Traditional political economy emphasizes the difficulty of conducting simultaneous transitions toward market economy and democratic government. There are two major theories that seek to explain why some reform programs are never fully implemented or are reversed shortly after their inception. The J-Curve model (JCM) (Przeworski 1993) implicates the short-term losers from reform as the major opposition, and the Partial Reform Equilibrium model (PREM) (Hellman 1998) implicates the winners. I subject the models to empirical analysis with data from 25 post-communist countries and find that the data do not support the contention of the JCM. High unemployment rates do not threaten the survival of reform programs, and government instability does not necessarily translate into bad economic policies. These results suggest that the common concern that socially costly economic reforms endanger the consolidation of democratic norms may be misplaced.
During periods of strained economic circumstances, voters who are most hurt by their governments policies will punish elected officials by removing them from office and replacing them with ones more likely to enact policies sympathetic to the voters' plight. Because the benefits of economic reforms are dispersed while the costs are concentrated, disadvantaged voters are likely to be effective in undermining economic reform efforts by destabilizing their government under democratic regimes.
The alternative view holds that economic reforms generate a pattern of concentrated benefits and diffuse costs, which enables the groups most favored by the reform to capture the government and freeze the reform programs in a state most beneficial to them.
Despite many case-studies and statistical analyses of related questions, there has been no attempt to construct a statistical test that will examine the hypotheses generated by the two different models directly. This article bridges this gap and addresses several questions: Are the losers from economic reforms threatening to the progress of these reforms? Does their influence vary across countries? What are the policy implications that we can derive from the results? Can a democracy sustain socially costly economic reforms?
The results from the statistical analysis of 25 former communist countries are strong, internally consistent, and startling: democracies do universally better than nondemocracies when unemployment is low; when unemployment is high, democracies do as well as nondemocracies in short-term reforms, but consistently outperform nondemocracies in long-term reforms. Government stability has no impact on the performance of democracies, but has a weak beneficial impact on the performance of nondemocracies. As a whole, the results provide strong support for the view that losers from reform do not endanger the success of reforms, and that it is possible (and probably necessary) to create democratic institutions for these reforms to succeed.
These results have special relevance in theorizing about simultaneous political and economic transitions in former communist countries. Traditionally, it is assumed that building and maintaining a stable democratic regime require an advanced capitalist society. The problem in former communist states is especially acute because their governments are conducting a transition from command to market economy while building democratic norms and institutions at the same time. Depending on which of the two theoretical views about the pattern of gains and losses is correct, one would prescribe diametrically opposed policies. In the first case, one would seek to isolate the government from the pressure of reform losers to enable it to conduct economic policies that hurt a substantial segment of voters. In the other case, one would attempt to open the government to voter pressure and minimize the influence of reform winners.
MODELS OF SIMULTANEOUS TRANSITIONS
Whatever the long-term implications for sustainable economic growth and high living standards, the immediate effect of reforms is unemployment, rampant inflation (Marer and Zecchini 1991), resource misallocation1 (Roland 1994), volatility in income distribution (Milanovic 1995), declining output (Kolodko 1999), and a faltering social safety net (DeMelo, Denizer, and Gelb 1996). …