Skills and Inequality, winner of the 2015 Stein Rokkan Prize for Comparative Social Science Research, aims to reintegrate the analysis of education and training systems into comparative welfare state research. Because “education is special” (Wilensky 1975: 3), it has long been neglected by the political science literature (although not necessarily by the sociological literature). However, in recent years, scholarly interest has increasingly turned to issues related to education and skills, for instance in research on “active social policy” (Bonoli 2013) or the “social investment state” (Morel et al. 2012). In this literature, education and training is considered key for the reconciliation of generous welfare states with competitive liberal economies. But so far the literature has rarely analyzed education policy as such or how exactly education and other parts of the welfare state relate to each other. It is the great contribution of Skills and Inequality to identify these multiple linkages and connections. The book's substantial focus is exceptionally broad. First, it analyses the politico-economic coalitions that have shaped the institutional design of education and training systems, thereby demonstrating that these coalitions are the same ones that have also supported the expansion of the welfare state in the post-war decades. Second, the book examines how variations in the institutional setup of education and training systems – in interaction with welfare states – influence the distribution of income, wealth and employment in the political economy. Third, it explores the effects of educational institutions on individual attitudes vis-à-vis the education system and the welfare state. In all this, special emphasis is given to the role of vocational education and training, which is of utmost importance in many European states but often neglected in political science research. Methodologically, Busemeyer combines comparative historical analyses of a diverse set of countries (Germany, Sweden and the United Kingdom) with a quantitative analysis of up to 22 OECD countries, using both macro-level and micro-level data. The book is organized in two main sections. Following an introductory chapter, Part 1 analyses the political and institutional determinants of education policy. In Chapter 1, Busemeyer first conceptualizes education and training systems along two dimensions. Based on the seminal work by Esping-Andersen (1990), he differentiates between the de-commodification (understood as the extent to which the education system is protected from market forces) and stratification (understood as the extent to which the education system maintains class-related inequalities of access to higher levels of education) of education systems. What emerges resembles the three welfare regimes identified by Esping-Andersen: an Anglo-Saxon group characterised by low levels of de-commodification and stratification, a Nordic group characterised by high levels of de-commodification and low levels of stratification, and a continental European group characterised by high levels of de-commodification but also high levels of stratification. However, unlike Esping-Andersen's welfare regimes, most Southern European countries are grouped together with the Nordic instead of the continental European ones. Busemeyer subsequently develops a theoretical framework to explain the development of these three worlds of education, combining the literatures on varieties of capitalism, the power resources approach and historical institutionalism. Most emphasis, however, is given to the role of partisan politics. In particular, Busemeyer argues that both the varieties of capitalism and historical institutionalist literatures do not sufficiently appreciate the role of political parties in determining the interests of state actors. To counteract this tendency, he therefore develops hypotheses on how the long-term balance of power between political interests shapes state preferences vis-à-vis a specific policy output (e.g. spending levels) but also with regard to how the political process should be organized (e.g. access to decision-making arenas). These hypotheses are subsequently put to an empirical test in Chapters 2 (comparative historical analysis of education policies since 1945) and 3 (cross-national quantitative analysis of up to 22 OECD countries). Comparing the path initialisation, path formation and path consolidation of education and training policy in Germany, Sweden and the United Kingdom, Busemeyer concludes that partisan conflicts were decisive, in particular during the path formation stage (typically around the 1970s), but he also emphasizes how the institutional and historical context shaped parties' reform strategies. These findings are further substantiated by the quantitative analysis presented in Chapter 3, which shows that (secular) conservative governments are likely to increase the private share of education financing, that Christian-democratic ones are likely to expand apprenticeship training, and that Social-democratic ones are likely to increase both public spending on all sorts of education as well as enrolment in tertiary education. Part 2 turns to outcomes (Chapter 4) and individual attitudes (Chapter 5). In Chapter 4 on the effects of education systems on socio-economic inequality and employment performance, Busemeyer shows that higher levels of public involvement in the financing of any kind of education lowers the level of socio-economic inequality. However, in the case of vocational education and training, such public involvement might come at the price of higher rates of youth unemployment. In particular, Busemeyer postulates a trade-off in the case of vocational education and training: while workplace-based types of vocational education and training systems with strong employer involvement (such as in Switzerland) are more effective in reducing youth unemployment, school-based types with little role for employers (such as in Sweden) are more effective in reducing socio-economic inequality. Finally, Chapter 5 analyses the individual-level determinants of attitudes towards the education system but also how the education system shapes preferences on education spending and redistribution. Busemeyer observes important feedback effects (e.g. in countries with well developed vocational education and training systems, respondents are more likely to recommend vocational instead of academic education). However, some of his findings also reaffirm the distinctiveness of education in comparison to other social policies. For instance, he demonstrates that the effect of income on support for government spending on higher levels of education is conditional on the stratification of the education system. Where the children of the low-income groups struggle to get into university, the well-off are happy to support more public spending on higher education. In terms of content, Skills and Inequality is an exceptionally rich book. The author addresses so many questions that a short summary in the framework of a book review can hardly do justice to the book. However, the wide territory that the book covers is also its major weakness. Each of the book's sections raises several new questions that invite further research. For instance, the comparative historical case studies presented in Chapter 2 mainly emphasise state policies and partisan actors – although in education policy, in particular vocational education and training, non-state actors such as business associations also play central roles. In a similar vein, it remains somewhat unclear why the author decided to start his narrative in 1945 – although many countries were set on their respective development paths several decades earlier (Thelen 2004). In the quantitative analysis of cross-national differences in policy output, Busemeyer restricts his focus exclusively to de-commodification and stratification. However, other important aspects – such as the quality of education, the institutional structure, or the involvement of non-state actors in the political decision-making – are left aside. Finally, in the chapter on socio-economic inequality, the author does not consider the possibility that not only the distribution of incomes but also the distribution of skills might differ between countries. If, for instance, the distribution of skills is more compressed in countries with well developed vocational education and training systems, as indicated by the varieties of capitalism literature (Hall and Soskice 2001), then the relationship between education systems and socio-economic inequality must be seen in a new light. Overall, Busemeyer does an admirable job in charting new territory and bringing education back into comparative welfare state research. He succeeds in combining diverse literatures to facilitate exchange across different fields of research. Although the book sometimes raises more questions than it conclusively answers, it makes an important contribution to the literature by bringing an important topic back on the research agenda and stimulating further research. Thanks to scholars like Busemeyer, education and training policies can no longer be ignored.