Journal of Business Finance & Accounting, 28(1) & (2), January/March 2001, 0306-686X

Determinants of Capital Structure

and Adjustment to Long Run
Target: Evidence from UK
Company Panel Data
Aydin Ozkan*
1. INTRODUCTION

The major objective of this paper is to provide more insight into

the empirical determinants of target capital structure of firms
and the adjustment process toward this target. This study extends
the empirical research on the topic of capital structure by
focusing on the dynamics of capital structure decisions and the
nature of adjustment process. It also estimates the dynamic
capital structure model using a much stronger estimation
technique. In doing this, our empirical model captures at least
two important features of corporate borrowing behaviour. First,
firms have a long-run optimal target debt ratio which is assumed
to be a function of several firm-specific characteristics which vary
over time, over firms, or over both time and firms. Second, an
adjustment process takes place, which involves a lag in adjusting
to changes in the optimal target debt ratio. These features are the
results of relaxing perfect market assumptions of Modigliani and
Miller's (1958) analysis. The existence of market imperfections
such as corporate tax, bankruptcy and agency-related costs
implies an optimal capital structure (see, e.g., Kraus and
Litzenberger, 1973; Scott, 1976; and Kim, 1978). Also, market
* The author is from the Department of Economics and Related Studies, University of
York. (Paper received June 1999, revised and accepted February 2000)
Address for correspondence: Aydin Ozkan, Department of Economics and Related
Studies, University of York, Heslington, York YO10 5DD, UK.
e.mail: ao5@york.ac.uk
Blackwell Publishers Ltd. 2001, 108 Cowley Road, Oxford OX4 1JF, UK
and 350 Main Street, Malden, MA 02148, USA.

175

176

OZKAN

imperfections such as flotation costs (Marsh, 1982), adjustment

costs and constraints (Jalilvand and Harris, 1984; and Myers,
1984) may prevent firms from adjusting completely to their target
capital structures and immediately offsetting the effects of events
which take them away from their target ratios.
In order to investigate these issues we utilise a partial
adjustment model where the firm's financial behaviour is
characterised as partial adjustment to a long-term target debt
ratio. In this framework, we are able to analyse both the potential
determinants of target debt ratios and the nature of adjustment
to these targets. To estimate our dynamic model consistently
from a short and unbalanced panel the Generalised Method of
Moments (GMM) estimation procedure is used. The use of panel
data and GMM provide a more satisfactory basis for our purpose.
It is argued that , by providing a large number of data points and
blending characteristics of both cross-sectional and time seriesdata, panel data improves the efficiency of econometric estimates
(Hsiao, 1985). Moreover, compared to cross-sectional data, panel
data is more flexible in the choice of variables used as
instruments to control for endogeneity. The endogeneity
problem arises because observable as well as unobservable shocks
affecting corporate capital structure decisions are also likely to
affect some of other firm-specific characteristics such as the
market value of equity. It is also possible that observed relations
between leverage and firm-specific characteristics reflect the
effects of leverage on the latter rather than vice versa. The use of
GMM and panel data mitigates this problem by allowing us to
include firm-specific effects (which account for the crosssectional components of these unobservable shocks) and time
dummies (which control for the macroeconomic shocks
common to all firms). Hence it enables us to effectively choose
more efficient instruments to control for endogeneity.
This paper, exploiting the panel aspects of the data by
incorporating dynamic effects as well as controlling for
unobservable firm-specific effects and firm-invariant time-specific
effects, provides a more appropriate basis to analyse corporate
financial decisions. We control for firm-specific effects by
estimating the dynamic capital structure model in first
differences, rather than in levels. Also, time-specific effects are
controlled by including year dummies in the estimated models.
Blackwell Publishers Ltd 2001

DETERMINANTS OF CAPITAL STRUCTURE

177

The findings of this paper suggest that firms have target leverage
ratios and they adjust to the target ratio relatively fast. This
implies that the costs of being away from their target ratios and
the costs of adjustment are both important for firms. Another
major conclusion is that current liquidity and profitability of
firms exert a negative impact on their borrowing decisions
whereas there is a positive relation between past profitability and
debt ratio. Other firm-specific variables which appear to
influence leverage decision are non-debt tax shields and growth
opportunities of firms. There is only limited evidence that firm
size exerts an impact on capital structure decisions.
The rest of the paper is organised as follows. In the next
section we give a summary of the findings of the previous
empirical studies. Section 3 discusses the potential determinants
of target leverage ratios used in the study. In Section 4 the
empirical model is presented and a brief description of our data
is given. In Section 5 we present the estimation results and
discuss the robustness of results. The final section provides
concluding remarks.

2. EXISTING EMPIRICAL STUDIES

Much of the previous empirical work on the determinants of

borrowing decisions of firms has tended to concentrate on the
factors predicted by the static trade-off theory of capital structure,
which is based on a trade-off between the tax advantages of debt
financing and the costs of financial distress. As discussed in Myers
(1984), the static trade-off theory implies that the actual debt
ratio reverts towards a target or optimum, and it predicts a cross
sectional relation between debt ratios and asset risk, size,
profitability, tax status and asset type. The results of the empirical
literature confirm some of these predictions in that firms in the
same industry, facing similar conditions, risk characteristics and
tax status, have similar leverage ratios. Moreover, on balance the
evidence from these studies lends support to the negative impact
of business risk on corporate borrowing decisions. However,
there are conflicting conclusions on the impact of other firmspecific variables. For example, Bowen et al. (1982) and Kim and
Sorensen (1986) provide evidence on the negative relationship
Blackwell Publishers Ltd 2001

178

OZKAN

between non-debt tax shields and leverage. Conversely, Bradley et

al. (1984), Titman and Wessels (1988) and Homaifar et al. (1994)
fail to provide such a support. There are also conflicting results
on the relationship between size and leverage. Ferri and Jones
(1979), Kim and Sorensen (1986), and Chung (1993) show that
there is no systematic association between firm size and capital
structure. On the other hand, Homaifar et al. (1994) and Titman
and Wessels (1988) report results which are consistent with the
notion that larger firms have higher debt ratios. There is also
strong empirical evidence for the view that there is a negative
relation between profitability and debt ratios (pecking order
model of capital structure). For example, the findings of Kester
(1986), Titman and Wessels (1988) and Rajan and Zingales
(1995) lend strong support for the negative relationship.
However, Long and Malitz (1986) do not find such a relation
between leverage and profitability.
There are other studies in the literature which focus on the
determinants of adjustment to financial targets and provide more
direct evidence that firms adjust toward a target debt ratio. These
studies also shed some light on the likely determinants of speeds
of adjustment toward target debt ratios. For instance, Taggart
(1977) provides evidence that the speeds of adjustment to the
long-term capital targets are relatively slow and liquid assets and
short-term debt play an important role in the adjustment process.
Marsh (1982), using a logit model, analyses the choice of
financing instrument of companies and argues that this choice
depends on the difference between the company's current and
target debt ratios. His results suggest that companies try to
maintain their long-term target debt levels, although they deviate
from these targets in the short run in response to capital market
conditions. The paper also provides evidence that long-term
target debt levels are influenced by operating risk, company size
and asset composition. In a more recent work, Jalilvand and
Harris (1984) look at the determinants of speeds of adjustment
to long term financial targets where the speed of adjustment is
allowed to vary across companies and over time. Their results
suggest that firm size, interest rates and stock price levels affect
speeds of adjustment. Auerbach (1985), by allowing targets to
change by companies and over time, also estimates a targetadjustment model and provides support for target-adjustment
Blackwell Publishers Ltd 2001

DETERMINANTS OF CAPITAL STRUCTURE

179

behaviour of firms. Finally, in a related work, Shyam-Sunder and

Myers (1999), test a benchmark target-adjustment model against
a pecking order model and report that the target adjustment
model appears to be superior.
There is not a great deal of empirical work done on the capital
structure decisions of the UK firms. Bennett and Donnelly
(1993), Lasfer (1995), and Walsh and Ryan (1997) are notable
exceptions. Bennett and Donnelly (1993) provide an
examination of cross sectional determinants of leverage decisions
among non-financial UK firms. They find that non-debt tax
shields, asset structure, size and profitability exert an important
impact on the capital structure choice of firms. Lasfer (1995)
examines the impact of the corporation tax and agency costs on
firms' borrowing decisions by exploiting both the cross-sectional
and time-series variations in the capital structure of firms. He
provides evidence that firms with fewer growth opportunities
have more debt in their capital structure. He also finds that firms
that are more likely to have free cash flow problems have low debt
ratios and corporate tax does not seem to have a significant
impact on the capital structure choice of firms in the short run.
Walsh and Ryan (1997), on the other hand, test a binomial
choice model based upon observed debt and equity issues and
find that agency and tax considerations are significant in
determining debt and equity decisions of the UK firms.

3. FIRM-SPECIFIC CHARACTERISTICS THAT INFLUENCE CAPITAL

STRUCTURE

In what follows we discuss and motivate the choice of firmspecific characteristics which are thought to influence the
borrowing decisions of firms.
(i) Size
It has been suggested by a number of authors that firm size is
positively related to leverage ratio. The rationale for this belief is
the evidence provided by Warner (1977) and Ang et al. (1982)
that the ratio of direct bankruptcy costs to the firm value
decreases as the value increases, suggesting that the impact of
Blackwell Publishers Ltd 2001

180

OZKAN

these costs on the borrowing decisions of large firms might be

negligible. It is also argued that larger firms are more diversified
(Titman and Wessels, 1988), have easier access to the capital
markets, and borrow at more favourable interest rates (Ferri and
Jones, 1979). Furthermore, it is argued that larger firms may have
lower agency costs associated with the asset substitution and the
underinvestment problems (Chung, 1993). A further reason for
smaller firms to have lower leverage ratios might be that smaller
firms are more likely to be liquidated when they are in financial
distress (Ozkan, 1996). We use the natural logarithm of sales as a
proxy for the size of firms. This measure is in line with other
studies in this area (e.g., Titman and Wessels, 1988; Whited,
1992; and Rajan and Zingales, 1995).
(ii) Growth Opportunities
Myers (1977) suggests that the amount of debt issued by a firm is
inversely related to the growth opportunities consisting of future
investment opportunities, which would increase the value of the
firm when undertaken. It is argued that firms financed with risky
debt pass up some of these valuable investment opportunities in
some states of nature. Titman and Wessels (1988) also point out
that firms in growing industries incur higher agency costs since
they have more flexibility in taking future investments. It is also
suggested that although growth opportunities are capital assets
which add value to a firm, they cannot be collateralized and do
not generate current income. They are intangible in nature and
valuable as long as the firm is alive. Their value will fall
precipitously if the firm faces bankruptcy, which suggests that
expected bankruptcy costs for firms with greater growth
opportunities will be higher (Myers, 1984; Williamson, 1988;
and Harris and Raviv, 1990). Larger expected bankruptcy costs
would in turn imply lower financial leverage. As suggested in
Myers (1977), we use the market value of assets to the book value
of assets as a proxy for growth opportunities. Similar to other
studies (e.g. Smith and Watts, 1992; Whited, 1992; Barclay and
Smith, 1995; and Rajan and Zingales, 1995) this proxy is defined
as the ratio of book value of total assets minus the book value of
equity plus the market value of equity to book value of total
assets.
Blackwell Publishers Ltd 2001

DETERMINANTS OF CAPITAL STRUCTURE

181

(iii) Non-Debt Tax Shields

Some investments may generate non-debt tax benefits which are
unrelated to how firms finance these investments. Although
these investments do not consist of any debt related costs they act
as substitutes for tax shields. DeAngelo and Masulis (1980)
present a capital structure model where non-debt tax shields
serve as a substitute for the interest expenses which are
deductible in the calculation of the corporate tax. Therefore,
an inverse relationship is expected to exist between the amount
of the non-debt tax shields and leverage. Following Titman and
Wessels (1988), we use the ratio of annual depreciation expense
to total assets as a proxy for non-debt tax shields. However, it
should be noted that this may also be a proxy for other things
than non-debt tax shields. For example, firms with higher
depreciation ratios are also more likely to have relatively fewer
growth options in their investment opportunity sets and relatively
more tangible assets (Barclay and Smith, 1995; and Krishnaswami
et al., 1999). Following the argument given for growth
opportunities earlier, this then implies a positive relation
between the depreciation ratio and the leverage ratio.1
(iv) Profitability
Myers (1984) and Myers and Majluf (1984) point out that firms
prefer retained earnings as their main source of financing
investment (pecking order theory of capital structure) where the
second preference is debt financing, and last comes new equity
issues. It is suggested that observed capital structure of firms will
reflect the cumulative requirement for external financing. An
unusually profitable firm with a slow growth rate will end up with
an unusually low leverage ratio compared to the industry average
in which it operates. On the other hand, an unprofitable firm in
the same industry will end up with a relatively high debt-equity
ratio. In this sense, profitability allows the firm to use retained
earnings rather than external finance and a negative association
between profitability and debt ratios would be expected.
Following Titman and Wessels (1988) and Whited (1992), we
measure the profitability as ratio of the earnings before interest,
tax and depreciation (EBITD) to total assets.
Blackwell Publishers Ltd 2001

182

OZKAN

(v) Liquidity
Liquidity ratios may have a mixed impact on the capital structure
decision. First, firms with higher liquidity ratios might support a
relatively higher debt ratio due to greater ability to meet shortterm obligations when they fall due. This would imply a positive
relationship between a firm's liquidity position and its debt ratio.
On the other hand, firms with greater liquid assets may use these
assets to finance their investments. Therefore the firm's liquidity
position should exert a negative impact on its leverage ratio.
Moreover, as Prowse (1990) argues, the liquidity of the
company's assets can be used to show the extent to which these
assets can be manipulated by shareholders at the expense of
bondholders. We use the ratio of current assets to current
liabilities as a proxy for the liquidity of the firm's assets.

4. DATA AND EMPIRICAL MODEL

The data source for our analysis is the Datastream database which
provides both accounting data on firms and the market value of
equity. The sample data was constructed in the following manner.
First, firms which operate in the financial sector (banks, insurance
companies and investment trusts) and utilities (companies
providing a public service such as electricity, gas and telephone)
were excluded. All firms with any missing observations for any
variable in the model during the sample period were dropped.
Firms were then chosen that had at least five continuous time
series observations during the period 19841996, resulting in a
final sample of 390 firms and 4,132 observations.
In the next section we report alternative estimates of the
following specification
Dit Di;t1 ; Sizeit ; Liquidityit ; Ndtsit ; Profitit ; Growthit
i t "it

where Dit is the leverage ratio of company i in year t, and  allows

for lags in both dependent and explanatory regressors. i and t
represent time-invariant unobservable firm and/or industryspecific fixed effects and firm-invariant time-specific effects
respectively. Unobservable characteristics of the firm that have
Blackwell Publishers Ltd 2001

DETERMINANTS OF CAPITAL STRUCTURE

183

a significant impact on the firm's capital structure decision are

captured in i . They vary across firms but are assumed to be
constant for each firm. Examples of these effects include
managerial attitude such as ability and motivation, or attitudes
toward risk. They may also capture time-invariant industry
specific effects that are specific to the industry in which the firm
operates such as entry barriers, and factor market conditions. On
the other hand t is the same for all firms at a given point in time
but varies across time. These time-specific effects include
macroeconomic factors such as prices and interest rates.
Including t in the analysis controls for both observable and
unobservable aggregate time effects.
In the presence of firm-specific effects OLS coefficients are
biased assuming that i is unobservable and covariances between
regressors and i are nonzero (Hsiao, 1985). Also, OLS will result
in inconsistent estimation of the coefficient parameters since Di;t1
will be correlated with i which is constant. One way of dealing
with this problem is to take differences in order to eliminate the
firm-specific effects. Again OLS regression does not consistently
estimate the parameters because Di;t1 Di;t2 and uit ui;t1
are correlated through terms Di;t1 and ui;t1 . Anderson and Hsiao
(1982) suggested a consistent estimation technique. This requires
using
Di;t2 Di;t2 Di;t3 or Di;t2 as instruments for the first
difference of the lagged dependent variable where both are
correlated with Di;t1 Di;t2 but uncorrelated with uit ui;t1 .
The instrumental variable (IV) estimation will result in consistent
estimates if the error term uit in levels is not serially correlated.
However, the IV estimation does not necessarily lead to efficient
estimates of the model parameters as it fails to utilise all of the
available moment conditions. Arellano and Bond (1991) use the
Generalised Method of Moments (GMM) estimation technique,
which deploys additional instruments obtained by utilizing the
orthogonality conditions that exist between the lagged values of
the dependent variable and disturbances. They study the
performance of these estimators and show that the GMM estimates
result in smaller variances than those associated with the AH type
instrumental variable estimators.2 In what follows, we first present
the estimation results under each specification and then show that
the GMM is more appropriate to estimate our dynamic capital
structure model.
Blackwell Publishers Ltd 2001

184

OZKAN

5. RESULTS

Columns (1) and (2) of Table 1 present GMM and AndersonHsiao (AH) type estimates respectively, both in first differenced
form. Column (3) gives OLS estimates of leverage equation (1)
in levels. The estimation period is 198696 for both the GMM
and the AH estimates as a result of losing two cross-sections in
constructing one lag for each variable and taking first
differences. In both models, the fixed effects are eliminated by
first differencing and all the variables except the lagged
dependent variable are treated as exogenous. However, in model
(3) which gives the OLS estimates, the lagged dependent variable
is also treated as exogenous and the unobservable firm-specific
fixed effects remain. In all models the sample contains 390 firms
and 4,132 observations although usable observations vary
according to the estimation method. We report six test statistics:
(1) First order autocorrelation of residuals which is
asymptotically distributed as standard normal N(0,1) under the
null of no serial correlation; (2) Second order autocorrelation of
residuals which is distributed as standard normal N(0,1) under
the null of no serial correlation; (3) Wald test 1 is a Wald test of
joint significance of the estimated coefficients which is
asymptotically distributed as chi-square under the null of no
relationship; (4) Wald test 2 is a Wald test of the joint significance
of the time dummies; (5) Wald test 3 is a Wald test of the joint
significance of the industry dummies; and (6) Sargan test of
overidentifying restrictions which is asymptotically distributed as
chi-square under the null of instrument validity. All estimations
were carried out using the DPD program written in GAUSS
(Arellano and Bond, 1988).
Turning to the estimation results and comparing the GMM and
AH estimates it can be seen that the coefficient estimates
(including the lagged dependent variable) under AH are poorly
determined. The AH results reveal substantially larger variance
than that associated with the GMM, suggesting a loss in efficiency
compared to the GMM estimates. This is consistent with the
findings of Arellano and Bond (1991). There is also evidence that
the OLS level specification is inappropriate to estimate our
dynamic model. First, the serial correlation tests reveal that the
assumption of serially uncorrelated errors is violated and this
Blackwell Publishers Ltd 2001

185

DETERMINANTS OF CAPITAL STRUCTURE

Table 1
Alternative Estimates of Target Capital Structure
Dependent Variable: Dit
Independent
Variables
Di;t1
Sizeit
Sizeit1
Liquidityit
Liquidityit1
Ndtsit
Ndtsit1
Profitit
Profitit1
Growthit
Growthit1
Correlation 1
Correlation 2
Wald test 1 (df)
Wald test 2 (df)
Wald test 3 (df)
Sargan test (df)

(1)
GMM

(2)
AH

(3)
OLS

0.590***
(0.028)
0.077***
(0.009)
0.032***
(0.008)
0.055***
(0.011)
0.026***
(0.005)
0.059
(0.111)
0.020
(0.159)
0.244***
(0.026)
0.061**
(0.026)
0.024***
(0.005)
0.008**
(0.004)

0.295**
(0.132)
0.075***
(0.012)
0.012
(0.012)
0.035**
(0.015)
0.009*
(0.006)
0.064
(0.157)
0.190
(0.205)
0.225***
(0.037)
0.005
(0.044)
0.020***
(0.007)
0.006
(0.006)

0.781***
(0.015)
0.050***
(0.010)
0.047***
(0.010)
0.035***
(0.009)
0.030***
(0.008)
0.010
(0.124)
0.112
(0.133)
0.238***
(0.032)
0.133***
(0.027)
0.009*
(0.005)
0.003
(0.005)

11.560
1.258
979.68 (11)
124.40 (11)

88.06 (65)

3.966
0.547
2101.89 (11)
49.91 (11)

5.241
1.503
3772.02 (11)
46.74 (10)
5.884 (4)

Notes:
The dependent variable d is defined as the ratio of total debt to total assets. `Size' is the
natural logarithm of sales. `Liquidity' is the ratio of current assets to current liabilities.
`Ndts' is the ratio of annual depreciation expense to total assets. `Profits' is the ratio of
EBITD to total assets. `Growth' is the ratio of book value of total assets minus the book
value of equity plus the market value of equity to book value of total assets.
Time dummies are included in all models. Industry dummies are included only in model
3. Asymptotic standard errors robust to heteroscedasticity are reported in parentheses.
Model 1 gives the GMM estimates in first differences where Di;t2 , and further lags are
used as instruments. In Model 2 Anderson-Hsiao type estimates in first differences are
reported where
Di;t2 is used as instrument for
Di;t1 . Model 3 gives OLS estimates in
levels. ***, ** and * indicate coefficient is significant at the 1%, 5% and 10% levels,
respectively.

Blackwell Publishers Ltd 2001

186

OZKAN

suggests some degree of misspecification. Second, there is strong

evidence of an upward bias on the coefficient of the lagged
dependent variable in this OLS level specification. The estimated
coefficient of the lagged dependent variable under OLS is 0.78
compared to 0.59 under the GMM specification.3 This is
unsurprising since the lagged dependent variable is expected
to be biased upwards due to correlation with the unobservable
fixed effects. This result can also be seen as an indication of the
presence of firm-specific effects (Arellano and Bond, 1991). As
noted earlier, when these firm-specific effects exist and are
unobservable, OLS estimation in levels leads to an omitted
variables bias because of the potential correlation between fixed
effects and the included regressors.
GMM estimation is the preferred model but some of the test
statistics show evidence for the existence of misspecification. More
specifically, although the Wald tests of the joint significance of the
regressors and the time dummies are both satisfied, the test for the
absence of second order autocorrelation of residuals is only
marginally satisfied. Furthermore, the Sargan test reveals that the
instruments used in the GMM estimation may not be valid. The
null of instrument validity is rejected at the 5% level of
significance. We are marginally unable to reject the hypothesis at
the 1% significance level. Therefore, we conclude that it is
inappropriate to treat firm-specific characteristics as exogenous.
As mentioned above, it is unlikely that the firm-specific
variables are strictly exogenous because shocks affecting financial
choices of firms are also likely to affect, for example, market
value of equity, profitability, size and liquidity of firms. One
example of this potential endogeneity problem is as follows. If
the leverage of a firm increases one could then observe, assuming
leverage decreases the firm's market value since it increases
financial risk, a negative relation between leverage and the
market-to-book ratio. However, this could simply be due to the
effect of leverage on the ratio under investigation, rather than
vice versa. We therefore employ an instrumental variable
estimation technique, more specifically GMM, in an attempt to
reduce the likelihood of observed relations reflecting the effects
of leverage on firm characteristics.
Table 2 reports the GMM estimates, where all variables are
treated as endogenous. In the first two columns, we carry out the
Blackwell Publishers Ltd 2001

187

DETERMINANTS OF CAPITAL STRUCTURE

Table 2
GMM Estimates of Target Capital Structure
Dependent Variable: Dit
Independent
Variables
Di;t1
Sizeit
Sizeit1
Liquidityit
Liquidityit1
Ndtsit
Ndtsit1
Profitit
Profitit1
Growthit
Growthit1
Correlation 1
Correlation 2
Wald testr 1 (df)
Wald test 2 (df)
Sargan test (df)

(1)
GMM

(2)
GMM

(3)
GMM

0.431***
(0.036)
0.034
(0.024)
0.053**
(0.024)
0.025***
(0.006)
0.002
(0.003)
1.208***
(0.414)
0.240
(0.300)
0.457***
(0.096)
0.123***
(0.041)
0.044***
(0.014)
0.047***
(0.012)

0.485***
(0.035)
0.029
(0.024)
0.049**
(0.025)
0.019***
(0.006)
0.002
(0.004)
1.151***
(0.444)
0.090
(0.296)
0.375***
(0.092)
0.104**
(0.041)
0.040***
(0.014)
0.043***
(0.012)

0.443***
(0.034)

0.024*
(0.013)
0.020***
(0.005)

1.071***
(0.367)

0.495***
(0.093)
0.121***
(0.042)
0.040***
(0.013)
0.047***
(0.012)

11.478
0.192
477.73 (11)
55.314 (11)
63.25 (54)

11.492
0.996
502.84 (11)
55.07 (11)
70.36 (54)

11.666
0.758
468.03 (8)
66.08 (11)
66.35 (58)

Notes:
The dependent variable D is defined as the ratio of total debt to total assets in Models (1)
and (3) and the ratio of total debt plus preferred capital to total assets in Model (2). `Size'
is the natural logarithm of sales. `Liquidity' is the ratio of current assets to current
liabilities. `Ndts' is the ratio of annual depreciation expense to total assets. `Profit' is the
ratio of EBITD to total assets. `Growth' is the ratio of book value of total assets minus the
book value of equity plus the market value of equity to book value of total assets.
In all models Di;t2 , Liquidityi;t2 , Ndtsi;t2 , Profiti;t2 , Growthi;t2 are used as instruments.
The estimation period is 198696 for all models.
See also notes to Table 1.

Blackwell Publishers Ltd 2001

188

OZKAN

GMM estimation with alternative definitions of the dependent

variable (leverage). In the first column of Table 2, we define
leverage as the ratio of total debt to total assets and in the second
column we include preferred capital in the definition of total
debt. Preferred capital is included in the numerator since
preference shareholders are entitled to a fixed rate of dividend.
Finally, in the third column, we report the GMM results, where
the coefficients that are individually and jointly insignificant are
omitted. The definition of leverage in this specification is the
same as that in the first column. The results are virtually the same
when we use the alternative definition of leverage and hence they
are not reported separately. All three models show similarities
with the GMM estimator in the previous table in terms of the
estimated coefficients and the test-statistics. The signs of the
estimated coefficients are the same except those of Sizeit, Sizeit1
and Growthit. However, the models in Table 2 are better specified
than the GMM model in Table 1. More specifically, the test for
serial correlation in residuals provides evidence of negative firstorder serial correlation. However, second-order serial correlation
is absent.4 The Wald test for the joint significance of the
regressors is also satisfied. Time dummies are jointly significant
suggesting that aggregate factors exert a significant influence on
borrowing decisions of firms. More importantly, the Sargan test
indicates that the instruments used in the GMM estimation are
valid and this implies that the instruments are not correlated with
the error term.
Turning to the GMM estimates of the coefficients, the majority
of the estimated coefficients have the expected signs and are
significant. As expected, the coefficient of the lagged leverage is
positive and significant at the 1% level. The adjustment coefficient
 which is given by 1 0 is relatively large (it is greater than 0.5 in
all cases) possibly providing evidence that the dynamics implied by
our model are not rejected and firms adjust their leverage ratios
relatively quickly in an attempt to have their target debt ratios.
One possible explanation for this adjustment speed could be that
the costs deviating from the target debt ratio are significant and
firms' leverage ratios are persistent over time. However, as revealed
by the coefficient of the lagged leverage ratio being midway
between 0 and 1, this adjustment process is costly. This is
consistent with the view that firms may trade-off between two
Blackwell Publishers Ltd 2001

DETERMINANTS OF CAPITAL STRUCTURE

189

different types of costs: costs of making adjustment to their target

ratios and costs of being in disequilibrium (being off target). The
adjustment coefficient would be close to one if the costs of being
in disequilibrium were much higher than the costs of adjustment.
Alternatively, it would be close to zero if the costs of adjustment
were much higher than the costs of disequilibrium.
The coefficient of growth opportunities (as proxied by the
market-to-book ratio) is negative and significant. The negative
impact of growth opportunities on leverage might reveal several
features of borrowing behaviour of companies. It may give
support to the prediction that firms which have a relatively large
proportion of intangible assets cannot support a high leverage
ratio. This evidence is also consistent with the view that firms with
greater growth opportunities might have lower leverage ratios
due to the fear of debtholders that firms might pass up valuable
investment opportunities. However, it should be noted that there
may be other potential reasons for the negative coefficient of the
market-to-book ratio. For instance, this may stem from the
tendency of firms to issue stock when their stock price is high
relative to their earnings or book value. This would imply that the
negative correlation between leverage and the market-to-book
ratio is driven largely by firms that issue significant amounts of
equity (Rajan and Zingales, 1995). The coefficient for the lagged
growth is positive and significant. One explanation for the
opposite sign of the lagged growth might be that growth has a
transitory effect on leverage ratios.
There is little evidence that firm size (as proxied by the natural
logarithm of sales) has a positive effect on their leverage ratios.
The lagged size coefficient is significant at the 5% level in the
first two models and only at the 10% level in the third model in
Table 2. This result does not change when we estimate our model
by including the logarithm of total assets as an alternative proxy
for the size variable. Still, this limited evidence may reflect two
features. First, large firms might be more diversified and fail less
often, so past values of firm size may serve as an inverse proxy for
the probability of bankruptcy. Second, to the extent that the size
of firms is an inverse proxy for the direct costs of bankruptcy
small firms are expected to borrow less than large firms.
However, we have no evidence in support of the positive effect
of current sales on leverage decisions of firms. Although the
Blackwell Publishers Ltd 2001

190

OZKAN

estimated coefficient is not significant, the negative sign of this

variable is worth discussing. Size may be seen as an inverse proxy
for the degree of asymmetric information between the firm and
outside investors (Rajan and Zingales,1995). To the extent that
this is the case, one should expect the firm to increase its
preference for equity relative to debt financing, which implies a
negative relation between size and leverage.
As noted earlier firms with a high level of non-debt tax shields
which can be deducted from the taxable income are expected to
have less debt than other firms ceteris paribus. This prediction is
confirmed by the negative and significant coefficient of the
current non-debt tax shields ratio (proxied as the ratio of annual
depreciation expense to total assets). The estimated coefficient
for this variable is significant at the 1% level in all specifications.
There is no evidence that past values of non-debt tax shields exert
an influence on corporate borrowing decisions. Nevertheless,
one should be cautious in interpreting these results because our
measure may be a proxy for some other effects. For example,
firms with higher depreciation ratios are more likely to have a
higher proportion of tangible assets and fewer growth
opportunities in their investment opportunity set. This in turn
implies a positive relation between the non-debt tax shields ratio
and leverage. However, this prediction is not confirmed by our
results. In order to further explore this possibility, we replaced
`Ndts' by `Fixed assets' which is defined as the ratio of tangible
assets to total assets. We could not find any significant relation
between this variable and leverage.
The findings are in line with the view that liquidity of firms
exerts a negative impact on firms' borrowing decisions. The
coefficient estimate of the current liquidity ratio is significant at
the 1% level. This negative effect might be due to potential
conflicts between debtholders and shareholders of firms. As
noted earlier, the liquidity of firms' assets can be taken as
evidence to show the extent to which these assets can be
manipulated by shareholders at the expense of bondholders. The
positive effect of the liquidity position of the firm on its leverage
ratio is not realised.
Finally, evidence emerges that current profitability of firms
exerts a negative influence on firms' borrowing decisions. The
estimated coefficient is significant at the 1% level. The negative
Blackwell Publishers Ltd 2001

DETERMINANTS OF CAPITAL STRUCTURE

191

sign of profitability is consistent with the pecking order theory that

predicts a preference for internal finance rather than over
external finance. However, the coefficient on the lagged profit is
positive and significant. One possible explanation for the positive
influence of profitability on leverage might be that more
profitable firms can support a higher leverage ratio. Also, Jensen
(1986) explains when a positive relation between profitability and
leverage can be expected. He argues that this happens if the
market for corporate control is effective in forcing firms to
commit to paying out cash by levering up. In such a market
managers of profitable firms cannot avoid the disciplinary role of
debt financing and suppliers of debt could be more willing to
lend to firms with higher profits. However, this is inconsistent
with the view that the relation between past profitability and
leverage should also be negative as past profitability can be
viewed as proxy for future growth opportunities whose value
could be severely damaged in financial distress (Smith and Watts,
1992; and Shyam-Sunder and Myers, 1999).
(i) Robustness of the Results
We conducted a number of experiments to examine the
robustness of the results obtained in this paper. Firstly, in
addition to the Sargan test for the validity of the instrument set,
we carefully investigated whether the variables used in the
analysis are predetermined or strictly exogenous with respect to
error term. This is important because the GMM estimator
provides consistent estimates only if a valid instrument set is
used and also the form of the instrument set depends on the
relationship between regressors and uit. For example, supposing
that one regressor (say xit) is correlated with the fixed effects and
uit is serially uncorrelated, one must consider whether xit is
predetermined or strictly exogenous with respect to uit. These
possibilities are investigated following a similar approach to
Blundell et al. (1992). In order to test for the possibility that xit is
predetermined with respect to uit we start using instruments
dated t 2 for each variable included in the instrument set.
Later, xi;t1 is added to the existing instruments to investigate the
potential biases which arise from the correlation between xi;t1
and the first-differenced error term
uit . In the presence of
Blackwell Publishers Ltd 2001

192

OZKAN

measurement error, the estimate of the coefficients of x is

expected to fall, which suggests a downward bias due to the
simultaneous determination of xi;t1 and ui;t1 . This procedure
was carried out for each variable and it was concluded that no
variable is predetermined with respect to uit. Thus for all variables
instruments dated t 2 are chosen. We also investigate the
possibility of strict exogeneity of variables with respect to uit by
including current values. The effect is a fall in the estimates of
the coefficients so that we conclude that no variable is strictly
exogenous with respect to uit.
We also carried out the GMM estimation with alternative
definitions of the dependent variable (leverage) and some other
firm-specific characteristics. Firstly, to assess the robustness of the
results to alternate specification of the leverage variable, we
replaced our basic measure of leverage (defined as the ratio of
long-term plus short-term debt to total assets) with the ratio of
only long-term to total assets. The results using this definition of
the dependent variable were very similar to those reported above.
This may be due to the fact that short-term bank loans and
overdrafts may often be renewable and, therefore being
effectively a source of long-term finance. As mentioned earlier,
we also included the logarithm of total assets as an alternative
proxy for the size variable. The results were again very similar to
those reported below and, therefore, are not reported separately.
The results reported above are also insensitive to whether EBITD
is normalised by total assets or sales. Due to the opposite signs of
the estimated coefficients of current and past profitability and
growth opportunities ratios we investigated the possibility that
the change in these ratios rather than current profitability and
growth of firms may influence leverage but we could not find
such evidence.

6. CONCLUSIONS

This paper has investigated the determinants of target capital

structure of firms and the role of adjustment process. There are
several important features of the data set and the estimation
technique used in this paper, which, we believe, extends the
literature on the empirical determinants of borrowing decisions
Blackwell Publishers Ltd 2001

DETERMINANTS OF CAPITAL STRUCTURE

193

of firms. First, a panel data set for 390 UK companies was

constructed, which allows us to adapt a dynamic model which
sheds some light on the issues of long-term target debt ratio of
firms and adjustment process to this target. Second, panel data
analysis and the GMM estimation procedure have allowed us to
effectively control for firm-specific fixed effects which are
unobservable yet important in affecting financial decisions of
firms. Third, the analysis of this paper has dealt with the
endogeneity problem by allowing us to choose more efficient
instruments to control for endogeneity. Our investigation has
provided the following insights into the empirical determinants
of corporate borrowing. The evidence suggests that firms have
long-term target leverage ratios and they adjust to the target ratio
relatively fast, implying that the costs of being away from their
target ratios and the costs of adjustment are equally important for
firms. The results provide evidence that profitability, liquidity
and growth opportunities exert a negative effect on the capital
structure choice of firms. Our results are also consistent with the
prediction of the theory that there exists an inverse relationship
between non-debt tax shields and borrowing ratio of firms.
Finally, there is only a limited support for a positive effect arising
from size of firms.

APPENDIX

Empirical Specification
We investigate the role of adjustment costs by adopting a partial
adjustment model. Suppose that the target long-term debt ratio
of firms is taken to be a function of several variables:
X
Dit
k xkit "i t
A1
k

where firms are represented by subscript i 1; . . . ; N , and time

by t 1; . . . ; T . In the model leverage ratio, D, is explained in
terms of K explanatory variables x1 ; . . . ; xk . Disturbance term "it is
assumed to be serially uncorrelated with mean zero and possibly
heteroscedastic. k0 s which are common to each firm are the
unknown parameters of interest.
Blackwell Publishers Ltd 2001

194

OZKAN

Firms adjust their borrowing in order for their current leverage

ratio to be close to the target one. This leads to a partial
adjustment mechanism which is given by:
Dit Di;t1 Dit Di;t1

A2

where 0 <  < 1, Dit is the actual debt ratio and Dit is the target
debt ratio of firm i at time t. Dit Di;t1 can be interpreted as
the target change whereas only a fraction  of the target change
is achieved, which is equal to Dit Di;t1 . Combining (A1) and
(A2) yields:
X
Dit 1 Di;t1
k xkit it
A3
k

which can be written as:

Dit 0 Di;t1

k xkit uit

A4

k1

where 0 1 ; k k ; and "it uit (where uit has the same

properties as "it ).
Sample Characteristics and Descriptive Statistics
Our panel data set which includes 390 firms has the following
unbalanced structure:
Table A2 gives the number of observations (companies)
available in each year of the data sample:
Table A1
Structure of Panel
Number of Records on Each Company
5
6
7
8
9
10
11
12
13

Number of Companies
24
17
32
31
34
33
15
15
189

Blackwell Publishers Ltd 2001

195

DETERMINANTS OF CAPITAL STRUCTURE

Table A2
The Number of Companies in Each Year
Year

Number of Companies

1984
1985
1986
1987
1988
1989
1990
1991
1992
1993
1994
1995
1996

223
237
253
285
316
337
362
370
370
359
350
341
329

Table A3
Descriptive Statistics
Variable

Mean (Standard Deviation)

Levels

Leverage
Leveraget1
Sizet1
Growth
Growtht1
Ndts
Profit
Profitt1
Liquidity

0.168
0.167
11.234
0.288
0.274
0.036
0.103
0.107
1.641

(0.106)
(0.105)
(1.807)
(0.504)
(0.500)
(0.020)
(0.087)
(0.079)
(0.756)

First Differences
0.001
0.002
0.116
0.013
0.010
0.001
0.004
0.003
0.004

(0.071)
(0.071)
(0.247)
(0.302)
(0.312)
(0.012)
(0.078)
(0.070)
(0.467)

NOTES
1

Bennett and Donnelly (1993) and Walsh and Ryan (1997) use the ratio of
unprovided deferred tax to total assets as a proxy for non-debt tax shields.
Unfortunately, in our data set there were too many missing observations for
these variables so that we could not use it as an alternative proxy for nondebt tax shields.
The GMM estimator allows the instruments used in each period to increase
as one moves through the panel, whereas the AH type estimator uses only

Di;t2 to instrument
Di;t1 . The set of valid instruments change
depending upon the assumption concerning the correlation between xikt

Blackwell Publishers Ltd 2001

196

OZKAN

and uit . It is suggested that the valid instruments for period t for the
equation in first differences will be zit Di1 ; . . . ; Di;t2 ; xik1 ; . . . ; xik;t1
under the assumption that uit is serially uncorrelated, and xikt is predetermined. That is, Exkit uis 6 0 for s < t and zero otherwise. If, on the
other hand, xikt is strictly exogenous, i.e. Exikt uis 0 for all t,s, then all x's
are valid instruments. In this case zit becomes zit Dit ; . . . ; Dis ; xik1 ; . . . ; xikT
where s 1; . . . ; T 2.
One should be cautious in making comparison between alternative
estimates at this stage because this may be obscured by the likely
endogeneity of firm-specific characteristics. To explore this possibility, we
treated all variables as endogenous and derive the results accordingly. This
did not change the basic results in that the coefficient of the lagged
dependent variable is significantly biased upwards in OLS specification and
the coefficient estimates of the AH specification are poorly determined
compared to the GMM estimates.
Note that, if uit are serially uncorrelated in the model in levels of first
differences, transformation induces first-order serial correlation in the
differenced residuals, but second order serial correlation should be absent.
That is, as explained in Arellano and Bond (1991), Euit uit1 need not be
zero, but the consistency of the GMM estimators relies heavily on the
assumption that Euit uit2 0.

REFERENCES
Anderson, T.W. and C. Hsiao (1982), `Formulation and Estimation of Dynamic
Models Using Panel Data', Journal of Econometrics, Vol. 18, pp. 4782.
Ang, J.S., J.H.Chua, and J.J. McConnell (1982), `The Administrative Costs of
Corporate Bankruptcy: A Note', Journal of Finance, Vol. 37, pp. 33748.
Arellano, M. and S. Bond (1988), `Dynamic Panel Data Estimation Using DPD
A Guide for Users', Institute for Fiscal Studies, Working Paper 88/15
(London).
________ ________ (1991), `Some Tests of Specification for Panel Data: Monte
Carlo Evidence and an Application to Employment Equations', The Review
of Economics Studies, Vol. 58, pp. 27797.
Auerbach, A.J. (1985), `Real Determinants of Corporate Leverage', in Friedman
(ed.), Corporate Capital Structure in the United States (University of Chicago
Press, Chicago).
Barclay, M.J. and C.W. Smith (1995), `The Maturity Structure of Corporate
Debt', Journal of Finance, Vol. 50, pp. 60932.
Bennett, M. and R. Donnelly (1993), `The Determinants of Capital Structure:
Some UK Evidence', British Accounting Review, Vol. 25, pp. 4359.
Blundell, R., S. Bond, M. Devereux, and F. Schiantarelli (1992), `Investment
and Tobin's Q: Evidence from Company Panel Data', Journal of Econometrics,
Vol. 51, pp. 23357.
Bowen, R.M., L.A. Daley, and C.C. Huber (1982), `Evidence on the Existence
and Determinants of Inter-Industry Differences of Leverage', Financial
Management, Vol.11, pp. 1020.
Bradley, M., A. Jarrell and E. Kim (1984), `On the Existence of an Optimal
Capital Structure: Theory and Evidence', Journal of Finance, Vol. 39, pp.
85780.
Blackwell Publishers Ltd 2001

DETERMINANTS OF CAPITAL STRUCTURE

197

Chung, K.H. (1993), `Asset Characteristics and Corporate Debt Policy: An

Empirical Test', Journal of Business Finance & Accounting, Vol. 20, No. 1
(January), pp. 8398.
DeAngelo, H. and R.W. Masulis (1980), `Optimal Capital Structure Under
Corporate and Personal Taxation', Journal of Financial Economics, Vol. 8, pp.
329.
Ferri, M.G. and W.H. Jones (1979), `Determinant of Financial Structure: A New
Methodological Approach', Journal of Finance, Vol. 34, pp. 63144.
Harris, M. and A. Raviv (1990), `Capital Structure and the Informational Role of
Debt', Journal of Finance, Vol. 45, pp. 32149.
Homaifar, G., J. Zietz and O. Benkato (1994), `An Empirical Model of Capital
Structure: Some New Evidence', Journal of Business Finance & Accounting,
Vol. 21, No. 1 (January), pp. 114.
Hsiao, C. (1985), `Benefits and Limitations of Panel Data', Econometric Reviews,
Vol. 4, pp. 12174.
Jalilvand, A. and R.S. Harris (1984), `Corporate Behaviour in Adjusting to
Capital Structure and Dividend Targets: An Econometric Study', Journal of
Finance, Vol. 39, 12745.
Jensen, M.C. (1986), `Agency Costs of Free Cash Flow, Corporate Finance and
Takeovers', American Economic Review, Vol. 76, pp. 32339.
Kester, C.W. (1986), `Capital and Ownership Structure: A Comparison of
United States and Japanese Manufacturing Corporations', Financial
Management, Vol. 15, pp. 97113.
Kim, E. (1978), `A Mean-Variance Theory of Optimal Capital Structure and
Corporate Debt Capacity', Journal of Finance, Vol. 33, pp. 4563.
Kim, W.S. and E.H. Sorensen (1986), `Evidence on the Impact of the Agency
Costs of Debt on Corporate Debt Policy', Journal of Financial and Qualitative
Analysis, Vol. 21, pp. 13144.
Kraus, A. and R. Litzenberger (1973), `A State-Preference Model of Optimal
Financial Leverage', Journal of Finance, Vol. 28, pp. 91122.
Krishnaswami, S., P.A. Spindt, and V. Subramaniam (1999), `Information
Asymmetry, Monitoring and the Placement Structure of Corporate Debt',
Journal of Financial Economics, Vol. 51, pp. 40734.
Lasfer, M.A. (1995), `Agency Costs, Taxes and Debt', European Financial
Management, pp. 26585.
Long, M. and I. Malitz (1985), `The Investment Financing Nexus: Some
Empirical Evidence', Midland Corporate Finance Journal, Vol. 3, pp. 5359.
Marsh, P. (1982), `The Choice Between Debt and Equity: An Empirical Study',
Journal of Finance, Vol. 37, pp. 12144.
Modigliani, F. and M.H. Miller (1958), `The Cost of Capital, Corporation
Finance and the Theory of Investment', American Economic Review, Vol. 48,
pp. 26197.
Myers, S.C. (1977), `Determinants of Corporate Borrowing', Journal of Financial
Economics, Vol. 5, pp. 14775.
________ (1984), `The Capital Structure Puzzle', Journal of Finance, Vol. 39, pp.
57592.
________ and N.S. Majluf (1984), `Corporate Financing and Investment
Decisions When Firms Have Information that Investors Do Not Have',
Journal of Financial Economics, Vol. 13, pp. 187221.
Ozkan, A. (1996), `Corporate Bankruptcies, Liquidation Costs and the Role of
Banks', The Manchester School, Vol. 64, pp. 10419.
Prowse, S.D. (1991), `Institutional Investment Patterns and Corporate Financial
Blackwell Publishers Ltd 2001

198

OZKAN

Behaviour in the U.S. and Japan', Journal of Financial Economics, Vol. 27, pp.
4366.
Rajan R.G. and L. Zingales (1995), `What Do We Know About Capital Structure?
Some Evidence from International Data', Journal of Finance, Vol. 50, pp.
142160.
Scott, J.H. (1976), `A Theory of Optimal Capital Structure', Bell Journal of
Economics, Vol. 7, pp. 3354.
Shyam-Sunder, L. and S.C. Myers (1999), `Testing Static Tradeoff Against
Pecking Order Models of Capital Structure', Journal of Financial Economics,
Vol. 51, pp. 21944.
Smith, C.W. and R.L. Watts (1992), `The Investment Opportunity Set and
Corporate Financing, Dividend, and Compensation Policies', Journal of
Financial Economics, Vol. 32, pp. 26392.
Taggart, R.A. (1977), `A Model of Corporate Financing Decisions', Journal of
Finance, Vol. 32, pp. 146784.
Titman, S. and R. Wessels (1988), `The Determinants of Capital Structure
Choice', Journal of Finance, Vol. 43, pp. 119.
Walsh, E.J. and J. Ryan (1997), `Agency and Tax Explanations of Security
Issuance Decisions', Journal of Business Finance & Accounting, Vol. 24, Nos 7
& 8 (September), pp. 94361.
Warner, J.B. (1977), `Bankruptcy Costs: Some Evidence', Journal of Finance, Vol.
32, pp. 33748.
Whited, T. (1992), `Debt, Liquidity Constraints, and Corporate Investment:
Evidence from Panel Data', Journal of Finance, Vol. 47, pp. 142560.
Williamson, O. (1988), `Corporate Finance and Corporate Governance', Journal
of Finance, Vol. 43, pp. 56791.

Blackwell Publishers Ltd 2001