Introduction

The GDyn Data Base, used with the GDyn model, is based on the GTAP Data Base, which describes the global economy in a given year (2001 for the GTAP 6 Data Base; Dimaranan 2006). The GTAP Data Base is augmented with additional data required for GDyn, the dynamic version of the GTAP model. This chapter covers the preparation of these additional elements of the GDyn Data Base1 and the manipulation and aggregation of the GDyn Data Base with various utility programs. The focus is on the additional data requirements, with minimal discussion of the dynamic parameters (see Chapter 3). Readers interested in the underlying GTAP Data Base are referred to Dimaranan (2006).

The GTAP Data Base (Dimaranan, 2006) is changed in the following three ways to obtain a database for GDyn:

1. Five new arrays, listed in Table 4.1, are added to the standard GTAP data file. The values of the regional savings array also change from those in the standard GTAP Data Base, although the dimensions of the array do not change. These changes are discussed in Section 2.

2. A new parameters file contains the seven new parameters used in the dynamic theory (Table 4.2). The methodology for estimating econometrically, and in some cases calibrating these parameters, was examined in detail in Chapter 3. In this chapter the procedure for assigning values to these parameters in the standard GDyn Data Base is outlined. These changes are discussed in Section 3.

3. […]

Introduction

The use of welfare measures is now commonplace in many AGE applications. The calculation and decomposition of welfare have become important parts of the analytical tools used to evaluate and/or compare the impact of policies on regions. In dynamic models, in which the importance of capital ownership features significantly, welfare measures are clearly superior to the use of real GDP as a measure of assessing the benefits of a policy. This is because GDP is defined as goods and services produced in the country, rather than as ownership of produced goods and services. However, calculating and decomposing welfare in a dynamic model are much more difficult tasks than in the traditional comparative static models.

The aim of this chapter is to develop a method for decomposing welfare in the GDyn model presented in Chapter 2. In fulfilling this aim, a number of practical problems associated with decomposing welfare in the GDyn model arise. These problems are discussed and a solution is proposed. It involves a comparative static simulation that allows us to decompose the welfare change resulting from a policy shock at any year during the simulation horizon. The method is illustrated using an example.

The Welfare Decomposition

The welfare decomposition used in the GDyn model is based on the welfare decomposition outlined in Hertel and Huff (2001) for the comparative static GTAP model. Hanslow (2000) generalized it, showing that the decomposition can be applied to other models; for example, with multiple households and/or foreign income flows resulting from foreign capital ownership, remittances, or foreign aid.

The objective of the Global Trade Analysis Project (GTAP), launched in 1992, was to lower the cost of entry into the world of applied computable general equilibrium (CGE or AGE) modeling using a global, economy-wide framework. The birth of the GTAP project and the subsequent publication of the GTAP book (Hertel 1997), which documented the model structure, data, and software, were timely because there was an increasing demand for quantitative analyses of trade policy issues on a global basis. Most notably, the Uruguay Round negotiations under the auspices of the General Agreement on Tariffs and Trade (GATT) were a catalyst in moving forward the GTAP database and model, as were the heated debates over the North American Free Trade Agreement (NAFTA) and subsequently the World Trade Organization's (WTO) Doha Development Agenda. In response to this demand, the GTAP project grew from a few people in a handful of countries in 1992 to more than 8,500 people from 140 countries in 2010; the GTAP book has been widely cited; and the GTAP model and data have been actively used by a large number of public institutions around the world and in analyses on various topics published in numerous refereed journals, books, and reports.

With the launch of the GTAP project two decades ago it became possible for economists who are not specialists in AGE modeling to start using the general equilibrium approach for economic policy analysis, while at the same time avoiding the excessive costs of collecting data and programming. Not only the large number of citations of the first GTAP book documenting the standard GTAP model (Hertel 1997) but also the large number of individuals around the world using different versions of the framework and the database for policy research and analysis are testimony to the success of this project.

The GTAP network of researchers and analysts has been a key asset to the project: Members of the team not only benefited from the availability of data and advances in the modeling framework but also contributed models and data to the project. Knowledge generated by network members has been disseminated at annual conferences, and new ideas have been implemented quickly as the initial data and model lowered the costs of each new extension. In this respect GDyn is no different from other extensions of the GTAP model, although it did benefit substantially from the fact that there was a global database and standard model to start from.

Introduction

The world's rate of population growth is slowing, but total population is still increasing at about 80 million per year and is expected to reach 10 billion by the middle of the 21st century (World Bank 1999). Most of this growth will take place in developing countries, particularly in Asia and Africa. These projected increases in population, along with growth in per capita incomes and associated changes in demand for agricultural commodities, are expected to increase pressures on natural resources both through the expansion of land under cultivation and through more intense use of resources already employed in agricultural production. In support of these expectations, a recent study by Evenson et al. (1999) estimates that without the development of high-yielding varieties of crops, prices for developing country consumers would likely be much higher than they are today. For example, technological advances in the cultivation of rice have reduced costly food imports by 8% and have eliminated the need to convert millions of hectares of forestland to agricultural uses, as would have been required had yields remained at 1960 levels. This evidence and the rise of food and resource prices in 2008 highlight the central role of agricultural research in ensuring sustainable rural development and food security in high-growth developing countries and in reducing the strain on forest ecosystems.

Introduction

During the 1980s and 1990s, China carried out significant reforms to restructure and open up its economy to foreign trade. These reforms helped spur a period of rapid growth, with annual growth in real per capita GDP averaging 6.04% over the period 1978–95 (Maddison 1998). However, reforms to the legislative framework governing foreign investment in China have proceeded more slowly. Entry into certain sectors, as well as the share of foreign ownership, is significantly limited, and joint ventures typically remain the best option for foreign investors. In an effort to boost foreign investment, the Chinese government began offering a number of special incentives in the early 1990s, including duty drawbacks on imported intermediate inputs and capital goods used for the production of exports, exemptions and reductions in the rate of income taxes paid on profits,1 and preferential tax rates for foreign enterprises that reinvest their profits (China Council 2000). These incentives paid off, with foreign investment increasing sharply in 1992–3 (Fig. 8.1). Indeed, by 1994 China accounted for approximately 20% of all foreign direct investment (FDI) in developing countries (Garbaccio 1995).

However, by 1996 the rate of growth in FDI had begun to slow (Fig. 8.1). A survey of the Chinese economy by The Economist (2000) explored some of the reasons behind the FDI slowdown in China. In many cases investors' high hopes for this market were slow to materialize, with the absence of a rules-based economy making it difficult for outsiders to operate effectively in China.

Introduction

GDyn is a recursively dynamic AGE model of the world economy. It extends the standard GTAP model (Hertel 1997) to include international capital mobility, capital accumulation, and an adaptive expectations theory of investment. This chapter discusses the rationale behind the design decisions affecting GDyn and presents its technical features in detail.

The main objective of GDyn is to provide a better treatment of the long-run within the GTAP framework. In standard GTAP, capital can move between industries within a region, but not between regions. This inability of capital to move between regions impedes the analysis of policy shocks and other developments that affect incentives to invest in different regions. For a good long-run treatment, then, we need international capital mobility.

With capital mobile between regions, we need to expand the national accounts to allow for international income payments. Policies that attract capital to a region may have a strong impact on the gross domestic product, but if the investment is funded from abroad, the impact on the gross national product and national income may be much weaker. Therefore, to avoid creating spurious links between investment and welfare, we need to distinguish between asset ownership and asset location: The assets owned by a region need no longer be the assets located in that region, and the income generated by the assets in a region need no longer accrue to that region's residents.

Introduction

The purpose of this chapter is to introduce the reader to the publicly available software – RunDynam – that is used to carry out the applications presented in Part III of this book. The RunDynam program is based on the GEMPACK suite of software (Harrison and Pearson 1998),which is specifically designed to solve nonlinear general equilibrium models. Other general equilibrium models solved using the GEMPACK software suite include the standard GTAP model and the Monash model of Australia. The RunDynam program has been specially tailored to the needs of the GDyn model and other dynamic models. It offers the user a great deal of flexibility in constructing simulations. It is available from the Centre of Policy Studies at Monash University, Australia.

Using the RunDynam software, you can examine the data, construct and modify simulations, solve simulations, and examine results. Users who wish to alter the underlying theory of the model will need to acquire additional software from the Centre of Policy Studies at Monash University, Australia. In addition, those who wish to make their own aggregations will have to purchase the GTAP Data Base from the Center for Global Trade Analysis, Purdue University. Altering the standard model and the data aggregation process are not discussed in this chapter.