Ownership Structure and Financial Performance (Dependent Variable: ROE)
Source SS df MS
Model 0.5465 5 0.1093
Residual 5.5238 488 0.0113
Total 6.0703 493 0.0123
Number of observations 494
F (5, 488) 9.66
Probability > F 0.0000
Adjusted R-squared 0.0807
Root MSE 0.1064
roe Coefficient Std. Err. t P>|t| 95% Conf. Interval
director -0.0412 0.1961 -0.2100 0.834 -0.4266 0.3442
government -0.1840 0.2024 -0.9100 0.364 -0.5817 0.2137
institute -0.2729 0.1980 -1.3800 0.169 -0.6619 0.1161
foreign -0.0765 0.2023 -0.3800 0.705 -0.4741 0.3210
public -0.2199 0.1972 -1.1100 0.265 -0.6073 0.1676
_cons 0.2621 0.1952 1.3400 0.180 -0.1215 0.6456
Table 3: Ownership Structure and Financial Performance (Dependent Variable: ROE)
From Table 3, it can be stated that the ownership structure affects ROE negatively but insignificantly. However, similar to the results of ROA, P-values are greater than 0.01, 0.05, 0.10 inferring that ownership structure does not have any significant effect on ROE. All dependent variables have non-significant negative impact on ROE which indicates that any change in ownership structure will affect ROE negatively and non-significantly.
P-value of the model is 0.0000 (value less than 0.01, 0.05, 0.1) stating that at least one independent variable is significant. R-Squared value is 0.0900 which is lower but acceptable meaning that 9.00% of ROE is explained by all dependent variables together.
6.3 Multicollinearity Analysis
One of the most common methods to determine Multicollinearity among variables is Variation Inflation Factors (VIF). It tends to quantify the amount of inflation that occurs in the estimated coefficients’ variance when multicollinearity exists. VIF of 1 indicate that there is no correction among predictor variables. In general cases, rule of thumb states that VIF warrants exceeding 4 numbers require further investigation but VIFs greater than 10 indicate significant multicollinearity.
Variable VIF 1/VIF
director 53.97 0.018527
public 52.42 0.019078
government 18.3 0.054638
institute 17.89 0.055905
foreign 12.19 0.082038
Mean VIF 30.95
All the variables have VIF value which is greater than 4 and even 10 therefore it can be stated that the variables have significant multicollinearity amongst them.
. vce, corr
Correlation matrix of coefficients of regression model
e(V) director government institute foreign public _cons
government 0.9697 1.0000
institute 0.9675 0.9496 1.0000
foreign 0.9533 0.9256 0.9399 1.0000
public 0.9884 0.9679 0.9676 0.9544 1.0000
_cons -0.9964 -0.9726 -0.9782 -0.9596 -0.9958 1.0000
From the above depicted table of correlation matrix of coefficients of regression model it can be stated that all of the variables are highly correlated with each other as expected from the result of variation inflation factor (VIF). Most correlated are public ownership and directorial ownership with a coefficient of 0.9884. Whereas, lowest correlation exists between foreign ownership and government ownership with a coefficient of 0.9256. Correlation could be stated significant if coefficient is greater than 0.5000. Thus, it can be inferred that severe multicollinearity exists in the regression model.
6.4 Breusch-Pagan Test
Breusch-Pagan (BP) test is one of the most used tests for testing heteroskedasticity in a linear regression model. Heteroskedasticity occurs when standard deviation of any variable which is observed over a certain time period is not constant. It is also known as error variance within at least one independent variable in a sample that is used to calculate the margin of error between data sets i.e. actual results vs. expected results.
Breusch-Pagan / Cook-Weisberg test for heteroskedasticity
Ho: Constant variance
Variables: director government institute foreign public
chi2(5) = 447.41
Prob ; chi2 = 0.0000
Breusch-Pagan test results in a chi-squared distribution and for my data the statistic is 447.41. P-value is resulted from chi-squared test. Null hypothesis is rejected if p-value is less than 0.05. Results of the above test state a p-value ; 0.05 therefore there is significant evidence of heteroskedasticity.
6.5 Hypothesis Outcome
Hypothesis linked with directorial ownership affecting financial performance:
? As p-value of ROA and ROE (0.201 and 0.834 respectively) for directorial ownership is greater than 0.01, 0.05 or 0.10 for both ROA and ROE. Null hypothesis cannot be rejected, rather it is accepted stating that directorial ownership has no significant impact on financial performance.
H0: Directorial ownership has no significant impact on financial performance
Hypothesis linked with government ownership affecting organization’s financial performance:
? As p-value of ROA and ROE (0.707 and 0.364 respectively) for government ownership is also greater than 0.01, 0.05 and 0.10 for both ROA and ROE. Null hypothesis cannot be rejected, rather it is accepted indicating that government ownership has no significant effect on financial performance.
H0: Government ownership has no significant effect on financial performance
Hypothesis linked with institutional ownership affecting organization’s financial performance:
? As P-value of ROA and ROE (0.910 and 0.169 respectively) for institutional ownership is greater than 0.01, 0.05 and 0.10 for both ROA and ROE. Null hypothesis is accepted inferring that institutional ownership does not affect financial performance.
H0: Institutional ownership does not affect financial performance
Hypothesis linked with foreign ownership on financial performance:
? As p-value of ROA and ROE (0.066 and 0.705 respectively) for foreign ownership is less than 0.10 for ROA, Null hypothesis is rejected stating that foreign ownership has significant relationship with financial performance (ROA). However, P-value of foreign ownership is greater than 0.01, 0.05 and 0.10 for ROE.
Ha: Foreign ownership has significant relationship with financial performance
Hypothesis linked with public ownership and organization’s financial performance:
? As p-value of ROA and ROE (0.656 and 0.265 respectively) for public ownership is greater than 0.01, 0.05 and 0.10 for both ROA and ROE, Null hypothesis is accepted. This infers that public ownership does not affect financial performance.
Ho: Public ownership does not affect financial performance
6.6 Reasoning of Results
As per the results of our analysis, the most of the dependent variables are not significant to affect financial performance of the organization. There could be many potential reasons behind the insignificance, as stated below:
? Data from 4 different industries are used to analyze regression equation whereas, the business model or ownership structure (as per law) is not the same.
? Directorial ownership consists of 43.38% of the ownership structure on an average stating that most organizations have directors as their major shareholder but does not affect financial performance of organizations. This could be due to average of 37.64% shareholders being public and managers has to take decision regarding business aligning benefits to both the directors and public shareholders. Therefore directorial ownership does not affect financial performance of the organizations.
? Government ownership is only 1.79% of the total ownership concentration on an average and financial institutions have regulations to comply from regulatory bodies. This could be the major reason for insignificant relationship between government ownership and financial performance.
? Institutional ownership consists of 14.21% in ownership concentration on an average. Investors could be weak in Bangladesh and are not capable enough to provide monitoring services which is in accordance with studies of Sanchez and Garcia (2007) and Lee (2008).
? Foreign ownership consist of 2.76% of ownership concentration on an average. The amount of average foreign ownership is nominal and this empirical study states that financial ownership does not affect organization performance significantly. However, the relationship with ROA is significantly positive. Foreign owners are concerned more about long term profit, therefore, are expected to invest more in structured organizations.
? Public ownership is second largest in shareholding concentration, i.e. 37.64%, on an average. But according to this study it can be inferred that public ownership does not affect organization’s financial performance. Public ownership does not allow shareholders to control or influence any major decision of an organization. As public shareholders are not a major part of decision making process, it is expected not to have any significant relationship with organization’s financial performance.
? Taking all results in to account, the study shows that organization’s ownership structure does not affect financial performance.
Chapter 7: Limitations
? Data collected from 99 organizations and of five years only, which is not enough to determine relationship in such broader scale. Researches conducted across the globe suggests that all organizations from an industry is required to obtain a clear picture of relationship between variables.
? Data of organizations belong to different industries, whereas various industry has certain trends that is dependent on social or economic condition of the country. Different industries prefer different structure of ownership and impact on financial performance. Due to having data from 4 different industries, there is structural differences between ownership, business type, costs associated and possibility of profitability.
? Only secondary data was used to conduct the research. Collecting primary data could have provided a better outcome with in depth qualitative analysis. Quantitative analysis would allow to conduct research based on factual data only which is publicly available, therefore, in-depth analysis is not possible without availability of primary data like interview sessions, focus group discussions etc.
? The political instability of Bangladesh during 2013 had an impact on the economy due to which most of the organizations were affected. Bangladesh faced a decrease of GDP performance by 0.07 per cent which can be attributed to the political turmoil. Costs increased which in turn decreased profitability during this period of instability which is not accounted in the research.
? Data of only enlisted organizations were used to conduct this research. Therefore, the results cannot be generalized due to small sample size and absence of organization of all type.
Chapter 8: Conclusion
The research paper empirically investigates relationship between ownership structure and financial performance of an organization. It involves determining the importance of dependent variables like directorial ownership, government ownership, institutional ownership, foreign ownership and public ownership. Research is based on a sample of 99 enlisted organizations from 4 different sectors of Bangladesh and 5 years (2012-2016) of financial data from respective annual reports. Results obtained from regression analysis are similar to the findings of most studies in the literatures mentioned in this research.
Overall, the findings of research state that there is no significant relationship between the majority of variables determining ownership structure and Return on Assets, although the effect is positive but is insignificant. Foreign ownership has significant effect on Return on Asset meaning that ROA will increase with more foreign ownership.
Similar results have been observed for relationship between ownership structure and Return on Equity where the relationship is insignificant but effect is negative. No dependent variable is significant enough to cause negative effect on the relationship.
Hence, it cannot be concluded that ownership structure affects organizational performance.