(Archived document, may contain errors)
321 January 13, 1984 THE MYTH OF AMERICA'S DECLINING MANUFACTURING SECTOR INTRODUCTION Advocates of national industrial policy, to varying degrees calk for extraordinary government intervention in the economy.
Just look at the major industr ial policy proposals: increased I government spending in virtually every area, protectionism legislatively enacted worker representation in management, greater governmental control over the allocation of credit, plant-closing laws, and various national ec o nomic planning schemes.l One of the prime "justifications" for these renewed pleas to expand government intervention turns out to be a myth--that the manufacturing sector of the economy, long a major source of j0b.s and products, is experiencing a precipi t ous decline. Greater governmental control over resource allocation is needed, so it is said, to cope with this major transition from an industrial to a service-oriented economy. Yet these claims appear to be based largely on spotty evidence and personal i m pressions. A closer look at the post-World War I1 growth of the U.S. manufacturing sector reveals that, contrary to the dire warnings of industrial policy enthusiasts, the sector is not declining, but is continuing to evolve along its historical path. The fundamental premise behind many industrial policy proposals, in other words, turns out to be a myth For an overview of the issues see Richard B. McKenzie, "National Industrial Policy: 275, July 12, 1983.
An Overview of the Debate Heritage Foundation Backgrounder No. IS MANUFACTURING IN DECLINE?
The message offered by industrial policy proponents is often aimed at middle-class parents to an expanded welfare state their children, upon growing up will find no middle-class standard of living awaiting them. As journalist Robert Kuttner recently warned It is that unless they acquiesce There is a good deal of evidence that job opportunities in the United States are polarizing, and that, as a result, the country's future as a middle class society is in jeopardy. W hat the decline of the middle class would mean to the country can only be guessed at, but it presumably would be unwelcome to the millions of parents who hope that their children can move up the economic ladder; to American business, which needs a middle c lass to consume products; and to everyone who is concerned about fairness and social harmony. As the economy shifts away from its traditional manufacturing base to high technology and service industries, the share of jobs providing a middle class standard of living is shrinking An industrial economy employs large numbers of relatively well-paid production workers A service economy, however, employs legions of key punchers, salesclerks, waiters, secretaries, and cashiers and the wages of these jobs tend to b e comparatively low.2 Barry Bluestone of Boston University and Bennett Harrison of MIT in their book. The Deindustrialization of America,3 echo this gloomy theme The pattern of wages in the old, mill-based economy looked just like a norma.1 bell curve. It had a few highly-paid at the top, a few low-wage jobs at the bottom, and plenty of jobs in the middle. But .in the new services economy the middle is missing.
And Harvard lawyer Robert B. Reich warns that because of the alleged dismantling of the manufact uring sector 'I a growing portion of America's work force is...locked into deadend employ ment.It4 The U.S. is becoming, says Reich, a nation of dishwashers janitors, and fastfood workers Robert Kuttner The Declining Middle Atlantic Monthly, July 1983, p. 60.. Kuttner's "solution" to this "problem" is the enactment of laws to promote unionization and increased numbers of public employees in federal state and local bureaucracies.
Barry-Bluestone and Bennett Harrison, The Deindustrialization of America Plant Closings, Community Abandonment, and the Dismantling of Basic Indus tries" (New York: Basic Books, 1982).
Robert B Reich, The Next American Frontier (New York: Times Books 1983), p. 202 3 Manufacturing Output These pronouncements are always made with muc h confidence, but it is difficult to see why. The data do not show at all that the manufacturing sector is in a state of precipitous decline, as the industrial policy advocates maintain.
Department of Commerce production indexes for manufacturing and othe r industries from 1960 to 1982.5 As the table indicates, in- dustrial production has risen steadily over the past several decades, with minor reductions during periods of recession, such as in 19
80. Looking back even further, the index of industrial prbduction increased from 45 in 1950 to 59 in 19
55. Given the. increase to 151 in 1981, this means that there has been about a threefold increase in the past three decades. The manufacture of durable goods exhibits the same pattern of steady growth in every category, from metals to Ilmiscellaneous If The one thing missing from the decline of the manufacturing sector,Il therefore, is any evidence of a decline. It might have been expected that the rate of growth of industrial production would decline during th e past decade--as it did--in light of the vastly expanded costs of regulation and the inflation of the 1970s. The latter increased the effective rates of corporate taxation and.contributed to reduced savings. But this is quite different from the bottom dro pping out of the manufacturing sector In fact, the index of industrial production grew by 24 percent from 1975 to 1980, which is greater than the 21.6 percent average five-year growth from 1950 to 19
80. The five-year period 1970-1975) showed only a 9 percent growth rate, but overall the past decade has not indicated a general trend toward decline Table 1 lists U.S The Service Economy Another common theme among industrial policy proponents is that the supposed decline of the manufacturing sector is being paralleled by an expansion of the Ilservice economy." The U.S. is rapidly becoming a service-oriented economy, it is said, but these service jobs are largely "dead end." Again, this claim is pure ficti on. Table 2 compares manufacturing and service output in constant 1972 dollars) as a percentage of GNP in selected years from 1950 to 19
81. It can be seen that the manufacturing sector provided almost the same proportion of GNP in 1981 24 percent) that it did in 1950 (25 percent), with only minor varia tions.year by year. Yet service output as a percentage of GNP increased by only two percentage points, to 13 percent, over a period of 31 years The data were obtained from U.S. Department of Commerce, Burea u of the Census, Statistical Abstact of the United States (Washington, D.C U.S Government Printing Office, 1982-83). 4 0 U P 2 rl I3 rl m rl C 0 rl 0 U P U rl U P I4 U m U m rl m V rl 4 0 U P I4 E 0 P A m aJ aJ m n rl m 1 E m E mod muhl hldd d 00 m d rl 5 C 0 E 0 a 2 a00 mmw hldd U 1 a 0 pI rl m rl LI U V d H s ahlr-hl00uo oooooom dddddd Nmhl moo ddd mooamu ammmm ahlo am0 d rn M C P .rl U u25 E( 0 m 5 Table 2 Manufacturing and Service Output (In Constant 1972 Dollars As Percentage of GNP: Selected Years, 19 5 0-1981 Sector 1950 55 60 65 70 75 77 78 79 80 81 Manufacturing 25% 25 23 25 24 24 25 18 25 24 24 Services 11% 10 11 11 12 12 12 12 12 13 13 Source: Calculated from U.S. Department of. Commerce, Bureau of the Census Statistical Abstact of the U.S Washingto n , D.C U.S. Government Printing Office, 1982-1983 Manufacturing Employment A further claim made by industrial policy proponents is that employment in the manufacturing sector has fallen sharply, forcing large numbers of workers into Ildead-endll jobs in th e service sector or onto welfare. Robert Kuttner, in particular, uses this picture of impending despair to make a case for preferential legislative treatment of labor unions and increased numbers of jobs in government bureaucracies. Yet this claim, too, is false.
The data.show a rise, not a decline, in manufacturing employment over the past several decades--from 16.7 million workers employed in 1960 to 20.7 million in 1970, and 21.8 million in 1981.6 If data on manufacturing employment during the 1981-1982 recession are singled out and compared to employment, say, one year or so before the recession, then of course a reduction can be detected but that is no indication of a general trend In sum, the claim that there is currently a large shift from the tradit i onal manufacturing base to the service industries is a myth. A further distortion of the truth appears in statements made about the types of jobs resulting from this nonexistent shift in the composition of the economy. The game is played as follows. To ex a ggerate the notion of !!dead-end labor, emphasis is placed on the absolute number of openings in such service jobs as janitors, salesclerks, and nurses' aides.7 But if there are already large numbers of such jobs--as there are-=a relatively large increase will simply reflect the overall growth of the economy and not a major shift.
An additional 100,000 secretarial jobs is not extraordinary, for instance, if there are already say, 2 million such jobs. On the other hand, it is often claimed that Ilhigh-techl l jobs are the wave of the future by relying on the high percentage changes in these job categories pointed out that, since only a fraction of all high-tech jobs are It is then Ibid.
Both Kuttner and Reich, among others, engage in these delusions. See Rob ert Samuelson Swing to High-Tech Jobs Appears to b,e Modest Washing ton Post, June 28, 1983, p. D-14. 6 well paying (engineers, computer programmers, etc the workforce is becoming "further polarized.Il But if 1 million (in a workforce of 102 million) high -tech jobs are added to a relatively small base of, say 1 million, then the percentage change will of course be large, but will represent no major shift in the economy.
As Robert Samuelson concluded after examining Labor Department data, the actual rearran gement of jobs in the economy is modest T].he changes don't indicate a drifting either toward lower-skilled or high-skilled jobs. Growth and shri.nkage are crudely offsetting in both high-paying and low-paying categories To see the shifts as triggering th e collapse of middle-class society requires a large leap in logic 8 EMPLOYMENT AND PRODUCTIVITY One thing the data do reveal is that employment in the manu facturing sector, although rising, appears to be rising more slowly than output. This implies that p r oductivity has risen so fewer workers are needed to produce a given level of manufactured goods. Industrial pol.icy advocates have used this observation in a typically exaggerated way) to reiterate the hoary, Luddite notion that productivity growth produc es unemployment--precisely the opposite of what historically is true.g Technological change will increase 'employment and income, just as it always has.
Examples go as far back as Adam Smithls Wealth of Nations and the famous story of the pin factory noted Smith, llcould scarce, perhaps, with his utmost industry make one pin a day but with the introduction of machinery, he could make nearly 5,000 pins per day A s'ingle worker without machinery The Luddite view was that this should have created massive clos e to 99 percent) unemployment in the industry, but it did not. The greatly reduced price of pins expanded their existing use, and new uses were invented. Moreover, the increased real incomes of the users of pins enabled them to purchase more of other thing s, stimulating production and employment in those areas, including the production of pin-making machines. This is how technological change enhances economic growth and employment in the economy--a process ignored by the %eo-Luddites.
This basic economic truth was also ignored over a century ago when the economy was shifting from a predominantly agrarian to an industrial base. The sharp decline in the percentage of a Ibid..
The Luddites were a group of British textile workers in the early 19th Century who opposed the introduction of machinery by destroying the machines.
Ludd of Sherwood Forest and thus became known as Luddites.
They issued proclamations in the name of the mythical King 7 the labor force employed in the agricultural sector did not create ma ssive unemployment or the disappearance of middle-class society; job opportunit$es expanded and technological advances in agriculture enabled America to produce more food with fewer workers tions in the new industries. Fortunately, most modern economists u nderstand that technological change is perhaps the main determi nant of economic growth it probably accounts for as much as 80 to 90 percent of economic growth in the United States.lo this estimate.ll Other workers were able to pursue more lucrative occup a Robert M. Solow of MIT estimated that Several other studies support In sum, the distorted arguments of some industrial policy enthusiasts stem from their reliance on a static view of the world=-one that fails to recognize how technology and automation al t er relative prices and incomes and generate economic growth and change. There may be transitional problems, such as temporary unemployment due to technological change, but the overall benefi cial effect has always occurred. And there already exist efforts to ease the plight of the temporarily unemployed. But the massive new federal programs proposed by industrial policy proponents can only slow or eliminate the underlying changes that bring about economic advancement. Government control over the allocation of capital, and plant-closing laws, for example, would delay or frustrate the type of changes that must be made in a growing economy competition and the forces of change They would inevitably be used to protect politically powerful businesses, unions, and regions from the rigors of Americans are rapidly adapting to the changing conditions of a .healthy modern economy, but this point is misunderstood or ignored by industrial policy advocates. Barry Bluestone, for example, talks of 38 million jobs that were I ldestroyedll in the 1970s. Since there were not nearly 38 million unemployed people in the 1970s, however, it is clear that this statistic reveals simply that about 38 million Americans changed their jobs as the economy continued to evolve and grow. Some i ndustries grew others did not. Some businesses succeeded; others failed. Jobs have not been Indeed, there are more people working now than ever before. As Nobel Laureate Friedrich Hayek has stated, the benefits received from competitive markets and econom ic growth are the results of such changes, and will be main tained only if the changes are allowed to continue.
But every change of this kind will hurt some organized lo l1 Robert M. Solow, "Technological Change and the Aggregate Production Func tion," Review of Economics and Statistics, August 1957, pp. 312-320.
For a survey of these analyses, see Morton Kamien and Nancy Schwartz Market Structure and Innovation (Cambridge, England: Cambridge University Press, 1981 interests; and the preservation of the ma rket order will therefore depend on those interests not being able to prevent what they dislike. All the time it is thus the interest of most that some be placed under the necessity of doing something they dislike (such as changing jobs and this general i n terest will be satisfied only if the principle is recognized that each has to submit to changes when circumstances determine that he is the one who is placed under such a necessity.l Thus, since many o.f the proposed industrial policies (parti cularly pla n t-closing laws and government credit allocation inhibit change, they only hinder, not help, economic growth and job creation BASIC INDUSTRIES Another claim of industrial policy proponents is that during the past decade "basic industriesi1 (primarily steel , autos, and textiles) have been in decline. Based on this proposition, it is often suggested that the entire manufacturing sector may be in structural decline--a claim already shown to be false. Moreover a close look at the data, and some common sense, re v eals two things: First, even though output and employment in steel autos, and textiles have declined in recent years, the decline has been neither large nor rapid; and second, the overall growth of the manufacturing sector demonstrates that the industrial base is changing its composition (less steel, fewer cars but not withering.
Employment trends from 1970 to 1981 in the steel, automobile and textile and apparel industries are shown in Table 3. The data reveal that during the recession years of 1979-1981 employ ment did fall in these industries. But before the steep 1979 recessio n (which was followed by a very short recovery) employment levels either changed very little or increased. Furthermore, the changes that have occurred in recent years have been small and very slow to materialize. For example, from 1980 to 1981, employ ment in the primary metals industries fell by only 1 percent; it fell by 0.5 percent in the auto industry; by 2 percent in textiles and by 1 percent in apparel products.
Given that other areas of the manufacturing sector are expanding, the only conclusions tha t can be drawn are that the manufacturing sector .is evolving and growing and that there are no calamitous disruptions in the Ilbasicll industries, other than the routine fluctuations of the business cycle. The term "basic l2 Friedrich Hayek, Law, Legisla t ion, and Liberty, vol. 3 Order of a Free People (Chicago: University of Chicago p. 94 The Political Press 19791, 9 industries" is, in fact, misleading because it implies that change in these industries is necessarily bad. Agriculture was once a I'basicll i ndustry, employing more than half of the working population, but the decline of employment there (mainly because of technological advances) made the nation better off, not worse off, for workers channeled their efforts into more productive endeavors. When certain industries are called ltbasicl1 or !lessen tial it is not long before there is a plea for protectionism or assistance, which does not serve the cause of economic growth It is the process of economic change, not any particular industry, that is bas i c, or essential, to a healthy economy Table' 3 Employment in "Basic" Industries: Selected Years, 1970-1981 Thousands Industry 1970 1975 1976 1977 1978 1979 1980 1981 Primary Metals Blast Furnace and Basic Steel Iron and Steel Foundaries Primary Nonferrous Metals Nonferrous Rolling and Drawing Nonferrous Foundries Motor Vehicles Textile Mill Products Apparel Products 1,260 1,139 1,190 1,179 1,213 1,254 1,142 1,121 627 548 543 553 560 571 512 505 229 230 218 231 236 241 209 201 I 72 66 84 65 70 73 71 70 213 1 81 194 199 209 220 211 206 83 76 79 89 93 100 90 90 799. 792 851 938 997 990 789 784 975 868 966 914 900 885 848 823 1,364 1,243 1,299 1,312 1,333 1,304 1,264 1,244 Source: U.S. Department of Commerce, Bureau of the .Census, Statistical Abstract of the U. S Washington, D.C U.S. Government Printing Office, 1983 CONCLUSION A major premise of the 'advocates of a Ifnational industrial policyll is that government action is needed because of problems created by an alleged shift from a manufacturing to a service o r iented economy. Yet the evidence shows that there has been no major shift from manufacturing to service. Manufacturing output as a percentage of real GNP is approximately the same as it was thirty years ago, and service output as a percentage of total out p ut has edged up just two percentage points--to 13 percent--in 10 thirty-one years. The data reveal that manufacturing employment is not in a state of sectoral decline, as some claim. In fact it has continued to expand, even though its composition may be c h.anging--as it always has. And technological change has continued to produce labor-saving (and cost-reducing) devices in the manu facturing sector. This has been a spur to economic growth and job creation.
The nation's industrial capacity is not declining , it is growing and changing market forces must be permitted to redirect resources to their uses of highest value. Industrial policies designed to insulate politically influential groups from the forces of economic change may grant the recipients valuable benefits. But=these are gained at the high cost of inflicting great harm on the rest of the economy In order for this growth to continue Prepared for The Heritage Foundation by Thomas J. DiLorenzo, Ph.D.
Deparment of Economics George Mason University