PurposeThe purpose of this paper is to empirically examine the inequality perpetuated through social categories in accessing the social capital generated through the microfinance interventions in India as the country has pronounced economic inequality by social categories like many developing stratified societies.Design/methodology/approachThe study uses survey data collected from 75 villages in rural India and tests whether the formation and maximization of social capital through self-help groups (SHGs) is dominated by social categories, e.g. high-caste groups, males and superior occupation classes. Using logistic regression framework, the study assesses the formation and maximization of social capital through multiple SHG membership.FindingsThe paper finds that the microfinance approach of empowering weaker sections is considerably limited in its success, in the sense that it provides them with the opportunity to the credit access and support through SHGs. But, the empirical model further indicates that social capital in form of these SHGs may fall prey to the dominant social categories, and thus, these institutions may potentially enhance inequality.Research limitations/implicationsThe paper is derived from the secondary data set, so it is unable to comment field reality qualitatively.Practical implicationsMicrofinance policy makers will have an improved understanding of inherent social inequalities while implementing group-based programs in socially stratified societies.Originality/valueSocial capital, if treated as an outcome accumulated in form of groups, provides with an important framework to assess the unequal access through the microfinance interventions. Overlooking the inherent unequal access will deceive the purpose of social justice in the group-based interventions. The microfinance and other welfare policies engaged in group formation and generating the social capital need to be more sensitive to the disadvantageous sections while focusing on multiple group access by disadvantaged social groups.