Priorities and Sequencing in Privatization: Theory and Evidence from the Czech Republic

While privatization of state-owned enterprises has been one of the most important aspects of economic transition from a centrally planned to a market system, no transition economy has privatized all its firms simultaneously. This raises the issue of whether governments strategically privatize firms. In this paper we examine theoretically and empirically the determinants of the sequencing of privatization. First, we develop new and adapt existing theoretical models in order to obtain testable predictions about factors that may affect the sequencing of privatization. In doing so, we characterize potentially competing government objectives as (i) maximizing sales revenue from privatization or public goodwill from transferring shares of firms to voters, (ii) increasing economic efficiency, and (iii) reducing political costs due to layoffs. Next, we use an enterprise-level data set from the Czech Republic to test the competing theoretical predictions about which firm characteristics affect the sequencing of privatization. We find strong evidence that more profitable firms were sold first. This suggests that the government sequenced the sale of firms in a way that is consistent with our theories of sale revenue maximization and/or maximizing public goodwill from subsidized share transfers to citizens. Our results are also consistent with Shleifer and Vishny's (1994) prescription for increasing efficiency when there are political costs to employment losses caused by privatization. We also find that the Glaeser-Scheinkman (1996) recommendations for increasing efficiency by privatizing first firms subject to large informational shocks are consistent with our results. Finally, our findings are inconsistent with the government pursuing a static Pareto efficiency objective. In addition to enhancing the general understanding of privatization, our evidence suggests that many empirical studies of the effects of privatization on firm performance may suffer from selection bias since privatized firms are likely to have observable and unobservable characteristics that make them more profitable than firms that remain under state ownership.