4

The aim of this article is to study the impact of the institutional changes of 2001 and 2011 on economic activity on the island of Mayotte. To achieve this, the synthetic control method is used. This method allows us to estimate the best counterfactual for Mayotte’s economy by optimally allocating relative weights to each territory in a control group. To avoid the pitfalls associated with the long-term unavailability of national accounts data, particularly acute for small island economies, economic activity is approximated using harmonized satellite nightlight data. This strategy provides time series starting in the mid-1990s. The application of the synthetic control method successfully identifies credible counterfactuals for Mayotte’s economy, both for the transition to departmental collectivity in 2001 and the transition to departmentalization in 2011. The results suggest that these institutional changes had a positive effect on economic activity. However, the beneficial effects of these changes appear less clear when nightlight are adjusted for the territory’s population. This suggests that the wealth generated has difficulty translating into improved well-being for the local population. Finally, the results obtained are robust to the sensitivity analyses typically applied in the literature.JEL codes: C32, E01, F15, O47, O57.