334 February 28, 1984 DISPLACED, WORKERS A ROLE FOR THE FEDERAL GOVERNMENT Richard B. McKenzie Senior Fellow INTRODUCTION National industrial policy enthusiast s continually raise the disquieting prospect that the U.S. is being transformed into a nation of hamburger cooks and computer jocks--that middle-skilled industrial workers will go the way of stagecoach drivers as their jobs are pulled from underneath them . Indeed, Bob Kuttner, a frequent economics writer for the Atlantic Monthly, warns that the] country's future as a middle class society is in jeopardy As the economy shifts away from its tradi tional manufacturing base to high-technology and service indust ries, the share of jobs providing a middle-class standard of living is shrinking.l Nothing could be further from the truth.

According to the scenario laid out by such economists, the growing "dualism in the labor force" (meaning many workers with very low or very high skills, and few workers with medium-grade skills) will result in traditionally well-paid production workers being forced to 1) accept low-skill and low-paying jobs 2 become wards of the welfare state, or (3) retrain for high-skill jobs at gre a t personal expense As a result, continues this This paper is a revised and expanded version of a paper presente&:at a conference on "The Displaced Worker Problem: Implications for Tfaining and Educational Institutions," sponsored by the National Institute of Education and the National Center for Research in Vocational Education Washington D.C September 12-13, 1983 Bob Kuttner, "The Declining Middle," Atlantic Monthly, July 1983, p. 60 0 2 scenario: U.S. competitive capitalism will lose the large middle cla s s needed to consume its products; workers will be easily exploited by their employers, since many of the Ifnew jobsg1 will be nonunion; the nation will gradually become peopled with the Itpermanently displacedgg; and the employed citizenry will have to sh oulder ever greater tax burdens to finance retraining programs.

Before accepting this element of the industrial policy argu ment and jumping onto the retraining bandwagon, Congress should reevaluate carefully the economics of federally funded retraining programs--not of programs for the hard-core unemployed, but those for the hardworking, reasonably well paid Americans who find themselves in what are tho.ught to be declining industries. The central issue in such a reevaluation should be the wisdom of further federal intrusion into the retraining process of workers cern e d about people who suffer during recessions and industrial transformation, especially when that suffering is the inescapable by-product of an anti-inflation policy-as during the recent recession. Where proponents and opponents of national industrial polic y generally disagree is over the appropriate solution to this distress of the market process, proponents tend to assert that social justice can be pursued directly through legislation'and more federal spending As a general proposition, opponents contend su c h a policy course will most likely achieve the opposite--more injustice and social mischief Both supporters and opponents of industrial policy are con Being long on emotion and short on appreciation This second view stems, in-part, from an analysis of the data on displaced workers, but even more from the conceptual arguments underpinning the current political drive for more federal taxes and dollars to retrain displaced workers. For the debate, at its core, is not about numbers.

It concerns what to do abou t displaced workers, and probably would have been joined even if the number of displaced workers were half what it is, or if federal funding of retraining programs were doubled or quad rupled. The debate is fundamentally about the assignment of responsibi lity for relief of social ills, and the escalating tendency of people to seek public, rather than private, relief for observed social ills. In short, it is about how much can government really be expected to accomplish in relieving social problems.

The pro blem of displaced workers, of course, does not stand The political agenda is crowded with competing claims on Any argument for government alone. the limited revenues of government action that does not recognize the competing claims, and the limited capaci t y of government to do good, is .hardly worthy of sober consideration. Interest groups; nonetheless, continue to proliferate'in Washington and to press the federal government to spend more money on their causes attention, these groups fail to acknowledge t h at merely citing a social ill is no justification for a claim on federal government By narrowly focusing their 3 revenue the context of competing claims, and each interest group must demonstrate that its claim is more worthy than the claims of others The case for federal action ultimately must be made in THE EMPIRICAL CASE The case for more federal aovernment involvement in worker training and retraining programi can be attacked on both empirical and conceptual grounds.

The Structure of Employment As proof of the decaying job structure in the U.S statistics are usually cited on the hundreds of thousands of jobs expected for salesclerks, cashiers, janitors, typists, and stockhandlers during the 1980s. Kuttner, for instance, claims that According to the most recent comprehensive forecasts of the Bureau of Labor Statistics, released in 1982, the economy will generate some 19 million new jobs between 1980 and 1990, about 3:s million of which will be professional and technical Low-wage service and clerical work w ill account for almost 7 million new jobs.2 The presumed decay of the skilled sector is further illus trated, contends Kuttner, by the list of job openings in Table 1 reprinted from his article. Since so many menial jobs are being created, the reader is i nvited to conclude that America must be turning into a low-skilled workforce.

Boston College economics Professor Barry Bluestone, on whom Kuttner relies for scholarly support, compounds the confusion over the future job structure by asserting that A dramat ic transformation in the structure of the entire national job distribution is responsible for an extreme mismatch between the skills and income needs of displaced workers and the skill requirements and wage levels of the new, jobs Ibid., p. 62 Barry Blues t one, "Industrial Dislocation and the Implications for Public Policy a paper prepared for the Third Annual Policy Forum on Employa bility, Development entitled, "The Displaced Workers Problem: Implica tions for Training and Educational Institutions," spons o red by the National Institute of Education and the National Center for Research in Vocational Education, Washington, D.C., September 12-13, 1983 4 Table 1 Job Openings in 1980 Job Category Retail Clerks Managers and administrators (not elsewhere classifie d Cas hie rs Secretaries (not elsewhere classified Waiters and waitresses Cooks (except private household Stockhandlers Janitors and sextons Bookkeepers Miscellaneous clerical workers Nursing aids and orderlies Child-care workers, private household Buildin g -interior cleaners Typists Truck drivers Openings, 1980 757,750 711,793 617,973 599,216 465,628 437,341 358,393 333,309 304,789 299,940 284,332 277,525 295,528 250,276 245,377 Source: Bob Kuttner, "The Declining Middle," Atlantic Monthly, July 1983 p. 62 Circumstances are so llextreme,ll according to Bluestone, that only government expenditures on worker training can pull the U.S back from the brink of economic disaster.

This line of argument, however, is based on a warped percep- tion of reality. Advocate s of greater government expenditures, such as Kuttner and Bluestone, fail to point out that the hundreds of thousands of new jobs for clerks and janitors will be added to an already large workforce of clerks and janitors, meaning that the percentage chang e in these categories of workers will not be particularly significant-=no more future Americans, in other words, will be forced to become clerks and janitors than are now. These writers also fail to point out that during the last two 4 5 decades manufactur i ng employment in the U.S. has remained more or less level, after adjusting for the normal cyclical swings in the economy, at about 20 million. And they ignore the forecast, made by the Bureau of Labor Statistics, that manufacturing employment will grow-no t decline-at an average annual rate of 0.8 percent.

Table 2 shows that the structure of the U.S. labor force in 1990 will likely be little different from what it was in 1980 Granted, nearly 3 million manufacturing jobs were tllosttt during the recent reces sion. But as Brookings Institution Senior Fellow Robert Lawrence has argued, dips in manufacturing jobs historically Table. 2 The Distribution of the Work Force, 1979 and 1990 Average annual Actual Projected 1979-1990 1979 1990 I I I Change Distribution D i stribution TOTAL EMPLOYMENT GENERAL GOVERNMENT Federal Military Civilian State and Local Education Noneducation TOTAL PRIVATE Agriculture Nonagriculture Mining Construction Manufacturing Durable Goods Nondurable Goods Transportation Communications Public U tilities Wholesale and Retail Trade Finance, Insurance and Other Services Government Enterprise Private Households Real Estate 1.4 0 6 0.3 0.0 0.7 0.7 0.5 1.8 1.5 2.3 1.6 1.5 0.5 0.8 1.0 0.3 1.1 1.7 2.8 2.7 1.8 0.1 100.0 15.9 4.1 2.0 2.0 11.8 6.4 5.4 84.1 2.7 81.4 0.7 5.8 20.6 12.5 8.1 5.3 21.5 5;3 19.4 1.4 1.7 100.0 14.8 3.6 1.7 1.9 11.2 5.3 5.9 85.2 1.9 83.2 0.8 5.7 19.2 11.9 7.3 5.1 22.2 5.7 21.8 1.4 1.3 Source: U.S. Department of Labor, Bureau of Labor Statistics, Economic Pro jections to 1990, Bulleti n 2121 (Washington, D.C U.S. Government Printing Office, 1982 p. 31 6 have been more severe than the recessions they accompany.4 is every reason to believe that manufacturing employment will continue to rebound in the current recovery more rapidly than wil l general economic activity There The Permanently Displaced Worker Advocates of an expanded federal role in retraining claim that the problem of the permanently displaced worker with many years of service is threatening to stifle growth and prosperity No o n e can deny the hardship experienced by the displaced worker. But the number of unemployed workers in declining industries who have considerable job experience represents a tiny proportion of the labor force. The Congressional .Budget Office estimates that of the 8 million unemployed workers in January 1983, only 1.6 million (20 percent) were in declining industries, and that most of these were the victims of the recession, not structural change. Of the unemployed workers in the declining industries, only 2 4 0,000 had been out of work for 26 weeks or more, and only 60,000 of these, representing only five-hundredths of one percent of total unemployment, had ten or more years on the Job.5 Even if one defined a 'Idislocated worker" to include workers in "de- cli ning industriestt who had been unemployed for eight or more weeks in 1980, the number of dislocated workers would still be raised only to 412,0

00. This would mean about 3 percent Of the labor force in the declining industries, or less than 0.4 percent of the total labor force.6 but it is hardly overwhelming. The displaced worker problem exists THE CONCEPTUAL CASE The enthusiasm for expanding federal aid to displaced workers should be tempered for four important conceptual reasons, in addition to empirical evidence Robert Z. Lawrence, "Changes in U.S. Industrial Structure: The Roll of Global, Secular Trends and Transitory Cycles a paper prepared for the symposium on Industrial Change and Public Policy, organized by the Fed eral Reserve Bank of Kansas City, Jackson Hole, Wyoming, August 25-26 19

83. See also Charles Shultz, "Industrial Policy: A Dissent," Brookings Review, October 1983.

Congressional Budget Office as reported in A. F. Ehrban, "Grasping the New Unemployed," Fortune, May 16, 1983, p. 108.

See Marc Bendick, Jr., and Judith Radlinski Devine, "Workers Dislocated by Economic Change: Do They Need Federal Employment and Training Assis tance The Federal Interest in Employment and Training, seventh annual report (Washington, D.C 1981).

National Commission on Employment Policy 1 7 Money goes to firms, not workers.

Many backers of expanded retraining aid seem to harbor the notion that the workers would be the explicit targets and benefi ciaries of aid, whereas much of the assistance actually would be p ocketed by the firms that no longer had to cover the costs of training and retraining their own workers. Indeed, much industry support for federal retraining programs is founded upon the strictly business proposition that lobbying for training programs is likely to be profitable-that is, more profitable than alter native activities. These firms understand that, if more federal retraining monies are passed to the states and communities, the result will be enhanced competition among states and communities fo r industries-thanks to the federal government defraying train ing costs. More federal money thus would mean more competition among governments to subsidize industries 2. Retraininq expenditures lead to winners and losers expenditures on retraining would re s ult in an increase in the stock of retrained workers equal to the number of people who are retrained in retraining workers (and there are good reasons to assume otherwise nothing could be further from the truth. Many of the workers retrained under the aus p ices of a government program would have been retrained anyway by their firms, at company ex- pense, or would have sought retraining themselves, at their own expense Proponents of federal programs appear to believe that federal But even if such programs we r e completely effective In addition, many proponents of enhanced federal assistance clearly fail to comprehend the two-edged sword of government ex penditure policies. They ignore the negative effects on. employ ment of the federal taxes and deficits that a re necessary to finance the expenditures. Yet experience shows that higher taxes destroy jobs far more effectively than expenditures create them which means that tax money used for retraining workers in one sector of the economy is likely to create a need for more retrain ing in another.

Moreover, federal deficits crowd outi1 private investors from the bond markets. Therefore, if money for retraining dis placed workers were to be raised by federal borrowing rather than new taxes, the result would be more displaced workers in industries cr o wded out of the bond market by government credit demands. And additional deficit spending to fund expansive new government training programs would, through greater p.ressure on international exchange rates, destroy more jobs and exacerbate the country!s r e training needs. While the costs for a new federal retraining program might be a "drop in the bucket according to its advo cates, the road to $200 billion deficits has been paved with similar arguments made by virtually all interest groups 8 In order that Congress might weigh the balance of effects of a retraining program, its proponents should be required to sup port their recommendations with Ileconomic impact statements."

These would do three things a) provide estimates of the jobs saved and/or created i n identified industries by the policy change; (b) provide estimates for the jobs destroyed and/or jeopardized in other identified industries; and (c) provide compelling reasons why the jobs of the workers in the identified lflosing industriesII should be jeopardized for the benefit of the targeted workers 3. Federal retraining assistance fosters inequities.

Proponents of federal retraining of displaced workers argue that widely accepted ethical and social priorities dictate that more federal funds should b e allocated to retraining workers in sectors where jobs supposedly are being destroyed by new tech nology or foreign competition or by the economyfs presumed trans formation from manufacturing to service industries.

This appeal to justice and fairness is questionable, because many of the workers in such industries have enjoyed above average wages and fringe benefits. Indeed, the market wages of many of these workers mounted in part because of the growing risk that their jobs would be eliminated and that t h ey would have to retrain themselves. By covering the retraining costs for these workers with federal funds, the general public would be paying double for retraining--once through higher prices for products, reflecting the higher wages, and again through h i gher taxes. One must ques tion the reasoning of those who contend that workers who have earned far less over the years than the beneficiaries targeted for retraining and paid for their own retraining needs should be asked to help pay for the retraining of these higher paid workers.

Increased federal retraining assistance would tend to dis courage workers from remaining abreast of the opportunities available for self-improvement. The workers' incentive instead would be to negotiate noncompetitive wages--in the knowledge that, if their firms were forced out of business, they could fall back on government subsidized retraining. In short, federal retraining subsidies could give rise to worker dksplacement.

Similarly, additional federal subsidies for retraining would, on the margin, temper employers' interest in keeping their workers skills current, since the responsibility and cost of retraining workers would be shifted to taxpayers 4 the area of worker training Externalities do not justify retraining.

Much is made of how the marketplace supposedly Iffailsit in The argument put forward is that For details of the proposal, see Richard B. McKenzie, "Name Winners Losers of Industrial Policy," New York Times, October 31, 1983, p. 25 9 labor markets fail because of the .l1externalities1' to education.

That is to say, proponents of government retraining programs con tend that people other than the employer and employee benefit from retraining, and these benefits are not captured by the, pric- ing system.

Few human activities, of course, are totally devoid of such llexternalities.ll By the same token, training for the displaced worker is designed mainly to produce jobs and income for workers.

The National Commission on Employment Policy, for instance, stated clearly that "[elmployment and training programs are intended to raise the earnings of those who participate.Il8 So the alleged market failuret1 argument should be treated with some suspicion.

The alleged barriers that prevent most benefits of retraining pro gram s from being captured through the pricing system are by no means obvious, and such barriers must be present to make the ex ternality case for retraining programs Even if externalities could be shown to exist, federally financed retraining would still not necessarily be justified.

The externalities must be shown to be widespread and sufficient to justify the cost of government action. And negative as well as positive externalities exist. Some workers, for instance might be harmed by the retraining of others , since the retrained workers might compete for their jobs others might feel that the retrained workers were the beneficiaries of welfare, which was simply unfair retraining of others should be compensated. would greatly increase the cost of retraining sa r ily imply that there would be more training undertaken than if the service were provided pri~ately Proponents of federal re training programs invoke the idea of externalities, yet they do not wish to give everyone's preferences equal treatment in their co nceptual framework. Their argument, in other words, becomes largely a set of assertions, not logically deduced conclusions.

If both negative and positive externalities were considered it could well be that the current level of federal funding may have alre ady surpassed the amounts required to achieve the elusive goal of economic efficiency Consistency implies that the people harmed by the And such compensation The collectivization of retraining, therefore, does not neces Steven G. Cecchetti, Daniel H. Saks , and Ronald B. Warren, Jr Employ ment and Training Policy and the.Nationa1 Economy," The Federal Interest in Employment and Training, seventh annual report (Washington, D.C National Commission on Employment Policy, 1981 p. 12.

For further details on this argument, see Richard B. McKenzie, "The Con struction of the Demand for Public Goods and the Theory of Income Redis tribution," Public Choice 1981 pp. 337-344.

EXTERNALITY ARGUMENT 10 EVALUATED Those using the externality argum ent to support federal pro This is because the subsidies and grams should also realize that pressing for additional federal funding can be self-defeating incentive effects of federal retraining programs can create exter nalities of their own. By failing t o retrain themselves, for instance, workers can impose a heavier tax burden on the rest of the taxpaying population. From the perspective of market failure economics, there is no guarantee that, on balance, social effici ency will be enhanced by retraining programs.

The case for reliance on markets in establishing the retrain ing needs is, in fact, also a case for minimizing externalities.

By holding workers and firms responsible for their own retraining needs, the market imposes the costs on those who ben efit from the retraining. When the responsibility for keeping workers' skills current is transferred from the worker to the community, however the potential for externalizing the costs of retraining is ex panded. So government funding is likely to increas e-not reduce externalities.

The externality argument has been taken to remarkable lengths by proponents of retraining training is not provided for unemployed workers, they will be forced onto the welfare roles. Hence, the argument goes, a federal retrainin g program would reduce the externality of taxes collected to pay for welfare expenditures These advocates contend that if re The argument has an appealing ring, especially in the absence of empirical evidence. It should be understood, however, that it tur ns not on some presumed market failure, but on a failure of government policy--and on an admission of the tremendous disin centive effects of welfare.

There are other reasons why this broad externality argument First, it implies that the cost of welfare pr ograms is suspect should be viewed much more broadly than as just the funds that are spent specifically on welfare. The total cost should include expenditures on a host of other programs designed to prevent people from going on welfare. So retraining cost s should be included in the calculation. By the logic of the externality argument, there fore, expenditures of retraining do not Irsave1I on welfare costs.

Those who use this revised externality argument are clearly not willing to generalize it in this way are not articulating a broad and logical principle for policy formulation, but rather a highly selective rationalization for the scale of importance they choose to apply to externalities To that extent, they A second reason for suspicion regarding the ex ternality argu ment is the assumption that any additional federal expenditures on retraining will supplement, not reduce, the welfare budget.

Part of the cost of training may, in fact, come from money that 11 would otherwise be spent on welfare services so there might be transfers from lower-income people, who would have received the welfare benefits, to higher-income people who would receive re training benefits CONCLUSION The arguments that expanded federal spending to retrain dis placed workers would ne t overall social benefits are unconvincing.

The efficiency of such a proposal is doubtful because it appears that the market is already well equipped to handle the nation's retraining needs. Wages do adjust upward to reflect the training costs incurred by workers To say that retraining costs are llhighI* is not to say that the market cannot adjust, especially in the case of non-poor workers.

The ethics of many of the current proposals should also be questioned. The retraining proposals before Congress have every characteristic of a political con game. In the name of the un fortunate poor (who probably will obtain little benefit from the programs), political support is being generated for a program that will effectively' transfer income, through the guise o f the educational system, from lower-income to higher-income workers from workers who have kept their wages competitive to workers who have not. Legislators must resist the pressure on them to make the taxpayer pick up the tab for yet another program that sounds very worthy in principle, but would turn out to be very different in practice Richard B. McKenzie is a professor of economics at Clemson University.