protectionist measures limit the realization of potential gains from trade within the domestic market and impose real losses on consumers. In the case of China, low state-sector prices are not so problematic. Many enterprises under local jurisdiction are obligated to sell a specified amount of their final output at low and fixed state prices. As quota enforcement is generally effective, price controls in China do not result in a massive diversion of low-priced goods from basic users in the local market. However, as the analysis by Murphy et al. (1992) suggests, because quotas are poorly enforced in Russia, price controls can lead to massive diversion of goods that can be easily resold.
Why do some local governments continue to try to regulate prices? What are their incentives for maintaining policies that appear to impose tremendous local and national welfare losses?
One explanation, based on an influential article by Shleifer and Vishny (1992), is that these price controls generate rents for local governments and the firms under their jurisdiction.6 The basic idea of the model is captured in Figure 8-1. Price and quantity are on the vertical and horizontal axes, respectively. The lines marked d and mr represent a local demand schedule and its corresponding marginal revenue curve, respectively. Normal costs are constant at a level c. A firm earns a "normal profit" when it charges the price c, and c therefore represents a marginal or average cost. However, c may also include per unit subsidies supplied by local or other government sources.
Shleifer and Vishny argue that price controls provide a mechanism for managers and their regulators in local governments to enrich themselves. During the transition, many firms, especially suppliers of consumer goods and services, have come under the jurisdiction of local governments. Nevertheless, the federal government is still able to impose a high tax on reported profits that exceed the normal rate. Thus, managers and their regulators in the local government can make money only if they can conceal excess profits. The spread between the posted price of c and the actual market price, multiplied by the sales quantity, represents potential "bribe revenue" that can be concealed from federal tax collectors. This is illustrated by the shaded rectangle in Figure 8-1. The posted price of c is the firm's marginal cost, which generates the normal per unit profit. The manager and local government officials equate marginal revenue with marginal cost and pocket the difference between the market and posted prices.
There are, however, two problems with using the Shleifer-Vishny model to explain locally initiated price controls. First, the federal tax rates applied to enterprises under local jurisdiction have fallen, while the local tax shares on profits, value-added taxes, and payroll taxes have increased. Thus, there is less incentive to raise money in this concealed fashion. Second, a local poli-