April 1996 € Volume 23 € Number 2

Promising Future, Disappointing Present

Figures from U.S. Bureau of Labor Statistics present mixed bag for OR/MS employment

By Douglas A. Samuelson

Note: Figures 1-6, which appear throughout this article, represent actuals (dots), mid-point projections (lines) and projection ranges (vertical bars) of employment for the years indicated, based on BLS data.

If your job opportunities in the past few years have seemed less exciting than you expected, there's a good reason why. New figures recently released by the U. S. Bureau of Labor Statistics (BLS) indicate that employers still expect to hire many more operations research/management science analysts by 2005, but actual employment in the occupation declined from 54,490 in 1990 to 43,728 in 1992, about a 20 percent decrease, and was virtually the same -- 44,370 -- in 1994.

The figures also imply that the vast majority of OR/MS practitioners, including those in most of the sectors with the largest projected growth, are not members of INFORMS. This mismatch makes it more difficult for job hunters and prospective employers to connect through INFORMS meetings and publications.

The news is not all bad, however. More recent, if less solid, information shows signs of an upturn in the commercial sector and a slower, more limited improvement in government-related analysis. For example, this reporter's personal leading indicator swung sharply in the "good" direction in September 1995. Starting then and continuing since, the number of calls I've received from people seeking employees has exceeded the number of calls from people seeking jobs. Since the sudden turn of this indicator in the other direction proved to be one of the first warnings of the profession's woes of the early 1990s (as reported in OR/MS Today, April 1990), it deserves attention.

Figure 1

Recruiters familiar with the profession have similar impressions. Pierre Dowd, president of Ability Resources (Alexandria, Va.), states that the government-centered sector is "slow" but commercial areas are "hot," especially market research, credit scoring, financial risk analysis, financial forecasting, yield management for airlines, fleet management for trucking, and econometrics for litigation support.

"The huge modeling efforts of the 1970s are no more," he adds, "but the opportunities are there and improving in the new hot areas."

Dowd says that prospects in the Washington, D.C., area have been poor, as the federal shutdown and downsizing have had a large effect. "New York is always hot," he notes, "partly because it's a financial center and partly because it's always hard to get people to move there. Los Angeles and Dallas have been hot lately, too, especially in transportation."

Lauren Kamens, of New York City-based Analytic, agrees that "the financial sector is where we see the most opportunity now. Consumer credit cards, retail banking, especially mortgages and home equity lines, database marketing and investment risk management are the high-growth areas for our clients. Now business is all about special niche marketing. Advertising has gotten expensive, so businesses want to target segments of the population. People with good statistics and quantitative skills are needed for this work.

Figure 2

"We also see transportation companies wanting people who can help with yield management, and telephone call centers trying to forecast staffing and resource needs," she adds. "And there's a little in manufacturing."

In contrast to Dowd's comments, Kamens says, "New York is our quietest market. I guess the grass is always greener somewhere else for employers, too. Delaware, North Carolina, California and Texas are the places we hear about with the most opportunities."

This latest BLS update indicates declines in OR/MS employment in almost every major industrial sector, although employers in most sectors still expect large increases in OR/MS jobs over the next 10 years. In many sectors, the projected increases would only bring OR/MS employment back to around its 1988-1990 level. Only engineering and management services firms have experienced a big increase in actual OR/MS employment; not surprisingly, they forecast a bright future as well.

Figure 3

Services account for the most OR/MS actual and projected employment: nearly 48 percent of the 1994 total and 55 percent of the total projected for 2005. The industries which anticipate adding the most OR/MS jobs are engineering and management services (9,000 to 10,000 new jobs by 2005) and computer and data processing services (4,300 to 6,300). In the finance, insurance and real estate sector, the industries with highest anticipated growth of OR/MS jobs are insurance carriers (3,200 to 3,700) and commercial banks (1,000 to 1,300).

Note, however, that actual OR/MS employment in most of these industries significantly decreased since 1990: from 8,408 in 1990 and 8,831 in 1992 to 7,317 in 1994 in computer and data processing services; from 5,684 in 1990 to 3,655 in 1994 in commercial banks; and from 9,370 in 1990 to 7,752 in 1994 in insurance carriers. Therefore, the projections should be interpreted with some skepticism.

Figure 4

The decline of OR/MS employment has been most pronounced in the manufacturing sector, which went from 9,167 OR analyst jobs in 1990 to 4,461 in 1994, with projected employment nearly flat from the new, lower base. This trend, combined with the decline in both actual and projected employment in the federal government and U. S. Postal Service, appears to indicate not only a change in employment levels, but also a shift in what kinds of OR/MS activities will be most in demand.

Figure 5

Large-scale analyses of logistics and production may be giving way to market analyses and business restructuring. If this is so, "softer" interpersonal and communication skills and an understanding of marketing may take on greater importance for job-seekers, while proficiency with the traditional tools and techniques of the profession may be less valued.

The employment figures for the federal government and the U. S. Postal Service, while relatively small, also portend a long-term shift in emphasis within the profession. Federal contracts have supported large numbers of OR/MS analysts; the decline of "in-house" OR/MS analysts indicates a devaluing of this kind of work in the government and also reduces agencies' ability to assess the quality and benefits of contractors' efforts.

Despite these problems, however, even the government-related picture may be better than it seems, since the federal shutdown delayed much work already planned and many agencies have yet to receive authorization to let new contracts. The scramble to meet deadlines in spite of the delay has led some local firms to start preparing to add people, and the release of the pent-up demand by other agencies this spring and summer could lead to a surge of additional hiring then, as agencies attempt to commit Fiscal Year 1996 money before Sept. 30, when the fiscal year ends.

Figure 6

If this overview seems confusing, one reason is that the sources quoted here are looking at different parts of the situation. Informal contacts often run a few months ahead of orders -- or lack of orders -- to recruiters. Recruiters see employers' current short-term needs, not their long-term forecasts. The BLS figures are compiled from state-run establishment surveys in which employers are given definitions of occupations and asked to give actual and forecast employment for each occupation. An "operations research analyst" is broadly defined, and includes anyone who does quantitative analysis, preferably computer-assisted, of complex organizational and technical problems. No educational requirements are specified, nor is the use of any specific techniques.

This broad definition also helps to explain why the BLS total of approximately 44,000 OR/MS practitioners (college and university faculty are listed as a separate occupational category, as are government and military) far exceeds the total membership (about 12,000) of INFORMS, let alone the non-government, non-academic practitioner membership, which is around 5,000. Unfortunately, this also means that 80 to 90 percent of the employer-defined OR/MS practitioners in the United States do not belong to INFORMS. Of course, this calculation leaves unexplored the question of how many employer-defined practitioners INFORMS would choose to include in its definition of the profession - a question which is well outside the scope of this report.

The patterns of employment are pretty much consistent with earlier BLS reports (see, for example, Samuelson, 1990 and Geoffrion, 1992) but the declining numbers of actual jobs also indicate that past forecasts have been optimistic, at least in the short term. In addition, both actual and projected employment in the computer and data processing services have declined since the 1990 report, which may indicate that some jobs classified as OR analysts in 1990 in those sectors are now called something else. This in turn implies that similar redefinitions may have occurred in other sectors as well.

Readers should also note that occupational misclassifications apparently occur both ways. While some people who clearly would not fit INFORMS' usual definition are counted as members of the profession, some people who do fit INFORMS' definition are not counted. In Delaware, for example, long-time INFORMS member Sid Hess points out that no OR analysts are listed in the chemical industry, although several of the state's chemical companies have OR departments. This could mean simply that none of these firms were among those sampled, or it could mean (more likely) that the OR analysts were counted in a different category.

On the other hand, the increase in both actual and projected employment in the engineering and management services, and the decline in such sectors as commercial banking, may indicate that, over time, employees are being classified somewhat more consistently with INFORMS' definition of the profession.

In any case, the BLS statistics clearly imply that the profile of the profession, as seen by employers, is changing significantly: market-oriented and process-oriented skills seem to be gaining in importance, as do the sectors in which these skills are paramount. This implication is consistent with recruiters' impressions. The profession has endured a serious downturn, but the future is promising for those OR/MS analysts who can meet the market's new demands.

References
1. Bureau of Labor Statistics, U. S. Department of Labor, "The Workforce in 2005," Monthly Labor Review, Vol. 118, No. 11, November 1995; and supplementary data tables available by request from the Office of Employment Projections, BLS, USDOL, 2 Massachusetts Ave., N.E., Washington, DC 20212.

2. Geoffrion, Arthur M., "Forces, Trends, and Opportunities in OR," Operations Research, Vol. 40, No. 3, May-June 1992, pp. 423-445. This was a revised and extended version of the Omega Rho Lecture delivered at the November 1991 national meetings.

3. Samuelson, Doug, "Basic Research Threatened -- Squeeze on Analytical Work Worsens," OR/MS Today, Vol. 17, No. 2, April 1990, p. 18.

4. Samuelson, Doug, "Labor Department Study: Major Changes in OR Opportunities," OR/MS Today, Vol. 17, No. 4, August 1990, p. 22.

Doug Samuelson is president of InfoLogix, Inc., a consulting firm in Annandale, Va., near Washington, D.C. He is also a past president of the Washington OR/MS Council and a frequent contributor to OR/MS Today.

For more information about this article, input the number 8 in the appropriate space on the Reader Service Form

Click here to return to the table of contents.