This class is intended to help graduate students and professional economists interested in how to do trade policy analysis with applied general equilibrium models calibrated to economic data for the Ukraine. The dataset employed in this workshop was assembled as part of a recently completed USAID project to assess the Ukraine-Turkey Free Trade Agreement. The database parameterizes a modern 45-sector small open economy computable general equilibrium model of Ukraine with seven external regions.
The workshop will provide guidance on the underlying policy issues, data sources and the model’s implementation in an algebraic modeling language (GAMS).
Lectures will be presented live and recorded in Kyiv. (We hope to make arrangements for live participation for students in Kyiv.)
This day will focus on a few policy issues which might be addressed in a general equilibrium model based on Ukrainian data. We could choose to introduce issues which arose in the Free Trade Agreement (FTA) between the Ukraine and Turkey, but I would prefer to keep this fairly simple. The objective of the first day is to get students oriented to the course objective: using a CGE model to simulate policy.
In our first day we will introduce an economic issues and overview the steps involved in conducting a policy analysis on a single topic. This could be Ukraine-Turkey FTA, but perhaps something else. This first day will only provide a top-down view of the underlying theory and the model results, although we want to get a model on their machine which they can run. Then we outline the model code focusing on how we specify policy simulations, and we will introduce the use Excel pivot charts to compare results.
Calibrated choice (applied microeconomics)
This material is based on Varian’s undergraduate and graduate textbooks and my lecture notes on CES functions. We should put together a ten page set of lecture notes.
GDP accounting and social accounting matrices.
Basic GDP identities with an introduction to the BEA tables and the methods involved in acccessing those data. This should help orient students to what is being provided by WiNDC.
We go through three examples: small, medium and large
Markusen-style example (2 sector)
Single Excel worksheet example in which there are multiple sectors but all the numbers are visible.
Larger example based on the publically-available IO table.
The same larger example perhaps based on the FTA database.
Algebraic modeling with GAMS (MCP and CNS)
Based on Jim Markusen’s lecture notes.
Introduction to MPSGE based on small examples derived from Jim Markusen’s computational economics course. These include a variety of “theory with numbers” models which illustrate the range of applications of computable general models.
Working with the database: aggregation to a given base year, sectors and regions.
Sample applications (each with short write-up, GAMS model and code for general simulations and pivot reports).
The course fee depends on payment date. The following fee schedule applies:
Ukrainian participants - if payment received 14 or more days prior to start of class: Eur 500
normal EU participant - if payment received 14 or more days prior to start of class: Eur 2000
EU students (full time at the university being paid at a student rate or unfunded) - if payment received 14 days prior to start of class: Eur 1000