Omitted Variable Bias and Wartime Legacies

A Response to Malesky and Nguyen (2024)

Abstract

Malesky and Nguyen (2024, Journal of Comments and Replications in Economics, MN) reassess a recent study on the long-term effects of wartime violence on civic engagement in Vietnam, attributing discrepancies in previous findings to coding and historical errors. They argue that prewar party strength, rather than wartime exposure, drives Vietnam's contemporary engagement and membership in social organizations. I address this critique by demonstrating that their re-analysis omits a crucial geographic covariate--prewar residence in North or South Vietnam--which is essential for the validity of the original study's quasi-experimental design and included in the model specification by Barcel\'o \citeyearpar{barcelo2021long} and, also, in Miguel and Roland \citeyearpar{miguel2011long}. This omission leads to a violation of the instrumental variable model's exclusion restriction, as the border distance treatment becomes conflated with the South-North distinction due to the asymmetrical distribution of Vietnamese residents across the 17th parallel.

Addressing such complex modeling challenges is crucial to ensure that causal pathways are properly accounted for. In the context of the study, omitting key pre-treatment covariates can lead to misleading conclusions. Once this omission is corrected, the originally reported results withstand all of their proposed modifications. These adjustments include (a) controlling for their measures of prewar party strength, whether zone IV provinces or the birthplace of second committee members, (b) excluding zone IV provinces or provinces of the birthplace of second committee members, (c) reweighting the sample to reflect the distribution of party and non-party members in Vietnam, (d) controlling for respondents who are party members, (e) excluding party members, (f) removing organizational types linked to the VFF, (g) applying various clustering structures for the standard errors, (h) recoding the control variable of South, (i) subsetting the effect only in the North or only in the South; and, (j) controlling for ethnicity.

Thus, this manuscript confirms that wartime exposure indeed enhanced long-term civic engagement and participatory values across Vietnam, countering MN’s critique. Furthermore, it contributes to legacy studies and border design literature by highlighting how improper model specifications can produce biased estimates that obscure rather than clarify our understanding of historical impacts.