Debt capital and financial performance: A study of South African companies

Debt capital and financial performance: A study of South African companies K.M.R.Magoro 1 and D.K.Y. Abeywardhana 2 1 kmr2175@gmail.com, 2 Department of Accountancy, University of Kelaniya, Sri Lanka. dilyapa@kln.ac.lk Corresponding author: D.K.Y. Abeywardhana, Tel:+94714601643 Abstract Purpose The purpose of this study is to examine how debt capitals of the listed companies operating in the wholesale and retail sector of South Africa affect their financial performance. Design/methodology/approach The study used a panel data sample of 25 South African wholesale and retail sector companies to examine the impact of debt capital on the financial performance of companies over the 2011-2015 period. Fixed-effects (within) regression modelwas used on the accounting-based-measures of profitability and financial performance. Findings The study confirms that debt capital, in terms of short-term debt and long-term debt, has a negative impact on the financial performance of wholesale and retail sector companies of South Africa. Practical Implications The findings of this research will help South African wholesalers and retailers to understand the impact of debt capital on company performances. This study will help them make decisions that will ensure profit maximization and reduction of costs associated with debt, and ultimately, maximization of shareholders wealth. Originality/Value This study gives special focus to the wholesale and retail sector as it seeks to pioneer the addressing of root causes and reasons of research contradictions in this study area. Keywords: Debt capital; Financial Performance; South Africa; Capital Structure. 71

1. Introduction The interplay between capital structure and firm performance is a crucial and frequently discussed issue, and has been the subject of many studies. A great number of researchers have come up with solutions to this particular problem. It was in an attempt to find the best solution to firms that capital structure theories were introduced. Modigliani and Miller (1958) were the pioneers of this study area, and developed a proposition highlighting the irrelevance of capital structure, thereby hypothesized that in perfect markets, it matters not what capital structure a company uses to finance its operations as a firm s financing choice does not affect its capital cost, value or real operations, including performance. The scholars theorized that the market value of a firm is determined by its earning power and by the risk of its underlying assets, and that its value is independent of the way it chooses to finance its investments or distribute dividends. For this view to hold, Modigliani and Miller (1958) based their proposition on a number of assumptions; for example, that no taxes or transaction costs exist, that both companies and investors have equivalence in borrowing and lending money, that companies and investors have the same market information (symmetry) and that agency costs exist. Although Modigliani and Miller (1963) suggested that firms can gain the benefits of tax-deductible interest payments by increasing the amount of debt in their capital structure, this view has been questioned due to its assumption of market perfection and its limited applicability to small firms (Grabowski and Mueller, 1972; Chaganti et al., 1995). Different researchers followed with theories that contributed to the literature. Now there are well-known capital structure theories, which include Pecking Order Theory, Trade-Off, Agency Costs, Corporate taxes, financial distress, and Signalling. The capital structure literature is varied with mixed results in different times and locations. However, the few studies (Gleason et al., 2000; Yazdanfar and Öhman, 2015) which have investigated multiple sectors; including wholesale and retail, have found common results. The researchers found that, although the results of the impact of capital structure on firm performance could be mixed in other sectors, the relationship is however negative in the wholesale and retail sector. While a large body of supporting empirical evidence is available for most of the developed countries in America and Europe (Claessens and Laeven, 2006), very little research has been conducted in developing countries like South Africa, particularly in the currently researched wholesale and retail sector. Most of the previous researches conducted under the capital structure (Goddard et al., 2005; Abor, 2005; Abu-Tapanjeh, 2006; Nimalathasan and Brabete, 2010; Sheikh and Wang, 2011; Salim and Yadav, 2012), specifically those that examined the impact of debt and equity mix on firm performance did not incorporate the wholesale and retail sector, with the exception of Gleason et al. (2000) and Yazdanfar and Öhman (2015), who, to the knowledge of the researcher, are the among the few scholars that studied and examined this impact in the wholesale and retail sector among other sectors. The current study addresses the debt-performance relationship from the pecking order theory perspective among the South African wholesalers and retailers which are the fourth largest contributors to GDP growth of South Africa (Statistics South Africa, 2013).This study uses the Fixed-effects (within) regression model on the accounting-based-measures of debt capital and financial performance to analyse the impact thereof. The results indicate that debt capital has a significant impact on the financial performance of the firms operating in the South African wholesale and retail sector in terms of profitability. Companies which have lower debt ratios are performing well and are able to achieve higher profits. 72

The remainder of the paper is structured as follows: Section 2 discusses the literature review of capital structure and financial performance. Section 3 presents the research design, hypotheses and methodology employed to address the research questions of the study. It also explains data collection and sampling, as well as the measurement of research variables. Section 4 presents data analyses and discusses the results and section 5concludes the paper. 2. Literature Review 2.1 Theoretical Framework Capital structure theory is based on the seminal work of Modigliani and Miller (1958). The scholars assumed a perfect market and theorized that the market value of a firm is determined by its earning power and by the risk of its underlying assets, and that its value is independent of the way it chooses to finance its investments or distribute dividends. The theorem was also based on the assumptions that no taxes or transaction costs exist, that both companies and investors have equivalence in borrowing and lending money, that companies and investors have the same market information (symmetry) and that agency costs exist. However, in the real world, there are no perfect markets. Taxes and transaction costs do exist, there are differences in borrowing and lending, and there s information asymmetry in all markets (imperfect) which are associated with agency conflicts and moral hazards (Jensen and Meckling, 1976; Greenwald et al, 1984; Myers and Majluf, 1984; Stiglitz, 1988; Greenwald and Stiglitz, 1993). In papers following their 1958 seminal work, Modigliani and Miller included both the effects of taxes and bankruptcy costs. Although Modigliani and Miller (1963) suggested that firms can gain the benefits of tax-deductible interest payments by increasing the amount of debt in their capital structure, this view has been questioned due to its assumption of market perfection and its limited applicability to small firms (Grabowski and Mueller, 1972; Chaganti et al., 1995). In an attempt to criticize the irrelevance theory of Modigliani and Miller (1958) which assumed the perfect market and information symmetry among other things, Donaldson (1961) suggested and introduced the Pecking order theory in an interview survey of 25 large firms based in the United States (US). Donaldson (1961) made a conclusion based on the results of the survey that management of firms prefer to use internal funds when available, and that if internal sources of funds are available, management would opt against external funding. The pecking order theory was modified by Myers and Majluf in 1984. Myers and Majluf (1984) added that managers and investors depend upon information as their base of making decisions regarding debt or equity financing. Most companies would be reluctant to issue new equity if they feel that it is undervalued by the market. The study showed that asymmetric information favoured the issue of debt as it indicates the confidence of the companies boards that an investment is profitable and the stock price is undervalued. Otherwise, equity issue would be preferred only when the stock price is over-valued. Based on this argument, if managers tend to issue undervalued equity (low priced equity); the wealth will be transferred to the investors against the shareholders benefits and wealth. In this situation, internal funds and debt will be preferred to equity. Myers (1984) referred to this as the pecking order theory of financing. This states that firms prefer to finance new investment first internally with retained earnings, second with debt, and last by issuing new equity (Al-Tally, 2014). 73

Examinations of the pecking order theory have not been able to show that it is of first-order importance in determining a firm's capital structure. However, several researchers (Fama and French, 1998; Shyam-Sunder and Myers, 1999) found that certain features of the financial data are better explained by the pecking order theory than the trade-off one. On the contrary, Frank and Goyal (2007) showed that Pecking Order theory fails where it should hold, namely for small firms where information asymmetry is presumably an important problem. The drawbacks of this theory prompted the need and desire for further research and the development of more theories. 2.1 Review on the Location of Trade While most of the previous studies ignore the wholesale and retail sector in favour of manufacturing and other sectors (Goddard et al., 2005; Abor, 2005; Abu-Tapanjeh, 2006; Nimalathasan and Brabete, 2010; Sheikh and Wang, 2011; Salim and Yadav, 2012), the current study highlights the significance of the less researched sector and the importance of location of trade for the companies being researched with special focus given to Africa. Abor (2005) adapted regression analysis to test the relationship between capital structure and performance in terms of profitability among 22 listed firms in Ghana over the 1998-2002 period. The study of Abor (2005) revealed a significantly positive relation between the ratio of short-term debt to total assets and Return on Equity (ROE). On the other hand, Abor (2005) indicated that there was a negative relationship between the ratio of long-term debt to total assets and ROE. Going further, Abor (2007) used a generalized least squares regression to study a sample of 160 Ghanaian and 200 South African Small and Medium Sized Enterprises (SMEs) over the 1998-2003 period. The results of Abor (2007) indicate that capital structure influences financial performance of Ghanaian and South African firms, although not exclusively. On the other hand, Abor (2007) found that by and large, the results indicate that capital structure, especially long-term and total debt ratios, negatively affect performance of SMEs. The two studies show that there was a definite negative relationship between long-term debt ratios and the financial performance of the firm; while the short-term debt ratios impact the firm performance positively. Ebaid (2009) used multiple regression analysis to estimate the relationship between the leverage level and firm s performance of all publicly traded firms on Egyptian stock exchange during the period of 1997-2005. The results revealed that capital structure choice decision, in general terms, has a weak-to-no impact on firm s performance. David &Olorunfemi (2010) studied the impact of capital structure on corporate performance in the Nigerian Petroleum Industry and employed panel data analysis by using Fixed-effect estimation, Random-effect estimation and Maximum likelihood estimation. Firm performance was measured by both Earnings Per Share (EPS) and Dividend Per Share (DPS) The study found leverage to impact firm performance positively. Booth et al. (2001) found that debt ratios and profitability are negatively correlated for a set of ten developing countries including Brazil, Mexico, South Korea, Zimbabwe and Malaysia among others. The researchers emphasize this negative relationship by stating that a consistent result in both the country and pooled data results is that the more profitable the firm, the lower the debt ratio, regardless of how the debt ratio is defined. 74

The review of the literature based on African studies reveals that even in the closest of regions, the findings are still not in agreement. The studies were conducted in different sectors, used different statistical methods and proxies for firm performance, and reported varying results. This can partly lead to the conclusion that it is important to study each sector in each region (country) given the fact that varying socioeconomic factors may play a role in the relationship between debt capital and financial performance. This is why it is relevant to study this relationship in the context of South Africa s wholesale and retail sector. 3. Research Methodology 3.1 Measurement of Variables The empirical literature shows that there are a number of different measures of firm performance which can be used to test its relationship with capital structure. Firm performance as a dependent variable can be measured and expressed in various ways. Several researchers (Krishnan and Moyer, 1997; Abor, 2007; Ebaid, 2009; Alanazi et al., 2011; Salim and Yadav, 2012; Dawar, 2014; Abeywardhana, 2015) have used multiple variables including accounting based ratios from balance sheet and income statements such as ROA, ROCE, ROE, ROS, gross profit margin and EPS; while a few others (Goddard et al., 2005; Yazdanfar and Öhman, 2015) have used only one variable ROA. Stock market returns and their volatility have also been used as performance measure by Cole and Mehran (1998) as well as Welch (2004). In the current study we use a similar approach to that of Alanazi et al. (2011), which used ROA and ROS as proxies for firm performance and they are defined as follows: 1. Return on Assets (ROA) - Net Profit/Total Assets 2. Return on Sales (ROS) - Net Profit/Sales For the purpose of the current study, as was the case in the study of Dawar (2014), the researcher used the following debt capital variables which were divided into two categories in order to identify the relationship between debt ratios and profitability in detail as well as to assess the impact that debt financing has on firm performance,: 1. Short-term debt - Debt repayable within one year/total Assets OR (short-term debt (STD) to total assets) 2. Long-term debt - Debt repayable beyond one year/total Assets OR (long-term debt (LTD) to total assets) Gleason et al. (2000) as well as Yazdanfar and Öhman (2015) found that agency conflicts may be primarily responsible for overleveraging of retailers, resulting in a negative relationship between capital structure and performance. On that basis, the following hypothesis is developed: H1: Debt capital has a negative impact on the financial performance of the wholesale and retail sector companies in South Africa. H1.1: Short-term debt has a negative impact on the financial performance of the firm. H1.2: Long-term debt has a negative impact on the financial performance of the firm. 75

Previous studies (Penrose, 1959; Stinchcombe, 1965; Shepherd, 1989;Majumdar and Chhibber, 1999; Cabral and Mata, 2003; Berger and Di Patti, 2006; Jermias, 2008; Ebaid, 2009; Sheikh and Wang, 2011; Alanazi et al., 2011; Yazdanfar and Öhman, 2015)have indicated that the size and age of the firm can have a significant influence on its financial performance. Accordingly, size and age are included in this study s model as controlled variables. Based on that, the second and third hypotheses are developed as follows: H2: Firm size has a positive impact on the financial performance of the wholesale and retail sector companies in South Africa. H3: Firm age has a positive impact on the financial performance of the wholesale and retail sector companies in South Africa. 3.2 Sample Data were collected from the listed companies on the Johannesburg Stock Exchange for 2011-2015 periods. For a homogeneous selection and accurate results of the analysis, the researcher excluded companies formed after the year 2010, i.e., the year before beginning of the year under study. The analysis is based on 25 publicly listed firms in the wholesale and retail sector of South Africa. As is the case in Hu and Ansell s (2007) research, Only publicly listed companies were chosen. Given that listed companies had to abide by regulations in the financial market, their financial information tended to be more open and transparent than that of private companies. 3.3 Model Specification Fixed-effects (within) regression model was considered for the analysis in this research as it takes into effect systematic differences between the firms as compared to ordinary least squares regression method which assumes that model parameters remain constant across all firms.the relationship between debt capital and firm performance as well as the impact thereof was tested by the following regression models:, = +, +, +, +, +,, = +, +, +, +, +, Where:, = net profit to total assets for firm i in year t;, = net profit to total sales for firm i in year t;, = short term debt to total assets for firm i in year t;, = long term debt to total assets for firm i in year t;, = logarithm of sales for firm i in year t;, = age of firm i at time t, measured as the natural logarithm of the number of years since firm inception, as of the year of data collection; and = error term., 76

4. Empirical results 4.1 Descriptive Statistics Table 4.1 below shows the summary of the descriptive statistics of both the dependent and independent variables. Descriptive statistics show the mean, standard deviation, minimum, maximum and the number of observations. The most important measure that shows the balance point and is the exertion center of distribution is arithmetic mean (Azar et al, 2006; Sadeghian et al., 2012). The sampled companies are characterized by an average age of just above 63 years. The mean age show that these companies have been in business for long and are generally well established. The mean value for company size is 9.8. The mean values of both company size and age indicate that the sampled companies are fairly large and have reached a significant scale of operations. The ROA and ROS reveal that companies operating in this sector achieve modest profitability, with mean profitability of 8.85% and 5.52% for ROA and ROS respectively. The descriptive statistics indicate that profitability of the sampled firms (as measured by ROA and ROS) as well as long-term debt are highly volatile, as the standard deviations of these variables are greater than their mean values. However, levels of short-term debt, size and age indicate that they are less volatile, and thereby showing the quality of being similar or comparable in nature (homogeneousness) of the sampled companies. As shown in the descriptive statistics table below, 37% of total assets of the sampled companies were on average financed using short-term debt, while long-term debt was used as a means offinancing for approximately 15% of the total assets. This means that just about 52% of the total assets in the sampled companies are financed by debt, and thereby revealing that companies in this sector generally use internal financial sources (i.e. equity capital and retained earnings) to finance around 48% of the total assets. The proportion between debt use and equity use is not much (approximately 4%), and that indicates the balance of financing options maintained when using these two components of the capital structure by companies in the wholesale and retail sector, with slight preference to debt over equity. Companies in this sector seem to prefer short-term debt as their external source of financing. They use long-term debt less often or minimally. The reason for this was better explained by Yazdanfar and Öhman (2015) when the researchers indicated that Financial institutions such as banks usually agree to issue short-term debt, because it can be cancelled on short notice. Therefore, firms can use short-term debt as an additional funding source to increase internal sources of capital. The mean values of all the variables are significant at 5 percent level. 77

Table 4.1: Descriptive statistics for the dependent and independent variables Variable Mean Std. Dev. Min Max Observations ROA Overall 0.0885 0.1056-0.2799 0.3312 N = 125 Between 0.0986-0.1644 0.3063 n = 25 Within 0.0416-0.0270 0.2823 T = 5 ROS Overall 0.0552 0.0758-0.1834 0.2278 N = 125 Between 0.0728-0.0994 0.2152 n = 25 T = 5 Within 0.0246-0.0334 0.1770 STD Overall 0.3721 0.1943 0.0985 0.8699 N = 125 Between 0.1841 0.1281 0.7163 n = 25 T = 5 Within 0.0704 0.1676 0.6336 LTD Overall 0.1478 0.1631 0.0109 0.9469 N = 125 Between 0.1591 0.0145 0.8329 n = 25 T = 5 Within 0.0459 0.0064 0.3927 Size Overall 9.8033 0.8625 7.7526 11.0557 N = 125 Between 0.8741 7.8606 10.9615 n = 25 T = 5 Within 0.0668 9.6394 10.0099 Age Overall 63.16 35.6808 8 130 N = 125 Between 36.2418 10 128 n = 25 T = 5 Within 1.4199 61.16 65.16 Notes: ROA = Return on Assets; ROS = Return on Sales; STD = Short-term debt; LTD = Long-term debt; Size = natural logarithm of total sales; Age = Number of years since establishment. 4.2 Correlation Analysis Correlation analysis was conducted in order to identify the relationships between the variables included in the main models, and to examine potential multicollinearity among the independent variables. The results of correlation analysis between the dependent variables, which are represented in terms of ROA and ROS, as well as each independent and control variable are reported in Table 4.2 below. As shown on the table below, firm performance defined in terms of ROA is negatively and significantly correlated with all the independent variables (Short-term debt and Long-term debt), and positively correlated with the control variables (Size and Age). Although ROA is positively related to the control variables, the correlation is only significant with Size and not with Age at 5 percent level. Age shows a significance level of 21.16%, as far as correlation with ROA is concerned. Firm performance in terms of ROS shows almost similar results, with ROS being negatively and significantly correlated with both Short-term debt and Long-term debt (independent variables). The relationship with the control variables is not significant at 5 percent level, with Size and Age showing significance levels of 13.22% and 8.52% respectively. The insignificant correlation results between firm performance (ROA and ROS) and the control variables indicate that age of the sampled companies matter less when it comes to the enhancement or depletion of firm performance, while size has little significance. 78

The relationship between short-term debt and long-term debt is not significant at 5 percent level. Short-term debt is positively and significantly related to size, while it is negatively and significantly correlated with the age of the firm. This indicates that as firms become larger or bigger (in terms of size), they tend to make use of short-term debt; and that the older the firms become, the less they use short-term debt. Long-term debt is positively and significantly correlated with the age of the firm. This is consistent with the short-term debt correlation results, as it indicates that older firms tend to use long-term debt over short-term debt, the older they get. The relationship between long-term debt and the size of the firm is not statistically significant at 5 percent level. Size show a significance level of 11.22% as far as association with long-term debt is concerned. The results also confirm that all the coefficients between variables are fairly low, so there is no indication of multicollinearity among the variables included in the model. Table 4.2: Correlation of analyses of the variables included in the study ROA ROS STD LTD Size Age ROA 1.000 0.862** -0.238** -0.496** 0.271** 0.113 Significance 0.0000 0.0075 0.0000 0.0023 0.212 ROS 0.862** 1.000-0.452** -0.428** 0.135 0.155 Significance 0.0000 0.0000 0.0000 0.1322 0.085 Short-term debt -0.238** -0.452** 1.000-0.162 0.282** -0.396** Significance 0.0075 0.0000 0.0714 0.0015 0.0000 Long-term debt -0.496** -0.428** -0.162 1.000 0.143 0.237** Significance 0.0000 0.0000 0.0714 0.1122 0.0078 Size 0.271** 0.135 0.282** 0.143 1.000 0.200** Significance 0.0023 0.1322 0.0015 0.1122 0.0255 Age 0.113 0.154-0.3955** 0.237** 0.200** 1.000 Significance 0.2116 0.0852 0.0000 0.0078 0.0255 Notes: ** Correlation is significant at 5 percent level. ROA = Return on Assets; ROS = Return on Sales; STD = Short-term debt; LTD = Long-term debt; Size = natural logarithm of total sales; Age = Number of years since establishment. 4.3 Results of the Fixed-effects (within) regression model Table 4.3 and 4.4 below present the regression results using the Fixed-effects (within) regression model to test the impact of debt capital on the financial performance of companies measured by ROA and ROS. As indicated by the results of the complete model and consistent with H1, debt capital - i.e. short-term debt and long-term debt has a negative impact on firm performance in the case of both ROA and ROS. In terms of ROA, the slope coefficients of long-term debt and short-term debt are both negative and statistically significant at 5 per cent level. The slope coefficient of the long-term debt ratio variable (β = -0.3663613; P> t = 0.000) represents the highest debt ratio, while the slope coefficient of the short-term debt ratio variable (β = -0.2283428; P> t = 0.000) represent the lowest. 79

In terms of ROS, the results are somewhat similar with the slope coefficients of both long-term debt and shortterm debt being negative and statistically significant at 5 per cent level. However, the slope coefficients of both debt ratio variables are comparatively lower than in ROA. In the case of ROS, the slope coefficient of the long-term debt ratio variable (β = -0.1719025; P> t = 0.002) still represents the highest debt ratio, while the slope coefficient of the short-term debt ratio variable (β = -0.1188747; P> t = 0.001) represent the lowest. These results suggest that an increase in the amount of both or either short-term debt and long-term debt is associated with a decrease in the financial performance of the firm. Size of the firm is proved to have a significant positive impact on firm performance as measured by both ROA and ROS. This is consistent with H2, indicating that larger companies in the wholesale and retail sector operating in the South African market are, on average, more likely to be profitable or perform well financially. This could also mean that larger companies in South Africa are able to achieve the economies of scale and are also able to exercise considerable influence in product and factor markets. Contrary to H3 put forward, the firm s age was found to have significantly negative associations with both ROA and ROS, and thus implying that younger firms tend to perform well financially than their older counterparts, as they may be flexible in adjusting or adapting to rapid changes that have been one of characteristics of this particular industry in recent years. All the associations between the variables are confirmed by the t-values which tested the hypotheses, and the results show that all the independent and control variables affect both ROA and ROS negatively, with the exception of Size (control variable) which was proved to impact firm performance positively. The Fixedeffects (within) regression model indicate that independent variables explain approximately 40.28 percent of the change in ROA, and 30 percent of the change in ROS, as shown by R-squared in their respective regression results. This suggests that other firm-level, industry and/or macroeconomic variables obviously affect the financial performance of the firm. Approximately 99% of the variance in both regression results is due to differences across the panels, as depicted by rho which is known as the intraclass correlation or the percentage of the variation that s explained by individual specific effects. The high percentage of rho in the two regression results (in terms of ROA and ROS) is good as it means that it s not just idiosyncratic. Overall, the empirical results in Table 4.3 and 4.4 below indicate that after controlling for factors such as firm size and firm age, debt capital (debt ratios) has a negative impact on the financial performance of the firms operating in the South African wholesale and retail sector in terms of profitability (ROA and ROS). Companies which have lower debt ratios appear to be performing well and are able to achieve higher profits. The evidence provided by the current study is in accordance with the postulates of the pecking order theory. This suggests that managers of profitable companies operating in the South African wholesale and retail sector can use equity capital and retained earnings efficiently, thereby minimizing conflicts of agency or agency costs and remaining independent of external financiers. The empirical findings of the current study support those of previous studies from other countries and different sectors, e.g. Majumdar and Chhibber (1999), Gleason et al. (2000); Goddard et al. (2005), Abor (2007), Sheikh and Wang (2011); Salim and Yadav (2012); Dawar (2014); and Yazdanfar and Öhman (2015). 80

Table 4.3: Regression results of debt capital and firm performance measured by ROA ROA Coefficient Std. Err. t-statistics Significance Short-term debt -0.2283 0.0516-4.42** 0.000 Long-term debt -0.3664 0.0821-4.46** 0.000 Size 0.2270 0.0843 2.69** 0.008 Age -0.0158 0.0039-4.03** 0.000 Constant -1.0009 0.6552-1.53** 0.130 R² within 0.4028 R² between 0.0021 R² overall 0.0029 Sigma_u 0.5641 Sigma_e 0.0365 rho 0.9958 Prob> F 0.0000 Notes: ** Significant at 5 percent level. ROA = Return on Assets; Size = natural logarithm of total sales; Age = Number of years since establishment. Table 4.4: Regression results of debt capital and firm performance measured by ROS ROS Coefficient Std. Err. t-statistics Significance Short-term debt -0.1189 0.0331-3.59** 0.001 Long-term debt -0.1719 0.0526-3.27** 0.002 Size 0.1565 0.0540 2.90** 0.005 Age -0.0089 0.0025-3.55** 0.001 Constant -0.8461 0.4199-2.01** 0.047 R² within 0.3007 R² between 0.0013 R² overall 0.0007 Sigma_u 0.3305 Sigma_e 0.0234 rho 0.9950 Prob> F 0.0000 Notes: ** Significant at 5 percent level. ROA = Return on Assets; Size = natural logarithm of total sales; Age = Number of years since establishment. 81

5. Conclusion, Recommendations and Limitations Contrary to the agency-cost theory which suggests that high leverage levels can help firms increase market value and performance, the overall results of this study indicate that debt capital, in the form of both long-term debt and short-term debt affects the financial performance of the South African firms operating in the wholesale and retail sector significantly and negatively. These results are consistent with the pecking order theory and confirm the findings of Gleason et al. (2000) and Yazdanfar and Öhman (2015) which indicated that debt financing affects the financial performance of wholesalers and retailers negatively. The results also support the most telling evidence provided by Myers (1993) against the static tradeoff theory. The study by Myers (1993) found a strong inverse correlation between profitability and financial leverage. Other researchers that have found similar results are, amongst others, Booth et al. (2001); Fama and French (2002); Goddard et al. (2005); Dawar (2014), etc. The results from the South African firms confirm the study of Dawar (2014)which suggested that the assumptions of the agency cost theory may be commonly received and accepted in developed and other emerging markets, but this is not the case in many developing and/or underdeveloped markets. The sampled companies seem to prioritize the use of short-term debt before longterm debt, and this could be due to the ease of access of short-term debt as opposed to long-term debt which usually has unattractive terms and conditions for businesses. As pointed out by Yazdanfar and Öhman (2015), Such a strategy can reduce the costs related to information asymmetry and agency conflicts. The lower the leverage level, the lower the agency costs of external debt and the higher the firm profitability. The size of the firm has a significant and positive impact on company performances, which is a contradiction to the findings of Yazdanfar and Öhman (2015) which indicated that size had a significantly negative relationship with profitability in the retail trade and wholesale sectors. However, the size-performance relationship supports the findings of Gleason et al. (2000) which indicated that firm size also influences performance, with larger retailers earning higher return on assets compared to smaller retailers. This indicates that larger companies in the wholesale and retail sector are, on average, more likely to be profitable or perform well financially than their small counterparts. Younger firms seem to be performing well financially than their older counterparts, as they may be flexible in adjusting or adapting to rapid changes that have been one of the characteristics of this particular industry in recent years. This is shown by the significant negative relationship between age of the sampled South African firms and the financial performance of the firm. This confirms the results of Warusawitharana (2014), Dawar (2014) and Yazdanfar and Öhman (2015). Warusawitharana (2014) used the life cycle to explain the negative relationship between age and profitability. The researcher stated that as firms age, they tend to be more likely to transition towards being less-profitable firms. Although the South African firms seem to be trying to balance the use of debt and equity in terms of percentage (Debt use = 52%; Equity use = 48%), the empirical evidence provided by the current study shows that the negative impact of debt capital on firm performance suggest that debt levels in these firms need to be cut down to ensure improved performances. This suggests that managers of profitable companies operating in the South African wholesale and retail sector can use equity capital and retained earnings efficiently, thereby minimizing conflicts of agency or agency costs and remaining independent of external financiers. Future research can explore and study the impact of debt capital on the financial performance of companies in other sectors individually for in-depth understanding of this impact in specific sectors. This would address the root causes associated with the contradictions in this study area with special focus given to individual specific 82

sectors at a time. Others should look to contribute to the literature by studying the wholesale and retail sector individually in other regions (countries) as location has proved to be a significant factor when it comes to the choice of capital structure and its impact on the financial performance of the companies. There are several limitations associated with the current study. One of them is that the data was collected for only the wholesale and retail sector, which was considered to be a less researched industry in the study of capital structure. So, further researchers should attempt to study other less researched sectors for more sectoral detailed analysis. Due to the limited duration of this study, data was collected for only a 5 year period, 2011-2015. Future researches could consider covering a longer period than this. Only two performance measures were used in the current study, so attempts should be made to use more performance measures to improve the understanding of the impact of debt capital on the financial performance of companies. Since literature on the wholesale and retail sector under the capital structure study area is limited, the researcher has compared the results with the few that were available at the time of the current study. Wholesalers and retailers were not separated in the current study as some companies were both wholesalers and retailers. References Abeywardhana, D. K. Y. (2015). Capital Structure and Profitability: An Empirical Analysis of SMEs in the UK. Journal of Emerging Issues in Economics, Finance and Banking (JEIEFB), 4(2), 1-14. Abor, J. (2005). The effect of capital structure on profitability: an empirical analysis of listed firms in Ghana. The journal of risk finance, 6(5), 438-445. Abor, J. (2007). Debt policy and performance of SMEs: Evidence from Ghanaian and South African firms. The Journal of Risk Finance, 8(4), 364-379. 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. Alanazi, A. S., Liu, B., & Forster, J. (2011). The financial performance of Saudi Arabian IPOs. International Journal of Islamic and Middle Eastern Finance and Management, 4(2), 146-157. Booth, L., Aivazian, V., Demirguc Kunt, A., &Maksimovic, V. (2001). Capital structures in developing countries. The journal of finance, 56(1), 87-130. David, D. F., &Olorunfemi, S. (2010). Capital structure and corporate performance in Nigeria petroleum industry: panel data analysis. Journal of Mathematics and Statistics, 6(2), 168-173. Dawar, V. (2014). Agency theory, capital structure and firm performance: some Indian evidence. Managerial Finance, 40(12), 1190-1206. Donaldson, G. (1961). Corporate debt capacity. El-SayedEbaid, I. (2009). The impact of capital-structure choice on firm performance: empirical evidence from Egypt. The Journal of Risk Finance, 10(5), 477-487. 83

Fama, E. F., & French, K. R. (1998). Taxes, financing decisions, and firm value. The Journal of Finance, 53(3), 819-843. Frank, M. Z., &Goyal, V. K. (2007). Trade-off and pecking order theories of debt. Available at SSRN 670543. Gleason, K. C., Mathur, L. K., &Mathur, I. (2000). The interrelationship between culture, capital structure, and performance: evidence from European retailers. Journal of Business Research, 50(2), 185-191. Goddard, J., Tavakoli, M., & Wilson, J. O. (2005). Determinants of profitability in European manufacturing and services: evidence from a dynamic panel model. Applied Financial Economics, 15(18), 1269-1282. Hu, Y. C., & Ansell, J. (2007). Measuring retail company performance using credit scoring techniques. European Journal of Operational Research, 183(3), 1595-1606. Mahmoud Abu-Tapanjeh, A. (2006). An empirical study of firm structure and profitability relationship: The case of Jordan. Journal of Economic and Administrative Sciences, 22(1), 41-59. Majumdar, S. K., &Chhibber, P. (1999). Capital structure and performance: Evidence from a transition economy on an aspect of corporate governance. Public Choice, 98(3-4), 287-305. 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. Modigliani, F., & Miller, M. H. (1963). Corporate income taxes and the cost of capital: a correction. The American economic review, 53(3), 433-443. Myers, S. C., &Majluf, N. S. (1984). Corporate financing and investment decisions when firms have information that investors do not have. Journal of financial economics, 13(2), 187-221. Nimalathasan, B., &Brabete, V. (2010). CAPITAL STRUCTURE AND ITS IMPACT ON PROFITABILITY: A STUDY OF LISTED MANUFACTURING COMPANIES IN SRI LANKA. Young Economists Journal/RevistaTinerilorEconomisti, 8(15). Sadeghian, N. S., Latifi, M. M., Soroush, S., &Aghabagher, Z. T. (2012). Debt Policy and Corporate Performance: Empirical Evidence from Tehran Stock Exchange Companies. International Journal of Economics and Finance, 4(11), 217. Salim, M., &Yadav, R. (2012). Capital structure and firm performance: Evidence from Malaysian listed companies. Procedia-Social and Behavioral Sciences, 65, 156-166. Shyam-Sunder, L., & Myers, S. C. (1999). Testing static tradeoff against pecking order models of capital structure. Journal of financial economics, 51(2), 219-244. Yazdanfar, D., &Öhman, P. (2015). Debt financing and firm performance: an empirical study based on Swedish data. The Journal of Risk Finance, 16(1), 102-118. 84