ABSTRACT. The impact of the Internet on crime is a priori ambiguous. I use a panel of state-level yearly data on broadband (or high-speed Internet) penetration and criminal activity in the US to empirically estimate this relationship. I tackle the omitted variable problem by instrumenting high-speed Internet penetration with a measure of early-times telephone and cable TV diffusion, multiplied by year indicators. This approach originates from the consideration that two broadband technologies rely on the pre-existing telephone and cable TV networks to transmit data. These technologies accounted for almost the totality of connections at the beginning of the broadband era, and became relatively less prevalent over time. Then, the pre-Internet penetration of telephone and cable TV is arguably correlated with broadband diffusion – and such correlation varies over time, providing a source of exogenous variation in high-speed Internet penetration. Results show a negative and significant relationship between high-speed Internet penetration and total and property crimes. I rely on a theoretical framework derived from the findings of previous literature and interpret this outcome as reflective of a negative time substitution effect: time spent online crowds out alternative activities that would more likely lead to crime.

ABSTRACT. This paper estimates the effect of local labor market conditions on crime in a developing country with high crime rates. Contrary to the previous literature, which has focused exclusively on developed countries with relatively low crime rates, we find that labor market conditions have a strong effect on homicides. We exploit the 1990s trade liberalization in Brazil as a natural experiment generating exogenous shocks to local labor demand. Regions facing more negative shocks experience large relative increases in crime rates in the medium term, but these effects virtually disappear in the long term. This pattern mirrors the labor market responses to the trade shocks. Using the trade liberalization episode to design an instrumental variables strategy, we find that a 10% reduction in expected labor market earnings (employment rate × earnings) leads to a 39% increase in homicide rates. Our results highlight an additional dimension of adjustment costs following trade shocks that has so far been overlooked in the literature.

ABSTRACT. This paper investigates the influence of ambient light on number of homicides by firearms in one of the most violent countries in the world. As exogenous variation in daylight hours we exploit the transition from Standard to Daylight Saving Time (DST) in some but not all Brazilian states to provide both within and between-states comparison around the transition. Using non-parametric regression discontinuity techniques, we find a sharp decrease in the number of homicides by firearms in Brazilian treated states (14.4%), while no statistical relationship is found among states not affected by the policy. The uncovered effect is mostly concentrated on hours directly affected by the shift in daylight. Moreover, comparing treated to untreated states in a Differences-in-Differences approach, we find that homicides by firearms decreased during DST months by about 3%. These estimates imply Daylight Saving Time is responsible for saving about 3,850 potential victims from 2006-2011.

ABSTRACT. This paper assesses the role of education quality in the convergence process of GDP per capita through teachers quality impact in human capital formation. The simple two-period OLG model suggests initial level of teacher's human capital is important to explain non-convergence, even when education quality return is decreasing. This non-convergence arises because an initially low level of teachers' human capital translates into a low level of human capital transfered to students, which means a low level of teachers' human capital in the next period, and so on. It is also shown an education quantity-quality trade-off, despite all dynamics coming from quality evolution. This trade-off helps to explain why developing countries did not reached high GDP levels, despite recent evolution of average years of schooling in these countries. The paper, therefore, provides an alternative explanation for why countries income do not converge, even when differences in other inputs, such as capital stock, are not accounted for.

ABSTRACT. We study the impact of the different stages of human capital accumulation on the evolution of labor productivity in a model calibrated to the U.S. from 1961 to 2008. We add early childhood education to a standard continuous time life cycle economy and assume complementarity between educational stages. There are three sectors in the model: the goods sector, the early childhood sector and the formal education sector. Agents are homogenous and choose the intensity of preschool education, how long to stay in formal school, labor effort and consumption, and there are exogenous distortions to these four decisions. The model matches the data very well and closely reproduces the paths of schooling, hours worked, relative prices and GDP. We find that the reduction in distortions to early education in the period was large and made a very strong contribution to human capital accumulation. However, due to general equilibrium effects of labor market taxation, marginal modification in the incentives for early education in 2008 had a smaller impact than those for formal education. This is because the former do not decisively affect the decision to join the labor market, while the latter do. Without labor taxation, incentives for preschool are significantly stronger.

ABSTRACT. This paper studies the feasibility of implementing paternalistic taxation in the presence of health and human capital accumulation. Early decisions made in life can be present-biased and distort optimal decisions. Agents can incorrectly discount differently future educational choices and health investments. This paper characterizes the optimal linear tax (subsidy) on health and education investments for individuals with heterogeneous misperception of future implications of such decisions. Our results suggest that a paternalistic government must subsidize educational and health investments but tax the consumption of unhealthy goods of present-biased households in the first-best policies. We show that the larger the present bias, the larger the distortion necessary to correct it. Second-best results suggest that the optimal level of distortion of health and human capital investments must capture two terms, namely, a Pigouvian term to correct for the degree of misperception of future benefits of more education and better health and an optimal tax term, which captures the redistributive impact of taxation. We show that the larger the correlation between present-bias and earnings, the lower the distortion on investment decisions, but larger has to be the distortion of agent's time and resources allocation, the larger consumption inequality among individuals. Lastly, a subsidy to human (health) capital accumulation can achieve similar welfare improvement without subsidizing health (education).

ABSTRACT. Financial innovations in recent decades have vastly expanded portfolio choice. We theoretically investigate how greater choice affects investors savings and asset returns, when investors speculate on their heterogeneous beliefs. Under mild assumptions, we establish a choice channel by which greater portfolio choice induces investors to save more. The effect on asset returns depends on the type of financial innovation. Portfolio customization (access to risky assets other than the market portfolio) reduces the risk-free rate and leaves the risk premia unchanged. We also analyze participation (access to the market portfolio) and securitization (ability to issue risk-free debt to purchase risky assets).

ABSTRACT. We analyse bank runs under uncertainty over asset liquidity, and show how the risk of losses upon premature loan termination produces an unique run equilibrium, where inefficient runs occur even under minimal fundamental risk. The model refines the bank run framework by rationalizing actual bankruptcy rules on mandatory stay, such that default occurs before all illiquid assets are sold. Since less liquid assets are not readily available for running depositors, asset liquidity risk has a concave effect on run incentives, unlike fundamental risk. Runs are rare when asset liquidity is abundant, become more frequent as it falls and decrease again under very low asset liquidity. The socially optimal demandable debt contract limits inessential runs by offering a higher rollover yield and improves welfare if asset liquidity is sufficiently high. However, the private choice minimizes funding costs, tolerating more frequent runs.

ABSTRACT. The considerable literature on “gaming” incentive systems highlights an anomaly in the way economists think about the design of  incentive systems such as contracts, organizations, or regulatory regimes. In the standard framework, agents behave as their principals intend; bad outcome signal only bad luck, not badly designed incentives. Empirically, however, principals sometimes behave as though bad outcomes do imply bad design: an old incentive scheme is replaced with something else, which may lead to new bad outcomes, or to satisfactory performance. More strikingly, sometimes schemes that work well but are costly to implement, get replaced by less costly ones which ultimately fail to perform. Similar organizational failures recur across time and space.

We rationalize these  patterns  with a simple social learning model of the evolution via “loopholes" of incentive systems. Young principals, who are uncertain about what actions will be feasible for their agents, get limited observations of earlier agents' behavior, and rationally update their beliefs about feasible actions. When old, they offer contracts that deter only the harmful actions they deem sufficiently likely to exist; their agents then cheat if it is both feasible and undeterred by the contract, i.e., if there is a loophole.

In a loophole equilibrium, principals who observe cheating close loopholes when they offer contracts. But loophole-free contracts deter all cheating, thereby conveying little information about feasible actions to the next generation of principals; they rationally update in favor of the view that cheating is unlikely or infeasible and may therefore offer contracts with loopholes. The result is cycling of contract types that alternately deter and encourage undesired behavior, yielding heterogeneity across principals. There are also ``bureaucratic'' equilibria in which contracts deter behavior that is actually infeasible.

Depending on whether principals sample concurrently or historically, population dynamics in a loophole equilibrium may display aggregate cycling or convergence to a stationary, non-degenerate distribution of contracts or rules.

Some implications for economic development and institutional design are discussed.

ABSTRACT. This paper develops estimation and inference for quantile regression models with measurement errors. We propose an easily-implementable semiparametric two-step estimator when repeated measures for the covariates are available. Building on recent theory on Z-estimation with infinite-dimensional parameters, consistency and asymptotic normality of the proposed estimator are established. We also develop statistical inference procedures and show the validity of a bootstrap approach to implement the methods in practice. Monte Carlo simulations assess the finite-sample performance of the proposed methods. We apply the methods to the well-known example of returns to education on earnings using a data set on female monozygotic twins in the U.K. We document strong heterogeneity in returns to education along the conditional distribution of earnings. The returns are relatively larger at the lower part of the distribution, providing evidence that a potential economic redistributive policy should focus on such quantiles.

ABSTRACT. Dynamic discrete choice models (DDC) are nonparametrically not identified. However, the non-identification of DDC models does not necessarily imply non-identification of their associated counterfactuals. We provide necessary and sufficient conditions for the identification of counterfactual behavior and welfare for a broad class of counterfactuals. The conditions are simple to check and can be applied to virtually all counterfactuals in the DDC literature. To explore how robust counterfactuals can be to model restrictions in practice, we consider a numerical example of a monopolist’s entry problem, as well as an empirical application of agricultural land use. For each case, we provide relevant examples of both identified and on-identified counterfactuals.

ABSTRACT. Na edição Inaf-2015, foi incluído um conjunto de questões que permitem qualificar os respondentes de acordo com três competências socioemocionais: autogestão, autoconceito e abertura ao novo. Graças a essa iniciativa - realizada em parceria com o Instituto Ayrton Senna - temos disponíveis pela primeira vez no País uma base com representatividade nacional sobre essas competências na população adulta. Qual o papel das competências socioemocionais no alcance das realizações individuais? Esse papel difere a depender do contexto de vida? Para quem o socioemocional mais importa? O socioemocional pode contribuir à superação das desigualdades no Brasil? Quais aprendizados podemos extrair para a formulação de políticas públicas? Responder a essas perguntas é o desafio lançado aos especialistas convidados e também um convite endereçado ao público presente.

ABSTRACT. This article analyzes optimal monetary management with exogenous changes in the money supply using a dynamic macroeconomic model with money. Money supply shocks deviate the inflation rate from the target, and a monetary authority evaluates the inflationary losses and cost of intervention to re-establish the equilibrium interest rate. We show that the monetary authority will optimally allow for variation in the nominal interest rate within a small range around the target interest rate, thereby intervening in the monetary market when the nominal interest rate trends toward a point outside that range. We propose an empirical strategy to test the reactions of the Brazilian Central Bank to liquidity shifts caused by changes in the Treasury Single Account (TSA) balance. Different metrics for comparing TSA and money stock changes are created, and the results confirm the predictions of the theoretical model.

ABSTRACT. This paper quanties and compares the macroeconomic effects of mon- etary policy shocks under different fiscal policy rules. We analyze the case in which the rise in the interest rate following a monetary policy shock increases public debt interest payments, thereby making a fiscal adjust- ment necessary to guarantee debt sustainability. The analysis is based on a medium-sized DSGE model developed and calibrated to represent the Brazilian economy, in which the effect of the interest rate on the public debt service tends to be pronounced. The model incorporates a detailed public sector capable of intervening in the economy through several chan- nels. Our simulation results show that the magnitude of the reduction in GDP following a monetary policy shock varies considerably depending on the fiscal rule adopted. In particular, economic performance tends to be better under rules that adopt public employment as a fiscal adjustment tool, and tends to be worse when fiscal adjustment relies on reductions in public investment.

ABSTRACT. Empirical evidence suggests that the magnitude of the negative comovement between real stock returns and inflation declined during the Great Moderation in the U.S. To understand the role of monetary policy credibility in this change, I study optimal monetary policy under loose commitment in a macroeconomic model in which stock price movements have direct implications for business cycles. In line with the data, a calibration of the model featuring a significant degree of credibility can replicate the weakening of the negative relationship between real stock returns and inflation in the Great Moderation era.

ABSTRACT. This paper evaluates the claim that transparency increases the effectiveness of monetary policy. We present a simple New Keynesian model where firms do not observe the time-varying inflation target and the monetary policy shock. Two informational assumptions are considered: (i) Firms observe interest rate decisions only (standard assumption); (ii) firms observe interest rate decisions and an idiosyncratic signal about the inflation target (that may be very noisy). Greater transparency about the target implies in lower costs of disinflation, but it only increases the effectiveness of monetary policy -- in terms of the effects of monetary shocks -- when the second set of assumptions is considered.

We show that rational expectations about, say, inflation target and monetary shocks differ qualitatively under those two set of assumptions in a large class of DSGE models. Moreover, the differences still persist in the limiting case that the idiosyncratic signal is uninformative as long as agents learn from endogenous variables. Key to these differences is the fact that under the second set of assumptions, the ability to extract information from the endogenous variables changes. This is a result that individuals are prevented to infer the aggregate endogenous variables from their own decisions as agents do not know others' decisions. This has important implications to how to incorporate learning in macroeconomics.

ABSTRACT. Do people acquire more information when they are encouraged to participate in elections? This paper presents empirical evidence on the effects of compulsory voting laws on the consumption of TV news. In Brazil, the law determines that every literate citizen over the age of eighteen at the day of the election is subject to a number of penalties if they don't attend the ballots. This provides a natural experiment which allows us to identify the causal effect of being under a compulsory voting regime on information acquisition. Using national survey data on the consumption of media we find that compulsory voting has a significant and substantial positive impact on the probability of an individual to watch Brazil's main newscast. This result is restricted to young voters who just turned eighteen and is robust to different polynomials and bandwidth lengths.

ABSTRACT. In a seminal work about strategic voting, Duverger suggested that single-ballot electoral systems generate incentives for voters to behave strategically, which leads to two-party systems. Dual-ballot electoral systems, in turn, allow voters to behave sincerely, which favors multi-party systems. Even though several works have found results consistent with strategic voting behavior, some findings also suggests that not all voters behave the same way. This leads to the question on who exactly are the voters behaving strategically and, more importantly, why some voters do not behave this way. This paper investigates whether voters with different levels of education and, arguably, different level of access to information, behave differently when choosing their candidates in Brazilian mayoral elections. We construct a simple model to motivate our empirical study where voters are not able to observe society's preferences. They only receive signals about those preferences. In our model the more informative the signal received, the higher the incentive to vote strategically. We test the predictions of our model exploiting an exogenous variation in Brazilian electoral rules and using a unique database with information disaggregated at the polling station level, which allows us to identify votes and characteristics of small groups of voters. Our results show that more educated voters, those that arguably have access to more information, are more likely to vote strategically.

ABSTRACT. Países da América Latina investem relativamente pouco em segurança pública. A incerteza com relação ao retorno político de programas de combate a grupos armados pode ser a chave para explicar os gastos reduzidos. Este trabalho analisa a estratégia de pacificação, iniciada em 2008 no Rio de Janeiro, para estimar a magnitude da recompensa que pode ser obtida nas urnas em função da adoção de uma política bem-sucedida de combate a violência. O projeto de responsabilidade do governo do estado tem o objetivo de retomar o controle de áreas da capital que estão sob o domínio de grupos armados. Na prática, a polícia invade e estabelece bases permanentes de operação nestes territórios. O estudo utiliza dados do Tribunal Superior Eleitoral (TSE). A partir da metodologia de diferenças em diferenças é feita uma comparação entre votos no governador em áreas incluídas no projeto e em territórios não selecionados para participar do programa. Os principais resultados indicam que a pacificação é consistente com um aumento significativo dos votos deste político.

ABSTRACT. We study optimal labor income taxation in non-competitive labor markets. Firms offer screening contracts to workers who have private information about their productivity. A planner designs a tax system aimed at maximizing a Paretian objective. We provide necessary and sufficient conditions for tax implementation of constrained efficient allocations. For a Utilitarian objective, if an allocation is implementable almost all workers face negative marginal tax rates. Not all allocations that are implementable in a competitive setting are implementable in this non-competitive environment. Keywords: Mirrlees’ problem; Non-competitive labor markets

ABSTRACT. This paper studies a committee decision-making problem. Committee members are heterogeneous in their competence, they are biased towards one of the alternatives and career oriented, and they can choose whether to vote or abstain. The interaction between career concern and bias a¤ects the voting behavior of members depending on transparency of individual votes. We show that transparency attenuates the pre-existing biases of competent members and exacerbates the biases of incompetent members. Public voting leads to better decisions when the magnitude of the bias is large, while secret voting performs better otherwise. We provide experimental evidence supporting our theoretical conclusions.

ABSTRACT. In the aftermath of the Taper Tantrum episode, the Central Bank of Brazil announced a major program of intervention in foreign exchange markets on August 2013, with daily sales of foreign exchange (FX) futures settling in domestic currency swaps that provided insurance against a depreciation of the real. We analyze the effect of that program on the level and volatility of the exchange rate using a synthetic control approach. Our counterfactual results suggest that program led to an appreciation of the Brazilian in excess of 10 percent. Some of our estimates also point to a decline in the option-implied volatility. A second announcement extending the program had a more muted effect on the exchange rate. Subsequent extensions had little or no impact, likely because they were already expected and priced-in by the market.

ABSTRACT. The Import Substitution Process in Latin America was an attempt to enhanc e GDP growth and productivity by rising trade barriers up on capital-intensive products. Our main goal is to analyze how an increase in import tariff on a particular type of good affects the production choices and trade pattern of an economy. We develop an extension of the dynamic Heckscher-Ohlin model – a combination of a static two goods, two-factor Heckscher-Ohlin model and a two-sector growth model – allowing for import tariff. We then calibrate the closed economy model to the US. The results show that the economy will produce less of both consumption and investment goods under autarky for low and high levels of capital stock per worker. We also find that total GDP may be lower under free trade in comparison to autarky.

ABSTRACT. A sessão terá como objetivo discutir os principais resultados da avaliação de impacto da eficácia dos modelos de intervenção apoiados pelo BID que apoiam as pequenas e médias empresas no setor industrial no país; bem como, à luz desses resultados, debater os principais desafios e perspectivas do país para aumentar a produtividade de suas empresas.

ABSTRACT. Differences-in-Differences (DID) is one of the most widely used identification strategies in applied economics. However, how to draw inferences in DID models when there are few treated groups remains an open question. We show that the usual inference methods used in DID models might not perform well when there are few treated groups and errors are heteroskedastic. In particular, when there is variation in the number of observations per group, inference methods designed to work when there are few treated groups tend to (under-) over-reject the null hypothesis when the treated groups are (large) small relative to the control groups. We provide evidence from Monte Carlo simulations and from placebo DID regressions with the American Community Survey (ACS) and the Current Population Survey (CPS) datasets to show that this problem is relevant even in datasets with large numbers of observations per group. We then derive an alternative inference method that provides accurate hypothesis testing in situations where there are few treated groups (or even just one) and many control groups in the presence of heteroskedasticity. Our method assumes that we can model the heteroskedasticity of a linear combination of the errors. We show that this assumption can be satisfied without imposing strong assumptions on the errors in common DID applications. Importantly, we do not need to specify the structure of the serial correlation of the errors. Our inference method can also be combined with feasible generalized least square (FGLS) estimation. This way, it is possible to attain an uniformly most powerful (UMP) test if the FGLS assumptions are valid, while still provide a test with correct size if the serial correlation is misspecified. With many pre-treatment periods, we provide an alternative inference method that relies on strict stationarity and ergodicity of the time series instead of relying on the correct specification of the heteroskedasticity. Finally, we extend our inference methods to linear factor models when there are few treated groups.

ABSTRACT. The synthetic control (SC) method has been recently proposed as an alternative to estimate treatment effects in comparative case studies. The SC relies on the assumption that there is a weighted average of the control units that reconstructs the potential outcome of the treated unit in the absence of treatment. If these weights were known, then one could estimate the counterfactual for the treated unit using this weighted average. With these weights, the SC would provide an unbiased estimator for the treatment effect even if selection into treatment is correlated with the unobserved heterogeneity. In this paper, we revisit the SC method in a linear factor model where the SC weights are considered nuisance parameters that are estimated to construct the SC estimator. We show that, when the number of control units is fixed, the estimated SC weights will generally not converge to the weights that reconstruct the factor loadings of the treated unit, even when the number of pre-intervention periods goes to infinity. As a consequence, the SC estimator will be asymptotically biased if treatment assignment is correlated with the unobserved heterogeneity. The asymptotic bias only vanishes when the variance of the idiosyncratic error goes to zero. We suggest a slight modification in the SC method that guarantees that the SC estimator is asymptotically unbiased and has a lower asymptotic variance than the difference-in-differences (DID) estimator when the DID identification assumption is satisfied. If the DID assumption is not satisfied, then both estimators would be asymptotically biased, and it would not be possible to rank them in terms of their asymptotic bias.

ABSTRACT. The synthetic control (SC) method has been recently proposed as an alternative method to estimate treatment effects in comparative case studies. Abadie et al. (2010) and Abadie et al. (2015) argue that one of the advantages of the SC method is that it imposes a data-driven process to select the comparison units, providing more transparency and less discretionary power to the researcher. However, an important limitation of the SC method is that it does not provide clear guidance on the choice of predictor variables used to estimate the SC weights. We show that such lack of specific guidances provides significant opportunities for the researcher to search for specifications with statistically significant results, undermining one of the main advantages of the method. Considering six alternative specifications commonly used in SC applications, we calculate in Monte Carlo simulations the probability of finding a statistically significant result at 5% in at least one specification. We find that this probability can be as high as 13% (23% for a 10% significance test) when there are 12 pre-intervention periods and decay slowly with the number of pre-intervention periods. With 230 pre-intervention periods, this probability is still around 10% (18% for a 10% significance test). We show that the specification that uses the average pre-treatment outcome values to estimate the weights performed particularly bad in our simulations. However, the specification-searching problem remains relevant even when we do not consider this specification. We also show that this specification-searching problem is relevant in simulations with real datasets looking at placebo interventions in the Current Population Survey (CPS). In order to mitigate this problem, we propose a criterion to select among SC different specifications based on the prediction error of each specifications in placebo estimations.

ABSTRACT. This paper estimates the short- and medium-run effects of the construction of large hydroelectric power plants (HPPs) on the economic development of Brazilian municipalities. We use the synthetic control method (Abadie et al., 2010) to construct the counterfactual trajectory of the economy of 82 municipalities that were first affected by a HPP between 2002 and 2011. By effectively performing one case study for each affected municipality, we are able to characterize the distribution of the HPP effects. Two main findings emerge from this analysis. First, the median impact of the construction of HPPs on the local economy is modest and follows an inverted U-shape over a five-year horizon. Second, the estimated effects display a lot of dispersion: despite the median impacts being typically small and following similar patterns, the effects on some municipalities are much more severe than in others. These results do not provide support for the view that large construction works can unequivocally spur local development.

ABSTRACT. This study develops and implements a measurement strategy of the economic impact of foreign-aid projects at the Uganda district level (first level administrative unit - ADM1). The study entails two stages of implementation. The first stage requires the harmonization of four data sources: AidData's geocoded project mappings, living standard measurement type surveys (LSMSs), satellite nighttime luminance (nightlight) data, and multi-spectral satellite data that captures land-cover/land-use change (LCLUC). LCLUC data allows measurement of nonlight-generating economic activity associated with land-use transitions that may be associated with aid projects. The second stage will utilize this integrated data to estimate ODA-impact on growth. Two estimation strategies are employed to address endogenous aid allocation: panel vector autoregression (PVAR) produce structural estimates of impulse response of growth to ODA. Second, we use an interacted variables instrument (IIV) strategy to produce a powerful excludable instrument. The study demonstrates promise for both approaches. With the PVAR we nd relatively large and statistically significant positive impact upon nightlight of a one standard deviation shock to ODA, though the shock e effect is not persistent. With the IIV we find ODA generates a marginally significant (10% confidence level) positive efeect on night light. This project will be highly scalable and exible, with straightforward application to most low-income countries.

ABSTRACT. This paper investigates the role of infrastructure improvements on regional development, especially the improvements that relate to the transport network. This paper attempts to verify if the greater proximity to markets, by reducing transportation costs, provides greater regional development, which causes urbanization and changes in the local labor markets that lead to improvements in the living conditions of the population. We adopt an empirical strategy that is based on a difference-in-difference matching estimator. We focus on four different investments in the Brazilian transport network, namely, the Belém-Teresina (BR 010/316), the Cuiabá-Porto Velho (BR 364), the Cuiabá-Santarém (BR 163/230) and the Highway of Coffee (BR 376). Our analysis uses data from the Brazilian Census Demographic (IBGE) from 1970 to 2000. Considering the indicators of income, labor market and education, the main results show that the construction of new roads in areas of limited access has not necessarily contributed to the development of the municipalities that are benefited by the highway.