In the 1980s, many scholars of both comparative and American politics argued that states often act autonomously from social demands. Rejecting reductionist assumptions regarding the primacy of social groups for public policy, both groups of scholars examine how government actors and preexisting institutional constraints influenced policy implementation. Since then, however, while the state has been retained as the primary unit of analysis for most studies of American political development, interest in the autonomy of the state has dwindled, and scholars have increasingly focused on how social groups and electoral outcomes explain state formation and public policy, especially in the nineteenth century. In some instances, scholars have even denied that state autonomy is a relevant concept for the study of American political development.