A new research paper titled “Climate Migration in Serbia: Quantification Using a Difference‑in‑Differences Approach” has been published in Stanovništvo (2026). The study, authored by Vera Gligorijević, Damjan Bakić and Filip Mijušković, provides the first systematic panel‑based assessment of whether the May 2014 floods, the most severe hydrological disaster in modern Serbian history, were associated with changes in formally registered internal out‑migration.
Using a balanced municipality‑year panel (2009–2018) and a Difference‑in‑Differences design, the authors compare 40 flood‑affected municipalities with 125 non‑flooded municipalities. The findings show:
- The primary estimate of the differential post‑flood change is positive but imprecise (0.239 per 1,000 residents; 95% CI: –0.423 to 0.901).
- A matched‑sample sensitivity analysis yields a larger estimate, but also with wide confidence intervals.
- A specification allowing NUTS2 region‑specific annual shocks produces a statistically significant positive estimate, highlighting sensitivity to regional time‑varying factors.
- Event‑study and placebo tests show no statistical evidence of differential pre‑flood trends, though confidence intervals remain wide.
- Exploratory analyses suggest heterogeneity: responses may differ depending on pre‑flood demographic structure and labor‑market conditions.
Overall, the study concludes that there is no specification‑invariant evidence of a uniform positive flood effect on registered internal out‑migration. Instead, the estimated association is highly sensitive to the comparison group and regional shocks, underscoring the complexity of climate‑related mobility and the importance of local context.
This research contributes to the growing empirical literature on climate mobility and provides valuable insights for post‑disaster planning, demographic analysis and climate‑migration policy design in Serbia.
