NCEE Brief No. 4, November 1989
Roger J. Vaughan and Sue E. Berryman
Employer-sponsored training in the United States is a substantial human-capital-producing and wealth-producing system. The distribution of that training, its consequences, and the forces that are changing it have important public policy implications. How much and how well we invest in human capital will shape how fast national income grows, how fast we expand our capacity to produce, and how the benefits of development are shared.
Investments in employer-sponsored training account for about 40 percent of our annual human capital investments, totalling about $200 billion (4.4 percent of Gross National Product) in 1985. These investments are about half as large as investments in plant and equipment.
Most people need training to get and keep jobs. Employers are a major source for this training. In 1983, 42 percent of the men and 34 percent of the women in the labor force needed training to obtain their current job and got some or all of it either in formal company programs or informally on the job. Employers were also a major source of training for employees needing to improve job skills more than a fourth of the men and women in the labor force received training, some or all of it from their employer, to improve their job skills.
Employers do not train all of their employees. Seventy-nine percent of college graduates receive training from their employers, while 71 percent of high school graduates and only 45 percent of those who failed to complete high school receive training from their employers. Further, employees who are trained in one job are more likely than other new employees to be trained in subsequent jobs. In short, employers train the trainable and the trained. Employer training accentuates differences in educational attainment and achievement among employees differences that account for most of the differences in income among workers.
Lack of education and training is the single most distinguishing characteristic of the poor. Lacking preparation, they are unlikely to be trained by their employers. Less than three percent of disadvantaged men and women report receiving any training from their employers. Those entering the workplace without sound academic and problem-solving skills find it difficult to remedy their deficiencies on the job.
Occupations that require the greatest amount of formal, company-sponsored training—managerial and professional or technical jobs—also require the greatest amount of training in schools. This further accentuates the differences among employees.
Since new entrants to the labor market are likely to change jobs frequently, employers invest less in training employees during their first five years in the labor market. As workers demonstrate their commitment, employers increase their training. At the other end of the scale, employers invest less in older workers because they can recapture less of their investment.
Employers are less likely to train non-white than white employees. Employers trained 39 percent of white employees and only 27 percent of non-white employees during their first 13 years in the labor force. Racial differences were even greater for management, professional, and technical training: 19 percent of whites received managerial training, compared with only 8 percent of non-whites; 44 percent of whites received professional and technical training, but only 24 percent of non-whites.
Even when the analysis controls for many observable worker and industry characteristics, older white male workers are still more likely than older black male workers to receive employer-sponsored training. Racial differences are much reduced for younger workers. However, the gap between black and white female workers has closed—they have equal chances of receiving employer training.
But women are less likely to receive company training than men. Even well-educated women are less likely to receive employer training than comparably-educated men. Female college graduates are only 137 percent more likely than high school graduates to receive additional training on the job, but male college graduates are 148 percent more likely than high school graduates to receive such training.
The sectors that are growing rapidly—financial services, public administration, and professional services—employ better-educated workers than agriculture, mining, wholesale trade, and manufacturing. The overall shift of employment from goods production to services means that more education and training are needed both to get jobs and to keep them.
For companies, the benefits of training are substantial. The productivity of trained workers increases by about twice as much as wages. Workers who receive training are less likely to leave the company. Workers who exhibited prior mobility exhibit less mobility once they receive company training. Trained workers also change jobs within their company less often than untrained workers.
For employees with less than 12 years of work experience, trained workers enjoy wages that are almost ten percent higher than wages of untrained workers, controlling on other wage-relevant characteristics. For those with more than 12 years experience, trained workers receive about 3.5 percent higher wages than untrained workers. The impact of training on wages endures for over ten years. Since employers select the employees who receive training, and since company-sponsored training can be more job-specific, employer training raises wages more than post-secondary training.
Well-educated and trained workers are less likely to be laid off and experience shorter periods of unemployment when they are. Workers with less than 12 years of schooling are 170 percent more likely to experience periods of unemployment and suffer spells of unemployment that are 30 percent longer than workers with 16 or more years of schooling. Employer training further reduces the probability and duration of unemployment. The reduced likelihood of unemployment for those with company training is noticeable for 12 years after training.
Training reinforces the differences among employees. Well-educated people are the most likely to find employment and to receive training from their employers. Once trained, they earn more, switch jobs less frequently, and are rarely or briefly unemployed. If they change jobs, they find another one more easily and are likely to receive further training from their new employer. Those who enter the labor force lacking sound academic and problem-solving skills fall further and further behind.
Since World War II, the replacements for each generation of retiring workers have been drawn from a larger and better-trained generation of workers. Today, however, new entrants are fewer and the growth in education has slowed. In 1970, the workforce was growing at about 2.5 percent annually. Today, its rate of growth is less than one percent annually, and continues to fall. For increasingly complex jobs, employers have fewer qualified applicants. The U.S. birth rate is falling, and the workforce is aging. Between 1985 and 2000, population growth will be mainly in the older ages. The number of people between ages 35 and 47 will grow 38 percent, while the number between 48 and 53 will grow 67 percent. The number of 16- to 19-year-olds in the workforce is declining absolutely, and the share of the workforce held by people between the ages of 20 and 34 is falling. In the future, therefore, a growing share of the new skills needed in industry will have to be met by retraining existing workers, not by new hires.
A variety of forces have displaced jobs requiring little education and created jobs that require higher education and skill levels. The most potent force is the expanded importance of international trade. The U.S. produces less of what it can acquire relatively cheaply abroad and increases its production of goods and services in which it enjoys a comparative but only a temporary advantage. When new nations enter the trading arena and other nations compete, the advantage can be lost. Since these shifts are unpredictable, international trade creates an imperative: adapt quickly or fall behind. Shifts in trade have eliminated many well-paid jobs for uneducated workers, particularly among non-whites, and created new jobs requiring more extensive formal education in high technology industries and in professional services. As a result, the rates of return from education have risen.
New technologies have not, as many feared, reduced overall employment. After a century of technological progress, a higher share of the population is employed today and is earning more than at any time in our history. However, technological progress has reduced the number of jobs in goods-producing activities, increased the relative importance of higher-skill occupations within sectors, and broadened skill requirements within occupations.
First, it has redistributed employment among sectors, expanding opportunities in services while reducing jobs in manufacturing.
Second, within sectors, technology has reduced the need for unskilled and semi-skilled employees and increased the need for technically trained people. In automobile production, computer-assisted design and manufacturing has halved manual labor's share of the jobs while doubling the number of technical jobs.
Finally, within occupations, new technology has demanded new competencies and broadened skill requirements. A secretary, for example, must operate complex communications and data processing equipment.
In response to accelerating technological change, employers first hire people with more, and more recent, education. Better-educated workers appear better able to deal with technical problems and with the unstable environment created by rapid technological change. But hiring highly-educated workers is expensive: their general skills command higher wages. Therefore, when new technologies become routinized, employers can be more selective and specific in the skills that they hire. At that point, firms expand training and hire fewer well-educated, and expensive, workers.
A hidden consequence of computer-based technological change is the decentralization of economic activity. Computer-based technologies are flexible, allowing firms to produce more diversified and customized products. New businesses enter markets easily, and large corporations contract out for more of the products and services that they need. For the three decades following World War II, the Fortune 1000 companies employed nearly two-thirds of the workforce. Today, those companies employ only one-third of the workforce. Three-quarters of the people entering the workforce today find their first job in a company with less than 100 employees. But new, small businesses are more volatile employers than large businesses. People must be able to recycle their skills faster than in the past. In fact, many of today's workers will get on only by broadening their workplace skills so they can operate their own businesses.
Both service and manufacturing industries have moved from a production-oriented world to a customer-oriented world, and from mass production to flexible production. These changes in orientation demand fast retooling, shorter production runs, and customized production and generate the need for different, better, and more generic skills, such as:
Computers in the workplace force a replacement of observational learning with learning acquired through symbols, either verbal or mathematical. For example, a family of technological systems known as manufacturing resource planning (MRP) allows American businesses to compete. MRP systems support such manufacturing innovations as "just-in-time" inventory and small-batch custom production. MRP, which is spreading from large corporations to middle-and small-size firms, is a content-free, formal, closed conceptual system that workers at all skill levels within the firm have to use. It more nearly resembles "school" subjects such as math or grammar than the traditional systems of knowledge accumulated by experienced business or production managers.
The shift to flexible production and changes in the time-frame for production combine to increase the need for higher-order cognitive thinking even for jobs that are usually thought of as lower skill. Speed of response is an important competitive weapon. Since flexible production multiplies the number of decisions that must be made, the need to respond quickly means that decisions can no longer be bucked up supervisory lines but may be made on the shop floor level. Employees in both higher and lower skill jobs are required to deal with uncertainty, discontinuity, and the unfamiliar.
All these changes eliminate supervisory and middle management positions. Previously-supervised workers are required to make decisions previously handled by supervisors and regulate and direct themselves.
The resulting volatility in job tasks implies the need to know how to learn—how to identify and organize resources, how to ask germane questions, and how to penetrate poor documentation.
Under mass production, employees often worked alone, albeit in physical proximity to each other. Broadening job responsibilities requires collaboration among workers, and the need for collaboration generates the further need for communication and conflict-resolution skills.
The growing acknowledgement of the economic importance of education demands a reexamination of all aspects of federal and state educational policy. What should be the public policy toward employer-sponsored training? Should it be subsidized?
Current federal and state tax codes already subsidize employer-provided training. The costs of training can be expensed (written off when they are incurred), while investments in plant and equipment must be depreciated (written off over time). Estimates of the value of this subsidy (expensing) range between $13.2 billion and $58.3 billion annually.
One-third of employer-trained workers are enrolled in external, post-secondary education or training institutions. The fees employers pay to most of these trainees are much less than the full cost of training. There are no data on the true costs of public programs (public agencies rarely include depreciation, for example) to compare with the revenues received for customized training programs. Public benefits from employer-sponsored training, beyond those enjoyed by employers and their employees, are hard to demonstrate. The case for further public support of employer-sponsored training rests on five flawed arguments:
If it is true that the U.S. saves and invests too little, then it is true of all assets, not merely of human capital on the job. The appropriate policy would be to subsidize all productive investments, not just one asset already heavily subsidized.
The best guide to the appropriateness of employers' investments is the rate of return earned by those investments. Unfortunately, we cannot measure the rate of return from training because we do not have any good measures of the costs of the investments. We are therefore unable, statistically, to identify which firms or industries systematically under-invest in training or which types of employees might profitably use additional training.
An association between higher rates of training and higher rates of economic growth—as in major Japanese manufacturing firms—does not necessarily mean that increased training increases economic growth. Training must be connected to new economic opportunities. These new opportunities may be new technologies, new markets, or new products. Training in the absence of new economic opportunities will have no economic impact.
Employers already have many ways of recapturing their investments in training, and the lower mobility rates for employer-trained workers show that these measures are successful.
Regarding these two arguments, the evidence is clear—further subsidies would increase rather than reduce inequalities since those trained by employers are already relatively well-educated and economically advantaged. Further, employers train firm-specific skills that have limited use outside of the firm.
Even if strong evidence for under- or mis-investment could be shown, and policymakers wanted to subsidize employers' investments, it would be difficult to direct those subsidies effectively. The market already provides a more efficient corrective to companies' training investments mistakes than public policy can. Employers who consistently under- or mis-invest will succumb to the superior productivity and flexibility of competitors who make better training investments.
Because employer-sponsored training builds upon rather than fills in for skills learned in school, the best way to encourage employers to invest more in training is to insure that those who leave school possess stronger verbal, quantitative, and problem-solving skills.
Public post-secondary training institutions already provide a large share of employer-sponsored training, and demographic and economic changes indicate that this role will grow in importance. States should give fiscal and administrative incentives to their post-secondary institutions to be more responsive, such as letting educational institutions keep the proceeds from fees charged employers for customized training.
Those who do not start the labor market race with good skills fall further and further behind. For the most part, employers will not remedy their deficiencies. Special training programs for the disadvantaged, such as the Job Training Partnership Act, have not provided training in fundamental skills. These programs give students the minimal skills required for the minimum placement. They equip graduates for the bottom rungs of the ladder—the most unstable rungs, in perpetual risk of being automated or organized out of existence. In today's economy, moving off the bottom rungs requires solid academic, problem-solving, and teamwork skills. To stop the ominous evolution of a dual labor market, public training programs and post-secondary occupational training programs need to focus more on developing stronger academic skills in their disadvantaged clients. This process cannot be completed as quickly as giving an individual minimal preparation for a minimal placement.
We end with five questions: First, do federal and state student grant and loan programs adequately cover the costs of post-secondary training for disadvantaged students? Second, can current grant and loan programs be used to finance remedial education? Third, how might post-secondary institutions' eligibility for student grants and loans be tied to their record in remedial education? Fourth, how might institutions' success rates in equipping students with solid academic and problem-solving skills be publicized to inform students' choices? Fifth, what are the academic and problem-solving skill payoffs of integrating remedial and occupational training?
This NCEE Brief summarizes the papers and discussions presented in December 1988 at a conference on employer-sponsored training held in Alexandria, Virginia. The conference was funded by the National Assessment of Vocational Education of the U.S. Department of Education, and organized by the Institute on Education and the Economy, Teachers College, Columbia University. Several of the conference papers were based on research supported by a grant from the Office of Educational Research and Improvement of the U.S. Department of Education.
Bailey, T. Changes in the Nature and Structure of Work: Implications for Employer-Sponsored Training.
Bartel, A. P. Utilizing Corporate Survey Data to Study Investments in Employee Training and Development.
Mincer, J. Labor Market Effects of Human Capital and of Its Adjustment to Technological Change.
Noyelle, T. Skills, Skill Formation, Productivity and Competitiveness: A Cross-National Comparison of Banks and Insurance Carriers in Five Advanced Economies.
Tan, H. Private Sector Training in the United States: Who Gets It and Why.
Vaughan, R. J. Public Subsidies and Private Training.
Vaughan, R. J., & Berryman, S. E. Employer-Sponsored Training: Current Status, Future Possibilities. Synthesis of Findings.
Institute on Education and the Economy, Teachers College, Columbia University
525 West 120th Street, Box 174, New York, NY 10027
Phone: (212) 678-3091 | Fax: (212) 678-3699 | email@example.com