In this paper we tried to identify classes and class fractions in the social space in Serbia and Croatia and to determine the social mechanisms which shape them. If in Bourdieu’s conception, classes and the fractions within them are defined as clusters of positions, or regions in the social space, by using fuzzy cluster analysis we tried to discern the regions of social space which are characterized by a similar volume and composition of capital. And we tested whether these statistical constructs truly capture “real” groups, or groups which could through activities of political mobilization be “transformed” into classes and class fractions in two different ways. Building, on the one hand, on the tradition of the Cambridge Stratification Group (Prandy, Stewart, Blackburn, Bottero, Holmwood, Siltanen), we studied whether there are interactions between members (“classes” and “class fractions”) within the identified clusters in social space (in the form of marriages or friendships) and whether any social boundaries emerge in relation to the “members” of other clusters (a lower level of interaction or absence of marriages and friendships). On the other hand, following the tradition of Bourdieusian analyses of “classification struggles”, we tested these same clusters in relation to similarities and differences in their lifestyles, that is, whether “within” the identified clusters (classes and class fractions) their “inhabitants” have similar lifestyles, and whether they establish symbolic boundaries in relation to the members of other clusters. That was how we identified three clusters in Croatia and Serbia which we labeled as Capital Rich Classes (CRC), Classes with Average Capital (CAC) and Capital Poor Classes (CPC) and six sub-clusters which differ based on the level of interaction with them (marriages and friendships) and based on the lifestyles of their “inhabitants”. In the end, using a multinomial logistic regression, we tried to identify the social mechanisms which create social inequalities in the two countries under analysis and define the objective structural relations between classes, starting from the position in the division of labor and the education level of all the members of the household, to non-economic factors (such as membership in political parties, the difference between belonging to ethnic majority and minority groups, between urban and rural areas, regional differences) ; economic factors outside the sphere of formal economy (moonlighting, self-provisioning), but also the influences of the composition of the household itself (the gender and age composition of the household, that is, economically active and dependent household members). If in his class analysis Bourdieu distinguishes between two aspects: the analysis of class condition (the material conditions of existence of their members and the type of work which the members of a particular class perform) and the analysis of class position (their relations with other classes), then in this text we mostly dealt with this second aspect. And what these studies of the interaction of agents within the classes and class fractions and objective, structural relations between them brings we denoted as the move from the existential to the social class.