Analysis of microfinance performance and development of informal institutions in Cameroon

IV.4- Regression approach

It is important to underline that, our research focuses on the study of the link between

social and financial performance on one hand and MFIs performance and development of the informal sector on the other hand. In fact the repetitive verb in our dissertation is «to link». This verb implies that we are studying the correlation among the variables which characterizes each

56

Analysis of microfinances' performance and development of informal institutions in Cameroon

By Djamaman Brice Gaétan

indicator. Therefore, the regression approach, based on the Ordinary Least Square (OLS) is used

in this research.

To best describe our hypothesis, the research requires the following regression models

? Financial performance regression;

? Social performance regression;

? Informal sector regression.

Notes: the mission drift regression is included in the social and financial regression (H1), which is the consequence of negative link of the influence of social performance on financial performance. The list of variables and indicators are given in the appendix

General multiple regression models are used to analyse the explanatory function of the control variables and independent variables. The selected financial and social performance indicators are first used as the dependent variables for testing hypothesis 1 and 2. Concerning hypothesis 3 and 4, the indicators of the development of the informal sector function as dependent variables, whereas social performance, financial performance and control variables are used as independent variables. It is important to notice that even the informal sector and control variables are considered as independent variables in the financial and social performance regression.

The next chapter provides an insight in the descriptive statistics of the variables and indicators presented in the dataset. Preliminary, the minimum and maximum values suggest a wide range for many of the variables. Hence, outliers may be a concern in the regression analyses. Woolridge (2003, p. 312) stated «OLS is susceptible of outlying observations because it minimizes the sum of squared residuals: large residuals (positive or negative) receive a lot of weight in the least squares minimization problem». Cull et al. (2007, p. 17) faced the same concern and applied a robust estimation technique. The authors found that «those results are similar to the base results, although there are a few minor differences». The next chapter also provides an insight in the correlation and coefficients of regression between the selected variables and indicators.

Analysis of microfinances' performance and development of informal institutions in Cameroon

By Djamaman Brice Gaétan

IV.5- conclusion

A selection of variables and indicators used for the financial performance, social performance and informal institutions has been presented. For robustness, a selection of control variables has been added to the regression models. We have given the different hypotheses underlying this research. From the hypothesis one assumes that «social performance influences financial performance of MFIs» The next hypothesis assumes that: «financial performance influences social performance» The third hypothesis assumes that «financial performance influences the development of the informal sector» and the last hypothesis supposes that «social performance influences the development of informal sector»

In this research the OLS regression approach is used. The regression approach has been successful in previous studies. In line with the hypotheses, the research contains three general regression models: financial performance regression, social performance regression and informal sector regression.

57

58

Analysis of microfinances' performance and development of informal institutions in Cameroon

By Djamaman Brice Gaétan

CHAPTER V- PRESENTATION AND ANALYSIS OF DATA

This chapter contains four sections. Section 1 provides an insight in the various sources and the process of data collection. In section 2, multiple sources have been combined in order to collect general information, financial and social performance data of 45 active MFIs in Yaoundé (Cameroon). In section 3, we will give preliminary data analysis and section 4 explains the different regressions analysis that we will perform in our research.

V.1- Data collection

Data are obtained from various sources and can be classified into primary data and secondary data.

? Primary data: These data were obtained largely through the survey undertaken by COBAC and the Ministry of Finance. Indeed, the objective of investigation had as finality the setting-up of a report which was followed by an assessment of Cameroon?s MFIs, carried out within the framework of the implementation of the Microfinance Activity Evaluation and Supervision System (SESAME). This implementation was carried out from March to September 2011. We have also obtained primary information from some surveys undertaken by the Ministry of Economy, through their data base on MFIs.

? Secondary data: this category of information was obtained firstly from the Microfinance Information eXchange database (MIX MARKET, web site: www.mixmarket.org) and secondly from Cameroon National Institute of Statistics web site: www.ins.cm.

It is important to underline that it is inside the SESAME report that we have found the most important and available information concerning the 45 MFIs which will constitute our sample. The list and different acronyms for the concern is given in the appendices

V.2- The data set

The dataset contains general information, financial performance data, social performance and non-formal institutions data from 45 MFIs of Yaoundé. All the observations are from the year 2010. Let us mention that the sample was drawn from the population of Cameroon MFIs which is about 488 microfinances. Table 7 shows the distribution of microfinances based on their categories

59

Analysis of microfinances' performance and development of informal institutions in Cameroon

By Djamaman Brice Gaétan

Table 7: Distribution of microfinances based on their categories

Categories	Center region	East region	Littoral south region	and west	West region	Far region	North	North region	west	Total
First	45	13	32		52	79		221		442
Second	17	0	18		2	1		4		42
Third	0	0	2		0	2		0		4
Total	62	13	52		54	82		225*		488

* CAMCCUL network is integrated in North-West region. Source COBAC

Concerning the network we have CAMCCUL: 191, A3C: 34, UCCGN: 9, CMEC West: 19, CMEC North West: 9 and MUCADEC: 3

From the 488 approved MFIs in 31^st December 2010, there are 442 in the first category, 44 in the second category and 4 in the third category. This sector is predominated by MFIs of first category which represents 90.1% of total MFIs, follows by the structures of second category (9%). MFIs of third category that are essentially of old project are established in far North and West areas. Despite the fact that the proportion of MFIs of second category is low, they still control almost half the market especially in terms of fixed deposits and loans from clients. Another network received in 2011 is the agreement of monetary authority with three of its affiliated, they include the MUCADEC network.