We have launched E-mail Alert service,subscribers can receive the latest catalogues free of charge


Two Decades of Experience of Policy Modeling of DRC


Wang Huijiong Li Boxi Li Shantong

I. Introduction

Development Research Center of the State Council was established in early 80's which was one of the products of economic reform and opening to the outside world of China in the late 70's. The role of this Center is to provide policy consultative service for the government. Therefore, the purpose of studying and application of mathematical model in our work focused very much on policy analysis. We must confess the fact that China is a late comer in modern economic science as well as themes of economic models for policy analysis. Our Center had tried with full effort to catch up and bridge the gap of this field with the contributions of efforts from all our collaborators in this development process. It should also be emphasized in the very beginning of this paper that these are technical difficulties to get exact results from economic mathematical modeling, this issue can be seen clearly from the criticism given by Keynes to the pioneering research of macroeconomic model of Tinbergen, that there were problems of miss classification, multi-collinearity functional form, dynamic specification, structural stability, and the difficulties associated with the measurement of theoretical variables etc. The crucial issue in China is that mathematical economic models should be based on certain economic theory, while the transitional economics of the former Central planned economy to a market oriented economy is a new field of economic study.

In spite of these facts, we have the recognition that "Mathematical models are important took of forecasting and policy analysis, and it is unlikely that they will be discarded in the future. The challenge is to recognize their limitations and to work towards turning them into more reliable and effective tool. There seems to be no alternatives." That is exactly the efforts of our Center focused on our work in the past two decades.

This paper will be divided into two parts: the first part will give a brief retrospect of the policy modeling projects that had been done by us within the context of background of economic situation of China in different stages of economic reform. This will give a general picture of evolution of policy modeling in China.

II. A Retrospect of Policy Modeling of DRC

Policy modeling work of DRC can be roughly divided into two periods based upon the broad context of progress of economic reform as well as the experience of our staffs.

2.01 First Period (1982-1990)

In this period, China is in the initial stage of economic reform and opening, the planning system is undergoing changes and reform. 1. Initial Period of Application of mathematical modeling in policy analysis

(1) The first pioneering effort to the application of mathematical modeling in policy analysis was incorporated in the project of "Comprehensive Planning of Shanxi Province". This project was assigned by the Central government to study the long term development planning of Shanxi Province. This is a natural process of development of policy analysis described by Chenery that 'the earliest form of policy analysis in underdeveloped countries was typically described as "development planning, since one of its purpose was to assess the consistency of policy instruments and objectives. This term is perhaps unfortunate, since it is after taken to imply greater government control of economic activity." In fact, it is well known globally that a gradualist approach had been adopted by China in the process of reform and opening. Therefore, in the project of "Comprehensive Planning of Shanxi Province", China hadn't implemented the reform of the planning system in this initial period. The concept of planning was still traditional in sense, and the only available policy instrument was "Investment allocation".

But this pioneering project implemented in Shanxi province from Feb. 1982-June 1983 has two unique features: the first, the planning period is extended to 20 years (1981-2000) compared to past practice of 5-year planning period, the second, our Center had the privilege to organize various government organizations, research institutions and academic field to work together in national key project. There are around 1400 people involved in this project. We try to experiment, policy analysis with new technique in this project. A wide variety of modeling exercise were experimented in this project. There are around 100 people involved in the exercise of policy modeling. 26 models are prepared. The scope of study covered very broad areas, The comprehensive planning of Shanxi province, investment on coal sector, electric industrial planning, water resource utilization, optimal plantation, population model, environmental projection and planning, education planning and projection, investment on science and research, input output of light industry, output projection of electronic industry etc. The techniques used include input-output, econometrics, state vector differential equation, linear programming, multiple goal programming, decision analysis etc.

(2) The result was finally edited into a book titled "Compilation of Economic Mathematical Models of Comprehensive Planning of Shanxi Province." This pioneering project represents a very nascent stage of policy modeling of DRC. But we have achieved the purpose to get acquaintance to the application of various mathematical tools to the economic policy analysis, a team and organizational relationship has also been created among the researchers. An appendix with the brief introduction of seven mathematical tools are included in the above publication. In spite of the traditional concept of the planning, the important types of interaction among the policy variables (objectives and instruments) and the constraints on the economic system of Shanxi province had been correctly identified in the specification of the planning model.

(3) Due to the nascent stage of economic mathematical modeling, and also we understand the key issue of economic mathematical modeling is an interdisciplinary study between economics, mathematics and the real world. The scholars involved in the Task Oriented Transient Organization have different background, automation, economics and others. Therefore, it is commented in the preface of the above edition that "Economic modeling and mathematical "model" are two fashionable terms currently in China. Model is designed through abstraction and simplification of the system of real world. Economic modeling is a simplified abstraction of economic activity. It can be in the form of flow chart, statistical table, bar chart. Hydraulic model to simulate economic system had once been used abroad, but it was short lived. Mathematical model to represent the economic system became widely adopted both domestically and in abroad. It should been emphasized that our "model" is a combination of economic activity and mathematical means. The precondition to design a relatively useful model is a detail objective observation of economic phenomena, with appropriate analysis and synthesis to understand clearly the interrelationship of variables of various economic activity, to compare this with established theory, and express these relationship with appropriate mathematics."

(4) Project Evaluation The question of how to allocate investment and other scarce resources among sectors and projects was also an urgent issue of development policy of China with governmental investment allocation playing a dominant role. Our Center had proposed to the State Planning Commission the application of 'Feasibility Study of Industrial Projects". We had organized a meeting in national scale, and a book had also been published. (Reference 2)

2. Second Period of Policy Modeling of DRC

(1)China Toward the Year 2000

A pioneering project of strategic planning of China.

With the experience gained in the development planning at provincial level, we had got the awareness the weakness of traditional Soviet model of the planning system which became inappropriate in current stage of China's development. This point has also been correctly summarized in a recent IMF report that "In the case of Russia and China, at the outset predominantly agrarian economies where the majority of citizens were illiterate, the transformation to an industrialized and educated society was achieved roughly within a generation. However, once these economies entered intermediate or higher stages of development and resource allocation choices became more complicated was unable to cope." Therefore, in the national priority project of "China toward the Year 2000" initiated in 1983 and completed in 1985, a strategic development planning was developed which is totally different from traditional model of Soviet planning. This study includes a main report and 13 sub-reports. Two of these sub-reports are: Macro-Economic Model Projections of China toward the year 2000 and Summary of Data of China Toward the Year 2000.

(2) China: Economic Development and Modeling

Fourteen models had been prepared in the project, which covered the study of development strategy and policy analysis, macro-economic model with application of econometrics, macro-economic model based upon production function and analysis of TFP, population and coordinated economic development planning model, quantitative analysis of economic structures, reproduction of two major sectors, long term trend of development model, I-application of system dynamics, long term trend of development model, II-China's social economic development model, medium and long term macro-economic model, education planning mathematical model, energy system planning and decision model, energy demand model, China's environmental projection model, production structure plantation model etc. The number of people involved in this subreport is also around 100. The collection of these models are openly published in 1990 (See reference 3). This project represents the policy modeling with collaboration of people of different organizations at national level. The mathematical tools used are the same as the first project described in 1-(1), in addition, system dynamics and recursive programming have been added. This effort of modeling exercise had got the 1st class national award of Application of mathematical modeling in the national exhibition of computer application.


If you need the full context, please leave a message on the website.