Privatization in Poland
What was the government trying to achieve?¹

Abstract

This article uses the sequencing of privatization to infer the objective pursued by the Polish government in the privatization of its large manufacturing firms in the second half of the 1990s. We construct a model of mixed oligopoly and use it to evaluate the privatization process; our analysis is based on the assumption that firms which furthered the government's objective function the most would be chosen to be privatized first. Based on the features of the firms that were chosen for early privatization, our empirical analysis suggests that welfare maximization was more important than the desire to maximize the revenues from privatization and the government's budget or to minimize employment losses.JEL classifications: D63, L33, P35.1 Authors thank a referee of this journal for many helpful comments on an earlier draft. 532De Fraja and Roberts

Political objectives and privatization decisions

“…, 2007). Contrary to our results, both Dinc and Gupta (2011) and De Fraja and Roberts (2009) find that large firms are more likely to be privatized in India and Poland, although the Polish data contain only the largest SOEs. Gupta et al.…”

“…Szentpéteri and Telegdy (2010) study the same question in Romania by estimating the effect of privatization from real data and constructing a counterfactual effect of non‐privatizable firms. De Fraja and Roberts (2009) and Gupta et al. (2008) study factors that explain sequencing of privatization in Poland and the Czech Republic, respectively.…”

Political Selection of Firms Into Privatization Programs. Evidence From Romanian Comprehensive Data

“…Ã Significant at the 10% level. 11 The 75th percentile of the employment distribution of privatizable firms is larger than the employment size of the median non-privatizable firm, and the 25th percentile of the To illustrate the effects of all variables at the same time, we estimate a probit specification similar to those used in other studies (De Fraja and Roberts, 2009;Gupta et al, 2008;Liu et al, 2007). The dependent variable indicates whether the firm is privatizable or not, and the regressors are firm characteristics in 1992 (employment, value of assets, labor productivity, wage, ROA, net income, overdue payments, and export share) as well as two-digit industry dummies.…”

“…To take the effect of all variables into account at the same time, we estimate a probit specification similar to those used in other studies (De Fraja and Roberts 2008;Gupta et al, 2008;Liu et al, 2007). The dependent variable indicates whether the firm is privatizable or not, and the regressors are firm characteristics in 1992 (employment, labor productivity, wage and industry dummies).…”

“…To assess politicians' objectives, we propose an alternative method to those papers that have analyzed this question with microeconomic data (De Fraja and Roberts, 2008;Dinc and Gupta, 2007;Guo and Yao, 2005;Gupta et al, 2008;Liu and Woo, 2007). Instead of estimating the effects of pre-privatization characteristics on the probability of being selected into the privatizable group (or on sequencing of privatization), we use information on the effect of actual privatizations on firm behavior and simulate the relationship between preprivatization firm characteristics and the effect of privatization on both privatizable and nonprivatizable firms.…”