Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
7 Current Policies for Worker Adjustment In this chapter, we survey the structure and adequacy of the public and private policies that deal with the needs of entrants to the labor force and workers displaced by technological change. We include in this discussion a consideration of human resources policies, including training for en- trants to the labor market, job search assistance, training and retraining for workers who have been or may be displaced by technological change, and income support for those displaced workers. The primary emphasis of this chapter is on public policies; Chapter 6 discusses private sector policies that encourage cooperation between managers and labor. Be- cause of the complexity of the adjustment problem, however, we also address here the human resources policies that can be used by firms to deal with worker displacement policies such as advance notice of plant closings or large permanent layoffs, severance pay, and employer- provided training. There are important deficiencies in current policies in all of these areas. Many of our recommendations in Chapter 10 follow from this analysis. As noted in previous chapters, technological change will create unem- ployment in some occupations, industries, or regions, and it appears likely that it will increase the demands on the existing system through which workers retrain and acquire job-related skills. Indeed, the broader issue of worker training and skills requires attention. Where do workers currently receive their job-related training? How is the employer invest- ment in training allocated between blue-collar and white-collar employ- ees? Can the U.S. training system respond to surges in the demand for 137
138 TECHNOLOGYAND EMPLOYMENT new skills for example, can it train robot technicians or software engineers in response to increased demand for these specialists? Building on the discussion in Chapters 3 and 5, we also discuss briefly the skills needed by labor force entrants and the ability of training institutions to provide them. Finally, we consider displaced workers and the public and private programs for dealing with displacement. JOB-RELATED TRAINING Some evidence from other nations supports the hypothesis that job- related training contributes to international competitiveness, although these data are qualitative rather than quantitative. Japan, Sweden, and West Germany each have unique systems for job-related skills training, and all of them appear to provide higher levels of such training to the blue-collar work force than does the U.S. system. These nations also appear to be adopting some advanced manufacturing technologies more rapidly than the United States (see Chapter 21. There is no direct evidence of a link between these phenomena, nor do we have internationally comparable data on investments in job-related skills training within the United States and other countries. Nonetheless, the work of Nelson et al. (1967) and Bartel and Lichtenberg (1987), discussed in Chapter 2, suggests that such investments may play a role in the more rapid adoption of new technologies within these foreign economies. Where Do Workers Receive Their Training? In the United States, occupational training is provided by a large, decentralized "system" that defies neat description. A. P. Carnevale of the American Society for Training and Development has explored some of the system's proportions; as he states in a 1986 article in the Training Detailed descriptions of the national training systems in these countries can be found in such sources as the International Labour Office (1985), the Organisation for Economic Co-operation and Development (1973), and the U.S. Congress's Office of Technology Assessment (1986b). The West German system relies heavily on apprenticeship programs for its labor force entrants; 50 percent of the individuals completing compulsory education entered such programs in 1977. The Japanese system, which since the 1950s has been characterized by long-term employment commitments to workers from large corporations, features heavy investment by such corporations in training employees through formal and informal means and job rotation. Public funding in Japan also supports upgrade training for employed workers. In Sweden, large public investments (retraining and labor adjustment programs account for 2-3 percent of Swedish GNP) in training and retraining of the employed work force are combined with an elaborate system of vocational and technical education for labor force entrants and the unemployed.
CURRENT POLICIES FOR WORKER ADJUSTMENT 139 and Development Journal: "Workplace training and development is roughly equivalent in size to the entire elementary, secondary, and higher education systems" (p. 184. Carnevale estimated that the annual costs of the "total learning enterprise"2 in the United States were over $450 billion in 1985; of this total, he estimated that about $210 billion (or 47 percent) goes for formal and informal employee training. The most important source of skill improvement training, which is acquired once a worker is employed, is the employer (Table 7-11. More than 70 percent of skill improvement courses ("Formal Company Programs" and "Informal On-the-Job Training," in Table 7-1) are conducted at a worker's place of employments (Carnevale and Gold- stein, 19834. The dominant role of employers in funding or providing such training also affects its distribution between white-collar and blue-collar workers; white-collar workers receive much more employer- provided training. College graduates are twice as likely as the average worker to receive employer-provided adult education, whereas those with less than a high school education are only one-fourth as likely to receive such education. Employer-provided adult education thus tends to increase any differences in the educational attainment of workers that are already present when they enter the labor force. These conclusions about the skewed distribution of employer-provided adult education were confirmed by Tierney (1983), who found that those workers with some graduate training were seven times as likely to have received employer-provided training as those with less than a high school education. This divergence also means that whites are almost twice as likely to receive employer-provided training as nonwhites. According to Tierney, "formal education and training programs appear to be directed toward those workers who already have substantial levels of educational achievement" (p. 161. Employees in large firms also receive more employer-provided train- ing than employees in small firms. Lusterman (1977) reports that firms with 1() non or more emr~lovees spent an average of $86 per worker on ~ 7 _ ~ -~ - _ ~ - - - - ~ 2Carnevale defines the '`enterprise" as comprising all public and private expenditures on elementary, secondary, and postsecondary education; informal and formal employee training; and government training for civilians. Although the U.S. military is a major investor in training in a wide range of basic and job-related skills (it spent $18 billion in fiscal year 1987 for formal training alone), the impact of this investment on the civilian work force appears to be small, as Marcus (1987) has noted. 3The duration of training differs in school-based and workplace programs. In 1983 over 70 percent of the workers in formal company programs trained for less than 12 weeks; two-thirds of the workers in school-based training (an unknown portion of which is employer financed) trained for over 12 weeks, with approximately 38 percent in programs lasting more than a year (Carey and Eck, 1985, Tables 32 and 41).
140 TECHNOLOGY AND EMPLOYMENT TABLE 7-1 Sources of Skill Improvement Training by Occupation (1983) Workers Formal Informal Who Took Company On-the-Job Occupational GroupTraining SchoolPrograms Training Other Executive, administra tive, and managerial5,098 (47) 1,916 (18)1,884 (17) 1,688 (16) 836 (8) Professional specialty7,802 (61) 4,352 (34)1,936 (15) 1,756 (14) 1,408 (11) Technicians and related support1,588 (52) 600 (20)550 (18) 585 (19) 166 (5) Sales3,578 (32) 769 (7)1,411 (13) 1,642 (15) 487 (4) Administrative support, including clerical5,152 (32) 1,547 (10)1,565 (10) 2,423 (15) 392 (2) Private household33 (3) 10 (1)7 (1) 14 (1) 10 (1) Service workers, ex cept private house hold3,151 (25) 814 (7)955 (8) 1,528 (12) 360 (3) Farming, forestry, and fishing500 (16) 164 (5)51 (2) 203 (7) 142 (5) Precision production, craft, and repair4,133 (35) 863 (7)1,654 (14) 1,860 (16) 353 (3) Machine operators, as semblers, and inspec tors1,639 (22) 228 (3)286 (4) 1,151 (16) 78 (1) Transportation and ma terial moving706 (18) 84 (2)235 (6) 376 (9) 50 (1) Handlers, equipment cleaners, helpers, and laborers520 (14) 57 (2)92 (2) 381 (10) 19 (<0 5) Total 33,901 (35) 11,404 (12)10,625 (11) 13,606 (14) 4,301 (4) NOTE: All employment figures are in thousands. Percentages, which are calculated on the basis of total employment in each occupational group, appear in parentheses. Many workers reported more than one source of training, so percentages may not sum to 100. An unknown fraction of skill improvement training in school is employer financed. SOURCE: Carey and Eck (1985). training annually. Smaller firms (500-999 employees) invested an aver- age of $27 per worker annually. Of particular relevance to this study is the fact that, according to Lillard and Tan (1986), reliance on employer-provided training is even greater in technologically dynamic industries. In sum, employers provide the bulk of skill improvement training in the United States, and such training tends to reinforce, rather than offset, inequalities in preemployment educational attainment within the work force. Inasmuch as technological change is likely to inflict the most serious dislocations (measured, for example, in terms of duration of
CURRENT POLICIES FOR WORKER ADJUSTMENT 141 unemployment) on the least educated workers, this tendency of employer-provided training to reinforce inequality in the educational attainment of employees is likely to exacerbate the severity of technolog- ical displacement for low-skill, poorly educated workers. Why do U.S. blue-collar workers (as used here, a category that includes white-collar female clerical employees) receive less employer- provided training than U.S. white-collar workers and less than blue-collar workers in some other industrial nations? One tentative explanation for U.S.-Japanese differences in such training investments focuses on the contrasting structures of U.S. and Japanese labor markets (Hashimoto, 1979; Hashimoto and Raisian, 1987~. U.S. markets for blue-collar labor tend to display relatively inflexible nominal wages, and the supply of and demand for labor are equated through cyclical layoffs. Japanese labor markets, on the other hand, historically have exhibited greater nominal wage flexibility. A large share of the annual earnings of Japanese production workers consists of bonuses, the amount of which is tied to corporate performance, and Japanese labor markets also rely on longer- term employment relationships.4 Both employer and employee therefore face less uncertainty about turnovers and layoffs and can share the costs of training in job-related skills more efficiently. For example, employees can contribute a share of these costs by accepting starting wages that are lower than would be the case in the absence of the training investment by the employer. In the United States, where blue-collar employee turnover is relatively high (Levy et al., 1984, estimated that average annual turnover rates in manufacturing were more than 20 percent) and layoffs are more common, sharing training costs between employer and employee is more difficult; there is, among other things, greater uncertainty about the probability of layoffs and a firm's ability to retain the returns from its investment in job-related skills training. Thus, the levels of investment in such training appear to be lower, although better data are needed to support more strongly or reject the predictions of this heuristic model. Interestingly, some recent union contracts in the United States contain provisions resembling those of Japanese labor contracts. Several recent collective bargaining contracts in the U.S. aerospace and automotive industries, for example, combine annual bonuses or profit sharing for workers with corporate commitments to greater employment security and larger invest- ments in worker retraining. 4Although this contrast has characterized U.S. and Japanese labor markets since the early 1950s, recent structural changes in the Japanese economy, which have resulted in some large layoffs, may alter the structure of Japanese labor markets and labor contracts in the future.
142 TECHNOLOGYAND EMPLOYMENT The Adequacy of the Training System Can the U.S. system we have described respond to changes in the demands for specific job-related skills? Based on the evidence reviewed by this panel, the U.S. training system for job-related skills appears to be responsive to the demands of employers. This measure of adequacy does not address the issue of the distribution of such education, nor does it confront the adequacy of worker preparation in basic skills. Neverthe- less, a serious "structural" mismatch between the supply of and demand for skills in the American economy should be revealed in recurrent skill shortages or in the unemployment of workers with particular skills in low demand. In fact, there is little evidence of significant skill shortages for any extended period, although unemployment within the population receiving little job-related or basic skills training (the disadvantaged and unskilled) has been a persistent problem for much of the post-1945 period. The formal training system (comprising elementary, secondary, and higher education) for labor market entrants (1~24 years old) also appears to respond rapidly to changes in the skills demanded by employers (Berryman, 19851. Especially within higher education, perceptions of oversupply or excess demand in specific fields heavily influence students' intended fields of study, their actual fields of study, and the level of completed degrees. Comparable data for employer-provided training almost certainly would reveal similar flexibility. Operating in conjunction with the enormous size of the decentralized U.S. training system, this flexibility offers considerable opportunities to alter rapidly the mix of job-related skills in the U.S. work force. About 6 million students completed programs of study at the secondary and postsecondary levels during the 198(~1981 academic year; the annual output from these formal channels was approximately 5.7 percent of the work force. As Berryman puts it, "the sheer number of these completions per year represents a remarkable opportunity to rapidly re-configure the skill profile of the American labor force" (p. iii). This evidence suggests that the enormous system that trains workers in job-specific skills can respond to changes in the demand for skills of different types resulting from the adoption of new technology or structural change in the economy. In fact, in some instances the supply mechanism may respond too rapidly to anticipated changes in demand as in the rush to secure robotics technician training in 1982-1983, when there was one student enrolled in an introductory robotics course for every robot likely to be purchased (Hunt and Hunt, 1983~. Consequently, we see little if any need for additional investments in forecasting skill requirements or training needs, especially since the reliability of these forecasts re- flecting their high levels of uncertainty and the weak methodologies they
CURRENT POLICIES FOR WORKER ADJUSTMENT 143 use has been low. (See Binkin, 1987, for a critical assessment of the U.S. military's experience in forecasting occupational trends and training requirements.) TRAINING IN BASIC SKILLS FOR LABOR FORCE ENTRANTS5 Labor force entrants acquire basic skills largely within the U.S. public educational system, and educational attainment is a rough index of a worker's level of preparation in these skills. As we noted earlier, basic skills currently are important for obtaining entry-level jobs and will be even more important in the future for obtaining better-paying jobs and for climbing the economic ladder toward higher earnings. As indicated by the evidence in Chapter 3 on growth during the past decade in the returns to higher levels of educational attainment, the distribution of educational attainment within the U.S. population will have a great influence on the distribution of the economic fruits of technological change. Although educational attainment is improving within the labor force entrant popu- lation, the large remaining gaps in attainment among whites, blacks? and Hispanics are a cause for concern. More limited evidence, noted in Chapter 3, suggests that the quality of basic skills training for U.S. labor force entrants is lower than that provided to the labor force entrants of other nations (e.g., Japan). Significant deficiencies in the quality of such training for U.S. labor force entrants relative to other nations will impede the ability of this nation to generate and adopt new technologies with sufficient speed and effectiveness to remain competitive in the world economy. DISPLACED WORKERS Although we believe that employment displacement resulting from technological change will not be widespread, there will be transitional, regional, and occupational dislocation. As we noted in Chapter 3, the BLS estimates that the number of experienced workers suffering per- manent job loss from all causes is roughly 1 million persons per year. Currently, there are no analyses of the number or characteristics of the smaller population of workers who have been displaced by technological change. Such data are important because the needs of these workers may differ substantially from those of the larger displaced worker population. 5The data on educational attainment in Chapters 3 and S underlie this assessment of the adequacy of basic skills training for labor force entrants.
144 TECHNOLOGYAND EMPLOYMENT Lacking this information, in the discussion later in this section we do not try to differentiate among workers according to cause of displacement. As we note in Chapter 10, there are also strong arguments against making such distinctions in the design and implementation of adjustment assistance programs for displaced workers. Workers displaced by technological change receive greater attention than firms suffering dislocations from the same cause because of the greater adjustment problems of workers. Over a working career a worker accumulates a stock of skills and "human capital" that may be highly specific to a single occupation within a firm. When technological change or other forces transform or result in the loss of that job, the value of a worker's skill-based assets may suddenly vanish. A firm facing the loss of markets for its products generally can redeploy its assets and working capital more easily than an individual worker. Machinery and inventories can be sold (albeit at a discount), allowing a firm to realize some return on these assets and enabling it to adjust. Adjustment to technological or structural change is difficult for both workers and firms, but the unique problems faced by workers have drawn attention to their needs. Much of the following discussion of programs for displaced workers focuses on training in both basic and job-related skills. Although we believe that additional resources should be devoted to retraining dis- placed workers, we must stress that retraining is not a panacea. The ability of retraining to restore the earning power of some displaced workers, especially those previously employed in durables manufacturing (e.g., basic steel), may be limited. Furthermore, many displaced workers may not participate in retraining programs; they are interested primarily in rapid reemployment, rather than retraining. The most important function of public adjustment programs often is to provide job search assistance and information, rather than retraining, for individuals who may not have changed jobs for 20 years. Characteristics To assess the adequacy of policies that address their needs, reliable information on the characteristics of displaced workers is essential. The description in Chapter 3 of the displaced worker population noted the following characteristics: · Younger workers are more likely to be displaced, but the duration of unemployment after displacement is greater for older workers. · Both the duration of unemployment and the magnitude of earnings losses associated with reemployment after displacement are higher in regions with relatively high unemployment.
CURRENT POLICIES FOR WORKER ADJUSTMENT 145 · A large fraction of this population either is ineligible for unemploy- ment compensation or exhausts these benefits prior to reemployment. · Many displaced workers, especially those from durables manufac- turing, experience serious financial losses as a result of displacement and receive lower wages in their new jobs. · A large portion of the displaced worker population suffers from serious basic skill deficiencies. Displaced workers are a heterogeneous population made up of groups with different histories of employment and earnings and consequently different needs for retraining and employment assistance. This fact complicates the design of programs to serve them, as illustrated by a discussion of two particular groups. During the early 1980s, permanent job loss affected a segment of the U.S. work force that previously had experienced limited long-term unemployment that is, unionized, high-wage workers in curable s man- ufacturing. By 1984 more than 800,000 of the 5.1 million experienced workers identified by BLS as displaced during the previous 5 years came from the nonelectrical machinery, automobile, and primary metals (largely steel) manufacturing industries. Displacement of workers in these industries has been a particular concern for policymakers and the public alike, due to the unprecedented (by post-1945 standards) scope of long-term unemployment within this group, as well as the financial hardships associated with job loss in these industries. Displaced workers from this sector generally are well-paid men with years of steady employment in the same plant (Flaim and Sehgal, 19851. Retraining in job-related skills is not likely to restore the previous earnings of many members of this group, and most programs to aid them focus on the transition between jobs. Job transition needs are addressed by union contract provisions in many of these industries, provisions whose aims are to reduce the short-term financial hardships of displace- ment. Despite such transitional help, however, a substantial portion of this group of displaced workers may face difficulties in finding new employment because they lack the basic skills that ease the search for new jobs. Designing policies to aid this group is complicated by the difficulty of determining whether their displacement is permanent. The high wages these workers were able to secure in their original jobs, the likelihood that alternative employment will mean a substantial cut in these wages, and the availability of supplemental unemployment benefits (SUB) in many of these industries all encourage these workers to avoid any action that might divert them from reemployment in their former occupation. Many of these workers choose to avoid the costs of
146 TECHNOLOG Y AND EMPLO YMENT retraining and searching for a new job, in hopes of being recalled to work. Another group within the displaced worker population comprises individuals who are poorly educated, female, black or Hispanic, and nonunion. According to the January 1984 displaced worker survey, nearly 900,000 of the 5.1 million experienced workers displaced during 1979-1983 were black or Hispanic; these workers experienced signifi- cantly longer unemployment than did white workers (Flaim and Sehgal, 1985; Podgursky, 19871. The characteristics just noted (low education, female, nonwhite, and nonunion) often are associated with lower-wage employment; for these individuals, the acquisition of basic or job-related skills may enhance their prospects for reemployment and also lead to jobs at higher wages. Thus, this group is more likely to benefit from retraining or training. These two groups of displaced workers have different needs. The electiveness of retraining in job-related skills to restore the earning power of a displaced steelworker or autoworker may well be limited, although training in basic skills is essential to some of these individuals in finding new employment and job search counseling and assistance also may be beneficial. For members of the second group, a combination of basic skills training, job-related retraining, and job search counseling and assistance could be of substantial help. Federal Training and Income Support Programs Since 1945, federal programs to aid displaced workers have consisted of the Manpower Development and Training Act (MDTA) of 1962, super- seded in 1973 by the Comprehensive Employment and Training Act; Title III of the Job Training Partnership Act (JTPA) of 1982; Trade Adjustment Assistance (TAA), passed in 1962; unemployment insurance, which provides short-term income support to unemployed workers; and the Perkins Vocational Education Act of 1984.6 MDTA was passed on the wave of public concern over technological change and unemployment that gave rise to the National Commission on Technology, Automation, and Economic Progress. The law was intended to help technologically displaced workers with 3 or more years of work experience by retraining these individuals for occupations believed to be 6Numerous sectoral federal adjustment programs have been developed during the past century, ranging from the Interstate Commerce Act of 1887 to the Redwood National Park legislation of 1978. These programs typically provided income support and retraining to workers in industries affected by legislative or other governmental actions. All of them were very narrowly focused.
CURRENT POLICIES FOR WORKER ADJUSTMENT 147 in high demand. After 1964, however, declining aggregate unemployment resulted in a reorientation of the program toward disadvantaged rather than displaced workers; it was eventually replaced by the Comprehensive Employment and Training Act (CETA), which focused primarily on disadvantaged workers. In 1982 CETA was replaced by the JTPA. THE JOB TRAINING PARTNERSHIP ACT After the reorientation of MDTA in the early 1960s, no federal programs other than TAA addressed the problems of displaced workers until the passage in 1982 of JTPA. JTPA's Title III, Employment and Training Assistance for Dislocated Workers, provides federal funds to the states for training and related employment services for workers who have been laid off or have received notice of layoff and who are unlikely to return to their previous industry or occupation. Workers who have been laid off or who are about to be laid off because of a permanent plant closing and the long-term unemployed are also covered. Although the actual services available to workers under Title III vary by state, they generally emphasize job search assistance. Support for job-related retraining is modest, and there is little training offered in basic skills. The U.S. General Accounting Office's recent survey of JTPA Title III programs (1987a) found that in 1985 only 6 percent of program participants received basic skills training (with a median duration of only 2 weeks), whereas 42 percent received some form of job-related training through classroom or on-thejob learning. Eighty-four percent of Title III program participants, on the other hand, received job counseling, and 66 percent received job search assistance. Overall, the General Accounting Office has estimated that Title III serves only 6-7 percent of the annual flow of displaced workers. TRADE ADJUSTMENT ASSISTANCE Trade adjustment assistance was first provided under the Trade Expansion Act of 1962 to assist workers displaced as a result of the bill's reduction of trade barriers. As originally enacted, workers were eligible for the income support and training benefits offered by TAA only when tariff reduction-and the import penetration of U.S. markets that often followed was the single most important cause of displacement. The Trade Act of 1974 broadened participation in the program by extending benefits to any worker for whom imports "contributed importantly" to his or her displacement. ("Contributed importantly" meant that import penetration was an important cause of job loss but not necessarily more important than any other cause.) As a result of this change in eligibility
148 TECHNOLOGYAND EMPLOYMENT criteria, as well as surging auto imports, the program grew from approximately 14,000 workers and $15 million in benefits in fiscal year 1973 to a peak in fiscal year 1980 of 532,000 workers and approximately $1.6 billion in benefits, much of which went to displaced auto workers. In 1981, Congress redefined and limited TAA income support pay- ments. In fiscal year 1986, TAA served an estimated 42,000 workers at a cost of approximately $148 million. Projected outlays for fiscal year 1987 are approximately $206 million. The President's budget for fiscal year 1988 has proposed that the program be replaced with a broader displaced worker program (Aho and Bayard, 1984; U.S. House of Representatives, Committee on Ways and Means, 1987; unpublished 1987 data obtained from the U.S. Office of Trade Adjustment Assis- tance). Despite the program's stated goal of providing both income support and retraining, TAA has emphasized retraining only since 1985. Of the more than $4 billion in program outlays since the passage of the Trade Act of 1974, less than 5 percent has been allocated to retraining (U.S. House of Representatives, Committee on Ways and Means, 19871. One reason for the minor role accorded to retraining stems from the design of TAA. Throughout its life, the program has been hampered by the requirement to determine the contribution of import competition to a worker's displacement. According to one evaluation, during the mid- 1970s, this requirement so slowed the delivery of TAA services that the first payment of TAA funds was received by a worker on average 14 months after layoff (Corson et al., 19791. Such delays meant that in many cases workers received payments well after they had found new jobs and largely precluded any meaningful role for retraining within TAA. UNEMPLOYMENT INSURANCE The unemployment insurance system is the only public income support program for which most displaced workers are eligible. Previous employ- ment can qualify workers for a maximum of 26 weeks of benefits.7 Benefit levels vary across states, with the highest weekly benefits ranging Washington and Massachusetts have a 30-week maximum benefit period. In addition, Massachusetts provides up to 13 weeks of supplemental reemployment assistance benefits to workers unemployed as a result of a large layoff or plant closing who did not receive at least 90 days' advance notice and/or separation pay (Commonwealth of Massachusetts, Division of Employment Security, 1987). Federal law provides for a 13-week extension of benefits during periods of very high unemployment. As of January 1987, only Alaska and Puerto Rico qualified for this extension.
CURRENT POLICIES FOR WORKER ADJUSTMENT 149 between 50 and 70 percent of the average weekly wage. Because those workers with above-average wages receive a lower proportion of their former salary in unemployment benefits, the national average weekly benefit was only 36 percent of the average wage in fiscal year 1986. Average total benefits for fiscal year 1986 were $1,956; maximum total 1987 benefits range among the states from $3,120 to $9,900 (U.S. House of Representatives, Committee on Ways and Means, 19874. For a worker undertaking retraining, these benefits could provide as much as 26 weeks of income support; but many states prohibit or otherwise discourage individuals from receiving unemployment insurance payments while enrolled in retraining (Barton, 1986) in order to minimize the time an unemployed worker is drawing unemployment compensation.8 FEDERAL ASSISTANCE FOR UPGRADE TRAINING FOR EMPLOYED WORKERS The Perkins Vocational Education Act of 1984, the most recent revision to legislation authorizing federal support for state-level programs in vocational education, mandated that more than $90 million of the roughly $882 million in federal support for vocational education be spent by states on "adult" education, including "upgrade" training for employed work- ers and education in basic skills for adults. In 1986 the U.S. Department of Education began a large-scale evaluation of the Perkins Act; when its results are available, they should inform the development of other federal policies to improve the job-related and basic skills of the U.S. work force (U.S. Department of Education, 19861. Indeed, rigorous program evalu- ation is critical to all efforts to design new programs to help displaced workers. Evaluation of Federal Displaced Worker Programs Have federal displaced worker programs improved the reemployment prospects of participants? Unfortunately, research on the effectiveness of adjustment assistance, including job training for experienced adult workers, suffers from methodological problems and a shortage of data and funding. The evaluations of MDTA (Ashenfelter, 1978; Bloom, sCalifornia, through the California Training Benefit Program, provides an additional 26 weeks of state-funded unemployment benefits for displaced workers undertaking training in an occupation for which there is demand in their geographical area (California Employment Development Department, 1986a, 1986b). Displaced workers enrolled in training or other adjustment assistance programs supported by funds from Title III of the JTPA also are permitted to collect unemployment compensation throughout the United States.
150 TECHNOLOGYAND EMPLOYMENT 1982; Sommers, 1968) do not provide sufficient evidence to support broad statements about the effectiveness of these programs. Only three recent evaluations of displaced worker programs the Downriver Com- munity Conference Economic Adjustment Program (Kulik et al., 1984), the Buffalo Dislocated Worker Demonstration Program (Corson et al., 1985), and that sponsored by the Texas Department of Community Affairs (Kulik and Bloom, 1986) have yielded reliable evidence on their effectiveness.9 This small number of rigorous evaluations is disturbing, considering that over $500 million has been spent on JTPA Title III programs. Two of these evaluations, supported by the U.S. Department of Labor, examined demonstration and experimental programs that had been un- dertaken immediately prior to the implementation of JTPA. The third was supported by the Texas Department of Community Affairs, which used a portion of its JTPA Title III program funds to sponsor and evaluate several demonstration programs. These evaluations found that most workers who received services from the programs were unemployed for shorter periods and had higher wages in jobs found immediately after completing the program than was true of workers in the control group (Bowman, 1986~. Additional evaluations of workers over longer periods of time, however, are necessary to determine whether these gains in wages persist over the long run. Despite their positive general findings, these evaluations provide much less information on the effects of the specific services provided by the programs. All three evaluations had design and implementation problems that do not allow definitive conclusions about the relative merits of different services within adjustment programs for displaced workers. For example, the effect of job training, as opposed to job search assistance, cannot be ascertained. The conclusions of these studies also may not be applicable to displaced worker populations that are significantly different from those of the study populations. Finally, although case study evi- dence suggests that a large percentage of displaced workers suffer from basic skill deficiencies, none of these programs offered this type of training; thus, we have minimal evidence on the design and effectiveness of programs providing basic skills training to displaced workers. Clearer insights into the effectiveness of displaced worker programs and the appropriate design of such programs must await the results of further evaluations, one of which currently is being planned by the U.S. 9We exclude evaluation of TAA from this discussion because of the program's design and certification requirements. The results of a large-scale evaluation of TAA are reported in Corson et al. (1979); Aho and Bayard (1984) provide an overview and summary of a number of evaluations of TAA.
CURRENT POLICIES FOR WORKER ADJUSTMENT 151 Department of Labor. Yet present levels of support for evaluation are insufficient. The magnitude of projected expenditures on displaced worker programs (as proposed in the President's fiscal year 1988 budget, as much as $980 million annually), the paucity of evidence concerning the most effective mix of services, and the significant differences in the costs of various services all suggest that greater investments in program evaluation are needed (see Chapters 8 and 10 for further discussion). Although additional evaluations are needed to measure the effective- ness of various program services, the bias in the existing JTPA Title III program mix toward job search and placement services suggests that workers with basic skill deficiencies may be underserved. Generally speaking, we believe that there is too little emphasis on basic skills training within Title III. It is difficult to accept the proposition that the needs of workers with basic skills deficiencies can be addressed in 2 weeks of training, the average amount provided by Title III programs (U.S. General Accounting Office, 1987a). In addition, Title III programs, by virtue of the design of their contracts with service providers (contracts that are based on performance in placing participants in jobs), tend to process and train only those displaced workers who are easy to place. Workers with basic skills deficiencies thus may be excluded. The absence of income support within most Title III programs (beyond any unemployment compensation for which displaced workers are eligible) means that any training must generally be completed in 26 weeks or less to avoid exhausting their unemployment insurance benefits. This re- duces the attractiveness and efficacy of retraining and may make it completely infeasible for the many displaced workers who do not receive unemployment compensation. State Programs: The California Employment Training Panel Most state-administered displaced worker training programs rely in part on JTPA Title III funds. The California Employment Training Panel (ETP), however, a training and employment assistance program, is funded entirely with state monies. (ETP is also of interest because of its size and innovative design.) The program was established in 1982; it receives an annual allocation of approximately $55 million from a small state payroll tax imposed on employers who pay unemployment insur- ance.~° ETP provides training for both displaced workers and employed reimposition of the ETP tax was made less onerous by an offsetting reduction in the state unemployment insurance tax, a move made possible by the large surplus in California's unemployment insurance fund.
152 TECHNOLOGY AND EMPLOYMENT workers threatened with layoff. Unlike JTPA, however, the worker need not have received notice of an impending layoff to qualify for training assistance. By investing in training for workers before they are laid off, ETP supporters argue that the program reduces the costs of personnel turnover for employers and reduces outlays from the state unemployment insur- ance fund. The state's support of retraining for employed workers also is based on the belief that such retraining improves the competitiveness of the state's industry by encouraging the adoption of new technologies; ETP's goal is to avoid immediate job losses while strengthening long-term employment opportunities (California Employment Training Panel, 19851. A preliminary study of 1,200 ETP project participants who completed training and were placed in jobs during 1983-1985 suggests that these goals have been met: the average number of weeks of unemployment for participants declined by more than 60 percent after ETP training, and average earnings rose by more than 50 percent (Moore, 19861. This evidence, however, is preliminary rather than definitive; it was not compiled from a rigorous experimental design that included a control group, nor were the workers graduating from ETP training tracked over an extended period of time. Without question, ETP represents an imaginative response to the imped- iments to greater investment by firms in training their blue-collar work force. The program's effectiveness in delivering retraining to the displaced worker population, however, is limited. Basic skills training for employed and displaced workers alike is not supported by the program as a matter of policy. Moreover, ETP's performance requirement (in its contractual agree- ments with training providers, the program requires that displaced workers be placed in jobs for 90 days prior to payment for the training provided) discourages many potential external providers of training for displaced workers (e.g., community colleges) from participating. Although it was initially intended to provide retraining for experienced workers who were actually unemployed as well as threatened with unemployment, ETP in- creasingly appears to be financing the training by firms of their employed work force. Thus, the program may in some cases substitute public funds for training investments that would have been made in the absence of the program, which means that there is no net increase in the delivery of training services to workers threatened with displacement. ETP and other state-level programs that fund training for the employed work force offer a rich set of policy experiments for analysis and evaluation. (See Stevens, 1987, for a descriptive survey of such programs.) Federally supported evaluations of a sample of state programs could yield useful information about the design and effectiveness of these programs for decision making at both the state and federal levels.
CURRENT POLICIES FOR WORKER ADJUSTMENT 153 Displaced Workers and Adult Education The majority of federal and state programs for displaced workers rely on the existing adult and community college educational infrastructure to deliver basic skills and job-related training. These postsecondary educa- tional institutions have compiled a mixed record of success in meeting such demands. A recent study by Bruno (1986) noted that many commu- nity colleges historically have served college-age students rather than more mature displaced or employed workers. Historically, many of these institutions have not pursued curriculum development and staffing poli- cies to meet the needs of displaced workers for training in basic or job-related skills. This pattern of development has hampered the ability of many of these institutions to meet the needs of displaced workers. As Bruno observes: . . . the community colleges and vocational-technical schools must reexamine their philosophy and policies if they are to meet the growing need for training created by dislocated workers. Unlike traditional college students, they have neither the time nor financial resources after being laid-off to enroll in a one- or two-year certificate program. They are in a state of crisis that must be addressed quickly. (p. 59) The very technologies whose development has created concern over worker displacement have great potential for the innovative delivery of training in basic and job-specific skills. For example, one of the greatest impediments to participation in adult education programs is the reluc- tance of displaced workers to subject themselves to a classroom environ- ment. Advanced teaching technologies, such as self-paced instructional workstations, can support individualized learning outside the classroom, thereby enhancing the attractiveness and effectiveness of basic skills and job-related skills training. Although the existing educational system exhibits a number of deficien- cies in meeting the needs of displaced workers, many innovative commu- nity colleges have shown themselves capable of doing so when offered financial incentives. Better financing of worker access to this system through income support and other financial assistance for retraining can motivate community colleges and other adult education institutions to address the special needs of displaced workers. Careful program over- sight and evaluation, however, also are necessary. Private Adjustment Programs In recent years, federal and state income support and retraining programs have been supplemented by a growing number of privately
154 TECHNOLOGYAND EMPLOYMENT funded, joint union-management programs. In a number of manufacturing industries, unions have negotiated supplements to regular unemployment benefits. These benefits are coordinated with regular unemployment insurance and provide higher benefit levels as much as 95 percent of take-home pay for up to 1 year (Jacobson, 1986~. Through collective bargaining, a number of firms have also established early retirement programs. (Such plans, however, typically require giving up the right to be recalled to work.) In some unionized industries (e.g., steel), workers who are displaced can qualify for severance pay or pension supplements based on a combination of years of service and age. Severance pay is not widely available, mainly because workers do not want to give up recall rights, and it is often offered as part of a plant closing settlement. Moreover, recent events (e.g., the bankruptcy of the LTV Corporation, a major steel producer, in 1986) suggest that these payments are not com- pletely reliable. Firrn bankruptcies can jeopardize these supplemental or severance benefits. An analysis of data from a recent survey by the U.S. General Accounting Office (1987b) of layoffs and plant shutdowns involving establishments with more that 100 workers found that severance pay was one of the most common forms of private adjustment assistance-37 percent of the respondents provided severance pay to their blue-collar employees, and 57 percent provided such compensation to white-collar employees. Another common form of assistance was employer continu- ation of health insurance coverage. Thirty-eight percent of firms provided this benefit for blue-collar employees, and 48 percent did so for white- collar employees (the published survey data do not specify the length of the period of health insurance continuation). Other common forms of employer-provided adjustment assistance include the continuation of life insurance coverage (22 percent of the firms provided this benefit for blue-collar employees and 30 percent did so for white-collar workers) and job search assistance (provided by 26 percent of firms to blue-collar employees and by 35 percent to white-collar employees). The survey does not provide a breakdown of the benefits coverage of unionized and nonunionized workers. Nevertheless, this evidence suggests that white- collar employees receive somewhat more generous adjustment assistance from employers after plant shutdowns or layoffs. A number of programs have been established recently through collec 'iGAO mailed survey forms to 500 establishments chosen from an initial sample of 2,400. Data incorporating responses from 60 percent of the firms surveyed are published in its 1986 report (U.S. General Accounting Office, 1986); data incorporating responses from 80 percent of the firms are published in a 1987 report (U.S. General Accounting Office, 1987b). Brown (1987) presents date from the GAO survey that includes responses from 80 percent ofthe firms.
CURRENT POLICIES FOR WORKER ADJUSTMENT 155 tive bargaining to assist displaced workers and provide training services. The best-known examples of these programs are the United Auto Workers (UAW)-Ford and UAW-GM "nickel" and "dime" funds. The UAW-Ford program, which began with the 1982 contract, provides services to displaced hourly employees that are similar to those pro- vided under JTPA; job search assistance, counseling, tuition assistance for education and retraining, and, beginning in 1984, relocation assist- anceintheformofloans. Between 1982 and 1985, approximately 12,600 laid-off Ford workers took part in one or more of the programs. Based on industry and union estimates that at some point during this period approximately 100,000 Ford hourly workers were laid off, this yields a participation rate of 13 percent. Although these programs are primarily funded by employer contributions, they also receive JTPA funds (Pascoe and Collins, 19851. Advance Notice of Plant Shutdowns or Permanent Layoffs as a Mechanism for Adjustment The best time to undertake programs of job search assistance, coun- seling, and retraining for workers is prior to their displacement. In most cases, this can occur only with the cooperation of the employer cooperation that includes advance notice to workers of impending plant shutdowns or large permanent layoffs. The efficacy of predischarge help reflects the tendency of workers to disperse after layoff, the greater effectiveness of programs that have the cooperation of management and workers, and the greater willingness of workers to enroll in such programs when they are available prior to layoff. Data on the operation of several adjustment assistance programs for displaced workers confirm that pre-layoff assistance is used more inten- sively by workers. According to the Downriver Community Conference, 50 percent of workers participated in adjustment programs that were available prior to plant closure, 35 percent participated in programs made available up to 1 year later, and only 17 percent participated in programs offered after 2 years (U.S. Congress, Office of Technology Assessment, 1986a). The Philadelphia Area Labor-Management Committee found that employee participation in worker assistance workshops ranged between 70 and 80 percent when these workshops were provided prior to layoff; '2There exists no consensus on the "optimal" period of advance notice, but discussions of best-practice methods for plant shutdowns (Driever and Baumgardner, 1984) suggest that 1-6 months' notice is helpful. Section 283 of the Trade Act of 1974 urges firms moving production facilities to other countries to provide 60 days' notice to employees (cited in U.S. Congress, Office of Technology Assessment, 1986a).
156 TECHNOLOG Y AND EMPLO YMENT when the workshops were offered after layoff, participation dropped to less than 20 percent (Berenbeim, 19861. Statistical evidence on the effects of advance notice suggests in addition that advance notice of shutdowns or layoffs is associated with shorter spells of unemployment after these events. A study by Folbre et al. (1984) of the impact of advance notice in Maine found, on average, that unemployment in local labor markets in which plant closings were announced in advance was lower than unemployment in regions that experienced plant shutdowns without advance notice. The study also found that providing at least 1 month's advance notice reduced the average duration of unemployment per worker from 5 to 4 months. Addison and Portugal (1987), who applied a model of unemployment duration associated with advance notice to national data from the BLS 1984 displaced worker survey, confirmed Folbre et al.'s results, conclud- ing that advance notice reduced the average spell of unemployment after layoff by roughly 27 percent, or 4 weeks. Data in Flaim and Sehgal (1985) suggest that advance notice of layoffs did not have a substantial impact on the probability that a worker displaced during 1979-1983 was employed as of January 1984. The data on which their analysis was based, however, were flawed. They were drawn from the BLS 1984 survey of displaced workers, which did not specify the period of advance notice. In addition, because the survey asked whether workers received advance notice or "expected" that layoffs were imminent, the methodology may have introduced significant recall bias. Because this evidence suggests that advance notice of plant closures and large-scale layoffs increases the effectiveness of publicly supported programs of worker adjustment assistance and reduces the average duration of unemployment after layoff, advance notice may reduce the costs to taxpayers of such actions. Plant shutdowns and large-scale layoffs that occur without advance notice appear to impose substantial additional costs on both workers and the public sector, by comparison with situations in which notice is given. Such costs are "externalities," costs imposed on individuals and society that are not borne by the firms closing the plant or laying off the workers. In other areas (e.g., health and safety regulation, pollution controls), public regulations have been devel- oped to ensure that a portion of these costs are also borne by the organizations that contribute to them. In addition to externalities, plant closures and layoffs in which workers do not receive information available to managers concerning the immi- nence of displacement may cause workers to make career decisions based on defective or incomplete information. Hamermesh (1987) estimated that plant shutdowns without advance notice impose an average cost on workers (measured as the difference in the level of worker investments in
CURRENT POLICIES FOR WORKER ADJUSTMENT 157 job-related skills for a specific employer in a situation in which notice is provided and a situation in which it is not provided) of $4,50~$15,000.~3 Competitive markets function most efficiently when all of the actors in them have equal access to information concerning their environment and the consequences of their actions. The incentives for participants in markets to disclose freely all information, however, often are minimal strategic use or selective disclosure of information (as in the many recent "insider trading" cases in securities markets) can prove extremely profitable. The disclosure of relevant information to all parties to a transaction or contract is one of the primary motivations for statutory and regulatory control of securities markets, health and safety regulations, and consumer protection regulations and laws. Employers are often concerned about employee behavior and produc- tivity after advance notice. Do employees facing permanent layoffs react by sabotaging products or otherwise lowering productivity substantially? There is widespread agreement among business and union leaders (Berenbeim, 1986, p. 14; Driever and Baumgardner, 1984, p. 14; the Secretary of Labor's Task Force on Economic Adjustment and Worker Dislocation, 1986, p. 23; U.S. Congress, Office of Technology Assess- ment, 1986b, quoting business and union participants in a workshop, p. 22) that advance notice does not lower worker productivity after the announcement. Indeed, the recent study by Ronald Berenbeim for the Conference Board reported that "all industrial plants studied [five] noted improvements in quality and productivity in the final phase of the facility's operations" (1986, p. 141. These results corroborate those of an earlier study by Weber and Taylor (1963) of 32 plant shutdowns. The productivity and quality improvements that occur appear to reflect the reaction of employees to the evidence that management is concerned about their welfare, the operation of counseling and job search programs that begin prior to layoff, the resolution of anxieties and uncertainties, and the desire of workers concerned about reemployment to demonstrate to new employers that the quality of the work force in the closed plant was high. The evidence on the benefits of advance notice for worker adjustment and the distribution of social costs, combined with evidence on its productivity effects, has led a large number of employer organizations, labor-management task forces, and public commissions to endorse voluntary advance notice. Private sector groups endorsing advance notice include the Business Roundtable (1983), the National Association i3This cost measures only the loss in the value of workers' investments in such training and ignores the loss of earnings associated with unemployment and the possibility that reemployment will involve reductions in wages.
158 TECHNOLOG Y AND EMPLO YMENT of Manufacturers (1987), the National Alliance of Business (1987), and the National Center on Occupational Readjustment (19841. Recent public commissions, study groups, and officials endorsing advance notice include the President's Commission on Industrial Competitive- ness (1985), the Secretary of Labor's Task Force on Economic Adjust- ment and Worker Dislocation (1986), and the Assistant Secretary of Labor for Employment and Training (Semerad, 19871. Despite this support for advance notice, however, many of these groups oppose or have failed to endorse public actions to mandate advance notice of plant closures and permanent, large-scale layoffs, citing as justification the high costs and counterproductive employment effects of mandatory regulations. The costs of mandatory advance notice cited by these analysts include the additional burdens that these regulations place on small firms, the impracticality of advance notice under all circumstances in an economic environment characterized by uncertainty and change, the need to preserve managerial discretion in decision making, and the possibility that such regulations will raise the costs of operating production establishments within the United States, further encouraging the movement of production to offshore locations and discouraging domestic job creation. As an example of the negative effects of mandatory advance notice, analysts point to Western Europe, where advance notice regulations are relatively common. Opponents of mandatory advance notice cite these regulations as contributing to low rates of job creation, due to their tendency to raise the costs of operating a business. (See the National Alliance of Business, 1987, and Semerad, 1987; Balassa, 1984, presents related evidence on European job creation and regulatory costs.) There is little or no evidence that would allow a systematic estimation of the magnitude or significance of these costs. Federal legislative proposals to require advance notice of plant shutdowns and permanent layoffs have accommodated several of these objections, typically ex- empting small firms (those with fewer than 5~100 employees) and allowing exemptions for "unforeseen circumstances." Although a num- ber of Western European nations enforce mandatory advance notice regulations, in many nations these are combined with requirements for substantial severance payments to displaced workers, making a precise comparison with the impacts of notification alone very difficult.~4 '4The report of the Subcommittee on the Foreign Experience of the Secretary of Labor's Task Force on Economic Adjustment and Worker Dislocation (1986) has disputed the argument that social regulations have contributed to low rates of job creation in Western Europe, arguing that macroeconomic policies and inefficient nationalized industries, among other factors, are more significant.
CURRENT POLICIES FOR WORKER ADJUSTMENT 159 The primary issue in the debate over advance notice does not concern its merits there appears to be widespread agreement that such notice aids worker adjustment but the most elective way of ensuring its widespread application at reasonable cost. A number of U.S. employers (both union and nonunion firms) currently provide voluntary advance notice of plant closures and layoffs to their employees. Evidence from a recent national survey administered by the U.S. General Accounting Office (1986, 1987b), however, strongly suggests that voluntary advance notice is not functioning effectively few workers are receiving even 30 days' advance notice. The GAO survey data, which appear in Brown (1987), suggest that 31 percent of the respondents provided no specific notice of plant closure or layoff (i.e., informing workers in advance of a specific date of layoff), 34 percent provided 1-14 days' notice, 15 percent provided 15-30 days' notice, and 20 percent provided 31 or more days' notice. On average, blue-collar workers received 7 days' notice and white-collar workers received 14 days' notice.~5 In unionized establish- ments, blue-collar workers received an average of 2 weeks' specific notice; blue-collar workers in establishments without unions received an average of only 2 days' advance notice of plant closure or lay of (U.S. Congress, Office of Technology Assessment, 1986a; U.S. General Ac- counting Office, 1986, 1987b). i5The first set of survey results published by the General Accounting Office (1986) found that 76 percent of establishments provided "general" advance notice, defined as notice to groups of workers that some or all of them might be laid offin the future. "Specific" advance notice was defined as telling individual workers of the date of their impending layoff. A more restrictive definition of general advance notice was used by Brown (1987) who defined it as notification of individual workers that layoffs or a plant closing were likely. Drawing on data from the BLS Permanent Mass Layoff and Plant Closing Program (Secretary of Labor's Task Force, 1986, Appendix C), which defined general advance notice as notification to individual employees that they would be laid off, Brown found from a survey of establish- ments employing 50 or more workers that only 36 percent of the establishments provided general notice. The most recent tabulation of a more complete set of responses to the GAO survey (U.S. General Accounting Office, 1987b) has eliminated any analysis of "general" advance notice.
