Optimal two-sided tests for instrumental variables regression with heteroskedastic and autocorrelated errors

This paper considers two-sided tests for the parameter of an endogenous variable in an instrumental variable (IV) model with heteroskedastic and autocorrelated errors. We develop the nite-sample theory of weighted-average power (WAP) tests with normal errors and a known long-run variance. We introduce two weights which are invariant to orthogonal transformations of the instruments; e.g., changing the order in which the instruments appear. While tests using the MM1 weight can be severely biased, optimal tests based on the MM2 weight are naturally two-sided when errors are homoskedastic. We propose two boundary conditions that yield two-sided tests whether errors are homoskedastic or not. The locally unbiased (LU) condition is related to the power around the null hypothesis and is a weaker requirement than unbiasedness. The strongly unbiased (SU) condition is more restrictive than LU, but the associated WAP tests are easier to implement. Several tests are SU in nite samples or asymptotically, including tests robust to weak IV (such as the Anderson-Rubin, score, conditional quasi-likelihood ratio, and I. Andrews' (2015) PI-CLC tests) and two-sided tests which are optimal when the sample size is large and instruments are strong. We refer to the WAP-SU tests based on our weights as MM1-SU and MM2-SU tests. Dropping the restrictive assumptions of normality and known variance, the theory is shown to remain valid at the cost of asymptotic approximations. The MM2-SU test is optimal under the strong IV asymptotics, and outperforms other existing tests under the weak IV asymptotics.

New evidence on trade liberalization and productivity growth

Although the subject of a large number of studies, the debate on the links between trade reform and productivity growth is still unresolved and most studies at the micro level have not been able to establish a relationship between the two phenomena. Brazil provides a natural experiment to study this issue that is seldom available: it was one of the closest economies in the world until 1988, when trade reform was launched, and intra-industry data are available on an annual basis before, during and after liberalization. Using a panel of industry sectors this paper tests and measures the impact of trade reform on productivity growth. Results confirm the association between the former and the latter and show that the magnitude of the impact of tariff reduction on the growth rates of TFP and output per worker was substantial. Our data reveal large and widespread productivity improvement, so that the estimations in this paper are an indication that liberalization had an important effect on industrial performance in the country. Cross-sectional differences in protection are also investigated.

Labor rights, formality and spillovers: evidence from Brazil

Nós analisamos o efeito da emenda constitucional 72/13 no Brasil, que igualou direitos trabalhistas de empregadas domésticas a aqueles de outros empregados. Mostramos que, após a legislação, uma considerável cobertura midiática e um interesse público intensificado aumentou o conhecimento geral de direitos trabalhistas de empregadas domésticas. Como consequência, o não-seguimento de legislações trabalhistas no setor de serviços domésticos ficou mais difícil. Ao mesmo tempo, a necessidade de regulamentar adicionalmente a emenda fez com que custos trabalhistas ficassem praticamente inalterados. Usando uma abordagem de diferença-em-diferenças que compara ocupações selecionadas ao longo do tempo, mostramos que a emenda -- e a discussão que ela causou -- levou a um aumento na formalização e nos salários de empregados domésticos. Então, usando a heterogeneidade do impacto da emenda em grupos demográficos, nossos resultados mostram que emprego doméstico foi reduzido e que mulheres pouco qualificadas saíram força de trabalho e foram para empregos de menor qualidade. Testes de placebo e análises de robustez indicam que nossos resultados não são explicados por diversas interpretações alternativas.