The imperative of open institutionalism

The overall scope of social reform across the Member States of course is heterogeneous, disparate, and uneven. On balance, however, I observe trajectories of welfare reform in many countries that are becoming proactive and reconstructive, rather than purely defensive and deconstructive. Alongside serious retrenchments there have been deliberate attempts to rebuild social programmes and institutions to accommodate policy repertoires to the new economic and social realities of the twenty-first century, roughly in line with social investment priorities, in many advanced European welfare states, in the fields of family services, gender equality, employment policy and labour market regulation, and human capital formation.

Because of the emergence and proliferation of these novel welfare mixes, preceding the global financial crisis, we are in dire need of an alternative analytical perspective that allows for a better and nuanced understanding of these more positive, multidimensional, and dynamic trajectories of social reform in comparison to the biased 'new politics' conjecture of change-resistant social insurance provision. As we observe more profound changes on the output side of the dependent variable of policy responses, we are confronted with a theoretical quest to find a more adequate theoretical perspective for explaining profound social policy reform. The theoretical challenge is twofold. What we are looking for in the first place is a theoretical perspective that is more dynamic, better able to gain leverage on social policy innovation and institutional transformation across time. Second, and most important, such a dynamic perspective should be able to conceptualize policy actors as more open and responsive to adaptive challenges. If policymakers, contrary to received wisdom, do engage in major social reforms, in spite of highly relevant political and institutional obstacles, to the extent of potentially undermining (re-)election chances, what distinguishes them and the institutional setting within which they operate from the purported general case of welfare state inertia?

Institutional inertia is an ordinary part of the post-formative welfare state development. Welfare reform is difficult, but it happens. Paul Pierson's pathbreaking studies on the political incentives of blame avoidance and organized interest opposition continue to be extremely valuable contributions to our understanding of the modern social politics in affluent democracies. But when taken too far, Pierson's conjectures become excessively deterministic, and unlikely to offer much analytical purchase on complex processes of profound post-formative welfare state change. He is not alone. Over the past two decades, institutional analysis has become unnecessarily overgeneralized, with institutions increasingly understood as self-reinforcing equilibria delimiting the parameters of creative political choice (see Crouch, 2005; Streeck andThelen, 2005). According to Paul Pierson (2004: 157), recognizing the inertial quality of institutions is important to the understanding of long-term processes of institutional development. Institutions are subject to change, but multiple sources of inertia suggest that in many circumstances stability prevails. In the first place, institutional structures often constitute 'deep equilibria'. Once multiple policy actors have come to coordinate their collective action strategies around particular institutional arrangements, they often have little incentive to alter their policy strategies unilaterally in favour of revision. Second, and especially relevant to social policy, is that welfare systems are usually made up of portfolios of interlocking, complementary, and mutual reinforcing policy arrangements. Where such institutional complementarities are 'tightly coupled', with each policy arrangement having a positive effect of the operation of neighbouring policy structures, substantial reform in one policy domain is inherently difficult and most likely inefficient, as it would disturb the effective operation in complementary settings (ibid. 162). Finally, there is the imprint of time. When institutions have been in place for a long time, change becomes inherently more difficult. Over time, institutions become increasingly 'locked in' (ibid. 157). This is not to suggest that institutional inertia precludes change, but more that developments unfavourable to institutional reproduction must reach a critical threshold level before reform is possible. A critical juncture of institutional redesign, according to Pierson, will often follow a long build-up of external pressure.

The inputs of the irresistible forces welfare states are facing, together with the outputs of welfare states as immovable objects, are well defined in the 'new' politics literature; their mutual interaction, however, is unduly neglected. Fiscal strain coming from demographic ageing, intensified market competition from low-wage manufacturing around the globe, the removal of national economic boundaries through European integration, and the shift from passive social protection to more active social servicing are practically all understood in regressive terms, ready to undermine the redistributive scope of the welfare state. On the output side, the welfare status quo is preserved through self-serving, blame-avoiding politics, reform-opposing collective action, and institutional inertia per se. There is hardly any feedback effect between outside pressures and policy responses through relevant policy actors.

It is my contention that the past two decades have not been ones merely of straightforward regressive institutional liberalization. On the input side of external challenges, both demographic ageing and greater gender equality should count as important welfare successes. Population ageing, to the extent that it reflects increased life expectancy in good health, through the introduction of old age pensions, is one of the great achievements of post-war welfare state innovation. More recently, gender equality, also pressed for by the EU, liberating women to choose a career and decide under which conditions they form families and raise children, also represents real social progress. Moreover, the effects of social and economic change are never uniform. Pressures of economic internationalization, labour market change, and adverse demography, trigger different reform agendas, but they do not determine policy content or the timing and scope of policy change. Moreover, the term 'permanent austerity', suggesting long-term economic stagnation, is somewhat misleading. With an average growth rate of above 2.3 per cent of GDP per year over the past three decades of alleged austerity, the EU15 economy has nearly doubled. This has been largely sufficient to sustain comprehensive welfare commitments.

On the output side of institutional responses, in the decade leading up to the crisis, the Nordic welfare states have achieved the highest rates of growth and employment participation, while preserving generous unemployment, disability benefits, old age pensions, and expensive active labour market policy and expanding quality family servicing and long-term care. Other European countries, Germany, the Netherlands, the United Kingdom, Ireland, and Spain followed suit with important social reforms in labour markets, retirement policy, and family servicing, supporting women and lone parents' employment, early childhood education and care, and specific forms of labour market regulation and social protection institutions that promote flexible security. These reform experiences suggest forms of welfare reconstruction, which—in conjunction—are difficult to locate on a straight line between the poles of radical retrenchment and reform resistance. Beyond the Scandinavian welfare states, whose social investment profile can be traced back to the 1930s (Morel, Palier, and Palme, 2009), very few countries can be said to have implemented social investment policies in a comprehensive fashion, enough to stem the tide of income polarization, rising in-work poverty, precarious work, and social and economic dualization (Cantillon, 2011).

What is important for the theoretical purpose of this chapter is that many of the above reform examples suggest a more 'open' relationship between adaptive challenges (independent variables) and policy responses (dependent variables), shaped—but not overdetermined—by concrete institutions. There is a fundamental need to understand in a dynamic fashion how 'closed' or 'open' welfare states as comprehensive policy portfolios can be to adaptive challenges. In the 'new politics' of the welfare state literature there is, theoretically, hardly any attempt to study the shifting ways in which concrete policy provisions intersect with key changes in the wider policy environment of the welfare state. All we are presented with is regressive pressures favouring risky retrenchment meeting the defensive political postures of blame-avoiding politics and rent-seeking collective action. A political space for constructive and purposeful collective action, aimed at solving social problems before they fester and grow, is theoretically ruled out. However, mechanisms of institutional self-reinforcement are particularly ill equipped to explain the recent examples of more proactive social policy innovation since the middle of the final decade of the twentieth century.

In this context, it is important here not to conflate institutional resilience and institutional inertia. Institutional inertia describes a tendency to remain in a fixed position or, more generally, the disinclination to move or act, to resist change, unless moved by some overriding external force. In contrast, institutional resilience denotes a propensity to bounce back, to recover strength, to regain balance after being pressed or stretched. There is an element of endurance and elasticity in institutional resilience that a condition of inertia, understood as immobility or resistance to change, completely lacks. Resilience is about absorbing as well as responding to outside pressures, not merely the capacity to resist change by defensively diverting blame or attention. If institutional inertia is aptly described as constraint on change, the mere resistance of adverse conditions, then resilience is best understood as an institutional resource, providing leverage and mobilizing adaptive capacities, which, in turn, enable institutional actors to adjust to outside pressures in order to restore health and regain balance—not by inaction, resistance, playing safe, or diverting attention. When facing adaptive challenges, institutional actors need a holding environment and time to see their condition in a different light, which will allow them to develop and harness their own response strategies. The idea of a 'holding environment' comes from the literature of organizational leadership (see Heifetz, 1994). I see it as an institutional resource that provides protection, regulates stress, mobilizes confidence, renders a sense of purpose, focuses attention, channels conflict, and fosters compromise. It also provides reliable information in times of need, with a view ultimately to generating satisfying policy perspectives and responses. Throughout this book I will deploy the concept of resilience in terms of the institutional capacity to accompany and accomplish change, and institutional inertia, more narrowly, as defensively holding steady.

The theoretical conclusion I draw from the fairly consistent redirection in welfare policy provision in recent decades is that the key insights of the 'new politics' of the post-formative welfare state, have to be complemented with a more 'possibilist' and purposeful view of political behaviour. Relevant policy actors thus seem able to mobilize creative capacities as they try out novel approaches to new social policy problems, including the unintended consequences of existing policy programmes under new social and economic conditions (Beland and Cox, 2011). There is a need for an open institutional theoretical perspective for understanding welfare state change and continuity, which takes seriously social and economic pressures as delineating, but not determining, the direction, scope, and dynamic of social policy development, filtered through institutions and their political support bases. As a constructive view of collective action, however, requiring external inputs—power, mobilization, ideas, coordination, deliberation, concertation, and conflict resolution, it is necessary to reflect on the whereabouts of these inputs and the institutional preconditions of their effectiveness. How should one articulate a coherent theoretical account of policymakers seeking to adapt social programmes to changed environmental conditions, while remaining faithful to the fundamental insight that 'history matters' and existing institutions have a life of their own in politics and policy? Fundamental here is information-feedback between dependent and independent variables, between profound social policy change and adaptive pressures.

What I observe in policymakers engaging in major reforms, despite obvious institutional obstacles and political costs, is a readiness to use information from past performance, and new ideas and expertise, together with the diffusion of inspiring reform successes in various countries. I therefore believe that the dynamics of policy learning should count as potentially important explanatory factors. In other words, evolving cognitive understanding of policy elites, changing beliefs of politicians, and changing normative orientations with respect to issues of social justice more broadly, should be taken seriously as important factors explaining changing welfare states. Historically, embedded actors such as politicians, administrators, and their expert advisers draw lessons from relevant, often past, policy experience that informs current decisionmaking. Ideally, the more sophisticated information feedback routines are, encouraging social learning and effective adaptation to environmental change, the less likely sudden breakdowns, followed by the need for punctuated institutional change, will be.

Of key theoretical importance to a more open institutional approach to postformative welfare state development, to which I ascribe, is that it should allow for reflexivity in actor orientations. We have to bring back the ideational components, both cognitive and normative, of the policy press in our study of contemporary welfare state development (Hemerijck, 2008: 47). Recent changes in social policy can be explained only if cognitive and normative reorientation by policy actors in complex policy environments is taken seriously. It is imperative to rehabilitate the role of actors' reflexive disposition vis-a-vis concrete social policy problems, which theoretically have been sidelined to the profit of institutional self-reinforcement and rather crude material interest-based political strategies shaped by the welfare status quo. There is, however, no need to depart from the basic theoretical tenets of institutional policy analysis.

Social reality is always mediated by actors' perceptions of environmental conditions, the institutions they inhabit, the worldview they adhere to, and the interests they defend (Berger and Luckman, 1966). Institutional policy analysis generally portrays political behaviour as governed by standard routines, accepted ideas, and rules of conduct, specifying relatively stable repertoires of collective action (Cyert and March, 1963). Institutions do not always favour continuity over change. Routines, ideas, and rules can be both resources of stability and vehicles for change. The question becomes when, how, and why routines, ideas, and rules are challenged and become subject to comprehensive change (Olsen, 2009). Political interest is triggered when environmental pressures alter the effectiveness and legitimacy of prevailing policy. When actors encounter indeterminate situations that confront them with conditions they experience as conflicting with institutional conventions and standing rules of procedure, when available standard operating procedures are perceived to be unsatisfactory to solve problems, this can, potentially, unleash a search process aimed at transforming a problem situation into a determinate policy choice to restore or restructure institutional order (Whit- ford, 2002). As such, creativity, invention, and imagination on the part of policy actors, hypothesizing and dramatizing problems, probing new combinations, seeking new instruments, sometimes even articulating new policy objectives, become key ingredients of the policy process. Reflexive actors in vexing problem situations, associated with high levels of uncertainty, are unlikely to continue to abide by standard conceptions of utilitarian rationality based on fixed preferences and stable material interests. In other words, the problem situation thus carries important explanatory weight, suggesting that increasing deficiencies in existing social policy repertoires, as a consequence of demographic ageing, intensified global economic competition, and family and labour market transformations, are not being ignored.

To the extent that welfare state development requires the exercise of public authority and incurs distributive consequences, social policy change and continuity remain inherently political phenomena, shaped by institutions, policy legacies, and power relations. Without the political articulation of these new conditions, social needs and risks, and disenchantment with existing policy routines, policy change will not take place. External and internal pressures shape, but do not determine, the kinds of reform policy actors conceive intellectually and are willing to pursue politically. Very likely, actors wishing to push through reform are willing to confront opponents by suggesting that their resistance is problematic for reasons of effectiveness and fairness. To be ultimately successful, however, change agents will have to build political consensus to gain support for proposed reforms (Stiller, 2010).

Welfare institutions fulfil certain functions, such as macroeconomic stabilization, labour market integration, social insurance, poverty alleviation, and social service provision, but it is political contestation over these functions that (in part) drives policy change in times of fundamental social and economic restructuring. Moreover, policy actors reflect not only on policy problems and their resolution, but also on the causal links between institutions and their power positions. We need a better understanding of how policyrelevant ideas get selected, modified, or ignored depending on constellations of power. Political parties and organized interests, to be sure, continue to matter; however, their roles in social policy change have become more heterogeneous. Reflexivity plays a more critical role when parties are confronted with uncertainty and this is indeed the case when new types of social risks and new cleavages enter the political realm.

Parties and interest groups are miniature political systems in which internal coalitions compete over policy options. In his dissertation, Gerben Korthouwer studied the positions of social and Christian democratic parties with respect to 'old risk' pension and 'new risk' family policy reform in Austria, Germany, and the Netherlands. He found that usually failure to assure participation in coalition governments, rather than electoral failure per se, have led to shifting positions with respect to welfare reform. Failure to participate in government triggered moments of reflection within both social democratic and Christian democratic parties, with the effect of changing social policy orientations, which, in turn, were put into practice after having secured government participation after subsequent elections (Korthouwer, 2010). A similar argument has been advanced on the proliferation of 'new' social pacts in the 1990s in many European countries, as they reveal changing social partnership dispositions and compromises on issues such as wages, pensions, labour market regulation, and employment policy (Avdagic, Rhodes, and Visser, 2011).

Reflexivity on social policy problems is probably a stronger attribute of institutional actors, such as expert committees or bureaucrats, than of political parties or organized interests who have their ideological and interest-based constituencies to cater for. For this reason, empirically, we have to expand our research to a broader range of policy actors, and especially penetrate the interface of expertise and politics, where expert commissions, central planning agencies, strategy units, and international peer review forums, such as the OECD and the European Union, act as important 'idea brokers' and 'bridge builders' to facilitate the deliberation and dissemination of policy relevant ideas (Lindblom, 1978). It is particularly relevant to examine how certain policy recipes travel from intellectual 'production' to ultimate policy 'consumption'. Hence, social learning processes must necessarily have an important place in our understanding of welfare state change.

Close to four decades ago, Hugh Heclo drew our attention to the critical intellectual components of the policy process in the development of modern social policy in his classic study Modern Social Politics in Britain and Sweden (1974). In this book, he portrayed the evolution of the modern welfare state 'from poor relief to universal coverage' as a political learning process by addressing the complex relationship between expert policy analysis and the exercise of power in the policy process. Paul Pierson has always been critical of examining social policy development in mature welfare states under conditions of 'permanent austerity' from a policy learning perspective. Short time horizons, multiple goals, power asymmetries, and unclear accountability, pressed by austerity, do not conjure an image of a 'learning friendly environment' (Pierson, 1994, 2004). I disagree. It is my contention that there is now overwhelming evidence that fundamental rethinking, re-examination, and reappraisal of the European welfare state far better captures the reform momentum of the past two decades than the 'new politics' of change-resistant welfare states (see also Weyland, 2006; Gilardi, 2010; Gilardi, Fuglister, and Luyet, 2009; Sabel and Zeitlin, 2010).