Rudolf Klein has build up an impressive track record in writings about health policies and politics, more specifically, about the (new) politics of the NHS. His seventh edition of The New Politics of the NHS is another milestone in his academic career. The most important contribution of this book and its predecessors is probably that he has given successive generations of English stakeholders a collective memory of ‘their’ NHS. The value of this should not be underestimated, for, if path dependency is so important for constituting and defining the discretionary action space that policy-makers have at their disposal at a certain moment in time when dealing with new challenges, then they need to have some knowledge and understanding of where they came from. True, policy-makers may also feel a bit desperate when reading Klein’s books, ‘haven’t we been here before’ might be one of the rhetorical reactions of those who wish to revolt against any new reforms (e.g. Hirschman, Reference Hirschman1991)? But collective memories are important in solving dilemmas of collective action. They not only remind us of where we came from, they also remind us of what we consider to be important values that we should care for and that we (wish to) share and sustain with other members of the collectivity. Collective memories contribute to the establishment of moral and ethical norms in defense of universal and impartial political institutions, such as the NHS. Where those shared norms and morals are forgotten, social capital will wither and as a consequence, we may end up in a zero-sum game when confronted with new challenges (Rothstein, Reference Rothstein2005: 148).

In this review, the aim is not to analyze Klein’s book in terms of its practical contribution to NHS policy-making, that is, its contribution to problem-oriented policy research and knowledge for policy-making. Instead, I wish to reflect on his contribution to our knowledge of policy-making, that is, interaction-oriented policy research; a discipline that is now central to the field of institutional policy analysis (Scharpf, Reference Scharpf1997). If I would have to choose a disciplinary label for the work of Rudolf Klein, I would position it in the field of political sociology (which are actually two labels). It is not difficult to identify many important insights from political sociology in his work, although Klein himself is never explicit about this. His book is full with theoretical notions but they serve the heuristic goal of reconstructing the developing policy logics behind the evolution of the NHS and by doing so, they help us to better understand these developments, to make sense of what is going on: ‘Verstehen’ as the German sociologist Max Weber called it (Weber, Reference Weber1968). His book is full with fascinating theoretical insights, but since Klein dislikes scholarly niggling (Klein, Reference Klein2013: vii), these are often written between the lines. For a student who is interested in the analysis of policy-making and who hopes to find a repertoire of theoretical ideas and causal mechanisms, his book might at first appear to be treacherously simple, but after closer reading, the student might feel as equally desperate as the policy-maker who aims to find practical solutions for the problems at hand. The analysis turns out to be much more complex and therefore difficult to replicate.

Making sense of policy-making and political complexities is not an easy undertaking. One needs to reconstruct an evolutionary logic, if there is a logic, out of many seemingly contradictionary and paradoxical developments. Or, in the words of Klein himself, how the history of the NHS is one of conflict contained in consensus (Klein, Reference Klein2013: 317). Enter institutions (to paraphrase an expression that Klein often uses in his book)! Klein maintains that the constituting institutional parameters of different health care systems still matter: “Inherited institutions define the landscape within which those policy experiments take place: perceived problems and possible solutions will be different in a tax-funded centralized system like the NHS from those in a social insurance based, pluralistic system like Germany’s or in a system which is considered to be underfunded from one which is seen as overspending. But while institutions may constrain policy choices, they leave space for autonomous decision-making by policy-makers: were it not so, there would be no scope for political debate about the course of action to be taken and policy would be on automatic pilot”(Klein, Reference Klein2013: 309). Institutional stability at one level goes hand in hand with change at another level. The crux for explaining institutional change has to be found in the evolutionary details. Or, in the words of Hirschman and Lindblom: “Detailed descriptions of types of incremental meandering would also be interesting; perhaps this would more clearly differentiate between a sequence that lead to reform and another that leads to revolution” (Hirschman and Lindblom, Reference Hirschman and Lindblom1962: 221).

From a methodological point of view, the problem is that we have many different variables and only a few comparable cases. In other words, there are only a few empirical settings that have a more or less comparable configuration of institutions. What matters even more are the interaction effects of the high number of different constellations of institutional and non-institutional contingencies that are important for understanding the evolution of policy in a particular empirical domain. We necessarily have to deal with multi-causal complexity. Therefore, instead of relying on forward-looking hypotheses in which we isolate one particular independent variable, it is better to take the opposite road of a backward-looking research design (Scharpf, Reference Scharpf1997: 25). We start with a remarkable outcome (the dependent variable: the new politics of the NHS) and then go all the way back in order to identify the relevant combinations of explanatory independent and intervening variables. Needless to say that we need to stop somewhere in our search for founding moments and this is most often an arbitrary decision. For Klein, however, the creation of the NHS in 1948, with some of its pre-war history, was a logical point to start because his object of study was the politics of the NHS, not the history of medicine. Moreover, given the differences between NHS systems in the United Kingdom, especially after devolution in the 1990s, he has chosen to concentrate on the English NHS.

What follows in Klein’s book is what I consider to be one of the best pieces of case-analytical description. Political scientists that have chosen a backward-looking research design often opt for historically descriptive studies of individual cases in order to represent the full complexity of the particular case in-depth, but this is extremely difficult to do. The value of this work lies in the unraveling of the relational logic of connected, but often loosely coupled, events and trajectories, and the reconstruction of how precisely one thing led to another thing over a long period of time. One may identify separate episodes, as Klein does in his book, but the serial analysis of how one episode leads to another episode has to be accompanied by an analysis of parallel developments in one particular episode and the spillover effects of these serial and parallel developments. Hence, the descriptive analyst has to juggle with many different empirical storylines. What makes this analytical work even more difficult is that there is no standard in helping the analyst to assess the value of his claims. In his book Games Real Actors Play. Actor-Centered Institutionalism in Policy Research, published in 1997, the German political sociologist Fritz Scharpf writes (I quote him at length): “the insights obtained in this fashion are not ‘self-correcting’ (as would be true of quantitatively testable hypotheses) but rather depend very much on the personal capabilities and visions of the researchers who conduct this sort of study – not only on their diligence and resourcefulness in data collection but even more on their intuitive or experience-based understanding of how things may, and may not, happen in that particular province of reality, on their detective skills in constructing a coherent story that accounts for the available evidence, on their intellectual honesty in exploring alternative interpretations, and last but by no means least, on their literally skills in communicating what they have learned” (Scharpf, Reference Scharpf1997: 28). I could not have written better praise for the work of Rudolf Klein, who possesses all the capabilities that according to Scharpf are needed when undertaking such a piece of case-descriptive analytical work.

In the prefaces of his later editions, Klein states that he has not tried to re-write the past with the benefit of hindsight simply because he does not feel that it is necessary to modify his earlier interpretations of events, nor does he want to engage in scholarly niggles (Klein, Reference Klein2013: vii). Having written seven editions of his book, he has reason to be confident about his earlier interpretations. Although his perspective shifted somewhat when new reforms were added to the history of the NHS, he sticks to his original assumption that the NHS can be seen as a laboratory for a whole range of social, institutional and organizational experiments with implications for other policy areas and perhaps other countries as well. Klein adopts the model of the NHS as a quasi-experiment, a laboratory for social policy, policy as a learning process. Policy-making is a practice of collective puzzle-solving: a quasi-experimental strategy, with policy-makers learning from both success and failure. Think of the way New Public Management has changed the operation of the NHS or how Enthoven’s model of management competition has been reshaped and implemented in the NHS and the quasi-market model that evolved out of this, ending up in the competition-and-choice policies of New Labour. Klein describes how Enthoven was disappointed when he returned to Britain to see what had come of his ideas. The internal market that was created was, according to Enthoven, the work of politicians in a hurry, more concerned with the election cycle and blame avoidance than with building the foundations for long-lasting evolutionary change (Klein, Reference Klein2013: 173).

Yet, this is exactly the sort of reality that characterizes the politics of the welfare state, especially within the context of health politics in the British Westminster model. Policy learning takes place with the context of politics: enter politics! What policy-makers learn, and the implications of the lessons they draw, depends on who they are: the lenses through which they see the world (Klein, Reference Klein2013: 310). Two decades later, NHS policy-makers were eager to learn from the success of Kaiser Permanente, another US role model, and again, although it was admitted that NHS policy-makers could ‘learn’ from the integrated approach of the Kaiser model, it was less clear how the Kaiser model could be translated into the very different context of the NHS. Finally, lessons are not only ambiguous, they are also used selectively by Ministers and other policy-makers, simply because some lessons fit in better within their ideological frames than others or because different moments in the political election cycle ask for different evidence. This is probably even more so in the polarized political setting of the British Westminster model than in the more consensus-oriented setting of the consociational democracy (with a long tradition of coalition governments) in my own country (The Netherlands). Hence, Klein’s analysis of the new politics and policies of the NHS is as much about puzzling as it is about powering. Nevertheless, by continuing his analysis over time, Klein is able to unravel the ideational transformations that take place at a much slower pace and that model-makers like Alain Enthoven tend to neglect, although these same model-makers do play a role in this process.

If Klein would be willing to engage a little bit into scholarly ‘niggling’, then he would perhaps have used the conceptual language of the new literature on gradual institutional change to describe the incremental transformations that he has been looking for in the NHS. Only recently, scholars have begun to explore and conceptualize the often far more gradual character of institutional reform and the transformative character of incremental change (Hacker Reference Hacker2002, Reference Hacker2005; Thelen Reference Thelen2004; Streeck and Thelen, Reference Streeck and Thelen2005; Mahoney and Thelen, Reference Mahoney and Thelen2010; Helderman and Stiller, Reference Helderman and Stiller2014). Klein agrees with these authors that ‘drift’, rather than inertia, is the default condition of institutional change (Klein, Reference Klein2013: 309). He could have used ‘layering’ to show how new institutions in the NHS have been created alongside old institutions (enter the regulators!). He could have used ‘conversion’ to show how existing institutions have intentionally been ‘redeployed’ by adding new goals and functions to them. Klein’s book is full with examples of ambiguous situations that allowed for different interpretations and for gradual institutional change within the NHS. His analysis fits perfectly within the approaches and concepts of the gradual institutional change school (an ever-expanding academic industry as he calls it), but he is not interested in scholarly niggling about how to explain institutional change, simply because his work has always been about stability and change (and conflicts contained in consensus) and, because his subject is not the theoretical problem of how to explain institutional change but the new politics of the NHS embedded within a very specific historical social order.

This brings me, very briefly, to the work of a second German political sociologist: Wolfgang Streeck. Since Streeck has built up an impressive oeuvre on corporatist institutions and organized capitalism, his work has always been very important for my understanding of reforms in the Dutch corporatist social health insurance system. Streeck also belongs to the ever-expanding industry of gradual institutional change scholars (e.g. Streeck and Thelen, Reference Streeck and Thelen2005). But in what I consider to be one of his best works, Reforming Capitalism (2009), Streeck stresses that the subject of this book is not institutions, but capitalism. He writes: “[…] the language of institutional analysis, while extensively employed in the construction of the empirical accounts on which theoretical reflections draws, is suggested to be of limited value when it comes to explaining institutional change and identifying its driving causes. This can only be achieved, it is claimed, if institutional analysis is systematically specified with respect to time and place, shifting from a formal to a substantive perspective and from a concern with institutions in general to one with unique features of historical social orders” (Streeck, Reference Streeck2009: 2). And he then continues: “Why institutional change in today’s capitalist political economies proceeds the way it does rather than some other way cannot, I suggest, be explained in terms of an institutional theory as such but only in terms of a theory of capitalism as a substantive, that is, historical social order” (Streeck, Reference Streeck2009: 3). Streeck aimed to write a theory of (organized) capitalism and he succeeded very much in doing so because he decided to give the primacy back to substantive policy analysis. Rudolf Klein never departed from his belief in the strength of substantive policy analysis. Besides his contribution to the collective memory of successive generations of NHS stakeholders, he has given us nothing less than a theory of the politics of the NHS, explaining how the more irreconcilable the policy aims become, the more complicated the institutional structures (and design) of the NHS become. What Rudolf Klein has taught us with his seven editions of The New Politics of the NHS is that there is reason for caution when we expect too much and too fast revolutionary change or when we hope for straightforward linear change and explanations of change. When adhering to an evolutionary perspective with respect to policy substance, the analyst needs to go back and forward to pick up the different storylines, and he needs to reconstruct the logic behind all these lines that lead him and his audience to a continuously moving plot.

For policy-makers and politicians, there is reason for caution as well, when they aim at revolutionary change. If there is one lesson in this book that is of a universal nature, it is that politicians and policy-makers should take better care of maintaining the laboratory that they need for keeping their social policies (and solidarity) on track. Whether we are adherents to the evolutionary perspective of the garage, or believers in intelligent design (the church), we should by now know that evolution and design are not mutually exclusive paradigms in the social sciences, nor in the politics of the NHS. It is a bit of both with a high degree of still muddling, not yet through (Lindblom, Reference Lindblom1979). Klein’s book will perhaps be frustrating for the student who hopes to find a toolbox for doing this sort of analysis. The analytical skills conducted in this book cannot easily be learned, nor can the analysis easily be replicated. It is truly about the art and craft of policy analysis (Wildavsky, Reference Wildavsky1979). But hopefully the student will be as much inspired as I and many of us are by Klein’s work. Every policy domain that we care for and that we value as much as the English value their NHS (although occasionally, they seem to have an uneasy time with it) deserves an academic like Rudolf Klein who reminds us of its crucial values by making sense of complexity.