Policy Debate: Is a "jobless recovery" a necessary part of the "new economy?"

The divergent paths of output and employment in 1991-92 and 2002-03 suggest the emergence of a new kind of recovery, one driven mostly by productivity increases rather than payroll gains. The fact that no influx of new workers occurred in the two most recent recoveries means that output grew because workers were producing more. Although one might speculate that output increased because workers were putting in longer days, average hours worked by employees actually changed little during this and the previous jobless recovery.
~Erica Groshen and Simon Potter

While some unemployed individuals try to find a new job as quickly as possible, others respond to the low net cost of remaining unemployed by spending more time searching for a better job or simply using the time for household activities. Statistical studies show a substantial effect of unemployment benefits on the duration of benefits and a significant surge in job finding in the weeks just before benefits run out.

When Congress extended the maximum duration of benefits from the usual 26 weeks to 39 weeks, it slowed the rate at which the unemployed found jobs. While no one should begrudge unemployed workers the protection of unemployment benefits, it is important to recognize that those benefits raise the unemployment rate and delay the rise of employment in every economic recovery.
~Martin Feldstein

The 1980s and 1990s contained the two longest business cycle expansions that have been recorded for the U.S. Both of these expansions were accompanied by falling inflation and unemployment rates. The prolonged expansion of the 1990s lead to extensive discussion of the possibility of a "new economy" in which the Phillips curve tradeoff between inflation and unemployment had seemingly disappeared. In this new economy scenario, technological change resulted in rising productivity, lower prices, and increased wages and employment. (The arguments concerning the "new economy" are addressed in an online debate on this topic.) While each of these expansions was followed by a relatively short recession (each recession lasted only for 8 months), the subsequent expansions were characterized by a much slower rate of job growth than in other post-World War II expansions.

Some economists have argued that the long expansions preceding these brief recessions had resulted in excessively high employment levels. Joseph Schumpeter has argued that the business cycle results in a process of "creative destruction" in which recessions force the elimination of less efficient firms and less efficient business and employment practices. During an unusually length expansion, it is argued, a greater than usual amount of inefficient labor usage might occur. The resulting labor market adjustments might still continue for some time after the recovery begins. The very high rates of productivity growth during the initial stages of these two expansions are consistent with this argument.

Until the 1990-1991 recession, an inverse relationship had seemed to exist over the course of the business cycle between the growth rate of output and the unemployment rate. Results reported by Arthur Okun in 1962 indicated that a 1 percent reduction in the unemployment rate was associated with a 4 percent output growth rate. Later results for the 1960s through 1990s suggested that a 1 percent reduction in the unemployment rate was associated with a 3 percent output growth rate. This inverse relationship was so persistently found that it became known as "Okun's Law." The economic recoveries following the 1990-1991 and 2001 recessions have case doubt on the reliability of Okun's Law.

A complicating issue is the large differences between employment levels recorded by the establishment and household surveys conducted by the Bureau of Labor Statistics during the period from 2001 through 2003. The establishment survey showed substantial employment declines during this period while the household survey recorded modest employment growth. One possible explanation for the discrepancy is a shift to outsourced employment in large established firms. If these jobs are outsourced to small new companies or to self-employed individuals, the establishment survey will understate employment and result in overstated labor productivity gains.

Will "jobless recoveries" be more common as a result of structural changes in the economy? Is a "jobless recovery" primarily an artifact of measurement error based on a reliance on establishment payroll data? Or is the a "jobless recovery" the result of technological change and globalization that has resulted in an increase in structural unemployment? These questions are likely to be the source of substantial research effort in the near future.

• National Bureau of Economic Research, "Business Cycle Expansions and Contractions"
  http://www.nber.org/cycles.html
  The National Bureau of Economic Research provides estimates of the dates of the peaks and troughs of the business cycle. This website contains a list of these dates from 1854 through the present.

• Business Cycle Dating Committee, National Bureau of Economic Research, "The NBER’s Recession Dating Procedure"
  http://www.nber.org/cycles/recessions.html
  This document describes the procedure used by the National Bureau of Economic Research (NBER) to date the peaks and troughs of the business cycle. In this document, it is noted that the NBER bases its dates of business cycle turning points on changes in output, not changes in unemployment rates.

• Bureau of Labor Statistics
  http://www.bls.gov
  The Bureau of Labor Statistics is the best source for data on current labor market conditions. site contains a wide variety of statistics and data, online working papers dealing with the labor market, and information about the programs and publications of this agency.

• Bureau of Labor Statistics, "Occupational Outlook Handbook"
  http://www.bls.gov/oco/
  The Occupational Outlook Handbook is a superb source for information about current and expected future labor market conditions for a wide variety of occupations. For each occupational title, this handbook provides information about the type of work involved, training and educational requirements, and salary information. Projections of future labor market conditions (based upon survey responses) are also included for each occupation.

• Monthly Labor Review
  http://stats.bls.gov:80/opub/mlr/welcome.htm
  The online edition of Monthly Labor Review contains current labor force statistics and studies of trends in labor markets. This is a good place to search for recent studies of employment and unemployment patterns.

• Bureau of Labor Statistics, "How the Government Measures Unemployment"
  http://www.bls.gov/cps/cps_htgm.htm
  This online document contains a detailed description of the process by which the government constructs unemployment statistics.

• Bureau of Labor Statistics, "The Employment Situation"
  http://www.bls.gov/news.release/empsit.toc.htm
  This document contains the most recent release of statistics on the state of the U.S. labor market. A detailed breakdown of labor force, employment, and unemployment statistics is provided by age, educational attainment, race, gender, and industry.

• Federal Reserve Bank of St. Louis, "Federal Reserve Economic Data"
  http://research.stlouisfed.org/fred2/
  This website, provided by the Federal Reserve Bank of St. Louis, contains current data on an extensive collection of economic data. You may download the data or view an online plot of each of the variables. This site is a good place to visit to find current GDP, inflation, and unemployment data.

• Joint Economic Committee, "A Tale of Two Employment Surveys"
  http://jec.senate.gov/_files/TwoSurveysUpdated.pdf
  This October 2003 Joint Economic Committee Report describes differences between the employment statistics resulting from the establishment payroll survey and the household survey. The payroll survey indicated a substantial decline in jobs while the household survey indicated a substantial rise in jobs from the end of the recession in 2001 through late 2003. The difference seems to be related to growing self-employment and job growth in new small businesses that are not immediately included in the payroll survey, Some of the discrepancy is also the result of an upward adjustment in the population estimate used for the household survey. (The Adobe Acrobat viewer plugin is required to view this document. You may download this viewer by clicking here.)

• The Concise Encyclopedia of Economics, "Arthur M. Okun"
  http://www.econlib.org/library/Enc/bios/Okun.html
  The Concise Encyclopedia of Economics provides a discussion of Arthur M. Okun's role in formulating economic policy. A succinct discussion of Okun's law is provided as part of this discussion.

• Erica L. Groshen and Simon Potter, "Has Structural Change Contributed to a Jobless Recovery?"
  http://www.newyorkfed.org/research/current_issues/ci9-8/ci9-8.html
  Erica L. Groshen and Simon Potter investigate the reasons for the "jobless recovery" following the 2001 recession. They present evidence that a larger than usual share of unemployment during this period is due to structural unemployment. They note that temporary layoffs were the cause of a smaller than usual share of the unemployment that occurred during this period. Permanent layoffs were much more common than usual during a recession. The industries that lost the most jobs during the recession were also those that were declining in employment shares during the prior expansion. Groshen and Potter argue that the sluggish employment response after the close of the recession is due to the longer time delay required for adjustments to structural employment changes. During most expansions, employment growth is partly the result of workers returning from layoffs. It takes much more time for workers to shift to new occupations (or to other geographical regions).

• Martin Feldstein, "There's No Such Thing As a 'Jobless' Recovery"
  http://www.nber.org/feldstein/wj101303.pdf
  In this October 13, 2003 Wall Street Journal article, Martin Feldstein argues that the "jobless recovery" is partly the result of the Congressional decision to extend unemployment benefits from 26 to 39 weeks. He indicates that a worker receiving $600 a week in income "loses less than $200 in spendable income by remaining unemployed for an additional week." Feldstein suggests that the the unemployment rate would have fallen more rapidly if this benefit extension had not been passed. (The Adobe Acrobat viewer plugin is required to view this document. You may download this viewer by clicking here.)

• Mark Schweitzer, "Another Jobless Recovery"
  http://www.clevelandfed.org/Research/Com2003/0301.pdf
  Mark Schweitzer discusses the recovery from the 2000-2001 recession in this Federal Reserve Bank of Cleveland Economic Commentary article. He provides detailed comparisons of the behavior of employment, unemployment rates, and labor force participation rates during the two most recent business cycles with those occurring during earlier business cycles. Schweitzer notes that there were more voluntary withdrawals from the labor force during the 1990-1991 and 2001 recessions. This appears to be due to a larger than usual number of people choosing to engage in child care or educational activities during this period. This suggests that one of the reasons for the "jobless recovery" is a reduction in job seeking by those who left the labor force in response to perceptions of limited job opportunities. (The Adobe Acrobat viewer plugin is required to view this document. You may download this viewer by clicking here.)

• Ben S. Bernanke, "Remarks by Governor Ben S. Bernanke"
  http://www.federalreserve.gov/boarddocs/speeches/2003/200311062/default.htm
  Federal Reserve Board of Governor Ben S. Bernanke provides several possible explanations for the jobless recovery in this November 6, 2003 speech:
  • firms expanded employment excessively during the extended economic boom of the 1990s.
  • rising health care and pension costs have resulted in higher labor costs and reduced hiring (although this has been partly offset by lower rates of wage increase).
  • an atmosphere of increased political and economic uncertainty following the 9/11/01 attacks and the subsequent military involvement in Afghanistan and Iraq.
  • structural unemployment have have risen in response to a faster rate of structural change (caused by technological change and/or increased global trade).
  • rapidly rising productivity has allowed firms to reduce the amount of labor used in production.

• Allan H. Meltzer, "A Jobless Recovery?"
  http://www.aei.org/publications/pubID.19367/pub_detail.asp
  Allan H. Meltzer argues, in this October 29, 2003 article, that the alleged "jobless recovery" is at least partly the result of measurement problems. He notes that household surveys indicate greater job growth than do establishment surveys during the expansion following the 2001 recession. He notes that: "For the year ending in August, the Establishment Survey shows a loss of 463,000 jobs. The Household Survey shows that the economy added 313,000 new jobs in the same period." Meltzer attributes the difference in these measurements to the outsourcing of employment by large established firms to smaller new firms or to self-employed contractors. Since new firms and self-employed workers are not counted by the establishment survey, he suggests that the household survey provides a better measure of employment change.

• National Center for Policy Analysis, "The Current 'Jobless" Recovery"
  http://www.ncpa.org/iss/eco/2003/pd072103f.html
  The National Center for Policy Analysis argues, in this July 21, 2003 Daily Policy Digest, that the "jobless recovery" is due to increased reliance on the use of temporary workers, the replacement of full-time workers with part-time workers, and increased use of overtime. The use of these strategies by firms have allowed them to expand output while reducing the number of full-time workers.

• Jerry Bowyer, "Not a Jobless Recovery After All"
  http://www.renewamerica.us/columns/bowyer/031112
  In this November 12, 2003 article, Jerry Bowyer argues that claims of a "jobless recovery" are not reasonable. He notes that the unemployment rates during the recovery from the 2001 recession were "lower than the average for: Clinton over his first three years (6.23%), George H.W. Bush's single term (6.28%), Reagan's 2nd term (6.48%), Carter's single term (6.56%), Ford/Nixon (6.64%), and Reagan's 1st term (8.58%)." He suggests that Bureau of Labor Force Statistics often undercounts employment growth during an expansion since much of the job creation is in new firms that are not immediately included in the employment survey used by the Bureau of Labor Statistics.

• Jared Bernstein, "The Jobless Recovery: Suffering from the Recession's Aftershocks, Labor Market Conditions Continue to Worsen"
  http://www.epinet.org/content.cfm/issuebriefs_ib186
  Jared Bernstein describes, in this January 24, 2003 Economic Policy Institute Issue Brief, the poor labor market conditions during the expansion following the 2001 recession. He details the rising unemployment levels, the reduction in private sector employment, the increase in the average duration of unemployment, and the reduction in labor force participation rates that occurred during the first two years of this expansion.

• Martin A. Regalia, "The U.S. Economy, The Jobless Recovery, and the State of our Manufacturing Sector"
  http://www.uschamber.com/issues/testimony/2003/031114regalia.htm
  In this November 14, 2003 testimony before the Senate Democratic Policy Committee, Martin A. Regalia discusses the state of the U.S. economy. He analyzes the discrepancy between the employment data from establishment and household surveys. Regalia cites evidence that the slow improvement in employment is partly the result of an output growth rate that is below the potential output growth rate. He also cites evidence that there was more structural unemployment during the 2001 recession than had occurred in most earlier recessions.

• Rajeev Dhawan, "The Beguiling Recovery"
  http://www.robinson.gsu.edu/magazine/fall2003/rajeev_reports.html
  In this Fall 2003 article, Rajeev Dhawan attributes the jobless recovery to a modest rate of output growth combined with a relatively high rate of productivity growth. He suggests that high employee benefit costs and low growth in revenue also resulted in levels of investment spending that are below those that usually occur during an expansion.

• Tim Kane, "The Myth of a Jobless Recovery"
  http://www.heritage.org/Research/Labor/em917.cfm
  Tim Kane discusses the "jobless recovery" in this March 25, 2004 Heritage Foundation Executive Memorandum. He argues that the reported jobless recovery in the early 2000s was the result of an error in how jobs are counted in the Bureau of Labor Statistics' payroll survey. He notes that this survey overcounts the number of many workers when they change jobs. Since labor turnover declined in the late 1990s and early 2000s, he argues that the reported reduction in employment was the result of a smaller overcount of employment (due to the lower labor turnover rate). He suggests that this explains approximately 1 million of the job loss that were reported to be lost between 2001 and 2003. Kane suggests that the household survey is more accurate and observes that this reported substantial employment growth during this period.

• Charles L. Schultze, "Offshoring, Import Competition, and the Jobless Recovery"
  http://www.brookings.edu/comm/policybriefs/pb136.htm
  Charles L. Schultze investigates the cause of the 2001-2003 "jobless recovery" in this August 2004 Brookings Institution policy brief. He argues that increased use of offshore labor and increased global competition can only explain a small portion of the employment loss during this period. He notes that this period was characterized by a higher than usual productivity increase.

• PBS NewsHour, "The Jobless Recovery"
  http://www.pbs.org/newshour/bb/economy/jan-june03/jobs_6-23.html
  This website contains the transcript of a June 23, 2003 episode of NewsHour, hosted by Paul Solman. The focus of the discussion is on the impact of the "jobless recovery" following the 2001 recession. A RealAudio broadcast of the episode is also available.

• Bureau of National Affairs, "Jobless Recovery"
  http://ww.bna.com/webwatch/joblessrecovery.htm
  This website, provided by the Bureau of National Affairs, contains links to online resources dealing with the "jobless recovery."

• Mathew Ingram "Is a 'Jobless Recovery' a Good Thing?"
  http://www.theglobeandmail.com/servlet/story/.../BNStory/Business/
  Mathew Ingram discusses the "jobless recovery" that followed the 2001 recession in this online article. He notes that the unemployment rate would have been much higher during this period if there had not been an unusually large reduction in labor force participation rates. He suggests that future employment prospects will be affected by the ability of unemployed consumers to continue spending borrowed money. Ingram also notes that much of the unemployment appears to be structural unemployment. This type of unemployment is less responsive to output growth than is cyclical unemployment.

• Larry Kudlow, "The Sixth Supply-Side President"
  http://www.nationalreview.com/kudlow/kudlow071503.asp
  In this July 18, 2003 article, Larry Kudlow discusses possible reasons for the slow recovery following the 2001 recession. He notes that economic growth following the recession was substantially below the usual pattern for a recovery. Rising productivity rates, it is argued, also reduced the employment effects resulting from the economic expansion.

• Alan Kohler, "Blame India for that Jobless Recovery"
  http://www.smh.com.au/articles/2003/10/01/1064988273910.html
  Alan Kohler argues that "offshoring" provides an explanation for at least a part of the jobless recovery. He cites evidence of increased use of foreign labor by American Express, JP Morgan, General Electric, HSBC, and Ford.

• Robert Reich, "This Jobless Recovery is Worse Than The Last"
  http://www.robertreich.org/reich/20030116.asp
  Robert Reich argues that the jobless recovery following the 2001 recession is more severe than that following the 1991-1992 recession. He notes that: "Manufacturers have been cutting jobs now for 29 straight months -- the longest period of manufacturing cutbacks since the Great Depression."

• David Altig, Terry Fitzgerald, and Peter Rupert, "Okun's Law Revisited: Should We Worry about Low Unemployment?"
  http://www.clevelandfed.org/research/com97/0515.htm
  David Altig, Terry Fitzgerald, and Peter Rupert discuss Okun's Law in this May 15, 1997 Federal Reserve Bank of Cleveland Economic Commentary article. They note (as did Okun) that this relationship is not likely to be stable through time.

• Leopold Soegner and Alfred Stiassny, "A Cross-Country Study on Okun's Law"
  http://www.wu-wien.ac.at/inst/vw1/gee/papers/gee!wp13.pdf
  In this September 2000 working paper, Leopold Soegner and Alfred Stiassny examine the robustness of Okun's Law across countries and time by estimating this relationship for 15 OECD countries. They find that the magnitude of this relationship has declined over time for most countries. Substantial variation in the magnitude of this relationship is also found across countries. (The Adobe Acrobat viewer plugin is required to view this document. You may download this viewer by clicking here.)

• Gert Schnabel, "Output Trends and Okun's Law"
  http://www.bis.org/publ/work111.pdf
  In this April 2002 Bank for International Settlements Working Paper, Gert Schnabel estimates a version of Okun's model for a variety of OECD countries. He finds that trend growth in the U.S. increased from 2.3% to 3.3% from the first half of the 1990s to the second half of this decade. Schnabel's estimates indicate substantial variation in the coefficients of this model across countries and across time periods. (The Adobe Acrobat viewer plugin is required to view this document. You may download this viewer by clicking here.)

• Democratic Policy Committee, "President Bush's Failed Economic Record"
  http://democrats.senate.gov/dpc/dpc-new.cfm?doc_name=fs-108-2-284
  The Democratic Policy Committee critiques the labor market outcomes that occurred in 2001-2004 in this online document.They argue that the jobless recovery was accompanied by rising poverty and a decline in median income.

• Genevieve Roja, "Take This Jobless Recovery and Shove It"
  http://www.alternet.org/story.html?StoryID=16572
  Genevieve Roja provides a personal statement of the social and psychological costs associated with unemployment in this August 11, 2003 article.

• David R. Francis, "Mystery of the 'Jobless Recovery'"
  http://www.csmonitor.com/2003/0829/p01s02-usec.html
  In this August 29, 2003 Christian Science Monitor article, David R. Francis discusses the distribution of the labor market costs associated with the "jobless recovery." He notes that African-Americans and college graduates have been affected relatively more than the general population. Francis also observes that real wages have fallen for males and risen only slightly for females during the first two years of the recovery.

• V. Anantha Nageswaran, "Jobless Recovery in US — World Staring Down a Gun Barrel"
  http://www.thehindubusinessline.com/2004/02/02/stories/2004020200210800.htm
  V. Anantha Nageswaran examines the implications of the 2001-2003 "jobless recovery" in the U.S. He raises concern that job losses to overseas firms are being blamed for the slow recovery in U.S. employment. As part of this discussion, Nageswaran notes that the median duration of unemployment in the U.S. increased from 7.2 weeks in the third quarter of December 2001 (when the recession officially ended) to 10.4 weeks in December 2003. He also notes that it is likely that the number of discouraged workers also substantially increased (since almost 2 million workers left the labor force during this period).

• William M. Rodgers III, "The African-American Experience in the Recent Recession and Job Loss Recovery"
  http://www.americanprogress.org/issues/2004/01/b19510.html
  In this January 1, 2004 online article, William M. Rodgers III examines the differential effect of the "jobless recovery" of 2001 to 2003 on white and African-American workers. He notes that, while the initial job losses disproportionately affected white males (due to job losses in IT industries), a larger effect was felt by minority groups as the jobless recovery developed. African-American households experienced a larger decline in median income and a larger increase in unemployment during this period. By the end of 2003, the unemployment rate for African-American teenagers was 28.2% (twice the rate for white teenagers).