Corporate Governance in Microfinance Institutions

We also conduct the test for detection of the correlation of type AR(1) (see the appendix 1, test 3). We apply the Wooldridge test, where the null hypothesis is the absence of first order autocorrelation. The null hypothesis is rejected that means that the model faces the correlation of type AR(1).

We also find that the residuals are not normal (see the appendix 1, test 4). It is one more reason to use the another model.

As we find the heteroscedasticity and the correlation of type AR(1) in the random effect regression, we decide to apply the Prais-Winsten regression adjusted to these problems.[Prais and Winsten, 1954]. This model is based on specific case of feasible generalized and is the modification Cochrane-Orcutt model [Johnston, 1972]. This model has two strong advantages. The first one is that it does not require the special type of panel data and can be implemented to every panel. And the second one is that the model takes into account two problems of the regression in comparison to other models, which usually consider only one problem. That is why, the Prais-Winsten regression is most suitable model for our sample.

We do the same steps for other regressions. Namely, we compare FE and RE models, choosing the best one. After that we test our regression for different kind of problems: multicollinearity, heteroscedasticity and autocorrelation. It should be mentioned that if the best model is FE regression, we check the heteroscedasticity using the modified Wald test, where the null hypothesis is the existence of homoscedasticity. We figure out that there is no multicollinearity in our regressions, but there are heteroscedasticity and first order autocorrelation. Therefore, we adjust our models and use the Prais-Winsten regressions. All tests and regressions you can find in appendix 1.The summary tables you can see appendix 2.

3. Explanation of results

Let us discuss the results of the regressions. Firstly, we turn our attention to the first group of regressions. This group consists of two regressions, where the dependent variables are the portfolio at risk > 90 days ratio and gross loan portfolio to total assets ratio. We build these regressions in order to find out, whether the women influence the MFI's risk. And we find that the female CEO and female loan officers mitigate the risks, it may happen because women are more hardworking and diligent, so our hypotheses are confirmed. However, women managers are positively related to the risks. If we remember the description statistics of indicators, we should point that the maximum value of the percent of the women managers is 33,8% and this indicator is not so high. This fact is reflected in our model. It is probable that the small percent of women and their zeal cannot cover the effect of risks.

We also can see that average loan affect positively the risks, because the risks of loan default increase. But this influence is too small. The total expenses to assets have positive impact on risks by means of the proxy - gross loan portfolio to total assets ratio. While the operating expense to loan portfolio reduces risks.

Our expectations about the relationship between the firm size and risks are found the conformation. At that time we cannot say the same statement about the age.

The interesting fact we find that if the bank decide to change the legal status to non-bank financial institution, credit union or non-governmental organization, it might increase its risks.

To sum up, it should be stated the only women in management from different levels affect the MFI's risk.

The second regression demonstrates the relationship between the determinants and financial performance. Let us focus on the key points. According to the common sence, the risk diminishes the financial effectiveness. And the result of our regression confirms this statement. We find that women in the board are positively influence the portfolio yield and the profit margin. This point is contrary to our hypothesis. The explanation can be following: women can bring a new view and experience in order to solve some issues. The percent of women managers have positive impact, as we assume in previous section.

We see the four regressions in this group. Only one contains the significant women-related and risks-related variables at the same time. It should be also mentioned that the significant women-related indicator is the percent of women managers, which have the positive impact on the risks. The result is quite controversial, because in the first regression women managers affects positively to the risks and in the second regressions women influence positively financial performance and at the same time an increase in risks decrease the dependent vatiable. Therehore, we can't say that women affect the financial performance through the risks.

Analyzing the third group of the regression we cannot clearly conclude that women influence the social outreach trough the risks. In spite of the significant figures (women CEO and portfolio at risk>90 day ratio) they influence the social outreach in different directions. It is probable that generally women influence the social performance by means of other characteristics.

Conclusion

In the thesis we have set the goal to determine how the MFIs' corporate governance, managed by the women in different levels, influences the financial and social performance in the countries, which are located in Eastern Europe and Central Asia. In order to achieve it, we build three regressions on panel data from 2010 to 2014. We obtain that it is quite difficult to make a direct conclusion and claim that the women influence the financial and social performance only through the risks. It is more likely that women affect the financial effectiveness and the social outreach by the means of other their characteristics.

References

1. Adams, R. B. and Ferreira, D. (2012). Regulatory pressure and bank directors' incentives to attend board meetings // International Review of Finance, 12(1), 227-248.

2. Adams, R.B. and Mehran, H. (2012). Bank board structure and performance: Evidence for large bank holding companies // Journal of Financial Intermediation, 21, 243-267.

3. Aebi, Vi., Sabato, G. and Schmid, M. (2012). Risk management, corporate governance, and bank performance in the financial crisis // Journal of Banking and Finance, forthcoming.

4. Alexa, A., Perryman, A., Guy, D., Fernando, B., Tripathy, A.C. (2016). Do gender differences persist? // An examination of gender diversity on
firm performance, risk, and executive compensation

5. Berger, A.N., Imbierowicz, B. and Rauch, C. (2012). The roles of corporate governance in bank failures during the recent financial crisis, mimeo.

6. Cheffins, B.R. (2012). The History of Corporate Governance // ECGI Law Working Paper N° 184/2012.

7. Christopher. (2011) Corporate governance - a multi-theoretical approach to recognizing the wider influencing forces impacting on organizations // Critical Perspectives on Accounting, 21, 683-695.

8. Dalton, D.R., Daily, C.M., Johnson, J.L. and Ellstrand, A.E. (1999). Number of directors and financial performance: a meta-analysis// Academy of Management Journal, 42(6), 674-686.

9. de Andres, Vallelado, P. and Vallelado, E. (2008). Corporate governance in banking: The role of the board of directors // Journal of Banking and Finance, 32(12), 2570-2580.

10. E-Dubreuilґ, E. and Torreguitart-Mirada, (2015). The confirmation of this fact we see in the Estap` E-Dubreuilґ and Torreguitart-Mirada article.

Размещено на Allbest.ru