Purpose: This study measures the local extinction index for Korean local governments and analyzes factors associated with the index. We also examines how the estimation results change when the elderly population ratio is included in or excluded from the model. This is because the index incorporates elderly population information.Research design, data, and methodology: Using official KOSIS statistics, we construct a four-year (2021–2024) panel dataset of 243 local governments (972 observations). We estimate panel regression models and adopt a fixed-effects specification based on the F-test and the Hausman test.Results: In Model 1, elderly population ratio, old housing ratio, and crude marriage rate are negatively associated with the local extinction index, while fiscal self-reliance ratio and total fertility rate are positively associated with the local extinction index. Green space ratio is not statistically significant. In Model 2 (excluding elderly population ratio), the signs and significance of old housing ratio, crude marriage rate, and total fertility rate remain unchanged, but their coefficient magnitudes increase. Fiscal self-reliance ratio becomes less statistically significant, and green space ratio remains statistically insignificant.Implications: This study explicitly highlights the potential overlap between the local extinction index and the covariates. It also compares models with and without the elderly population ratio to assess whether the results reflect substantive associations or mechanical components arising from the index's construction. This approach examines how sensitive the findings are to model specification and suggests that the results should be interpreted with caution. From a policy perspective, raising crude marriage rates alone is unlikely to be sufficient. Policies should prioritize the conditions that convert marriage into sustained settlement and fertility—especially stable housing, quality jobs, and childcare support. The finding on the crude marriage rate also suggests that local extinction risk is closely tied to broader regional socio-economic structures rather than a single demographic event.
전주열 et al. (Sat,) studied this question.