DETERMINANTS OF FINANCIAL PERFORMANCE OF NON-FINANCIAL SECTOR: EVIDENCE FROM PUBLIC TEXTILE SECTOR OF SYRIA

Dania Ebrahem Ghaia 1 and Dr. Radwan Al Ammar 2 . 1. Phd cand., dep. Of financial and banking sciences, faculty of economics, tishreen university, lattakia, syria. 2. Professor at the Dep. of Financial and Banking Sciences, Faculty of Economics, Tishreen University, Lattakia, Syria. ...................................................................................................................... Manuscript Info Abstract ......................... ........................................................................ Manuscript History

The objective of this study is to examine the determinants of financial performance of public textile firms in Syria. The analysis of present study is based on unbalanced panel data of 7 public textile firms for the period 2000-2016. Panel multiple regression analysis is used for estimation. The empirical results show that debt financing has a negative and significant effect on profitability of public textile firms operating in Syria, while both company size and sales growth have a positive and significant effect on financial performance.
In order to increase the firm's performance, management showed be aware of the factors that affect the firm's financial performance. Financing decision or capital structure is considered as one of the most important factors that has association with financial performance. Modigliani and Miller (1958) were the pioneers of the capital structure theory, followed by a number of theories among which are more common: Trade-off theory (Modigliani and Miller, 1963), Agency theory (Jensen and Meckling, 1976), and Pecking order theory (Myers, 1984;Myers and Majluf, 1984). All these theories emphasized the importance of capital structure in its relation with financial performance. Not only financing decisions, but also investing decisions play a great role in determining the firm's financial performance. In addition, firm's characteristics (i.e. firm size, firm age, liquidity, sales growth, etc.) have been tested by a lot of empirical studies. The macroeconomic environment (business cycle, crises, etc.) also has its impact on the firm's financial performance.
The aim of this study is to investigate the determinants of financial performance of public textile companies in Syria. The textile sector in Syria plays a key role in the Syrian economy, that is Syria is one of the world's top cotton producers and textiles and clothing represent over 25% of industrial GDP making textile manufacturing one of the largest industrial sectors in Syria (Bisso, 2009;IMF Country Report, 2006). The textile industry of Syria is composed of a few large public companies and a large number of small and medium private companies. There are 26 public firms supervised by the General Organization for Textile Industry (GOTI), which was established by ISSN: 2320-5407 Int. J. Adv. Res. 6(7), 167-178 169 legislative decree No. 2174 dated 31/7/1975. On the other hand, there are approximately 24000 small and medium enterprises (SMEs) owned by private sector. Our concentrate is on public companies because they are large companies and have the major part of the country's cotton and wool yarn spinning capacities and blanket production (Bisso, 2009). In addition, most of private companies are micro or small 1 , therefore there are difficulties regarding getting financial statements from them.
The rest of the paper is organized as follows. The next section reviews the past empirical literatures, followed by a section of the methodology applied in the present study. Next section presents the empirical results and discussion. The final sections draws conclusions and recommendations.
Literature Review:-A lot of studies have examined the determinants of financial performance or have studied the relationship between capital structure and financial performance. The majority of the studies have focused on the insurance and manufacturing sectors in the developed economies (Ayako et al., 2015). Therefore, there still a gap regarding financial performance of non-financial sector in developing countries.   As the debt-to-asset ratio increases, initially the ROE increases until an optimal debt level is reached, after that it starts decreasing.
Sales growth has positive and significant effect whereas the firm size has no significant impact on its ROE. The independent variables: total debt, long-term debt, liquidity, and size.

Pooled OLS
Total debt and long-term debt have negative effect on ROA and ROE. While size and liquidity have positive effect on ROA and ROE.
In case of MBR model, total debt, long-term debt and liquidity have positive effect. The independent variables: short-term debt, long-term debt, total debt, size, sales growth, and growth opportunities.

Random effects model
Short term debt and total debt have negative effect on ROA.
Both size and growth opportunities have positive effect on ROA, while sales growth has negative effect. Return on Assets (ROA) was used as proxy of financial performance, ln of total assets was used as proxy of investing decision, and debt to equity was used as proxy of financing decision. The study found that investing decision positively affected financial performance, whereas there was a negative effect of financing decision on financial performance. Ayako et al. (2015) analyzed the factors influencing the performance of non-financial companies listed on the Nairobi Securities Exchange (NSE) using panel data of 41 companies over the period 2003 to 2013. The researchers used ROA and ROE as measures of financial performance, and they used the following independent variables: board size, board independence, leverage, liquidity, and firm size. Random effects model was applied for ROA estimation and a fixed effects model was applied for ROE estimation. The empirical results showed that board size and board independence have positive effect on financial performance, while leverage has negative effect on financial performance. Liquidity and firm size did not play significant role in determining financial performance. There are no research findings available in Syria context. Thus, this study contributes to the literatures by providing new evidence of determinants of financial performance of textile sector in Syria. In addition, most of literatures have focused on listed companies, but this study has taken non-listed companies.

Methodology:-Population and Sample:
The population of this study consists of /26/ public sector textile firms supervised by the GOTI. The GOTI firms distributed between seven sections: spinning, textile, apparel, garments, wool carpet, wool yarn and other ( Table 2). All these firms are public owned firms and they are non-listed in Damascus Securities Exchange (DSE) market.
Due to data limitation, a sample of /7/ firms that accounts for 27% of all public textile firms was chosen. Table (3) shows the sample under study which geographically distributed among several governess in Syria.

Results and Discussion:-Descriptive Statistics:-
The descriptive statistics results showing mean, standard deviation, minimum and maximum values of public textile sector performance are presented in table (5). As shown in this table, the average value of return on assets has been (-1.2 percent) with standard deviation (3.2 percent) over the period of 2000 to 2016 for the textile firms in the sample. It could be inferred to the negative value of average that public textile firms are achieving loses in average over the period of study. The average value of total debt to total assets is (55.3 percent), which shows that total debt (total liabilities) of public textile firms in Syria exceeds the half of total assets on overage.
Tangibility of assets has an average of 11 percent which imply that only 11 percent of total assets of these firms are tangible assets. This could imply that current assets represent a very high percentage of total assets of firms under study. In addition, it could be inferred according to the high standard deviation values of fixed assets growth and sales growth rates that these rates have been distinctly varying among public textile firms under study. Note: ROA stands for return on assets; TD represents total debt to total assets ratio; FAG represents fixed assets growth rate; TAN represents tangibility of assets; Size represents the company size; SG represents sales growth rate. Source: processed data based on EViews V. 9 Correlation Analysis:-Correlation analysis results are depicted in table (6). Correlation matrix gave us predictions about the association between ROA and other variables. It can be noticed that ROA has negative association with total debt, fixed assets growth, tangibility and Syria's crisis. While it has a positive association with both company size and sales growth.
In addition, correlation matrix enable us to see if the explanatory variables are highly correlated with one another, and thus to check for multi-collinearity problem. Such problem occurs when the correlation coefficients between explanatory variables are high, and most researchers appear to consider the value of 0.9 as the threshold beyond which this problem is likely to occur (Asteriou and Hall, 2007). It can be noticed that the highest degree of correlation (absolute value) exists between tangibility of assets and company size with value (0.70) which is below (0.9).
To be sure about not having a multi-collinearity problem, Variance Inflation Factor (VIF) and Tolerance (TOL) also have been used. If the value of VIF exceeded 10 and the value of TOL is close to zero, then the variable is said to be highly collinear (Gujarati and Porter, 2009). From table (7), it can be observed that all VIF values are below 10 and all TOL values are far from zero, hence there is no multi-collinearity problem for the study variables. Note: ROA stands for return on assets; TD represents total debt to total assets ratio; FAG represents fixed assets growth rate; TAN represents tangibility of assets; Size represents the company size; SG represents sales growth rate; CRISIS represent Syria's crisis of 2011. Source: processed data based on EViews V. 9 Note: TD represents total debt to total assets ratio; FAG represents fixed assets growth rate; TAN represents tangibility of assets; Size represents the company size; SG represents sales growth rate; CRISIS represent Syria's crisis of 2011.

Panel Unit Root Test:-
In the presence of non-stationarity, the results obtained from a regression are totally spurious and such regression is called spurious regression (Asteriou and Hall, 2007). Spurious regression or nonsense regression lead to misleading R 2 and t statistic, and the t statistics cannot be used for testing hypotheses about the parameters (Gujarati and Porter, 2009). The results of panel unit root tests are displayed in table (8). These results report the rejection of the null hypothesis of all variables. Therefore, it can be inferred that all the variables under study are stationary variables at theirs level form. Thus, the regression results would not be spurious. Note1: ROA stands for return on assets; TD represents total debt to total assets ratio; FAG represents fixed assets growth rate; TAN represents tangibility of assets; Size represents the company size; SG represents sales growth rate; CRISIS represent Syria's crisis of 2011. Note2: represents the most general model with a drift and trend; is the model with a drift and without trend;  is the most restricted model without a drift and trend. *,**,*** denotes the rejection of null hypothesis in 10%, 5% and 1%, respectively. Lag length is based on Schwarz Information Criterion (SIC).

Panel Multiple Regression Analysis:-
Panel regression results are presented in table (8). Because there are two proxies of investing decisions (FAG and TAN), three models have been estimated, and thus we could see if the results would change. The probability values of F-statistics are less than the traditional significant levels, indicating that the overall models are statistically fit and statistically significant. R square values indicate that more than 65% of variation in dependent variable has been explained by variation in independent variables for models (1) and (2), while the R square value of model (3) is 27.2%.
Table (8) reports also the Hausman test results. The probability values for the test is less than the traditional significant levels for regressions (1) and (2), indicating that the random effects model is not appropriate and that the fixed effects specification is to be preferred for these models. While, in regression (3), the probability value is above the traditional significant levels, indicating that random effects model performs better than fixed effects model.
Present study also applied panel cross-section dependence test in table (8). The null hypothesis state that there is no cross-section dependence (correlation) in residuals. The probability values for all tests are above the traditional significant levels, indicating that the null hypothesis is accepted.
The results of estimated regressions indicate that debt is significant at 10% level and showing negative impact on financial performance (profitability) of public textile firms in Syria. The negative impact of debt on financial performance is steady with the results of pecking order theory that firms tend to borrow less because firms maintain the sufficient amount of funds internally. Firm's size is significant at 1% and has a positive impact on financial performance. It can be inferred from this result that firms with large size have higher financial performance as compared to small firms, because they benefit from economies of scale and obtain funds at lower costs. This positive impact of size on financial performance is consistence with other empirical studies such as Abbas  Sales growth has a positive and significant effect on financial performance that is the increase in the sales of textile products will increase the profitability of firms. This positive impact is consistence with other studies such as Tauseef et al. (2015).
Tangibility is not significant at any level, and thus it does not play a significant role in firm's financial performance in public textile sector in Syria. This non-significant impact is consistence with other studies such as Abbas et al. (2013) and Azarmi (2014). In addition, fixed assets growth does not play a significant role in firm's financial performance. Syria's crisis also does has a significant effect on financial performance. This could be due to the fact that public textile sector in Syria was achieving losses before the crisis. Processed data based on EViews V. 9 Note: TD represents total debt to total assets ratio; FAG represents fixed assets growth rate; TAN represents tangibility of assets; Size represents the company size; SG represents sales growth rate; CRISIS represent Syria's crisis of 2011.

Conclusion:-
This study aimed to investigate the determinants of financial performance of public textile sector in Syria. The analysis of present study is based on unbalanced panel data of public textile firms operating in Syria. In this respect, dataset of 7 public textile firms operating in Syria for the period 2000-2016 is used. The dependent variable is the financial performance that is measured by Return on Assets (ROA). While the independent variables were: total debt to total assets ratio, fixed assets growth rate, tangibility of assets, company size, sales growth, and Syria's crisis.
The results of panel multiple regression analysis showed that profitability of public textile sector of Syria is affected by debt financing, company size and sales growth. The debt financing has a negative and significant effect, while both company size and sales growth have a positive and significant effect on profitability of public textile firms 177 operating in Syria. On the other hand, tangibility of assets, fixed assets growth and Syria's crisis do not play significant role in determining profitability of these firms.
The findings of this study could enable the managers of public textile sector in Syria to use their financing and investing decisions more efficiently to enhance the profitability of public textile firms. Precisely, the researchers recommend the management of public textile companies to employ minimal debt level and increase their sales of textile products in order to enhance their financial performance by benefiting from economies of scale as they are large companies. This requires a good, qualified and responsible management to ensure achieving better financial performance.
The main limitation of present study is that it is limited and applicable to textile industry of Syria only, and it is not applicable to financial sector that its capital structure different from non-financial sector.