Back to homepage

Jump to topic:

Determinants of trade and specialization

Trade theory had identified several determinants of trade and specialization. I am interested in quantifying and ranking the influence of these determinants. I look at trade costs, technology, consumption preferences, and factor endowments. Research in progress expands the scope of this investigation to include more countries, sectors, and determinants.

Relevant results (oldest to latest)
Papers that contains these results
This paper estimates the effects of technology and factor endowments on specialization using panel data. The effects of technology are found to be consistently significant while the effects of factor endowments are harder to pin down and are often insignificant. Technology, factor endowments, and specialization in OECD countries: The role of variable capital utilization (first draft: 2004). Applied Econometrics and International Development 10-2, December 2010, pp. 25-46.
Using simulations, this paper finds that the magnitudes and signs of the effects of capital stock and technology on specialization are very country-specific, which explains the difficulties in obtaining such average effects in regressions (as in the paper above). The volume of trade is not significantly affected by capital endowments and industry-level comparative advantages, but is greatly affected by the producer-level productivity differences. Capital, technology, and specialization in the neoclassical model (Matlab programs and data; working paper). Journal of International Economics, 83(2), March 2011, pp. 229-242.
This paper uses several metrics to rank the importance of various determinants of the pattern (direction) of trade and specialization. Productivity and taste differences are found to be the first and second most significant determinants of trade and specialization. Factor endowments are found to be the least influential for the average country in the data set, but their effects are found to be greater in the poorer than richer countries. The paper also shows a number of important ways in which trade costs affect trade and specialization. Determinants of trade and specialization in the OECD countries (abstract, working paper; first draft: 2008). Economic Inquiry, 51(1), January 2013, pp. 138-158.
This work in progress employs the same methodology as the paper above, but uses a significantly larger dataset, which includes many low-income countries. Preliminary results show that industry-level productivity differences are a more influential determinant of trade and specialization than factor endowment differences, even in the poorest countries. Determinants of trade and specialization (slides with preliminary results).
This paper investigates the sources of competitiveness of the U.S. motion picture industry. The focus of the paper is on the institutional aspects that affect this competitiveness (work in progress). Institutional determinants of specialization in the motion picture industry.
Work in progress. Institutional determinants of trade and specialization.


Effects of trade and investment liberalizations and trade wars (policy-related research)

Government policies regarding international trade and investment can have large effects on the economy, including employment and prices. These effects are usually different across industries and countries. The papers listed below study the effects of free-trade agreements and trade wars on trade, prices, off-shoring, and welfare.

Relevant results (oldest to latest)
Papers that contains these results
The paper predicts the effects of industry-specific trade wars between the European Union and the United States. Some trade wars turn out to be more beneficial to the U.S. than E.U. and vice versa. The results are linked to the recent trade disputes between the U.S. and E.U. The paper also predicts the effects of lowering of trade barriers between rich and poor countries. The results show which industries are good targets for barrier reduction depending on the policy goal. Putting industries into the Eaton-Kortum model (working paper; first draft: 2004). Journal of International Trade and Economic Development, 21(6), December 2012, pp. 807-837. Winner of the Best Paper Prize for 2012.
This paper analyzes the effects of NAFTA on international trade. It compares the results of the forecasts made at the time of NAFTA signing to the actual changes in trade. It uses a new CGE model of trade to make a prediction of the effects of NAFTA starting with 1989 data. Predicting the effects of NAFTA: Now we can do it better! (working paper; first draft: 2007). Journal of International and Global Economic Studies, 5(2), December 2012, pp. 32-59.
This paper investigates the effects of free-trade agreements on offshoring. It estimates the barriers to international productions and international trade and quantifies the welfare gains from international production and trade. International production, technology diffusion, and trade (Preliminary draft: 2007; first draft: 2009). Journal of International Trade and Economic Development, forthcoming.
This paper forecasts and analyzes the effects of the U.S.-Korea free-trade agreement. Special emphasis is placed on the predicted reallocation of employment across industries due to this trade agreement. The effects on welfare are also quantified. What would KORUS FTA bring? (with Ozlem Yaylaci) (online appendix). International Economic Journal, forthcoming.


Technology diffusion and spillovers through trade and international production

Benefits of a technological improvement are not limited to the country where the improvement has taken place. International trade and international investment carry the benefits to other countries. For example, other countries benefit from lower prices of final and intermediate goods through trade. Multinational production enables technology transfer across countries. I quantify these benefits.

Relevant results (oldest to latest)
Papers that contains these results
This paper quantifies technological spillovers from the U.S. and Japanese machinery industries to other industries within those countries as well as other countries using a multi-industry version of the Eaton-Kortum model of trade. The effects vary depending on the strength of inter-industry linkages, trade costs, and the pattern of comparative advantage. Putting industries into the Eaton-Kortum model (working paper; first draft: 2004). Journal of International Trade and Economic Development, 21(6), December 2012, pp. 807-837. Winner of the Best Paper Prize for 2012.
This paper develops a general equilibrium model of international production and technology diffusion. Producers establish foreign affiliates in order to minimize the cost of production and getting their product to market. Technology diffusion occurs through foreign direct investment. The magnitude of this technology diffusion is quantified. International production, technology diffusion, and trade (Preliminary draft: 2007; first draft: 2009). Journal of International Trade and Economic Development, forthcoming.


Development and evaluation of new models of international trade and multinational production.

I am interested in developing new models of trade and multinational production. I am equally interested in evaluating the performance of these models. Complex computable models, such as the ones that I use, typically have very good in-sample fit due to the large number of the parameters in the model. Therefore, it is better to evaluate such models by their out-of-sample performance, using historical forecasts or backcasts.

Relevant results (oldest to latest)
Papers that contains these results
This paper builds a computable general equilibrium model of trade that uses Eaton-Kortum methodology to explain intra-industry trade instead of the usual Armington assumption. The model is able to predict and explain changes in bilateral trade flows, industry outputs, employment, prices, and wages. The paper develops a procedure for calculating industry-level technology parameters in a multi-industry model. Putting industries into the Eaton-Kortum model (working paper; first draft: 2004). Journal of International Trade and Economic Development, 21(6), December 2012, pp. 807-837. Winner of the Best Paper Prize for 2012.
This paper extends the model above to several factors of production. The ability of the model to accurately predict changes in specialization in response to changes in factor endowments is investigated. The evaluation procedure uses backcasting and historical forecasting. The model is found to make accurate predictions. Capital, technology, and specialization in the neoclassical model (Matlab programs and data; working paper). Journal of International Economics, 83(2), March 2011, pp. 229-242.
This paper evaluates the ability of the model above to predict changes in bilateral trade in response to trade liberalization. Historical forecasts are compared to the actual data and predictions of other models. The model is found to make substantially more accurate forecasts relative to the existing models. Predicting the effects of NAFTA: Now we can do it better! (working paper; first draft: 2007). Journal of International and Global Economic Studies, 5(2), December 2012, pp. 32-59.
Develops a general equilibrium model of international trade, international production and technology diffusion. Producers establish foreign affiliates in order to minimize the cost of production and getting their product to market. Technology diffusion occurs through foreign direct investment. The model is used to study the relationship between international production and trade. International production, technology diffusion, and trade (Preliminary draft: 2007; first draft: 2009). Journal of International Trade and Economic Development, forthcoming.


Improved measures of key industry-level variables

Relevant results (oldest to latest)
Papers that contains these results
This paper carefully builds an extensive dataset that includes accurate measures of industry value addeds and factor shares. It also discusses why the measures of value added and factor shares reported by the surveys of manufactures are typically inaccurate. The dataset has since been used in Nunn and Trefler (2010) and Reshef (2007). An improved measure of industry value added and factor shares: description of a new dataset of the U.S. and Brazilian manufacturing industries (unpublished manuscript).
This paper shows the importance of accounting for variable factor utilization when using times series data on factor inputs. Not correcting for changes in utilization over business cycles can lead to spurious results in some regression models. Technology, factor endowments, and specialization in OECD countries: The role of variable capital utilization (first draft: 2004). Applied Econometrics and International Development 10-2, December 2010, pp. 25-46.


Research by doctoral students

Parul Deswal - trade in services.

Ozlem Yaylaci - historical changes in trade costs; effects of WTO on trade costs.

Abdoulaye Seck - international technology diffusion and economic growth.

 

Home CV Contact Research Teaching