10.1108@jaar 03 2012 0023

Journal of Applied Accounting Research
Determinants of the adjustment speed of capital structure: Evidence from developing

economies
Tesfaye T. Lemma Minga Negash
Downloaded by KING MONGKUT UNIVERSITY OF TECHNOLOGY THONBURI At 10:38 12 October 2014 (PT)
Article information:
To cite this document:
Tesfaye T. Lemma Minga Negash , (2014),"Determinants of the adjustment speed of capital structure",
Journal of Applied Accounting Research, Vol. 15 Iss 1 pp. 64 - 99
Permanent link to this document:
http://dx.doi.org/10.1108/JAAR-03-2012-0023
Downloaded on: 12 October 2014, At: 10:38 (PT)
References: this document contains references to 92 other documents.
To copy this document: [email protected]
The fulltext of this document has been downloaded 101 times since 2014*
Users who downloaded this article also downloaded:
Tesfaye Taddese Lemma, Minga Negash, (2013),"Institutional, macroeconomic and firm-specific
determinants of capital structure: The African evidence", Management Research Review, Vol. 36 Iss 11 pp.
1081-1122 http://dx.doi.org/10.1108/MRR-09-2012-0201
Dr Haizhi Wang, Jian Chen, Chunxia Jiang, Yujia Lin, (2014),"What determine firms’ capital structure in
China?", Managerial Finance, Vol. 40 Iss 10 pp. 1024-1039 http://dx.doi.org/10.1108/MF-06-2013-0163
Robert Rugimbana, Sulagna Mukherjee, Jitendra Mahakud, (2010),"Dynamic adjustment towards target
capital structure: evidence from Indian companies", Journal of Advances in Management Research, Vol. 7
Iss 2 pp. 250-266
Access to this document was granted through an Emerald subscription provided by 465057 []
For Authors
If you would like to write for this, or any other Emerald publication, then please use our Emerald for
Authors service information about how to choose which publication to write for and submission guidelines
are available for all. Please visit www.emeraldinsight.com/authors for more information.
About Emerald www.emeraldinsight.com
Emerald is a global publisher linking research and practice to the benefit of society. The company
manages a portfolio of more than 290 journals and over 2,350 books and book series volumes, as well as
providing an extensive range of online products and additional customer resources and services.
Emerald is both COUNTER 4 and TRANSFER compliant. The organization is a partner of the Committee
on Publication Ethics (COPE) and also works with Portico and the LOCKSS initiative for digital archive
preservation.
*Related content and download information correct at time of download.

The current issue and full text archive of this journal is available at
www.emeraldinsight.com/0967-5426.htm
JAAR
15,1
Determinants of the adjustment
speed of capital structure
Evidence from developing economies
64
Tesfaye T. Lemma
School of Accountancy, University of Limpopo, Polokwane, South Africa, and
Minga Negash
Department of Accounting, Metro State University of Denver, Denver,
Colorado, USA and School of Accountancy, University of the Witwatersrand,
Johannesburg, South Africa
Abstract
Purpose – The purpose of this paper is to examine the role of institutional, macroeconomic, industry,
and firm characteristics on the adjustment speed of corporate capital structure within the context
of developing countries.
Design/methodology/approach – The authors considers a sample of 986 firms drawn from nine
developing countries in Africa over a period of ten years (1999-2008). The study develops dynamic
partial adjustment models that link capital structure adjustment speed and institutional,
macroeconomic, and firm characteristics. The analysis is carried out using system Generalized
Method of Moments procedure which is robust to data heterogeneity and endogeneity problems.
Findings – The paper finds that firms in developing countries do temporarily deviate from (and
partially adjust to) their target capital structures. Our results also indicate that: more profitable firms
tend to rapidly adjust their capital structures than less profitable firms; the effects of firm size, growth
opportunities, and the gap between observed and target leverage ratios on adjustment speed are
functions of how one measures capital structure; and adjustment speed tends to be faster for firms in
industries that have relatively higher risk and countries with common law tradition, less developed
stock markets, lower income, and weaker creditor rights protection.
Research limitations/implications – Future research should focus on examination of the
adjustment speed of debt maturity structure. Identification of industry-specific characteristics that
affect the pace with which firms adjust their capital structure to the optimum is another possible
avenue for future research.
Practical implications – Our findings have practical implications for corporate managers,
governments, legislators, and policymakers.
Originality/value – The study focuses on firms in developing countries for which the literature on
adjustment speed of capital structure is virtually non-existent. Furthermore, unlike previous works on
capital structure, it explicitly models industry variable as one of the determinants of adjustment speed.
Therefore, it contributes to the literature on capital structure and adjustment speed in general and to
the literature on developing countries in particular.
Keywords Developing countries, Capital structure, Macroeconomic conditions, Industry,
Adjustment speed, Firm characteristics, Instituitions
Paper type Research paper
JEL Classification — G32

Journal of Applied Accounting The authors gratefully acknowledge the helpful suggestions of Dr Julia Mundy (Editor) and
Research
Vol. 15 No. 1, 2014 two anonymous referees. The authors also wish to thank Addis Ababa University for the
pp. 64-99 financial support it provided to the first author. The authors also gratefully acknowledge the
r Emerald Group Publishing Limited
0967-5426 helpful suggestions of the participants of the International conference of Southern Africa
DOI 10.1108/JAAR-03-2012-0023 Accounting Association of the 2011 edition.
1. Introduction Evidence from
Capital structure research is epitomized by conditional theories that posture developing
contrasting views about the target adjustment behaviour of firms. For instance,
traditional trade-off theory stresses various costs and/or benefits of debt, and thus, economies
implies the existence of an optimal capital structure. On the other hand, market timing
theory suggests that a firm’s capital structure is a cumulative result of its historical
efforts to time equity issuances with high market valuations rather than a result of 65
a dynamic optimizing strategy (Baker and Wurgler, 2002). Similarly, inertia theory
suggests that equity price shocks have a persistent effect on leverage which it
considers as evidence against firms rebalancing their capital structures towards an
optimum (Welch, 2004). Likewise, pecking order theory considers capital structure to
be primarily a result of firms’ historical profitability and investment opportunities and
hence firms have no strong tendencies to reverse shocks to capital structure caused by
financing needs and earnings growth (Myers, 1984; Myers and Majluf, 1984).
Recent empirical literature documents evidence which show that market
imperfections and adjustment costs and/or benefits oblige firms to operate at a debt
level that is sub-optimal. It further suggests that, in perfect markets where there is
no friction, the adjustment of capital structure towards a target is costless, and thus,
a firm can instantly adjust its capital structure towards the optimum. However, in
imperfect markets, the adjustment of capital structure towards the optimum is costly,
and hence, a firm may not adjust its capital structure instantly, but adjusts partially
(see Drobetz and Wanzenried, 2006; Heshmati, 2001; Leary and Roberts, 2005).
The literature’s attempt to empirically discriminate between trade-off and other
competing theories by using dynamic trade-off theory framework and partial
adjustment models has rather become promising (Elsas and Florysiak, 2008). The
findings, almost invariably, confirm the argument that there is a substantial dynamic
component in a firm’s capital structure decisions, and that the dynamism depends on
firm, industry, macroeconomic, and institutional factors (Drobetz et al., 2007; Drobetz
and Wanzenried, 2006; Flannery and Hankins, 2007). Although the institutional and
macroeconomic setup of developing countries differs in many ways from those
of developed countries, we are not aware of any published work that investigates the
dynamic partial adjustment (DPA) of a firm’s capital structure within the context of
developing countries. This paper aims to fill this void by investigating whether firms
in developing countries adjust their capital structures to certain target levels and, if
they do, how firm, industry, macroeconomic, and institutional factors influence the
adjustment speed.
The contributions of this paper are fourfold. First, it provides an “out-of-sample-test”
for the theoretical and empirical research carried out in the context of advanced
economies. Second, it explicitly models the industry variable as one of the determinants
of capital structure adjustment speed. Third, it helps identify the institutions and
macroeconomic policies that are conducive for enhancing the convergence of capital
structure of firms in sample countries to optimum levels. Fourth, it helps policymakers
and other stakeholders in crafting policies and legislations that enhance firms’ ability to
adjust to optimal capital structures.
The paper applies DPA models on ten-years (1999-2008) data pertaining to 986
non-financial firms drawn from nine developing economies from Africa that have
functioning stock exchanges. Model parameters were estimated using system
Generalized Method of Moments (sys-GMM) estimator proposed by Blundell and Bond
(1998). The main findings of the study are: capital structures of firms in our sample
JAAR countries do temporarily deviate from and partially adjust to target; more profitable
15,1 firms tend to more rapidly adjust their capital structures than less profitable firms; the
effects of firm size, growth opportunities, and the gap between observed and target
leverage ratios on adjustment speed are functions of how one measures capital
structure; and adjustment speed tends to be faster for firms in industries that have
relatively higher risk and countries with common law tradition, less developed stock
66 markets, lower income, and weaker creditor rights protection.
The practical implications of our findings are twofold. First, the influence of
firm-level variables on adjustment speed suggests that managers have some sway
over the speed with which firms adjust their capital structures to the optimum.
Second, the influences of institutional and macroeconomic factors imply that
governments, legislators, and policymakers could influence the pace at which firms
adjust their capital structures to the optimum through regulations and policymaking.
The paper proceeds as follows. Section 2 presents a brief review of the literature on the
adjustment speed of capital structure. Section 3 proffers the empirical setup for our
analysis. Section 4 presents the results and discussions and Section 5 concludes.
2. Literature review
Recent literature criticizes studies on determinants of capital structure by pointing to
the fact that researchers do not take into account the typical rebalancing behaviour
of firms as far as their capital structure is concerned. It draws on dynamic trade-off
theory and develops a theory for dynamic capital structure (Flannery and Hankins,
2007). This literature could be grouped into two succinct clusters: those investigating
whether firms adjust towards target capital structures, and those investigating the
factors that influence the pace at which firms adjust their capital structure. In what
follows, we attempt to briefly review these two clusters of the literature[1].
2.1 On the existence of a target capital structure

The literature alludes that detection of target behaviour in capital structure of a firm is,
arguably, central to discriminating between trade-off and alternative theories (Myers,
1984). Trade-off theories imply the existence of target capital structure and assume
that firms make financing choices that minimize the cost of deviating from their target
(Chang and Dasgupta, 2009). On the other hand, alternative theories mentioned earlier
suggest that firms do not have target capital structure. In a rebuttal of the trade-off
view, Miller (1977) showed that there would be no optimum capital structure at firm
level since bankruptcy costs are “trivial” and tax advantage of debt financing at firm
level is exactly offset by the tax disadvantage of debt at personal level[2]. Haugen
and Senbet (1978) and Barnea et al. (1980) point out that bankruptcy “penalties” are too
small to offset the effect of tax advantage of debt. Nonetheless, DeAngelo and Masulis
(1980), Kim (1982), and Modigliani (1982) argue that bankruptcy costs are not the only
costs against which the tax advantages of debt ought to be weighed – there are other
costs of debt such as agency costs, loss of non-debt-related tax-shield, etc., that should
be considered in the determination of optimal capital structure.
The literature shows that capital structure decisions reflect not only the optimal
leverage ratio but also the rebalancing exercises towards an optimum position for the
firm. Myers’ (1984) view that trade-off theory suggests a target capital structure was
corroborated by Jalilvand and Harris (1984) who reported that a firm’s financial
behaviour is characterised by partial adjustment to long-run financial targets.
However, Jalilvand and Harris’s work was criticized for exogenously specifying the
long-run financial target to which firms adjust. In all, 17 years later, De Miguel Evidence from
and Pindado (2001) attempted to improve on Jalilvand and Harris’s work by developing
endogenizing the target capital structure in their model and concluded that firms
adjust their capital structure towards an optimum. At about the same time, in a survey economies
of corporate finance practice, Graham and Harvey (2001) reported that 81 per cent
of the CFOs in their sample had either a target range of debt ratio or a “strict” target
debt ratio. 67
In a similar vein, Fama and French (2002) note that firms slowly adjust their debt
ratios towards a target. This observation is consistent with the suggestion by Myers
(1984) that firm’s may take long time to return to their target capital structures in the
presence of costs of adjustment. In a rebuttal of the works of Baker and Wurgler (2002)
and Welch (2004), Leary and Roberts (2005) show that firms actively rebalance their
capital structure to stay within the optimal range. In a further push, more recent
literature, employing models and procedures that are more robust, confirms not only
that firms adjust their capital structure but also that adjustment costs and/or benefits
enhance or mitigate the speed at which firms adjust their capital structures towards
the optimum (Faulkender et al., 2012; Banerjee et al., 2004; Drobetz and Wanzenried,
2006; Flannery and Rangan, 2006; Gaud et al., 2005; Huang and Ritter, 2009; Frank
and Goyal, 2009).
Although copious empirical studies endeavour to investigate adjustment speed of
capital structure, Shyam-Sunder and Myers (1999) and Chen and Zhao (2007) caution
against overly interpreting adjustment coefficients by pointing to the possibility that
“firm’s leverage ratios tend to mechanically revert to the mean regardless of the firm’s
financing preferences”. However, a number of other widely employed tests are also
susceptible to mechanical effects that could arise when firms do not follow target
behaviour (e.g. Chang and Dasgupta, 2009).
2.2 On the determinants of adjustment speed of capital structure

In a recent paper, Faulkender et al. (2012) re-iterate a contemporary question in capital
structure research: whether firms have a target capital structure, and if so, what
factors enhance (or hinder) the speed of adjustment towards a target? Various research
endeavours suggest that the speed at which firms adjust their capital structures
towards target varies from study to study. Part of the dissention has to do with the
econometric procedures employed and part of it could be ascribed to differences in
adjustment costs and/or benefits. The latter view was reflected in Flannery and
Hankins (2007) who remark that capital structure decisions reflect not only the level of
the optimal leverage ratio but also both the costs of deviating from the target and the
costs of adjusting towards that target. According to these authors, whilst adjustment
costs hinge on external financing expenses, stock price movements and financial
constraints, adjustment benefits depend on potential costs of distress and the value
of tax shields. In what follows, we present a synthesis of the influence of firm, industry,
and country characteristics on adjustment costs and/or benefits, and thereby, on
adjustment speed.
2.2.1 Inter-firm heterogeneity in adjustment speed of capital structure. As the
adjustment costs and/or benefits are likely to vary from firm to firm, so does the
optimal capital structure adjustment process (e.g. Flannery and Hankins, 2007).
Studies as early as Fischer et al. (1989) propose a model of dynamic capital structure
choice in the presence of adjustment costs and show that the swings in capital
structure are functions of firm-specific factors. In the following paragraphs, we present
JAAR a synthesis of the relationship between firm characteristics and adjustment speed by
15,1 using adjustment costs and/or benefits as a framework.
The literature customarily suggests that larger firms tend to have lower
information asymmetry which enables them to have lower financing costs as they
are likely to enjoy better access to external finance. The lower the financing cost of a
firm is the lower its capital structure adjustment cost. Thus, we expect larger firms to
68 have smaller adjustment costs and, thus, faster adjustment speed (e.g. Banerjee et al.,
2004; Drobetz and Wanzenried, 2006; Flannery and Hankins, 2007). On the other hand,
one could argue that larger firms tend to have less cash flow volatility, which reduces
the potential costs of distress (e.g. Flannery and Rangan, 2006). A reduction in potential
cost of distress in turn reduces a firm’s benefit of adjusting toward a target capital
structure thereby reducing the adjustment speed (e.g. Flannery and Hankins, 2007).
Although Flannery and Hankins (2007) report a positive relationship between firm
size and adjustment speed, Haas and Peeters (2006) and Banerjee et al. (2004) observe
an inverse relationship between the two variables.
According to Flannery and Hankins (2007), profitability influences both the costs
and/or benefits of capital structure adjustment of a firm. A more profitable firm is
likely to have more flexibility (i.e. lesser constraints) in financing decisions and likely
to enjoy issuance of securities at more attractive rates (i.e. lower cost of external
financing). This signifies that firms that are more profitable are likely to experience
lesser costs of rebalancing their capital structure towards a target. In addition,
higher profit may also increase the value of debt tax shields or minimize asset
substitution concerns (i.e. increase benefits of adjustment); especially if the firm is
under-leveraged (see Flannery and Hankins, 2007). Thus, we conjecture that
profitability of a firm positively influences the pace at which a firm adjusts its
capital structure to a target.
Both Banerjee et al. (2004) and Drobetz and Wanzenried (2006) suggest that growing
firms tend to have more flexibility in choosing the sources of finance than no-growth
firms which can only change their capital structure by swapping debt against
equity. This in turn implies that growing firms are likely to enjoy lesser financing
constraints and hence are likely to more rapidly rebalance their capital structures
towards target levels. Although Drobetz and Wanzenried’s (2006) empirical results
corroborate this conjecture, Banerjee et al.’s (2004) results indicate that firms with
higher growth opportunity adjust more slowly towards the optimal capital structure.
The inconsistent result obtained by Banerjee et al. (2004) could partly be due to the
non-linear least square estimation technique (they) used to analyse the data which
usually leads to biased and inconsistent estimators (Drobetz and Wanzenried, 2006).
The theoretical predictions regarding the relationship between the magnitude of the
distance between target and observed capital structures of a firm (i.e. the distance
variable) and adjustment speed are indeterminate. If fixed costs (e.g. legal fees and
investment bank fees) constitute a major portion of the rebalancing cost, only firms
that moved significantly far away from the optimal capital structure will change their
capital structure. Hence, we expect a positive relationship between adjustment speed
and the distance variable (e.g. Banerjee et al., 2004, Drobetz and Wanzenried, 2006)[3].
On the other hand, if the fixed costs of adjustment are prohibitively high, firms may
avoid using capital markets to raise funds and manipulate their dividend policy to
rebalance their capital structure. In this case, cost of adjustment tends to be increasing
with increase in the distance variable implying slower adjustment speed. While
Drobetz and Wanzenried (2006) found a statistically weak but positive relationship,
Banerjee et al. (2004) reported mostly insignificant relationship between the two Evidence from
variables. Finally, a recent strand of research also reports links between cash flows developing
and adjustment speed. Both Byoun (2008) and Faulkender et al. (2012) note that a firm’s
financial need is a critical determinant of capital structure adjustment speed. economies
2.2.2 Inter-industry heterogeneity in adjustment speed of capital structure. Studies
that explicitly examine inter-industry heterogeneity in adjustment speeds are scant.
However, some studies control for industry effects (e.g. Flannery and Rangan, 2006, 69
Hovakimian et al., 2004) to account for inter-industry differences in adjustment speeds.

Roberts (2002) showed that the speed at which firms revert back to their target capital
structures varies dramatically across industries suggesting the existence of significant
inter-industry variation in adjustment costs and/or benefits. Likewise, Banerjee et al.
(2004) compared the adjustment speed of firms in various industries and reported
that there are substantial differences in adjustment speed across industries in the USA
and UK. In a similar vein, Smith et al. (2010) estimate adjustment speeds of capital
structures of firms in 15 industries in New Zealand and showed that firms in
agriculture and fisheries, mining, forestry, media and communications, and investment
industries move towards their target capital structures relatively rapidly, in
comparison to those in other industries. They suggest that risk characteristics of the
industry in which a firm operates might be an underlying factor explaining
inter-industry differences in adjustment speed.
In addition, Stoja and Tucker (2007) classify industries into “new economy” group
which include biotechnology, IT, and leisure industries and “old economy”
group which includes oil and mining, construction, textiles, and real estate
industries. These authors propose that adjustment costs for firms in “old economy”
industries are likely to be higher as they are fixed-assets-intensive with a low level of
service element whereas firms in “new economy” industries are likely to adjust
faster since they are R&D-intensive with a high level of service element. Thus, we
expect that industry characteristics might influence the costs and/or benefits of
adjusting to or deviating from target capital structure.
2.2.3 Cross-country heterogeneity in adjustment speed of capital basic structure. Most
prior empirical works were based on single country databases. However, the speeds of
adjustment reported vary considerably around the world. While a number of papers
find evidence of relatively slow adjustment speeds, others report faster speeds. Studies
which report low adjustment speeds include Fama and French (2002) who report that
US firms move towards target debt ratios at speeds ranging from 7 to 18 per cent
each year. Similarly, Hovakimian and Li (2011) estimate adjustment speeds ranging
from 5 to 13 per cent. Consistent with faster speeds reported in earlier studies such
as Jalilvand and Harris (1984), Shyam-Sunder and Myers (1999), and Flannery and
Rangan (2006) report adjustment speeds of 41 and 34 per cent, respectively, for firms in
the USA. According to Ozkan (2001), the adjustment speed of capital structure of UK
firms is in the vicinity of 43 per cent. There are also studies (e.g. De Miguel
and Pindado, 2001 and Gaud et al., 2005 for Swiss firms and Lemmon et al., 2008 and
Huang and Ritter, 2009 for US firms) which report adjustment speeds that lie at the
middle of the spectrum. As we alluded to earlier, part of the dissension in the
adjustment speeds stems from econometric issues. But econometric issues may not
explain all of the variation in the speed of adjustment. This variation in the observed
adjustment speeds in different countries opened a further research direction in which
many researchers attempted to examine the nexus between country characteristics and
capital structure adjustment speed.
JAAR La Porta et al. (1997, 1998, 1999a, b) extensively document that the quality of law
15,1 and the extent of its enforcement are important determinants of the shape and
complexity of financial contracts. At the heart of their argument is the protection
afforded by legal systems to mitigate agency problems between insiders and outsiders.
Investors’ disposition towards providing funding for firms partly depends on the
protection they receive from the legal system. The authors show that legal systems
70 based on the English common law provide a stronger protection to investors
(i.e. shareholders and creditors) than those based on the French civil law. The empirical
literature confirms this prediction. Thus, firms in countries with legal systems
based on the English common law tend to have lesser agency-associated problems
compared to those in countries with legal systems based on the French civil
law. Therefore, we expect firms in the first group of countries to more quickly
adjust their capital structure to a target than those in the latter group. Following
Clark et al. (2009) and Wanzenried (2006), we also anticipate that firms located
in countries with stronger creditor and shareholder rights protection and
relatively higher levels of contract enforcement efficiency would adjust their capital
structures more quickly than firms located in countries characterized by lower levels
of creditor and shareholder rights protection and less efficient means of enforcing
contracts.
The literature also attempts to explain variations in adjustment speeds by invoking
cross-country divergence in financial systems. Developed stock markets and banking
sectors make it easier for firms to raise capital. The likely smaller transaction costs and
reduced agency costs associated with developed stock markets and banking sectors
would mean that firms find it easier to adjust their capital structures to a target
(e.g. Clark et al., 2009; Demirgüç-Kunt and Maksimovic, 1999; Grossman and Stiglitz,
1980; Wanzenried, 2006). Hence, we conjecture that the size and liquidity of stock
markets and the size of banking sector have positive effects on the speed at which
firms adjust their capital structures to a target.
To test the impact of macroeconomic conditions on the speed of capital structure
adjustment, prior empirical literature employs such factors as overall size of the
economy, GDP growth rate, inflation rate, and taxation. The GDP growth rate is
usually considered as an indicator of financing needs of firms (e.g. Demirgüç-Kunt
and Maksimovic, 1999). Thus, in line with Cook and Tang (2010), Drobetz and
Wanzenried (2006), and Wanzenried (2006), we expect firms to adjust their capital
structure to a target at a faster rate as the economy goes through higher GDP growth.
According to Mills (1996), higher inflation rates lead to higher costs of capital and
changes in the cost of capital are paid closer attention by firms so that they can
optimize their capital structure. Hence, consistent with Wanzenried (2006), we expect
higher inflation rates to have positive influence on adjustment speed. The dynamic
trade-off theory predicts that adjustment speed is positively related to benefits of
being at a target capital structure. Thus, the higher the benefit resulting from
untapped tax benefits, the faster the pace at which a firm adjusts its capital structure
(Clark et al., 2009). As in the legal institutions, we examine the effect of macroeconomic
conditions on the speed of adjustment in two stages. We first examine if there
are variation in adjustment speeds across broadly defined income groups (i.e. upper-
middle-income, lower-middle-income, and low-income countries) to which the country
belongs. Second, we examine the effect of more-narrowly defined macroeconomic
variables (i.e. taxation, inflation, size of economy, and growth rate of GDP) on
adjustment speed.
3. The empirical framework Evidence from
Until recently, empirical work on capital structure imposes the implicit, but unrealistic, developing
assumption that firms are always at their target capital structure. In an imperfect
environment where there are a set of adjustment costs and/or benefits, a firm’s capital economies
structure may not necessarily be at the target level. In an effort to properly account for
the dynamic nature of capital structure, recent literature adopted a DPA capital
structure model that allows target capital structure to vary across firms and over time 71
(Drobetz and Wanzenried, 2006; Fischer et al., 1989; Hovakimian et al., 2001).
As we noted earlier, measures of adjustment speed are very sensitive to the
econometric design. The econometric challenges include, among others, problems of
model specification, unobservable variables, heterogeneous panel data, short panel
biases, autocorrelation, and unbalanced panels (e.g. Zhao and Susmel, 2008). Two
distinct strands of econometric modelling approaches stand out in the study of
adjustment speed of capital structure: the two-stage and integrated DPA capital
structure models (e.g. Cook and Tang, 2010). Although both approaches are widely
used, Flannery and Rangan (2006) convincingly show that the two-stage dynamic
panel adjustment model results in abnormally smaller estimates of adjustment speed
than theory would predict. Further, this approach does not allow us to examine the
determinants of adjustment speed while the integrated approach enables us to jointly
determine the adjustment speed along with its determinants. Hence, this study adopts
the integrated DPA model.
3.1 Model specification

In line with De Miguel and Pindado (2001) and Hovakimian et al. (2001), we define
target capital structure as a leverage ratio that a firm would desire to have in a
frictionless environment. To analyse the impact of firm, industry, and country
characteristics, pursuant to Drobetz and Wanzenried (2006) and Öztekin and Flannery
(2012), we specify target capital structure by using a dynamic capital structure model.
Let the optimal or target capital structure of firm i in period t, labelled as Levi;t , be
a linear function of a set of N explanatory variables, Kj,i,t (where j ¼ 1, 2, 3, y. N) that
have been used in past cross-sectional studies of capital structure:
X
Levi;t ¼ N
j¼1 yj K j;i;t ð1Þ
where y j denotes a column vector containing the coefficients of explanatory variables.
In a frictionless environment where information asymmetries, transaction costs,
and other adjustment costs and/or benefits are absent, firms may instantly adjust their
capital structure to a target. Hence, in such an environment, observed capital structure
ðLevi;t Þ is expected to be the same as target capital structure ðLevi;t Þ . In other words, in
a perfect environment, the difference between the current and the previous periods’
observed capital structure should be the same as the difference between target capital
structure and the pervious period’s capital structure. That is, Levi;t Levi;t1 should be
equal to Levi;t Levi;t1 . However, in the presence of all sorts of adjustment costs
and/or benefits (which is more likely in the real world), Levi,t is not necessarily the
same as Levi;t . That is, firms may not fully adjust their capital structure to the target
capital structure. They may rather adjust partially. Thus, the equality is disrupted and
a more realistic partial adjustment model may be specified as:
ðLevi;t Levi;t1 Þ ¼ gi;t ðLevi;t Levi;t1 Þ þ ei;t ; where 0ojgi;t jo1 ð2Þ
JAAR where gi,t denotes the adjustment parameter representing the magnitude of adjustment
15,1 towards a target capital structure between two consecutive periods, Levi,t1 represents
capital structure of firm i, in period t1, and ei,t denotes the idiosyncratic error term.
Rearranging the terms in Equation (2), we obtain:
Levi;t ¼ ð1 gi;t ÞLevi;t1 þ gi;t ðLevi;t Þ þ ei;t ; where 0ojgi;t jrmo1 ð3Þ
72
Our model follows Drobetz and Wanzenried (2006) and Hovakimian et al. (2001), where
firms adjust to a target capital strucure to an endoegenously determined capital
structure as specified in Equation (1). Following prior empirical work, we specify
adjustment speed (gi,t) as a linear function of factors affecting the costs and/or benefits
of adjustment and the unobserved firms-specific effects as follows:
gi;t ¼ o0 þ o1 Xi;t ð4Þ

When firm-specific variables are used to explain the speed of adjustment, Xi,t has both
time and cross-sectional dimensions. In contrast, in the case of country-level variables,
Xi,t has only time dimension as country-level variables do not vary across firms.
Substituting Equations (4) and (1) in Equation (3), we obtain:
X
N
Levi;t ¼ ½1 ðo0 o1 Xi;t ÞLevi;t1 þ ðo0 o1 Xi;t Þð j¼1 yj Kj;i;t þ ei;t Þ ð5Þ
Partly multiplying Equation (5) out, we obtain:

X X
N N
Levi;t ¼ ð1 o0 ÞLevi;t1 o1 Xi;t Levi;t1 þ o0 j¼1 yj Kj;i;t þ o1 j¼1 yj Kj;i;t þ ei;t ð6Þ
When Equation (6) is estimated, interest is mainly in o1 which is the coefficient of

the interaction term between the determinant variable of adjustment speed, Xi,t,
and the lagged leverage, Levi,t1. Thus, we formulate the null hypothesis that o1 ¼ 0,
i.e. the speed of adjustment is independent from firm, industry, and/or country
characteristics.
We observe, in the econometrics literature, that varying econometric procedures
could be used to estimate Equation (6) and results are non-robust. However, after
comparing the results from estimation procedures ranging from OLS to fixed and
random effects, from Instrument Variables to GMM, Antoniou et al. (2008) and
Deesomsak et al. (2009) demonstrate that sys-GMM is the most appropriate method to
estimate Equation (6). Hence, this study uses sys-GMM for the purpose of estimation.
We check presence (or absence) of second-order serial correlation in the first differences
of the error term as the consistency of the sys-GMM estimator requires that this
condition be satisfied. Further, we verify the validity of instruments using the Sargan
test of over identifying restrictions.
3.2 Measuring capital structure

Similar to the competing theories, there is no universally accepted definition of capital
structure in the literature. Researchers agree that measures of capital structure should
vary depending on the purpose of analysis (e.g. Bevan and Danbolt, 2002; Rajan
and Zingales, 1995; Lemma and Negash, 2011, 2013a). In addition, empirical studies
show that different measures of capital structure produce different results, hence, can
affect the interpretation of results (Harris and Raviv, 1991). Further, competing theories
have different implications for capital structure depending on how it is defined Evidence from
(e.g. Bhaduri, 2002a, b; Frank and Goyal, 2009; Titman and Wessels, 1988). Hence, the developing
literature emphasizes the importance of considering: both short-term and long-term;
and market-based and book-based measures of capital structure. economies
Ostensibly, most studies do not use market-based measures of capital structure
since: most theoretical predictions apply to book values (see Fama and French, 2002);
book-based measures may better reflect management’s target capital structure 73
since market values of equity depend on a number of factors that often cannot be
controlled by the firm; information obtained from financial statements is more credible;
and market values of debt are often not available (see Thies and Klock, 1992). Many
researchers report that the use of book value delivers similar results to market value as
the two, as Bowman (1980) demonstrates, are highly correlated. Further, Welch (2010)
shows how the common use of financial-debt-to-asset ratio as a measure of leverage is
fundamentally flawed. In cognizance of all these, the present study employs three
book-based measures of basic capital structure: short-term leverage (STL); long-term
leverage (LTL); and total leverage (TL)[4]. The specific definition of each measure
of capital structure is indicated in the explanatory notes accompanying the
various tables.
3.3 The sample and data

This study focused on firms drawn from nine developing economies in Africa –
Botswana, Egypt, Ghana, Kenya, Mauritius, Morocco, Nigeria, South Africa, and
Tunisia. The choice of these countries was motivated by several factors. First, they
are all in Africa where the literature on the role of firm, industry, institutional, and
macroeconomic factors on capital structure adjustment speed is sparse. Second, these
countries have different institutional setups, such as financial markets, legal traditions,
and level of economic development. In particular, Botswana, Ghana, Kenya, Nigeria,
and South Africa are members of the British Commonwealth, and thus, have some
common attributes in corporate governance and corporate control whereas Egypt,
Mauritius, Morocco, and Tunisia are civil law based countries. In addition, while the
stock exchanges in Botswana, Ghana, Kenya, Nigeria, Mauritius, Morocco, and Tunisia
are recently emerging exchanges those in South Africa and Egypt are more established
markets. Furthermore, although not as wide, there is considerable difference in
the level of economic development of these countries. This diversity offered the
opportunity to assess the effects of different institutional and macroeconomic
environments on capital structure adjustment speeds.
Firm-specific data used in this study were extracted from financial statements
of listed firms in sample countries. The financial statements were sourced from
OSIRIS database of Bureau DIJK that maintains a comprehensive financial database of
over 70,000 firms across the globe. We started with all the firms listed in all of the
functioning stock exchanges of 17 developing economies in Africa that had data in
the OSIRIS database as at 31 December 2009. Following precedence in similar studies
(De Jong et al., 2008), we require that firms in our sample should have at least three
years of available data over the study period and countries should have at least ten
firms. We dropped firms in the financial industry (US SIC code 6000B) as their capital
structure is subject to different set of regulations. The final dataset comprised of
ten-years (1999-2008) data pertaining to 986 non-financial firms. The sampled firms
represent ca. 56 per cent of the average number of firms listed in sample countries over
the sample period (see Table I). We adjusted differences in fiscal years of firms to
JAAR Country All
15,1 Industry
South All firms
Egypt Africa Botswana Ghana Kenya Mauritius Morocco Nigeria Tunisia firms (%)
Non-durables 107 26 1 3 8 9 8 14 3 179 18

Durables 18 9 1 0 1 1 0 1 1 32 3
Manufacturing 114 31 0 2 4 1 7 11 3 173 18
74
Oil and Gas 7 41 0 0 3 1 4 2 1 59 6

Chem. and
Constriction 75 16 0 1 1 0 3 5 4 105 11
Business equipment 11 35 0 1 0 0 5 2 2 56 6
Regulated 23 15 0 0 5 2 2 1 2 50 5
Wholesale and Retail 51 38 6 2 4 7 6 10 3 127 13
Health 38 5 0 1 0 0 1 6 2 53 5
Service and other 80 36 3 0 6 4 3 19 3 153 16
Firms in the sample 522 252 11 10 32 25 39 71 24 986 100
% of sample 53 26 1 1 3 3 4 7 2 100 –
Average number of
firms listed 828 470 18 27 53 50 60 203 46 1,754 –
% of listed firms 63 54 63 37 61 50 66 35 52 56 –
Notes: Percentage of sample refers to the percentage of the number of firms drawn from each country in the sample
considered for the study. Average number of firms listed refers to the average of the number of firms listed in the
stock exchanges of the individual countries as reported in the World Development Indicators database on the World
Bank. The table provides a country-by-country and industry-by-industry composition of the sampled firms. Non-
durables (IND1) include industries which fall within the following US SIC classifications: 0100-0999, 2000-2399,
2700-2799, 3100-3199, and 3940-3989. Durables (IND2) include industries which fall within the following US SIC
classifications: 2400*, 2500-2519, 2590-2599, 3630-3659, 3710-3711, 3714-3714, 3716-3716, 3750-3751, 3792-3792,
3900-3939, and 3990-3999. Manufacturing (IND3) includes industries which fall within the following US SIC
classifications: 2520-2589, 2600-2699, 2750-2769, 3000-3099, 3200-3569, 3580-3629, 3700-3709, 3712-3713, 3715-3715,
3717-3749, 3752-3791, 3793-3799, 3830-3839, and 3860-3899. Oil and Gas industry (IND4) includes industries which
fall within the following US SIC classifications: 1000*, 1400*, 1200-1399, and 2900-2999. Chemical and Construction
industries (IND5) include industries which fall within the following US SIC classifications: 1500*, 1600*, 1700*,
2800-2829, 2840-2899. Business equipment industry (IND6) includes industries which fall within the following US
SIC classifications: 3570-3579, 3660-3692, 3694-3699, 3810-3829, 7370-7379. Regulatory industries (IND7) include
industries which fall within the following US SIC classifications: 4000*, 4400*, 4500*, 4600*, 4800-4899, 4900-4949.
Wholesale and Retail industries (IND8) include industries which fall within the following US SIC classifications:
Table I. 5000-5999, 7200-7299, 7600-7699. Health industries (IND9) include industries which fall within the following US SIC
Composition of the sample classifications: 2830-2839, 3693-3693, 3840-3859, 8000-8099. Service and, etc., industries (IND10) include all others
provide a more accurate empirical work. That is, if the date of preparation of financial
statements for a firm is on or before 30 June, its year was stamped as one-year prior to
its fiscal year and if a firm’s fiscal year is after 30 June, that same year was stamped as
the firm’s fiscal year.
Data on country specific variables were collected from various sources. Data on
legal variables, except for rule of law data, were downloaded from the web page
of Andrei Shelifer[5]. The rule of law data were taken from Kaufmann et al. (2009).
All the data on country’s macroeconomic and market conditions were taken either from
World Development Indicators or Financial Structure Database of the World Bank.
Additional country-level data were obtained from previous studies including
Berkowitz et al. (2003).
To provide further insights about the sample, we present an overview of the number
of firms available in the final data set by country and industry (see Table I). In terms of
country distribution, we note that firms from Egypt and South Africa may heavily
influence the sample; they constitute ca. 79 per cent of firms included in the sample. Evidence from
On the other hand, those from Botswana and Ghana have little influence on the sample developing
as they constitute only 2 per cent of firms included in the sample. Nonetheless, the fact
that our sample (except for Ghana and Nigeria) consists of more than 50 per cent of economies
firms listed in each of the individual countries permits us to have a reasonable picture
of the capital structure decisions of firms in those countries (see Table I).
From an industry perspective, we observe that firms in non-durable, 75
manufacturing, and service industries may dominate the results with participation
of 18, 18, and 11 per cent, respectively. Firms from durables and health industries are at
the other end of the spectrum, with only 3 and 5 per cent participation, respectively.
4. Results and discussions

4.1 Descriptive statistics
Prior cross-country studies on capital structure report that firms in developing
countries exhibit lower levels of gearing compared to those in developed countries
(e.g. De Jong et al., 2008). Thus, we assess whether the gearing levels in our sample
countries are comparable with those for developed and other developing economies[6]
reported in Cheng and Shiu (2007)[7]. Table II – panel A shows that the overall
mean ratios of sample firms is 49.3, 11.8, and 37.5 per cent for TL, LTL, and STL,
respectively. We also note from the same table (panel C) that the average TL-ratios for
our sample firms vary from a low of 44.1 per cent in Morocco to a high of 64.9 per cent
in Nigeria[8]. On the other hand, Cheng and Shiu (2007) report the average TL-ratios
vary from a low of ca. 41.9 per cent in Taiwan to a high of 66.9 per cent in Indonesia
for developed countries and from a low of 31.8 per cent in Venezuela to a high of ca. 62.9
per cent in Pakistan for other developing countries. Thus, unlike the allusions in other
studies, the level of leverage-ratios of our sample firms is more or less similar to those
in other developing and developed economies.
We further observe three salient features of capital structure of sample firms. First,
all measures of capital structure were varying over time (see Table II – panel A).
This could be an initial indication that firms might be attempting to adjust their capital
structure towards a target. Second, we observe an overall upward trend in all the three
measures of capital structure throughout the sample period. TL-ratio, for example,
increased from 41.3 per cent in 1999 to 47.6 per cent in 2008 while LTL-ratio went from
9.9 to 13.9 per cent over the same period. This might, arguably, be ascribed to
a confluence of increasing size and growth rate of the economies of sample countries
(e.g. Booth et al., 2001; De Jong et al., 2008); increasing size and liquidity of stock
markets (Demirgüç-Kunt and Maksimovic, 1999; Wanzenried, 2006); and increasing
inflationary situations (e.g. Frank and Goyal, 2009; Taggart, 1985) over the same period
(see Table II – panel A). This phenomenon may also be due to the steady increase in
profitability, growth opportunities, and dividend payout variables that we noted in
unreported results (Antoniou et al., 2008; Barclay and Smith, 1999; Benito, 2003;
Deesomsak et al., 2004; Mazur, 2007).
Third, STL-ratio was on the decline over the second half of the sample period which
suggests that firms in developing African countries are gradually moving towards
long-term finance and away from the traditionally dominant short-term finance. This
phenomenon could be due to the increase in the size and liquidity of stock markets over
the sample period which may have enticed quoted firms to switch to using more
long-term than short-term debt (e.g. Deesomsak et al., 2009). It could well be due to the
hike in firm size, profitability, and growth opportunities that we observed in
76
15,1
JAAR
Table II.
Capital structure and
country characteristics
over the sample period
Panel A: descriptive statistics of institutional and macroeconomics characteristicsa Size of
Total Long-term Short-term Size of Growth of Size of stock Liquidity of banking Creditor Shareholder Rule of
Year leverage leverage leverage Taxation Inflation economy economy market stock market sector rights rights law
1999 0.413 0.099 0.314 35.108 4.098 3.188 2.332 73.484 26.960 0.660 2.384 3.550 –
2000 0.448 0.100 0.348 34.985 4.213 3.199 2.621 58.206 28.824 0.657 2.384 3.550 0.077
2001 0.488 0.121 0.367 34.985 4.821 3.206 1.677 46.577 18.948 0.691 2.384 3.550 –
2002 0.501 0.115 0.386 34.985 5.363 3.210 1.034 61.606 30.713 0.702 2.384 3.550 0.102
2003 0.500 0.109 0.392 34.863 5.797 3.220 2.206 62.971 20.428 0.699 2.384 3.550 0.125
2004 0.500 0.112 0.388 34.863 8.252 3.233 3.202 85.285 23.278 0.705 2.384 3.550 0.036
2005 0.499 0.115 0.384 34.863 5.530 3.246 2.980 112.525 35.167 0.709 2.384 3.550 0.030
2006 0.498 0.121 0.377 34.531 7.001 3.266 4.609 125.792 44.854 0.691 2.384 3.550 0.099
2007 0.490 0.131 0.359 23.404 8.021 3.285 4.592 144.504 42.829 0.679 2.384 3.550 0.119
2008 0.476 0.139 0.337 23.404 NA NA NA NA 51.166 – 2.384 3.550 0.100
Overall 0.493 0.118 0.375 32.599 5.899 3.228 2.806 85.661 32.317 0.688 2.384 3.550 0.086
b
Panel B: summary of country characteristics
Size of Size of Share-
overall Growth rate of Income Stock Stock market banking Creditor holder Rule of
Country Taxa. Inflat. economy real GDP group market size liquidity sector rights rights law Origin
Egypt 36.00 5.38 3.20 2.91 LMI 53.74 32.97 0.78 2.00 3.00 0.04 0.00
South Africa 29.50 5.31 3.51 2.53 UMI 201.47 48.02 0.73 3.00 5.00 0.12 1.00
Botswana 15.00 8.26 3.60 4.40 UMI 27.01 3.21 0.18 3.00 3.50 0.62 1.00
Ghana 29.90 17.93 2.43 2.82 LI 16.56 3.07 0.24 1.00 5.00 0.10 1.00
Kenya 30.30 8.82 2.62 1.15 LI 25.79 7.35 0.33 4.00 2.00 0.95 1.00
Mauritius 23.00 6.03 3.62 3.36 UMI 42.15 6.65 0.84 2.25 3.50 0.85 0.00
Morocco 35.00 1.66 3.17 2.93 LMI 44.57 18.76 0.64 1.00 2.00 0.03 0.00
Nigeria 25.00 11.76 2.61 2.92 LI 17.88 14.05 0.18 4.00 4.00 1.31 1.00
Tunisia 31.34 2.92 3.35 3.93 LMI 12.00 17.44 0.62 0.00 3.00 0.20 0.00
21.06 6.38 2.38 2.39 NA 20.66 7.84 0.34 1.56 2.48 0.08 NA
Panel C: summary of firm characteristics by countryc
Total Long-term Short-term Earnings Growth Asset Dividend
Country leverage leverage leverage Firm size volatility Profitability opport. tangibility payout Tax shield
Egypt 0.471 0.083 0.377 4.912 0.220 0.095 0.055 0.362 0.714 0.030
(0.296) (0.149) (0.235) (0.816) (0.246) (0.189) (0.191) (0.254) (0.898) (0.027)
South Africa 0.523 0.167 0.349 5.343 0.241 0.123 0.072 0.278 0.462 0.037
(0.261) (0.176) (0.199) (1.187) (0.238) (0.569) (0.233) (0.232) (0.981) (0.029)
Botswana 0.442 0.151 0.291 5.112 0.235 0.171 0.070 0.248 0.665 0.035
(0.167) (0.167) (0.173) (0.678) (0.265) (0.163) (0.213) (0.178) (0.805) (0.030)
Ghana 0.608 0.085 0.483 4.428 0.229 0.099 0.120 0.428 0.258 0.036
(0.418) (0.169) (0.249) (1.513) (0.203) (0.181) (0.086) (0.256) (0.303) (0.037)
(continued )
Kenya 0.509 0.200 0.309 5.322 0.184 0.121 0.054 0.410 0.487 0.036
(0.202) (0.157) (0.186) (1.060) (0.194) (0.140) (0.142) (0.220) (0.634) (0.025)
Mauritius 0.467 0.181 0.286 5.514 0.203 0.081 0.040 0.490 0.554 0.040
(0.211) (0.113) (0.188) (1.021) (0.223) (0.076) (0.092) (0.187) (0.560) (0.035)
Morocco 0.441 0.085 0.356 5.405 0.204 0.104 0.047 0.271 0.587 0.044
(0.221) (0.121) (0.182) (0.943) (0.238) (0.093) (0.140) (0.205) (0.587) (0.032)
Nigeria 0.649 0.098 0.504 5.449 0.234 0.206 0.056 0.600 0.248 0.018
(0.298) (0.155) (0.256) (0.971) (0.230) (0.629) (0.201) (0.354) (0.597) (0.026)
Tunisia 0.475 0.155 0.319 4.566 0.188 0.077 0.040 0.327 0.693 0.054
(0.241) (0.147) (0.182) (0.532) (0.213) (0.066) (0.092) (0.154) (0.677) (0.028)
Notes: aTotal leverage refers to the average of the ratio of total liabilities total assets. Long-term leverage refers to the average of the ratio of non-current liabilities to total assets.
Short-term leverage denotes the average of the ratio of current liabilities to total assets. Taxation refers to the average of the highest corporate marginal tax rate (%). Inflation
refers to the average of the consumer price index which is the annual percentage change in the cost to the average consumer of acquiring a fixed basket of goods and services
that may be fixed or changed at specified intervals, such as yearly. Size of economy is measured by the average of the logarithm of GDP per capita (constant 2,000 US$). Growth
of economy denotes the average of the logarithm of GDP per capita growth (constant 2,000 US$). Size of stock market refers to the average of the value of listed shares to GDP,
calculated using the following deflation method: (0.5) [Ft/P_et þ Ft-1/ P_et-1]/[GDPt/ P_at] where F is stock market capitalization, P_e is end-of period CPI, and P_a is
average annual CPI. Liquidity of stock market refers to the average of ratio of the value of total shares traded to average real market capitalization, the denominator is deflated
using the following method: Tt/ P_at/(0.5) [Mt/P_et þ Mt-1/P_et-1] where T is total value traded, M is stock market capitalization, P_e is end-of period CPI. P_a is average
annual CPI. Size of banking sector denotes the average of claims on domestic real non-financial sector by deposit money banks as a share of GDP, calculated using the following
deflation method: (0.5) [Ft/P_et þ Ft-1/P_et-1]/[GDPt/ P_at] where F is deposit money bank claims, P_e is end-of period CPI, and P_a is average annual CPI. Creditor rights
protection index refers to an index aggregating creditor rights, following La Porta et al. (1998). A score of one is assigned when each of the following rights of secured lenders is
defined in laws and regulations: first, there are restrictions, such as creditor consent or minimum dividends, for a debtor to file for reorganization. Second, secured creditors are
able to seize their collateral after the reorganization petition is approved, i.e. there is no “automatic stay” or “asset freeze”. Third, secured creditors are paid first out of the
proceeds of liquidating a bankrupt firm, as opposed to other creditors such as government or workers. Finally, if management does not retain administration of its property
pending the resolution of the reorganization. The index ranges from 0 (weak creditor rights) to 4 (strong creditor rights) and is constructed as at January for every year from 1978
to 2003. Shareholder rights protection index refers to an index of Anti-director rights is formed by adding one when: the country allows shareholders to mail their proxy vote;
shareholders are not required to deposit their shares prior to the General Shareholders ¼ Meeting; cumulative voting or proportional representation of minorities on the board of
directors is allowed; an oppressed minorities mechanism is in place; the minimum percentage of share capital that entitles a shareholder to call for an Extraordinary
Shareholders ¼ Meeting is less than or equal to 10 per cent (the sample median); or when shareholders have pre-emptive rights that can only be waived by a shareholders
meeting. The range for the index is from 0 to 6; bthe table presents average values for country level characteristics. All variables are averaged over the period 1999-2008. In the
Income group column, LMI refers to lower middle income group, UMI refers to upper middle income group, and LI refers to low-income group. The exact definition of the other
variables is as in panel A; cthe table presents mean (standard deviation in parenthesis) values for firm characteristics. All variables are averaged over the period 1999-2008, in
which data are required to be available at least for three years. Firm size refers to the average of the natural logarithm total sales. Earnings volatility refers to the average of
absolute value of first difference of the natural logarithm of profit after tax. Profitability refers to the average of the ratio of earnings before interest and taxes to total assets.
Growth opportunities refer to the average of the first difference of the natural logarithm of sales. Asset tangibility refers to the average of the ratio of tangible fixed assets to total
assets. Dividend payout refers to the average of the ratio of cash dividend paid to profit after tax. Tax shield refers to the average of the ratio of depreciation, amortization, and
depletion to total assets. The definitions of the other variables is as indicated in panel A
economies
developing
Evidence from
77
Table II.
JAAR unreported results (e.g. Antoniou et al., 2006; Barclay and Smith, 1995; Chen et al., 1999;
15,1 Deesomsak et al., 2009; Ozkan, 2002; Smith and Warner, 1979). Overall, the three salient
features identified signify that the capital structure of sample firms exhibited a
dynamic behaviour during the period under study.
Firm-specific determinants of adjustment speed of capital structure were picked
based on those often suggested in the literature (e.g. Drobetz et al., 2007; Song and
78 Philippatos, 2004). A perusal of Table II – panel C indicates that there is between and
within country variation in firm-specific variables. Whether these differences in

observed firm-characteristics have led to differences in adjustment speed is discussed
in a later section. Table II – panel C also shows that firms in developing African
countries on average are relatively smaller and yet more profitable compared to those
reported in Drobetz et al. (2007) for firms in France, Germany, Italy, and UK. Compared
to firms investigated in Cheng and Shiu (2007), the average profitability of firms in
developing African countries was somehow higher. As the theoretical prediction about
the influence of firm size on adjustment speed is inconclusive, what role has the
relatively smaller size of firms in developing countries played on their capital structure
adjustment speed is covered in a latter section. On the other hand, we anticipate that
the higher profitability of sample firms should lead to a relatively more rapid
adjustment.
A review of Table II – panel B reveals that sample countries varied considerably in
terms of their macroeconomic conditions. Furthermore, in unreported results based on
the World Development Indicators database, we observe that the size of overall
economy of sample countries was much smaller than the G-7 countries commonly
covered in capital structure studies while their GDP growth rates and inflation were
remarkably higher. We also note from this same table that there were considerable
variations in shareholder and creditor rights protection and the rule of law indices and
the sample countries exhibited comparatively lower levels of shareholder and creditor
rights protection and rule of law compared to the G-7 countries.
In Table III, we present Pearson’s pairwise correlation coefficients of variables
along with their statistical significance. We note that firm size is positively and
significantly associated with capital structure independent of how the latter is
defined. While earnings volatility is positively and significantly correlated, dividend
payout ratio is negatively and significantly associated with all the three measures of
capital structure. Also apparent is the inverse and statistically significant association
between profitability and all the three measures of capital structure. Not surprisingly,
the association between asset tangibility and capital structure is sensitive to how
the latter is defined; it is positively related with LTL-ratio and inversely related
with STL-ratio.
Our results also indicate that the association between most of the macroeconomic
and institutional variables and capital structure is dependent on which measure is
used in the analysis. For example, the highest marginal corporate tax rate, size of the
overall economy, and rule of law are negatively and significantly related with TL- and
STL-ratios while they are positively and significantly associated with LTL-ratios.
We also observe that creditor and shareholder rights protection indices are positively
associated with TL and LTL-ratios. The correlation matrices also show that the relative
size of a country’s banking sector is negatively and significantly associated with all
the three measures of capital structure. On the other hand, we note that the association
between measures of stock market development (i.e. its size and liquidity) and capital
structure is sensitive to how the latter is measured. Specifically, both measures of stock
Panel A: leverage and firm characteristicsa

Total Long-term Short-term Earnings Growth Asset
leverage leverage leverage Firm size volatility Profit. opprtun. tangibility Div. pay Tax shield
Total
leverage 1.000 *** 0.436 *** 0.744 *** 0.104 *** 0.030 * 0.085 ***0.002 0.085 ***0.095 ***0.009
Long-
term
leverage 0.436 *** 1.000 ***0.181 *** 0.023 * 0.061 ***0.052 *** 0.055 *** 0.230 ***0.099 *** 0.130 ***
Short-
term
leverage 0.744 ***0.181 *** 1.000 *** 0.120 ***0.008 0.039 ***0.019 0.309 ***0.039 * 0.096 ***
Panel B: leverage and macroeconomic variablesb
Size of Grwth of Size of stk Liq. of stk Size of Creditor Sharehol
Taxation Inflation economy economy mkt mkt bnk0 g rights rights Rule of law
Total
leverage 0.026* 0.043*** 0.033** 0.019 0.049*** 0.000 0.081*** 0.100*** 0.097*** 0.075***
Long-
term
leverage 0.052*** 0.037*** 0.122*** 0.045*** 0.155*** 0.068*** 0.032** 0.123*** 0.168*** 0.049***
Short-
term
leverage 0.130*** 0.068*** 0.114*** 0.019 0.050*** 0.049*** 0.042*** 0.012 0.011 0.111***
Panel C: pairwise correlation analysis of independent variablesc
Firm Earnings Growth Asset Div. Tax Size of Grwth of Size of Liq. of Size of Creditor Shareholder Rule of
size volatility Profit opprt. tang. pay shield Taxation Inflation economy economy stk mkt stk mkt bnk0 g rights rights law
[1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13] [14] [15] [16] [17]
[1] 1.000***
[2] 0.044*** 1.000***
[3] 0.077*** 0.011 1.000***
[4] 0.111*** 0.077*** 0.124*** 1.000 ***
[5] 0.020 0.009 0.016 0.018 1.000 ***
[6] 0.031 0.134*** 0.040** 0.121 ***0.015 ** 1.000 ***
[7] 0.029** 0.020 0.010 0.003 0.288 *** 0.045 ** 1.000 ***
(continued )
economies
developing
Evidence from
79
Correlation matrices
Table III.
80
15,1
JAAR
Table III.
[8] 0.186*** 0.006 0.050*** 0.039 ***0.042 *** 0.090 ***0.005 1.000 ***
[9] 0.023* 0.011 0.034** 0.048 *** 0.169 ***0.115 ***0.117 ***0.394 *** 1.000 ***
[10] 0.034** 0.022 0.031** 0.018 0.258 *** 0.127 *** 0.128 ***0.070 ***0.398 *** 1.000 ***
[11] 0.081*** 0.024 0.008 0.041 ***0.072 ***0.008 0.006 0.235 *** 0.150 *** 0.106 *** 1.000 ***
[12] 0.145*** 0.041** 0.029** 0.057 ***0.196 ***0.019 0.023 0.347 ***0.051 *** 0.649 *** 0.161 ***1.000 ***
[13] 0.092*** 0.022 0.008 0.029 ** 0.152 *** 0.034 * 0.010 0.213 ***0.026 ** 0.513 *** 0.363 ***0.696 ***1.000 ***
[14] 0.096*** 0.011 0.068*** 0.001 0.188 *** 0.157 *** 0.042 *** 0.533 ***0.475 *** 0.669 ***0.053 ***0.245 ***0.330 *** 1.000 ***
[15] 0.178*** 0.034** 0.061*** 0.026 * 0.078 ***0.124 ***0.101 ***0.384 *** 0.350 ***0.200 ***0.135 ***0.325 ***0.062 ***0.532 *** 1.000 ***
[16] 0.127*** 0.052*** 0.033 ** 0.044 ***0.084 ***0.111 *** 0.028 ** 0.346 *** 0.125 *** 0.435 ***0.041 ***0.747 ***0.388 ***0.073 *** 0.515 *** 1.000 ***
[17] 0.035** 0.021 0.011 0.005 0.176 *** 0.097 *** 0.096 *** 0.129 ***0.457 *** 0.852 ***0.080 ***0.338 ***0.217 *** 0.769 ***0.516 *** 0.144 *** 1.000 ***
Notes: The exact definition of the variables is as presented in Table II.aThe table reports the correlation coefficients between the three measures of leverage and firm-specific variables; bthe table
reports the correlation coefficients between the three measures of leverage and macroeconomic and institutional variables. cThe table reports the Pairwise correlation coefficients between the
independent variables. ***,**,* Correlation coefficients that are significantly different from 0 at the 1, 5, and 10 per cent levels respectively
market development are inversely related with STL-ratio while they have the opposite Evidence from
association with the other two leverage-ratios. developing
Finally, we note that the correlation coefficients between country-level determinants
of capital structure are very high. To keep the estimation problem tractable and economies
avoid problems of multicollinearity when estimating Equation (6) in the presence
of high correlations, we develop slightly different specifications of Equation (6) by
excluding highly correlated variables. 81
4.2 Determinants of target capital structure

Reasonable estimation of the dynamic model in Equation (6) crucially hinges on the
correct specification of the model for target capital structure. Hence, we estimate
Equation (1) using a two-way fixed effect (FE) regression procedure to examine if the
model is accurately specified. We also estimate Equation (1) using Seemingly
Unrelated Regression (SUR) for robustness check. Estimation results are presented in
Table IV[9].
Table IV shows that firm size is positively related with leverage and most
coefficients are significant at the 1 per cent level of significance. We also note that the
relationship between asset tangibility and capital structure is a function of the measure
of capital structure used in the analysis. Furthermore, the results indicate that firm
profitability, dividend payout ratio, and tax-shield are inversely related with capital
structure. Broadly, these findings are in tandem with main stream capital structure
theories. They also sit well in the club of many other efforts within the context of
developed and developing countries.
What is more, all the F-statistic and w2-statistic reject the null hypothesis that all
explanatory variables are simultaneously equal to zero. Also, the Hausman test rejects
the null hypothesis that the fixed effects estimator and the random effects estimator are
equivalent. We interpret this as favouring the fixed effects estimator. Overall, our
Short-term leverage Total leverage Long-term leverage

FE SUR FE SUR FE SUR
Earnings vol. 0.0140 0.0217 0.0002 0.0270 0.0109 0.0489***

Firm size 0.1030*** 0.0371*** 0.1360 *** 0.0457*** 0.0112 0.0057
Profit. 0.1780*** 0.1260** 0.3170 *** 0.2170 *** 0.1350*** 0.1100***
Growth oppt. 0.0095 0.0381 0.0275 0.0413 0.0024 0.0086
Asset tang. 0.0544 0.2820*** 0.0794 0.0894 ** 0.0907** 0.1750***
Div. payout 0.0039 0.0068 0.0032 0.0266 *** 0.0065** 0.0198***
Tax shield 0.4950** 0.1930 0.7900*** 0.0228 0.2380 0.2950
Constant 0.157 0.2470*** 0.2600 0.2520*** 0.0221 0.0279
F-statistic 3.54*** – 3.2 *** – 2.04*** –
w2 – 339.45*** – 121.67*** – 340.12***
Hausman test 58.28*** – 21.94 – 12.13 –
n 1,695 1,695 1,696 1,696 1,743 1,743
Notes: The table reports the results from fixed effects (FE) regressions (two way error component) Table IV.
and seemingly unrelated regressions (SUR) of leverage ratio on firm-specific capital Fixed effects and
structure determinants. F-test statistic for FE and w2 statistic for SUR are reported. seemingly unrelated
***,**,* Coefficients that are significantly different from 0 at the 1, 5, and 10 per cent levels are regressions for capital
marked respectively structure determinants
JAAR results are comparable to those in other similar studies and they indicate that the
15,1 explanatory variables are appropriate to model a time varying target capital structure
in a dynamic adjustment model.
4.3 Determinants of adjustment speed of capital structure

In this section, we report dynamic panel estimation results from Equation (6).
82 Dynamic panel estimation using sys-GMM allows estimation of all coefficients in
Equation (6) simultaneously. We begin our analysis by perusing the results for our
baseline regression model (Model 1) which specifies only firm-specific factors as the
independent variables. Table V presents estimates of Model 1.
As indicated in an earlier section, our focus is on the estimates of (1o1) and o1.
While (1o1) shows the movement of leverage-ratio to its target, o1 indicates whether
the speed of adjustment is independent of firm-specific characteristics included
in Model 1. The estimates of (1o1) for STL, LTL, and TL-ratios were 0.461, 0.410, and
0.606, respectively. This implies that firms in sample countries close by 53.9, 59.0,
and 39.4 per cent the gap between current and target STL, LTL, and TL ratios,
respectively, within one year. This means that a firm takes somehow less than two
years to reach its optimal STL (100 per cent divided by 53.9 per cent) and LTL (100 per
cent divided by 59.0 per cent) while it takes ca. two and half years (100 per cent divided
by 39.4 per cent) to reach it optimal TL. The difference in adjustment speeds is perhaps
Dependent variable Short-term leverage Long-term leverage Total leverage
LVi,t1 0.461*** 0.410*** 0.606 ***

(0.086) (0.097) (0.086)
LVi,t1 Sizei,t 0.033 0.116* 0.062**
(0.022) (0.062) (0.025)
LVi,t1 Profiti,t 0.810** 0.453 0.767**
(0.403) (0.944) (0.359)
LVi,t1 Grwthti,t 0.164 0.257 0.048
(0.185) (0.528) (0.231)
LVi,t1 Disti,t 0.128 8.038*** 0.814**
(0.361) (1.231) (0.354)
Constant 0.150** 0.042 0.095
(0.062) (0.031) (0.062)
Wald test 49.21*** 317.18*** 386.16***
Z2 1.148 0.974 0.753
Sargan test 96.738(107) 127.715(113) 100.641(107)
n 1,067 1,130 1,070
Notes: The table reports the results of estimating Equation (6) using sys-GMM estimator proposed by
Blundell and Bond (1998). Variations in sample size are due to data limitations. The table shows the
coefficients on the lagged leverage ratio and on the interaction term of the determinant of adjustment
speed with the lagged leverage ratio. Robust standard errors are in brackets. The Wald test statistic
refers to the null hypothesis that all coefficients on the determinants of target leverage ratio are jointly
equal to zero. The test statistic Z2 tests the null hypothesis of no second-order correlation in the
residuals. The Sargan test statistic refers to the null hypothesis that the overidentifying restrictions are
valid and uses the Blundell and Bond (1998) sys-GMM estimator. In parenthesis are the w2. Sizei,t refers
Table V. to size of firm i at time t. Profiti,t refers to profitability of firm i at time t. Grwthti,t refers to growth
Firm-specific factors and opportunities of firm i at time t. Disti,t refers to the value of the distance variable of frim i at time t. The
capital structure exact definition of the variables is as presented in Table II. ***,**,* Coefficients significantly different
adjustment – Model 1 from 0 at 1, 5, and 10 levels are marked respectively
due to the fact that TL includes STL and LTL. Such a rapid adjustment towards Evidence from
a target capital structure suggests the dominance of trade-off theory over rival theories developing
proposed by Baker and Wurgler (2002) and Welch (2004). It also suggests the presence
of costly and non-instantaneous adjustment towards target capital structure economies
(e.g. Flannery and Hankins, 2007; Leary and Roberts, 2005).
The relatively fast adjustment speed reported in Table V is consistent with the
comparatively higher profitability and growth that epitomized sample firms. It is also 83
consistent with the relatively high economic growth and inflation that characterized
the macroeconomic condition of sample countries. It is also in tandem with the
relatively faster adjustment speeds reported in Shyam-Sunder and Myers (1999) and
Flannery and Rangan (2006) for firms in the USA and Mukherjee and Mahakud (2010)
for Indian companies. Nonetheless, such an adjustment speed is in contrast with
what we would have expected given the relatively weaker legal institutions and less
developed financial institutions that characterized sample countries. Our interpretation
of this phenomenon is that the influence of macroeconomic conditions has, perhaps,
overwhelmed that of legal and financial institutions.
4.3.1 Firm-specific determinants of adjustment speed of capital structure. A perusal
of the estimates of o1 in Table V indicates that the nexus between distance variable
(Disti,t) and speed of adjustment is dependent on how capital structure is defined. The
further the observed STL and TL-ratios are from the target, the faster is the speed
of adjustment. This suggests that firm’s cost of maintaining a sub-optimal STL and TL
is higher than the cost of adjustment and the fixed costs of adjustment are not
significant. On the other hand, the negative relationship between adjustment speed of
LTL-ratio and the Disti,t variable suggests that adjustment costs (i.e. cost of external
financing, transaction cost, etc.) for long-term financing were prohibitively high for
sample firms. In situations where firms sidestep capital markets to adjust their capital
structure, they may take “extended excursions away from the optimal capital
structure” and only adjust their capital structures slowly as part of their normal
operations while larger adjustments require new issues of securities (Loof, 2004)[10].
This finding is consistent with our preliminary result which suggested that firm
financing in sample countries was dominated by short-term sources. Our result
corroborates the findings reported in Drobetz and Wanzenried (2006) and Mukherjee
and Mahakud (2010).
Furthermore, the estimated coefficients of profitability (see Table V) are all negative,
indicating a positive association between firm profitability and the pace at which
sample firms adjust their capital structure to the optimum. This is consistent with the
conjecture that more profitable firms have the flexibility and better access to raising
external finance and hence adjust their capital structure more rapidly than less
profitable firms. Similar results were reported in Flannery and Rangan (2006) and
Song and Philippatos (2004).
We observe that the nexus between firm size and adjustment speed is sensitive to
how we measure capital structure. The results indicate that firm size enhances
adjustment speed of LTL-ratio while it deters those of STL and TL-ratios. This
suggests that the potential cost of distress that is likely to be lesser in larger firms is an
important factor in the adjustment decisions of sample firms. As larger firms tend to
have lesser cost of financial distress, they tend to be reluctant to adjust their long-term
capital structure. On the other hand, the lower transaction costs seem to play
a dominant role in the adjustment decision of short-term, and thereby total, leverage
of sample firms. Also evident in Table V is that the growth opportunities variable
JAAR has a statistically weak and definitionally sensitive relationship with adjustment
15,1 speed. This result is in contrast with what was reported in Drobetz and Wanzenried
(2006). Specifically, the finding in Drobetz and Wanzenried (2006) and Mukherjee and
Mahakud (2010) that growing firms adjust faster than no-growth firms could not be
confirmed.
4.3.2 Inter-industry heterogeneity of adjustment speed of capital structure. We now
84 attempt to examine whether the adjustment speeds reported in Table V persist when
we estimate Model 1 on an industry-by-industry basis. Estimation results for each

industry are reported in Table VI – panel A. For reasons of brevity, we report
only coefficients of lagged leverage-ratios along with the corresponding robust
standard errors and number of observations.
The results of industry-by-industry analysis for ten industries indicate that
adjustment speeds vary across industries. On a STL-ratio basis, firms within the
Durables and Chemicals and Construction industries move towards their target capital
structures relatively rapidly than those in other industries. The adjustment speed of
short-term capital structure for these industries is about 57.5 per cent per year. On a
LTL-ratio basis, firms within the Health, Oil and Gas, and Regulated industries move
towards their target capital structures relatively rapidly than those in other industries.
The adjustment speeds of long-term capital structure for these industries are between
66.7 and 91.6 per cent per year. These industries generally had high levels of leverage-
ratios, which may indicate that they were comparatively riskier than other industries
(see Table VI-Panel B). When firms in these industries deviate from their target capital
structure, in particular take on additional debt, they might increase their risk even
further. Consequently, they might try to adjust back towards their target capital
structure faster than firms in comparatively less risky industries (Smith et al., 2010).
In contrast, on a STL-ratio basis, firms within the Business Equipment, Wholesale
and Retail, and Health industries adjust their capital structures relatively slowly
towards their target (Table VI – panel A). On a LTL-ratio basis, on the other hand,
firms within the Durables, Service, and Wholesale and Retail industries adjust their
capital structures relatively slowly towards their target. While firms within the Health
industry have low short-term leverage-ratios, those in Wholesale and Retail industry
have low LTL-ratios (Table VI – panel B). This might point that these industries are
relatively less risky. Therefore, when firms in these industries deviate from their target
capital structure, and in particular take on additional debt, they might feel less
pressure to adjust back to the target quickly (Smith et al., 2010).
Broadly speaking, we observe that a majority of the industries in our sample
exhibited an inverse relationship between level of leverage-ratios (i.e. financial risk)
and speed of adjustment. Taking a cue from Smith et al. (2010), our evidence indicates
that industry’s financial risk may be the underlying factor determining adjustment
speed. That is, firms in industries that have relatively high financial risk tend to revert
more quickly to their target capital structure than those in industries that have
relatively low risk. Nonetheless, this conclusion should not be taken as anything more
than a pointer since only a proper modelling of more specific measure of industry risk
and its relationship with adjustment speed can better establish the influence of the
former on the latter.
4.3.3 Cross-country heterogeneity of adjustment speed of capital structure. To gain an
idea about cross-country variation in adjustment speeds, we estimate Model 1 for each
country included in our sample. For reasons of brevity, we report only coefficients of
lagged leverage-ratios[11]. On a STL-ratio basis, our results (Table VII – panel A) show
Panel A: inter-industry heterogeneity in adjustment speeds of capital structurea
Evidence from
Short-term Long-term Total developing
Dependent variable leverage leverage leverage
Non-durable industry 0.616 *** 0.549 ** 0.725 *** economies
(0.165) (0.259) (0.141)
1,006 1,055 1,011
Durable industry 0.424 * 0.929 ** 0.686 ***
85
(0.284) (0.408) (0.192)

167 170 167
Manufacturing industry 0.563 *** 0.564 *** 0.662 ***
(0.134) (0.162) (0.102)
921 958 922
Oil and Gas industry 0.523 0.255 0.208
(0.386) (1.414) (1.302)
385 383 386
Chemicals and
Construction industry 0.426 0.370 *** 0.550 ***
(0.852) (0.131) (0.174)
523 536 523
Business Equipment
industry 0.902 ** 0.537 0.865 ***
(0.387) (0.894) (0.223)
346 350 346
Regulated industry 0.728 0.333 ** 0.839
(2.279) (0.152) (1.494)
304 310 305
Wholesale and Retail
industry 0.764 0.592 *** 0.841
(0.882) (0.138) (0.819)
697 748 705
Health industry 0.739 0.084 0.467
(0.503) (3.134) (0.791)
283 294 283
Service industry 0.453 *** 0.643 *** 0.649 ***
(0.199) (0.208) (0.156)
814 862 814
Panel B: summary statistics of leverage by industry
Short-term leverage Long-term leverage Total leverage
Mean SD Obs Mean SD Obs Mean SD Obs
Non-durables 0.345 0.209 1006 0.109 0.159 1055 0.467 0.288 1011
Durables 0.342 0.178 167 0.088 0.115 170 0.432 0.212 167
Manufacturing 0.357 0.194 921 0.124 0.176 958 0.482 0.245 922
Oil and Gas 0.265 0.233 385 0.197 0.206 383 0.477 0.321 386
Chem. and Construction 0.445 0.224 523 0.108 0.164 536 0.555 0.230 523
Business Equipment 0.429 0.243 346 0.078 0.105 350 0.526 0.316 346
Regulated 0.367 0.200 304 0.182 0.194 310 0.546 0.226 305
Wholesale and Retail 0.428 0.229 697 0.095 0.119 748 0.545 0.309 705
Health 0.352 0.189 283 0.074 0.138 294 0.435 0.232 283
Service and others 0.318 0.226 814 0.132 0.160 862 0.462 0.293 814
Notes: SD, standard deviation; Obs, number of observations. The exact definition of the industries
is as reported in Table I. aThe table reports the parameter estimates of the one-period lagged dependent
variable and the corresponding robust standard errors and number of observations. Equation (6) was
estimated using sys-GMM estimator proposed by Blundell and Bond (1998) for each industry in the sample.
For reasons of brevity, we do not report parameter estimates and related details of firm-specific variables Table VI.
included in the model. Robust standard errors are in brackets. And the figure in a third raw is the number of Capital structure and
observations. ***,**,* Coefficients significantly different from 0 at 1, 5, and 10 per cent levels are marked its adjustment speed
respectively by industry
JAAR Dependent variable Short-term leverage Long-term leverage Total leverage
15,1
Egypt 0.531*** 0.442***` 0.658***
(0.096) (0.098) (0.133)
2,685 2,702 2,697
South Africa 0.816 0.178 0.773***
86 (0.853) (0.645) (0.136)
1,664 1,663 1,665

Kenya 0.349*** 0.178*** 0.022
(0.135) (0.056) (0.092)
150 163 151
Morocco 0.750 0.949*** 0.647***
(0.833) (0.141) (0.318)
288 289 288
Tunisia 0.539*** 0.131 0.360
(0.179) (0.181) (0.452)
176 177 176
Notes: The table reports parameter estimates of the one-period lagged dependent variable and the
corresponding robust standard errors and number of observations. Equation (6) was estimated using
sys-GMM estimator proposed by Blundell and Bond (1998) for each country in the sample. The results
Table VII. of four countries including Botswana, Ghana, Mauritius, and Nigeria were not included owing to
Cross-country sample size issues. For reasons of brevity, we do not report parameter estimates and related details of
heterogeneity in firm-specific variables included in the model. Robust standard errors are in parentheses. And the
adjustment speed of figure in a third raw is the number of observations. ***,**,* Coefficients significantly different from 0
capital structure at 1, 5, and 10 per cent levels are marked respectively
that firms in Kenya adjust at the fastest rate (1 0.349 ¼ 0.651) while those in South
Africa adjust at the slowest rate (10.816 ¼ 0.184). On a LTL-ratio basis, firms in
Kenya adjust at the fastest rate (10.178 ¼ 0.822) while those in Morocco adjust at the
slowest rate (10.949 ¼ 0.051)[12]. These results suggest two important points. First,
they suggest that there is, indeed, a cross-country variation in capital structure
adjustment speeds. Second, it appears that the positive influence of inflation on capital
structure adjustment speed dominates that of other macroeconomic and institutional
variables. A joint consideration of results in Table VII and Table II – panel B reveals
that firms in countries with higher inflation tend to more rapidly adjust their capital
structure to the optimum than those in a less inflationary situation regardless of the
influence of other country- and firm-specific factors.
Earlier, we hypothesized that legal institutions should determine the adjustment
speed of firm’s capital structure. To this end, we examine the dynamics by splitting
the sample into firms from countries with common law and civil law traditions[13].
We estimate Model 1 for firms in each sub-sample. Table VIII reports the adjustment
speeds for each sub-sample. As in the previous analyses, we report only coefficients of
lagged leverage-ratios.
Consistent with our expectation, we observe that firms in the common law sub-sample
adjust to target capital structures at a relatively faster speed than is the case with the civil
law sub-sample (see Table VIII). The difference in the adjustment speeds is sharper when
one considers STL and LTL-ratios separately than TL-ratio. Similar results were reported
in other studies (e.g. Öztekin and Flannery, 2012). This variation in adjustment speeds
does strengthen the hypothesis that legal institutions influence the adjustment costs
and/or benefits, and hence, the adjustment speed of capital structure of firms.
Evidence from
developing
Common Law 0.430 *** 0.282 0.619 ** economies
(0.151) (1.309) (0.291)
3,322 3,341 3,334
French Law 0.527 *** 0.469 *** 0.641 ***
(0.086) (0.100) (0.080) 87
2,122 2,325 2,128

sys-GMM estimator proposed by Blundell and Bond (1998) for each legal family. For reasons of
brevity, we do not report parameter estimates and related details of firm-specific variables included in Table VIII.
the model. Robust standard errors are in parentheses. And the figure in a third raw is the number of Heterogeneity in
observations. ***,**,* Coefficients significantly different from 0 at 1, 5, and 10 per cent levels are adjustment speeds
marked respectively across legal origin
We further examine capital structure adjustment speeds by trifurcating our sample

into sub-samples of income groups: upper-middle-income; lower-middle-income; and
low-income countries. We consider these sub-samples because the results might help us
uncover the influence that economic development has on capital structure dynamics
that we could not capture through other variables. We carry out separate estimates of
Model 1 for each sub-sample. Table IX reports the variation in adjustment speeds of
capital structure of firms in all the three sub-samples. Again, only coefficients of the
lagged leverage-ratio variable are reported.
Consistent with the hypothesis that the level of development of a country influences
adjustment speed of capital structure of firms, our results show that adjustment speeds
vary across income groups. Specifically, on a STL-ratio basis, firms in richer countries
tend to have a slower adjustment speed than is the case in poorer countries. We observe
similar results for TL-ratio (see Table IX). This variation in speeds of adjustment is
consistent with the view that the relative costs and/or benefits of deviating from target
Upper middle income countries 0.802 0.099 0.775***

(0.618) (0.221) (1.379)
1,911 1,910 1,912
Lower middle income countries 0.539*** 0.471*** 0.648***
(0.136) (0.105) (0.081)
3,149 3,168 3,161
Low income countries 0.190 0.310*** 0.533
(0.432) (0.080) (0.597)
388 588 389
sys-GMM estimator proposed by Blundell and Bond (1998) for each income group family. For reasons
of brevity, we do not report parameter estimates and related details of firm-specific variables included Table IX.
in the model. Robust standard errors are in parentheses. And the figure in a third raw is the number of Heterogeneity in
observations. ***,**,* Coefficients significantly different from 0 at 1, 5, and 10 per cent levels are adjustment speeds across
marked respectively income groups
JAAR capital structure vary across income levels. Hence, the (net) adjustment cost, based on
15,1 Table IX, is highest for upper-middle-income countries, followed by lower-middle-income
countries and low-income countries. This result strengthens the notion that
macroeconomic factors influence the speed of adjustment (Hackbarth et al., 2006;
Wanzenried, 2006). However, it is not consistent with the proposition that firms in less
developed countries actually adjust their capital structure at a slower rate than those in
88 developed countries.
We now proceed to examine the influence of more-narrowly-defined aspects of

macroeconomic and institutional environment on adjustment speed by introducing a
set of more specific set of legal, financial, and macroeconomic variables into Equation
(6) – Model 2. Specifically, we include such variables as size of the overall economy,
growth rate of the economy, corporate tax rate, inflation, stock market size, stock
market liquidity, size of banking sector, creditor rights protection, shareholder rights
protection and rule of law. As has been pointed out earlier, most of these country-level
variables were severely correlated. Thus, putting all of them into one model would
result in multicollinearity problem (see Table III – panel C). To avoid the
multicollinearity problem, we develop variants of Model 2 (i.e. Model 2a-Model 2g)
each of which encompass only less severely correlated independent variables. Table X
presents model estimates for each measure of capital structure.
The estimated coefficient on the interaction term with the highest marginal
corporate tax rate indicates that firms in countries with higher marginal corporate tax
rates adjust faster towards their target leverage-ratio implying that higher untapped
tax benefits enhance the pace at which firms adjust to their target capital structure.
Empirical works by Öztekin and Flannery (2012) and Clark et al. (2009) report similar
results. In line with predictions by Mills (1996), our results indicate a statistically weak
but positive relationship between inflation and speed of capital structure (see Table X).
This confirms the hypothesis that inflationary situations lead to increased cost of
capital for sub-optimal capital structure, and hence, lead to higher adjustment speed
towards optimal capital structure. This result is in sync with the findings reported in
Wanzenried (2006).
Our results show that the nexus between the overall size of the economy and its
growth rate, on the one hand, and capital structure, on the other, is a function of how
the latter is measured. In particular, we observe a negative but weak relationship
between GDP per capita growth rate and adjustment speed of STL-ratio and a positive
relationship for LTL and TL-ratios (see Table X). This partially confirms Hackbarth et
al.’s (2006) argument that lower restructuring thresholds during periods of high GDP
per capita growth lead to faster capital structure adjustment speeds. In a study of firms
drawn from four European countries, Wanzenried (2006) reports similar results. What
is more, we note that overall size of the economy and adjustment speed of short-term
leverage-ratio are negatively related while the former positively influences the
adjustment speed of LTL-ratio (see Table X).
Dependable legal systems, which assure investors that they receive promised cash
flows, enhance capital market transactions (e.g. Öztekin and Flannery, 2012). Although
statistically weak, we find that adjustment speeds of both STL and LTL-ratio are faster
in countries with stronger shareholder rights protection. However, an opposite
relationship emerges when TL-ratio is considered (see Table X). These partially
confirm the hypothesis that firms in countries with better protection to shareholder
rights exhibit faster capital structure adjustment speed than is the case in countries
with poor shareholder rights protection.
Panel A:dependent variable – short-term leverage ratioa

LVi,t1 Model 2 (a) Model 2 (b) Model 2 (c) Model 2 (d) Model 2 (e) Model 2 (f) Model 2 (g)
0.456*** (0.093) 0.452*** (0.083) 0.521*** (0.100) 0.462*** (0.097) 0.469*** (0.087) 0.498*** (0.112) 0.480*** (0.105)
LVi,t1 Profiti,t 0.609* (0.343) 0.626* (0.337) 0.906** (404) 0.458 (0.384) 0.537 (0.380) 0.626 (0.411) 0.455 (0.349)
LVi,t1 Grwthti,t 0.251 (0.186) 0.123 (0.178) 0.143 (0.219) 0.129 (0.204) 0.226 (0.209) 0.198 (0213) 0.231 (0.213)
LVi,t1 Disti,t 0.440 (0.388) 0.061 (0.286) 0.176 (0.348) 0.213 (0.462) 0.392 (0.421) 0.150 (0.449) 0.432 (0.353)
LVi,t1 GDPGi,t 0.003 (0.010)
LVi,t1 SRi,t 0.077 (0.052)
LVi,t1 RULi,t 0.055 (0.084) 0.075 (0.094) 0.105 (0.090)
LVi,t1 TAXi,t 0.001 (0.001)
LVi,t1 STKLIQi,t 0.050 (0.135)
LVi,t1 INFLi,t 0.001 (0.004) 0.001 (0.004)
LVi,t1 STKSIZi,t 0.097** (0.048)
LVi,t1 BNKSIZi,t 0.169 (0.181)
LVi,t1 CRi,t 0.080* (0.048)
LVi,t1 LOGGDPi,t 0.015 (0.070)
Constant 0.170***(0.047) 0.204***(0.046) 0.168*** (0.051) 0.161** (0.063) 0.154* (0.086) 0.174** (0.072) 0.229*** (0.079)
Wald test 42.99*** 43.68*** 58.28*** 41.35*** 37.99*** 31.21*** 54.09***
Z2 0.155 1.471 0.295 0.483 0.264 0.262 1.184
Sargan test 82.186 103.280 90.687 80.335 85.404 89.233 110.194
n 5,444 5,441 5,435 5,440 5,429 5,437 5,428
Panel B:dependent variable – long-term leverage ratiob
0.463*** (0.135) 0.396*** (0.118) 0.476*** (0.152) 0.460* (0.244) 0.428*** (0.124) 0.492*** (0.095) 0.413*** (0.095)
LVi,t1 Profiti,t 1.188 (1.197) 0.551 (0.881) 1.500* (0.859) 0.989 (0.244) 1.220 (1.012) 1.360* (0.740) 0.692 (0.834)
LVi,t1 Grwthti,t 0.527 (0.596) 0.615 (0.391) 0.471 (0.494) 0.421 (0.379) 0.363 (0.407) 0.331 (0.378) 0.555 (0.453)
LVi,t1 Disti,t 6.460*** (1.461) 8.109*** (1.424) 7.713*** (2.173) 7.153*** (2.318) 7.334*** (1.597) 6.750*** (1.404) 7.690*** (1.347)
LVi,t1 GDPGi,t 0.001 (0.021)
LVi,t1 SRi,t 0.095 (0.101)
LVi,t1 RULi,t 0.140 (0.165) 0.074 (0.145) 0.104 (0.155)
(continued )
economies
developing
adjustment speed of
Determinants of
Evidence from
capital structure – Model 2

Table X.
89
90
15,1
JAAR
Table X.
LVi,t1 TAXi,t 0.011*** (0.004)
LVi,t1 INFLi,t 0.013 (0.013) 0.003 (0.017)
LVi,t1 STKSIZi,t 0.276** (0.133)
LVi,t1 CRi,t 0.050 (0.160)
LVi,t1 LOGGDPi,t 0.211* (0.131)
Constant 0.004 (0.026) 0.022 (0.016) 0.010 (0.011) 0.041* (0.024) 0.059* (0.034) 0.006 (0.020) 0.023 (0.022)
Wald test 294.04*** 299.46*** 331.24*** 237.02*** 241.09*** 286.83*** 371.00***
2
Z 0.812 1.171 0.787 1.034 0.833 0.950 0.986
Sargan test 97.645 118.615 105.800 96.548 102.111 102.877 132.440
n 5,666 5,662 5,645 5,658 5,661 5,659 5,652
Panel C: dependent variable – total leverage ratiob
0.612*** (0.096) 0.634*** (0.101) 0.699*** (0.077) 0.568*** (0.109) 0.586*** (0.090) 0.664*** (0.083) 0.620*** (0.076)
LVi,t1 Profiti,t 0.594* (0.309) 0.553 (0.366) 0.799** (0.393) 0.610** (0.300) 0.766* (0.446) 0.630* (0.355) 0.740* (0.434)
LVi,t1 Grwthti,t 0.132 (0.179) 0.078 (0.188) 0.064 (0.201) 0.076 (0.132) 0.090 (0.179) 0.083 (0.165) 0.086 (0.162)
LVi,t1 Disti,t 0.180 (0.246) 0.449* (0.251) 0.256 (0.236) 0.320 (0.257) 0.516* (0.285) 0.280 (0.206) 0.435 (0.307)
LVi,t1 GDPGi,t 0.003 (0.012)
LVi,t1 SRi,t 0.008 (0.031)
LVi,t1 RULi,t 0.089 (0.089) 0.070 (0.094) 0.040 (0.083)
LVi,t1 TAXi,t 0.001 (0.001)
LVi,t1 INFLi,t 0.001 (0.004) 0.001 (0.003)
LVi,t1 STKSIZi,t 0.073** (0.030)
LVi,t1 CRi,t 0.124** (0.055)
LVi,t1 LOGGDPi,t 0.050 (0.051)
(continued )
Constant 0.201*** (0.058) 0.221*** (0.061) 0.167*** (0.050) 0.184*** (0.067) 0.168** (0.083) 0.185*** (0.065) 0.204*** (0.061)
Wald test 105.49*** 104.08*** 176.15*** 196.97*** 83.37*** 125.91*** 118.35***
Z2 0.288 1.432 0.347 0.102 0.159 0.234 1.110
Sargan test 83.319 109.630 85.574 81.504 80.504 82.596 116.237
n 5,462 5,448 5,432 5,455 5,461 5,460 5,441
Notes: aGDPGi,t refers to the growth rate of real GDP of the country in which firm i operates at time t. SRi,t refers to the shareholder rights protection index of the
country in which firm i operates at time t. RULi,t refers to the rule of law index of the country in which firm i operates at time t. TAXi,t refers to the highest corporate
marginal tax rate of the country in which firm i operates at time t. STKLIQi,t refers to stock market liquidity of the country in which firm i operates at time t. INFi,t refers
to inflation rate of the country in which firm i operates at time t. STKSIZi,t refers to stock market capitalization of the country in which firm i operates at time t. BNKSIZi,t
refers to the relative size of banking sector of the country in which firm i operates at time t. CRi,t refers to creditor rights index the country in which firm i operates at time
t. LOGGDPi,t refers to natural logarithm of the GDP of the country in which firm i operates at time t. The exact definition of the other variables is as presented in Table II.
The table reports the results of estimating Equation (6) using sys-GMM estimator proposed by Blundell and Bond (1998). Variations in sample size are due to data
limitations. The table shows the coefficients on the lagged leverage ratio and on the interaction term of the determinant of adjustment speed with the lagged leverage
ratio. Robust standard errors are in parentheses. The Wald test statistic refers to the null hypothesis that all coefficients on the determinants of target leverage ratio are
jointly equal to zero. The test statistic Z2 tests the null hypothesis of no second-order correlation in the residuals. The Sargan test statistic refers to the null hypothesis
that the overidentifying restrictions are valid and uses the Blundell and Bond (1998) sys-GMM estimator. ***,**,* Coefficients significantly different from 0 at 1, 5, and
10 per cent levels are marked respectively; bthe exact definition of the other variables is as presented in Tables II and X – panel A. The table reports the results of
estimating Equation (6) using sys-GMM estimator proposed by Blundell and Bond (1998). Variations in sample size are due to data limitations. Disti,t is constructed as
the fitted values from a fixed effects (two way error component) regression of the respective measures of leverage on the eight capital structure determinants. The table
shows the coefficients on the lagged leverage ratio and on the interaction term of the determinant of adjustment speed with the lagged leverage ratio. Robust standard
errors are in parentheses. The Wald test statistic refers to the null hypothesis that all coefficients on the determinants of target leverage ratio are jointly equal to zero. The
test statistic Z2 tests the null hypothesis of no second-order correlation in the residuals. The Sargan test statistic refers to the null hypothesis that the overidentifying
restrictions are valid and uses the Blundell and Bond (1998) sys-GMM estimator. ***,**,* Coefficients significantly different from 0 at 1, 5, and 10 per cent level are
marked respectively
economies
developing
Evidence from
Table X.
91
JAAR In contrast to Öztekin and Flannery (2012) and Clark et al. (2009), a negative
15,1 relationship is revealed between adjustment speeds of STL and TL-ratios and creditor
rights protection (see Table X). On the other hand, although statistically weak, we
observe that creditor rights protection positively influences the adjustment speed of
LTL-ratio. The statistically significant negative relationship does not support our
hypothesis that a stronger protection of creditor rights leads to a faster capital
92 structure adjustment speed. In addition, albeit statistically weak, we observe that
better law enforcement tends to positively impact on the adjustment speed of capital
structure (see Table X). This is in harmony with the hypothesis that better law
enforcement positively affects adjustment speed and also in agreement with results
reported in Öztekin and Flannery (2012) and Clark et al. (2009).
We observe that stock market size has a statistically strong but definitionally
sensitive influence on adjustment speed. To be exact, it has a negative influence on
adjustment speed of STL and TL ratios while it has a positive influence on that of
LTL-ratio. We observe, more or less, similar results for the stock market liquidity
variable. These results vindicate Deesomsak et al. (2009) who argue that developed
stock markets, by reducing information asymmetry, may trigger firms to switch to
long-term debt. As such, firms may rapidly adjust their LTL than STL ratio in
countries with bigger and developed stock markets.
In line with extant literature (e.g. Öztekin and Flannery, 2012), our results show a
statistically weak but positive relationship between relative size of banking sector and
capital structure adjustment speed. Thus, our result supports the proposition that
firms in countries with more developed banking sector adjust their capital structure
more rapidly than is the case in countries with less developed banking sector.
5. Conclusions
In this paper, we extended the debate on capital structure decisions of firms in
developing countries along the lines of empirical endeavours in advanced
economies. We contended that capital structure of firms in developing economies
displays target behaviour and the pace at which firms adjust their capital structure
to a target is a function of not only firm characteristics but also of industrial,
institutional, and macroeconomic factors. We examine the data using sys-GMM
panel data estimator, which is robust to firm heterogeneity and variable
endogeneity problems.
The paper presented evidence that capital structure of firms in developing countries
not only converges to a target but also that it faces varying degrees of adjustment costs
and/or benefits in doing so. This suggests not only that dynamic trade-off theory
explains capital structure decisions of firms but also rules out the dominance of
information asymmetry-based theories within the context of firms in developing
countries.
Also, the study established that the extent of costs and/or benefits of adjustment
that firms in developing countries face is determined, inter alia, by firm-specific factors
such as firm profitability, size, growth opportunities, and the gap between observed
and target capital structure. Furthermore, except for firm profitability, which
positively influences adjustment speed, we observe that the nature of influence that
firm-specific characteristics exert on adjustment costs and/or benefits is a function
of how we measure capital structure. The role that firm-specific characteristics play in
the determination of adjustment speed suggests that financing costs, financial
flexibility, access to external finance, the potential cost of distress and the value of
debt-related tax-shields are at play in aggravating or mitigating adjustment costs Evidence from
and/or benefits. developing
In terms of inter-industry differences in adjustment costs and/or benefits, we note
once again that the relationships are sensitive to how one defines capital structure. economies
On a STL-ratio basis, firms within the Durables and Chemicals and Construction
industries move towards their target capital structures relatively faster than is the case
in other industries. In contrast, on a STL-ratio basis, firms within the Health, Oil and 93
Gas, and Regulated industries move towards their target capital structures relatively
quicker compared to other industries. A further investigation points to the tendency
that firms in riskier industries adjust faster than those in less risky industries. This
implies that cost of bankruptcy has an important place in determining adjustment
costs and/or benefits of sample firms. However, as the present study did not include
variables defining industry characteristics (such as industry risk) into the models,
the conclusions pertaining to inter-industry variation in adjustment speed should be
considered with caution.
In addition, consistent with the view that adjustment costs should be lower and/or
adjustment benefits should be higher in common law origin countries; we observe
evidence that firms in countries with common law tradition tend to more rapidly
adjust their capital structure than is the case in countries with civil law system. In
terms of more-narrowly defined institutional variables, we observe that shareholder
rights protection and rule of law, in contrast to creditor rights protection, have positive
influence on capital structure adjustment speed. The implication of these findings is
that investor protection and contract enforceability are important matters in the
determination of adjustment costs and/or benefits of sample firms.
The present study also proffers evidence that more developed banking sectors and
stock markets deter the pace at which firms adjust their STL and TL-ratios. Contrary
to expectation, adjustment speeds of ST and LTL-ratios are slower in richer countries
than is the case in poorer countries. Furthermore, firms in countries which have higher
marginal corporate tax rate and inflation tend to have faster adjustment speed. Put
together, the evidences again suggest that access to external finance and tax issues are
central to the determination of adjustment costs and/or benefits of sample firms.
Overall, the study points out that the adjustment speed of capital structure of
sample firms towards the optimum is influenced by a host of firm, industry,
and country-level variables. We draw two important implications for corporate
finance. First, to the extent that firm managers could exert their influence on firm
characteristics, they could influence capital structure adjustment speed and hence
optimum cost of capital. Second, to the extent that regulators could exert their
influence on country level variables, they could influence the rate at which firms
rebalance their capital structure towards the optimum level and, hence their cost
of capital.
Notes
1. While there is a separate and growing literature on the adjustment speed of specific
dimensions of capital structure including debt maturity structure, the review in this section
has purposely focused on the literature pertaining to the adjustment speed of “basic capital
structure” in line with the aim of the paper.
2. See Haugen and Senbet (1978), Barnea et al. (1980), DeAngelo and Masulis (1980), Kim (1982),
and Modigliani (1982) for elaborate discussions on the [ir]relevance of bankruptcy costs to
capital structure decisions.
JAAR 3. Leary and Roberts (2005) provide an elaborate discussion on the implication of the structure
of adjustment costs on the adjustment speed of capital structure.
15,1
4. These measures are different from those commonly used to measure debt maturity structure
of a firm. For elaborate discussions on various measures of debt maturity structure, see
Barclay and Smith (1995), Stohs and Maur (1996), Antoniou et al., 2006, Deesomsak et al.
(2009), and Lemma and Negash (2012, 2013a).
94
5. We thank Andrei Shelifer for making data pertaining to creditor rights, shareholder rights
and legal origin freely available on his page (www.economics.harvard.edu/faculty/shleifer/
dataset).
6. The classification of countries by income groups is a contentious issue and surrounded by
fierce debate. Different institutions (e.g. the World Bank, IMF, the economist, CIA, etc.) use
different criteria for different purposes to classify countries. The classification in this study
was based on the World Bank’s income group of countries. Hence, Botswana, Mauritius, and
South Africa fall under upper-middle-income countries; Egypt, Morocco, and Tunisia fall
under to lower-middle-income; and Ghana, Kenya, and Nigeria fall under low-income
countries.
7. Comparisons in most studies make reference to Rajan and Zingales (1995). However, since we
note that Cheng and Shiu (2007) is more recent and comprehensive, we opted to compare our
results with Cheng and Shiu (2007).
8. Average leverage ratio of firms in our sample countries appear to be invariably greater than
the five countries sampled in Gwatidzo and Ojah (2009). These differences may probably
have resulted due to the bigger sample we examined and some differences in definitions of
leverage ratios.
9. The inconsistency in the number of observations for the three measures of leverage in
Table IV and subsequent tables is due to missing data points.
10. Note that Equation (6) specifies a negative sign on n o1, and therefore the signs of the
estimated coefficients on the respective interaction terms must be interpreted accordingly.
11. The figures in parenthesis are robust standard errors.
12. The results of the other four countries including Botswana, Ghana, Mauritius, and Nigeria
were not reported owing to sample size issues.
13. While Botswana, Ghana, Kenya, Nigeria, and South Africa have common law legal systems,
Egypt, Mauritius, Morocco, and Tunisia have civil law legal systems.
References
Antoniou, A., Guney, Y. and Paudyal, K. (2006), “The determinants of debt maturity structure:
evidence from France, Germany and the UK”, European Financial Management, Vol. 12
No. 2, pp. 161-194.
Antoniou, A., Guney, Y. and Paudyal, K. (2008), “The determinants of capital structure: capital
market-oriented versus bank-oriented institutions”, Journal of Financial and Quantitative
Analysis, Vol. 43 No. 1, pp. 59-92.
Baker, M. and Wurgler, J. (2002), “Market timing and capital structure”, The Journal of Finance,
Vol. 57 No. 1, pp. 1-32.
Banerjee, S., Heshmati, A. and Wihlborg, C. (2004), “The dynamics of capital structure”,
Monetary Integration, Markets and Regulation, Vol. 4 No. 1, pp. 274-297.
Barclay, M. and Smith, C. Jr (1995), “The maturity structure of corporate debt”, Journal of
Finance, Vol. 50 No. 2, pp. 609-631.
Barclay, M. and Smith, C. Jr (1999), “The capital structure puzzle: another look at the evidence”, Evidence from
Journal of Applied Corporate Finance, Vol. 12 No. 1, pp. 8-20.
developing
Barnea, A., Haugen, R. and Senbet, L. (1980), “A rationale for debt maturity structure and
call provisions in the agency theoretic framework”, Journal of Finance, Vol. 35 No. 5, economies
pp. 1223-1234.
Benito, A. (2003), “The capital structure decisions of firms: is there a pecking order?”, Documento
de Trabajo, No. 310, pp. 1-41. 95
Berkowitz, D., Pistor, K. and Richard, J.F. (2003), “Economic development, legality, and the
transplant effect”, European Economic Review, Vol. 47 No. 1, pp. 165-195.
Bevan, A.A. and Danbolt, J. (2002), “Capital structure and its determinants in the United
Kingdom – a decompositional analysis”, Applied Financial Economics, Vol. 12 No. 3,
pp. 159-170.
Bhaduri, S.N. (2002a), “Determinants of capital structure choice: a study of the Indian corporate
sector”, Applied Financial Economics, Vol. 12 No. 9, pp. 655-665.
Bhaduri, S.N. (2002b), “Determinants of corporate borrowing: some evidence from the Indian
corporate structure”, Journal of Economics and Finance, Vol. 26 No. 2, pp. 200-215.
Blundell, R. and Bond, S. (1998), “Initial conditions and moment restrictions in dynamic panel
data models”, Journal of Econometrics, Vol. 87 No. 1, pp. 115-143.
Booth, A., Dermirguc-Kunt, A. and Makismovic, V. (2001), “Capital structures in developing
countries”, Journal of Finance, Vol. 56 No. 1, pp. 87-129.
Bowman, R. (1980), “The importance of a market-value measurement of debt in assessing
leverage”, Journal of Accounting Research, Vol. 18 No. 1, pp. 242-254.
Byoun, S. (2008), “How and when do firms adjust their capital structures toward targets?”, The
Journal of Finance, Vol. 63 No. 6, pp. 3069-3096.
Chang, X. and Dasgupta, S. (2009), “Target behaviour and financing: how conclusive is the
evidence?”, The Journal of Finance, Vol. 64 No. 4, pp. 1767-1796.
Chen, L. and Zhao, X. (2007), “Mechanical mean reversion of leverage ratios”, Economics Letters,
Vol. 95 No. 2, pp. 223-229.
Chen, S., Ho, K. and Yeo, G. (1999), “The determinants of debt maturity: the case of bank
financing in Singapore”, Review of Quantitative Finance and Accounting, Vol. 12 No. 4,
pp. 341-350.
Cheng, S.R. and Shiu, C.Y. (2007), “Investor protection and capital structure: International
evidence”, Journal of Multinational Financial Management, Vol. 17 No. 1, pp. 30-44.
Clark, B.J., Francis, B.B. and Hasan, I. (2009), “Do firms adjust toward target capital structures?
Some international evidence”, February, available at: http://ssrn.com/abstract¼1364095;
http://dx.doi.org/10.2139/ssrn.1364095
Cook, D.O. and Tang, T. (2010), “Macroeconomic conditions and capital structure adjustment
speed”, Journal of Corporate Finance, Vol. 16 No. 1, pp. 73-87.
De Jong, A., Kabir, R. and Nguyen, T.T. (2008), “Capital structure around the world: the roles of
firm-and country-specific determinants”, Journal of Banking and Finance, Vol. 32 No. 9,
pp. 1954-1969.
De Miguel, A. and Pindado, J. (2001), “Determinants of capital structure: new evidence from
Spanish panel data”, Journal of Corporate Finance, Vol. 7 No. 1, pp. 77-99.
DeAngelo, H. and Masulis, R. (1980), “Optimal capital structure under corporate and personal
taxation”, Journal of Financial Economics, Vol. 8 No. 1, pp. 3-29.
Deesomsak, R., Paudyal, K. and Pescetto, G. (2004), “The determinants of capital structure:
evidence from the Asia Pacific region”, Journal of Multinational Financial Management,
Vol. 14 Nos 4-5, pp. 387-405.
JAAR Deesomsak, R., Paudyal, K. and Pescetto, G. (2009), “Debt maturity structure and the 1997 Asian
financial crisis”, Journal of Multinational Financial Management, Vol. 19 No. 1, pp. 26-42.
15,1
Demirgüç-Kunt, A. and Maksimovic, V. (1999), “Institutions, financial markets, and firm debt
maturity”, Journal of Financial Economics, Vol. 54 No. 3, pp. 295-336.
Drobetz, W., Pensa, P. and Wanzenried, G. (2007), “Firm charcteristics, economics conditions and
captial structure adjustments”, January, available at: http://ssrn.com/abstract¼924179;
96 http://dx.doi.org/10.2139/ssrn.924179
Drobetz, W. and Wanzenried, G. (2006), “What determines the speed of adjustment to the target
capital structure?”, Applied Financial Economics, Vol. 16 No. 13, pp. 941-958.
Elsas, R. and Florysiak, D. (2008), “Empirical capital structure research: new ideas, recent
evidence, and methodological issues”, Discussion Paper No. 2008-10, Munich School of
Management, Munchen, 30 July, available at: http://ssrn.com/abstract¼1634932
Fama, E. and French, K. (2002), “Testing trade-off and pecking order predictions about dividends
and debt”, Review of Financial Studies, Vol. 15 No. 1, pp. 1-33.
Faulkender, M., Flannery, M.J., Watson Hankins, K. and Smith, J.M. (2012), “Cash flows and
leverage adjustments”, Journal of Financial economics, Vol. 103 No. 3, pp. 632-646.
Fischer, E.O., Heinkel, R. and Zechner, J. (1989), “Dynamic capital structure choice: theory and
tests”, Journal of Finance, Vol. 44 No. 1, pp. 19-40.
Flannery, M. and Hankins, K. (2007), “A theory of capital structure adjustment speed”,
unpublished manuscript, University of Florida, Gainesville, FL.
Flannery, M. and Rangan, K. (2006), “Partial adjustment toward target capital structures”,
Journal of Financial Economics, Vol. 79 No. 3, pp. 469-506.
Frank, M.Z. and Goyal, V.K. (2009), “Capital structure decisions: which factors are reliably
important?”, Financial Management, Vol. 38 No. 1, pp. 1-37.
Gaud, P., Jani, E., Hoesli, M. and Bender, A. (2005), “The capital structure of Swiss companies: an
empirical analysis using dynamic panel data”, European Financial Management, Vol. 11
No. 1, pp. 51-69.
Graham, J.R. and Harvey, C.R. (2001), “The theory and practice of corporate finance: evidence
from the field”, Journal of Financial Economics, Vol. 60 Nos 2-3, pp. 187-243.
Grossman, S.J. and Stiglitz, J.E. (1980), “On the impossibility of informationally efficient
markets”, The American Economic Review, Vol. 70 No. 3, pp. 393-408.
Gwatidzo, T. and Ojah, K. (2009), “Corporate Capital Structure Determinants”, African Finance
Journal, Vol. 11 No. 1, pp. 1-23.
Haas, R. and Peeters, M. (2006), “The dynamic adjustment towards target capital structures of
firms in transition economies”, Economics of Transition, Vol. 14 No. 1, pp. 133-169.
Hackbarth, D., Miao, J. and Morellec, E. (2006), “Capital structure, credit risk, and macroeconomic
conditions”, Journal of Financial Economics, Vol. 82 No. 3, pp. 519-550.
Harris, M. and Raviv, A. (1991), “The theory of capital structure”, The Journal of Finance, Vol. 46
No. 1, pp. 297-355.
Haugen, R. and Senbet, L. (1978), “The insignificance of bankruptcy costs to the theory of optimal
capital structure”, Journal of Finance, Vol. 33 No. 2, pp. 383-393.
Heshmati, A. (2001), “The dynamics of capital structure: evidence from Swedish micro and small
firms”, Research in Banking and Finance, Vol. 2 No. 1, pp. 199-241.
Hovakimian, A., Hovakimian, G. and Tehranian, H. (2004), “Determinants of target capital
structure: the case of dual debt and equity issues”, Journal of Financial Economics, Vol. 71
No. 3, pp. 517-540.
Hovakimian, A. and Li, G. (2011), “In search of conclusive evidence: how to test for adjustment to
target capital structure”, Journal of Corporate Finance, Vol. 17 No. 1, pp. 33-44.
Hovakimian, A., Opler, T. and Titman, S. (2001), “The debt-equity choice”, The Journal of Evidence from
Financial and Quantitative Analysis, Vol. 36 No. 1, pp. 1-24.
developing
Huang, R. and Ritter, J.R. (2009), “Testing theories of capital structure and estimating the
speed of adjustment”, Journal of Financial and Quantitative Analysis, Vol. 44 No. 2, economies
pp. 237-271.
Jalilvand, A. and Harris, R. (1984), “Corporate behaviour in adjusting to capital structure and
dividend targets: an econometric study”, Journal of Finance, Vol. 39 No. 1, pp. 127-145. 97
Kaufmann, D., Kraay, A. and Mastruzzi, M. (2009), “Governance matters VIII: aggregate and
individual governance indicators, 1996-2008”, World Bank Policy Research Working
Paper No. 4978, 29 June, available at: http://ssrn.com/abstract¼1424591
Kim, E.H. (1982), “Miller’s equilibrium, shareholder leverage clienteles, and optimal capital
structure”, Journal of Finance, Vol. 37 No. 2, pp. 301-319.
La Porta, R., Lopez-de-Silanes, F., Shleifer, A. and Vishny, R.W. (1997), “Legal determinants of
external finance”, Journal of Finance, Vol. 52 No. 3, pp. 1131-1150.
La Porta, R., Lopez-de-Silanes, F., Shleifer, A. and Vishny, R.W. (1998), “Law and finance”, Journal
of Political Economy, Vol. 106 No. 6, pp. 1113-1155.
La Porta, R., Lopez-De-Silanes, F. and Shleifer, A. (1999a), “Corporate ownership around the
world”, Journal of Finance, Vol. 54 No. 2, pp. 471-517.
La Porta, R., Lopez-De-Silanes, F., Shleifer, A. and Vishny, R.W. (1999b), “Investor protection:
origins, consequences, and reform”, NBER Working Paper No. w7428, December, available
at: http://ssrn.com/abstract¼227587
Leary, M. and Roberts, M. (2005), “Do firms rebalance their capital structures?”, The Journal of
Finance, Vol. 60 No. 6, pp. 2575-2619.
Lemma, T.T. and Negash, M. (2011), “Rethinking the antecedents of capital structure of
Johannesburg Securities Exchange listed firms”, Afro-Asian Journal of Finance and
Accounting, Vol. 2 No. 4, pp. 299-332.
Lemma, T.T. and Negash, M. (2012), “Debt maturity choice of a firm: evidence from African
countries”, Journal of Business and Policy Research, Vol. 7 No. 2, pp. 60-92.
Lemma, T.T. and Negash, M. (2013a), “Institutional, macroeconomic and firm-specific
determinants of capital structure: the African evidence”, Management Research Review,
Vol. 36 No. 10, pp. 1-38.
Lemmon, M.L., Roberts, M.R. and Zender, J.F. (2008), “Back to the ‘beginning: persistence and
the cross-section of corporate capital structure’”, The Journal of Finance, Vol. 63 No. 4,
pp. 1575-1608.
Loof, H. (2004), “Dynamic optimal capital structure and technical change”, Structural Change
and Economic Dynamics, Vol. 15 No. 4, pp. 449-468.
Mazur, K. (2007), “The determinants of capital structure choice: evidence from Polish
companies”, International Advances in Economic Research, Vol. 13 No. 4, pp. 495-514.
Miller, M.H. (1977), “Debt and taxes”, Journal of Finance, Vol. 32 No. 2, pp. 261-275.
Mills, G.T. (1996), “The impact of inflation on capital budgeting and working capital”, Journal of
Financial and Strategic Decisions, Vol. 9 No. 1, pp. 79-87.
Modigliani, F. (1982), “Debt, dividend policy, taxes, inflation and market valuation”, Journal of
Finance, Vol. 37 No. 2, pp. 255-273.
Mukherjee, S. and Mahakud, J. (2010), “Dynamic adjustment towards target capital structure:
evidence from Indian companies”, Journal of Advances in Management Research, Vol. 7
No. 2, pp. 250-266.
Myers, S. (1984), “The capital structure puzzle”, The Journal of Finance, Vol. 39 No. 3, pp. 575-592.
JAAR Myers, S. and Majluf, N. (1984), “Corporate financing and investment decisions when firms have
information that investors do not have”, Journal of Financial Economics, Vol. 13 No. 2,
15,1 pp. 187-221.
Ozkan, 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 Nos 1-2,
pp. 175-198.
98 Ozkan, A. (2002), “The determinants of corporate debt maturity: evidence from UK firms”,
Applied Financial Economics, Vol. 12 No. 1, pp. 19-24.

Öztekin, Ö. and Flannery, M.J. (2012), “Institutional determinants of capital structure adjustment
speeds”, Journal of Financial Economics, Vol. 103 No. 1, pp. 88-112.
Rajan, R.G. and Zingales, L. (1995), “What do we know about capital structure? Some evidence
from international data”, Journal of Finance, Vol. 50 No. 5, pp. 1421-1460.
Roberts, M. (2002), “The dynamics of capital structure: an empirical analysis of a partially
observable system”, working paper, Fuqua School of Business, Duke University, Durham, NC.
Shyam-Sunder, L. and Myers, S. (1999), “Testing static trade-off against pecking order models of
capital structure”, Journal of Financial Economics, Vol. 51 No. 2, pp. 219-244.
Smith, C.W. and Warner, J.B. (1979), “On financial contracting: an analysis of bond covenants”,
Journal of Financial Economics, Vol. 7 No. 2, pp. 117-161.
Smith, D.J., Chen, J. and Anderson, H.D. (2010), “Partial adjustment towards target capital structure:
evidence from New Zealand”, Research Paper No. 15, Massey U. College of Business,
available at: http://ssrn.com/abstract¼1640324; http://dx.doi.org/10.2139/ssrn.1640324
Song, J. and Philippatos, G. (2004), “Have we resolved some critical issues related to international
capital structure? Empirical evidence from the 30 OECD countries”, working paper,
University of Tennessee, Knoxville, TN.
Stohs, M. and Maur, D. (1996), “The determinants of corporate debt maturity structure”, Journal
of Business, Vol. 69 No. 3, pp. 279-312.
Stoja, E. and Tucker, J. (2007), “Long and short-run capital structure dynamics in the UK – an
industry level study”, available at: http://ssrn.com/abstract¼1045101; http://dx.doi.org/
10.2139/ssrn.1045101.
Taggart, R.A. (1985), “Secular patterns in corporate finance”, NBER Working Paper No. w0810,
June, available at: http://ssrn.com/abstract¼351354.
Thies, C. and Klock, M. (1992), “Determinants of capital structure”, Review of Financial
Economics, Vol. 1 No. 2, pp. 40-53.
Titman, S. and Wessels, R. 1988, “The determinants of capital structure choice”, Journal of
Finance, Vol. 43 No. 1, pp. 1-19.
Wanzenried, G. (2006), “Capital structure dynamics in the UK and Continental Europe”, The
European Journal of Finance, Vol. 12 No. 8, pp. 693-716.
Welch, I. (2004), “Capital structure and stock returns”, Journal of Political Economy, Vol. 112
No. 1, Part1, pp. 106-132.
Welch, I. (2010), “A bad measure of leverage: the financial-debt-to-asset ratio”, discussion paper,
Brown University, Providence, RI.
Zhao, T. and Susmel, R. (2008), “Testing the trade-off theory of capital structure: a Kalman filter
approach”, 13 December, available at: http://ssrn.com/abstract¼130209; http://dx.doi.org/
10.2139/ssrn.1303209.
Further reading
Frank, M.Z. and Goyal, V.K. (2007), “Trade-off and pecking order theories of debt”, in Eckbo, B.E.
(Ed.), Handbook of Corporate Finance: Empirical Corporate Finance, Elsevier BV,
Amsterdam.
Kim, E.H. (1997), “Legal determinants of external finance”, Journal of Finance, Vol. 52 No. 3, Evidence from
pp. 1131-1150.
developing
Kim, E.H. (1998), “Law and finance”, Journal of Political Economy, Vol. 106 No. 6, pp. 1113-1155.
Lemma, T.T. and Negash, M. (2013b), “The adjustment speed of debt maturity structures:
economies
evidence from African Countries”, Investment Analysts Journal, No. 78, forthcoming.
99
About the authors

Tesfaye T. Lemma is an Associate Professor and Head of Financial Management Department in
the School of Accountancy at the University of Limpopo Turfloop Campus, South Africa. He
obtained a BA (Accounting) and an MBA from the Addis Ababa University, Ethiopia and a PhD
(Finance) from the University of the Witwatersrand, Johannesburg. He has published in academic
journals, such as Afro-Asian Journal of Finance and Accounting, Journal of Business and Policy
Research, Management Research review, and Investment Analysts Journal. His current research
interests are capital and debt maturity structures, dividend policy, financial institutions and
markets, corporate ownership and governance. Tesfaye T. Lemma is the corresponding author
and can be contacted at: [email protected]
Minga Negash is a Tenured Professor at the Metropolitan State University of Denver and a
Part Time Continuous Personal Professor at the University of Witwatersrand (Wits). He has been
teaching Accounting and Finance at both undergraduate and postgraduate programmes at a
number of universities. He supervises a number of PhD research projects at Wits and has
published widely. Negash studied at the Kokebe Tsebah Haile Selassie I Secondary School, the
Old Commercial School of Addis Ababa, Addis Ababa University, Catholic University of Leuven
and at the Free University of Brussels.
To purchase reprints of this article please e-mail: [email protected]

Or visit our web site for further details: www.emeraldinsight.com/reprints

10.1108@jaar 03 2012 0023

Uploaded by

Copyright:

Available Formats

10.1108@jaar 03 2012 0023

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

10.1108@jaar 03 2012 0023

Uploaded by

Copyright:

Available Formats

Journal of Applied Accounting Research

Determinants of the adjustment speed of capital structure: Evidence from developing

*Related content and download information correct at time of download.

JEL Classification — G32

2.1 On the existence of a target capital structure

2.2 On the determinants of adjustment speed of capital structure

Hovakimian et al., 2004) to account for inter-industry differences in adjustment speeds.

3.1 Model specification

gi;t ¼ o0 þ o1 Xi;t ð4Þ

Partly multiplying Equation (5) out, we obtain:

When Equation (6) is estimated, interest is mainly in o1 which is the coefficient of

3.2 Measuring capital structure

3.3 The sample and data

Non-durables 107 26 1 3 8 9 8 14 3 179 18

Oil and Gas 7 41 0 0 3 1 4 2 1 59 6

4. Results and discussions

within country variation in firm-specific variables. Whether these differences in

Panel A: leverage and firm characteristicsa

4.2 Determinants of target capital structure

Short-term leverage Total leverage Long-term leverage

Earnings vol. 0.0140 0.0217 0.0002 0.0270 0.0109 0.0489***

4.3 Determinants of adjustment speed of capital structure

Dependent variable Short-term leverage Long-term leverage Total leverage

LVi,t1 0.461*** 0.410*** 0.606 ***

we estimate Model 1 on an industry-by-industry basis. Estimation results for each

(0.284) (0.408) (0.192)

1,664 1,663 1,665

2,122 2,325 2,128

We further examine capital structure adjustment speeds by trifurcating our sample

Dependent variable Short-term leverage Long-term leverage Total leverage

Upper middle income countries 0.802 0.099 0.775***

We now proceed to examine the influence of more-narrowly-defined aspects of

Panel A:dependent variable – short-term leverage ratioa

capital structure – Model 2

Applied Financial Economics, Vol. 12 No. 1, pp. 19-24.

About the authors

To purchase reprints of this article please e-mail: [email protected]

You might also like

LVi,t1 0.461* 0.410* 0.606 ***