The political goals and strategies of employers have received increasing attention in comparative political economy in recent years. A focus on the role of employers in the introduction and expansion of welfare state programs and industrial relations institutions has come to challenge the conventional view of these institutions as the product of left-wing political power and class conflicts (Shalev 1983; Korpi 1983; Esping-Andersen 1985; Huber and Stephens 2001), a view that assumes employers’ opposition to social policy. Employer-centered studies have questioned the validity of this assumption, arguing that employers often backed policies protecting workers because they thought that these policies would contribute to labor productivity and profitability (Jenkins and Brents 1991; Swenson 2002; Mares 2003a; Gordon 1994; Martin 2010: 565-6, 571; Martin and Swank forthcoming: 9). Research on the benefits of social policy and employee participation rights for the development of a skilled and cooperative workforce (Addison 2009; Estévez-Abe et al. 2001; Hall and Soskice 2001; Thelen 1991) has given additional credibility to the argument that business interests can be a source of social protection. I call this argument the “business interests thesis” because it contends that business interests do, in part, explain the generosity of the modern welfare state. Against this view, I argue in this book that the extent of employers’ support for social policy varies with the type of political challenge they face. When confronted by what they perceive as a revolutionary challenge from the left, employers respond by promoting an expansion of social protection or bargaining rights. I call this the politics of pacification. When confronted by a reformist challenge, that is, challenge demands for the expansion of social protection in the absence of a revolutionary, they respond by promoting more moderate types of reform. They try to contain the scope and generosity of social protection by endorsing what they perceive as “lesser evils.” For this reason, I call this the politics of containment. In the absence of a challenge to their interests, employers typically show no interest in the expansion of social policy or bargaining rights. My analysis shows that challenges from the political sphere explain the variation in employer support for reforms expanding social protection better than differences in production strategies or economic interests. These challenges may come, for instance, from the labor movement, a reformist government

bureaucracy, or political parties competing for the votes of pro-welfare citizens. Employers adapted their social policy positions to these challenges. In the early days of the labor movement they backed social protection and bargaining rights when they thought that private ownership of industry was under threat from revolutionary forces out to expropriate them. During later periods, characterized by social policy expansion, they strategically backed those policy options they considered the least costly or least disadvantageous. In both cases, their goal was to ward off developments that they thought were even less desirable. In short, my argument is that while employers do not always oppose social protection, they do not promote it in the absence of political challenges that render social protection a lesser evil. I call this the political accommodation thesis. The emphasis in this book on how political challenges shaped employers’ social policy positions contradicts the hypothesis that the economic interests of firms are an important source of social protection. Even when employers supported reforms extending social benefits, they thought that these benefits would undermine labor productivity and work incentives. When they supported social reforms they did so to guard against outcomes that they thought would be even worse, rather than because they thought that social policy would help them to create the kind of labor force they needed. On balance, the impact of employers on welfare state development was a constraining one. In the absence of employers’ interventions, social reforms would probably have been more far-reaching. This is because employers’ social policy proposals, while not generally hostile to social protection, were typically less generous than those of other powerful actors. After the waning of the revolutionary challenges of the early days of the labor movement, employers lost interest in the expansion of social protection. Their focus shifted to containing the further growth of social expenditures. To the extent they succeeded in doing so, employers thus limited welfare state expansion. My conclusions are based on an in-depth case study of the development of the German welfare state. I chose Germany as a crucial case, which needs to fit a theory closely if we are to have confidence in it (Eckstein 1975: 118). For the political accommodation thesis to hold, we would need to find evidence for it in Germany, because this is a case where the alternative business interests thesis has greatest plausibility. This is because Germany constitutes a paradigm case for a type of economy that benefits firms relying on high-quality/high-skill production (for example, Thelen 2001; Hassel 2007; Hall 2007; Streeck 1992; Hall and Soskice 2001). Given that it is firms of this type that benefit most from social policy and institutions for class cooperation (Estévez-Abe et al. 2001; Iversen 2005: 8), we should be most likely to find evidence of business support for social policy and bargaining institutions in Germany if the business interests thesis is right. In other words, Germany constitutes a difficult case for the political accommodation thesis because of the particular credibility of the alternative business interests thesis. If we find support for the political accommodation thesis in this case, it is likely to be relevant also in cases where the economic benefits of the institutional setting are less developed.