Cette thèse analyse l'effet de la législation de protection de l'emploi sur le taux de chômage, les salaires et la productivité des entreprises. En particulier, je m'intéresse dans cette thèse à l’effet de la réglementation des licenciements et des contrats de travail temporaires. Cette question de recherche est motivée par le fait que dans de nombreux pays de l’OCDE, la législation combine des coûts de licenciements élevés et des restrictions faibles sur les contrats temporaires, ce qui entraîne, d’après un certain nombre d’économistes, une segmentation du marché du travail. Le premier chapitre défend l’idée qu’il est important de comprendre les mécanismes qui expliquent le choix des entreprises de signer des contrats temporaires ou permanents afin d'évaluer l’effet de la protection de l’emploi. Ce chapitre analyse un problème de contrat dynamique entre un travailleur averse au risque et un employeur neutre vis-à-vis du risque. Dans ce chapitre, je soutiens notamment que le choix du type d'emploi est déterminé par un arbitrage entre les gains associés au partage du risque qu’offre un emploi permanent et les gains associés à la flexibilité qu’offre un emploi temporaire. Le deuxième chapitre construit un modèle du marché du travail caractérisé par des frictions de recherche et d’appariement, dans lequel le contrat dynamique du chapitre 1 est intégré. Je propose ainsi un modèle dans lequel l’allocation des agents au sein des différents types d’emplois est déterminée de façon endogène par des considérations reliées au partage du risque. Le modèle, calibré pour reproduire les caractéristiques du marché du travail en France durant les années 2000, suggère que les contrats temporaires ont tendance à augmenter la productivité des entreprises mais également le taux de chômage. Le dernier chapitre propose un modèle de cycle de vie visant à évaluer les effets des coûts de licenciement sur l’emploi et les salaires en fonction du niveau d’éducation et d’expérience. Le modèle est calibré sur les données d’enquête sur la main d’œuvre en France durant les années 2000. Une série d'expériences contrefactuelles indiquent que les coûts de licenciement ont un effet négatif sur l’emploi, concentré principalement sur les jeunes travailleurs avec un niveau d’éducation faible. En revanche, cet effet semble être négligeable pour les travailleurs avec un niveau d'expérience et d'éducation élevé. / This thesis analyzes the effect of employment protection on labor market outcomes. The thesis focuses on the impact of firing restrictions and the regulation of temporary contracts. In many OECD countries, the employment protection legislation combines high firing restrictions and relatively lax regulation of temporary jobs which is, according to several economists, a source of labor market segmentation. The first chapter argues that analyzing the effect of employment protection requires to understand how economic agents choose between permanent and temporary contracts. This chapter examines a dynamic employment contract between a risk-averse worker and a risk-neutral firm. I argue in this chapter that the choice between a permanent and a temporary contract is driven by a trade-off between efficient risk-sharing and flexibility. The second chapter builds a model of the labor market with search frictions, in which the contracting problem of chapter 1 is embedded. Thus, this chapter proposes a model in which the allocation of agents into permanent and temporary jobs is endogenous to risk-sharing considerations. The model is calibrated to the features of the French labor market during the 2000s and indicates that temporary contracts tend to increase productivity but unemployment as well. The third chapter proposes a life-cycle model to evaluate the effect of firing costs across different experience and education groups. The model is calibrated using a French labor force survey dataset. Policy experiments suggest that firing costs have a negative effect on employment, which is concentrated on low experience and education workers.

Several interesting findings emerged from this study. First, strong labour movements still failed to successfully bargain for employment protections due to resistance from employers to encroachments on their institutionalized managerial prerogatives. Second, governments favoured a policy of abstentionism and acquiescence to the collective-laissez-faire tradition until the critical juncture of the 1960s and 1970s. Third, the increasing power resources of trade unions and a deteriorating socio-economic climate created a window of opportunity for bold government action to improve industrial relations, albeit without the consent of employers, and at first, unions. Fourth, contrary to the liberalizing pressures one would expect to find in an archetypical free market economy, the UK has implemented far more statutory protections than deregulatory reforms. Fifth, in contrast to its traditional non-intervention in industrial relations and reputation for worker-protective regulations, Swedish governments have enacted numerous statutes, both restricting and freeing managerial prerogatives in the hiring and firing process. Sixth, statutory employment protections became an independent set of institutional power resources for unions in the long run, serving their organizational and representational interests in important ways. Seventh, unions and left parties consistently defended and advanced the policy preferences of their core constituencies in secure employment by privileging the job security of regular contracts. Eighth, employers and parties on the right of the political spectrum consistently opposed restrictions on the managerial capacity to hire and fire at will, especially for small businesses. Nineth, to increase flexibility without threatening the stability of regular contracts, reforms over the years had to foster atypical forms of work, creating a regulatory gap between permanent and temporary employment, particularly in Sweden. Tenth, differences exist between job security in the statute books and job security in action, particularly in the UK where this gap pervades all aspects of the unfair dismissal system. These findings suggest employment protection legislation has developed in ways far more complex, dynamic and contradictory than is commonly assumed by prominent theories of comparative political economy. / Dissertation / Doctor of Philosophy (PhD) / This thesis examines how and why employment protection legislation developed in the United Kingdom and Sweden in the ways that it did from its early beginnings to the present period. It hopes to offer answers to questions about the initial impetus for statutory regulation, the number, content and impact of significant legislative changes and the preferences of key stakeholders with material interests in the policymaking process. It does this by drawing on a variety of both primary and secondary source materials, including employment protection databases, parliamentary records and research publications. At the same time, it assesses the explanatory merit of dominant theories in the political economy literature by testing them against voluminous empirical evidence and provides a multi-factorial account to fill the gaps in the existing body of knowledge.