We focus on a sample of Colombian researchers active between 1990 and 2021. We build a unique dataset, reconstructing mobility patterns and scientific performance from national centralized CV repository data and national funding agencies’ records, combined with bibliometric data from OpenAlex. Our empirical strategy combines two methods – Nearest Neighborhood Matching (Rubin, 1973) and a Difference-in-Differences with multiple time periods estimator (Callaway & Sant’Anna, 108 2021). Establishing a causal link between co-authoring with a foreign-educated researcher and scientific outcomes (if any) for non-mobile researchers is challenging because these foreign-educated researchers do not randomly select their co-authors. We leverage Propensity Score Matching (PSM) to build comparable groups of non-mobile researchers who co-authored with at least one foreigneducated researcher (treated) and those who never co-authored (control). We use variables on common performance and socio-demographic characteristics (Jones et al., 2008; Yadav et al., 2023). We next use a Difference-in-Differences estimator to assess the dynamic effects (i.e., an aggregation scheme that observes how the effect changes over time given exposure to the treatment) of such co-authorships, with the first co-authorship with the foreign-educated researcher as the treatment year. Our analysis focuses on four outcome variables: the number of publications, the cumulative number of citations per publication, the number of publications in top journals (all logged), and the share of publications in English.