Staffing Industry Economic Analysis

Rebound

ASA’s Annual Economic Analysis of the Staffing Industry

Editor’s note: This article provides an overview of the size, scope, and dynamics of the staffing industry. It is intended as a general reference for staffing firms, staffing customers, industry analysts, journalists, and policy makers.

By Steven P. Berchem

After weathering four arduous years of lowered demand and a 28% loss of jobs, the U.S. staffing industry rebounded to its prerecession employment and sales levels in 2004, ending the year with accelerating double-digit annual growth rates.

The staffing industry benefited from a strengthening economy. Real gross domestic product grew by 4.4% in 2004, up from 3.0% in 2003 and the strongest since 1999, the U.S. Bureau of Economic Analysis reported earlier this year. Business investment—particularly in inventories, equipment, and software—was the largest contributor to the growth, helped by increased consumer spending, especially in the purchases of services and nondurable goods.

Figure 1: America’s Staffing Companies Match Millions of People to Millions of Jobs Every Day. Figure 2: Over the Course of the Year, U.S. Staffing Firms Hired 11.7 Million Employees in 2004. Figure 3: Staffing Employee Turnover Declined Slightly in 2004, Increasing Average Tenure to 11.3 Weeks. Figure 4: Temporary Staffing Sales Increased by 12.5% in 2004.

Besides increased capital investment, businesses also began to spend more on people. The U.S. Bureau of Labor Statistics estimates that nonfarm employment increased by 1.5 million last year. U.S. job numbers had fallen sharply in 2001 and remained flat for most of 2002 and 2003, but went up nearly every month in 2004 and have continued to grow this year. Unemployment also declined in 2004, from 5.7% in January to 5.4% in December, and the rate has been relatively stable so far this year.

Staffing in 2004

America’s staffing companies employed an average of 2.55 million temporary and contract workers per day in 2004, according to the American Staffing Association’s quarterly employment and sales survey (see Figure 1). With the addition of an average of 280,000 jobs per day, staffing employment grew by 12.4% in 2004, marking the second consecutive year of double-digit growth and returning the industry to employment levels last seen in 2000.

Over the course of the year—taking into account the high turnover (360% last year) inherent with temporary work—U.S. staffing firms hired 11.7 million employees in 2004 (see Figure 2). Of those, approximately 8 million transitioned to permanent jobs, based on previous ASA research that demonstrates nearly three-fourths of temporary employees go on to permanent jobs when they leave the staffing industry.

While on any given day, the staffing industry employs 1.9% of the U.S. work force, approximately one in 11 nonfarm workers had a job with a staffing company at some point last year.

Temporary and contract staffing employee turnover declined slightly last year, from 373% in 2003 to 360% in 2004, thereby increasing average tenure from 11.0 weeks in 2003 to 11.3 weeks in 2004 (see Figure 3).

The staffing industry began to see a boost in business late in the second quarter of 2003, but it wasn’t until 2004 that the momentum appeared sustainable. Employment was growing in 2003 (see Figure 1), but sales were relatively flat (see Figure 4)—presumably because most of the new hires were in sectors with lower hourly bill rates. In the first quarter of 2004, both employment and sales increased at rates in the high single digits, then burst into the mid-teens in the second quarter and for the remainder of the year.

Temporary and contract staffing revenues for 2004 were $63.3 billion, according to the ASA quarterly employment and sales survey, 12.5% more than in the previous year and nearly on par with the industry’s prior peak year in 2000 (see Figure 4). After dropping for several years, permanent placement revenues increased about 20% in 2004 to an estimated $7.3 billion. Staffing industry revenues for the year totaled $70.6 billion, up 13.1% from 2003.

The staffing industry normally rebounds robustly during economic recoveries (see Figure 5). In the previous three decades, the industry’s periods of greatest growth generally occurred immediately after each recession. For example, after the 1991 recession, staffing industry employment increased by 7.8% the next year and then surged by 25% in each of the following two years.

But this most recent cycle was different. Staffing employment fell half a year before the recession began and stayed down for a year and a half after the recession ended.

The Recession That Almost Wasn’t

The Business Cycle Dating Committee of the National Bureau of Economic Research, a nongovernmental organization that is viewed by most economists as the arbiter of U.S. economic cycles, declared July 17, 2003, that the recession occurred March to November 2001. With the U.S. employment slump extending well into 2003, some economists questioned whether NBER called the end too early. The criticism prompted the NBER committee to take the unusual step of issuing a statement on Oct. 21 of that year to defend its dating of the 2001 recession.

"A recession is a significant decline in economic activity spread across the economy, lasting more than a few months," the statement said. "In choosing the dates of business-cycle turning points, the committee follows standard procedures to assure continuity in the chronology. Because a recession influences the economy broadly and is not confined to one sector, the committee emphasizes economywide measures of economic activity. The committee views real GDP as the single best measure of aggregate economic activity."

The committee "places considerable weight on the estimates of real GDP issued by the Bureau of Economic Analysis of the U.S. Department of Commerce." But those estimates are based on calendar year quarters. The committee seeks to determine the exact months of economic turning points, so it refers to several monthly measures, primarily real personal income and employment, but also industrial production and wholesale-retail sales.

NBER’s October 2003 statement charted how "GDP reached a peak in the fourth quarter of 2000. This was followed by a contraction during the first three quarters of 2001 and growth since then."

Figure 5: Temporary Staffing Employment Usually Rebounds Robustly During Economic Recoveries

But the NBER dating committee was basing its analysis on apparently inaccurate data. Last July, BEA issued revised real GDP growth rates for the previous three years, including 2001 (see Figure 6). The first quarter of that year was down half a point, the second quarter was up more than one point, and the third quarter was down nearly one and a half points. A seesaw—down, up, down, three months at a time—with a net annual rate during those three quarters of -0.7%.

In retrospect, the GDP figures do not appear to support the "significant decline… lasting more than a few months" specified by NBER as criteria for defining a recession. NBER has been silent on the matter since October 2003.

Even if there was a recession, when did it end? The NBER dating committee—using evidence from the previous six business cycles—cited two other key factors it considered in making its determination regarding the recession’s end: personal income and employment.

Income had grown since the end of the recession, but only about one-fifth as strongly as the average of the previous six postrecession periods.

And the employment response had been the opposite, the dating committee acknowledged. After the previous six recessions, employment, on average, grew steadily. Unlike the past, though, employment had fallen after the 2001 recession ended (see Figure 7)—indicating, at the very least, an employment recession.

So maybe we had an economic recession. Maybe not. If so, maybe it ended in November 2001. Or maybe not.

Or perhaps there was something else going on.

The Error of Relying On Past Cycles

David Malpass, chief global economist at Bear Stearns, argues that the current cycle—beginning with "the downturn started in late 2000"—is unique. "There has been a systematic underestimate of the strength and durability of the expansion," he wrote in an April 6, 2004, commentary published in the Wall Street Journal. "The underestimation stems from an insistence among analysts on comparing this cycle to the previous ones, including the jobscreation process. Macro-economics seems wedded to cycle theory, the idea that business cycles are similar. The problem is that the world is changing too fast to rely heavily on past cycles."

He cites the declining trend in the unemployment rate as an example. The 10-year average has been falling for more than 20 years (see Figure 8). The peaks following recessions have been lower, as have the troughs in expansions. "It doesn’t make much sense to compare today’s job growth rate, when unemployment is 5.7%, to 1984’s recovery, when unemployment was at 7.5%," he said, noting that the rate was 3.9% in late 2000 and peaked at 6.3% in June 2003.

Figure 6: Was There Really a Recession? Revised GDP Data Show a Seesaw, But No "Sustained Decline…Lasting More Than a Few Months." Figure 7: A Recession Like No Other? U.S. Payroll Employment Declined After the 2001 Recession-Opposite the Average of the Previous Six Recessions. Figure 8: The 10-Year Average U.S. Unemployment Rate Has Been Falling For More Than 20 Years. And Peaks and Troughs Have Been Lower.

"The big-picture error is that previous recoveries took place at higher unemployment levels, were caused by inflation, and aren’t comparable." The current expansion is durable, he said, because it is "built on small businesses, a flexible labor force, innovation, and lower tax rates."

This cycle is different in other ways, too. Structural changes—the permanent relocation of workers from some industries to others—have contributed significantly to the sluggishness of the job market, according to two economists with the Federal Reserve Bank of New York, Erica L. Groshen and Simon Potter.

Using BLS data, they show that in previous recession-recovery cycles, workers would be laid off and then recalled to their jobs some months later—indeed, employers often helped workers apply for unemployment benefits to encourage them to return to their jobs, rather than find new ones, when the employer resumed stepped-up production. The layoffs were cyclical and temporary.

In the two most recent recessions, Groshen and Potter argue, the layoffs had been permanent—employees were let go with no expectation on the part of the employer that the workers would ever return to their former jobs. This occurred modestly in the 1990-91 recession, they said, and more so in 2001, when 79% of employees worked in industries affected more by structural shifts than by cyclical shifts.

In the August 2003 edition of Current Issues in Economics and Finance, the New York bank’s monthly newsletter, Groshen and Potter outlined the considerable obstacles employers must overcome to create new jobs. "Uncertainty and financial headwinds likely constrain new job creation," they wrote. "Creating jobs takes longer than recalling workers to their old positions and is riskier in the [then] current uncertain environment."

The structural changes may be occurring in part because of management innovations toward leaner staffing, they argued. "Instead of furloughing permanent workers, firms increasingly hire temporary help when they are busiest and then cut back when demand falls. Indeed, firms’ use of temporary or contract employees to smooth labor needs has grown substantially."

Groshen and Potter—along with New York University statistician Rebecca Paul Sela—took another look at the data and presented their findings at the American Economics Association annual meeting in January. They reviewed the last five recessions, from 1970 to 2001, and concluded that "the 2001 recession deviated starkly from previous recessions in a number of ways."

They found that at the onset of the earlier three recessions, many fast-growing industries suddenly reversed course for a year. At the trough of those recessions, many of the industries that were losing jobs the fastest reversed course and started adding jobs. That pattern was largely absent in 1991 and muted in 2001.

Importantly, the pattern showed unusual timing in 2001. Fast-growing industries reversed course in 2000, a year earlier than in previous cycles. And there were two distinct periods when industries that had been losing jobs began adding them. The first, weaker period roughly coincided with the time when fast-growing industries began contracting. The second, stronger (though substantially less so than in the first three recessions studied) period occurred with the official onset of the 2001 recession. In other recessions, this period occurred at the end of the recession, not the beginning.

In other words, the authors conclude, the economy appeared to be poised for recovery in early 2001, only to be hit with the combined impact of the Sept. 11 attacks, geopolitical uncertainty, and corporate and accounting scandals.

Without disputing NBER’s business cycle dating committee, Groshen et al. say that when measured in terms of employment—rather than output (GDP)—the 2001 recession was long and deep.

Key Factors Affecting Job Growth

Federal Reserve Board chairman Alan Greenspan has repeatedly cited the recent "shocks" to the U.S. economy and subsequent "highly uncertain environment" as key factors behind the slow turnaround in job growth. In testimony before Congress last July, he said, "Over much of the previous three years, managers had seemed to pursue every avenue to avoid new hiring despite rising business sales. Their hesitancy to assume risks and expand employment was accentuated and extended by the corporate accounting and governance scandals that surfaced in the aftermath of the decline in stock prices and also, of course, by the environment of heightened geopolitical tensions."

In congressional testimony Feb. 16, he noted that business executives had become more optimistic, as evidenced by increased capital spending and corporate borrowing. "Even in the current much-improved environment, however, some caution among business executives remains," he said. "In contrast to the typical pattern early in previous business-cycle recoveries, firms have appeared reluctant to take on new workers and have remained focused on cost containment."

Economists also cite strong growth in productivity—the output per hour of work—as a key factor is the slow recovery of employment. The NBER dating committee noted as such in its October 2003 defense of the recession end date: "Real GDP has risen substantially since November 2001," the committee statement said. "However, this growth in real GDP entirely took the form of productivity growth. As a result, the growth in real GDP has been accompanied by falling employment."

The productivity proposition is simple: more output, less work. Ultimately, productivity creates prosperity. But surges in productivity without corresponding increases in consumption yield surplus labor. From 2001 through 2003, productivity grew faster than the economy (see Figure 9). So businesses needed fewer workers to keep up with demand. Indeed, businesses could shed workers and still keep up with increased demand. In 2004, GDP grew slightly faster than productivity, and hiring increased. Indeed, productivity growth seems to have slowed markedly in the second half of last year, which would suggest continued strengthening of employment. But, "Productivity numbers are prone to gyrations from quarter to quarter," remarked Wall Street Journal columnist Timothy Aeppel shortly after BLS released the 2004 figures. "One reason job growth may continue to disappoint is that productivity isn’t really expected to fall all that much when viewed over the longer term, which is really the best way to interpret the meaning of productivity anyway." He said that most economists predict that productivity growth will ultimately stabilize at between 2% and 2.5%, which would be strong by historical standards (the average annual rate for the 1990s was just under 2%).

Another key factor affecting job growth is the supply of labor. Greenspan is becoming increasingly vocal about his concerns regarding the quality of the U.S. work force. "For the past 20 years, the supply of skilled, particularly highly skilled, workers has failed to keep up with a persistent rise in the demand for such skills. Conversely, the demand for lesser skilled workers has declined, especially in response to growing international competition. The failure of our society to enhance the skills of a significant segment of our work force has left a disproportionate share with lesser skills. The effect, of course, is to widen the wage gap between the skilled and the lesser skilled." Large income disparities fuel resentment and political polarization that lead to misguided economic policies that work to the detriment of the economy and society as a whole, he said.

Skill disparity not only leads to income disparity, it may also be driving some people out of the work force. Greenspan’s colleague, Federal Reserve Board vice chairman Roger W. Ferguson Jr., presented an analysis of recent trends in labor force participation at a convention of economists and social scientists in Philadelphia in January. Labor force participation declined as the economy weakened. Importantly, the decline was most pronounced among teenagers, young adults, and workers aged 25 to 44—not near-retirement baby boomers, as would have been expected. Also important, the proportion of workers who have exited the work force because they could not find a job increased only slightly. "Instead," he said, "the rise in nonparticipation since 2000 is associated with an increase in the proportions of individuals attending school, reporting themselves as ill or disabled, or retired." Whether these people will return to the labor force—either because they have obtained new, marketable skills or because they are attracted by economic incentives—may dramatically affect job creation, inflation, and the economy.

The Flexibility Factor

America’s work force is changing as more and more people are looking for flexibility in their employment. America’s staffing companies match millions of people to millions of jobs—in virtually all occupations, from day laborer to CEO (see Figure 10). Results from a series of surveys conducted by BLS indicate that staffing industry jobs are shifting toward occupations that require higher levels of skills and education.

Figure 9: The Recession Ended When GDP Resumed Growth—But Productivity Grew Faster, So Employment Fell in 2002 and 2003.

Much like the "regular" work force, the vast majority of temporary and contract employees work full time. According to BLS, 79% of temporary employees and 90% of contract employees work full time, compared with 83% of workers in traditional arrangements who work 35 or more hours per week. BLS data also show that a plurality of temporary and contract employees prefer their status over traditional arrangements.

Why? Flexibility. In an ASA survey of temporary employees, 64% said flexible work time is important to them, 28% said they work for a staffing firm because it gives them the flexibility to pursue other interests, and 43% said needing time for family was an important factor to them in their job decisions.

Businesses seek flexibility, too. It helps them compete in a fast-paced, ever changing global economy.

"Developing an agile, adaptable work force that embraces change and aligns itself quickly will be tomorrow’s competitive differentiator," said PeopleSoft vice president Carl Steffen. His remarks were based on a 2004 Conference Board survey of 540 global business leaders who designated speed, flexibility, and adaptability as top priorities for their companies.

Companies that embrace work force flexibility and engage staffing firms do better. "Increased reliance on contingent (i.e., temporary/part-time) labor…is associated with superior subsequent performance…[and] no increase in systematic risk," concluded a study published in Decision Sciences journal.

Economists Nandkumar Nayar of Lehigh University and G. Lee Willinger of the University of Oklahoma compared firms in a carefully constructed sample and found that earnings (before interest, taxes, depreciation, and amortization), gross margins, and stock returns improved after the increased use of contingent labor.

Figure 10: Temporary and Contract Employees Work in All Occupations. Figure 11: 15% of U.S. Businesses With 25 or More Employees Obtained Staffing Services in 2004.

The larger the company, the more likely it is to use staffing services, according to various surveys. In an ASA poll of 500 businesses conducted last year, 12% of companies with 25 to 99 employees used staffing services, compared with 24% of companies with 100 or more employees

Board members—mostly global companies—found that 90% use staffing services. And a survey of large employers in San Diego found that 95% use staffing services. Businesses that obtain temporary and contract employees through staffing firms cite three main reasons, according to the ASA poll (see Figure 12):

• To fill in for absent employees or to fill a vacancy temporarily
• To provide extra support during busy times or seasons
• To staff special short-term projects

Flexibility is not only good for businesses and workers, it’s also good for the economy. In remarks delivered at the European Banking Congress last November, Fed chairman Greenspan said that labor market flexibility helped the United States weather the recent shocks to its economy. He urged other countries to more fully embrace such flexibility.

Labor market flexibility helps create jobs, economists say. A study published by the Employment Policies Institute determined that "the temporary help industry helped to increase employment in manufacturing by allowing firms to expand their labor forces in the face of uncertain demand conditions." While BLS reported an increase of 570,000 manufacturing jobs from 1992 to 1997, EPI estimated that manufacturing employment actually increased by 1,075,000. Temporary help workers accounted for the difference—about half a million jobs. In the absence of a flexible staffing alternative, the study concluded, manufacturers would not have hired aggressively in response to rapid increases in demand.

The administrations of presidents Bill Clinton and George W. Bush have both cited the staffing industry as an important contributing factor in creating jobs and reducing unemployment. Economists Lawrence Katz of Harvard University and Alan Krueger of Princeton University studied the dramatic drop in the unemployment rate in the 1990s. They concluded that the growth of the staffing industry was responsible for half of the reduction in the unemployment rate.

Staffing in 2005 and Beyond

Figure 12: U.S. Businesses Turn to Staffing Firms to Fill Work Force Gaps, Augment Their Own Staff, and Find New Employees.

Staffing industry growth is likely to continue in 2005 and well beyond. Current economic conditions bode well for another year of double-digit growth for the staffing industry. Besides the structural shift toward flexibility, three factors stand out:

First, with the improved economy, inflation is increasing, prompting the Federal Reserve to raise interest rates. Bear Stearns analyst Andrew Steinerman has shown that, historically, rising interest rates are associated with staffing industry growth.

Second, the unemployment rate is falling and the labor market is tightening. It is going to become increasingly difficult for employers to recruit workers, especially skilled workers.

Third, three-quarters of employees are looking for new jobs, according to a survey conducted by the Society for Human Resource Management and Career-Journal.com, suggesting a likely boost in employee turnover. More turnover means more vacancies.

In short, here’s the problem businesses will face in the coming months: As the economy grows, they will need more help. As the economy strengthens, employee turnover will grow, so businesses will be short of help just as they need more help. Meanwhile, it will be harder for them to find help as the labor market tightens. Solution? Staffing companies.

Harris Nesbitt analyst Jeffrey M. Silber estimates that the staffing industry will grow 10% to 13% this year, well above the 3% to 4% GDP growth expected by most economists. Staffing Industry Analysts Inc. forecasts 10.4% growth this year.

Growth is expected to continue well beyond 2005. BLS predicts that more jobs will be created in personnel supply services (which includes staffing) than in any other industry in this decade, and that it will be the fifth fastest-growing industry through 2012 (see Figure 13).

It looks like the staffing industry rebound will continue its upward trajectory for years to come.

Figure 13: Employment Services—Mostly Staffing—to Create More New Jobs Than Any Other Industry Through 2012.