L’objectif de cette thèse est de présenter quatre essais en économie comportementale et expérimentale sur prise de décision dans le risque et l’ambiguïté. Le premier essai présente une synthèse et un point de vue sur la représentativité des résultats expérimentaux en matière de préférences : préférences sociales et préférences concernant le risque et le temps dans les pays développés ainsi que dans les pays en voie de développement. Le deuxième essai explore expérimentalement l’effet du risque et de l’ambiguïté sur le comportement de recherche d’emploi en horizon infini. Les résultats montrent qu’en risque et ambiguïté, les salaires de réservation sont inférieurs aux valeurs théoriques et diminuent au cours du processus de recherche. De même, les sujets se comportent comme des agents neutre à l’ambiguïté. Le troisième et quatrième essai étudient l’effet du contexte social et la corrélation des paiements sur les attitudes face au risque et à l’ambiguïté respectivement dans le domaine de gain, perte et le domaine mixte. Les résultats montrent que l’introduction du contexte social a un effet significatif sur les attitudes face au risque dans les trois domaines. Néanmoins, la corrélation des risques a un effet sur les attitudes face au risque seulement dans le domaine mixte. Les attitudes face à l’ambiguïté varient selon le domaine. De même, la corrélation des paiements diminuent l’aversion à l’ambiguïté. / The objective of this thesis is to present four essays in behavioral and experimental economics on decision-making under risk and ambiguity. The first essay presents a synthesis and a point of view on the representativeness of experimental results regarding individual preferences: social preferences and risk and time preferences, in developed countries as well as in developing countries. The second essay explores experimentally the effect of risk and ambiguity on job search behavior in an infinite horizon. The results show that in risk and ambiguity, reservation wages are lower than the theoretical values and decrease during the search process. Similarly, subjects behave as ambiguity neutral agents. The third and fourth essay examine the effect of the social context and the correlation of payments on attitudes towards risk and ambiguity respectively in gain, loss and mixed domain. The results show that the introduction of the social context has a significant effect on attitudes towards risk in all three domains. Nevertheless, the correlation of risks has an effect on risk attitudes only in the mixed domain. As for ambiguity, ambiguity attitudes vary across domains. The correlation of payments decreases ambiguity aversion.

This study examines how recent graduates with a Bachelor of Science in Media Technology use LinkedIn as a tool for job search and networking. It examines their thoughts around theirbehaviour in relation with how their behaviour can affect their future career opportunities. Thepurpose is to gain a deeper understanding of how LinkedIn is used which is based on the uses and gratifications theory. To examine this an abductive approach has been used, and four semistructured interviews have been done to gather data from the recent graduates. The study shows that recent graduates with a Bachelor of Science in Media Technology use LinkedIn as a tool for job search, but do not use the opportunity to network to its fullest potential. The recent graduates in this study are also passive users who use LinkedIn as a place to gather information that can be used later. This strengthens the earlier studies done in the field that students and recent graduates use LinkedIn passively and do not use the functions “liking”, “commenting”, “sharing” and “posting” to be more visible to the fullest. The recent graduates are aware of the opportunities LinkedIn provide and that it is beneficial for them to be more active there than they are today but find it difficult to know what they would contribute with by being more active.

No description available.

No description available.

No description available.

No description available.

This thesis covers two areas of the labour market not commonly studied in the context of discrimination: potential bias of job seekers against employers based on ethnicity and gender, and discrimination against employment seekers in the context of the unemployment insurance system. Utilizing survey experiments, both studies yield robust null results. Overall, these studies contribute to the understanding of discrimination dynamics in the labour market and welfare systems. Paper I shows that job seekers may not be motivated by discriminatory practices when seeking employment. However, more research is needed, and future work should be focused on natural experiments to prevent limitations similar to those in our study. Paper II highlights the importance of strict legal frameworks and of maintaining rigorous standards in public service delivery to mitigate discriminatory practices. / <p><strong>Funding:</strong> Swedish Unemployment Insurance Inspectorate</p>

Ce manuscrit comporte trois essais qui partagent l’objectif commun d’évaluer l’impact de la confiance en soi sur les décisions des agents économiques à l’aide de la méthode expérimentale. Ce travail se concentre sur trois thèmes relatifs à l’économie comportementale du travail : le travail au noir, la recherche d’emploi et le travail en équipe. Le premier chapitre analyse les biais d’optimisme dans le contexte du travail irrégulier. Ce travail fournit une mesure des biais d’optimisme à travers un processus de décision. Les résultats montrent que les modalités d’annonce du contrôle altèrent la perception du risque : la désignation du nombre d’agents aléatoirement contrôlés tend à encourager l’optimisme des fraudeurs. Le second chapitre étudie comment l’incertitude quant à l’habileté et l’estime que les demandeurs d’emploi ont d’eux-mêmes affectent leurs décisions de recherche. Les résultats montrent qu’en moyenne les agents peu habiles ne modifient pas leur salaire de réserve, alors que les sujets très habiles tendent à diminuer leurs exigences salariales et donc à stopper plus rapidement leur recherche. Cependant, les décisions des agents peu habiles ne sont pas homogènes : les agents peu compétents ont des exigences salariales d’autant plus élevées qu’ils ont une haute estime d’eux-mêmes. Le troisième chapitre vise à évaluer dans quelle mesure l’image que les travailleurs ont d’eux-mêmes conditionne leur choix d’effort lorsqu’ils travaillent en groupe. Les résultats montrent que les agents qui sur évaluent (sous-évaluent) leur habileté exercent plus (moins) d’effort que les sujets qui ont une perception correcte de leurs compétences. Les résultats révèlent également que les individus bénéficient de la sur-confiance de leur partenaire, mais pas de leur propre biais, alors que la sous-confiance détériore le bien-être de tous les membres de l’équipe. / This dissertation contains three essays that estimate the effects of self-confidence on economic agents’ decisions. An experimental approach is used for those contributions. This work is interested in three topics concerning behavioral labor economics: moonlighting, job search and teamwork. The first chapter investigates the existence of optimism biases in the context of irregular work. This essay proposes a measure of optimism biases through a decision process. The results show that the way the monitoring policy is announced deeply affects the perception of the risk at stake: the designation of the number of randomly controlled agents tends to foster the cheats’ optimism. The second chapter studies how the uncertainty on ability and self-esteem of job-seekers affect their search behaviors. The results show that on average the low ability agents’ decisions are not affected by the uncertainty about their ability, whereas the high ability agents tend to decrease their reservation wage and thus to stop their search faster. However, the low ability agents’ decisions are not homogeneous: the higher the worker’s self-esteem is, the higher his reservation wage is. The third chapter aims at estimating how workers’ self-image biases affect effort choices and team production. The results show that the workers who overestimate (underestimate) their ability provide higher (lower) effort levels than the unbiased. The results also reveal that the agents benefit from their partner’s confidence, but not from their own bias. Conversely, the presence of underconfident agents in the team damages the welfare of both teammates.

Analyses about aggregate employment, unemployment, and inactivity rates frequently disregard labor market trends of specific household members, which may explain some puzzles in the relationship between economic activity and labor market participation. The relevance of family approaches of labor supply transcends the aggregate macroeconomic trends and addresses important micro-level analysis concerning the behavior and intrahousehold decisions of members and policy-relevant results. Despite the consensus about the joint labor supply decisions of household members, studies are typically at the individual level and disregard sons and daughters as decision-makers in a family. Therefore, in this thesis, we investigate these questions for Brazil, in two studies. In the first study, we analyze the labor supply decisions of sons daughters aged 14 to 24 years living with their parents, in a reduced form exercise. We collaborate to the empirical literature about intrahousehold impacts of policies, testing whether the minimum wage, which affects the income of parents, impacts the final labor supply decisions of sons and daughters. We also verify if the policy has distinct effects depending on whether the eligible person is the father or the mother, aiming to test the income-pooling hypothesis. Our identification strategy is based on an intention-to-treat approach, and in a differences-in-differences estimator. Another innovation is the use of the PNADC (IBGE) for 2012-2016. We find that the direction and magnitude of the minimum wage effects affecting fathers and mothers, on the labor supply of sons and daughters, depend on who is and how many eligible members there exist in the household: it is negative, whether the eligible person is the mother or the father, and it is positive, whether both are eligible. Therefore, our results strengthen the argument in favor of household approaches, since the income pooling hypothesis does not seem to be valid in this context. In the second study, we investigate how the decisions about labor supply could determine the aggregate results of unemployment and inactivity of the secondary household earners. We develop and estimate a structural household job search model with on-the-job search. We extend Dey and Flinn (2008) to allow for unemployment and inactivity of mothers and sons and daughters who are subject to shocks to their employment and income shocks to fathers. These shocks may determine different search behavior and job acceptance, depending on the other household member\'s labor market situation. The model is estimated using the PME (IBGE) for 2004-2014. We perform counterfactual simulations, and we verify that the decreasing unemployment rate of sons/daughters would not have changed between 2004 and 2014 if the labor market opportunities and conditions of this member remain the same. The unemployment rate of mothers does not alter a lot in this period. The increasing trend in the inactivity of sons/daughters is mostly determined by a decreasing encouragement rate and the increasing dropout rate observed among these members in this period. These exogenous factors that determine the move to or the permanence in the inactivity could be related to the lower cost of education. We conclude that the use of individual job search models to understand aggregate unemployment and inactivity can be misleading, since the household search behavior matters in the labor supply decisions of secondary household earners. / As análises sobre taxas agregadas de emprego, desemprego e inatividade frequentemente ignoram a dinâmica dos membros das famílias no mercado de trabalho, a qual pode explicar puzzles na relação entre atividade econômica e participação no mercado de trabalho. A relevância das abordagens familiares para a oferta de trabalho está nas análises macroeconômicas sobre tendências agregadas, e também em análises microeconômicas do comportamento, decisões intrafamiliares e resultados de políticas. Apesar do consenso sobre as decisões conjuntas de oferta de trabalho dos membros da família, grande parte dos estudos são abordagens individuais e desconsideram filhos jovens como tomadores de decisão. Nesta tese, organizada em dois estudos, investigamos essas questões para o Brasil. No primeiro estudo, analisamos a decisão de ofertar trabalho de jovens entre 14 e 24 anos vivendo com os pais, em um exercício de forma reduzida. A tese colabora com a literatura empírica sobre os efeitos intrafamiliares de políticas, ao testar se o salário mínimo que afeta a renda dos pais impacta a decisão final dos filhos de ofertar trabalho. Também testamos a hipótese de agregação da renda, ao verificar se se a política tem efeitos distintos caso a pessoa elegível na família seja a mãe ou o pai. A estratégia de identificação é baseada em uma abordagem de intention-to-treat, e no uso do estimador de diferenças-em-diferenças. Outra inovação é o uso da PNADC (IBGE) para 2012-2016. Verificamos que a direção e a magnitude dos efeitos do salário mínimo dos pais, na oferta de trabalho dos filhos, dependem de quem é e de quantos são os membros elegíveis na família: o efeito é negativo, se a pessoa elegível é a mãe ou o pai, e é positivo, se ambos são elegíveis. Esses resultados reforçam o argumento em favor das abordagens intrafamiliares, uma vez que a hipótese de income-pooling não parece ser válida neste contexto. No segundo estudo, investigamos como as decisões de oferta de trabalho poderiam determinar os resultados agregados de desemprego e inatividade dos membros secundários. Desenvolvemos e estimamos um modelo estrutural de busca por emprego familiar com on-the-jobsearch. Estendemos o modelo de Dey e Flinn (2008), para permitir desemprego e inatividade de mães e filhos, sujeitos a choques em seus empregos e choques na renda dos pais. Esses choques podem determinar diferentes comportamentos de busca e aceitação de emprego, dependendo da situação do outro membro no mercado de trabalho. O modelo é estimado com a PME (IBGE) para 2004-2014. Realizamos simulações contrafactuais e verificamos que a taxa de desemprego dos filhos, decrescente entre 2004 e 2014, não teria se alterado no período, caso as condições e oportunidades de mercado de trabalho dos filhos tivessem continuado as mesmas de 2004. Já a taxa de desemprego das mães não sofre grandes alterações no período. A tendência crescente na inatividade dos filhos é determinada por uma taxa de encorajamento decrescente e uma taxa de desistência crescente, que refletem fatores exógenos que levam jovens trabalhadores à inatividade. Esses fatores exógenos podem estar relacionados ao menor custo da educação no período. Concluímos que o uso de modelos individuais de busca por emprego para entender o desemprego e a inatividade agregados deve ser desencorajado, pois o comportamento de busca familiar importa para as decisões de oferta de trabalho dos membros secundários da família.

La thèse étudie les liens qui existent entre les recommandations issues de la littérature sur l'assurance chômage optimale et la qualité des emplois repris par les chômeurs. Nous nous intéressons en particulier à une dimension de la qualité de l'emploi qu'est la stabilité des emplois dans un contexte où les contrats de courte voire de très courte durée sont en pleine expansion en France. En utilisant les modèles théoriques de recherche d'emploi et d'appariement, nous analysons la manière dont les caractéristiques de l'assurance chômage affectent la stabilité des emplois repris. La particularité de notre analyse consiste à intégrer la manière dont les employés sont influencés par les paramètres de l'assurance chômage. En effet, si l'on souhaite analyser de manière globale l'impact des paramètres du système d'indemnisation sur l'évolution du taux de chômage, il convient de déterminer comment ces derniers influencent le taux de sortie du chômage (analyse du comportement des demandeurs d'emploi) mais aussi comment ils affectent le taux d'entrée au chômage (analyse du comportement des employés). Pour étudier le comportement des employés nous considérons que ces derniers influencent leur probabilité de conserver leur emploi en fournissant des efforts de rétention d'emploi. Nous montrons notamment que les chômeurs qui quittent rapidement le chômage retrouvent fréquemment des emplois peu stables. Une fois en emploi, ils exercent relativement peu d'efforts pour conserver leur emploi augmentant par là même leur probabilité de retourner rapidement au chômage. L'impact final d'une réduction du montant de l'indemnisation sur l'évolution du chômage est donc indéterminé dès lors que l'on intègre les employés dans l'analyse. / The thesis studies the relationships between the Optimal Unemployment Insurance (UI) literature recommendations and post unemployment job stability. We focus on one particular job quality dimension that is job stability within a context of a huge increase of very short duration job contracts in France since the 2000’s. Using job search and matching frameworks, we analyse how the features of the UI system affect job stability. The particularity of our approach is that we account for employees’ behaviors while the majority of the literature on optimal UI focuses only on jobless workers behaviors. We show notably that job-seekers who leave quickly unemployment tend to find unstable jobs. Once employed they have a greater probability to return to unemployment because the job-retention efforts they exert are not sufficient.