Does WTO membership influence unrelated economic outcomes? This article replicates ninety-four specifications from sixteen different studies, revealing that the mere existence of a country in the WTO has statistically significant associations with approximately 34 percent of dependent variables, many lacking theoretical justification. Similarly high false positive rates emerge when examining membership in a low-impact environmental treaty (CITES). The authors argue this highlights a critical issue: without theoretically informed sensitivity analysis, seemingly robust findings can mask selection on unobservables.

Research Design & Methodology

The study employs replication techniques across diverse political science research specifications. It specifically examines econometric relationships involving international institutions by reanalyzing previously published empirical models using the same dataset but adjusting for potential biases.

Key Findings: High False Positive Rates

The analysis demonstrates two crucial points:

* A statistically significant relationship exists between WTO or CITES membership and many dependent variables, despite a lack of theoretical connection to these institutions' core mandates.

* These findings are not unique to the WTO but also manifest in other institutional contexts.

This high incidence of spurious relationships raises concerns about how international memberships affect research conclusions in political economy.