The economic effect of splitting a region in a centralized country: A case from Chile

• Provides a systematic analysis of region splitting in a centralized context. Previous studies find growth gains from territorial splits primarily in highly decentralized countries; this paper studies Chile’s 2007 reform at a time when regional governments had no fiscal or legislative autonomy. It studies also the effect of providing new administrative status such as regional and provincial capital to cities. • Robust quasi-experimental approach. The paper combines synthetic control, difference-in-differences, and event-study methods to isolate causal effects on economic activity, carefully controlling external shocks. • At the regional level, the new regions did not imply a boost in economic activity. • Heterogeneous effects at municipal level. Granting provincial-capital status generated large increases in economic activity in La Unión, a small, low-urbanization city which aligns with current research. • Administrative boundary changes alone do not spur growth; they must be paired with providing more competence to local government. The economic impact of territorial reforms has been discussed in the literature but mostly looking at amalgamations rather than de-amalgamations. Little is known about the impact of territorial splits and even less in the context of a centralized and emerging country. This paper addresses this gap by analyzing the 2007 territorial reform in Chile, where two new regions were created at opposite ends of the country. By using a synthetic control approach, a difference in differences and event studies, the causal effect of such reform over GDP and private sales on the affected regions is estimated, including the effect of the new administrative status for the cities of Valdivia, La Union, Arica and Putre. After accounting for all potential external shocks, no significant impact of the division on the affected regions was found. However, at the municipal level, while regional capital status had no significant effect, transitioning from a regular municipality to a provincial capital had a notable positive impact in a small, low-urbanization city such as La Union. We believe these results are very important as illustrate that policies aimed at splitting regions or relocating the capital status to promote growth should consider the context and the scope of competences of the subnational governments as well as the socioeconomic characteristics of the places.