Second, market entry by non-state enterprises was subject to severe restrictions. Production remained monopolized and international trade was centrally regulated. The existing state-owned enterprises therefore never faced real competition pressure from domestic or international sources and lacked the incentives to improve productivity.

Third, in the traditional Soviet-type system, to prevent managerial discretion under the distorted macro-policy environment, state-owned enterprises were not allowed to set their workers' wage level. In the Chinese case, after the profit-sharing arrangement was introduced to the state-owned enterprises, wages were still controlled by the state. A worker's wage would increase only if the enterprise's profits exceeded a preset level. In Poland, Hungary and the former Soviet Union, however, partial reforms gave the enterprises the autonomy to set their workers' wages. The weakening of state control on wages gave managers and workers an opportunity to increase their incomes at the expense of the state by absorbing whatever income flow and whatever assets they could obtain from state-owned enterprises. The state's revenues were thus greatly curtailed.

Fourth, wage inflation caused the shortage to become even more acute; governments in Poland and in the former Soviet Union then tried to play a populist game. They increased the imports of consumer goods and forced a heavy burden of foreign debt on their countries. Because of this, instead of bringing continuous growth and a gradual transition to a market economy-as in China and Viet Nam-the partial reforms led Poland and the former Soviet Union to the brink of bankruptcy and hyperinflation.

The transition from a CAD-type economy to a market economy in socialist and developing countries proved difficult. A transitional economy's institutions must be weak and there will be severe distortions in prices and production structures. Shock therapy-which characterizes a macro-first approach to building up the requisite market institutions-cannot deliver a rapid jump to a prosperous market economy. The experiences in China and other East Asian economies show that deep and extensive reforms are not required for dynamic growth at the onset of the transition.

As such, the crucial issue in transition is to have a strategy of sequencing reforms that identifies the most pressing bottlenecks and concentrates resources on the relaxation of binding constraints, removing the suppression of incentives and inspiring people to improve performance to achieve a better life through their own efforts. The IMF/World Bank's macro-first reform approach might be appropriate for an economy in which market institutions are more or less intact and the structural imbalance is small. To use the famous analogy in a somewhat different version, "When the chasm is narrow, it's all right to jump over it." The stabilization program can achieve its goal immediately and the economy can soon operate in a normal market environment. In a country that has pursued a CAD strategy for a long time with severe distortions and a large number of non-viable enterprises, the chasm will be too wide and too deep. A jump without careful preparation will result in a disastrous fall. In such a situation, it is desirable to fill and narrow the chasm before making the jump.

The East Asian experience suggests that with a small change that provides the right incentives for people it is possible to unleash dynamic growth on a weak institutional base, leading to an eventual transition to a fully-fledged, well-functioning market economy. For a developing country that follows a CAD strategy, there must be distortions in the incentive system, which suppress individual efforts in production, and there must be industries that are consistent with the economy's comparative advantages but which are repressed. The useful lessons from the gradual, dual-track, micro-first approach to transition in East Asia can be summarized as follows.

- The government can take measures to improve individual incentives by granting partial managerial autonomy and profit-sharing to farms and state-owned enterprises in order to improve incentives and allow the economy to move closer to the production frontier, which will induce a new stream of output growth.

- The government can introduce a dual-track price and allocation system to replace the old single-track plan. It can remove market entry restrictions to allow resources to be allocated increasingly by the non-state sector to the previously suppressed, more productive industries, while maintaining the quota obligations of state-owned enterprises and farms in order to secure adequate resources to subsidize the existing non-viable enterprises.

- When the products in a sector are allocated largely by the market track, it is time for the government to introduce full market liberalization in the sector.

- The government should introduce continually the necessary regulations and laws to strengthen market institutions during the above process.

The above principles or experiences of other countries should not be applied in a dogmatic way. One example is China's 1979 reform of its household responsibility system, which leased collectively owned land to farm households for 15 years. Like many reforms in China, it was initiated by farmers, sanctioned by the government and introduced nationwide only after its performance was demonstrated. This reform resulted in a dramatic increase in agricultural productivity and output growth. The government of the former Soviet Union under Mikhail Gorbachev adopted similar reforms of its state farms with 50-year leases. Theoretically, the Russian reforms seemed to be better than the Chinese reforms because of their longer and more secure tenure arrangements; however, the Soviet government had a hard time finding farmers willing to accept this arrangement. In hindsight, the failure of the Soviet Union's reforms might have been because its state farms were highly mechanized, depended heavily on purchased inputs, such as chemical fertilizers and fuel, in the production process and were far away from markets. As such, a small individual household farm was not viable. The opposite was true in China.

In a gradual, piecemeal reform, therefore, the government should not have a predetermined, grand blueprint. Instead, it should follow a diagnostic approach, finding out the most crucial binding constraints on incentives and resource allocation and introducing reform measures that are effective but which can be regarded as "halfway measures" by market fundamentalists. In the process, the government should encourage and pay attention to local and private initiatives in institutional innovations. In this regard, political wisdom derived from Chinese culture--shishiqiushi, jiefangsixiang and yushijujin--could be relevant to reform-minded governments in other developing and transitional countries.