Research

Work In Progress

• Bayesian Causal Inference in the Presence of Endogenous Selection into Treatment and Spillovers.

Working Papers

• Causal Inference on Quantiles in High Dimensions: A Bayesian Approach. (2024).

  abstract pdf

  This paper proposes a novel approach, Bayesian Analog of Doubly Robust (BADR) estimation, to estimate unconditional Quantile Treatment Effects (QTEs) in observational studies. By augmenting the proposed estimator with shrinkage priors, this framework can account for high-dimensional covariates and feature a flexible Bayesian modeling strategy with favorable frequentist properties in finite samples, even when either the treatment assignment or outcome models are misspecified. The proposed approach offers a straightforward and adaptable implementation for incorporating probabilistic machine learning techniques to fit the propensity score and conditional cumulative distribution function, followed by combining posterior draws. This enables the effective handling of high-dimensional covariate spaces or nonlinear relationships to achieve better accuracy and appropriate uncertainty quantification. The simulation results show that BADR estimators yield a substantial improvement in bias reduction for QTE estimates compared with popular alternative estimators found in the literature. We revisit the role of microcredit expansion and loan access on Moroccan household outcomes, demonstrating how the new method adds value in characterizing heterogeneous distributional impacts on outcomes and detecting changes in overall economic inequality, which is also appealing to other applied contexts.

• A Comparative Review of Bayesian Shrinkage and Variable Selection in Econometrics. (joint with Dimitris Korobilis, Kenichi Shimizu) (2022).

  abstract notes

• Government disclosure in the pandemic. (2020).

  abstract pdf

  Motivated by the pandemic COVID-19, this paper aims to explore the optimal policy for public information release during an epidemic by employing the framework of Information Design and Bayesian Persuasion. The paper formulates a model of government information disclosure to the public and, based on theoretical analysis, predicts that when the Government possesses commitment power, any partial information disclosure with a partition structure is better than no information disclosure but not as good as full information disclosure, in terms of ex-ante social welfare.

• Early career gender wage gap: First evidence in Vietnam. (joint with Anh Nguyen Ngoc, Hai Doan Ma, Thuc Hoang Kim). Paper presented at 11th Vietnam Economist Annual Meeting (2018).

  abstract

  This paper examines the gender wage gap during the transition from school to work period in Vietnam, utilizing the Oaxaca decomposition method and the Heckman sample selection procedure. Leveraging the unique data from the Vietnam School-to-Work Transition Surveys 2015 (VSWTS 2015) for youngsters, we uncover some meaningful findings about early career wages. Firstly, some determinants of early career wages are confirmed, consistent with human capital theory. Actual working experience, highest level of education, and job tenure significantly impact hourly wages. In line with job shopping theory, the level of job mobility has a negative effect on wages in several cases. Secondly, the analysis reveals a wage gap between young males and females. Although young female workers tend to possess advantageous endowments that positively impact earnings, their average wage is lower than that of their male counterparts. The wage gap even widens when work characteristics are considered. This unexplained wage gap is typically implied as gender discrimination in the literature. Lastly, there is evidence that two gender groups self-select differently into the wage employment sector. Nonemployed women might possess higher productivity-related characteristics versus employed counterparts, and if they worked as salaried employees, they would earn more than actual wage workers on average. Taking this into account slightly reduces the wage gap.

Publications

• Firm export and the impact of foreign ownership in Vietnam: A micro-data analysis. (joint with Vinh Nguyen Thi Thuy), Journal of Economic Development (2020), 45(1).

  abstract pdf

  This paper investigates the impact of foreign ownership on firm exports in Vietnam and analyzes how export participation and export intensity vary with ownership status by comparing Foreign Direct Investment enterprises (FDI enterprises) versus domestic firms, and wholly-foreign-owned enterprises (WFs) versus foreign joint ventures (JVs). Using data from 2010-2015 Vietnamese Enterprise Survey (VES), we document that after controlling for firm characteristics, industry, and region, FDI enterprises have higher export participation and higher export intensity than local firms. The finding supports the hypothesis that FDI enterprises inherit from foreign firms competitive advantages and therefore become superior in exports. We also find that while export participation is similar between JVs and WFs, export intensity is significantly higher for WFs than for JVs. This suggests that export-oriented foreign investors tend to establish 100% foreign-owned companies to exploit advantages of labor costs or natural resources, while domestic-oriented firms tend to form joint ventures to penetrate the domestic market.

• The impact of exchange rate volatility on exports in Vietnam: A bounds testing approach. (joint with Vinh Nguyen Thi Thuy), Journal of Risk and Financial Management (2019), 12(1).

  abstract pdf

  This paper investigates the impact of exchange rate volatility on exports in Vietnam, using data spanning from the opening quarter of 2000 to the final quarter of 2014. The paper applies the autoregressive distributed lag (ARDL) bounds testing approach to the analysis of level relationships between effective exchange rate volatility and exports. Employing the demand function of exports, the paper also considers the effect of depreciation and foreign income on exports in Vietnam. The results unveil that exchange rate volatility negatively affects the export volume in the long run, as anticipated. A depreciation of the domestic currency affects exports negatively in the short run, but positively in the long run, which aligns with the J curve effect. Surprisingly, an increase in the real income of a foreign country actually leads to a decrease in Vietnamese export volume. These findings suggest some policy implications in managing the exchange rate system and promoting exports in Vietnam.