Working Papers
Trade Reform and Structural Transformation in South Korea
(with Carline Betts and Rahul Giri)
July 2013 draft
Feb 2011 draft
A two country, three sector hybrid model of structural change with distortionary government policies is used to quantify the impact of international trade and trade reform for industrialization. The model features Armington motivated trade in agriculture and industry, and a novel representation of trade reform as a time sequence of import tariffs, export subsidies and lump sum government transfers of net tariff revenue. We calibrate our economy to data on South Korea and the OECD, inputting time series of country and sector specific labor productivity, tariffs and export subsidies which determine evolution of the effective pattern of comparative advantage. The model’s predicted reallocations of Korean labor from agriculture into industry and services from 1963 through 2000 are quantitatively similar to those in the data. Incorporating trade and measured Korean trade reform are both important for the accuracy of this predicted structural change, although international real income differences under non-homothetic preferences primarily determine trade and specialization patterns rather than comparative advantage. Counterfactually eliminating a) international trade b) international labor productivity differentials c) post 1967 Korean tariff reform and d) post 1967 industrial export subsidy reform increase the model’s SSE by 91 percent, 56 percent, 27 percent, and 62 percent respectively.
(with Carline Betts and Rahul Giri)
July 2013 draft
Feb 2011 draft
A two country, three sector hybrid model of structural change with distortionary government policies is used to quantify the impact of international trade and trade reform for industrialization. The model features Armington motivated trade in agriculture and industry, and a novel representation of trade reform as a time sequence of import tariffs, export subsidies and lump sum government transfers of net tariff revenue. We calibrate our economy to data on South Korea and the OECD, inputting time series of country and sector specific labor productivity, tariffs and export subsidies which determine evolution of the effective pattern of comparative advantage. The model’s predicted reallocations of Korean labor from agriculture into industry and services from 1963 through 2000 are quantitatively similar to those in the data. Incorporating trade and measured Korean trade reform are both important for the accuracy of this predicted structural change, although international real income differences under non-homothetic preferences primarily determine trade and specialization patterns rather than comparative advantage. Counterfactually eliminating a) international trade b) international labor productivity differentials c) post 1967 Korean tariff reform and d) post 1967 industrial export subsidy reform increase the model’s SSE by 91 percent, 56 percent, 27 percent, and 62 percent respectively.
Export Intensity and Firm Productivity
(with Bruce McWilliams)
Draft is available on request
To identify the premium from exporting, researchers typically estimate a fixed difference in productivity between exporters and non-exporters, ignoring how productivity can differ between firms exporting at different intensities. Using World Bank Enterprise Survey data of emerging markets, we do a systematic analysis of the productivity and export relationship across the spectrum of export intensity. We find that pure (100%) exporters are distinctly lower in productivity, on average, than regular exporters, i.e., firms that both sell domestically and export, and that among regular exporters, there is a U relationship between productivity and export intensity. Firms exporting at low (10% of output or less) and high (90% of output or more) intensities are significantly more productive than their non-exporting and pure exporting neighbors, indicating that participating in a second market even at a marginal level implies distinctly different firm productivity. We theoretically explain observed behavior by introducing heterogeneous costs for serving both domestic and export markets into a Melitz model. A simulation of the model yields estimates of export intensity patterns similar to that observed in real world data. The consistency between the empirical results and the model provides support for a modification of the self-selection hypothesis. While standard models assume low costs for selling domestically and high costs for exporting, the widespread presence of relatively less productive pure exporters in emerging markets suggests that the reverse can also be true. Thus the self-selection of productive firms should be into both export and domestic markets, rather than just exporting.
(with Bruce McWilliams)
Draft is available on request
To identify the premium from exporting, researchers typically estimate a fixed difference in productivity between exporters and non-exporters, ignoring how productivity can differ between firms exporting at different intensities. Using World Bank Enterprise Survey data of emerging markets, we do a systematic analysis of the productivity and export relationship across the spectrum of export intensity. We find that pure (100%) exporters are distinctly lower in productivity, on average, than regular exporters, i.e., firms that both sell domestically and export, and that among regular exporters, there is a U relationship between productivity and export intensity. Firms exporting at low (10% of output or less) and high (90% of output or more) intensities are significantly more productive than their non-exporting and pure exporting neighbors, indicating that participating in a second market even at a marginal level implies distinctly different firm productivity. We theoretically explain observed behavior by introducing heterogeneous costs for serving both domestic and export markets into a Melitz model. A simulation of the model yields estimates of export intensity patterns similar to that observed in real world data. The consistency between the empirical results and the model provides support for a modification of the self-selection hypothesis. While standard models assume low costs for selling domestically and high costs for exporting, the widespread presence of relatively less productive pure exporters in emerging markets suggests that the reverse can also be true. Thus the self-selection of productive firms should be into both export and domestic markets, rather than just exporting.
Work in Progress
Markets or Exports? Understanding Innovation in Emerging Markets (with Bruce McWilliams)
Informality and Jobless Growth in India (with Rahul Giri)
Informality and Jobless Growth in India (with Rahul Giri)
Published Work
Can Total Factor Productivity Explain Value Added Growth in Services?
published in Journal of Development Economics, Volume 99, Issue 1, September 2012, Pgs 163-177
working paper version
This paper accounts for the rapid growth of the service sector observed in India during 1980-2005. A sectoral growth accounting exercise shows that total factor productivity (TFP) growth was the fastest for services; moreover this TFP increase was significant in accounting for the service sector value added growth. A growth model with agriculture, industry and services as three principal sectors is calibrated to Indian data using sectoral TFP growth rates. The model performs well in accounting for the evolution of value added shares and the growth rates of these shares from 1980-2005. The performance of the model improves significantly when the post 1991 increase in service sector TFP growth is accounted for. It is argued that market-based liberalization policies led to the services’ productivity increase. A modified version of the model is used to qualitatively assess the impact of a sector specific tax policy on sectoral labor and output reallocations.
published in Journal of Development Economics, Volume 99, Issue 1, September 2012, Pgs 163-177
working paper version
This paper accounts for the rapid growth of the service sector observed in India during 1980-2005. A sectoral growth accounting exercise shows that total factor productivity (TFP) growth was the fastest for services; moreover this TFP increase was significant in accounting for the service sector value added growth. A growth model with agriculture, industry and services as three principal sectors is calibrated to Indian data using sectoral TFP growth rates. The model performs well in accounting for the evolution of value added shares and the growth rates of these shares from 1980-2005. The performance of the model improves significantly when the post 1991 increase in service sector TFP growth is accounted for. It is argued that market-based liberalization policies led to the services’ productivity increase. A modified version of the model is used to qualitatively assess the impact of a sector specific tax policy on sectoral labor and output reallocations.
Structural Transformation and Jobless Growth in the Indian Economy
published in The Oxford Handbook of the Indian Economy; ed. Oxford University Press, 2012
Historical growth patterns of contemporary advanced nations highlight the manufacturing sector to be the forerunner of economic growth. In contrast, India has witnessed a very significant role played by the service sector which accounted for a large and rapidly growing share of gross domestic product during the 1970-2007 period. The analysis in this chapter describes this atypical pattern of sectoral growth being witnessed by India. In the first part of our analysis, we present and describe empirical facts on the three principal sectors of the Indian economy- agriculture, industry and services during the sample period. Combining sectoral output, employment, education and factor income data, a growth accounting exercise is conducted which reveals that the principal factor driving growth in Indian services is attributable to rapid growth in total factor productivity (TFP) in this sector. In the second part a general equilibrium growth model with three sectors is developed and calibrated to Indian data during the 1970-2007 period. The results suggest that the model can replicate the evolution of value added shares over the sample period and can also quantitatively match the growth rates of the value added shares of these sectors. Therefore, the model is a suitable candidate to describe the process of structural transformation of the Indian economy. We further explain how the described model can be used to conduct counterfactual experiments, illustrating two examples of such exercises that elicit the importance of differential TFP growth rates in the Indian growth experience. Furthermore, an analytical discussion of how we can relax certain assumptions in the basic set up is also discussed. Some examples of how calibrated growth models such as the one described here can be employed for future research are suggested.
published in The Oxford Handbook of the Indian Economy; ed. Oxford University Press, 2012
Historical growth patterns of contemporary advanced nations highlight the manufacturing sector to be the forerunner of economic growth. In contrast, India has witnessed a very significant role played by the service sector which accounted for a large and rapidly growing share of gross domestic product during the 1970-2007 period. The analysis in this chapter describes this atypical pattern of sectoral growth being witnessed by India. In the first part of our analysis, we present and describe empirical facts on the three principal sectors of the Indian economy- agriculture, industry and services during the sample period. Combining sectoral output, employment, education and factor income data, a growth accounting exercise is conducted which reveals that the principal factor driving growth in Indian services is attributable to rapid growth in total factor productivity (TFP) in this sector. In the second part a general equilibrium growth model with three sectors is developed and calibrated to Indian data during the 1970-2007 period. The results suggest that the model can replicate the evolution of value added shares over the sample period and can also quantitatively match the growth rates of the value added shares of these sectors. Therefore, the model is a suitable candidate to describe the process of structural transformation of the Indian economy. We further explain how the described model can be used to conduct counterfactual experiments, illustrating two examples of such exercises that elicit the importance of differential TFP growth rates in the Indian growth experience. Furthermore, an analytical discussion of how we can relax certain assumptions in the basic set up is also discussed. Some examples of how calibrated growth models such as the one described here can be employed for future research are suggested.
The Service Sector Revolution in India: A Quantitative Analysis
published in The Rise of China and India: Impacts, Prospects and Implications, UNU-WIDER Studies in Development Economics and Politics, editors Palgrave Macmillan, October 2010
Following the trade liberalization in 1991, the Indian economy witnessed a high growth rate of service sector output while that of industry was relatively muted. As a result, the sectoral composition of GDP resembles that of a rich country while its per capita income still remains that of a poor country. In this paper, I identify the service sector as important in two respects: it witnesses unusually high TFP growth, as compared to the other sectors, and experiences rapid expansion in exports and imports of services, especially after liberalization. I develop a three-sector open economy growth model with two important inputs: productivity growth in each sector and trade in the industrial and services sectors. I focus on two steady state years, 1980 and 2003, and assume trade to be balanced in these two years. The model is calibrated to Indian data and can account for the levels as well as the change in composition of domestic output and in factor allocations across the sectors for the two steady states. A counterfactual experiment suggests that growth in productivity has a relatively more important role than growth in trade in accounting for the growth in the share of services value added in aggregate GDP.
published in The Rise of China and India: Impacts, Prospects and Implications, UNU-WIDER Studies in Development Economics and Politics, editors Palgrave Macmillan, October 2010
Following the trade liberalization in 1991, the Indian economy witnessed a high growth rate of service sector output while that of industry was relatively muted. As a result, the sectoral composition of GDP resembles that of a rich country while its per capita income still remains that of a poor country. In this paper, I identify the service sector as important in two respects: it witnesses unusually high TFP growth, as compared to the other sectors, and experiences rapid expansion in exports and imports of services, especially after liberalization. I develop a three-sector open economy growth model with two important inputs: productivity growth in each sector and trade in the industrial and services sectors. I focus on two steady state years, 1980 and 2003, and assume trade to be balanced in these two years. The model is calibrated to Indian data and can account for the levels as well as the change in composition of domestic output and in factor allocations across the sectors for the two steady states. A counterfactual experiment suggests that growth in productivity has a relatively more important role than growth in trade in accounting for the growth in the share of services value added in aggregate GDP.