This dissertation examines multiple state and local expenditure categories in the United States to expand understanding of fiscal federalism and spatial interactions among governments. First, the author investigates the relationship between police expenditures and crime rates from a spatial perspective. Both police expenditures and crime rates in one state are found to exhibit a similar pattern to that in neighboring states. Spatial correlation is also detected between police expenditures and crime rates. As police of neighbors in fact deter crime at home, there are positive externalities present among the states. Second, the author conducts new tests on the Leviathan hypothesis, i.e., more competition, smaller government. While cost efficiency is used in place of government size to capture the idea that fiscal decentralization reduces wasteful expenditures, spatial interaction is taken as another measure for decentralization. The hypothesis is supported by some evidence from total, police, highway, and welfare expenditures.

My dissertation consists of three essays focused on identifying the strategic responses of governments and individuals following changes in the tax and transfer system. Two essays contribute to the literature on fiscal competition, focusing on state level polices aimed at redistributing income. A third essay contributes to the literature estimating the responsiveness of individual’s incomes to changing marginal tax rates. A better understanding of these responses contributes to our ability to design an optimal tax and transfer system in a federalist nation. In essay 1 I employ a spatial dynamic approach to investigate interstate welfare competition across multiple policy instruments and across three distinct welfare periods - the AFDC regime, the experimental waiver period leading up to the reform, and the TANF era. Results suggest the strategic setting of welfare policy occurs over multiple dimensions of welfare including the effective benefit level and the effective tax rate applied to recipient's earned income. Furthermore, strategic behavior appears to have increased over time, a finding consistent with a race to the bottom after welfare reform. Another form of interstate competition examined in Essay 3 is the spatial patterns in state level estate tax policy. My examination follows a major reform which greatly altered both the state and federal estate tax landscape. This study develops a model in which a state’s tax base and rate are simultaneously determined. Results indicate a state’s estate tax base is negatively influenced by its own tax rate and positively influenced by the tax rate set in neighboring jurisdictions. A state’s own tax rate is also found to be positively influenced by the tax rates set in neighboring jurisdictions. Last, Essay 2 uses matched panels from the Current Population Survey for survey years 1980-2009 to estimate the elasticity of taxable income (ETI) and how it varies in response to measurement of the tax rate, heterogeneity across education attainment, selection on observables and unobservable, and identification. Substantial variation in the ETI across all key economic and statistical decisions is found.

This dissertation focuses on how distributive politics influences the geographic allocation of federal grants to state and local governments. A secondary focus is the role of social trust in the growth of government. In the first essay, I test the degree to which the earmark ban of 2011 prevented legislators from directing federal competitive grants to their home congressional districts and whether earmarking distorted equality in the distribution of federal grants across demographic groups. I find that earmarking skewed the distribution of federal grants toward wealthy congressional districts and away from poor congressional districts. This is a groundbreaking finding, considering that no literature has addressed the impact of earmarking on economic inequality. In the second essay, I estimate the returns to lobbying for local governments in terms of federal earmarked grants, and I find that local governments in counties with higher levels of income per capita were more likely to engage in lobbying. I also find evidence of a causal link between lobbying and federal earmarks to local governments. Given that local governments in wealthy areas tend to have larger tax bases, which allows them to more easily fund public infrastructure projects, my findings imply that lobbying and earmarking hampered the ability of federal grant programs to promote equality in the distribution of federal funds. The third essay utilizes time series econometrics to examine the relationship between government regulation, spending, interest group activity, and social trust in government.

Ever since the first Industrial Revolution, during which many textile artisans lost their jobs to weaving machines, the relationship between technological progress and unemployment has been explored and examined by researchers and policy makers. Existing empirical research, mostly at the microeconomic level, has presented ambiguous results. Procuring data on 51 U.S. states for a period of 19 years and a large number of controls, this paper studies the employment effect of technological innovations with a novel state-level macroeconomic analysis. Using commercially-supplied Research and Development expenditure as a proxy, this paper finds that although technological innovations have a non-significant effect on employment at the general state level, there are a few factors that determine how well each state’s labor market responds to technological changes. More specifically, non-urbanized, non-tech-savvy, or states with a large number of workers employed in Manufacturing or Accommodation and Food Services industry experience a more severe unemployment effect than the other states. The results also suggest that unemployment rate is more negatively affected by technological innovations during the Obama Administration, compared with the Clinton and Bush Administration. This paper adds to the limited, macroeconomic literature on technological unemployment, and provides policy makers with important implications on how to prepare citizens for the imminent waves of technological changes.

For fragile families, external shocks to parents such as incarceration, illness, abuse, death, and divorce can be enough to push the entire family into homelessness. Using data from the Fragile Families and Child Wellbeing Study, I find statistically significant evidence that paternal incarceration, spousal abuse, and parental illness all increase the probability of childhood homelessness. However, I am unable to find a significant correlation between divorce and child homelessness. These findings imply large external shocks to the father, as well as both major external shocks and daily life habits such as drug abuse in the mother, can increase a child’s chance of homelessness. Additionally, I find that these results have a greater effect when they happen earlier in the child’s life, suggesting that policy directed at supporting fragile families with incarcerated fathers, abused mothers, and sick parents would be effective for decreasing childhood homelessness in the United States.

Why is there significant variation in rent extraction amongst high-income democracies? A large number of political economy investigations into this research question have found that a long period of democratic rule and high per capita income are associated with less rent extraction amongst public policy-makers. However, attempts to explain the residual, yet significant, variation in rent extraction amongst countries that possess both these characteristics have been significantly more circumspect and disputed. The thesis explores how the distribution of policy-making responsibilities between electorally accountable decision-makers (EDD) and their electorally unaccountable (NEDD) public policy-making counterparts, determines the optimal level of rents extracted in any given high-income democracy context. Specifically, the thesis formally models how: (1) variation in the EDD/NEDD ratio, by altering (2) voters’ evaluation of incumbent competency, changes (3) the incentives that policy-makers, wishing to remain in office, have to minimize their short term level of rent extraction in order to signal their competency and hopefully retain office. Given these ‘career concerns’ the theoretical model predicts that an increase or decrease in the EDD/NEDD ratio will be associated with more or less rent extraction. This hypothesis is then tested empirically, primarily using an augmented version of Persson and Tabellini’s (2003) dataset. Specifically, the thesis tests whether (1) the EDD/NEDD ratio can predict variation in rent extraction only amongst high-income democracies; (2) whether voters, and not just elites, use the EDD/NEDD ratio to update their beliefs regarding the determinants of rent extraction; and (3) whether the EDD/NEDD ratio affects the level of rent extraction, once controlling for other institutional variables (Efficacy of Elections) also associated with variation in voter evaluation of incumbents’ competency. Establishing that the EDD/NEDD ratio does robustly predict variation in rent extraction is a significant finding, as it can enable analysts to predict how changes in policy-making contexts may affect the incentive for good governance in this sub-set of countries. However, the results are (1) exploratory in nature, and also (2) contingent on other factors (regime type and institutional variation), meaning that while significant, they cannot be generalized to non-democratic contexts.

Cette thèse est une étude appliquée de la taxation optimale et du risque du revenu en amérique latine, en particulier pour l'argentine, le chili, la colombie et le mexique. En faisant face a des niveaux élevés d'inégalité de revenu, l'impôt sur le revenu est pour l'amérique latine un outil essentiel de redistribution qui jusqu'à présent, n'a pas été pleinement exploité. L'un des objectifs de cette thèse est de voir jusqu'où ces pays sont à leur niveau optimal de taxation afin d'explorer la capacité d'amélioration que ce type de charge peut avoir dans chacun des pays de l'étude. Le risque du revenu est une autre caractéristique importante des économies en développement tels que celles trouvées en amérique latine. Etant donné leur vulnérabilité aux chocs macroéconomiques externes, ces économies ont tendance à être particulièrement volatiles. Dans ces conditions, les individus en amérique latine sont sujets à des revenus plus risques que les individus qui vivent des pays développés. La présence du risque a un effet sur comment les agents répondent aux changements divers de l'économie. C'est pourquoi, l'etude des niveaux de risque et en particulier comment les revenus plus risqués affectent la taxation optimale, est l'un des piliers de cette thèse. Le risque du revenu peut être décomposé en deux éléments, permanents ou transitoires. La dernière partie de cette thèse est consacrée à évaluer dans quelle mesure le risque de revenu présente dans chaque pays est causée par un composant permanente ou une composante temporelle. / This thesis is an applied study of optimal taxation and income risk in latin america, in particular for argentina, chile, colombia and mexico. Having to deal with high levels of income inequality, income taxation is for latin america a key tool of redistribution that until now has not been fully exploited. One of the aims of this thesis is to see how far are these countries from their optimal level of taxation, in order to explore the capacity of improvement that this type of tax may have in each one of the countries in the study. Income risk is another important characteristic of developing economies such as those found in latin america. Given their vulnerability to outside macroeconomic shocks, these economies tend to be particularly volatile. under these conditions, individuals in latin america are prone to riskier incomes than the individuals living in developed countries. The presence of risk has an effect on how the agents respond to divers changes in the economy. That is why, studying the levels of risk, and in particular how the riskier incomes affect optimal taxation, is one of the pillars of this thesis. Income risk may be decomposed into two components, permanent or transitory. The last part of this thesis is devoted to estimate how much of the income risk present in each country is caused by a permanent or a temporal component. Given the lack of panel data for latin american countries, these estimations were performed introducing pseudo-panel techniques into traditional panel data methodologies. Comparisons between pseudo-panel results and panel data results were made when the data was available.

The costs incurred by the community resulting from Western Kentucky University’s location – cost of educating children of university related personnel, cost of services provided by the local governments to the university community, and the lost property tax revenue resulting from the tax exemption of university property – were estimated as were the benefits enjoyed by the community as a result of the University’s location – local expenditures, local employment generated by the university’s presence, the university community’s impact on the local banking system’s credit base, and the occupational license tax receipts levied by the city on the university payroll. The local university related expenditures plus the university community’s local tax expenditures were compared to the university related costs to the community to compute a benefit-cost ratio of 6.99 to 1.0 for the Bowling Green – Warren County community. The expenditure habits of university employees and students were estimated from the responses received to questionnaires administered to university employees and students, while the bulk of the information concerned with the university’s costs to the community was obtained from local university and government officials. A business survey was used to determine the economic base of the local economy and to compute a suitable multiplier for the Warren County area. A business survey provided the needed information to estimate the local income generated by the university community. The local income generated was estimated to be $15,440,000. It was established that Western Kentucky University was the largest single generator of jobs, with 19.1 percent of the available local jobs being directly or indirectly attributed to the University.

Thesis advisor: Hideo Konishi / My doctoral dissertation consists of three chapters on political economy and public economics. The first chapter discusses the effect of media bias on a voting competition. The second chapter focuses on how residents respond to increasing natural disaster risks in a multi-community framework. The third chapter investigates a coalition formation game with congestion effects. In chapter 1, I present a two-party election model with media noise. The media may provide polarized messages instead of those that explain the parties' actual policies. The rational voter relies on the media as an imperfect information source regarding a party's platform. Given this framework, I show that Downsian policy convergence is not valid. Moreover, when a party's ideology is relatively strong and the media bias is significant, one-sided polarization can occur: the party with more imprecise reports may adopt a more extreme strategy, whereas its opponent is more of a centrist in a perfect Bayesian equilibrium. This occurs when one party is misrepresented more often, causing the voter to think that the other party has more incentive to polarize. Therefore, the voter may favor the highly misrepresented party, which gives that party more room to polarize. I also show that parties never gain from these increasing misrepresentations, and a biased media environment can negatively affect the voter's welfare. My results suggest that the public should pursue a balanced media environment. Global warming and climate change have become increasingly important. In chapter 2, I investigate a local public goods economy using a new element: location-specific risks of disasters. Agents in this economy ``vote with their feet'' by choosing their favorite location as their residential base. In each location, all residents use majority rule to decide the local wealth tax rate and the amount of local public goods provision that can reduce the loss caused by disasters. I show that the equilibrium is wealth stratified if preferences are represented by a homothetic Stone-Geary utility function. Moreover, when disaster risks at a location increase, the population usually moves away from that location and the housing rents consequently decrease. Meanwhile, the housing rents and tax rates increase at the location the residents shift to. Moreover, I use this framework to numerically evaluate two policies: foreign donation and inter-jurisdiction transfer. If developed countries provides subsidies to a location with greater risks in a developing country, wealthier agents in the recipient country may move into the said location and force poorer agents to move out. This effect makes the wealthier the direct beneficiary of the foreign subsidy. Furthermore, I find that the inter-jurisdiction transfer may harm the poorer by rising housing rents. In chapter 3, I consider a coalition-formation problem, in which there is a set of feasible alternatives for each coalition and each player's payoff is affected by the coalition she belongs to and by its chosen alternative. In this chapter, I focus on ``congestion effects'': an agent's payoff goes down as an additional player joins the coalition other things being equal. The equilibrium notion considered is ``stability": a stable allocation (pairs of coalition structure and alternatives chosen by each coalition) is an allocation such that no coalition has an incentive to deviate from it. I find quite robust counterexamples to show that stability may fail to exist even under strong preference conditions such as the intermediate preference property and single peakedness. Nevertheless, I show a sufficient condition for the nonemptiness of stability: congruent-pair solvability. I also provide some results on the ``Nash-like" equilibrium notion. / Thesis (PhD) — Boston College, 2014. / Submitted to: Boston College. Graduate School of Arts and Sciences. / Discipline: Economics.

En France, en 2014, les dépenses de santé s’élèvent à 190 milliards d’euros. Ce montant, qui croît années après années, est financé à hauteur de 76,6% par une assurance sociale, obligatoire et proportionnelle au revenu, prenant en charge l’ensemble des résidents - l’Assurance Maladie ; 13,5 % sont financés par les assurances complémentaires et 8,5 % directement par les ménages sous la forme de restes-à-charge. La relation entre assurances publique et privée caractérise les systèmes mixtes.Au sein des systèmes mixtes, les assurances peuvent se complémenter mais aussi être à l’origine d’interactions inefficaces. Dans une première partie, on étudie théoriquement un système où l’assurance publique peut être complétée par une assurance complémentaire et/ou supplémentaire. Alors qu’il existait une confusion entre ces deux types d’assurance dans la littérature (Petretto, 1999), on trouve des effets parfois opposés entre assurances complémentaire et supplémentaire. L’utilité marginale des plus pauvres étant plus élevée que celle des plus aisés, on trouve qu’en utilisant simplement comme fonction d’utilité de bien-être social la somme des utilités individuelles, le taux d’assurance publique optimal est positivement lié au caractère redistributif de l’assurance.Dans cette première partie, on remarque que le partage entre ce qui est du domaine de l’assurance complémentaire et supplémentaire est fonction de la définition du panier de biens publics. Le second chapitre étudie la composition optimale de ce panier public. Au cœur du choix de critères à employer pour sélectionner les biens remboursés socialement, se pose la possibilité de comparer des préférences, longuement débattue au sein de la théorie du bien-être et formalisée par le théorème d’incompatibilité d’Arrow. Le principe d’équivalent-revenu de Fleurbaey et al. (2013) propose de dépasser cette limite. Ce critère ordinal, défini comme le revenu en pleine santé équivalent au revenu en mauvaise santé, permet des comparaisons interpersonnelles. En adaptant un modèle théorique étudiant la définition d’un panier optimal social (Hoel, 2007) et en utilisant le principe d’équivalent-revenu, on trouve que la présence d’assurance privée modifie le classement des biens devant être couverts socialement et réduit le budget social optimal. Les individus sans assurance privée peuvent alors être dans l’impossibilité d’accéder à des traitements auxquels ils auraient eu accès en l’absence d’un marché de l’assurance privée.Alors que la seconde partie a permis de révéler quels traitements l’Assurance Maladie devrait renoncer à rembourser en priorité dans un contexte de budget limité, le dernier chapitre étudie un marché caractérisé par une très faible participation de l’assurance sociale. Le marché de l’optique est un marché caractérisé par des asymétries d’information et de la différentiation des produits. Au-delà de financer les dépenses de santé, on se demande si les assurances privées sont, en mesure de réduire ces défaillances de marché et de maitriser les dépenses de santé. S’inscrivant dans la littérature sur le managed-care et la concurrence pour le droit de servir une demande, la dernière partie étudie l’impact des réseaux mis en place par les assurances privées sur le marché de l’optique pour diminuer les dépenses. A partir d’une base de données exclusive assemblée par l’auteur sur un réseau d’opticiens conventionnés mis en place par la mutuelle de santé de l’Éducation Nationale (MGEN), on teste empiriquement l’effet du réseau sur le nombre de ventes et les prix exercés. L’effet de la concurrence pour le réseau et dans le marché sur les prix en verres unifocaux et bifocaux est estimé au sein de 450 regroupements de communes françaises. On constate empiriquement que la concurrence pour le réseau réduit significativement les prix exercés dans le réseau alors que la concurrence dans le marcher affecte les prix à l’extérieur du réseau. / In France, in 2014, health expenditures represented EUR 190 billion. This figure grows year after year; 76.6% is financed by a compulsory social insurance (Assurance Maladie), with contributions proportional to income; 13.5% is financed by private complementary insurances and 8.5% is financed directly by households in the form of out-of-pockets. The relationship between Social Health Insurance (SHI) and Private Health Insurance (PHI) is what characterises a mixed system.Within mixed systems, insurances can complement each other but also interact in inefficient ways. In a first part, I study a system where SHI can be complemented by a complementary or supplementary private insurance. Whereas there was a confusion in the theoretical literature between complementary and supplementary insurances, we find that these insurances can have opposing effects. This model underlines the importance of the nature of the health good (in terms of elasticity) insured by SHI on the optimal rate of social insurance. The higher the rate of low income individual purchasing the socially insured good, the higher the redistributive effect of insurance will be. Marginal utility of poor individuals being higher than high income individuals, I find that using an unweighted additive welfare function, the optimal social insurance rate of insurance is positively related to the redistributive characteristic of insurance.In this first part we underline that the selection of goods that should be insured privately depends on the definition of social insurance. The second part studies what criteria should be used to select the goods to socially insure. At the heart of the selection of goods to socially insure is the possibility of comparing individual preferences debated extensively within welfare economics and formalised by Arrow’s incompatibility theorem. The equivalent income principal developed by Fleurbaey et al. (2013) offers to overcome this limitation. This ordinal criteria, defined as the income in perfect health which yields the same satisfaction as the income in a sick state (i.e. the income in good health minus the willingness to pay to be in good health), allows making interpersonal comparisons. By adapting a theoretical model studying the optimal selection of goods to insure socially (Hoel, 2007) and by using the equivalent income criteria, we find that the introduction of private health insurance decreases the marginal benefit of social insurance. This modifies the ranking function and decreases the optimal social budget, leaving uninsured individuals facing the impossibility to use certain efficient treatments.Whereas the second part revealed what treatment social insurance should first renounce reimbursing in a within a limited budget, the last chapter studies a market characterised by minimal social participation. The market of eyewear (glasses) is characterised by strong asymmetric information and product differentiation. Beyond financing health expenditures, we ask whether, similarly to social insurance, PHI are able to reduce the effects of market failures and manage health expenditures. Following the literature on managed-care and competition for the right to serve a demand, the effect of networks of preferred provides on prices is analysed. Using an exclusive dataset of all purchases in eyewear made by MGEN (Mutuelle Générale de l’Education Nationale) enrollees between 2012 and 2014, we test empirically the effect of the network on the number of purchases and the prices of lenses. The effect of competition for the network and in the market on prices of unifocal and bifocal lenses, within 450 areas of France, is estimated. We find that competition for the network reduces significantly prices of purchases made inside the network and competition in the market reduces prices outside the network.