Printer-friendly version

Recent Modifications of Employee Benefits Data in the National Compensation Survey

by John E. Buckley
Bureau of Labor Statistics

Originally Posted: May 29, 2009

BLS recently discontinued the collection of data on five types of employee benefits in its National Compensation Survey, allowing limited resources to be used to provide more pertinent and timely data on benefits.

Introduction

The Bureau of Labor Statistics (BLS) has collected and published data on employee benefits for decades. In 1981, after conducting pilot studies in 1979 and 1980, BLS launched the Employee Benefits Survey program (EBS), initially publishing data on the percentage of workers provided specific types of benefits. These were prominent, widely recognized benefits such as health care, life insurance, and retirement, as well as less prominent types of benefits that were beginning to appear frequently in employee benefit packages (thrift savings plans and paid military leave, for example). These emerging benefits were originally listed in EBS publications under the category “Other benefits.” The EBS conducted annual surveys from 1981 to 1998, at which time BLS introduced the National Compensation Survey (NCS).1 Throughout this article, references to years indicate the year in which the survey was conducted. Typically, the data were published 1 or 2 years after they were collected.

Over the years, BLS modified the list of other benefits many times, adding new types of benefits that were showing up in the workplace with some consistency. Some types of benefits grew substantially and therefore were moved from the other benefits category to a category with a more permanent status. For example, in the mid-1980s, thrift savings plans became part of Defined contribution plans, as various types of contributory retirement plans became a prominent form of retirement savings. Benefits that showed no growth, or remained rare after their introduction, were subsequently dropped from the survey. Under the NCS, other benefits were listed under a variety of table headers: “Family-related benefits,” “Selected benefits,” and “Quality-of-life benefits.” The shifting of categories is evidence of the ever-changing nature of employee benefits in the workplace. For the remainder of this article, these emerging benefits are referred to simply as “other benefits.”

BLS strives to publish data that reflect the current labor market; hence, it is important that the types of benefits surveyed remain current. The NCS monitors developing trends in compensation practices by researching benefits literature and by relying on reports from BLS field economists, who have constant contact with human resources staff in sampled establishments. Also, BLS is sensitive to the burden placed on survey respondents and attempts to reduce that burden whenever possible.2 The most recent adjustment to the list of benefits reduced the number of other benefits studied from 28 to 23, with the following five benefits dropped: 1) adoption assistance, 2) educational assistance, 3) employer-provided home computers, 4) recreation benefits,3 and 5) travel accident insurance.

The statistical history of data in the other benefits category is complex, not only because of the introduction and removal of various benefits over the years, but also because of the changing scope, measurement concepts, and definitions of these benefits used by the surveys. As a result, comparing the estimates of benefits over time should be done with caution. This article provides a resource for understanding these changes in the surveys and a guide for understanding the data. For all the benefits analyzed in this article, the surveys measure the percent of workers who had the benefit available to them, not the percent who used the benefit. The first part of the article provides a general overview of BLS benefits surveys over time, and the second part focuses on the five benefits that were recently dropped from the NCS.

Overview of BLS benefits surveys

Background. In 1979 and 1980, BLS conducted pilot surveys on employer-provided benefits, leading to the establishment of the Employee Benefits Survey (EBS) program, which began publishing annual survey data in 1981. In 1981, BLS published EBS data on the percent of all full-time workers in medium and large private industry establishments (for most years, those with 100 or more workers) providing specific types of benefits.4 The types of benefits included prominent benefits such as health insurance and retirement plans.5 EBS data were also published that showed the percent of workers receiving other benefits, which included separate estimates on the incidence of paid military leave; paid funeral leave; profit sharing, savings, and stock plans; severance pay; employee discounts; gifts; in-house infirmary; relocation allowances; recreational facilities; subsidized meals; educational assistance; parking; and use of company automobile. From 1981 to 1984, no estimates for “all full-time workers” were published for other benefits; only separate estimates for full-time “professional and administrative,” “technical and clerical,” and “production” employees were published.

Changes in scope and level of detail. The EBS changed in scope over the years, expanding the survey to small private industry establishments and to State and local government establishments. To reduce costs, the EBS conducted surveys in alternate years, with data for medium and large private establishments published one year and those for small private and State and local government establishments the next year.

The National Compensation Survey (NCS) published its first benefits estimates in 2001.6 The estimates, which were for survey year 1999, were on employer-provided benefits for workers in private industry, with separate estimates for full-time workers in establishments of any size; part-time workers in establishments of any size; all workers in establishments with 1 to 99 employees; and all workers in establishments with 100 or more employees.7 Since its inception in 1999, the NCS has expanded the scope of workers for the benefits portion of the survey. The NCS began publishing estimates on employee benefits in State and local government establishments in March 2008 (with a reference date of September 2007). The NCS began publishing estimates for all civilian workers (those in private industry and in State and local government, as defined by the NCS), in addition to separate estimates for private industry workers and State and local government workers, in August 2008 (with a reference date of March 2008).8 (See appendix tables A and B for a summary of the differences in survey scope for the 1981-2008 period. For all the benefits analyzed in this article, the surveys measured the percent of workers who had the benefit available to them, not the percent who used the benefit.)

Five benefits recently dropped

As noted previously, changes in scope and level of detail over many years require that comparisons be made with caution.9 To simplify this discussion, the analysis of recently dropped benefits focuses on the benefits estimates for full-time workers in medium and large private industry establishments. One major limitation to comparisons between EBS and NCS data is the fact that published EBS data on full-time workers in medium and large private establishments are available for 1981-86, 1988-89, 1991, 1993, 1995, and 1997 (all years except those in which surveys of small private establishments or State and local governments were conducted), while published NCS data on full-time workers in all establishments (with no subcategory by establishment size) are available for 1999-2000 and 2003-2008. In addition, the two surveys do not provide published data within the same establishment and work categories. To make an imperfect, but practical, comparison, Table 1 includes data for full-time workers in medium and large private establishments for the period from 1981 to 1997 and full-time workers in all private industry establishments for the 1999-2008 period.

Table 1. Percent of full-time workers in private industry establishments(1), selected employer-provided benefits, 1981-97 and 1999-2008
Year Adoption Assistance Educational assistance(2) Employer-provided home computers Recreation benefits Travel Accident Insurance
Fully paid by employer(4) Employer and employee share cost(4) Work related Non-work related Recreational facilities(3) Fitness centers(5)
Fully paid by employer(4) Employer and employee share cost(4) Employer and employee cost not specified
1981-97

1981

 -- 30 48 -- -- -- 10 10 21 -- --

1982

 -- 29 53 -- -- -- 12 14 -- -- --

1983

 -- 26 56 -- -- -- 13 16 -- -- --

1984

 -- 29 53 -- -- -- 17 16 -- -- --

1985

 -- 27 49 -- -- -- -- -- 33 -- 52

1986

 -- -- -- -- -- -- -- -- -- -- --

1988

 5 -- -- 70 18 -- -- -- 25 -- 49

1989

 5 -- -- 69 19 -- -- -- 28 -- 53

1991

 8 -- -- 72 23 -- -- -- 26 -- 42

1993

 7 -- -- 72 22 -- -- -- 27 -- 44

1995

 11 -- -- 65 18 -- -- -- -- 19 41

1997

 10 -- -- 67 20 -- -- -- -- 21 42
1999-2008

1999

 6 -- -- 47 12 -- -- -- -- 10 22

2000

 6 -- -- 44 11 -- -- -- -- 10 17

2003

 10 -- -- -- -- 3 -- -- -- -- --

2004

 11 -- -- -- -- 3 -- -- -- -- --

2005

 11 -- -- 56 16 3 -- -- -- 14 26

2006

 12 -- -- 56 16 3 -- -- -- 14 25

2007

 12 -- -- 56 17 3 -- -- -- 14 25

2008

 13 -- -- 56 17 3 -- -- -- 15 26

Footnotes:
(1) Estimates from 1981 through 1997 include only full-time workers in medium and large establishments (those with 100 or more workers); estimates from 1999 through 2008 include full-time workers in establishments with 1 worker or more.
(2) From 1981 through 1985, data on educational assistance were provided on the basis of whether the benefit was fully or partially paid for by the employer. Beginning in 1988, data were provided on whether the benefit was work related or nonwork related.
(3) After 1984, method of funding (e.g., all or partial payment by employer) was no longer separated.
(4) Small percentages of workers were in firms in which some, but not all, workers were eligible for the benefit. These numbers are not included here.
(5) Fitness center benefits include those fully paid by the employer and those for which the employer and employee share the cost. Fitness centers and recreational facilities are defined differently; see Appendix for NCS definitions.

Note: The 1981–84 data are for professional and administrative workers only. Dashes indicate no data were collected or published.

Educational assistance. Educational assistance is the only special benefit that was studied in the inaugural year of the survey (1981) and remained in the list of “other benefits” through 2008. EBS data on educational assistance from 1981 to 1984 (shown in table 1) include only full-time, professional and administrative workers10 in medium and large private establishments who were eligible to participate in plans for which the employer either paid all or part of the expenses. The percent of workers offered full or partial reimbursement for educational benefits was relatively constant from 1981 to 1985. The collection of data on whether educational assistance was fully or partially funded by the employer was dropped in 1988, and the EBS began collecting data on whether the assistance was job-related or not. Both types of educational assistance remained fairly steady through 1997, when 67 percent of workers were offered job-related assistance and 20 percent were offered non-job-related assistance. In 1999, after the NCS was introduced, estimates that include full-time workers in small, medium, and large establishments showed that 47 percent of these workers had access to work-related educational assistance; by 2005, that figure had risen to 56 percent and remained steady through 2008. Twelve percent of full-time workers in private industry had access to non-work related educational assistance in 1999; by 2008, that figure had increased to 17 percent.

Recreational benefits. Recreational facilities as an employer-provided benefit was collected in 1981 and remained on the list through the mid-1990s. For the first four survey years (1981–84), separate estimates were published on the percent of professional and administrative workers in plans where the employer either fully or partially defrayed the expenses. The percent of professional and administrative workers eligible to use employer-subsidized recreational facilities increased from about 20 percent in 1981 to 33 percent for all full-time workers in medium and large private establishments in 1985; by 1993, that figure had decreased to 27 percent. In 1994, the EBS dropped the collection of data on recreational facilities benefits and began collecting data on fitness center benefits.

In 1995, 19 percent of full-time workers in medium and large private establishments were eligible for fitness center benefits, a figure that climbed to 21 percent in 1997. In the 1999 estimates, which include full-time workers in establishments of any size, the comparable figure dropped to 10 percent. This can be explained in part by the inclusion of small private establishments, which typically are less likely than larger establishments to offer such benefits. The 2005–08 estimates show a slightly higher rate of access, at 14 to 15 percent. Still, fitness center benefits have remained somewhat uncommon among full-time private industry workers.

Adoption assistance. Adoption assistance was introduced into the EBS in 1988. That year, 5 percent of full-time workers in establishments with 100 or more workers were eligible for this benefit. The percent of workers provided the benefit ranged from 5 to 11 percent between 1988 and 1997. From 1999 to 2008, the percent ranged from 6 percent to 13 percent. Although survey scope and size of establishments covered differ in the two surveys, the EBS and NCS estimates show that adoption assistance has remained a relatively rare benefit.

Employer-provided home computers. In the short period that information on employer-provided home computers was collected (2003 through 2008), the percent of workers with access to the benefit never exceeded 3 percent for full-time workers. Hence, it was dropped from the survey.

Travel accident insurance. Information on travel accident insurance was first collected in 1985. In that year, 52 percent of all full-time workers in medium and large private establishments were provided this benefit. The percentage dropped to 42 percent in 1997. In 1999, with smaller private establishments included in the estimate, only 22 percent of full-time workers had access to travel accident insurance. The percentage remained relatively steady, with 26 percent of full-time workers in all private industry being offered the benefit in 2008.

Conclusion

As a result of monitoring compensation trends in the labor market, efforts to comply with the directive to reduce respondent burden whenever possible, and the careful consideration of years of published data have compelled BLS to drop five types of benefits from the National Compensation Survey. The benefits dropped remained rare among workers in private industry, were not in great demand, or showed little growth in recent years. Comparisons of the employer-provided benefits estimates from year to year must be made with caution because of the changing scope and definitions used by the surveys. At the same time, the changing definitions of employer-provided benefits over the years, first in the EBS and then in the NCS, attests to the fact that the compensation field is constantly changing. The benefits studied under the NCS will no doubt continue to change in the future, as BLS strives to provide the public with data that are of the utmost relevance.11

 

Appendix

Table 1 provides selected published estimates for the five benefits being dropped from the National Compensation Survey. Because the scope of the surveys varied over the period from 1981 to 2008, only broad conclusions should be made from these data when analyzing changes over time. Historical benefits data for the NCS and its predecessor surveys can be found in the 2009 National Compensation Survey Publications List, on the Internet at http://www.bls.gov/ncs/ncspubs.htm.

The data in tables A and B show some of the differences in the survey scope over the 1981–2008 period. The gaps in the annual data indicate that there were no comparable data for the missing years or no survey was conducted that year.

Table A. Scope of survey, full-time workers in medium and large private industry establishments, Employee Benefits Survey, 1981–97.
Year Establishment size Occupational group Sector Geography

1981-1986
The minimum employment size was 50, 100, or 250, depending on the industry(1)
Professional and administrative; technical and clerical; and production workers
Private
US except Alaska and Hawaii

1988-89
100 or more workers
Professional and administrative; technical and clerical; and production and service workers
Private
US except Alaska and Hawaii

1991, 1993, 1995, 1997
100 or more workers
Professional and technical; clerical and sales, and blue-collar and service workers
Private
United States

Footnotes:
(1) For industry size details, see Employee Benefits in Medium and Large Private Establishments in the United States, 1981, Bulletin 2140 (Bureau of Labor Statistics, August 1982), p. 43; available on the Internet at http://www.bls.gov/ncs/ebs/sp/ebbl0038.pdf.

NOTE: The EBS produced data from 1981 through 1997. This study uses data from only medium and large private establishments surveyed under the EBS; to simplify the analysis, small private establishments and State and local government establishments are excluded. The EBS published data on medium and large private establishments in 1986; however, no benefits in the "Other benefits" category were published that year. Beginning in 1999, when the National Compensation Survey took over the collection and publication of benefits data, establishments with 1 to 99 workers were added to the survey.


Table B. Scope of survey, full-time workers in private industry establishments with one or more workers, National Compensation Survey, 1999 - 2008
Year Establishment size Occupational group Sector Geography

1999-2000
1 or more workers
Professional and technical; clerical and sales, and blue-collar and service workers
Private
United States

2003-06
1 or more workers
White-collar, blue-collar and service workers
Private
United States

2007-08
1 or more workers
Management and professional; service; sales and office; natural resources, construction, and maintenance; and production, transportation, and material moving.
Private
United States

Definitions for the five discontinued NCS benefits

Adoption assistance. Financial aid given to either single or married employees for the purpose of covering all or part of the cost of adopting a child.

Educational assistance. Educational allowances provide employees with assistance in paying for tuition and/or books for training and education courses that permit the employee to acquire additional general knowledge or to develop particular knowledge or skills.

Employer-provided home computers. This employer-provided benefit helps the employer by giving the employee access to company data and the employee’s work projects. If the employee cannot go into the office they may still be productive by using his or her home computer.

Employers may purchase the computers outright and provide them to employees. Other options include the following:

  • Allowing employees to lease computers at a nominal rate with the employee owning the computer at the end of the lease


  • Loans at low or no interest rates


  • Computer subsidies or grants

If an employer only provides a home computer as part of a flexible workplace arrangement, employees are not considered as having employer-provided home computers.

Fitness Center Benefit. A program where the employer fully or partially pays the cost of membership in a fitness center or health club. The club or center may be on or off the employer’s premises.

Note: The former benefit type “recreation facilities” is a more inclusive category than the more recent “fitness centers” benefit. Employer-subsidized recreational facilities can include golf clubs, swimming pools, tennis courts, and other similar facilities, whether they are provided on-site, off-site, or whether membership dues are reimbursed in full or in part by the employer.

Travel accident insurance (also called “traveler’s insurance”). A specific form of accidental death and dismemberment insurance that provides payments in the event of death or injury of an employee who is traveling on company business.

 

John E. Buckley
Economist, Division of Compensation Data Analysis and Planning, Office of Compensation and Working Conditions, Bureau of Labor Statistics.
Telephone: (202) 691-6299; E-mail: Buckley.John@bls.gov.

 

Notes

1 The National Compensation Survey started collecting data on occupational wages and employer cost of total compensation (wages and benefits, according to NCS definition) in 1996.

2 The Paperwork Reduction Act of 1980 is the Federal legislation that established the principle that surveys should minimize reporting burden. The text of the Paperwork Reduction Act can be found on the National Archives website at http://www.archives.gov/federal-register/laws/paperwork-reduction/3501.html.

3 Recreation benefits, a category used only for discussion purposes in this article, includes recreation facilities and fitness centers. See appendix for definitions.

4 See appendix for details.

5 EBS published more extensive incidence data on paid holiday, vacation, and personal leave; work schedules; paid lunch time and rest time; sick leave; health insurance for employees; health insurance for dependents; retirement pensions; life insurance; accident and sickness insurance; and long-term disability insurance, including the percent of employees offered various plan provisions. Plan provisions included a wide variety of measures, for example, the number of days of paid holidays per year, the annual coinsurance limit of major medical coverage, and whether the minimum service requirement of a pension plan is based on age, years of service, or both.

6 See Employee Benefits in Private Industry, 1999, USDL 01-43 (U.S. Department of Labor), December 19, 2001; available on the Internet at http://www.bls.gov/ncs/ebs/sp/ebnr0006.pdf.

7 The NCS published benefits estimates for additional establishment size categories in 1999, 2007, and 2008.

8 See Employee Benefits in State and Local Government--September 2007, USDL 08-0408 (U.S. Department of Labor) March 25, 2008; available on the Internet at http://www.bls.gov/news.release/ebs3.nr0.htm; see also Employee Benefits in the United States, March 2008, Bulletin 2715 (Bureau of Labor Statistics, September 2008); available on the Internet at http://www.bls.gov/ebs/#bulletins.

9 Although standard errors have been calculated for the 2008 estimates presented in this article, no standard errors have been calculated for the 1981–2007 estimates; therefore, the quality of comparisons made cannot be verified with a statistical test.

10 EBS published separate estimates for professional and administrative employees, technical and clerical employees, and production employees in the early years of the survey. Estimates for all occupations were not published for other benefits until 1985.

11 The following link provides the most recent data on employer-provided employee benefits: www.bls.gov/ncs/ebs.