The services industry in the 'good' versus 'bad' jobs debate

Monthly Labor Review

Volumn 121, Issue 2, 1998, Pages 22-47

  Meisenheimer II, Joseph R ^a  

^a NONE

Author keywords

[No Author keywords available]

Indexed keywords

EID: 0001939460     PISSN: 00981818     EISSN: None     Source Type: Journal    
DOI: None     Document Type: Article

References (47)

1. These estimates are compiled by the U.S. Department of Commerce, Bureau of Economic Analysis. An industry's gross product equals its gross output - sales or receipts and other operating income plus inventory change -minus the goods and services purchased from other industries or imported.
2. See Adam Smith, An Inquiry into the Nature and Causes of the Wealth of Nations, Book II, Chapter III (originally, Glasgow, Scotland, 1776).
3. Ironically, Smith criticized a group of French thinkers of his era, called the Economistes or Physiocrats, for their view that the only productive members of society were "proprietors" and "cultivators," that is, the people who owned and farmed land. According to the Physiocrats, "artificers, manufacturers, and merchants" were unproductive in the sense that they never added any new value that exceeded the cost of their labor. For Smith's criticism of the Physiocrats, see The Wealth of Nations, Book IV, Chapter IX. Many observers today might argue that Smith's view of the value of service providers also was too narrow.
4. See, for example, Lawrence Mishel and Jacqueline Simon, "The State of Working America," Challenge, November-December 1988, pp. 50-51; and Maggie Mahar, "Blue Collar, White Collar: Good Jobs Are Vanishing Throughout the Economy," Barrons, May 11, 1992, pp. 8-24.
5. See, for example, Lawrence Mishel and Jacqueline Simon, "The State of Working America," Challenge, November-December 1988, pp. 50-51; and Maggie Mahar, "Blue Collar, White Collar: Good Jobs Are Vanishing Throughout the Economy," Barrons, May 11, 1992, pp. 8-24.
6. See, for example, Robert S. Smith, "Compensating Wage Differentials and Public Policy: A Review," Industrial and Labor Relations Review, April 1979, pp. 339-52; Charles Brown, "Equalizing Differences in the Labor Market," The Quarterly Journal of Economics, February 1980, pp. 113-34.
7. See, for example, Robert S. Smith, "Compensating Wage Differentials and Public Policy: A Review," Industrial and Labor Relations Review, April 1979, pp. 339-52; Charles Brown, "Equalizing Differences in the Labor Market," The Quarterly Journal of Economics, February 1980, pp. 113-34.
8. For a thorough discussion of the various factors that affect pay levels, see "The Structure of Earnings," Report on the American Workforce (U.S. Department of Labor, 1994), pp. 49-93.
9. BLS surveys that provide data on earnings include: the Current Population Survey (CPS), the Current Employment Statistics (CES) survey, the Covered Employment and Wages program, the National Compensation Survey, the Employment Cost Index (ECI), the Consumer Expenditure Survey, and the National Longitudinal Surveys.
10. It should be noted that the category "production workers" in manufacturing encompasses different occupations from those in the category "nonsupervisory workers" in services. This difference in occupational coverage complicates the comparison of wages between the two industries. For a more detailed discussion of wages and occupational coverage in the Current Employment Statistics survey, see Katharine G. Abraham, James R. Spletzer, and Jay C. Stewart, "Divergent Trends in Alternative Wage Series," in John Haltiwanger, Marilyn Manser, and Robert Topel, eds., Labor Statistics Measurement Issues (University of Chicago Press, 1998) (in press).
11. The Current Employment Statistics survey does not obtain earnings figures for each individual worker. Rather, from each surveyed establishment, an aggregate weekly payroll figure is obtained for all production or nonsupervisory workers. Average weekly earnings are computed by dividing the aggregate payroll by the number of workers. Average hourly earnings are computed by dividing the average weekly earnings by the average weekly hours of production or nonsupervisory workers. Because these methods are used, it is not possible to obtain information on the distribution of earnings.
12. In the CPS, however, the industry designation might not be as precise as that collected in the CES, and the earnings data reported by household respondents might lack the precision of payroll reports from establishments.
13. Firms might provide health benefits to their employees if they feel that doing so will help to keep employees healthy and productive. Firms might offer vacation benefits if they feel that employees will be more productive if they periodically take time off from work. Some firms offer stock options or employee stock ownership plans to give employees a financial stake in meeting corporate goals. Firms also might provide retirement benefits, specifically, defined-benefit pensions, to encourage workers to remain with the firm long enough to receive the benefits. Providing pensions can help firms to keep productive workers and reduce turnover costs, such as hiring and training new workers to replace those who leave the firm. Another reason firms tend to provide pensions is to encourage employees to perform at an acceptable level of productivity so that they can keep their jobs long enough to receive retirement benefits, rather than being dismissed for poor performance. In addition, because it is illegal in the United States for firms to require mandatory retirement at a certain age, firms instead may use pension plans to encourage workers to retire. For a further discussion of these reasons, see Alan L. Gustman, Olivia S. Mitchell, and Thomas L. Steinmeier, The Role of Pensions in the Labor Market, working paper no. 4295 (Cambridge, MA, National Bureau of Economic Research, Inc., 1993). Firms also may prefer to provide benefits instead of cash compensation to avoid incurring larger expenses for legally required benefits that cost employers a fixed percentage of cash compensation. Of course, firms can essentially shift the entire cost of these benefits onto workers by reducing their cash compensation to cover the additional cost of the legally required benefits.
14. See "Health Care and the Workplace: Recent Trends and Current Status," Report on the American Workforce (U.S. Department of Labor, 1995), pp. 101-35.
15. Three BLS surveys provide data on benefits. The Employee Benefits Survey (EBS) obtains data from private-sector establishments and State and local governments on the incidence and detailed provisions of a variety of employee benefit plans. The Employment Cost Index (ECI) obtains information from private-sector establishments and State and local governments on the cost to employers of employee benefit plans. The Current Population Survey (CPS), the monthly survey of households that provides information on employment, unemployment, and other labor force measures, periodically has included supplemental questions on participation in employee benefit programs. The most recent CPS supplement on employee benefits was conducted in April 1993. The CPS is used in this analysis because its large sample size enables it to provide greater industry detail than the Employee Benefits Survey.
16. See "Employer Costs for Employee Compensation-March 1997," USDL 97-371 (U.S. Department of Labor), Oct. 21, 1997.
17. Health coverage differs from other employee benefits because it is a benefit not just to workers but, in many cases, their families as well. In the April 1993 CPS, workers were asked whether they received employer health coverage for themselves and their families, or simply for themselves. Among health plan participants, workers in services were less likely than those in other industries to receive family health coverage. There are, however, several possible explanations for the lower rate of family coverage in services industries. Employers in services may be less likely to offer family coverage. Alternatively, workers in services may be less likely to need or accept family coverage, either because they are less likely to have families or because their family members may receive health coverage from other sources, such as their own employers. Because it is not possible to determine the specific reasons for the different coverage rates across industries, family health coverage is not examined in this article.
18. For further discussion on health insurance coverage from all sources in addition to one's own employer, see Health Insurance Coverage: 1996, Current Population Reports, Series P60-199 (Bureau of the Census, September 1997).
19. For further discussion on flexible compensation arrangements, see Joseph R. Meisenheimer II and William J. Wiatrowski, "Flexible benefits plans: employees who have a choice," Monthly Labor Review, December 1989, pp. 17-23. For further discussion of the value employees may place on benefits, see Melissa Famulari and Marilyn E. Manser, "Employer-provided benefits: employer cost versus employee value," Monthly Labor Review, December 1989, pp. 24-32.
20. For further discussion on flexible compensation arrangements, see Joseph R. Meisenheimer II and William J. Wiatrowski, "Flexible benefits plans: employees who have a choice," Monthly Labor Review, December 1989, pp. 17-23. For further discussion of the value employees may place on benefits, see Melissa Famulari and Marilyn E. Manser, "Employer-provided benefits: employer cost versus employee value," Monthly Labor Review, December 1989, pp. 24-32.
21. For further discussion of health plan options, see Michael Bucci and Robert Grant, "Employer-sponsored health insurance: what's offered; what's chosen," Monthly Labor Review, October 1995, pp. 38-44.
22. There are two major types of retirement plans: defined-benefit pensions and defined-contribution plans. Defined-benefit pensions legally obligate employers to provide payments to retirees according to a specified formula, which usually includes the retiree's salary and length of service with the firm. Defined-contribution plans generally specify the level of employer contributions to a plan, but not the formula for determining eventual benefits. Instead, individual accounts are established for participants of defined-contribution plans, and benefits depend on the amounts employers and employees contribute, plus the investment earnings of the contributions. In contrast to defined-benefit pensions, defined-contribution plans place the entire financial risk on employees if plan investments perform poorly. If investments perform well, however, participants of defined-contribution plans could receive sizable returns on their investments. Defined-benefit pension participants generally receive no extra benefits if plan investments perform well. The April 1993 CPS asked retirement plan participants to specify whether they were in a defined-benefit pension, defined-contribution plan, or both. Nearly 1 in 5 respondents did not identify either a defined-benefit or defined-contribution plan, however, limiting the usefulness of the data. Because of this problem, this article does not compare which plan types are most common in each industry.
23. There are other types of defined-contribution plans that do not include these tax-deferral features. Such plans are included in the overall figures on retirement plan participation, but they are not examined separately.
24. The most that a single individual can contribute on a tax-deferred basis to an IRA is $2,000 annually, and the most a married couple can contribute is $4,000. Under 401(k), 403(b), and Section 457 plans, the maximum tax-deferred employee contribution permitted by the plan is often 10 percent or 15 percent of the employee's salary, subject to a dollar maximum specified by the Internal Revenue Code. In 1997 this maximum was $9,500 for 401(k) and 403(b) plan participants and $7,500 for Section 457 plan participants. Since enactment of the Tax Reform Act of 1986, single persons who are not eligible to participate in an employer-sponsored retirement plan can contribute up to $2,000 annually on a tax-deferred basis to an Individual Retirement Account (IRA). Single persons who are eligible to participate in an employer-sponsored retirement plan can contribute up to $2,000 annually if their annual adjusted gross income is under $25,000. Single persons with income between $25,000 and $35,000 can contribute less than $2,000 to an IRA, and those with annual income above $35,000 cannot contribute on a tax-deferred basis to an IRA. Married couples can contribute up to $4,000 annually if both spouses work and neither is eligible to participate in an employer-sponsored retirement plan. If either spouse is eligible to participate in an employer-sponsored plan, the couple can contribute up to $4,000 on a tax-deferred basis to an IRA if their annual adjusted gross income is under $40,000. Couples with incomes between $40,000 and $50,000 can make smaller IRA contributions, and those with incomes above $50,000 cannot contribute on a tax-deferred basis. Tax legislation that was passed in 1997 and became effective on January 1, 1998, permits single persons and married couples with higher incomes to contribute to IRA'S. Also, nonworking spouses can contribute up to $2,000 annually on a tax-deferred basis to an IRA, even if the working spouse has an employer sponsored retirement plan; this tax deferral phases out for couples with annual income above $150,000.
25. For work-related illnesses or injuries, most workers are covered by State workers' compensation laws.
26. See Employee Tenure in the Mid-1990s, USDL 97-25 (U.S. Department of Labor), Jan. 30, 1997.
27. See "Employment Relationships in a Changing Economy" in Report on the American Workforce, 1995, pp. 3-49.
28. In most BLS analyses of data from the Displaced Worker Surveys, displacement rates have been calculated as the number of persons displaced who had been with their employer at least 3 years as a proportion of the average number of workers with at least 3 years of tenure with their current employer. In this analysis, the 3-year tenure restriction has been removed to provide a more comprehensive assessment of worker displacement in each industry.
29. See Steven Hipple, "Worker displacement in an expanding economy," Monthly Labor Review, December 1997, pp. 26-39.
30. See Contingent and Alternative Employment Arrangements, Report 900 (Bureau of Labor Statistics, August 1995). BLS has devised three measures of contingency. The narrowest measure includes wage and salary workers who have worked for their current employer 1 year or less and expect to work for this employer an additional year or less. The intermediate measure of contingency expands the first measure to include self-employed workers and independent contractors who had been and expect to be in these employment arrangements for 1 year or less. The broadest measure of contingency removes the 1-year restriction on both current tenure and expected future job duration. This broadest measure, which is used in this analysis, includes all workers who expect their jobs will end in the near, although unspecified, future for an economic reason.
31. The rate of contingent employment in business services also would have been lower if the calculation for workers in personnel supply services had been based on their expectation of future employment with the temporary-help firm, rather than the firm's client. In this analysis, workers in the personnel supply services industry were counted as contingent if they did not expect their assignment with the present client to last much longer. Many of these workers may expect their relationship with their temporary-help firm to continue well into the future, however.
32. See, for example, Henry S. Farber, Are Lifetime Jobs Disappearing? Job Duration in the United States: 1973-1993, working paper no. 5014 (Cambridge, MA, National Bureau of Economic Research, Inc., 1995). Also see Robert E. Hall, "The Importance of Lifetime Jobs in the U.S. Economy," American Economic Review, September 1982, pp. 716-24; Manuelita Ureta, "The Importance of Lifetime Jobs in the U.S. Economy, Revisited," American Economic Review, March 1992, pp. 322-35; Francis X. Diebold, David Neumark, and Daniel Polsky, Job Stability in the United States, working paper no. 4859 (Cambridge, MA, National Bureau of Economic Research, Inc., 1994); and Kenneth A. Swinnerton and Howard Wial, "Is Job Stability Declining in the U.S. Economy?" Industrial and Labor Relations Review, January 1995, pp. 293-304.
33. See, for example, Henry S. Farber, Are Lifetime Jobs Disappearing? Job Duration in the United States: 1973-1993, working paper no. 5014 (Cambridge, MA, National Bureau of Economic Research, Inc., 1995). Also see Robert E. Hall, "The Importance of Lifetime Jobs in the U.S. Economy," American Economic Review, September 1982, pp. 716-24; Manuelita Ureta, "The Importance of Lifetime Jobs in the U.S. Economy, Revisited," American Economic Review, March 1992, pp. 322-35; Francis X. Diebold, David Neumark, and Daniel Polsky, Job Stability in the United States, working paper no. 4859 (Cambridge, MA, National Bureau of Economic Research, Inc., 1994); and Kenneth A. Swinnerton and Howard Wial, "Is Job Stability Declining in the U.S. Economy?" Industrial and Labor Relations Review, January 1995, pp. 293-304.
34. See, for example, Henry S. Farber, Are Lifetime Jobs Disappearing? Job Duration in the United States: 1973-1993, working paper no. 5014 (Cambridge, MA, National Bureau of Economic Research, Inc., 1995). Also see Robert E. Hall, "The Importance of Lifetime Jobs in the U.S. Economy," American Economic Review, September 1982, pp. 716-24; Manuelita Ureta, "The Importance of Lifetime Jobs in the U.S. Economy, Revisited," American Economic Review, March 1992, pp. 322-35; Francis X. Diebold, David Neumark, and Daniel Polsky, Job Stability in the United States, working paper no. 4859 (Cambridge, MA, National Bureau of Economic Research, Inc., 1994); and Kenneth A. Swinnerton and Howard Wial, "Is Job Stability Declining in the U.S. Economy?" Industrial and Labor Relations Review, January 1995, pp. 293-304.
35. See, for example, Henry S. Farber, Are Lifetime Jobs Disappearing? Job Duration in the United States: 1973-1993, working paper no. 5014 (Cambridge, MA, National Bureau of Economic Research, Inc., 1995). Also see Robert E. Hall, "The Importance of Lifetime Jobs in the U.S. Economy," American Economic Review, September 1982, pp. 716-24; Manuelita Ureta, "The Importance of Lifetime Jobs in the U.S. Economy, Revisited," American Economic Review, March 1992, pp. 322-35; Francis X. Diebold, David Neumark, and Daniel Polsky, Job Stability in the United States, working paper no. 4859 (Cambridge, MA, National Bureau of Economic Research, Inc., 1994); and Kenneth A. Swinnerton and Howard Wial, "Is Job Stability Declining in the U.S. Economy?" Industrial and Labor Relations Review, January 1995, pp. 293-304.
36. See, for example, Henry S. Farber, Are Lifetime Jobs Disappearing? Job Duration in the United States: 1973-1993, working paper no. 5014 (Cambridge, MA, National Bureau of Economic Research, Inc., 1995). Also see Robert E. Hall, "The Importance of Lifetime Jobs in the U.S. Economy," American Economic Review, September 1982, pp. 716-24; Manuelita Ureta, "The Importance of Lifetime Jobs in the U.S. Economy, Revisited," American Economic Review, March 1992, pp. 322-35; Francis X. Diebold, David Neumark, and Daniel Polsky, Job Stability in the United States, working paper no. 4859 (Cambridge, MA, National Bureau of Economic Research, Inc., 1994); and Kenneth A. Swinnerton and Howard Wial, "Is Job Stability Declining in the U.S. Economy?" Industrial and Labor Relations Review, January 1995, pp. 293-304.
37. See Stephen J. Rose, Declining Job Security and the Professionalization of Opportunity, research report no. 95-04 (National Commission for Employment Policy, 1995).
38. See Survey of Occupational Injuries and Illnesses, 1995, Summary 97-7 (Bureau of Labor Statistics, May 1997). Incidence rates are calculated as: (N/H) X 200,000, where N = Number of injuries and illnesses H = Total hours worked by all employees during the calendar year 200,000 = Base number of annual work hours for 100 full-time equivalent workers, who are assumed to work 40 hours per week for 50 weeks. In the Survey of Occupational Injuries and Illnesses, reports are collected from a sample of about 250,000 private-sector establishments. The survey includes job-related injuries, as well as illnesses that are recognized, diagnosed, and reported during the year. Some latent illnesses, such as those caused by long-term exposure to carcinogens, are difficult to attribute to the workplace and likely are underreported in the survey. Most of the illnesses reported in the survey are easier to relate directly to work activity. Examples of these illnesses are contact dermatitis and carpal tunnel syndrome. In 1995, there were 6.6 million nonfatal work-related injuries and illnesses in the private sector. Of these, nearly 6.1 million were injuries that resulted in lost worktime, medical treatment other than first aid, loss of consciousness, restriction of work or motion, or transfer to another job. Illnesses totaled 495,000 in 1995.
39. See, for example, Occupational Injuries and Illnesses: Counts, Rates, and Characteristics, 1994, Bulletin 2485 (Bureau of Labor Statistics, April 1997).
40. See "National Census of Fatal Occupational Injuries, 1996," USDL 97-266 (U.S. Department of Labor, Aug. 7, 1997).
41. See Fatal Workplace Injuries in 1995: A Collection of Data and Analysis, Report 913 (Bureau of Labor Statistics, April 1997).
42. See, for example, Erica L. Groshen and Alan B. Krueger, "The Structure of Supervision and Pay in Hospitals," Industrial and Labor Relations Review, February 1990, pp. 134-S-146-S. See also, Mahmood Arai, "Compensating Wage Differentials versus Efficiency Wages: An Empirical Study of Job Autonomy and Wages," Industrial Relations, April 1994, pp. 249-62. Groshen and Krueger tried to estimate the level of autonomy by examining the ratio of managers to nonmanagerial employees. Such a measure does not necessarily indicate the level of autonomy, however. A factory, for example, may have only 1 supervisor for every 50 production workers, but because much of the work is done on an assembly line, the employees have little discretion in how they do their jobs. By comparison, an economic research organization may have 1 manager for every 8 nonmanagerial economists, but the nature of the work may provide these economists with considerable autonomy in applying their technical and creative skills. It should be emphasized that Groshen and Krueger limited their study to hospitals, rather than comparing the autonomy of jobs in different industries. Because work activities tend to be similar across firms within a narrowly defined industry such as hospitals, comparing the ratio of supervisors to nonsupervisory personnel may be an acceptable measure of the level of supervision or autonomy. Such a measure is not appropriate when making comparisons across different industries, which is the focus of this article. Even within a specific industry, however, the ratio of supervisors to nonsupervisory workers may not be a good indicator of autonomy because workers who are not in managerial or supervisory occupations sometimes perform supervisory duties. For example, in many hospitals, registered nurses oversee the work of licensed practical nurses. In the study by Arai, the author examined data from Sweden and defined a job as having autonomy if three criteria were met: (1) the worker had a flexitime arrangement in which he or she could choose the beginning and ending times of the work day; (2) the worker did not have to use a time clock; and (3) the worker could decide the pace of the work. This measure misses other aspects of autonomy, such as the ability to choose different work methods and control the overall quality of the output. Furthermore, as defined in the study, this measure only allows for a job to be autonomous or not autonomous; it does not reflect the fact that different jobs have different degrees of autonomy.
43. See, for example, Erica L. Groshen and Alan B. Krueger, "The Structure of Supervision and Pay in Hospitals," Industrial and Labor Relations Review, February 1990, pp. 134-S-146-S. See also, Mahmood Arai, "Compensating Wage Differentials versus Efficiency Wages: An Empirical Study of Job Autonomy and Wages," Industrial Relations, April 1994, pp. 249-62. Groshen and Krueger tried to estimate the level of autonomy by examining the ratio of managers to nonmanagerial employees. Such a measure does not necessarily indicate the level of autonomy, however. A factory, for example, may have only 1 supervisor for every 50 production workers, but because much of the work is done on an assembly line, the employees have little discretion in how they do their jobs. By comparison, an economic research organization may have 1 manager for every 8 nonmanagerial economists, but the nature of the work may provide these economists with considerable autonomy in applying their technical and creative skills. It should be emphasized that Groshen and Krueger limited their study to hospitals, rather than comparing the autonomy of jobs in different industries. Because work activities tend to be similar across firms within a narrowly defined industry such as hospitals, comparing the ratio of supervisors to nonsupervisory personnel may be an acceptable measure of the level of supervision or autonomy. Such a measure is not appropriate when making comparisons across different industries, which is the focus of this article. Even within a specific industry, however, the ratio of supervisors to nonsupervisory workers may not be a good indicator of autonomy because workers who are not in managerial or supervisory occupations sometimes perform supervisory duties. For example, in many hospitals, registered nurses oversee the work of licensed practical nurses. In the study by Arai, the author examined data from Sweden and defined a job as having autonomy if three criteria were met: (1) the worker had a flexitime arrangement in which he or she could choose the beginning and ending times of the work day; (2) the worker did not have to use a time clock; and (3) the worker could decide the pace of the work. This measure misses other aspects of autonomy, such as the ability to choose different work methods and control the overall quality of the output. Furthermore, as defined in the study, this measure only allows for a job to be autonomous or not autonomous; it does not reflect the fact that different jobs have different degrees of autonomy.
44. See, for example, Christopher Jencks, Lauri Perman, and Lee Rainwater, "What Is a Good Job? A New Measure of Labor Market Success," American Journal of Sociology, May 1988, pp. 1322-57; Maury B. Gittleman and David R. Howells, "Changes in the Structure and Quality of Jobs in the United States: Effects by Race and Gender, 1973-1990," Industrial and Labor Relations Review, April 1995, pp. 420-40.
45. See, for example, Christopher Jencks, Lauri Perman, and Lee Rainwater, "What Is a Good Job? A New Measure of Labor Market Success," American Journal of Sociology, May 1988, pp. 1322-57; Maury B. Gittleman and David R. Howells, "Changes in the Structure and Quality of Jobs in the United States: Effects by Race and Gender, 1973-1990," Industrial and Labor Relations Review, April 1995, pp. 420-40.
46. In some cases, unions provide workers with benefits that they might not receive in the absence of a union. However, there is an inverse relationship between union effect and skill level. That is, unionization does the least for the jobs that, in one important sense (level of skill), are "good" jobs. See Albert Rees, The Economics of Work and Play (New York, Harper & Row, 1973), pp. 155-57.
47. For a discussion on surveying workers to determine how different job characteristics affect workers' assessments of job quality, see Jencks and others, "What Is a Good Job?" American Journal of Sociology, May 1988, pp. 1322-57.