Abstract
This study examines income convergence across Turkish provinces over the period 2004-2023 by combining traditional convergence analysis with spatial econometric methods. Both unconditional and conditional convergence models are estimated using provincial data. To account for structural heterogeneity across regions, the empirical framework incorporates net migration rates, trade openness, educational attainment, and digital infrastructure measured by per-capita fiber-optic cable length. The results show that income convergence slowed considerably over time and varied across sub-periods. The estimated speed of convergence declined from 1.77% per year during 2004-2013 to 0.76% during 2013-2023, while convergence is not statistically significant for 2020-2023. Moran’s I values remain high (0.64 0.69) throughout the sample period, indicating persistent positive spatial dependence among provinces. Spatial panel models further reveal significant regional spillovers, with the spatial autoregressive coefficient ranging from 0.295 to 0.776 across estimation periods. Overall, the findings suggest that regional income dynamics in Türkiye are shaped by both structural factors and spatial interactions, highlighting the need for region-specific and spatially coordinated development policies.

This publication has 23 references indexed in Scilit: