Inference in differences-in-differences with few treated groups and heteroskedasticity

Differences-in-Differences (DID) is one of the most widely used identification strategies in applied economics. However, how to draw inferences in DID models when there are few treated groups remains an open question. We show that the usual inference methods used in DID models might not perform well when there are few treated groups and errors are heteroskedastic. In particular, we show that when there is variation in the number of observations per group, inference methods designed to work when there are few treated groups tend to (under-) over-reject the null hypothesis when the treated groups are (large) small relative to the control groups. This happens because larger groups tend to have lower variance, generating heteroskedasticity in the group x time aggregate DID model. We provide evidence from Monte Carlo simulations and from placebo DID regressions with the American Community Survey (ACS) and the Current Population Survey (CPS) datasets to show that this problem is relevant even in datasets with large numbers of observations per group. We then derive an alternative inference method that provides accurate hypothesis testing in situations where there are few treated groups (or even just one) and many control groups in the presence of heteroskedasticity. Our method assumes that we can model the heteroskedasticity of a linear combination of the errors. We show that this assumption can be satisfied without imposing strong assumptions on the errors in common DID applications. With many pre-treatment periods, we show that this assumption can be relaxed. Instead, we provide an alternative inference method that relies on strict stationarity and ergodicity of the time series. Finally, we consider two recent alternatives to DID when there are many pre-treatment periods. We extend our inference methods to linear factor models when there are few treated groups. We also derive conditions under which a permutation test for the synthetic control estimator proposed by Abadie et al. (2010) is robust to heteroskedasticity and propose a modification on the test statistic that provided a better heteroskedasticity correction in our simulations.

Inference in differences-in-differences with few treated groups and heteroskedasticity

Differences-in-Differences (DID) is one of the most widely used identification strategies in applied economics. However, how to draw inferences in DID models when there are few treated groups remains an open question. We show that the usual inference methods used in DID models might not perform well when there are few treated groups and errors are heteroskedastic. In particular, we show that when there is variation in the number of observations per group, inference methods designed to work when there are few treated groups tend to (under-) over-reject the null hypothesis when the treated groups are (large) small relative to the control groups. This happens because larger groups tend to have lower variance, generating heteroskedasticity in the group x time aggregate DID model. We provide evidence from Monte Carlo simulations and from placebo DID regressions with the American Community Survey (ACS) and the Current Population Survey (CPS) datasets to show that this problem is relevant even in datasets with large numbers of observations per group. We then derive an alternative inference method that provides accurate hypothesis testing in situations where there are few treated groups (or even just one) and many control groups in the presence of heteroskedasticity. Our method assumes that we know how the heteroskedasticity is generated, which is the case when it is generated by variation in the number of observations per group. With many pre-treatment periods, we show that this assumption can be relaxed. Instead, we provide an alternative application of our method that relies on assumptions about stationarity and convergence of the moments of the time series. Finally, we consider two recent alternatives to DID when there are many pre-treatment groups. We extend our inference method to linear factor models when there are few treated groups. We also propose a permutation test for the synthetic control estimator that provided a better heteroskedasticity correction in our simulations than the test suggested by Abadie et al. (2010).

Escândalos políticos e o clima econômico dos países sul-americanos: evidências baseadas na confiança de especialistas

Qual o impacto dos escândalos políticos sobre o clima econômico dos países sul-americanos? O presente trabalho busca responder essa pergunta ao avaliar a confiança de especialistas na economia de sete países sul-americanos durante a ocorrência de escândalos políticos em um recorte temporal de 10 anos (de 2005 até 2014). Entendemos os escândalos políticos como sendo eventos noticiados pela mídia envolvendo os presidentes das repúblicas sul-americanas em episódios de corrupção ou abuso de poder. Já o clima econômico é medido a partir da avaliação da economia por especialistas regularmente consultados pela Sondagem Econômica da América Latina, uma pesquisa que gera a construção do Índice de Clima Econômico da América Latina. Evidências apontam a influência de determinantes políticos sobre a avaliação econômica realizada pelo público geral. Poucos estudos exploram o processo de formação da confiança econômica de especialistas. Utilizamos o modelo de regressão em painel para verificar a correlação entre escândalos políticos e o Índice de Clima Econômico. Nenhuma correlação pôde ser verificada quando adotamos um modelo relacionado à economia internacional. Surpreendentemente, encontramos uma correlação significante e positiva quando adicionamos variáveis econômicas domésticas à análise. Acreditamos que futuras contribuições para o tema devam levar em conta a importância do papel das instituições como elemento fundamental na confiança de especialistas.

Exogenous shocks and misattribution of responsibility for economic performance: results from survey experiments

Recent models of economic voting assume that citizens can discount exogenous factors when assessing government's economic performance. Yet there is evidence that Latin American voters do not behave in such way, and attribute to presidents outcomes that are beyond their control. This paper presents three survey experiments designed to explore mechanisms that could potentially correct such misattribution, and therefore contribute to debiasing individual behavior towards government evaluation. Our results provide individual-level evidence of the misattribution found in aggregate studies of electorate behavior, and reinforce psychologist's skepticism towards prospects of mental decontamination, as we found very scant evidence that providing information, raising awareness, or increasing motivation to correct biases infuenced individual's evaluation of president's performance.