The paper uses two panel unit root procedures to test stochastic convergence for 14 states in Malaysia, for the period 1960-2003.  Stochastic convergence implies that idiosyncratic country-specific factors cannot explain long-run economic growth, and that shocks to relative regional real per capita Gross Domestic Product (GDP) have temporary effects. In this case, real per capita GDP differentials between states are stationary. Thus, stochastic convergence implies that regional differences across states are not persistent, and that long-run movements in a state's GDP are driven by common technology shocks. The result indicates that the null hypothesis of no convergence in GDP per capita across the 14 states in Malaysia can be rejected. From the policy perspective, this study suggests that previous regional development policies were able to reduce income disparity between the 14 states in Malaysia.  Source: ICFAI Journal of Industrial Economics; May2008, Vol. 5 Issue 2, p21-30, 10p.  Authors: Habibullah, Muzafar Shah & Sivabalasingam, V.