Search

We adopt an artificial counterfactual approach to assess the impact of lockdowns on the short-run evolution of the number of cases and deaths in some US states. To do so, we explore the different timing in which US states adopted lockdown policies, and divide them among treated and control groups. For each treated state, we construct an artificial counterfactual. On average, and in the very short-run, the counterfactual accumulated number of cases would be two times larger if lockdown policies were not implemented.

The number of Covid-19 cases is increasing dramatically worldwide, with several countries experiencing a second and worse wave. Therefore, the availability of reliable forecasts for the number of cases and deaths in the coming days is of fundamental importance. We propose a simple statistical method for short-term real-time forecasting of the number of Covid-19 cases and fatalities in countries that are latecomers – i.e., countries where cases of the disease started to appear some time after others. In particular, we propose a penalized (LASSO) regression with an error correction mechanism to construct a model of a latecomer in terms of the other countries that were at a similar stage of the pandemic some days before. By tracking the number of cases in those countries, we forecast through an adaptive rolling-window scheme the number of cases and deaths in the latecomer. We apply this methodology to four different countries: Brazil, Chile, Mexico, and Portugal. We show that the methodology performs very well. These forecasts aim to foster a better short-run management of the health system capacity and can be applied not only to countries but to different regions within a country, as well.

The Covid-19 crisis has lead to a reduction in the demand and supply of sectors that produce goods that need social interaction to be produced or consumed. We interpret the Covid-19 shock as a shock that reduces utility stemming from 'social' goods in a two-sector economy with incomplete markets. We compare the advantages of lump-sum transfers versus a credit policy. For the same path of government debt, transfers are preferable when debt limits are tight, whereas credit policy is preferable when they are slack. A credit policy has the advantage of targeting fiscal resources toward agents that matter most for stabilizing demand. We illustrate this result with a calibrated model. We discuss various shortcomings and possible extensions to the model.

Speculation, in the spirit of Harrison and Kreps [1978], is introduced into a standard real business cycle model. Investors (speculators) hold heterogeneous beliefs about firm growth. Firm ownership, and thus, the firm's discount factor varies with waves of optimism and leverage. These waves ripple into firm investments in hours. The firm's discount discount factor links the equity premium and labor volatility puzzles. We obtain an upper bound to the amplification that can be generated by speculation for any model of beliefs - a factor of 1.5. A calibration based on diagnostic beliefs amplifies hours volatility by a factor of 1.15 and produces a bubble component of 20 percent.

Financial constraints on Brazilian firms are very high compared to advanced economies. In Brazil, 59% of firms have access to a bank loan or a credit line. In developed countries, the average percentage is 95%. Loan collateral requirements are much higher in Brazil (95% of the loan value) than the developed country average (50% of the loan value). The interest rate spread in Brazil is very high in relation to the average of developed countries (12% in Brazil and 3% in developed countries). In this context, the objective of this work is to evaluate the impact of reductions in three different financial constraints on the gross domestic product (GDP) per capita in Brazil. In more specific terms, our main goal is to assess which of the three financial frictions (cost of participation in the credit market, borrowing limit or cost of monitoring) is the most important to affect per capita GDP in Brazil. To reach our goal, we used the theoretical framework developed in Dabla-Norris et al. (2015). This structure consists of a version of the neoclassical growth model with heterogeneous agents and three financial frictions. The model is calibrated for the Brazilian economy in 2009 and we perform simulation exercises in which we evaluate the impacts of reductions in financial frictions. In this text, we perform three exercises. In the first one, the reduction in the cost of participation in the credit market (which increases financial access to the level of developed countries) generates an increase in per capita GDP of 3.6%. In the second, the reduction in the cost of monitoring (which raises system efficiency to the level of developed countries) generates a rise in GDP per capita of 1.7%. Finally, in the third, we evaluated a relaxation in the indebtedness constraints that would be obtained if the collaterals as a proportion of the loans in Brazil were equal to the average of the developed countries. The results show that the reduction of collaterals in Brazil would raise GDP per capita by 12%. In this context, the financial constraint with the greatest impact on GDP per capita, in the Brazilian case, is the borrowing limit. In this sense, policies aimed at reducing borrowing constraints will have a greater impact on GDP than policies aimed at reducing the spread or increasing participation in the credit market.

We document that a huge frustration shock, clearly unrelated to government's actions, was perceived to lead to substantial punishment at the polls months later. In particular, we provide evidence that Brazil's 7-1 humiliating defeat to Germany in the 2014 World Cup was perceived by financial market participants as a political shock against the incumbent president. To do so, we explore an empirical strategy that allows us to extract daily political news content from stock market data. Among the many events that affected the convoluted 2014 presidential election, the 7-1 defeat to Germany was among those perceived to have high political impact. We propose an explanation whereby this huge frustration shock triggered a transfer of domains that led Brazilians to update their beliefs regarding government's poor performance.

We assemble a novel dataset on transitions in central bank leadership in several countries, and study how monetary policy is conducted around those events. We find that policy is tighter both at the last meetings of departing governors and first meetings of incoming leaders. This finding cannot be fully explained by endogenous transitions, the effects of the zero lower bound, surges in inflation expectations, omitted variables such as fiscal policy and uncertainty nor electoral cycles. We conclude by offering two possible, perhaps complementary, explanations for these results. One based on a simple signalling story, another based on career and reputation concerns.

In a repeated unobserved endowment economy in which agents negotiate long- term contracts with a financial intermediary, we study the implications of the interaction between incentive compatibility and participation constraints for risk sharing. In particular, we assume that after a default episode, agents consume their endowment and remain in autarky forever (one-sided commitment). We find that, once away from autarky today, if the probability of drawing the highest possible endowment shock is small enough, the optimal contract prevents agents from reaching autarky tomorrow and, thus, from being \impoverished". Moreover, an invariant cross-sectional distribution of life-time utilities (or values) exists. Our numerical simulations suggest that the mass of agents living in autarky is zero in the limit.

This paper augments a relatively standard dynamic general equilibrium model with financial frictions in order to quantify the macroeconomic effects of the credit deepening process observed in many Latin American (LA) countries in the last decade, most notably in Brazil. In the model, a stylized banking sector intermediates credit from patient households to impatient households and firms. The key novelty of the paper, motivated by the Brazilian experience, is to model the credit constraint faced by (impatient) households as a function of future labor income. In the calibrated model, credit deepening generates only modest abovetrend growth in consumption, investment, and GDP. Since Brazil has experienced one of the most intense credit deepening processes in Latin America, it is argued that the quantitative effects for other LA economies are unlikely to be sizeable.

A public job can be seen as a source of insurance against income risk. Indeed, many public employees have job stability, which is compounded with less volatile and more compressed wages. Hence, by increasing its number of public employees, the government enhances the overall degree of insurance in the economy. In this paper, we introduce public employment in a standard incomplete markets model with overlapping generations. The aim is to explore the welfare gains or losses due to a larger government, accounting for this extra source of insurance. In a model economy calibrated to Brazil, where public employment is around 13.5 percent of the workforce, we find that if the government relies on consumption taxes to balance its budget, the optimal size of public employment is nearly flat, ranging from 8 to 12 percent of the workforce. However, if the public employment is reduced from 12 to 8 percent, welfare losses due to a reduction in the degree of insurance are 2 percent, which are compensated by welfare gains due to level and inequality effects. This insurance effect is robust to a missepecification of the production technology associated with the public sector

Neste artigo, introduz-se ao modelo de crescimento neoclássico a defasagem necessária para que o investimento público se consolide em capital público (processo time-to-build), assim como alíquotas tributárias distorcivas que se ajustam de acordo com a dívida acumulada pelo setor público. O objetivo é isolar quantitativamente o efeito do aumento do investimento público observado no Programa de Aceleração do Crescimento (PAC). Dependendo da defasagem associada ao processo time-to-build e da política de ajuste fiscal consideradas, o PAC pode ter induzido uma queda do produto entre 0,2% e 0,4% em um horizonte de até quatro anos.

We use data on immigrants who live in the United States to study the effects of exposure to hyperinflation on occupational choice. To do so, we calculate the number of years an individual had lived under hyperinflation before arriving to the US. We find that its marginal effect on the probability of being self-employed instead of wage-earner is 0.87 percentage point. This effect depends on the age individuals had when exposed to hyperinflation. In particular, it is stronger for individuals who experienced hyperinflation at an early age, but it vanishes for those over the age of 40. These results suggest that the macroeconomic environment an individual grows up in permanently affects his economic behavior.

This paper introduces cash transfers targeting the poor in an incomplete marketsmodel with heterogeneous agents facing idiosyncratic risk. These transfers change the degree of insurance in the economy and affect precautionary motives asymmetrically,leading the poorest households to decrease savings proportionally more than their richer counterparts. In a model economy calibrated to Brazil, once the cash transfer program is adopted, wealth inequality and social welfare increase, poverty decreases,while employment and income inequality remain about the same. Imperfect access to financial markets is important for these results, whereas whether the program is funded with lump sum or distortive taxes is not.

This paper introduces cash transfers targeting the poor in an incomplete markets model with heterogeneous agents facing idiosyncratic risk. These transfers change the degree of insurance in the economy and affect precautionary motives asymmetrically, leading the poorest households to decrease savings proportionally more than their richer counterparts. In a model economy calibrated to Brazil, once the cash transfer program is adopted, wealth inequality and social welfare increase, poverty decreases, while employment and income inequality remain about the same. Imperfect access to financial markets is important for these results, whereas whether the program is funded with lump sum or distortive taxes is not.

I investigate the optimal auditing scheme for a revenue-maximizing tax-collection agency that observes not only reported pro ts, but also the level of employment at each firm. Each firm is owned by a single entrepreneur whose managerial ability is random. The optimal auditing scheme is discontinuous and non-monotone in ability. In intermediate audit costs, less-productive entrepreneurs face auditing probabilities that increase in ability, whereas the ablest ones are not audited. Finally, the effecive tax rate is higher in the middle of the managerial ability distribution; thus, the overall regressive (or progressive) bias that arises from evasion is unknown.

While most economic studies of crime have focused on its determinants, we study the reverse question: does crime affect economic behavior? Being such an important social phenomenon, one would expect crime to affect economic decisions. Using local data on crime rates and savings per capita in a high-crime environment, we document a striking empirical relationship: crime induces savings. Our paper is one of the first to successfully relate crime to an economic outcome. This result is robust to an extensive sensitivity analysis, which include: 1) controlling to a large set of demographic covariates; 2) accounting for the fact that crime and savings may be determined jointly; 3) measuring savings in different ways; 4) accounting for the presence of possible outliers; 5) weighting the data according to population; 6) accounting for spatial correlation; and, finally, 7) estimating the model for different sub-samples of cities. Our estimates indicate that only property, not violent, crime induces savings, which is consistent with the theoretical explanations on why crime would increase thriftiness

Esta tese estuda o comportamento da inflação no Brasil e consiste em três ensaios. O primeiro capítulo introduz um arcabouço de projeção de inflação de curto prazo usando um conjunto de 5147 séries relacionadas, principalmente, a índices e subíndices de inflação e a preços. As séries tem frequências diferentes e representam a inflação e os preços em diferentes estágios da cadeia de produção. Usamos várias classes de modelos e métodos, MQO, regularização, fatores e modelos de árvore, para projetar o IPCA e também vários dos seus subitens que são posteriormente agregados. Nossos resultados indicam que as escolhas do melhor modelo e nível de agregação dependem do horizonte de projeção. Em média, os modelos de fatores e os métodos de regularização tem melhor desempenho, bem como níveis de agregação mais granulares. As combinações de projeções são mais precisas que projeções individuais e apresentam bom desempenho relativo em relação aos benchmarks de mercado, principalmente na véspera da divulgação dos dados de inflação. O segundo capítulo corrige o viés de projeções de inflação explorando a estrutura dos erros fora de amostra. O arcabouço introduzido analisa os erros no cross section dos horizontes de projeção e na série temporal para estimar o erro de projeção. Consideramos diferentes modelos de fatores para os erros: Diebold-Li 3 fatores, Componente Principal e Modelo de Fatores Dinâmicos. Estimamos a dinâmica dos fatores usando modelos autorregressivos. Nosso arcabouço melhora a precisão das projeções de inflação. A melhora depende do modelo e da agregação. Os melhores resultados são para os métodos de regularização e para as agregações mais granulares. A combinação de projeções ajustadas pelo erro esperado também apresenta resultados positivos. Em comparação com analistas profissionais, o erro da combinação das projeções originais está entre os menores 20%, enquanto o erro da combinação das projeções ajustadas está entre os menores 12%. O terceiro capítulo analisa as relações entre séries macroeconômicas e expectativas de inflação. O objetivo é avaliar como a estrutura a termo da inflação implícita responde a variações exógenas em variáveis macro. O impacto das inovações estruturais na curva da inflação implícita é determinado em dois estágios. Primeiro, estimamos as relações entre a curva de inflação e as séries macroeconômicas. Na segunda etapa, examinamos os impactos dos choques macroeconômicos usando um SVAR. Em seguida, calculamos o efeito dos choques sobre a inflação implícita combinando as IRFs das variáveis macro e suas relações com os fatores da curva de inflação. Em geral, as respostas seguem o sinal esperado pela teoria econômica: choques positivos na inflação corrente e na taxa de câmbio elevam as expectativas de inflação. Também como esperado, a resposta da parte longa da curva da inflação implícita é mais tímida do que a resposta da parte curta. This thesis studies inflation dynamics in Brazil and consists of three essays. The first chapter proposes a short-term inflation-forecasting framework. The dataset consists of 5,147 series, related mostly to inflation indices and sub-indices, and prices. Series have different frequencies, and represent inflation and prices at different stages of the supply chain. We estimate several classes of models and methods: OLS, regularization, factors, and tree models. We generate forecasts for the headline IPCA index, and for several sub-items that we then aggregate. Our results indicate that the choice of the best model and the best level of aggregation depend on the forecast horizon. On average, factor models and regularization methods perform better, as do more granular levels of aggregation. Projections generated through combinations of individual forecasts are more accurate and perform relatively well against market benchmarks, especially on the eve of the release of inflation data. The second chapter corrects the bias of inflation forecasts by examining the out-of-sample error structure. We introduce a framework to analyze the errors in the cross section of forecast horizons, and in time series to estimate the forecast error. We consider different factor models for the errors: Diebold-Li 3-factor, principal component, and dynamic factor model. We estimate the factor dynamics using autoregressive models to forecast the error. Our framework improves the accuracy of inflation forecasts. This improvement depends on the model and aggregation. Regularization methods and the most granular aggregations give the best results. A combination of adjusted forecasts performs well. Compared to professional analysts, the combination of original forecasts’ error is among the smallest 20%, while the combination of adjusted forecasts’ error is among the smallest 12%. The third chapter analyzes the relationships between macroeconomic series and inflation expectations. Our goal is to assess how the term structure of breakeven inflation responds to exogenous changes in macro variables. We determine the impact of structural macroeconomic innovations on the breakeven inflation curve in two stages. First, we estimate the inflation curve’s relationship to the macroeconomic series. In the second stage, we examine the impacts of macro shocks, using an SVAR. We then calculate the effect of the shocks on breakeven inflation, by combining the IRFs of the macro variables, and their relations with the factors of the inflation curve. In general, responses have the signal expected by economic theory: positive current inflation and exchange rate shocks raise inflation expectations. Also as expected, the response of the long end of the breakeven inflation curve is more muted than that of the short end.

Esta tese consiste em três artigos independentes em macroeconometria. No primeiro artigo, nós propomos um modelo de rational-inattention, a fim de se analisar e estimar o comportamento de forecasters profissionais do Focus Survey, um sistema administrado pelo Banco Central do Brasil (BCB). Já no segundo artigo nós realizamos o nowcast da inflação do Brasil, oficialmente medido pelo IPCA. Nesse artigo, nós usamos métodos de aprendizado de máquina (machine learning), modelos de alta dimensionalidade e combinação de forecasts para melhorar a precisão do nowcast de inflação. Finalmente, o terceiro artigo investiga o potencial impacto que o problema de viés de seleção dos ativos pode apresentar em explicar o Equity Premium Puzzle. Nossa abordagem aplica simulação de Monte Carlo e estimação do coeficiente de aversão relativa ao risco via Método dos Momentos Generalizados. This thesis consists of three independent essays in macroeconometrics. In the first paper, we propose a novel rational-inattention approach to model and estimate the forecasting behavior of professional forecasters in the Focus Survey, gathered by the Brazilian Central Bank (BCB). In the second paper we nowcast Brazilian inflation. In this paper we use machine learning methods, high dimensional models and combination of forecasts in order to improve the accuracy of inflation nowcasting. The third paper investigates the potential impact a selection-bias problem for assets has in explaining the Equity Premium Puzzle. Our approach uses Monte-Carlo simulations and we estimate the relative risk aversion coefficient using the Generalized Method of Moments.

Esta tese é composta de três ensaios relacionados a economia brasileira. O primeiro deles estuda a evolução da incerteza fiscal, bem como os seus efeitos e mecanismos de transmissão sobre os agregados macroeconômicos. O segundo ensaio analisa os efeitos da política de crédito direcionado sobre as variáveis macroeconômicas e a potência da política monetária. O arcabouço teórico desses ensaios é o modelo DSGE. No último capítulo, o modelo de Laubach e Williams (2003) é adaptado para uma pequena economia aberta, e nós estimamos a taxa de juros natural do Brasil. This thesis is composed of three essays related to the Brazilian economy. The first one studies the evolution of fiscal uncertainty, as well as its impacts and transmission mechanism on macroeconomic aggregates. The second essay analyzes the effects of earmarked credit policies on macroeconomics variables and on the power of monetary policy. The theoretical framework of these essays is the DSGE model. In the last chapter, the Laubach and Williams (2003) model is adapted to a small open economy, and we estimate the Brazilian natural rate of interest.

Como os choques agregados são otimamente compartilhados entre trabalhadores e aposentados? Em primeiro lugar, como os riscos são compartilhados entre dois grupos de agentes: um que deve ser incentivado a trabalhar e outros, que não são precisam mais incentivados? Segundo, como os incentivos podem feitos por meio de promessas de aposentadoria, como essas promessas devem variar otimamente em diferentes estados da natureza? Depois de formalizar esses problema, mostramos que o compartilhamento de riscos é ótimo para utilidade log e quando o crescimento de produtividade é i.i.d.. Em outros casos, o compartilhamento de não é analiticamente ótimo, com melhorias sendo possíveis ao se atribuir mais risco aos aposentados (resp. trabalhadores) quando o crescimento de produtividade persistente (resp. reversão a média). Por fim, resolvemos numericamente os modelos e mostramos que se escolhermos parâmetros suficientemente próximos da utilidade log ou crescimento i.i.d., os ganhos de desviar-se do compartilhamento ótimo de riscos é desprezível. No entanto, com parâmetros mais e mais distantes desse conjunto, os ganhos dos desvios se tornam cada vez maiores e mais relevantes. How is aggregate risks optimally shared between workers and retirees? We can break this question in two parts. First, how is risk shared between two groups of agents: one which must be provided incentives to make effort and other, which no longer be incentivized? Second, since incentives may be backloaded through pension entitlements, how should backloading optimally vary across states of nature? After formalizing these two aspects of the problem, we show that perfect risk sharing is optimal for log utility and when productivity growth is i.i.d.. For all other cases, perfect risk sharing is not analytically optimal, with improvements taking place by assigning more risk to retirees (resp. workers) when productivity growth is persistent (resp. mean reverting). Finally, we numerically solve the model and show that if we choose parameters close enough to log utility and i.i.d. growth, gains from deviating from perfect risk-sharing are meaningless. However, once we go far and far away from this set, the optimal will also become far away from perfect risk-sharing.

O objetivo dessa dissertação é avaliar os efeitos que certas metodologias usadas na construção de índices e portfolios têm sobre suas performances e como elas diferem entre si. Para isso, analisaremos o caso do índice Ibovespa cuja metodologia foi alterada em 2014. Esse trabalho analisa os indicadores de performance dos principais índices de ações brasileiros durante o período que engloba os anos 2003 a 2013, incluindo também uma simulação para o que teria sido o Ibovespa caso as mudanças implementadas em 2014 estivessem em vigor durante esse período. Posteriormente, por meio de testes e regressões, é feita uma tentativa de encontrar fatores que expliquem as diferenças encontradas nos comportamentos dos índices. Os principais resultados encontrados são que a metodologia antiga do Ibovespa é menos eficiente e gera performance inferior aos outros três índices avaliados, todos utilizando uma ponderação por market cap. Além disso, encontra-se evidências de que períodos de piores rendimentos relativos do Ibovespa coincidem com momentos de grande atividade no mercado acionário, indicando uma relação positiva entre os diferenciais de retorno dos índices e o nível de transações realizadas no período.

The purpose of this work is to study the role for government in mitigating capital misallocation. We develop an entrepreneurship model in which heterogeneous producers face collateral constraints on production, but can hedge idiosyncratic shocks. Hedging works as a tool for reallocating resources to states in which they are more productively deployed, and can alleviate the effect of the financial frictions and be a counteracting force to capital misallocation. Government incentives to hedging improve workers’ welfare in steady state through an increase in TFP and wages. The intervention leads to a reduction in the rate of return of entrepreneurs and an increase in wealth dispersion. These two effects cause entrepreneurial welfare to decrease.