Tracking Small-Firm Coverage, 1989 1996 An increasing number of small employers say that their workers aren t interested in health insurance. by P au l B. Gin sburg, Jon R. Ga bel, a n d Kell y A. Hun t Am id r ob u s t econo mi c expansion with falling rates of unemployment, the number of uninsured Americans has grown. Recent data from the Current Population Survey (CPS) show that the number of uninsured persons in the United States increased by about one million between 1995 and 1996. Most insured persons obtain their coverage through their employer. Small employers, however, are much less likely than large employers to offer health insurance to their employees. In 1996, 49 percent of firms with fewer than 200 employees offered coverage to their employees, while 99 percent of larger firms did. 1 In this paper we examine trends in small employers coverage and reasons for not offering coverage, for those employers that do not do so. We also examine trends in employee enrollment in plans that are offered by employers. Our major finding is that in the 1990s, small employers have become more likely to offer coverage, but their employees have become less likely to enroll. The net effect is a decline in the rate of employmentbased health coverage for persons working in small firms. The data for this analysis are drawn from a 1989 survey of employers conducted by the Health Insurance Association of America (HIAA) and a 1996 survey conducted by KPMG Peat Marwick. The two surveys used similar questionnaires, sampling design, and weights. Although all sizes of employers were included in these surveys, our analysis concentrates on those firms with fewer than 200 employees. This analysis complements a recent study, published in Health Affairs, by Philip Cooper and Barbara Schone, which used data from 1987 and 1996 household surveys. 2 The advantage of household surveys is that they can collect data on coverage of employees and their families under employment-based insurance and other sources of insurance. They also can describe the characteristics of persons who work for firms that do not offer coverage or who do not enroll when coverage is offered. Employer surveys, on the other hand, retrieve information on firm s reasoning about whether or not to offer coverage and can provide more accurate information on firm size, characteristics of the health plan, eligibility rules, and required contributions from employees. Together, the two survey approaches offer different angles on the same issue, in this case producing consistent results. This paper draws on the results of Cooper and Schone s work on households to enrich the story told by our employer survey data. METHODS This study used the same methods used in our study of trends in the types of health plans 167 Paul Ginsburg is president of the Center for Studying Health System Change in Washington, D.C. Jon Gabel is director of the Center for Survey Research at KPMG Peat Marwick in Arlington, Virginia. Kelly Hunt is a senior consultant at KPMG Peat Marwick in New York City. 1998 ThePeople-to-People Downloaded from Health HealthAffairs.org Foundation, Inc. on January 06, 2019.
168 offered by small firms. 3 The major difference is that in this study we used 1989 HIAA data in place of 1988 data, which more than doubled the sample size for the early year in the period. The response rate for the 1989 survey was 66 percent (the same as for the 1988 survey), but the rate for the 1996 KPMG survey was 50 percent. 4 In cases where a question did not appear on the HIAA survey, we incorporated data from the 1993 KPMG Peat Marwick survey instead. Because the response rate for the smallest firms (one to nine employees) was particularly low in 1996, we conducted a special survey of nonrespondents with only one question: Do you offer health benefits to your employees? Nonrespondents were far less likely to offer coverage, so we adjusted our data to account for nonresponse bias. We found a much greater proportion of firms with one to three workers in the 1989 survey than in the 1996 survey, probably the result of more aggressive screening of self-employed persons in the 1996 survey. 5 Since the smallest firms are much less likely to offer insurance, we constructed a common size distribution of firms in the one-to-nine size category. 6 For different segments of the analysis, we used different weights to extrapolate the results from the sample to the national population. 7 MARKET CONTEXT Perhaps the most important factor behind declining employee enrollment is the sharp increase in the size of contributions required from workers. With respect to the business cycle, 1989 and 1996 represent comparable years. Unemployment was low (5.3 and 5.4 percent, respectively), and the economy had expanded for the previous four years in 1996 and the previous seven in 1989. Hence, the major changes in coverage patterns from 1989 to 1996 are likely to reflect long-term changes in the U.S. economy rather than the business cycle. For both large and small employers, the health insurance marketplace changed dramatically during the study years. The percentage of Americans with employer-based coverage who were enrolled in managed care plans grew from 33 percent to 73 percent. 8 Premiums for a health insurance policy were rising at a rate of 18 percent in 1989, but this rate declined every year thereafter, hitting a historical low in 1996 of less than 1 percent. Employees were faced with increased contribution rates for single and family coverage, particularly in small firms. For example, employees in small firms paid 12 percent of the cost of single coverage in 1988; by 1996 they paid 33 percent of the premium cost. 9 Between 1989 and 1996 most states enacted legislation to reform the small-group insurance market. The most common reforms included rating bands on pricing of insurance to small employers, prohibition of preexisting condition exclusions for persons moving from one firm with insurance to another, and guaranteed renewal of insurance coverage. 10 BENEFITS IN SMALL FIRMS The percentage of small firms offering coverage to workers grew slightly, from 46 percent in 1989 to 49 percent in 1996 (Exhibit 1). Almost all of the growth was among firms with fewer than twenty-five workers. The strongest gains were made among firms with fewer than ten workers; the percentage of those firms offering coverage increased from 35 percent in 1989 to 42 percent in 1996. Ninetynine percent of firms employing 200 or more workers offered health benefits to their workers in 1989. Cooper and Schone reported the same trend in their study of households, although their analysis did not find the stronger gains in the smaller size categories. 11 Small-firm health benefits became more uniformly available across the United States during the period. The percentage of small firms in the East (the region where coverage H E A L T H A F F A I R S ~ V o l u m e 1 7, N u m b e r 1
EXHIBIT 1 Percentage Of Small Employers Offering Health Insurance, 1989 And 1996 Firm size and region 1989 1996 All firms with 1 199 employees 46% 49% a Size 1 9 employees 10 24 employees 25 49 employees 50 99 employees 100 199 employees Region Northeast Midwest South West SOURCES: Health Insurance Association of America, 1989; and KPMG Peat Marwick, 1996. 35 90 96 95 56 42 44 46 42 a 78 a 90 91 a 96 50 a 49 a 48 a 49 a historically is most likely to be offered) providing health benefits fell from 56 percent to 50 percent. In contrast, more small firms in the South and the Midwest (where the fewest small firms offered health benefits in 1989) offered coverage in 1996 than in 1989. Both surveys asked about the importance of various factors in a firm s decision not to offer coverage, although phrasing of the questions changed slightly. 12 In each year the high cost of insurance is the reason most frequently identified as very important (Exhibit 2). But the importance of this and the two other questions concerning costs of insurance declined sharply between 1989 and 1996. In contrast, responses relating to lack of employee interest in health insurance ( not interested and not necessary to attract employees ) became more common over time. WHO ACCEPTS EMPLOYER-BASED COVERAGE? Although the proportion of small firms offering coverage increased between 1989 and 1996, the proportion of employees enrolled in 169 EXHIBIT 2 Small Employers Reasons For Not Offering Health Insurance, 1989, 1993, And 1996 Percent indicating very important Reason 1989 1993 1996 Insurance premiums are too expensive The firm s profits won t cover the cost of health insurance Employees are not interested in health insurance 76% 61 29 85% a 65 a 44 a 58% a 43 a 33 a Employee turnover is too high to warrant offering health insurance Health insurance coverage is not necessary to attract employees Concern over future health care costs 21 24 55 18 a 39 a 53 a SOURCES: Health Insurance Association of America, 1989; and KPMG Peat Marwick, 1993 and 1996. b Not asked. 11 a b 46 a
170 small-employer plans decreased from percent in 1989 to 67 percent in 1996 (Exhibit 3). This decline occurred relatively uniformly across all small firms and among large firms (200 or more employees) as well. The decline is more likely to be attributable to decisions by employees than to changes in firms eligibility policies. Although data were not collected in 1989, waiting periods declined from an average of 2.7 months in 1993 to 2.4 months in 1996 in small firms. Likewise, trends in part-time and temporary employment do not appear to be a major factor. Although part-time and temporary workers constituted a larger percentage of the workforce in 1996, the magnitude of change is not large enough to have an appreciable effect on enrollment rates. The proportion of parttime workers in the employed population grew from 17 percent to 18 percent, and the number of temporary workers increased from 1 percent to 2 percent of the workforce. Between 1993 and 1996 there was no change in the percentage of small firms granting parttime workers eligibility for health benefits (30 percent). 13 Perhaps the most important factor behind declining employee enrollment is the sharp increase in the size of contributions required from workers. For example, a combination of premium increases and reduction in the proportion of premiums paid by employers caused the average monthly contribution for family coverage for workers in small firms (fewer than 200 workers) to increase from $34 to $175 between 1988 and 1996. 14 A related factor is declining inflation-adjusted incomes for less skilled workers, such as those without college degrees. The net effect is that more workers have declined coverage, opting to limit coverage to their spouse s policy, or may have forgone coverage altogether. Cooper and Schone s results confirm that this decline in enrollment has led to more employed persons and their families not having health insurance coverage. When we combine the phenomenon of more small firms offering coverage with fewer enrolled employees in those firms, the net result is a decline in the proportion of workers who are enrolled in employment-based coverage through their job. The percentage of workers covered by an employment-based health plan fell from 57 percent to 52 percent for small firms and from percent to 67 percent for large firms (Exhibit 3). These declines were generally sharpest in the larger firms, although larger firms consistently have a higher proportion of enrolled workers. DISCUSSION For years the debate over health insurance coverage has revolved around policies to encourage small firms to offer coverage. Policymakers have considered everything from mandates and subsidies to reforms of the small-group EXHIBIT 3 Percentage Of Employees Enrolled In A Health Plan, 1989 And 1996 In firms offering health plans In all firms All firms with 1 199 employees 1 9 employees 10 24 employees 25 49 employees 50 99 employees 100 199 employees All firms with 200 or more employees 1989 1996 1989 1996 % 79 71 78 75 76 73 66% a 74 67 74 70 57% 35 50 69 73 52% a 36 a 52 a 66 a 64 a 64 a SOURCES: Health Insurance Association of America, 1989; and KPMG Peat Marwick, 1996. NOTE: All figures weighted by number of employees. H E A L T H A F F A I R S ~ V o l u m e 1 7, N u m b e r 1
insurance market. Reforms have gone forward, but the jury is still out on their impact. The news that small firms are increasingly willing to offer health insurance coverage to their employees is very encouraging. Our study suggests that the dramatic slowing of premium increases in recent years may have played a role. But a new issue has arisen: employees willingness to enroll. The proportion of employees enrolling in employer plans has declined, and this decline has outweighed the increase in the proportion of firms offering coverage. In addition, firms that do not offer coverage are increasingly citing lack of interest on the part of their employees as a major reason behind their decision. The two phenomena are not independent. Accompanying firms greater willingness to offer coverage is the requirement that employees pay a larger portion of the premium. With larger personal contributions (meaning smaller employer subsidies), insurance is bound to be less attractive. Indeed, for those with the lowest expected use of medical care, purchasing care through employment-based insurance could well be more expensive than paying for it directly. The problem of lack of insurance coverage may increasingly be one of basic affordability. Years of health care costs that have grown more rapidly than personal incomes have made private health insurance unaffordable to an increasing proportion of working people. The absence of real income gains for those with the least education and fewest skills exacerbates this disparity. Progress in reforming markets for small-group coverage is encouraging, but the problem of basic affordability has more than offset whatever advances have been made. The United States appears to be losing ground in the quest for universal insurance coverage through a predominantly employment-based system. The research for this project was conducted as a collaboration between the Center for Studying Health System Change and KPMG Peat Marwick. It was funded by the Robert Wood Johnson Foundation. NOTES 1. KPMG Peat Marwick Survey of Employer- Sponsored Health Plans, 1996 (calculations by the authors). 2. P.F. Cooper and B.S. Schone, More Offers, Fewer Takers for Employment-Based Health Insurance: 1987 and 1996, Health Affairs (November/December 1997): 142 162. 3. See J.R. Gabel, P.B. Ginsburg, and K.A. Hunt, Small Employers and Their Health Benefits, 1988 1996: An Awkward Adolescence, Health Affairs (September/October 1997): 103 110. 4. The 1989 HIAA survey includes 2,521 firms, of which 1,241 firms employ 199 or fewer workers. The 1993 KPMG sample includes 1,953 firms, including 950 firms with fewer than 200 workers. The 1996 KPMG survey consists of 1,965 firms, including 810 with fewer than 200 workers. 5. In both 1989 and 1996 interviewers regarded oneworker firms as those with an owner and a worker. However, there was more aggressive application of this criterion in 1996 than in 1989. 6. We calculated this common distribution through a computation of the geometric mean of the employer weight from 1989 and 1996. 7. When employer weights are used, the smallest firms (one to nine workers) constitute the majority of weights. When employee weights are used (to show the situation from the perspective of employees), the smallest firms account for only 17 percent. 8. J. Gabel et al., Employer-Sponsored Health Insurance, 1989, Health Affairs (Fall 1990): 161 175; and Gabel et al., Small Employers and Their Health Benefits, 1988 1996. 9. Gabel et al., Small Employers and Their Health Benefits, 1988 1996. 10. Alpha Center, State Health Reform: Recent Public Sector Activity (Washington: National Institute for Health Care Management, May 1995), A3 A10. 11. The much lower percentage of firms offering coverage in our analysis reflects employer versus employee weighting. Using employee weights, 73 percent of employees of small firms were offered health plans by their employers in 1996. 12. In addition, the KPMG survey asked a more extensive set of questions, so for many cases there is no comparison question for 1989. 13. In 1996 only 8 percent of employees who worked in small firms (9 percent in large firms) worked for an employer that extended health benefits to temporary workers. 14. Gabel et al., Small Employers and Their Health Benefits, 1988 1996. 171