Originally Posted: January 30, 2003
The 1970s were immediately followed by a period of economic transition, a time when there were periods of high inflation and unemployment. Around the mid-1980s, the economy rebounded and the United States began a period of prosperity that extended throughout almost all of the 1990s. The period was characterized by changes in the structure of the economy and in American society that had started years earlier--a continuation of both the growth in importance of women in the workforce and the aging of the workforce, a change in the pattern of immigration, and a continuation of the shift towards the service sector. These changes led to an evolution in the way workers were compensated--aligning pay to organizational goals, tailoring compensation to employees’ needs, and reconfiguring employee benefit plans.
The 1980s began on an uncertain note, with worries about the country’s ability to compete in world markets and fears that high inflation rates would never end.1 From 1980-82, the unemployment rate jumped from 7.1 to 9.7 percent but then dropped from year to year, to 5.3 percent in 1989. The 1982-83 recession, however, did seem to curb inflation, as the Consumer Price Index dropped from an annual change of 13.3 percent in 1979 to 3.8 percent in 1982. By 1984, the economy had rebounded, and there ensued a long period of sustained growth. Millions of new jobs were created, and there was a resurgence of American confidence.
During this decade, a number of forces worked to limit the influence of labor unions. Foreign competition grew in industries where unionism historically has been strong--especially the automobile and steel industries.2 Additionally, employment growth had occurred in sectors--such as in services--where unions had typically not been dominant. As a result of these and other factors, trade union membership in the United States declined sharply as a share of employment (union density):
The trend toward benefits accounting for a higher proportion of compensation costs continued, though at a slower pace than earlier. The slowdown in the growth of benefits as a proportion of compensation can be attributed primarily to health insurance and employers retirement costs. Over-the-year increases in health costs peaked at 23.5 percent in March 1983, dropped to 3.5 percent in June 1985, then rose to about 13 percent in 1989. Employers introduced a number of cost containment arrangements, including shifting more of health insurance cost to their employees.
The decline in the relative importance of retirement costs reflects the shift from defined benefit to defined contribution plans and a rising stock market that enabled employers to meet their defined benefit obligations with smaller outlays than before.
The Bureau’s compensation program was influenced by these changes occurring in compensations plans, particularly the growing depth and breath of, and public interest in, data relating to benefit plans. Additionally, budget cuts in the late 1970s and early 1980s led to tough decisions regarding which BLS programs to be scaled back or eliminated. The criterion increasingly used by Congress during this time when deciding what surveys to fund was whether the survey was of broad national interest.
BLS already had extensive experience in surveying and publishing wage data; but, by 1975, the Bureau realized that it also needed to capture and publish data on benefit cost trends to produce total compensation cost trends. This initiative presented the Bureau with the challenges of identifying, measuring, and publishing data on benefit cost data every quarter, while continuing to publish timely, high-quality wage data.
To realize its objectives, BLS enhanced the ECI program. In 1980, rates of change in benefits costs were published for the first time for the private nonfarm economy and for a selected number of subseries. In 1981, wage and benefit indexes for State and local governments were added, as well as indexes for the combined private nonfarm and State and local government work-forces.
In the mid-1980s, Congress provided the Bureau additional funds to expand the ECI sample of establishments, in order to increase the number of series published, particularly in the service sector. As a result of this initiative, new series were published for health services, including hospitals, that reflected the growing national interest in information about health care costs and their potential inflationary effects. At the same time, however, major cuts were made in the IWS and AWS programs, with the surviving surveys targeting major metropolitan areas and industries of special interest, such as temporary help supply companies.
Partly as a result of the ECI sample expansion, it was determined in 1987 that it was possible to begin publishing estimates of compensation cost levels—the employer cost per hour worked for employee compensation and its components—from data collected for the ECI.3 This new data source, called Employer Costs for Employee Compensation (ECEC), replaced the Employer Expenditures for Employee Compensation that was abolished after its 1977 survey.
During the 1990s, the resurgence in American confidence begun in the 1980s continued. Except for a mild recession in 1990-91, the economy expanded continuously through the 1990s. By the end of the decade, there were large budget surpluses. Over this decade, employment in the private sector grew by more than 20 million, to about 110 million. The largest employment gains occurred in retail trade (especially eating and drinking places) and the service industry (especially business and health services).
The unemployment rate declined steadily after 1992; but, surprisingly, there was no resurgence of inflation, as had occurred in other periods of sustained growth. The unemployment rate was at a 30-year low in 1999. Despite this growing tightness in the labor market, the inflation rate, too, declined, from 6.1 percent in 1990 to 1.6 percent in 1998. The CPI increased 3.4 percent in 2000, the highest since 1990, but still low given the unemployment rate.
One of the explanations given for the low rate of price increases was moderation in wage gains. Production workers’ average hourly earnings increased 3.5 percent in 1990 and only 2.1 percent in 1992. Wage increases were in the 2.6- to 2.9-percent range during 1993-95 and in the 3.8- to 4.2-percent range during 1996-2000. Despite the relatively low rate of wage increases during the 1990s, real average earnings rose slightly, because prices increased even less.
During the first half of the decade, benefit costs rose faster than wages and salaries, but in the second half that relationship was reversed. This pattern largely reflected what was happening to employer costs for health insurance. The net effect of these changes was to return the structure of compensation in 2000 to about what it was in 1990. Dominant features of compensation in the 1990s were pay for performance and other forms of flexibility in what workers were paid.4 At this time, pay reflected stock options, profit sharing, choices among benefits, and individual awards.
The decade saw several changes in the Bureau’s core compensation programs to meet a broad set of administrative and programmatic needs, to capture changes in compensation prac- tices, and to adjust to resource constraints. Most importantly, during this time, the Bureau began planning and implementing the development of a comprehensive, integrated compensation program, the National Compensation Survey.
A major change in the Bureau’s wage survey program came with passage by Congress of the Federal Employees Pay Comparability Act of 1990 (FEPCA), which changed the pay-comparability process by creating a combination of national and local pay adjustments.5
The FEPCA provided that Federal white-collar worker pay include a national adjustment (based on the ECI) and a locality adjustment. The latter required creation of a locality-based system to replace the single General Schedule that largely disregarded locality pay differences found in the private sector. The President’s Pay Agent6 was given primary responsibility for administering FEPCA, and FEPCA named the Bureau of Labor Statistics as the agency to conduct surveys for use in determining locality pay levels.
In the early 1990s, the Bureau combined its existing occupational wage surveys by area and industry—AWS, PATC, IWS, and SCA—into a single survey, the Occupational Compensation Survey (OCS), to fulfill its part in implementing FEPCA.7 Given the tight budgetary environment and various needs of users of these existing surveys, it was decided to pursue three goals: Provide data required by FEPCA, continue to provide as much of the traditional data as possible, and streamline and cut back on the overall cost of collecting occupational wage data. The end result was development of a single survey that retained as many of the features of existing programs as possible.
After several years of collecting locality pay data in OCS, it became clear that, to gain maximum efficiencies, BLS would have to further coordinate the collection and processing of compensation data—that is, combine the OCS and ECI, ECEC, and EBS surveys. What were the driving forces that led the Bureau to adopt this umbrella approach to compensation? In effect, changing Federal pay requirements already had resulted in the integration of the AWS and PATC estimates for white-collar occupations and work levels. Also, health reform initiatives in 1993 pointed to the need for further integration of the Bureau’s compensation program. While BLS produced substantial data on employers’ health care costs and employees’ health care costs, these data could not be combined, and plan costs could not be compared to plan provisions. This led the Bureau to re-examine its compensation programs and resulted in the formulation of the National Compensation Survey (NCS). The OCS was the first program included in the NCS, in 1997. The ECI, ECEC, and EBS8 surveys are now being incorporated into the NCS.
The NCS is designed to meet a broad set of administrative and programmatic needs. It is a flexible, integrated, comprehensive effort that retains the best features of the previous surveys and does so in an efficient way, by minimizing the burden on establishments to provide wage and benefits data and by reducing duplication in data processing. The NCS’s flexible design allows BLS to adjust the survey to changing administrative and programmatic needs and to capture changes in compensation practices that the survey must reflect.
The survey sample provides wage distributions and information on wages by occupation and work level, by area. The wage distributions show, for example, average earnings in the bottom and top quartiles as well as the mean and median. Work levels show earnings for different types of job requirements within each occupation, based on a factor evaluation system that makes use of nine factors, such as knowledge, supervision required, and complexity. Because these factors are also used in the factor evaluation system to grade Federal General Schedule workers, this information can be used to derive grade level equivalents for Federal workers.
In addition to wage data, the NCS provides information on employer costs of benefits, as well as benefit incidence and provisions. This information will enable analysts to evaluate the cost of particular benefits, in addition to tradeoffs of wages for benefits. The large sample size for this wages and benefits portion of the NCS will permit the publication of new measures, such as compensation indexes for major metropolitan areas, as well as publication of more detailed industry and occupational series at the national level.
"Truth in our ideas means their power to work."
— William James
How will employee compensation programs evolve during the 21st century? Predicting developments in this field is difficult for many of the same reasons that making economic predictions is difficult. We live in a vibrant economy that routinely outpaces our ability to understand it fully and in a world where outside factors often change a system before we can model it precisely.
Like the economy as a whole, the compensation field is affected by forces working in opposite directions. Employers seek to curb labor costs to remain competitive in supplying goods and services, but at the same time may need to upgrade compensation programs to attract and retain skilled workers. Additionally, an aging population, by placing increased demands on employer health care and retirement plans, may prompt employers to adopt cost containment measures. At the same time, however, a small supply of young workers may prompt employers to enhance compensation packages to compete for qualified staff.
These opposing forces will challenge efforts to maintain correct and relevant statistics on compensation in this new century. The voluntary nature of most data collection relies on the cooperation of employers and especially human resource professionals. As these individuals face the difficult task of developing competitive compensation packages while limiting costs, their ability to comply with requests for detailed data may be strained.
Three major trends characterized employee compensation in the last years of the 20th century, and these trends will probably shape employee compensation in the early years of the new century. As with the last century, however, it is unlikely that the ways employees are compensated will evolve along a straight path. Companies and governments will try many alternative programs; some will work and become the paradigms of the 21st century; others will not and will be discarded or will be adopted in only a few workplaces.
Aligning pay to organizational goals. The first of these three major compensation trends is aligning pay to organizational goals. As our Nation’s econ-omy becomes increasingly tied to world economic conditions, competitive pressures will prompt employers to seek ways to efficiently use their work-forces. Employee compensation, in this environment, will increasingly be viewed as a tool for promoting increased productivity and innovation among workers. Compensation programs are, thus, likely to be geared to employee performance or desired characteristics such as skills or knowledge. Examples of these compensation programs include variable pay schemes that tie pay to individual or group performance and salary plans that reflect the possession or acquisition of knowledge or skills deemed critical to the success of the organization. Compensation in the 21st century is also likely to evolve in ways that tie employee pay and benefits to corporate performance. Examples include stock options and profit-sharing plans.
Tailoring compensation to employee needs. The second major trend is tailoring compensation to employee needs. This is a way of efficiently delivering compensation to employees by giving them a choice in what they want or need, rather than providing a universal program that meets the needs of the average employee. Examples include choices among health care and within retirement savings plans, flexible work schedules and telecommu-ting arrangements, and reimbursement accounts. Implicit in this flexibility is the increase of employee responsibility in making prudent choices. On the other hand, this flexibility may be constrained, particularly if significant numbers of employees make poor choices. Social policy concerns about the consequences of unwise choices, however, are less likely to stymie—than to shape—the evolution of this flexibility.
Reconfiguring employee benefit plans. The third major trend is reconfiguring employee benefit plans to provide for a defined level of employer contributions, rather than a defined level of ultimate benefit. This has been the trend in retirement plans over the 1980s and 1990s and may spread to other types of employee benefits plans. Examples include defined contribution employee health insurance plans, defined contribution retiree health insurance plans, and employer-funded reimbursement accounts. These arrangements give employers greater control over costs than in the past and greater ability to predict costs. On the other hand, employees are required to absorb more risks associated with insuring against future events than formerly. Coupled with the trend to charge employees with more responsibility for retirement savings and other benefits, the move towards defined contribution insurance arrangements may spur counter-measures to insulate employees from these risks. For example, a defined contribution health insurance plan might be required to include a core set of benefits that guards employees against catastrophic expenses.
In the 1990s these three major com-pensation trends were seen as helping to meet the needs of a mobile work-force. With the expected labor shortages of the early years of the new century, however, these trends are likely to be tempered in ways thought to increase employee incentives for remaining with the organization. Employer drives to increase efficiency and curb costs may have to be balanced with one of the traditional goals of compensation programs–to acquire and maintain an adequate supply of skilled labor.
Capturing and reporting data that adequately illuminate these major trends will be a challenge for the Bureau of Labor Statistics and others. The movement toward compensation that is based more on individual performance—and, thus, less standard—will require more data collection and innovative means of reporting results. No longer can pay be captured and reported as an hourly rate; all manner of pay such as individual bonuses, group bonuses, gainsharing, and stock options might need to be included in the new concept of pay.
The trend toward greater employee choice in compensation has already posed data collection and tabulation challenges; expansion of such choices will only compound the challenges. For example, where once employers offered only one health insurance plan, the often present choice of several plans means more data must be collected. And employee choice to substitute one benefit for another makes it more difficult to identify how much of a benefit cost is paid by the employer versus the employee. Additionally, flexibility in work hours (called flextime or flexitime) and work location (telecommuting or sometimes flexplace) make the traditional concept of compensation per hour less meaningful than before. Just as the 20th century saw an evolution in compensation statistics to address changes in the law and the growth of benefits, it is likely that statistics at the close of the 21st century will little resemble data available today.
Wages and salaries
Retirement and savings
Legally required benefits
1 Floyd Norris, Christine Bockelmann, and Paul A. Vocker, The New York Times Century of Business (New York, McGraw-Hill, 1999), p. 253.
2 See, for example, Albert E. Schwenk, "Trends in the Differences Between Union and Nonunion Workers in Pay Using the Employment Cost Index," Compensation and Working Conditions, September 1996, pp. 27-33.
3 Felicia Nathan, "Analyzing employers’ costs for wages, salaries, and benefits," Monthly Labor Review, October 1987, pp. 3-11.
4 George T. Milkovich and Jennifer Stevens, "Back to the Future: A Century of Compensation," Working Paper 99-08 (Ithica, NY, Cornell University Center for Advanced Human Resources Studies, July 1999), p. 6.
5 Public Law 101-509, Nov. 5, 1990.
6 The President’s Pay Agent consists of the Secretary of Labor, the Director of the Office of Management and Budget, and the Director of the Office of Personnel Management.
7 A more detailed description of OCS is provided in John Buckley and Elizabeth Dietz, "The Occupational Compensation Survey: A Retrospective," Compensation and Working Conditions, Fall 1997, pp. 40-46.
8 Beginning in 1990, EBS, ECI, and ECEC data were collected from the same set of establishments using consistent definitions.
9 Social Security benefits are permanently reduced based on the number of months that benefits are received prior to age 65. For example, if one individual retires at age 62, benefits will be reduced 20 percent, whereas if that same individual retires at age 64, benefits will be reduced 6-2/3 percent. This may serve as an incentive to delay retirement.