The transition from socialism to capitalism, and the comprehensive transformation of the economic institutions of the countries concerned, is surely one of the most important economic events of the twentieth century. According to Gérard Roland of the Free University of Brussels, speaking at an IMF Institute seminar on July 6-7, future economic historians will probably remember the twentieth century for the contest that took place between the socialist and the capitalist economic systems.
The large-scale institutional changes that transition requires are among the most complex economic and social processes one can imagine, Roland said, but because the collapse of central planning in many countries was sudden and unexpected, no theories or empirically tested models for dealing with transition existed to which policymakers and strategists could turn for guidance. Given these complexities and the important stakes for hundreds of millions of people, economists cannot claim to know the right answers. The experience in policy advice to transition countries, Roland acknowledged, has been "particularly humbling." It is important, he said, that research on transition issues be collated and systematized for the benefit of policymakers in transition countries as well as for the insights such research can provide to the study of economics at large.
Roland devoted the first part of the seminar to the political economy of reforms. The second part focused on the role of soft budget constraints (see box, pages 232-33).
Transition is an economy-wide process involving winners and losers. Thus, even if aggregate welfare is enhanced, political constraints are of crucial importance. These constraints, Roland said, have played an important role in the transition process in many countries, influencing both the design and the implementation of policy and affecting its outcome as well. Privatization strategies, for example, have differed markedly, depending on the political situation. In the Czech Republic, privatization was undertaken quickly, the object being to distribute assets and put ownership into the hands of outsiders as quickly as possible, leaving the new owners to undertake the necessary restructuring. In Russia, assets and ownership of privatized state enterprises were merely redistributed among the insiders, who were too strong a faction to oust.
Do political constraints matter? Some analysts have suggested that the efficiency gains from a reform program could be used to compensate the losers. If everyone is convinced that the outcome of a full-steam-ahead reform will be positive, then no one will attempt to derail or delay the reform. In this case, political constraints should not affect the speed and sequencing of reforms, only the level of compensation. The main flaw in the theory of compensation from efficiency gains, Roland noted, is that uncertainty about the outcome—either individual or aggregate—can lead to strong resistance from potential losers that may stall or subvert the reform process. In actual experience, political constraints appear to have played a major role in the transition process. In Russia, for example, the Duma has blocked stabilization plans, and continued stalemate over key issues of reform has been a characteristic of reform politics. In all transition countries, political constraints have also played an important part in blocking parts of the restructuring process—for example, drastic cuts in state-owned enterprises and heavy worker resistance to closing inefficient state enterprises. For this reason, Roland said, political constraints do matter.
Ex ante and ex post political constraints
Resistance to reform can block decision making before a reform is implemented (ex ante political constraint), Roland said, or act as backlash after decisions have been made and outcomes observed, leading to pressure for reversals (ex post political constraints). Reformists need to deal with each type of constraint differently. Ex ante constraints imply either that compromises must be made on reform plans, that credible compensation for losers must be devised, or that decisions on radical programs must be delayed. Ex post political constraints require that irreversibility be built into the reform process.
A built-in "status quo bias" operates in the case of both ex ante and ex post constraints, Roland noted. Depending on the level of uncertainty about the proposed reform or its outcome, this bias can effectively block reform. Successful reforms must, therefore, satisfy both ex ante and ex post constraints.
Big bang or gradualism?
Because of the status quo bias, policymakers are likely to face resistance to reform, either from certain segments of the population (that is, the potential losers) or, more universally, as in the case of aggregate uncertainty. A gradualist approach—say, when one reform is proposed first and a second reform is proposed one period later—may simply give rise to what Roland termed an interim status quo bias. When this happens, vested interests (the losers) can block the reform process.
In this situation, Roland said, the "big bang" approach—when both reforms are proposed simultaneously—may be preferable. Under the big bang, the choice is narrowed down to maintaining the status quo (that is, not proceeding) or moving ahead with both reforms. In the face of individual uncertainty, the cost of reversal under a big bang strategy may be high enough to deter policy reversals, thus making this strategy preferable to gradualism.
Gradualism has one option that the big bang lacks, however, especially in the face of aggregate uncertainty. It can foster experimentation—that is, it allows for a period during which partial lessons about the viability of reform outcomes can be learned—and it holds out the possibility of early reversal of the first reform if the prospects for full reform look bad enough. This option of early reversal may make it easier for policymakers to start the reform process.
There is thus a trade-off between big bang and gradualist strategies. In general, it is the low reversal costs under gradualism that make this strategy more initially attractive or acceptable than the status quo. The high reversal costs under the big bang approach, however, may make this strategy more advantageous from the point of view of ex ante political constraints, because it reduces the reversibility of enacted reforms. Thus, Roland suggested, reformist politicians will use any window of opportunity to get big bang packages through to create irreversibility.
However, irreversibility under these windows of opportunity can sometimes be more of a curse than a blessing, Roland noted. With high reversal costs of reform, policymakers may get stuck in an economically inefficient situation, whereas lower reversal costs would allow them to undo a reform that has proved unsuccessful and then try something else in its place.
Sequencing and complementarities
Complementarities between reforms are particularly important in the sequencing of reforms in the gradualist approach, Roland said. For example, a first reform might be to develop a small private sector, while a complementary reform might be to reform the state sector. Together, both reforms will create entry and competition and increase output supply as well as allocative efficiency. However, the development of the small private sector without a reform in the state sector may lead to an inefficient diversion of inputs from the state sector if the private sector can bid up prices for inputs and the state-owned enterprises are not allowed to do so. This inefficient diversion may include collapses of production chains in the state sector and create a situation that is worse for social welfare than the status quo.
Thus, Roland suggested, correct sequencing can exploit the complementarities between reforms to create momentum effects and build constituencies for further reforms. While experimentation under gradualism can make proposed reforms more acceptable, without complementarities, partial reforms are unsustainable in the long run, forcing politicians either to go ahead with further reforms or to go backward.
In transition strategy, Roland said, uncertainty needs to be taken into any consideration of a reform program, because, as experience has shown, the efficiency gains from reform are not virtually certain in the medium run. One should always be cautious, he warned, when drawing policy recommendations from theoretical models. It is especially damaging for the reputation of the economics profession as a whole—and even more for the populations of the countries concerned—when economists present reform blueprints promising sure efficiency gains, and then the opposite materializes, he said. Unilaterally emphasizing the irreversibility created by reforms may be misleading and damaging in social welfare terms when uncertainty is important, as it virtually always is in the transition context. The theory of reform under uncertainty thus emphasizes the importance of experimentation, trial and error, and pragmatism.