INVESTIGATING THE LEVERAGE COMPOSITION OF PAKISTANI FIRMS THROUGH THEIR DETERMINANTS

INVESTIGATING THE LEVERAGE COMPOSITION OF PAKISTANI FIRMS THROUGH THEIR DETERMINANTS FARAH YASSER* Abstract To have an ideal mix of debt and equity in a balance sheet of an entity is till to date a very complicated issue for managers as there is no such rule to predict an optimal capital structure. An in-depth understanding is required for the corporate culture, the degree of the development of the capital market and the economy in which the firms operate. This study seeks to investigate the leverage composition of Pakistani corporations through their determinants. Fixed effect regression is used to show the relationship of determinants of capital structure on leverage corporations listed on Karachi Stock Exchange (KSE) for the period of 2006 to 2013. The results suggest that agency cost, growth, age, and size are significantly and negatively associated with the capital structure of Pakistan firms, however, collateral value of asset is significantly but positively associated with the capital structure of the firm. On the other hand, free cash flows, non debt tax shield, profitability, business risk and bankruptcy cost are not significantly associated with leverage composition of the firms and are against the signaling theory and peaking order theory. The key importance of this study is that no prior research was done for determinants like agency cost, free cash flows, bankruptcy cost and age as determinants of capital structure for Pakistani firms among other determinants. Further, this study does not confine to a particular sector rather it covers all companies listed by Karachi Stock Exchange. Key Words: Leverage, Determinants of Capital Structure, Agency cost, Collateral value of assets, Bankruptcy cost. ** Assistant Professor, School of Commerce and Accountancy, University of Management and Technology, Lahore, Pakistan. Email: farah.yasser@umt.edu.pk Journal of Management and Research Volume 3 Number 1 2016 1. Introduction Capital structure irrelevance theory was introduced by Modigliani and Miller (1958) and since then arguments have progressed for the leverage decision of the firm. MM theory suggests that the value of the firm is independent of its capital structure under certain assumptions. Even though this theory is based on unrealistic assumptions, there are a number of variables that narrate the value to the firm and often identify as determinants of the capital structure e.g. agency cost, collateral value of asset, growth, free cash flows, and age of the firm, business risk, bankruptcy risk, profitability and non debt tax shield. Hence, the main purpose of the firm is to maximize the wealth of the stockholders by evaluating a suitable finance mix. Combination of debt and equity capital becomes the most controversial corporate issue over the past four decades. The capital structure decisions directly influence the market value of the firms and the cost of the firm. How to have an ideal mix of debt and equity in a balance sheet of an entity is till to date a very complicated issue as there is no such rule to predict an optimal capital structure. An in-depth understanding is required for the corporate culture, the degree of the development of the capital market and the economy in which the firms operate. Firms can only achieve their objectives through skillful intellectual managers and the management can perform better without thinking about finance shortage or finance mix. So soon after the arrival of (Modigliani & Miller, 1958), the western world started financing their corporations without bothering the mode of finance i.e. debt or equity and forgot about the equity debt mix and just focused on growth and achievement of commercial business objectives. Companies achieved local and international remarkable growth. Due to rapid growth historic numbers soon became meaningless. Corporations started focusing more about their market values and future fund flows rather than the book value of assets. Financial ratios are also designed on the market values of the company. Human capital also becomes an important part of the company. Pakistan is one of the developing countries with lot of issues involving unstable micro and macroeconomic situations, political crisis, social behavior, geographical Journal of Management and Research Volume 3 Number 1 2016 structure, complicated tax mechanism and non robust legal system. It is equally important to find out those factors that influence a firm’s capital structure choice. However, in case of developing economies, inadequate literature is available regarding capital structure of corporations in Pakistan. Eldomiaty (2008) mentioned that because of the insufficient information problems, the capital markets of the developing economies are not efficient enough to compare with developed market. Therefore, the outcomes of the developed countries cannot be generalized with the developing countries like Pakistan where political risk is very high, foreign currency fluctuations are very frequent, business risk is lofty, and the capital market is in developing phase. Most of the debate in the country on low investment ratios has been centered around factors such as infrastructure, law and order, skill shortages and bureaucratic hassles (Hussain, 2006). The key importance of this study is that no prior research was done for determinants like agency cost, collateral value of assets, free cash flows, bankruptcy cost and age as determinants of capital structure for Pakistani firms. Further, this study does not confine to a particular sector rather it covers all companies listed by Karachi Stock Exchange. Therefore, this study shows an in depth analysis of determinants of capital structure of Pakistani firms. Moreover, this study is very functional for the managers of the corporations and provides guidelines for efficient use of the determinants of capital structure in order to maximize firm performance. Government authorities, taxation bodies and policy makers can also be benefited from the findings of the study. Hence, this study also served the purpose of a rich contribution in the existing literature regarding the determinants of capital structure of firms in Pakistan. 2. Literature Review Theories of corporate structure draw a closer attention in the world of finance when Modigliani and Miller (1958) presented capital structure irrelevance principle. This principle states that in a perfect market, whatever the mode of finance the firm uses, it is irrelevance to the firm’s value. And since then, many other theories were presented with the help of basic idea provided by Modigliani and Miller (1958) like trade off theory, pecking order theory, OLI theory, signaling theory, etc. These theories highlight different Journal of Management and Research Volume 3 Number 1 2016 determinants of capital structure and draw attention to their importance in the perspective of capital structure. Pecking order theory is based on asymmetric information of the firm. In this theory, Myers (1984) explained that how a company prioritized its financing decisions. The main theme of the theory is that the firm takes finances from easier sources first i.e. internal finance. Pecking order theory also suggests that there is an inverse relationship between profitability and leverage. Modigliani and Miller (1958) theorem was a very simple utopian kind of model where it was assumed that there are no tax benefits, no agency costs and bankruptcy cost. And it tried to convince its reader about being inert to whether a capital is used or debt is employed for a firm’s operations. However, M&M proposition II did take into account these costs and explained that although the firm’s WACC decreases with debt inclusion, yet it becomes more risky. The signaling theory talks about firm’s management decisions taken as being signals by the other stakeholders to it. This allows good firms managers to signal to stakeholders about their firm’s value and thus enable them to take decisions accordingly. Such popular signals involve dividend policy, issue of bonus shares, stock splits. 2.1 Determinants of Capital Structure - Around the World Al Amri and Al Ani (2015) examined determinants of capital structure for three sectors (food, construction and chemical) of Omani industrial companies for the period of 2008 -2012. They found that there is a significant and positive relationship between risk and tangibility and leverage and there is a significant but negative relationship between growth rate and profitability and leverage, while there is no association with size. Baltacı and Ayaydın (2014) explored Turkish banking sector and found that capital structure is positively and significantly related with size, industry leverage and GDP growth. They further find that capital structure is significantly but inversely related with financial risk, profitability, tangibility and inflation. Forte, Barros, and Nakamura (2013) investigated capital structure of Brazilian firms and found that profitability is significantly and negatively related to capital Journal of Management and Research Volume 3 Number 1 2016 structure. Also, growth is positively and significantly related to leverage. Further, size is positively related and age is negatively related to the leverage. Mac an Bhaird and Lucey (2010) explored 299 Irish enterprises and found that age, intangibility, collaterals and size are the significant determinants of capital structure. They further found that ownership structure, size, age and collateral are similar across industry. Kouki and Said (2011) conducted research on 244 French listed companies and found that trade off theory, pecking order theory and market timings are not significant. Huang (2006) revealed an inverse relationship between leverage and profitability, non debt tax shield, growth opportunities and managerial shareholdings. 2.2 Determinants of Capital Structure – Pakistan Khan, Jan, and Khan (2015) explored cement sector of Pakistan. By using pooled regression model, they found that there is an inverse relationship of firm size and leverage of the firm which is against static trade off theory. Qadri (2015) investigated Pakistani non financial firms listed on Karachi stock exchange for the period of 2004 to 2012. This study showed a significant and negative association between profitability and leverage, supporting peaking order theory. Moreover, this study showed a significant and positive association of tangibility and size with leverage supporting trade off theory. Masnoon and Saeed (2014) examined automobile sector of Pakistan for the period of 2008 to 2012. This study found that leverage has an inverse and significant relationship with profitability and liquidity, whereas, leverage has a positive insignificantly relationship with earning variability. Ahmed Sheikh and Wang (2011) examined firms listed on Karachi stock exchange for the period of 2003 to 2007 and found that profitability, liquidity, earnings volatility, and tangibility are negatively associated to leverage, while firm size is positively associated to leverage. Non-debt tax shields and growth opportunities do not appear to be significantly related to leverage. Ahmad and Zaman (2013) analyzed textile sector listed on Karachi stock exchange of Pakistan and revealed that profitability and size is significantly and negatively related to leverage while tangibility and growth were positively related to leverage. Journal of Management and Research Volume 3 Number 1 2016 Afza and Hussain (2011) investigated three sectors (automobile, engineering and cable and electric goods) listed on Karachi stock exchange by using pooled data regression model. The results of this study supported static trade off theory and pecking order theory. Hijazi and Tariq (2006) investigated listed firms of cement industry of Pakistan and found that firm size, tangibility , growth and profitability is associated to leverage.Shah and Khan (2007) examined non financial firms listed on Karachi stock exchange for the period of 1994 to 2002 and found that tangibility, volatility and non debt tax shield are significantly related with leverage, hence confirming trade off theory and profitability is significantly associated with leverage confirming pecking order theory, whereas, size is insignificant to leverage. From the above literature review, it can be wrap up that no prior research was done for determinants like agency cost, collateral value of assets, free cash flows, bankruptcy cost and age as determinants of capital structure for Pakistani firms. However, these variables show a significant relationship with leverage in developed countries Baltacı and Ayaydın (2014), Forte et al., (2013), Mac an Bhaird and Lucey (2010) etc. Therefore this study includes all those variables that have already tested in all developed economies studies but first time include in developing ecomomy senario like Pakistan. 3. Theoretical Framework And Hypothesis Development  Agency Cost Agency cost is one of the most important determinants of capital structure. It begins with the conflicts of interest between debt holders and equity holders (Myers, 1977). Firm having high agency cost have high cost of debt and thus leads to have lower debt ratio (Fama, 1980; Jensen, 1986; Titman, 1984). Ho1: There is no significant relationship between agency cost and leverage of Pakistani firms. Journal of Management and Research Volume 3 Number 1 2016  Bankruptcy Cost Bankruptcy cost depends on costs like legal fees, loss of sales, employees and suppliers and the probability of its happenings. If financing through debt increase, the probability of bankruptcy also increases and as a result bankruptcy cost increase. Firms with higher bankruptcy cost have lower debt. Ho2: There is no significant relationship between bankruptcy cost and leverage of Pakistani firms.  Non Debt Tax Shield It is usually argued that company with more non debt tax shields should have less debt since the tax advantage of debt are comparatively less important (Akhtar & Oliver, 2009). Ho3: There is no significant relationship between non debt tax shield and leverage of Pakistani firms.  Profitability Myers (1984) argued that if a firm is more profitable then it will have more internal financing than external sources according to pecking order theory of capital structure. Therefore it can be proposed that the firms with higher profit have higher internal finance and hold less debt. Internal finance is less costly and easier whereas external finance is more costly. Hence it can hypothesize that there is an inverse relationship between profitability and leverage. Ho4: There is no significant relationship between profitability and leverage of Pakistani firms.  Size Bigger firms usually have larger exposure to the public than smaller firms and consequently need to provide more information to consumers, creditors, suppliers, Journal of Management and Research Volume 3 Number 1 2016 forecasters and government personals (Cooke, 1991). Larger firms have more resources to provide relevant information to stakeholders and as a result, larger firms have more debt with more attractive terms as compare to smaller firms. Hence, a direct association is expected between firm size and leverage. Empirical studies suggested size as a determinant of capital structure (Ferri & Jones, 1979; Scott Jr & Martin, 1975) and (Aggarwal, 1990). Ho5: There is no significant relationship between size and leverage of Pakistani firms.  Collateral Value Of Assets Rajan and Zingales (1995) found that tangibility of assets or collateral value of assets is a determinant of capital structure. Corporations with more tangible assets are expected to have more debt because having more tangible assets gets debt easily on more favorable terms. On other side, Graham Jr. (1988) suggested that the corporations having high intangible assets have lower cost of borrowings cause better security for debt holders. Ho6: There is no significant relationship between collateral value to assets and leverage of Pakistani firms.  Business Risk Business risk can be defined as the risk related with the future operations of the company. Firms with less business risk, (the risk that is connected with the upcoming business operations) are assumed more financial risks. Ho7: There is no significant relationship between business risk and leverage of Pakistani firms.  Growth Theoretically it is suggested that the firm with higher growth rate will have lower debt in their capital structure. A company that grows fast invests huge amount in research and development. Ho8: There is no significant relationship between growth and leverage of Pakistani firms. Journal of Management and Research Volume 3 Number 1 2016  Age As the firm grows, more information is available for the firm’s probable viability in the future. More information cause less leverage in the capital structure of the firm. Ho9: There is no significant relationship between age and leverage of Pakistani firms.  Free Cash Flows Jensen (1986) define free cash flows as the cash flow that is left after all positive NPV projects are funded. Harris and Raviv (1991) argued that firm with greater free cash flows will have lesser debt and vice versa. Ho10: There is no significant relationship between free cash flows and leverage of Pakistani firms. Table 1: Proxies for Variables Variables Leverage Agency Cost Free Flows Cash Growth Age Non Debt Tax Shield Size Proxies (Long term debt) / (long term debt +Market value of Equity) (Cash and Marketable Securities) / 3 years of average total Assets (EBIT+Depreciation+Amortization -Tax-Interest-Dividends) / (Average Total Assets) (Change in Total Assets) / (Total Assets) Natural log (age of firm in years from date of incorporation) (Total Annual Depreciation Expense) / Total Assets Natural log of Total Sales Collateral Value of Assets (Fixed Assets) / Total Assets Profitability (Net Income) / Total Sales Business Risk Volatility of Net Operating Income Bankruptcy Cost SD of first Difference in EBIT / Total Assets Journal of Management and Research Volume 3 References (Burgman, 1996),(Chkir & Cosset, 2001) (Titman & Wessels, 1988) (Jensen, 1986), (Akhtar, 2004) (Jensen, Solberg, & Zorn, 1992) (Mehran, 1992), (Shah & Hijazi, 2004) (Bracdley, Jarrell, & Kim, 1984), (Chaplinsky, 1984), (Lee & Kwok, 1988) (Bradley, Jarrell, & Kim, 1984), (Titman & Wessels, 1988) (Ferri & Jones, 1979),(Scott Jr & Martin, 1975), (Aggarwal, 1990),(Rajan & Zingales, 1995) (Rajan & Zingales, 1995), (Friend & Lang, 1988) (Chittenden, Hall, & Hutchinson, 1996) (Doukas & Pantzalis, 2003), (Rajan & Zingales, 1995), (Shah & Hijazi, 2004) (Burgman, 1996), (David M Reeb et al., 1998) (Lee & Kwok, 1988) (Bracdley, Jarrell, & Kim, 1984), (Lee & Kwok, 1988), (Chaplinsky, 1984) Number 1 2016 Table 1 shows the proxies for dependent variable leverage and independent variables agency cost, free cash flows, growth, age, non debt tax shield, size, collateral value of asset, profitability, business risk, bankruptcy cost and foreign exchange risk. 4. Research Methodology 4.1 Population The initial data is collected from the publication of State Bank of Pakistan titled as “Financial Statement Analysis of the Companies (Non-Financial) listed at Karachi Stock Exchange”. The publication includes only non financial firms. Some data is also collected from companies’ websites and annual reports. 4.2 Sample In this study, random stratified sampling technique is used and data is collected for the period of 2006 to 2013. SBP Publication contains 399 firms of different sectors. Slovin’s sampling technique is used to determine the sample size for this study as used in different previous studies. (Meyer, Mudambi, & Narula, 2011; Onimisi, 2010; Sharif, Naeem, & Khan, 2012; Yasa et al., 2013). It is usually impossible to survey every member in the population because of time or money constraint therefore, Slovin formula is useful to determine the sample size at a given error tolerance. Slovin’s formula is: n = N/(1+Ne2) Where; n = Number of samples in the data N = Total population e = Error tolerance I considered error tolerance 5% Therefore: n = __399_ _ 1+399(.05)2 n = 199.77 = Approximately 200 firms Journal of Management and Research Volume 3 Number 1 2016 Table 2: Industry-wise Sample Size by Using Slovin's Formula Industries TOTAL SLOVIN’S FORMULA SAMPLE Textiles Sugar and Other Food Products Chemicals, chemical products and Pharmaceuticals Electrical Machinery and Other manufacturing Cement and other Mineral Products Motor vehicles, trailers and auto parts Fuel & Energy Information, Communication, transport and other services Refined petroleum products Paper, paperboard and products Total: 155 54 43 (155*200)/400 = 77.5 (54*200)/400 = 27 (43*200)/400 = 21.5 78 27 22 39 28 22 18 22 (39*200)/400 = 19.5 (28*200)/400 = 14 (22*200)/400 = 11 (18*200)/400 = 9 (22*200)/400 = 1 20 14 11 9 11 9 9 399 (9*200)/400 = 4.5 (9*200)/400 = 4.5 (400*200)/400 = 200 5 5 202 From the table 2 it can be analyzed that Pakistan’s non financial sector is a diversified sector with different nature of firms however, textile industry is the biggest industry in Pakistan with highest number of companies and it ranges from spinning and weaving to make up textile items. Sugar and other food is the second largest industry of Pakistan and chemicals and pharmaceutical are the third largest industry. As Pakistan is a developing country, high tech industries are smaller in number like motor vehicles, auto parts, refined petroleum products, paper and board products, etc. Table 3: Firms’ Years* Distribution of companies by Economic Groups Economic Groups Textiles Sugar and Other Food Products Chemicals, chemical products and Pharmaceuticals Electrical Machinery and Other manufacturing Cement and Mineral Products Motor vehicles, trailers and auto parts Fuel & Energy Information, Communication, transport and other services Refined petroleum products Paper, paperboard and products Total: *Firms’ Years can be defined as the number of firms multiplied by the number sample, i.e. 202 firms multiplied by 8years = 1616 firms’ years. Total 624 216 176 160 112 88 72 88 40 40 1616 of years % 38.0 13.4 10.9 10.4 6.9 5.4 4.5 5.5 2.5 2.5 100 in the Table 3 represents the distribution of firm’s years with respect to economic groups. Hence textile is the biggest economic group of Pakistan with total 624 firms’ years with 38% in total. The second largest economic group is sugar and other food Journal of Management and Research Volume 3 Number 1 2016 products which have 216 firms’ years in total. Chemicals and pharmaceuticals and electric machinery and other manufacturing are also big economic groups with 176 and 160 firm’s years respectively. In contrast refined petroleum products and paper, and paper board products are the smallest economic groups with 40 firms’ years each. Table 4: Descriptive Statistics of Leverage and Determinants of Capital Structure LEV AC FCF GRO BC AGE NDT S PRO F CVA BR SIZE N Mean Minimum 0.172 0.050 0.026 0.692 9.321 3.341 0.107 Std. Deviatio n 0.246 0.150 0.344 11.849 31.628 0.556 1.796 1612 1613 1615 1613 1613 1614 1613 Maximu m 0.000 0.000 -0.277 -1.000 0.020 0.000 0.000 1.000 4.915 12.042 440.803 823.977 4.913 57.930 1595 -1.512 71.583 -2802.000 550.000 1613 1612 1595 0.473 -6.476 14.93 3 0.241 214.014 1.705 0.000 -6035.644 0.000 1.973 651.989 20.819 Table 4 shows the descriptive statistics of dependent variable (leverage) and independent variables (determinants of capital structure) of six years from 2006 to 2013. Leverage is 0.172 means on average generally Pakistani firms hold 17.2% debt in their capital structure. Agency cost is 0.050 while free cash flows are 0.026. There is a positive growth of corporations (69.2%) which is a good sign for Pakistan. Profitability is negative (-151% ) which is a worrying sign on other hand. Financial crises, energy crises and other unfavorable factors may be the reasons for negative profitability. Further, average collateral value of asset is 0.473 and business risk of Pakistani firms is -6.476. Average bankruptcy cost is 9.3321. 4.3 Ethical Issues Journal of Management and Research Volume 3 Number 1 2016 As data is the secondary data, publically available on websites, no confidentiality or anonymity issues will arise. 4.4 Data Analysis Initial data is collected and entered into Microsoft excel worksheet. The collected data has been entered accurately and systematically. The data for this research is the panel data means the data is the combination of time series data and cross sectional data therefore it has been organized accordingly and panel was created. In order to obtain the accurate empirical results, this study is using STATA 12 which is a very useful statistical tool for panel data. Different test have been applied in this research like descriptive statistics for the comparison of mean of variables and regression analysis for the relationship of variables. 5. RESULTS AND DISCUSSION Empirical model LEV = α+ β1ACit+ β2BCit +β3NDTSit +β4PROFit +β5SIZit +β6CVAit + β7GROit +β8FCFit+ β9AGEit+ β10BRit +uit Where, LEV = Leverage AC = Agency Cost BC = Bankruptcy Cost NDTS = Non Debt Tax Shield PROF = Profitability SIZ = Size CVA = Collateral value of Assets GRO = Growth FCF = Free Cash Flows AGE = Age BR= Business Risk 6. Results And Finding Journal of Management and Research Volume 3 Number 1 2016 Sample for this study contains data across firms and over time so panel data analysis is appropriate. Panel data analysis has many advantages like it provide a hefty number of data points, increasing the degree of freedom, also decreasing the co-linearity among variables and helps in developing well-organized economic estimate (Hsiao, 1986). Further, panel data has advantage of make out and determine those effects that are simple not deductable in exclusive cross sectional or exclusive time series data (Baltagi, 1995). Moreover, Hsiao (1986) mentioned that panel data allows the application of variable intercepts models that initiate firm/industry type and/or time specific effects into the regression equation that minimize or evade the omitted variable bias. The most popular tools for analysis of panel data are fixed effect model and the random effect model. In this thesis the author is using the following decision making criteria for selection of the model presented by (Dougherty, 2011). According to Figure 1, first of all, there is a need to perform both the fixed effect regression and the random effect regression if the data is selected randomly. Then Durbin-Wu Hausman’s (DWH) specification test is required. If the result of DWH test rejects the null hypothesis, then fixed effect model should be used otherwise the random effects model is required. And then, in case of random effects model, a further test is required called Breusch Pagan Lagrange Multiplier (BPLM) Test to decide between the random effects model and Pooled Ordinary Least Square (OLS) regression. Again, if null hypothesis is rejected in case of BPLM test, the random effect model should be used otherwise pooled OLS regression is required. Journal of Management and Research Volume 3 Number 1 2016 Figure 1 Source: Adapted from (Dougherty, 2011). Decision making criteria for selection of model. Source: Adapted from (Dougherty, 2011). Decision making criteria for selection of model. Table 5.1: Fixed Effect Model Journal of Management and Research Volume 3 Number 1 2016 LEV Coef. Std. Err. t P value AC -0.146 0.078 -1.860 0.063*** FCF 0.007 0.018 0.360 0.718 GRO -0.002 0.001 -3.960 0.000* AGE -0.241 0.039 -6.220 0.000* NDTS 0.000 0.004 -0.020 0.986 SIZE -0.022 0.009 -2.580 0.010* CVA 0.230 0.039 5.910 0.000* PROF 0.000 0.000 0.760 0.447 BR 0.000 0.000 -0.380 0.705 BC 0.000 0.001 0.920 0.356 _cons 1.203 0.164 7.360 0.000 R-square within 0.0878, between = 0.1398, and overall = 0.098 F Statistics = 13.27, and Prob > F = 0.000, Variable is significant at * 1, ** 5, and ***10% level of significance (two-tailed) Table 5.2 : Random Effect Model LEV Coef. Std. Err. z P Value AC -0.1475 0.0618 -2.3800 0.017* FCF 0.0051 0.0186 0.2700 0.7850 GRO -0.0019 0.0005 -3.8800 0.000* AGE -0.0666 0.0161 -4.1400 0.000* NDTS -0.0001 0.0036 -0.0300 0.9780 SIZE -0.0202 0.0056 -3.6100 0.000* CVA 0.3210 0.0301 10.6700 0.000* PROF 0.0000 0.0001 0.3300 0.7400 BR 0.0000 0.0000 -0.5200 0.6000 BC 0.0003 0.0004 0.7500 0.4550 _cons 0.5495 0.0990 5.5500 0.0000 R-square within 0.0711, between = 0.3148, and overall = 0.1920 Wald Chi2 = 210.19, and Prob > Chi2 = 0.000, Variable is significant at * 1, ** 5, and ***10% level of significance (two-tailed) Table 5.3 : Hausman Test Fixed Random Difference AC -0.1458 -0.1475 0.0016 FCF 0.0066 0.0051 0.0016 GRO -0.0022 -0.0019 -0.0003 AGE -0.2414 -0.0666 -0.1748 NDTS -0.0001 -0.0001 0.0000 SIZE -0.0221 -0.0202 -0.0019 CVA 0.2298 0.3210 -0.0912 PROF 0.0001 0.0000 0.0000 BR 0.0000 0.0000 0.0000 BC 0.0005 0.0003 0.0002 Chi2 = 50.16, Prob > chi2 = 0.000 As mentioned earlier, data selected randomly for this study, there is a need to perform both the fixed effect regression and the random effect regression according to Journal of Management and Research Volume 3 Number 1 2016 Figure 1. Therefore, first of all these tests are applied to the sample of firms in Pakistan. Both models are overall statistically good fit model as F test is significant in fixed effect model and Chi2 is significant in random effect model in table 5.1 and 5.2. However, R 2 (within) is higher in fixed effect model as compare to random effect model (0.0878 vs 0.0711) and R2 between and overall is higher in random effect model as compare to fixed effect model (0.1398 and 0.098 vs 0.3148 and 0.1920). In table 5.3, DWH test reject the null hypothesis and shows significance at 1% level which means that this study entail the fixed effect model and there is no need to further BPLM test and OLS test. According to table 5.1, agency cost is significantly but negatively related to the leverage (p-value = 0.063). Bankruptcy cost is not a significant determinant of capital structure for Pakistani firms (p-value = 0.356). Non debt tax shield is not a significant determinant of capital structure (p-value = 0.968). DCs (p-value = 0.453) or MNCs (pvalue = 0.591). Profitability is not significant (p-value = 0.447) which is against Pecking Order Theory of Myers (1977). Size is a significant determinant of capital structure for all Pakistani firms (p-value = 0.010). Collateral value of assets is a significant determinant of capital structure (p-value = 0.000) and shows a positive relationship (coefficient = 0.023) with leverage indicating that if collateral value of asset increases, the leverage of the company also increases. Growth is a significant determinant of capital structure for the sample of Pakistani firms (p-value = 0.000) at the significance level of 1%. The results show a negative relationship with the leverage for firms (coefficient = -0.002). Age is significant determinant of capital structure for firms (p-value = 0.0000) and shows a negative relationship with the leverage. Business risk is not significant (p-value = 0.705). Free cash flows is a significant determinant of capital structure for all firms (p-value = 0.0000) and for DCs (p-value = 0.0010). Free cash flows shows no relationship with leverage for (p-value = 0.705). Table 6: Pearson Coefficient Correlation LEV LEV AC FCF GRO AGE NDTS SIZE 1 Journal of Management and Research Volume 3 Number 1 2016 CVA PROF BR BC AC -0.1488 1 FCF -0.0079 -0.0089 1 GRO -0.0236 0.1248 -0.0024 1 AGE -0.127 -0.076 0.0054 -0.0078 1 NDTS 0.011 -0.0021 0.7021 0.0194 -0.0018 1 SIZE -0.1965 0.0365 -0.0183 -0.2376 0.0603 -0.0528 1 CVA -0.0419 0.004 0.002 0.1836 -0.0054 0.0001 0.1809 1 PROF 0.4075 -0.0936 -0.0386 0.0606 -0.1035 -0.0108 -0.2405 0.0273 1 BR -0.003 0.0093 -0.0018 0.0036 -0.0196 0.0016 -0.048 0.104 0.0107 1 BC 0.0065 0.6143 0.0456 0.3642 -0.0251 0.0942 -0.4449 -0.1493 -0.0048 -0.0087 6.1 1 Pearson coefficient correlation Correlations among variables can cause multicolinearity which may create problems in regression analysis. Table 6 shows correlations above 0.6 which explains that there is no multicolinearity among variables in however a modest correlation between free cash flows and non debt tax shield (0.702). Therefore, a multicolinearity test is further required to check any dependence among these variables in case of MNCs. Table 7: Variance Inflation Factor Variable 6.2 VIF BC AC NDTS FCF SIZE GRO PROF CVA AGE BR 2.88 2.12 2.00 1.98 1.61 1.28 1.14 1.10 1.03 1.00 Mean VIF 1.61 1/VIF 0.34683 0.47216 0.50092 0.50571 0.61978 0.78052 0.87522 0.91215 0.97221 0.99622 Multicollinerity test When correlation among variables is high i.e. more than 0.6, either positive or negative, then the problem of multicolinearity may arise. To check whether Journal of Management and Research Volume 3 Number 1 2016 multicolinearity exists among variables, a variance inflation factor (VIF) test is applied and hence the results show that (Table 7) that there is no multicolinearity exists among variables. 7. Conclusion This study used 202 Pakistani companies and analyzed eight years data for the annual periods of 2006 to 2013 and investigated the leverage composition of firms and the determinants of capital structure namely agency cost, bankruptcy cost, profitability, age, growth, collateral value of assets, non debt tax shield, free cash flows, size, business risk and foreign exchange risk. Fixed effect model was used to regress the variables. This study found that agency cost, growth, age, and size are significantly and negatively associated with the capital structure of Pakistan firms, however, collateral value of asset is significantly but positively associated with the capital structure of the firm. On the other hand, free cash flows, non debt tax shield, profitability, business risk and bankruptcy cost are not significantly associated with capital structure of the firms and are against the signaling theory and peaking order theory. From this study, one can conclude that firms in Pakistan are using collateral values for getting more leverage out of their assets. Further, as agency cost, growth, age and size of the firm increases, shareholders prefer to invest rather than taking external debt. Therefore, this study demonstrates imperative policy implications for managers and investors of the firms. This fact must be taken into account while making the finance mix decisions for securing the benefits of stakeholders. Further research is required on several other factors that affect the capital structure of the companies like diversification, i.e. product diversification and geographical diversification. Both kinds of diversification not only affect capital structure of the companies but also have an impact on the profitability of the firms. Political risk is also a very important determinant of capital structure. Further, human capital is also a very vital determinant of capital structure in modern world. Very less work has been done on human capital. Firm’s specific factors may also affect the leverage of the firm therefore Journal of Management and Research Volume 3 Number 1 2016 one can explore those firms’ specific factors. Impact of inflation on leverage can also be assessed. References Afza, T., & Hussain, A. (2011). Determinants of Capital Structure across selected Manufacturing sectors of Pakistan. International Journal of Humanities and Social Science, 1(12), 254-262. Journal of Management and Research Volume 3 Number 1 2016 Aggarwal, R. (1990). Capital structure differences among large Asian companies. ASEAN Economic Bulletin, 39-53. Ahmad, Y., & Zaman, G. (2013). Determinants of Capital Structure: A case for the Pakistani Textile Composite Sector. Abasyn Journal of Social Sciences, 6(1), 9110. Ahmed Sheikh, N., & Wang, Z. (2011). Determinants of capital structure: An empirical study of firms in manufacturing industry of Pakistan. Managerial Finance, 37(2), 117-133. Akhtar, S. (2004). ANU-Digital Collections: Australian multinational and domestic corporations capital structure determinants. Made available in DSpace on 201101-05. Akhtar, S., & Oliver, B. (2009). Determinants of Capital Structure for Japanese Multinational and Domestic Corporations*. International review of finance, 9(1‐ 2), 1-26. Al Amri, M., & Al Ani, M. (2015). The determinants of capital structure: an empirical study of Omani listed industrial companies. Verslas: teorija ir praktika(2), 159167. ALIU, N. O. (2010). Effect of capital structure on the performance of quoted manufacturing firms in nigeria (Doctoral dissertation, AHMADU BELLO UNIVERSITY, ZARIA). Baltacı, N., & Ayaydın, H. (2014). Firm, Country and Macroeconomic Determinants of Capital Structure: Evidence from Turkish Banking Sector. Burgman, T. A. (1996). An Empirical Examination of Multinational Corporate Capital Structure. Journal of International Business Studies, 27(3), 553-570. Chittenden, F., Hall, G., & Hutchinson, P. (1996). Small firm growth, access to capital markets and financial structure: Review of issues and an empirical investigation. Small Business Economics, 8(1), 59-67. Dougherty, C. (2011). Introduction to econometrics: Oxford University Press. Eldomiaty, T. I. (2008). Determinants of corporate capital structure: evidence from an emerging economy. International Journal of Commerce and Management, 17(1/2), 25-43. Fama, E. F. (1980). Agency Problems and the Theory of the Firm. The Journal of Political Economy, 288-307. Journal of Management and Research Volume 3 Number 1 2016 Ferri, M. G., & Jones, W. H. (1979). Determinants of financial structure: a new methodological approach. The Journal of Finance, 34(3), 631-644. Forte, D., Barros, L. A., & Nakamura, W. T. (2013). Determinants of the capital structure of small and medium sized Brazilian enterprises. BAR-Brazilian Administration Review, 10(3), 347-369. Friend, I., & Lang, L. H. (1988). An Empirical Test of the Impact of Managerial Self‐ interest on Corporate Capital Structure. The Journal of Finance, 43(2), 271-281. Graham Jr, J. E. (1988). DOMESTIC AND STRUCTURE: THEORY AND EVIDENCE. MULTINATIONAL CAPITAL Harris, M., & Raviv, A. (1991). The theory of capital structure. The Journal of Finance, 46(1), 297-355. Hijazi, S. T., & Tariq, Y. B. (2006). Determinants of capital structure: A case for Pakistani cement industry. Lahore Journal of Economics, 11(1), 63-80. Hsiao, C. (1986). Analysis of Panel DataCambridge University Press. Cambridge, UK. Huang, G. (2006). The determinants of capital structure: Evidence from China. China Economic Review, 17(1), 14-36. Hussain, I. (2006). Pakistan's investment scenario and capital markets. Jensen, M. (1986). Agency cost of free cash flow, corporate finance, and takeovers. Corporate Finance, and Takeovers. American Economic Review, 76(2). Khan, I., Jan, S. U., & Khan, M. (2015). Determinant of Capital Structure: An Empirical Study of Cement Sector of Pakistan. Asian Journal of Management Sciences & Education Vol, 4, 3. Kouki, M., & Said, H. B. (2011). Capital structure determinants: new evidence from French panel data. International journal of business and management, 7(1), p214. Lee, K. C., & Kwok, C. C. Y. (1988). Multinational corporations vs. domestic corporations: International environmental factors and determinants of capital structure. Journal of International Business Studies, 195-217. Mac an Bhaird, C., & Lucey, B. (2010). Determinants of capital structure in Irish SMEs. Small business economics, 35(3), 357-375. Masnoon, M., & Saeed, A. (2014). Capital Structure Determinants of KSE Listed Automobile Companies. European Scientific Journal, 10(13). Journal of Management and Research Volume 3 Number 1 2016 Meyer, K. E., Mudambi, R., & Narula, R. (2011). Multinational enterprises and local contexts: the opportunities and challenges of multiple embeddedness. Journal of Management Studies, 48(2), 235-252. Modigliani, F., & Miller, M. H. (1958). The cost of capital, corporation finance and the theory of investment. The American economic review, 48(3), 261-297. Myers, S. C. (1977). Determinants of corporate borrowing. Journal of financial economics, 5(2), 147-175. Myers, S. C. (1984). The capital structure puzzle. The Journal of Finance, 39(3), 574592. Qadri, S. Q. S. U. (2015). An Empirically Study of Determinants of Capital Structure of Non Financial Listed Companies in KSE. International Journal of African and Asian Studies, 12, 50-55. Rajan, R. G., & Zingales, L. (1995). What do we know about capital structure? Some evidence from international data. The Journal of Finance, 50(5), 1421-1460. Scott Jr, D. F., & Martin, J. D. (1975). Industry influence on financial structure. Financial Management, 67-73. Shah, A., & Hijazi, S. (2004). The determinants of capital structure of stock exchangelisted non-financial firms in Pakistan. The Pakistan Development Review, 43(4), 605-618. Shah, A., & Khan, S. (2007). Determinants of capital structure: Evidence from Pakistani panel data. International review of business research papers, 3(4), 265-282. Sharif, B., Naeem, M. A., & Khan, A. J. (2012). Firm’s characteristics and capital structure: A panel data analysis of Pakistan’s insurance sector. African Journal of Business Management, 6(14), 4939-4947. Titman, S. (1984). The effect of capital structure on a firm's liquidation decision. Journal of financial economics, 13(1), 137-151. Yasa, N. N. K., Sukaatmadja, P. G., Jawas, A., Budhi, M. K. S., Marhaeni, A., Ndibe, L. O., Rathnamani, V. (2013). SME performance improvement and its effect on the poverty reduction in Bali. Journal of Management and Research Volume 3 Number 1 2016