Entropy Balancing for Continuous Treatments

Interest in evaluating the effects of continuous treatments has been on the rise recently. To facilitate the estimation of causal effects in this setting, the present paper introduces entropy balancing for continuous treatments (EBCT) – an intuitive and user-friendly automated covariate balancing scheme – by extending the original entropy balancing methodology of Hainmueller, J. 2012. “Entropy Balancing for Causal Effects: A Multivariate Reweighting Method to Produce Balanced Samples in Observational Studies.” Political Analysis 20 (1): 25–46. In order to estimate balancing weights, the proposed approach solves a globally convex constrained optimization problem, allowing for computationally efficient software implementation. EBCT weights reliably eradicate Pearson correlations between covariates (and their transformations) and the continuous treatment variable. As uncorrelatedness may not be sufficient to guarantee consistent estimates of dose–response functions, EBCT also allows to render higher moments of the treatment variable uncorrelated with covariates to mitigate this issue. Empirical Monte-Carlo simulations suggest that treatment effect estimates using EBCT display favorable properties in terms of bias and root mean squared error, especially when balance on higher moments of the treatment variable is sought. These properties make EBCT an attractive method for the evaluation of continuous treatments. Software implementation is available for Stata and R.

An assessment of the Extractive Industries Transparency Initiative (EITI) using the Bayesian corruption indicator

Advocated across the international community for more than 15 years, the Extractive Industries Transparency Initiative (EITI) is now widely recognised as a hallmark anti-corruption scheme in the extractive sector. This study presents an assessment of the relationship between EITI membership and countries’ progress in tackling corruption. It provides the first study that looks at this issue using a ‘state-of-the-art’ indicator called the Bayesian Corruption Indicator. It also introduces an innovative estimation strategy combining entropy balancing with a difference-in-difference framework to address the baseline inequalities that exist between member and non-member countries. Contrary to the findings of many leading studies, this analysis finds corruption scores have improved significantly among EITI member countries. In particular, the evidence is strongest when we examine a sub-group of EITI members designated fully compliant with the initiative's transparency standards.

Efectos del Programa de Adquisición de Alimentos (PAA) sobre los Apicultores del Estado de Ceará, Brasil: Un Abordaje por Entropy Balancing

El Programa brasileño de Adquisición de Alimentos (PAA) compra productos de la agricultura familiar y los distribuye entre las personas socialmente vulnerables. Este trabajo busca evaluar los efectos del PAA sobre la producción y los ingresos de la actividad apícola en Ceará, Brasil. Se utilizó la metodología Entropy Balancing con regresiones OLS y se tuvieron en cuenta las características de los productores de miel participantes y no participantes del programa en 2011. Los resultados revelaron impactos positivos y significativos sobre los aspectos económicos para los beneficiarios del programa. Además, el total de colmenas pobladas es un mecanismo de transmisión importante para explicar estos efectos.

Does working at a start-up pay off?

Using representative linked employer-employee data for Germany, this paper analyzes short- and long-run differences in labor market performance of workers joining start-ups instead of incumbent firms. Applying entropy balancing and following individuals over ten years, we find huge and long-lasting drawbacks from entering a start-up in terms of wages, yearly income, and (un)employment. These disadvantages hold for all groups of workers and types of start-ups analyzed. Although our analysis of different subsequent career paths highlights important heterogeneities, it does not reveal any strategy through which workers joining start-ups can catch up with the income of similar workers entering incumbent firms.

175Bleeding and cardiovascular risk of New Oral Anti-Coagulants (NOACs) versus warfarin in type 2 diabetes

Background New oral anticoagulants (NOACs) (vs warfarin) in the general population are associated with 34% less major bleeding events. We compared the risk of bleeding and cardiovascular disease (CVD) events between NOACs and warfarin in people with type 2 diabetes (T2DM). Methods Incident NOAC users (cases) and matched controls (incident warfarin users) with T2DM were identified from within 40 UK general practice records (1/4/2017-31/3/2018). Coarsened Exact Matching restricted the comparison to areas of sufficient overlap in characteristics (i-iv) and missingness. Entropy Balancing sequentially balanced NOAC and warfarin users on their distribution of (i) age, sex, general practice, first anticoagulant prescription year, diabetes duration, (ii) body mass index, systolic blood pressure, total cholesterol, HbA1c, (iii) bleeding/CVD history, (iv) treatment (Antiplatelet, Corticosteroids, statins, anti-hypertensives, anti-diabetes treatment number). Continuous covariates were balanced for mean, variance, and skewness. Pre-processing/balancing occurred without reference to outcomes. Multivariable logistic regression modelling estimated bleeding and CVD relative risks (RRs: incident non-elective hospitalisation within 12 months since first anticoagulant prescription), identified from linked hospitalisation data (Secondary Uses Service dataset), using entropy balancing weights from steps i-iv. Results The twelve month RRs of bleeding with NOAC (n = 967) vs matched/balanced warfarin (n = 911) were 0.94(0.62-1.45), 0.95(0.62-1.46), 0.98(0.65-1.48) and 0.98(0.67-1.43) through steps i-iv. CVD RR was increased with NOAC treatment through steps i-iv: 1.15(1.00-1.31), 1.22(1.06-1.39), 1.34(1.13-1.60) and 1.43(1.20-1.70). Conclusion Among T2DM patients, risk of bleeding among NOAC and warfarin incident users was similar. However, NOAC use significantly increased the risk of CVD events. Further external replication studies are warranted.

Rural extension and technical efficiency in the Brazilian agricultural sector

In this paper, we identify the effect of rural extension on the productive performance of Brazilian agricultural establishments, using technical efficiency as a measure of farm performance. The data used is drawn from the microdata of the 2006 Agricultural Census, accessed directly from the IBGE secrecy room. For this, we use an approach that combines the stochastic production frontier, considering selection bias in the adoption of rural extension (Heckman’s approach) and entropy balancing. We find that rural extension increases efficiency in the use of the productive inputs, with more technical efficiency found among adopting producers than non-adopters. When considering the differences according to farm size, an even greater effect is observed for larger producers. In addition, public rural extension generates higher technical efficiency scores than those obtained by privately-operated establishments.