Zhi Wang U.S. International Trade Commission E-Mail: zhi.wang@usitc

1. Computable General Equilibrium Modeling and Its Application in Trade Policy Analysis Lecture 1: Introduction Zhi Wang U.S. International Trade Commission E-Mail: zhi.wang@usitc.gov

2. Outline • Structure and Contents of the course • Importance, feasibility and limitations of economic modeling • Current difficulties in developing policy modeling systems • What is a Computable General Equilibrium Model and its General Structure • Why we need CGE model in Trade policy Analysis

3. Why We Emphasis “Applied” “Economic” Modeling ? • There is no clear division between theoretical and applied modeling • From theory to “theory with numbers”, to “numbers with theory” • Applied CGE modeling: • A clear policy orientation • A concern for accurate and current dada as basis for the modeling exercise • Model structure determined by the data, not selective use of data to fit a theoretical structure

4. Sequences of lectures • Introduction • Basic Structure for CGE Models • Data Structure: An Accounting Framework and Benchmark Data Set for CGE Analysis • GAMS Lab 1: Basic Programming in General Algebraic Modeling System

5. Sequences of lectures (cont.) • Constraint matrix balance procedure and its applications in data reconciliation • Application of static CGE model in trade policy analysis: the impact of economic integration of a “greater China” • Application of recursive dynamic CGE model in trade policy analysis: the economic impact of China and Taiwan joining the WTO • GAMS Lab 2: calibration and simulation of CGE models

6. The Importance of Economic Policy Modeling and Simulations(cont.) • Unlike most physical scientists, who can test their ideas on controlled experiments in laboratories, economists have to rely primarily on natural experiments for their data. • Classical economists, like K. Max, believes that study economics, one can use neither the scalpel nor the chemical reagent, the only mean is the abstract ability of human beings.

7. The Importance of Economic Policy Modeling and Simulations (cont.) • The Rapid development of modern computer and IT provide the means for today’s economists to examine their ideas by a computer-based simulation model before they put into practice. Simulation and economic modeling has become a major field in applied economics • Numerous historical examples show that any implementation of improper economic or social policy on a large scale would lead to economic disaster and social chaos, requiring years to readjustment at a very high cost.

8. Feasibility and Limitations of Economic Policy Modeling • We could neither include every aspect of the world economy in to a mathematical model, nor could we quantify every step of certain policy implementation precisely in a computer simulation model. What the best we may do is to estimate roughly what shock a policy may bring about on the world economy under certain assumptions and simplifications. • Our logic is simple: if a policy can not prove its feasibility and consistency in a computer simulation model which properly captures the major stylized facts of that economy, it is unlikely to be feasible and consistency in the real economy. Similar to jet designers always implement their new design in physical models and through "wind tunnel" tests before permitting the jet go into the air.

9. The History and Development of CGE Models • One of the major advances in applied economics since the 1970s is converting the well-known Walrasian general-equilibrium structure from an abstract representation of an economy into realistic models of actual economies to conduct policy evaluations by specifying production and demand functions and incorporating data of the real world. • Hundreds of such models have been built and applied to a number of policy issues, ranging from public finance and taxation, economic integration, GATT negotiations, and issues of North-South trade, to the evaluation of development strategies and energy and environment policies for almost all the major countries in the world.

10. Current Problems in Developing CGE Policy Modeling Systems • In the earlier days of CGE modeling, algorithmic and computer power were the main constraints. Both the technical difficulties and computing costs were associated with the process of finding numerical equilibrium of the model, while they are no longer problems today. Statistics show that the numerical solution does not account for more that 15% of the modeling effort • Model formulation, implementation, data preparation and transformation, and interpretation of simulation results have become major burdens of the modeling process • The CGE models built by professionals in academic institutions are still some distance from the kind of practical policy evaluation tool that can be easily understood and used by policy maker in the government

11. Current Problems in Developing CGE Policy Modeling Systems (cont.) • The major difficulties come from the fact that three representations of the same model are needed before any policy simulation can be carried out: • Any policy problem has a semantic description. It describes the policy issue and the model to address it in intuitive terms. • Formal representation. The intuitive model are transformed into formal (mathematical) models by skilled analytical professionals. Policy makers, who are often not familiar with math notations, find those models difficult to understand. • Computer and Algorithm representation. Any formal model needs an algorithm to solve, and any algorithms need a data structure and a problem representation, while problem representations that are meaningful to humans, are not acceptable to machines.

12. Components in CGE Models • A set of economic agents such as firms, households and government whose behavior is to be analyzed. Each agent has a set of endowments that can be used as production factors, such as labor and capital, and an economic account that records his revenues and expenditures ; • Behavioral rules for these agents that reflect their assumed motivation such as profit maximization for firms and utility maximization for consumers. • A set of signals observed by these agents on which they make their economic decisions, such as market prices or government rationing quotas.

13. Components in CGE Models (cont.) • Institutional structure of the model economy, which are the rules of the game by which various agents interact. For example, perfect competition implies that each agent is a price taker and prices are flexible. • A set of explicit definitions of equilibrium conditions which are "system constraints" that must be satisfied for the whole economy but which are not taken into account by each individual agent in making his decisions.

14. Equilibrium in CGE Models • An equilibrium: can be defined as a set of signals such that the resulting decisions of all agents jointly satisfy the system constraints. The signals represent the equilibrating variables of the model. For example, in a perfectly competitive CGE model the assumption that excess demand equals zero in all markets is a system constraint that defines the nature of equilibrium.

15. Do CGE models only be applied to perfect competitive market economies? There are different equilibrium concepts: • in the product and factor markets (Walrasian or micro) • in financial or nominal flows (Keynesian or macro) • in asset markets (another macro) in an inter-temporal process • non-Walrasian equilibrium which depend upon the properties of demand and supply function under rationing. • the behavior assumptions, the institution structure, the signals, and the system constraints or macro closures all can be specified by various completely different economic theories and under a wide variety of institutional assumptions.

16. Justification - Why Should We Use CGE Analysis ? • Theoretical consistency. Comparative statics and counterfactual simulations. • Accounting consistency. Closed system with no leak. • Capture both direct and indirect inter-sectoral, inter-regional, and inter-temporal effects induced by trade policy changes.

17. Justification - Why Should We Use CGE Analysis ? (cont.) • Able to provide more concrete welfare analysis that influence real policy making. • Able to analyze the tradeoff between efficiency and equity/distribution issues • Able to analyze large, discrete, policy changes that far away from the baseline.

18. Justification - Why Should We Use CGE Analysis ? (cont.) • the clear microeconomic structure with links between micro and macro aspect of the economy makes CGE model the soundest tool for quantitative policy analysis • help analysts to understand the essential relationships relevant to particular policy • very useful to build a bridge between economists and policy makers, and provide them with a base for dialogue.

19. Knowledge required for CGE Analysis • Knowledge of general equilibrium theory • Knowledge of real world data. Be able to manipulate and convert it into a model admissible form • Knowledge of computer programming. Be able to implement the model in computer • Knowledge of policy issues and institutional structure

20. Gains to Economic Growth from Trade Liberalization • More efficient allocation of production factors, including the migration of agricultural labor to manufacture activities, which increase labor productivity; • More rapid physical capital accumulation from a "medium-run growth bonus" which compounds the efficiency gain from trade liberalization, induces higher income for economic agents and lower price for capital goods, lead to higher saving and investment so that more physical capital stock available in the economy; • More rapid growth of total factor productivity (TFP) due to speeding technology transfer via expansion of capital and intermediate goods imports from other countries, especially from advanced industrial countries.