During the last decades, the regional convergence process in Europe has attracted a considerable interest as a highly significant issue, especially after EU enlargement with the New Member States from Central and Eastern Europe. The most usual empirical approaches are using the β‑ and σ‑convergence, originally developed by a series of neo-classical models. Up‑to-date, the EU integration process was proven to be accompanied by an increase of the regional inequalities. In order to determine the existence of a similar increase of the inequalities between the administrative counties (NUTS3) included in the NUTS2 and NUTS1 regions of Romania, this paper provides an empirical modelling of economic convergence allowing to evaluate the level and evolution of the inter-regional inequalities over more than a decade period lasting from 1995 up to 2011. The paper presents the results of a large cross-sectional study of σ‑convergence and weighted coefficient of variation, using GDP and population data obtained from the National Institute of Statistics of Romania. Both graphical representation including non-linear regression and the associated tables summarizing numerical values of the main statistical tests are demonstrating the impact of pre-accession policy on the economic development of all Romanian NUTS types. The clearly emphasised convergence in the middle time subinterval can be correlated with the pre-accession drastic changes on economic, political and social level, and with the opening of the Schengen borders for Romanian labor force in 2002.