THE TAX REFORM IN CHINA : AN ASSESSMENT USING CGE ANALYSIS

Abstract

In 1994 China implemented extensive reform of its tax system with the objective of promoting market-oriented economic development. This reform mainly involves the transition from turnover taxes to value-added taxes (VAT) and the integration of income tax rates for different type of enterprises. Presumably, these reforms effect resource allocation, government revenue, and income distribution. This thesis applies a Computable General Equilibrium (CGE) model to analyze the effects of the 1994 tax reform. The model is static and neo-classical. It has fifteen production sectors, two primary factors and eleven household groups. Major distortionary features of the Chinese tax system are captured by the model. The production and demand decisions of all agents in the economy are determined under the assumption that consumers choose the bundle of commodities which maximizes their utility, and producers select the combination of factors of production which maximizes their profits. The model then brings together the demand and production decisions which determine the equilibrium prices, uses of factors, and activity levels as well as tax revenues. A benchmark equilibrium data set is constructed to provide the benchmark equilibrium for the model. Comparative equilibria, calculated under alternative tax regimes, are compared with the benchmark equilibrium to evaluate the impact of the tax changes. Concentrating on the study of resource allocation and income distribution effects of the 1994 tax reform and alternative tax/subsidy policies, four sets of simulations are performed using the CGE model. The first set of simulations examines the impact of the 1994 tax reform by separately considering each component of tax change in three cases. The second set of simulations considers the same tax changes as the first set but uses yield-preserving tax rates rather than statutory rates to calculate counterfactual equilibria. The third set of simulations examines the potential impact of two proposed reforms in value-added tax. Some alternative subsidy policies are evaluated in the fourth set of simulations. Finally, the sensitivity of the results to substitution elasticity and tabor supply specifications are tested by simulating the overall tax changes introduced in the 1994 tax reform. Major findings concluded from the results of these simulations are • Small aggregate welfare gains are obtained from the 1994 tax reform, however, household groups are worse off because of the redistribution effects between household and government sectors. There will be a substantial increase in the government revenue and prudent and productive use of the revenue could improve the welfare of the households. • The statutory rates introduced in 1994 may be too high from the equal yield standpoint. In particular, the rate of 17% for the existing income-type VAT seems more reasonable for a consumption-type VAT. • Small improvements in income distribution from the tax reform are obtained, and this comes mainly from the changes in enterprise income taxation. • Extension of VAT or adoption of a consumption-type VAT is worth considering in future reform as long as the government can deal with the inflation and administrative problems.