NOTE: The Prevalence of Employer-Provided Benefits by Industry of Employment and Implications for Social Security Disability Insurance Claiming Behavior

by
Social Security Bulletin, Vol. 79 No. 1, 2019

Policymakers seek effective ways to restore or maintain the labor force participation of current and potential Social Security Disability Insurance (DI) beneficiaries. The availability of certain types of employer-provided benefits may affect whether workers with health impairments are able to maintain employment. In this research note, we use National Compensation Survey data to estimate the availability of employer-sponsored health insurance and paid leave by industry. We find that most types of employer-provided benefits were offered to greater shares of state and local government workers than to private-sector workers. However, private short- and long-term disability insurance offerings were more prevalent in the private sector.


Ӧzlen Luznar is with the Division of Modeling, Office of Analytics and Improvements, Office of Analytics, Review, and Oversight, Social Security Administration (SSA). Jackson Costa is with the Office of Research and Demonstration, Office of Research, Demonstration, and Employment Support, Office of Retirement and Disability Policy, SSA.

Acknowledgments: We thank Lawrence Surowitz, Mark Trapani, and Jesus Ranon for their helpful comments and suggestions; Kristin Monaco and John Bishow at BLS for enabling us to access the NCS data (and for clearing our data analyses); and Susan Wilschke for providing both. We also thank David Weaver, who supported and encouraged this research.

The findings and conclusions presented in the Bulletin are those of the authors and do not necessarily represent the views of the Social Security Administration.

Introduction

Selected Abbreviations
BLS Bureau of Labor Statistics
DI Disability Insurance
LTD long-term disability
NAICS North American Industry Classification System
NCS National Compensation Survey
SSA Social Security Administration
STD short-term disability

The two largest federal income assistance programs for individuals with disabilities are Social Security Disability Insurance (DI) and Supplemental Security Income (SSI), both administered by the Social Security Administration (SSA). The Social Security Act defines disability as “the inability to engage in substantial gainful activity (SGA) by reason of any medically determinable physical or mental impairment which can be expected to result in death, or which has lasted or can be expected to last for a continuous period of not less than 12 months.” Although both programs use that definition, their eligibility requirements differ. DI provides cash benefits to workers who leave the labor force because of a mental or physical disability.1 These individuals have a work history and an earnings record, they have paid Federal Insurance Contributions Act (FICA) taxes, and they are unable to earn at the SGA level because of their disabling condition(s). DI benefits replace some of the earnings lost because of disability. SSI, on the other hand, provides means-tested payments to working-age individuals with low income and few resources who are not able to work because of disability. Some SSI recipients also qualify for a modest DI benefit.

Applications for DI disabled-worker benefits increased from about 1.3 million in 2000 to a peak of 2.8 million in 2010. About 2.2 million workers applied for DI benefits in 2016 (SSA 2018, Table 60). Various factors may have contributed to the fluctuating trends in DI applications and enrollment over the years, such as changes in the prevalence of employee access to employer-sponsored health insurance (Livermore, Wittenburg, and Neumark 2014). Employers may voluntarily offer health insurance and paid leave to provide their employees with security against unforeseen events that might arise during their productive years. Such benefits offer economic stability for workers and their families, which in turn helps to insulate the wider U.S. economy against potential shocks. Employer-provided benefits can be seen as a partnership among businesses, individual workers, and the government, in that employer costs for many such benefits are partially offset by tax advantages. In addition to providing employee benefits voluntarily, employers are obligated to contribute to government-administered programs via taxes. Social Security coverage (for disabled and retired workers and their dependents), unemployment insurance, workers' compensation, and family and medical leave2 are, for most jobs, mandatorily provided (Employee Benefit Research Institute 2011, Chapter 1). One would expect to observe little or no variation in the coverage of most workers under these mandatory programs. However, the voluntary provision of employee benefits, such as employer-sponsored health insurance, paid leave, and short- and long-term disability insurance, might vary widely.

Analyzing the potential interactions between elective employer-provided benefits and application for DI benefits can provide useful insights to employers, workers, and policymakers. In this research note, we examine the extent to which employers in different industries offered selected types of health insurance and paid leave to employees and compare those patterns with the industry distribution of DI beneficiaries' last employment. Such statistics could be used to test the assumption that employees who have access to employer-sponsored health insurance and paid leave would use those benefits to delay or forestall a workforce exit because of a disability. Depending on the nature and severity of the disability, access to timely care also could dissuade some workers from filing disability claims (Dement and others 2015; Schimmel Hyde and Livermore 2016).

Background

Several studies have explored the relationship between the health condition of U.S. workers, the availability of health insurance, and DI application volume (Livermore, Stapleton, and O'Toole 2011; Du Bois and Donceel 2008). Salkever and others (2000) study DI beneficiaries who have a mental disorder and examine the severity of their diagnoses, their sociodemographic characteristics, their access to health care, and their long-term disability (LTD) benefit claiming behavior. The authors observe that LTD claimants may consider several factors in deciding whether to return to work, including the generosity of the LTD benefits, the severity of the disabling condition, and the availability of supplemental compensation. They further observe that workers who leave their jobs because of a severe mental health problem may need ongoing treatment from their providers and adequate supports from employers after they return to work. The authors conclude that greater access to specialized mental health treatment and generous health insurance coverage could reduce the number of disability claims and thereby allow some workers to remain employed or return to work after claiming LTD benefits. Nonetheless, for some workers with severe mental health conditions, DI application may be inevitable.

Schimmel Hyde and Livermore (2016) assess the effect of the Affordable Care Act (ACA) of 2010 on timely access to health care among employed individuals with work-limiting disabilities. The authors find that many disabled workers work part-time or intermittently, which makes them ineligible for employer-sponsored health insurance and other benefits. Although the ACA expanded access to health insurance, the authors find that disabled workers have greater difficulty obtaining timely health care than individuals without work-limiting conditions. The authors note that disabled workers' timely access to health care may need to be reexamined now that the ACA is fully implemented.

O'Leary, Walker, and Roessel (2015) present a table that reports the distribution of DI disabled-worker beneficiaries in 2013 by industry of last employment. The authors identify industries using North American Industry Classification System (NAICS) designations.3 We reprint that table here as Table 1, which shows that most of the beneficiaries in 2013 had previously worked in the services, retail trade, or manufacturing industries. The least common industries of previous employment among 2013 DI beneficiaries were wholesale trade, mining, and public administration.4 It is worth noting that the availability, variety, and generosity of health insurance and paid leave benefits vary across industries. Knowledge of these variations in employer-provided benefits may help researchers to better understand the composition of the DI beneficiary population and may improve efforts to help workers with impairments.

Table 1. DI disabled-worker beneficiaries in 2013: Percentage distribution by industry of last employment, with detail by sex and summary comparisons to total employment in 2002 and 2012
Industry Disabled workers (2013) Total employment
Men Women Total 2002 2012
Agriculture, forestry, and fishing 3.7 1.2 2.5 1.7 1.5
Mining 1.1 0.1 0.6 0.4 0.6
Construction 7.8 0.7 4.4 5.0 4.1
Manufacturing 15.3 9.2 12.4 11.4 8.7
Transportation, communications, electric, gas, and sanitation 6.9 3.3 5.1 3.6 3.6
Wholesale trade 3.5 1.9 2.7 4.2 4.2
Retail trade 16.2 19.4 17.8 11.3 10.9
Finance, insurance, and real estate 6.2 8.1 7.1 5.9 5.7
Services 38.6 55.6 46.8 40.3 44.6
Hotels, rooming houses, camps, and other lodging places 1.2 2.3 1.7 a a
Personal services 0.7 1.6 1.1 a a
Business services 12.9 11.3 12.1 a a
Automotive and miscellaneous repair services and parking 3.2 1.3 2.3 a a
Amusement, recreation, parks, and museums 3.5 3.5 3.5 a a
Health services 4.2 16.3 10.1 a a
Legal services 0.2 0.8 0.5 a a
Educational services 1.9 3.6 2.7 a a
Social services 2.7 5.8 4.2 a a
Membership organizations 1.2 1.6 1.4 a a
Engineering, accounting, research, management, and related services 3.8 4.3 4.1 a a
Miscellaneous services 3.0 3.4 3.2 a a
Public administration 0.6 0.6 0.6 16.1 16.1
SOURCE: Reprinted from O'Leary, Walker, and Roessel (2015, Table 3); based on SSA 2013 Disability Analysis File, SSA 2012 Continuous Work History Sample, and Bureau of Labor Statistics (2013).
NOTES: "Industry of last employment" reflects the most recent job (within 10 years) prior to the date of current eligibility for disabled-worker benefits. If a beneficiary worked more than one job in the most recent year, the highest-paying of those jobs determines the industry.
Calculations include adjustments to account for slight interagency differences in industry category definitions as well as to intra-agency revisions to those definitions over time.
Rounded components of percentage distributions do not necessarily sum to 100.0.
a. Inconsistencies in the service-industry subcategory definitions used by SSA and the Bureau of Labor Statistics (and between those used by the Bureau of Labor Statistics in 2002 and in 2012) prevent useful cross-column comparisons.

We extend the findings of O'Leary, Walker, and Roessel (2015) by presenting estimates of the prevalence of employer-sponsored health insurance and paid leave benefits by NAICS industry sector. In 2016, more than 2.2 million workers applied for DI benefits claiming a qualifying medical condition preventing them from working, and nearly 500,000 applicants were awarded benefits (SSA 2018, Table 60). To what extent might access to employer-sponsored health insurance and paid leave affect DI application rates? Some researchers and advocates suggest that increased availability of health insurance and paid leave increases the likelihood that employees receive adequate preventive and routine care, keeping them healthier for the duration of their working years. Workers with such access would presumably be better able to remain productive, maintain a higher standard of living, and avoid applying for DI or public welfare programs (Stapleton and others 2006). If workers are healthier, the DI award rate might decrease as well. Additionally, workers who have access to paid (or unpaid) leave are better positioned to keep their jobs through a prolonged absence caused by a medical condition.

We examine four types of health insurance: medical, prescription drug, dental, and vision coverage. We likewise examine four categories of paid leave: sick; personal; “other,” comprising paid time off for funerals, military service, or jury duty; and “any,” which includes all types in the preceding three categories as well as paid holidays and vacations. We also present data on access to private short-term disability (STD) and LTD insurance plans.

Data and Methodology

Under the authority of an interagency agreement between SSA and the Bureau of Labor Statistics (BLS), we use data collected by BLS in the 2013 National Compensation Survey (NCS). The NCS is a voluntary establishment-based survey that produces nationally representative estimates of employer-provided benefit availability, provisions, and costs. NCS respondents include private-sector employers from each NAICS industry sector except the agriculture, forestry, and fishing sector. The NCS also includes public-sector employers at the state and local level but excludes the federal government (Pierce 1999).5,6

Our variables of interest are industry of employment, identified by 2-digit NAICS code; whether the employer is a private-sector establishment, a state government, or a local government; and the availability of four types of health insurance, four categories of paid leave, and private STD and LTD insurance for employees. Although the NCS excludes the agriculture, forestry, and fishing sector, the scope of our industry coverage is otherwise consistent with that of O'Leary, Walker, and Roessel (2015), which accounts for most but not all U.S. employment. Note that our industry-level results do not correspond directly with those in O'Leary, Walker, and Roessel (2015). Those authors used administrative records from SSA as their primary data sources while we use the NCS. In addition, the level of detail in the industry designations differ in some instances between their study and ours.

Our data set covers 28,065 establishments in local government, state government, and the private sector. To make our estimates nationally representative, we use BLS sample weights in our statistical analysis. To eliminate the possibility of identifying individual survey respondents from the estimates, BLS suppressed data in some instances prior to our analysis.

Results

Table 2 presents the distribution of the employees of the respondent establishments in our sample by employer type, with detail for full- and part-time workers. Most of the employees reported by the sample establishments worked in the private sector (88.8 percent). State governments employed 7.8 percent of the sample and local governments employed the remaining 3.4 percent. Full-time workers dominated nearly every subgroup in the sample. Almost 98 percent of local government employees worked full time in 2013. Among state government and private-sector employees, about 88 percent and 86 percent, respectively, worked full time. More than two-thirds of employees in the sample worked in one of three industries: finance, insurance, and real estate (24.3 percent); manufacturing (24.0 percent); and retail trade (19.2 percent).

Table 2. Employment in NCS respondent establishments: Full- and part-time workers, by employer type, 2013
Employer type NAICS code All workers Full time Part time Percentage full time
Number Percentage distribution
Total . . . 28,065 100.0 24,195 3,870 86.2
Local government . . . 946 3.4 922 24 97.5
State government . . . 2,185 7.8 1,916 269 87.7
Private sector . . . 24,934 88.8 21,357 3,577 85.7
Industry
Mining 21 262 0.9 259 3 98.9
Construction 23 2,022 7.2 1,921 101 95.0
Manufacturing 31–33 6,725 24.0 6,623 102 98.5
Wholesale trade 42 1,619 5.8 1,526 93 94.3
Retail trade 44–45 5,390 19.2 2,921 2,469 54.2
Transportation, communications, and utilities 48–49 1,515 5.4 1,240 275 81.8
Finance, insurance, and real estate 52 6,831 24.3 6,516 315 95.4
Services 81 821 2.9 572 249 69.7
Public administration 92 2,880 10.3 2,617 263 90.9
SOURCE: Authors' calculations based on the 2013 NCS.
NOTES: The NCS does not cover the agriculture, forestry, and fishing sector (NAICS code 11) or the federal government.
Employment in public administration (NAICS code 92) does not equal the sum of employment in local and state government. The NCS may code certain government employees according to job tasks or responsibilities that are typically associated with a particular private-sector industry.
. . . = not applicable.

Table 3 shows the proportions of workers who had access in 2013 to various types of employer-provided benefits in state and local government and the private sector. For each type of health insurance, the proportions were highest for state government workers and were lowest for private-sector workers. For example, 97.3 percent of state government workers had access to medical insurance, whereas only 83.2 percent of local government workers and 76.9 percent of private industry workers had such access. Prescription drug coverage was nearly as prevalent as medical coverage for all three employer types. Dental insurance was available to 63.5 percent of state government workers, 61.7 percent of local government workers, and 50.7 percent of private-sector workers. Among the types of health insurance, vision benefits were the least prevalent (47.3 percent or lower).

Table 3. Percentage of workers with access to selected benefits, by employer type, 2013
Benefit type State government Local government Private sector
Health insurance
Medical 97.3 83.2 76.9
Prescription drugs 93.9 81.2 75.4
Dental 63.5 61.7 50.7
Vision 47.3 44.6 25.2
Paid leave
Any (holiday, vacation, sick, personal, or other leave) 98.1 91.9 90.3
Sick 97.9 88.3 84.4
Personal 66.2 47.5 37.2
Other (leave for funerals, military service, or jury duty) 98.0 89.7 74.9
STD insurance 29.3 24.8 45.3
LTD insurance 30.8 31.3 34.2
SOURCE: Authors' calculations based on 2013 NCS.

The patterns of access to paid leave were similar to those for health insurance, with the proportions highest for state government workers and lowest for private-sector workers. About 98 percent of state government workers had access to sick or other paid leave in 2013, a proportion no lower than that for access to any type of paid leave. The proportions of private-sector workers with access to sick leave, other leave, and any leave were 84.4 percent, 74.9 percent, and 90.3 percent, respectively. Access to private STD and LTD insurance was much lower than access to almost all the other benefit types. Among state government workers, only 29.3 percent had access to STD insurance and 30.8 percent had access to LTD insurance. The respective access rates for local government workers were comparable, at 24.8 percent and 31.3 percent. Private-sector workers had the highest access rates, at 45.3 percent and 34.2 percent, respectively.

O'Leary, Walker, and Roessel (2015) reported that 46.8 percent of 2013 DI disabled-worker beneficiaries had last worked in the services industry, which accounted for 44.6 percent of total employment in 2012. However, only 0.6 percent of DI beneficiaries had worked in public administration, in stark contrast with that industry's 16.1 percent share of total employment in 2012. Variations in the industry-of-employment distributions of current workers and of disabled-worker beneficiaries—and the causes of those variations—are of research interest. Different levels of access to health insurance, paid leave, and private disability insurance for employees across industries may be one of the factors. As a preliminary exploration, we present the proportions of workers with access to employer-sponsored health and leave benefits, by industry sector and selected subsector,7 in Tables 4 and 5.

Table 4 covers access to health insurance in 2013. Regardless of industry, access to medical insurance was the most prevalent, followed in descending order by access to prescription drug, dental, and vision coverage.

The industry sectors in which employers were most likely to offer medical coverage were mining; manufacturing; and finance, insurance, and real estate. Medical coverage was available to about 92 percent of employees in each of those NAICS sectors.8 Workers in the services and retail trade industry sectors were the least likely to have access to medical coverage, at 49.5 percent and 60.7 percent, respectively.9 Access to prescription drug coverage ranged from 91.6 percent for workers in the finance, insurance, and real estate sector to 48.3 percent in the services sector.10 The availability of dental insurance ranged from a high of 75.4 percent for workers in the finance, insurance, and real estate industry to a low of 27.9 percent for workers in the services industry. The industry sector with the highest proportion of workers who had access to vision coverage (45.0 percent) was public administration.11 The retail trade industry sector offered vision coverage to 14.6 percent of its workers, the lowest sector-level access rate.12

Table 4. Percentage of workers with access to health insurance, by coverage type and selected NAICS industry sector and subsector, 2013
Industry sector and subsector NAICS code Medical Prescription drugs Dental Vision
Mining 21 91.5 89.7 62.5 24.3
Construction 23 70.8 68.7 39.4 25.3
Manufacturing 31–33 92.3 90.5 63.6 35.6
Food, beverage, and tobacco products; textiles; apparel; and leather and allied products 31 87.5 83.8 60.2 34.3
Wood, paper, and printing and related support; petroleum and coal; chemicals, plastics, and rubber products; and nonmetallic mineral products 32 90.6 88.8 59.7 26.5
Primary metals; fabricated metals; machinery; computer and electronic products; electrical equipment; appliances and components; transportation equipment; furniture; and miscellaneous 33 94.5 93.3 66.3 39.5
Wholesale trade 42 86.6 84.6 51.1 28.0
Retail trade 44–45 60.7 59.0 37.3 14.6
Motor vehicles and parts; furniture and home furnishings; electronics and appliances; building materials and garden equipment and supplies; food and beverages; health and personal care items; gasoline; and clothing 44 56.5 54.9 32.7 17.1
Sporting goods; general merchandise; and miscellaneous and nonstore retail 45 68.3 66.6 45.9 9.9
Transportation, communications, and utilities 48–49 84.0 83.2 58.1 31.9
Air; rail; water; truck; transit and other ground passenger service; pipelines; scenic and sightseeing tours; and support activities 48 77.6 77.0 51.4 21.5
Postal service; couriers and messengers; and warehousing and storage 49 98.1 97.3 73.1 55.1
Finance, insurance, and real estate 52 92.5 91.6 75.4 29.2
Services 81 49.5 48.3 27.9 16.9
Public administration 92 88.3 86.0 62.2 45.0
SOURCE: Authors' calculations based on 2013 NCS.
NOTES: The NCS does not cover the agriculture, forestry, and fishing sector (NAICS code 11) or the federal government.
For consistency with O'Leary, Walker, and Roessel (2015, Table 3), certain NAICS industry sectors are omitted.

Table 5 covers paid leave and private disability insurance. By definition, access to any leave equals or exceeds access to the specific types of leave. Access to any paid leave ranged from 98.5 percent of workers in the manufacturing sector to 79.1 percent for services sector workers.13 In all industries, sick leave was available to greater shares of workers than was personal leave. More than 90 percent of employees in the following industries had access to sick leave: finance, insurance, and real estate (97.7 percent); manufacturing (96.6 percent), including each of the three NAICS subsectors of the manufacturing sector; mining (95.5 percent); wholesale trade (92.4 percent); and public administration (91.4 percent).14 Workers in the finance, insurance, and real estate industry sector had the most access to personal leave (67.2 percent)15 and those in the construction sector had the least (18.6 percent).

Access to STD and LTD benefits may delay or lower the likelihood of a worker's application for DI benefits. For many workers, DI might be the only potential source of income if a disability requires an extended absence from work. However, if a worker has enrolled in employer-sponsored disability insurance, he or she may choose not to apply for DI. In 2013, employees in the finance, insurance, and real estate industry sector had the highest rate of access to both STD and LTD insurance (72.4 percent and 74.5 percent, respectively). Employees in the construction, services, and retail trade industry sectors had among the lowest rates of access to STD and LTD insurance.

Table 5. Percentage of workers with access to paid leave and private disability insurance, by type and selected NAICS industry sector and subsector, 2013
Industry sector and subsector NAICS code Paid leave STD insurance LTD insurance
Any a Sick Personal Other b
Mining 21 96.0 95.5 21.1 83.6 60.3 60.2
Construction 23 86.7 72.9 18.6 51.3 29.2 18.4
Manufacturing 31–33 98.5 96.6 38.1 87.8 63.2 44.5
Food, beverage, and tobacco products; textiles; apparel; and leather and allied products 31 95.9 92.9 27.3 77.9 56.5 31.4
Wood, paper, and printing and related support; petroleum and coal; chemicals, plastics, and rubber products; and nonmetallic mineral products 32 (X) 97.0 35.4 87.4 62.1 46.6
Primary metals; fabricated metals; machinery; computer and electronic products; electrical equipment; appliances and components; transportation equipment; furniture; and miscellaneous 33 (X) 97.6 42.6 91.1 65.9 47.9
Wholesale trade 42 95.7 92.4 38.9 81.5 50.4 44.2
Retail trade 44–45 80.8 71.2 30.7 64.4 26.1 14.6
Motor vehicles and parts; furniture and home furnishings; electronics and appliances; building materials and garden equipment and supplies; food and beverages; health and personal care items; gasoline; and clothing 44 83.3 72.9 26.5 62.9 29.5 17.1
Sporting goods; general merchandise; and miscellaneous and nonstore retail 45 76.0 68.2 38.5 67.3 19.9 10.1
Transportation, communications, and utilities 48–49 94.4 89.4 47.8 82.5 47.1 36.1
Air; rail; water; truck; transit and other ground passenger service; pipelines; scenic and sightseeing tours; and support activities 48 93.0 86.3 35.0 77.1 37.0 32.5
Postal service; couriers and messengers; and warehousing and storage 49 97.7 96.4 76.3 94.7 69.8 44.0
Finance, insurance, and real estate 52 (X) 97.7 67.2 95.1 72.4 74.5
Services 81 79.1 72.6 30.2 57.3 27.4 17.3
Public administration 92 93.7 91.4 53.7 93.1 26.0 30.2
SOURCE: Authors' calculations based on 2013 NCS.
NOTES: The NCS does not cover the agriculture, forestry, and fishing sector (NAICS code 11) or the federal government.
For consistency with O'Leary, Walker, and Roessel (2015, Table 3), certain NAICS industry sectors are omitted.
(X) = suppressed by BLS to prevent possible disclosure of information about specific respondents.
a. Holiday, vacation, sick, personal, or other leave.
b. Leave for funerals, military service, or jury duty.

Conclusion

Our analysis of the 2013 NCS provides descriptive statistics on the availability of employer-sponsored health insurance and paid leave benefits by NAICS industry sector. We find that greater proportions of state and local government employees had access to medical, dental, and vision benefits than private-sector workers, yet the latter had greater access to LTD and STD insurance. Access to certain types of benefits varied widely across industries. Among benefit types, vision coverage, STD insurance, and LTD insurance were the least commonly offered. We believe this finding is important because workers might have to rely on STD or LTD benefits to retain employment while addressing a temporary health-related work limitation. If such benefits are not widely available, workers may be more likely to apply for DI benefits. This may account for higher-than-average DI application rates among workers in certain industries. The potential relationship between the availability of STD and LTD insurance and DI application is worth further exploration.

This note provides preliminary steps toward understanding the potential relationship between workers' access to health and leave benefits and their likelihood of DI application. Between O'Leary, Walker, and Roessel (2015) and this study, the statistics reveal differences by industry in both the access to employer-provided benefits and the prevalence of DI beneficiary status. Whether those two differences are coincidental or related is a subject for future studies.

Notes

1 Under certain circumstances, family members of workers may also be eligible for DI benefits.

2 The Family and Medical Leave Act (FMLA) of 1993 requires employers to offer leave to qualifying employees, but such leave can be unpaid. The law permits an employee to elect, or the employer to require the employee, to use accrued paid leave (such as vacation or sick leave) for some or all of the FMLA-covered period.

3 The NAICS is the standard system used by the federal statistical agencies of the United States, Canada, and Mexico. For a full description, see Census Bureau (2018).

4 Not all public-sector employees are covered under Social Security. For example, federal employees hired before 1984 are covered under the Civil Service Retirement System and not Social Security (unless they opted to convert to the Federal Employee Retirement System) and thus may not be eligible for DI. This may partially explain the low proportion of DI beneficiaries who had worked in public administration.

5 The NCS omits the federal government because the Office of Personnel Management (OPM, the federal “employer” agency) uses BLS data to determine federal compensation costs (including benefits). Thus, a BLS survey of OPM's costs would essentially collect BLS' own data.

6 For more information on the NCS, see https://www.bls.gov/ncs/home.htm.

7 “Industry sector and subsector” is NAICS nomenclature; this use of “sector” should not be confused with the broader “public sector” and “private sector” contexts.

8 More than 98 percent of workers in the transportation, communications, and utilities industry subsector of NAICS code 49 had access to medical coverage, as did 94.5 percent of workers in the manufacturing industry subsector of NAICS code 33.

9 Almost 57 percent of workers in the retail trade industry subsector of NAICS code 44 had access to medical coverage.

10 More than 97 percent of workers in the transportation, communications, and utilities industry subsector of NAICS code 49 had access to prescription drug coverage, as did 93.3 percent of workers in the manufacturing industry subsector of NAICS code 33.

11 About 55 percent of workers in the transportation, communications, and utilities industry subsector of NAICS code 49 had access to vision coverage.

12 Only 9.9 percent of workers in the retail trade industry subsector of NAICS code 45 had access to vision coverage.

13 Paid leave of any kind was available to only 76.0 percent of workers in the retail trade industry subsector of NAICS code 45.

14 Sick leave also was offered to 96.4 percent of workers in the transportation, communications, and utilities industry subsector of NAICS code 49.

15 About 76 percent of workers in the transportation, communications, and utilities industry subsector of NAICS code 49 had access to paid personal leave.

References

[BLS] Bureau of Labor Statistics. 2013. “Employment Projections: 2012–2022.https://www.bls.gov/news.release/archives/ecopro_12192013.htm.

Census Bureau. 2018. “North American Industry Classification System.” https://www.census.gov/eos/www/naics/index.html.

Dement, John M., Carol Epling, Julie Joyner, and Kyle Cavanaugh. 2015. “Impacts of Workplace Health Promotion and Wellness Programs on Health Care Utilization and Costs: Results From an Academic Workplace.” Journal of Occupational and Environmental Medicine 57(11): 1159–1169.

Du Bois, Marc, and Peter Donceel. 2008. “A Screening Questionnaire to Predict No Return to Work Within 3 Months for Low Back Pain Claimants.” European Spine Journal 17(3): 380–385.

Employee Benefit Research Institute. 2011. EBRI Databook on Employee Benefits. Washington, DC: EBRI.

Livermore, Gina, David Stapleton, and Meghan O'Toole. 2011. “Health Care Costs Are a Key Driver of Growth in Federal and State Assistance to Working-Age People with Disabilities.” Health Affairs 30(9): 1664–1672.

Livermore, Gina, David Wittenburg, and David Neumark. 2014. “Finding Alternatives to Disability Benefit Receipt.” IZA Journal of Labor Policy 3(1): 14.

O'Leary, Paul, Elisa Walker, and Emily Roessel. 2015. “Social Security Disability Insurance at Age 60: Does It Still Reflect Congress' Original Intent?” Issue Paper No. 2015-01. Washington, DC: SSA.

Pierce, Brooks. 1999. “Using the National Compensation Survey to Predict Wage Rates.” Compensation and Working Conditions 4(4): 8–16.

Salkever, David S., Howard Goldman, Mohankumar Purushothaman, and Judith Shinogle. 2000. “Disability Management, Employee Health and Fringe Benefits, and Long-Term-Disability Claims for Mental Disorders: An Empirical Exploration.” The Milbank Quarterly 78(1): 79–114.

Schimmel Hyde, Jody, and Gina A. Livermore. 2016. “Gaps in Timely Access to Care Among Workers by Disability Status: Will the Patient Protection and Affordable Care Act Reforms Change the Landscape?” Journal of Disability Policy Studies 26(4): 221–231.

[SSA] Social Security Administration. 2018. Annual Statistical Report on the Social Security Disability Insurance Program, 2017. SSA Publication No. 13-11826. Washington, DC: SSA. https://www.ssa.gov/policy/docs/statcomps/di_asr/2017/index.html.

Stapleton, David C., Bonnie L. O'Day, Gina A. Livermore, and Andrew J. Imparato. 2006. “Dismantling the Poverty Trap: Disability Policy for the Twenty-First Century.” The Milbank Quarterly 84(4): 701–732.