The establishment of democratic regimes in East European countries may obstruct radical economic reforms that hurt the majority of their populations. If the economic gains from reform suffice to compensate the potential losers, institutional change supported by appropriate compensation schemes will be welfare-improving, and such resistance cannot be explained as the behaviour of rational agents. The required compensation payments may prove very costly to the government, however, if it cannot observe the potential losses of different categories of workers, particularly if workers do not believe the government's commitment to compensation schemes.

In Discussion Paper No. 538, Research Fellow Mathias Dewatripont and Gérard Roland assess the possible virtues of a gradualist transition in the case of a sectoral restructuring requiring massive lay-offs and redeployment. If a government cannot distinguish the individual workers' opportunities after the transition, it can induce only the workers with the best opportunities to move by offering better wages, productivity and exit bonuses. Its inability to discriminate between worker types in setting wages or exit bonuses may impose severe costs. If the approval of a majority of workers is required for a reform plan to be accepted, the necessary wage and exit compensations may be very costly to the state budget, so there is a trade-off between allocative efficiency and the budgetary costs of reform. Full reforms achieve allocative efficiency immediately, at the cost of massive and immediate redundancies, which require substantial compensation. Partial reforms, which entail laying off only those workers with the best relative outside opportunities, will generally cost less. When budgetary considerations are sufficiently important to make partial reforms preferable to full reforms, gradualism is the optimal policy; and a government can offer a plan which enables not only workers with `good' but also those with `average' labour market opportunities to redeploy. Allocative efficiency is thus reached gradually, at a lower budgetary cost than under a full or `big bang' reform.

Dewatripont and Roland argue that a government in control of the reform agenda may win majority approval for plans that hurt majority interests intertemporally, if it can use a credible threat of future reforms to extract concessions `today' from groups that will be in the minority `tomorrow' and use their votes to hurt other groups `today'. Such a `divide and conquer' strategy may be profitable for either gradualist or `big bang' reforms. If a democratic government has sufficient legitimacy to maintain full control of the reform process, this may enable it to secure acceptance of transitional measures that hurt a majority of the population, and hence to circumvent the political constraints to the implementation of efficiency-enhancing reforms.

The Virtues of Gradualism and Legitimacy in the Transition to a Market EconomyMathias Dewatripont and Gérard Roland

Discussion Paper No. 538, April 1991 (AM)