Introduction

In recent decades, social policy analysis has emerged as a central issue in Canadian and international sociology. Interdisciplinary by nature, social policy studies are the sites of crucial scholarly debates about welfare state retrenchment and restructuring. These debates are centred on the following questions: How much has the welfare state changed since the end of the "golden age" of post-war welfare state expansion from the 1950s to the 1970s? Is the "new politics of the welfare state" (Pierson) exclusively about retrenchment and institutional continuity, or is it also about a long-term restructuring of social provisions? What are the obstacles to policy change? What factors facilitate widespread change?

Reflecting the diversity of the current sociological and political literature on social policy reform, this special issue is not aimed at offering definitive answers to these questions. Drawing on the international literature on social policy, the six articles presented here illustrate the fragmented nature of the modern welfare state by focusing on issues as diverse as childcare, decentralization, public pensions, social assistance, and provincial social spending. And while half of the papers discuss the Canadian situation at both the provincial and the federal level, the remaining of the special issue surveys other national cases: the United States, the United Kingdom, and three smaller European societies,Denmark, the Netherlands, and Sweden. Yet the six papers assembled in this special issue all contribute to the abovementioned debate over the respective scope of continuity and change in welfare state development.

In the comparative literature on social policy, the concept of path dependence is at the centre of the debates concerning institutional change. According to institutionalist scholars such as Paul Pierson (2000), welfare states create vested interests and economic incentives that militate against path-departing processes. Although external shocks may radically alter the course of a policy's development, social programs themselves create powerful economic and political constraints (constituencies, transition costs) that complicate reform. Ultimately, these feedback effects from existing social programs favour the enactment of path-dependent changes that seldom depart from existing institutional logics. What is debated in the current social policy literature is the frequency of path-departing processes and the specific conditions under which they may occur.

In 1989 the House of Commons passed a unanimous resolution "to seek to achieve the goal of eliminating poverty among Canadian children by the year 2000." Though often dismissed as so much rhetoric, Jane Jenson makes a persuasive case that this event is a symbolic marker of a real and profound paradigm shift in Canadian social policy. She supports this claim first by describing change over time, characterizing it as a shift from a public policy paradigm in which parents have full responsibility for their children's well-being to one that can be labelled an investing-in-children paradigm, in which responsibility for children's well-being is shared by families and the broader community. Jenson then moves on to account for the change, one which attributes it not only to new social and economic risks but also to the work of a social-learning network, one that provides significant space to actors promoting social knowledge about children and advocating for them as well as state institutions.

Discussing the development of childcare in Manitoba, Susan Prentice deals with the issue of policy change at the provincial level. Surveying recent childcare reforms enacted in that province, she underlines the institutional and ideological continuity of social liberalism there. Simultaneously, Prentice argues that Manitoba's case complicates determinist assumptions of "path dependency" while highlighting the degree to which political mobilization impacts policy outcomes. Policy redesign is a political process, not a mere technocratic logic.

Exploring variations in provincial social spending, Paul Bernard and Sébastien Saint-Arnaud present the results of a cluster analysis of the situation prevailing in the four largest Canadian provinces — Québec, Ontario, Alberta, and British Columbia — in the middle of the 1990s. Amending Esping-Andersen's well-known typology, the authors underline the existence of four welfare regimes before discussing their relevance to the comparative study of provincial social spending. The results of their analysis indicate modest, though important...