THE HENRY J. KAISER FAMILY oma FOUNDATION medicaid - uninsured October 2006 Changes in Employees' Health Insurance Coverage, 2001-2005 Lisa Clemans-Cope Bowen Garrett Catherine Hoffman The majority of Americans under age 65 receive their health insurance coverage through their own employer or the employer of a family member, but the long-standing link between work and employer-sponsored insurance is weakening. Because employer-sponsored insurance is voluntary on the part of businesses and employees, not all firms offer health benefits, not all workers are eligible for coverage, and not all employees choose to participate or can afford their share of the health premium. Employer-sponsored insurance (ES!) coverage rates have been falling. In 2000, 66% of non- elderly Americans were insured through the workplace, but by 2004 only 61% were covered by ESI. Both children and adults experienced steady declines in job-based coverage through this period; however, all of the growth in the number of uninsured was among adults. Were it not for Medicaid and S-CHIP, the number of children without health insurance would have increased also. Instead, as unemployment spells lowered families' incomes, public insurance filled in the gap for children up until 2004, but not adults.1. By 2005 however, as employer-sponsored insurance continued to erode, the number of low-income uninsured children began to grow again, along with the number of uninsured adults.2 This issue brief focuses primarily on how ESI coverage has changed among employees (i.e., workers who are not self-employed). Based on a larger report entitled, Changes in Employer- Sponsored Health Insurance Sponsorship, Eligibility, and Participation: 2001 to 2005, it begins with a brief description of major forces driving employer-sponsored insurance: changes in the workforce and the rising costs of health insurance over the four year period. 1330 G STREET NW, WASHINGTON, DC 20005 PHONE: 202-347-5270, FAX: 202-347-5274 WEBSITE: WWW.KFF.ORG Trends in Health Coverage of Working-Age Adults Health insurance coverage changed substantially between 2000 and 2004 for working-age adults. The share covered by employer-sponsored insurance decreased by five percentage points, from 68% to 63%. Increases in Medicaid, other public coverage, and private nongroup coverage were not enough to offset this loss and the share of working-age adults without health insurance grew from 18% to 21% (Figure 1). Figure 1 Health Insurance Coverage Changes Among Working-Age Adults (age 19-64), 2000-2004 Percent 2000 @ 2004 100% 75%) 67.9% 63.1% 50% 20.6% 25% , 17.9% eee 8.6% 10.5% ° of 5.8% 5.6% o% | aoa Employer- Medicaid/ Private, Uninsured* Sponsored* Other Public* Non-Group * Changes are significantly different (p<.05). SOURCE: Kaiser Commission on Medicaid and the Uninsured/Urban Institute, Health Insurance Coverage in America, 2004 Data Update. Nov, 2005. The number of uninsured nonelderly adults grew by 6.3 million between 2000 and 2004 and 3.8 million (60%) were working adults. The majority of the growth in uninsured workers was among poor and near-poor workers (Figure 2).3 Figure 2 Growth in Uninsured Workers, by Family Income as a Percent of the Federal Poverty Level, 2000-2004 High (400%+) 16% Poor (<100% FPL) 28% Mid-Level (200-399% FPL) 31% Near-Poor (100-199% FPL) 24% Total Growth in Uninsured Workers = 3.8 Million Data do not total 100% due to rounding. SOURCE: Kaiser Commission on Medicaid and the Uninsured/Urban Institute. Unpublished data. 2005. THE KAISER COMMISSION ON Medicaid and the Uninsured By 2005, there were nearly 19 million uninsured workers who were employees of a business and another nearly 4 million who were self-employed. While there are many fewer self-employed workers, they have always been at greater risk of being uninsured with no direct access to group health benefits for themselves. Employment Dynamics Affecting Health Coverage, 2001-2005 Health insurance coverage is directly affected by shifts in employment patterns. Although the economic recession that started in early 2001 was short in duration, the negative impact on employment was more persistent. A substantial part of the reason so many working-age adults lost health coverage between 2001 and 2005 can be explained by underlying workforce changes. « Most of the decline in employment was due to 5.1 million adults leaving the labor force altogether (e.g., early retirees, disabled, caregivers, students, and discouraged workers) while an additional 1.4 million became unemployed (e.g., laid-off workers and those available for and looking for work). « Family incomes shifted downward. By 2005, more workers came from families with incomes below the federal poverty level. The number of workers grew by 2.2 million over the four years, but 1.8 million of them were in poor families. = By 2005 more of the workforce was self-employed, working part-time, or held temporary or contract positions. Compared to 2001, workers in 2005 were more likely to come from families supported only by part-time workers and slightly less likely to have two full-time workers in them. = By 2005 more employees worked in the smallest of businesses (with fewer than 10 employees) and fewer worked for large firms (with 100 or more employees). The share of employees working in large firms decreased from 66% in 2001 to 64% by 2005. « Occupations shifted with fewer working in production jobs, service work, and administrative support positions. Professional occupations and jobs related to professional work increased among women. Construction and mining jobs also increased somewhat among men-but with more self-employed and contingent workers. THE KAISER COMMISSION ON Medicaid and the Uninsured THE Trends in the Cost of Health Insurance Employer-sponsored insurance is sensitive to changes in health insurance premiums and marked increases in premiums beginning in 2000 adversely affected whether businesses offered health benefits, who was eligible, and whether employees participated in health benefit options. Between 2001 and 2005 employer-sponsored health insurance premiums grew by no less than nine percent each year, ranging between 9.2% and 13.9% annually for premiums for a family of four. Employer-sponsored family health insurance premiums averaged $10,880 in 2005. The share of all businesses offering health benefits declined from 69% in 2000 to 60% by 2005, driven largely by decreases among small to mid-size firms (with 3 to 199 employees). Employees' earnings grew slowly between 2001 and 2005, ranging between 2.2% and 4.0% each year-mirroring the range of overall inflation rates of 1.6% to 3.5% annually-making health insurance even less affordable relative to their incomes. The average share of a family premium employees were required to pay themselves stayed fairly flat between 2001 and 2005, around 27%. However, given the large increases in premiums, that share amounted to an increase of nearly $1,000 over this period, from $1,788 a year in 2001 to $2,712 by 2005.4 KAISER COMMISSION ON Medicaid and the Uninsured Trends in Health Coverage and Access to ESI Among Employees The share of employees who were covered by ES! decreased by almost four percentage points between 2001 and 2005 (81.2% to 77.4%). In turn, the uninsured rate among employees increased by nearly three percentage points to 16.6%-with almost 19 million uninsured employees by 2005 (Figure 3). Figure 3 Health Insurance Coverage Changes Among Employees, 2001 - 2005 Percent of Employees 2001 2005 100% 81.2% 77.4% 13.7% 16.6% 3.69 3.1% 98% 4.94% 2.4% 0% a -- Employer- Medicaid/ Private, Uninsured Sponsored Other Public Non-Group All changes are significantly different (p<.05). Data may not total 100% due to rounding. Source: Urban Institute analysis of the February 2001 and 2005 Contingent Work Supplement of the Current Population Survey (CPS) and the March 2001 and 2005 Annual Social and Economic (ASEC) Supplement of the CPS. By 2005, nearly 15% of employees had no offer of ESI within their family, an increase of 2.5 percentage points over 2001. For those with an offer of ESI, participation in employers' health benefits dropped off slightly, with about 7.7% declining health coverage in 2005 compared to 6.5% in 2001. The main reason for the decline in ESI among employees between 2001 and 2005 was because fewer employees worked for employers who sponsored health benefits. Almost half of the decline in ESI rates among employees (1.8 points of the 3.8 percentage point drop in ESI coverage rates) was due to loss of sponsorship. Changes in employees' eligibility for health benefits and their access to ESI coverage as a dependent of another employee accounted for smaller shares of the decrease in employer-sponsored coverage. About a quarter of the drop in ESI rate was due to employees not participating, i.e., not taking-up health benefits offered to them (Figure 4). THE KAISER COMMISSION ON Medicaid and the Uninsured Figure 4 Reasons for Decline in ESI among Employees, 2001- 2005 Employee Take-up Decline 27% Employer Sponsorship Loss of ESI Oeeon Dependent fo Coverage 11% Employee Eligibility Decline - 14% Source: Urban Institute analysis of the February 2001 and 2005 Contingent Work Supplement of the Current Population Survey (CPS) and the March 2001 and 2005 Annual Social and Economic (ASEC) Supplement of the CPS. By 2005, the majority (56%) of uninsured employees had no access to ESI within their family (e.g., from either their own or their spouse's employer). Another 14% worked in businesses where health benefits are offered, but they were not eligible and in addition they did not have access to ESI through a family member. Only a third of uninsured employees had access to ESI, but chose not to participate or "take-up" the health insurance benefit being offered by an employer (Figure 5). Figure 5 Access to ESI among Uninsured Employees, 2005 Declined Own or Family Member's Health Benefits 30% No ESI Sponsorship from Own or Family Member's Employer 56% Not Eligible, And No ESI Available Through Family Member's Employer 14% 18.8 Million Uninsured Employees Source: Urban Institute analysis of the February 2005 Contingent Work Supplement of the Current Population Survey (CPS) and the March 2005 Annual Social and Economic (ASEC) Supplement of the CPS. THE KAISER COMMISSION ON Medicaid and the Uninsured Growing Differences in Employer-Sponsored Insurance Income Differences. Decreases in ESI coverage-and increases in the share who are uninsured-were greatest among low-income workers, those who were already the most likely to be uninsured. For example, the share of poor employees who had ESI dropped from 37% in 2001 to 30% by 2005 and among the near-poor dropped from 59% to 52%, while among those with the highest incomes (400% of the poverty level and higher) ESI rates stayed at over 92% (Figure 6). Figure 6 Changes in Employees' ESI Coverage and Uninsured Rates, by Family Income Levels, 2001-2005 Percent with ESI 100% 92.9% 92.2% 12001 M2005 7.44, 80% - 58.7% 60% " 51.7% 40% | 36.8% 39.4% 0% <100%FPL 100-199%FPL 200-399% 400% 46.7% BERG 33.9% 13.4% Ezz 3.6% Bau Uninsured Rates * Statistically significant changes for both ESI and uninsured rates for these groups (p<.05). Source: Urban Institute analysis of the February 2001 and 2005 Contingent Work Supplement of the Current Population Survey (CPS) and the March 2001 and 2005 Annual Social and Economic (ASEC) Supplement of the CPS. Declines in employer sponsorship over the four year period were deepest among poor and near- poor employees (Figure 7). Lower income employees also are less likely to be eligible or participate in offered benefits and these differences grew over the four years. Figure 7 Decreases in ESI Due to Changes in Employer Sponsorship and Employee Participation, by Income Levels, 2001-2005 MB <100% [ 100-199% [] 200-399% [_] 400%+ Employer Sponsorship Employee Take-Up 0 (Percentage Point Changes) 0 (Percentage Point Changes) -3 "27 .29 Source: Urban Institute analysis of the February 2001 and 2005 Contingent Work Supplement of the Current Population Survey (CPS) and the March 2001 and 2005 Annual Social and Economic (ASEC) Supplement of the CPS. THE KAISER COMMISSION ON Medicaid and the Uninsured By 2005, 55% of employees from poor families had no access to ES! from either their own employer or a family member's employer and only 30% were covered by an employer. Near-poor employees were somewhat better off; still over a third had no access to ESI. In contrast 92% of the highest income employees had health coverage through an employer (Figure 8). Figure 8 Employee Access to ESI within the Family by Family Income, 2005 O Covered by Own or Family Member's Employer B Declined offer from Own or Family Member's Employer Not offered through Own or Family Member's Employer 0, 79% - [<8% 4% 13% S 4% <100% 100-199% _200-399% 400% + Poverty Level Data may not total 100% due to rounding. Source: Urban Institute analysis of the February 2005 Contingent Work Supplement of the Current Population Survey (CPS) and the March 2005 Annual Social and Economic (ASEC) Supplement of the CPS. Differences Across Firm Sizes. Decreases in employer coverage were deepest among employees working in smaller firms (with fewer than 25 employees)-those who are the most likely to be uninsured. Employees working in firms with 10 to 24 workers experienced a nearly five percentage point decrease in their ESI rate, with an increase in their uninsured rate from 26% to 30%. Even employees in large businesses (with 100 or more employees) experienced a substantial decline in ESI, raising their uninsured level to over 10% (Figure 9). Figure 9 Changes in Employees' ESI Coverage and Uninsured Rates, by Firm Size, 2001-2005 Percent with ESI 100% 2001 m 2005 88.2% 95.1% 77.2% ° 80% 74.9% . 68.0% 65 1%, 60% | 768% 59 6% 40% 20% 0% <10 10-24 "25-99 100+ 33.1% 25.6% 17.0% za 3.1% PLE Uninsured Rates * Statistically significant changes for both ES| and uninsured rates for these groups (p<.05). Source: Urban Institute analysis of the February 2001 and 2005 Contingent Work Supplement of the Current Population Survey {CPS) and the March 2001 and 2005 Annual Social and Economic (ASEC) Supplement of the CPS. THE KAISER COMMISSION ON Medicaid and the Uninsured Declines in employer sponsorship over the four year period were deepest for employees of small firms. Declines in sponsorship occurred both in firms with fewer than 10 employees and in firms with 10 to 24 employees. However, increases in take-up and eligibility rates offset what would have been a larger decline in total ES! coverage among employees of the smallest firms (<10 employees; Figure 10). Figure 10 Decreases in ESI Due to Changes in Employer Sponsorship and Employee Participation, by Firm Size, 2001-2005 Mm <10 HS 10-24 [5 25-99 [-) 100+ Employer Sponsorship Employee Take-Up 0 (Percentage Point Changes) 1 (Percentage Point Changes) 0.6 -3 4-28 °27 Source: Urban Institute analysis of the February 2001 and 2005 Contingent Work Supplement of the Current Population Survey (CPS) and the March 2001 and 2005 Annual Social and Economic (ASEC) Supplement of the CPS. By 2005, half of employees working in the smallest of firms (<10 employees) had no access to ES| from either their own employer or a family member's employer and only 43% were covered by an employer. The situation was somewhat better for those working in businesses with 10 to 24 employees, where 35% had no access to ESI. In contrast just 8% of the employees in the largest firms had no access to ES! at all (Figure 11). Employee Access to ESI within the Family by Firm Size*, 2005 O Covered by Own or Family Member's Employer @ Declined offer from Own or Family Member's Employer @ Not offered through Own or Family Member's Employer 439 % 56% 71% ° «7% 85% 9) Ui ; 11% . 7% 8% Less than 10 10-24 25-99 100+ Firm Size sLargest employer in the family Source: Urban Institute analysis of the February 2005 Contingent Work Supplement of the Current Population Survey (CPS) and the March 2005 Annual Social and Economic (ASEC) Supplement of the CPS. THE KAISER COMMISSION ON 10 Medicaid and the Uninsured Differences Among Age Groups. Younger employees experienced the greatest decreases in employer coverage, with the greatest increases in their chances of being uninsured. Just over half of employees who are 19 to 24 years old have ES! and 36% are uninsured. In contrast, middle-age employees, who earn more and are more likely to be married with two potential sources of ESI, experienced a substantially smaller decrease in ESI and over 80% have ESI coverage (Figure 12). Figure 12 Changes in Employees' ESI Coverage and Uninsured Rates, by Age Group, 2001-2005 Percent with ESI 2001 2005 100% 9 86.8% 84.79 78.5% 83.6% 99.4% ro 84.7% 80% 71.9% 61.7% 60% - 54.4% 40% 20% 0% 19-24 25-34 35-44 45-64 * * * * 29.7% EEE 16.7% EEX 12.2% EE 8.3% Uninsured Rates * Statistically significant changes for both ESI and uninsured rates for these groups (p<.05). Source: Urban Institute analysis of the February 2001 and 2005 Contingent Work Supplement of the Current Population Survey (CPS) and the March 2001 and 2005 Annual Social and Economic (ASEC) Supplement of the CPS, Declines in employer sponsorship and employee participation over the four year period were deepest for employees under age 35 and explained the majority of their loss in ESI. In contrast, the reasons for ESI loss among employees between 45 and 64 years old, which was significant but much smaller than that of younger employees, were largely due to declining sponsorship (Figure 13). Figure 13 Decreases in ESI Due to Changes in Employer Sponsorship and Employee Participation, by Age Groups, 2001-2005 Mi 19-24 GB 25-34 [>] 35-44 [ 45-64 Employer Sponsorship Employee Take-Up (Percentage Point Changes) or ercentage Point Changes) -3.1 Source: Urban Institute analysis of the February 2001 and 2005 Contingent Work Supplement of the Current Population Survey (CPS) and the March 2001 and 2005 Annual Social and Economic (ASEC) Supplement of the CPS. THE KAISER COMMISSION ON Medicaid and the Uninsured 11 By 2005, one third of employees who were 19 to 24 years old had no access to ESI from either their own employer or a family member's employer; 54% were covered by an employer. Middle- aged employees fared much better with just 10% having no access to ESI at all and 85% being covered by ESI (Figure 14). Employee Access to ESI within the Family by Age Groups, 2005 OO Covered by Own or Family Member's Employer Bi Declined offer from Own or Family Member's Employer @ Not offered through Own or Family Member's Employer 54% ° 72% 80% 85% 14% 5% Eyay) 1% 7% " 18% 13% <- 10% 19-24 25-34 35-44 45-64 Age Groups Source: Urban Institute analysis of the February 2005 Contingent Work Supplement of the Current Population Survey (CPS) and the March 2005 Annual Social and Economic (ASEC) Supplement of the CPS. Differences in health insurance coverage by income, age, and the size of the employer have grown. Gaps in access to ES! have widened and the chances of being uninsured have grown the most among those with low incomes, younger employees, and those working in small businesses. Figure 15 Growth in Uninsured Employees, by Income and Firm Size, 2001 - 2005 Income Firm Size 400% + 6% 100+ Less than 10 29% <100% 37% 200-399% 27% 10-24 100-199% 25-99 17% 30% 71% Growth in uninsured employees = 3.4 million Data may not total 100% due to rounding. Source: Urban Institute analysis of the February 2001 and 2005 Contingent Work Supplement of the Current Population Survey (CPS) and the March 2001 and 2005 Annual Social and Economic (ASEC) Supplement of the CPS. THE KAISER COMMISSION ON 12 Medicaid and the Uninsured THE Figure 16 Growth in Uninsured Employees, by Age and Hours Worked, 2001 - 2005 Age Hours Worked 19-24 45-64 17% Part-Time Employees EullTi 39% ull-Time Employees 59% 35-44 y 25 - 34 11% 11% Other Employees* 2% Growth in uninsured employees = 3.4 million 'Work hours data missing, cannot determine if these employees are full- or part-time. Data does not add to 100% due to rounding. Source: Urban Institute analysis of the February 2001 and 2005 Contingent Work Supplement of the Current Population Survey (CPS) and the March 2001 and 2005 Annual Social and Economic (ASEC) Supplement of the CPS. Between 2001 and 2005 the number of uninsured employees grew by 3.4 million. Two-thirds (67%) of that growth was among low-income employees, with family incomes less than 200% of the poverty level and almost half of the growth (46%) occurred among those working in firms with fewer than 25 employees (Figure 15). Nearly 60% of the growth in uninsured employees was among full-time employees, and a large majority were also employees under age 35 (Figure 16). KAISER COMMISSION ON Medicaid and the Uninsured 13 Conclusion Employer-sponsored insurance declined between 2001 and 2005 among nonelderly adults, employees, and children. The loss can be explained in large part by these underlying trends: lower employment rates, shifts in the labor force and the kinds of jobs available, increases in the number of low-income families, and marked increases in the cost of health insurance. These changes, taken together, significantly reduced access to ESI in just four years. The share of employees with ESI decreased from about 81% in 2001 to 77% by 2005. The main reason for the decline in ESI among employees between 2001 and 2005 was because fewer employees worked for employers who sponsored health benefits. Almost half of the decline in ESI rates among employees was due to loss of sponsorship. Another quarter of the drop in ES! rate was due to employees not participating in health benefits that were offered to them. Changes in employees' eligibility for health benefits and their access to ESI coverage as a dependent accounted for smaller shares of the decrease in employer-sponsored coverage. Gaps in access to ES! between high and low-income families, between younger and older employees, and those working in small vs. large businesses widened. To compensate, workers turned to Medicaid if they were eligible and to private nongroup coverage-neither of which was able to offset the loss of ESI for most workers, leaving them uninsured. Between 2001 and 2005, the number of uninsured employees grew by 3.4 million-two-thirds of whom were from low- income families. The trends in part-time and temporary contract work and industry/occupational shifts are likely to continue. Moreover, the pressure on job-based coverage from rising health insurance premiums and greater employee cost-sharing is likely to increase. If health insurance reforms are to bolster the current system of employer-based coverage, they will need to compensate for the erosion of jobs with health benefits and address the affordability of health insurance for both employers and employees. THE KAISER COMMISSION ON Medicaid and the Uninsured Data and Methods This issue brief excerpts key findings from a forthcoming report entitled, Changes in Employer- Sponsored Health Insurance Sponsorship, Eligibility, and Participation: 2001 to 2005, prepared by Urban Institute researchers Lisa Clemans-Cope and Bowen Garrett for the Kaiser Commission on Medicaid and the Uninsured. The estimates in this issue brief are based on analyses from the larger report using two matched data sets: the February Contingent Workers and Alternate Employment Supplement of the Current Population Survey (CPS) and the March Annual Demographic Survey of the CPS from the Census Bureau. While the March survey is available annually, the most recent February Contingent Worker Supplement surveys were conducted only in 2001 and 2005. The February supplement provides information not available from the March survey, on employer sponsorship of ESI coverage, worker eligibility for the coverage, and whether the worker takes-up an offer of coverage. The March supplement contains demographic information and other information, such as firm size and health status, which is not available from the February survey. Matching workers across the two surveys allows a more detailed analysis of the patterns of ES| coverage across different types of workers and jobs. Detailed specifics of the analyses are available in the full report's appendix. Endnotes " Hoffman, C, A Carbaugh, H Yang Moore, and A Cook. 2005. Health Insurance Coverage in America, 2004 Data Update. Kaiser Commission on Medicaid and the Uninsured Report #7415, November. Available at: http:/;www.kff.org/uninsured/7415.cim. 2 U.S. Census Bureau. 2006. Income, Poverty, and Health Insurance Coverage in the United States: 2005. Report P60-231. August 2006. 3 Hoffman, C et al., 2005. 4 Kaiser Family Foundation and Health Research and Educational Trust. 2005. Employer Health Benefits, 2005 Summary of Findings. Report No. 7316. THE KAISER COMMISSION ON Medicaid and the Uninsured 15 1330 G STREET NW, WASHINGTON, DC 20005 PHoone: (202) 347-5270, FAx: (202) 347-5274 WEBSITE: WWW.KFF.ORG/KCMU Additional copies of this report (#7570) are available on the Kaiser Family Foundation's website at www.kff.org. THE HENRY J. GaN FAMILY FOUNDATION