Revision Requests
"Structural Change within the Service Sector and the Future of Cost Disease" (with Georg Duernecker and Ákos Valentinyi). CEPR Discussion Paper. Slides Revised and resubmitted: Journal of the European Economic Association, May 2022
Baumol (1967) observed that developed economies suffer from cost disease, i.e., aggregate productivity growth falls because structural change reallocates production to services with low productivity growth. We document that past cost disease importantly contributed to the productivity growth slowdown in the U.S. To assess how severe future cost disease may become, we build a model of structural change among the goods sector and broad services sectors. The calibrated model implies that broad categories of services are substitutes and the services with low productivity growth do not take over production. Simulating the calibrated model forward suggests that future cost disease will be less severe than past one.
Working Papers
"New Evidence on Sectoral Labor Productivity: Implications for Industrialization and Development" (with Richard Rogerson and Ákos Valentinyi). CEPR Discussion Paper, 2022. Slides
We ask whether moving labor from agriculture to manufacturing--"industrialization"--helps the development of poor countries by closing their aggregate labor productivity gaps with rich countries. To answer our question, we leverage recent data releases by the Groningen Growth and Development Centre and build a new dataset of comparable labor productivity levels in agriculture and manufacturing for 64 mostly poor countries during 1990--2018. We find two key results: (i) cross-country productivity gaps were larger in manufacturing than in the aggregate and (ii) there was no tendency for manufacturing productivity to converge. While our results challenge the notion that expanding manufacturing \emph{employment} is essential for closing the productivity gaps, we find the additional result that the initially poor countries that reduced their productivity gaps had higher productivity growth in manufacturing along with several other core sectors.
"Markups, Input-Output Linkages, and Fixed Costs – Evidence from Advanced Economies" (with Benjamin Bridgman). 2022. Slides
We study markups in 17 advanced OECD countries during 1984--2016. Using NIPA data and assuming constant returns to scale, we find that gross-output markups increased in 13 countries but usually remained well under 10\%. We also find that final-output markups were two to three times larger than gross-output markups because of double marginalization resulting from input-output linkages. Lastly, we calculate the fixed costs that reconcile our macro estimates with the much larger micro estimates for firm-level data from Compustat. We find large fixed costs that more than doubled in many countries. We discuss the implications of our findings for competition policy.
"Labor Share, Markups, and Input-Output Linkages – Evidence from the U.S. National Accounts" (with Benjamin Bridgman). CEPR Discussion Paper, 2021. Slides
The literature has suggested many driving forces behind the recent decrease in the U.S. labor share. We build a multi-sector model with input-output linkages to measure the importance of the key driving forces. We find that the decrease in the U.S. labor share reflects both sectoral forces, which can be identified with micro or NIPA data, and aggregation effects, which can be identified only with NIPA data. Specifically, we find that the main driving forces behind the decrease in the U.S. labor share were increases in sectoral markups of similar magnitude, which were importantly amplified by input-output linkages.
"Endogenous Sector-Biased Technological Change and Industrial Policy" (with Ákos Valentinyi). CEPR Discussion Paper. Slides
We build a model of structural transformation with endogenous sector-biased technological change. We show that if the return to specialization is larger in the goods sector than in the service sector, then the equilibrium has the following properties: aggregate growth is balanced; structural transformation takes place from goods to services; the service sector receives more innovation but the goods sector has more productivity growth. We show that compared to the efficient allocation the laissez-faire equilibrium has too much labor in the goods sector. This suggests that optimal industrial policy should aim to increase the pace of structural transformation.
Work in Progress
"Occupations, Skills, and Barriers to Labor Reallocation" (with Georg Duernecker)