summaries

4

LAURA D ANDREA TYSON

LAURA D'ANDREA TYSON is associate pro^ssor in economics at the University of California, Berkeley, and a research associate of the Berkeley Roundtable on the Intemational Economy. She has served as a consultant for several organizations, including the International Bank for Reconstruction and Development, the congressional O£Rce of Technology Assessment, Wharton Economic Forecasting Associ- ates, the Rand Corporation, the President's Commission on Indus- trial Competitiveness, and the Council on Competitiveness. She was assistant professor in economics at Princeton University from 1974 to 1977 and has published extensively on international economic issues.

The Meaning of Competitiveness DI ٧ |

后 ing concera over American competitiveness. Indeed, competi- 인 tiveness, a concept that did not even exist in national policy K discussions until the last few years, has become a buzzword, r Business, labor, education, and goverament kaders speak of the competitiveness challenge confronting the United States ع & and offer a potpouιτi of sometimes convicting policy solu-

။ Competitiveness: An Analysis 好 of the Problem and £ a Perspective on Future Policy

96 LAURA D'ANDREA TYSON COMPETITIVENESS： PROBLEM & POLICY 97

lions, initiatives in such diverse areas as trade legislation, edu- cational reform, and taxes are defended or criticized on the basis of their likely effects on U.S. competitiveness.

Like most buzzwords, competitiveness has symbolic signifi∙ cance. It draws national attention to the unassailable fact that the position of the United States in the world economy is weakening. America can no longer rest comfortably in the belief that it will continue to be the premiere economic power in the world. Although still the largest and one of the richest economies, the United States has lost position compared to many countries with which it competes in world trade. To some extent, of course, this was inevitable. At the end of World War IL as Paul Kennedy points out in the Atlantic Monthly, the U.S. share of world manufacturing output was close to one-half：—a share never beCre attained by a single nation. As many developed countries rebuilt from war destruc∙ tion and as many developing countries introduced ambitious devdopment programs, some catching up was inevitable. But the pace and extent of the catch-up—or to put it differently, the pace and extent of the decline in the U.S. position—were not inevitable. They were inRuenced by actions at home and abroad. And, significantly for the future, there is no inevitabil- ity to a continued decline in the U.S. position, especially at the rate observed in recent years.

f If competitiveness is to have more than symbolic signifi∣ cance, if it is to become a reliabk guide £or policy by public and private actors, it must be properly de£ned. The need for an appropriate definition is evident in current policy debates. Some peop!e define competitiveness as a trade problem. The trade deficit is an indicator of competitive difficulties» and trade policy changes, even ones that threaten to reduce our access to foreign markets and our standard of living, are the solution. Others define competitiveness as a macroeconomic problem. Low U.S. saving rates and huge fiscal deficits have

। kept the U.S. cost of capital relatively high, and U.S producers ٧have not been able غ0و1٦：۴ع compeütive inv：tment rates required for improvements in product and ；cess techn٠l٥٠ ；s：1；e Mätien lies in； reformulaüon of غ mon2ry an；

Productivity, Technology, and Competitiveness The critica! determinants of competitiveness as defined here √

are productivity improvements and technological innovation. The key to the ability of US producers to offer competitive prices while increasing rather than cutting real wages for U.S. workers is productivity growth. When inputs become more productive, ests decline and producers can pay higher wages and charge competitive prices without sacrificing profits，Be- hind productivity improvements, in turn, lie both human and technological forces. Managerial and labor skills are the foun- dation for productivity performance—the better trained and more flexible the work force, other things being equal, the

fiscal policies to get interest rates down and national saving and investment up. Still others de£ne the competitiveness probkm as a labor cost problem، U.S. firms cannot compete because U.S. wages are too high, and the solution is a decline in US wages and perhaps, incidentally, continued erosion in union influence.

Γ For me Durpose ofdiscu：sion her^ competinveness is de ید ∕fined^ it ：;：in th^rep：s；;：fthePr;side：t'S〇C;m：：sionc〇; 시 Industrial ؟ompetidveness and more recently by the^Councü on٦〇mpet：tiveness ؛١ i ״；ƒdee٢ree：〇؛h1ch a nation, under free^andafai1r ن： įditions, pr¦duces 1goodΓand sen׳ices ；：：』：：；♦三~：Ođntematæna；markets while ٦imultane؛

ously maintaining and expanding the real incomes or its citi־ ге”； ?。,״ ：，：］：［«：： P。：： :《,, <tenm<i^n. Firil, أ٠ ：［：：”：,：：：： ：*：：; ：Dde ; ၊।<^national таг^и, ^<h ?*：：%：：：：：;»：,。；& ；：, :, ؟ ش^^ل;।aled d٢।ine in real »^ک, and M√lΛoui : contwuXd de-：：：״：；would d：：；। ול he dط ٧شرالم ］٥٢ In دم

¿»:¿w ;^^^į ئت۴ ;：,^ذ:^اغؤ ^ة1,,ق«^ةث ：: 1^^11 ”।ίιγ щ compele ni ħL)ι∣d ope：；；；： ك।；ائ

］^^；।^^：；：：?!^¡；¿^^：՛¿^^¿，；《^^!¿^^o ̂, ^^^^^^^；^；^^：；í^í^^： ：：^^^^|^^^o ̂j٠^؛¿^^^^^^^；^^ س^ad^m^^^^^ competitiveness.

COMPETITIVENESS： PROBLEM & POLICYLAURA D’ANDREA TYSON98 99

greater its productivity. And the more rapid the diffusion of technological improvements in production processes, the faster will productivity improve. Innovative new machinery and new materials are both driving forces behind productivity growth.

Technological improvements also directly affect the com- petitiveness of US producers in product markets. Improve- ments in product quality or serviceability and product innova- tions are often the source of competitive success in international markets. For products that compete on such fea- tures, including many of the high-technology products that account fbr about two-Rflths of U.S. trade in manufactures, technological improvements in process or product design are signi£cant inHuences on competitive outcomes.

In productivity performance and technological innovation, there are clear signs that the United States has been losing ground relative to its competitors. For the period from 1960 to 1985, U.S. productiv"y growth was the bwest among the advanced industrial countries by a substantial margin. In manufacturing, U.S. productivity growth picked up sharply in the economic recovery beginning in 1982-83 but still fell short of rates achieved by most of the other developed countries and many developing ones as well. As a result of more than twenty- five years of relatively poor productivity growth, the substan- tial advantage in productivity levels enjoyed by US producers has narrowed dramatically and in some instances disappeared altogether. In several sectors, including steel, autos, machine tools, and semiconductors, productivity levels in several com- petitor countries now equal or exceed U.S. levels. The erosion in the relative U.S. position has been concentrated in manure turing and has been particularly dramatic vis-a-vis Germany and Japan. When the United States had an apparently over- whelming advantage in productivKy, the pace at which that advantage declined did not seem of concern. Now that the advantage has narrowed, the long-term differential in the pace of productivity growth is of paramount importance.

There are also several indicators of a relative decline in U.S. technological capabilities in recent years. Research and devel-

opment (R&D) spending is the principal mechanism fbr gener- ating commercial innovations. In the United States, according to Martin Baily's article in the October 1986 issue of 5⅛z^, real industry-funded R&D spending rose by 6.2 percent per year from 1960 to 1969, but by only 2.4 percent per year between 1969 and 1977. Significantly, the slowdown in R&D spending was associated with a slowdown in industrial produc- tivity, suggesting that a slower pace of innovation was behind weakening productivis performance. R&D spending as a per- cent of sales by U.S. firms has increased fiom the low level of the 1970s, and U.S. industríal productivity growth also has picked up, once again suggesting a link between innovation and productivity. But the growth in R&D spending by Japa- nese and German firms since the late 197os has been even faster than that by American firms. At the national level, non- defense R&D spending as a percentage of gross national prod- uct (GNP) has grown more rapidly in Japan and Germany than in the United States during the same time period.

In the United States there has been a dramatic increase in real federal spending on defense related R&D while real fed- eral spending on nondeSense R&D has fallen as a percentage of GNP since 1979. Defense oriented research cannot replace civilian R&D. Past evidence suggests that America's reliance on defense as a technological engine can hurt rather than help competitive outcomes in commercial applications. In comput- erized machine tools, fbr example, U.S. suppliers concen- trated on unusual aerospace and defense applications, thereby losing a dominant share in commercEl markets in such cHtical areas as robotics. Many observers now argue that civilian and defense needs are becoming even more divergent and that the current boom in defense research is diverting resources away from industrial pursuits.

Lower rates of civilian R&D spending in the United States have been associated with continued erosion in the U.S. patent position. The number of patents granted to U.S. inventors is an indicator of the How of new commercial technology. The number of patents granted to U.S. applicants peaked in 1972 and declined substantially thereafter. During the same period.

,ي .־

٠।： !بم ئ

： ：أي

س ي. خذق ؤا

।ج٠ ب

ي> ؛^

٠ح٠٠ ।ى।׳آل

ر؛ ئ. حح خب ب

٢ ت :ذ

； ٦ء:ل־

ت. ٠٠

ث ׳ذ ٠

غئ --

ا !غك

حم :؛؛ غ ；٦لج

م^ ।

شى لمب ،< ذ

٨ س«

سس مفي

№ كد

၊ س

ةد دف

। ।ظ^

س جك

!^ د

।ك। ^ك

^ 0

د3̂

COMPETITIVENESS： PROBLEM & POLICY IQỈ

How РоНсу Has Weakened U.S. Competitiveness in Recent Years

Despite supply-side rhetoric, US policies during the last several years have failed to address the long-term competitive- ness difficulties of the United States. Investment rates, while rising from cyclically depressed levds in 1981-82, have re٠ mained low compared to our major competitors, and the U.S. saving rate, already low by international standards in the 】970s, fell to record lows in the 1980s∙ Most economists agree that at least in the short run, the tax law introduced in 1987 will depress investment and saving rates still further. The qual∣ ity of the U.S. labor force continues to suffer from long-term difficulties, resulting in higher illiteracy rates, poorer math and science training, and higher dropout rates than those reported in other advanced industrial economies. And despite the in- crease in civilian R&D spending noted above, the share of such spending in GNP remains lower in the United States than in Germany or Japan, as has been the case since the 1960s.

Indeed, our national macroeconomic policies, in the name of supply⅜side economics, have actually aggravated the com- petitive decline of the United States in recent years. The sharp appreciation of the dollar between 1980 and 1985, caused by the growing boιτowing needs of the federal government and low U٠S٠ saving rates, significantly weakened the already pre- carious competitive position of U.S. producers in world mar” kets. In short, our macro policy choices superimposed a short- term erosion in U.S. competitiveness on a more fundamental long-term erosion. The existence of the long-term erosion is one reason why, even though the dollar has fallen sharply since late 1985, wiping out most of the appredation against several major currencies, the U.S. trade imbalance has not improved.

A particularly disturbing aspect of the erosion in U.S. com- petitiveness, especially in recent years, has been its effects on the composition of^obs and wages in the U.S. economy. High-

LAURA D’ANDREA TYSON100

the share of U.S. patents held by U.S. residents also declined, from over 80 percent in 1965 to 55 percent in 1986. At the same time, the share of U.S. patents held by foreigners in- creased, with two-thirds of the increase awarded to Japanese naUonals.

At the industrial level, perhaps the most dramatic evidence of the erosion in U.S. technobgical leadership comes from the semiconductor industry. In 1977, the U.S، technological lead in semiconductors appeared unchallenged. But studies by the Defense Science Policy Board and the Berkeley Roundtable on the International Economy indicate that the Japanese have now achieved leadership in a range of critical areas, and in others they are fast gaining on their American competitors. There are warning signs of similar trends in the computer and telecommunications industries as well.

The apparent relative weakening in the U.S. technological position is particularly disturbing because it has come at a time of seemingly basic innovation in the machines and know-how that constitute manufacturing practice. In this period of tech- nobgical transition, made possible by the application of mi٠ croelectronics to producHon techniques through a range of industrial sectors, if US firms fail to design, adapt, and diffuse new technologies of production as rapidly and effectively as their foreign competitors, they will face even more difficult struggles in international markets in the future.

Fragmentary but suggestive evidence indicates that even when U.S. firms imroduce new, more Rexible automation sys- tems into their production processes, they encounter difficul- ties. For example, a study by Jay Jakimar in the Harvard Вгшпш Review (1986) compares the use of flexible manufacturing sys٠ tems (FMS) in Japan and the United States f6r comparable products. The study reveals that the number of parts made by an FMS in Japan is almost ten times greater than in the United States, while the rate of new product introduction is twenty- two times greater in Japan than in the United States. Jakimar concludes that with few exceptions the FMSs instated in the United States show an astonishing lack of Rexibility in use،

حم هة -ح ت عه

၊٠ه ع؛ :،ا

٠

x： べ三液

怨治チ 籍ノモ

๘،£ ร3

٠٠ โ٢٠

». 2

. س >ها .،م

٦ بق س־ ظخ ءتك

٢；٦ لئ ل؛ل جئ غتة ذغ غغ ه

COMPETITIVENESS： PROBLEM & POLICY 103

in product lines in world markets would not spill over into an erosion in related services، In other words, the health of the most valuable services the United States has to offer depends on the health of the manufacturing base that provides the major market £or such services.

During the last couple of years, mainly as a result of the intransigence of the trade imbalance in the face of the da matic fall in the dollar's value, concern over U.S. competitive- ness has intensified· Supply-side economics has been largely discredited, leaving a legacy of huge fiscal deceits without the anticipated boost in U.S. growth and productivity perform- ance. Confidence that market forces are pushing us to a brighter information-based service economy has weakenp replaced by growing anxiety that US. manufacturing ρroduc- ers can no longer compete effectively ،،on the shop Hoor." There is growing understanding that our huge foreign debt, reHecting several years of living beyond our means as a nation, represents an inescapabk foreign claim on our fUture goods and assets. And, as the technologies of production and com- municaticn change rapidly, there is recognition that the com- petitive dynamics of the world economy are in a period of transition where the positions of individual producers and nations in the world's hierarchy of e8nomic power and wealth are subject to rapid change. .

In these new circumstances, the United States must fashion a new set of economic policies if it is to retain its position as a m^or economic power, perhaps no longer ahead of the pack as it was earlier, but as first among equals. We need a strong national commitment to bolster our national competitiveness and policies, institutions, and values that reRect this commit- ment. In short, we are after a true "supply-side” economics that strengthens our national productive potential over the long run.

Luckily, as we seek to design a competitiveness agenda for the nation, there are lessons to be learned from other nations that have started out earlier than we on paths to build national competitiveness. Perhaps the most basic lesson from foreign experience is that the old distinction between free trade and

102 LAURA D'ANDREA TYSON

wage jobs in the United States are still disproportionately in the manufacturing sectors where the US competitive decline has been the most severe. Many studies, including a recent one by Tyson and Zysman, show that surging imports and f%ltering exports played an important role in the rapid destmction of manufacturing jobs after 1979. Since manufacturing jobs on average pay much higher wages than jobs in the rest of the economy, the loss of manufacturing jobs meant a loss in high- wage job opportunities for U.S. citizens. Thus, it is not surpris, ing to find that between 1979 and 1985, the average weekly wage of jobs lost—mainly in manufacturing—was $444, while the average weekly wage oFjobs gained—mainly in services- was $272. This pattern of job destruction and job creation is one of several factors contributing to the surge in income inequality in the United States since the late 1970S・

At the national level, we have consoled ourselves about the decline in manufacturing jobs by talking about a "market- driven” transition to an in£。rmation・based service economy. The problem with this line of reasoning is twofold: the pace of the transition has been accelerated by our eroding competi- tiveness in manufacturing; and many service activities, espe∙ cially the high-wage, high・skill services in finance, software, business consulting, engineering, and the like, are what Ste∣ phen S. Cohen andJohn Zysman call "tightly linked" to manu- facturing activities.

Where linkages between services and products are tight, services are often specElized for particular products or even for particular suppliers, and these services, together with the products they support or use, are best thought of as an inter- related system rather than as separate commodities. This is especially true for many high-technology products and their specialized support services. Under these circumstances, close Sequent communkation between producers of output and specialized services is common, and such communication in- volves more than just the exchange of price information and often involves a substantial element of risk on both sides that results in lθng-leι∙m contractual relationships. For tightly linked services, it is hard to imagine how a competitive erosion

١٠ ٥٠ ٠١ ٠

٠٠ ١٠ ٠٦

匸

.・ニ ご宀丹

口茨で

104 LAURA D'ANDREA TYSON

protection, which still holds sway in many US policy debates, is outmoded. Protection is not the only or even the most important way governments affect the national competitive- ness of producers over time. The experience of many nations, including the United States, shows that a whole array of gov- ernment policies, none of them necessarily protectionist in the usual sense of the term, can "create" or shape the competitive advantage of national producers over time.

Creating Advantage by Macroeconomic Policy

State policies can create competitiveness at both the macro- economic and the microeconomic levels. At the macroeco- nomic level, policies can affect competitiveness over time by influencing the quantity and quality of labor, capital, and tech- nology. An advantage in capital-intensive or technology∙inten∙ sive industries is not an immutable fact of nature, but the result of a host of inteιτelated e8nomyzide policies that affect the incentives to save, invest, acquire human capital, and innovate and diffuse technology. In each of these areas, US policy can be shaped to strengthen U.S. competitiveness over the long run.

U.S، saving and investment rates are persistently lower than those of our most successful competitors in both the devel- oped and the developing world. Despite widespread euphoria about the U.S. investment boom encouraged by tax cuts and economic recovery after 1982, the U.S. investment rate has remained below that of our major 8mpetitors throughout the 1980s, continuing a trend that characterizes the entire postwar period. As a result, the pace at which we supply our work force with modem equipment and technology has been slower, and this is rejected in our lower rate of growth of productivity. Nations with rapid productivity growth invest in plant and equipment at a much higher rate than nations with slow pro- ductiv^y growth. The United States has been a low-investment nation, and the effects on our productivity growth have ac־ cumulated over time.

COMPETITIVENESS： PROBLEM & POLICY 105

Since there is a high correlation between national saving and national investment, despite the greater international mobility of capital, the low US saving rate acts as a long-tem barrier to a higher national investment effort. In this respect, US saving behavior in the 1980s is disturbing. The gross private saving rate in the United States has remained well below the levels achieved by our major competitors, and the US. per- sonal saving rate has fallen to new lows. On the govemment side, the growing federal deficit has induced very high rates of government dissaving, despite the fact that state and local governments have generated substantial saving. The financing requirements of the govemment deficit and private invest- ment have exceeded the saving generated by the private sec- tor. The excess has been covered by the inRow of funds from the rest of the woHd, an inHow that equalled about 15 percent of U.S. nondomestic investment between 1983 and 1985, and about 40 percent of the 】985 increase in GNP. Without the inRow of fbreign funds, at unchanged rales of government dissaving, the result would have been much higher interest rates crowding out private investment to make room for gov- ernment borrowing needs.

A sustained reduction in the federal deficit is a necessary condition for improving national competitiveness over the long run. Without a fundamental change in our macroeco∙ nomic situation, the United States will continue to run massive trade and current ac8unt imbalances. Under these circum∙ stances, any possible salutary effects of competitiveness poll cies on U.S. trade performance will continue to be swamped by national borrowing needs. To the extent that policy initia- tives, such as the proposed trade bill or a continued drop in the dollar's value, succeed in improving the trade imbalance and stemming the inHow of foreign capital without a decline in the borrowing needs of the federal govemment, farther increases in the inHation rate and in interest rates can be expected.

Seen from this perspective, a necessary but by no means sufficient condition for an improvement in long-term competi- tiveness is a sustained, substantial reduction in the federal

歩

雄居X 主之二

磽コ匸 劣։つ

点モュ 冃マ=

士キ話 ンサ®

42 三=

稣~ 二=

キ -τ ..-

1∙ L∙

∙ 느亠 セ,目

二.：

106 LAURA D’ANDREA TYSON COMPETITIVENESS： PROBLEM & POLICY 107

cates that by curtailing individual retirement accounts (IRAs) and increasing the tax burden on capital income, the 1986 tax refom bill will probably discourage private saving as well.

If the debate about tax reform had focused on the need to restore US competitiveness over the long run, a very different tax bill might have emerged. The investment tax credit, rather than being eliminated, might have been strengthened and tar- geted toward particular activities or industries. The same might have been true for the tax credit on R&D expenditures. Reductions in personal income tax rates might have been ac- companied by the introduction of a value-added tax. The vir- tues of such a tax are that it falls on consumption, not saving, and that it can be rebated on sales to foreign markets under the General Agreement on Tariff and Trade (GATT) rules.

Government policy can also afEect .competitiveness at the economy-wide or macroeconomic level over time by inHuenc- ing the quality of labor resources. In a world where capital and technobgy are increasingly mobile, national competitiveness ultimately depends on the quality and productivity of national human resources. If the United States is to be able to compete in world markets with rising real incomes forits domestic work force, then this work fbrce must have the quality, Rexibility, and productivity required to make and sell products at prices competitive with those of similar products produced by lower paid workers elsewhere.

As a result of many well-known problems in our educational system, especially in grades κ through 12, there has been a relative erosion in the quality of our labor resources. This shows up in a variety of indicators: compared to its major fbreign competitors, the United States has more %nctional illiterates, graduates a smaller percentage of its population from high school, has a greater percentage of high school graduates uneducated in math, and produces a smaller per- centage of engineers among its college graduates. Only 75 percent of high school students in the United States graduate on time. Estimates of the number of functionally illiterate adults in the United States run to the tens of millions. Forty percent of the thirteen-year-olds in the United States are read-

budget de£cit. Without such a reduction, improvements in the trade balance are likely to occur at the expense ٠€ rather than in support o£ long-term competitiveness and are likely to be accompanied by worsening macro conditions at home and abroad. With competitiveness as the long-teιτn policy ol"ec- tive, not just any mechanism to reduce the fiscal imbalance will do. Tax increases that further reduce business incentives to invest and innovate, that reduce private incentives to save, or that increase the cost of labor will hurt U.S. competitiveness, as will expenditure cuts concentrated on education, R&D sup- port, worker dislocation assistance, and the like.

Given outstanding spending commitments on Social Secu- rity and defense and growing interest payments on the na- tional debt, the only way to reduce the deficit significantly without eliminating not only spending programs that promote competitiveness but almost all other discretionary spending programs as well is some kind of revenue-raising measure. In 1986 only $282 billion was spent by the federal government for all items other than Social Security, defense, and interest payments on the national debt, while the total budget deficit was $220 billion. Thus to eliminate the deficit solely by cutting these other expenditure items would necessitate the virtual elimination of government.

Unfortunately the 1986 tax refbrm does not address our long-term competitiveness needs. Because of transitionally higher tax rates in 1987, the new tax law was projected to reduce the federal deficit in 1987, but this effect is not ex- pected to persist in fUture years. The tax refbma was avowedly revenue-neutral at a time when the country desperately needed a revenue-raising tax bill. It also eliminated important incentives to invest. According to Lawrence Summers in the Harvard Business Review, the tax bill raises the effective tax rate on most classes of investment by about 20 percentage points and increases the rate of return required on a typical invest” ment project by 10 percent to 15 percent. In the long run. Summers estimates that this may well reduce the stock of plant and equipment by 10 percent to 15 percent and the economy's potential output by about 3 percent. And new evidence indi-

COMPETITIVENESS： PROBLEM & POLICY 109

One promising development at the federal level is the pro- posed program of training and assistance for displaced work- ers٠ including but not limited to workers displaced by foreign competition. In new, more competitive international condi- tions, and in the presence of major technolo^cal changes in production processes, Hexibility is the key to competitive SUC- cess. And Hexibility requires that workers be able to move quickly and easily from one task to another or one machine to another in the workplace, and from one workplace to another as market conditions change. As the Japanese and Swedish experience indicates, well-organized and well-financed pro- grams to train and relocate workers support the kind of Hexi- bility that is required, thereby enhancing the competitiveness of the overall economy.

Finally, government policy can a^ect competitiveness at the economy-wMe level by its effects on the pace and direction of technological innovation and diffusion. In the United States, the federal govemment fUnds about 47 percent of all R&D spending. Federal funds for R&D are overwhelmingly defense oriented. During the Reagan years, real federal spending on nondefense R&D as a percentage of GNP fell, and the gap between federal spending on defense R&D and nondefEnse R&D hit a twenty-year high. R&D de^nse spending has fbcused mainly on "mega-prqịects” like the strategic defense initiative (SDI) that are likely to have limited spillover eff⅛cts in commercial areas, at least in the short run, and that divert scienti£c and engineering talent from commercially oriented R&D programs. In contrast to the U.S. situation, in Japan, where govemment financing is a much smaller percentage of the total national R&D effort, it is expressly fbcused on com- mercial areas such as new materials, semiconductors, and bi- 。technology. The Japanese approach has been one of financ- ing programs to promote cooperative research on so-called generic technologies that seem to have promising commercial applications. These programs have been financed to a signifi- cant extent by the companies participating in the cooperative programs, leading to a shared burden of the R&D effort be- tween public and private actors. Joint financing and coopera-

108 LAURA D’ANDREA TYSON

ing below the skill levels for their age. A recent study of read- ing skills of seventeen-year-olds found that less than half of them performed at higher than basic or İnteı־mediate levels. A recent study of math students in several countries showed American twelfth graders well behind their Japanese counter- parts and students from many other countries. Even the besi U.S. students—those who leam calculus in the twelfth grade- performed poorly against Japanese students and were only on a par with other participating countries.

In the United States, much of the responsibility for develop- ing a more skilled, motivated work fbrce lies with the states and local communities that fUnd and oversee the nation's pub- lie schools and universities and with private employers that are responsible for fbrging better worker-management relations, more effective on-the^ob training programs, and better em٠ ployee incentives. That actions by both state and local com- munities on the one hand and private employers on the other can make a critical difference to worker quality and perform- ance is suggested by a variety of evidence. New educational inidatives in Tennessee, New Jersey, and Massachusetts, for example, have had notable successes in raising indicators of educational excellence. And as a growing number of examples of Japanese takeover and management of U.S. firms and work- ers indicate, substantial productivity improvements in U.S. labor performance can be realized by changes in workplace organization and manager-empbyee relations.

Growing interest in proposals to strengthen the nation's schools has encouraged a needed infusion of new mnds at the state level. From 1982 to 1987, support for education by state governments increased by 26 percent in real te։٠ms, although some of this increase went to make up for federal budget cuts. Adjusted for inHation, average teacher pay in 1987 was up 2 percent from its previous peak in 1972, while average wages in the rest of the economy remained lower than those realized in the early 1970s. At the federal level, the budget fbr the Department of Education has borne a dispr〇p。rti。nate share of cuts in order to preserve defense and Social Security and to meet outstanding interest obligations without tax increases. แ।

4， ジ与れ

® 1

^٠ حج ه

ج؛ ။اع

COMPETITIVENESS： PROBLEM & POLICY III

experience indicates that promotional schemes of this type can speed the introduction of critical technologies with beneficial effects for the competitiveness of both the using industries and the supplier industries.

Creating Advantage at the Microeconomic Level

In addition to using policy to create 8mpetitive advantage at the economy-wide level, government can use policy to сте- ate competitive advantage in individual industries or activities. Although this is a widely accepted idea and practice in many countries around the world, in the United States it is fraught with theoretical and political difficulties.

A large and growing part of world trade consists of ex- changes that do not reHect national diferences in resource endowments, even those resources whose quantity and quality are inHuenced by government policy. Instead, such trade re- Hects apparently arbitrary or temporary advantage resulting &om economies of scale, shifting positions in technological leadership, or from product differentiation and other f61־ms of nonprice competition. For trade in these products, it seems obvious that national policies can have an enduring effect on trade Hows and national welfare. This proposition has spawned a whole new area of trade theory among economists. The so-called new trade theory has concentrated on demon- strating that under certain conditions, national policies to pro- mote or protect domestic producers in international smpeti- tion can improve national welfare. There are essentially two different types of conditions that give rise to results of this kind. First, industries that are "imperfectly” competitive, most often as a result of economies of scale, eam higher retums than those available in the rest of the e8nomy. Under these enditions, national welfare may be improved by govemment policy to win larger market shares fbr domestic producers in world markets and hence a larger share of world profits f^r the domestic popubtion. A second set of conditions that provide a justification f6r welfare-improving policies draws on stan-

LAURA D’ANDREA TYSONIIO

tion make economic sense, since the social returns to such R&D efforts are likely to exceed the private returns appropria- ble by any single Rrm.

Concern over competitiveness has led to a rash of new proposals for federal programs to support commercial R&D effbrts in the United States. Examples of federal govemment programs to support generic research in commercial areas include National Science Foundation support fbr the express switching program and the national laboratory program in ceramics centered in the Oakville National Laboratory. Proposals to find a mechanism for the Department of Defense to help finance a major program o^oint R&D in semiconduc- tor manufacturing equipment and techniques are consistent with the need to find effective means for govemment to pro- mote generic research, the benefits of which will be shared by a large number of commerdal users. President Reagan called for the establishment of a number of new university-based science and technology centers for long-term research in criti- cal generic technology areas, like new-materials processing and robotics for automated manufacturing.

In addition to devoting more of its research budget to com- mercial R&D activity, the U.S. govemment should consider policies to speed the diffusion of new technologies. President Reagan and Congress made several proposals to improve the diffusion of technological information from a variety of gov- emmental agencies, including the federal laboratories and the Department of Defense. At the same time that better institu- tional mechanisms for the division of technological infoma- tion are designed, however, the incentive of those in the pri- vate sector to learn about and insrporate such information in their own production processes and products must be strengthened. R&D tax credit policy is an important determi- nant of this incentive as well as the incentive of the private sector to engage in its own R&D efforts. In addition to strengthening tax incentives for R&D investment by industry, the government should evaluate the efficacy and cost of pro- motional tax or credit schemes targeted at the more rapid diffusion of critical technologies, such as robotics. Japanese

*جهح: عا هء

ع٠ ٠٠

س ءام ىج آلي ،ج

٢ج٠ بج

11 جو بلم حيع عح عءل ةية ءر ع

⅜ ⅛ J⅜

ι ٠ثج

ًائ جخ

ذالب تئم

٠ سع

ز، ذع

־ ٠

٠؛تم ص

رر هذ

ذ و٦٠

٠٠ مذ

بم ।

جذ لمئ

ئ ٠:

٠7 |ب

٠٠ ٠

تئ ئذ

٠٠ ؛ب

,ج لح

؛ج بي

٦֊ذك٠٣ اح

ددة ءتغ

LAURA D’ANDREA TYSON COMPETITIVENESS： PROBLEM & POLICYI12 113

that both the extent and the strength of policy intervention have declined since the early 1970S・ Nonetheless, the long- term effects of earlier or continuing policy intervention on trade patterns in targeted sectors persist because of the long, term nature of the advantages created by such intervention،

A careR11 examination ofjapanese policy indicates that it has been what might be called "market conforming” or "market promotional” in its objectives، It has been designed to pro- mote or accelerate the development of sectors deemed to be both privately and socially profitable and to manage the de- cline of sectors deemed to be privately and socially unprofita∙ ble. In the United States there is some support of and eχperi- ence with policies to aid national producers as they adjust to competitive decline. In industries as diverse as apparel, steel, and automobiles, the United States has responded to severe import 8mpetition by a variety of protectionist measures de٠ signed I;o give the domestic industry time to restructure. Trade ∙ a^ustment assistance programs for workers who lose their jobs fr٠m import competition also have been part of the policy response at various times. In no case have protection and adjustment assistance been linked to 8nditions specifying the pace and direction of industry restructuring. This has been left to "market” forces, despite the fact that market conditions have been distorted by protectionist measures, and despite the fact that there is evidence in the business literature suggesting that the most productive, efRcient firms in a weakened industry may be the first to leave, weakening it still further. The U.S. approach stands in sharp contrast to the Japanese approach, where decline is managed by legally mandated recession and rationalization cartels in which government assistance to as- fected firms, workers, and communities is linked to a delated program for reducing capacity at the industry and often the firm level. The French, too, have introduced explicit restruc∙ turing programs for industries in decline, such as the steel indust^ in the 1970s.

Even those who might be willing to entertain the idea that industq-specific policies may be market conforming and beneficial when used to moderate the adjustment costs as-

dard notions of externally or spillover effects- In lay terms, certain industries may be more important than others because they generate benefits for the rest of the economy, and gov- ernment policy to promote them can improve welfare by sos־ tering these spillover effects. Under both sets of conditions, the industries involved are defined to be "strategic" either in the sense that resources employed by them earn higher re־ turns than they would earn elsewhere or in the sense that they generate special benefits for the rest of the economy.

Much of the empirical research on how government policy creates advantage at the micro level has £ocused on high- technology sectors and on Japan for case study evidence (see, for example, the works of the scholars at the Berkeley Round- table on the International Economy—BRIE). This research indicates that the Japanese government has used a host of interrelated policies, some aggregate, some industry specifİG some formal, some informal, to target certain industries and that these policies have had an enduring effect on Japanese trade patterns in speci£c sectors. Case study work by BRIE scholars has demonstrated the role of policy in the evolution of Japanese production and trade in the snsumer electronics industry, the semiconductor industry, and the telecommunica： tions industry. This work suggests that an interrelated set of protectionist and promotional policies has given Japanese producers the advantages of large-scale production and cum۴٠ iative production and research experience that have been criti٠ cal to their export successes in related products♦

The Japanese approach to these and other industries is motivated by the goal of guiding or influencing the structure or composition of the economy in specific directions♦ In pur٠ suit of this goal, the Japanese have used trade policies, such as tarifs, tangible and intangible nontari^ barriers, and relaxed policies, such as controls over direct foreign investment. But these kinds of policies have been used in conjunction with a variety of other policies, including tax and subsidy polic% Ո٠ nancial and interest rate policy, research and development policy, and antitrust policy. The actual mix of policies has changed significantly over time, and some observers argue

LAURA D'ANDREA TYSON114 COMPETITIVENESS： PROBLEM & POLICY 115

Paul Krugman coined the phrase "linkage extemality" to apply to this type of spillover effect resulting from increasing returns in the production of inputs and their effects on the costs of downstream producers. Such a notion underlies argu∣ ments for policies to promote infrastructure in the usual sense of transportaron and communication networks.

In recent years, Japanese market promoting policies have concentrated on high1technology sectors, such as semiconduc∙ tors, computers, and telecommunications. The Japanese view these industries, like steel and shipbuilding in the past, as providing infrastructure on which the future 8mpetitive sue- cess of a variety of sectors depends. A policy of promoting R&D, investment, and growth in these new ،٠infrastrucíuraľ٠ activities is viewed as generating beneficial effects throughout the economy. As the arguments above make clear, these indus- tries certainly satisfy some of the conditions required for a linkage-externality argument for market promoting policies. Seen from a narrow perspective, they provide inputs for pro- duction throughout a broad spectrum of the economy, and they enjoy both dynamic and static increasing returns. Indeed, increasing returns have been nothing short of spectacular in semiconductor production in the last decade, with spillover effects on increasing retums in related computer and telecom- munications equipment. Private increasing returns in this complex of industries, in turn, are the basis for social increas- ing returns throughout the economy, as the standard linkage- externality argument suggests.

Seen fiom a broad perspective, this group of industries pro- vides the foundation for a fundamental revolution in produc∙ tion and communications technologies that is transforming how work is done and how life is lived. The sρiUover effects of this revolution are likely to be so profound and so wide- spread that any attempt to define them precisely, much less to quantify them, at this point in time is an exerdse in false formalism. What seems certain is that such industries are "strategic” in the sense of providing both linkage-externalities in the form of spillover effects from R&D and innovation throughout the economy.

sociated with industrial decline are often unwilling to acknowl∙ edge that such policies may have beneficial effects when ap- plied in industries that are growing and profitable. If indus- tries or activities are profitable by market criteria, why might govemment policy to promote them be welfare improving? The answer to this question lies in the notion of strategic industries or strategic activities introduced earlier. If an ec٠٠ nomic activ^y is strategic in the sense that it eams higher retums than those available elsewhere in the economy or in the sense that it generates significant positive spillover effects in other parts of the economy, then market promoting policies to support such an activity can improve national economic welfare.

Japanese industrial policy seems to be motivated by such a perspective. For example, in the 1950s and 1960s steel and shipbuilding were promoted because they were believed to provide substantial spillover benefits in the fbrm ofinfrastruc־ ture. From a traditional economics perspective, the spillover elects provided by the steel and shipbuilding industries are "pecuniary” in the sense that they are rejected in lower input prices to their downstream users. Because pecuniar externali∙ ties are rejected in market prices, there is no need for govern- ment policy—markets left to themselves will provide the opti∙ mal amount of investment and production. This conclusion holds, however, only as long as there are no ،،impeģctions।) in product markets and no ،،distortions|> in capital markets. And these conditions are often at odds with economic reality.

Take, for example, the case of the Japanese steel industry in the 1950s∙ Because steel was an important inteιdiate input produced with economies of scale, policies to promote rapid expansion of high-quality domestic steel resulted in lower prices and reduced costs f^r steel-using industries. This en- couraged the expansion of these industries which in turn fed back into further expansion of the steel industry and still lower costs. This virtuous interdependence between the steel indus- try and downstream users gave rise to a true externality- increasing private retums in the steel industry, resulting in increasing social returns in the downstream user industries.

116 LAURA D’ANDREA TYSON COMPETITIVENESS： PROBLEM & POLICY แ7

Even if one discounts the linkage-externalities and trans- formative effects of these high-technology industries—a post tion that is misguided in its shortsightedness—standard exter- nality arguments about the retums to R&D and innovation provide a traditional case for market promoting policies. Economists generally focus externality arguments of this vari- ety on the issue of appropriability. As long as the returns to innovation and R&D are appropriable, there is no divergence between private retums and social returns and hence no ration- ale for policy intervention. Appropriability can be better un- derstood by examining different kinds of knowledge gener- ated by R&D and innovation. Three kinds of knowledge, all of which are present in the high-tech electronics industries, can be distinguished: production process knowkdge reflected in firm-specific leasing curves that can be internalized within a firm; product design knoMedge that can be reverse engi- neered, which, once generated, is available intemationally; and knowledge that spreads beyond the firm but not necessar- ily easily beyond national or sometimes even regional bounda- ries. This third kind of knowledge seems to be the reason behind the development of geographically concentrated "high-technology” centers, with information embodied in people and spread through social and academic networks.

The economic literature on innovation and diffusion has focused on the second kind of knowledge—knowledge that is only partly appropriable by the innovating Arm. For such knowledge, the evidence suggests overwhelmingly that the social retums to R&D and innovation are significantly greater than the private retums, and this is the most widely accepted rationale among economists for govemment support for R&D. The argument is particularly power&H for basic and generic R&D, which, by its nature, is likely to generate knowledge with benehts that extend beyond the innovating firm.

Until recently discussions of the spillover eifects ofR&D did not address the issue of the geographical concentration or dispersion of knowledge. Recently, however, both because of the apparent tendency of high-technobgy firms to cluster to- gether in distinct communities and because of concem about

the extent and pace at which technological knowledge diffuses across паНопаї boundaries, the issue has received considera- ble attention. Of particular interest to the question of how policy creates competitive advantage in trade is the idea that govemment support for R&D and innovation helps to create a national pool of innovative talent. The history of technologi- cal change in a variety of industries and nations indicates that technological change both supports and is supported by the creation of an ability to innovate embodied in a pool of special- ized knowledge and in a specialized labor force. By its nature, this ability is not easily contained within Arms or sectors but is much more easily contained within nadonal boundaries. A nation that promotes R&D and investment in its high~tech industries is encouraging the development of a highly skilled pool of innovative talent, which, in the long run, given the mobility of goods, technology, and capital across national bor- ders, may be the single most important factor on which na- tional competitive advantage rests.

Finally, if high-technology industries are strategic because of their linkage and knowledge extemalities, they are also strategic in the sense that they are characterized by imperfect competition. A variety of characteristics cause these industries to diverge sharply from the competitive model: significant leaming curve economies and the advantages they yield to early entrants; product heterogeneity based on changing tech- nological positions and changing product standards; the im- portance of marketing-distribution channels to market pene- tration; and implicit or explicit preferential or protectionist treatment for domestic producers in many national environ~ ments,.most notably the Japanese.

From the point of view of U.S. Arms and U.S. policy makers, the concentrated nature of these industries, the vertical link- ages among them, and the governments role in coordinating joint activities among them in Japan are cause for particular conceE. The Japanese computer industry consists of three large Erms (Fι"itsu, NEC, and Hitachi) that are also three of Japan's four largest semiconductor producers. These same firms control more than two-thirds of all telecommunications

COMPETITIVENESS： PROBLEM & POLICYLAURA D’ANDREA TYSON118 Z19

need a strategic commercial initiative that funds research in the most promising commercial directions.

equipment production in Japan and dominate domestic and global markets in many consumer electronics items. They are also tightly linked to the largest Japanese producers of semi- conductor capital equipment. Given this market structure, it is reasonable to conclude that even in the absence of promo- tional policy measures, U.S. firms as sellers would face signifi= cant barriers to market entry in Japan, and U.S. firms as buyers would be unable to purchase Kontier technology inputs from Japanese producers to compete with the same producers in downstream, higher value-added product markets.

The constellation of arguments indicating the "strategic- transfbmative” nature of high-technoSgy industries pevides a powerful prima facie case in support of market promoting policies in the United States. The evidence suggesting that the competitive strength of Japanese producers in these industries has been bolstered by such policies to the disadvantage of U.S. producers indicates that this case should be seated seriously in U.S. policy discussions of whether and how the United States should respond. It is not enough to argue glibly that closure of the Japanese market to U.S. firms simply worsens consumer welfare in Japan or thatjapanese promotional poli- cies simply lower prices to US. consumers, with a net welfare gain for the United States. These policy prescriptions apply only in a static, perfectly competitive world without a<ijust- ment cost and without externalities, a world that does not fit the high-technology industries.

We need policy prescriptions that reflect the strategic sig- nificance of high-technology industries and activities, such as semiconductors, computers, new materials, superconduc- tivity, and Motechnology, to our fhture economic well-being. An obvious starting point is the introduction of cooperative R&D programs, financed in part by the federal govemment, and in part by firms in a sew critical areas. The United States must redirect a significant fraction of its sederai R&D dollars to nondesense, commercial programs if our technology lead is to be sustained. Instead of a strategic desense inidative that ⅛nds research in a weapon system that will never be used, we

Conclusion

The U.S. economy confronts a crisis in its international com- petitive position. The staggering trade deficits of the last sev- eral years and the foreign borrowing that has financed them cannot be sustained over the long run. At some point during the next sew years, the U.S. trade position will have to adjust dramatically, with profound consequences fbr the kvel and composition of production and employment at home and abroad. Both domestic and foreign policy makers will have to grapple with ways to improve U.S. export performance and to reduce U.S. import dependence. Macroeconomic policy choices will play a critical role in these adjustments, as they did in the spectacular erosion in the U.S. trade deficit after 1982. Given the magnitude of the adjustment required and the un- derlying longer-term weaknesses in U.S. productivity per- fbrmance, however, reHance on macroeconomic policy alone will prove risky. Reliance on exchange rate corrections, with or without a deceleraron in U.S. growth rates compared to growth rates abroad, runs the risk of higher inflation rates, recessionary conditions, or more likely a combination of both, in the United States and in the rest of the world.

The adjustment process can be made easier by policies to improve the underlying competitiveness of U.S. producers. Higher productivity growth, better quality products, and inno- vations in product and process technobgy, like exchange rate adjustments and a recession in domestic demand, can improve the U.S. trade imbalance but with a far lower cost in relative living standards and in fbrgone output and consumption in the United States. What is needed to make the required adjust- ment of the U.S. trade position less costly is the development of a national competitiveness policy. Such a policy would rep- resent the true realization of the supply-side otjectives of the

LAURA D'ANDREA TYSON120

late 1970s rather than the distortion of such objectives as a result of the “supply-side” fiscal policies of the last six years.

The process of strengthening the competitiveness of the U.S. economy is a long-term one. There are no quick Exes Just as erosion in the U.S. competitive position was a long-term phenomenon reflecting years of policy inattention and failure, any improvement from changes in policy will be realized only gradually. The danger is that policy makers, frustrated by the apparent lack of success of prudent policy measures in the short run, may adopt misguided policies of protection that promise to improve the trade balance quickly but are destruc∙ tive of real competitiveness in the long run٠

Finally, in designing a national competitiveness policy, we face three principal constraints. First, as a nation, we cannot compete in world markets by cutting real wages: this approach will not work because there are millions of foreign workers who are willing to work at wages lower than those we can pay. It also will not work because it will threaten the fbundations of our economic and political system. Second, a retreat to defensive protection will not serve as a long-term policy to support high wages and sustained productivity. Third, policies that are radically inequitable are unlikely to generate the broad political support required for a national commitment to long-term growth and innovation.

The opportunities we face are equally constrained. Ours is a world in which science and technology, capital, and manage- ment know-how are widely available. Consequently, we can only compete internationally in how effectively we develop, diffuse, and use technology and product and production know-how in our firms. Effectively using technical possibilities depends on management vision and worker skills. Simply put, in the long term, investment in science, education, technologi- cal innovation, and technological diffusion is all that will Sus- tain US.

٠ س

!ت ؛٠

।٠٠ ।

٠س ة٩ب

٠٠ ။։

يج ייי

ייי י