Contributors: David DeRemer, Jeffrey C. Fuhrer, Kristina Johnson, Jane Sneddon Little, Radoslav Raykov, Scott Schuh, Geoffrey M. B. Tootell, Robert Triest, and Anne van Grondelle
No. 1, January 2004 - August 2004

Motivation for the Research
The recovery from the 2001 recession has been very unusual. The typical pattern of aggregate employment change over the business cycle is for employment to decrease during a recession, but then to rebound fairly rapidly during the subsequent recovery. In August 2003, 20 months following the latest recession’s end, nonfarm payroll employment represented roughly 2.7 million fewer jobs than the pre-recession peak of employment.Despite improved employment growth since then, in early June nonfarm payroll employment stood roughly 1.3 million jobs below the pre-recession peak level, although the recession ended more than two-and-a-half years earlier. This Public Policy
Brief examines possible causes of the “employment gap.”

Research Approach
The research looks at the issue from three separate but complementary perspectives. Section I describes the unusual characteristics of the current recovery to date, explains why it has been termed the “job-loss recovery,” examines the alternative data sources that are used to measure employment, and analyzes the magnitude of the “employment gap” and the rate at which employment must grow to close the gap by the start of 2006. Section II addresses outsourcing, which some have blamed for the job malaise. Section III explores the role of productivity and costs as possible explanations for sluggish employment growth.

Key Findings

• The recent data are encouraging, but we still have quite a way to go in closing the employment gap. Employment growth will need to average between 262,000 and 430,000 jobs per month between April 2004 and January 2006 in order to eliminate the employment gap.
  
  
• The anemic rate of net employment growth during much of the current economic expansion has been caused by a lack of job creation, not by an unusually high rate of job destruction.
  
  
• Although outsourcing of manufacturing jobs is nothing new, what is new is that outsourcing now involves services and the export of moderately high-skill, white-collar jobs. As many as 300,000 business and professional services jobs may have been lost to import competition and overseas job relocation over the past three years.
  
  
• According to BLS data on large, long-lasting layoffs, import competition and job relocation overseas explain just 2.4 percent of all such layoffs in 2001 through 2003. Although the estimated gross number of jobs lost to outsourcing overseas — 1.3 million — may seem large, it become quite small when seen in the context of the total job loss.
  
  
• While this country is running a huge trade deficit overall, the United States continues to enjoy a surplus in services trade — even a growing surplus in other private services vis-à-vis the rest of the world and vis-à-vis developing Asia. In other words, U.S. workers remain highly competitive in high-value-added services — even in Asia.
  
  
• What about the jobs the United States failed to create because of outsourcing? Although the limited job creation data available do show that hiring rates in professional and business services have fallen more than hiring rates for the average U.S. industry since 2001, hiring rates have also fallen more than average in many industries not well suited to outsourcing, suggesting that it is domestic forces that have discouraged job creation.
  
  
• The impact of job relocation overseas is, and will likely remain, modest, in part because the trade and investment flows that facilitate foreign outsourcing trigger offsetting equilibrating forces. Moreover, the integration of dynamic new regions like Japan and Korea into the world economy has historically always benefited other countries.
  
  
• Productivity — output per worker — has essentially been the dominant engine of growth in nonfarm business output during the current recovery. From the recent recession trough through the end of 2003, productivity grew even faster than output — by 9.9 percent while output grew by 8.5 percent. The other components of output — hours per worker and employment — both declined. It is quite unusual for productivity to be the only growing component of output.
  
  
• In this recovery (through June), firms have chosen to respond to increasing demand for output entirely through higher productivity rather than by raising employment or hours per worker. While we are not certain of the reason for this, there appears to be some validity to the hypothesis that firms have been reluctant to hire workers because they are uncertain about the sustainability of recent growth.
  
  
• Recent gains in productivity occurred in all three components: higher trend productivity, higher cyclical productivity, and higher unexplained productivity. Higher trend productivity does not “explain” the “job-loss” recovery. However, it does help us understand why relatively strong recent GDP growth has not had a greater impact on employment and unemployment.
  
  
• The relatively high level of unexplained productivity remains a puzzle. This brief offers modest evidence suggesting that uncertainty about the economic recovery and growth may explain a piece of the puzzle.

Implications
If economic uncertainty continues to fade quickly and productivity growth reverts to trend, the long-awaited emerging recovery of employment may remain in a brisk “job gain” mode for the remainder of this year or so.

Full text of Public Policy Brief 04-1