Abstract

Binscatter is very popular in applied microeconomics. It provides a flexible, yet parsimonious way of visualizing and summarizing large data sets in regression settings, and it is often used for informal evaluation of substantive hypotheses such as linearity or monotonicity of the regression function. This paper presents a foundational, thorough analysis of binscatter: we give an array of theoretical and practical results that aid both in understanding current practices (i.e., their validity or lack thereof) and in offering theory-based guidance for future applications. Our main results include principled number of bins selection, confidence intervals and bands, hypothesis tests for parametric and shape restrictions of the regression function, and several other new methods, applicable to canonical binscatter as well as higher-order polynomial, covariate-adjusted and smoothness-restricted extensions thereof. In particular, we highlight important methodological problems related to covariate adjustment methods used in current practice. We also discuss extensions to clustered data. Our results are illustrated with simulated and real data throughout. Companion general-purpose software packages for Stata and R are provided. Finally, from a technical perspective, new theoretical results for partitioning-based series estimation are obtained that may be of independent interest.

Professor of Operations Research and Financial Engineering (ORFE) at Princeton University. His research spans over mathematical statistics, econometrics, data science, and quantitative methods in the social, behavioral and biomedical sciences, with particular interest in program evaluation and causal inference. He currently serves in the editorial boards of the Journal of the American Statistical Association, Econometrica, Operations Research, the Review of Economics and Statistics, Journal of Business & Economic Statistics, Econometric Theory, the Econometrics Journal, and the Journal of Causal Inference.

-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------

Abstract

Are teachers’ impacts on students’ test scores (“value-added”) a good measure of their quality? This question has sparked debate because of disagreement about (1) whether value-added (VA) provides unbiased estimates of teachers’ impacts on student achievement and (2) whether high-VA teachers improve students’ long-term outcomes. I present the results from published research analyzing school district data from grades 3-8 for 2.5 million children linked to tax records on parent characteristics and adult outcomes. First we test for, and find little evidence of, bias in VA estimates. Then we show that students assigned to high-VA teachers have better outcomes in a variety of ways, such as being more likely to attend college and earning higher salaries. We estimate that replacing a teacher whose VA is in the bottom 5% with an average teacher would increase the present value of students’ lifetime income by more than $250,000 for the average classroom in our sample. Additionally, I discuss recent findings regarding other measures and dimensions of teacher quality and their relationship with students’ long-term outcomes.

Professor of Economics at the Columbia Graduate School of Business and a Research Associate at the National Bureau of Economic Research. Professor Rockoff’s interests center on the finance and management of public schools. He received his Ph.D. in Economics from Harvard University and a B.A. in Economics from Amherst College.

-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------

Abstract

We study a propaganda campaign sponsored by the government against the main political challenger in the days preceding the 2015 Argentine presidential ballotage. Subjects in the treatment group watched an “ad” initially aired during soccer transmissions that was part of this campaign and were then asked about their political views. Relative to subjects in the control group, their declared preference for the challenger drops by 6.5 percentage points. We find no effects of the three types of defenses employed by the challenger (a positive message unrelated to the “ad”, an answer to the accusations in the “ad”, and a counter-attack). The propaganda effect is mainly driven by women.

Ph.D. in Economics, Harvard University. Fue rector de UTDT, previamente decano de la Escuela de Negocios, profesor visitante en Stanford e investigador visitante en Harvard. Sus estudios analizan el efecto del otorgamiento de títulos de propiedad de la tierra en áreas marginales, el impacto de la presencia policial sobre el crimen, el efecto de la utilización de sistemas de monitoreo electrónico de detenidos sobre la reincidencia criminal, el efecto de la privatización de las empresas de agua sobre la mortalidad infantil, el análisis del apoyo a las privatizaciones en la opinión pública, el impacto del servicio militar obligatorio sobre el delito y la relación entre salarios de los funcionarios públicos y la corrupción. Sus trabajos han sido publicados en American Economic Review, Journal of Political Economy, Quarterly Journal of Economics, American Economic Journal, Journal of Law and Economics, entre otros.

-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------

Factores disposicionales y situacionales en el trabajo: validación de escalas y análisis de sus influencias sobre la satisfacción laboral

Abstract 

Este artículo persigue tres objetivos. En primer lugar, examina las propiedades psicométricas de las versiones en español de la Escala de Autoevaluaciones Esenciales (i.e., Core Self-Evaluations Scale, CSES) y el Índice Breve de Satisfacción Laboral Afectiva (i.e., Brief Index of Affective Job Satisfaction, BIAJS) en términos de consistencia interna, estructura factorial, validez convergente, validez discriminante y validez empírica. En segundo lugar, analiza y compara la influencia relativa de un conjunto de factores disposicionales (i.e., autoevaluaciones esenciales, CSE) y situacionales (i.e., factores psicosociales del trabajo, FPST) sobre la satisfacción laboral. En tercer lugar, examina los efectos indirectos (i.e., mediados) de ambos conjuntos de determinantes sobre la satisfacción laboral a través del contraste de modelos alternativos de ecuaciones estructurales. A tales fines, se recolectan y analizan datos provenientes de dos muestras independientes de profesionales argentinos pertenecientes a ocupaciones altamente demandantes (209 académicos y 116 directivos de nivel medio y alto). Los resultados revelan que ambas escalas presentan propiedades psicométricas satisfactorias y consistentes con las reportadas en contextos laborales no hispano-parlantes, lo que brinda evidencia sobre la universalidad y validez transcultural de los constructos involucrados. Por otro lado, los análisis de regresiones múltiples demuestran que tanto las CSE como dos de los FPST, i.e. la autonomía y la estima, se relacionan significativamente con la satisfacción laboral. Más aún, los resultados de los análisis de ecuaciones estructurales muestran que aquellos empleados con CSE más elevadas tienden a percibir las características de su trabajo de manera más favorable (i.e., menor exposición percibida a riesgos psicosociales), lo que los vuelve más proclives a experimentar mayores niveles de satisfacción laboral. Estos hallazgos aportan evidencia al controversial debate individuo-situación y demuestran que la inclusión simultánea tanto de factores situacionales como disposicionales (i.e., perspectiva interaccionista) resulta fundamental en futuras investigaciones que examinen la satisfacción laboral. Se discuten las implicancias de este estudio para la práctica profesional y se proponen líneas futuras de investigación. 

Es doctor en Administración, MBA, especialista en Docencia Universitaria, investigador de posdoctorado en CONICET, investigador asociado en la Universidad Católica del Maule (Talca, Chile), profesor de Metodología de la Investigación, Métodos Cuantitativos y Taller de Tesis en el MBA de la Facultad de Ciencias Económicas y Sociales de la U. Nacional de Mar del Plata, referato externo en Academia: Revista Latinoamericana de Administración (Colombia), integrante del Grupo de Investigaciones en Administración (Facultad de Ciencias Económicas, U. Nacional de Mar del Plata) y del Grupo de Investigación en Psicología Laboral (Facultad de Psicología, U. Nacional de Mar del Plata), colaborador externo del proyecto Anillos SOC180040 – Conicyt Chile (2019-2021).

-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------

Abstract

In negotiations the objectives of parties are generally in conflict. Facing this conflict can trigger negative emotions: A sentiment of nervousness, embarrassment or awkwardness. In this paper, by means of a lab experiment, I explore the existence and implications of these confrontation costs. First, I show that a significant proportion of participants avoid bargaining even when it delivers higher payoffs. I find that the avoidance rate is 50% higher in face-to-face negotiations relative to electronic negotiations. This offsets a higher rate of agreements conditional on bargaining, making the unconditional probability of agreement equal across communication channels. Second, after shutting down alternative channels, I attribute the decrease in bargaining propensity to an increase in confrontation costs. Both things together make e-negotiations welfare improving in my design, casting doubts on the general belief that face-to-face communication increases efficiency by fostering transactions. Finally, consistent with previous literature, I observe that women haggle less than men, and I find that confrontation costs are a reason to explain it.

Ph.D. Candidate in Economics, CEMFI.

-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------

Abstract

I empirically investigate the role of downstream competition in the use of exclusive dealing and quantify its effects in the formation of the car retailing networks. I estimate a model where exclusivity impacts both supply and demand, and dealers choose which brand to offer in a strategic manner. These choices frame product and brand availability in the market and the retail networks for manufacturers. In my model, dealers have incentives to add more brands in order to sell a wider set of products, but their interest to differentiate from local rivals limits this option. Moreover, manufacturers could raise costs anticompetitively to deter dealers from selling products of rival brands. I analyze the potential for this foreclosing channel by estimating fixed cost differences between exclusive and non-exclusive stores using moment inequalities. I find that multi-dealing has an average cost advantage between 10,000 and 620,000 euros. These numbers indicate that downstream competition, instead of anticompetitive motives, explains a more substantial part of the prevalence of exclusive dealing in the market.

Ph. D. candidate in Economics, University of Mannheim.

---