This thesis consists of three chapters. In the first one we investigate whether the provision of information about research findings regarding effectiveness of policies cause political leaders to enact policy change. To do so, we use two types of experiments to measure elected heads of government: (1) demand for research and (2) policy responses to supply of research. We find that policymakers are willing to pay relatively high amounts to learn the results of impact evaluations, update their posterior over the expected impact of a policy in the “correct” direction if informed of research findings, and pay more for types of studies that subsequently affect their beliefs more. Correspondingly, providing information about research findings indicating positive, cost-effective impact of a policy increases the probability that mayors implement the policy in their own municipality by 10 percentage points. In the second chapter, we study the role that revenue shocks play on government investment in fiscal capacity. Using a differencein-difference event-study design, we analyze local budget adjustments to an exogenous revenue shock in formula transfers to Brazilian municipalities. We find that positive revenue shocks translate into additional spending, while the adjustment after a negative shock depends on local characteristics. On average, municipalities increase tax collection, but this effect disappears in jurisdictions with low-educated mayors, which rather tend to cut expenditures. We show that hiring tax related workers is the main mechanism behind the increase in tax revenues. In the third chapter, we follow the theory of knowledge-based hierarchies to study the internal organization of municipal governments in Brazil. Using detailed matched employer-employee data, we construct bureaucrats’ hierarchies within municipalities and show that the empirical patterns match the theoretical predictions. We then present suggestive evidence linking the organization structure in which bureaucrats operate and public sector productivity

This dissertation assesses the aggregate and distributional effects of the recent transport infrastructure expansion in Rio de Janeiro (Brazil) triggered by 2014 Football World Cup and 2016 Olympic Games. In preparation for the sports events, the city invested more than 4.5 billion dollars in its public transport system, which included the extension of a subway line, the construction of a light-rail system and two BRT corridors that stretch  approximately 108 kilometers. Chapter 1 provides a  escription of new transport infrastructure and estimates its potential effects on commuting times. I compute travel times in the absence of the investments using random forest regression methodology and data from 2011 and 2018 travel times. Estimates suggest that the new infrastructure significantly reduced travel times. The remaining chapters explore two different methodologies to account for the impacts of the transport investments. Chapter 2 explores the timing of announcement and inauguration of new BRT and subway stations in Rio  de Janeiro City to investigate the effects of the expansion of transport  infrastructure on growth and reorganization of economic activity. Firm’s addresses were geocoded to construct a panel data set  hat contains information on number of firms and jobs per 100 meter’s grid cell from 2006 to 2016. Applying a difference-in-differences methodology on this novel data set, I estimate the heterogeneous effects of the transport expansion according to workers’ characteristics and industry. All effects are obtained for eight different distance rings ranging from 250m to 2km. Chapter 3 aims to measure the effects of transportation infrastructure on the city’s wages, productivity and welfare, investigating heterogeneous impacts for high and low skilled workers. To answer these questions, I construct an extensive database for the Rio de Janeiro Metropolitan Area that combines information on residence and employment for each skill group inside each city block. In order to measure general equilibrium effects, I develop a model ofinternal city structure that features heterogeneous workers and production externalities across worker’s skill levels. I estimate structural parameters using generalized method of moments. Finally, I perform contrafactual exercises to assess the impacts of the recent transport infrastructure expansion in Rio de Janeiro using 2018 travel times collected from Google Maps API and travel times computed in the first chapter. Results show that  connecting new areas to the central business district results in lower residential concentration and higher employment  concentration. The improvement of transportation services allows citizens to work in high productivity locations and live in high amenity locations, which leads to higher overall welfare. Nevertheless, benefits are not evenly split. High-skilled workers benefit twice since they have higher benefits from agglomeration and, consequently, they are able to pay for higher residential prices from lower commuting costs. Moreover, areas in the vicinity of the new transport stations saw an increase in economic activity. The bulk of the impact is characterized by small firms, from the commerce and service sectors. Additionally, most of the workforce employed by these firms are low-skilled. 

This thesis consists of three papers examining the impact of history on long-run development processes through the channels of institution and culture. The first paper studies land revenue institutions in colonial India and identifies a multi-channel mechanism through which variations in that institution have long-run consequences for agricultural investment and productivity. The second paper examines the relationship between  various dimensions of cultural diversity and growth in Indian districts using an instrumental variables strategy. These results find the strongest impacts for religious diversity. The significant impact of religious diversity in increasing productivity and reducing poverty may be due to increased emphasis on secular institutions in the face of religious competition. The last paper studies the formation of cultural values as a channel through which development outcomes may be impacted by initial conditions. We find that inherent geographical traits render certain regions more likely to be agricultural, male-dominated societies with a lower propensity to worship female deities, which in turn leads to worse female literacy outcomes.

This thesis discusses the role of agents’ expectations regarding the conduction of monetary and fiscal policies in determining policy outcomes, economic dynamics and the volatilities of macroeconomic variables. The first Chapter shows that accounting for agents’ expectations of a possible regime change has critical effects in the responses of macroeconomic variables to shocks, even if this switch does not materialize itself along the path observed after the shock. Recognizing the possibility of regime switches also have important consequences for the volatilities of endogenous variables, which are higher than those obtained in the linear model and very dependent on the policy parameters chosen by monetary and fiscal authorities in each regime. In the second Chapter, I discuss the role of expectations in determining the depth of a crisis when the economy hits the zero lower bound on nominal interest rates. I show that when analysing the impact of a fiscal stimulus during a zero interest rate episode, there is more than just short-run multipliers. To have larger positive effects on output and inflation, monetary and fiscal policies should last longer than the duration of the shock and be coordinated in their actions. The third Chapter presents a thoughtful evaluation of a fiscal stimulus in terms of the implied welfare losses making clear that it should account not only for the effects of policies on short-run output and inflation, but also for the present discounted value of output and inflation in future periods as well. It also analyses how to obtain the optimal level for the nominal interest rate once the economy gets out of the crisis state, if the monetary authority wants to use the expectations channel to undermine the depth of the crisis.

This thesis is comprised of three chapters. In the first two I analyze how fertility decisions and the presence of young children in the household affect the labor supply of women on the intensive margin and their labor market allocation across different types of work such as formal, informal and self-employment. To do so, I use Brazilian data, motivated by the relative rigidity of the labor legislation in Brazil, the high prevalence of informal work and the scarcity of part-time jobs in the formal sector. In the first chapter I specify a reduced-form, dynamic discrete choice model of sequential labor supply decisions in two margins, labor force participation and work in the formal sector. I use longitudinal household survey data representative of the Brazilian population for the years 2012- 2017 to estimate the model. The results show that, among low-education women, observed fertility is endogenous with respect to the decision to work in the formal sector, and that the presence of children in the household has a significant negative effect on the probability of having a formal job conditional on labor force participation. No such evidence is found for college-educated women. In light of these results, in the following chapter I develop an estimable structural model of life-cycle labor supply and fertility decisions and estimate it using data for the years 2002-2015. I then perform counterfactuals on the estimated model in order to isolate the effect of fertility on female labor market informality and to show the partial equilibrium effects of increasing the availability of part-time work in the formal sector and of increasing the duration of maternity leave. In the third chapter, I present a new approach  to handle the “initial conditions problem” in dynamic binary choice models with individual unobserved heterogeneity. I assess the performance of my method relative to existing approaches using a set of simulation experiments and show that it displays relatively better precision and only slightly worse accuracy.

Esta tese é composta por 3 capítulos. No primeiro capítulo mostro que quando um banco central não é totalmente apoiado financeiramente pelo  tesouro e enfrenta uma restrição de solvência, um aumento no tamanho ou uma mudança na composição de seu balanço pode servir como um mecanismo de compromisso em um cenário de armadilha de liquidez. Em particular, quando a taxa de juros de curto prazo está em zero, operações de mercado aberto do banco central que envolvam compras de títulos de longo prazo podem ajudar a mitigar a deflação e recessão sob um equilíbrio de política discricionária. Usando um modelo simples com produto exógeno, mostramos que uma mudança no balanço do banco central, que aumenta seu tamanho e duração, incentiva o banco central a manter as  taxas de juros baixas no futuro, a fim de evitar perdas e satisfazer a restrição de solvência, aproximando-se de sua política ótima de commitment. No segundo capítulo da tese, eu testo a validade do novo mecanismo desenvolvido no capítulo 1, incorporando um banco central financeiramente independente em um modelo DSGE de média escala baseado em Smets  e Wouters (2007), e calibrando-o para replicar principais características da expansão do tamanho e composição do balanço do Federal Reserve no período pós-2008. Eu observo que os programas QE 2 e 3 geraram efeitos positivos na dinâmica da inflação, mas impacto modesto no hiato do produto. O terceiro capítulo da tese avalia as consequências em termos de bem-estar de regras fiscais simples em um modelo de um pequeno país exportador de commodities com uma parcela da população sem acesso ao mercado financeiro, onde a política fiscal assume a forma de transferências. Uma constatação principal é que as regras orçamentárias equilibradas para as receitas de commodities geralmente superam as regras fiscais mais sofisticadas, em que as receitas de commodities são salvas em um Fundo de Riqueza Soberana. Como os choques nos preços das commodities são  tipicamente altamente persistentes, a renda atual das famílias está próxima de sua renda permanente, tornando as regras orçamentárias equilibradas próximas do ideal. 

This thesis consists of three essays on firm dynamics. The first essay evaluates the effects of supply and demand shocks on firm dynamics and selection in Brazil. We explore the fact that China’s recent growth has led not only to an increase in import competition, but also to higher export demand for commodities, which is especially relevant in developing countries. We find that firms facing greater competition from Chinese imports suffer from an increase in exit probability, while firms in industries benefiting from increased export demand have lower probability of exit. In both cases, these effects are concentrated among smaller firms. In the second article, we describe the relationship between energy misallocation and resource misallocation across manufacturing industries in Brazil, and quantify the extent to which distortions affecting energy use result in output losses at the aggregate level. We find that these two measures of misallocation are positively related across industries, which suggests that energy is an important component of resource allocation efficiency. We show that reallocating resources between firms would result in substantial aggregate output gains. However, capital distortions account for most of the potential gains in manufacturing from reallocating resources between firms. The third essay compares firm life cycle dynamics in manufacturing band services, and finds that life-cycle growth is slower for service firms, even when controlling for initial size. We show that this result arises because of the selection pattern and weaker relationship between productivity and size in service industries. Finally, we assess the role of two potential explanations  for these results: age-related distortions and monitoring costs.

This dissertation assesses policy effects of conservation efforts adopted within the scope of the federal action plan to combat Amazon deforestation in Brazil. Chapter 1 provides a description of key policy changes and surveys the associated effectiveness literature. It finds evidence that supports the action plan’s efficacy in reducing aggregate deforestation levels, but notes that indirect impacts of conservation policies have received little attention.

The remaining chapters explore direct and indirect impacts of action plan policies using a georeferenced ten-year panel dataset to account for spatial dynamics. Chapter 2 tests whether legal territorial protection grants actual protection against advancing deforestation. Using a measure of neighboring clearing activity to capture local deforestation risk, the analysis compares forest clearing outcomes in unprotected and protected territory under equivalent deforestation pressures. The empirical strategy draws on the dataset’s raster structure to mitigate concerns of potentially confounding unobservables via the use of raster cell fixed effects. Results document protection’s efficacy in a high-risk context, with significantly less forest being cleared in protected cells than in unprotected ones. Yet, although protected territory effectively shields vegetation under its domain from advancing deforestation, it appears to deflect clearings to unprotected areas. Protection therefore affects regional forest clearing dynamics, but not the overall level of deforestation. Chapter 3 investigates whether changes in tropical regeneration constituted a spillover effect from law enforcement targeting forest loss. Secondary vegetation was vulnerable during the first decade of the action plan, which neither promoted tropical regeneration nor sought to conserve existing secondary vegetation. Moreover, regeneration remained undetected in satellite-based forest monitoring systems. Still, during this period, the extent of Amazon secondary vegetation increased by nearly 7 million hectares. The final part of this dissertation examines whether law enforcement contributed to this growth, albeit unintentionally. The empirical strategy uses a ten-year cross-sectional difference in observed regeneration outcomes to address the intrinsically time-consuming nature of this phenomenon. Results are shown to be robust to the inclusion of a host of raster cell-level controls, mitigating concerns about omitted variable bias. 

Findings indicate that the intensity of enforcement in a location’s close surroundings is associated with both increased probability of secondary vegetation expansion and increased area of secondary vegetation in that location. This lends support to the hypothesis that environmental offenders, once faced with a higher perceived cost of engaging in illegal deforestation, abandoned the area they were operating in and thereby allowed a natural process of forest regrowth to occur. The spillover effect of enforcement on regeneration appears largest in places that have undergone neither too much nor too little deforestation: in the former, forest clearings and non-forest land use are probably more consolidated, and regrowth is therefore less likely; in the latter, there is still relatively little area for the forest to grow back in.

Counterfactual exercises shed light on the magnitude of this effect. An enhanced satellite-based monitoring system for targeting enforcement would have resulted in nearly 300 thousand additional hectares of secondary vegetation.

This thesis is composed of three papers. The first one studies the relation between monetary policy power and the availability of earmarked loans. To that end, we estimate the responses of sectoral macroeconomic variables to monetary policy shocks identified through sign and equality restrictions in a factor-augmented VAR (FAVAR). We find that monetary policy loses power in sectors with a larger share of earmarked loans among its bank debt. The second paper proposes the introduction of new restrictions to identify monetary policy shocks in SVARs. In particular, besides standard sign restrictions on interest rates and inflation, we propose to add as an identification restriction the inability of monetary policy to have real effects ten years after the shock. This paper presents evidence of the model consistency of this neutrality restriction both for the canonical 3- equation new keynesian model and the Smets and Wouters (2007) model. In a simple empirical application, this paper shows that this restriction may be important to recover real effects of monetary policy. The third paper shows that foreign banks can mitigate informational barriers vis-a-vis private domestic banks by observing their peers’ behavior. Conditional on a loan application being filed by a SME firm, we find that the existence of past loans of this firm with private domestic banks constitute a more important predictor that a loan will be granted by foreign banks in comparison to private domestic banks. Our results are compatible with the view that the higher ability of private domestic banks to access informationally opaque business credit risks makes past loans with these banks a more valuable signal for foreign lenders

This thesis is comprised of three articles. The first two investigate the relationship between monetary policy power and the prevalence of governmental credit (featuring interest rates that are insensitive to the monetary cycle) in the economy. The first shows that the available microeconometric evidence is not necessarily informative about the macroeconomic phenomenon of interest, as this depends not only on the average micro effect but also on external effects that capture general equilibrium forces. As an illustration it shows a simple New-Keynesian model with working capital credit, where (i) the macroeconomic effect is usually significantly weaker than the associated micro effect; and (ii) inflation becomes more responsive to monetary policy when governmental credit is present, as this mitigates the cost-channel. Giving sequence, the second article extends the analysis with a medium-sized DSGE model where governmental credit is used to finance the acquisition of physical capital by firms. Not relying on a cost-channel, it shows that under some calibrations the presence of governmental credit increases monetary policy power over inflation. The model is then estimated to Brazil using Bayesian techniques. The results suggest that monetary policy power is indeed reduced for both output and inflation, but more for the former than for the later, implying a lower sacrifice ratio. The estimated effects are small, nonetheless. The model is also used to show that the subsidized governmental credit is little effective in boosting steady-state aggregate investment as the government is unable to distinguish between investment projects that would or would not be done without subsidies.  Finally, the third article studies to what extent the effects of capital flows on a small open economy's business cycle depend on the type of the inflow (e.g., whether a bond or a stock inflow, a liability or an asset flow). For such it build an open economy New-Keynesian model with financial frictions. In a first general analysis it identifies direct mechanisms through which inflows may have differentiated effects depending or their type. It is shown how these differences have implications for the conduct of sterilized FX interventions. The relevancy of these mechanisms is then examined using a calibrated version of the model, and it is concluded that they are probably of little significance. 

This thesis consists of three papers. In the first one, we study the endogeneity bias on Taylor rule estimation. We show that monetary policy shocks explain only a small fraction of inflation and the output gap. Hence the endogeneity bias is small. We show analytically in the 3-equation NK model that the asymptotic bias is a function of the fraction of the variance of endogenous variables accounted for by monetary policy shocks. We use Monte Carlo methods to show that this result survives in larger DSGE models. In the second paper, we estimate a DSGE model to assess the effects of forward guidance in a framework with endogenous time-varying price of risk. We investigate how forward guidance impacts the term structure of interest rates, and document how different monetary policy news can impact macroeconomic variables. We find that forward guidance, through isolated news shocks, has limited impact on long term rates, and our results suggest that the "forward guidance puzzle" cannot be eliminated even within a framework in which forward guidance has limited impact on long term rates. Finally, in the third paper, we exploit information from changes in the yield curve to identify monetary news shocks in a macro-financial DSGE model constrained by ZLB. We find evidence that the standard macro-financial model is not able to satisfactorily recover monetary policy news shocks.

This thesis contains three essays on culture and early childhood. The first essay studies the effect of ethnic beliefs/preferences on breastfeeding practices in Peru and Bolivia. Comparing the breastfeeding practices of rural-to-urban migrants and their descendants by ethnicity, we find that Aymara indigenous mothers breastfeed longer than non-indigenous ones. The evidence suggests that vertical transmission of breastfeeding beliefs occurs essentially through female family members, and that breastfeeding differences are larger in regions where Indigenous and Non-indigenous people historically lived longer together. The latter is consistent with cultural distinction theory.

The second essay investigates the effect of Chile crece Contigo - ChCC hereinafter -, a national-scale early childhood development policy implemented in Chile. The policy intended to improve children development by enhancing children's family environment and parents’ childcare abilities. Consistent with the policy objectives, we show that ChCC increased parental time investment and children cognitive and non-cognitive skills. Parents spend more time doing activities that foster children skills, such as reading, drawing, singing, visiting the park, etc. This is associated with greater psycho-motor skills in children under two, and greater vocabulary and better external behaviour in children between three to four years of age.

The third essay delves on the effect of ChCC using the human skill formation model of Cunha and Heckman (2006). We estimate a production function of skills for pre- and post-ChCC cohorts, and find that for children under two years of age the improvement in cognitive and non-cognitive skills is not only associated with an increase in parental investment but also an increase in the average marginal product of this variable.

This thesis consists of three essays on human capital formation from gestation to adolescence. The first two essays use the Pelotas 1993 Birth Cohort Study, from Pelotas, Brazil. The introductory essay looks at the relationship between household income at birth, late childhood, and late adolescence and variables that reflect human capital at age 18. Our results show that income at birth, during childhood and adolescence affect human capital formation. The estimate points at highest impacts being felt at childhood and birth years. These are consistent with the existence of family borrowing constraints, which are particularly pervasive in earlier years. 

In the second article we investigate the long-term determinants of human capital, from birth until early adulthood. We use the human skill formation model of Cunha, Heckman and Schennach (2010) to estimate a production function of abilities at birth and at age 11. We follow to measure how these abilities combine to produce human capital outcomes. We find that parental investments have strong effects on all our dimensions of child development and at all ages. Further, we use exogenous shifts in income during pregnancy to show that income shocks can have long lasting effects on child abilities and hence on adult human capital levels. Finally, our results show a high degree of complementarity between parental investments and parental abilities and child abilities.

The third essay investigates Chile Crece Contigo a national-scale early childhood development policy implemented in Chile. The policy intended to improve children development by enhancing children's family environment and parents’ childcare abilities. We estimate a production function of skills for pre- and post-ChCC cohorts, and find improvements in cognitive and non-cognitive skills for children under two years of age. This increase is not only associated with an increase in parental investment but also with an increase in the average marginal product of this variable.

This thesis consists of three essays on crime and justice. The first essay evaluates the impact of legal capacity on violent crimes in Brazil. The district classification elevation increases judicial performance and decreases homicide rate due to the selection of more productive and experienced judges. The second essay assess the impact of women's protection laws on domestic violence. The Maria da Penha law, introduced in Brazil in 2006, reduced significantly the female household homicide rate. The policy effect was concentrated in small municipalities. It was greater in places without women’s police station and with low women’s income and labor market participation, low female educational level and low rate of divorce. The third essay investigates the determinants of judicial productivity. More developed localities in terms of income, inequality, educational level and urbanization concentrate the more productive courts

Esta tese é composta de três artigos em Desenvolvimento Econômico. No primeiro artigo avaliamos o impacto de reduções em três diferentes restrições financeiras sobre o PIB per capita no Brasil. Neste caso, analisamos qual das três fricções financeiras (custo de participação no mercado de crédito, limite de endividamento ou custo de monitoramento) é a mais importante para afetar o PIB per capita no Brasil. Para tanto, utilizamos uma versão do modelo de crescimento neoclássico com agentes heterogêneos e três fricções financeiras. O modelo é calibrado para a economia brasileira em 2009 e fazemos exercícios de simulação onde avaliamos os impactos de reduções nas fricções financeiras. No primeiro exercício, a eliminação do custo de participação no mercado de crédito gera um aumento no PIB per capita de 7%. No segundo exercício, a eliminação do custo de monitoramento gera uma elevação no PIB per capita de 2,4%. No terceiro exercício avaliamos um relaxamento nas restrições de endividamento. Os resultados mostram que a redução dos colaterais no Brasil elevaria o PIB per capita em 12%. Neste contexto, a restrição financeira com maior impacto sobre o PIB per capita, no caso brasileiro, é o limite de endividamento. Por fim, fazemos um exercício de decomposição do crescimento do PIB per capita gerado pela redução das fricções financeiras em termos das margens extensiva e intensiva. No segundo artigo buscamos analisar os efeitos macroeconômicos do aumento do crédito com recursos livres no Brasil no período 2001-2011. Em termos mais específicos, avaliamos os impactos do aumento do crédito sobre o PIB per capita. Para tanto, utilizamos uma versão do modelo de crescimento neoclássico com agentes heterogêneos, restrições de crédito e escolha ocupacional, calibrado para a economia brasileira em 2001 e simulamos, no modelo, o aumento do crédito com recursos livres ocorrido no período. Nossos resultados mostram impactos significativos do aumento do crédito para as firmas sobre o PIB per capita. No exercício realizado, o aumento no crédito com recursos livres para as firmas de 9,5%, em 2001, para 15% do PIB em 2011, gerou um aumento de 1,5% no PIB per capita no período. Além disso, o aumento do crédito também gerou impactos significativos sobre o consumo, a produtividade total dos fatores e sobre o estoque de capital. O terceiro artigo analisa os efeitos do aumento do crédito com recursos livres para as firmas sobre o percentual de empresários na economia brasileira, no período 2004-2008. Para tanto, utilizamos uma versão do modelo de crescimento neoclássico com agentes heterogêneos, restrições de crédito e escolha ocupacional. Na implementação do modelo, utilizamos uma abordagem mista de estimação e calibração para a economia brasileira em 2004. Nós simulamos, no modelo, o aumento do crédito para as firmas, ocorrido no período. Nossos resultados mostram impactos significativos do aumento do crédito para as firmas sobre o percentual de empresários na economia. No exercício realizado, o aumento no crédito com recursos livres para as firmas gerou um aumento de cerca de 5% no percentual de empresários na população total entre 2004 e 2006. Além disso, o aumento no acesso ao crédito elevou o PIB per capita em 2% entre 2004 e 2008. Neste caso, a elevação do crédito explica cerca de 14% da elevação do PIB per capita no período.

In the first chapter, we analyze the impact of a community based preschool program on children’s development and schooling in rural Mozambique. Preschools were randomly assigned to 30 out of 76 eligible communities in the study area. Children who attend preschool experience gains in cognitive development, communication, fine motor skills and socio-emotional skills, scoring 0.37 standard deviations higher on a child development screening test. Preschoolers are 21 percentage points more likely to be enrolled in primary school and are 14.9 percentage points more likely to enroll at the appropriate age. The preschool intervention also had positive spillovers on the schooling of older siblings and labor supply of caregivers. At a cost of about $3 dollars per child per month, these results suggest that community led preschools are a promising policy option for helping children meet their development potential. 

In the second chapter, I study the impacts of the Renda Melhor Jovem Program, an attainment award targeted to disadvantaged secondary students in the state of Rio de Janeiro, Brazil. While high school attainment awards have recently been implemented in many places in Latin America, there is still little evidence on whether they can positively impact students’ outcomes when implemented at scale. By exploiting the phased in expansion of the program, I find that program eligibility substantially decreases dropout and increases test scores and high school completion. The program is particularly effective for boys and students with high risks of dropping out. 

In the third chapter, we analyze an empirical case of loss aversion in public policy. While previous papers have identified evidence of loss aversion in laboratory experiments or in sports such as basketball or golf, it is still unclear whether these findings could generalize to other domains. We test for loss aversion by analyzing teacher reactions to receiving a bonus that is based on a continuous underlying measure of school performance. As there are no sanctions or rewards for teachers or schools and the index of school performance is publicly available, winning or losing the bonus should introduce no new information and, consequently, should generate no reactions by teachers. Consistent with loss aversion, however, we find sizable improvements on student scores at schools failing to receive the bonus by a little margin. We investigate the mechanisms behind this results and we find significant changes in teachers’ pedagogical practices. Finally, contrary to claims that behavioral biases should be attenuated by experienced and informed agents, we find stronger results for experienced principals and for schools with high proportion of teachers who report understanding the teacher bonus rules. 

We consider a new method to conduct counterfactual analysis when one "treated'' unit suffers a shock or an intervention, such as a policy change. The proposed approach is based on the construction of an artificial counterfactual from a pool of "untreated'' peers, and is inspired by different branches of the literature such as: the Synthetic Control method, the Global Vector Autoregressive models, the econometrics of structural breaks, and the counterfactual analysis based on macro-econometric and panel data models. 
We analyze the asymptotic properties of the artificial contrafactual constructed either via a conditional expectation model or conditional quantile model. In both cases we derive an asymptotic Gaussian estimator for the average effect (quantile effect ) of the intervention and present a collection of companion hypothesis tests. We also discuss finite sample properties and conduct a detailed Monte Carlo experiment.
Finally we consider the implications of applying the our framework to integrated process of order 1. We found that for the case without drift where at least one cointegration relation exists we have a square root consistent estimator for the intervention effect albeit with a non-standard distribution. Whereas without a cointegration relation (spurious) the estimator diverges. For the case when the drift is present the estimator for both cases (cointegrated and spurious) either diverges or is not well define asymptotically. As a recommendation we suggest to work in first diference reducing the estimation to the stationary case.

This thesis is composed of three papers, the first in organizational economics and culture; the last two in political economics. In the first chapter, we model the relation between dissemination of social norms of work ethic and incentives proposed by firms, which we motivate using evidence from three different datasets. In our model, when the effort of workers in a firm is complementary, high coworker's effort increases an agent's productivity, and consequently firms (endogenously) choose high-powered incentives, demanding more effort. Predicting this, parents prefer to transmit a cultural trait that favors effort in societies where firms choose high-powered incentives, because the dissemination of work ethic is wider, producing (dynamic) multiple equilibria.
In the second chapter, we examine whether neighbors' income affects voting, using data of election results for 2004-2012 Presidential Elections in Unites Stades, by precinct and block group. That way, we try to contribute to understanding the reason why there are different demands for income redistribution. As an identification strategy, we use year fixed-effects and tract year dummies; tract is the smallest geographic unit larger than block group (on average, each tract contains 4 block groups). Our hypothesis is that, within a tract and electoral cycle, living decisions are essentially random. We find that rich voters vote more progressively when living closer to the poor, while poor voters vote more conservatively when closer to the rich. We discuss theoretical explanations for these results.
In the third chapter, we study patronage, investigating the effect of a mayoral candidate's victory on the probability that members of his party (or parties in the same presidential coalition) occupy public jobs in the government, or on their income accrued from government, in case they are already public employees. We also analyze the effect of a party's victory over the number of registered members of that party in future years, which would indicate that voters affiliate to political parties as a way to signal support to the office holder. We estimate plausibly causal effects of a party holding mayoral position by comparing municipalities where that party nearly won with places where it nearly lost. We find no evidence of patronage or of support signaling in our database, which covers the 2008 election in Brazilian Northeast.

This thesis consists of three chapters. In the first one, we study how countries choose the exchange rate regime to adopt, specifically to fix or to floa. Using a model in which the policymaker is in constant learning about the regimes impacting growth and inflation in other countries, we find intuitive connections between the arrangement's choices and related economic variables. For instance, low international reserves encourage the adoption of floating regime. Furthermore, policymakers are more certain about the information coming from geographically closer countries. In the second one, we model how consumers form inflation expectations. Every period, the consumer weighs his expectation between the forecast from professionals and from its own model. For the latter, we assume an adaptive learning approach, which is essential to deal with the inflationary memory present on consumers' behavior. The data generated by the model fits well the observed data. Moreover, we find evidence that the more the consumer absorbs news about inflation, the more he weights on the inflation prediction made by professionals. Finally, in the third chapter, we explore the model proposed in the second one to investigate the relationship between central bank transparency and predictability of inflation of professionals and consumers. Using a panel data, we find evidence that a higher degree of transparency of the central bank improves inflation forecast of consumers, but does not affect the forecast of professionals.

Esta tese é composta por três artigos. O primeiro deles estima a elasticidade preço do consumidor residencial de energia elétrica no Brasil, explorando as regras regulatórias do setor. As datas em que ocorrem os reajustes e as revisões tarifárias e o custo de geração de energia são utilizados como instrumentos para a tarifa de fornecimento. O segundo artigo estuda o comportamento do consumidor industrial de energia elétrica. Diferente do consumidor residencial, a maioria dos consumidores industriais é atendida em níveis de fornecimento de média e alta tensão e, assim, é faturada pela distribuidora pelo consumo de energia elétrica e pela potência máxima disponibilizada. O modelo proposto nesse artigo resolve o problema de maximização da utilidade gerada pelo uso da energia elétrica descontada o custo faturado pela distribuidora em função do consumo mensal e da potência máxima contratada e verificada. De acordo com o modelo, o cliente industrial não toma a decisão de consumo de energia e potência de forma separada como dois bens com preços diferentes. A sensibilidade da demanda em relação a uma variação no preço médio de eletricidade para os consumidores industriais cativos é calculada através da combinação dos dois preços existentes na fatura. O terceiro artigo investiga se as tarifas de energia elétrica são socialmente ótimas utilizando como referência o modelo de Ramsey. As elasticidades preço para os três maiores grupos de consumidores, residencial, comercial e industrial são combinadas com o custo da distribuidora para simular os preços de Ramsey. Os resultados indicam que uma diminuição na tarifa para o segmento industrial e um aumento para o residencial e comercial aumentariam o bem-estar social.