WEST LAFAYETTE – More than 225 leading economists and policy makers, including representatives from the World Trade Organization, various United Nations and government offices and other leading international institutions and universities, are convening at Purdue University for the 20th Annual Conference on Global Economic Analysis through Friday.
The annual gathering started in 1998, six years after the founding of the Global Trade Analysis Project (GTAP), a collaboration of more than 15,000 researchers in 170 countries that revolutionized the analysis of trade policies and their global impact.
Whether economists around the world are studying the far-reaching impact of Brexit or new trade agreements, they universally consider the GTAP Model and GTAP Data Base as some of the most critical tools for sound economic analysis. Under GTAP, analysts from those 170 countries starting sharing resources and data to conduct more comprehensive quantitative analysis of international policy issues. That type of collaboration was unprecedented.
“GTAP has done nothing less than revolutionize the way we perform applied economic research,” said Dr. Frank van Tongeren, Head of Division, Policies in Trade and Agriculture, Organisation for Economic Co-operation and Development (OECD) in Paris, France. “Harnessing the power of network economies was not an obvious thing to do in the trade research community, or indeed in economic research more generally, when GTAP started 25 years ago.”
“There is hardly a trade minister in the world who has not heard of GTAP,” said Dominique van der Mensbrugghe, director of GTAP and research professor in agricultural economics at Purdue University. “No trade agreement is made without some quantitative assessment using a GTAP-based model.”
An idea born out of frustration
GTAP was started by Thomas W. Hertel, currently a distinguished professor of agricultural economics at Purdue, who recognized the importance of harnessing data from countries around the world for one comprehensive and consistent database.
That revelation wasn’t immediately evident. It followed Hertel becoming disillusioned with inconsistencies in computable general equilibrium (CGE) modeling throughout Europe and North America about 30 years ago. He identified several challenges, including the lack of data available to the public and the inability to verify the results economists were presenting at professional conferences.
“Large, multi-year projects undertaken by forward-looking research managers in public institutions had to start from scratch and ended up reinventing the wheel - particularly when it came to obtaining and reconciling global economic data,” Hertel said. “This proved very time-consuming and limited the effectiveness of most policy-oriented projects in the global CGE arena. By the end of the one to two years it typically took to come up with a credibly parameterized model, sympathetic managers and decision makers had generally moved on to other interests, and even other positions.”
After being inspired by a project in Australia, Hertel collaborated with other economists to bring an open source format to the field of CGE modeling. He created a small team based at Purdue to develop the GTAP Data Base that vastly improved researchers’ ability to assess the global impact of trade policies. With that database, analysts are all using the same data and advanced tools to conduct research. Also, Hertel pointed out, since it is offered as an open source model, there are no barriers to students and smaller countries to gain access to information.
"By providing a common, widely accepted, analytical data base for global economic analysis, GTAP allows economists to respond quickly to emerging issues, focusing their scarce resources on what really matters,” he said.
To achieve the far-reaching acceptance of the shared database model required breaking some strongly held conventions, according to van Tongeren of OECD in France.
“Individual researchers and institutions would typically guard their tools, data and recipes they used in their analysis, often resulting in wasteful duplication of efforts in data collection, model building and a lack of transparency and reproducibility of results,” he said. “The open collaboration in the GTAP Network has enormously improved the quality of the analysis and has made large scale applied economic modelling accessible for many. It has contributed to better policy choices by enabling a better informed use of data and models and has created a collective intellectual capital that is unparalleled.”
“The biggest advance was having an agency on a regular basis provide a high-quality database at the global level. It just didn’t exist from a user’s point of view,” said van der Mensbrugghe. “It created an explosion of analysis.”
Prior to the availability of GTAP, a 2- to 3-year project would devote up to 90 percent of the time to develop the data, leaving only little time for analysis. “Now researchers can quickly integrate a GTAP Data Base into their work and focus more efforts on actual analysis,” he said.
Also, the problems with inconsistencies were addressed. “Everyone is working with the same database; if you have a different outcome, it’s because you have a different perspective on economic interactions,” he said.
Though initially largely focused on international trade policies, GTAP-based analysis of other key economic policies with an international dimension has exploded including the economics of climate change policies, international migration, foreign direct investment, international transmission of communicable diseases and much more. Irrespective of one’s policy preferences, a thorough understanding of the tradeoffs, and winners and losers from policy alternatives, can only lead to better-informed decisions.
Additional comments from longstanding GTAP Network members:
Dr. Robert Koopman, Chief Economist, Economic Research and Statistics Division, World Trade Organization, Geneva, Switzerland — “The Global Trade Analysis Project has been instrumental in advancing policy analysis and insights on economy-wide effects for over 20 years. The initial objective of a common database and economic model to assess trade policy changes, which has evolved into energy and environmental questions, filled a major need and truly lowered to "cost of entry" for many researchers and organizations. The collective efforts to develop, refine and update the database and model ensures up to date data and techniques, and helps ensure policy makers benefit from top notch analysis. One important advantage the GTAP approach provides over econometric and partial equilibrium analysis is the ability to illustrate tradeoffs across the economy and to prevent "magical" results so many policy makers would like to see and some other models deliver. The fact is that applied general equilibrium analysis keeps you grounded in the real constraints an economy faces in terms of resource endowments and technologies. Partial views of policy changes often ignore these tradeoffs and seem to suggest to policy makers that economic growth and efficiency are simple things to be released when policy changes. While GTAP and applied general equilibrium has its very real limitations and challenges having this data and toolkit in your analytical portfolio is critical."
Dr. Stephen Karingi, Director, Capacity Development Division, United Nations Economic Commission for Africa, Addis Ababa, Ethiopia "GTAP has helped to not only structure but also liven inter-governmental discussions among African countries on matters of regional integration and trade agreements at bilateral, regional and global level. Regional integration happens to be one of the most critical development pillar for Africa. At the heart of the integration agenda is market integration. Because of GTAP, it has become possible to facilitate high level discussions among Government Senior Officials and Ministers, because they are able to appreciate results from GTAP applications that show the potential gains of deeper integration and how they are shared. By helping bring science into the inter-governmental discussions and negotiations, GTAP has helped many countries support the idea of accelerating the market integration at sub-regional and continental level. It is through use of GTAP that the idea of an ambitious Continental Free Trade Area has been cemented, after it was possible to demonstrate the possibility of doubling intra-African trade based on results that looked beyond tariffs, to trade facilitation. Beyond regional integration, GTAP has enabled African countries form common positions in their engagement with the developed and emerging economies, in the different bilateral and multilateral negotiations. African countries have therefore been able to have a common voice, given the prioritisation of issues that analysis using GTAP has been able to deliver. Without doubt, policy making in Africa has benefited significantly from GTAP. It is the African researchers' hope that the versatility of the GTAP database will continue to support the efforts of Africa to harness the potential of the main mega-trends in the continent's favour."
Dr. Thomas Hertel, Distinguished Professor and Executive Director, Center for Global Trade Analysis, Department of Agricultural Economics, Purdue University, West Lafayette, Ind., United States "GTAP was born out of frustration. Those of us seeking to make forward progress in global CGE modeling were frustrated with the lack of replicability and transparency that reigned in this field (outside of Australia) circa 1990. It was also the case that large, multi-year projects undertaken by forward-looking research managers in public institutions had to start from scratch and ended up reinventing the wheel – particularly when it came to obtaining and reconciling global economic data. This proved very time-consuming and limited the effectiveness of most policy-oriented projects in the global CGE arena. By the end of the one to two years it typically took to come up with a credibly parameterized model, sympathetic managers and decision makers had generally moved on to other interests, and even other positions. By providing a common, widely accepted, analytical data base for global economic analysis, GTAP allows economists to respond quickly to emerging issues, focusing their scarce resources on what really matters. This is typically an accurate characterization of the proposed policies and an improved specification of economic behavior in the sector or region of interest. As a consequence, the GTAP Database is now used on a regular basis by economists and decision makers in many of the world’s leading institutions dealing with international trade and environmental policies. This widespread adoption carries the added benefit that results can be readily communicated across departments, ministries and nations, with virtually no loss of content through translation. Today, 25 years after the project’s founding at Purdue University, GTAP is a language spoken around the world!"
Dr. John Reilly, Co-Director, MIT Joint Program on the Science and Policy of Global Change, Cambridge, Mass., United States - "Coming from a perspective of natural resources and the environment, first at the USDA and then at MIT with focus on climate change, I saw the value of an economy-wide approach to evaluating resource and environment issues. However, the challenge with standard input-output, social accounting frameworks is that the wonderful economic assumption that prices represent the marginal value of inputs and outputs, allowing aggregation across ‘apples and oranges’, does no hold for environmental externalities that by definition are not incorporated into market prices. Each dollar of energy or land use does not have the same environmental implication. Thus, what one needed was enough disaggregation to add supplemental physical accounts to the economic data—how many tons of coal (and resultant carbon dioxide emissions), how many hectares of land (lost carbon, albedo change), and what improvements or limits on improvements in efficiency in energy conversion are possible. Substitution elasticities in the power sector, if relative prices were pushed far enough, could end up with more energy coming out than was present in the fuel input, a violation of basic laws of thermodynamics. GTAP has gradually added physical accounting of energy, greenhouse gases, land and water, and more recently greater detail in the power generation sector to the underlying economic data, while demonstrating methods for incorporating that data into analysis. This has been absolutely crucial to understanding the connections between the world economy and the earth system—or more aptly, to allow the creation of models of the earth where human activity is an active part of the system. And, by making this data and research widely available, it has facilitated different avenues of analysis and model development. Advances in computing power have been phenomenal over the past decades—however those advances would be useless without advances in data and analytical approaches to using it. GTAP is the major (only) data resource of for economy-wide modeling."