Capital Structure and Firm Efficiency

Journal of Business Finance & Accounting, 34(9) & (10), 1447–1469, November/December 2007, 0306-686X
doi: 10.1111/j.1468-5957.2007.02056.x
Capital Structure and Firm Efficiency
Dimitris Margaritis and Maria Psillaki
∗
Abstract: This paper investigates the relationship between firm efficiency and leverage. We
consider both the effect of leverage on firm performance as well as the reverse causality
relationship. In particular, we address the following questions: Does higher leverage lead
to better firm performance? Does efficiency exert a significant effect on leverage over and
above that of traditional financial measures of capital structure? Is the effect of efficiency on
leverage similar across different capital structures? What is the signalling role of efficiency to
creditors or investors? Using a sample of 12,240 New Zealand firms we find evidence supporting
the theoretical predictions of the Jensen and Meckling (1976) agency cost model. Efficiency
measured as the distance from the industry’s ‘best practice’ production frontier is positively
related to leverage over the entire range of observed data. The frontier is constructed using the
non-parametric Data Envelopment Analysis (DEA) method. Using quantile regression analysis
we show that the reverse causality effect of efficiency on leverage is positive at low to mid-leverage
levels and negative at high leverage ratios. Firm size also has a non-monotonic effect on leverage:
negative at low debt ratios and positive at mid to high debt ratios. The effect of tangibles and
profitability on leverage is positive while intangibles and other assets are negatively related to
leverage.
Keywords: capital structure, agency costs, firm efficiency, DEA
1. INTRODUCTION
In an influential paper Leibenstein (1966) showed how different principal-agent objec-
tives, inadequate motivation and incomplete contracts become sources of (technical)
inefficiency measured by the discrepancy between maximum potential output and the
firm’s actual output. He termed this failure to attain the production or technological
frontier as X-inefficiency.
Leibenstein’s work fits well with recent theories that emphasize the disciplinary role
of leverage in agency conflicts and the importance of contracting and information costs
in the determination of the firm’s capital structure policy (see Jensen and Meckling,

∗
The authors are respectively Professor in the Department of Finance, AUT, Auckland, New Zealand
and Associate Professor, University of Nice-Sophia Antipolis and GREDEG-CNRS, Valbonne, France. They
acknowledge financial support from the New Zealand Foundation for Research, Science and Technology.
They are grateful to the editor of this journal and an anonymous referee for their valuable comments.
They would also like to thank Gary Feng and Boram Lee for excellent research assistance. (Paper received
December 2005, revised version accepted May 2007. Online publication August 2007)
Address for correspondence: Dimitris Margaritis, Department of Finance, Faculty of Business, AUT, Private
Bag 92006, Auckland 1020, New Zealand.
e-mail:
C

2007 The Authors
Journal compilation
C

2007 Blackwell Publishing Ltd, 9600 Garsington Road, Oxford OX4 2DQ, UK
and 350 Main Street, Malden, MA 02148, USA.
1447
1448 MARGARITIS AND PSILLAKI
1976; Myers, 1977; Myers andMajluf,1984;Harrisand Raviv, 1990; and Walsh and Ryan,
1997). While theoretical work on capital structure has seen remarkable progress since
the seminal Modigliani and Miller (1958) contribution, the practical applications of
capital structure theories are far from satisfying (see Rajan and Zingales, 1995; Ross
et al., 2005; and Beattie et al., 2006).
The main problem applied researchers face is that several of the variables which are
hypothesized to affect capital structure are not directly observable or are difficult to
measure (see Barclayand Smith, 2001). Likewise, Berger and Bonaccorsi di Patti(2006)
argue that the lack of conclusive evidence in support of the agency cost hypothesis may
in part be attributed to the difficulties researchers face in obtaining a measure of firm

performance that is closely related to the theoretical definition of agency costs.
The first objective of this paper is to consider explicitly the role of production
and cost decisions in determining the extent of firm leverage. To do that we will
need to reconstruct the black box details of production technology which have been
traditionally left out in modern finance representations of the firm. We rely on
Leibenstein’s framework and more recent theoretical work on production frontiers
and performance measurement to establish the link between productive efficiency and
leverage.
1
Using this apparatus, we interpret thedistance from the best practice frontier
as a summary measure of incomplete contracts, differentprincipal-agent objectives and
inadequate motivation or more generally for the types of market imperfections that
are often used as reasons to explain why firm performance may have an effect on
equilibrium capital structure. We then proceed to analyze the effects of efficiency on
capital structure using two competing hypotheses. Under the efficiency-risk hypothesis,
more efficient firms may choose higher debt to equity ratios because higher efficiency
reduces the expected costs of bankruptcy and financial distress. On the other hand,
under the franchise-value hypothesis, more efficient firms may choose lower debt to equity
ratios to protect the economic rents derived from higher efficiency from the possibility
of liquidation (Berger and Bonaccorsi di Patti, 2006).
Potential conflicts of interest between owners and managers impact on the value of
the firm. For example, managers with free cash flows in mature industries with limited
profitable growth opportunities may overinvest in their own companies or diversify
in unchartered territories both of which will negatively impact on profitability and
the value of the firm (see Jensen, 1986).
2
Similarly, firm values especially those of
growth companies facing financial difficulties may be adversely affected when there are
conflicts between debt and equity holders. In these situations managers acting in the
interests of shareholders are likely to underinvest (see Myers, 1977) as they recognize

that most of the value created by capital investments would eventually benefit the
creditors rather than the owners. The second objective of this paper is thus to assess
the extentto which leverageacts as a disciplinary device in mitigating the agency costs of
outside ownership and thereby contributes to an improvement on firm performance.
3
We also allow for the possibility that at high levels of leverage the agency costs of outside
1 Fried et al. (2007) and F¨are et al. (2007) provide an extensive review of the literature on efficiency and
productivity.
2 Clearly the best strategy for adding value in these situations is to distribute the free cash flow to investors
either by increasing dividends or preferably by substituting debt for equity through stock repurchases
(see Barclay and Smith, 2001). There is evidence to suggest that leveraged buyouts and other leveraged
recapitalisations have been associated with improvements in operational efficiencies (see Chew, 2001).
3 We thank an anonymous referee for suggesting that we incorporate the effect of leverage on efficiency.
C

2007 The Authors
Journal compilation
C

Blackwell Publishing Ltd. 2007
CAPITAL STRUCTURE AND FIRM EFFICIENCY 1449
debt may overcome those of outside equity whereby further increases in leverage lead
to an increase in total agency costs.
This paper contributes to the literature in two directions: (1) by using X-efficiency
as opposed to financial variables to measure firm performance and test the predictions
of the agency cost hypothesis; (2) by examining whether the distance from the best
practice frontier is an important determinant of capital structure. We view the best
practice frontier as a benchmark for each firm’s performance that would be realized if
agency costs were minimized. As articulated by Berger and Bonaccorsi di Patti (2006),
efficiency measures are closer to the theoretical definition of agency costs and have

the advantage over the more traditional measures of firm performance based on
financial indicators in that they control for firm-specific factors outside the control
of management which are not part of agency costs. More specifically, we address the
following questions: Does higher leverage lead to better firm performance? Does
efficiency exert a significant effect on leverage over and above that of traditional
financial measures? Is the effect of efficiency on leverage identical across different
capital structures? What is the signalling role of efficiency to creditors or investors?
This is to our knowledge one of the first studies to consider the association between
productive efficiency and leverage. In a recent study Berger and Bonaccorsi di Patti
(2006) examine the bi-directional relationship between capital structure and firm
performance for the US banking industry using a parametric measure of profit
efficiency as an indicator of (inverse) agency costs. They find evidence in support
of the agency cost hypothesis whereby the effect of the agency cost of outside equity
dominates the effect of outside debt over almost the entire range of observed data. For
the reverse causality from efficiency to capital structure they report a negative effect of
efficiency on equity capital at high levels of efficiency and a positive effect at low levels
of efficiency.
In a different but related context, Zavgren (1985), Keasey and Watson (1987) and
Becchetti and Sierra (2003) have emphasized the importance of non-financial data as
predictors of company failures. For instance, Zavgren (1985) argues that econometric
models that solely rely on financial statement information will not predict accurately
business failures. Using a measure of efficiency obtained from a stochastic frontier
model, Becchetti and Sierra (2003) find that productive inefficiency is a significant
ex-ante indicator of business failure while Keasey and Watson (1987) report that better
predictions for small company failures are obtained from models using non-financial
data rather than conventional financial indicators. We adopt a similar reasoning in
considering the relationship between financial structure and X-efficiency in this paper.
The reminder of the paper is organized as follows. The next section details the
methodology used in this study to construct the ‘best practice’ frontier and establish
the link between efficiency and capital structure. Section 3 describes the variables used

in the econometric analysis of efficiency and leverage. Section 4 describes the data, the
empirical specification and estimation procedure and reports the empirical results.
Section 5 concludes the paper.
2. METHODOLOGY
(i) Firm Performance and Capital Structure
The agency cost theory is premised on the idea that the interests of the company’s
managers and its shareholders are not perfectly aligned. In their seminal paper Jensen
C

2007 The Authors
Journal compilation
C

Blackwell Publishing Ltd. 2007
1450 MARGARITIS AND PSILLAKI
and Meckling (1976) emphasized the importance of the agency costs of equity in
corporate finance arising from the separation of ownership and control of firms
whereby managers tend to maximize their own utility rather than the value of the
firm. This leads us to Jensen’s (1986) ‘free cash flow theory’ where as stated by Jensen
(1986, p. 323) ‘the problem is how to motivate managers to disgorge the cash rather
than investing it below the cost of capital or wasting it on organizational inefficiencies.’
In other words complete contracts cannot be written. A higher level of leverage may be
used as a disciplinary device to reduce managerial cash flow waste through the threat
of liquidation (Grossman and Hart, 1982) or through pressure to generate cash flows
to service debt (Jensen, 1986).
Agency costs can also exist from conflicts between debt and equity investors.
These conflicts arise when there is a risk of default. The risk of default may create
what Myers (1977) referred to as an ‘underinvestment’ or ‘debt overhang’ problem.
Alternatively, it could lead to increased risk-taking activity as managers acting on their
shareholders’ behalf have incentives to take excessive risks as part of risk shifting

investment strategies (see Jensen and Meckling, 1976). Whereas the agency costs
of outside equity underpin a positive relationship between firm performance and
leverage, the agency costs of outside debt result in a negative effect of leverage on
firm performance as highly leveraged firms especially those on the brink of default are
more likely to pass up profitable investment opportunities or shift to riskier operating
strategies.
Thus a firm’s ability to achieve best practice relative to its peers will be compromised
in situations where it is forced to forego valuable investment opportunities, participate
in uneconomic activities that sustain growth at the expense of profitability or is subject
to other organizational inefficiencies. Following Leibenstein (1966) we use technical or
X-inefficiency as a proxy for the (inverse) agency costs arising from conflicts between
debt holders and equity holders or from different principal-agent objectives. These
conflicts will give rise to a discrepancy between a firm’s potential and actual output
so that individual firms with similar technologies can be benchmarked against their
best performing peers. As in Berger and Bonaccorsi di Patti (2006) we view these best
practice firms as those which minimize the agency costs of outside equity and outside
debt.
In line with Jensen and Meckling (1976) we expect the effect of leverage on agency
costs to be negative overall. We do however, allow in our model specification for the
possibility that this effect may be reversed at the point where the expected costs of
financial distress outweight any gains achieved through the use of debt rather than
equity in the firm’s capital structure. Therefore, under the agency cost hypothesis (H
1
)
higher leverage is expected to lower agency costs, reduce inefficiency and thereby lead
to an improvement in firm’s performance with the proviso that the direction of this
relationship may switch at a point where the disciplinary effects of further increases in
leverage become untenable.
But firm performance may also affect the choice of capital structure. As Berger
and Bonaccorsi di Patti (2006) pointed out, regressions of firm performance on

leverage may confound the effects of capital structure on performance with the reverse
relationship from performance on capital structure. This reverse causality effect was
in essence a feature of theories considering how agency costs (Jensen and Meckling,
1976; Myers, 1977; and Harris and Raviv, 1990); corporate control issues (Harris and
Raviv, 1988); and in particular, asymmetric information (Myers and Majluf, 1984; and
C

2007 The Authors
Journal compilation
C

Blackwell Publishing Ltd. 2007
CAPITAL STRUCTURE AND FIRM EFFICIENCY 1451
Myers, 1984) and taxation (DeAngelo and Masulis, 1980; and Bradley et al., 1984) are
likely to affect the value of the firm.
Although corporate capital structure theory lacks consensus, it is arguably the case
that financial distress costs, signaling and information costs all play some role in
determining a firm’s capital structure (see Barclay et al., 1995; and Myers, 2001). As
in Berger and Bonaccorsi di Patti (2006) we put forward two reasons explaining why
there may be a relationship between capital structureand firm performance. Under the
efficiency-risk hypothesis (H
2
), more efficient firms choose higher leverage ratios because
higher efficiency is expected to lower the costs of bankruptcy and financial distress.
Berger and Bonaccorsi di Patti (2006) note that more efficient firms are more likely to
earn a higher return for a given capital structure, and that higher returns can act as
a buffer against portfolio risk so that more efficient firms are in a better position
to substitute equity for debt in their capital structure. In effect, the efficiency-risk
hypothesis is a spin-off of the trade-off theory of capital structure whereby differences
in efficiency, other things constant, enable firms to alter their optimal capital structure

a notch up or down.
Yet even highly efficient firms face a trade-off between the advantages of more debt
and the costs of financial distress when they find that they have borrowed too much.
A value-maximizing firm would operate at the point where these expected benefits
and costs equate. This point will correspond to a higher debt to equity ratio for more
efficient firms equipped with assets that would escape serious damage in the event of
financial distress. However, it is also possible that firms which expect to sustain high
efficiency rates into the future will choose lower debt to equity ratios in an attempt
to guard the economic rents or franchise value generated by these efficiencies from
the threat of liquidation (see Demsetz et al., 1996; and Berger and Bonaccorsi di Patti,
2006). Thus in addition to the substitution effect, the relationship between efficiency
and capital structure may also be characterized by the presence of an income effect.
Under the franchise-value hypothesis (H
2A
) (see Berger and Bonaccorsi di Patti, 2006)
more efficient firms tend to hold extra equity capital and therefore, all else equal,
choose lower leverage ratios to protect their future income or franchise value.
Thus the efficiency-risk hypothesis (H
2
) and the franchise-value hypothesis (H
2A
) yield
opposite predictions regarding the likely effects of firm efficiency on its choice of
capital structure. Although we cannot identify the separate substitution and income
effects, our empirical analysis is able to determine which effect dominates the other
across the spectrum of different capital structure choices.
(ii) Benchmarking Firm Performance
In this section we develop the model that underpins the relationship between efficiency
and capital structure. First, we explain how we benchmark firm performance. To
do that we rely on duality theory and the use of distance functions. The difference

between maximum potential output and observed output while maintaining a given
level of input use, or actual and minimum potential input for given output, or some
combination of the two is attributed to technical or X-inefficiency. We interpret these
inefficiencies to be the result of contracting costs, managerial slack or oversight. They
differ from allocative inefficiencies which are due to the choice of a non-optimal mix
of inputs and outputs.
C

2007 The Authors
Journal compilation
C

Blackwell Publishing Ltd. 2007
1452 MARGARITIS AND PSILLAKI
Distance functions are alternative representations of production technology which
readily model multiple input and multiple output technological relationships. They
measure the maximum proportional expansion in outputs or contraction in inputs
that firms would be able to achieve by eliminating all technical inefficiency. They are
the primal measures; their dual measures are the more familiar value functions such
as profit, cost and revenue.
(a) The Production Structure
Following F¨are et al. (1985 and 1994) we assume that firms employ N inputs denoted
by x = (x
1
, , x
N
) ∈ R
N
+
to produce M outputs denoted by y = (y

1
, , y
M
) ∈ R
M
+
.
Technology may be characterised by a technology set T, which is the set of all feasible
input and output combinations, i.e., T = f(x, y):x can produce yg.
The technology is illustrated in Figure 1. The graph consistsof the input-output (x, y)
combinations that are bounded by the curved line and the x-axis. The technology set is
assumed to satisfy a set of reasonable axioms. Here we assume that T is a closed, convex,
nonempty set with inputs and outputs which are either freely or weakly disposable.
4
The Shephard output and input distance functions (see Shephard, 1970) are
defined, respectively, as:
D
o
(x, y) = min{θ :(x, y/θ) ∈ T} (1)
D
i
(y, x) = max{λ :(x/λ, y) ∈ T}. (2)
They scale in either the output direction by seeking the minimum value of θ to achieve
the maximum radial expansion of y so that y/θ remains feasible; or, separately, in the
Figure 1
The Distance Function and Technical Efficiency.
y
x
0
y

*
x
a
T
x
*
y
a
(
x
a
,
y
a
)
4 Input weak disposability means that if all inputs increase proportionally then output will not decrease.
This allows for the possibility that increases in some of the inputs may congest output; that is, the marginal
product for the specific input(s) may be negative. This is particularly useful for modelling production in the
presence of undesirable input use; for example, that may be caused by undesirable capital structure choices
should we decide to model those explicitly as part of the firm’s technology set.
C

2007 The Authors
Journal compilation
C

Blackwell Publishing Ltd. 2007
CAPITAL STRUCTURE AND FIRM EFFICIENCY 1453
input direction by seeking the maximum feasible contraction λ of the input vector.
5

In
Figure 1 the Shephard output distance function would project (x, y) due North onto
the boundary of T (yielding an efficiency measure of 0y
a
/0y
∗
); the Shephard input
distance function would project due West. If a firm is efficient the Shephard distance
function is equal to one; if a firm is inefficient D
i
will be greater than one (D
o
will be
less than one).
6
Note that we only require a weak form of optimality for measuring X-efficiency,
viz., radial (or hyperbolic) optimization. Revenue maximization and cost minimization
are stronger forms of optimization. Profit maximization entails an even stronger
optimization requirement (see F¨are and Grosskopf, 1996: p. 24). It implies cost
minimization given the profit maximizing choice of output(s); it also implies revenue
maximization given the optimal choice of inputs (but not the converse). In this regard,
the proposed methodology is general enough to encompass a variety of behavioural
assumptions and market structures.
(b) Data Envelopment Analysis (DEA)
The input or output distance functions may be estimated using linear programming
methods. In particular, Data Envelopment Analysis (DEA), a non-parametric mathe-
matical programming technique can be used to construct the empirical technological
or ‘best practice’ frontier and obtain the efficiency measures as distances from this
frontier. This can be implemented as follows:
Suppose we have k observations of inputs and outputs for k firms. From these we can

construct a reference technology under constant returns to scale (CRS) as:
T ={(x, y):
K
k=1
z
k
y
km
≥ y
m
, m = 1, ,M; 
K
k=1
z
k
x
kn
≤ x
n
, n = 1, ,N;
z
k
≥ 0, k = 1, K }.
(3)
The input distance function can be computed as:
D
i
(y, x) = min
λ,z
λ (4)

s.t.
K
k=1
z
k
y
km
≥ y
m
, m = 1, ,M; 
K
k=1
z
k
x
kn
≤ λx
n
, n = 1, ,N;
z
k
≥ 0, k = 1, K.
Alternatively, an output distance function D
o
can be computed by max (θ ) subject to a
similar set of constraints.
5 Alternatively, we could employ a hyperbolic efficiency measure, defined as D
h
(x, y) = minfλ :(λx, y/λ)
∈ Tg. This measure simultaneously contracts inputs and expands outputs proportionally. It does this along

a hyperbolic path to the frontier. Under constant returns to scale (CRS) the hyperbolic distance function
equals the square root of the Shephard output distance function or the square root of the reciprocal of the
Shephard input distance function. Also under CRS technical efficiency is dual to a profitability measure (see
F¨are and Grosskopf, 2004).
6 An important property of the distance functions is the homogeneity condition. The input and output
distance function D
i
and D
o
are homogeneous of degree +1 in inputs and outputs, respectively. Under CRS
the output distance function value is the reciprocal of the input distance function.
C

2007 The Authors
Journal compilation
C

Blackwell Publishing Ltd. 2007
1454 MARGARITIS AND PSILLAKI
The right-hand sides of the inequalities in (3) above represent all of the outputs
and inputs that are feasible given the observed inputs and outputs that are on the
left-hand side. In other words, the N +M inequality constraints restrict the technology
in that for a particular firm no more output can be produced using no less input than
a linear combination of all observed inputs and outputs. The z’s are usually referred
to as intensity variables and serve to construct convex combinations of the observed
data points. Alternatively, a variable returns to scale (VRS) technology which allows for
increasing, constant and decreasing returns to scale can be constructed by restricting
the z’s to add up to one. In this case maximum profits can be positive, negative, or zero.
Scale efficiency, a measure of the optimum scale of operations, can be estimated as
the ratio oftechnicalefficiencymeasuredunderCRS overtechnicalefficiencymeasured

under VRS. This ratio takes values between zero and one with a value of one indicating
scale efficiency. Values of less than one are either due to decreasing or increasing
returns to scale, with CRS defining the optimum scale.
7
3. THE EMPIRICAL MODEL
We use a two equation cross-section model to test the agency cost hypothesis (H
1
) and
the reverse causality hypotheses (H
2
and H
2A
).
(i) The Agency Cost Model
The regression equation for this model for each firm is given by:
D
i
= a
0
+ a
1
L
i
+ a
2
L
2
i
+ a
3

Z
1i
+ u
i
(5)
where D is the firm efficiency measure obtained from (4) above; L is the debt to total
assets ratio; Z
1
is a vector of control variables; and u is a zero mean error term.
According to the agency cost hypothesis the effect of leverage (L) on efficiency should
be positive, i.e. α
1
> 0, α
1
+2α
2
L > 0. However, the possibility exists that at sufficiently
high leverage levels, the effect of leverage on efficiency may be negative. The quadratic
specification in (5) is consistent with the possibility that the relationship between
leverage and efficiency may not be monotonic, viz. it may switch from positive to
negative at higher leverage. Leverage will have a negative effect on efficiency for values
of L < −α
1
/2α
2
. A sufficient condition for the inverse U -shaped relationship between
leverage and efficiency to hold is that α
2
< 0.
The variables included in Z

1
control for firm and market characteristics. More
specifically, we assume that risk, size, growth opportunities, market power and exposure
to international trade are likely to influence firm efficiency.
8
Firmrisk (SV) is measured
by the standard deviation of earnings before tax during the period 2000 to 2004. The
7 An estimate of allocative efficiency (AE) may be obtained under CRS by taking the ratio of observed
revenue over cost and dividing it by our measure of technical efficiency (see F¨are and Grosskopf, 2004). The
allocative efficiency measure may take values above or below one, with one indicating allocative efficiency.
8 Most of these variables are used as determinants of firm efficiency in previous studies – see for example,
Becchetti and Sierra (2003) and Berger and Bonaccorsi di Patti (2006). We would also have liked to control
for ownership structure but information on this variable is not available in the data provided by Statistics
New Zealand.
C

2007 The Authors
Journal compilation
C

Blackwell Publishing Ltd. 2007
CAPITAL STRUCTURE AND FIRM EFFICIENCY 1455
effect of this variable on firm efficiency is expected to be negative. Riskier firms tend
also to be those which are poorly organized (see Berger and Bonaccorsi di Patti, 2006).
Firm Size (SIZE) is measured by the logarithm of the firm’s sales. The effect of this
variable on efficiency is likely to be positive as larger firms are expected to use better
technology, be more diversified and better managed. A negative effect may be observed
in situations where there will be loss of control resulting from inefficient hierarchical
structures in the management of the company (see Wiliamson, 1967).
Intangible assets (OA) are measured by the ratio other assets to total assets.

9
This
variable may be considered as an indicator of future growth opportunities (see Titman
and Wessels, 1988; Michaelas et al., 1999; and Ozkan, 2001). We would generally expect
that companies with substantial intangible investment opportunities will tend to adopt
faster to better technology, be better managed and thereby be more efficient and more
profitable.
Market power is proxied by the concentration index (CI) representing the share
of the largest four firms in the industry. We would expect the effect of this variable
on efficiency to be negative, as competition forces firms to operate more efficiently.
It is however, arguable that higher concentration does not necessarily indicate lower
competition, but rather reflects market selection and consolidation through survival
of more efficient companies (see Demsetz, 1973). In this case, higher concentration
will have a positive effect on efficiency.
Exposure to international trade (TRADE) is measured by a dummy variable
indicating whether the firm belongs to the tradables or the non-tradables sector. The
hypothesis here is that firms which compete against imported goods or those in the
export sector should on average be more efficient than firms which operate in the
non-tradables sector.
(ii) The Leverage Model
The capital structure equation relates the debt to assets ratio to our measure of
efficiency as well as to a number of other factors that have commonly been identified
in the literature to be correlated with leverage (see Harris and Raviv, 1991; and Myers,
2001). The leverage equation is given by:
L
i
= β
0
+ β
1

D
i
+ β
2
Z
2i
+ v
i
(6)
where Z
2
is a vector of factors other than efficiency that correlate with leverage and v
is an error term. Under the efficiency-risk hypothesis, efficiency has a positive effect on
leverage, i.e. β
1
> 0; whereas under the franchise-value hypothesis, the effect of efficiency
on leverage is negative, i.e. β
1
< 0. We use quantile regression analysis to examine
the financing choices of different subsets of firms in terms of these two conditional
hypotheses. This is in line with Myers (2001) who emphasized that there is no universal
theory but several useful conditional theories describing the firm’s debt-equity choice.
These different theories will depend on which economic aspect and firm characteristic
we focus on.
The variables included in Z
2
control for firm characteristics such as size, asset
structure, profitability, risk and growth and for industry characteristics such as market
9 Statistics in New Zealand do not provide a break down of other assets into intangibles and investments.
C


2007 The Authors
Journal compilation
C

Blackwell Publishing Ltd. 2007
1456 MARGARITIS AND PSILLAKI
power that are likely to influence the choice of capital structure (see Harris and Raviv,
1991; and Rajan and Zingales, 1995). Firm size (SIZE) is measured by the logarithm of
the firm’s sales (see Titman and Wessels, 1988; Rajan and Zingales, 1995; and Ozkan,
2001). As larger firms are more diversified and tend to fail less often than smaller ones,
we would expect that size will be positively related to leverage. However Rajan and
Zingales (1995) discuss the possibility that size may also be negatively correlated with
leverage. They argue that size may act as a proxy for the information outside investors
have, and that informational asymmetries are lower for large firms which implies that
large firms should be in a better position to issue informationally sensitive securities
such as equity rather than debt.
Asset Tangibility (TA) is measured by the ratio of fixed tangible to total assets (see
Titman and Wessels, 1988; Rajan and Zingales, 1995; Frank and Goyal, 2003; and
Hall et al., 2004). The existence of asymmetric information and agency costs may
induce lenders to require guarantees materialized in collateral (Myers, 1977; Scott,
1977; and Harris and Raviv, 1990). For example, if a firm retains large investments in
land, equipment and other tangible assets, it will normally face smaller funding costs
compared to a firm that relies primarily on intangible assets. We would thus expect that
tangibility should be positively related to debt.
Profitability (PR) is measured by pre-interest and pre-tax operating surplus divided
by total assets (see Titman and Wessels, 1988; and Fama and French, 2002). There
are conflicting theoretical predictions on the effects of profitability on leverage (see
Harris and Raviv, 1991; Rajan and Zingales, 1995; Barclay and Smith, 2001; and Booth
et al., 2001). Myers (1984) and Myers and Majluf (1984) predict a negative relationship

because they argue that firms will prefer to finance new investments with internal
funds rather than debt. According to their pecking order theory, because of signalling
and asymmetric information problems, firms financing choices follow a hierarchy
in which internal cash flows (retained earnings) are preferred over external funds,
and debt is preferred over equity financing. Thus, they argue, we should expect a
negative relationship between past profitability and leverage. On the other hand,
using arguments based on the trade-off and contracting cost theories we can predict a
positive relation between profitability and leverage. For example, the trade-off theory
suggests that the optimal capital structure for any particular firmwill reflect the balance
(at the margin) between the tax shield benefits of debt and the increasing agency
and financial distress costs associated with high debt levels (Jensen and Meckling,
1976; Myers, 1977; and Harris and Raviv, 1990). Similarly, Jensen (1986) argues that
if the market for corporate control is effective and forces firms to pay out cash
by levering up, then there will be a positive correlation between profitability and
leverage.
Intangible assets (OA) can be considered as future growth opportunities (Titman
and Wessels 1988; Michaelas et al., 1999; and Ozkan, 2001). Following Myers (1977)
the underinvestment problem becomes more intense for companies with more growth
opportunities. The latter pushes creditors to reduce their supply of funds to this type
of firms. Firms with expected growth opportunities would keep low leverage in order
to avoid adverse selection and moral hazard costs associated with the financing of
new investments with new equity capital. In theory, and according to Myers’ assertion,
there should be a negative relationship between debt and growth opportunities. As
noted above, we use in this study a broader measure of ‘other assets’ which includes
intangibles, shares, mortgages and debentures.
C

2007 The Authors
Journal compilation
C


Blackwell Publishing Ltd. 2007
CAPITAL STRUCTURE AND FIRM EFFICIENCY 1457
Risk (SV) is measured by the standard deviation of annual earnings before taxes
(Castanias, 1983; and MacKie-Mason, 1990). Several authors have included a measure
of risk as a determinant of leverage (e.g., Jordan et al., 1998; Titman and Wessels, 1988;
MacKie-Mason, 1990; and Booth et al., 2001). A negative relation between risk and
leverage is expected from a pecking order theory perspective: firms with high volatility
on earnings try to accumulate cash to avoid underinvestment problems in the future.
From anagency costsor asymmetricinformationtheory perspectivewe expecta positive
relationship (see Ross, 1977; and Harris and Raviv, 1990).
Growth is measured as the annual percentage change on earnings. Growth is likely
to put a strain on retained earnings and push the firm into borrowing (Michaelas et al.,
1999). On the other hand, Myers (1977) argues that firms with growth potential will
tend to have lower leverage. He argues that growth opportunities can produce moral
hazard effects and can push firms to take more risk. This may explain why firms with
ample growth opportunities may be considered as risky and thus face difficulties in
raising debt capital on favorable terms.
Market power is proxied by the concentration index (CI). This index represents the
market share of the largest four firms in the industry. Berger and Bonaccorsi di Patti
(2006) argue that this variable captures the effect of expected future rents from local
market power on the firm’s capital structure decision so it is expected to be positively
associated with leverage.
4. EMPIRICAL RESULTS
In this section we provide answers to the questions in Section 1. As mentioned in
the Introduction we are interested in examining how capital structure choices affect
firm value as well as the reverse relationship between efficiency and leverage. More
precisely, we want to examine if leverage has a positive effect on efficiency and whether
the reverse effect of efficiency on leverage is similar across the spectrum of different
capital structures.

We use a sample of 12,240 New Zealand firms from the 2004 Annual Enterprise
Survey. New Zealand is an interesting case study for the purposes of this investigation.
Small and medium enterprises (SMEs) constitute the majority of all enterprises in
the country. The Ministry of Economic Development (MED) in New Zealand defines
SMEs as enterprises with 19 or fewer employees.
10
According to Devlin (1984) small
businesses in New Zealand share the following features:
r
they are personally owned and managed;
r
the owner/operator made most management decisions;
r
they do not have specialist staff at management level;
r
they are not part of a larger business or group of companies with access to
managerial expertise.
SMEs also play a major role in firm dynamics. Births and deaths among small firms have
increased by 142 and 126 percent, respectively, over the last decade. Thus, financial
distress and liquidation are particularly important issues for the firms in our sample.
Arnold et al. (2003) in a study of the New Zealand industry structure report that
New Zealand in comparison with other developed countries has the highest industry
10 Report of the Ministry of Economic Development (2005) on SMEs in New Zealand.
C

2007 The Authors
Journal compilation
C

Blackwell Publishing Ltd. 2007

1458 MARGARITIS AND PSILLAKI
concentration, the highest capital intensity across most industries, the highest total cost
to revenue (with smaller firms having a relatively higher ratio than larger firms) and
consequently significant diseconomies of scale.
Distance from trading partners may be relatively more important than the size of
the economy for industry structure. According to Caves et al. (1980) the small country
handicap has implications for allocative and technical efficiency. In small countries
markets can often only support a few firms in industries where scale is important;
either because incumbent firms can profitably raise prices above competitive levels to
earn excess profits; or because small markets can impede the ability of firms to achieve
minimum efficient scale (see Arnold et al., 2003). As a result, unit costs may be higher.
In addition, small markets affect the incentives to, and resources for, innovation and
appropriate investment which has implications for allocative and technical efficiency
over time.
Technical efficiency determines the level of cost at any level of output. It may be
affected by the smallness of an economy, but it is as likely to be affected by the barriers
to entry of that economy. While profit maximisation provides strong incentives for
technical efficiency, competition arguably strengthens them (see Arnold et al., 2003).
Companies with listed shares are pressured to be productively efficient by the threat
of takeover: the market for control. Inefficiencies may also be exacerbated by high
compliance costs imposed on businesses which contribute to high cost to revenue
ratios.
As explained in Section 2, we use non-parametric frontiermethods (DEA)to evaluate
the technical and scale efficiency of firms in our sample. The DEA model is constructed
using a single output (value-added) and two inputs (capital and labour) technology.
The labour input is measured by the total number of full-time equivalent employees
and working proprietors whereas capital is measured by the firm’s fixed tangible assets.
Firms that do not perform as well as the benchmark firms lie inside the frontier. We
estimate efficiency under both CRS and VRS technologies. Scale efficiency is measured
by the ratio of (optimum scale) CRS over VRS technical efficiency. We impose minimal

requirements about firm behaviour and market structure.
Table 1 presents balance sheet informationfor the entire New Zealand market sector
and for selected industries.
11
The high leverage industries are trade and construction
and the low leverage industries are mining and utilities.
Table 2 gives the descriptive statistics of the firms in the sample. On average the
long-term debt to total assets ratio is 23 percent which is lower than the aggregate
(market sector) average of 34 percent reported in Table 1. Also the tangible assets and
other assets ratios are considerably different from the aggregate averages of Table 1.
The reason for this discrepancy is that the finance and insurance sector is not included
in our sample of companies. As expected the sales distribution is highly skewed. Table
2 also presents descriptive statistics for the efficiency measures that we estimated using
DEA. The mean efficiency score for the firms in the sample is 56.5 percent if we allow
simultaneous scaling of inputs and output; 33.5 percent if we assume a CRS technology
with scaling in either the output or the input direction; and 62.1 percent for a VRS
technology which scales in the input direction only. The CRS efficiency score average
11 Dairy farming along with livestock and horticulture were not part of the sample of firms as data on these
firms is not consistently available in the AES survey. Financial firms, Government Administration, Education,
Health and Community Services were also excluded from the sample.
C

2007 The Authors
Journal compilation
C

Blackwell Publishing Ltd. 2007
CAPITAL STRUCTURE AND FIRM EFFICIENCY 1459
Table 1
Financial Ratios (Percentage of total assets)

Agriculture, Transport &
All Industries Forestry & Fish Mining Manufacturing Communication
2003 2004 2003 2004 2003 2004 2003 2004 2003 2004
Assets
Current Assets
∗
26.7 23.3 18.6 18.8 – 43.7 43.6 30.6 33.1
Tangible Assets 26.4 26.0 51.3 54.0 43.3 49.1 36.2 34.9 49.4 44.7
Other Assets
∗
47.0 50.7 30.1 27.2 – – 20.1 21.6 20.0 22.1
Equity and Liabilities
Owners Equity 38.6 39.0 53.4 52.2 71.6 70.3 45.6 45.7 43.6 45.7
Current Liabilities 29.2 26.9 10.0 9.9 13.0 13.8 30.6 33.5 30.5 28.2
Long-term Liabilities 32.2 34.0 36.6 37.9 15.4 15.8 23.8 20.9 25.9 26.1
Utilities Construction Hotels & Restaurants Wholesale Trade Retail Services
2003 2004 2003 2004 2003 2004 2003 2004 2003 2004
Assets
Current Assets
∗
12.1 10.1 – – 18.7 20.0 74.7 74.5 58.7 57.0
Tangible Assets 73.8 79.8 30.8 31.4 66.4 63.3 14.0 14.6 27.2 25.3
Other Assets
∗
14.0 10.1 – – 14.9 16.7 11.3 10.9 14.1 17.7
Equity and Liabilities
Owners Equity 54.5 58.1 34.9 35.4 46.4 45.7 36.1 38.8 36.3 35.5
Current Liabilities 21.0 13.2 45.5 44.6 26.0 28.4 48.7 46.7 38.5 43.0
Long-term Liabilities 24.5 28.8 19.6 20.0 27.6 25.9 15.3 14.5 25.2 21.5
Notes:

∗
Current Assets include cash and deposits, debtors, short-term bills and bonds, stocks and inventories. Other Assets include investments (such as shares, mortgages,
debentures etc.) and intangible assets (such as goodwill and brands).
C

2007 The Authors
Journal compilation
C

Blackwell Publishing Ltd. 2007
1460 MARGARITIS AND PSILLAKI
Table 2
2004 Descriptive Statistics
Mean Median St. Dev.
Long-term Debt
∗
0.229 0.208 0.237
Short-term Debt
∗
0.261 0.188 0.227
Tangibles
∗
0.482 0.502 0.273
Other Assets
∗
0.143 0.068 0.195
Profit (BTI)
∗
0.275 0.256 0.617
Sales

∗
3.074 1.955 5.017
Profit Margin
∗
1.505 1.272 0.731
Hyperbolic Technical Efficiency 0.565 0.529 0.125
Allocative Efficiency 0.483 0.374 0.359
CRS Technical Efficiency 0.335 0.280 0.168
VRS Technical Efficiency 0.621 0.500 0.288
Scale Efficiency 0.620 0.608 0.263
Notes:
∗
Ratio over total assets.
BTI = before tax and interest; Profit margin = ratio of revenue over cost.
Full efficiency = 1.
seems to be very low but it may not come as a surprise for a small distant economy
for reasons that we have already explained.
12
Average scale efficiency is estimated at
62 percent. According to our estimates this is due to increasing returns to scale for
the majority of firms in our sample. The average score for allocative efficiency is 48.3
percent.
We turn next to empirically assess the role of leverage on efficiency as well as to
investigate whether differences in efficiency are related to leverage. The simultaneous
equation system given by (5) and (6) above requires adequate structure to be properly
identified. An obvious way to deal with the identification problem is by imposing
relevant restrictions on the structural system. For example, we may impose at least
one exclusion restriction on each of the two Z vectors and proceed to estimate the
two equation system via an instrumental variables method. These restrictions should
be theoretically consistent in that the variable(s) excluded from Z

2
but included in Z
1
should in some meaningful way affect efficiency but not the choice of leverage. For
example, it may be argued that exposure to international trade may affect efficiency
but not the choice of leverage. Therefore, we can use this restriction to identify the
capital structure equation.
A similar requirement will apply for the variable(s) excluded from Z
1
. It may be
argued – albeit less convincingly – that fixed tangible assets do not have an effect
on efficiency. This condition will help identify the efficiency equation. Additionally,
we will need to verify that the Z variables included in the two equations are indeed
exogenous. Undoubtedly, the task of both properly identifying the system of equations
for efficiency and leverage, and ensuring that the conditioning variables entering these
two equations are indeed exogenous, is fraught with difficulty.
12 The DEA efficiency measures were constructedfor groups of firms that belonged to industries with largely
similar technological characteristics; this avoids pooling together firms with distinctly different technologies
to construct an overall best practice frontier. We did not account for the effect of corporate structure in the
DEA measures as this would have biased the regression results in Table 4. Efficiency scores were on average
very low for firms in the traditional services (wholesale; retail; accommodation and restaurants) industries.
C

2007 The Authors
Journal compilation
C

Blackwell Publishing Ltd. 2007
CAPITAL STRUCTURE AND FIRM EFFICIENCY 1461
There are two more issues worth considering. Both have ramifications for the

specification of the agency cost - leverage model. As we shall explain, they also provide a
more effective way to deal with the identification and simultaneity bias problems. First,
the various theories on capital structure focus either on the levels of debt and equity
(e.g., contracting-cost theories) or on the flows of new debt and equity issues (e.g., the
pecking order theory). Since both flows and stocks are expected to play an important
role in financing decisions, neither of these two theoretical approaches provide on
their own accord a complete guide to optimal capital structure. We would expect that
firms will adjust their capital structure by balancing the costs of adjustment with the
benefits of operating at or near their leverage target level. As short-term debt or bank
loans incur the least transactions and information costs, it is likely that smaller bank
financed firms will spend considerable time away from their target capital structures
(see Barclay and Smith, 2001).
Secondly, we would expect that both the effect of leverage on efficiency and the
reverse effect from efficiency on leverage will not be instantaneous. Time lags are
also likely to prevail when considering the effect of other conditioning variables on
efficiency and leverage. For example, the pecking order theory explicitly states that it
is past, not current profitability, that is expected to have an effect on leverage.
An explicit account of the dynamics in the relationship between efficiency and
leverage would thus help solve the identification problem while rendering a structure
which is not prone to simultaneity bias problems. Based on this we have proceeded to
estimate a variety of static and dynamic models.
13
They include models that emphasize
the importance of flows (viz., specifying the leverage equation in first differences),
adjustments towards an optimal capital structure (i.e., combining a dynamic stock
adjustment model with a target leverage equation) as well as exploring more general
forms of dynamic model specifications. We have estimated the structural forms using
instrumental variables techniques and their dynamic or reduced form specifications
using OLS or quantile regressions. The results we have obtained from the different
models or estimation techniques appear to be quite robust, in particular those used

to assess the predictions of the agency-cost and efficiency hypotheses (H
1
and H
2
).
We only report the results obtained from estimating dynamic models for both the
efficiency and leverage equations using a VRS efficiency measure.
14
The regressors
in these equations are predetermined (lagged endogenous or exogenous) variables
thereby circumventing the simultaneous bias problem. Parsimonious forms of these
equations were obtained by applying a standard general to specific methodology (see
Hendry, 1995) starting with models that used variables with up to four year lags.
Table 3 reports the estimates of the efficiency equation. The results show that both
the linear and quadratic leverage terms have a significant effect on efficiency. This
effect is positive at the mean of leverage as well as it remains positive over the entire
relevant range of leverage values. Thus, we find support for the predictions of the agency
13 Given the limited number of time periods for which data is available, we have opted to estimate cross-
section not panel models. This ensures sufficient dynamic conditioning of the agency cost and leverage
equations and also allows us to compute measures of risk and growth using lagged values of these variables.
In addition, it would have been difficult to apply quantile regression methods to panel data as quantiles of
convolutions of random variables are highly intractable objects (see Koenker and Hallock, 2001).
14 Table A1 in the Appendix reports 2SLS results for the leverage equation in levels and first difference
form. The instruments are the exogenous variables (Z) and lagged values of the efficiency, leverage and
profitability variables.
C

2007 The Authors
Journal compilation
C


Blackwell Publishing Ltd. 2007
1462 MARGARITIS AND PSILLAKI
Table 3
The Determinants of Firm Efficiency
Variable Coefficient Estimate t-Ratio p-value Adj R-Sq
Intercept 0.8071 23.18 <0.0001 0.204
OA 0.2652 8.42 <0.0001
S −0.0468 −9.09 <0.0001
L0.0589 4.00 <0.0001
LSQ −0.0038 −2.85 0.0044
SV 0.0000 −1.51 0.1318
CI 0.9781 14.13 <0.0001
TRADE 0.0551 5.87 <0.0001
Intercept 0.7589 16.17 <0.0001 0.211
TA 0.2904 3.53 0.0004
TASQ −0.3312 −4.44 <0.0001
OA 0.2040 5.57 <0.0001
S −0.0441 −8.19
<0.0001
L0.0711 4.77 <0.0001
LSQ −0.0045 −3.35 0.0008
SV 0.0000 −1.63 0.1042
CI 0.9516 13.75 <0.0001
TRADE 0.0495 5.23 <0.0001
Notes:
TA = tangibles to total assets ratio; OA = intangibles & other assets to total assets ratio; S = (log) size; L =
leverage; SV = risk; CI = concentration index; TRADE = dummy variable which takes on the value of one
if the firm belongs to the tradables sector and zero, otherwise; SQ indicates the square of the variable. All
right hand side variables aside from TRADE enter with a three period lag.

cost hypothesis in that higher leverage is associated with improved efficiency. We find
that industry concentration is positively associated with efficiency. Our interpretation
of this result is that higher concentration reflects the benefits of market selection
and consolidation on efficiency rather than lack of competition. We also find that
firms which are exposed to international trade or those with more intangible assets
appear to be more efficient. Firms that operate in the tradables sector and those with
substantial intangible investment opportunities are in general expected to adopt better
technologies and are able to employ better managers, thereby achieving improved
efficiencies relative to their peers.
The negative effect of size on efficiency may reflect the loss of control resulting from
inefficient hierarchical structures in the management of the company. As Berger and
Bonaccorsi di Patti (2006) point out, the negative effect of size on efficiency should be
interpreted within the partial regression context. This effect is conditioned on leverage
and the other regressors. Larger firms may be more efficient than they appear to be
because they may be more leveraged. This may in turn improve their efficiency for
reasons that we have already explained. The effect of risk on efficiency is negative but
not statistically significant.
In Table 4 we report least squares (OLS) as well as quantile regression results
for the leverage equation. We use quantile regression to account for asymmetries in
the distribution of the dependent variable (the leverage distribution exhibits excess
C

2007 The Authors
Journal compilation
C

Blackwell Publishing Ltd. 2007
CAPITAL STRUCTURE AND FIRM EFFICIENCY 1463
Table 4
The Determinants of Firm Leverage

OLS Quantile Regression
Coeff. t-ratio p-value Coeff. t-ratio p-value
tau = 0.5
Intercept 0.532 6.280 <0.0001 Intercept 0.226 31.691 0.000
TA(-2) 0.174 3.480 0.001 TA(-2) 0.114 14.788 0.000
PR(-2) 0.161 4.630 <0.0001 PR(-2) −0.032 −0.755 0.450
OA(-1) −0.218 −3.140 0.002 OA(-1) −0.154 −30.455 0.000
S −0.023 −2.090 0.036 S 0.006 4.941 0.000
D(-1) −0.131 −2.590 0.010 D(-1) 0.023 6.472 0.000
CI −0.386 −3.380 0.001 CI 0.036 1.664 0.096
R-square 0.210 R-square 0.244
Quantile Regression
tau = 0.1 tau = 0.6
Intercept 0.218 8.291 0.000 Intercept 0.261 12.636 0.000
TA(-2) 0.138 8.178 0.000 TA(-2) 0.087 6.089 0.000
PR(-2) 0.102 0.848 0.397 PR(-2) 0.091 0.898 0.369
OA(-1) 0.100 5.540 0.000 OA(-1) −0.190 −13.221 0.000
S
−0.029 −7.495 0.000 S 0.009 3.362 0.001
D(-1) 0.056 3.540 0.000 D(-1) 0.010 1.169 0.243
CI −0.058 −0.870 0.384 CI 0.026 0.635 0.526
R-square 0.106 R-square 0.288
tau = 0.2 tau = 0.7
Intercept 0.248 11.509 0.000 Intercept 0.246 7.562 0.000
TA(-2) 0.116 7.525 0.000 TA(-2) 0.110 4.889 0.000
PR(-2) 0.122 5.165 0.000 PR(-2) 0.502 3.901 0.000
OA(-1) −0.018 −0.916 0.360 OA(-1) −0.190 −6.341 0.000
S −0.026 −9.421 0.000 S 0.017 3.544 0.000
D(-1) 0.060 5.018 0.000 D(-1) −0.021 −1.692 0.091
CI 0.140 3.199 0.001 CI 0.003 0.052 0.958

R-square 0.159 R-square 0.374
tau = 0.3 tau = 0.8
Intercept 0.248 16.597 0.000 Intercept 0.300 3.601 0.000
TA(-2) 0.151 14.084 0.000 TA(-2) 0.102 1.693 0.091
PR(-2) 0.073 2.645 0.008 PR(-2) 0.841 3.228 0.001
OA(-1) −0.094 −7.416 0.000 OA(-1) −0.198 −2.480 0.013
S −0.017 −6.793 0.000 S 0.042 3.379 0.001
D(-1) 0.039 5.716 0.000 D(-1) −0.141 −4.070 0.000
CI 0.086 3.176 0.002 CI −0.113 −0.804 0.422
R-square 0.201 R-square 0.489
tau = 0.4 tau = 0.9
Intercept 0.215 29.768 0.000 Intercept 0.476 3.084 0.002
TA(-2) 0.151 20.642 0.000 TA(-2) 0.168 1.607 0.108
PR(-2) 0.008 0.322 0.748 PR(-2) 0.978 2.567 0.010
OA(-1) −0.095 −7.308 0.000 OA(-1) −0.165 −1.016 0.310
S −0.003 −2.620 0.009 S 0.065 3.727 0.000
D(-1) 0.038 14.846 0.000 D(-1) −0.343 −5.138 0.000
CI 0.043 3
.837 0.000 CI −0.315 −0.955 0.340
R-square 0.223 R-square 0.613
Notes:
TA = tangibles to assets ratio; PR = profits to assets ratio; OA = intangibles & other assets to total assets
ratio; S = (log) size; D = efficiency score; CI = concentration index. All right hand side variables aside
from firm size, profits and CI are lagged one or two periods as indicated in brackets. Profits enter with one
and two periods lags so the reported coefficients are the sum of the coefficients of the two lags. Tau values
indicate the τ -th sample quantile.
C

2007 The Authors
Journal compilation

C

Blackwell Publishing Ltd. 2007
1464 MARGARITIS AND PSILLAKI
kurtosis and is skewed to the right).
15
In particular we are interested in testing the
predictions of the efficiency-risk hypothesis and the franchise-value hypothesis across the
spectrum of different leverage levels. An advantage of quantile regressions is that there
is no need to segment the sample in accordance with the unconditional distribution of
the dependent variable and run separate regressions on each segment.
16
We find no
evidence to suggest that industry effects captured by individual industry dummies or a
set of consolidated dummies for low, medium and high leverage sectors are significant
determinants of leverage in our sample so we do not include any dummy variables.
Similarly, we find no evidence to suggest that risk (aside from the debt flow equation
reported in Table A1) or growth, play a significant role in determining capital structure
choices for firms once we control for efficiency and the other factors that affect this
choice. Therefore, these variables are also omitted from the regressions we report in
Table 4. We specify the leverage equation as a function of fixed tangible assets to total
assets ratio with a two period lag; the ratio of other assets to total assets lagged one
period; profits (operating surplus before tax and interest) to total assets ratio with
one and two period lags; (log) sales; the concentration ratio; and efficiency measured
by the distance from the empirically constructed production frontier with one
period lag.
There are some notable differences between the conditional mean (least squares)
and conditional median (τ = 0.5) regression results arising partly from the asymmetry
of the conditional leverage distribution and also attributed to the strong effect exerted
on the least squares estimates by high-leverage data points. There are also interesting

patterns that arise when we consider the effect of the different factors on leverage at
other quantiles. The OLS results indicate a significant negative relationship between
efficiency and leverage thereby providing support for the franchise-value hypothesis over
the efficiency-risk hypothesis. Fixed assets are positively related to leverage which is to be
expected as these assets serve as a proxy for collateral. Profitability is also positively
related to leverage which is contrary to the predictions of the pecking order theory
but consistent with the trade-off theory of capital structure. According to this theory
more profitable firms are able to move their optimal debt to equity ratio up (at the
margin) because higher profitability lowers, other things constant, the expected costs
of bankruptcy and financial distress associated with higher debt levels. In addition, if
past profitability is a good proxy for future profitability, profitable firms will be less
likely to be subject to rationing and other supply constraints.
We also find in the OLS results that intangibles and other assets are negatively
correlated with leverage which is consistent with the Myers’ (1977) view that firms with
expected growth opportunities would maintain low leverage levels in order to avoid
adverse selection. Size proxied by (log) sales has a negative relationship with leverage.
An explanation for the negative effect of size on leverage is provided by Rajan and
15 The quantile regression estimator is obtained by min
β

n
i=1
ρ
τ
(y
i
− x

i
β) where ρ

τ
(z) = z*(τ − 1(z ≤0))
is a parametric function that yields the τ-th sample quantile as its solution (see Koenker and Bassett, 1978).
The median regression which is estimated by the least absolute deviations (LAD) estimator is the solution of
the quantile regression when τ = 0.5. Mean or median regressions minimise sums of symmetric squared or
absolute residuals; other quantile regressions involve the minimisation of sums of asymmetrically weighted
absolute residuals. For example, by setting τ = 0.9, the ρ
τ
function will weigh positive residuals much more
heavily than negative residuals. We carried out quantile regression estimation and inference using Koenker’s
quantile regression software package ‘quantreg’ in R.
16 Heckman (1979) has expressed some serious misgivings about the practice of ‘truncating’the dependent
variable and running a series of segmented regressions.
C

2007 The Authors
Journal compilation
C

Blackwell Publishing Ltd. 2007
CAPITAL STRUCTURE AND FIRM EFFICIENCY 1465
Zingales (1995, p. 1457) on the basis of informational asymmetries as explained above.
The effect of industry concentration on leverage is negative. As Berger and Bonaccorsi
di Patti (2006, p. 1089) suggest this effect is ‘consistent with the influence of expected
future rents from local market power on the capital structure decision.’
The quantile regression results of Table 4 show that the effect of efficiency on
leverage is positive and significant in the low to medium range of the leverage
distribution thereby supportingthe predictions ofthe efficiency-risk hypothesis. This result
suggests that more efficient firms which do not carry very high levels of debt in their
capital structure can choose, other things constant, higher debt ratios because higher

efficiency lowers the expected costs of bankruptcyand financialdistress. Thesefirmsare
also considered from lenders as more solvent and consequently they can be expected
to have better access to debt finance than their less efficient peers. Our results also
show that at higher leverage levels the income effect of higher rents dominates the
substitution effect of higher efficiency over equity capital thereby providing support for
the predictions of the franchise-value hypothesis. Thus more efficient but highly leveraged
firms are expected to choose lower debt levels to protect the rents or franchise value
associated with higher efficiency from the prospect of liquidation.
17
The effect of tangibles appears to be consistently positive and significant throughout
the entire range of the leverage distribution. Profitability appears to have a positive
and significant effect on leverage for firms at low and at high leverage quantiles.
Intangibles and other assets have a negative effect on leverage aside from a positive
effect we estimate at the lowest quantile.
18
To the extent that intangibles such as real
options or patents, brands and goodwill represent franchise value and franchise value
increases the costs of financial distress we would expect that firms with substantial
franchise value will have an incentive to hold less debt and more equity in their
capital structure (see Titman, 1984; and Titman and Wessels, 1988). This variable
may thus be interpreted as controlling for the effect of franchise value on leverage
that is not already fully captured by efficiency. The concentration index is positively
associated with leverage for low debt firms.This effect is not significantfor more levered
firms.
We find the effect of firm size on leverage to be positive for firms in the middle
to upper half of the leverage distribution consistent with the view that size may be a
proxy for the (inverse) probability of default. As larger firms are generally expected to
be better diversified they can carry, all else equal, less capital relative to debt in their
capital structure to safeguard against the expected costs of bankruptcy or financial
distress.

19
We also find that the effect of firm size on leverage is negative at low debt
ratios. In this case firms are likely to face lower costs of financial distress so that we
would not expect to find a strong positive relationship between size and leverage. As
17 An obvious consequence of the non-robustness of the least squares fit is that it provides a rather poor
estimate of the conditional mean for low leverage firms in the sample.
18 Intangibles are not generally considered to be good collateral for loans so they are not likely to be
positively associated with leverage; they may proxy instead asymmetric information costs or represent future
growth opportunities (real options) and in this sense may be expected to be negatively related to leverage
(see Bartholdy and Mateus, 2005).
19 New Zealand is a British law country with strict insolvency legislation. There is currently no provision for
businesses to be protected under bankruptcy provisions (as for example in the US) so firms in distress face
a great danger of immediate liquidation. There is also very little protection for creditors although there are
statutory procedures in place whereby insolvent business can enter into agreements with their creditors as
an alternative to liquidation/bankruptcy (Law Commission, 2001, p. 69).
C

2007 The Authors
Journal compilation
C

Blackwell Publishing Ltd. 2007
1466 MARGARITIS AND PSILLAKI
Rajan and Zingales (1995) argue the effect of size on leverage may be dominated
by informational asymmetries between insiders in a firm and the capital markets. We
surmise that informational asymmetries dominate default considerations at low debt
levels thereby yielding a negative relationship between firm size and leverage.
20
5. CONCLUSION
This paperinvestigates therelationship betweenefficiency and leverage. Using asample

of 12,240 firms from the 2004 New Zealand Annual Enterprise Survey we consider both
the effect of leverage on firm performance as well as the reverse causality relationship.
We find evidence supporting the theoretical predictions of the Jensen and Meckling
(1976) agency costmodel. More precisely, wefind support for the core prediction of the
agency cost hypothesis in that higher leverage is associated with improved efficiency over
the entire range of observed data. This analysis is conducted using distance functions
to model the technology and obtain X-efficiency measures as the distance from the
efficient frontier. We interpret these measures as a proxy for the (inverse) agency
costs arising from conflicts between debt holders and equity holders or from different
principal-agent objectives.
We also investigate the reverse causality relationship from efficiency to leverage by
putting forth two competing hypotheses: the efficiency-risk hypothesis and the franchise-
value hypothesis. Using quantile regression analysis we show that the effect of efficiency
on leverage is positive at low to mid-leverage levels and negative at high leverage
ratios. Thus our results suggest that in the upper range of the leverage distribution
the income effect resulting from the economic rents generated by high efficiency
dominates the substitution effect of debt for equity capital thereby rendering support
to the franchise-value hypothesis. We surmise that the managers of leveraged but more
efficient companies have an incentive to limit risk exposure and thereby choose lower
debt toequity ratios relativeto their less efficientpeers toprotect the ‘rents’ or franchise
value generated by these efficiencies from the danger of liquidation. The choice of a
lower leverage ratio should be interpreted in this case as a signal to outside investors of
the company’s strength rather than weakness. On the other hand, our results support
the predictions of the efficiency-risk hypothesis for firms in the low to middle range of the
leverage distribution. In this case more efficient firms will choose higher debt ratios
because higher efficiency acts as a buffer against the expected costs of bankruptcy or
financial distress.
The effectoftangibles and profitabilityonleverage ispositive.Firmsizehasa negative
effect on leverage at the lower half of the leverage distribution and a positive effect at
the upper half of the distribution. The effect of intangibles and other assets is estimated

to be negative. The concentration index is positively associated with leverage for low
leverage firms. This effect is not significant for firms with higher leverage ratios.
We find our results are fairly robust to different model specifications. An important
contribution of this paper is in the use of productive efficiency as a measure of firm
performance to test the predictions of the agency cost hypothesis and the reverse
relationship from performance to capital structure. Our methodology has gone some
20 Rajan and Zingales (1995) and Kremp et al. (1999) report a negative overall relationship between size
and leverage for Germany. These authors argue that the negative relationship is not due to asymmetric
information, but rather reflects the characteristics of the German bankruptcy law and the Hausbank system,
which offer good protection to creditors.
C

2007 The Authors
Journal compilation
C

Blackwell Publishing Ltd. 2007
CAPITAL STRUCTURE AND FIRM EFFICIENCY 1467
way in reconciling some of the empirical irregularities reported in prior studies. In
particular, we have shown how competing hypotheses may dominate each other at
different segments of the relevant data distribution thereby cautioning the standard
practice of drawing inferences on capital structure choices using conditional mean
(least squares) estimates. It will be useful to consider in future research the relevance
of the proposed methodology in the study of the efficiency and capital structure link
in cross country comparisons thereby subjecting our findings to further scrutiny. This
will also help to assess how different economic and/or institutional characteristics may
impact on capital structure decisions and how in turn these choices may affect firm
performance.
APPENDIX
Table A1

The Determinants of Firm Leverage
2SLS Dependent Variable L 2SLS Dependent Variable L
Coeff. t-ratio p-value Coeff. t-ratio p-value
Intercept −0.259 −9.03 <0.0001 Intercept 0.839 3.99 < 0.0001
TA 0.042 4.92 <0.0001 TA 0.140 2.47 0.014
PR −0.028 −13.02 <0.0001 PR 0.286 18.11 < 0.0001
OA 0.041 3.59 <0.0001 OA −0.142 −1.95 0.052
S 0.019 8.24 <0.0001 S −0.047 −2.88 0.004
D 0.328 9.92 <0.0001 D −0.588 −2.59 0.009
CI −0.136 −7.54 <0.0001 CI 0.129 0.99 0.325
SV −0.002 −3.54 <0.0001
R-square 0.017 R-square 0.198
Notes:
 = First difference operator; L = leverage; TA = tangibles; PR = profits; OA = intangibles and other
assets; S = (log) size; D = efficiency; CI = concentration index; SV = risk.
REFERENCES
Arnold, T., D. Boles de Boer and L. Evans (2003), ‘The Structure of New Zealand Industry:
Its Implications for Competition Law’, in M. Barry and L. Evans (eds.), New Zealand
Competition Law at the Year 2000 (Wellington: Victoria University of Wellington).
Barclay, M.J. and C.W. Smith, Jr. (2001), ‘The Capital Structure Puzzle: Another Look at the
Evidence’, in D.H. Chew (ed.), The New Corporate Finance: Where Theory Meets Practice (New
York: McGraw-Hill).
——— ——— and R.L. Watts (1995), ‘The Determinants of Corporate Leverage and Dividend
Policies’, Journal of Applied Corporate Finance, Vol. 7, No. 4, pp. 4–19.
Bartholdy, J. and C. Mateus (2005), ‘Debt and Taxes: Evidence from Bank-Financed Small
and Medium-Sized Firms’, Paper Presented at the 12th Annual Conference of the
Multinational Finance Society (Athens, Greece).
Beattie, V., A. Goodacre and S.J. Thomson (2006), ‘Corporate Financing Decisions: UK Survey
Evidence’, Journal of Business Finance & Accounting, Vol. 33, Nos. 9&10, pp. 1402–34.
Becchetti, L. and J. Sierra (2003), ‘Bankruptcy Risk and Productive Efficiency in Manufactur-

ing Firms’, Journal of Banking and Finance, Vol. 27, pp. 2099–120.
Berger, A. N. and E. Bonaccorsi di Patti (2006), ‘Capital Structure and Firm Performance: A
New Approach to Testing Agency Theory and an Application to the Banking Industry’,
Journal of Banking and Finance, Vol. 30, pp. 1065–102.
C

2007 The Authors
Journal compilation
C

Blackwell Publishing Ltd. 2007
1468 MARGARITIS AND PSILLAKI
Booth, L., V. Aivazian, A. Demirguc-Kunt and V. Maksimovic (2001), ‘Capital Structure in
Developing Countries’, Journal of Finance, Vol. 56, pp. 87–130.
Bradley, M., G. Jarrell and E. Kim (1984), ‘On the Existence of an Optimal Capital Structure:
Theory and Evidence’, Journal of Finance, Vol. 39, pp. 857–78.
Castanias, R. (1983), ‘Bankruptcy Risk and Optimal Capital Structure’, Journal of Finance, Vol.
38, No. 5 (December), pp. 1617–35.
Caves, R.E., M. Porter and A.M. Spence (1980), Competition in an Open Economy: A Model Applied
to Canada (Cambridge, MA: Harvard University Press).
Chew, D.H. (2001), The New Corporate Finance: Where Theory Meets Practice (New York: McGraw-
Hill).
DeAngelo, H. and R. Masulis (1980), ‘Optimal Capital Structure Under Corporate and
Personal Taxation’, Journal of Financial Economics, Vol. 8, No. 1, pp. 3–29.
Demsetz, H. (1973), ‘Industry Structure, Market Rivalry, and Public Policy’, Journal of Law and
Economics, pp. 1–9.
Demsetz, R., M. Saidenberg and P. Strahan (1996), ‘Banks with Something to Loose: The
Disciplinary Role of Franchise Value’, Federal Reserve Bank of New York Economic Policy Review
(October), pp. 1–14.
Devlin, M. (1984), The Small Business Sector in New Zealand: An Introductory Perspective, Small

Business Research Series Paper No. 1 (Washington: Development Finance Corporation).
Fama, E. and K. French (2002), ‘Testing Tradeoff and Pecking Order Predictions About
Dividends and Debt’, Review of Financial Studies, Vol. 15, pp. 1–33.
F¨are, R. and S. Grosskopf (1996) Intertemporal Production Frontiers: With Dynamic DEA (Boston:
Kluwer Academic Publishers).
——— ——— (2004), New Directions: Efficiency and Productivity (Boston: Kluwer Academic
Publishers).
——— ——— and C.A.K. Lovell (1985), The Measurement of Efficiency of Production (Boston:
Kluwer-Nijhoff).
——— ——— ——— (1994), Production Frontiers (Cambridge: Cambridge University Press).
——— ——— and D. Margaritis (2007), ‘Efficiency and Productivity: Malmquist and More’, in
H. O. Fried, C.A.K. Lovell and S.S. Schmidt (eds.), The Measurement of Productive Efficiency
and Productivity Growth (New York: Oxford University Press).
Frank, M. and V. Goyal (2003), ‘Testing the Pecking Order Theory of Capital Structure’,
Journal of Financial Economics, Vol. 67, pp. 217–48.
Fried, H.O., C.A.K. Lovell and S.S. Schmidt (2007), ‘Efficiency and Productivity’, in H. O.
Fried, C.A.K. Lovell and S.S. Schmidt (eds.), The Measurement of Productive Efficiency and
Productivity Growth (New York: Oxford University Press).
Grossman, S. J. and O. Hart(1982),‘CorporateFinancialStructure and Managerial Incentives’,
in J. McCall (ed.), The Economics of Information and Uncertainty (Chicago: University of
Chicago Press).
Hall, G.C., P.J. Hutchinson and N. Michaelas (2004), ‘Determinants of the Capital Struc-
tures of European SMEs’, Journal of Business Finance & Accounting, Vol. 31, No. 5,
pp. 711–28.
Harris, M. and A. Raviv (1988), ‘Corporate Control Contests and Capital Structure’, Journal of
Financial Economics, Vol. 20, pp. 55–86.
——— ——— (1990), ‘Capital Structure and the Informational Role of Debt’, Journal of
Finance, Vol. 45, No. 2 (June), pp. 321–49.
——— ——— (1991), ‘The Theory of Capital Structure’, Journal of Finance, Vol. 46,
pp. 297–355.

Heckman, J. (1979), ‘Sample Selection Bias as a Specification Error’, Econometrica, Vol. 47,
No. 1, pp. 153–61.
Hendry, D.F. (1995), Dynamic Econometrics (Oxford: Oxford University Press).
Jensen, M. (1986), ‘Agency Costs of Free Cash Flow, Corporate Finance, and Takeovers’,
American Economic Review, Vol. 76, pp. 323–29.
——— and W. Meckling (1976), ‘Theory of the Firm: Managerial Behavior, Agency Costs and
Capital Structure’, Journal of Financial Economics, Vol. 3, pp. 305–60.
Jordan, J., J. Lowe and P. Taylor (1998), ‘Strategy and Financial Policy in UK Small Firms’,
Journal of Business Finance & Accounting, Vol. 25, pp. 1–27.
C

2007 The Authors
Journal compilation
C

Blackwell Publishing Ltd. 2007
CAPITAL STRUCTURE AND FIRM EFFICIENCY 1469
Keasey, K. and R. Watson (1987), ‘Non-Financial Symptoms and the Prediction of Small
Company Failure: A Test of Argenti’s Hypotheses’, Journal of Business Finance& Accounting,
Vol. 17, No. 1, pp. 119–35.
Koenker, R. and G. Bassett (1978), ‘Regression Quantiles’, Econometrica, Vol. 46, pp. 107–12.
——— and K.F. Hallock (2001), ‘Quantile Regression: An Introduction’, Journal of Economic
Perspectives, Vol. 15, pp. 143–56.
Kremp, E., E. St¨oss and D. Gerdesmeir (1999), ‘Estimation of a Debt Function: Evidence from
French and German Panel Data’, in A. Sauv´e and M. Scheur (eds.), Corporate Finance in
Germany and France (Frankfurt-am-Main- and Paris: Deutsche Bundesbank and Banque
de France).
Law Commission (2001), Insolvency Law Reform: Promoting Trust and Confidence (An Advisory
Report to the Ministry of Economic Development, Wellington, New Zealand).
Leibenstein, H. (1966), ‘Allocative Efficiency vs. ‘X-Efficiency’’, American Economic Review,

Vol. 56, pp. 392–415.
Mackie-Mason, J. (1990), ‘Do Taxes Affect Corporate Financing Decisions?’, Journal of Finance,
Vol. 45, pp. 1471–92.
Michaelas, N., C. Francis and P. Poutziouris (1999), ‘Financial Policy and Capital Structure
Choice in UK SMEs: Empirical Evidence from Company Panel Data’, Small Business
Economics, Vol. 12, pp. 113–30.
Ministry of Economic Development (2005), SMEs in New Zealand: Structure and Dynamics
(August).
Modigliani, F. and M. H. Miller (1958), ‘The Cost of Capital, Corporation Finance and the
Theory of Investment’, American Economic Review, Vol. 48, pp. 261–97.
Myers, S. (1977), ‘Determinants of Corporate Borrowing’, Journal of Financial Economics,
Vol. 5, pp. 147–75.
——— (1984), ‘The Capital Structure Puzzle’, Journal of Finance, Vol. 39, No. 3 (July), pp.
575–92.
——— (2001), ‘Capital Structure’, Journal of Economic Perspectives, Vol. 15, No. 2 (Spring),
pp. 81–102.
——— and N. Majluf (1984), ‘Corporate Financing and Investment Decisions When Firms
Have Information That Investors Do Not Have’, Journal of Financial Economics, Vol. 13, pp.
187–221.
Ozcan, A. (2001), ‘Determinants of Capital Structure and Adjustment to Long Run Target:
Evidence from UK Company Panel Data’, Journal of Business Finance & Accounting,
Vol. 28, pp. 175–98.
Rajan, R.G. and L. Zingales (1995), ‘What Do We Know About Capital Structure? Some
Evidence from International Data’, Journal of Finance, Vol. L, No. 5, pp. 1421–60.
Ross, G.C. (1977), ‘The Determinants of Financial Structure: The Incentive Signalling
Approach’, Bell Journal of Economics and Management Science (Spring), pp. 232–40.
Ross, S.A., R.W. Westerfield and J. Jaffe (2005), Corporate Finance (7th ed., Irwin Series in
Finance).
Scott, J. (1977), ‘Bankruptcy, Secured Debt, and Optimal Capital Structure’, Journal of Finance,
Vol. 32 (March), pp. 1–19.

Shephard, R.W. (1970), Theory of Cost and Production Functions (Princeton: Princeton University
Press).
Titman, S. (1984), ‘The Effect of Capital Structure on a Firm’s Liquidation Decision’, Journal
of Financial Economics, Vol. 13, pp. 137–51.
——— and R. Wessels (1988), ‘The Determinants of Capital Structure Choice’, Journal of
Finance, Vol. 43, pp. 1–19.
Walsh, E.J. and J. Ryan (1997), ‘Agency and Tax Explanations of Security Issuance Decisions’,
Journal of Business Finance & Accounting, Vol. 24, pp. 943–61.
Williamson, O.E. (1967), ‘Hierarchical Control and Optimum Firm Size’, Journal of Political
Economy, Vol. 75, pp. 123–38.
Zavgren, C.V. (1985), ‘Assessing the Vulnerability to Failure of American Industrial Firms: A
Logistic Analysis’, Journal of Business Finance & Accounting, Vol. 12, No. 1, pp. 19–45.
C

2007 The Authors
Journal compilation
C

Blackwell Publishing Ltd. 2007