SBE39: 39TH MEETING OF THE BRAZILIAN ECONOMETRIC SOCIETY
PROGRAM FOR THURSDAY, DECEMBER 14TH
Days:
previous day
next day
all days

View: session overviewtalk overview

08:30-10:15 Session 10A: Sessão Especial FIPE

EconomistAs [Brazilian Women in Economics]

Chair:
Renata Narita (FEA/USP, Brazil)
Location: Jade
08:30
Renata Narita (FEA/USP, Brazil)
Apresentação do grupo de pesquisa EconomistAs [Brazilian Women in Economics]
SPEAKER: Renata Narita
08:45
Cecília Machado (FGV/EPGE, Brazil)
Valdemar Pinho Neto (FGV/EPGE, Brazil)
The Labor Market Consequences of Maternity Leave: Evidence from Brazil

ABSTRACT. This paper examines the labor market consequences of paid maternity leave for women who participate in the Brazilian formal labor market (and thus qualify for the leave policy). We take advantage of rich administrative data and follow women 47 months before
and after leave-taking. Using an event study approach, we nd an inverted U-shape employment pattern, which peaks at the first month of leave-taking. Employment is stable until the fourth month, indicating compliance with the legislation, but drops sharply after the job protection period and stabilizes again at around one year. Almost half of the women are out of the formal labor market 47 months after the leave. As most employment effects are due to separations that occur from employer's initiative, our results suggest that further policies are needed to promote higher attachment of women in the labor market, especially for the less educated workers. We also restrict our analysis to firms that have extended the leave period by combining the event study analysis with a difference-in-difference strategy. We nd that the extended leave policy alleviates part of the negative employment effects of maternity leave-taking.

09:30
Daniel Santos (FEA-RP/USP, Brazil)
Quem se beneficia da expansão de creches no Brasil?
SPEAKER: Daniel Santos
08:30-10:15 Session 10B: Theory II
Chair:
Wilfredo Maldonado (UCB, Brazil)
Location: Mezanino Bossa Nova
08:30
Fernando Barros Jr (FUCAPE, Brazil)
Ricardo Cavalcanti (FGV/EPGE, Brazil)
On the optimality of inside-money inflation in random-matching models

ABSTRACT. We study the optimality of inflationary policies in a model of inside money. We build on Cavalcanti and Wallace (1999) random-matching model allowing for a monetary intervention. Then, using numerical simulations, we unveil some properties of the optimal allocation featuring inside money. Our findings indicates that although inflationary policies decreases the return of money, there are some net benefits of money creation in matching models. This results holds for a broad set of parameters, however, when people are patient and the fraction of monitored people is high enough, then inflation is not part of the optima.

09:00
Liev Maribondo (IMPA, Brazil)
Susan Schommer (UERJ, Brazil)
Aloisio Araujo (IMPA e FGV/EPGE, Brazil)
Collateral Requirement Regulation in a General Equilibrium Model - Job Market Paper

ABSTRACT. This paper proposes and analyzes, as an alternative policy to be used in moments of crisis, the relaxement of the collateral requirement with government compensation in the second period. The numerical results shows that this policy can be more effective than the unconventional monetary policy. In addition to unvealing greater regions of Pareto improvement than the ones obtained with the unconventional policy, it also shows the it is possible to produce Pareto improvement in economies short sale constrained, which is an impossibility with the unconventional policy.

09:30
Wilfredo Maldonado (UCB, Brazil)
Luciana Fiorini (University of Western Australia, Australia)
Richard Cornes (Australian National University, Australia)
Expectational Stability in Aggregative Games

ABSTRACT. Using the replacement function associated with aggregative games, we analyze the expectational dynamics of the aggregate strategy of the game. We can interpret the Nash equilibrium of the game as the rational expectations equilibrium (REE) of the system, and we examine the expectational stability of the REE. We characterize local stability in terms of fundamentals and the REE itself. We illustrate the results through well-known aggregative games (Cournot games, Bertrand competition with dfferentiated goods, rent seeking games, and the public goods provision game) and analyze their global expectational dynamics.

08:30-10:15 Session 10C: Education I
Chair:
Isabela Furtado (World Bank e FGV/EESP, Brazil)
Location: Ametista
08:30
Vitor Pereira (CPI, Brazil)
Claudio Ferraz (PUC-RIO, Brazil)
Can students benefit if teachers lose their bonus? Behavioral biases inside the classroom - Job Market Paper
SPEAKER: Vitor Pereira

ABSTRACT. While previous papers have identified evidence of loss aversion in laboratory experiments or in sports, 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 these results and we find significant changes in teachers’ pedagogical practices.

09:00
Priscilla Tavares (FGV/EESP, Brazil)
Vladimir Ponczek (FGV/EESP, Brazil)
Efeitos de aumentos salariais dos professores sobre o desempenho dos estudantes: evidências para a rede estadual paulista

ABSTRACT. Neste artigo, nós fornecemos evidências dos efeitos de aumentos salariais de professores da educação básica sobre o aprendizado dos alunos, no contexto de países em desenvolvimento (rede pública estadual de São Paulo). Nós exploramos a variação na remuneração dos professores, dada pela regra dos adicionais salariais por tempo de serviço (regra dos quinquênios). Observamos a elegibilidade de cada professor ao aumento de salário e exploramos as diferenças na data de contratação dos docentes ao longo do ano para calcular o tempo de exposição dos professores tratados aos salários mais elevados. Empregamos uma estratégia de diferenças-em-diferenças para controlar por características não-observáveis de professores pertencentes a coortes de admissão distintas. Nossos resultados estão em linha com o que se encontra na literatura empírica internacional: aumentos salariais para professores já contratados não parecem afetar sua produtividade e, portanto, não são capazes de impactar o aprendizado dos estudantes na educação básica.

09:30
Luis Felipe Oliveira (IPEA e UnB, Brazil)
Rafael Terra (UnB, Brazil)
Impacto do Programa Ensino Médio Inovador em indicadores educacionais

ABSTRACT. Esse artigo estima o impacto, em indicadores educacionais, de transferências federais para escolas estaduais, por meio do Programa Ensino Médio Inovador (ProEmi), nos anos de 2013 e 2015. A transferência de recursos, escalonada por patamar de matrículas de ensino médio, traz em sua regra uma variação descontínua no tratamento. Ou seja, tem-se um fator exógeno para a determinação da transferência recebida. Garante-se, portanto, uma estratégia de identificação que estima o efeito causal da intervenção similar aos trabalhos de Ferraz e Finan (2011) e Angrist e Lavy (1999). Assim sendo, o artigo se enquadra na discussão sobre a efetividade de políticas que aplicam mais recursos em educação e sobre descentralização de recursos na provisão de serviços educacionais. No entanto, a maioria das evidências encontradas, em quatro abordagens econométricas distintas, indica efeito causal nulo.

This paper examines whether higher direct federal transfers for high schools, controlled by Brazilian states, improve educational performance. To identify these effects, we exploit non-linear step-function pattern in the transfers rule. The exogenous variation is addressed by four different econometric models including similar approaches to Ferraz e Finan (2011) and Angrist e Lavy (1999) identification strategies. We show that the program has barely no effect on educational outcomes in 2013 and 2015, indicating that it needs to update its accountability.

10:00
Isabela Furtado (World Bank e FGV/EESP, Brazil)
Caio Piza (World Bank, Brazil)
How Early Should Financial Education Be Taught? Evidence from a Large-Scale Experiment

ABSTRACT. This paper presents the findings of a large-scale experimental evaluation of a financial education pilot program for primary school students in Brazil. The pilot was carried out during the school year of 2015 and included students from four different grades (3rd, 5th, 7th and 9th) in 101 municipal schools of Manaus and Joinville. The pilot’s objective was to increase students’ financial proficiency and, consequently, change attitudinal and behavioral outcomes regarding consumption and savings. To your knowledge, this is the second randomized controlled trial of a financial education program targeting primary school students in a developing country, though the first to report gains in financial proficiency and changes in attitudinal outcomes. The results suggest that the program increased financial literacy by 0.07 SD for the overall sample and by 0.1 SD for middle school students. We also found positive results on attitudes towards consumption and savings of 0.09 and 0.05 SD respectively. Heterogeneous analyses were carried out seeking to understand: (1) for which subgroup the pilot had the largest effect, and (2) the extent to which the pilot’s implementation influenced the program’s efficacy. In fact, the program had positive average effects on knowledge only for middle school students. We found that the program had larger average effects among students with more educated mothers, but curiously found suggestive evidence that the program might have been slightly more effective among the poor. Finally, we exploited information on program’s execution and found positive and statistically significant effects only when at least 60 percent of the syllabus was covered in classroom. These results indicate that (i) knowledge is not a sufficient condition to change attitudes and behavior, (ii) the program might have larger effects if better targeted, and (iii) the quality of program’s implementation plays a key role in program’s effectiveness.

08:30-10:15 Session 10D: DSGE
Chair:
Celso José Costa Junior (UEPG e FGV/EESP, Brazil)
Location: Água Marinha
08:30
Lucicleyton Farias (UFPE, Brazil)
Marcelo Eduardo Alves (UFPE, Brazil)
Commodity Prices and Business Cycles in Small Open Economies: The Role of News Shocks

ABSTRACT. In this paper, we explore the hypothesis that some movements in commodity prices are anticipated and can trigger aggregate fluctuations in the context of a dynamic stochastic general equilibrium model. The model is a multi-sector small open economy model, with endogenous commodity production, featuring three real rigidities: internal habit formation, capital adjustment costs and working capital constraint. There are five exogenous processes: a productivity shock for each sector, a country-specific interest rate shock that responds to commodity price fluctuations, and a commodity price shock composed by an unanticipated and anticipated component. Agents receive news about future movements in commodity prices with two and four periods in advance. We show that "News" shocks lead to higher real output, investment and consumption, and to a counter-cyclical movement in the trade-balance-to-output ratio. We also show that "News" shocks are a non-negligible source of business cycles in small open economies, accounting for around 15% of fluctuation in output and consumption and even larger shares of the fluctuations in investment, labor and the trade balance-to-output ratio.

09:00
Silvio Costa (BCB, Brazil)
Structural Trends and Cycles in a DSGE Model for Brazil
SPEAKER: Silvio Costa

ABSTRACT. This paper builds and estimates a structural growth model with micro-founded specifications of trends and cycles to allow a consistent multivariate filtering of key macroeconomic variables. Emerging countries like Brazil show expressive trend dynamics that can blur the vectors determining real business cycles, but most models cannot consider this additional complexity, and so they usually throw out meaningful dynamics of scarce data. Our basic DSGE model cares for the raw data dynamics and aims to disentangle endogenously trends and cycles and unveil the underlying growth vectors. Thereby, historical interpretation and forecasts can hold internal consistency, which improves storytelling and calls for an in-depth forecasting. We report model evaluation summaries to support fully integrated models as a highly valuable tool for applied macro exercises and policy advising. Yet, we present an application for the measurement of potential output and the output gap.

09:30
Celso José Costa Junior (UEPG e FGV/EESP, Brazil)
Alejandro Garcia-Cintado (Pablo de Olavide University, Spain)
Carlos Usabiaga (Pablo de Olavide University, Spain)
Fiscal Adjustments and the Shadow Economy in an Emerging Market

ABSTRACT. We build a DSGE model that allows us to: (i) derive a time series for the shadow economy in Brazil spanning the period 2002-2014, which successfully replicates the behavior of the main series provided by Pesquisa Nacional por Amostra de Domicílios; (ii) run dynamic simulations showing that expenditure-based fiscal consolidations, in the presence of a large informal labor markets (around 50% of the total labor force), lead, at worst, to mild recessions in the formal sector and are likely to foster public debt sustainability. On the contrary, increases in distortionary taxation worsen both public debt dynamics and fiscal collection, but do not necessarily trigger a recession due to the offsetting increase in the informal economy. Thus, in countries with large informal economies such as Brazil, expenditure-cutting measures should be relied upon when trying to put the fiscal house in order.

08:30-10:15 Session 10E: Finance III
Chair:
Bruno Giovannetti (FGV/ESSP, Brazil)
Location: Topázio
08:30
Fernando Barbosa (FGV/EPGE, Brazil)
Marco Bonomo (Insper, Brazil)
Lira Mota (Columbia University, United States)
Loan Fee Dispersion and the Cross-Section of Returns
SPEAKER: Marco Bonomo

ABSTRACT. The over the counter (OTC) structure of the market to borrow stocks gives rise to loan fee dis- persion across contracts for the same stock at the same day. Unlike other countries, in Brazil all stock loan transactions must be registered at the stock exchange. In other words, investors face an opaque market, but researchers can observe every single transaction. Using data from 2007 to 2013 we show that loan fee dispersion is the best predictor of the cross-section of stock re- turns, when compared to traditional short-sale related measures in the literature: loan fee average, short-interest, days to cover,and loan fee volatility.

09:15
Fernando Chague (FGV/EESP, Brazil)
Rodrigo De-Losso (FEA/USP, Brazil)
Bruno Giovannetti (FGV/EESP, Brazil)
The Short-selling Skill of Institutions and Individuals

ABSTRACT. Using market-wide data from the Brazilian stock lending market, we find strong evidence of short-selling skill among institutions and individuals. Skilled short-sellers present out-of-sample performance persistence, both over time and across stocks. Performance persistence is robust: by randomly splitting the sample across stocks, we show that performance in a group of stocks often predicts performance in another group of stocks. We then study how skilled short-sellers trade. We find that most of their profit does not come from firm-specific private information, they follow short-term momentum strategies, and they do not display the disposition effect.

10:30-12:15 Session 11: Keynote lecture
Chair:
Ariaster Chimeli (FEA/USP, Brazil)
Location: Jade
10:30
Christopher Timmins (Duke University, United States)
Environmental Justice
13:45-15:30 Session 12A: Sessão Especial Instituto Escolhas

Economia e Meio Ambiente: o papel do sistema tributário

Chair:
Sergio Leitão (Instituto Escolhas, Brazil)
Discussant:
Claudio Lucinda (FEA-RP/USP, Brazil)
Location: Jade
13:45
Sergio Leitão (Instituto Escolhas, Brazil)
Fernanda Estevan (FEA/USP e SBE, Brazil)
Programa de Bolsas Edital 2018 da Cátedra Escolhas Economia e Meio Ambiente
14:00
Bernard Appy (Centro de Cidadania Fiscal e Instituto Escolhas, Brazil)
Economia e Meio Ambiente: o papel do sistema tributário
SPEAKER: Bernard Appy
13:45-15:30 Session 12B: Sessão Especial Prêmio SBE Econometria
Chair:
Cristine Pinto (FGV/EESP, Brazil)
Discussants:
Sergio Firpo (Insper, Brazil)
Luiz Renato Lima (The University of Tennessee e UFPB, United States)
Cristine Pinto (FGV/EESP, Brazil)
Location: Topázio
13:45
Otavio Bartalotti (Iowa State University, United States)
Quentin Brummet (U.S. Census Bureau, United States)
Steven Dieterle (The University of Edinburgh, UK)
A Correction for Regression Discontinuity Designs with Multiple Mismeasurement Types in the Running Variable

ABSTRACT. We propose a simple procedure to correct estimates obtained in a regression discontinuity framework in the presence of multiple types of measurement error in the running variable. Our procedure extends the prior literature on measurement error on the running variable by leveraging auxiliary information from (potentially) multiple datasets in order to account for more general forms of measurement error. Additionally, we propose new adjusted asymptotic variance and standard errors that take in consideration the variability introduced by the estimation of nuisance parameters from multiple auxiliary datasets required to implement the measurement error correction. Simulations provide evidence that the proposed procedure adequately corrects for measurement error bias and tests using the new adjusted formulas exhibit empirical coverage closer nominal test size than ``naive'' alternatives. We provide two empirical applications, which demonstrate that correcting for measurement error can either reinforce the results of a study or provide a new empirical perspective on the data.

14:30
Lara Nassar Rangel (UFRGS, Brazil)
Flávio Ziegelmann (UFRGS, Brazil)
Seasonal Weighted Lag Adaptive LASSO

ABSTRACT. The main purpose of this paper is to propose a new LASSO type penalty aiming to improve out-of-sample forecasting performance for seasonal time series in high dimensionality scenarios. We also present results concerning parameter estimation performance. Our approach leads to what we call SWLadaLASSO (seasonal weighted lag adaptive LASSO) which assigns larger penalties for higher-lagged covariate coefficients - based on the idea of WLadaLASSO by Konzen and Ziegelmann (2016) - but those associated to the seasonal lags of the variable being estimated. It can be considered a generalization of the WLadaLASSO. In our Monte Carlo studies, the SWLadaLASSO is superior in terms of forecasting, parameter estimation and also covariate selection in most of the cases when compared to other LASSO-type penalty models. An empirical application is conducted to evaluate the capability of the proposed approach to forecast Brazilian GDP growth. Additionally, a set of forecast combinations is implemented in search of forecast accuracy improvement.

13:45-15:30 Session 12C: Sessão Especial Prêmio SBE Micro Aplicada
Chair:
Gustavo Gonzaga (PUC-Rio, Brazil)
Discussants:
Gustavo Gonzaga (PUC-Rio, Brazil)
Rodrigo Moita (Insper, Brazil)
Vladimir Ponczek (FGV/EESP, Brazil)
Location: Ametista
13:45
Renata Narita (FEA/USP, Brazil)
Gabriella Conti (University College London, UK)
Rita Ginja (University of Bergen, Norway)
Non-Contributory Health Insurance and Household Labor Supply: Evidence from Mexico
SPEAKER: Renata Narita

ABSTRACT. A central topic in the global health agenda is universal health coverage (UHC). The primary goal of social health insurance schemes is to protect beneficiaries from the health and financial consequences of adverse health events. While in this sense there is scope for government intervention in providing insurance, the impacts of UHC on labor markets in developing countries are less clear. We study this issue using the case of Mexico, which introduced in 2002 a non-contributory health insurance scheme directed to the half of the country's population uncovered by Social Security protection (the Seguro Popular, SP). Since before SP uninsured individuals could only access affordable health care through their employer, the introduction of a non-contributory public health insurance scheme could have resulted in large effects on the labor market. In practice, SP is a transfer(tax) to the informal(formal) sector workers and to the nonemployed. On the one hand, if the value placed on SP benefits is high, the introduction of fully subsidized health insurance can lead to negative impacts on employment and/or formality. On the other hand, wages in equilibrium might compensate the increase in benefits in the informal sector, in which case the impact on formality and employment is ambiguous. We start analyzing the effects of SP on labor market outcomes by exploiting its staggered introduction across municipalities using a difference-in-differences strategy on the Mexican Labor Force Survey data. We show that the implementation of SP in a municipality is associated with an increase in informality by 4% for low-education families with children. Then, to study why the policy change had limited impacts on the labor market, we develop and estimate a novel household search model which incorporates the value of SP as well as the pre-reform valuation assigned to the amenities in the formal sector relative to the alternatives (i.e., informal sector and non-employment), in order to understand whether access to free health services is valued by household members when they make their labor market decisions. Our structural model is able to replicate both the stocks of household types by Social Security coverage and the transitions in and out of employment and between formal and informal jobs in the pre-reform period. The results show that the steady-state marginal willingness to pay for the health insurance coverage provided by SP is very low, amounting to only 1.3% to 4.2% of the mean wage in the informal sector. Lastly, using the model to simulate counterfactual scenarios of employment and labor formality under different valuations of the new health system implemented in Mexico, we find that the willingness to pay for SP would have had to be significantly greater than it was to have substantial impacts on the economy.

14:30
Rafael P. Ribas (University of Amsterdam, Netherlands)
Erasmo Giambona (Syracuse University, United States)
The External Cost of Prostitution: Evidence from Downsizing Red Light Districts in the Netherlands

ABSTRACT. We measure the externalities of prostitution by quantifying the discount that households require to live next to a brothel. In our tests, we exploit a unique feature of Amsterdam's Red Light District (RLD), where private homes are next to prostitution windows and inside a perimeter naturally delimited by canals. Using a two-dimensional difference-in-discontinuity (DiD) estimator, we find that households pay as less as 24% on homes inside the RLD. We also find that this discount disappears once prostitution windows are forcibly closed by local authorities. Notably, by incorporating the exact coordinates of brothel closings, our empirical design allows us to establish a direct link between these closings and changes in price discontinuities. To estimate the economic impact on households out of the RLD, we also exploit the closing of all brothels in Utrecht (the fourth largest city in the Netherlands) in 2013. Households are found to paid up to 12% of their home value to be far from prostitution. In both cities, the contraction of the paid-sex industry is also associated with a drastic reduction in crime rates. Overall, our findings suggest that prostitution does far more harm than good for residents due to the nuisances that it creates.

13:45-15:30 Session 12D: Macroeconomics II
Chair:
Felipe Iachan (FGV/EPGE, Brazil)
Location: Água Marinha
13:45
Mariana Fialho Ferreira (UFES, Brazil)
Pedro Cavalcanti Ferreira (FGV/EPGE, Brazil)
João Victor Issler (FGV/EPGE, Brazil)
On Deriving Evidence from Input-Output Linkages: A Multi-Country Sectoral Total Factor Productivity Analysis

ABSTRACT. In this paper, we offer an alternative - and perhaps more appropriate - analytical setup for estimating sectoral TFPs. We employ the seminal contribution of Long and Plosser (1983) to the real business cycle literature to construct total factor productivity (TFP) measures that are multi-country and multi-sector. This is a novel approach, since the previous literature has used only labor-productivity measures to discuss long-run growth. The artificial economy uses inputs from different sectors in producing every commodity, which is either used as an input or consumed by the representative agent in a general equilibrium framework. This input-output structure is suitable for a wide range of exercises, which can be performed using datasets that only recently became available. By combining worldwide data from the WIOD and the input-output linkages of the proposed setup, we are now able to calculate sectoral total factor productivity for a set of countries that respond to more than 85% of world's GDP. We construct both 35-activity TFP indices and level series for a group of forty countries, which responds for more than 85% of world GDP. We then aggregate data to construct country-level and 3-sector indices and level series, so that we can extract some new insights and compare our country-level data. Our findings suggest that, in most of the countries, the services sector not only has presented higher productivity than industry, but also that these differences are getting larger over time. These results are at odds with some widely known conclusions, which have been drawn from labor productivity data by previous work in the field. It occurs mainly because our framework enables us to quantify the influence of all productive inputs on total factor productivities. Therefore, by constructing this new dataset on TFPs, we aim to to encourage a wide range of new disaggregate sectoral analysis for a large number of countries.

14:15
Filipe Vasconcelos (FEA/USP, Brazil)
Firm dynamics and productivity learning: When wrong incentives can lead to losses Aggregated.

ABSTRACT. This project aims to present a model that shows that, in a single economic environment,government stimulus to sectors, which features high productivity, can generate short-term productivity gains aggregates, but misallocation and long-term aggregate productivity loss. This may occur if a long-term productivity component is imperfectly observablein the short term due to temporary shocks, and capital is specific and irreversible insome sectors. The ideal be learn about long-term components of the productivity before investing. However accelerate investments in the sectors of higher productivity generates productivity gains in the short term, since higher productivity in the short term this correlated with higher productivity in the long run. As discussed, this fact would motivate governments to encourage high productivity observed sectors, although these incentives were socially suboptimal.

14:45
Felipe Iachan (FGV/EPGE, Brazil)
Dejanir Silva (University of Illinois at Urbana-Champaign, United States)
Ana Luiza Dutra (Princeton University, United States)
Insuring firms, ensuring misallocation
SPEAKER: Felipe Iachan

ABSTRACT. We study entrepreneurial risk taking in an economy in which collateral constraints lead to misallocation. We show that hedging bad states for firms lowers aggregate productivity and wages. It also increases the dispersion in returns on capital investment and, through pecuniary externalities, leads to strategic complementarities and amplification of inefficiencies. Optimal policy tries to reduce distortions by facilitating the allocation of capital towards the most productive firms.

13:45-15:30 Session 12E: Industrial Organization
Chair:
Nathalie Gimenes (PUC-Rio, Brazil)
Location: Mezanino Bossa Nova
13:45
Rodrigo Raad (UFMG, Brazil)
Intertemporal Consistence and Recursivity in a Stackelberg Discounted Stochastic Game
SPEAKER: Rodrigo Raad

ABSTRACT. This article analyzes a Stackelberg environment with a benevolent central planner as a leader and markets with global heterogeneity as a follower. Prescott et al. (Journal of Political Economy 1997 and Journal of Economic Dynamics and Control 1980) point toward an intertemporal inconsistency problem given that central planner's current choices influence all subsequent periods of the game. This work presents conditions on the fundamentals in order to recover the intertemporal consistency even for the case of global heterogeneity. It is also shown the existence of a recursive structure with a minimal state space that implements a Stackelberg subgame perfect equilibrium. The recursive structure also allows the implementation of an algorithm for computing the equilibrium.

14:15
Mateusz Mysliwski (University College London, UK)
Fabio Miessi Sanches (PUC-Rio, Brazil)
Daniel Silva-Junior (City University of London, UK)
Sorawoot Srisuma (University of Surrey, UK)
Dynamic Pricing and Consumer Loyalty in a Multiproduct Oligopoly

ABSTRACT. We study dynamic pricing in a multiproduct oligopoly setting, in which consumers exhibit inertia in their choices and rms face costly price adjustments. Motivated by the empirical observation that most retail price movements are changes between regular (high) and sale (low) price, we formulate the problem as a discrete dynamic game of incomplete information to guarantee equilibrium existence. To recover the structural parameters in rms' payo functions, we use a novel identification strategy that exploits the fact that adjustment costs can be interpreted as lump-sum promotional fees paid by the manufacturers to the retailers. We apply the model to the UK butter and margarine industry and estimate it using a rich scanner data set. The results show that the cost of setting up a price promotion for one product can range from 2.5 to 13% of weekly prots and that the existence of these frictions might be limiting competition from smaller players. Our counterfactual exercises explore the consequences of breaking down the rms into single-product entities for their prots and pricing strategies, as well as the eects of reducing the costs of changing prices for consumer welfare.

14:45
Jayeeta Bhattachary (Queen Mary University of London, UK)
Nathalie Gimenes (PUC-Rio, Brazil)
Emmanuel Guerre (Queen Mary University of London, UK)
A Quantile Approach for Ascending Auctions with Asymmetric Bidders

ABSTRACT. Asymmetry is a relevant question when bidders of different size and technological capability are competing at the same auction. Especially if policy recommendation based on a misspecified symmetric setup causes significant effect on the aution outcome, as for example the seller's expected revenue. Therefore, a parsimonious estimation approach that accounts for bidders asymmetry can be very useful for policy analysis. In this paper, we suggest an identification and estimation approach to recover the private value conditional quantile function when bidders are asymmetric. The estimation strategy is done in two steps: in the first-step, the asymmetric parameters are estimated; and then used in the second-step to estimate the private value conditional quantile functions of each type. Simulations show a good performance of the estimator even in small samples. We also investigate the impact of asymmetry on the optimal screening levels.

15:45-17:30 Session 13A: Sessão Especial Instituto Unibanco

Implicações da flutuação dos resultados de proficiência por escola para tomada de decisão

Chair:
Mirela de Carvalho (Instituto Unibanco, Brazil)
Location: Ametista
15:45
Mirela de Carvalho (Instituto Unibanco, Brazil)
Reynaldo Fernandes (FEA-RP/USP, Brazil)
Ricardo Madeira (FEA/USP, Brazil)
Sergio Firpo (Insper, Brazil)
presenters
15:45-17:30 Session 13B: Sessão Especial FGV-CLEAR & OVE-IDB

Produtividade das Firmas Brasileiras: Como Avançar?

Chair:
André Portela Souza (FGV/EESP, Brazil)
Location: Jade
15:45
José Cláudio Linhares (OVE/BID, United States)
Angelo Duarte (Ministério da Fazenda, Brazil)
Marcelo Fabricio Prim (CNI, Brazil)
presenters

ABSTRACT. O OVE concluiu o estudo “Avaliação de Programas de Apoio a Firmas no Brasil” , que fornece uma perspectiva de diversos programas brasileiros de apoio a empresas – incluindo finanças produtivas, consultoria de negócios, cadeia de valor, promoção de exportação e apoio à inovação – bem como uma avaliação dos efeitos de um subconjunto de programas relacionados a produtividade, emprego e salários reais. Partindo dos principais achados do estudo, os palestrantes apresentarão suas visões acerca da produtividade no Brasil e como avançar para incrementá-la entre as empresas.

15:45-17:30 Session 13C: Sessão Especial FGV EPGE: Prêmio SBE Teoria
Chair:
Wilfredo Maldonado (UCB, Brazil)
Discussants:
Jefferson Bertolai (FEA-RP/USP, Brazil)
Wilfredo Maldonado (UCB, Brazil)
Jaime Orrillo (UCB, Brazil)
Location: Topázio
15:45
Braz Camargo (FGV/EESP, Brazil)
Fabian Lange (McGill University, Canada)
Elena Pastorino (Stanford University, United States)
Earnings Dynamics: The Role of Learning, Human Capital, Accumulation, and Performance Incentives
SPEAKER: Braz Camargo

ABSTRACT. We document evidence from both public individual survey data and proprietary firm personnel records suggesting that the importance of performance pay declines at the end of a career. This evidence is at odds with the implication of models of implicit and explicit performance incentives that have tried to explain the life-cycle profile of the variable component of wages. We provide a novel model that integrates uncertainty and learning about ability, human capital acquisition, and performance incentives to provide a rationale for the declining importance of the pay-for-performance component of wages. We analytically characterize the optimal compensation contract in this environment, derive its qualitative properties, and decompose the pay-for-performance component of the optimal contract into different terms. These terms capture the standard trade-off between risk and incentives arising in situations of moral hazard, the desire to hedge against the risk inherent in learning about ability, and the changing strength of reputational and human capital investment motives over time. We prove the model is identified based on a panel of wages. We derive estimators of the model's primitive parameters, estimate the model, and use the estimated parameters to measure the relative contribution of the three sources of wage dynamics we nest to the life-cycle profile of earnings and variable pay. According to our preliminary estimates, performance incentives seem to be qualitatively and quantitatively key to explaining the life-cycle profile of earnings and variable pay as well as individual wage growth, despite the declining importance of variable pay relative to fixed pay over time.

16:30
Kazuhiro Hara (FGV/EPGE, Japan)
Efe Ok (NYU, Turkey)
Gil Riella (FGV/ENAE, Brazil)
Coalitional Expected Multi-Utility Theory
SPEAKER: Gil Riella

ABSTRACT. This paper begins by observing that any reflexive binary (preference) relation (over risky prospects) which satisfies the Independence Axiom admits a form of expected utility representation. We refer to this representation notion as coalitional minmax expected utility representation. By adding the remaining properties of the expected utility theorem, namely, continuity, completeness and transitivity, one by one, we find how this representation gets sharper and sharper, thereby deducing the versions of this classical theorem in which any combination of these properties are dropped from its statement. This approach also allows us to weaken transitivity in this theorem, rather than eliminating it entirely, say, to quasitransitivity or acyclicity. Apart from providing a unified dissection of the expected utility theorem, these results are relevant for the growing literature on boundedly rational choice in which revealed preference relations often lack the properties of completeness and/or transitivity (but often satisfy the Independence Axiom). Finally, and perhaps more importantly, we show that our representation theorems allow us to answer many economic questions that are posed in terms of nontransitive/incomplete preferences, say, about the maximization of preferences, existence of Nash equilibrium, preference for portfolio diversification, and possibility of the preference reversal phenomenon.

15:45-17:30 Session 13D: Econometrics I
Chair:
Luiz Renato Lima (The University of Tennessee e UFPB, United States)
Location: Mezanino Bossa Nova
15:45
Ricardo Masini (FGV/EESP, Brazil)
Marcelo Medeiros (PUC-Rio, Brazil)
Carlos Carvalho (PUC-Rio e BCB, Brazil)
The Perils of Counterfactual Analysis with Integrated Processes

ABSTRACT. Recently, there has been a growing interest in developing econometric tools to conduct counterfactual analysis with aggregate data when a “treated” unit suffers an intervention, such as a policy change, and there is no obvious control group. Usually, the proposed methods are based on the construction of an artificial counterfactual from a pool of “untreated” peers, organized in a panel data structure. In this paper, we investigate the consequences of applying such methodologies when the data are formed by integrated process of order 1. We find that without a cointegration relation (spurious case) the intervention estimator diverges resulting in the rejection of the hypothesis of no intervention effect regardless of its existence. Whereas, for the case when at least one cointegration relation exists, we have a √ T-consistent estimator for the intervention effect albeit with a non-standard distribution. However, even in this case, the test of no intervention effect is extremely oversized if nonstationarity is ignored. When a drift is present in the data generating processes, the estimator for both cases (cointegrated and spurious) either diverges or is not well defined asymptotically. As a final recommendation we suggest to work in first-differences to avoid spurious results.

16:15
Alan André Borges Costa (FEA/USP, Brazil)
Sergio Firpo (Insper, Brazil)
Saturação e spillover do tratamento para quantis

ABSTRACT. O objetivo deste trabalho consiste em identificar o efeito da saturação do tratamento nos diversos quantis da variável de resultado na presença de spillover. Explorando a variação advinda da aleatorização em dois estágios é proposto um estimador que depende da proporção de tratados permitindo analisar, a nível de quantil, os efeitos diretos, indiretos e totais. A Stable Unit Treatment Value Assumption (SUTVA) é reescrita para se ajustar ao contexto de spillover. Por fim, propõem-se a estimação das derivadas dos quantis de forma flexivel não assumindo forma funcional e não separabilidade do erro.

16:45
Alessandro Casalecchi (none, Brazil)
Sergio Firpo (Insper, Brazil)
Antonio Galvao (University of Arizona, United States)
Improvements for external validity tests in fuzzy regression discontinuity designs

ABSTRACT. When the average treatment effect (ATE) for an entire population equal those for a subpopulation, we say that the estimate of the ATE for this subpopulation is externally valid, in the sense that the estimate is a valid estimate for the entire population as well. Under imperfect compliance researchers are typically able to estimate the local average treatment effect (LATE) but are not able to estimate the ATE for the entire population. This paper provides formal justification for the use of goodness-of-fit (GOF) test statistics to test for external validity of LATE in fuzzy regression discontinuity designs (FRDDs), requiring to this end assumptions of continuity, strict monotonicity and pointwise convergence. We focus on the FRDD, but the reasoning can be extended to other treatment-control settings. The test statistic per se can be anyone used for two-sample GOF tests. To our knowledge, external validity tests proposed so far effectively test for hypotheses involving conditional means (BertanhaImbens, 2016; Angrist and Fernández-Val, 2013), and not entire conditional distributions. Under our assumptions, GOF tests allow for detection of more alternative hypotheses and allow the researcher to learn more about the data generating process, like the joint distribution of potential outcomes and the distribution of unobservable heterogeneity, within every compliance subpopulation. Another advantage of our proposal is that, while the null hypotheses effectively tested by some authors can be true even if external validity is false, under our assumptions equality between conditional distributions hold if and only if external validity is true.

17:15
Luiz Renato Lima (UFPB e The University of Tennessee, United States)
Fanning Meng (ScotiaBank-Toronto, Canada)
Quantile Forecasting with Mixed-Frequency Data

ABSTRACT. We analyze the quantile combination approach (QCA) of Lima and Meng (2017) in situations with mixed-frequency data. Estimation of quantile regressions with mixed-frequency data leads to a parameter proliferation problem, which can be addressed through extensions of the MIDAS and soft (hard) thresholding methods towards quantile regression. We apply the proposed approach to forecast the GDP growth rate and our results show that including high-frequency information into the QCA produces a substantial gain in terms of forecasting accuracy.

15:45-17:30 Session 13E: Macroeconomics III
Chair:
Emerson Marçal (FGV/EESP e Mackenzie, Brazil)
Location: Água Marinha
15:45
Diogo Guillen (Itau Asset Management, Brazil)
Bernardo S. C. Ribeiro (PUC-Rio, Brazil)
A Structural Time-Varying Index for Business Cycle Synchronization in Small Open Economies

ABSTRACT. We provide a new index of business cycle synchronization coming from a time-varying structural vector auto regression of a two country small/large open economy model. The index is composed by the sum of impulse response functions of a small open economy after shocks in the large economy. The transmission of each shock and the size of it is allowed to change across time. We find that this index changes across time in lower and higher frequencies and is correlated with institutional variables, such as capital account openness. Our results shed light on the "decoupling" debate of developing countries after the crisis, suggesting that these countries are not affected by windfalls in developed countries as they used to be.

16:15
Ricardo Sabbadini (BCB e FEA/USP, Brazil)
Overcoming the original sin: Gains from local currency external debt

ABSTRACT. Emerging markets increasingly rely on external debt denominated in local currency. In order to understand if this is a better situation than using US dollar denominated debt, I present a small open economy model with two sectors, and endogenous determination of both default risk and real exchange rate. Using a calibrated version of the model that replicates default frequency and debt level of emerging economies, I compare the two possibilities of debt denomination: local or foreign currency. I find that welfare gains from issuing debt in local currency derive from less frequent defaults, higher sustainable debt levels, and less volatile consumption and real exchange rates. However even an economy issuing debt in local currency still faces counter-cyclical interest rate spread and real depreciations around default episodes.

16:45
Emerson Marçal (FGV/EESP e Mackenzie, Brazil)
Diogo de Prince (UNIFESP, Brazil)
Giovani Merlin (FGV/EESP, Brazil)
Beatrice Zimmermann (World Bank, Brazil)
Addressing econometric issues on how to construct theoretical based exchange rate misalignment estimates: searching for a comprehensive approach.

ABSTRACT. Abstract There is no consensus on what norms apply best to assess real effective exchange rate misalignment. In this paper we highlight the importance of constructing estimates based on a well specified and congruent econometric model satisfying economic restrictions. Moreover, we can discriminate or conciliate different approaches and decompose estimates into economic meaningful pieces. We criticize Behavioral Equilibrium Exchange Rate approach (BEER) for disregarding a priori trade balance information and provide evidence that it is cointegrated with net foreign asset position for a great number of economies. Finally, we assess Brazilian and Australian misalignment and show that trade balance information is important for first but not for second case.

17:45-19:30 Session 14A: Sessão Especial Bradesco: Prêmio SBE Macro Aplicada
Chair:
Marco Bonomo (Insper, Brazil)
Discussants:
Marco Bonomo (Insper, Brazil)
Fábio Gomes (FEA-RP/USP, Brazil)
Eduardo Zilberman (PUC-Rio, Brazil)
Location: Topázio
17:45
Sergio Lago Alves (BCB, Brazil)
Monetary Policy, Trend Inflation and Unemployment Volatility

ABSTRACT. The literature has long agreed that the canonical DMP model with search and matching frictions in the labor market can deliver large volatilities in labor market quantities, consistent with US data during the Great Moderation period (1985-2005), only if there is at least some wage stickiness. I show that the canonical model can deliver nontrivial volatility in unemployment without wage stickiness. By keeping average US inflation at a small but positive rate, monetary policy may be accountable for the standard deviations of labor market variables to have achieved those large empirical levels. Solving the Shimer (2005) puzzle, the role of long-run inflation holds even for an economy with flexible wages, as long as it has price dispersion induced by staggered price setting and search and matching frictions in the labor market.

18:30
Cezar Santos (FGV/EPGE, Brazil)
Tiago Cavalcanti (University of Cambridge e FGV/EESP, UK)
Georgi Kocharkov (University of Konstanz, Germany)
Family Planning and Development: Aggregate Effects of Contraceptive Use
SPEAKER: Cezar Santos

ABSTRACT. What is the role of family planning interventions on fertility, savings, human capital investment, and development? To examine this, endogenous unwanted fertility is embedded in an otherwise standard quantity-quality overlapping generations model of fertility and growth. The model features costly fertility control and families can (partially) insure against a fertility risk by using costly modern contraceptives. In the event of unexpected pregnancies, households can also opt to abort some pregnancies, at a cost. Given the number of children born, parents decide how much education to provide and how much to save out of their income. We fit the model to Kenyan data, implement several family planning policies and decompose their aggregate effects. Our results suggest that given a small budget (up to 0.5 percent of GDP), legalizing and subsidizing the price of abortion is a more cost-effective policy for improving long-run living standards and reducing inequality than policies that either subsidize the price of modern contraceptives or subsidize basic education.

17:45-19:30 Session 14B: Minicurso Itaú - Aula 2
Location: Ametista
17:45
Sergio Almeida (FEA/USP, Brazil)
Economia Comportamental: Desenvolvimentos Teóricos e Implicações para Políticas Públicas