Computable General Equilibrium (CGE) Modelling and Its Applications to Policy Impact Analysis

Institution: see Organisers & Supporters

Programme of study: International Research Workshop

Lecturer: Hans Kremers (Independent Researcher)

Date: Thursday, 29/09/16 (09:30 – 18:00)

Max. number of participants: 20

Credit Points: 5 CP for participating in the whole IRWS

Language of instruction: English


Computable General Equilibrium (CGE) modelling has become a popular tool for policy impact analysis at many government, policy oriented, and academic institutions such as the EU, economic university departments, or policy assessment institutes such as the ZEW in Mannheim, CPB in The Netherlands. It even looks ’trendy’ to have your own CGE model. During the negotiations between the Greek government and the IWF, EU, and EZB, negotiatiors often call for quantitative assessments of the proposals, which might well be based on an application of the EU’s computable general equilibrium models. CGE models, like many other quantitative economic models are often unjustly considered to be the main culprit of economists supposed to be unaware of a financial crisis in the global finance system before 2008. In this short course, I want to provide more background information on what these models are and how they are applied to policy impact analysis. The course attendants should get some idea on what these models are good for and about their limitations. I refer to existing courses on CGE modelling regularly given by institutes such as GTAP (, ECOMOD (, by the Gempack community at the Center of Policy Studies (CoPS) of Victoria University in Melbourne (, and by the GAMS community ( among many others. Furthermore, I refer to Shoven and Whalley (1992) and Ginsburgh and Keyzer (1997) as underlying standard literature.

We consider three significant developments in economics during the 20th century that have lead to the rise of CGE modelling within policy impact analysis. CGE models are calibrated on a social accounting matrix, comparable to a much extended input output table. This hence refers to long time developments in input-output modelling pioneered by the Russian economist Wassily Leontief, see Leontief (1936). Parallel to these developments, a mathematical theory of general equilibrium has been developed by well-known economists like Arrow, Debreu, Hahn, using insights from mathematical programming, often related to so-called fixed point proofs and related algorithms to prove the existence and uniqueness of an equilibrium. I refer to the PhD thesis of Gerard Debreu which builds up the general equilibrium model in all its mathematical detail, Debreu (1959), or to Arrow and Hahn (1971). The latter idea points us to the third development in economics, namely in developments of mathematical programming algorithms to compute an economic equilibrium in a general equilibrium model. The work of Herbert Scarf, Scarf and Hansen (1973), was seminal here, and formed the basis from which John Shoven and John Whalley built their CGE models. The introduction of computing equipment provided the means to be able to solve large models efficiently. The morning part of the course in CGE modelling is dedicated to a more detailed description of these three developments in economics and how they cooperate in what we nowadays call CGE modelling. We also describe how we perform a CGE analysis to assess the impact of a policy.

The afternoon is dedicated to introduce several existing CGE models and their applications of CGE modelling. We do so by presenting an existing study on the application of each model. Originally, following the Uruguay trade rounds, CGE models were applied to assess the impact of trade and tax policies until the Kyoto Protocol was signed to support a global effort to curb carbon emissions, which was expected to have significant effects on international trade flows. I again refer to Shoven and Whalley in Shoven and Whalley (1984) and Shoven and Whalley (1992) for applications on trade and taxes. Hence, the application of CGE models was extended to the assessment of climate policies. The GTAP model and underlying Social Accounting Matrix at Purdue University originated as a pure trade CGE model and database following the Uruguay trade rounds, but has, over time been extended to include climate related issues such as economy related carbon emissions, energy substitution, land use. The research, models and data can be found on their website, To further improve its application on this area, a demand arose to link, among others, CGE models with models from other, climate related areas such as meteorology, into so-called integrated assessment models. The increased attention of policy makers to the climate as well as signals that our current dependency on fossil fuel energy and issues of energy supply security endanger the economy also raised an interest in applying CGE models. Applying a CGE model to assess the impact of climate policies required an extension of the model. Again, a lot on this can be found in the extensive research database at GTAP. There exists an energy substitution variant of the GTAP model, referred to as GTAP-E (see Burniaux and Truong (2002)), which is often applied and extended to such issues as energy substitution, emission permits and carbon taxes, land use.

We also look at applications of CGE modelling to assess the impact of transport policies on the economy following the rise in transport problems such as congestion with the growth of many economies. An example of such transport issues is the inclusion of road pricing to stop congestion around big cities. There is a single country CGE model for Austria that looks at road pricing from a tax point of view. The model is referred to in Steininger and Friedl (2004). In Kalinowska, Kremers, and Truong (2008), we apply this model to the German case.

We will look at the application of a CGE model to a developing economy like Mongolia, where two large mines have been discovered, with a large impact on the local underdeveloped post-communist economy and neighbouring China and Russia. This regional single country CGE model is known as the Mon-CGE model and has been applied to the Mongolian economy to assess the impact of introducing an Energy Master-Plan within a project by the Asian Development Bank (ADB). For a description of the Mon-CGE model, as well as an application of the model to the Mongolian economy, I refer to Corong et al. (2011). Enkhbayar et al. (2010) also provide an interesting application of a regional CGE model to the Mongolian economy, within project based research.

Last but not least, we are currently looking at the construction and application of a regional CGE model to Sønderborg and the Southern Denmark regions within project zero ( This project intends to offer a platform for initiatives in the Sønderborg region to introduce emission reduction measures such as renewable energy technologies into the local and regional economy of Sønderborg.

Attendants of this course are expected to have some background in economics, in particular micro economic theory, although I do not intend to go very deep into economic theory. I would like to ask interested PhDs to send an email to with a description of their background and what would interest them (models, applications, political issues etc.) in this course.


  • Arrow, K., and F. Hahn (1972), General Competitive Analysis, San Francisco, Holden-day.
  • Burniaux, J.M., and T.P. Truong (2002), “GTAP-E: An Energy-Environmental Version of the GTAP Model”, GTAP Technical Paper No. 16, GTAP, Purdue.
  • Corong, E., B. Decaluwé, and V. Robichaud (2011), “Assessing the Impact of Increased Foreign Direct Investment in Mongolia: A Computable General Equilibrium (CGE) Approach”, Mimeo, Asian Development Bank.
  • Debreu, G. (1954), Theory of Value, New-York, Wiley.
  • Enkhbayar, S., D. Roland-Holst, T. Oi, and G. Sugiyarto (2010), “Mongolia’s Investment Priorities from a National Development Perspective”, Mimeo, Asian Development Bank.
  • Ginsburgh, V., and M. Keyzer (1997), The structure of applied general equilibrium models, Cambridge, Massachusetts Institute of Technology Press.
  • Kalinowska, Kremers, and Truong (2004), “Fitting passenger travel into a CGE model”, mimeo, DIW Berlin.
  • Leontief, W. (1936), “Quantitative input and output relations in the economic system of the United States”, Review of Economics and Statistics.
  • Scarf, H., and T. Hansen (1973), The Computation of Economic Equilibria, New Haven, Yale University Press.
  • Shoven, J.B., and J. Whalley (1984), “Applied general equilibrium models of taxation and international trade”, Journal of Economic Literature 22, 1007-1051.
  • Shoven, J.B., and J. Whalley (1992), Applying General Equilibrium, New York, Cambridge University Press.
  • Steininger, K. and B. Friedl (2004, June), “Economic and distributional impacts of nationwide car road pricing: a CGE analysis for Austria”, Paper submitted to the Thirteenth Annual Conference of the European Association of Environmental and Resource Economists, Budapest.

You have to register for the 10th International Research Workshop to participate in this course.