Proceedings of World Business, Finance and Management Conference
14 - 15 December 2015, Rendezvous Grand Hotel, Auckland, New Zealand
ISBN: 978-1-922069-91-7
Determinants of Capital Structure: Some Australian Service
Sector Evidence
Rafiuddin Ahmed1
Capital structure is one of the widely researched finance topics which
generated a healthy body of literature in different countries. The key aim of
this paper is to examine the determinants of leverage of seven service sector
listed companies in the Australian stock exchange during the years 20122014. The study uses 57 individual firms and a total of 171 firm-year
observations. The analyses of pooled and panel regression models reveal
significant relationship between firm-specific and industry-specific
determinants (factors) affecting leverage decisions of these firms. The factors
are, however, dissimilar to factors observed in country-specific studies
undertaken elsewhere. The reasons for differences in observed and extant
factors in capital structure studies are explained, and the implications for
research are outlined in this paper.
Keywords: Capital structure, Australian service sector, panel data analysis, industry effect,
firm characteristics.
1. Introduction
Capital structure has been extensively researched since the seminal work of Modigliani and
Miller (1958) was published. Since then, capital structure research has moved to examine
firm value, managerial actions, and firm strategies to maximize income, shareholder value
and growth of firms. Theoretical refinements to MM’s work (1958) are later clustered into
different theories such as Trade-Off Theory, Pecking Order Theory, and Agency Theory
(Jensen and Meckling 1976). The general results of these theories suggest an optimal capital
structure has less than 100% leverage due to bankruptcy risks and optimal tax benefit
threshold. The theoretical developments raised questions if costs and benefits are
economically significant enough to have an appreciable impact on capital structure
(Michaelas et al. 1999). This question is later empirically tested by relating firm characteristics
such as size, profitability, growth rate, firm risk, and industry characteristics.
The empirical studies range from robust cross country comparisons of capital structure to
single country-specific studies. Multiple country studies are, however, not very recent. Fan et
al (2012) examines capital structure differences of countries and observed intuitional and
bond markets. Song et al (2004) examines the impact of country-specific factors on capital
structure of 30 countries. John (2008) examines capital structures of 42 countries from all
continents and finds different country-specific statistics of capital structures. However,
Gitman’s (2004) study examines two clusters of seven countries each and finds a separate
set of statistics for both sets.
In single country-specific studies, researchers have used listed and non-listed companies or
industry sectors and examined capital structures of firms in cross sections of an economy.
For example, Abor (2007) uses firm characteristics of listed firms in Ghana, Shah and Khan
1
College of Business, Law and Governance, James Cook University, Townsville Campus, Australia. Email:
rafiuddin.ahmed@jcu.edu.au
1
Proceedings of World Business, Finance and Management Conference
14 - 15 December 2015, Rendezvous Grand Hotel, Auckland, New Zealand
ISBN: 978-1-922069-91-7
(2007) use firm characteristics of Pakistani firms, and finally, De Vries and Erasmus (2010)
use firm characteristics of listed and unlisted firms in South Africa to explore the determinants
of capital structure. There are many more studies in other country contexts but only some
recent ones are listed here. Compared to studies in other countries, Australian studies are
limited and very old. The Australian studies are confined to studying the imputation tax system
and its effect of capital structure choice, firm value and shareholder values.
Compared to the studies outside Australia, capital structure studies in Australia are still at a
nascent stage and less conclusive (Qiu and La 2010). Australian capital structure studies are
different and interesting Qiu and La (2010) due to the introduction of double taxation in 1987.
Allen’s (1991) study on Australian firms’ capital structure reports financial flexibility and tax,
Allen’s (1993) study examines theoretical assumptions of capital structure theories in the
context of Australian firms (see also, Twite 2001). The study by Cassar and Holmes (2003)
examines characteristics of Australian firms on capital structure choices (see also,
Deesomsak et al. 2004). Finally, the most recent study by Qui and Ba (2010) reports the
effect of firm characteristics of Australian firms over an extended period of time, over a 15year period (1992-2006). The Australian studies are too old, and none of the studies has
examined the firm characteristics of the service sector and industry effects on capital
structure. This current study is aimed to fill in this gap in the literature.
The main contributions of this study to the literature on capital structure are the determination
of firm characteristics of the service sector listed firms in the Australian Stock Exchange (ASX)
and industry effects of three service sectors on leverage decisions. The findings are similar
to the characteristics observed in earlier studies in other countries and the industry sectors
are different from other studies. Recent evidence of service’s sector effects on capital in the
Australian context is non-existent; this study is an addition to the global research on capital
structure.
The rest of the paper is organized as follows. In the next section, the theory and evidence of
capital structure are presented. This follows the three consecutive sections on data, results
and discussions. The final section rounds of the paper with conclusions and directions for
further research.
2. Theoretical developments in capital structure research
Since the development of MM theory in 1958 (Modigliani and Miller 1958) that capital structure
has no relevance to firm valuation, other theories on capital structure and firm value have
emerged. In 1963, MM (Modigliani and Miller 1963) refined their 1958 theory by adding taxes
and relating it to firm value by stating that debt in a capital structure reduces a firm’s tax
payable amount, improves income after taxes, resulting in a higher value of a firm than a firm
without any debt in the capital structure. This development has later paved the way for the
development of the trade-off theory which states that a firm’s optimal leverage is achieved by
minimizing taxes, cost of financial distress and agency costs. Baxter (1967) argues that
increased debt levels increase the chances of bankruptcy, and increase interest payable to
the debtholders. A firm’s optimal leverage is where tax advantage from debt exactly equals
the cost of debt. Kraus and Litzenberger (1973) argue that a firm’s market value declines if
debt obligations are greater than its earnings. Deangelo and Masulis (1980) propose the static
trade off theory and include other tax-minimizing offsets such as depreciation and investment
2
Proceedings of World Business, Finance and Management Conference
14 - 15 December 2015, Rendezvous Grand Hotel, Auckland, New Zealand
ISBN: 978-1-922069-91-7
tax credits. They argue that firms weigh between tax advantages of debt with business risk
(a cost). Their theoretical model proposes that a firm’s optimum debt level is where the
present value of tax savings from debt equals the present value of costs of distress.
Myer’s (1984) theoretical explanation of an asymmetric information hypothesis proposes
different information held by firms’ internal and external stakeholders about firms’ income
distribution plans (Ross, 1977). Thus, firms’ leverage level movements signal firms’
confidence levels, suggesting lower leverage as a poor signal about income and its
distribution potential and vice versa. Pettit and Singer (1985) discuss the problems of
asymmetric information and possible agency costs affecting firms’ demand and supply of
credit. They argue that small firms possess higher level of asymmetric information due to
financial constraints for sufficient disclosure of financial information to outsiders. This theory
has laid the foundation for Pecking Order Theory (PET).
Donaldson (1984) introduced the concepts and ideas of Pecking Order Theory (POT) which
was later refined by Meyers (1984) and Myers and Majluf (1984). The fundamental premise
of this theory is firm’s preferences for funding is stacked by a pecking order of risk preferences
and corresponding costs. Thus, firms use the cheapest source of internal funds such as
retained earnings, debt, convertible debt and preference shares) and the external equity
(Myers 1984). The cost of sourcing extra funding is dependent on the extent of information
asymmetries of risk perceptions emanating from differential information needs held by inside
management and potential investors. In addition to a firm’s desire to source the cheapest fund
to finance its needs, other factors such as the stage development of a firm such as a startup,
a mature firm etc. influence the supply of funds (Mac an Bhaird and Lucey 2010).
Agency theory (Jensen and Meckling 1976) addresses the fundamental problem of managing
a firm’s capital structure from the cheapest source of funds. While common equity is an
expensive source of fund, its use results in suboptimal firm value when equity holders insist
on risk reduction from lower leverage usage. If managers’ and shareholders’ interests are not
aligned, it is highly unlikely that optimal firm value is ever going to eventuate from managerial
actions. The debt holders risk perceptions encourage them to ask for debt covenants or other
costly debt shielding instruments. The tensions between the two subgroups of owners impose
increased risk of monitoring by management resulting in costly monitoring and hence agency
costs. A number of remedial measures such as reduction in consumption of resources when
debt and hence, bankruptcy risks increase, increasing the stake of managers in a firm or
increasing the leverage (Jensen 1986), commonly packed as ‘free cash flow hypothesis’. Free
cash flow hypothesis proposes the adoption of measures to reduce free cash flow at
managers’ disposal by increasing leverage (Stulz 1990) so that less cash flow is available for
desired investment choices.
3. Empirical literature
Empirical literature on capital structure covers robust cross country comparisons to single
country-specific studies. First, some important cross country comparison studies are
reviewed. Fan et al. (2012) compare the capital structure of 39 countries and find that degree
of development of banking sector, equity and bond markets explained the heterogeneity of
capital structures among the countries studied (see also, Hall et al. 2004). Song et al.’s (2004)
study 30 OECD countries, however, find no support for country-specific variations as
important determinants of heterogeneity of capital structure. Giannetti (2003) contends that
3
Proceedings of World Business, Finance and Management Conference
14 - 15 December 2015, Rendezvous Grand Hotel, Auckland, New Zealand
ISBN: 978-1-922069-91-7
variations in findings of cross country comparisons of capital structure may be due to
inappropriate sample selection such as consideration of only large firms in samples. John et
al. (2008) criticize these studies for common country-level and firm level slopes and study for
country-specific and firm level unique slopes and co-efficients. Their study uses 42 countries,
from all continents and firms of all sizes. They find different country-specific slopes and
different firm level slope and coefficients in these countries. In an earlier study, Gitman (2004)
reports different slopes and co-efficient for seven firms each from developed and developing
countries. While the studies reviewed above are aggregated country level studies, single
country studies are also important in understanding firms operating in a unique regulator,
economic and instructional frames.
A large body of research examines single country-specific firm level factors affecting capital
structure choices. Drobetz, Penza and Wanseried (2006) find that profitability, investment
opportunities, tangibility of assets and volatility are important firm level determinants of capital
structure (see also, Rajan and Zingales 1995). Vries and Erasmus (2010) examined firm and
economic characteristics of listed South African firms affecting capital structures in the
industrial sector. The study uses panel data over a period of 14 years (1995-2008) and 2684
firm year observations (170 listed and 110 unlisted firms). The results suggest that asset
structure (Fixed assets/Total assets) and natural log of sales [ln(sales)] are two most
important determinants of capital structure of South African listed and unlisted industrial firms.
The study also reports different slopes and coefficient values by subsets of listed and unlisted
firms, suggesting the differences due to ‘survivorship bias’ (De Vries and Erasmus 2010, p
6).
Australian capital structure studies are different and interesting (Qiu and La 2010) due to the
introduction of double taxation in 1987. Allen’s (1991) study on Australian firms’ capital
structure reports financial flexibility and tax as two main concerns of capital structure choices
of Australian firms. Allen (1993) finds profitable Australian firms’ inclination to use less debt,
confirming signaling and pecking order theory and contradicting trade-off theory. Twite (2001)
reports significant debt ratio reductions, more reliance of capital sourcing following the tenets
of pecking order theory and signaling theory but not so much about trade-off theory. The study
by Cassar and Holmes (2003) reveals fast-growing and larger firms use more debt but firms
with less tangible assets use less debt. Their survey fails to find any significant effect on a
firm’s capital structure (debt ratio). Deesomask et al’s (2004) study fails to find profitability,
growth, and risk as explanatory factors of capital structure differences in Australia. Qiu and
La (2010) reports in their study of Australian firms over a 15 year period (1992-2006) that size
is not statistically significant but tangibility of assets and profitability have positive relation and
growth prospects and business risks have negative relationship with leverage. The Australian
studies are too old and none of the studies have examined the firm characteristics of the
service sector and industry effects on capital structure. This current study is aimed to fill in
this gap in the literature.
4. Data and procedures
This study is based on data collected from Datanalysis database. The database provides
financial information about Australian companies operating in different sectors. The current
study uses only three years of complete data (2012-2014) on 60 service companies that
match the search criteria. Three of these companies are later excluded due to incomplete
4
Proceedings of World Business, Finance and Management Conference
14 - 15 December 2015, Rendezvous Grand Hotel, Auckland, New Zealand
ISBN: 978-1-922069-91-7
information for analysis. Thus our final sample comprises 57 companies (panel data) and 171
firm year observations (pooled data). The companies comprise seven sectors of the
Australian economy and have positive values for all search entries. Only listed firms from
these seven sectors are included, so the sample may have survivorship bias (De Vries and
Erasmus 2010). Michaelas et al. (1999) argue that surviving firms comprise a material
component of an economy, so only listed firms will provide important insights to our analyses
of capital structure information. Table below shows the simple descriptive statistics of the
variables (n =171), disregarding the panel information.
5
Proceedings of World Business, Finance and Management Conference
14 - 15 December 2015, Rendezvous Grand Hotel, Auckland, New Zealand
ISBN: 978-1-922069-91-7
Table 1: Descriptive statistics of dependent and independent variables
Longterm
liability
Total
liability
Current
liability
Asset
growth
Operating
revenue
Profitability
Asset
tangibility
Depreciation
ratio
Growth
Mean
18.0949
18.9479
17.9582
0.2577
19.4497
0.9033
19.8949
0.0179
1.5834
Median
17.7616
19.1021
18.1484
0.1443
19.0705
0.9065
20.0000
0.0140
0.9334
Maxim um
22.4236
23.5444
23.4492
2.3135
23.9159
0.9752
24.1617
0.0893
25.000
Minim um
13.2375
13.5424
9.3955
0.0000
14.6831
0.7723
14.5960
0.0001
0.0000
Std. Dev.
2.1879
2.5513
3.2371
0.3161
2.1633
0.0309
2.3366
0.0144
2.6384
Probability
0.0351
0.0141
0.0428
0.0000
0.0694
0.0000
0.0311
0.0000
0.0000
4.1
Measurement of variables:
Dependent variables
There are different ways of calculating leverage in the capital structure.
Total debt ratio: Total debt to total assets (Michaelas et al. 1999) but Delcoure (2007
, p 269) has modified it by suggesting the book value of debt to book value of debt plus
the book value of equity.
Long term debt ratio: long term debt by total assets (Michaelas et al. 1999, Delcoure
2007)
Short term debt ratio:
short term debt to total assets (Michaelas et al. 1999,
Delcoure 2007)
Independent variables
Size: Natural logarithm of Sales (Titman and Wessels 1988, Michaelas et al. 1999,
Delcoure 2007, De Vries and Erasmus 2010)
Profitability: Earnings before interest and taxes divided by Total Assets (Michaelas
et al. 1999, De Vries and Erasmus 2010)
Growth: Percentage growth in total assets over a three-year period (Titman and
Wessels 1988, Michaelas et al. 1999, Delcoure 2007)
Depreciation ratio Depreciation expenses divided by total Assets (Michaelas et al.
1999, Delcoure 2007)
Tangibility of assets: Total fixed assets divided by total assets (Michaelas et al. 1999,
Delcoure 2007)
Industry dummies: Selected sector dummy is set to ‘1’ value, otherwise a ‘0’ value
for each subject. The process is repeated for six sectors and one sector is kept as a
benchmark sector for comparison purposes (no dummies assigned (Delcoure 2007).
4.2
Model specification
The study uses two different models, a constant co-efficient model (pooled regression model)
and a panel regression model (time series regression). The pooled regression model ignores
time and industry effects, assumes a constant value for the intercept for the entire sample,
and the slope co-efficient are the same for the sample (171 observations in this study). The
model is specified below.
LGit = β0 + β Xit + ϵit
6
Proceedings of World Business, Finance and Management Conference
14 - 15 December 2015, Rendezvous Grand Hotel, Auckland, New Zealand
ISBN: 978-1-922069-91-7
where LGit = the measure of leverage of a firm I at time t; β0 = the intercept of the equation; β
= the change coefficient of Xit = the different independent variable for a leverage of a firm i at
time t; i = the number of the firms (sample size) = 1, 2, …57, and t = the time periods (1,
2...N).
In order to examine the relationship between leverage and firm level variables, panel data is
used. Panel data uses time and cross-sectional data, increases data points in a single study,
increases data points, increases degree of freedom which in turn reduces multi-collinearity
among explanatory variables leading to improved econometric estimates (Shah and Khan
2007). In order to choose the most appropriate model that takes into account industry and
time effects, Hausman Test is used for model selection. The equations of the models, the
fixed effect and the random effects model, are as follows:
Fixed effect model:
LGit = aoi + β 1 X 1it + β2 X 2 it…+ U it
The subscript i in the intercept represents different intercepts across sectors but the ‘i’ used
with the slopes assumes the same slope across different industry sectors. Therefore, the
following Random effects model is developed for this study.
Random effect model:
LGit = μ+ + β X it + (Ui + ϵit)
Where LGit is the i th individual at time, X it is a vector of independent variable, β is a vector
of parameters, and (Ui + ϵit) is the error term.
5.
Results and discussions
The table below shows the best pooled regression model (backward method) including the
industry dummies. The best pooled model in Table 2 below shows only the significant
determinants of capital structure (at 5% level of significance).
Table 2: Pooled regression analysis
Model
(Constant)
Operating_revenue
TangibiltyofAssets
Dep_ratio
Asset_growh
Materials
softwareequop
Telecom
Co-efficientsa
Unstandardized
Standardized
Coefficients
Coefficients
Std.
B
Beta
t-value
Error
-1.663
0.294
-5.664
0.517
0.040
0.511
12.961
0.480
0.037
0.513
13.009
4.676
2.003
0.031
2.334
0.160
0.089
0.023
1.796
-0.148
0.070
-0.027
-2.117
0.186
0.082
0.031
2.263
0.235
0.111
0.028
2.115
Probability
0.0000***
0.0000***
0.0000***
0.02100**
0.074000*
0.03600**
0.02500**
0.03600**
7
Proceedings of World Business, Finance and Management Conference
14 - 15 December 2015, Rendezvous Grand Hotel, Auckland, New Zealand
ISBN: 978-1-922069-91-7
a. Dependent Variable: B_LONGTERMLIABILITY
*** represents 1% level of significance, **represents 5% level of significance, *
represents 10% level of statistical significance, and * represents significance
at 10% level.
The best model in Table 2 shows four firm level explanatory variables and three industry
sector specific factors at different significance levels marks with the number of asterisk/s (*).
A pooled regression model does not consider time effect, so two additional models are run in
the Eviews software. The first model uses only firm-specific panel data and reports three
statistically significant independent variables. Earnings Before Interest and Taxes (EBIT),
Depreciation Ratio, and profitability, explain the variability of long term liability (in absolute
terms) of sample. This model, however, has not considered time and industry effects at the
same time, so panel models are used to determine time and industry effects.
Table 3: Panel regression analysis results (without industry dummies)
Variable
t-statistic
Probability
Constant
Coefficient
-2.076712
-2.261851
0.0250
Depreciation ratio
4.893158
2.237506
0.0266**
Size
Tangibility
0.525015
0.457126
12.70309
12.28565
0.0000**
0.0000**
Profitability
0.890949
0.901469
0.3687
Growth
-0.016724
-1.589905
0.1138
** represents 5% level of statistical significance
Table 3 above shows that three independent variables, size, depreciation ratio (DEP_RATIO)
and tangibility of assets, are important determinants of capital structure of Australian service
sector firms. In the table, only long term liability is used as the best descriptor of leverage.
The model above, however, does not consider industy and time effects. Table 4 below reports
the output of a panel model with industy dummies. Based on Hausman test, the output of the
random effects model is presented (null hypohesis that random model is appropriate is
accepted at 5% level, p= 0.57).
Table 4: Panel regression analysis results (with industry dummies)
Variable
Co-efficient
t-statistic
probability
C
12.605
1.8614
0.0645
EBIT
0.9279
2.3206
0.0216**
ASSET_GROWH
0.0671
0.8512
0.3959
TANGIBILTY_OFASSET...
0.2867
0.8019
0.4238
DEP_RATIO
15.6194
2.1387
0.0340**
PROFITABILITY
-17.8994
-2.3637
0.0193**
PHARMA
0.2699
1.3153
0.1903
SOFTWAREEQUOP
0.3510
2.2022
0.0291**
UTILITIES
0.3172
1.7542
0.0813*
TELECOM
0.3329
1.6533
0.1002
8
Proceedings of World Business, Finance and Management Conference
14 - 15 December 2015, Rendezvous Grand Hotel, Auckland, New Zealand
ISBN: 978-1-922069-91-7
ENERGY
-0.0876
-0.5497
0.5833
HEALTHCAREEQUIPMEN...
0.1308
0.9270
0.3553
*represents 10% level of significance, ** represents 5% level of significance
Table 4 above shows that three independent variables, EBIT, depreciation ratio
(DEP_RATIO) and profitability (PROFITABILITY), are found as important determinants of
capitl structure of Australian service sector firms. In the table, only long term liability is used
as the best descriptor of leverage. The model above, considers industy effects but no time
effect due to an insufficient number of samples. Out of seven industry sectors, only software
sector (significant at 5% level) and utilities sector (significant at 10% level) affects capital
structure signifcantly.
The results of the tables above suggest EBIT, depreciation ratio and profitability as three
generic significant determinant of capital structure of Australian service sector firms. The first
factor, profitability, is related to a firm’s choice of sourcing capital during the times of need.
Following Myers (1984) and his Pecking Order Theory (PET), a firm prefers internal to
external financing when it comes to raising additional capital. Therefore, a negative correlation
is expected between a firm’s leverage and profitability. In the first model above, profitability
has a positive co-efficient, confirming the PET expectations. Whereas the tradeoff theory
(Jensen 1986)(Jensen, 1986) suggests tax deductibility and tax bill reductions as the prime
reasons for debt usage. Firms with high profitability are more affluent and hence a positive
association between profitability and leverage is expected. The result confirms the trade-off
theory hypotheses. The co-efficient of profitability is not statistically significant at any level of
statistical significance as three standardized beta coefficients are negative. However, when
sector dummies are used in the second panel model in Table 4, the results show a negative
co-efficient for profitability in line with the predictions of trade-off theory of capital structure.
The second important determinant is the depreciation ratio in Table 3 and 4. Depreciation and
other non-debt tax deductible items are substitutes for tax shields on debts (DeAngelo and
Masulis 1980) So a negative association between depreciation ratio and leverage is expected.
The outputs of the two models in Table 3 and 4 reveal positive beta coefficients for the debt
ratio (all significant at 5% level) suggesting that Australian firms in the sample are highly
levered, that is, use more long term debts to fund their assets.
The third determinant, tangibility of assets, is statistically significant (co-efficient value 0.4571
and p = 0.0000). The positive coefficient means for each percentage point increase in tangible
assets, debt also increases by 0.4571%. This finding is consistent with the predictions of
Jensen and Meckling’s (Jensen and Meckling 1976) agency theory and Myer’s (1984)tradeoff theory. The finding is also consistent with findings in other countries and different from the
findings in other countries. The results suggest that Australian firms are still insufficiently
levered. As a result, the owners are not earning enough from their investments, and the value
of the firms in the service sector has not reached to the optimal level. A mirror image of this
explanation is that Australian firms are using more equity to fund their operations, which
means less debt is used, implying suboptimal earning from the existing configuration of capital
structure. This can be improved by increased borrowing level.
The fourth factor, growth, has a negative co-efficient (-0.016721) and is not statistically
significant at 5% level (p = 0.1138) but insignificant in the second model in Table 3 and Table
4 (coefficient value is 0.2867, p= 0.4238). This result suggests that Australian service sector
9
Proceedings of World Business, Finance and Management Conference
14 - 15 December 2015, Rendezvous Grand Hotel, Auckland, New Zealand
ISBN: 978-1-922069-91-7
firms are not funding their operations from long-term debt. The finding is dissimilar to Shah
and Khan (2007). Some plausible explanation can be the developing nature of the sector and
increased risk perceptions by debtors to lend money to the service-sector firms. Borrowing
by rising firms may be perceived risky and may attract a high risk premium, and resultant high
debt interest rate demands by creditors. The second explanation may be the risk perception
of managers and reluctance to take risks when the service-sector firms are growing. This is
to save managers’ own jobs as well and to invest in a stable project only ( see, for example,
Shah and Khan 2007).
Size is an important determinant of capital structure in almost all studies. In this study, size is
measured by natural log of sales. The coefficient value is positive (0.5250) and statistically
significant at 5% level (p= 0.0000). The results suggest that Australian service sector
organizations are enjoying the benefits of increased revenue from leverage, that is, for each
percentage point of an increase in revenue, long-term debt increases by 0.52%. Prior
literature has two opposing expectations, a positive relationship (Titman and Wessels 1988)
and a negative relationship (Rajan and Zingales 1995). A positive association is expected
when investors have confidence in a firm and are willing to lend without the fear of bankruptcy,
so debt is cheaper (Shah and Khan 2007) whereas, when a negative relation is expected, it
implies less confidence and more asymmetric information held by firms or a rising sector
(Shah and Khan 2007). The results support the view of Titman and Wessels (1988),
suggesting that the Australian service sector is quite established, it has the confidence of the
debt market, and that the firms in the sector are able to utilize debt effectively for expansion
of their operations.
Sector dummies are used to study industry effects on leverage decisions. Industries from two
different sectors reveal a significant level of leverage in their capital structure. These two
sectors, software and utilities, are approximately 35% more levered than the benchmark
sector (the materials sector), suggesting the firms in this sector are more established and
profitable.
6.
Conclusions
The current study examines firm characteristics and industry sector effects of the Australian
service sector listed firms in the ASX. Using pooled and panel regression models, the study
reports some significant firm characteristics and industry sectors that affect capital structure
choices. Due to variations in the way capital structure is defined in extant research, the results
of this study also report different determinants when a capital structure is defined differently,
and control variables are changed to get the results. The pooled regression model reports
some control variables, long term debt and some industry sectors as significant determinants
of capital structure of the firms. The panel regression (without industry effect) reports
tangibility, size and depreciation as important determinants of capital structure of the firms.
When industry effect is considered, a different set of control variables is used to get a
meaningful result. The random effect model suggests that Earnings before interest and taxes
(EBIT), profitability and depreciation ratio are significant firm level determinants and Software
and Utilities sectors are significantly more levered than the benchmark sector.
The study is based on 57 firms, over two years, 2012 to 2014 and a total of 171 firm-year
observations. A larger sample size, either by including more firms or more number of years,
may reveal more robust results than reported in this paper. The study uses only listed
10
Proceedings of World Business, Finance and Management Conference
14 - 15 December 2015, Rendezvous Grand Hotel, Auckland, New Zealand
ISBN: 978-1-922069-91-7
companies in the service sector, so it suffers from ‘survivorship bias’ (De Vries and Erasmus
2010). To overcome this limitation, inclusion of non-listed companies may be considered.
Time dummies tell important information about the movements of leverage structure over
time. Due to an insufficient number of observations, time dummies cannot be included in
panel regression analysis. Inclusion of more time periods may overcome the time dummy
related data processing problems. The exploratory nature of this study is based on only six
independent variables. Data availability on more independent variables will obviously
overcome this shortcoming.
7.
REFERENCES:
Abor, J., 2007. Corporate governance and financing decisions of Ghanaian listed firms.
Corporate Governance: The international journal of business in society, 7, 83-92.
Allen, D.E., 1991. The determinants of the capital structure of listed Australian companies:
the financial manager's perspective. Australian Journal of Management, 16, 103-128.
Allen, D.E., 1993. The pecking order hypothesis: Australian evidence. Applied Financial
Economics, 3, 101-112.
Baxter, N.D., 1967. Leverage, risk of ruin and the cost of capital*. the Journal of Finance, 22,
395-403.
Cassar, G. and Holmes, S., 2003. Capital structure and financing of SMEs: Australian
evidence. Accounting and Finance, 43, 123-147.
De Vries, A. and Erasmus, P., 2010. Determinants of capital structure: A South African study.
Corporate Ownership and Control, 8, 591-599.
Deangelo, H. and Masulis, R.W., 1980. Optimal capital structure under corporate and
personal taxation. Journal of financial economics, 8, 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,
14, 387-405.
Delcoure, N., 2007. The determinants of capital structure in transitional economies.
International Review of Economics & Finance, 16, 400-415.
Donaldson, G., 1984. Managing corporate wealth. New York: Praeger.
Drobetz, W. and Wanzenried, G., 2006. What determines the speed of adjustment to the
target capital structure? Applied Financial Economics, 16, 941-958.
Fan, J.P., Titman, S. and Twite, G., 2012. An international comparison of capital structure
and debt maturity choices. Journal of Financial and quantitative Analysis, 47, 23-56.
Giannetti, M., 2003. Do better institutions mitigate agency problems? Evidence from
corporate finance choices. Journal of Financial and quantitative Analysis, 38, 185-212.
Hall, G.C., Hutchinson, P.J. and Michaelas, N., 2004. Determinants of the capital structures
of European SMEs. Journal of Business Finance and Accounting, 31, 711-728.
Jensen, M.C., 1986. Agency cost of free cash flow, corporate finance, and takeovers.
Corporate Finance, and Takeovers. American Economic Review, 76.
Jensen, M.C. and Meckling, W.H., 1976. Theory of the firm: Managerial behavior, agency
costs and ownership structure. Journal of financial economics, 3, 305-360.
Kraus, A. and Litzenberger, R.H., 1973. A state‐preference model of optimal financial
leverage. the Journal of Finance, 28, 911-922.
Mac an Bhaird, C. and Lucey, B., 2010. Determinants of capital structure in Irish SMEs. Small
business economics, 35, 357-375.
11
Proceedings of World Business, Finance and Management Conference
14 - 15 December 2015, Rendezvous Grand Hotel, Auckland, New Zealand
ISBN: 978-1-922069-91-7
Michaelas, N., Chittenden, F. and Poutziouris, P., 1999. Policy implications arising from the
impact of the last recession on the capital structure of UK SMEs. Environment and
Planning C: Government and Policy, 17, 411-430.
Modigliani, F. and Miller, M.H., 1958. The cost of capital, corporation finance and the theory
of investment. The American economic review, 261-297.
Modigliani, F. and Miller, M.H., 1963. Corporate income taxes and the cost of capital: a
correction. The American economic review, 433-443.
Myers, S.C., 1984. The capital structure puzzle. the Journal of Finance, 39, 574-592.
Myers, S.C. and Majluf, N.S., 1984. Corporate financing and investment decisions when firms
have information that investors do not have. Journal of financial economics, 13, 187221.
Pettit, R.R. and Singer, R.F., 1985. Small business finance: a research agenda. Financial
management, 47-60.
Qiu, M. and La, B., 2010. Firm characteristics as determinants of capital structures in
Australia. International Journal of the Economics of Business, 17, 277-287.
Rajan, R.G. and Zingales, L., 1995. What do we know about capital structure? Some evidence
from international data. the Journal of Finance, 50, 1421-1460.
Shah, A. and Khan, S., 2007. Determinants of capital structure: Evidence from Pakistani panel
data. International review of business research papers, 3, 265-282.
Song, J. and Philippatos, G., 2004. Have we resolved some critical issues related to
international capital structure? Empirical evidence from the 30 OECD countries.
Stulz, R., 1990. Managerial discretion and optimal financing policies. Journal of financial
economics, 26, 3-27.
Titman, S. and Wessels, R., 1988. The determinants of capital structure choice. the Journal
of Finance, 43, 1-19.
Twite, G., 2001. Capital structure choices and taxes: evidence from the Australian dividend
imputation tax system. International Review of Finance, 2, 217-234.
12