L’itinérance est un domaine à la fois passionnant et exigeant de la politique publique. C’est un domaine nouveau, très complexe, mal défini et mal compris. Du milieu des années 1990 au milieu des années 2000, l’itinérance chronique a augmenté au Canada, et jusqu’ici, tant le gouvernement fédéral que les provinces n’ont pas réussi à la contrer sérieusement. En l’absence d’initiatives de la part du fédéral et des provinces, les groupes locaux de partout au pays se sont unis pour lutter contre ce qui était de plus en plus appelé la crise de l’itinérance. L’ampleur de l’itinérance chronique est très similaire dans les grandes villes du Canada. Confrontés au même problème, les décideurs locaux des quatre plus grandes et plus importantes villes du Canada – Vancouver, Calgary, Toronto et Montréal – se sont unis pour constituer différents modèles de gouvernance de l’itinérance. En d’autres termes, il existe différents pourvoyeurs de protection sociale pour les itinérants chroniques, soit l’autre 1 %, dans chacune de ces villes. Les modèles de gouvernance locale présentent deux différences principales : le rôle du gouvernement local, et la centralisation ou la fragmentation du modèle. À Vancouver et à Toronto, le gouvernement local est très impliqué dans la gestion de l’itinérance et y a fait d’importants investissements politiques et financiers. Tandis qu’à Montréal et à Calgary, le gouvernement local joue un rôle bien moins important. Ensuite, la gouvernance de l’itinérance est centralisée dans un seul organisme ou une seule agence à Calgary et à Toronto, alors qu’elle est divisée en plusieurs intervenants à Vancouver et à Montréal. Je me penche sur ce qui pourrait expliquer cette grande différence entre les modèles de gouvernance de l’itinérance, et j’analyse les conséquences théoriques et pratiques que cela peut avoir sur la protection sociale au Canada. Je conclus que le rôle du gouvernement local dans la coalition gouvernante est déterminé par les pouvoirs des villes en matière de logement et par l’engagement des politiciens locaux pour lutter contre l’itinérance. À Vancouver et à Toronto, il y a soit des pouvoirs importants en matière de logement, soit un engagement politique solide à l’égard de l’itinérance, soit les deux. À Montréal et à Calgary, il y a comparativement moins de pouvoirs formels en matière de logement, et l’engagement politique à l’égard du problème est relativement faible. Dans chaque ville, c’est l’organisation des forces sociales locales qui détermine la fragmentation ou la centralisation de la coalition gouvernante. À Vancouver et à Montréal, les forces locales et sociales sont fortes et organisées, mais elles sont divisées, ce qui fait que la gouvernance de l’itinérance est fragmentée. À Calgary, les forces sociales locales sont dominées par le secteur privé, alors qu’à Toronto, les forces sociales locales sont mal organisées et la Ville est un intervenant fort, et en quelque sorte dominant. Cela explique la centralisation de la gouvernance de l’itinérance dans ces deux villes. Malgré leur engagement et leur créativité, aucun des modèles de gouvernance locale n’a réussi à réduire fortement l’itinérance. Aucun ordre de gouvernement seul ne peut résoudre le problème de l’itinérance, et l’absence du gouvernement fédéral des discussions concernant les politiques en matière de logement et d’itinérance était suffisante pour limiter le succès des initiatives menées à l’échelle locale. Ces deux conclusions à la fois confirment et remettent en question les théories existantes de l’État-providence. D’une part, cela confirme l’argument que l’évolution de l’État providence est le reflet l’évolution du fédéralisme, et qu’il y a de plus en plus un nouveau concept du capital social et humain en politique sociale. D’autre part, toutefois, il met au défi ces écrits, en soulignant le rôle que joue le niveau local dans la production de la protection sociale. Les études sur l’itinérance et l’État providence devraient accorder une attention particulière non seulement aux paliers de gouvernement fédéral et provincial, mais également au niveau local aussi. / Homelessness is a challenging and fascinating area of public policy; it is new, very complex, poorly defined and poorly understood. From the mid-1990s to mid-2000s, chronic homelessness was growing throughout Canada, yet federal and provincial governments failed to respond to it in any meaningful way. In the absence of federal or provincial leadership, local groups across the country have come together to fight against what was increasingly called a crisis of homelessness. The scale of chronic homelessness is very similar in big cities across Canada, yet facing the same problem, local actors in Canada’s four biggest and most important cities – Vancouver, Calgary, Toronto and Montreal – came together to form different models of governance of homelessness. In other words, there are very different producers of social protection for the chronically homeless, the other 1%, in each of these cities. There are two main differences in the local governance models: the role of the local government and the centralization or fragmentation of the model. In Vancouver and Toronto, the local government is highly involved in governing homelessness and has made significant political and financial investments, whereas in Montreal and Calgary the local government plays a much smaller role. Further, the governance of homelessness is centralized in one single body or agency in Calgary and Toronto, whereas it is divided among a number of actors in Vancouver and Montreal. I ask what explains these very different models of governance of homelessness, and I consider the theoretical and practical consequences this has for social protection in Canada. I conclude that the role of the local government in the governing coalition is determined by its housing related powers and the local political commitment to homelessness. In Vancouver and Toronto, there are either significant local housing related powers, a strong political commitment to homelessness, or both. In Montreal and Calgary, there are comparatively few housing related powers and the political commitment to the issue is relatively weak. The fragmentation or centralization of the governing coalition is determined by the organization of local social forces in each city. In Vancouver and Montreal local social forces are strong and organized, but divided, making the governance of homelessness fragmented. In Calgary, local social forces are dominated by the private sector whereas in Toronto, local social forces are poorly organized and the city is a strong and somewhat domineering actor. This explains the centralization of the governance of homelessness in these two cities. Despite their commitment and creativity, none of the local governance models has been successful at significantly reducing homelessness. No one level of government alone can solve homelessness, and the absence of the federal government from policy discussions regarding housing and homelessness has been enough to limit the local level successes. These conclusions both confirm and challenge existing theories of welfare state. On the one hand, it confirms the argument that the evolution of the welfare state has mirrored the evolution of federalism, and that there is increasingly a new human or social capital paradigm of social policy. It challenges this literature, however, by highlighting the role that is played by the local level in the production of social protection. Studies of homelessness and the welfare state should pay careful attention not just to federal and provincial governments, but to the local level as well.

Korea and Japan stand out in the group of OECD countries for their rapid increase in, and high levels of, non-standard employment. The empirical evidence leads us to a two-part puzzle: Why are there so many precarious workers in Korea and Japan? And what are the institutional determinants of such labour market risks? This thesis commences by introducing the concept of 'risk shift', and the fuzzy-set ideal type approach is employed to conduct a comparative study of 18 countries. The labour market risks in Korea and Japan are then compared in an international context with 16 selected OECD countries. Fuzzy-set Qualitative Comparative Analysis is employed to investigate the institutional determinants of labour market risks. It then focuses on the increase in non-standard employement in Korea and Japan. Taiwan is also included as a contrasting case, the study taking an institutional approach employing Comparative Historical Analysis. Chapters employing CHA examine how the different welfare production regimes evolved and how they matter in explaining the high rate of non-standard employment in East Asia. The new risk discussion, the argument on the definition and impact of deindustrialization and lastly theories on East Asian welfare states are revisited in the conclusion of this thesis. Finally, I critically discuss the notion of precarious workers and highlight the centrality of social policy that their organizational configuration affects political culture, the formation of the production system, the structure of the labour market and the kind of risk a country could experience.

In this dissertation I explore the redistributive effects of second-dimensional identity politics in parliamentary democracies. Specifically, I focus on parties’ electoral incentives to manipulate the salience of the territorial-identity cleavage. My main argument is that a greater electoral salience of the second dimension distorts the nature of redistributive outcomes. Although the redistributive effects of second dimensions of political competition have been explored in majoritarian democracies, much less is known about their effects in democracies with proportional representation (PR). The dissertation brings “bad news” in that regard: when the territorial second dimension is salient, it is no longer true that parliamentary democracies with proportional electoral systems redistribute more –which is the prevalent view in the existing literature. In fact, the so called “left-bias” of PR systems vanishes when the territorial-identity cleavage is politically activated. This key insight therefore offers a fundamental qualification to the institutionalism literature, by making an effort to understand the way in which regional diversity interacts with institutions through multidimensional political competition. The dissertation is divided in two parts: one theoretical and one empirical. First, I develop a formal model that illustrates the way in which parties’ second-dimension electoral incentives affect both the electoral stage and the subsequent post-electoral coalition bargaining among parties in national parliaments. The reason is that both right-wing and regionalist parties have incentives to increase the salience of the second dimension at the electoral stage to attract voters, and subsequently the coalition bargaining among parties in parliaments offers new opportunities for legislative coalitions. In the second part of the dissertation, I test the empirical implications at the macro-level, the meso-level and the individual-level. The main empirical results can be summarised as follows. First, I present empirical evidence according to which the legislative salience of the second dimension induces a negative effect on redistribution and a positive effect on the regionalisation of public policy. Second, I provide evidence which shows that both right-wing and regionalist parties strategically increase the electoral salience of the second dimension when they are “losers” on the first dimension. Finally, I illustrate the way in which the salience of the second dimension affects the formation of individual preferences for redistribution. In sum, this dissertation provides new arguments and empirical evidence that demonstrates how second dimensional politics can have profound redistributive consequences in parliamentary democracies.

This thesis contributes to the growing EU-US literature by comparing and explaining the evolution of urban policy in these two federal systems. The thesis begins with a puzzle: after introducing two similar and equally short-lived regeneration schemes—Model Cities (MC) (1967) and URBAN (1994)—the US and the EU followed different paths: the former replaced MC with the durable Community Development Block Grant (CDBG) in 1974, while the latter ended urban policy by ‘mainstreaming’ URBAN in its regional policy in 2006. To solve the puzzle I formulate a two-part argument: first, I explain the similarities between MC and URBAN as resulting from three factors: a favourable political context, holistic urban policy ideas, and centre-periphery mistrust. I then explain subsequent trajectories by looking at the interplay of policy and politico-constitutional institutions. While both MC and URBAN were unable to ‘stick’ because of their inherent weaknesses, the result of their demise depended on the existence of a federal ‘city welfare’ state. In the US, the Housing and Urban Development Department (HUD) embodied this state, and channelled Nixon’s attacks on MC into the creation of the structurally stronger CDBG. In the EU, conversely, DG Regio could not provide a comparable anchor for urban policy: when URBAN was attacked by regions and cities, the DG just reverted to its ‘business as usual’ by mainstreaming the programme. I test my argument with a macro-historical comparison of the two cases and four in-depth city studies—Arlington, VA and Baltimore, MD on the US side, and Bristol, UK and Pescara, Italy on the EU side—aimed at analysing micro-level institutional dynamics. In both parts of the study I use a wide range of sources: secondary and grey literature, statistical sources and, especially, archival material and elite interviews. At both levels of analysis the test confirms my argument.

Au Québec, depuis les 25 dernières années, l’enjeu de la privatisation dans le secteur de la santé revient constamment dans le débat public. Déjà dans les années 1980, lorsque le secteur de la santé a commencé à subir d’importantes pressions, faire participer davantage le privé était présenté comme une voie envisageable. Plus récemment, avec l’adoption de la loi 33 par le gouvernement libéral de Jean Charest, plusieurs groupes ont dénoncé la privatisation en santé. Ce qui frappe lorsque l’on s’intéresse à la privatisation en santé, c’est que plusieurs textes abordant cette question ne définissent pas clairement le concept. En se penchant plus particulièrement sur le cas du Québec, cette recherche vise dans un premier temps à rappeler comment a émergé et progressé l’idée de privatisation en santé. Cette idée est apparue dans les années 1980 alors que les programmes publics de soins de santé ont commencé à exercer d’importantes pressions sur les finances publiques des États ébranlés par la crise économique et qu’au même moment, l’idéologie néolibérale, qui remet en question le rôle de l’État dans la couverture sociale, éclipsait tranquillement le keynésianisme. Une nouvelle manière de gérer les programmes publics de soins de santé s’imposait comme étant la voie à adopter. Le nouveau management public et les techniques qu’il propose, dont la privatisation, sont apparus comme étant une solution à considérer. Ensuite, par le biais d’une revue de la littérature, cette recherche fait une analyse du concept de privatisation, tant sur le plan de la protection sociale en général que sur celui de la santé. Ce faisant, elle contribue à combler le flou conceptuel entourant la privatisation et à la définir de manière systématique. Ainsi, la privatisation dans le secteur de la santé transfère des responsabilités du public vers le privé dans certaines activités soit sur le plan: 1) de la gestion et de l’administration, 2) du financement, 3) de la provision et 4) de la propriété. De plus, la privatisation est un processus de changement et peut être initiée de manière active ou passive. La dernière partie de cette recherche se concentre sur le cas québécois et montre comment la privatisation a progressé dans le domaine de la santé au Québec et comment certains éléments du contexte institutionnel canadien ont influencé le processus de privatisation en santé dans le contexte québécois. Suite à une diminution dans le financement en matière de santé de la part du gouvernement fédéral à partir des années 1980, le gouvernement québécois a privatisé activement des services de santé complémentaires en les désassurant, mais a aussi mis en place la politique du virage ambulatoire qui a entraîné une privatisation passive du système de santé. Par cette politique, une nouvelle tendance dans la provision des soins, consistant à retourner plus rapidement les patients dans leur milieu de vie, s’est dessinée. La Loi canadienne sur la santé qui a déjà freiné la privatisation des soins ne représente pas un obstacle suffisant pour arrêter ce type de privatisation. Finalement, avec l’adoption de la loi 33, suite à l’affaire Chaoulli, le gouvernement du Québec a activement fait une plus grande place au privé dans trois activités du programme public de soins de santé soit dans : l’administration et la gestion, la provision et le financement. / For the last 25 years, the issue of health care privatization has frequently been a subject of public debate in Québec. Already during the 1980s, when the health sector was starting to be under pressure, the use of the private sector was presented as a possible alternative. More recently, with the adoption of bill 33 by the liberal government of Jean Charest, many groups in the civil society have denounced the privatization of health care. What is striking when we study health care privatization is that the concept is rarely clearly defined. Through a closer look at the Quebec case, this research first recalls how the idea of health care privatization has emerged and progressed. This idea appeared during the 1980s when health care public programs started to represent important pressures on the public finance of states that were also experiencing an economic crisis. At the same time, neoliberalism and its questioning of the welfare state was slowly eclipsing keynesianism. A new way of managing health care public programs seemed necessary. New public management and the techniques it proposes, privatization being one of them, appeared to be the solution to consider. Then, through a literature review, this research analyzes the concept of privatization, in the realm of social protection in general but also more specifically in health care. Through this process, the research clarifies the concept of privatization and defines it systematically. Thus, privatization in the health care sector transfers responsibility from the public to the private in different areas. These areas are: 1) management and administration, 2) financing, 3) provision, 4) ownership. Adding to this definition, privatization is a process of change and can be initiated actively or passively. The last part looks more particularly at the Québec case and shows how privatization has evolved in the Québec health care sector. It also highlights how some aspects of the Canadian institutional context have influenced health care privatization in the province. Québec has not been impervious to privatization. After the cuts in health care funding by the federal government that began in the 1980s, the Québec’s government reacted by actively privatizing some complementary health services through de-insurance. It has also put in place the ambulatory shift policy, which has lead to a passive privatization of health care. The ambulatory shift represented a new trend in health care provision that consists in sending patients home more quickly after their hospitalization. The Canadian Health Act, considered as an obstacle to privatization, does not limit this kind of privatization. Finally, with the passing of bill 33, after the Chaoulli case, the Québec’s government has actively opened the door to the private sector in three health care activities: administration and management, provision, and financing.

Thèse numérisée par la Division de la gestion de documents et des archives de l'Université de Montréal.

L’objectif de ce mémoire est d’identifier la nature des changements des politiques sociales adoptés entre 1988 et 1998 en Suède et ainsi de mieux comprendre l’état actuel du modèle suédois, caractérisé dans la littérature par l’institutionnalisation des valeurs d’égalité sociale et d’universalisme. À l’aide d’une grille d’analyse inférée à partir de la typologie des changements de politiques publiques de Peter Hall, l’auteure pose l’hypothèse selon laquelle un changement de paradigme est survenu dans les secteurs de l’éducation obligatoire et de la santé. À l’issue de cette étude, il est démontré que, si le paradigme a effectivement été contesté au cours de la période de réforme étudiée, il est toujours en vigueur à la fin des années 1990. Toutefois, l’effet cumulé des changements d’instruments l’ont miné peu à peu durant les décennies suivantes. Les conséquences de ces changements, devenues des « anomalies », menacent aujourd’hui l’existence du modèle suédois. / The goal of this Master’s thesis is to identify the nature of the social policy reforms in Sweden between 1988 and 1998, in order to understand the current Swedish model better. Using three criteria inferred from Peter Hall’s typology of policy change, the author analyzes policy reforms in the compulsory education and healthcare sectors in order to whether a paradigm shift has occurred or not. The conclusion is negative: the social equality and universalism paradigm was indeed contested during the 1990s, but was still actively promoted by the social-democratic government at the end of the decade. The reforms were thus identified as only changes to policy instruments, characterized as second order changes in the typology. However, the cumulative effect of these changes undermined the Sweden model during the next decades. The consequences of the instrument changes thus became anomalies that now threaten the paradigm.

This thesis explores continuity and change in the developmental welfare approach in Thailand following the 1997 Asian financial crisis. It examines both the exogenous and endogenous forces that generated change as well as both the ‘process’ and the ‘content’ of transformation or responses to the crisis. It uses the One Tambon One Product (OTOP) policy as a case study to explore these changes. The principle research question is: To what extent did the post 1997 crisis policy on social protection in Thailand represent a shift from its existing institutional path of developmental welfarism? Extending from this overarching question are subsidiary questions, which guided the thesis. They include: To what extent did the OTOP policy address the social protection gaps that became apparent in the Asian financial crisis? To what extent did the OTOP policy benefit its target population? The thesis uses historical institutionalism (HI) and the role of ideas as the analytic frameworks in analyzing change. The thesis argues that the exogenous shock of the 1997 financial crisis contributed to some departure from the institutional path of developmental welfarism in Thailand. However, the change did not follow the conventional punctuated equilibrium (PE) model under the HI framework in the sense of moving from one equilibrium to another after an exogenous shock. Rather, the radical change that took place after the exogenous shock was gradual. The new set of institutional arrangement prompted significant ideational and institutional transformations. They involved both intended and unintended consequences of incremental shifts in the forms of ‘layering’ ‘drift’ and ‘conversion’ (Streeck and Thelen, 2005). In addition, the thesis argues that the transformation in Thailand after the 1997 financial crisis lies in an intermediate order of change that is found between shifts in policy instrument and a wholesale ‘paradigm shift’ (Hall, 1993). Here, apart from having introduced a new policy such as OTOP, the Thai government engaged in a broader rethinking of Thailand’s developmental welfare path. Moreover, the study finds that the structure of economic development in a developing country context can both promote and impede social protection, rather than only subordinate the latter. The claim is based on the finding that the expansion of economic policy goals in Thailand supported local development and increasing inclusiveness of the informal sector after the 1997 financial crisis. Finally, the thesis argues that social protection delivery or lack thereof reflects contestation of ideas as well as material interests. Both the state and the policy beneficiaries in the OTOP context pushed for their interests when there were gaps between policy formulation and implementation. As a result, changes occurred both in the policy goals and in who benefited from OTOP.

The collapse of the communist regimes in Central and Eastern Europe in 1989-1990 heralded the beginning of an economic transition from central planning to market economies. The subsequent period was marked by malfunctioning of these countries’ social sectors, including their health care systems, raising serious issues of equity. This thesis examines the impact of the transition period and the introduction of social insurance on equity in health care provision in Bulgaria. Equity in health care is investigated with respect to function - i.e. financing (according to ability to pay) and delivery (according to need) - and outcomes - i.e. health status, income inequality and poverty. Differences in health, health care financing and delivery are explored by income, education, ethnic, employment, marital status, age and sex groups. Furthermore, the thesis outlines the impact of health care provision, in particular social insurance, on poverty and health inequalities. The thesis employs empirical analysis based on household data. Its methodology includes concentration and decomposition analysis, and provides new ways of modelling health care financing and delivery, as well as the link between health and health care delivery. The thesis concludes that social insurance does not provide a uniform means of improving equity and that the root cause of the problem lies in the large proportion of out-of-pocket payments and the rather limited size of the health insurance sector. Inequity in health care provision leads to poverty and untreated illness. The data suggests that there are differences between socio-economic groups as regards their likelihood to seek treatment for their ill health, which result in differences in their health status. The social factors that have impacted the most on health are low education and low income.

Varieties of Capitalism theory predicts that the skill specificity of workers determines their demand for social protection. In this thesis, I test this assumption using a measure of occupational mobility between pre- and post-unemployment, which I apply to European workers in different skill groups as defined by Fleckenstein et al., (2011). Using this measure as an indicator of the portability of workers' skills, I then evaluate whether the lower marketability of human capital investments is associated with greater demand for unemployment protection. The findings demonstrate that whilst this relationship is apparent in certain countries, notably Coordinated Market Economies such as Germany, the assumptions do not apply across institutional settings. Consequently, skill specificity cannot explain variation in attitudes towards unemployment protection policies between countries.