Regional Convergence in Tunisia: Evaluating the New District Structure

Tunisia's 2023 presidential decree reorganized the country's 24 governorates into five horizontal districts, each deliberately pairing coastal and interior units under shared governance-an institutional response to a coastal-interior divide that has structured the country's economic geography since antiquity. This paper asks whether that regrouping is coherent with Tunisia's actual spatial economy, using a balanced panel of 24 governorates over 2002-2021 and a productivity measure built from confidential firm-level turnover data at the National Institute of Statistics, supplemented with satellite-derived built-up area and nighttime light intensity as infrastructure proxies. A two-way fixed-effects convergence regression finds robust conditional β-convergence in private-sector productivity: interior governorates are catching up to coastal ones, at a 5-year-interval-adjusted speed of roughly 12% a year and a half-life under six years. Spatial error models estimated under both a geographic contiguity matrix and a matrix encoding the district reform's own boundaries detect significant spatial dependence in the residuals-but only at medium-run (5-year) horizons, and only through correlated shocks rather than direct spillovers across governorate lines. Sigma convergence tests within each of the five districts complicate the picture further: only the Greater Tunis district shows significant internal convergence, while two districts straddling the coastal-interior line are significantly diverging internally. The reform's spatial logic is empirically defensible-but coherence with existing geography is not the same as a guarantee of convergence, and the districts most in need of active policy support are precisely the ones showing the least sign of closing internal gaps on their own.