THE HECKSCHER-OHLIN MODEL IN THEORY AND PRACTICE

1 PRINCETON STUDIES IN INTERNATIONAL FINANCENo. 77, February 1995 THE HECKSCHER-OHLIN MODELIN THEORY AND PRACTICEEDWARD E. LEAMERINTERNATIONAL FINANCE SECTIONDEPARTMENT OF ECONOMICSPRINCETON UNIVERSITYPRINCETON,NEW JERSEYPRINCETON STUDIESIN INTERNATIONAL FINANCEPRINCETONSTUDIES ININTERNATIONALFINANCE arepublished by the International Finance Section of theDepartment of Economics of Princeton University. Al-though the Section sponsors the Studies, the authors arefree to develop their topics as they wish. The Sectionwelcomes the submission of manuscripts for publicationin this and its other series. Please see the Notice toContributors at the back of this author of this Study, Edward E. Leamer, is Profes-sor of Economics and Chauncey J. Medberry Professor ofManagement at the John E. Anderson Graduate School ofManagement at the University of California, Los is a Fellow of the American Academy of Arts andSciences and of the Econometric Society and is a ResearchAssociate of the National Bureau of Economic Leamer has published several books and numer-ous articles in the fields of econometrics and internationaleconomics.

2 His most recent book isSturdy Econometrics:Selected Essays of Edward E. Leamer(1994). This Studywas presented as the Frank D. Graham Memorial Lectureon March 3, 1994. A complete list of Graham MemorialLecturers is given at the end of this ,DirectorInternational Finance SectionPRINCETON STUDIES IN INTERNATIONAL FINANCENo. 77, February 1995 THE HECKSCHER-OHLIN MODELIN THEORY AND PRACTICEEDWARD E. LEAMERINTERNATIONAL FINANCE SECTIONDEPARTMENT OF ECONOMICSPRINCETON UNIVERSITYPRINCETON,NEW JERSEYINTERNATIONAL FINANCE SECTIONEDITORIAL STAFFP eter B. Kenen,DirectorMargaret B. Riccardi,EditorLillian Spais,Editorial AideLalitha H. Chandra,Subscriptions and OrdersLibrary of Congress Cataloging-in-Publication DataLeamer, Edward HECKSCHER-OHLIN MODEL in THEORY and PRACTICE / Edward E. cm. (Princeton studies in international finance, ISSN 0081-8070 ; no. 77)Includes bibliographical 0-88165-249-0 (pbk.) : $ HECKSCHER-OHLIN principle.

3 2. Comparative advantage (International trade). II. 1995382 dc2094-49591 CIPC opyright 1995 by International Finance Section, Department of Economics, rights reserved. Except for brief quotations embodied in critical articles and reviews,no part of this publication may be reproduced in any form or by any means, includingphotocopy, without written permission from the in the United States of America by Princeton University Printing Services atPrinceton, New JerseyInternational Standard Serial Number: 0081-8070 International Standard Book Number: 0-88165-249-0 Library of Congress Catalog Card Number: 94-49591 CONTENTS1 INTRODUCTION12 THE LERNER-PEARCE DIAGRAM IN ACTION5 The Factor-Price-Equalization Theorems5 Applications of the Learner-Pearce Diagram83 ALGEBRA OF THE(EVEN) HECKSCHER-OHLIN -VANEK MODEL17 Neutral Technological Differences and Home Bias194 LEAMER TRIANGLES22 Effects on the United States of Physical-CapitalAccumulation in Japan24 Effects on the United States of Capital Accumulationin Germany255 EVIDENCE26 Patterns of Four Countries26 Patterns over Time33 Trade Patterns and Resource Supplies366 THE HECKSCHER-OHLIN MODEL AND INCOME INEQUALITY39 Three Mistaken Notions41 REFERENCES44 FIGURES1 Factor-Price Determination in a Lerner-Pearce Diagram52 Two-Cone Lerner-Pearce Diagram93 High-Wage and Low-Wage Equilibria with a NontradedSector94 Effects of a Minimum Wage with One UncoveredTradeable Sector105 Capital-Scarce Equilibrium with Talented Workers on Farms126 Capital-Abundant Equilibrium with Talented WorkersNot on Farms127 Technological Change in the Capital-Intensive Sector138 Capital Flow from a Technologically Backward Country159 Foreign Direct Investment into a

4 TechnologicallyBackward Country1510 Growth Paths in a Three-Factor Model2311 Net Exports per Worker, 19582912 Net Exports per Worker, 19653013 Net Exports per Worker, 19743114 Net Exports per Worker, 19883215a Net Exports of Forest Products per Worker, 1965 and 19883415b Net Exports of Forest Products per Worker, 1965 and 1988(Zoomed View)3416a Net Exports of Labor-Intensive Manufactures per Worker,1965 and 19883516b Net Exports of Labor-Intensive Manufactures per Worker,1965 and 1988 (Zoomed View)3517 Net Exports of Labor-Intensive Manufactures per Workerversus Capital per Worker, 19883818 The Global Labor Pool, 198940 TABLES1 Effects of Price Changes on Factor Earnings242 Components of Ten Commodity Aggregates273 Highest Correlations between Net Exports and FactorSupplies per Worker in 1988371 INTRODUCTIONAble research assistance for this study was performed by Robert Murdock. Researchsupport was provided by a National Science Foundation to the HECKSCHER-OHLIN factor-proportions THEORY of compar-ative advantage, international commerce compensates for the unevengeographic distribution of productive is obvious insome respects but not so obvious in others.

5 It is not a great theoreticaltriumph to identify conditions under which countries rich in petroleumreserves export crude oil, and it would not be a great surprise to findsupportive evidence. But it is a theoretical triumph to find conditionsunder which countries that are richer in labor than land export labor-intensive agricultural products and, as a result of trade, have wages thatapproach levels prevailing in high-wage labor-scarce countries. And itwould be a great surprise to find supportive basic insight of the HECKSCHER-OHLIN (HO) MODEL is that tradedcommodities are really bundles of factors (land, labor, and capital). Theexchange of commodities internationally is therefore indirect factorarbitrage, transferring the services of otherwise immobile factors ofproduction from locations where these factors are abundant to loca-tions where they are scarce.

6 Under some circumstances, this indirectarbitrage can completely eliminate factor-price differences. Perhaps themost important implication of the HO MODEL is that the option to sellfactor services externally (through the exchange of commodities)transforms a local market for factor services into a global market. As aresult, the derived demand for inputs becomes much more elastic, andalso more similar across feature that goes hand in hand with an elastic labor-demandfunction is an aggregate gross domestic product (GDP) with a relativelyconstant marginal productivity of capital. This is a critical propertybecause growth induced by capital accumulation is generally limited bythe declining marginal productivity of an HO MODEL of a1 Flam and Flanders offer an account of the MODEL s intellectual history in theirintroduction to Heckscher and Ohlin (1991).2 Romer (1986) sparked a boomlet of sustainable-growth models that emphasizeexternalities but that depend critically on a constant or increasing long-run marginalproductivity of open economy, however, the potential decline in the marginalproductivity of capital is completely offset by a shift in the product mixtoward capital-intensive products.

7 In a closed economy, by contrast,shifts in the product mix are necessarily more limited because every-thing has to be sold internally. Thus, growth is more easily sustained inopen economies than in closed s the THEORY . What about the facts? Is there any substantialevidence that international commerce compensates for the unevengeographical distribution of factors of production? If there is an associ-ation between trade and factor abundance, which is the direction ofcausation? What resources should be considered internationally immobileand over what period of time? Is the derived demand for labor actuallymore elastic in an open economy than a closed one? Is growth sustain-able in open economies but not in closed ones?Facts casually and not so casually collected seem to be adding up toa convincing case against the HO MODEL . The first was Leontief s(1953) troubling discovery that imports in 1947 were more capitalintensive than exports.

8 For several decades, this blow to the HOmodel was thought to have knock-out power, but Leamer (1980)showed that it missed the mark because of a misreading of the , Leamer, and Sveikauskus (1987) did not intend to attack theHO MODEL but, although doing the correct calculation, found whatseems to be a disappointingly small association across countries betweenfactors embodied in trade and factor set of troubling facts was provided by Grubel and Lloyd(1975), who cataloged the surprising amount of two-way trade in evenfinely disaggregated trade data. Furthermore, trade among the industrialcountries has been growing much more rapidly than output, even asthese countries have apparently become more similar in their these troubling facts have been accumulating, a formidablegroup of trade theorists led by Brander, Dixit, Grossman, Helpman,and Krugman have been crowding the HO MODEL out of academicdiscourse by publishing a vast array of interesting models that focus oneconomies of scale and strategic the HO MODEL remains very much alive and well, residinghappily and prominently in every textbook on international economicswritten by authors fond of the artistic diagrams and simple, remarkabletheorems associated with the HO viewpoint.

9 Without saying so explicitly,these textbook writers remind us that theories are neither true nor are sometimes useful and sometimes not so useful. These2authors understand that data analysts may hit the HO MODEL so hardthat it hollers false, and that theorists may pin the MODEL so firmly tothe mat that it squeals impressed, but the authors have not heard,nor do they imagine ever to hear, the HO MODEL scream useless. Infact, the HO MODEL is extraordinarily useful pedagogically, politically,and , the MODEL offers a loud wake-up call to the limitationsof partial-equilibrium thinking. Does an increase in the supply of laborlead to a lower wage rate? Of course, is the partial-equilibrium answer. Not necessarily, is the HO answer, because trade allows the potentialeffect of an increase in the labor supply to be partly, and sometimesfully, offset by a shift of the product mix in favor of sectors that uselabor intensively.

10 The increased supply of labor-intensive products canbe absorbed externally with little or no effect on product , the HO MODEL lends intellectual support to the warning keep your crummy government mitts off international trade. Thiscontrasts greatly with the appeal inspired by the modern trade theo-rists, fight the decline of America; support an industrial policy beforeit s too late. According to the HO MODEL , tariffs and quotas haveredistributive effects but reduce efficiency. When redistribution is thelegitimate goal, there is generally a better way to accomplish it. Forexample, Ross Perot s social tariff to offset the wage advantage ofMexico, Latin America, and Asia might help to maintain high wages forunskilled workers in the United States, but it would be a terribly costlyway to achieve only modest changes in the income new trade theorists may see the issue differently; they havebombarded us with models showing that the interest of the UnitedStates lies in protecting certain sectors, thereby encouraging produc-tion on a more efficient scale and also giving firms a strategicadvantage over foreign rivals.