Determinants of Capital Structure:

A Case of Listed Energy Sector Companies in Pakistan

Kashif Ghani* and Syed Husnain Bukhari**

Abstract:
This study is aimed at determining the capital structure of listed energy sector companies in
Pakistan, with a view to finding out the impact of four variables, i.e. Tangibility, Size, Growth
and Profitability of the firms on their leverage. The sample included data for 20 companies for
the period spanning 2004 to 2008. Our results show that all these factors affect the leverage of a
firm in some degree. We found that Tangibility and Size have positive association with leverage
which supports the predictions of Static Trade-off theory. On the other hand, Profitability was
found to have negative relationship with a firms level of debt, a finding that supports the
viewpoint presented by Pecking Order Theory. Growth had positive relationship with leverage
thus supporting the simple version of Pecking Order Theory.

Key words: Capital Structure, Leverage, Static Tradeoff Theory, Pecking Order Theory

________________________________
* MS Student SZABIST, Karachi (email: kashif.ghani@live.com)
** Joint Director, State Bank of Pakistan

Acknowledgement
In the name of Allah, the Most Beneficent, the Most Merciful.
This humble effort is dedicated to all those individuals whose contribution has made this
study possible.
At the outset, I would like to express my gratitude to my research supervisor, Dr. S.M.
Husnain Bokhari, who was abundantly helpful and whose guidance, valuable inputs and
continuous support during the course of this study made this work possible.
I am also grateful to Mr. Muhammad Akbar (Lecturer, Bahria University Islamabad)
whose useful ideas and insights helped me immensely during the completion of this
work.
The author is also deeply indebted to Mr. Saghir Ghauri who was kind enough to help
me out in any difficulty or predicament that I experienced during the course of this study
especially on the data analysis front and the use of E-views.
Deepest gratitude is due to my colleagues at Securities and Exchange Commission of
Pakistan and especially my supervisor at the Commission Mr. Shahid Nasim for being
very considerate and cooperative.
Special acknowledgement is owed to my teachers at SZABIST, Karachi from whom I
learnt the qualitative and quantitative research tools that helped me enormously during
this research work.
I would like to make special mention of all those educationists, practitioners, researchers
and academic stalwarts whose prior work guided me right throughout the course of this
project and from whose work I have undoubtedly benefited in plenty.
And finally, thanks and accolades are due to my family members whose relentless
facilitation and untiring support made this work possible and without whose
contribution towards giving me mental tranquility I would not have been able to
complete this work within the time bounds.

1.

INTRODUCTION

1.1 Background and overview

The capital structure of any firm comprises of two basic components, i.e. equity and debt.
How a firm decides to use a combination of debt and equity to design its capital structure
depends on various considerations that managers take into account. Choice of the right
capital structure is important because the organization wants to maximize its returns to
various stakeholders and at the same time desires to be competitive in its external
environment.
The pioneering work in the area of capital structure was done by Modigliani and Miller
(MM) (1958) who proposed that: 1) the market value of a firm is not affected by its
capital structure and 2) the required rate of return on equity increases as the firms debtequity ratio increases. MMs basic argument was based on the premise that an optimal
capital structure is a result of efforts to balance the bankruptcy risk with the tax savings
of debt.
MMs study was a breakthrough study and served as starting point for further research in
the area of capital structure choice. Subsequent researchers have attempted to eliminate
some of the simplistic assumptions in the MM study to further improve the understanding
of this subject.
Research has focused, inter alia, on various factors and considerations that determine the
level of debt a firms uses in its capital structure; and the decisions of the managers in
choosing a certain mix of equity and debt have been explained with the help of various
theories. Three major theories have emerged as a result of research in this area; these
include Static Trade-off Theory, Pecking Order Theory and Agency Cost Theory
(Buferna et al, 2005).
This research focuses on the Energy Sector listed companies of the Karachi Stock
Exchange to find out what factors affect the capital structure decisions of these firms and
which theory of capital structure better explains the capital structure decisions of the
managers of these companies.

Energy Sector of any country plays the pivotal role of fueling the operation and growth of
industry. Role of energy providing companies in the economic development of a country
can hardly be overemphasized. Provision of sufficient energy resources at reasonable
prices is one of the major aims of governments in their quest for promoting growth and
development of industry, domestic output and exports. In Pakistan, energy sector firms
operate both in public and in private sector. The Karachi Stock Exchange has four
different classifications for energy sector companies, which include Refineries, Oil &
Gas Marketing Companies, Oil & Gas Exploration Companies and Power Generation
Companies.

1.2

Problem statement

The capital structure of any firm is the result of a complex decision process followed by
its financial managers. The research problem involves the need to determine which
factors are more influential in determining the level of leverage of listed companies in the
energy sector of Pakistan. We intend to find out whether a firms tangibility of assets,
size, growth and profitability significantly affect its leverage and the direction of this
influence (i.e. positive or negative association).
1.3

Purpose of the study

This study attempts to examine the determinants of capital structure in listed energy
sector companies of Pakistan with an aim to ascertain whether the determinants used by
Rajan and Zingales (1995) and others apply to capital structure decisions made by this
sample of Pakistani listed firms and to find out which of the capital structure theories
better explains the capital structure of these firms.
1.4

Organization of the Study

This study is organized into five sections. Section-1 gives an introduction and overview
of the subject and purpose of present study. Section-2 provides the literature review and
theoretical framework. Section-3 deals with Data and Methodology of the study. Section4 discusses the empirical analysis and Section-5 summarizes the conclusions and
suggests areas for further research.

2.

2.1

LITERATURE REVIEW AND THEORETICAL FRAMEWORK

Theories of Capital Structure

Various theories of capital structure have been presented which attempt to explain the
factors that determine the capital structure of firms. Three major theories having
relevance to our study are discussed below:

i.

Static Trade-off Theory

The static tradeoff theory focuses on the advantages and disadvantages of issuing

debt versus equity. According to this theory, firms target an optimal debt ratio believing
that such ratio will maximize the value of the firm. The optimal point is achieved when
the marginal benefit of issuing debt equals the increase in the costs associated with
issuing more debt (Myers 1977). Since interest payments are tax deductible, therefore use
of debt provides tax benefit to the firms. However, personal taxes and non-debt tax also
enter a managers consideration making the decision a little more complicated. It is also
said that debt reduces the agency problem as debt financing limits the availability of free
cash-flow to the firm managers. Modigliani and Miller (1958) argued that costs of
financial distress are one of the costs of issuing more debt. Thus when a firm uses high
level of debt, it increases the chances of default on principal and interest payments and
thus leads to financial distress.

ii.

Pecking Order Theory

The other major theory found in corporate finance literature is the pecking-order

theory. According to this theory, firms prefer to finance their investments with internally
generated funds as opposed to external financing and this very fact determines a firms
capital structure. When external financing is required, managers tend to prefer debt
financing over equity. Pecking order theory is generally expressed in terms of
asymmetric information costs and transaction costs. When a firm decides to forego an
investment project rather than opting to use external financing, asymmetric information
costs are said to occur.

The other explanation in support of Pecking Order Theory is the existence of transaction
costs linked with arranging external financing. Since internal financing (use of retained
earnings) does not involve any transaction costs, it is preferred over external financing.
Similarly, external debt financing involves lower transaction costs as compared with
external equity financing. Thus the former is preferred over the latter.

iii.

Agency Theory
The third theory of capital structure is the Agency Theory. This theory states that

an optimum capital structure results from minimization of the costs arising from conflicts
between shareholders and debt-holders (Buferna et al, 2005). Due to the conflict that
exists between these two groups of stakeholders, agency costs attain considerable
importance. The theory states that when companies are in danger of being bankrupt,
shareholders can exert their influence on management to take decisions in favor of
shareholders, potentially depriving the debt-holders. Thus informed debt-holders will
require higher return if there is a chance of shareholders using their influence in the
above said manner to the detriment of bond-holders (Jensen and Meckling, 1976).

2.2

Summary of Literature Review

Capital structure of corporations has been an area of interest for researchers in the field of
finance. A large body of literature is found on various aspects of capital structure choice
decisions and the mechanics involved. Some of these studies are discussed below:

i.

Capital Structure Choice Research in the Developed World

Myers and Majluf (1984) in their famous study published in 1984 present a model
that deals with managers issue of capital and investing decisions in the situation where
managers possess superior information. They argue that it is generally better to issue debt
that is less risky for the firm than the more risky debt. Thus external debt financing is
better than external equity financing. They further contend that when managers possess
superior information and managers choose to finance their investment through issue of
stock, the price of companys stock is likely to fall, ceteris paribus.

Titman and Wessels (1988) analyzed the explaining power of some of the theories
of optimal capital structure. This study takes the prior empirical work on capital structure
theory a little further. It examines the set of capital structure theories that have not been
touched upon in depth by previous empirical work. They used the linear structural
modeling in this study and found that companies with unique products have relatively
low debt ratios. They also found that smaller firms generally use more short term loans
than larger firms. They found no evidence that debt ratios are related with a firms
expected growth, tax shields, volatility or tangibility of assets. However, they did find
evidence that profitability of the firm was associated with debt, as more profitable firms
tended to have less debt relative to market value of equity.

Harris and Raviv (1990) studied the informational role of debt and postulated a
theory stating that managers are not willing to forgo power and not willing to provide
such information to shareholders that would lead to reduction in the control the managers
have on the affairs of the firm. They contended that managers do not always act in the
best interests of the shareholders and debt provides a mechanism that helps keep things in
order. They argue that the fear of default on a companys debt keeps the managers honest
as not being able to meet its liabilities could lead a firm to an undesirable state of being
forced to liquidate.

Chiarella et al (1991) studied the determinants of capital structure choice in 226

Australian firms using data from 1977 to 1985. They found that companys non-debt tax
shield has a negative relationship with debt ratios. The relationship between leverage and
profitability was found to be negative, supporting the idea that firms prefer using internal
sources to finance their assets as opposed to external sources. Size was also found to be
positively related with leverage suggesting that larger firms tended to use more debt.
Further, the researchers also studied the relationship between cash holdings and leverage
and found that there was some evidence in favor of the free-cash flow hypothesis
proposed by Jensen (1976), however, this relationship was not significant. Interestingly,
the researchers did not find support for the relationship between growth opportunities and
7

collateral value, suggesting that this factor did not play a significant role in guiding
managers capital structure decisions.
Rajan and Zingales (1995) studied the capital structure decisions in a sample of
public firms in G-7 countries. They concluded that on an aggregate level, leverage of
firms in G-7 countries is relatively similar. Further, they found that tangibility was
positively correlated with leverage across all countries. Size was found to be positively
correlated with leverage in all countries except Germany. The researchers also concluded
that variables that were found to be correlated with debt-ratios in United States were also
found to be correlated in other G-7 countries.

ii.

Capital Structure Research in Developing countries

Booth et al (2001) analyzed capital structure decisions across 10 developing

countries in their study and found that capital structure decisions in developing countries
are taken in a similar manner as in developed countries, meaning that similar
considerations are taken into account by the financial managers in these two sets of
countries. They found differences among different countries from which data were taken,
indicating that decision factors did vary across individual countries. Further, they found
negative relationship between profitability and debt ratio, from which the researchers
draw the conclusion that external financing is avoided by firms because it is more costly.
Another explanation was that profitable firms required less external financing due to
availability of internally generated funds that can finance firms growth. They also found
that longer-term debt ratios tended to be higher and shorter-term debt ratios tended to be
lower in firms with high asset tangibility.

Buferna et al (2005) took data from a set of Libyan firms to study the
Determinants of capital structure with an additional aim of trying to understand the
impact of lack of a developed secondary market. They found evidence supporting the
Static Tradeoff and Agency Cost theory in the context of Libyan firms, with little
evidence suggesting the presence of information asymmetries. The agency cost factors
are linked by the researchers with the absence of a secondary market as the shareholders

who are unable to sell-off their shares in the market may try to influence the firms to take
actions deemed appropriate by those shareholders.

iii.

Research focusing the European firms

Chen et al (1998) studied the capital structure decisions in the light of various

theories using a sample of Dutch firms. They contended that a thorough understanding of
capital structure choice decisions was important as it would contribute towards a better
theoretical understanding of the subject and would also allow cross country comparisons
of companies in terms of their valuations. Using panel data to assess the impact of
various factors on absolute and year-wise change in leverage, the researchers found
evidence supporting the premises put-forth by the Pecking-order theory. They concluded
that for Dutch firms, agency cost factors and corporate control factors are less relevant in
explaining capital structure shoice.
Bancel and Mittoo (2002) studied capital structure in a selection of European
firms. However, their study was unique in the sense that they surveyed financial
managers in a variety of European firms to unearth the link between the theory and
practice of finance with reference to capital structure decisions. They found that financial
flexibility and dilution of per share earnings were the most important considerations of
finance managers while taking capital choice decisions. They further concluded that in
European firms reasonable evidence was available supporting the presence of factors
described by the static trade-off theory and little evidence supporting pecking-order or
agency theory. The researchers also found that there were no significant differences in the
way these decisions were made by European finance managers and their counterparts in
United States. They suggest that capital choice decisions are made after an interaction of
a complex set of factors and business practices.
Drobetz and Fix (2003) used Swiss data to study the predictions of Static Tradeoff
and Pecking order theories of capital structure. They found that Swiss firms tended to use
comparatively less leverage (compared with Anglo-American countries) and more
profitable firms also exhibited less reliance on debt. Evidence was found that firms with
more investment opportunities have less leverage, a finding that supported static tradeoff

model as well complex version of pecking order theory. A very important finding of this
study was that the Swiss firms appeared to strive for maintenance of leverage ratios
lending support to the basic idea behind static tradeoff model.
Sevil et al (2005) studied capital structure decisions in a sample of Turkish firms
and their results were mostly in conformity with the previous studies on the subject. They
tested the impact of tangibility, growth and profitability on leverage of Turkish firms with
an additional bifurcation of the data between inflationary and disinflationary periods.
They found that debt ratio was positively related with size of the firm and was negatively
related with asset tangibility and profitability. Moreover, they concluded that there was a
structural change in data between the inflationary and disinflationary periods.
iv.

Transition Economies
Devic and Krstic (2001) examined the capital choice decisions of a sample of

firms in two of the so called Transition Economies, namely Poland and Hungary. They
found that size was the most important determinant for Poland; whereas profitability was
significant only when book value of equity was taken for the purpose of computing
leverage. From this fact the authors draw the conclusion that managers of Polish firms
consider book value of equity while making their capital structure choice decisions.
Researchers found negative relationship between profitability and leverage giving
support to pecking-order theory whereas positive relationship between size and leverage
lends support to trade-off theory.
For Hungary, the researchers found that profitability was the major determinant
for explaining capital choice decisions. Tangibility was found to be positively related to
leverage giving weight to trade-off theory.
vi.

Capital Structure Research in Pakistan

Shah and Hijazi (2004) studied the capital structure in listed firms in Pakistan

(other than financial institutions) in 2004. They found that asset tangibility was positively
correlated with leverage (though not significantly); size was positively related while
growth was negatively related with level of debt.

10

Tahir and Hijazi (2006) studied the capital structure of listed cement companies in
Pakistan. They mainly used the variables used earlier by Rajan and Zingales to examine
what determines the level of leverage in the cement companies in Pakistan. Their aim
was to study the unique attributes of this sector and how its capital structure decisions
differ from other listed firms. They found that growth, tangibility and profitability had
significant impact on leverage, whereas size did not seem to impact leverage
considerably.

11

3.

DATA AND METHODOLOGY

This section includes a discussion of the data sources for this study, explanation of the
variables used and the model used for this study.
3.1

Data Sources

Data for this study were obtained from Balance Sheet Analysis of Joint Stock
Companies compiled by State Bank of Pakistan for the period 2004-2008 (five years).
We started with all companies included in the energy sector, which includes:
a. Oil & Gas Marketing Companies
b. Oil & Gas Exploration Companies
c. Refineries
d. Power Generation Companies
Initially all the companies in the sector were included, however, after scrutiny the
companies with incomplete data were removed from the sample. The data for 20
companies was found to be complete over the period of 2004-2008 which was included in
our sample.
3.2

Explanation of variables used in the study

The following section presents the explanation of the variables used in this study and how
these variables are measured. The same variables have been used in the past for similar
studies by Rajan and Zingales (1995) for their study of G-7 countries and Tahir and
Hijazi (2006) in their study of listed Cement companies in Pakistan.
i.

Leverage
Leverage is the dependent variable in this study, defined as the percentage of assets

financed by debt. Previous studies on the subject have used different measures of
leverage such as total liabilities to total assets, interest bearing debt to total assets, total
debt to net assets; and total debt to total capital (Drobetz and Fix, 2003). We use total
debt as the numerator because as pointed out by Shah and Hijazi (2004), most firms in
Pakistan rely on short term debt to meet their capital requirements as the average firm

12

size is small and banks are more inclined to lend for short term than for long term. Thus
our measure for leverage of the firm is total liabilities divided by total assets. Further, this
is also the broadest definition of leverage and covers all short term and long term
liabilities. Our objective here is to study the impact of our independent variables
(tangibility, size, growth and profitability) on leverage.

The second consideration regarding leverage is whether to use book-value of debt

or market value of debt as a measure of leverage. We have used the book value of debt in
our study on the grounds that in Pakistan the corporate debt market is not well developed
and price discovery is not possible for different types of debt securities thus making it
difficult to use market value as the measure of debt. Further, Fama and French (2002)
pointed out that both Pecking Order Theory and Static Trade-off Theory relate to bookvalue of debt and leverage based on market value of debt may not be a correct indictor.
ii.

Tangibility
It is believed that a firm with large amount of fixed assets (i.e. greater tangibility

of assets) finds it easier to raise debt due to the collateral value of its assets as compared
to a firm with lesser tangible assets. Thus firms with large amount of fixed assets are
expected to have higher leverage due to their ability to raise large amount of debt at
relatively cheaper rates (Sevil et al, 2005).
Drobetz and Fix (2003) argue that tangibility of assets also reduces shareholderbondholder conflict because if debts are secured against tangible assets the borrower is
forced to use them for specific projects thus avoiding over-investment or investment in
projects with negative NPV. Further, creditors get added assurance regarding the
repayment of their loans if the loans are secured against fixed assets since fixed assets
have greater liquidation value. Thus the Static Trade-off theory expects a positive
relationship between leverage and tangibility of assets.

We measure tangibility of assets as a ratio of fixed assets (net of depreciation) to total

assets. Thus the first hypothesis for this study is:
Hypothesis 01: a firm with higher tangibility of assets will have higher leverage
13

iii.

Size
Static Trade-off Theory postulates that larger firms are likely to have higher

amount of debt due to relatively lower risk of bankruptcy. Thus there should be a positive
relationship between size of the firm and leverage. Titman and Wessels (1988) argue that
since large firms face lesser threat of bankruptcy than the smaller firms, therefore they
are likely to give less weight-age to bankruptcy costs than the smaller firms when
undertaking capital structure decisions. Bankruptcy costs considerations discourage a
firms managers to use more debt in a firms capital structure (Chiarella et al, 1991).
However, the larger firms are also expected to hold a relatively well diversified portfolio
thus reducing the risk of bankruptcy. Some researchers like Chiarella et al (1991) have
argued that there may be a negative relationship between size of the firm and leverage, on
the premise that it is relatively easier for larger firms to issue equity due to lower
transaction costs. However, such views have not been proven empirically. Thus for our
study, we expect a positive relationship between the size of the firm and leverage.

We measure size of the firm by natural Logarithm of sales as done, among others,
by Chen et al (1998) and Booth et al (2001). The advantage being that using this method,
the trend over a period can be measured as opposed to contribution of the absolute size.
This approach also smoothes out the difference in size between large and small firms
(Daskalakis and Psillaki, 2003). This gives us our next hypothesis:
Hypothesis 02: there is positive relationship between size of a firm and its leverage
iv.

Growth
The Static Trade-off model postulates that firms with greater investment

opportunities will have less leverage because they have strong incentive to avoid underinvestment (Drobetz and Fix, 2003). Abor (2008) mentions that growing firms require
funds to finance the growing operations and internally generating funds may not be
sufficient to finance that growth, thus growing businesses resort to heavy leverage. On
the other hand Myers (1977) argued that growth firms will have less amount of debt in
their capital structure. He contended that growth firms face an agency problem between
shareholders and debt-holders in that the former have an incentive to take risks. Since

14

debt-holders will not enjoy any additional benefit from the higher returns anticipated
from future growth opportunities (if realized), they are reluctant to lend to growth firms,
leading to lower level of leverage. In some situations, this may also result in increased
reliance on short term debt.
The empirical evidence regarding the effect of growth on leverage is rather
mixed. Some researchers like Chen et al (1998) and Abor (2008) found positive
relationship between growth and leverage while some others such as Myers (1977), Rajan
and Zingales (1995) and Shah and Hijazi (2004) found growth and leverage to be
negatively related. Titman and Wessels (1988) found the evidence regarding the impact
of growth of a firm on leverage inconclusive.
Growth has been measured in different ways by various researchers, including by
research expenditure to total sales (Titman and Wessels, 1988), market-to-book ratio
(Rajan and Zingales, 1995) and annual percentage change in total assets (Buferna et al,
2005). Following Buferna et al (2005), we measure growth as the annual percentage
change in total assets for the reason that data on total assets is easily available in
Pakistan, while compiled data on research expenditure by firms is not readily available.
We expect positive relationship between growth and leverage. This leads us to our third
hypothesis, that says:
Hypothesis 03: firms with higher rate of growth are expected to have higher degree of
leverage
vi.

Profitability
Profitability is generally expected to be negatively related to leverage of the firm.

There is contradictory view given by static trade-off theory and pecking-order theory in
relation to the impact of profitability on firm leverage. The trade-off theory expects a
positive relation between profitability and leverage on the grounds that with increase in
profitability, possible bankruptcy costs are likely to decline encouraging more borrowing.

15

Secondly, tax deductibility of interest payments induces profitable firms to use

more of debt financing (Drobetz and Fix, 2003).
Pecking order theory on the other hand expects negative relationship between the
two variables. It argues that firms initially look to use internal sources of funds (retained
earnings) when in need of funds. Once retained earnings are exhausted, the firms turn to
debt financing and fresh equity issue is only used when the first two options are no longer
available. Thus firms with higher profitability will have more internally generated funds
available, thus reducing their requirement of external debt or equity financing (Chen et
al, 1998).
We measure profitability as Earning before taxes divided by firms total assets. So our
next hypothesis is:
Hypothesis 04: firms with higher profitability will have lower leverage
3.3
Regression Model
We have used pooled data analysis for this study since our data are panel data (the same
cross sectional variable is surveyed over time). Data on our dependent variable (leverage)
and four independent variables was obtained and proxies were used to define those
variables. Using the pooled regression model, we obtained the association each of our
independent variables has on leverage. Later, these relationships were analyzed and
compared with the predictions made by various capital structure theories. The model used
for this study is given below:
L = 1 (T) + 2 (S) + 3 (G) + 4 (P) +
Where
L = Leverage
T = Tangibility of assets
S = Size of the firm
G = Growth
P = Profitability
= the error term

16

4.

EMPIRICAL ANALYSIS

In this section, we present the summary statistics of the data used for our study, followed
by analysis and interpretation of results.
4.1

Descriptive Statistics

The following table summarizes the descriptive statistics for the data used in our study. It
may be noticed that minimum leverage for any firm was 0.014 (1.4%) while maximum
was 0.99 (99%). Mean leverage was 56.3%. Similar results can be seen for other
variables. Size had the maximum Standard Deviation which is understandable since our
sample included firms of various sizes, ranging from very large to relatively small.
Leverage Tangibility Size Growth Profitability
Mean
Median
Maximum
Minimum
Std. Dev.
Observations (N)

0.563
0.631
0.990
0.014
0.246
100

0.458 4.080
0.472 4.428
0.939 5.695
0.041 0.857
0.249 0.995
100 100

0.155
0.144
0.701
-0.354
0.191
100

0.109
0.066
0.554
-0.212
0.164
100

Correlation Co-efficient
The following table gives correlations amongst the variables used in our study. In
general, variables having correlation of 0.70 or above are not included in the analysis in
order to avoid the problem of multicollinearity. The maximum correlation was found to
be -0.634 between tangibility and growth, thus there is no problem of multicollinearity.
The correlation between tangibility and growth was found to be negative suggesting that
rapidly growing firms tend to have less tangible assets. This finding is consistent with the
finding of Tahir and Hijazi (2006) who found that growth and tangibility were negatively
related in the Cement companies of Pakistan. Another interesting association was
observed between size and profitability. It was observed that size was positively
correlated with profitability suggesting that larger firms tend to be more profitable (when
profitability is measured by earnings as percentage of assets).

17

Leverage
Growth
Tangibility
Size
Profitability

4.2

Leverage Growth Tangibility Size Profitability

1.000 0.321
-0.039 0.345
-0.390
0.321 1.000
-0.634 0.576
0.359
-0.039 -0.634
1.000 -0.436
-0.448
0.345 0.576
-0.436 1.000
0.398
-0.390 0.359
-0.448 0.398
1.000

Regression Analysis Results

The following table presents the results of our regression analysis. R-squared (R2) has a
value of 0.886 which means that our independent variables (tangibility, growth, size and
profitability) explain about 88% of the variation in our dependent variable (leverage).
Durbin Watson statistic is close to 2.00 which shows that there is no problem of
autocorrelation.
R-squared
0.886 Mean dependent var 0.563
Adjusted R-squared 0.869 S.D. dependent var 0.246
S.E. of regression
0.089 Akaike info criterion -1.874
Sum squared resid
0.679 Schwarz criterion
-1.509
Log likelihood
107.708 Durbin-Watson stat 2.020
Regression Coefficients and their significance

The following table shows coefficients of our regression analysis. It may be noticed that
coefficients of all independent variables (other than growth) were found to be significant
at 1% level, while growth was significant at 5% level.
Variable Coefficient Std Error t-Statistic Probability
Tangibility 0.268128 0.064496 4.157302
0.0001
Size
0.136057 0.009180 14.82100
0.0000
Growth
0.121275 0.055101 2.200961
0.0304
Profitability -0.467314 0.126065 -3.706937
0.0004

We can see that tangibility is significant at 1% significance level (B1=0.268) and has
positive association with leverage. This is as expected because availability of tangible
assets makes it easier for firms to obtain loans from banks and other lenders. This finding
is consistent with the findings of Meckling (1976), Myers (1977), Drobetz and Fix (2003)
and Sevil et al (2005). Thus we do not reject our first hypothesis at 1% level.
Size was found to be positively associated with leverage and was significant at 1% level
(B2=0.136). This finding is consistent with the predictions of trade-off theory which

18

postulates that larger firms are expected to have higher leverage due to lower bankruptcy
costs, more diversified portfolio and ease in obtaining financing. Our finding is consistent
with the finding of Titman and Wessels (1988), Chiarella (1991) and Chen at al (1998).
Thus we do not reject our second hypothesis at 1% significance level.
Our third independent variable - growth was found to be positively related with leverage.
It was significant at 5% level (B3= 0.121). This finding does not support the prediction of
static trade-off theory which argues that growth firms will have low level of leverage
owing to the agency problem between shareholders and bondholders where the latter will
be reluctant to take excessive risks associated with a growing firm. In this regard, our
finding is consistent with the finding of Chen et al (1998) and Abor (2008) who argued
that growth firm will have higher leverage due to inability to finance growing operations
with internally generated funds. Thus we do not reject our second hypothesis at 5%
significance level.
Our final independent variable - Profitability was found to be negatively associated with
leverage at significance level of 1% (B4= 0.467). Our finding supports the view putforward by Pecking Order theory that firms prefer to use internally generated funds to
meet their funding requirements, and when retained earnings are not sufficient the firms
resort to borrowing while issue of fresh equity is considered the last option. Our results
are consistent with the findings of Rajan and Zingales (1995) and Chen et al (1998). Thus
we do not reject our fourth hypothesis at 1% significance level.
The following table gives a summary of our independent variables; the proxies used,
expected relationship with dependent variable and whether the results of our regression
were significant.
Variable
Tangibility

Measure

Fixed Assets (net)/ Total

Assets
Size
Log of sales
Growth
Earnings before taxes/
total assets
Profitability Percentage change in
total assets over the
previous year

Expected
Relationship
Positive

Observed
Relationship
Positive

Significant/
Insignificant
Significant

Positive
Positive

Positive
Positive

Negative

Negative

Significant
Significant
(Only at 5%)
Significant

19

5.

CONCLUSION AND RECOMMENDATIONS

The table below summarizes the finding of our study in the light of Static Trade-off and
Pecking Order Theories. Our findings in relation to the impact of Tangibility and Size on
leverage were found to be in conformance with the static tradeoff theory. The impact of
Profitability was found to be in accordance with the prediction of Pecking Order Theory.
Our finding with relation to the impact of Growth was found to be in conformance with
the simple version of Pecking Order Theory (which states that growth firms would
finance the expanding operations by incurring more debt).
Variable

Tangibility

Measure

Fixed Assets
(net)/ Total
Assets
Size
Log of sales
Growth
Earnings
before taxes/
total assets
Profitability Percentage
change in
total assets
over the
previous year

Expected
Relationship

Observed
Relationship

Expected
Relationship
in Pecking
Order Theory

Positive

Expected
Relationship
in Static
Trade-off
Theory
Positive

Positive

Positive
Positive

Positive
Positive

Positive
Negative

Negative
Positive/
Negative*

Negative

Negative

Positive

Negative

Negative

* Simple and Complex version of Pecking Order Theory

Thus our results show that predictive capability of these capital structure theories is rather
mixed, as no single theory completely explains the behavior of financial decision makers
of these firms. We may conclude that in Pakistan, power sector firms with more tangible
assets use greater debt which may be due to the collateral value of their assets and ease of
obtaining financing. Firms with more tangible assets find it easier to obtain financing as
lenders would be more willing to lend against the security of fixed assets. It was also
observed that for Pakistani energy sector firms, size was found to be positively associated
with leverage suggesting that larger firms tend to use higher leverage. The reasons could
be lower bankruptcy risks in larger firms, more diversified portfolio and ease of obtaining
financing. Our third independent variable, Growth, was found to be positively associated
20

with leverage which supports the simple version of pecking order theory. Our results do
not support the prediction of static tradeoff theory that growing firms will tend to have
less leverage as they would find it difficult to invest sub-optimally. Similarly, the
viewpoint presented by Agency theory also does not hold true because Agency theory
predicts a negative relationship between growth and leverage. Our fourth variable,
profitability - was found to be negatively associated with leverage supporting the view of
Pecking Order Theory that profitable firms tend to have less leverage due to availability
of internally generated funds.

Suggested areas for further research

The following areas may be considered for future research:

Using different measures of leverage e.g. Short term and Long term debt

Incorporating both book value and Market value of debt

Including a larger sample of companies and doing intra-sector comparisons

Differentiating between different ownership structures

21

6.

REFERENCES

1. Bancel, F. and Mitto, U. (2002), The Determinants of Capital Structure Choice:

A Survey of European Firms, AFA 2003 Washington, DC Meetings; EFMA
2002

London

Meetings,

Available

at

SSRN:

papers.ssrn.com/sol3/papers.cfm?abstract_id=299172 (Accessed: June 05, 2009)

2. Booth, L., Aivazian, V., Demirguc-Kunt, A. and V. Maksmivoc (2001), Capital
structures in developing countries, Journal of Finance 56, pp.87-130.
3. Brealy, R., Myers, S. and Marcus, A. (1995), Fundamentals of Corporate
Finance, International Edition, McGraw Hill, Inc.
4. Buferna, F., Bangassa, K. and Hodgkinson, L. (2005), Determinants of Capital
Structure Evidence from Libya, University of Liverpool Research Paper Series
No.2005/08, ISSN 1744-0718
5. Chen, L., Lensink, R. and Sterken, E. (1998), The Determinants of Capital
Structure: Evidence from Dutch Panel Data, European Economic Association
Annual Congress, Berlin, September 2-5, 1998
6. Chiarella, C., Pham, T., Sim, A.B, and

Tan, M. (1991), Determinants of

Corporate Capital Structure: Australian Evidence University of Technology,

Sydney School of Finance and Economics Working Paper
7. Daskalakis, N. and Psillaki, M. (2003), The Determinants of Capital Structure of
the SMEs: Evidence from the Greek and the French firms, Available at:
http://www.univ-orleans.fr/DEG/GDRecomofi/Activ/psillaki_strasbg05.pdf
(Accessed: June 10, 2009)
8. Devic, A. and Krstic, B. (2001) Comparatible Analysis of The Capital Structure
Determinants in Polish and Hungarian Enterprises Empirical Study, part of the
project "Economic Efficiency of Development Strategies of Enterprise in Market
Economy - Possibilities and ways for improvement" (No. 1406) - Ministry of
Science, Technology and Development - Republic of Serbia.
9. Drobetz, W. and Fix, R. (2003), What are the determinants of capital structure?
Some evidence for Switzerland; University of Basel, WWZ/ Department of
Finance, Working Paper No. 4/03

22

10. Harris, M. and A. Raviv, (1990) Capital structure and the informational role of
debt, Journal of Finance 45, pp. 321-349
11. Jensen, M., and W. Meckling, (1976), Theory of the Firm: Managerial Behavior,
Agency Costs and Ownership Structure, Journal of Financial Economics 3, pp.
305-360.
12. Modigliani, F. and Miller, M., (1958), The Cost of Capital, Corporation Finance
and the Theory of Investment, The American Economic Review, Vol. 48, pp.
433-443
13. Myers, S., and N. Majluf, (1984), Corporate Financing and Investment Decisions
When Firms Have Information Investors Do Not Have, Journal of Financial
Economics, 13, pp. 187-222.
14. Myers, S.C., (1977), Determinants of Corporate Borrowing, Journal of
Financial Economics, Vol. 5., pp. 147-175
15. Rajan, R. and Zingales, L., (1995), What Do We Know about Capital Structure?
Some Evidence from International Data, The Journal of Finance, Vol. 50.
pp.1421-1460
16. Sevil, G., Saylr, O. and Yldrm, S. (2005), Determinants of Capital Structure:
Evidence from Turkish Manufacturing Firms, Available at:
http://home.anadolu.edu.tr/~gsevil/capital.pdf (Accessed May 26, 2009)

17. Hijazi, T. and Tariq, Y. (2006), Determinants of Capital Structure: A Case for
the Pakistani Cement Industry, The Lahore Journal of Economics, 11:1,
(Summer 2006), pp. 63-80
18. Shah, A. and Hijazi, T. (2004), The Determinants of Capital Structure of Stock
Exchange-listed Non-financial Firms in Pakistan, The Pakistan Development
Review, 43:4, Part II, pp. 605-618
19. State Bank of Bank of Pakistan, Balance Sheet Analysis of Joint Stock
Companies Listed on The Karachi Stock Exchange - 2008
20. Titman, S. and Wessels, R., (1988), The Determinants of Capital Structure
Choice, The Journal of Finance, Vol. 43, No.1, pp. 1-19

23

7.

GLOSSARY:

Agency Theory: the theory that deals with conflict of interest between the firms
owners and managers
Capital Structure: a firms mix of long term and short term financing
Equity: the total value of assets less the current and long term liabilities (equity
represents the owners interest in an enterprise)
Leverage: the percentage of assets financed by debt
Static Tradeoff Theory: The theory that capital structure of a firm is based on a
tradeoff (substitution) between tax savings and distress costs of debt (bankruptcy
costs)
Pecking Order Theory: Theory stating that firms prefer to use debt rather than
equity if internal financing is sufficient
(Source: Fundamentals of Corporate Finance, By: Brealy et al, 1995)

24

ANNEXURES

25

12
Series: LEVERAGE
Sample 1 100
Observations 100

10

Mean
Median
Maximum
Minimum
Std. Dev.
Skewness
Kurtosis

8
6
4
2

Jarque-Bera
Probability

0.562675
0.631376
0.990283
0.013902
0.245692
-0.447354
2.186511
6.092781
0.047530

0
0.000 0.125 0.250 0.375 0.500 0.625 0.750 0.875 1.000

14
Series: GROWTH
Sample 1 100
Observations 100

12
10

Mean
Median
Maximum
Minimum
Std. Dev.
Skewness
Kurtosis

8
6
4
2

Jarque-Bera
Probability

0.155073
0.144092
0.700757
-0.353794
0.191081
0.458615
3.226513
3.719244
0.155731

0
-0.4

-0.2

0.0

0.2

0.4

0.6

26

20
Series: SIZE
Sample 1 100
Observations 100

15
Mean
Median
Maximum
Minimum
Std. Dev.
Skewness
Kurtosis

10

Jarque-Bera
Probability

4.080190
4.428313
5.694850
0.857332
0.994915
-0.957824
3.663007
17.12203
0.000191

0
1.0 1.5 2.0

2.5 3.0 3.5

4.0 4.5 5.0

5.5

8
Series: TANGIBILITY
Sample 1 100
Observations 100

6
Mean
Median
Maximum
Minimum
Std. Dev.
Skewness
Kurtosis

Jarque-Bera
Probability

0.458034
0.472447
0.938817
0.041047
0.249219
-0.016770
2.011177
4.078732
0.130111

0
0.125 0.250 0.375 0.500 0.625 0.750 0.875

27

14
Series: PROFITABILITY
Sample 1 100
Observations 100

12
10

Mean
Median
Maximum
Minimum
Std. Dev.
Skewness
Kurtosis

8
6
4
2

0.109416
0.066026
0.553641
-0.212054
0.163694
0.803492
3.488478

Jarque-Bera
Probability

11.75419
0.002803

0
-0.125 0.000

0.125

0.250

0.375

0.500

1.0
0.8
0.6
0.4

0.4

0.2
0.2
0.0
0.0
-0.2
-0.4
10

20

30

40

Residual

50

60
Actual

70

80

90

100

Fitted

28

Breusch-Godfrey Serial Correlation LM Test:

F-statistic
Obs*R-squared

0.404843
5.680331

Probability
Probability

0.957414
0.931331

Test Equation:
Dependent Variable: RESID
Method: Least Squares
Pre-sample missing value lagged residuals set to zero.
Variable

Coefficient

Std. Error

t-Statistic

Prob.

TANGIBILITY
SIZE
GROWTH
PROFITABILITY
OGDC_D
POL_D
PPL_D
IDE_D
JPG_D
KESC_D
KOHE_D
HUBCO_D
MA(1)
MA(2)
RESID(-1)
RESID(-2)
RESID(-3)
RESID(-4)
RESID(-5)
RESID(-6)
RESID(-7)
RESID(-8)
RESID(-9)
RESID(-10)
RESID(-11)
RESID(-12)

-0.025386
0.003777
0.005029
0.056972
0.006762
-0.009118
-0.014563
0.003989
0.005433
0.016790
-0.007560
-0.063045
0.242646
0.187050
-0.282541
-0.030809
0.101259
-0.148338
-0.043236
0.005121
0.008096
-0.023266
-0.140334
-0.000852
-0.184628
-0.104808

0.074764
0.010993
0.060818
0.144355
0.085840
0.079380
0.084910
0.070018
0.080625
0.079874
0.080441
0.086033
0.630230
0.640129
0.632093
1.012402
0.560982
0.142940
0.210353
0.191589
0.127538
0.133773
0.135644
0.120052
0.140416
0.126673

-0.339549
0.343613
0.082690
0.394668
0.078771
-0.114865
-0.171516
0.056968
0.067382
0.210203
-0.093987
-0.732800
0.385012
0.292206
-0.446993
-0.030432
0.180504
-1.037763
-0.205540
0.026729
0.063482
-0.173923
-1.034572
-0.007098
-1.314869
-0.827388

0.7352
0.7321
0.9343
0.6942
0.9374
0.9089
0.8643
0.9547
0.9465
0.8341
0.9254
0.4660
0.7013
0.7709
0.6562
0.9758
0.8573
0.3028
0.8377
0.9787
0.9496
0.8624
0.3042
0.9944
0.1926
0.4107

R-squared
Adjusted R-squared
S.E. of regression
Sum squared resid
Log likelihood

0.056803
-0.261844
0.092805
0.637342
110.8871

Mean dependent var

S.D. dependent var
Akaike info criterion
Schwarz criterion
Durbin-Watson stat

0.005881
0.082617
-1.697741
-1.020397
1.979443

29

Pairwise Granger Causality Tests

Sample: 1 100
Lags: 12
Null Hypothesis:

Obs

F-Statistic

Probability

GROWTH does not Granger Cause LEVERAGE

LEVERAGE does not Granger Cause GROWTH

88

1.02710
2.07816

0.43597
0.03129

SIZE does not Granger Cause LEVERAGE

LEVERAGE does not Granger Cause SIZE

88

1.00806
1.03999

0.45247
0.42498

TANGIBILITY does not Granger Cause LEVERAGE

LEVERAGE does not Granger Cause TANGIBILITY

88

1.38888
0.57830

0.19504
0.85146

PROFITABILITY does not Granger Cause LEVERAGE

88

0.86385
1.16606

0.58655
0.32663

SIZE does not Granger Cause GROWTH

GROWTH does not Granger Cause SIZE

88

2.06371
1.30367

0.03258
0.23911

TANGIBILITY does not Granger Cause GROWTH

GROWTH does not Granger Cause TANGIBILITY

88

0.69847
1.38044

0.74688
0.19908

PROFITABILITY does not Granger Cause GROWTH

GROWTH does not Granger Cause PROFITABILITY

88

1.65167
1.12906

0.10010
0.35372

TANGIBILITY does not Granger Cause SIZE

SIZE does not Granger Cause TANGIBILITY

88

0.60572
1.15797

0.82933
0.33242

PROFITABILITY does not Granger Cause SIZE

SIZE does not Granger Cause PROFITABILITY

88

0.95968
0.56355

0.49585
0.86283

PROFITABILITY does not Granger Cause TANGIBILITY

88

1.41085
0.60394

0.18485
0.83080

LEVERAGE does not Granger Cause PROFITABILITY

TANGIBILITY does not Granger Cause PROFITABILITY

30

ADF Test Statistic

-4.369729

1% Critical Value*
5% Critical Value
10% Critical Value

-3.4986
-2.8912
-2.5824

*MacKinnon critical values for rejection of hypothesis of a unit root.

Augmented Dickey-Fuller Test Equation

Dependent Variable: D(LEVERAGE)
Method: Least Squares
Sample(adjusted): 4 100
Included observations: 97 after adjusting endpoints
Variable

Coefficient

Std. Error

t-Statistic

Prob.

LEVERAGE(-1)
D(LEVERAGE(-1))
D(LEVERAGE(-2))
C

-0.358603
0.177347
0.079200
0.197369

0.082065
0.102883
0.102858
0.049215

-4.369729
1.723769
0.769990
4.010361

0.0000
0.0881
0.4433
0.0001

R-squared
Adjusted R-squared
S.E. of regression
Sum squared resid
Log likelihood
Durbin-Watson stat

0.178159
0.151648
0.168917
2.653561
36.90517
2.000099

Mean dependent var

S.D. dependent var
Akaike info criterion
Schwarz criterion
F-statistic
Prob(F-statistic)

-0.003608
0.183394
-0.678457
-0.572283
6.720170
0.000374

31