Professional Documents
Culture Documents
To cite this article: Tanveer Ahsan, Wang Man & Muhammad Azeem Qureshi (2015): Mean
reverting financial leverage: theory and evidence from Pakistan, Applied Economics, DOI:
10.1080/00036846.2015.1080802
Article views: 21
Download by: [University of Sussex Library] Date: 13 September 2015, At: 01:05
APPLIED ECONOMICS, 2015
http://dx.doi.org/10.1080/00036846.2015.1080802
a
School of Accounting, Dongbei University of Finance and Economics, Dalian, P.R. China; bOslo Business School, Oslo and Akershus
University, College of Applied Sciences, Oslo, Norway
ABSTRACT KEYWORDS
Grounding concepts of the two competing theories of capital structure (trade-off theory, Target capital structure;
pecking order theory) are quite opposite to each other. Trade-off theory claims that there is Pakistan; unit root analysis;
an optimal (target) capital structure and firms try to achieve that optimal (target) point. panel data
Whereas pecking order theory argues that there is no optimal (target) capital structure but JEL CLASSIFICATION
the firms follow a specific pattern of financing. Using the two competing theoretic frame- C23; G30; G32
works, this study applies Fisher-type panel unit root test to an unbalanced panel data of
Downloaded by [University of Sussex Library] at 01:05 13 September 2015
13 115 firm-year observations of nonfinancial firms listed on Karachi Stock Exchange Pakistan
spread over 38 years (1973–2010). Overall panel test results, for short-term, long-term, as well
as total leverage support trade-off financing behaviour while individual firm results do not.
Individual firm results show that only 16% of the firms have short-term target, 25% of the
firms have long-term target and 12% of the firms have total target leverage ratio. Further,
industry results explain that most of the industries do have target leverage ratios and
classification of data into profitable and lossmaking firm-year observations explains that
profitable firms clearly follow trade-off financing behaviour while the results for lossmaking
firms do not support trade-off financing behaviour. Our study indicates that it is important for
the government to ensure policies to develop well-balanced financial markets and to improve
accountability systems.
Wang, and Qureshi 2015). All of these studies use used. Leverage of firms reverts towards its mean
regression analysis (multiple regression, panel data (target) value if the data are stationary, and hence
regression) and none has used unit root test which is the firms follow TOT. However, nonstationary data
a useful way not only to typify the available data and suggest that leverage of firms does not revert towards
test specific theory but also prevents researchers its mean (target) value and the firms may follow POT.
from producing spurious relationships (Granger We use this concept to gather empirical evidence
and Newbold 1974). To overcome this common about financial behaviour of Pakistani listed nonfi-
limitation observed in earlier studies carried out in nancial firms. We choose only nonfinancial firms
Pakistan, we propose to apply panel data unit root for two reasons. First, financial firms in Pakistan are
test to a unique large unbalanced panel data set of regulated and their capital structure decision may be
Pakistani listed nonfinancial firms to check if their affected by the regulations. Second, financial firms in
capital structure is mean-reverting. Pakistan are subject to different tax rate as compared
The financial as well as judicial institutions to the tax rates applicable to the nonfinancial firms.
are weak in Pakistan (www.govindicators.org). The TOT builds its arguments on the tax rate and
Downloaded by [University of Sussex Library] at 01:05 13 September 2015
Operating with three stock exchanges and Karachi consequent debt tax shield. As such, excluding finan-
Stock Exchange (KSE) as the biggest stock exchange, cial firms from our data set was necessary considering
Pakistan currently has a listed capital of around US$ the objective of the study. Moreover, we classify the
12 billion, market capitalization of around US$ 74 data into profitable and lossmaking firm-year obser-
billion, whereas the listed debt capital is only US$ vations as according to POT internal funds are most
150 million (www.kse.com.pk). These figures clearly preferable funds and profitability is a main source of
suggest that Pakistan is a bank-based economy and internal funds.
capital market of Pakistan is underdeveloped. That is Instead of sample selection, we use an extended
why raising bank debt is easy for firms here as com- data set of all Pakistani nonfinancial firms listed on
pared to raising equity. Furthermore, higher total KSE taken from the State Bank of Pakistan (SBP)
leverage ratios with the mean of 0.8028 (Table 2) publications for the period 1973 to 2010 (SBP
also confirm it. Moreover, higher average short-term 1973–2010) to examine whether Pakistani firms fol-
leverage ratio (mean = 0.5960) as compared to aver- low target capital structure or not. Our data set
age long-term leverage ratio (mean = 0.2068) suggests comprises of 13 industries and 670 listed firms.
that short-term debt financing is easier to avail as Consequently, it produces 13 115 firm-year observa-
compared to long-term debt financing (Table 2). tions (after excluding potential outliers) over the
Apart from the introduction presented in Section study period. In Table 1, we summarize the data
I, we organize rest of the article in the following used in this study.
manner: Section II describes theoretical framework
and the data as well as introduces the methodology;
Section III presents the empirical results and their Table 1. Number of firms and observations in each industry.
Cumulative
discussion; and finally, Section IV is about the con- Number Number of number of
clusions drawn, policy implications and suggested S. no. Industry of firms observations observations
directions for future research. We provide the refer- 1 Cement 25 482 482
2 Chemical 56 1037 1519
ences at the end. 3 Engineering 67 1290 2809
4 Fuel and energy 34 571 3380
5 Jute 09 210 3590
6 Miscellaneous 88 1666 5256
II. Theoretical framework, data and 7 Paper and board 17 389 5645
methodology 8 Sugar 42 1074 6719
9 Textile 255 4940 11 659
10 Ancillary textile 37 690 12 349
Empirical studies (Frank and Goyal 2009; Jõeveer products
2013) use mean leverage ratio as a proxy of target 11 Tobacco 07 178 12 527
12 Transport and 14 193 12 720
leverage for firms and stationarity property of the communications
data necessitates reversion towards its mean value. 13 Vanaspati 19 395 13 115
Notes: We create 100 percentiles of the data set and drop 1st and 100th
This property can be used to identify if there is target percentile as being extreme. The table presents the firm-year observa-
leverage. For this purpose, unit root analysis can be tions after omitting 1st and 100th percentiles.
APPLIED ECONOMICS 3
The growing use of panel data in empirical studies III. Empirical results and their discussion
has forced econometrics to extend time series unit root
In this section, first we present descriptive statistics
tests to panel data. Consequently, there are a number
of leverage ratios used in the study. Then we present
of unit root tests available for panel data such as LLC
and discuss overall, industry-level and firm-level
(Levin, Lin, and Chu 2002), IPS (Im, Pesaran, and Shin
empirical results of unit root analysis for three dif-
2003) and Fisher type (Choi 2001; Maddala and Wu
ferent measurements of leverage ratio.
1999) tests. These panel data unit root tests are differ-
ent from each other in their technique and assump-
tions. Considering its advantages (Maddala and Wu Descriptive statistics
1999; Baltagi 2005), we choose Fisher test. Moreover, it Table 2 presents three leverage measures along with
allows time dimension to be different for each i and as their descriptive statistics for overall data set and
such it does not require a balanced panel (Choi 2001) separately for profitable and lossmaking firms.
and considered a preferred choice for our study that Overall mean value of 0.8028 for total leverage
uses unbalanced panel of 13 115 firm-year observa- ratio explains that Pakistani listed nonfinancial
Downloaded by [University of Sussex Library] at 01:05 13 September 2015
tions. In the following equation, we present our panel firms are highly leveraged. Further, overall mean
data unit root (Fisher test) model: value of 0.5960 for short-term leverage ratio as com-
pared to the mean value of 0.2068 for long-term
X
N
leverage ratio explains that short-term financing is
P ¼ 2 lnρi
i¼1
the main source for these firms. Furthermore, the
mean leverage ratios of lossmaking firms (0.7999 for
where N is cross-sectional (firm-year) dimension short-term leverage; 0.2969 for long-term leverage
and ρi is the combination of p-values of unit root and 1.0969 for total leverage) are higher than those
tests carried out for each cross section i. Furthermore, of profitable firms (mean value of 0.4939 for short-
−2 ln ρi follows χ2 distribution with 2 degrees of term leverage; 0.1616 for long-term leverage and
freedom. 0.6556 for total leverage). This indicates that in a
This model contains the following hypothesis: bank-based economy the only financing source with
HO : j ρi j ¼ 1 for all i’s (all panels are lossmaking firms is the banks.
nonstationary)
HA : j ρi j < 1 for some i’s (some panels are non- Unit root analysis
stationary while others are not)
We use Eviews® 7.2 to apply Fisher-type panel unit In this section, we present the results of unit root
root test on different leverage ratios used in this analysis for three leverage ratios separately. We car-
study. We measure leverage using three proxies: ried out unit root analysis for complete data set
short-term leverage as the ratio of short-term debt (13 115 firm-year observations) and then for 13
to total assets; long-term leverage as the ratio of industries separately to overcome any plausible effect
long-term debt to total assets; and total leverage as of different number of firm-year observations for
the ratio of total debt to total assets. different industries.
Table 3. Overall, industry- and firm-level unit root results for short-term leverage.
Number of firms Number of observations Probability Number and
Lag length percentage of firms
based on ADF-Fisher ADF- Choi with stationary
Short-term leverage Available Included Dropped Available Included Dropped AIC χ2 z-stat behaviour.
Overall 670 650 20 13 115 11 954 1161 0–9 0.000 0.000 104/650 (16%)
Cement 25 23 02 482 423 59 0–5 0.000 0.000 5/23 (22%)
Chemical 56 54 02 1037 948 89 0–5 0.000 0.003 8/54 (15%)
Engineering 67 63 04 1290 1174 116 0–7 0.010 0.115 7/63 (11%)
Fuel and energy 34 31 03 571 519 52 0–4 0.012 0.006 4/31 (13%)
Jute 09 09 00 210 194 16 0–5 0.275 0.726 1/09 (11%)
Miscellaneous 88 83 05 1666 1523 143 0–7 0.007 0.275 8/83 (9.6%)
Paper and board 17 17 00 389 369 20 0–1 0.007 0.010 2/17 (12%)
Sugar 42 41 01 1074 980 94 0–9 0.000 0.000 6/41 (15%)
Textile 255 252 03 4940 4510 430 0–9 0.000 0.000 52/252 (21%)
Ancillary textile products 37 37 00 690 627 63 0–9 0.000 0.014 7/37 (19%)
Tobacco 07 07 00 178 169 09 0–2 0.193 0.226 1/07 (14%)
Transport and 14 14 00 193 176 17 0–1 0.021 0.064 2/14 (14%)
communications
Vanaspati 19 19 00 395 342 53 0–6 0.484 1.000 1/19 (5.3%)
Downloaded by [University of Sussex Library] at 01:05 13 September 2015
Table 4. Overall and industry-level unit root results of lossmaking firms for short-term leverage.
Number of firms Number of observations Lag length Probability
based on ADF-Fisher ADF- Choi
Short-term leverage Available Included Dropped Available Included Dropped AIC χ2 z-stat
Overall 573 415 158 4374 3550 824 0–3 0.000 NA
Cement 22 16 06 159 127 32 0–2 0.153 0.083
Chemical 38 25 13 206 154 52 0–1 0.102 0.529
Engineering 53 36 17 371 292 79 0–3 0.444 0.592
Fuel and energy 19 04 15 62 34 28 0–1 0.599 0.984
Jute 09 07 02 66 55 11 0–1 0.032 0.074
Miscellaneous 73 49 24 502 392 110 0–3 0.000 0.206
Paper and board 15 12 03 121 103 18 0–1 0.000 0.017
Sugar 41 34 07 330 274 56 0–1 0.028 0.222
Textile 238 187 51 2034 1692 342 0–3 0.001 NA
Ancillary textile products 30 20 10 240 196 44 0–2 0.764 0.977
Tobacco 05 03 02 31 23 08 0–0 0.225 0.364
Transport and communications 11 06 05 63 47 16 0–0 0.693 0.751
Vanaspati 19 16 03 189 161 28 0–3 0.615 0.999
Downloaded by [University of Sussex Library] at 01:05 13 September 2015
Table 5. Overall and industry-level unit root results of profitable firms for short-term leverage.
Number of firms Number of observations Lag length Probability
based on ADF-Fisher ADF- Choi
Short-term leverage Available Included Dropped Available Included Dropped AIC χ2 z-stat
Overall 638 533 105 8741 7761 980 0–9 0.000 0.000
Cement 24 21 03 323 279 44 0–5 0.001 0.001
Chemical 51 50 01 831 760 71 0–5 0.000 0.000
Engineering 66 51 15 919 807 112 0–7 0.000 0.000
Fuel and energy 34 31 03 509 458 51 0–4 0.001 0.000
Jute 09 08 01 144 129 15 0–1 0.084 0.419
Miscellaneous 85 68 17 1164 1041 123 0–4 0.004 0.022
Paper and board 14 13 01 268 242 26 0–4 0.000 0.000
Sugar 42 39 03 744 677 67 0–7 0.000 0.000
Textile 239 191 48 2906 2543 363 0–9 0.000 0.000
Ancillary textile products 36 30 06 450 400 50 0–5 0.000 0.000
Tobacco 07 06 01 147 136 11 0–2 0.176 0.163
Transport and communications 12 10 02 130 113 17 0–1 0.019 0.008
Vanaspati 19 15 04 206 176 30 0–4 0.080 0.192
Table 6. Overall industry and firm-level unit root results for long-term leverage.
Number of firms Number of observations Probability Number and
Lag length percentage of firms
based on ADF-Fisher ADF- Choi with stationary
Long-term leverage Available Included Dropped Available Included Dropped AIC χ2 z-stat behaviour.
Overall 670 642 28 13 115 11 828 1287 0–8 0.000 0.000 161/642 (25%)
Cement 25 23 02 482 427 55 0–6 0.000 0.000 7/23 (30%)
Chemical 56 54 02 1037 946 91 0–7 0.000 0.000 16/54 (30%)
Engineering 67 61 06 1290 1166 124 0–4 0.000 0.000 15/61 (25%)
Fuel and energy 34 31 03 571 510 61 0–8 0.000 0.003 6/31 (19%)
Jute 09 09 00 210 199 11 0–1 0.005 0.001 1/09 (11%)
Miscellaneous 88 81 07 1666 1462 204 0–8 0.000 0.000 20/81 (25%)
Paper and board 17 17 00 389 351 38 0–5 0.000 0.000 6/17 (35%)
Sugar 42 41 01 1074 1001 73 0–6 0.000 0.000 12/41 (29%)
Textile 255 250 05 4940 4495 445 0–7 0.000 0.000 61/250 (24%)
Ancillary textile products 37 35 02 690 601 89 0–4 0.000 0.000 10/35 (29%)
Tobacco 07 07 00 178 159 19 0–8 0.005 0.004 1/07 (14%)
Transport and 14 14 00 193 177 16 0–1 0.003 0.033 3/14 (21%)
communications
Vanaspati 19 19 00 395 334 61 0–7 0.002 0.861 3/19 (16%)
Vanaspati (in loss) 19 13 06 189 128 61 0–1 0.318 0.959
Vanaspati (profitable) 19 15 04 206 183 23 0–1 0.000 0.000
analysis find no unit root in long-term leverage ratio behaviour of Pakistani listed nonfinancial firms. But
and reject the null hypothesis. It shows stationary long- individual firm results show that only 25% of the over-
term leverage ratio and trade-off long-term financing all firms follow trade-off financing behaviour.
6 T. AHSAN ET AL.
Table 7. Overall, industry- and firm-level unit root results for total leverage.
Number of firms Number of observations Probability Number and
Lag length percentage of firms
based on ADF-Fisher ADF- Choi with stationary
Total leverage Available Included Dropped Available Included Dropped AIC χ2 z-stat behaviour.
Overall 670 650 20 13 115 11 909 1206 0–9 0.000 0.000 76/650 (12%)
Cement 25 23 02 482 425 57 0–8 0.005 0.015 4/23 (17%)
Chemical 56 54 02 1037 948 89 0–5 0.000 0.022 8/54 (15%)
Engineering 67 63 04 1290 1175 115 0–7 0.013 0.134 7/63 (11%)
Fuel and energy 34 31 03 571 521 50 0–3 0.003 0.043 5/31 (16%)
Jute 09 09 00 210 199 11 0–1 0.313 0.538 0/09 (00%)
Miscellaneous 88 83 05 1666 1498 168 0–8 0.003 0.377 7/83 (8.4%)
Paper and board 17 17 00 389 361 28 0–3 0.403 0.820 2/17 (12%)
Sugar 42 41 01 1074 968 106 0–9 0.001 0.001 5/41 (12%)
Textile 255 252 03 4940 4512 428 0–8 0.000 0.000 33/252 (13%)
Ancillary textile products 37 37 00 690 620 70 0–6 0.000 0.283 3/37 (8.1%)
Tobacco 07 07 00 178 161 17 0–8 0.466 0.522 1/07 (14%)
Transport and 14 14 00 193 172 21 0–3 0.494 0.842 0/14 (00%)
communications
Vanaspati 19 19 00 395 349 46 0–6 0.981 1.000 1/19 (5.3%)
Downloaded by [University of Sussex Library] at 01:05 13 September 2015
these industries do not demonstrate trade-off beha- the null hypothesis significantly for industries such
viour for total financing. Individual firm results for as cement, chemical, engineering, miscellaneous,
these industries also support that these industries do sugar, textile and ancillary textile products and
not have total target leverage ratios. We obtain show that the total leverage ratio of profitable firms
ambiguous results for the engineering, ancillary tex- of these industries is stationary. This means profit-
tile products and miscellaneous industries: χ2 prob- able firms of these industries follow trade off-theory
ability rejects null hypothesis but z-stat probability and do have total target leverage ratio. For fuel and
accepts it. To clarify this ambiguity, we further clas- energy, jute, paper and board, tobacco, transport and
sify our data into profitable and lossmaking firm- communications and vanaspati, the analysis accepts
year observations and run the same analysis again. the null hypothesis significantly and shows that the
Table 8 presents the results of unit root analysis of total leverage ratio of profitable firms of these indus-
lossmaking firms for total leverage ratio. According tries is nonstationary. Accordingly, we observe a
to the results, both the probabilities (χ2, z-stat) sig- trade-off financing behaviour in profitable firms of
nificantly accept the null hypothesis and explain that the cement, chemical, engineering, miscellaneous,
Downloaded by [University of Sussex Library] at 01:05 13 September 2015
lossmaking firms do not follow any target leverage sugar, textile and ancillary textile products indus-
ratio for total financing irrespective of the industry tries. On the other hand, there is no target total
type they belong to. leverage ratio even in profitable firms of the fuel
Table 9 presents the results of unit root analysis of and energy, jute, paper and board, tobacco, transport
profitable firms for total leverage ratio. According to and communications and vanaspati industries.
the results, both the probabilities (χ2, z-stat) reject Therefore, we can claim that even profitable firms
Table 8. Overall and industry-level unit root results of lossmaking firms for total leverage.
Number of firms Number of observations Lag length Probability
based on ADF-Fisher ADF- Choi
Total Leverage Available Included Dropped Available Included Dropped AIC χ2 z-stat
Overall 573 415 158 4374 3547 827 0–4 0.013 NA
Cement 22 16 06 159 129 30 0–2 0.516 0.607
Chemical 38 25 13 206 155 51 0–1 0.217 0.601
Engineering 53 36 17 371 295 76 0–2 0.059 0.643
Fuel and energy 19 04 15 62 35 27 0–0 0.887 0.930
Jute 09 07 02 66 55 11 0–1 0.561 0.529
Miscellaneous 73 49 24 502 394 108 0–3 0.069 0.666
Paper and board 15 12 03 121 101 20 0–2 0.018 0.490
Sugar 41 34 07 330 275 55 0–1 0.313 0.886
Textile 238 187 51 2034 1685 349 0–4 0.177 NA
Ancillary textile products 30 20 10 240 192 48 0–2 0.109 0.789
Tobacco 05 03 02 31 23 08 0–0 0.331 0.439
Transport and communications 11 06 05 63 45 18 0–1 0.972 0.997
Vanaspati 19 16 03 189 163 26 0–3 0.488 1.000
Table 9. Overall and industry-level unit root results of profitable firms for total leverage.
Number of firms Number of observations Lag length Probability
based on ADF-Fisher ADF- Choi
Total Leverage Available Included Dropped Available Included Dropped AIC χ2 z-stat
Overall 638 533 105 8741 7769 972 0–6 0.000 0.000
Cement 24 21 03 323 279 44 0–3 0.000 0.000
Chemical 51 50 01 831 750 81 0–5 0.000 0.003
Engineering 66 51 15 919 818 101 0–3 0.000 0.000
Fuel and energy 34 31 03 509 461 48 0–3 0.055 0.077
Jute 09 08 01 144 129 15 0–2 0.327 0.509
Miscellaneous 85 68 17 1164 1020 144 0–6 0.029 0.021
Paper and board 14 13 01 268 247 21 0–4 0.352 0.303
Sugar 42 39 03 744 686 58 0–3 0.007 0.003
Textile 239 191 48 2906 2557 349 0–6 0.000 0.000
Ancillary textile products 36 30 06 450 395 55 0–4 0.000 0.001
Tobacco 07 06 01 147 134 13 0–2 0.291 0.269
Transport and communications 12 10 02 130 112 18 0–1 0.065 0.032
Vanaspati 19 15 04 206 181 25 0–1 0.837 0.902
8 T. AHSAN ET AL.
of these industries do not demonstrate trade-off ratio, five industries appear to have a mean-reverting
behaviour for total financing. leverage ratio, suggesting trade-off leverage beha-
viour of these five industries. Whereas five other
industries do not appear to have mean-reverting
IV. Conclusions and policy implications
total leverage ratio, suggesting that these industries
We apply panel unit root test developed by Fisher on may not observe trade-off leverage behaviour to
a unique and large data set comprising of 13 115 make their total financing decisions. The remaining
firm-year observations of Pakistani nonfinancial three industries have ambiguous results.
firms listed on KSE spread over 38 years (1973– Classification into profitable and lossmaking firm-
2010) to provide empirical evidence on their lever- year observations clarifies this ambiguity and
age behaviour. We use three leverage proxies to explains that lossmaking firms do not have any
investigate corporate leverage behaviour at three total target leverage ratio irrespective of industry
levels: overall, 13 different industries and at firm type while for profitable firms seven industries
level. (cement, chemical, engineering, miscellaneous,
Downloaded by [University of Sussex Library] at 01:05 13 September 2015
Overall analyses for all three leverage ratios depict sugar, textile, ancillary textile products) do have
stationary behaviour, suggesting that Pakistani listed target total leverage ratio and six industries (fuel
nonfinancial firms generally have target leverage and energy, jute, paper and board, tobacco, transport
ratios that are mean-reverting in line with the and communications, vanaspati) do not have any
TOT. Therefore, these firms, in general, carry out target total leverage ratio. Individual firm results
long-run capital structure planning and try to stick show that only 12% of the firms follow trade-off
to it. Further, industry-level analysis explains differ- total financing behaviour.
ent results for three different ratios of leverage. For The results of the study suggest that firms of most
short-term leverage ratio, seven industries have of the industries do have target short-term and total
mean-reverting leverage and as such these seven leverage ratios and follow TOT. For long-term lever-
industries demonstrate trade-off leverage behaviour. age ratio, the results explain that all the firms have
On the other hand, three industries do not have long-term target leverage ratio, except for those
mean-reverting short-term leverage ratio, suggesting belonging to the vanaspati industry. These findings
that these three industries do not depict trade-off suggest that most of the industries demonstrate
leverage behaviour. For the remaining three indus- trade-off leverage behaviour. They have target lever-
tries, the results are mixed. Individual firm results age and try to achieve that target over time.
show that only 16% of the firms follow trade-off Moreover, classification of firms into profitable and
short-term financing behaviour. Moreover, the clas- lossmaking firm-year observations explains the
sification into profitable and lossmaking firm-year results more clearly. The lossmaking firms do not
observations explains that lossmaking firms do not have any target leverage ratio and they may follow
have any target for short-term leverage ratio irre- POT. Further, the leverage ratios of lossmaking
spective of the industry type. Whereas profitable firms are dangerously high such as the mean value
firms do have short-term target leverage ratio, except (1.0969) of total leverage ratio is even greater than 1.
for the firms in the jute, tobacco and vanaspati The peculiar nature of credit market in Pakistan
industries. For long-term leverage ratio, all the provides plausible explanation for this anomaly.
industries except for vanaspati have mean-reverting The debt market in Pakistan is underdeveloped. As
long-term leverage ratio, suggesting a trade-off such the banking sector is the only option to provide
financing pattern. Classification of vanaspati indus- debt to the corporate sector, especially lossmaking
try into profitable and lossmaking firm-year obser- firms. It is pertinent to note that majority of the
vations shows that lossmaking firms of vanaspati banks are either owned by the government or own-
industry do not have any long-term target leverage ers of the corporate sector otherwise a borrower.
ratio while profitable firms of this industry do have Moreover, the majority of the Pakistani firms are
long-term target leverage ratio. Individual firm closely held family-owned firms. As such we argue
results show that 25% of the firms follow trade-off that the weak financial and judicial institutions and
long-term financing behaviour. For total leverage environment of corporate arena in Pakistan, coupled
APPLIED ECONOMICS 9
with deep-rooted family-based politics, help develop Baltagi, B. H. 2005. Econometric Analysis of Panel Data. 3rd
and sustain the collusive networks of debtor and ed. New York: John Wiley & Sons.
Brounen, D., A. de Jong, and K. Koedijk. 2006. “Capital
creditor. Such a nexus, along with poor mechanisms
Structure Policies in Europe: Survey Evidence.” Journal
of corporate governance and the regulatory bodies of Banking & Finance 30: 1409–1442. doi:10.1016/j.
mired by political interferences and nexuses of self- jbankfin.2005.02.010.
serving myopic gains, helps provide them easy access Choi, I. 2001. “Unit Root Tests for Panel Data.” Journal of
to debt which may become nonperforming loan International Money and Finance 20: 249–272.
before being written off using their nexuses.1 doi:10.1016/S0261-5606(00)00048-6.
Donaldson, G. 1961. Corporate Debt Capacity: A Study of
According to a previous study, politically connected
Corporate Debt Policy and the Determination of Corporate
firms borrow 45% more and have 50% higher default Debt Capacity. Boston, MA: Harvard, Graduate School of
rates (Khwaja and Mian 2005). Under these circum- Business Administration.
stances, it may not only be difficult for these firms Fama, E. F., and K. R. French. 2002. “Testing Trade-Off and
but also for the banks to survive, having serious Pecking Order Predictions about Dividends and Debt.”
implications for the development of the capital mar- Review of Financial Studies 15: 1–33. doi:10.1093/rfs/
15.1.1.
Downloaded by [University of Sussex Library] at 01:05 13 September 2015
1
http://www.sbp.org.pk/ecodata/NPL.pdf of State Bank of Pakistan reports PKR 594 billion (approximately US$ 7 billion) as nonperforming loans of banking
sector in Pakistan. Apex court of Pakistan is currently hearing a case in this regard.
10 T. AHSAN ET AL.
Journal of Econometrics 108: 1–24. doi:10.1016/S0304-4076 Qureshi, M. A., M. Imdadullah, and T. Ahsan. 2012. “What
(01)00098-7. Determines Leverage in Pakistan? A Panel Data Analysis.”
Maddala, G. S., and S. Wu. 1999. “A Comparative Study of African Journal of Business Management 6: 978–985.
Unit Root Tests with Panel Data and a New Simple Test.” Ross, S. A. 1977. “The Determination of Financial Structure:
Oxford Bulletin of Economics and Statistics 61: 631–652. The Incentive-Signalling Approach.” The Bell Journal of
doi:10.1111/obes.1999.61.issue-S1. Economics 8: 23–40. doi:10.2307/3003485.
Modigliani, F., and M. H. Miller. 1958. “The Cost of Capital, Shah, A., and T. Hijazi. 2004. “The Determinants of Capital
Corporation Finance and the Theory of Investment.” Structure of Stock Exchange-listed Non-financial Firms in
American Economic Review XLVIII: 261–280. Pakistan.” The Pakistan Development Review 43: 605–618.
Modigliani, F., and M. H. Miller. 1963. “Corporate Income Sheikh, N. A., and M. A. Qureshi. 2014. “Crowding-Out or
Taxes and the Cost of Capital: A Correction.” American Shying-Away: Impact of Corporate Income Tax on
Economic Review 53: 433–443. Capital Structure Choice of Firms in Pakistan.” Applied
Myers, S. C. 1977. “Determinants of Corporate Borrowing.” Financial Economics 24: 1249–1260. doi:10.1080/
Journal of Financial Economics 5 (2): 147–175. 09603107.2014.925053.
doi:10.1016/0304-405X(77)90015-0. Sheikh, N. A., and Z. Wang. 2010. “Financing Behaviour of
Myers, S. C., and N. S. Majluf. 1984. “Corporate Financing Textile Firms in Pakistan.” International Journal of
and Investment Decisions When Firms Have Information Innovation, Management and Technology 1: 130–135.
Downloaded by [University of Sussex Library] at 01:05 13 September 2015
that Investors Do Not Have.” Journal of Financial Sheikh, N. A., and Z. Wang. 2011. “Determinants of Capital
Economics 13 (2): 187–221. doi:10.1016/0304-405X(84) Structure: An Empirical Study of Firms in Manufacturing
90023-0. Industry of Pakistan.” Managerial Finance 37: 117–133.
Qureshi, M. A. 2009. “Does Pecking Order Theory Explain doi:10.1108/03074351111103668.
Leverage Behaviour in Pakistan?” Applied Financial Stulz, R. M. 1990. “Managerial Discretion and Optimal
Economics 19: 1365–1370. doi:10.1080/0960310090281 Financing Policies.” Journal of Financial Economics 26:
7592. 3–27. doi:10.1016/0304-405X(90)90011-N.