Liaoning Province, Dynamic Financial Cge Model And Policy Simulation

Over the past 30 years, computable general equilibrium (CGE) models have become a standard tool of empirical economic analysis. In recent years, improvements in computer technology and model specification have improved the payoffs and reduced the costs of policy analysis based on CGE models, paving the way for their widespread use throughout the world. Because of the assumed money neutrality, most CGE models do not take financial markets into account. The interaction between the financial and the real sector of the economy is ignored. However, financial markets play a crucial role in the stabilization and structural adjustment policies. The traditional CGE models are therefore not adequately equipped to properly simulate these policies. Financial markets should be explicitly modeled in order to respond to this inadequacy. During the last two decades, some attempts have been made to make modifications to the Walrasian framework and extend empirical Walrasian CGE analysis with financial markets. Whereas, the research on financial CGE models and provincial CGE models are just at the beginning stage in China.A social accounting matrix (SAM) of Liaoning province is established in this dissertation. The SAM, which includes the financial markets, is based on the input-output table of Liaoning province in 2002. We try to build the Dynamic Financial CGE Model for Liaoning Province (LDF_CGE), and then analyze the possible impact on Liaoning province economy of a raise of the interest rate. The whole dissertation is organized in 5 chapters. The structure arranges as follows.Chapter 1 is the background introduction, which illustrates the theoretical basis of CGE models and reviews the relevant research on CGE models. Then the purpose and contents of the research are briefly introduced.The variables, parameters and model specification of the LDF_CGE are described in Chapter 2. The LDF_CGE contains seven production sectors, they are: agriculture, industry, architecture industry, transportation and postal service, commerce, food industry and the non-material department. Two primary factors of production are assumed for each sector: capital and labor. Two groups of residents are specified: urban resident and rural resident. The behavior of enterprises, government, financial markets, other provinces and other countries are described in the model. The LDF_CGE contains eight blocks of equations: price, production, trade, income, expenses, finance, equilibrium and dynamic recursion. There are 81 groups of altogether 4883 independent equations, 4883 independent variables, 19779 non-zero elements and 9272 non-linear n-z elements.Chapter 3 describes the data and the parameter specifications. To solve the CGE model, a benchmark data set should be constructed. This data set needs to be in equilibrium, consistent, and closed. A Social Accounting Matrix (SAM) provides a natural framework to organize the flow-of-value statistical data to satisfy those conditions. This section studies at the structure of SAM, and constructs a SAM of Liaoning province. Then balances the SAM using cross entropy methods. In addition, the calibration process is discussed in this section.Chapter 4 is the application part of the model. Policy simulation begins with an assumption that the economy under study is in equilibrium (not necessarily Walrasian type) in the presence of an existing policy regime and for the data in a chosen year. Using that data, parameter values are chosen so that the model will replicate this benchmark equilibrium through a model solution. We can get basecase forecasts by updating the exogenous variables and parameters. With variable shocks, the model is re-solved. Any hypothetical policy changes can be quantitatively measured by comparing the new solution with the initial one. This section analyzes the possible impact on Liaoning province economy of a raise of the interest rate.Chapter 5 provides a summary of this dissertation and gives some discussions of its limitations and possible extensions.