The IT Labor Shortage: Fact or Fiction?

These debates sometimes make dubious uses of statistical information. With this problem in mind, in 1998, the United Engineering and Alfred P. Sloan Foundations arranged for a small project to provide better guidance to the statistical facts about IT careers. The goal of the IT Workforce Data Project was to identify and disseminate authoritative facts about information technology job markets. The project has produced four brief reports:

• The overall size and makeup of the IT workforce.
• The production of degrees in IT specialties.
• The role of foreign workers in U.S. IT jobs.
• An assessment of supply and demand (the "shortage" question).

This article summarizes the findings of these reports and examines some of their implications for people in the profession. The full text of the reports is available online at http://www.uefoundation.org/ itworkfp.html.

Conclusions of the IT Workforce Data Project

Reliable data from the Bureau of Labor Statistics, Bureau of the Census, National Center for Educational Statistics, and National Science Foundation provide a consistent picture of professional employment in information technology, centered on a group of core occupational specialties -- computer scientists, computer engineers, and systems analysts. Programmers were added to this group; federal databases treat programming as a technician's field, not a professional specialty, but the industry's practice of using terms like "programmer" and "software engineer" interchangeably means that programmers should be included. In some cases (for example, in assessments of degree production), electrical engineers with specific computer hardware or software specialties were also included in counts of people with core IT skills.

Federal statistics on these occupations support the following broad conclusions (many more details are in the original reports of the IT Workforce project):

• IT specialties are already the fastest growing occupations in the United States; see Figure 1. They are also among the largest professional specialties. If programmers are included in the count, by 1999 there were more than 2.2 million people in the core IT occupations, compared to less than 750,000 just 15 years earlier. This makes IT as large as nursing, elementary and secondary teaching, and engineering. The Bureau of Labor Statistics expects that rapid growth of the IT specialties will continue; its most recent forecast, released in November 1999, predicts that by 2008 there will be 3.2 million people in these fields.
• IT workers are found in every employment sector in the economy; see Figure 2. Service industries predominate, especially computer and data processing services, a single group that includes the packaged software industry, applications and systems consulting, information management, and other services. Like the occupations in it, this employment sector is the fastest growing industry in the nation. But many other employment sectors are also important centers of IT work, including finance, wholesale and retail trade, and government.
• Less than half of those in the core IT occupations have bachelor degree or higher with completed major or minor concentrations of study in an IT-related discipline. Overall, about 1 million U.S. college graduates can be identified whose education has included formal preparation for IT work at either the undergraduate or the graduate level, but even that number is well short of half of the current workforce; and well over a third of those who do have such academic backgrounds are not working in the core IT occupations (some are undoubtedly employed in closely related jobs, such as information systems management). At least a third of the IT workers without formal credentials in the core areas do have training in closely related fields, such as math, physics, or other kinds of engineering. In addition, academic backgrounds in business or social science are common for IT workers. The project also looked into data on industry certification programs such as those supported by Novell and Microsoft. The number of certifications issued to date is approaching at least half a million, but at this time there is no way to estimate the overlap among either those with certificates (many persons may have more than one of these credentials) or those with both certificates and more traditional academic degrees.
• Persons with academic IT credentials who are not employed in the core occupations are significantly older than those who do hold such jobs. Some of these persons may have moved up to management. In general, the age distribution of core IT workers is distinctly more youthful than that of the rest of the U.S. scientific and engineering workforce; see Figure 3. Such a distribution is to be expected in a field that has grown as rapidly in recent years as has IT, so this fact alone is not enough to establish age discrimination, although it certainly would be consistent with that accusation. If there is age discrimination and if those practices continue, then the industry's recruitment problems will certainly get worse, because the generation of "Baby Bust" people is now entering the workforce, and during the next 10 years the number of people in the labor force who are in the 25-to-34 age group is going to decline.
• Five of every six U.S. IT workers are native-born citizens. Of the foreign-born, well over half are naturalized citizens and most of the rest are permanent residents of the United States. The foreign-born workers tend to be better trained than native workers, at every level of the industry. The number of IT specialists who are in the U.S. on temporary workers' visas is still relatively small, but the Congress has greatly expanded the number of these visas that can be issued each year, and further expansions are now being proposed. The path to permanent residency often begins with graduate study; companies recruit foreign students and then arrange for temporary work visas so that they can remain in the United States.
• A review of five different kinds of indicators of demand yields no compelling evidence that there is a current shortage of IT workers or that such a shortage threatens to damage the industry in the near future. The indicators show that levels of employment have increased; levels of unemployment have also increased, consistently, since 1997; earning levels have increased, but not significantly more than those for other kinds of professionals, and only marginally more than the increases obtained in the same time periods by the entire U.S. civilian labor force; industry estimates of the number of vacant positions are precisely matched by the record of annual growth in employment, indicating that supply is meeting demand; and the supply of labor has increased significantly, especially if alternative sources of talent are considered.

This last factor is of particular interest. First, participation in academic IT degree programs has increased sharply. Second, there is a substantial reserve of trained people who are not working in core IT jobs. Third, Congress has expanded access to temporary foreign workers and is considering expanding it even more. The U.S. accounts for less than 10 percent of the world's annual production of high-tech degrees; globally, more than 1 million bachelor's degrees in engineering and related fields like math and computer science are awarded every year. This enormous offshore pool of talent is increasingly available to U.S. employers, who have an understandable interest in "creaming" it. And fourth, immigration is not the only way that U.S. employers can use foreign talent. Data from the Department of Commerce's import/export statistics show that the dollar value of foreign outsourcing of computer, data processing, and other IT services rose nearly eightfold between 1986 and 1997, to $434 million per year; see Figure 4. The services of people with doctoral degrees in electrical engineering can be purchased in India for about a tenth of what they cost in the U.S., suggesting that this level of outsourcing may displace as much as $4 billion a year in domestic employment.

Implications for Career Planning

So what does all this mean for individual IT specialists? A healthy degree of skepticism is appropriate when the industry asserts that there are shortages of qualified people. More objective evidence says that this is not the case. As the final report of the IT Workforce Project put it, "...several indicators -- rising numbers of experienced unemployed workers, the flat compensation results reported by Computerworld, increasing enrollments in computer science -- suggest that if anything, pressures of demand on the available supply may have eased somewhat during the past year... It may seem contradictory, but we suggest that there is no general shortage of workers; many employers still can't find the people they seek; and some persons with IT training and experience have difficulty finding work. How can this be? One answer may be that there are signs of a strong preference for recent graduates in the IT job market."

In other words, if employers are experiencing shortages, they are not shortages of qualified people in general but rather shortages of particular kinds of qualified people. Much of the IT literature addresses this matter of idealized expectations on the part of some IT employers. As Reginald Charney indicated in his article on hiring practices in the Fall 1999 issue of the "Software Careers" supplement to DDJ, "The ideal candidate is a self-motivated, self-starting, team player who has just done exactly the desired job successfully in a closely related business -- preferably at the competition..." When searches for such ideal candidates fail -- and as Charney suggests, usually they will -- then many employers may fall back on the next best alternative, newly trained graduates who may make up in energy and a willingness to do whatever is asked of them for what they lack in previous experience. Similar motivations may explain the appeal of foreign labor. If an employee's presence in the United States depends on keeping his bosses happy, that gives management an edge that it does not have over other workers.

This raises an aspect of the IT job market that is underscored by the IT Workforce Data Project's findings: It is a hot field, the fastest growing employment sector in the nation. Hot fields promise lots of openings, which is why career guidance writers love them. But this is a poor way to make career decisions. People are much better off choosing to do what they enjoy and what they're good at, preferably both. In any case, the general level of supply and demand for a profession has little to do with the success of particular individuals; indeed, the greatest successes come to people who are astute or lucky enough to get into a field early, long before it is hot.

Hot fields have other pitfalls for the unwary. They tend to be linked to business sectors that attract investors who are looking for quick returns and who do not care much about the longer-run fate of employees; they may offer enormous rewards, but they also may entail serious risks. This does not mean that you should get out of IT; it just says that people in the profession need to be canny players and to learn that technical skills will not be all that it takes to survive and prosper, especially in the more risky high-wire ventures.

Another neglected characteristic of IT job markets is that because they are very large and present to at least some degree in virtually all employment sectors, they are also quite diverse. The frantic pace of Internet startups or prestigious software houses such as Microsoft and Oracle has a tendency to capture public attention and to define what the entire IT marketplace looks like, but this is illusory. Many large employers of IT talent represent more mature market sectors such as consulting, banking, manufacturing, government, and insurance. These types of employers are willing to look at a wide range of job candidates and may pay more attention to building long-term careers for their employees. A larger point is that the IT job market cannot be accurately described by any single set of characteristics. Instead, there are enormous variations by region, types of employers, and types of job candidates.

The findings of the IT Workforce project raise other issues. One is the fate of people as they age. Career tracks into management are one option, but they will not appeal to all, nor can management fill all of the needs for employment that an aging IT workforce has. Other technical career tracks may be blocked by age discrimination, which probably exists in many IT shops, and at the moment it does not appear that anyone is prepared to do much about it. As noted earlier, the supply of youthful talent will shrink during the next 10 years as the "Baby Bust" generation continues to move out of school and into the workforce. To the extent that employers depend on young talent, their recruitment problems are going to escalate. This might lead some organizations to rethink their hiring guidelines. To judge by anecdotal reports of discriminatory practices in the industry, some IT employers are going to need to make drastic changes in the habits of their employees. For example, it has been reported that youthful managers, who are common in IT, are uncomfortable supervising people who are older and more experienced than they are. This kind of attitude may become a luxury, if not a weakness, in the future, when the choice may be to supervise some older people or supervise no one at all.

In general, all IT specialists need to deal with the problem of the short life of particular skill sets, and with the fact that at least some employers assume that when people move on to more middle-aged life styles that include families and other interests beyond the job, then it is time for those people to work somewhere else. The rapid turnover of IT technology does give youth an edge; they have the most recent training. This raises the question of what to do when you are over the hill (which may occur as early as age 30, certainly by age 40). Many people may need to think about retraining or starting second careers. It may be possible to combine IT skills with substantive experience in other areas and then spin off in those directions.

Finally, the rising involvement of workers from all over the world in U.S. information technology reminds us once again that for employers, the job market is global. Nativists -- those who would restrict if not eliminate foreign access to U.S. jobs -- may try to oppose this trend, but the record of economic history suggests that they will fail. The U.S. workforce is going to have to compete with the top performers of the rest of the world. Our advice is to get used to it.

Further Reading

R.A. Ellis and B. Lindsay Lowell, IT Workforce Data Project Reports I-IV (New York: United Engineering Foundation, 1999). Web link to full text and Acrobat PDF files for all four documents: http://www .uefoundation.org/itworkfp.html.

Burt Barnow, John Trutko, and Robert Lerman, Skill Mismatches and Worker Shortages: The Problem and Appropriate Responses, the final report to the Office of the Assistant Secretary for Policy, U.S. Department of Labor (Washington: The Urban Institute, 1998. This important and highly useful document includes a case study of the labor market for IT workers. Unfortunately, it is not available on the Web. Interested readers may be able to obtain a copy by contacting Dr. Barnow at the Institute for Policy Studies, Johns Hopkins University).

Also see Carolyn M. Veneri, "Can Occupational Labor Shortages be Identified Using Available Data?" Monthly Labor Review [Washington: Department of Labor, March 1999 (http://stats.bls.gov/opub/mlr/ 1999/03/art2exc.htm)]; Carol Ann Meares and John F. Sargent, Jr., The Digital Workforce: Building Infotech Skills at the Speed of Innovation [Washington: Office of Technology Policy, U.S. Department of Commerce, 1999; (http://www.ta.doc .gov/ Reports/itsw/digital.pdf)]; and Peter Freeman and William Aspray, The Supply of Information Technology Workers in the United States [Washington: Computing Research Association, 1999; (http://www.cra .org/reports/wits/cra.wits.html)].

DDJ