonen, V., 1998. The foreign exchange exposure management practices of Finnish industrial firms. Journal of International Financial Management and Accounting 9, 3457.
Haushalter, G.D., 2000. Financing policy, basis risk, and corporate hedging: evidence from oil and gas producers.
Journal of Finance 55, 107152.
Knopf, J.D., Nam, J., Thornton, J.H., 2002. The volatility and price sensitivities of managerial stock option
portfolios and corporate hedging. Journal of Finance 57, 801813.
Lel, U., 2004. Currency risk management, corporate governance, and financial market development. Working
Paper, Indiana University, Bloomington, IN.
Leland, H.E., 1994. Corporate debt value, bond covenants, and optimal capital structure. Journal of Finance 49,
12131252.
Mallin, C., Ow-Yong, K., Reynolds, M., 2001. Derivatives usage in UK non-financial listed companies. European
Journal of Finance, 6391.
Marston, R.C., 2001. The effects of industry structure on economic exposure. Journal of International Money and
Finance 20, 149164.
Martin, A.D., Madura, J., Akhigbe, A., 1998. A note on accounting exposure and the value of multinational
corporations. Global Finance Journal 9, 267277.
Martin, A.D., Madura, J., Akhigbe, A., 1999. Economic exchange rate exposure of US-based MNCs operating in
Europe. The Financial Review 34, 2136.
Martin, A.D., Mauer, L.J., 2004. Scale economies in hedging foreign exchange cash flow exposures. Global
Finance Journal 15, 1727.
Mian, S.L., 1996. Evidence on corporate hedging policy. Journal of Financial and Quantitative Analysis 31,
419439.
Nance, D.R., Smith, C.W., Smithson, C.W., 1993. On the determinants of corporate hedging. Journal of Finance
48, 267284.
Pramborg, B., 2005. Foreign exchange risk management by Swedish and Korean non-financial firms: a comparative
survey. Pacific Basin Finance Journal 13, 343366.
Roberts, G.S., Viscione, J.A., 1984. The impact of seniority and security covenants on bond yields: a note. Journal
of Finance 39, 15971602.
Skinner, D.J., 1996. Are disclosures about bank derivatives and employee stock options value relevant? Journal
of Accounting and Economics 22, 393405.
Smith, C.W., Stulz, R.M., 1985. The determinants of firms hedging policies. Journal of Financial and Quantitative
Analysis 20, 391405.
Smith, C.W., Warner, J., 1979. On financial contracting: an analysis of bond covenants. Journal of Financial
Economics 7, 117161.
Tufano, P., 1996. Who manages risk? An empirical examination of risk management practices in the gold mining
industry. Journal of Finance 51, 10971137.