Health Insurance Data Briefs #3:  
Workers Receiving Employer-Provided
Health Insurance



By Heather Boushey and Joseph Wright[1]

April 13, 2004


This is the third of a five-part series by the Center for Economic and Policy Research (CEPR) on access to health insurance in the United States. Other Data Briefs in this series examine access to health insurance generally, changes in the share of Americans receiving employer-provided health insurance as a dependent on another family member’s plan, and interactions between the private and public health insurance systems. The first Brief details policy options available to expand coverage to include the nearly 70 million Americans who did not have health insurance during all of 2002.

The data used in this series come from CEPR’s analysis of the Survey of Income and Program Participation. CEPR creates user-friendly Data Sets from this survey and makes the data and programs available to the public via our website (www.cepr.net).

David Maduram provided valuable research assistance on this project.

This project was funded by a generous grant from the Rockefeller Foundation.


Executive Summary

            Over the economic expansion of the late 1990s, a greater share of individuals received health insurance coverage from their employer. However, over the most recent contraction, that share fell back to where it was in 1992, with two adults in five (38.5 percent) receiving coverage from their own employer.

The health insurance system in the United States is individualized, and private coverage is generally tied to employment. Most adults receive health insurance in their own name and have traditionally received health insurance from their employer. Among those with health insurance, the share of individuals who have insurance in their own name – that is, people who receive health insurance from their own employer, independently purchase their own plan, or receive government health insurance coverage under their own name, rather than a spouse’s, rose over the decade from 1992 to 2002. The increase is relatively small—a 2.7 percentage point increase up to 62.5 percent, but it does indicate that the trend is toward more individuals receiving health insurance on their own, rather than through a spouse or other family member. The largest increases in own coverage were for women and Latinos.


            Workers are not all equally likely to receive health insurance coverage from their employer. Low-wage workers and workers in small firms are far less likely than high-wage workers and those in large firms to have employer-provided health insurance, and workers in small firms are less likely than those in large firms. Within the group of low-wage workers, Latinos are the least likely to have health insurance coverage. In 2002, less than one-quarter (22.0 percent) of low-wage Latino workers had health insurance coverage from their own employer, compared to 42.5 percent of low-wage white workers.

Workers receiving employer-provided health insurance from their employer

The health insurance system in the United States is individualized, and private coverage is generally tied to employment. Most adults receive health insurance in their own name and have traditionally received health insurance from their employer. For most of the latter half of the 20th  century, employers extended coverage to their employees’ dependents as well. However, because most women now work outside the home, there has been increased movement toward individuals accessing coverage under their own name (Employer-provided health insurance as a dependent on another worker’s plan within the family is covered in Health Insurance Data Brief #4). How an individual accesses health insurance coverage has become increasingly important as employers are now more likely to either not offer dependent coverage for family members of their employees or to charge much more for coverage of dependents. If individuals cannot get health insurance coverage on their own or through a family member, it will create serious hardships for many families.

            This report uses data from the Survey of Income and Program Participation to examine which workers receive employer-provided health insurance and which do not. Although most workers have health insurance in their own name, less than two adults in five adults have employer-provided health insurance fully paid by their own employer. Women are less likely than men to have employer-provided health insurance in their own name, and Latinos are less likely than other racial/ethnic groups. Employer-provided health insurance is a rarity for young adults (ages 18 to 24).

            Low-wage workers are far less likely than high-wage workers to have employer-provided health insurance in their own name and workers in small firms are less likely than those in large firms to have such coverage. Over the most recent recession, firms have pushed the cost of providing health insurance on to their employees, and as a result the share of workers whose premiums were fully paid for by their employer fell. Further, Latinos – both high-wage and low-wage earners – have lost employer-provided health insurance, while coverage rose for other low-wage workers and fell for high-wage workers.

Fewer women than men have health insurance coverage in their own name

            The trend between 1992 and 2002 has been for adults with health insurance to increasingly receive it in their own name, although the changes have been small overall. The share of adults with health insurance in someone else’s name fell by 0.4 percentage points between 1992 and 2002, from 32.6 percent to 32.2 percent. Nearly two-thirds of covered adults have health insurance in their own name; that is, they receive coverage from their own employer or through the government on their own plan (Table 1).

 

Table 1. Source of health insurance coverage

Source of health insurance coverage


Own name

Other's name

Own & other's name
 
1992
1999
2002
 
1992
1999
2002
 
1992
1999
2002
All adults (18 to 64)
59.8%
61.8%*
62.5%*

32.6%
33.0%*
32.2%*

7.6%
5.2%*
5.3%
Women
46.5
49.0*
50.7*

44.9
45.2*
43.4*

8.5
5.8*
5.9
Men
73.6
75.0*
74.8*

19.8
20.4*
20.4*

6.6
4.7*
4.8
White
59.3
60.8*
61.3

33.3
33.8
33.3

7.4
5.3*
5.4
African-American
65.8
68.8*
69.6

25.0
26.3
25.1

9.2
5.0*
5.3*
Latino
59.2
63.8*
66.3*

32.8
31.5*
29.2*

8.0
4.8*
4.5*
Other
57.5
59.5*
60.8*

35.3
35.5
32.9*

7.3
5.0*
6.3*
Married
52.7
53.5*
54.0*

37.4
39.2*
38.6

10.6
7.2*
7.4
Previously Married
93.7
94.0*
93.4*

5.4
5.5*
5.6*

1.0
0.5*
1.0*
Never Married
63.8
66.5*
68.2*

33.7
31.2*
29.7*

2.5
2.3*
2.1
Cohabitating

80.6*
85.3*


17.0*
11.7*


2.4*
3.0
Age Group
18 - 24
32.3
34.2*
35.5*

63.7
61.9*
61.2*

4.0
3.9
3.3*
25 - 34
66.1
69.2*
68.8

26.5
26.5*
26.8*

7.4
4.2*
4.5
35 - 44
62.4
63.0*
64.2*

28.9
31.4*
30.8*

8.7
5.6*
5.0*
45 - 54
61.1
64.2*
65.0*

29.8
29.3*
28.3*

9.1
6.6*
6.7*
55 - 64
64.9
66.1*
65.5*

28.2
29.0*
28.5*

6.9
4.9*
6.1*
Source: CEPR analysis of Survey of Income and Program Participation, 92, 96, and 01 panels.
Note: Universe is all adults (ages 18 - 64) with health insurance. To increase sample size, we pooled estimates for Latinos across two years (92/93, 98/99, 01/02).
* Indicates that change from past year is significant at the 5 percent level.


            The decline in accessing health insurance coverage through someone else’s plan was more pronounced for women. Between 1992 and 2002, the share of women who had health insurance coverage through another person’s plan fell by 1.5 percentage points. Nonetheless, women are still less likely than men to have health insurance in their own name. In 2002, 50.7 percent of women had health insurance in their own name, compared to 74.8 percent of men. Similarly, men are less than half as likely as women to have health insurance in someone else’s name. In 2002, 20.4 percent of men had health insurance in someone else’s name, compared to 43.4 percent of women.

Similarly, the share of young adults who had health insurance coverage through someone else’s plan fell from 63.7 percent in 1992, down to 61.2 percent in 2002, or a 2.5 percentage point decline. This decline was greater than for other age groups, who saw only slight changes in how they access health insurance coverage. Even so, young adults (ages 18 to 24) remain less likely to have health insurance in their own name and the majority gets their health insurance from someone else. In 2002, 61.2 percent of young adults had their health insurance in someone else’s name, compared to 30.8 percent of adults ages 35 to 44.

Less than two adults in five receive employer-provided health insurance in their own name

In 2002, less than two adults in five (38.5 percent) had health insurance all year from their employer (Table 2).[2] Over 2002, among adults, nearly half (48.1) had employer-provided health insurance in their own name in any month. Some of the lack of health insurance coverage may be due to waiting periods when workers change jobs; however, the large share of workers without health insurance from their employer cannot be due entirely to job changes.

There is a significant gender gap in employer-provided health insurance. In 2002, 31.9 percent of women had employer-provided health insurance all year, compared to 45.3 percent of men. The gender gap in employer-provided coverage remains even when we control for employment status and other demographic characteristics (Figure 1). At all ages, men are more likely to receive health insurance from their employer than are women, even controlling for employment status and wages.[3]

            There is also a gap across racial/ethnic groups: Latinos are least likely to have employer-provided health insurance, with about one quarter (18.9 percent) receiving it, compared to 42.1 of whites and 33.4 percent of African Americans. Again, this holds true even once we control for other characteristics of Latinos, including employment status and wages (Figure 2).

Table 2: Employer-provided health insurance coverage

Share of individuals covered by their employer's health insurance plan


At all during month

All year
 
1992
1999
2002
 
1992
1999
2002
All adults (18 - 64)
44.5%
47.3%*
48.1%

35.5%
39.6%*
38.5%*
Women
36.3
38.8*
41.0*

28.1
31.9*
31.9*
Men
53.1
56.1*
55.7*

43.1
47.6*
45.3*
White
46.8
50.0*
51.4*

38.0
42.6*
42.1*
African-American
40.2
45.1*
45.5

30.7
36.8*
34.4*
Latino
33.7
35.1*
34.1*

18.8
22.7*
18.9*
Other
38.0
38.9
44.2*

28.5
30.2*
33.4*
Age group
18 - 24
18.7
21.3*
26.3*

9.7
10.7*
13.2*
25 - 34
45.9
50.0*
49.1*

35.1
40.6*
37.9*
35 - 44
51.4
51.6
51.5*

43.1
44.6*
42.5*
45 - 54
51.0
54.3*
54.3*

42.4
48.3*
46.1*
55 - 64
46.4
48.3*
49.9*

38.6
42.0*
41.7*
Source: CEPR analysis of Survey of Income and Program Participation, 92, 96, and 02 panels.
Note: Universe is all adults. To increase sample size, we pooled estimates for Latinos across two years (92/93, 98/99, 01/02).
* Indicates that change from past year is significant at the 5 percent level



Figure 1: Health insurance coverage from own employer, by gender (2002)

Source: CEPR Analysis of Survey Income and Program Participation (2001 Panel).
Notes: values plotted are median expected values. See Technical Documentation: Health
Insurance Data Briefs.
Figure 2: Health insurance coverage from employer, by race/ethnicity

Source: CEPR analysis of Survey of Income and Program Participation (2001 Panel).
Notes: values plotted are median expected values. See Technical Documentation.
Table 3: Premiums paid by employers

Among those working who have employer-provided health insurance coverage


Premium fully paid

Premium partially paid
 
1992
1999
2002
 
1992
1999
2002
All adults (18 - 64)
34.1%
32.1%*
30.0%*

61.4%
64.8%*
66.2%*
Women
32.8
29.9*
27.9*

62.2
66.7*
68.0*
Men
35.1
33.7*
31.5*

60.9
63.4*
64.9*
White
35.5
32.8*
30.5*

60.1
64.0*
65.7*
African-American
25.9
25.6
24.9*

68.9
71.2
71.1
Latino
29.1
33.4*
31.1*

66.5
63.5*
65.1*
Other
34.7
34.9
31.5*

63.2
63.1
65.7*
Age group
18 - 24
33.1
29.5*
28.5*

62.4
66.9*
68.7*
25 - 34
33.0
31.2*
30.0*

63.4
65.8*
66.7*
35 - 44
33.7
30.0*
28.4*

62.6
67.3*
68.2*
45 - 54
35.8
33.9*
30.1*

59.9
63.4*
66.2*
55 - 64
35.4
35.0
32.4*

57.1
60.0*
62.0*
Source: CEPR analysis of Survey of Income and Program Participation, 92, 96, and 02 panels.
Note: Universe is all adults. To increase sample size, we pooled estimates for Latinos across two years (92/93, 98/99, 01/02).
* Indicates that change from past year is significant at the 5 percent level

Young adults are least likely to have employer-provided health insurance: in 2002, among young adults, only about one in ten (13.2 percent) had coverage from their employer, compared to 42.5 percent of adults ages 35 to 44.

Most workers pay for this coverage
             In 2002, less than one-third of workers (30.0 percent) with employer-provided health insurance from their employer had their premium fully paid for by that employer (Table 3). Another two-thirds (66.2 percent) of those with employer-provided health insurance had their premium partially paid for by their employer, and the remaining 3.8 percent fully paid for their premiums. Women insured by an employer are less likely than men to have their premiums paid for: 27.9 percent of women’s premiums were fully paid for, compared to 31.5 percent of men’s. African American workers with employer-provided health insurance are less likely than whites or Latinos to have their premiums fully paid for: 24.9 percent of African Americans’ premiums were fully paid for, compared to 30.5 percent of whites’ and 31.1 percent of Latinos’.
Employer-provided health insurance grew over the economic expansion and declined over the economic contraction

The share of Americans who received health insurance all year from their employer grew slightly over the 1990s expansion and fell slightly over the early 2000s recession. Over the economic expansion, from 1992 to 1999, the share of those working full-year and having employer-provided health insurance all year increased by 4.1 percentage points, up to 39.6 percent. However, from close to the peak of the business cycle 1999 to 2002, the share fell back down to 38.5 percent. The declines over the economic contraction were larger among men than among women and coverage fell especially for Latinos. Between 1999 and 2002, the share of Latinos with employer-provided health insurance all year fell by 3.3 percentage points, compared to no change for whites and an increase of 2.4 percentage points for African Americans.

            Even over the economic boom, however, fewer employers fully paid their employees’ health insurance premiums (Figure 3). Between 1992 and 1999, the share of workers whose premiums were fully paid for by their employer decreased by 2.0 percentage points, down to 32.1 percent. This share fell further over the economic contraction – down to 30.0 percent in 2002. However, while whites and African



Table 4. Employer-provided insurance by wage level and firm size

Wage level

Firm size
Low-wage

High-wage

Less than 100 employees

100 employees or more
 
1992
1999
2002
 
1992
1999
2002
 
1999
2002
 
1999
2002
Working adults (18 - 64)
31.2%
34.9%*
38.9%*

84.2%
82.2%
82.1%

52.3%
55.7%*

71.1%
72.5%*
Women
25.7
29.4*
34.5*

76.4
74.7
75.3

44.6
48.3*

63.9
66.7*
Men
37.2
42.7*
42.5*

89.5
87.0*
86.2*

60.0
62.7*

77.8
77.9
White
32.5
37.1*
42.5*

84.9
82.9*
82.6*

54.2
59.7*

72.5
74.5*
African-American
31.3
38.4*
41.7*

83.1
81.4*
81.1

51.7
49.0

69.9
69.6
Latino
22.7
24.0*
22.0*

77.4
77.4
72.8*

43.5
36.4*

63.3
61.6*
Other
32.0
29.3
37.8*

81.8
77.4*
81.8*

38.1
48.5*

67.3
72.6*
Source: CEPR analysis of Survey of Income and Program Participation, 92, 96, and 02 panels.
Note: Universe is working adults (ages 18 to 64) with health insurance. To increase sample size, we pooled estimates for Latinos across two years (92/93, 98/99, 01/02).
* Indicates that change from past year is significant at the 5% level

            Americans saw their share of fully paid premiums decline over both the expansion and contraction, Latinos actually saw their shares rise slightly, up to 31.1 percent during the contraction. This fell back down by 2002.  High-wage workers are much more likely to have employer-provided health insurance than low-wage workers.

Low-wage workers – those earning at the 30th percentile or below – are far less likely to have employer-provided health insurance than are high-wage workers – those earning above the 70th percentile (Table 4).

            Among low-wage workers in 2002, about two in five (38.9 percent) had employer-provided health insurance, compared to four in five (82.1 percent) high-wage workers. There is a gender gap in employer-provided health insurance among both low-wage and high-wage workers. In 2002, low-wage men were 23 percent more likely than low-wage women to have employer-provided health insurance and high-wage men were about 14 percent more likely than high-wage women (Figures 4 and 5).

 




            There is also a large gap in coverage between Latinos and other racial/ethnic groups. In 2002, 22.0 percent of low-wage Latino workers received employer-provided health insurance, compared to 42.5 percent of low-wage white workers and 41.7 percent of low-wage African Americans workers. Looking across wage rates, once we control for other demographic factors, the gap between Latinos and other racial/ethnic groups remains (Figure 4). At each wage level, Latinos are least likely to have employer-provided health insurance.

            Firm size also plays a role. Workers in large firms (100 employees or more) are more likely than those in small firms (less than 100 employees) to have employer-provided health insurance. In 2002, 55.7 percent of workers in small firms had employer-provided health insurance, compared to 72.5 percent of those in large firms. The gender gap and racial/ethnic gaps are also in evidence across firm size, with women and Latinos being less likely to receive employer-provided health insurance than men and other racial/ethnic groups, respectively.

Seemingly odd, given the declining labor market conditions between 1999 and 2002, the share of those in low-wage jobs with employer-provided health insurance increased while the share of those in high-wage jobs with employer-provided health insurance stayed virtually the same. However, if firms are increasingly cutting back on family coverage, the lower-paid worker in a family will have to switch to coverage from their own employer, rather than remain a dependent on the higher-paid spouse’s better plan. Thus, the increase in low-wage workers with employer-provided health insurance may be due to a shift away from high-wage workers covering other working adults in the family. (See Health insurance Data Brief #4 for more information on dependent coverage).

Conclusion

            Most workers do not get health insurance from their employer: less than two in five Americans have coverage in their own name from an employer. Tying health insurance to employment has meant that high-wage workers receive health insurance as a benefit while low-wage workers do not. The likelihood of a low-wage worker - earning below the 30th percentile of earners – having employer-provided health insurance is less than half the likelihood of high-wage earners – those earning at the 70th percentile of earners or above. In 2002, most high-wage workers (82.1 percent) received employer-provided health insurance, compared to slightly more than two in five (38.9 percent) low-wage workers.

Women are less likely than men to have employer-provided health insurance from their own employer, and Latinos are less likely to have coverage compared to other racial/ethnic groups. However, it is young adults (ages 18 to 25) who are the least likely to have employer-provided health insurance in their own name, with about one in ten having such coverage.

Young adults are not likely to receive health insurance from their employer. However, since employer-provided health insurance is the primary mechanism for accessing the private health insurance market, this puts this group of Americans at great risk for being uninsured. Findings in Health Insurance Data Brief #2 showed that this group is the least likely among adults to have access to health insurance, even though these are prime child-bearing years.


[1] Heather Boushey is an economist and Joseph Wright a research assistant at the Center for Economic and Policy Research.
[2] An individual is assumed to have employer-provided health insurance from their own employer if their health insurance is employment-related in their own name.
[3] See Health Insurance: Health Insurance Data Briefs for a complete description of our regression analysis.