Abstract: This study examined, the role of social capital in access to micro credit in Ekiti State. Multi-stage stratified random sampling procedure was used to collect data from 116 households in the study area. The analytical methods used in this study include Descriptive and Regression analyses (Probit regression method). Results revealed that aggregate social capital index positively affects the probability of members of networks obtaining micro credit at 1% significant level and the marginal effect shows that increasing social capital by a unit will lead to 0.22% in probability of members of associations obtaining micro credit. Disaggregating the social capital index, the variables that significantly influence probability of obtaining micro credit includes memberships in associations, cash and labour contributions by members to various associations. Nevertheless, the study supports findings that in addition to information and other benefits derived from networks, it can be a source of obtaining credit. The study concluded that belonging to networks or associations improves the probability of obtaining credit for members, which can be channeled towards improving their livelihood activities.

INTRODUCTION

Finance remains a real obstacle for many rural dwellers where a combination of agricultural risk, scarce borrower information, cumbersome legal procedures and high transaction cost means that many formal financial service providers are reluctant to serve the rural people. For these reasons, the Nigerian government and donors set up credit programmes aimed at improving rural household’s access to credit. The vast majorities of these programmes that provide credit at subsidized interest rates failed and were rarely sustainable (Olowononi, 1997). However, in the same environment, credit from informal sources has been of mixed blessings. It exhibits very low loan default rate but interest charged on credit is very high crowding out very poor people. The most common response of the rural poor to their marginalization (from the mainstream credit) has been to ‘turn inwards’ that is to take initiative to mobilize their own resources as a method of meeting their credit needs and of reducing their dependence on the outside and one approach that has met with widespread acceptance is micro credit the idea of loaning very small amount of money to the poor in order to smoothen consumption or promote entrepreneurial endeavours.

However, statistics attest that the demand for micro finance financial services remains largely unmet (Zeller and Sharma, 1998; Buchenau, 2003; UNDP, 2004). This will undermine the first Millennium Development Goal (MDG) on the eradication of extreme poverty and one of the reasons is incomplete information equilibrium in credit markets (Stiglitz, 1990). One of the factors contributing to informational uncertainties includes inability of poor people to provide individual collateral, in reaction an increasing number of microfinance institutions provide credit on the basis of social collateral through, which the social networks to which, they belong replace physical collateral. This study seeks to answer the research questions: what are the characteristics of the different networks? Can existing social capital significantly improve access to micro credit? The aim of this study is to identify the characteristics of the networks and empirically measure the impact of social capital in access to micro credit in Ekiti State using a social capital index constructed from data collected with the aid of Principal Component Analysis (PCA).

Some empirical studies relates social capital with indirect outcomes such as welfare proxied by households expenditure (Grootaert, 1999; Yusuf, 2006) however, this study focuses on the relationship between social capital and one of its direct outcomes (access to credit).

One of the policy objectives of the new Nigerian Microfinance Policy is to enhance service delivery by microfinance institutions to micro, small and medium entrepreneurs (CBN, 2005). The findings of this study can serve as guide to policy makers on how social capital can reduce information uncertainties between the borrowers and the microfinance institution using the case study of Ekiti State, which was rated, as the second poorest state in the South-Western zone of Nigeria after Lagos State (Nbs, 2005).

Conceptual framework: The term social capital found its way into economic analysis only recently, although for long various elements of the concept have been in existence under different names social capital represents the degree of social cohesion in communities. It refers to the processes between people that establish networks, norms and social trust and facilitate coordination and cooperation for mutual benefits.

The narrowest concept of social capital is associated with Putnam (1993). He viewed, it as a set of horizontal associations between people. Social capital consists of social networks and associated norms that have an effect on the productivity of the community. The key feature of social capital in this definition is that it facilitates coordination and cooperation, for the mutual benefits of the members of the association. Another encompassing concept of social capital includes the social and political environment that enables norms to develop and shape the social structure.

There is growing evidence that social capital can have impact on development outcomes growth, equity and poverty alleviation. Bambang (2000) cited that associations and institutions provide an informal framework to organize information sharing, coordination of activities and collective decision making it beneficial to the welfare of the members.

The concept of Micro credit is as much about money as it is about information with sustainability and non-dependence on external resources being key to the growth of micro credit programmes. It is fairly, well understood how micro credit programmes use the existing networks of horizontal associations to lower some information and other transaction costs (Besley and Coate, 1995; Stiglitz, 1990).

First, the task of credit evaluation is done by the borrowers, rather than the bank, since members choose to be in a group with others whom they believe to be credit-worthy and can rely on to make timely payments. This is called the peer screening effect and it reduces transaction costs, because community members have much better information about each other than the bank has. Varian (1990) has analyzed this effect in a model where, banks can obtain perfect knowledge of borrower types by incurring some screening costs. Second, there is a peer monitoring effect that induces each group member to apply the loans in a productive way, so that the probability of success increases. Stiglitz (1990) and Varian (1990) have already developed models that illustrate the working of the peer monitoring effect. Third, the desire to preserve valuable social ties may induce borrowers to spend extra effort if necessary to secure timely payment because social ties are valuable for future purpose, advice on your business and because they may provide connections (customers, suppliers, etc.) useful for your business. We can call this the social relations effect. Fourth and very important, there is a collateral effect. The banks losses due to unsuccessful projects are greatly reduced because successful entrepreneurs within each group will cover part of these losses. In effect, a kind of collateral is created in the groups, even when none of the individuals had any collateral to offer. The reduction in losses allows banks to reduce interest rates, thus attracting applicants with lower risk projects. This improvement of the pool of applicants will reduce average risk and thus, allow a further reduction in interest rates.

Social capital and micro credit: Oke et al. (2007) analyzed the socio-economic variables that affect micro credit repayment of Clients of NGOs in South Western Nigeria. Some of the variables that significantly influence repayment include, amount of loan borrowed, penalty for lateness to group meetings, membership of cooperative society, number of days between loan application and disbursement. Thierry Van Bastelaer (1999) examined how social capital reduced the cost of imperfect information that is congenital to micro credit. He argued that the main source of social capital was the patron-client trust between loan officers and borrowers. Karlan (2001) found that social capital generates higher repayment and higher savings. Olomola (2000) examined the role of social capital in transforming the rural finance system in South Western Nigeria; he noted that relationship based on trust between lenders and borrowers is crucial in developing a sustainable financial system at the grass root level.

MATERIALS AND METHODS

This study was carried out in Ekiti State and was based on primary data. The data were collected through the use of structured questionnaires from a cross-section of households. Data collected at the household level captured information on the demographic characteristics of household members, participation and involvement in local level institutions and provision of micro credit in local level institutions.

A multistage stratified random sampling technique was used to select representative households. The state was stratified into three senatorial districts: Ekiti central, Ekiti Southern and Ekiti Northern zones. One local government area was randomly selected each from the three zones, respectively. Ado Ekiti Local Government Area from the Ekiti Central Zone, Ikere Ekiti Local Government Area from the Ekiti Southern Zone and Ikole Ekiti Local Government Area from the Ekiti Northern Zone. Three enumeration areas were randomly selected from each of the three Local Government Areas using the Ekiti State’s Enumeration Map sourced from the State’s Chapter of the National Bureau of Statistics. Sixteen respondents were randomly selected from each of the nine enumeration areas to ensure that all households had an equal probability of being selected. In all; 144 respondents were randomly selected, however, due to incomplete information from some households, data analysis was based on 116 households with complete information.

Analytical techniques: The analytical methods used in this study include Descriptive and Regression analysis (Probit regression methods). Descriptive statistics such as frequency tables, mean, mode and percentages were employed in the analyses of the characteristics of the local level institutions. Aggregate Social Capital Index was obtained by using Principal Component Analysis on the original variables defining social capital to build dimensions, which was then indexed into an aggregate variable, which was used to run the logistic models. Principal Component Analysis is appropriate for multivariate analysis of a continuous variable, it was used for the transformation because the social capital variables are plenty, which makes computation difficult, the reliability of the estimates may not be possible to assess sensibly due to the loss of degree of freedom and it also solve, the problem of multicollinearity (Koutsoyiannis, 1973). Whiteley (2000) produced a social capital index based on trust variables through PCA in his study of economic growth and social capital also, Narayan and Vella (2006) used PCA to transform the original data into dimensions of social capital in their study of building indices of social capital.

The probit model: Probit regression models are appropriate, when the response takes one of only two possible values representing presence or absence; it was used to measure the impact of social capital and other demographic variables on the probability of access to micro credit in local level institutions by households in the study area.

The Probit Model is expressed as (Gujarati, 2003):

Where,

(1)

(2)

(3)

i = 1, 2, … 116 households

where:

RESULTS AND DISCUSSION

Table 1 shows, the activities of households in Local Level Institutions, with an average of 1.74 associational memberships per households; density is highest in Ikole Ekiti Local Government Area. In Ado Ekiti Local Government Area, each household belongs to an average of 1.6 groups and to 1.17 groups in Ikere Ekiti Local Government Area. In Ado Ekiti Local Government Area, some households do not belong to any group.

Table 1:	Distribution by activities of households in local level institutions

Table 2:	Distribution by monthly cash contributions of households to their associations ()
Field survey (2006)

In terms of meeting attendance, it seems that meetings are most frequent in Ado Ekiti Local Government Area occurring on the average of 5 times in a month, while it is 3 and 2 times a month, respectively in Ikole Ekiti and Ikere Ekiti Local Government Areas. The index of active participation in decision making is significantly lower in Ikere Ekiti Local Government Area than in the other two Local Government Areas. Respondents from Ado Ekiti and Ikole Ekiti Local Government Areas are relatively active in their involvement in the Local Level Institutions, while respondents in Ikere Ekiti Local Government Area are slightly active.

Associations in Ikole Ekiti Local Government Area are most heterogeneous than in the other two Local Government Areas. Associations in Ikere Ekiti Local Government Area are the most homogenous of the three local government areas, on the average, associations in the overall sample are heterogeneous. The cooperative societies are common to all the three local government areas, while the religious societies, community based associations and occupational groups also represent important associations to the sampled households in this study in Ekiti State.

Table 3:	Distribution by monthly labour contribution of households to their associations (days year-1)
Field survey (2006)

Cash contributions are made by households to their associations, part of this savings are used for the general running of the associations and loaned as micro credit to members, who signify interest in loans. Table 2 shows the monthly contribution by members of different associations in Ekiti State. Across the Local Government Areas, association members contribute most to cooperative societies with respondents from Ado Ekiti Local Government Area leading with an average contribution of 4512.50 per month. Given the heavy cash contribution to cooperative societies, occupational groups and religious organizations, most household would seem to partake in these associations for economic gains and spiritual benefits.

Some associations require some form of labour activities from their members towards the development of the associations and their communities. Active members are usually recognized and considered when they require assistance from the associations. Labour contribution by members of association indicates a low level in the overall sample as shown in Table 3. Respondents in Ikole Ekiti Local Government Area do not contribute in terms of labour to their association and this may have adverse effect on the sustainability of these institutions.

Cooperative societies are used in this study to show some characteristics of a typical credit Local Level Institution. The results are revealed in Table 4. The average amount of credit requested by members of the cooperative societies in the overall sample is about 119,375, which is less than the average amount requested in Ikole Ekiti Local Government Area (140,000). The credit requested for ranged from 10,000-400,000, which is also the range for credit obtained. Ikere Ekiti local Government Area had the least amount of credit obtained (85,714.29) among the three local government areas. The interest rate charged on these loans ranged from 0-10%. In Ado Ekiti Local Government Area, 50% of the loan beneficiaries responded that no interest rate was charged while, in Ikole Ekiti Local Government Area, all loan beneficiaries were charged between one and nine percent interest rate. Overall, 35% of the loan beneficiaries paid interest rate ranging from 1-9%.

Table 4:	Selected characteristics of loan from a credit oriented association cooperative, societies
Field survey (2006)

The monthly mode of repayment schedule is common to the respondents in the three local government areas, specifically in Ikole Ekiti Local Government Area; all beneficiaries repaid the loans on a monthly basis. Loans are paid up within a year as shown in the Table 4.

Overall, 35% of the loan beneficiaries paid interest rate ranging from 1-9%. The monthly mode of repayment schedule is common to the respondents in the three local government areas, specifically in Ikole Ekiti Local Government Area; all beneficiaries repaid the loans on a monthly basis. Loans are paid up within a year as shown in the Table 4. A 82.6% of the beneficiaries in this study paid up their loans within a year. Sanctions imposed for failure to meet repayment deadline ranged from no sanction to doubling of interest rates. In Ikere Ekiti Local Government Area, 100% of the loan beneficiaries responded that no sanctions were imposed, so also 81.8% of the beneficiaries in Ado Ekiti Local Government Area. This could give rise to high loan defaults, which makes the institutions non-sustainable. In Ikole Ekiti Local Government Area, all loan beneficiaries responded that the loan was used/will be for economic purposes. Overall, 36.58% of the loan beneficiaries in the state listed loan delays as a major problem encountered in obtaining credit from their associations and this is closely followed by inadequacy of funds in their associations (34.1%).

Social capital and access to micro credit: The relationship between social capital and access to micro credit is revealed in Table 5. Three Probit regression equations were run to explain the effect of social capital on access to micro credit in local level institutions. In the 1st model: the basic model, the log-likelihood function of (-60.99) and the Chi-square χ² (30.97); the log-likelihood function (-55.23) and Chi-square χ² (42.49) of the second model with Aggregate Social Capital Index; the log-likelihood function (-44.66) and Chi-square χ² (63.24) of the third model with additive Social Capital variables were all significant at 1% level, respectively with different variables being significant in the 3 models. This shows that all the three probit models were well fitted.

In the basic model, five explanatory variables were statistically significant at different levels and these include sex, age, household’s credit history, occupation and households located in Ikole Ekiti local government area.

Table 5:	Probit regression analysis of the effect of social capital on access to micro credit at local level institutions
Computer printout of the data in September 2006; The dependent variable is access to micro credit by households at local level institutions (1 = yes, 0 = no); *Significant level at 1%; **Significant level at 5%; ***Significant level at 10%

Male headed household’s increases probability of access to micro credit at 10% significant level, this is expected in a society where, women have less access to resources due to cultural issues. The age of household heads increases the probability of access to micro credit at 1% significant level, this is because, the average household head in this study is at his economic active age, households located in Ikole Ekiti Local Government Area increases the probability of access to micro credit at 5% significant level, which might be connected to the high level of diversity among members of the local level institutions, which enables them to pool resources. On the other hand, household’s credit history reduces probability of obtaining micro credit at 1% significant level, households engaged in farming activities has a negative regression coefficient of 0.811 at 5% significant level, being a farmer reduces probability of obtaining micro credit, this may be due to uncertainties associated with agricultural activities in Nigeria, which makes farming a risky business. Also, the marginal effect shows that if farmers increase by a unit percent, the probability of obtaining micro credit decreases by 2.87%.

With the introduction of the Aggregate social capital index to the model, male headed households increases the probability of access to micro credit at 5% significant level, this shows that males are more likely to receive micro credit from local level institutions than females. Also, age of household head and households located in Ikole Ekiti Local Government Area increases the probabilities of access to micro credit at 1% significant levels, respectively. Household size and lifecycle of household’s access to micro credit reduces the probabilities of obtaining credit at 5 and 1% significantly levels, respectively. This implies that households with large family size are less likely to receive micro credit. The Aggregate social capital index increases the probability of obtaining micro credit at local level institutions at 1% significant level. This is perhaps, the sense, in which social capital is truly social in that the buildings of net works and trust among members in the context of a social setting spills over into financial benefits by easier access to credit. The marginal effect shows that if household size increases by a unit percent, the probability of obtaining micro credit decreases by 0.50% and if the aggregate social capital index increases by 1%, probability of obtaining micro credit increases by 0.22%.

The third model reveals the inclusion of six social capital variables in other to find the dimensions of social capital, which truly increases the probability of obtaining micro credit. These are: household memberships in associations, index of participation, heterogeneity index, meeting attendance, cash contribution score and labour contribution score. This disaggregating shows that the effect of social capital on access to micro credit are traceable to memberships of households in associations, cash contribution of households to associations, household’s labour contribution to associations. In line with the view of Grootaert (1999) and Yusuf (2006), additional memberships of households in associations increases the probability of access to credit at local level institutions at 1% significant level in this study. The marginal effect shows that a 1% increase in membership of association by household will lead to a 0.43% increase in probability of obtaining micro credit.

Household’s cash contribution to association is presumably a sign of greater interest in the association and serves as a collateral effect for households wanting to borrow money. Cash contribution to associations by households increases the probability of access to credit at 10% significant level. In term of labour contribution, a 1% increase will lead to a 0.25% increase in probability of obtaining micro credit. Labour contribution to associations is minimal so households willing to contribute man hours to their association are likely to be regarded with respect to loan issuance.

CONCLUSION

Provision of micro credit on a sustainable basis for the poor and low income groups in Nigeria is important in order to achieve the Millennium Development Goal of halving poverty. However, the size of the unserved market by existing financial institutions is large and one of the reasons adduce for this is incomplete information equilibrium in credit markets, which can be bridged by social capital due to it’s peer screening effects, peer monitoring effects and the peer collateral effects. The study proved that aggregate social capital index positively affects the probability of members of networks obtaining micro credit. Disaggregating the social capital index, the variables that significantly influence probability of obtaining micro credit includes memberships in associations, cash and labour contributions by members to various associations.

The study showed that the heterogeneity of associations were relatively low for those in Ado Ekiti and Ikere Ekiti Local Government Areas. In formation of Local Level Institutions, associations and networks should target members, who have diverse needs as well as multidimensional interventions to help members.

The loans received by households from networks and associations were untimely and this may lead to diversion of loans for non economic purposes, which will affect repayment schedules. Associations and groups should process loan requests on time so that members can make use of loans for economic purposes. Investments in social capital deserve to be part of poverty alleviation programmes since the return to investment in social capital are larger for the poor than for others and this will form part of the economic interventions for states such as Ekiti whose poverty level is high.

This study supports findings that in addition to information and other benefits derived from networks, it can be a source of obtaining credit, belonging to networks or associations will improve the probability of access to credit for members, which can be channeled towards improving their livelihood activities.