A r c h i v e d  I n f o r m a t i o n

Are Employers' Recruitment Strategies Changing:
Competence Over Credentials?

Laurie J. Bassi
Vice President, Research
American Society for Training and Development

The field cannot well be seen from within the field.
— Ralph Waldo Emerson

Human capital, mercifully, is the only form of capital that corporate America cannot buy or sell. As a result, it is the only form of capital that does not have well-defined (if admittedly grossly imperfect) accounting procedures and reporting rules associated with it. In part because of this, firms' procedures for assessing and valuing the competencies of prospective and incumbent employees range from primitive to nonexistent. Another factor contributing to the sorry state of assessing workplace competence is that in times of rapid change (such as we are currently in), employers' concepts of what constitutes competence also change rapidly; these concepts, however, are rarely well defined and articulated. Still another factor complicating the difficulty of assessing and valuing workers' competencies is that the methods by which these competencies can be created are expanding faster than is our collective capacity for evaluating the effectiveness of these methods. Finally, human beings are extraordinarily complex; participating in well-honed, highly effective learning experiences is no guarantee that any given participant actually learns anything at all.

In sum, there are many reasons why assessing competence is inherently difficult and becoming even more difficult with the passage of time. There are also reasons to believe that the imperative to tackle this difficult task is becoming more pressing.

This paper examines evidence from a variety of perspectives that are relevant to understanding the changing nature of workplace competence. Four perspectives are considered: the operation of the labor market, surveys of what employers say, the market for employer-provided training, and the stock market.

What the Labor Market Is Telling Us

Over the past 25 years, 2 striking trends have dominated all others in the labor market. First, real (i.e., inflation-adjusted) wages have fallen for the majority of workers. Second, dispersion of wages—both within and between groups—has increased.

Historically, real wages have been driven by productivity changes.¹ The period since 1973 has been no exception. Between 1961 and 1972, productivity grew at an annual rate of slightly more than 3 percent; between 1973 and 1994 the productivity growth rate fell by more than 60 percent, to an annual rate of less than 1.2 percent.²

Consequently, understanding the productivity slowdown is tantamount to understanding the real wage decline. Unfortunately, while a wide variety of explanations have been put forth, the slowdown in U.S. productivity remains largely unexplained. What is clear, however, is that widely heralded changes in technology have not yet manifested themselves either in growth of productivity or wages.

The combined effect of the increase in wage dispersion, along with the decline in real wages has been most pronounced for males. The real hourly earnings of men with less than a high school education, for example, fell by 28 percent between 1973 and 1995. This decline has been most pronounced for young men.

The only groups that have enjoyed increases in real earnings are women with at least a college education and men with advanced degrees (see table 1).

Table 1.— Percent change in real hourly wage by education: 1973-1995

Source: Lawrence Mishel, Jared Bernstein, and John Schmitt. The State of Working America 1996-97. Armonk: M.E. Sharpe, Inc., 1997.

These shifts in earnings patterns have been occurring at the same time that the average education level was increasing quite sharply. For example, between 1975 and 1994, the percentage of adults (age 25 years old and over) who had graduated from high school rose from 63 to 81 percent with the corresponding figures from college being 14 and 22 percent (U.S. Bureau of the Census, Statistical Abstract of the United States 1995, page 157). In other words, the supply of highly educated workers increased. Since the relative (and in the case of women, the absolute) wages of college graduates rose as the supply rose, demand for educated workers must have risen even faster than their supply; otherwise the wages of highly educated workers would have fallen.

It is less clear, however, what is behind changes in demand for highly educated workers. Like the mystery of the slowdown in productivity growth rates, there is no widely accepted explanation of what accounts for the rise in U.S. wage inequality. Outcomes in the labor market are a complicated interaction of not only supply and demand but also institutional factors (e.g., the minimum wage and unionization). Table 2 summarizes the results of a recent survey (conducted by Alan Krueger as a part of the New York Federal Reserve Colloquium) in which a group of economists was asked to assign responsibility for rising wage inequality to a variety of factors (table 2).³

Table 2.— Causes for the rise in U.S. wage inequality? (In Percent)

Table 2 indicates that, as a group, economists believe that technological change (which affects the demand for various groups of workers) is the single most important source of rising inequality of wages, accounting for an estimated 40 percent of the increase. Looked at more expansively, however, even more of the increase in inequality might be attributable to technological change. Richard Coopers, for example, argues that trade (i.e., increasing globalization) is inseparable from technological change.⁴ Similarly, at least part of the decline in unionization is undoubtedly attributable to trade, and therefore, by extension to technological change. Adding together the effects of these forces suggests that technological change might account for as much as 60 percent (40 + 11 + 9) of the rise in wage inequality in the United States.

In sum, the consensus of the economics profession is that technology has had a very significant (perhaps even unprecedented) effect on the demand for highly educated workers.

What these statistics do not tell us, however, is why this is so. Unfortunately, very little is known on a systematic basis that enables us to identify in a rigorous manner exactly what is behind the shift in demand for educated workers. Levy and Murnane have, however, uncovered one tantalizing shred of empirical evidence. And that is that across occupations, the return (in the form of a wage premium) on mathematical skill has increased significantly over the time period under consideration (Levy and Murnane 1992). Since it is unlikely that the use of mathematics has increased significantly over all occupations, this suggests that mathematics skills may be a proxy for some other skill increasingly in demand by employers, perhaps the ability to solve problems (which is, essentially, what mathematics teaches).

Another tantalizing fact that has emerged is that the increased dispersion in earnings across age/gender/education groups has been accompanied by an equally large increase in the dispersion of wages within each of these groups (Levy and Murnane 1992). That is to say, while education is an increasingly important determinant of wages, so too is something else. But we don't know what that something else is. There are many possible explanations. It could be that the return on competence has increased. Or it could be that in an economy that is undergoing rapid change there is an increasing payoff to being in the right place at the right time (i.e., luck).

A recent analysis of wage patterns over the course of decades (Autor, Katz, and Kreuger 1997) has detected patterns between people's use of computers at work and variations in wages. The authors conclude that "computer technology may `explain' as much as 30 to 50 percent of the increase in the rate of growth of the wage-bill share of more-skilled workers since 1970."⁵ Consequently, use of computers at work "explains" both the increase in wage dispersion across educational groups as well as within educational groups.⁶

Taken together, these labor market trends suggest the following conclusions. First, education credentials are an increasingly important determinant of demand for labor, which in turn, affects wages. Second, it is not clear whether employers are increasingly relying on education credentials as a method of screening for the skills they need, or whether these credentials are merely a proxy of increasing importance for some necessary skill (perhaps the ability to learn quickly). Third, the demand (as evidenced by the growing wage premium) for mathematics skills has grown. It may be that these skills serve as a proxy for some other important skill (such as problem-solving ability). Fourth, since wage inequality has also increased within educational categories, some aspect of supply and demand (above and beyond educational credentials) is at work in the labor market. This could be some unmeasured competencies. Or it could be luck. Or it could be something else. Fifth, the use of computers is likely to be an important part of the "something else." Finally, it is simultaneously true that both educational credentials and something beyond educational credentials have become increasingly important in determining employers' demand for workers, and therefore, the wages that workers earn. It is likely that both competence and credentials are increasingly in demand by employers.

To paraphrase Emerson, it isn't easy to figure out what's going on when you are in the middle of what's going on ("The field cannot well be seen from within the field"). As the conclusions outlined above make clear, an analysis of labor market data can reveal much about what is happening, but little about why things happen. The review of opinion surveys that follows represents one method for developing hypotheses about why these labor market trends are occurring.

What Employers Are Telling Us

Insight sometimes has to be gained at the expense of rigor. Although simply asking people to share their opinions about why things are happening the way they are is not terribly rigorous, nonetheless it can be insightful. The advantage of this approach is that it is possible to get much more timely and detailed information than that which can be gleaned from the operation of the labor market. The disadvantages of this approach are: (1) it is based on a limited sample (of either employers or employees) and (2) people's perceptions and their actions (i.e., the actual choices they make in the labor market) are not necessarily entirely consistent with one another.

Nonetheless, in survey after survey about skill requirements, employers report skill deficiencies. The severity of these complaints varies over the business cycle, inevitably worsening as the labor market tightens. What is not clear from these surveys is whether employers' complaints about skill deficiencies have become increasingly severe over the course of recent decades, or whether these complaints merely reflect employers' age-old complaints that workers' wages are higher than employers would prefer to pay.

Surveys of employers consistently find that computational and problem-solving skills are among the most serious "hard" skill deficiencies, although inadequate writing and verbal skills are also reported to be a problem. Among entry level workers, however, deficiencies in interpersonal skills, poor attitudes toward work, and difficulty "fitting in" are consistently reported as serious (chapter 5, Cappelli et al 1997).

The results of a recent survey sponsored by the Society for Human Resource Management (SHRM) are reported in table 3 (Aon Consulting and SHRM, The 1997 Survey of Human Resource Trends Report, U.S. edition, 1997).

In general, employer surveys suggest that employers are experiencing a fair to moderate degree of difficulty in filling entry level jobs, and somewhat more difficulty in filling higher level jobs that require specific skills (see table 3).

Table 3.— Percent of firms that rated competencies as "weak" among nonmanagement job applicants

Source: Aon Consulting and the Society for Human Resource Management, The 1997 Survey of Human Resource Trends Report.

The range of tools that employers use to help them select among job applicants are summarized below in table 4 (based on SHRM survey). As is typical in this type of survey, the response rate was low — 12 percent. The total usable sample was 1,700. No analysis was presented of differences between respondents and non-respondents, but it is clear that large firms are over-represented in the sample. As a result, it is likely that the respondent firms are more sophisticated in their human resource policies and strategies than a representative sample of firms.

Table 4.— Prevalence of selection procedure use for nonmanagement and management candidates (percent of firms that report using the certain techniques)

Despite the probable nonrepresentativeness of the sample, it is nonetheless of interest that a majority of responding firms (55 percent) report using skills tests to assess nonmanagement job candidates. A variety of other methods for assessing competencies (realistic job previews, job simulations, and assessment centers) are used by fewer employers. Unfortunately, there is no time-series data on the prevalence of these practices. As a result, it is not known whether their use is rising or falling. Nonetheless, the results suggest that firms use a variety of methods (above and beyond screening education credentials) to assess competence.

Interestingly enough, the SHRM survey results indicate that these methods may have only a modest effect on employers' satisfaction with the effectiveness of their selection procedures. These results, summarized in table 5, indicate that using skills tests increases by 14 percent the probability that employers report that they are satisfied with their selection procedure. (Note that the independent effect of skills test on satisfaction must be inferred from table 5 by solving a set of simultaneous equations, which yields the estimate of 14 percent.)⁷

Given the difficulty and expense often involved in developing and implementing valid skills tests, an incremental 14 percent increase in satisfaction with these (probably expensive) selection procedures is somewhat disappointing. This suggests that employers may face a substantial challenge if more and more skills are acquired through nontraditional means (such as the Internet), and fewer skills can be verified through traditional methods (such as checking credentials). On the other hand, the finding that only 36 percent of employers report they are satisfied with the effectiveness of their traditional basic selection tools (which includes checking credentials) indicates that employers are less than satisfied with traditional measures of assessing job applicants' qualifications (see table 5).

Table 5.— Percent of firms reporting satisfaction with selection procedure effectiveness as a function of procedures used

What the Training Market Is Telling Us⁸

Private employers spent approximately 1.8 percent of their payroll on training in 1995. This spending was split nearly equally between direct and indirect costs (compensation of employees while in training), with each amounting to approximately 0.9 percent of payroll.

The following listing ranks industries from high to low according to their spending on formal employer-provided training: transportation, communications, and public utilities; mining, finance, insurance, and real estate; nondurable manufacturing; wholesale trade; durable manufacturing; services; construction; and retail trades.

Household survey data indicates that in 1995, 16 percent of the civilian workforce—approximately 20 million workers—reported that they received formal employer-provided training within the previous year.⁹ Workers with high levels of education are more likely to receive employer-provided training than are workers with low levels of education. Mature workers are more likely to receive training than are younger workers. Men are somewhat more likely to receive training than are women. Whites are more likely to receive training than are minorities.

Throughout the 1980s and 1990s, spending on employer-provided training appears to have increased faster than the rate of inflation. This increase in spending, however, does not appear to have kept pace with the increase in the number of workers. Consequently, training expenditures per worker have fallen somewhat.

At the same time, the percentage of workers receiving employer-provided (or required) training has increased sharply, growing from 5 percent of workers in 1981 to 16 percent of workers in 1995 (once again, based on household surveys). This suggests that expenditures per worker trained have fallen significantly.

One possible explanation for at least a portion of this decline is that as more workers are trained, the type of worker being trained has undoubtedly changed. For example, if more production workers are being trained and this training is relatively inexpensive, this change could result in lower costs per worker trained.

Figure 1. — Training time by delivery system

Source: ASTD's 1997 Human Performance Practices Report.

Technology is another possible explanation for the decline in expenditures per worker trained. As technology has enabled training to be delivered through low-cost methods such as electronic performance support systems and computer-based training, the cost per worker trained would be expected to fall.

Unfortunately, there is very little reliable time-series evidence on the extent to which employers use learning technologies to deliver education and training to their workforces. A recent survey done by the American Society for Training and Development (ASTD) (figure 1), found that in 1996, 84 percent of all training time occurred in instructor-led classroom settings. Seven percent of all training was delivered through self-paced instruction, 6 percent through computer-based training, and 3 percent "other."

The percentage of companies that use a variety of instructional media is summarized in figure 2. Only a tiny fraction of companies (less than 3 percent) report using either the Internet or intranet as a vehicle for delivering training (see figure 2).

Figure 2.—Percent of companies using instructional media

Large firms are significantly more likely to rely on computer-based training than are smaller firms. But a more important determinant of firms' choice of the systems they use to deliver training is a variety of innovative training and human resource practices; the more innovative practices a firm uses, the more likely it is to rely heavily on alternatives to traditional classroom training.

Figure 3 summarizes how survey respondents answered the question "How will your organization's total training time delivered by the following technologies change from 1996 to 2000?" (See figure below)

Figure 3.—Predicted increase in media*

Putting the results of figures 2 and 3 together suggests that the increase in the use of electronic technologies for delivering workplace training will be modest over the next 3 years. (Even though some of the predicted increases in percentages are fairly large, the total change is predicted to be small because the current usage is so small.) It remains to be seen, of course, whether or not these predictions will become reality.

The limited time-series evidence available on employers' use of electronic learning technologies to deliver education and training to their workers suggests that change (at least in the recent past) has not been as sweeping as one might believe from reading the popular press. ASTD has been collecting data from a group of large (primarily Fortune 500) firms over the past few years. In 1994, this group of firms delivered 78 percent of all their training and education via traditional classroom training. In 1995, this amount declined to 69 percent, but in 1996 remained virtually unchanged at 70 percent (table 6).

Table 6. — Delivery systems, by percent of training time: 1994-1996

What can be said with certainty at this point is that the vast majority of employer-provided training is currently delivered via traditional classroom training. The use of computer-based technology is rising, albeit slowly. CBT and distance learning still represent a minor weapon in the training arsenal. Only a tiny fraction of training is delivered via the Internet or intranets, although its rate of growth is impressive (more than quadrupling in 2 years from a base of 0.4 percent in 1994). Although employers are not predicting sweeping change in the short term, over the long term it is possible that we will witness a significant change in the relative balance between computer-based, distance education and classroom instruction.

What the Stock Market Is Telling Us¹⁰

Wall Street, however, appears to be betting that sweeping changes in the delivery of education and training is, indeed, on the horizon. The owners of a number of specialized training firms have recently made their fortunes by taking their previously private firms public. These initial public offerings are both capitalizing on and fueling the growth of technology-based education and training companies.

Computer Learning Centers, for example, is taking advantage of the available opportunities in the corporate world. It has recently spun off a new division, Advantec Institute, which offers specialized computer classes geared toward employers' interests. Based on the positive response to Advantec thus far, and the realization that the demand for these training services is greater than the supply, Computer Learning Centers management anticipates that the firm will continue its expansion into the domain of corporate training, particularly in the area of providing customized training packages for companies.

Like Computer Learning Centers, other education and training companies are emerging in corporate markets, and in doing so, are capturing the interest of Wall Street. CBT Systems, Learning Tree International, National Education Corporation, and Westcott communications, among others, have received heightened attention from business investors after recently making public stock offerings.

Smith Barney now advises its clients that "the growing importance of intellectual capital" suggests that a well-balanced portfolio should include investments in the emerging sector of publicly traded education and training providers. Smith Barney segments its coverage of this sector into three groups: (1) educational management organizations, (2) training and development providers, and (3) instructional media companies. With regard to the latter two, Smith Barney bases its advice to investors on what it sees as six big drivers:

• the evolution from manufacturing-based to knowledge-based economy,

• the ubiquitous nature of technology,

• changes in the workplace,

• the advance of communications technology,

• the trend toward outsourcing, and

• the advent of a global economy.

In a presentation at the annual conference of the Instructional Systems Association in March 1997, Charles Hall (Smith Barney's director, Education Group) cited evidence that over the past 2 years, a composite of education and training stocks had appreciated in value at more than twice the rate of S&P industrials.

Nor is Smith Barney alone in its interest in the education and training sector. Goldman Sachs, Montgomery Securities, and Piper Jaffrey have also developed education and training investment lines, and others are poised to enter the market. Warren Buffett, a renowned investor, has taken an active interest in the for-profit training and education industry by buying FlightSafety International, Inc., one of the largest training companies in the world. Michael Milken has invested $125 million of his own money in Knowledge Universe, a training and consulting company which, by year end, is expected to reach $1 billion in sales.

It appears that the investment community has discovered the dawn of the age of knowledge, and subsequently recognized the significance of the training and development industry. These companies' stocks are selling for 30 to 50 times their expected yearly earnings per share, a phenomenon that also characterizes the behavior of technology stocks.

Wall Street, of course, has been known to make bad bets. Only time will tell if this is one of them. But the interest that the investment community has begun to show in the education and training sector, nonetheless, suggests an underlying emerging trend.

If Wall Street is right, then the need that companies have to assess competencies without credentials will only continue to grow as more and more education and training is provided through nontraditional, uncertified methods. Indeed, in private conversations with Wall Street research analysts in the education and training sector, company representatives concede that the absence of systems for assessing the value of education and training delivered through nontraditional means could well impede the growth of this emerging sector.

If, however, Wall Street is wrong, then not only may this sector fail to grow as quickly as many expect, it could disappear entirely from the horizon. It is certainly possible that a good deal of the media attention being paid to distance learning is fueled by Wall Street's current interest. Should profits in this sector fail to realize Wall Street's expectations for them, then interest in this sector could dissipate quickly. Only time will tell.

Summary and Conclusions

By all accounts, the importance of human capital to our individual and collective well-being will continue to grow into the foreseeable future. The threefold increase between 1981 and 1995 in the percentage of workers who reported that they received formal, employer-provided training speaks to the increasing significance of human capital within the world of work.

Given its increasing centrality to corporate profitability, it is predictable that the task of producing and enhancing human capital will increasingly occur outside traditional classroom settings. Community colleges, universities, and for-profit education and training providers are increasingly in the business of providing education and training to corporate America. By facilitating high-quality, cost-effective delivery of education and training, advances in technology could enable these opportunities to be made much more widely available than has heretofore been possible. To date, however, this possibility has yet to be realized on a broad-based scale. The vast majority of employer-provided training is still delivered through traditional classroom training, and employers are predicting that this will remain the case over the course of the next few years. The stock market, however, is betting that there will be a surge in the electronic delivery of education and training will surge.

If Wall Street is correct, then employers, who report that they are already less than content with their methods of assessing competence, will have to become more creative than in the past. Indeed, analysts on Wall Street already understand that the absence of a broadly accepted system for assessing the extent to which technology-based education and training enhances competence may well inhibit the growth of this sector of the economy.

This is not a task that employers working in isolation will be able to solve optimally. Rather, some collective action will be necessary if the full potential for this market to distribute learning opportunities is to be realized. This suggests a role for public policy.

Moreover, quite apart from the concerns of Wall Street and employers are those of individuals. The labor market analysis summarized herein confirms the by-now well-documented increase in the economic return on education credentials, and suggests that the return on undocumented competence is also on the rise. Consequently, individuals who are seeking to invest in their own education and training have a growing need for information that enables them to make well-informed purchasing decisions with regard to the effectiveness of an increasingly complex array of education and training options. The marketplace will almost surely fail to produce high quality, reliable information of this type—once again suggesting an important role for public policy.

Notes

1. While economists generally agree on the trend in real wages, other observers dispute it. Some have argued that the downward trend in real wages is overstated because the CPI (the Consumer Price Index), used to adjust for inflation, overstates the true rate of inflation. However, to the extent that the CPI has always overstated inflation, this argument does not account for the downturn that took place in 1973. Others argue that the trend in compensation (wages plus fringe benefits) has been less pronounced than the trend in wages (i.e., increases in fringe benefits have partially or totally offset the trend in wages). According to this perspective, wages have merely stagnated, rather than declined.

2. U.S. Government, Economic Report of the President (February 1997). As is the case with the wage trend, the low estimates of U.S. productivity growth rates have been challenged by some observers on the grounds that they are mismeasured — understating the true rate of productivity growth. [In fact, because the measurement of inflation and productivity are essentially codetermined, this argument is essentially equivalent to arguing that inflation is overestimated.] While mainstream economists generally agree that productivity growth rates are likely to be underestimated (perhaps by 1 percentage point per year), this mismeasurement cannot explain the post-1973 trends unless: (1) mismeasurement became significantly worse in 1973, and (2) has continued to worsen with the passage of time.
3. Each respondent was asked to assign a weight to these factors so that the factors summed to 100 percent.

4. Comments made at an Aspen Institute Conference on Work and the Future of Society, August 1997.

5. It should be noted that these authors use the term "more-skilled" to mean more highly educated. Autor, Katz, and Kreuger use the term "explains" with caution, noting that their data reveal significant correlations, but cannot prove causality.

6. Computer usage contributes to dispersion in wages across educational groups both because the likelihood of using computers at work increases with education, and because the wage premium to computer usage increases with education. Computer usage contributes to dispersion of wages within educational groups because of the wage premium that accrues to those who use them.

7. Let B=basic selection tools, A=structured applications, I=structured interviews, S=skill tests. Table 5 says that:
   (1) B=36
   (2) B + A + I = 58 = ==> A + I =58 - 36 = 22
   (3) B + A + S = 61 = ==> A + S = 61 - 36 = 25
   (4) B + I + S = 61 = ==> I + S = 61 - 36 = 25
   Equations (3) and (4) imply that A = I, and consequently equation (1) implies that A = 11, which can be used to solve for S = 14 in equation (3).

8. This section draws upon: Laurie J. Bassi, Anne L. Gallagher, and Ed Schroer, The ASTD Training Data Book (1996); Laurie J. Bassi, George S. Benson, Mark E. Van Buren, and Rosio Bugarin, The 1997 Human Performance Practices Report; Laurie J. Bassi and Scott Cheney, Benchmarks: Tools for Continuous Improvement (November 1997).

9. U.S. Department of Labor, Bureau of Labor Statistics Report (December 1996): 13. Recent evidence from the Bureau.
10. This section is taken from: "Trends That Affect Learning and Performance Improvement" (forthcoming), In Training & Development.Bassi, Laurie J., Scott Cheney, and Mark Van Buren.

References

Aon Consulting and the Society for Human Resource Management. The 1997 Survey of Human Resource Trends Report, United States Edition. Detroit: Aon Consulting, Inc., 1997.

ASTD Benchmarking Forum. 1997 Benchmarking Forum Comparative Report. August 1997.

Autor, David H., Laurence F. Katz, and Alan B. Kreuger. "Computing Inequality: Have Computers Changed the Labor Market?" Working Paper #377, Industrial Relations Section, Princeton University, March 1997.

Bassi, Laurie J. and Scott Cheney. "Benchmarks: Tools for Continuous Improvement." In Training and Development, November 1997.

Bassi, Laurie J., Scott Cheney, and Mark Van Buren. "Trends That Affect Learning and Performance Improvement," In Training and Development, November 1997.

Bassi, Laurie J., George S. Benson, Mark E. Van Buren, and Rosio Bugarin. 1997 Human Performance Practices Report, Executive Summary. Alexandria, VA: ASTD, 1997.

Bassi, Laurie J., Anne L. Gallagher, and Ed Schroer. The ASTD Training Data Book. Alexandria, VA: ASTD, 1996.

Cappelli, Peter, Laurie Bassi, Harry Katz, David Knoke, Paul Osterman, and Michael Useem. Change at Work. New York: Oxford University Press, 1997.

Levy, Frank, and Richard J. Murnane. "U.S. Earnings Levels and Earnings Inequality: A Review of Recent Trends and Proposed Explanations." In Journal of Economic Literature, volume xxx, no. 3, September 1992. Kingsport: American Economic Association, 1992.

Mishel, Lawrence, Jared Bernstein, and John Schmitt. The State of Working America 1996_97. Armonk: M.E. Sharpe, Inc., 1997.

U.S. Bureau of the Census, Statistical Abstract of the United States: 1995 (115th edition) Washington, DC, September 1995.

U.S. Department of Labor. Bureau of Labor Statistics Report on the Amount of Formal and Informal Training Received by Employees. Washington, DC: USDL, December 1996.

U.S. Government. Economic Report of the President. Washington, DC : U.S. Government Printing Office, February 1997.of Labor Statistics (BLS), however, suggests that the percentage of employees who receive formal education and training at work may be much higher. When asked to record any activity that enabled them to do their jobs better (which BLS then coded as either formal or informal training), a sample of more than 1,000 employees indicated that 70 percent of employees received formal training); American Society for Training and Development (ASTD) 1997 Human Performance Practices Survey (1997). In a recent survey done by ASTD, employers reported that 58 percent of their workers received formal training in 1996.

-###-

[Competence Without Credentials:. . .]

[High Tech vs. High Touch; The Potential Promise. . .]