8829

8/14/2019 8829

http://slidepdf.com/reader/full/8829 1/91

Paying the Bill: Manufacturing and America's Trade Deficit

June 1988

NTIS order #PB88-229539

8/14/2019 8829


Recommended Citation:

U.S. Congress, Office of Technology Assessment, Paying the Bill: Manufacturing

and America’s Trade Deficit, OTA-ITE-390 (Washington, DC: U.S. GovernmentPrinting Office, June 1988).

Library of Congress Catalog Card Number 88-600548

For sale by the Superintendent of DocumentsU.S. Government Printing Office, Washington, DC 20402

8/14/2019 8829


Foreword

In the 1980s, the United States has experienced large current account deficits, par-ticularly in manufactures trade. This special report analyzes the causes of the deteriora-tion in America’s trade performance and examines the importance of U.S. manufac-turing in helping the nation improve its position in international trade. The report was

requested by Senator John Heinz as part of an assessment of technology, innovationand U.S. trade requested by the Senate Committee on Finance; the Senate Commit-tee on Banking, Housing and Urban Affairs; and the House Committee on Banking,Finance and Urban Affairs. A final report will be published in 1989.

In recent years, the Federal budget deficit, the overvalued dollar, and high real in-terest rates have helped to boost domestic consumption and increase imports. Somecountries have concentrated on exporting to the U.S. market, while keeping their ownmarkets relatively closed. Another very important factor is that the United States haslost its once substantial edge in manufacturing technology. Reversing these trends willnot be easy.

The relative decline of U.S. manufacturing is worrisome. Although services tradeand employment is growing, manufacturing remains vitally important to the U.S.economy. Manufactured goods continue to dominate in international trade, many ser-vice industries depend heavily on manufacturing, and manufacturing remains an im-portant source of well-paid jobs. The United States has to improve its manufacturingperformance if it is to maintain its economic strength.

A weaker dollar has helped to increase exports of U.S. manufactures in the firstquarter of 1988, but counting on the lower dollar alone to sell American manufacturedgoods is a shaky strategy with risks of painful adjustments. Changes in fiscal and tradepolicies will be needed. Additionally, improved manufacturing competitiveness – the

ability to make high-quality goods at reasonable costs, without sacrificing our standardof living to get costs down – will be crucial for the United States to eliminate the tradedeficit.

OTA thanks the panel members, reviewers and other individuals in government,business, labor, and academia who provided data and advice. As with all OTA reports,the responsibility for content is OTA’s alone.

JOHN H. GIBBONS Director

Ill

8/14/2019 8829


Technology, Innovation, and U.S. Trade

Advisory Panel

Lewis Branscomb, ChairmanHarvard University

Michael AhoCouncil on Foreign Relations

Ralph GomoryIBM

Joseph GrunwaldThe Institute of the Americas

Thomas HoutBoston Consulting Group

Ramchandran JaikumarHarvard Business School

Franklin P. Johnson, Jr.Asset Management Company

Lester C. Krogh3M

Paul R. KrugmanNational Bureau of Economic Research

Alvin P. LehnerdSteelcase, Inc.

Ann MarkusenNorthwestern University

Ray MarshallUniversity of Texas

Regis McKennaRegis McKenna, Inc.

Richard S. MorseConsultant

David MoweryNational Academy of Sciences

Paula SternCarnegie Endowment for International Peace

Brian TurnerAFL-CIO

Gus TylerInternational Ladies Garment WorkersUnion

Lewis C. Veraldi

Ford Motor Company

Ezra F. VogelHarvard University

NOTE: The Advisory Panel provided advice and comment throughout the assessment, but the members do not necessarily approve,disapprove, or endorse the report for which OTA assumes full responsibility.

iv

8/14/2019 8829


OTA Project Staff - Paying the Bill: Manufacturing and America’s Trade Deficit

Lionel S. Johns, Assistant Director, OTA

Energy Materials, and International Security Division

Audrey Buyrn

Industry, Technology, and Employment Program Manager Julie Fox Gorte, Project Director

Katherine Gillman, Deputy Project Director

Philip Shapira, Analyst

Brenda A. Brockman, Analyst Carol Henriques, Research Assistant

Administrative Staff

Edna M. Thompson, Administrative Assistant Diane White, Secretary

Acknowledgments

This special report was prepared by the staff of the Industry, Technology, and Employ-ment Program of the Office of Technology Assessment. The staff wishes to acknowledge

the contribution of the Advisory Panel, and to thank the following organizations for theirgenerous assistance:

Congressional Budget OfficeCongressional Research ServiceU.S. Department of CommerceBureau of Economic Analysis

Office of Trade and Investment AnalysisU.S. Department of LaborBureau of Labor Statistics

8/14/2019 8829


Contents

Page

Summary . . . . . . . . . . . . . . . . . . . . . . .

The Trade Deficit: In What and To Whom?

Causes Of The Trade Deficit . . . . . . . . .Signs of Weakness in U.S. Manufacturing .A Manufacturing and Service Economy . . .Conclusion . . . . . . . . . . . . . . . . . . .A Note About the Special Report . . . . . .

U.S. Trade Performance . . . . . . . . . . . . .

What is the Trade Deficit? . . . . . . . . , .

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

Manufacturing and the Merchandise Trade Deficit

The Causes of the Deteriorating Trade Balance . . . .The Macroeconomic Forces . . . . . . . . . . . . .

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

q

The Declining Competitiveness of U.S. Manufacturing

U.S. Leadership in Technology . . . . . . . . . . . . . . . .

U.S. Manufacturing Performance . . . . . . . . . . . . . . .

The Share of Manufacturing in the U.S. Economy . . .Manufacturing Employment and Wages . . . . . . . .Productivity Growth: International Comparisons . . .

Why Manufacturing Matters . . . . . . . . . . . . . . . . .

Links Between Manufacturing and Services . . . . . .Manufacturing and the Quality of Jobs . . . . . . . . .High Technology Industries . . . . . . . . . . . . . . .

The Anatomy of Trade . . . . . . . . . . . . . . . . . . . . .

Products . . . . . . . . . . . . . . . . . . . . . . . . . .Countries . . . . . . . . . . . . . . . . . . . . . . . . . .International Companies . . . . . . . . . . . . . . . . .

Climbing Out: How To Reduce the Trade Deficit . . , . .

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

,

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

,

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

. 1

. 2

. 4

. 5

. 6

. 7

. 7

. 9

.10

.13

.17

.17

.22

.26

.36

.37

.42

.45

.53

.53

.58

.62

.66

.66

.68

.76

.79

vi

8/14/2019 8829


List of Tables

Table No. Page

1.2.

3.

4.5.

6.

7.

8.

9.

10.

11.

Simplified U.S. Balance of Payments Statement . . . . . . . . . . . . . . . . . . . .11Business-Funded R&D As a Percentage of Gross Domestic Product . . . . . . . .27Manufacturing Share of Gross National Product, 1979-86 . . . . . . . . . . . . . .39

Real per Capita Spending On Goods and Services . . . . . . . . . . . . . . . . . . .41Annual Percent Changes in Manufacturing Productivity, Seven Countries(1960-86) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..45Index of Manufacturing Output and Employment, 1986; and ProductivityGrowth Rates, 1979-86 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .47Average Annual Changes in Real Gross Domestic Product per EmployedPerson, 1960-86 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .48Workforce Involved in Manufacturing and Average Full-Time EquivalentCompensation, 1984 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .56Productivity in Manufacturing and All Businesses, 1960-87 . . . . . . . . . . . . . .59Value-Added per Hour, by Industry, 1986 . , . . . . . . . . . . . . . . . . . . . . .60

U.S. High Technology Manufacturing Industries . . . . . . . . . . . . . . . . . . . .6312. Trade Balance in Selected Manufacturing Industries . . . . . . . . . . . . . . . . .6713. Major U.S. Imports From and Exports to Japan, 1986 . . . . . . . . . . . . . . . . .7014. Major U.S. Imports From and Exports to Canada, 1986 . . . . . . . . . . . . . . . .7215. Major U.S. Imports From and Exports to Western Europe, 1986 . . . . . . . . . .7316. Major U.S. Imports From and Exports to East Asian NICs, 1986 . . . . . . . . . .7417. Major U.S. Imports From and Exports to Latin America, 1986 . . . . . . . . . . . .7518. Balance of Merchandise Trade, U.S. Parent Companies and Majority-

Owned Foreign Affiliates, 1977 and 1982-85 . . . . . . . . . . . . . . . . . . . . . .7619. Balance of Merchandise Trade, Foreign Companies and U.S. Affiliates,

1977-85 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .78

20. Average Prices, Imports to the United States . . . . . . . . . . . . . . . . . . . . . .8 1

List of Figures

Figure No. Page

1.2.

3.

4.

5.

6.

7.8.

9.

lo.

U.S. Current Account Balance, 1960-87 . . . . . . . . . . . . . . . . . . . . . . . . . 1U.S. Manufacturing Trade Balance, 1971-86 . . . . . . . . . . . . . . . . . . . . . . 3Largest U.S. Trade Deficits by Country, 1987 . . . . . . . . . . . . . . . . . . . . . . 3Goods and Services Trade, Percent of GNP (1960-87) . . . . . . . . . . . . . . . . 9Net U.S. International Investment Position, 1971-87 . . . . . . . . . . . . . . . . .13Composition of U.S. Merchandise Exports and Imports, 1967-86 . . . . . . . . . .15

All Government Purchases, Percent of GNP . . . . . . . . . . . . . . . . . . . . . .19Federal Government Purchases, Percent of GNP . . . . . . . . . . . . . . . . . . .19Personal Consumption, Percent of GNP . . . . . . . . . . . . . . . . . . . . . . . .19Short Term Real Interest Rates, United States, Japan, and West Germany . . . .21

vii

8/14/2019 8829


Figure No. Page

11. U.S. International Investment Position (Cumulative, 1971-86) . . . . . . . . . . .2112, U.S. Share of World Import and Export Markets, 1955-85 . . . . . . . . . . . . . .2313. Index of Effective Exchange Rages for the U.S. Dollar, 1976-86 . . . . . . . . . .2314. Non-Defense Research and Development, Percent of GNP . . . . . . . . . . . . .2715. Scientists and Engineers in Research and Development . . . . . . . . . . . . . . . 28

16.

17.

18.19.20.21.22.23.24.

25.26.27.28.29.

U.S. Science and Engineering Bachelors Degrees Granted, Percent of Total Degrees Granted . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .29U.S. Engineering Ph.D.s Granted to U.S. Citizens and Foreign Citizens,1960-86 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .29Trends in Technological Innovation . . . . . . . . . . . . . . . . . . . . . . . . . .31U.S. Patent Grants by Nationality of Inventor, 1960-86 . . . . . . . . . . . . . . .31External Patent Application by Nationality of Applicant . . . . . . . . . . . . . . .32Manufacturing Share of U.S. Gross National Product (current dollars) . . . . . .38Manufacturing Share of U.S. Gross National Product (constant dollars) . . . . . .38Distribution of U.S. Employment by Sector, 1870-1986 . . . . . . . . . . . . . . .43Gross Fixed Capital Formation in Manufacturing, 1973-85 . . . . . . . . . . . . .50

Manufacturing Productivity in Japan and the United States . . . . . . . . . . . . .51Balance of Trade in Automotive Products, 1976-85 . . . . . . . . . . . . . . . . . .66Volume of U.S. Exports, 1978-86 . . . . . . . . . . . . . . . . . . . . . . . . . . . .69Volume of U.S. Imports, 1978-86 . . . . . . . . . . . . . . . . . . . . . . . . . . . .69Average Price, Imported Motor Vehicles and Parts . . . . . . . . . . . . . . . . . 81

Vlll

8/14/2019 8829


Summary

In the 1980s, the United States livedbeyond its means to an extent unimaginablea few years before. Consumption rose — both

in absolute terms and as a percentage of GNP–with consumption of foreign-madegoods leading the way. Imports grew at anaverage rate of 8 percent per year between1980 and 1987, far outpacing exports. Invest-ment recovered soon after the 1982 reces-sion. Federal government spending surgedahead of reduced tax revenues, causing thebiggest peacetime budget deficits in U.S. his-tory. And in the process, the United States,a creditor nation since World War I, quickly

became the world’s leading debtor. Its net in-debtedness exceeded $400 billion in 1987,and could reach $1 trillion by the early 1990s.

The U.S. current account balance–themost comprehensive measure of trade ingoods and services —was stable throughoutthe 1950s and 60s and experienced sometremors in the 1970s. Then, beginning in

1981, it nosedived (figure1). The only waythe United States was able to sustain thisdeficit was with loans and investments from

abroad. A massive infusion of foreign capitalallowed Americans to live beyond theirmeans. It cannot continue, though, andtherein lies the problem.

No nation, not even one as rich as theUnited States, can go on forever paying itscurrent account deficit with foreign capital.A time of reckoning will come. As theUnited States sinks deeper into debt, foreigninvestors and creditors — central banks, in-

dividuals and firms –will be less inclined tocommit ever-increasing amounts of capitalto a $4 trillion economy on a spending spree.

The trade deficit will go away. As the floodof foreign capital ebbs, the United States willbe forced to rein in government spending,business investment, or consumption – or all

Figure 1.

U.S. Current Account Balance1960-87

Billions of dollars50

0

-50

-100

-150

-200

/’-

\ ,

1960 1965 1970 1975 1980

SOURCE: U.S. Department of Commerce, Bureau of Economic Analysis, Business Statistics:ment Printing Office, 1987) Appendix II, U.S. International Transactions, p. 250; U.S.Economic Analysis, Survey of Current Business, March, 1988, p. 31, table D.

1985

1986, (Washington, DC: U.S. Govern-Department of Commerce, Bureau of

1

8/14/2019 8829


2 q Paying the Bill

three. Whether this comes about throughslower growth, or through a recession thatcuts investment and consumption in ab-solute terms, will depend on how competi-tive American manufacturers are and how

fast other major economies are growing.One way or another, exports will have to ex-ceed imports. A recession could force this tohappen, by cutting consumption and thusrestraining imports. So could a further dras-tic fall in the value of the dollar, raising theprice of imported goods beyond the meansof many consumers, making video cassetterecorders, imported cars, and so on luxuryitems for the few.

A less painful course is not only to makeneeded macroeconomic adjustments, butalso to get better at manufacturing — to makea wide range of high quality goods at com-petitive costs. That is the most constructiveway to recapture some of our own marketsand raise exports. Such gains will not be easyto win, however; they will require con-centrated efforts on the part of U.S.producers to improve manufacturingproductivity and quality. And they will re-

quire redoubled efforts on the part of theU.S. government to promote Americanmanufacturing; for example, through exportpromotion and through policies that willease the pressures on manufacturers to pur-sue short-term profits at the expense of longer term investments in technology andmarket share.

The Trade Deficit: In What and To

Whom?

The U.S. trade deficit is mostly a deficit inthe trade of manufactured goods. Of the$161 billion current account deficit in 1987,85 percent was in manufacturing trade

(figure 2). The growing U.S. service sectorcannot generate sufficient trade to offsetcontinuing deficits in manufactured goodstrade. The services trade is simply not bigenough; goods can be stored and shipped

while services by and large cannot.Moreover, the surpluses the United Stateshas enjoyed in services trade are shrinking.Other nations have become more competi-tive in an array of services that are traded in-ternationally – from engineering to bankingand software design.

Nearly three-quarters of the U.S. manufac-turing trade deficit is in three product areas:

qmotor vehicles and parts (a $53 bil-lion deficit);

q textiles, apparel and shoes (a $28billion deficit);

qelectronics, especially semiconduc-tors, telecommunications equip-ment and consumer electronicitems (a $22 billion deficit).

The countries with which the United Statesruns the largest trade deficits are, in order:

Japan, Taiwan, West Germany, Canada,South Korea, Hong Kong, Italy, Mexico,Brazil, and Great Britain (see figure 3).

Japan accounted for 36 percent of the U.S.merchandise trade deficit in 1987— about$57 billion. From Japan came 21 percent($85 billion) of U.S. merchandise imports,but to Japan went only 11 percent ($28 bil-lion) of U.S. merchandise exports. The lead-ing Japanese import by far was motor

vehicles and parts – about 30 percent of allimports from that country. Other major im-ports from Japan include consumerelectronic products, telecommunicationsequipment, computers and their attach-ments, other office machinery (e.g., copyingmachines), and semiconductors.

8/14/2019 8829


Summary q 3

Figure 2.U.S. Manufacturing Trade

1971-86Billions of dollars

40

20 -

0

-20 -

-40

-60

-80

-1oo

-120I

-140 -

-160I 1 I I

1971 1973 1975 1977 1979 1981 1983 1985

‘ B E A OTIA

NOTE: Bureau of Economic Analysis firgures are merchandise trade less petroleum imports and agriculture exports,

SOURCE: U.S. Department of Commerce, Bureau of Economic Analysis, Survey of Current Business, Table 3, June 1982 and 1987;U.S. Department of Commerce, Office of Trade and Investment Analysis, Presentation by Allen Lenz, “U.S. Trade Deficitsand International Competitor. ”

Figure 3. Largest U.S. Trade Deficits

Mexico

Italy

Hong Kong

South Korea

Canada

West Germany

Taiwan

Japan

by Country,

I

1 1 I I 1 1

0 10 20 30 40 50 60

Tens of billions of dollars

1987

SOURCE: U.S. Department of Commerce, Office of Trade and Investment Analysis, 1988 unpublished data.

8/14/2019 8829


4 q Paying the Bill

The deficit with Japan has accounted forone-third to one-half of the U.S. merchan-dise trade deficit for the last decade, growingtenfold in that time from $5.5 billion to al-most $57 billion. The U.S. merchandise

trade deficit with Asian countries other thanJapan has also grown significantly over thepast decade. By 1987 it had reached $47 bil-lion, of which nearly three-quarters was withTaiwan, South Korea, Hong Kong, and Sin-gapore.

In 1980, the United States had a merchan-dise trade surplus of $20 billion withWestern Europe. By 1987, this surplus hadturned into a deficit of $27 billion, with West

Germany accounting for more than half ($15billion). Automotive products are the num-ber one item in the U.S. merchandise tradedeficit with Western Europe.

As the U.S. deficit declines, the countriesexporting most to the United States will haveto adjust to exporting less –or at the least,slowing the growth of exports. Although theadjustment will not be easy for anyone,countries that can expand consumption in

their own economies, and that have low un-employment rates, strongly competitivemanufacturing industries, and healthy tradesurpluses, are best equipped to weather thechanges.

Causes of the Trade Deficit

There is no one cause and no single cure.

Macroeconomic policies certainly con-tributed to the deficit. In the 1980s, theUnited States has pursued expansionary fis-cal policies, while most other industrializednations acted to restrain their deficits. As aresult, the United States needed to borrowmoney, and real interest rates had to rise toattract it. In response countries such as Japan

and West Germany invested their savings inthe United States. This, in turn, increased thedemand for dollars and pushed up thedollar’s value. The strong dollar made goodsproduced in the United States more expen-

sive for foreigners and foreign goodscheaper for Americans.

But the strong dollar is only part of the storybehind the U.S. trade deficit. The dollarpeaked in the first quarter of 1985 and sincethen its value has fallen by one-third relativeto other major currencies. It is now atpostwar lows against the yen and the Germanmark. Despite this 3-year decline, anddespite the recent upsurge in exports, the

U.S. merchandise trade deficit was runningat an annual rate of well over $100 billion in1988. The deficit with Japan hit a new recordin 1987, and only began to decline in the firstmonths of 1988. It seems that the devalueddollar spurred U.S. exports, but it did notreduce merchandise imports until April1988.

There is further evidence that something inaddition to currency exchange rates is at

work here. U.S. manufacturers of productsas diverse as automobiles, integrated circuitsand color televisions began to lose theirworld market share well before the dollar’srise. Moreover, since about 1970, U.S.manufacturers have been able to hold on totheir shares of world markets only when thedollar’s value is falling, making U.S.-madegoods progressively cheaper compared togoods made in other countries. This suggestsloss of competitiveness.

Of course, the United States cannot expectto dominate world markets to the extent itdid in the first couple of decades after WorldWar II. War-damaged industrial countriesrecovered, and the diffusion of capital andtechnical knowledge made it possible forsome of the poorer countries to achieve

8/14/2019 8829


Summary q 5

vigorous economic growth. The worldeconomy became richer — a desirable result,and one which has long been the aim of U.S.policy.

The fact remains that the U.S. market – thelargest and richest in the world, and one of the most open to foreign goods – is the bestprospect for both developed and developingcountries to cultivate. Some of thesecountries have concentrated on exports andkept their own markets relatively closed, asa development strategy. Few nations havefaced the kind of competitive pressure theUnited States is under. While somedeveloped nations have labor costs com-

parable to those of the United States, mostnations have much lower wages. U.S. capitalcosts have also been higher than those of most other developed nations. The combina-tion of these disadvantages and the attrac-tiveness of the American market to mostforeign producers (in developed anddeveloping countries alike) means that theUnited States must do a great many thingsvery well, just to stay even with the competi-tion.

Quite a few signs indicate that U.S.manufacturing is not staying even.

Signs of Weakness in U.S.

Manufacturing

U.S. pre-eminence in many manufacturingindustries has evaporated. For example, onlyone U.S.-owned company is still makingcolor TV sets, and most of its productiontakes place in Mexico. No U.S. companymakes video cassette recorders or compactdisc players. Mass production of automobiles was invented in the UnitedStates, but others (especially the Japanese)

are now leaders in the technology andmanagement of auto manufacture. Of the10.3 million passenger cars bought byAmericans in 1987, 3.1 million came fromJapan, despite the quota on these imports.

Another 620,000 cars were built in NorthAmerica in Japanese-owned plants; stillanother 1 million cars were imported fromother countries.

What is behind these losses? There aresigns that the United States is losing its oncesubstantial edge in technology, a crucial fac-tor in competitiveness for an advanced, high-wage nation. For example, the United Statesis spending a smaller share of gross national

product on the kind of research and develop-ment likely to pay off commercially than itsmajor competitors; U.S. civilian R&Dspending was less than 1.9 percent of GNP in1985, compared to Japan’s 2.8 percent andWest Germany’s 2.5 percent. Japaneseprivate businesses are even farther ahead inspending on R&D, devoting 2.1 percent of GNP to the purpose in 1986, compared to 1.4percent for U.S. businesses.

In the human skills needed for technologi-cally advanced manufacturing, the UnitedStates is also losing ground. We are graduat-ing and using just over half as many en-gineers per capita as Japan; and our publicschools are turning out young people who donot measure up internationally, especially inmath and the sciences.

The heart of the matter, however, iswhether American manufacturers have fal-

len behind in the practical application of technology. The available evidence suggeststhat they have. One study of flexiblemanufacturing systems – computer control-led systems that are designed to make dif-f e r e n t k i nds o f pa r t s i n sm a l lbatches–concluded that American firms

8/14/2019 8829


have no edge at all in this advanced form of automated manufacturing. On the contrary,they used the technology far less effectivelythan the Japanese. The American flexiblemanufactured systems produced many fewer

kinds of parts, took longer to develop, andperformed less reliably.

Another example comes from auto designand manufacture. U.S. auto companiesspend, on average, over 5 years taking amodel from the initial concept to fullproduction. Japanese companies take only alittle over 3 1/2 years to do the same–andthey do it with about half as many engineer-ing hours. The Japanese advantage appears

to come from such things as putting a singleboss in charge of the project, getting thecompany’s research/development/designpeople and its production people to com-municate with each other, ironing out con-flicts early, and treating product and processdesign as simultaneous rather than sequen-tial.

There are other Japanese strengths.Among those most often cited are close at-

tention to product quality and reliability,consensus building, and emphasis on long-term market share rather than short termprofit. Not all Japanese firms share thesecharacteristics, and some American firmsdo. But firsthand observation, case studies,and the remarkable record of Japanese in-dustrialization and adaptation in the postwarperiod support the basic point: Japanesemanufacturers have moved into a command-ing position in many industries and have sur-

passed U.S. rivals in many importantmarkets, by developing and applying tech-nology.

U.S. manufacturers have responded to theJapanese challenge (and the challenges fromTaiwan, Korea, Germany, and so on) in avariety of ways, some effective, and some lessso. Overall, American manufacturing has not

yet recouped the losses of recent years. Asone departing chief executive officer of amajor U.S. manufacturer told the New York Times: “Yes, American industry has im-proved over the past four or five years, butso have our competitors.”

1

A Manufacturing and Service

Economy

Th e United States cannotstrong manufacturing sector.

do without aManufactured

good are indispensable for trade with othernations. It is also clear that America has notentered a post-industrial stage in whichdemand for goods gives way to demands forservices. Demand for manufactured goods isas great as ever – greater, for everything butthe basics, food and fuel. American con-sumers, businesses and government nowdevote over 30 percent of all their spending

to manufactured goods other than food andfuel, compared to 23 percent 30 years ago.

More fundamentally, to speak of servicesas taking the place of manufacturing is tooverlook the strong interdependence of thetwo kinds of activities and the blurring of dis-tinctions between them. Many service in-dustries depend heavily on manufacturersfor business. And some manufacturing in-dustries could hardly exist without allied ser-

vices –for example, the manufacture of computers and design of software, often byan independent firm.

I Robert Anderson, former chief executive officer, Rockwell International Corporation, quoted in Claudia H. Deutsch, “U.S. Industry'sUnfinished Struggle,” The New York Times, Feb. 21,1988, sec. 3.

8/14/2019 8829


Summary q 7

Manufacturing remains extremely impor-tant for employment in the U.S. economy.Nearly 28 million Americans–about one-quarter of the work force – make their livingin manufacturing, either directly or in

providing services or materials to manufac-turing. Far from replacing manufacturing assource of employment, in the manner thatmanufacturing took the place of agriculture,service industries include a good many jobsthat depend on manufacturing – and by andlarge, these are well-paying jobs. Not onlyare manufacturing wages, on average, higherthan wages in the service sector; most of the

jobs in services that are closely tied in withmanufacturing also pay better than average.

To keep those good jobs, America has tocompete effectively in the production of goods, as well as the provision of services.

High technology industries cannot take theplace of traditional manufacturing, any morethan services can simply replace themanufacturing sector as a whole. Certainly,high tech industries are vital to the genera-tion of jobs, wealth, exports, and the advanceof technology in other industries. But they do

not stand alone. The best customers of hightech industries such as semiconductors areother industries, both high tech (e.g., com-puters) and traditional (e.g., autos). Nor canthe high technology industries, by themsel-ves, compensate for trade deficits in declin-ing traditional industries. The trade balancein high technology products shrank from asurplus of $27 billion in 1980 to a surplus of only $600 million in 1987— with an interven-ing deficit of $2.6 billion in 1986. U.S. high

technology industries are still quite competi-tive, but they are unlikely to regain thedominance they enjoyed 10 years ago or togenerate the large trade surpluses of thatlime.

Conclusion

Counting on the lower dollar alone to sellAmerican manufactured goods is a shaky

and potentially painful strategy. If theUnited States is to maintain its standard of living and live within its means, it will haveto reduce the Federal budget deficit, in-crease its access to foreign markets, andmake its manufacturing sector more com-petitive. As yet, some progress has beenmade on some of these fronts, but moreground remains to be gained. Improvingmanufacturing competitiveness — the abilityto make high-quality goods at reasonable

costs, without sacrificing our standards of living to get costs down – will be crucial if theUnited States is to remain aeconomic power.

A Note About the Special Report

first-class

This special report is an interim product of the full assessment Technology, Innovationand U.S. Trade. This report describes thecauses and anatomy of the U.S. trade deficit,

and discusses the role and health of manufacturing within the U.S. economy.

The full assessment will analyze the recordof American manufacturing companies indeveloping and applying new product andprocess technologies, with particular em-phasis on how we have lost or could bolstertechnological advantages. The full assess-ment will also examine the extent to whichhigh capital costs, and relationships of

manufacturers with providers of capital,have limited the ability of U.S. manufac-turers to make needed investments in tech-nology development.

8/14/2019 8829


8 q Paying the Bill

In addition, the full assessment will discusshow Federal government policies promoteor hinder technology development and itsapplication to manufacturing. It will assesshow foreign government trade, industrial,

and technology policies affect U.S.manufacturers’ access to foreign marketsand their ability to compete in the U.S.market. That part of the assessment will con-centrate on Japan and Asia’s newly in-dustrializing countries, where the most

significant technological progress has beenand will likely be. The full assessment willalso evaluate how U.S. trade policies have af-fected American manufacturing, both inpromoting increased exports and in coping

with rapidly rising imports. Policy optionswill focus on fostering technology develop-ment and application in manufacturing,building technological advantage, promot-ing exports and opening foreign markets,and alternatives for dealing with imports.

8/14/2019 8829


U.S. Trade Performance

A nation’s economic health can bemeasured in many ways. Common measuresinclude Gross National Product, per capita

income, wages and unemployment rates, lifeexpectancy, literacy rates and educational at-tainment. The balance of international tradeis one important indicator of the ability of anation’s firms and industries to compete in-ternationally. A nation’s economic and tech-nological strength and weaknesses arereflected in its trade figures.

In the mid-1980s, for the first time in recenthistory, the trade accounts of the United

States have gotten seriously out of balance.In the 1950s and 1960s, the U.S. was accus-tomed to running modest trade surpluses. Inthe 1970s and early 1980s, small deficitsbegan to appear, but both deficits andsurpluses remained lower than one percent

of GNP. In the mid- 1980s, the trade deficitballooned; in 1987, the current accountdeficit was a record-high $161 billion, or 3.6

percent of GNP2 Before 1983, the currentaccount surplus or deficit had not exceeded1.2 percent of GNP.

3

Simultaneously, the importance of interna-tional trade to the American economy wasgrowing: imports of goods and services in-creased from 4.7 percent of GNP in 1960 to12.2 percent in 1987, while exports expandedfrom 5.8 percent to 9.5 percent (figure 4).

4

The expansion was not smooth. In 1980, ex-

ports totaled nearly 13 percent of GNP, andhave since fallen in percentage terms. Im-ports grew at about 8 percent per year, onaverage, from 1980 to 1987; meanwhile, ex-ports grew unevenly, falling and then risingagain for an average annual growth rate of

Figure 4Goods and Services Trade, Percent of GNP

1960-87 “Percent of GNP

14

12

10

8

6

41960 1965 1970 1975 1960 1985

—E x p o r t s Imports

SOURCE: U.S. Department of Commerce, Bureau of Economic Analysis, National Income and Product Accounts, Table 1.1, 1987electronic data,

2 The United States keeps account of trade balances using a variety of partial balances, as discussed below.3 U.S. Department of Commerce, 13ureauof Fkonomic Analysis, The National Income and Product Accounts of the United Slates, 1929-82,

(Washin on, IX: U.S. Government Printing Office, September, 1986); and U.S. Department of Commerce, Bureau of Ikonomic Analysis,F’Survey o Current 13usinexs, various issues.

4 Ibid.

9

8/14/2019 8829


10 q Paying the Bill

3.2 percent over the period. Before 1983, ex-ports and imports tended to grow or falltogether, as percentages of GNP. It is themarked divergence of imports and exportsthat accounts for the unprecedented deficits

of the 1980s.

The dominance of the United States inworld markets in the 1950s and 1960s wasnever expected to be a permanent condition.Europe and Japan were rebuilding their in-dustrial bases after the devastation of WorldWar II, often using newer and more efficienttechnologies. The international trading sys-tem of the General Agreement on Tariffsand Trade, and various programs of

economic development aid, were designedto help both war-ravaged industrial nationsand developing countries along. The fact thatmany newly industrialized nations in Asiaand Latin America were able to achieverapid growth in the past few decades is atleast partly testament to the success of suchprograms. Often, in order to develop orrebuild, developing and developed countriesalike controlled access to their own markets,using them as incubators for their owndeveloping industries. While these develop-ments can all be viewed positively, as con-tributing to world economic growth anddevelopment, they have also begun topresent problems for American industries.Limited access to many foreign marketspresents problems for U.S. exporters, whilerelatively open access to our own marketgiven to countries such as Japan, Taiwan,West Germany, South Korea, etc. increasesthe competition at home.

In short, the fact that American dominancein world goods markets has slipped is ex-

pected and even partly self-imposed. So whydo we view our trade deficits as a problem?In part, the speed of the decline in the late1970s and throughout the 1980s has been un-settling; but more fundamentally, we are

concerned that the responses U.S. manufac-turers and government have made to thedecline are inadequate to stem further los-ses. The losses are beginning to hurt. Manymanufacturing industries are in trouble,employment has fallen, whole communitiesin older industrial areas are in decline, andwages of manufacturing workers have stabi-lized well below their historical peak, in realterms. The trade deficit, then, is a manifesta-tion of a set of problems that could well be-come much worse.

Proposals for “solving” the deficit are noth-ing if not diverse, ranging from upgrading theskills of the workforce to crafting new waysof dealing with unfair trade. Some observerscounsel little action at all. They see thedeficit as self-correcting, and caution thatgovernment interference with trade regimesor factors determining trade will provecounterproductive in the end. Differentviews on what should be done — or notdone–about the trade deficit stem partlyfrom different opinions on the importance of its causes. Regardless of the policy prescrip-tion, however, an overview of the composi-tion of the trade deficit makes it clear wherethe potential problems are, and equally im-portant, where they are not.

What is the Trade Deficit?

Strictly speaking, there is no such thing as“the trade deficit.” Most often, “the trade

8/14/2019 8829


U.S. Trade Performance q 11

deficit” is synonymous with the merchandise ternational flows of goods, services, and(or goods) trade deficit, which is only one of capital are included in the U. S. balance of the partial balances commonly used to ex- payments statements (table 1). Partial balan-press the position of the United States in in- ces —such as the current account, the mer-ternational flows of goods and services. chandise trade account, and the balance onThere is no single indicator that accurately goods and services –reflect the net debit-

5Rather, in-and wholly reflects this position.

Table 1.– Simplified U.S. Balance of Payments Statement

Credits (receipts) Debits (payments)

Current accounts: Current accounts:

1 U S merchandise exports 1. U S. merchandise imports

2 U S services sold to foreign residents 2. Services purchased from foreign residentsa

b.c.

d

Foreign tourist expenditures in the U.S. a.

Fees and royalties from foreigners b.Transportation, insurance and other private and c.government servicesReceipts of income from U S (government d.and private) investments abroad

U.S. tourist expenditures abroad

Fees and royalties paid to foreignersTranspiration, insurance and other private andgovernment servicesPayments of income on foreign (governmentand private) investments in the U S

3. Unilateral transfers received from abroad 3. Unilateral transfers sent abroada. Private remittanceb Pension paymentsc Government grants

Capital account:

1 Net change in investment by foreigners in the U S.a Direct investment

b. Indirect investment

c Foreign bank loans to U S residentsd Deposits by foreigners in U S banks

e Other

2 Net change in foreign official reserve assets in the U S

a. Private remittanceb. Pension paymentsc. Government grants

Capital account:

1. Net change in U S. investment abroada. Direct investmentb. Indirect investmentc. U.S. bank loans to foreignersd. Deposits by U S. residents in foreign banks

e. Other

a U S Government securities held by monetary a.authorities b.

b. Other dollar and dollar -denominated assets held c.by foreign monetary authorities d.

3 Allocations of special drawing rights (SDRs)*

Total credits

2. Net change in U S official reserve assets abroadGold

-

Special drawing rights (SDRs)U S. reserve position in the IMFForeign currencies

Total debits

q Capital account 3 has an entry only in years when the International Monetary Fund allocates SDRs to member countries

SOURCE Arlene Wilson, “U S Trade and Payment Balances. What Do They Mean?” Congressional Research Service Report 85-26E (Washington, DC: Library of Congress, 1985)

5 Arlene Wilson, “U.S. Trade and Payments Balances: What Do They Mean?” Congressional Research Service Report No. 85-26 E,January 23, 198.5.

8/14/2019 8829



credit position of that part of U.S. interna-tional trade and transactions.

The entire balance of payments accountmust, as the name implies, balance. Its two

components, the capital account and the cur-rent account, mirror each other, at least intheory.

6A current account deficit must be

balanced by a capital account surplus of thesame amount; without capital funds comingin from abroad, something else would haveto give — consumption, investment imports,

7government spending, or all four. The cur-rent account measures international flows of goods, services, and unilateral transfers,while the capital account includes flows of

direct and indirect investment and changesin official reserve assets.

The current account– the most com-prehensive measure of trade in goods andservices — was relatively stable for twodecades following World War II, becomingmore volatile after 1970 and plunging deep-ly into deficit after 1981 (see figure 1). Thecapital account, therefore, had to show a cor-responding surplus –also unprecedented.

As a corollary, the international investmentposition of the United States has shiftedfrom surplus to deficit in the 1980s, roughlybalancing the shifts in the current account.That is, foreign investment in the UnitedStates exceeded American investment off-shore by nearly $424 billion in 1987 (figure5). This infusion of capital allows the UnitedStates to sustain its current account deficit,or to consume more goods and services thanit produces.

A nation’s ability to consume more than itproduces is attractive from the standpoint of the consumer–while it lasts. In this sense,the current account deficit has benefittedmany Americans in the short term. But a na-

tion cannot go on forever paying for its cur-rent account deficit through a surplus in thecapital account. The capital account surplusconsists of savings from other nations, whichare invested in the United States in order to

Figure 5.Net U.S. International Investment

Position

B

r — —

—--1

100

-100

-XXI

-300

-400

1971 1973 1975 1977 1979 1981 1983 1985 1987

SOURCE: U.S. Department of Commerce, Bureau of EconomicAnalysis, Survey of Current Business, June, 1987,U.S. International Transactions, table 1

provide future returns. Those returns willeventually drain away funds that, if they hadgone into the hands of U.S. nationals, mighthave been used for American consumption,investment, or public spending. Moreover,foreign investors cannot invest larger andlarger amounts of money in the UnitedStates indefinitely. At some point, con-

e In practice, there are differences (called statistical discrepancies) between the dollar amount of the capital and current accounts.Moreover J the capital and current accounts do not necessarily balance at any particular point in time it may take many months for theadjustments in one account to cause changes in the other to show up. I:or a discussion of these accounts and explanations of the items in eachaccount, see Arlene Wilson, op. cit.

7 For further discussion of the relation between the current account deficit and an influx of foreign capital, see the section on The Guscsof the Deteriorating Trade Balance, The Macroeconomic Argument.

8/14/2019 8829


U.S. Trade Performance q 13

fidence that U.S. investments can continueto yield higher returns, or more reliablereturns, will erode, or the supply of foreignsavings will be curtailed, and the massiveflows of foreign capital into the United

States will dry up.

No one can pinpoint the time when this willhappen. But most analysts expect that for-eigners will cease to finance our large cur-rent account deficit within a few years atmost.

The trade deficit for 1988 promises to besmaller than the one in 1987– the firstchange in this direction since 1980. Whilethis reduction in the trade deficit is relative-ly small, further, more consequential chan-ges in our current account are coming, andthey will necessitate adjustments on our part.What kind of adjustments? To get some in-sight on this question, it is helpful to look atthe components of international trade –what kinds of goods, services, or other ex-changes are most important to trade, andwhere the United States is running its biggestdeficits.

Manufacturing and the

Merchandise Trade Deficit

The current account measures what wecommonly think of as international trade –

exports and imports of goods and services,plus unilateral transfers.

8The merchandise

trade deficit, reflecting international flows of goods, is larger than the current account,mainly because the United States runs a

surplus in international trade in services. In1987, the current account deficit was $160.7billion, with a surplus of $14.3 billion in ser-vices trade and a deficit of $159.2 billion inmerchandise trade.

9

To reduce the current account deficit, theUnited States must reverse the deficit inmerchandise trade.

10Surpluses in services

alone cannot make much of a dent in the cur-rent account; they are dwarfed by the deficitin merchandise trade. Two kinds of activitiesare included in the services accounts: invest-ment income (e.g., dividends and interest),and trade in services such as banking, in-surance, travel, and license and royalty pay-ments. In 1987, investment income,according to Commerce Departmentfigures, produced a surplus of $14.5 billion,but trade in service activities was slightly indeficit, to the tune of $200 million.

In an earlier assessment, OTA found thatthe official figures have consistently under-stated the surplus from services trade (bank-ing, travel, and the like) .11 For example, theCommerce Department figures showed asmall surplus for services trade of $2 billionin 1984, whereas the OTA mid-range es-

a Unilateral transfers include U.S. C~ovemment grants (excluding milita~ grants of goods and seticcs), U.S. government pensions andother transfers, and private remittances and other transfers.

9 The remaining deficit of $12.8 billion was accounted for by unilateral transfers. In this section, trade figures are drawn from the national

income and prcduct accounts, which are calculated by the Commerce Department’s Bureau of E.conom ic Analysis on the free-along-side(f.a.s.) basis. Other trade figures, kept on a more current basis by the Commerce Department’s International Trade Adm inist rat ion, calculateimports on the cargo-insurance -freigh t (c.i. f.) basis. Imports figured on the c.i.f. basvs are higher, and thus make the U.S. t rade deficits appearhigher (or the surpluses lower).

10 Much of this section depends on a presentation entitled “U.S. Trade Deficits and International Competitive ness,” by Allen 1,.erw, formerdirector, Office of Trade and Investment Analysis, Department of Commerce.

I I U.S. Congress, Office of Technology Assessment, Trade in Services: Exports and Foreign Revenues, OTA-ITE316 (Washington, DC :U.S. Government Printing Office, .September 1986), ch. 4,

8/14/2019 8829



timate of the surplus for that year was $14 bil-lion. Nonetheless, even using OTA es-timates, the surplus for services is smallcompared to the merchandise trade deficit.Furthermore, the surplus from services

trade was shrinking in the years OTA madeits calculations (1982 to 1984). Investmentincome has been quite considerable in pre-vious years, peaking at $34.1 billion in 1981,but it too is declining. Because the UnitedStates is now the world’s leading debtor, itseems likely that investment income willcontinue to decline for some years.

Are services on the brink of assuming muchgreater importance in international trade,

perhaps eclipsing goods? OTA judges thatthey are not. Goods can be shipped andstored; services, by and large, cannot. Mostservices are produced very near the placethey are consumed. For that fundamentalreason, goods are much more important tointernational trade than services and arelikely to remain so for a longtime. This situa-tion may change as telecommunication be-comes cheaper and more reliable, but thechanges are likely to be gradual, not revolu-

tionary. Moreover, it is not realistic to think of services trade as a replacement for tradein goods. The manufacture of goods andprovision of services is highly interdepen-dent. If American-made goods becomemore in demand and sell better around theworld, many services will be bundled alongwith sales of those manufactured items. Forexample, the companies that have succeededbest in selling computers in the world marketare also very good at providing services suchas systems integration, training, main-

tenance of hardware and provision of up-to-date software.

For different reasons, agriculture cannotdo much either to reverse the current ac-

count deficit. Agriculture, where America isgenerally thought to be internationally com-petitive, contributed fairly strong tradesurpluses in the 1970s, helping to offset thepetroleum deficits of that time and to keepthe current account more or less in balanceduring the decade. But agricultural tradesurpluses have dwindled in recent years.Farm support programs and the widespreaddispersal of production-enhancing agricul-tural technology throughout the world have

reduced the potential for American exports.Even if U.S. agriculture were to recoversome foreign markets, agricultural trade,like trade in services, is too small to much af-fect the huge merchandise trade deficits of the 1980s.

Manufactured goods dominate merchan-dise trade. (Figure 6 shows the compositionof merchandise trade over the past twodecades.) About 80 percent of merchandise

trade, both imports and exports, is inmanufactured items. Thus, most of the mer-chandise trade deficit – and therefore, thecurrent account deficit — is in manufacturedgoods. The great deterioration of the 1980sin the merchandise trade balance was due al-most entirely to manufacturing. The deficitin petroleum trade, once a major drag onmerchandise trade balances, improved byover $40 billion between 1981 and 1986, asoil prices fell and U.S. production increased,though this situation is temporary. Theagricultural trade surplus declined from $25

12 U.S. Congress, Office of Technology Assessment, International Competition in Services, O’I’A-lTE-328 (Washington, DC: U.S.Government Printing Office, July 1987), ch. 1.

13 For reasons why this is so, see U.S. Con ess, Office of Technology Assessment, U.S. Oil Production: The Effect of Imw Oil Prices,Fpcial Report (Washington, DC: U.S. GPO, n Press).

8/14/2019 8829


8/14/2019 8829



billion to $3.4 billion. But the trade balanceon manufactured products, having fluc-tuated moderately during the 1970s, plungedinto deep deficit in the 1980s (see figure 2).Despite the upturn in exports in 1987, the

manufacturing trade balance dropped to arecord deficit of $138 billion as imports of manufactured goods continued apace.

One way to reverse the current accountdeficit is for U.S. exports to grow much fasterthan imports, and continue doing so for sometime. But this is now likely. If import growthcontinues unchecked, it is highly unlikelythat exports could grow fast enough to closethe trade deficit; this would require extreme-

ly rapid expansion of exports, and assumesan improbably high rate of growth in worldmarkets. It is more likely that U.S. import

growth will slacken or reverse, either be-cause of a recession that cuts consumption,or because the falling dollar makes importstoo expensive for Americans to afford, or be-cause we replace some imports with domes-

tic production. At the same time, exports arelikely to pick up, as foreign firms and con-sumers adjust to lower-priced Americanproducts. Indeed, U.S. merchandise exportsgrew consistently throughout 1987, rising to$258 billion. The degree to which exports canexpand further will depend on many factors,including the value of the dollar, the com-petitiveness of U.S. manufacturing firms,and the economic and trade policies of manycountries, not least our own.

14 The following

section considers how these same factorswere involved in causing the deepeningtrade deficits of the 1980s.

14 Why the trade deficit must turn around and how it may occur is discussed in more detail in the concluding sections of this report.

8/14/2019 8829


8/14/2019 8829


8/14/2019 8829


The Causes of the Deteriorating Trade Balance q 19

Figure 7.All Government Purchases, Percent of

GNP1947-87

Percent of GNP —

24%

SOURCE: U.S. Department of Commerce, Bureau of EconomicAnalysis, National Income and Product Accounts,Table 1.1, electronic data, 1987.

Figure 8.Federal Government Purchases,

Percent of GNP

Percent of GNP —

1796

15%

1947 1952 1957 1962 1967 1972 1977 1982 1987

SOURCE: U.S. Department of Commerce, Bureau of EconomicAnalysis, National Income and Product Accounts,Table 1.1, electronic data, 1987,

ly slower than the rate of growth of GNP, butthis is partly a result of the fact that grossprivate investment was already quite high in1981, compared with any previous year orwith 1982 and 1983. If 1980 is chosen as thestarting point, the rate of growth of private

investment was 7.8 percent annually; if 1979is the starting point, the rate of growth was 6percent. In short, private investment rough-ly kept pace with GNP in the 1980s, and fluc-tuated within a range that was normal for thepostwar period. What is surprising aboutthis, however, is that investment maintained

its share of GNP during a period of high (byhistorical standards) real interest rates.

Personal consumption is composed of ex-penditures on durable goods, nondurable

goods, and services. Personal consumption,as a percent of GNP, has risen sharply in the1980s. The percentage share fluctuatedwithout much sign of a long term trend fromthe 1950s through the 1970s, varying be-tween about 62 percent of GNP and 64 per-cent (figure 9). In 1982, after the recession,personal consumption expenditures shot upto about 65 percent, and rose again in 1984to two-thirds of GNP. According to manyeconomists, consumer spending has buoyed

the economic recovery since the 1982 reces-sion.

The story is different for net exports. Fromthe 1950s until 1983, net exports’ share of GNP fluctuated within historical norms.After 1983, as would be expected from theperformance of other trade accounts, thepercentage share plummeted, becoming adrain on-GNP to the tune of nearly -3 per-cent per year by 1987.

Figure 9.Personal Consumption,

GNPPercent of

Percent of GNP6 9 % —

66%

67%

66%

65%

64% ‘v’

6 3%

/’

1947 1952 1957 1962 1967 1972 1977 1982 1987

SOURCE: International Monetary Fund, International FinancialStatistics, 1987 Yearbook, and Vol. 41, No. 4,(Washington, DC: International Monetary Fund,1987, 1988) p. 229, 299, 523.

8/14/2019 8829


8/14/2019 8829



Figure 10.Short Term Real

United States, Japan,Interest rate

0 0 8 r — – — —

Interest Rates,and West Germany

I0.06

---0.02 .“

.’

o

-0.02 /

I

1978 1979 1980 1981 1982 1983 1984 1985 1986 1987

— U.S. — Japan --- W. Germany

SOURCE: International Monetary Fund, International Financial Statistics, 1987 Yearbook, and Vol. 41, No. 4, (Washington, DC: Inter-national Monetary Fund, 1987, 1988) p. 229, 299, 523.

Figure 11.U.S. International Investment Position

Cumulative, 1971-86

Billions of dollars14

12

10

8

6 -

4 .

1971 1974 1977 1980 1983 1986

Foreign Investment U S Investment

in United States overseas

SOURCE: U.S. Department of Commerce, Bureau of Economic Analysis, Survey of Current Business, June, 1986, U.S. InternationalTransactions, table 1,

8/14/2019 8829



curtailed U.S. exports to those countries. Itis tempting to zero in on one thing—mostcommonly, the value of the dollar or thebudget deficit –but this kind of over-simplification is misleading when it comes to

choosing the policies necessary to remedythe situation, and raises false hopes of asingle silver-bullet solution.

It is equally wrong to focus on macro-economic causes of the trade deficit, andmacroeconomic solutions, ignoring the com-petitiveness issue. Far too much analysis hasbeen devoted to trying to prove that eithermacroeconomic factors or competitivenessis the root of the trade deficit, without recog-

nizing the interplay and synergism betweenthem.

The Declining Competitiveness of

U.S. Manufacturing

Several features of the trade picture in thepast two decades indicate that the UnitedStates – more specifically, U.S. manufactur-ing – has lost competitive prowess. Since the

1970s, it appears that the United States hasbeen able to keep its international trade ac-counts out of the red only when the dollar isdeclining. The strength of the dollar in the1980s was not a unique occurrence in our his-tory: by some measures, the dollar’s peak value in 1985 was comparable to its ex-change-rate value in 1970. But in 1970 theUnited States had a current account surplusof $2.3 billion, compared with deficits of

$107 billion in 1984 and $116 billion in1985.

24

It is noteworthy that U.S. current accountand manufacturing trade performance

began deteriorating before the rise of the dol-lar (see figures 1 and 2). When agriculturalexports and petroleum imports — the twolargest sources of nonmanufactured items inthe merchandise trade account–are sub-tracted from merchandise imports and ex-ports, the picture that emerges is one of deepening U.S. trade deficits since the early1970s. Significantly, the few years of surplus — 1974, 1975, 1980 and 1981 –wereassociated either with serious recessions

(which generally jampen demand for im-ports) or with an exceptionally low dollar.

The trends in U.S. share of world marketstell much the same story. Americanmanufacturers have been losing their shareof both domestic and foreign markets forsome time. Between 1970 and 1980, the U.S.share of world imports rose slightly, from12.1 to 12.5 percent, but its share of world ex-ports dropped from 13.6 percent to 10.9 per-

cent (figure 12).25

Between 1980 and 1986,American exporters’ sales of manufactureditems fell 15 percent, while countries outsidethe United States were increasing their im-ports from all sources by 22 percent (in

Another calculation showsvolume terms).26

a general drop in the world market share of U.S. manufactures after 1975 (interruptedonly by a brief rise at the end of the decadewhen the dollar fell), and then a steep

24 Krugman and Baldwin, op. cit., pp. 2-S. ‘Ilese authors present evidence suggesting that the real value of the dollar that would bring ourtrade into balance has declined over the long term.m United Nations, 1980 Yearbook of International Tmde Statistics, Volume I: Trade by Count T, Department of International Economic

and Social Affaim Statistical Office, (New York: United Nations, 1981).‘m Rimmer De Vries and Derek I Iargreaves, “The Dollar’s Decline and Trade: Mission Accomplished?” Challenge, January-Feb~ary

1987, p. 39.

8/14/2019 8829



Figure 12.U.S. Share of World Import and Export Markets

1955-85

18%

17% ’ ,

\ 16%

15% - - - . . . .. . . -. .

14% - .,

13% -

12% -

11% -

10% , , , , , , ( , , I , I , 1

1955 1961 1967 1973 1079 1965

—I m p o r t s

--- Exports

SOURCE: United Nations, International Trade Statistics Yearbook (United Nations: New York), Volume 1, table A, various years1962-84.

decline of 8 percentage points from 1980 to1985.27

Even U.S. exports of high-technologyproducts–from the very sectors in whichAmerican firms are supposed to shine –have lost market share. Often high-technol-ogy sectors, only two – office, computing andaccounting machines, and agriculturalchemicals – gained in share of world exportsbetween 1965 and 1980. Seven high-technol-

ogy industries (engines and turbines, profes-sional and scientific instruments, electricalequipment and components, optical andmedical instruments, drugs and medicines,plastic and synthetic materials, and in-dustries chemicals) lost shares of world ex-ports, and one (aircraft and parts) remainedabout the same. These losses of marketshare occurred before the rise of the dollar inthe 1980s. Since the dollar’s fall after 1985,America’s high technology trade picture has

improved somewhat. Following a deficit in

1986, high technology goods trade showed asmall surplus of $600 million in 1987.

Another indicator of a decline in competi-tiveness is the remarkably slow response of U.S. imports and exports to the dollar’s fall.From its peak in the first quarter of 1985, thedollar has fallen back to the lows of the late1970s (figure 13). But the trade deficit has

Figure 13.Index of Effective Exchange Rates for

155

150

145

140

135

130

125

120

115

110

105

100

the U.S. Dollar, 1976-861980 = 100Exchange rate index

95’1970 1972 1974 1976 1978 1960 1982 1984 1986

27 Paul R Krugrnan and George N. Hatsor

UIOS, “The Problem of U.S. Competitiveness in Manufacturing,” New l~ngland FxonomicRc\lew, Jan./Feb. 1987, p. 20. Krugman and atsopoulos have adjusted U.S. world market share data to eliminate two extraneous factors,First, the adjustment screens out the effects different economic performance of different regions or countries. For example, an economicslump In F.urope might curtail European imports, thus reducing the American manufacturers’ share of world markets without reflecting afundamental improvement in competitiveness. Second, the adjustment includes the U.S. market is included in the world market,

28 Global Competition: The New Reality, Report of the President’s Commission on Industrial Competitiveness (Washington, DC : U.S.(~ovemrnent Printing Office, Janua~ 198S), p. 6.

8/14/2019 8829



only just begun to fall substantially. In 1987,over 2 years after the dollar began to fall, themerchandise trade deficit set anew record of $159 billion.

That trade deficits should continue to risefor a time after a drop in the dollar’s value isnot unexpected. Since firms buying fromoverseas suppliers tend to make extendedcommitments, U.S. importers would nor-mally continue to buy from offshore sup-pliers even after the dollar’s adjustment. Atthe same time, importers must pay more forforeign-made goods when the dollar is fall-ing, thus making imports more expensive.The case of exports is parallel. Even after the

dollar’s fall, U.S. firms wishing to sell off-shore have to make special efforts to over-come buyer-supplier relationships builtduring the time when U.S. product priceswere higher, and such efforts take time. Thisaccounts for the usual and expected lag–known as a J-curve –between the adjust-ment of currency value and and a turnaroundin the trade deficit.

However, a lag of 3 years since the dollar

peaked before seeing any really significantturnaround is unusual and surprising. Themerchandise trade deficit abated somewhatin early 1988, but the deficit was still runningat an annual rate of well over $100 billion. Incontrast, the response to the dollar’s rise — arise in imports of manufactured goods and adrop in exports – was much swifter than theopposite adjustment when the dollar fell.This fact, in combination with others, sug-gests that U.S.-made goods are less attractive

than foreign-made goods, the price effects of currency adjustment aside. In some cases,the attractiveness of foreign products

reflects very low labor costs or governmentsubsidy; in other cases it arises from highquality and reliability.

The trade picture outlined above is certain-

ly not what one would expect of a nationwhose manufacturing industries are holdingtheir own in international competition.While trade and market share figures do notindisputably prove the case for loss of com-petitiveness, they are signs of trouble – espe-cially since manufacturing trade slipped intodeficit in the 1970s, with surpluses appearingthereafter only when the dollar’s valuedropped, or in recession years. And behindthe aggregate trade figures are the experien-

ces of individual industries: Americanmanufacturers of consumer electronics,steel, automobiles, and semiconductors suc-cessively lost out to competitors who offeredbetter quality goods or lower prices, andthese losses began well before the damagingrise of the dollar.

Other indicators as well point to loss of competitiveness in manufacturing. There isevidence that the share of the manufacturing

sector in the U.S. economy has declined,while consumption of manufactured goods,as a share of total spending, is greater thanever – the difference, of course, being madeup by imports. Productivity growth of American manufacturing has lagged, espe-cially behind Japan’s. In addition, there aresigns that American leadership in technol-ogy-the foundation for high productivityand excellence in manufacturing – is erod-ing. Further discussion of these trends and

indicators appears in the following sections.

m It is important to note, however, that while exports have risen since the drop of the dollar, imports have not fallen.

8/14/2019 8829



To put the different indicators intoperspective, it is useful to define competi-tiveness. For a firm, competitiveness is theability to design, develop, manufacture, andmarket products at home and in other na-

tions, in competition with other firms.30

Fora nation, it means doing all this without adecline in the real standards of living of itscitizens.

31This means, for an advanced na-

tion like the United States, exploiting tech-nology, in its broadest sense, to provide therising productivity and superior productquality that make goods from high-wage na-

&tions attractive and affordable. Even withthe dollar’s fall, most nations have lowerwages than the United States. Moreover, the

range of products low-wage nations make israpidly expanding. It is very risky — in fact,probably infeasible – to limit our productionto only the most knowledge-intensive goodsand services and jettison traditional sectorswhere low-wage nations have productioncost advantages. We must compete effec-tively in many product lines with low-wagenations, and with producers from developednations who have excellent records inproduct design and performance.

Not all the signs are negative. SomeAmerican industries perform much betterthan others; the United States is by no meansat the bottom of the list among nations incompetitive performance and some of thesigns (e.g., growth in manufacturing produc-

tivity) have recently improved. Nor is itnecessary for the United States to outstripeveryone else. Economic growth and risingliving standards in other countries are in-evitable and desirable. However, a relative

decline in U.S. performance is a matter of concern, for that is the road to second-classeconomic status.

We have to recognize that it is difficult fora high-wage, highly productive nation likethe United States to make the cost-saving,productivity-enhancing, quality-improvingadjustments necessary to stay at the cuttingedge. Despite the difficulties, it is necessary,in view of the efforts many developed and

developing nations are making to catchup intechnology and penetrate the Americanmarket – the richest, largest, and one of themost open in the world. Catching up is noteasy, but is often a more straightforward andmanageable proposition than staying ahead.Moreover, development aid and parts of theinternational trade regime (e.g., theGeneralized System of Preferences, allow-ing special exemptions from tariffs todeveloping countries) are intended to help

the process along. Policies of individualcountries also have an important effect.Many nations –developed, less developed,and newly industrializing – have trade andindustrial policies aimed at promoting ex-ports while keeping their home markets rela-tively protected.

33

w U.S. Congress, Office of Technology Assessment, International Competitiveness in Electronics, OTA-ISC-XXl (Washington, IX: U.S.Government Printing Office, November 1983), p. 4.

31 Global ~mmpetition, op. cit. p. 13.w Technology is here used to mean not only hardware and machlneq’, but also the software, human sk]lls, and managerial know-how to

put together all the elements of production effectively.

33 The conduct and performance of policies aimed at industrial development and competitiveness in several nations, including Japan andnewly industrializing Asian countries, will be explored in the full assessment of Technology, 1nno\ation, and U.S. Trade.

8/14/2019 8829


U.S. Leadership in Technology

If the United States can maintain a com-petitive advantage, it is likely to be built on

The reason is simply ‘hat ‘hetechnology.34

United States has substantial competitivedisadvantages relative to most other nationsin some areas – for example, wage rates andcapital costs. U.S. wages are among thehighest in the world, and during the first half of the 1980s probably were the highest. Thefalling dollar has lowered American wagesvis-a-vis those of a few other developed na-tions – in particular, West Germany andJapan – but, in general, American wages arestill high compared with those of most of ourtrading partners. As for capital costs, U.S. in-terest rates were substantially higher in the1980s than those in much of the rest of theworld.

Technology has been a traditional source of U.S. strength, compensating for these disad-vantages. Our technological advantage in thepast rested on the invention of new products(e.g., Nylon, photocopy machines, integratedcircuits), ‘swift adoption and efficientmanufacture of products invented elsewhere(e.g., electric generators, stainless steel, jetengines), and improvements in the manufac-turing process. The last includes not onlydesigning and using better equipment butalso organizing work and managing peopleso as to make efficient use of the equipment.

The commonly used measures of tech-nological advantage or progress are not verysatisfactory. Most are indirect; for example,

many are measures of inputs, such as spend-ing on research and development, or they arerough proxies for outputs of R&D, such aspatent grants. In general, they do not tell usmuch about how well technology is beingused in the production of goods. It is impres-sive, however, that most of the conventionaltechnology indicators point in the samedirection, and so do case studies thatmeasure more directly the practical use of technology in manufacturing. In relation toother countries and to our own history, theUnited States is losing ground.

The dominating technological lead theUnited States enjoyed in the 1950s and 1960swas bound to narrow or disappear in manyfields, since our advantage was in part theresult of wartime destruction of Europeanand Japanese industry. There are indica-tions, however, that America’s relativedecline is not just the natural effect of growthin other countries but also reveals a fun-damental weakening in our ability to usetechnology to make things cheaply and well.

Japan and Germany are ahead of theUnited States in the kind of R&D spendingmost likely to pay off commercially. Spend-ing by American companies and governmentagencies for non-defense R&D rose quitesteadily (in constant dollars) in the 1970s and1980s, and in absolute terms the UnitedStates leads the world. But that lead simplyreflects the size of the U.S. economy. Incivilian R&D as a percentage of gross

w In a few industries, competitive advantage may also be built on unique endowments of natural resources. For example, the Americanpaper and lumber industries have substantial advantages over most other nations because of their access to a large, high quality softwoodresource.

26

8/14/2019 8829


U.S. Leadership in Technology q 27

domestic product, we are trailing Japan andGermany by increasing margins (figure14).

35

0ur civilian R&D spending was 1.9percent of GDP in 1985, compared to 2.8

Figure 14.Non-Defense ResearcDevelopment, Percent

1971-86Percent of GNP

3-

‘ - 12 . 8

2 . 6

2 . 4

2 . 2

2

1 . 8

1 . 6

1 4

1.2

1

1971 1973 1975 1977 1979 1961 1963 1965

U S Japan -- W. Germany

NOTE: Latest data for West Germany are NSF estimates based

on preliminary national data.

SOURCE: National Science Foundation, International Scienceand Technology Update 1987 (NSF 87-319)(Washington, DC: 1987). p.7.

percent in Japan and 2.5 percent in Ger-many. If defense R&D is included, total U.S.spending for R&D is about equal to Japan’s

and Germany’s, as a percentage of GNP.However, the commercial payoff fromdefense R&D is uncertain; although it hassometimes been seminal for commercial ap-plications, such spinoffs tend to be long-range and indirect.

Japan has spurted still farther ahead inprivate business spending for R&D. In theearly 1970s, the United States, Germany,

and Japan were about on a par in business-funded R&D, as a percentage of grossdomestic product (table 2). Today, Japanesecompanies are far ahead of their Americancounterparts, an indication of the serious-

ness of their commitment to technologicaleminence. German companies are also rais-ing their rates of R&D spending faster thanU.S. businesses, though not at the pace of theJapanese. Money spent on research anddevelopment is of course an imperfectmeasure of effective efforts toward tech-nological progress; the money spent may ormay not pay off in the marketplace. Even so,the fact that the Japanese and German leadsare widening is reason for concern about

America’s future technological prowess.

In human resources devoted to R&D–another input measure — the United States isahead, but the gap with other countries,especially Japan, is narrowing. In 1984, the

Table 2.-Business-Funded R&D As a Percentageof Gross Domestic Product

1972 1981 1983 1985 1986

United States 0.99% 1.22% 1 32% 1.3% 1.42%*Japan ., 1.15 1.73 1.99 2.09 2 14*Federal

Republicof Germany 1.08 146 1.56 1.64 1 69*

q Estlmatad

SOURCE: U S Congress, Office of Technology Assessment, International Com-petition Services , OTA-ITE-328 (Washington, DC: U S Governmen tPrinting Office, July 1987), p 19

35 .Some analysts argue that the total amount of R&D spending in a nation is more significant than the amount of spending relative toGNP. IIowever, spending as a share of GDP takes into account the size of the nation’s economy and indicates how R&E) ranks in importanceIn the nation’s total expenditures.

36 U.S. Clmgress, Office of Technology Assessment, “R&E) in the United States and in Other OECD Guntries,” staff paper prepared forthe Subcommlt tee on Ekonomic Stabilization, I louse Committee on IIanking, Finance and Urban Affairs, November 1’383.

8/14/2019 8829


28 qPaying the Bill

number of scientists and engineers engagedin R&D, as a percent of the labor force, wasstill higher in the United States than in othermarket-oriented countries (figure 15), butJapan had almost closed on the U.S. levels

’

(There is no international information onthe proportion of researchers working in thecivilian versus the defense sector, but theJapanese defense sector is relatively small;most resources devoted to R&D are on thecivilian side.)

Figure 15.

Force PopulationPer 10,000 labor force population

60 -

4 0 - - -- - - - - - -. --

- - - -- -- - - -- - - -

20 -

1965 1973 1 1977 1979

—U s . Japan

--- W. Germany

SOURCE: National Science Board, Science and Engineering in-

dicators -1987, (Washington, DC: National ScienceFoundation, 1987) p. 227, Appendix table 3-17

Other measures also document theJapanese challenge. For example, in 1983,Japanese universities graduated 69,600bachelor-level engineers, while only slightlymore – 73,000 engineers – receivedbachelor degrees in the United States.Japan’s labor force is barely more than half

the size of ours.38

University education of engineers in Japan may not be the equal of that in the United States; most Japanese en-gineers extensive additional training onthe job.

39Nevertheless, Japanese industry

has nearly twice the engineering graduates,per capita, to choose from and train if neces-sary. Moreover, in the United States,defense industries siphon off about 20 per-cent of the Nation’s engineers. Engineeringtalent, as opposed to scientific, is indispen-sable for applying research to the develop-ment of new products and manufacturingprocesses.

In terms of our own past history, the num-

ber of engineers and scientists graduatingfrom American universities is rising; in par-ticular, more engineers than ever are receiv-ing bachelor’s degrees (figure 16). Doctoraldegrees in engineering dropped off sharply,however, in the 1970s and despite a recoveryhad not regained the 1972 peak by 1985(figure 17). The recovery depended almostentirely on an infusion of foreign students. In1985, 57 percent of engineers getting doc-toral degrees were foreigners.

40

these foreign engineers remain in the UnitedStates, at least for a time, contributing espe-cially to university faculties and to non-defense technology, since most defense work is done by U.S. citizens. But eventually a sub-stantial number return home. ManyAmerican engineers see no need for a doc-toral degree, since they can get a good jobwith a bachelor’s or master’s degree. But thesharp dropoff in doctoral degrees awarded to

37 The Soviet Union claims a hi her share of scientists and engineers in the labor force than any other major count~. The Soviet Union’s$uneven record in technological ~ ormance (e.g., high in space exploration, low in production of consumer goods) reflects factom other than

human resources devoted to saence and technology.

m National Science Foundation, International Science and Technology Data Update 1986, NSF-307, p.28. In 1982, more engineem receivedbachelor level degrees in Japan than in the United States (74,000 vs. 67,000).

w See U.S. Con ss, Office of Technology Assessment, International Competitiveness in Electronics, OTA-ISC-200 (Washington, DC:rU.S. Government nnting Office, November 1983), pp. 314-17.

40 National Science Foundation, Foreign Citizens in U.S. Science and Engineering: History, Status, and Outlook NSF 86-305 Revised(Washington, DC, 1987).

8/14/2019 8829



Figure 16.U. S. Science and Engineering Bachelor Degrees Granted,

Percent of Total Degrees Granted

S/E bachelor degrees as percent of total14%

12%

.8% ‘

<

‘ . . . . .

6% -.‘ . . .

. .-.

4%. . .

.

1960 1962 1964 1966 1968 1970 1972 1974 1976 1978 1980 1982 1984 1986

Sciences-- Engineering

SOURCE: Office of Technology Assessment contractor report, “Federal Funding of Science and Engineering Education: Effect onOutput of Scientists and Engineers, 1945-85,” Betty M. Vetter (Commission on Professionals in Science and Technology)and Henry Hertzfeld (Consultant), NTIS order #PB 88-177 928/AS.

Figure 17.U.S. Engineering Ph.D.s Granted to U.S. Citizens

and Foreign Citizens, 1960-86

Thousands of Engineering Ph. D.s granted4 , 0 0 0

3,000 -

2,000-

I1,000

. . -

. . -

19 1964 1968 1972 1976 1980 1984

To U.S. Citizens--- To Foreign Citizens

NOTE: The totals do not equal U.S. recipients plus foreign recipients because the citizenship of some students is not known.

SOURCE: National Science Foundation, Foreign Citizens in U.S. Science and Engineering: History, Status and Outlook @Washington,D. C.: National Science Foundation, 1986), table B-21.

8/14/2019 8829



U.S. citizens may signal a serious problem infinding well-qualified engineers for researchand teaching in universities — the seedbedfor future engineering progress.

In other ways as well, Americans are lag-ging in the human skills needed to use tech-nology to improve manufacturing. Ourpublic schools are turning out graduates whodo not measure up internationally. This isespecially true in mathematics; for example,in an algebra test given to thousands of 12thgrade students in 1982, American studentscame in 14th, just ahead of Thailand and be-hind Hungary. Hong Kong ranked first,slightly ahead of Japan. Maintenance and

repair jobs, which are vitally important tocomputerized automation in manufacturing,require technicians with mathematicalabilities. People who operate the com-puterized equipment need certain basicskills. They have to be able to read instruc-tions, grasp the concept of statistical qualitycontrol, communicate with fellow workers,and understand their own part in a complexmanufacturing process. However, it is noteasy to measure how the lack of these skill;

exerts a dragon American manufacturing.A strong argument can be made that failureof managerial skills has also been a serioushandicap in the past 10 or 15 years, as oneU.S. industry after another has lost competi-tive position. It is axiomatic, though, that awell-trained, well-educated work force is apositive force in maintaining technologicaladvantage.

One way of evaluating the results of a

nation’s R&D efforts is to count up, in somefashion, the innovations it contributes. A

well-known attempt at a cross-country com-parison of innovativeness was the studysponsored by the National Science Founda-tion in the mid-1970s. Experts from sixcountries (the United States, Great Britain,

West Germany, France, Japan, and Canada)selected and examined 500 technological in-novations that were introduced into the in-ternational marketplace from 1953 to1973.

42Included on the list were such things

as lasers, disc brakes for autos, fiber optics, anew antibiotic, and a camera with self-developing color film. The great majority of the innovations the group considered oc-curred in the United States (319 of the 500),but the share of U.S. innovations showed a

declining trend over the 21 years (figure 18).No new international study of this kind hasbeen done.

Another conventional indicator of R&Dresults is patent applications or grants. Thesedata support the story of former Americandominance and current decline, with theJapanese as principal challengers. U.S.patent data are especially telling. Patentsgranted to U.S. inventors peaked in 1971

(figure 19). By 1985, patents of foreign originaccounted for 46 percent of the total grantedin the United States, with Japan — once againthe leader among foreign nations – repre-senting 19 percent. This record is all themore impressive in light of the fact that for-eigners tend to patent only their moreproven and useful developments in theUnited States, since it is expensive and in-convenient to apply for patents in countriesother than one’s own.

41There is, however, a strong correlation between higher income and higher education, and low levels of education are strongly correlated

~~!h%%~p@hin@on, DC: U.S.GPO,lNjent rates. See U.S. Cmgress Office of Technology Asessment, Technology and the American Ikonornic Transition,

a Gellman Research Associates, Inc., Indicators of International Trends in Technological Innovation, report prepared for the Nationa]Science Foundation under contmct no. NSF-(X89, April 1976.

8/14/2019 8829


U.S. Leadership in Technolog y q 31

Figure 18.Trends in Technological Innovation

Number of innovations

150

100

50

01953-59 1960-66 1967-73

Period of market introduction

SOURCE: Research Associates Inc., Indicators of International Trends in Technological Innovation, report prepared for the NationalScience Foundation, 1976, table 3-1.

Figure 19.bU.S. Patent Grants y Nationality of Inventor

1960-86

Thousands of patents granted90

60 -

70 -

60 -

40

30 - ,- - - - -

20- .- ,

.- ._.10

o I I I I I f I I 1 1 1 1 I 1 1 1 1

1 9 6 0 1 9 6 4 1 9 6 6 1972 1976 1960 1964

— Total — US Invtrs ---- Non-US Invtrs

SOURCES: National Science Board, science Indicators -1965, (Washington, DC: National Science Foundation, 1986) p. 256; U..S.Department of Commerce, United States Patent Office, Office of Technology Assessment and Forecast, Technology As-sessment and Forecast Report, (Washington, DC:U.S. Government Printing Office, 1973).

8/14/2019 8829



Not only has the domestic share of U.S.patents declined; patents to U.S. nationalshave fallen sharply in absolute numberssince 1971. In a recent assessment, OTA con-sidered possible reasons for this decline,

considering that R&D spending has risensteadily .43 Was the R&D process ineffectivein getting results, or had U.S. firms decideddeliberately not to seek patent protection?The analysis found evidence that the firstpossibility is more likely. In a recent survey,100 U.S. firms reported that they sought topatent a greater percentage of developmentsin the period 1980-82 than in 1965-69.

44If

the propensity to patent is greater, andspending is higher, then it appears that

spending has become less effective.Moreover, the National Science Foundationreports that, in thousands of influential jour-nals throughout the world, research publica-tions by American authors in the fields of engineering and technology fell steadilyfrom 42 percent of the total in 1973 to 38 per-cent in 1982.

45

Patenting in OECD countries by residentsof other countries shows a brighter picture

for the United States (figure 20). Externalpatenting, as mentioned above, is a good in-dicator of the value companies place on theirnew technical developments since the ex-pense and bother of applying in a foreigncountry tends to weed out trivial innovations.In OECD countries, U.S. nationals are theundisputed leaders in external patenting;they even had something of a surge in 1983while Japanese applications dropped slight-ly. The Germans, despite recent declines, are

still a strong second. Whether the U.S. surge

in 1983 represented a one-time backlog or areal trend can only be proven when data forlater years become available. The Japaneserecord remains impressive. Starting withabout 3,000 applications in 1960, the

Japanese advanced to more than 55,000 in1983.

Figure 20.External Patent Application By

Nationality of ApplicantThousands

140 -

120 -

100 -

80 -

60 -, -

40 -. .

. .-- -- - -

. - -- - -20 - - ‘

- -- - - - -

r

1972 1975 1978 1981 1983

— U . S .-- Japan Germany

SOURCE: Organization for Economic Cooperation andDevelopment, Science and Technology Indicators II:R&D, Inventiveness and Competitiveness (Paris:OECD, 1986), tables 24 and 26.

The main failing of patents as a measure of technological advance is that most patents

are not commercialized; even externalpatents may or may not lead to commercialdevelopment. Productivity, another com-monly used indicator, does not have thisdefect, since technology must be put to usein industry before it can contribute to a risein productivity. Although productivity is butone factor in competitiveness, it is an impor-tant one. The U.S. record in improvingmanufacturing productivity is, all-in-all, nota bad one compared to Europe, especially in

recent years; in the 1980s, our productivity

43 U.S. Congress, Office of Technology Assessment (1987), op. cit., p. 200.u Id., citing E. Mansfield, “Studies of Tax Policy, Innovation, and Patents: A Final Report,” report to the National Science Foundation,

October 198S, p. 6.45 National Science Foundation, op. cit., p. 38.

8/14/2019 8829



growth rates have been as good or betterthan those of most of the big Europeancountries. But Japan continues to beat all theadvanced countries in productivity growth.That story is told below.

The core question, however, is whetherAmerican manufacturers are falling behindin the practical application of technology–using it to produce high quality goods at af-fordable cost. There are no aggregate datathat really answer this question. The best wayto approach it is to analyze firms and in-dustries, case by case, to see how much andhow well technology is contributing to U.S.competitiveness. OTA is doing that for the

full assessment of Technology, Innovation,and U.S. Trade, of which this report is an in-terim product. A number of such case studieshave already been done, by OTA and others.It is fair to say from the work already com-pleted that the reputation of U.S.-madegoods for quality and reliability has sufferedin recent years and that American manufac-turing methods are no longer the paradigmfor the world.

One of the best examples of such work isJaikumar’s study of flexible manufacturingsystems (FMS) in the United States andJapan.

% A flexible manufacturing system isa production unit which is designed tomanufacture different kinds of parts (for ex-ample, transmission cases or clutch housingsfor trucks and farm machinery) in relativelysmall batches. The FMS is made up of semi-independent work stations (such as numeri-cally controlled machining centers),

connected by automated material handlingsystems (conveyor belts, robots) and control-

led by computer. Jaikumar compared howJapanese and American firms used FMSs,and concluded that American firms had usedthe technology far less effectively than theJapanese. The American systems produced

many fewer kinds of parts, took longer todevelop, and performed less reliably. For ex-ample, U.S. firms typically took 2.5 to 3 yearsto develop FMSs, compared with 1.25 to 1.75years in Japan; produced only 10 differentkinds of parts compared with the Japaneseaverage of 93; and produced an average of 88units per day compared with 120 in Japan. InJaikumar’s words, “[r]ather than narrowingthe competitive gap with Japan, the technol-ogy of automation is widening it further.”

47

Jaikumar attributed the relatively poorperformance of FMS in the United States tomanagement, not to differences in machinequality or performance, or in the complexityor size of parts produced. Americanmanagers tended to prevent workers frommaking changes to the system once it wasoperating, treating the flexible automatedtechnology in much the same way that dedi-cated, hard-wired automated equipment is

used for mass production, and losing both ef-ficiency and flexibility in the process. “If itain’t broke, don’t fix it,” was the attitudecommon among American managers.

48

Having spent much more time than theJapanese getting their FMSs up and running,American managers tried to nail down astandard operating procedure and stick to it.Japanese managers, on the other hand, werewilling to continue tinkering and changingand improving their FMS installations. This

constant emphasis on incremental redesignand improvement is in fact widely cited as a

@ Ramchandran Jaikumar, “Postindustrial Manufacturing,” Harvard Business Review, Nov-Dee, 198647 Ibid., p. 69.u Ibid., p. 71.

8/14/2019 8829


8/14/2019 8829


from spending more: Japanese automakersuse only about half as many engineeringhours to complete a comparable project(“clean sheet” design of a new automobileand i ts production) as American

Clark and his colleagues con-automakers.54

eluded that the Japanese automakers’ designprocesses are more efficient because theygive a single “heavy manager” authority overthe whole project; the people doing re-search, development and design are in con-stant communication with the peopleresponsible for manufacture; conflicts areaired and settled early; product and processdesign are treated as simultaneous ratherthan sequential activities.

There are other Japanese strengths.Among those most often cited are greater at-tention to product quality and reliability,consensus building, and emphasis on long-term market share rather than short termprofit. All are difficult to quantify, butfirsthand observations, case studies, and theremarkable record of Japanese in-dustrialization and adaptation in the postwarperiod support the basic point: Japanese

manufacturers have moved into a command-ing position in many industries and have sur-passed U.S. rivals in many importantmarkets by developing and applying tech-

5 5nology.

While the record of technology develop-ment and application is mixed in differentEuropean countries and industries, there arealso European examples of aggressive use of new technology to create a competitive ad-

vantage. One of the best known is textile in-dustry machinery. Nearly all new weavingmachines in American textile mills comefrom Europe (West Germany and Switzer-land) or Japan. Unlike American suppliers,European manufacturers have introduced anew generation of equipment every coupleof years. The new equipment is oftenprogrammable, can weave a variety of widths, and is faster and quieter than the bestAmerican weaving machines. Little wonder,

then, that import penetration in textilemachinery has increased from 7 percent of the U.S. market in 1960 to nearly 58 percentin 1986. Import penetration in weavingmachinery was nearly 85 percent.

56

The improvement in Japanese and otherforeign producers’ manufacturing efficiency,quality and performance has elicited a num-ber of responses from American firms. Someresponses have been helpful, and others

have not. Overall, however, the responsesmade by U.S. manufacturers have not stabi-lized or improved America’s position inworld manufacturing.

w Kim B. Clark, W. Bruce Chew, and Takahiro Fujimoto, “Product Development in the World Auto Industxy Strategy, Organization andPerformance,” paper presented to the Brookin~ Institution Macroeconomics tinference, December 3,1987.= We should not attribute too much of the Japanese record to this one set of factors, hcnvever. The Japanese home market is and has been

much Ie= peMous to imports, particular in sectors targeted for development, than the American market, despite such widely-cited exam Ies1 Aof growing American protectionism as t e Multifiber Arrangement and Volunta~ Restraint Agreements on Japanese auto imports. is

subject – how foreign governments use tmde and industrial @icies to promote industrial development and maria e competition fromAmerican and other developed -countxy products – %is taken up m the full assessment, Technology, Innovation and U.S. rade.w U.S. International Trade Commission, U.S. Global Competitiveness: The U.S. Textile Mill Industry, Report to the Committee on

Finance, U.S. Senate, USITC Publication 2048, December 1987,

8/14/2019 8829


U.S. Manufacturing Performance

This war–and it is a war–isbeing fought not with dollars, oroil or steel, or even with modernmachines. It is being fought withcreative imagination and or-ganizational talent.

This admiring, if slightly defiant, descrip-tion of a powerful foreign economic chal-lenger is not an American’s view of theJapanese competition in the late 1980s. It isa Frenchman’s view of America in 1969.

57

Twenty years later, this description of American industry as all-conquering has

come to sound quaintly out of date.American pre-eminence in a great manymanufacturing industries is gone. Take con-sumer electronics. Only one major U. S.-owned company is still making color TV sets,and most of its production takes place inMexico; no American-owned companymakes video cassette recorders or compactdisc players. Mass production of automobiles was invented in the UnitedStates, but others (especially the Japanese)

are now leaders in the technology andmanagement of auto manufacture. Despitethe U.S.-Japanese agreement restrictingJapanese imports and despite the rise of theyen, 21 percent of the passenger cars sold inthe United States in 1987 were Japanese-made (another 6 percent were made byJapanese companies in North America);another 9 percent were imports from other

58 In semiconductors,foreign countries.

another native born American product andindustry, U.S. companies are still strong,especially in microprocessors and advanced,

custom designed chips. Yet, overall, U.S.companies have continually lost marketshare to Japanese competitors since the late1970s. By 1987, they had almost cededdynamic random access memory devices(DRAMs) –a large market segment that hasbeen both cash cow and technology driverfor the industry-to the Japanese. In all of these industries, trouble started before therise of the dollar.

Against the evidence of a decline, somehave argued that U.S. manufacturing isfaring quite well, that productivity growthhas been strong in the 1980s, and that thehigh dollar – not poor performance bymanufacturing — is responsible for the mas-sive manufacturing trade deficits of thedecade. The prescription that usually followsfrom this argument is to do nothing in tradeor industrial policy to support U.S. manufac-turing. One part of the argument is the state-

ment that manufacturing output, measuredin constant dollars, has not declined as ashare of gross national product, and that if iteventually does, that alone is not an “omenof decay or loss of competitiveness.”

59In-

stead, it may simply reflect a natural evolu-tion to a different pattern of demand in amaturing economy, and to the successfuleconomic development of our tradingpartners.

57 Jean-Jacques .Seman-schreiber, The American Challenge (New York: Atheneum, 1969), p. xiii.5a The remaining 64 percent were made in the United States and Canada. Production in both countries is considered “traditional North

American’ ’because of the U.S.-Canadian agreement establishing free trade in motor vehicles and parts.w Molly McUsic, “U.S. Manufacturing :

$Any Cause for Alarm?” New En land Economic Review, JanuaV/Februaty 1987. For other

examples of this point of view, see Robert . . Lawrence, Can America Cornr

Fte. (Washington, DC: The Brookings Institution, 1984); Officeof the U.S. Trade Representative, Annual Report of the President of the nited States on the Trade Agreements Program, 1985, p. 20.

36

8/14/2019 8829


U.S. Manufacturing Performance q 37

This argument does not really stand up toscrutiny. The United States is not and givesno sign of becoming a post-industrialeconomy. But because the question of manufacturing share is closely linked withpolicy, it is worth examining.

The Share of Manufacturing in the

U.S. Economy

Though the record is not entirely clear,there is evidence that the share of manufac-turing in gross national product (GNP) isfalling. And while it is falling (or at best stay-ing even), the demand for manufactured

goods by American consumers, businesses,and government is rising.

In current dollars, the share of manufactur-ing in GNP fell from 29 percent in 1960 to

just under 20 percent in 1986, and the rate of decline has been faster since 1979 thanformerly (figure 21). However, this currentdollar measure has the defect that it does nottake rising productivity into account.Manufacturing has performed better than

the economy as whole in raising productivity,and some of that productivity growth hasbeen passed on to consumers in lower-than-average price increases. In fact, CommerceDepartment data on the constituents of GNPdata show the share of manufacturing, inconstant 1982 dollars, hovering quite steadi-ly around 21 or 22 percent of total outputsince the late 1940s (figure 22). This series,prepared by the Department’s Bureau of Economic Analysis (BEA), is the onlyregularly published official set of data onconstant dollar shares of GNP. It is the basis

for the statement that manufacturing’s shareof GNP has held steady for many years.

However, estimating the size of variousparts of the economy in constant dollars is adifficult task; the uncertainties are greatenough to cast doubt on the constant-shareconclusion. In the BEA series based on con-stant 1982 dollars, one difficulty in particularlooms large. That is the unique role assignedto the non-electrical machinery industry,which includes computers, in pulling up thewhole manufacturing sector.

According to the BEA series, 15 of the 21major manufacturing industries in theUnited States experienced a declining shareof GNP from 1979 to 1986, while five stayedeven or rose only moderately (see table 3) –not enough of a rise to offset the decline inthe majority of industries. The only major in-dustry showing a big increase in share wasnon-electrical machinery; and more than 100percent of that industry’s increase was due tothe zooming sales, rapidly improving qualityand productivity, and falling real price of computers. By the logic of the numbers, itwould appear that computers, which con-tribute only 2 or 3 percent of manufacturingoutput, singlehandedly held up the share of the whole manufacturing sector.

Another difficulty is that the choice of baseyear for constant dollars greatly influences60

The more recent the base yearthe results.chosen, the smaller appears the share of manufacturing in past years (see figure 22).For example, when 1958 is used as the baseyear, the share of manufacturing in real GNPfor the year 1948 appears as 29.7 percent; on

m Nicholas S. Pema, “The Shift from Manufacturing to !$ewices: A Concerned View”, New England Fxonomic Review, JanuaV/Feb~aY1987.

8/14/2019 8829


36 qPaying the Bill

Figure 21U.Manufacturing Share of .S. Gross

Percent of GNP(current dollars)

35

30

25

20

National Product

I I I I I 1 I 1 1 I I I 1 1 1 I 1 I 1 1 I I I I 1 1 I 1 1 I I I 1 I I I

1947 1951 1955 1959 1963 1967 1971 1975 1979 1963

SOURCE: U.S. Department of Commerce, Bureau of Economic Analysis, National Income and Product Accounts, electronic data,Table 6.1

Figure 22Manufacturing Share of Gross National Product

Percent of GNP(constant dollars)

35

30

25

20

‘ . . ’

15I I 1 1 I I I I 1 I I 1 I 1 I I I 1 1 I I I I I 1 I 1 I I 1 I I 1 I 1 I I I

1947 1951 1955 1959 1963 1967 1971 1975 1979 1983

— $ 1 9 5 6 ‘ --- $ 1 9 7 2 — $ 1 9 6 2

SOURCE: U.S. Department of Commerce, Bureau of Economic Analysis, National Income and Product Accounts, electronic data, table6.2

8/14/2019 8829



0

*u

0)

0

u)

co

8/14/2019 8829



a 1972 base, the share in 1948 appears as 24.8percent; with the 1982 base, the 1948 shareshrinks to 21.5 percent – just about the sameas the 1982 share, which was 21.8 percent.The difficulty with applying an updated con-

stant dollar base to earlier years is that thenew base contains new weights for the inputsto industries, and these new weights do notrepresent the economy as it really was in ear-lier years. Perna, in discussing this problem,said: “The further one gets from the baseperiod, the less representative it is of theeconomy’s actual structure.

In order to analyze the changing structureof the economy for its assessment Technol-

ogy and the American Economic Transition,OTA independently prepared estimates of various parts of the economy in constant1980 dollars for selected years.

63The OTA

estimates show manufacturing’s sharedeclining by 2.5 percentage points from 1972to 1984, with an accelerated decline after1977. The complications and uncertainties of constructing these constant dollar estimatesare great; OTA’s estimates have their shareof flaws. The point is that constant dollar es-

timates are not graven in stone, but must betaken with a degree of caution.

Suppose it is true that manufacturing isfading as a contributor to the economy as awhole. The next question is: does it matter?It is not ordained that the share of manufac-turing in GNP should remain constant. Infact, the current dollar figures show it declin-ing gradually in the 1950s and 1960s, whenAmerican manufactured goods were still

dominant in the world (however, the declinehastened in the 1970s and 1980s, asAmerican products lost world market share).Moreover, agriculture is often held up as anexample of a sector of the economy that grew

greatly in output and productivity whiledeclining from a 22 percent share of the na-tional economy at the turn of the century to2.2 percent in 1986.

A critical difference between manufactur-ing and agriculture is that the latter has con-tinued to fulfill domestic demand (moreprecisely, to produce enough that importsare fully covered by exports, and sometimesto generate sizable trade surpluses as well).

Over the years, Americans have devoted suc-cessively smaller shares of their total pur-chases to products of farms, forests, andfisheries, and more to other goods and ser-vices. The same is not true of manufacturedgoods. While per capita spending for ser-vices has grown greatly in the past 40 years(table 4), it was not at the expense of demandfor manufactured products. While con-sumers spent smaller shares of their growingincomes on food and fuel, they spent more

on items such as cars, television sets, andsports gear. Altogether, American con-sumers, businesses, and government in-creased their share of spending onmanufactured goods other than food andfuel items from 23.4 percent of all their pur-chases in 1948 to 30.7 percent in 1986. Clear-ly, the U.S. economy is not passing into apost-industrial state in which demand formanufactured goods is giving way to demandfor services.

61 The weights are used to construct price deflators, which are the basis for constant dollar estimates of GNP and its constituents.w Id. For another study that questions BEA’s methods for developing the constantdollar series, and concludes that manufacturing has

declined as a share of GNP, see Lawrence R Mishel, Manufacturing Numbers: How Inaccurate Statistics Conceal U.S. Industrial Decline(Washington, DC: Economic Policy Institute, 1988).m U.S. Congress, Office of Technology Assessment, Technology and the American Economic Transition: Choices for the Future

OTA-’IT7r-283 (Washington, DC: U.S. Government Printing Office, May 1988).

8/14/2019 8829


Table 4.– Real per Capita Spending on Goods and Services (1960-86)

Per capita spending in 1982 dollars Percent of real apparent consumption

1948 1960 1973 1979 1986 1948 1960 1 9 7 3 1979 1986

Apparent consumption $ 7 ,4 3 2 $ 9 , 2 3 3 $ 1 3 , 0 9 9 $ 1 4 , 1 6 6 $ 1 5 , 9 7 3 100.0% 100.0% 100 o% 100 o% 100.0%

Gross national product 7 ,5 63 9 ,2 1 1 1 2 ,9 5 0 1 4 , 18 2 15 ,3 70 1 0 1 8 998 989 1 0 0 1 9 6 . 2

G o o d s p u r c h a s e s 3,437 3,966 5,613 6,200 7,250 462 43.0 429 43.8 45.4Consumer manufactures, except food and fuel 1,038 1,253 2,076 2,320 2,873 140 136 159 164 18.0Producers’ durable equipment 525 461 942 1,150 1,298 71 5 0 7 2 81 81Government goods purchases 181 400 461 733 2 4 4 3 3.1 3.3 4.6Consumer food and fuel purchases 1,693 1,851 2,187 2,269 2,346 228 20.1 167 16.0 14.7

Service purchases 2,961 3,956 5,629 6,287 7,041 398 429 43.0 444 441Consumer services 1,920 2,455 3,710 4,315 4,925 258 266 28.3 305 30.8Government services 1,042 1,501 1,919 1,972 2,116 140 16.3 14.6 13.9 13.2

NOTE Apparent consum ption equals gross national product less exports plus imports

SOURCE: U S Department of Commerce, Bureau of Economic Analysis, National Income and Product Accounts, electronic data, consumer goods and services purchases from Table23, government goods and services

spending from Table 3 BB producers’ durable equipment spending from Table 57

8/14/2019 8829



Until about 1970, U.S. production of manufactured goods generally kept pacewith consumption, or stayed ahead; but thenoutput began to dip below consumption, asshown in the recurring manufacturing trade

deficits of the decade. In the 1980s, of course,manufacturing output fell far short of con-sumption, creating the mounting manufac-turing trade deficits. Foreign suppliers havefilled the ever-widening gap betweenproduction and consumption of manufac-tured goods in the United States in the 1980s.

Manufacturing Employment and

Wages

Other measures that may tell us somethingabout the performance of U.S. manufactur-ing are the number of people working in itand what they get paid. Jobs in manufactur-ing have declined in the past decade, not justin relative terms, but in absolute numbers.Real wages of production workers inmanufacturing (adjusted for inflation) havealso dropped, by about 6 percent over thepast 10 years. Real compensation per

manufacturing worker, including employer-provided benefits, has stayed almost flat – instriking contrast to Japan and majorEuropean countries, where manufacturingcompensation rose about 20 percent in thesame period.

64

The decline in manufacturing jobs has beenhard on millions of displaced workers andtheir families and scores of communities, butit does not necessarily signify weakness in the

manufacturing part of the economy. Sincethe nineteenth century, and throughout theperiod of American industrial dominance,the share of employment in services has beenlarger, and has grown faster, than in

manufacturing (figure 23). As the output of manufactured goods grew, employment roseless because of improving labor productivity.

While the share of employment in U.S.manufacturing started a gradual decline inthe 1950s, the absolute number of manufac-turing jobs kept growing until 1979, whenmanufacturing employment peaked at 21million. In 1986, it averaged 19.1 million.

With the strong expansion of exports andmanufacturing output toward the end of 1987, employment recovered to 19.4 mil-lion–still 1.6 million below the peak.

An absolute loss of manufacturing jobs isnot necessarily a sign of weakness either.Some of the shrinkage in employment wascertainly due to rising productivity. Somewas also certainly due to the enormousgrowth in net imports of manufactured

goods over the same period. And much of itwas due to a combination of the two factors,in which actions to improve productivity –automation, for example, or closure of older,less efficient plants – were forced by foreigncompetition. If demand for a product isgrowing fast enough, then imports, produc-tivity, and employment can rise simul-taneously. If not, rising net imports are likelyto cost jobs. For example, employment inthree traditional industries –steel, textiles,

tM U.S. Department of Labor, Bureau of Labor Statistics. The figure for wages is the real hourly wage for production workers, who makeup about two-thirds of manufacturing employment. The figure for compensation is real weekly compensation for all persons employed inmanufacturing, includin wage and salaV earners, the Self+ mployed, and unpaid family workers, in the United States, and for all employeesin other countries. The Eonsumer Price Index was used as the basis for calculating real wages and real compensation.

8/14/2019 8829



Figure 23Distribution of U.S. Employment, by Sector

1870-1986Percent of total U.S. employment

. .-60

t

..-

.’ I

40-

. ..

- - - - -. . --

. . -

01 I I I I I I I

1870 1885 1900 1915 1930 1945 1960 1975

—P r i m a r y

—M a n u f a c t u r i n g

-- - - S e r v i c e s

SOURCE:U.S. Department of Commerce, Bureau of the Census, Historical Statistics of the United States, (Washington, DC: 1975) p. 137,series D127-141

and motor vehicles –dropped by 600,000from 1979 to 1986. In each of these in-dustries, productivity improved and importsrose; at the same time, demand for theindustries’ products either declined or grewslowly.

One study, following the ripple effects of imports and exports through the economy bymeans of an input-output model, concludedthat the United States gains 7.5 percent more

jobs from a given amount of exports than itloses from the same amount of imports.

65

Every $10 billion of exports generates193,000 jobs, the study found, while 179,000

jobs are lost with $10 billion of imports.However, the trade deficits have been so big

in recent years that job losses due to importshave swamped the job-creating effect of ex-

ports. In 1987, for example, exports of goodsand services amounted to $428 billion whileimports were $547 billion. The deficit of $119 billion spelled a net loss of 1.5 million

jobs, according to the analysis.

It has been suggested that the loss of manufacturing jobs in recent years maybe atleast partly illusory, because it simply reflectsthe trend in many manufacturing companiesto contract out services that they formerlypaid their own employees to perform. Forexample, if General Motors lays off en-gineers and contracts with an engineeringdesign firm to do the work once done in-house, that shows up in national employmentdata as a loss of jobs in manufacturing and a

gain in the engineering and architectural ser-vices category. In the same way, if firms un-

m Richard S. Belous and Andrew W. Wyckoff, “Trade Has Job Winners, Too,” Across the Board, September 1987. The authors used theOTA input-output model for this study.

8/14/2019 8829



bundle legal, accounting, auditing, janitorial,or clerical activities, then the employmentfigures would show a shift from manufactur-ing to services. A recent analysis, done by aU.S. Department of Labor economist, con-

cludes that unbundling has been a very smallfactor in the growth of employment inproducer services in the last decade.

66

Within manufacturing firms, the proportionof workers in managerial, professional, andtechnical occupations has actually risen (andthe rise is not accounted for by a changingmix of manufacturing industries). While theproportion of clerical and service workers inmanufacturing has dropped slightly, theseoccupations are not very significant in the

growth of employment in producer services.Thus, unbundling is not happening in-dustrywide, though it may well be happeningin some individual firms. Firms may bebuying more services from outside, but notat the expense of already existing jobs in themanufacturing sector.

It is fair to conclude that the job losses inmanufacturing are real, not illusory. Andthough it maybe hard to calculate the exact

number of jobs lost to import competition,the number is probably large –above 1 mil-lion at the least.

It also seems evident that import competi-tion has been a powerful factor holding downthe wages of manufacturing workers. Untilthe 1970s, wages of manufacturing workers,

like wages of American workers generally,rose strongly and steadily. Since then,manufacturing workers have made few last-ing gains, and the real wages of productionworkers (i.e., blue collar workers on the shop

floor) had not regained their 1978 peak adecade later. While manufacturing workersin the other advanced industrial nations en-

joyed strong growth in real compensation(wages plus benefits) from 1977 to 1986–growth that ranged from 14 percent in Italy,to 19 percent in Japan and Germany, to asmuch as 24 percent in Britain — Americansemployed in manufacturing gained less than2 percent.

What happened to manufacturing wageshas happened to real wages and salaries of all Americans: the long-term, consistentgrowth of the postwar period came to a haltin 1973, and there has been an unsteady butoverall decline since 1978.

68Part of this

change may have been due to demographics;the surge of young people from the babyboom and the increased participation of women in the labor force probably helddown wage growth in the 1970s. However,

the rate of growth in the work force has beenfalling since 1978, and is now back to earliernorms. Since 1978, a combination of factorshas restrained real wage growth: first, infla-tion, and then the deep recession of 1981-83,the decline of labor unions and, not least, theloss of manufacturing jobs to foreign com-petition and the threat of further losses.

a6 John Tschetter, “Producer Sewices Industries: Why Are They Growing So Rapidly?” Monthly Labor Review, December 1987.67 Some analysts have argued that the loss of U.S. manufacturing jobs since 1979 simply matches improvement in manufacturing

productivity, and draw the conclusion that imports had no effect on job loss. However, rising imr

rts of manufactured goods during the 1980salmost certainly replaced some domestic production of these goods –and the jobs that wou d have been devoted to producing them. Inaddition, as discussed in a later section, the official figures may overstate the growth in manufacturing productivity in the 1980s.

aa Weekly earnings of full-time wage and sala~ eamem declined 3 ~rcent from their 1978K

ak to 1987; hourly earnings of productionand nonsupetisory workers on private nonagricultural payrolls declined 8 percent from 19 to 1987. Earnings figures are from U.S.Department of Labor, Bureau of Labor Statistics, Empl

T

entand Eamin ,tables A-73(published quarterly) and B-1 (published monthly).xReal earnings are figured on the basis of the Consumer rice Index for U an Consumers (CPI-U), 1982= 100.

8/14/2019 8829



Productivity Growth: International

Comparisons

Another measure of how U.S. manufactur-

ing is doing in comparison with other nationsis trends in productivity. The bare figuressuggest that, over the past quarter of a cen-tury, the United States did not measure upto other nations in raising labor productivity

69 The U.S.. growth ‘atein manufacturing.was less than 3 percent per year, on average,from 1960 to 1986; this compares with near-ly 8 percent for Japan, about 5 percent forFrance, Italy, and Germany, and over 3 per-cent for Great Britain and Canada (table 5).

Behind these 26-year averages lies a morecomplex story. Since 1979, the Americanrecord has been about as good as Europe’s —better than some major countries and not farbehind the leaders. But, as noted earlier,America’s number one trade competitor,Japan, has continued to excel, achievinghigher growth than any other industrialized

country in the 1980s, with an average of 5.6percent growth per year from 1979 to 1986compared to 3.5 percent for the UnitedStates. Another distinction for Japan is thatits productivity growth, more than that of any

other advanced country, continues to belinked with rising output and employment.

Faster productivity growth in other in-dustrialized countries was in part a catchupphenomenon. From 1960 to 1973, U.S.manufacturing productivity rose at the re-spectable rate of 3.2 percent per year; butthis rate was bettered by nearly all Europeancountries, most of which were repairing wardamage and investing in new industrialequipment. Japan, starting from a lowerprewar base and suffering more war destruc-tion than most European nations, was ad-vancing even faster, at the remarkableaverage annual rate of 10.3 percent.

70

From 1973 to 1979, productivity growthslowed to some degree in all the industrial-ized countries but (except for Britain’s dis-mal record) the U.S. growth rate dropped to

Table 5.–Annual Percent Changes in Manufacturing Productivity, Seven Countries (1960-86)

United UnitedYear states Canada Japan France Germany Italy Kingdom

Output per hour:1960 86 ., ., . . . . . 2,8 3.3 7.9 5.2 4.6 5 7 3 6

1960 73. . 3.2 4.5 10.3 6.5 5 8 7.5 4 21973 79 . . . 1.4 2.1 5.5 4,9 4,3 3 3 1,21979 -86..... . . . : : ., ., 3.5 2.3 5.6 3.1 2.7 4.3 4.5

NOTE: Rates of change based on the compound rate method

SOURCE: Arthur Neef and James Thomas, "Productivity in Manufacturing at Home and Abroad,. Monthly Labor Review, Decemer 1987 U S Department of Labor , Bureau of

Labor Statistics, Monthly Labor Review, December 1987, Table 47

m Productivity of other factors of production besides labor, especially capital, is also ve~ important to good manufacturing performance.I Iowever, international comparisons are usually limited to labor productivity since data on multifactor productivity are fragmentary.

70 A number of economic historians have discussed the political and social conditions that made it possible for many nations –especiallyGermany and Japan – to rebuild rapidly and catch u afterword War II; see, forexam Ie, Moses Abramovitz, “Catching U , Forgin Ahead,

1’and Falling Behind,” Journal of Fxonomic IIisto~, une 1986, XLV1 (2), p~. 38 S-406; d p “ ~ð•€)E!20nomicncur Olson, ‘Ile Rise and Fall of Nat Ions.Growth, Stagflation and Social Rigidities (New Haven, CT : Yale University Press, 1982).

8/14/2019 8829


8/14/2019 8829



Plants were closed, workers were laid off,and unemployment soared to 20 percent andabove in the industrial North. Through 1982,Britain’s manufacturing output declinedsharply; it has since turned backup, showing

a moderate overall loss for the 7 years of 4.4percent. Manufacturing employment fellsteadily with no recovery, for a loss of 27.5percent. France, with a productivity growthrate of 3.1 percent per year, had a 2 percentcut in manufacturing output and lost 16 per-cent of manufacturing employment. Italy’slarge rise in productivity went along with asharp drop in manufacturing employment.Germany and the United States were in themiddle, with medium to good productivity

growth, rising output, and moderatelydeclining employment.

While jobs in U.S. manufacturing droppedby 10 percent from 1979 to 1986, real outputrose over 16 percent (according to the BEAconstant dollar series). At least some of the

Table 6.– Index of Manufacturing Output andEmployment, 1986; and Productivity

Growth Rates, 1979-66

Annual averagemanufacturing

1986 productivityoutput Employment growth(1979 = 100)* 1979-86

United States 1165 906 3 5Canada 114.2 97.1 2.3West Germany 105.8 92.1 3.1France 978 84.4 2,7Italy 1122 825 4.3J a p a n 159,9 109.5 5 6United Kingdom 95.7 72.5 4 5

q Adapted from Labor Department data published on a 1977 = 100 basis

SOURCE: Arthur Neef and James Thomas, “Productivityy in Manufacturing atHome and Abroad,

-

Monthly Labor Review, December 1987, U S De-

partment of Labor, Bureau of Labor Statistics, Monthly Labor Review,

December 1987, Table 47

turnaround in American manufacturingproductivity was due to shutdown of older,less efficient plants. With this restructuringcame some massive worker displacement; anaverage of 2 million workers per year, half of

them in manufacturing, lost jobs due to plantclosures or production cutbacks from 1979to 1985.

72Steel is an extreme example. Jobs

in basic steel numbered 570,000 in 1979 andby the end of 1987 were down to 280,000. TheUSX company, formerly U.S. Steel, con-tracted from over 100,000 employees in 1980to fewer than 20,000 in 1987. But meanwhile,USX productivity improved from 10.8 man-hours per ton of steel shipped in 1983 to 3.8manhours in 1987.

73

In the last half of 1987, as exports finallybegan to rise briskly in response to the lowdollar, manufacturing employment climbeda little (but still remained 8 percent belowthe 1979 peak) while output grew to 30 per-cent above the 1979 level. During the expan-sion, productivity growth held up; the growthrate was 3.7 percent in 1986 and 3.3 percentin 1987. To some degree, this simplyreflected greater use of plant capacity, which

generally has the effect of raising produc-tivity. But there are some signs that it alsoreflects more fundamental changes — invest-ment in productive new equipment, more ef-ficient organization of work, and better useof people.

A measure for comparing levels of laborproductivity from one country to another (asdistinguished from growth rates) is gross

n These figures are from two surveys of worker displacement, designed and analyzed by the Bureau of Labor Statistics, U.S. Departmentof Labor, and conducted by the Bureau of the Census, U.S. Department of Commerce, in JanuaT 19M and January 1986.

73 David Ignatius, “What’s Left of Big Steel?” The Washington Post, Mar. 20, 1988, p. Cl.

8/14/2019 8829


48 qPaying the Bill

domestic product per employee. Thiseconomy-wide measure includes private ser-vices and government activities as well asmanufacturing. By this measure, the UnitedStates was still ahead of most other advanced

nations in 1986. Several European countriesstood at 80 to 90 percent of the U.S. leveland Japan had reached 69 percent.

74

However, the rate of productivity growth inthe whole U.S. economy has recovered onlyslightly from the doldrums of the 1970s.Other major industrialized nations are nowimproving at much faster rates, Japan thefastest of all (see table 7).

It is a common observation that although

agriculture, many kinds of services, andsome manufactures are not highly produc-tive in Japan, the Japanese have putprodigious effort into raising productivity inindustries such as steel, autos, and

electronics that have been central to their ex-port-led growth strategy. It would be helpful,in comparing productivity levels in theUnited States with Japan, to break outmanufacturing, by industry, from the rest of

the economy. However, various internation-al comparisons of levels of manufacturingproductivity have come to quite inconsistentconclusions; in some, Japanese manufactur-ing productivity is shown as barely 70 percentof the U.S. level, while in others it is over 90percent for all manufacturing and well above100 percent for certain industries.

75

Several case studies of individual industrieshave found that Japan has not only caught up

with the United States in productivity, buthas forged ahead. For example, the Interna-tional Motor Vehicle Program found that inthe mid-1980s it took, on average, 19.1 hoursto build a car in Japanese assembly plants

Table 7.-Average Annual Changes in Real Gross Domestic Product per Employed Person, 1960-86

United UnitedYear states Canada Japan France Germany Italy Kingdom

1960-66 . . . . . . 1.2% 1.9% 5.5% 3.6% 3.1% 37% 2.2%1960-73 . . . . . . . . . . . . . 1.9 2.6 8.2 4.9 4.1 5.8 2.9

1973-79 . . . . . . . . . . . . . . 0. 0 1,3 2 9 2.7 2 9 1.7 1.31979-86 . . . . . . . . . . . . . . 08 1,0 2 8 1,9 1.6 1.6 1.7

SOURCE: U S Department of Labor, Bureau of Labor Statistics, Office of Productivity and Technology, unpublished data, mimeo, August 1887

74 These cross country comparisons are based on purchasing power parity (PPP) exchange rates, which show what it costs in one unit of foreign currencies to buy goods and setices equivalent to what a dollar will buy. At market exchange rates, rather than PPP exchange rates,Japan’s GDP

&

r employee reached 90 percent of the U.S. level in 1986. The market exchange rate for 1986 was 168.5 yen to the dollav thePPP rate was yen to the dollar. Another measure ofeconomy-wide productivity is GDP per hour worked. Since Japanese workers ut inmom hours

r!year than U.S. or European workem, this measure shows Japan at only58 percent of the U.S. level in 1986, using PPP exc ange

rates, and 6 pereent using market rates.75 See, forcxample, Elliot S. Grossman and George E. Sadler, Comparative Productivity Dynamics: Japan and the United States (Houston,

TX: American Productivity Center, 1982); George E. Sadler, Update: International Productivity Comparisons (Houston, TX: AmericanProductive Center, 1986); Elliot S. Grossman, Pace Univemi

%’ 7

“Productivity and International Competition: United States and JapaneseIndustries, papcrprepared for conference on Interindustry Dif erences in Productivity Growth, Amertcan Enterprise Institute, Washington,DC, October 1984; Martin Bailyand Alok Chakrabarti, Innovation and the Productive Crisis (Washin on, DC: The Brookin Institution,1988); Japan Productivity Center, Productivity Research Institute, International & mpansons of Lbor Productivity @ho o: JapanProductivity Chter, 1988 ; Molly McUsic, o .

J

cit.; and calculations based on data in Organization for Economic Cooperation andDevelopment, Industrial tructure Statistics, 1k (Paris: OECD, 1987).

8/14/2019 8829


U.S. Manufacturing Performance • 49

and 19.5 hours in Japanese-managed plantsin America. In American-managed plantsthe average time for assembly was 26.5hours. The quality of the Japanese autos wasbetter too, judging by the record of defects

owners discovered in the first three monthsof use. The U.S. plants were improving, weregenerally more productive than Europeanplants, and had about as good a record as theEuropean car makers in freedom fromdefects; but the Japanese were getting bettertoo. 76 For another example, Japanese

productivity and quality is conceded to be su-perior in parts of the semiconductor in-dustry, especially in the manufacture of 256Kdynamic random access memory chips.

The solid conclusion that can be drawnfrom available data is that Japan remains theleader in productivity growth. It is normaland expected that countries developing froma rural past to an industrialized future shouldshow high rates of productivity growth; wit-ness Japan in the 1950s and 1960s and Koreanow. What Japan has accomplished in thepast decade is to keep on raising productivityat a rapid rate, after becoming world class in

many industries, raising output, employ-ment, and wages through times of a rising yenas well as a falling yen.

One element supporting Japan’s progressis a high rate of investment in manufacturing.As figure 24 shows, Japan consistently in-

vested more in manufacturing, per dollar oryen of manufacturing output, than the

United States did, from 1973 to 1985.77

Asfor capital invested each year per manufac-turing worker, the Japanese investment (ex-pressed in U.S. dollars) climbed rapidly from1978 on, and b 1985 was 11 percent abovethe U.S. level.

78These figures do not tell the

story for the whole economy. For example,taking services together with manufacturing,the Japanese rate of investment inmachinery and equipment, per employee,was about on a par with the U.S. rate in 1985.

If Japan’s rate of capital investment is higherin manufacturing, the U.S. rate is almost cer-tainly higher in many service industries. For

example, optical scanners of bar codes andcomputerized systems for inventory controlare now commonplace in American super-markets and retail stores. Japan has pouredmost of its investment and management ef-forts into the manufacturing industries thatits leaders see as critical for competing inworld markets. Many services have been

relatively neglected, though not all; certainservices important to international trade,such as banking, do very well.

79

76 Information provided by the International Motor Vehicle Program, Center for Technology, Policy, and Industrial Development ,Massachusetts Inst]tute of Technology.

77 Tle source of this information is the Organization for Ezonomic Cooperation and Development (see figure 24 for source details); 1985is the latest year for which data are available. Investment in manufacturing means gross capital formation, including buildings and producers’durable equipment. Manufacturing output is the share of manufacturing in gross domestic product, that is value added in manufacturing.

78 How U.S. and Japanese investment per worker compare depends a great deal on what exchange rate is chosen. ~lis is not i rue ofinvestment as a share of manufacturing output, which can be fi

8

red in each country’s own currency.) The figures here are in L’. S. doilan,

based on 1985 prices and the 198S purchasing-power-parity (PP exchange rate for fried capital formation in machinery and equipment, Pi>[)exchange rates are developed by the Organisation for Ezonomic Cooperation and Development (OECD) to show what it costs to bu y thesame amount of goods and semices in different currencies. The PPP exchange rate for machinery and equipment in 1985 was 2!46 yen to thedollar.

79 U.S. (km ess, Office of Technology Assessment, International Competition in Services, (1987), op. cit., see espcciai!y chapter 3,

“International ~ repetition in Banking and Financial SeMces.”

8/14/2019 8829



Figure 24Gross Fixed Capital Formation in Manufacturing, 1973-85

Percent of manufacturing GDP25 1

20

15

10

51973 1975 1977 1979 1981 1983 1985

— u s –— Japan Germany

Dollars per manufacturing worker

5000

1000 –

19 1981 1983 1985

— u s — J a p a n Germany

NOTE: Converted to U.S. dollars at 1985 Purchasing Power Parities for Machinery and Equipment Capital Formation

SOURCE: Organization for Economic Cooperation and Development, Flows and Stocks of Fixed Capital, 196085, (Paris: OECD, 1987);OECD, National Accounts, Detailed Tables, 1960-85, Volume 11, (Paris: OECD, 1987)

8/14/2019 8829



Large investment in equipment does not byitself assure either productivity growth orgood performance in manufacturing; howwork is organized and people used with thenew equipment make a big difference. Here,

the Japanese seem to excel. Managerialcompetence is an important source of productivity growth in Japan, especially incomplex manufacturing where many stepsare required and many operations must becoordinated. For example in automobilemanufacture, assembly requires over 1,000independent operations; a report of a fewyears ago found that Japanese auto assemb-

ly plants were twice as productive as inIn engine plants’

American plants.80

withabout 200 operations, Japanese laborproductivity was 50 percent higher. In ironfoundries, where only about 30 steps are

needed, the Japanese advantage disap-peared (see figure 25).

Since the turn of the century, America hasbeen in first place in the most generally usedeconomy-wide measure of productivity,GDP per employed person. If others, start-ing from a lower base, are to catch up andenjoy the same benefits Americans get from

Figure 25.

Manufacturing Productivity in Japan and the United States

--- US/Japan mfg. labor content per unit2 0 0

1 8 0

1 6 0

1 4 0

1 2 0

A u t o A ss e m b l y .

Fork Lift Assembly

Axle A s s e m b l yAuto Engines. .

Stampings

Iron Foundry

.

.

Color TVs

1 0 0.

Ž Men’s Dress Shirts Automatic Transmission

80 1 I I I I I I I -.–u - -l-L- -. !

10 1 0 0 1 0 0 -

Manufacturing Steps

NOTE: As the number of steps in manufacture increases, the ratio of total factory labor content per unit of output of U.S. factories in-creases relative to Japanese factories,

SOURCE: James Albegglen and George Stalks, Jr., Kaisha, Tha Japanese Corporation (New York: Basic Books, 1985)

so James Abegglen and George Stalk, Jr., Kaisha: The Japanese Corporation (New York: Basic Books, 198S). U.S. assembly plants havesince improved, according to the International Motor Vehicle Program of the Massachusetts Institute of Technology cited above. TheProgram’s recent suwey showed that the average Japanese assembly plant now has a 40 percent advantage in productivity over the averageU.S. plant.

8/14/2019 8829



rising productivity-economic growth andrising standards of living — their growth ratesmust be higher, at least for a time. Indeed, ithas been suggested that convergence of productivity levels among industrialized na-

tions is all but inevitable, due to the diffusionof technical knowledge all over the worldand to the application of that knowledge by

This idea con-those striving to catch up.

81

tains some truth but it does not justify com-placency. If U.S. productivity growth were tolag behind that of its trade competitors for

long, the consequences would be serious.The example of Great Britain, the formerworld leader, is cautionary. The U.K.productivity growth rate averaged less thanone percentage point below that of the

United States from 1870 to 1950, but duringthat time the output per capita of the Britisheconomy dropped to 60 percent, andAmerica eclipsed Britain in standard of living and industrial might.

82

61 For an eF

ition of this point of view, see William J. Baumol. “Productivity Grow-th, Convergence, and Welfare: What the Long-RunData Show,” e American Economic Review, vol. 76, no. 5, December 1986.w Angus Maddison, “Growth and Slowdown in Advanced Capitalist Fxonomies: Techniques of Quantitative Assessment,” Journal of

Economic Literature, vol. xxv, June 1987.

8/14/2019 8829


Why Manufacturing Matters

The evidence is substantial that U.S. per-formance in manufacturing has weakened,that several important American industries

do not measure up to the competition, andthat the trouble cannot all be laid at the doorof the high dollar. The next question to ask is whether it matters. Does the Nation real-ly need a strong manufacturing sector? Orhas the time come to gradually cede produc-tion of goods to other countries while in thiscountry manufacturing gives way, in anatural and desirable progression, to perfor-mance of services?

The answer, for now and the foreseeablefuture, is that there is no choice to made be-tween manufacturing and services. The na-tion needs both. As we have seen,manufactured goods are indispensable fortrade with other nations. It is also clear thatAmerica has not entered a post-industrialstage; the demand for manufactured goodsby consumers, businesses, and government isgreater than ever. Moreover, to speak of ser-vices as taking the place of manufacturing in

the economy is to overlook the strong inter-dependence of the two kinds of activities andthe blurring of distinctions between them.Many manufacturing industries could hardlyexist without allied services; the manufac-ture of computers and design of software(often by an independent firm) are an ob-vious example. It works the other way as well.For instance, manufactured hardware makesit possible for hospitals to offer highlysophisticated radiology services such as mag-netic resonance imaging and computerizedtomography.

There are worrisome aspects to this inter-dependence. With the great rise in importsof manufactured goods in the 1980s, a large

number of jobs were lost in manufacturing,and with them went some closely associated

jobs in the service sector. OTA’s analysissuggests that about 6.5 million service sector

jobs were tightly linked to manufacturing in1984. Altogether, some 27.7 million U.S.

jobs were involved in manufacturing, eitherdirectly or indirectly (i.e., producing servicesor material inputs for manufacturing). Jobsassociated with manufacturing are generallygood ones. Manufacturing wages, overall,

are higher than wages in the service sector.Most of the jobs in producer services that areclosely tied in with manufacturing are alsobetter than average. To keep these good

jobs, as well as good jobs in the manufactur-ing sector itself, America must compete ef-fectively in the production of goods.

Links Between Manufacturing and

Services

It is hardly novel to observe that manufac-turing and services are interdependent.When Adam Smith remarked in 1776 that“the labour . . . of artifices, manufacturersand merchants naturally does fix and realiseitself in some such vendible commodity” hewas noting that merchants, although in a ser-vice occupation, are among the workers in-volved in bringing goods to the final

83 There are many other sorts ‘f

purchaser.connections as well. Before reaching themerchant who sells it, the vendible com-

m Adam Smith, The Wealth of Nations, Book IV, chapter IX, cited in J.I. Gershuny and I.D. Miles, The New Sewices Economy TheTransformation of Employment in Industrial .Societies (New York, NY: Praeger, 1983).

53

8/14/2019 8829



modity must first pass through the hands of truckers and warehousers who providetransportation and storage services. Fartherupstream, bankers and venture capitalists,insurance companies, lawyers, engineering

consultants, temporary help agencies, andcomputer specialists all contribute to theproduction of commodities. The service in-dustries, in turn, are important customers of the manufacturing sector. For example, ac-cording to one study, 80 percent of the com-puting, communications, and relatedinformation processing equipment sold inthe United States in 1982 was purchased bythe service sector.

84

Specialized technical skills are particularlyin demand for the manufacture of innova-tive, high technology products. Inmicroelectronics the links betweenmanufacturing and services are exceptional-ly close. The highly successful U.S. computermanufacture industry could not havedeveloped without constant interaction be-tween hardware engineers and softwaredesigners. Software itself is an excellent ex-ample of the marriage of manufacture and

services, since it has the character of both agood (it can be stored and shipped) and a ser-vice (computer programs are not immutablyfixed).

Some kind of services, however, are notvery closely tied to the location of goodsproduction. In general, the service activitiesdownstream of manufacturing–trucking,warehousing, and wholesale and retail sales

of the final product –are not tightly linkedMost Of thesewith domestic manufacture.

85

services can just as well take place with goodsshipped halfway across the world.

86On the

other hand, upstream services —those that

manufacturing firms use as inputs in produc-ing goods —tend to be linked much moreclosely to the place where the goods aremade. These upstream services include suchthings as process engineering, machineryrepair, trucking of goods between related in-dustries, janitorial services, testing and labwork, payroll and accounting services. Solong as manufacturing stays home, so willthese services and the jobs and national in-come they generate. If domestic goods are

displaced by imported ones, or if U.S.-ownedmanufacturing operations are moved off-shore, then many of the tightly linkedupstream services will go with them.

Not all upstream services are so tightlybound. An obvious case is advertising;American agencies, for example, createtelevision ads for Japanese cars. Thanks totelecommunication, some software designhas now migrated overseas, e.g., to India,

where salaries for engineers are much lowerthan for their American counterparts. And itis quite possible for foreign banks to lendmoney to U.S. manufacturing enterprises.By and large, however, upstream servicesthat are inputs to manufacturing stay or gowith the manufacturing activity itself, for thesimple reason that most services are not verytransportable, and are produced near theplace where they are consumed.

u Cited in: James Brian Quinn and Christopher E. Gagnon, “Will services follow manufacturing into decline, ’’Harvard Business Review,November-December 1986, p.96).

as For a detailed discussion of Iinka es, not only of services and manufactunn but also of the reduction of various kinds of oods with5each other, see Stephen S. Cohen and ohn Z~man, Manufacturing Matters: A ? %

Basic Books, Inc., 1987), ch. 2.e Myth of the ost-Industnal Economy (New ork, NY:

aa It should be noted that foreign manufacturers selling in the United States do often set up their own wholesale distribution centem.; anexample is Japanese multinational investment in the United States, which is heavily weighted to wholesale establishments. Many of the jobs,and a good deal of the income, generated by these establishments go to Americans.

8/14/2019 8829


Why Manufacturing Matters q 55

So long as the upstream services areprovided by employees of manufacturingfirms, they are apportioned in the nationalaccounts to manufacturing output andemployment. Often, however, they are

provided by outside firms, in which case thenational accounts attribute to the servicesector activities that are really a part of thefabric of manufacturing. Among the fastestgrowing sectors in the U.S. economy arethose that provide services to companies,rather than consumers. While total employ-ment in all the private service sectors grew atan average rate of 2.6 percent between 1973and 1986, the number of jobs in business ser-vices — which includes advertising, computer

software and data processing, temporaryhelp agencies, management services, and re-search and development laboratories — grewby 7.5 percent a year. Likewise, miscel-laneous professional services (including ar-chitectural and engineering services andaccounting, auditing and bookkeeping) in-creased at the rate of 5.2 percent per year.

Some of this great expansion in businessand professional services in recent years was

tied to manufacturing. To get a quantitativeidea of the connections between manufac-turing and services, an input-output model ishelpful. The model can provide estimates of how much the manufacturing sector buysfrom service industries in the process of making goods and how many jobs are in-volved, and vice versa. In 1984, private ser-vice industries supplied 17 cents of inputs

toward each dollar of manufacturing output.Manufacturing in turn contributed 12 centsworth of inputs toward each dollar of outputof the private service industries.

87

The same kind of exchange held foremployment. Many of the jobs counted inthe service sector are really closely involvedin manufacturing. Estimates based onOTA’s input-output model indicate the jobsinvolved in producing services that are in-puts to manufacturing numbered about 6.5million in 1984.

88In addition, 1.8 million

jobs in agriculture, mining, and constructionwere linked to manufacturing in the sameway (table 8). There were 19.4 manufactur-

ing jobs in 1984. Add to that the 6.5 million jobs in service industries and 1.8 million innatural resources producing inputs formanufacturing, for a total of 27.7 million jobsinvolved, directly or indirectly, in manufac-turing. In turn, some 6.5 million manufactur-ing jobs were devoted to making inputs forthe service and natural resource sectors,Workers in these jobs produce goods rang-ing from tractors to sewer pipes to computersto CAT scanners to paper clips, needed for

the conduct of business by enterprises asdiverse as farms, sanitary services, banks,hospitals, and insurance offices.

89

At a finer level, the service industries thatare most closely involved with manufactur-ing are wholesale trade, transportation andwarehousing, business services, gas, electricand sanitary utilities, and radio and

67 This exchange is on the basis of gross out ut of the manufacturing and private service sectors. The figures do not include purchases of capital equipment or structures needed to pJuce industry output.

aa The OTA model was developed for the asse=ment Technology and the American Ikonomic Transition (op. cit.). OTA’S model is basedon the 1980 input-output tables and is updated to 1984 for employment and sectoral demand. It includes capital flows.m 13stmates of setices jobs closely linked to manufacturing, manufacturing jobs to services, and links of both these sectors wth natural

resources are adapted from the OTA model.

8/14/2019 8829



television broadcasting. Each contributesmore than 20 percent of its employment tomeeting manufacturing demand (table 8). Innumbers of jobs, wholesale trade and busi-ness services are most prominent, together

accounting for about 2.6 million of the 6.5million jobs involved in service sector inputsto manufacturing in 1984.

90

The picture emerging from this analysis isinterdependence — not primacy of manufac-turing as a solid base on which a ratherflimsy superstructure of services is erected,nor on the other hand a succession in which

services are ousting manufacturing from aplace of economic importance. One can alsoconclude that if manufacturing productionand employment is lost, services cannot

Table 8.–Workforce Involved In Manufacturing and Average Full-Time Equivalent Compensation, 1984

Average annualPercent of full-time

Wage and sector equivalentsalary workers employment compensation

involved in involved in (thousands ofmanufacturing manufacturing dollars)

Agriculture . . . . . . . . . . . . . . . . . . . . . . . . . . . . 792 50,4% $11,3

Mining ., . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45.5 37,0

575Construction . . . . . . . . . . . . . . . . . . . . . . . . . . . 13.3 26,8

Manufacturing . . . . . . . . . . . . ., . . 19,396 1000 28 7

All public and private services ., .All private services .,Wholesale trade . . . . . . . . . . . . .T r a n s p o r t a t i o n a n d w a r e h o u s i n gBusiness services . . . . . . . .Radio and IV broadcasting . . .Electric, gas, water and sanitary servicesCommunications, except radio and television . . . .

Automobile repair and services . . . . . . . . . . . . .Retail, except eating and drinking . . . . . .Finance and insurance . . . . . . . ... . . .Hotels, personal and repair services (exe. auto) . . . .E a t i n g a n d d r i n k i n g p l a c e sReal estate and rental* ... .,Amusements . .Health, educ. & social serv. and nonprofit org.

6,4926,3431,501

7041,276

50171129

791,176413207428

724689

9.411.926.324222.8218214116

1161039 08 57 96 74.50.9

24624427,630324,729,637.5397

17817127.415711021 119.920.2

Government ., ., ., ., ., ... 149 0.9 31,1

Total ., 27,697 29.0% $274

SOURCE. ‘Workers involved in manufacturing data derived from OTA Input-Output Model (1980 technical coefficients, 1984 estimated demand, 1984 BLS employment, adjusted

for capital flows, imports and duties) Compensation data dewed from Bureau of Economic Analysis, National Income and Product Accounts, electronic data, mapped toInput-output industry classifications

m The 1980 input-output tables, on which OTA’S model is based, cover only 8S industries, with wholesale and retail trade lumped together.A much finer mesh, covering 537 industries and posting wholesale and retail trade separately, was published for 1977, the benchmark year.For this report, OTA used the 19?7 in ut-output tables to separate wholesale from retail trade, and thus to derive estimates of employment

[associated with manufacturing for eac separately.

8/14/2019 8829


simply and directly replace them. A substan-tial number of service jobs depend directlyon the presence of manufacturing. Manufac-turing and services are strongly enoughlinked that they will prosper together or

decline together.

Links between suppliers and customersmay also be quite close among differentmanufacturing industries. Of course this isnot always the case, since goods can bestored and shipped much more readily thanservices. For example, U.S. automakers buycomponents and parts, from engines towindshield wipers, all over the world. At thesame time, some major automakers, and

other manufacturers as well, are developingstronger bonds with local suppliers. Havingsuppliers close by enables companies to use

just-in-time deliveries, and helps in develop-ing long-term, cooperative relations with thesuppliers–both key elements in Japanesemanufacturing strategy. In the textile/ap-parel business, for example, a leading U.S.maker of jeans has completely changed itsrelations with denim suppliers in the past fewyears. Instead of driving the hardest possible

bargain on price with competing suppliers,the company now buys most of its denim inlong-term arrangements from two or threetextile manufacturers, gaining the ad-vantages of consistent high quality and just-in-time delivery. In fact, the jeansmanufacturer now keeps virtually no inven-tory and has turned an entire warehouse intosewing space.

Different segments of whole industry com-

plexes may depend on each other to a greaterdegree than one might suppose, if relationsbetween supplier and manufacturing pur-

chaser were governed only by technical pos-sibilities and not at all by spatial bonds.Cohen and Zysman draw examples fromagriculture; they say:

92

It is technically possible, buteconomically improbable to millsugar cane in a country far fromthe sugar fields, or to processtomatoes far from the tomatopatch, or to dry grapes into raisinsor crush them for wine far from thevineyard. It is a forward linkagestarting with farming; foodprocessing is downstream in theproduction chain . . . In agriculture,both in theory and in what is too

often dismissed as mere real-world examples, tight linkagesbind in both directions. There aremany activities tightly bound tofarming that are backwardlinkages: crop dusters, animalvets, harvesters, tractor repairers,mortgage appraisers, fertilizersalesmen, blight insurers,agronomists, chemists, truckers,shuckers.

The fiber/textile/apparel complex providesanother example. It is conceivable thatAmerican textiles could be sold to HongKong apparel makers, but the U.S. chemicalcompanies that make fibers and the textilecompanies that spin and weave the fibers arenot counting on it. Both are taking a leadingrole in strategies to strengthen the U.S. ap-parel industry, partly by forging strongerlinks among all segments of the industry,from textiles to apparel to designers andretailers. (Some of the chemical companies

91 OTA intetiew with Thomas O“Gorman, President, Greenwood Mills,w Ckhen and 7,ysman, op. clt,

8/14/2019 8829



are also hedging their bets by producing fiberin Southeast Asia, near textile and apparelmanufacturing centers.)

An input-output model is not much help in

showing the strength of the ties betweenmanufacturing companies. It can show whatmaterials or intermediate goods one in-dustry buys from another, but not whetherone depends on the presence of another inthe same national economy. For an accurateview of the strength of these spatial bonds,empirical studies of individual manufactur-ing industries are needed; OTA’s full assess-ment of Technology, Innovation, and U.S.Trade will discuss these kinds of connections

in several manufacturing complexes.

Manufacturing and the Quality of

Jobs

The kinds of jobs associated with manufac-turing are important as well as the number.Pay is better in manufacturing than in theprivate services overall, and has consistentlybeen so for many years. Moreover, the ser-

vices jobs most closely connected withmanufacturing tend to pay better than ser-vices in general.

Total compensation –wages, salaries andbenefits –of people employed in manufac-turing in 1984 was $28,700; for all workers inthe services, it was $22,900, and in theeconomy overall, $24,300 (see table 8).

93

Jobs in transportation and warehousing,radio and TV broadcasting, and utilities paid

as much or more than the manufacturing sec-tor itself, and they are closely linked to it.Over 20 percent of their output goes intomanufacturing as inputs, compared to lessthan 12 percent in private services as a

whole. Business, legal, and professional ser-vices, a category that includes everythingfrom janitors to corporate tax lawyers, is alsoclosely tied to manufacturing; jobs in thisgroup of industries paid above average forthe services, but below manufacturing.Wholesale trade, which has a higher propor-tion and larger number of jobs (1.4 million)associated with manufacturing than anyother service industry, paid nearly as well asmanufacturing.

Some service industry groups that do notsell a large share of their output to manufac-turing still devote a large number of jobs tomanufacturing input. The most important of these is retail trade, which had 1 million jobsassociated with manufacturing in 1984, andretail pay is low; Yearly compensation (perfull-time worker) averages $17,100. Otherswith fairly large numbers of jobs linked tomanufacturing but low pay are eating and

drinking places ($11,000) and hotels andpersonal services ($15,700). These threelow-paying industry groups employed one-quarter of the U.S. service sector workersmaking inputs for manufacturing in 1984.OTA has calculated the average compensa-tion for jobs in all the service sectors tightlylinked to manufacturing at $24,600, com-pared to $22,900 in the services as a whole,in 1984.

94The difference in pay between

private (non-government) services linked to

83 All compensation figures are given on a full-time worker basis; this eliminates a downward bias in compensation for the sexv-ice industries,which have a greater share of part-time jobs than manufacturing. h’umbers of workers are also given as full-time equivalents. The reason forchoosing 1984 as the year for comparing compensation in different sectors is that OTA’S input-output analysis showing relations betweenservice and manufacturing jobs was done for that year. Pay in manufacturing jobs and in service jobs associated with manufacturing remainedbetter than in setice jobs generally in 1986.w This is a weighted average, based on the data in table 8.

8/14/2019 8829



manufacturing and all private service in-dustries is even more pronounced – $24,400versus $21,900.

What is the basis for the longstanding supe-riority of wages in manufacturing, and in ser-vices closely related to manufacturing, overthe rest of the economy? Possibly, higheroutput per employee hour. It has long beenconsidered a truism that productivity is bet-ter in manufacturing than in services. This isnot entirely true. What does seem to be trueis that manufacturing and the distributionand producer services with closer than

average links to manufacturing have higherthan average productivity as well as higherthan average pay. The obverse does not hold,however. Finance, insurance, and com-munications have high productivity and paywell, even though they have only an average,or below average, degree of association withmanufactur ing.

According to official figures compiled bythe Bureau of Labor Statistics (BLS),productivity growth in manufacturing hasbeen higher than in private business as awhole for many years, since 1960 at least(table 9). The discrepancy appears especial-ly remarkable in recent years. From 1979 to1987, manufacturing productivity rose at anannual rate of 3.4 percent, while for privatebusiness as a whole (including manufactur-

ing), the yearly growth rate was 1.3 percent.Leaving out agriculture, the growth rate forall private business was only 1.1 percent.These figures seem to mean that manufac-turing has carried the whole economy on its

back in raising productivity, especially sincethe 1970s. Recall, however, that the produc-

Table 9. – Productivity in Manufacturing andAll Business, 1980-87

(1977 = 100)

Year All business Manufacturing

1960 . . 67.31961 : : : : : : . . . . . 69.71962 ., 72.31963 : . . 75.2

1964 . . : 78.41965 ., ., : : : : 80.81 9 6 6 . . . . , 82.91967 . 85.21968 : : : 87.61 9 6 9 87.71970 . 88.41 9 7 1 91.31972 94.01 9 7 3 95.91974 . 93.81975 95.71 9 7 6 98.41977 100.01978 100.81979 99.51 9 8 0 99.2

1 9 8 1 100.61982 100.31 9 8 3 103.01984 105.61985 107.51986 109.51 9 8 7 110.5

Annual average growth rates1960 73 2 81 9 7 3 7 9 0 61979 87 13

62.264.066.771.2

74.676.677477.479.880.88 0 885.3

89.09 3 490 69 2 9971

100.0101510141014

1036105911201181124212881330

3 21 43 4

SOURCE. U S Department of Labor, Bureau of Labor Statistics, electronic data,

Monthly Labor Review, various Issues, table 44

8/14/2019 8829



tivity figures are based on a constant-dollarseries for gross national product that maysubstantially understate the share of manufacturing in GNP for earlier years —and thus overstate its growth in real output,

value added, and productivity.95

Another way to compare productivity of manufacturing and various service industriesis to look at their respective levels (notgrowth rates) in one recent year, thus avoid-ing the problems of using a constant dollarseries over time. For this purpose, produc-tivity can be calculated as value added in anindustry or sector, divided by the number of hours worked in that industry.

%On this

basis, manufacturing productivity in 1986was $20.27 an hour, compared to $18.08 forall private services averaged together (seetable 10).

97The average conceals a more in-

teresting story. Business and professionalservices and the transportation andwarehousing industry, all of which devoteover 20 percent of their output to manufac-turing, are virtually the same as manufactur-ing in productivity. Public utilities, anotherindustry with close links to manufacturing,

has exceptionally high productivity–over$65 of value added per hour. Wholesaletrade, with its high proportion and largenumber of jobs involved with manufacturing,has productivity well above average.

Table 10.-Value-Added per Hour, byIndustry, 1986

Percent ofValue sectoradded employment

involved in

hour manufacturingAgriculture . . . . . . . . . . . . . . . . . . $12.68

Mining . . . . . . . . . . . . . . . . . . . . 54.55

Construction . . . . 1563

Manufacturing . . . . . . . . . . . . . . . 20.27

Public and private services . . . . . . . 20.21All private services . . . . . . . . . . . . . 21.86Wholesale trade . . . . . . . . . . . . . . 24.31Transportation and warehousing 21,25Business services . . . . . . 19,60Radio and TV broadcasting . 27.49Electric, gas, water ands a n i t a r y s e r v i c e s 65.31

Communications, except radio

and television . . . . 45,93Automobile repair and services 15.16Retail, except eating and drinking 12.62Finance and insurance . . 20.58Hotels, personal and repairs e r v i c e s ( e x e . a u t o ) 10.63

Eating and drinking places 1541Real estate and rental* 161.28Amusements . ., . 14,51

Health, educ. & social serv. andn o n p r o f i t o r g . 1323

G o v e r n m e n t . 14.57

Private services excludingr e a l e s t a t e 18.08

Total . . . $1777

50.4%

45.5

13.3

100.0

9.411,926.324.222.821,8

21.4

11,611,610.39.0

8.57.96 74 5

0.90.9

12.0

22.1%

Value-added includes inputed rent from other sectors of the economy

SOURCE: Value-added by industry from U S Department of Commerce, Bureau ofEconomic Analysis, National Income and Product Accounts, GrossProduct Originating by Sector, electronic data, hours of all persons en-gaged from U S Department of Labor, Bureau of Labor Statistics, Office

of Economic Growth, Hours of All Persons, unpublished data, workersinvolved in manufacturing dewed from OTA Input-Output Model (1980

technical coefficients, 1984 dewed demand, 1984 BLS employment,

adjusted for capital flows, imports and duties)

m If the constantdollar value added figures are taken literally, the level of roductivity in manufactunn was much lower than in services,{ %and in the economy as a whole, until quite recently. For example, using the B ‘A series on value added in 1 2 dollars, manufacturing output

per full-time e uivdent employee in manufacturing appears to have been $20,900 in 1960, compared to $2S,700 for rivate services (excluding! freal estate),an $29,200 for the economy as a whole; for 1986 the comparable figures are S44,000 for manufacturing, 32,800 for private services

except real estate, and $38,700 for the whole economy. A series calculated on the basis of constant 1982 dollars shows the level of productivityin manufacturing lower than that of the economy as a whole from 1960 through 1982. In fact, since the figures are still relatively close together,the only way productivity could have grown so much faster in manufacturing than in the rest of the economy for such a long time is to havestarted at a much lower level.

EM Value added is the difference between the cost of materials, parts, and services that an industry buys to produce an item or semice, andthe sales revenues the industxy collects. The constituents of value added, as usually calculated, are wages, interest, rent, profit, depreciation,and indirect taxes.

97 As noted in table 8, real estate is excluded from this calculation, because in the national income and product accounts, real estate valueadded includes not only agency fees but also all rents and all imputed rents for owner-occupied dwellings. By this definition, value added inreal estate is not really equivalent to value added in other sectors, but is more like gross output, and is inconsistently high.

8/14/2019 8829



Highly productive enterprises and good jobs certainly exist in service industries otherthan those closely linked to manufacturing.The communications industry, includingtelephone and telecommunication services

but excluding radio and TV broadcasting, isnear the top in productivity, but has no morethan average links with manufacturing.Finance and insurance, a very large sectorwith employment of 4.9 million in 1986, hasproductivity equal to that of manufacturing,but is not at all strongly linked. It is also truethat none of the service industry groups atthe bottom of the heap in productivity – con-sumer, social, and retail services — is veryclosely tied to manufacturing (11 percent or

less of their output goes into manufactur-ing).

98

Value added per hour in these in-dustry groups is down in the range of $12 to$13.

All this said, it must be recognized thatthere is something quite unsatisfactoryabout comparing productivity from one sec-tor to another. Ideally, productivity would becalculated on the basis of how many man-hours it takes to produce a physical quantity

of a good or standard unit of service. TheBLS does produce productivity studies of this sort for specific industries. But goods areunlike each other, and services are more dif-ferent still. To look at productivity in theeconomy as a whole, or across sectors, theonly common unit of measurement is dollars.

It may seem straightforward enough tofigure productivity in both goods and ser-

vices as value added per hour. But what isvalue added? By definition, it is the sum of wages, interest, rent, profit, depreciation,and indirect business taxes in the sector orindustry under consideration. A large

proportion of value added, varying by in-dustry but generally about one-half to two-thirds, is wages and salaries plus corporateprofits. Thus, if wages and profits are rela-tively high in an industry, its value added, andtherefore its productivity, will show up ashigh. This may reflect genuinely highproductivity-that is, high physical outputper hour worked; indeed, the economicfoundation for good wages and living stand-ards is high productivity. But within an

economy, one sector’s wages may be higherthan another for reasons other than produc-tivity.

An industry with a lot of market power, i.e.,in a near-monopolistic position, can oftenextract higher prices, and therefore payhigher wages and profits, than one that ismore competitive but equally efficient inusing labor. Consider steel up until the late1970s, before international competition and

declining demand destroyed the industry’smarket power. The steel industry paidpremium wages, and on the basis of valueadded, had above average productivitygrowth in the 1970s. But on the basis of physi-cal output per hour (as calculated by theBureau of Labor Statistics), steel’s produc-tivity growth was below the all manufactur-ing average for most of the decade.

l00By

contrast, agriculture, which has shown strongproductivity growth in physical measure-

ea Note, however, that a ve~ large number of retail trade jobs is associated with manufacturing, even though the proportion of linked jobsin this veV large sector is only 11 percent.w Other factom alw affect the ability of near-monopolistic industries to set prices above a competitive level, and thus pay higher wages

and profits than they otherwise could. ‘1’hese factors include the degree of elasticity of demand for labor, the elasticity of substit ution betweenlabor and capital, and the elasticity of demand for the Industries’ output,

leoU.S. Con ress, Office of Technology Assessment, U.S. Industrial Corn4 T

titiveness: A Comparison of Steel, Electronics, andAutomobiles ( ashington, DC: US, Government Printing Office, 1981), pp. 56-

8/14/2019 8829


ments of output (such as bushels of wheatper hour worked), is highly competitive, payslow wages, and has low value added in dollarterms. Also, strong unions can raise wages;and social practices such as paying nurses

(female) less than truck drivers (male) canlower pay. Because of these other influenceson wages and profits, value added (dividedby employee hours) is no more than a roughguide to levels of productivity indifferent in-dustries.

The fact remains that manufacturing paysbetter than services, and so do linked ser-vices. This at least suggests that manufactur-ing is able to pay both its employees and its

service suppliers relatively well because of superior productivity. This does not meanthat other services cannot provide good jobs.Within the designation of “services’’ are verydifferent kinds of activities; all they have incommon is that they do not produce tangiblegoods, and even that distinction is blurred insome industries, such as software. Some in-dustries in this disparate collection do in-deed have low productivity and pay, andemployment in these industries (e.g., retail

trade) is so large that they pull down theaverage pay for services in general. Of theservice industries that are better paid andmore highly productive, several have in com-mon a substantial dependence on advancedtechnology (e.g., computers in banking, in-surance, and telephone communications), orhigh capital investment per worker (e.g.,public utilities), or both.

101These features

are also found in the services most closelylinked to manufacturing.

High Technology Industries

Manufacturing industries at the cuttingedge of technology, in products or processes,

help to buoy the economy, provide new jobs,improve the trade balance, and advancetechnology outside their own industry as wellas within it. Traditionally, whatever in-dustries were at the technological forefrontfor their time have helped to give the UnitedStates a competitive edge. The criteria wide-ly used to define high technology industriesare higher than average ratios of technology-oriented workers, and average or higher thanaverage research and development spend-

ing.

102

A list of 26 manufacturing industriesbased on these two criteria includes most of the ones that people intuitively select as hightech (table 11). Among them are computers,electronic equipment and components,communication equipment, precision in-struments, specialized engineering products,aerospace, chemicals, and drugs.

Clearly, high tech industries are vital to thenation’s future. The development of

knowledge-intensive, technologically ad-vanced products and methods of manufac-ture, from supercomputers to robotics tobiotechnology, is indispensable for a betterquality of life and rising incomes. The ques-tion, however, is whether high tech in-dustries can fill the gaps left by the declineof traditional manufacturing industries,creating high wage jobs and making goodsfor export to offset imports of standardproducts made by lower wage workers in

lol For an indepth discussion of different kinds of service industries and a classification based on knowledge-intensiveness, see U.S.Congress, Office of Technology Assessment, International Competition in Services, OTA-lTE-328 (Washington, DC: U.S. GovernmentPrinting Office, Juty 1987).

I@This is the definition of Grou 111 high technoloE. Hecker, and John U. Burgan, “ f F

industries developed by the Bureau of I,abor Statistics; see Richard W. Riche, Daniel{igh Technology oday and Tomorrow. A Small Slice of the Employment Pie,’’ Monthly bbor Review),

November 1983.

8/14/2019 8829



Table 11. - U.S. High TechnologyManufacturing Industries

Electrical components and accessoriesOffice computing and accounting machinesCommunication equipmentAircraft and parts

Measuring and controlling instruments

Surgical, medical, and dental instrumentsGuided missiles and space vehiclesDrugsMiscellaneous electrical machinerySoaps, cleaners and toilet preparations

Industrial organic chemicalsOptical instruments and lensesEngineering, laboratory, scientific and research instrumentsPhotographic equipment and suppliesAgricultural chemicals

Miscellaneous chemical productsIndustrial inorganic chemicals

Engines and turbinesPetroleum refining

Electrical industrial apparatus

Ordnance and accessoriesPaints and allied productsSpecial industry machineryElectrical, transmission and distribution equipmentRadio and IV receiving equipment

Plastic materials and synthetics

NOTE. High technology manufacturing industries are defined as those with a pro-portion of technology -oriented workers (engineers, Iife and physical scien-

tists, mathematical specialists, engineering and science technicians and

computer spec ialists) eq ual to or greater tha n the average for all manufactur-ing industries, and a ratio of R&D expanditures to sales close to or above the

average for all Industries

SOURCE Richard W Riche, et al , "High Technology Today and Tomorrow a

Small Slice of the Employment Pie,” Monthly Labor Review, November

1983

other countries, as some have suggested.103

The answer is no. High tech industries donot stand alone, any more than services.Though they are necessary to the generationof jobs, wealth, and exports, they cannot dothe job alone.

A great many of the products of high techindustries are intermediate goods used by

other industries, both other high tech in-dustries downstream (e.g., computers) andmore traditional industries (e.g., autos).

There is little consumer demand for semi-conductor chips, lasers, or programmablemachine tools. The big consumer demand isfor goods such as cars, compact disk players,microwave ovens, and washing machines

which, increasingly, contain advanced tech-nology products or are made by advancedmanufacturing methods. For example, theauto industry is one of the largest users of computer aided design and computer as-sisted manufacturing equipment (CAD-CAM), robots, and sophisticated machiningcenters, and is also one of the largest pur-chasers of semiconductor chips.

Semiconductors illustrate the point that

high tech industries depend on other in-dustries to buy their wares. Excluding cap-tive producers (e.g., IBM and AT&T) whomake chips mostly for their own use, 85 per-cent of the 1986 output of the U.S. industrywent to non-military industrial customers,who use semiconductors in the process of manufacture or embed them in autos orother consumer goods. About 40 percent of the chipmakers’ output went to manufac-turers of data processing equipment (includ-

ing computers), and another 15 percent toproducers of communications equipment.Sixteen percent went to industrial machineryand equipment, 7 percent to consumerelectronic goods and 8 percent to transpor-tation equipment.

104Strong demand from

the semiconductor-using industries, bothtraditional and high tech, is fundamental tothe strength of the semiconductor industryitself.

Of course, that demand need not all bedomestic demand. The United States does

IcMSee, for example, Robert Z. Lavrence, Can America Cmmpcte? (Washington, DC : The 13rookings Institution, 1984), ch. 4,l~National .Seience Foundation, The Semiconductor Industry, report of a I:ederal Interagency Staff Working Group (Washington, DC:

National Science Foundation, 1987), p. 6, chart 2, based on in format Ion from I)ataquest.

8/14/2019 8829


8/14/2019 8829



the necessary precision economically, bothfor VCRs and for miniaturized productssuch as Sony’s Walkman.

None of this is to underrate the past con-

tributions of high technology industries inAmerica and the importance of their con-tinued development. Several of these in-dustries have expanded much more stronglythan manufacturing in general in recentyears, and have added jobs even as theirproductivity soared. For example, employ-ment in computer and semiconductormanufacture combined grew from 520,000 in1979 to 679,000 near the end of 1987.

110In

the 26 high tech manufacturing industries al-together, employment also rose, but moremodestly, from 5.1 million to 5.3 million.

As discussed below, the surpluses of pastyears in hightech trade have nearly vanished.High technology industries face increasinglystiff competition, both from lower wage butrapidly industrializing Asian countries, and

from the higher wage but highly competitiveJapanese. World demand for goods relatedto microelectronics is expected to go onrising strongly, but American manufacturerswill have to scramble to keep their share of demand and their output growing. One in-gredient in success has to be reliable demandfor high tech goods by a strong Americanmanufacturing sector.

1 I oThe composition of employment changed in computer manufact ure, howeveq while total employment rose from 320,(MI0 to 406,000 from1979 to 1987, jobs for blue<ollar production workers dropped from 131,0CKI to 128,000.

8/14/2019 8829


The Anatomy of Trade

In thinking about different ways in whichthe United States might turn the trade deficitaround, it is useful to consider first what U.S.

trade consists of–what we trade, what arethe biggest items in the deficit, who are ourmost important trading partners, and whichof them run the biggest surpluses with theUnited States. These facts about theanatomy of trade as it is now point to the ad-

justments that will have to be made whenU.S. trade comes back into balance.

Products

Manufactured goods account for most of the merchandise trade deficit. Amongmanufactured goods, by far the most impor-tant deficit item is motor vehicles, parts andengines. The deficit in automotive importsalone was over $53 billion in 1987, havingrisen more than tenfold since 1976 (figure26); it now amounts to about one-third of theentire deficit in merchandise trade. WhenU.S. trade deficits fall, it is clear that much

of the reduction must be in automotiveproducts — either through importing less orexporting more or both.

Other industrial sectors are also runningsizable deficits, and also face pressures foradjustment (table 12). Of course, it is notnecessary to reach a balance in every in-dustry; some with surpluses can compensatefor others that are in deficit. But the deficitsare so great in a few industries that it is hard

to see what others could generate highenough surpluses to offset these deficits. Astable 12 shows, electronic equipment, in-

cluding items ranging from semiconductorsto television and radio sets, ran a $23 billiondeficit in 1987, mitigated only slightly by a $1billion surplus in computers and automaticdata processing equipment. The textile andapparel industry complex hit an all-time high(or low) of $21 billion in deficit. The industrygroups with the strongest trade performancewere aircraft and other transportation

Figure 26

Balance Of Trade in Automotive Products,

1 9 7 6 -8 5

_ Billions of dollarso

-lo -

-20 -

-30 -

-40

-50

-60

t-70 L

1976 1977 1978 1979 1980 19611962196319641965 19681967

SOURCE: U.S. Department of Commerce, Bureau of Economic

Analysis, Business Statistics, 1986, p. 80-81, ForeignTrade of the U.S. - Value of Exports and Imports

equipment (excluding autos) with a surplusof over $12 billion, and chemicals, with ahealthy and rising surplus of nearly $10 bil-lion.

It is noteworthy that the worsening tradebalances in manufacturing have not sparedhigh technology products.

111Between 1985

11 I For international trade, the Department of Commerce defines high technology products as those embodying high levels of research anddevelopment expenditures

~r unit of output; the set of industries producing these ~oods is similar to the list based on the Bureau of Labor

Statistics criteria. See U.S. epartment of Commerce, International Trade Administration, United States Trade Performance in 198.S andOutlcmk (Washington, DC: U.S. Government Printing Office, 1986).

66

8/14/2019 8829


The Anatomy of Trade q 67

and 1987, for example, the positive trade billion in 1986. The improvement in 1987balance in computers and automatic data was due mostly to rising surpluses in theprocessing machinery dropped by $2.8 bil- aircraft and chemical industries.

112

lion, to about $1 billion (table 12). Overall,the trade balance in high technology The record still illustrates something fun-products shrank from a surplus of $27 billion damental: high technology industries havein 1980 to a surplus of only $600 million in come under many of the same pressures af-1987, having gone through a deficit of $2.6 fecting other manufacturing industries.

Table 12.–Trade Balance In Selected Manufacturing Industries(billions of dollars)

Industry 1985 1986 1987

Total manufacturing

Durable goodsWood and cork manufactures

Furniture and partsNonmetallic mineral manufacturesIron and steelNonferrous metals

Misc. metal manufactures*Industrial machinery

Power generating machinerySpecial industrial machineryMetalworking machineryOther industrial machinery

Electronic, computing, and office machineryOffice and ADP machineryTelecomm and sound reproducing equipSemiconductors and other electrical equip

Motor vehiclesAircraft and other transport equipmentProf. , scientific and control inst

Photo equip , optical goods and timing equipMisc. manufacturing**Military arms, ammo, vehicles

Non durable goodsTextiles and apparel

Yarns, fabrics and textile articlesWearing apparel and accessories

FootwearPaper, paperboard and manufacturesChemicals

Organic chemicals and related products

Medicine and PharmaceuticalsFertilizersSynthetic resins, rubbers and plasticsOther chemical materials and products

Tires and tire tubes

Luggage, handbags, and similar articles

$101.6

$1.4

3.15.89.95.33.70.10.42.21.60.9

15.33.8

14.44.7

39.811.23.4

2.410.32.7

1532 8

1246 03 9671316112.10 617

15

$1289

$1.4

3 96 78 4

-634 55 40 50 21.92 8

20814

1625 9

5161083 0

2 91192.0

17.63 5

1416 74 07 21419102 40 618

16

$137.7

-$1.6

-4.4-6,8-8.5-6,04 9

-6,7-0.6-1.6-1,4

--3.1

21.61.0

15 6- 7 053.312.53,0

3.112.82.0

20 8- 3 916 97 4

-4 49 6

1 91.71.43.4111 9

2.0

* Not specified elsewhere** Not specified

SOURCES: U S Department of Commerce, Mice of Trade and Investment Analysis, unpublished data

1121nformation prcwldcd t~> the L’ S, Ikpartmcnt of (bmmc rce, ( )fflcc of “1’radc and Invest mcnt Analysis,

8/14/2019 8829



Trade in high technology products is sub-stantial, with imports and exports bothreaching just over $80 billion in 1987; hightech imports were almost one fourth of allmanufactured imports, and high tech exports

were 42 percent of all manufactured exports.While American high tech companies arestill quite competitive, it is unlikely that theycan regain the kind of dominance they had

just a decade ago, relative to producers inEurope, Japan, and a few developing Asiannations. It is therefore unlikely that hightechtrade can generate a large enough surplus tooffset substantial deficits in traditional sec-tors.

Countries

The United States trades mostly with otherdeveloped nations, although trade withseveral developing nations has expandedgreatly in recent years. The top ten suppliersof U.S. imports in 1987, in descending order,were Japan, Canada, West Germany,Taiwan, Mexico, the United Kingdom, SouthKorea, Italy, France, and Hong Kong.

113

When the trade deficit is reduced, most of the adjustment will fall on these countries(figure 3). The impact of the adjustment willvary by nation and by region, depending onhow important trade with the United Statesis to our trading partners. It will also dependon how fast the economies of other countriesare growing; the more other markets expand,the easier will be the adjustment to reducedsales (or slower growth in sales) to theUnited States.

The developed nations that are our largesttrading partners will probably have to bearmost of the adjustment costs. Six of the topten suppliers of U.S. imports are developednations: Japan, Canada, West Germany, the

United Kingdom, Italy, and France. Theirmerchandise trade surpluses with the UnitedStates totaled nearly $96 billion–about 60percent of the U.S. merchandise tradedeficit. Figures 27 and 28 show, in percent-ages, the part played in U.S. trade by ourleading trading partners from 1950 to 1986.

Adjustment costs will mean different thingsto different nations. If American exports areto grow faster than imports, nations that ex-

port to the United States can maintain exportlevels only if worldwide economic growth,including U.S. growth, is substantiallygreater than it has been in recent years.Faster growth” in the American economy–the world’s largest — is not likely, since wecannot continue to maintain consumption,investment, and government deficits at cur-rent levels indefinitely. Under these cir-cumstances, it will be difficult for foreignproducers to maintain their levels of exports

to the United States, and many will find thatthey must replace U.S. customers with othersor produce in the United States instead of inthe home country. The alternative tothese cutbacks would be rapid economicgrowth rates in the exporting countries, thusenabling them to substitute their ownmarkets as well as others for the U.S. market.Except for Japan these countries have so farshown little evidence of being able or willingto do SO.

113U.S. Department of Commeree, International Trade Administration, U.S. Merchandise Trade Position at Midyear 1987, (Washington,D. C.: U.S. Government Printing Office, October 1987).

I IdUntil 198S, however, investment by foreign countries in the United States had done nothing to improve the U.S. merchandise tradebalance, in fact quite the contrmy. This situation may already have begun to change with the falling dollar, but whether or how soon foreigndirect investment in American producion replaces imports is uncertain..

8/14/2019 8829



Figure 27Volume of U.S. Ex por ts , 1 9 7 8 -8 6

Percent of total U S merch. exports1 0 0 % - - - - - -

80%

60%

40%

20%

0%

1978 1979 1980 1981 1982 1983 1084 1985 1986 1987

SOURCE: U.S. Department of Commerce, Bureau of Economic Analysis, Survey of Current Business, table 3, June 1987 and March1988.

Figure 28SVolume of U. . Imports, 1978-86

Percent of total U.S. merch. imports100%

8 0 %

60%

4 0 %

20%

0%V

1978 1979 1980 1981 1982 1983 1984 1985 1988 1987

SOURCE: U.S. Department of Commerce, Bureau of Economic Analysis, Survey of Current Business, table 3, June 1987 and March1988,

8/14/2019 8829



The greatest bilateral merchandise deficitof the United States – $57 billion in 1987, orabout 36 percent– is with Japan. Twenty-one percent of U.S. merchandise imports, or$85 billion, were from Japan, and about 11

percent of our exports ($28 billion) are sentthere. The deficit with Japan has been one-third to one-half of the U.S. merchandisetrade deficit for about the last decade, grow-ing during that time approximately tenfold.In 1986, the leading import by far from Japanwas passenger motor vehicles (table 13).

115

This item accounted for $23 billion, or overone-fourth of all imports from Japan, andnearly half the U.S. deficit in motor vehicletrade. Besides motor vehicles, other major

imports from Japan include consumerelectronics, telecommunications equip-ment, automatic data processing and officemachinery, and electronic components.

Some adjustment has already taken place.For example, even before the fall of the dol-lar against the yen, beginning in early 1985,there were pressures on and within Japan tochange its postwar policies of export-leddevelopment.

116 Many nations were exhort-

ing Japan to reduce its trade surplus, and in-creasing saturation of some export marketswas apparent. The Japanese government hasannounced an official policy of lesserreliance on exports. But shifting to aneconomy more dependent on growth of domestic consumption is not simple; exportgrowth accounted for almost 40 percent of Japan’s economic growth between 1980 and

1985.117 As exports slackened in 1986,

Japanese GNP growth faltered somewhat,rising only 2.4 percent compared to 4 and 5percent in the earlier 1980s. Capital-invest-ment plans were revised downward.

118

Japanese firms and industries that were par-ticularly hurt by stagnating demand, like thesteel industry, began to diversify, enteringhigh technology fields like special chemicals,new materials and biotechnology.

119At the

same time, there were layoffs, especially inthe steel industry. Nippon Steel, for ex-

Table 13.-Major U.S. Imports From and Exports toJapan, 1986

CompoundC i f value* annual

(millions of growth ratedollars) 1982-86

Import categoryPassenger motor vehicles $22.8 21 .2%Phonographs, IV image &

sound reproducing equipment . 6.0 371Special purpose motor vehicles 51 316Telecommunications equipment,

nspf** 4.0 239Parts of motor vehicles, nspf** 3.1 508Automatic data processing machines 2.9 64.1

Export categoryGold (nonmonetary, except ores) 3.3Air and spacecraft, etc. 1.8Corn or maize, unmilled 0.9Oilseed and oleag. fruit ., 0.8

Wood (rough cut) ., ., 0.8Meat (fresh, chilled, frozen) 0.7ADP machines 0.7Parts for office machines 0.7Radioactive and assoc. material 0.6Organic chemicals and products 0 6Medicinal and pharmaceutical prod O 6Fish (fresh, chilled, frozen) 0 6Measuring and checking instruments O 6Coal and lignite 0 6Petroleum products (refined) 0 4

6951896 93 7

179 71691318 83 96 6

12374

40827

“C I f value of imports includes cost, insurance and f reight

q *Not specified

SOURCE U S Department of Commerce, International Trade Administration,

1986 U S foreign Trade Highlights, Office of Trade and Investment

Analysis, March 1987

I IsDetailed figures on trade by product and by country were not yet available for 1987 when this report was written.I la.lon Woronoff, “Japan’s Structural Shift from 13qmrts to Domestic Demand, ”in Japan’s F~onomy and Trade wilh the United States:

selected Papers, Subcommittee on Ronomic Goals and Intergovernmental Policy of t he Joint IZonomic Committee. Congress of the UnitedStates (Washington, D. C.: U.S. Government Printing Office, December 1985).

1 ITRobert J. Samuelson, “Japan’s Case of Malaise, ’’Newsweek, May 4, 1987.I laNo Big Deal,’ The Economist, November 8, 1986.I IiJapan: Steelmaker are Vigorously Restructuring, ’’Foreign Broadcast Information Setice, 2.5 l:ebrua~ 1987.

8/14/2019 8829



ample, announced a temporary layoff of 3,000 of its workers in October 1986.

120

After this rather rough year, the Japaneseeconomy bounded back in 1987, with a GNP

growth rate of 4.2 percent. The revival wasfueled by a housing boom, an increase inconsumer spending (spurred by a tax cut),and a $46 billion government spending pack-age, including a 20 percent increase in thepublic works budget. A tax deduction onmortgage loans and a cut in interest rates en-couraged the housing boom; housing startsrose over 18 percent in the first half of 1987.

121The construction activity spilled

over into greater demand for steel, which

staged a substantial recovery, and a wholerange of consumer and household goods.

122

Whether domestic demand will continue torise at the 1987 rate, compensating for slow-ing or declining external demand and ashrinking trade surplus, is yet to be seen. TheJapanese economy has proven extraor-dinarily resilient in difficult circumstancesbefore, notably after the oil shock of theearly 1970s. And Japanese manufacturersare responding to the high yen by paring

profit margins and redoubling efforts to raiseproductivity.

123

At the same time, Japanesecompanies are beginning to move somemanufacturing operations offshore, inresponse to the high yen, and these movesare bound to have some dislocating effectson employment and the economy. The ad-

justment to a higher yen is not yet over.

While the coming changes are not simpleand easy for Japan, they could be harder forsome of our other trading partners. Amongdeveloped nations, Canada and the UnitedKingdom – our second and fifth largest sup-

pliers of imports, respectively–are in themost difficult positions; both countries runtrade surpluses with the United States, butsustain overall trade deficits and relativelyshaky economies. Canada, whose economy isheavily dependent on the American market,may face great difficulty — even if the newlyestablished free trade agreement is effectivein further liberalizing trade between the twocountries.

Between 1976 and 1987, the U.S. merchan-dise trade deficit with Canada increasedfrom $316 million to $11.9 billion. The 1982recession and the rise in the value of the dol-lar were clearly the major factors accountingfor the change in trade deficits with Canada.Canada’s share of the U.S. merchandisetrade deficit rose from its normal 5 or 10 per-cent to 25 percent in 1982. In absolute terms,the U.S. merchandise trade deficit withCanada increased over 4000 percent, peak-ing at $15 billion in 1985. The leading importfrom Canada in 1986 was passenger motorvehicles ($11.9 billion). Canada’s top five ex-ports to the U.S. consist of motor vehiclesand parts and wood products (table 14).

It could be very costly to Canada to reduceexports to the United States. Nearly four-fifths of Canada’s manufactured productsare sent here.

124Finding other markets to

ldapanese Steelmaker, Blasted,’The Economist, Janua~ 3, 1987.121”A Shopping Spree Starts Turning Japan Around, ’’Business Week, August 17, 1987, p. 50.lzKharles Smith, “Under Its Own Steam, ’’Far Eastern Economic Review, Feb. 4, 1988.

l-e, for example, John Burgess and Fred Hiatt, “Toyota Finds Ways to Hold Down Prices,’ The Washington Post, Feb. 16, 1988,t pdMarc Levinson, “More Bucks Out of the Maple haf?’’Dun’s Business Month, July 1986, p. 45.

8/14/2019 8829


72 qPaying the Bill

replace lost opportunities will be difficult,particularly if other U.S. trade partners aretrying to do the same thing. Past Canadianefforts to diversify exports, and reduce theheavy reliance on the United States, have

125 Moreover,failed. Canada has begun torun current account deficits: about $7 billionin 1987 down from a surplus of $2 billion in1984.

126 If exports to the United States are

curtailed, Canada’s trade deficit could in-crease, putting further downward pressureon an already low Canadian dollar and onCanadian living standards. Canadian un-

Table 14.--Major U.S. Imports From and Exports toCanada, 1986

Compound

C.i.f. vaIue* annual

(millions of growth ratedollars) 1982-86

Import categoryPassenger motor vehiclesParts of motor vehicles, nspf**Paper and paperboard (not cut)Wood (shaped or simply workedSpecial purpose motor vehiclesCrude petroleum

Gas (natural and manufactured)Gold (nonmonetary, except ores)

Export category

$11.9. . 4.9. 4 5. . 3.1

3.1. 2 9

. 2.52.4

19.4%21.3

8.116.24.57,2

-15.121.5

Parts of r&d-vehicles and tractors 63 7.6

Passenger motor vehicles . 5.9 25.2General merchandise, low-value . . . 3.2 41.4Internal combustion & piston engines 1.8 3.0Trucks & special purpose motors 1.7 4 1 6Parts for office and ADP machines 1.2 17.1Gold (nonmonetary, except ores) 1.1 14.2

Coal and lignite . . . . . . . . . . . 0.7 -8.2

q C. i. f value of imports includes cost, insurance and freight**Not Specified

SOURCE: U S Department of Commerce, International Trade Administration, Of-fice of Trade and Investment Analysis, 1986 U.S. Foreign Trade HIgh-

lights, Much 1987

employment has been higher than that of theUnited States and many other industrialisedcountries in the 1980s, though it had declinedto 7.8 percent in early 1988. This comparesto the current rate of 5.6 percent in the

United States, 2.7 percent in Japan, and his-torical rates in Canada of 3 to 6 percent inthe 1960s and early 1970s.

127

In another break with the past, the UnitedStates is running large deficits with WesternEuropean countries. Our merchandise tradewith Western Europe fell from a surplus of $20 billion in 1980 to a deficit of $27 billionin 1987. Over one-half of the Europeandeficit – $15.3 billion – was with West Ger-

many; Italy accounted for $5.5 billion, or 20percent, and the United Kingdom for $3.4billion. Again, as with Japan and Canada,the largest import item from WesternEurope is passenger motor vehicles –$11.7billion in 1986–with West Germany themajor supplier. Motor vehicle imports dwarf the next most important European import,organic chemicals (table 15).

The cost of adjustments will vary among

European countries. Unemployment is highin France, the United Kingdom, West Ger-many and Italy, relative to historical stand-ards. The worst off is France with anunemployment rate of nearly 11 percent inearly 1988. The United Kingdom is recover-ing from a prolonged bout of unemploymentat around 12 percent; the rate is currently 9percent and declining. West Germany’s un-employment rate, over 7 percent, is lower,but high by historical standards; the rate

throughout most of the 1960s and early 1970s

l~Alan M. Rugman, “U.S. Protectionism and Canadian Trade Policy, ’’Journal of World Trade I~w, July-August, 1986.nization for Economic Co-Operation and Development, OECD Economic Outlook, (Paris: O~,CD-Publications, May 1986 and

Yu;~W , p. 58.tmU.S.-~psrtment of Labor, Bureau of Labor Statistics.

8/14/2019 8829



was less than 1 percent. In Italy the currentunemployment rate of about 7 percent, ismore than double the levels of the 1960s and1970s. In terms of trade balance, Germany’smerchandise trade surplus —exceeding $20

billion in the mid-1980s–puts that countryin better shape to handle a diminishingAmerican export market than Italy or theUnited Kingdom, both of which were indeficit in 1983.

128These deficits were small:

Table 15.– Major U.S. Imports From and Exports toWestern Europe, 1986

CompoundC.i.f. value* annual(millions of growth rate

dollars) 1982-86

Import categoryPassenger motor vehicles . . . . . . . . $11.7Organic and related chemicals . . . . 2.8B e v e r a g e s , a l c o h o l i c . , 2.7A i r a n d s p a c e c r a f t , e t c . 2.6Crude petroleum . . . . . . 2.3Motor vehicle parts, nspf** ... 2.2Special transactions, nspf** 2.2Gold (nonmonetary except ores) 2,2Petroleum products . . . . . 21Specialized industrial machinery 1.9

Export categoryAir and spacecraf t , e tc . . , 5.1Office and ADP machine parts ., ., 4.0ADP machines 4.0Measuring and checking instruments 2.3Internal combustion engines ., . . . 2.2Oilseed and oleag. fruit . . . . . 2.2Coal and lignite ... . . . . . . . . . . 1.8

23.7%12.05.1

17.4-23723,2

7.190.09.0

24.9

12.714,28 92.68.9

-14.5-10,0

q C I f value of imports includes cost, insurance and freightq *Not specified

SOURCE: U S Department of Commerce, International Trade Administration, Of-fice of Trade and Investment Analysis, 1986 U.S. Foreign Trade High-

lights, March 1987

$1.7 billion for Italy, and $0.8 for the UnitedKingdom. Nonetheless, since both countriesran substantial merchandise trade surpluseswith the United States, any loss of U.S.markets would almost certainly meandeeper deficits, and downward pressure onliving standards and currency values.

Certain developing nations are importantsuppliers of imports to the United States andmajor factors in the U.S. merchandise tradedeficit. Those facing the largest adjustmentcosts are the East Asian newly industrializ-ing countries (NICs) — Taiwan, South Korea,Hong Kong, and Singapore –and two Latin

American NICs, Mexico and Brazil. Like thedeveloped nations, different developingcountries vary in their abilities to cope withthe adjustments.

About one-quarter of the U.S. merchan-dise trade deficit in 1987–$47.2 billion–was with Asian countries, excluding Japan;the four East Asian NICs accounted forthree-quarters of this. The deficit withTaiwan was much the largest: $17.4 billion,compared with $9.4 billion with the Republicof Korea, $5.6 billion with Hong Kong, and$2.1 billion with Singapore. Table 16 lists themost important imports from and exports tothe Asian NICs in 1986. If all the separatecategories of apparel and footwear are ag-gregated, this is by far the largest category of imports, amounting to at least $14 billion.

129

Apparel and footwear top the list of importsfrom three of the Asian NICs (Hong Kong,

12aU.S. Department of Commerce, International Trade Administration, International Eronomic Indicators, op. cit.lmTrade data are ublished in ways that make it difficult to sum u

?Iar e categories of imports, such as apparel, so this estimate is on l

approximate. Eight o the 35 leading Imports fmm the East Asian N]&f {in 986 were apparel and footwear; they added up to $13.8 billion. tis likely that more articles of apparel were imported, but were not among the leading 35. In addition, detailed data on imports and e

T

rts bycountxy and by region are pubhshed by the International Trade Administration (ITA), U.S. Department of Commerce, and are on a ifferentbasis from the more general trade figures published by the Department’s Bureau of Economic Analysis (BEA). In the BEA data, both importsand exports are reported on a free alongwde ship (f.a.s.) basis, which means the price of the item as it is loaded for shipment. In the ITS data,exports are f.a.s., but im rts are reported on a cost, insurance, and freight (c.i.f.) basis, which adds the cost of insurance and frei t to theoriginal cost of the item.% us, in the ITA accounts, imports appear to be greater than in the BEA accounts. Where possible, the B & figureshave been used, because they present imports and exports on the same basis. }Iowever, some of the detailed data are available only from ITA.

8/14/2019 8829



Table 16.-Major U.S. Imports From and Exports toEast Asian NICs, 1986


dollars) 1982-86

Import categoryFootwear (new, exe. military) $39Toys and baby carriages, etc. . 3.0Sweaters and other outerwear . . . 2.9Outerwear apparel, (cotton & wool) 2.6Office and ADP machine parts . . . 2.2Telecommunications

equipment, nspf** . . . . . . . . . . . . 2.2Electronic components and parts 1.7ADP machinery . . . . . . . . . . . . . . 1.4Furniture and parts . . . 1.3

Television receivers, etc. 1.2

Undergarments ., . . . . . . . . . 12Rubber and plastic articles . . 1.1

Export category

Electronic components and parts . 1.5A i r and spacecra f t , e t c . 1.2Office and ADP machine parts 0.9Organic chemicals and products 08Hides and skins (except fur) 06Oilseed and oleag. fruit . 0.6

Rubber, plastic and syn. resin 05ADP machines . . . 0.5Measuring and checking instruments 05Telecommunication equipment 04Corn or maize (unmilled) . . . . 0.4Pulp and waste paper . . 0.4Wheat (unmilled) ., . . . . . . 0.4

20.5%13.026.214.444.7

23.88.3

128,434.320.212,630.4

10,69.5

23.914,932,2

7.110.218.14.6

- 45-8.519.1

4.3

q C I f value of imports Includes cost, insurance and freight

q *Not specified

SOURCE: U S Department of Commerce, International Trade Administration, Of

fice of Trade and Investment Analysis, 1986 U.S. Foreign Trade High-

lights, March 1967

Korea, and Taiwan) and place in the top tenfrom Singapore. The fastest-growing im-

ports are passenger motor vehicles, whichrose from $10 million to $854 million in oneyear, 1985-86, and automatic data processingmachines (including computers and cal-culators), which increased from $53 million

to $1.4 billion in the 4 years 1982-86. Almostall automobile imports from the Asian NICsare from Korea, while automatic dataprocessing machinery exports are from allfour nations, with Taiwan accounting for 52percent, Korea for 26 percent, Singapore for16 percent, and Hong Kong for 6 percent.

In general, developing nations are less ableto cope with a diminishing American marketfor their exports than developed nations.

Even Taiwan, with a healthy and growingtrade surplus and massive reserves of foreignexchange, could have problems with adjust-ment.

130Like many other developing na-

tions, it is highly dependent on export-ledgrowth, particularly in exports to the UnitedStates. America is the market for half of Taiwan’s exports, and over half of Taiwan’sGNP depends on exports.

131

South Korea’s economy may be more vul-

nerable, as Korea is only just beginning toreverse chronic trade deficits, and is still

.

pursued a strategy of export-led growth,which is successful as long as exports are ableto expand fairly rapidly. In 1987, for example,Korean GNP rose 24 percent, pulled by a 36percent expansion in exports. Korea wasable to run a current account surplus in 1986,for the first time in modern history, mainly

130Carl Goldstein, “Economic Monitor Taiwan: Exports Hit New Peaks,’’ Far Eastern Economic Review, 9 April 1987, p. 137.lsl Robert G. Sutter, ‘Taiwan: Recent Developments and Their Implications for the United States, ’’Congressional Researeh Sexvice Issue

Brief IB87092, Updated June 16, 1987; information provided by the Cm-ordination Council for North American Affairs.ldavrence A. Veit, “Time of the New Asian Tigers, ’’Challenge, July-August 1987. Korea’s international debt totaled $45 billion in

1986 –45 percent of Korea’s $100 billion GNP.

8/14/2019 8829



due to its increasing import penetration of U.S. markets. Lower oil prices, the apprecia-tion of the yen, and falling interest rateshelped, but Korea’s $9.4 billion trade surpluswith the United States in 1987 offset its $5.2billion deficit with Japan. However, Korea’sdependence on exports could backfire whenthe U.S. merchandise trade deficit shrinks.Exports account for 40 percent of KoreanGNP, and 39 percent of Korea’s exports goto the United States – up from 30 percent adecade ago.

133Reducing the U.S. trade

deficit might cause political as well aseconomic trouble in Korea.

The U.S. merchandise trade deficit withLatin America was $12.2 billion in 1987,amounting to only 8 percent of the total. Thisrepresented a steep deterioration for theUnited States, however, with U.S. merchan-dise trade having descended from a surpluswith Latin America of $1.3 billion 1981.Most of the deficit is with just threecountries: Mexico ($5.7 billion), Brazil ($4.1billion), and Venezuela ($2.0 billion) asshown in table 17. In contrast to importsfrom Asian developing countries, ordeveloped countries, imports from LatinAmerica are tilted heavily towards naturalresource commodities: petroleum andagriculture and fishery products account for40 percent. However, imports of internalcombustion piston engines have been grow-ing very rapidly in the 1980s, from $362 mil-lion in 1982 to $1.1 billion in 1986, mostly anindicator of the importance of Mexicanproduction. The trade deficit with LatinAmerica peaked in 1984, declining sincethen as a result of both modest increases inexports and contractions in imports. A part

Table 17.– Major U.S. Imports From and Exports toLatin America, 1986


dollars) 1982-86

Import categoryCrude petroleum . . . . . . .,Petroleum products . . . . . . . . . . .Coffee . . . . . . . . . . . . . . . . . . . .,Fruits and nuts (prepared), nspf** .,Shellfish (fresh, frozen, salted) . .Internal combustion engines . . . . . . .Footwear (new, exe, military) .,Electrical distributing equipment .,Special transactions, nspf** .Motor vehicle parts, nspf**

Export CategoryRoad vehicles and tractor partsOrganic chemicals and productsA i r a n d s p a c e c r a f t , e t c .

Telecommunication equipment .,Civil engineer and contractorsGeneral merchandise, low valueRubber plastic and syn. resins .,Internal combustion engines . .Petroleum products (refined) .,Office and ADP machine partsElectronic components and partsElectrical appar. (current carrying)Wheat, unmilledA D P m a c h i n e sMeasuring and checking instrumentsE l e c t r i c a l m a c h i n e r ySpecialized industrial machineryElectrical distributing equipmentFertilizers and materials .,Misc. chemical products . ,

P a p e r a n d p a p e r b o a r dNon-electric parts, nspf**

$6.94 83 416121.1100 80 80 8

1,7110 9

100 90 90 80.80.80 70 70 60 50 50 50 50 50 40 40 4

0 40 4

-11 .9%-12.013.19.07 0

31.120.43360.6

284

0 84.13.1

9.213,828.4

3,24 6

13913,31506 5

1935 0

-2.70 02 3

2766 4

- 1 9

- 430 5

*C I f value of imports Includes cost, insurance and freight

* Not specified

SOURCE U S Department of Commerce, International Trade Administration, Of-

flee of Trade and Investment Analysis, 1986 U S Foreign Trade High

hights, March 1987

of the drop in value of imports is due to thesharp drop in oil prices in 1986.

Much of the deterioration in our merchan-

dise trade balance with Latin America in the1980s has to do with indebtedness. Brazil,the largest Latin American debtor, owednearly $107 billion to foreign creditors in

lxXhristopher Madison, “Korea: A New Interest, ’’National Journal, April 5, 1986; information provided by the Korean Embassy

8/14/2019 8829


1985; Brazil’s foreign debt was equal to 51percent of its GNP. Mexico owed $97 billion,with debt at over 58 percent of GNP;Venezuela’s debt was $32 billion, or 66 per-cent of GNP. Nations facing heavy interna-

tional debt burdens have been forced bytheir major creditors (the InternationalMonetary Fund and U. S., European, andJapanese banks) to devalue their currenciesand institute austerity programs —whichboost exports and curtail imports–beforetheir creditors would refinance their debts.So far, these countries have made littleprogress in reducing their debt levels; oneresult is a conflict between the needs of theUnited States to curb its merchandise trade

surplus and needs of Latin American debtorsto run trade surpluses to pay off their debts.Many proposals have been made to deal withthe Latin American debt crisis, but whateverthe outcome, the difficulties of managingthese debts are sure to intensify as U.S. tradedeficits fall.

The U.S. trade deficits of the 1980s not onlyenabled Americans to consume beyond thenation’s means –for the time being. They

also helped to fuel the economic growth of anumber of countries that based their growthon rising exports to the world’s largestmarket. Leaving aside the costs that wentalong with the benefits of America’s buyingspree (e.g., job losses of American factoryworkers), the situation cannot last. As notedearlier, foreign capital will not indefinitelymake up a widening difference betweenwhat we buy from other nations and what wesell. The burdens of the inevitable adjust-

ment, when it comes, will fall both onAmerican consumers and on foreign ex-porters. The adjustment might take a num-ber of different forms, some easier thanothers, but none painless.

International Companies

American companies with affiliates inother countries, and foreign companies withaffiliates in the United States, are importantplayers in international trade. Their trade ef-fects (at least through 1985, the last year forwhich data are available) are not entirelywhat might be expected. From 1982 to

1985 –years of large and growing nationaltrade deficits – all American companies withaffiliates abroad showed consistent mer-chandise trade surpluses of $3 billion to $5billion with their affiliates (see table 18).

134

Foreign affiliates of U.S. manufacturingcompanies accounted for 43 percent of thetotal $898 billion sales of the foreign af-filiates of all American companies in 1985.Trade surpluses of manufacturing parents

and their affiliates amounted to $11 billionto $15 billion in 1977 and 1982-85. For theU.S. parent companies overall, trade inpetroleum inflated imports, resulting in asmall deficit in 1977, and in the 1980sdiminishing to some extent the surpluses dueto trade between U.S.-based manufacturingcompanies and their affiliates.

Since World War II, many U.S.-ownedcompanies have engaged in large scale

IWle data in table 18 are for U.S. parent companies in the manufacturing indust V and foreign affiliates in which they hold a majorityinterest.

8/14/2019 8829



Table 18.–Balance of Merchandise Trade, U.S.Parent Companies and Majority-Owned

Foreign Affiliates, 1977 and 1982-85(billions of U.S. dollars)

All U.S. U.S. manufacturing

companies companies

1977 . . ., $-1.6 $12.31982 . . . . . . 5.8 14.71983 ., . . . . 3.6 11.21984 . . . . . . . . . . 3.4 11.71985 ... , ... 5.7 14.4

NOTE: Majority-owned foreign affiliates are those in which the U S parent com-pany holds a mqo~-ownership

SOURCES U S Department of Commerce, Bureau of Economic Analysis, U.S.

Direct Investment Abroad, 1977 (Alexandria, VA: National Technical

Information Service, April 1981 ) tables Ill T 1 and Ill T 4, U.S. Direct In -vestment Abroad: 1982 Survey (Washington, DC: U S

Government Printing Offfice, December 1985), tables Ill P 1 and Ill P 4,U.S. Direct Investment Abroad: Operations of Parent Companies andTheir Foreign Affiliates, 1983-85, available from the Bureau of Eco-nomic Analysis, tables 57 and 58

manufacturing operations overseas, mostlyin the developed world. About three-quarters of the sales by foreign manufactur-ing affiliates of U.S. companies are indeveloped countries — nearly 50 percent inEurope and another 20 percent in Canada. Itis generally accepted that the main reasonAmerican companies produce goods inEurope is to sell the goods there; and theseoperations are associated with tradesurpluses for the U.S. parent companies.

135

Some production by affiliates of U.S. com-panies in developing and newly industrializ-ing countries is for the same purpose, butanother important reason is to reduce costsof producing goods to be sold back in theUnited States and in other markets. Whilemanufacture by U.S.-affiliated companies in

less developed countries is still on a muchsmaller scale than activities in the industrial-ized world, there is evidence of change in

regional patterns. Manufacturing by and forAmerican companies in Korea, SoutheastAsia, and Mexico is growing fast–muchfaster than production by U.S. affiliates indeveloped countries. These are the mostfavored locations for going offshore to lowerlabor costs. And these operations aregenerally associated with trade deficits forthe U.S. parent company. However, theamounts involved were not yet large enoughin 1985 to make much of a dent in thesurpluses from operations in Europe andCanada. The large and growing nationaltrade deficits of the 1980s were reduced, notaggravated, by operations of U.S.-basedcompanies abroad.

The opposite has been true of foreign com-panies and their affiliates in the UnitedStates. The deficit arises chiefly fromwholesale trade — evidence that the mainreason for foreign companies to operate inthe United States is to sell here, just as U.S.companies operate in other industrializedcountries principally in order to sell there.Merchandise trade between foreign parentsand U.S. affiliates showed sizable and grow-ing deficits on the U.S. side, from 1977 to1985 (table 19). Most of the deficit is due tosales of foreign goods – cars, VCRs, compactdisk players– through local wholesale out-lets of foreign companies. For example, in1985,$45 billion of the $54 billion deficit as-sociated with trade between U.S. affiliatesand their foreign parents was in wholesaletrade –$22 billion in motor vehicles andequipment and another $21 billion in other

durable goods.

l~Data published by the U.S. Department of Commeree, Bureau of Economic Analysis on U.S. Direct Investment Abroad, Operationsof U.S. Parent Companies and Their Foreign Affiliates, show that nearly half of all sales by foreign manufacturing affiliates of U.S. companiesare in Europe; that most goods produced by these European affiliates are sold in Europe; and that exports from the U.S. parents to theirmanufacturing affiliates in Europe are substantially greater than imports.

8/14/2019 8829



Table 19.-Balance of Merchandise Trade, ForeignCompanies and U.S. Affiliates, 1977-85

(bi l l ions of U.S. dollars)

All U.S. U.S. manufacturingaffiliates affiliates

1977 . . . . . . . . . . . . . . . . . $ -19.21978 . . . . . . . . . . . . . . . . -22.71979 . . . . . . . . . . . . . -23.21980 . . . . . . . . . . . . . . . . . -26.01981 . . . . . . . . . . . . . . . . . -25.31982 . . . . . . . . . . . . . . . . . -26.91983 . . . . . . . . . . . . . . . . . -32.21984 . . . . . . . . . . . . . . . . . -43.41985 . . . . . . . . . . . . . . . . . -54.1

$ -3.1-4.2-6.1-5.2-5.1-4,6-6.1-7.78.4

NOTE: US affiliates are those in which a single foreign person owns or controls

directly or in directly a 10 percent or greater share

SOURCES: US Department of Commerce, Bureau of Economic Analysis, For-eign Direct Investment in United States, Operations of U.S. Affili-

ates, 1977-60, table G-3; Foreign Direct Investment in the United

States, 1980, table G-3, Foreign Direct Investment in the United States, Opereations of U.S. Affiliates of Foreign Companies, 1981-85,table G-3, all available from the Bureau of Economic Analysts

U.S. manufacturing affiliates of foreignparents have also had persistent and growingdeficits in merchandise trade, but on a muchsmaller scale. Considering only trade be-tween the affiliates and their parents, thedeficit rose from $3.1 billion in 1977 to $8.4billion in 1985; if trade with unaffiliated for-eigners (on both the import and export sides)is added in, the deficits were smaller, risingfrom $2.1 billion to $5.6 billion. Thesedeficits are more or less comparable with thesurpluses associated with trade betweenforeign affiliates of U.S. manufacturing com-panies and their parents. One way that thesedeficits could arise in the affiliate’s countryis that the parent company exports parts andmaterials to its affiliate abroad, where more

value is added – but not enough to offset theimport of parts and materials. If the item issold in the affiliate’s country, then the salehelps the home country’s trade; if it is soldback in the parent country or in a third

country it has a negative effect on the homecountry’s trade balance.

The idea that foreign investment in theUnited States–specifically, investment inmanufacturing plants –will reduce mer-chandise imports very substantially is notnecessarily or always true. To the extentitems made in the foreign investor’s plantreplace imported goods, they do reduce im-ports, and improve the trade deficit. But if they replace goods made by a domesticmanufacturer, then they could increase im-ports and worsen the deficit.

The persistently low dollar may stimulateproduction of goods in foreign-owned plantsin the United States at the expense of im-ports. There was some evidence by mid-1988that higher prices, reflecting the high valueof the yen, was finally beginning to stem im-ports of Japanese cars. It would not besurprising if Honda, Toyota, and Mazdawere to switch as much production as pos-sible for the U.S. market (and possibly someproduction for other countries as well) totheir American plants. This would help toreduce the trade deficit--and the more U.S.suppliers replace Japanese suppliers, thegreater the effect.

8/14/2019 8829


Climbing Out: How To Reduce the Trade Deficit

It will be difficult to stop living beyond ourmeans. It will not be painless. But some wayswill be more painful than others.

First, the United States could reduce itsFederal budget deficit. Some deficit reduc-tion was achieved in 1987: the deficit fell to$150.4 billion, down from $221.2 billion in1986.

137 The Council on Economic Advisers

forecasts further reductions to less than $130se billion in 1989.

138 To reduce the deficitsubstantially requires either reduced outlaysor higher Federal revenues, and either mightalso reverse some other trends, including the

rising share of consumption in GNP. Highertaxes would also tend to restrict private in-vestment (with harmful repercussions on theperformance of manufacturers). Despitesome success in cutting the Federal deficit in1987, the deficit remains very large; furtherbudget cuts will be more difficult to ac-complish and raising taxes is unpopular. If there is an economic recession – which someanalysts are predicting for 1989 – it is likelythat the budget deficit will quickly bal-

loon. 139

Curtailing the growth of consumption andincreasing personal saving (which has fallento record low levels)

140means living less

well, for most Americans. If investment weredampened, efforts to improve competitive-

ness could be thwarted. Moreover, someforeign governments have been reluctant tospur economic growth and risk inflation, and

their disinclination is magnified byAmerica’s failure to make more progress indeficit reduction. In short, even reversing thereversible could prove elusive.

Added to this is the fact that there are somethings that the United States cannot affectdirectly, if at all. The prime example is theimproved manufacturing and export perfor-mance of other nations. The improved per-formance of many nations results both from

improved manufacturing productivity andquality, and from national industrial policiesand a world trade regime designed to stimu-late development. It would be foolish to ex-pect foreign companies to stop learning howto improve productivity and quality inmanufacturing; it is foolish to expect foreigngovernments to stop promoting their owneconomic development and exports. Wemight be able to impose barriers to the con-tinued access of foreign producers to the

U.S. market, but we can hardly expect ourtrading partners to accept such handicapswillingly. Moreover, while there may becases where such barriers are prudent, awholesale resort to trade barriers to improveour trade performance could be ruinous, and

136An excellent discussion of some of the choices confronting the United States and its tradinw

rtners can be found in I,ester C. Tlurowand Laura D’Andrea Tyson, “The Economic Black Hole, ’’Foreqq Policy, Summer 1987; and in arris, op. cit.

lwllonornic Re rt of the President, Transmitted to the Congress February 1988 (Washington, D. C.: U. S. Government Printing Office,

rFebruary 1988), p. 37.13aThis, of course, assumes that there will be no recession through 1989.~wFor example, the Federal budget deficit more than doubled during the 1982-3 recession, increasing from $79 billion in 1981 to $208

billion in 1983. Source, Economic Report of the President, op. cit.I@The personal saving rate, which fluctuated between 6 and 10

E 1?rcent throu out the 1960s and 1970s (dipping below 6 percent in only

5 quartem since 1%2) fell, b 1987, to between 3 and 4 percent. . urce:1’

U.S. epartment of Commerce, Bureau of Elconomic Analysis,Business Conditions Digest, anuaxy 1988, p. 83.

79

8/14/2019 8829



could undo the progress made in the postwarperiod toward more open international trad-ing.

The path of relying on further currency ad-

justments to achieve trade balance couldcause considerable pain. It is already ap-parent, both from research and from the per-formance of U.S. imports and exports, thatthe dollar is far from low enough to bring thecurrent account back to historical levels(within a range of $10 billion or so, deficit orsurplus). The dollar’s value, as of February1988, had fallen nearly 37 percent from itspeak in the second quarter of 1985, on atrade-weighted basis.

141 This has helped to

bring U.S. monthly merchandise tradedeficits down to $10-14 billion in the firstfour months of 1988, largely by stimulatingexports. It has only recently begun to reducemanufactured imports, although someproduct s — notably, Japanese motorvehicles — have begun to show lagging salesand rising inventoried, a result of price in-creases induced by the change in dollar-yenvalues.

Bringing the dollar down further wouldmean that more imports would move beyondthe means of more Americans. By the end of 1987, the unit value index for all U.S.manufactured imports had risen 12 percentfrom its 1985 level (table 20).

142Items that

are major purchases for most households –notably, motor vehicles — have becomemuch more expensive, rising 30 percent inprice above 1985 levels by 1987 (figure 29).While consumers might be expected to

switch to domestically produced vehicles,they apparently have not: importedautomobiles were expected to account fornearly 30 percent of American sales in 1987,up from 28 percent in 1986.

143In part, this is

because domestic automakers raised pricestoo, sometimes in response to the rising costsof imported vehicles. Less costly items likeVCRs, televisions, and CDs have also be-come more expensive, although many of these items come from countries like Taiwanand Korea whose currencies have not risenvery much relative to the dollar (althoughthere is a great deal of pressure on these na-tions to raise their currency values). Theprices of imported office machines and

automatic data processing equipment in-creased 45 percent between the first quarterof 1985 and the last quarter of 1987, and im-ported telecommunications equipmentprices rose 8 percent. Imported sound andimage tape recorders and players (includingVCRs) went up by 16 percent. Interestingly,the prices of imported television receiversdropped 2.5 percent. This shows the ef-fects of substituting imports from Korea, andfrom other nations whose currencies have

risen less against the dollar, for more expen-sive exports from Japan.

Consumers are not the only ones to suffer“as import prices rise. Imported capital goodshave become more expensive too, and manyproducers are finding it more difficult to af-ford imported machinery and equipment asa result. Imported capital goods cost almost9 percent more in 1987 than in 1985. Theprice of imported textile industry machinery

ldl International Monetaxy Fund, International Financial Statistics, April 1988, count~ pages.l@Jnit Value Index numbem are from the Department of Commerce, Office of Tmde and Investment [email protected]. Department of Commerce, 1988 U.S. Industrial Outlook, (Washington, D. C.: U.S. Government Printing Office, Janua~ 1988), p.

38-3.

1441mport price data for office and ADP machines, telecommunications equipment, televisions, and sound and image tape recorders andplaym are from the Bureau of bbor Statistics,

8/14/2019 8829


Climbing Out: How To Reduce the Trade Deficit q 81

Table 20.-Average Prices, Imports to the United States (1977= 100)

1980 1981 1982 1983 1984 1985 1986 1987

All commodities . . . ............161.4 170,3 167.5 160.6 163.5 159.4 154.0 164.6

Manufactured goods . .. ..140.6 145.5 148.1 143.6

Consumer goods, non-food, non-auto 131.1 137,1 141.0 133.5Autos ., ., . . 139.8 163.5 177.2 186.5Autos from West Germany . . 160.8 178.2 198.9 225.6Autos from Canada . . . . . . .129.3 153.3 166.8 176.0Autos from Japan .. ..143.6 172.6 185.4 199.2

Nondurable goods . ......1410 1428 156.1 120.5Woven cotton fabrics ., ... ., 126.4 143.5 147.9 144.4cap i t a l goods 126.7 123.1 126,4 122.0Industrial supplies and materials 192.1 2066 195,3 180.6Textile machinery except weaving . . .2155 2182 204,9 220.2

148.0

134.8200.3220.2189.8220.1115.5156.0128.6178.1246.1

148.1

128.9206.0

215.1197,5223.2104.5144.4131.0167.2247.4

153.0130.1236.9200.8205.9279.5103,9140,2136.2129,5283,6

163,2

133.9264.2361.4222.6306.3

99.2163.0142.4143,1356.2

NOTE: Average prices for imported items are expressed as unit value indexes

SOURCE: United States Department of Commerce, Bureau of the Census, unpublished data

rose 44 percent over the same period.145

Be-cause most textile industry machinery is im-ported, textile makers are, in most cases,unable to switch to cheaper domestically-made machines. This could handicap the ef-forts of the industry to improve productquality, raise productivity, and compete withless expensive, imported textiles.

Another danger of relying only on further

devaluation of the dollar to reduce the tradedeficit is the risk of a severe recession. If wetake no other action to reduce imports, in-crease world demand, reduce the budgetdeficit, and raise exports, foreign govern-ments and private investors will force a solu-tion by curtailing investment in Americanassets and securities. If that happens, we facea period of rising real interest rates as con-sumers, investors and the U.S. treasury com-pete for an increasingly limited supply of

capital. Exchange rate markets would also bein turmoil, as the dollar, no longer supportedby foreign demands for dollars, declines fur-ther, and sharply. These developments could

Figure 29Average Price, Imported Motor

Vehicles and Parts

1 977=100

370

320

270

220

170

120 1 1 I 1 t 1 J

1980 1981 1062 1983 1084 1085 1066 1987

— All motor vehicles and parts —

Motor vehicles from Germany

Motor vehicles from Canada--- Motor vehicles from Japan

NOTE: Average Prices for imported commodities are reportedas Unit Value Indices,

SOURCE: U.S. Department of Commerce, Bureau of the Cen-sus, various years, unpublished data.

force the U.S. economy into a recession,which would almost certainly engulf othernations whose welfare depends substantiallyon the American economy. According to

l&50urcc: Department of Commerce, unpublished data on Unit Value Indexes for Imports.

8/14/2019 8829



some analysts, such a recession would in-clude increasing inflation as well as rising in-terest rates–both of which would depressconsumption (and thereby, standards of living). Rising interest rates also choke off

investment, which would hamper the abilityof American firms to improve their competi-tiveness. In short, a recession generated by acutoff in foreign capital inflows would be asetback to our efforts to balance trade inmanufacturing by means other than protec-tion from imports.

Finally, relying on currency adjustments forfurther improvement in trade performanceis undependable at best. Many businesses

that are beginning to boom as export ordersrise are reluctant to add capacity or make sig-nificant long-term investments in plant andequipment solely on the strength of a cur-rency-induced upturn, since currency adjust-ments are beyond the direct control of manufacturers. According to one article,manufacturers still are unconvinced of thedurability of the dollar’s drop, and eventhose that are reaching production limits arereluctant to expand capacity. Some com-

panies are even passing up export business,preferring to serve domestic customers in-stead as they push production closer tocapacity. Many manufacturers see thedollar’s fall as a windfall, offsetting its dis-astrous rise — much as farmers welcome rainafter a drought. Long-term improvement inour trade picture cannot be based on suchwindfalls. Sustained improvement must bebased on something more reliable: improvedcompetitiveness.

In terms of the macroeconomic adjust-ments, the least painful course would besteady and substantial progress by theUnited States in reducing governmentdeficits, reducing the growth of consumption

and increasing savings; more expansionarypolicies and stimulatation of demand inmajor developed nations; efforts to find waysfor developing nations to reduce their debtburdens and begin to open their markets.

This is a tall order. At best, such changeswill take years, and an extraordinary degreeof cooperation between nations. Butprogress must be made if we are to restoresome degree of predictability and stability to

international markets.

Besides these changes, prompt and com-prehensive efforts to improve U.S. manufac-tur ing performance are needed.Technology-broadly defined–has been asource of strength. Promoting development,acquisition and diffusion of new product andprocess technologies will help to improvecompetitiveness. Other actions that govern-ment might undertake include improving

education and training workers andmanagers in new skills, helping firms to ex-port, encouraging investment in produc-tivity-enhancing machinery and qualifiedpeople, and providing information about ef-fective ways of organizing production anddeveloping new markets. The governmentcould also evaluate how other policies en-courage longer term investments in productand process improvement. This is not acatalogue of government policy options to

146Alan Murray, “Aided by Wc.ak Dollar, Facto~ Output Lzads Economy Once Again, ’’Wall Street Journal, Jan. 26, 1988.1471bid.

8/14/2019 8829


foster improved manufacturing perfor-mance; it is only a short list drawn from pastwork.

148

The trade patterns of the 1980s are a sig-

nificant departure from any past experience,and the enormous current account deficits of the United States have changed economicrelationships throughout the world. Whilesome remedies will make more of a dif-ference than others, solutions that aim at

Climbing Out: How To Reduce the Trade Deficit q 83

only one area — such as promoting competi-tiveness or reducing federal spending or fur-ther devaluing the dollar or opening foreignmarkets – cannot achieve more than limitedsuccess. We must make progress in many

areas to overcome the trade deficit whileminimizing the impact on standards of living.Whatever solutions we adopt, we are in un-charted waters: we have never had suchproblems to solve before.