Most scholars argue that US social policy has been late to develop and smaller in size and influence than that of other rich democracies. This underdevelopment of direct social policy has produced higher poverty rates than those of rich democracies. To explain the distinctiveness of US social policy, some argue that US labor has been comparatively weak, without its own political party, or that US capitalists have been comparatively powerful, dampening the forces for social policy advancement, or that pro-social policy coalitions of labor and farmers never emerged on the US political scene. Others argue that the unusual construction of US political institutions hamstrung potential change. Their extensive fragmentation made it easy for opponents to block social policy measures, and a state that was slow to bureaucratize and democratize also thwarted social policy advocates (see review in Amenta 2003).

Where does Michael Mann stand on these debates about the characteristics of US social policy and what accounts for them? How well do his characterizations of and arguments about the development of social policy in the twentieth century, especially regarding US social policy, work? Mann's volumes do not focus on social policy, but he does write extensively about it and the US case. And his take is of special interest. No expert on social policy has anything like his worldwide comparative and historical perspective, not to mention one that centers on major social changes and power conflicts that range far beyond domestic policy. To answer the first question, Mann largely follows the consensus characterization in the specialty literature, arguing that US social citizenship lagged. Mann argues that US social policy caught up with that of English-speaking and Nordic countries during the 1930s (mainly in Chapters 8 and 9 of Volume 3, as well as in Chapter 3 of Volume 4) and sees social citizenship advancing across the democratic North between the wars and immediately afterward. The United States did not gain as much from the Second World War and postwar periods, however, and fell behind again. From his vantage point, he sees the United States as truly exceptional, however, not in its social policy, but in its race relations.

In 1975, two important quantitative comparative analyses of social policy were published. Harold Wilensky's The Welfare State and Equality addressed the differences in social spending efforts among sixty-odd countries in the 1960s and found that the level of economic development and the age of populations accounted for most of the variance in efforts. This study was the best and latest in a series of cross-sectional analyses of countries and American states that suggested that economic and social modernization processes determined social policy (review in Skocpol and Amenta 1986). The second study, David Collier and Richard Messick's “Prerequisites versus Diffusion: Testing Alternative Explanations of Social Security Adoption,” examined a similar group of countries, but with a wider time horizon, and addressed the adoption of social insurance policies. Their findings cast serious doubt on the modernization hypothesis, as well as on the strategy of generalizing from studies of many countries on social spending at a single point in time. Just a year earlier, Hugh Heclo had published Modern Social Politics and Britain and Sweden, one of the first studies to take a comparative and historical approach to the development of social policy.

In the generation since then, we have learned a great deal about social policy – largely through comparative and historical work of the small-N variety. On the empirical side, we have learned when various countries adopted social programs, when these programs were expanded or contracted, the forms that social policy has taken, when countries completed their systems of social policy, differences in social spending among countries at different points in time, and which countries have seen significant retrenchment in social policy of what sort.

A generation ago few political sociologists placed states and other large-scale political institutions at the center of politics and understood states as sets of organizations. But now we do, transforming the way that political sociologists think about states and political processes. This alternative conceptualization of the field of study has opened up numerous questions and empirical terrains. If states and power are the central subjects of political sociology (Orum, 1988), in our understanding of these key concepts we political sociologists are now all “institutionalists.”

The rise of self-consciously state-centered scholarship was motivated in part by perceived inadequacies in Marxist, elitist, and pluralist theories and behaviorist approaches to politics, including their conceptions of states and their research programs. State-centered and political institutional scholars confronted these theoretical programs by contesting both what was worth explaining in political sociology and the dominant explanations for political sociological phenomena. Unlike the others, state-centered analysts tended to view states, in the manner of Weber, as a set of organizations, but with unique functions and missions. Thinking about states in this Weberian way shifted what was important to explain in political life, and this approach to politics opened up new research questions and agendas. This has especially been the case for analyses of revolutions and social movements, welfare states and social policy, and the development of states generally. Some of these new questions and research agendas promoted by state-centered scholars employing Weberian understandings of states have been taken up by proponents of varying theoretical persuasions, including Marxists and pluralists, who have provided explanatory answers different from those of state-centered scholars and political institutionalists.