Input-output economics

Monthly Labor Review, March, 2005 by Martin Kohli

Wassily Leontief and Input-Output Economics. Edited by Erik Dietzenbacher and Michael L. Lahr. New York, Cambridge University Press, 2004, 418 pp., $80/hardcover.

Many economists know that Wassily Leontief--the Russian-born, German-educated, decades-long member of the Harvard faculty--won a Nobel Prize, but few of them recognize how much his work has changed economics. After Leontief's death in 1999, the editors of this volume organized a memorial session and invited contributions from a number of his colleagues and former students. As the editors explain in a brief preface, the papers written for that conference form the first part of this book, and they offer a number of perspectives, not always harmonious, on Leontief's work. The second part, according to the editors, presents "new scientific research following the lines" pioneered by Leontief.

In the first chapter, Paul Samuelson reminisces about his professor's rigorous approach to price theory and his early work on international trade and nonlinear dynamics. Samuelson notes grudgingly that Leontief devoted his middle and later decades to input-output economics, work that was the basis for the Nobel Prize. Samuelson suggests that Leontief could have found better uses for his time--a criticism that tells us as much about Samuelson as it does about Leontief.

The other eight contributors to Part I avoid striking such discordant notes. Karen Polenske stresses his unusual views on methodology. Leontief rejected the separation of theory building and empirical work. For him, theories were frameworks for understanding how systems work. Polenske points out that this approach made possible the discovery of a number of facts about the U.S. economy. As one example, she mentions the BLS-constructed table for 1939, which the Bureau used to make projections about the postwar economy. To the surprise of many people, these indicated that the demand for steel would remain strong (because of the direct and indirect steel content in consumer durables), and that proved to be correct. Faye Duchin offers a nuanced account of how Leontief's thinking about international trade evolved away from equilibrium models toward models that could be used to explore alternative scenarios. She also summarizes the diverse reactions to Leontief's surprising discovery (made with the BLS-Constructed table for 1947) that American exports were labor-intensive--a finding that is still generating new models and empirical studies. Andrew Brody and Anne P. Carter take a personal approach, comparing Leontief with Joseph Schumpeter. While these two European intellectuals had different ways of doing economics, dynamics and changes in technology interested both of them. Brody and Carter describe some younger economists who use input-output relationships to track the diffusion of innovations. One of the economists they cite, Chris DeBresson, concludes the book's first part with an interview that he conducted with Wassily and Estelle Leontief. It touches on some of the methodological themes raised by Polenske, the relationship with the Schumpeters, and Leontief's enthusiasm for trout fishing. For reasons of space, it is not possible to comment on all the chapters.

The ten contributions to Part II shift the focus away from Leontief the person and toward current research along the lines he pioneered. The connections between present (as represented in Part II) and past research are clearer in some chapters than others. Exploring some of the same issues Leontief did, Edward Wolff's study of the factor intensity of U.S. trade confirms that U.S. exports remain relatively labor-intensive. Albert Steenge proposes an alternative to Leontief's path-breaking model of pollution. Lawrence R. Klein, who pioneered the use of input-output tables in econometric models, and two colleagues present estimates of KLEMI (capital, labor, energy, materials, and information) production functions. Although Leontief did not embrace regression models, he could hardly disagree with the epigraph that begins the chapter by Klein, Duggal, and Saltzman. It says that the quantification of inter-industry transactions can be "consequential" for practical purposes. Some of the other contributions to Part II are purely theoretical exercises that have little substantive connection with Leontief's research.

In short, the first part of the book presents stimulating perspectives on one of the giants of 20th-century economics. The second part is more specialized and less easy to evaluate.

Martin Kohli Bureau of Labor Statistics, New York region