DataWatch Trends In Employee Health Benefits by Pamela Farley Short This DataWatch describes the trends in employee health insurance benefits and out-of-pocket expenditures during the past decade. Employee health benefits have undergone changes during the 1980s, so that workers are increasingly required to share in the cost of their health care. Employers have instituted these changes to slow the increase in their own health costs and to instill more cost-consciousness among employees. Increased cost sharing is taking two forms. Employees are paying more in health insurance premiums, and, because of changes in deductibles and coinsurance requirements, they also are paying more for their health care. Employees now are sharing the effects of medical care inflation at a time when medical care prices are increasing much faster than wages and prices generally. For raising awareness of health care costs, the most significant trend is the introduction of cost sharing in two areas where the great majority of employees used to pay nothing at all: for hospital care and health insurance itself. Where cost sharing has been imposed, its level also has increased. According to surveys of the health benefits of full-time employees in medium and large firms by the U.S. Bureau of Labor Statistics (BLS), the premiums paid by employees for contributory plans (those requiring an employee contribution) have escalated faster than the rate of medical care inflation. 1 Deductibles under major medical insurance have increased as well, and, because employee cost sharing under major medical insurance is a percentage of covered expenses, out-of-pocket payments under such insurance plans keep pace with increases in covered expenses. While the trend for the average employee is an increase in out-ofpocket expenses, protection against financially ruinous expenses for employees faced with unusually large, catastrophic medical bills has Pamela Farley Short is a senior economist at the National Center for Health Services Research/ Health Care Technology Assessment (NCHSR) in Rockville, Maryland. She received her doctorate in economics from Yak University, with an emphasis in health economics.
DATAWATCH 1 8 7 improved substantially. In the 1980 BLS survey, little more than half of employees were protected by a stop-loss provision in their major medical insurance. Under such a provision, the insurer covers all expenses after the policyholder s share reaches a specified dollar amount, thereby limiting the potential out-of-pocket liability associated with coinsurance and deductibles. In 1986, four out of five employees with major medical insurance were covered by a stop-loss provision. Because of the comprehensiveness of the benefits offered by health maintenance organizations (HMOs), the continued growth of HMO enrollment also means improved catastrophic protection. Although the transformation of employee health benefits along these lines has been substantial, it is also far from complete. Most employees are still covered by basic hospital insurance that provides generally comprehensive coverage for hospital expenses. Most also have access to insurance for themselves, although not necessarily for their families, entirely at their employer s expense. As was the case a decade ago, major medical insurance most commonly involves a deductible of $100 and 20 percent coinsurance. The range of services under private insurance has expanded to reduce out-of-pocket expenses in areas that previously were financed entirely by employees. Finally, although the widespread introduction of major medical stop-loss provisions is an important improvement in catastrophic protection, one-fifth of employees could appropriately be considered underinsured because their major medical insurance lacks this protection. Cost Sharing For Health Services Since 1980, the BLS has conducted an annual survey of employee benefits in a national sample of medium and large private firms, generally defined as having a minimum of 100 or 250 employees, depending on the industry. The survey is limited to full-time workers, excluding part-time, seasonal, and temporary employees, and the employees of small firms. Company executives also are excluded. However, the survey is a good source of information regarding trends in employee health benefits, because of the consistency of the data collection procedures and sampling design over time. In addition, although limited in scope by design, the sample is randomly selected and nationally representative of all full-time employees in good-sized firms. These are the workers who enjoyed the most comprehensive health insurance benefits at the start of the decade. 2 Among the workers included in the BLS survey in 1980, just 12 percent had coverage for hospital room and board charges that was limited to major medical insurance (Exhibit 1), and therefore involved a deductible
188 HEALTH AFFAIRS Summer 1988 Exhibit 1 Active, Full-Time Workers In Medium And Large Firms With Hospital Room And Board Coverage, By Type Of Coverage, 1980-1986 Percent distribution Basic Basic and Major medical only a major medical only 1980 16% 72% 12% 1983 17 67 16 1986 23 44 32 Source: U.S. Bureau of Labor Statistics, Employee Benefits in Medium and Large Firms, Annual Bulletins. a Includes HMO plans. (typically $100) and coinsurance (typically 20 percent). By 1986, that figure had climbed to 32 percent, as coverage for inpatient services was brought under comprehensive major medical insurance instead of the traditional combination of basic inpatient and supplementary major medical insurance. Given that basic benefits almost always are set at 100 percent of the semiprivate room rate, the preponderance of basic benefits a decade ago was symptomatic of the complete absence of hospital cost sharing for nearly three-quarters of employees. 3 Indeed, at that time the deductible and coinsurance requirements of comprehensive major medical plans often were waived for inpatient services, thereby providing complete first-dollar coverage similar to basic hospital insurance. 4 Now, basic hospital coverage is changing to include deductibles more often and, in that respect, to resemble major medical insurance. Adding the 14 percent of employees with a basic hospital deductible to the 32 percent with major medical insurance only, 46 percent of employees were subject to a hospital deductible in 1986. 5 That is over four times the comparable figure in 1977 for employees in groups of 250 or more, similar to those in the BLS survey. 6 The upward trend in hospital cost sharing is evident in the national health accounts published each year by the Health Care Financing Administration (HCFA), as shown in Exhibit 2. As a percentage of consumer expenditures paid either directly out of pocket or indirectly by private insurance, hospital out-of-pocket expenditures declined during the 1970s to a low of 17 percent and then climbed to 21 percent in 1986. 7 HCFA projects this trend will continue through the year 2000, when the share of hospital expenditures paid out of pocket will return approximately to its 1970 level. In the meantime, the share of all consumer health expenditures paid out of pocket recently has been and is expected to stay around 48 percent. The latter observation, given increasing out-of-pocket expenditures for hospital care, implies that there are other health services for which
DATAWATCH 189 Exhibit 2 Percentage Of Consumer Expenditures For Health Care Paid Out Of Pocket, 1970-2000a Actual 1970 6 3 % 2 5 % 58% _ b 1975 5 5 1 8 4 6 _ b 1980 4 8 1 7 4 2 6 8 % 1 9 8 5 4 8 1 9 3 9 6 6 1986 4 9 2 1 4 0 6 6 Projected 1 9 9 0 4 8 2 1 3 9 6 5 2000 4 8 2 6.. 3 7 6 4 Sources: Actual 1970-1980 numbers are from D.R. Waldo, K.R. Levit. and H. Lazenby, National Health Expenditures, 1986, Health Care Financing Review (Fall 1986): 1-21; actual 1985-1986 and projected 1990-2000 numbers are from the Health Care Financing Administration, Division of Cost Estimates, National Health Expenditures 1986-2000, Health Care Financing Review (Summer 1987): 1-36. a Consumer expenditures are defined as the sum of private insurance payments and out-of-pocket payments. b Not available in published statistics. cost sharing has declined, A slow downward trend in the share of physician expenditures paid out of pocket is evident in the HCFA accounts, and since 1980 the same is true for dental care (Exhibit 2). Along similar lines, the BLS survey indicates that the scope of employee health benefits is expanding. Between 1982 and 1986, the percentage of workers with vision benefits increased from 25 to 40 percent. In 1982, the BLS reported that 9 percent of employees had coverage for hearing care, 50 percent for alcoholism treatment, and 37 percent for drug abuse treatment. In 1986, the corresponding figures were 20 percent with hearing benefits; 70 percent covered for alcoholism treatment, and 66 percent covered for drug abuse treatment. Thirty-seven percent of employees were insured for home health care in 1982 and 66 percent in 1986. Notably, the prevalence of dental insurance peaked at 77 percent of employees in 1984 and has declined since then to 68 percent in 1986. Although often cited as an indicator of increased cost sharing, the increase in major medical deductibles during the decade has served mainly to preserve the level of employee cost sharing (Exhibit 3). Because medical care prices increased about 60 percent between 1980 and 1986, a constant dollar deductible in 1986 corresponded to a smaller share of the cost of identical services in 1980. The actual increase in the average value of major medical deductibles (about 45 percent) has not been substantial in relation to the 60 percent increase in medical prices. Deductibles of $100 still are the most common, but $50 deductibles are becoming relatively rare while deductibles of $150 or $200 are far more frequent.
190 HEALTH AFFAIRS Summer 1988 Exhibit 3 Active, Full-Time Workers In Medium And Large Firms With Major Medical Insurance, By Amount Of Deductible, 1980-1986 Percent distribution a Amount of deductible 1980 1983 1986 Less than $100 32% 2 % 14% $100 53 52 44 $101-150 5 8 16 $151-200 3 5 15 More than $200 _b _b 6 Source: U.S. Bureau of Labor Statistics, Employee Benefit in Medium and Large Firms, Annual Bulletins. a Columns add up to less than 100 percent because workers without a deductible or a deductible based on earnings are not shown. b Included above. Nationally, per capita out-of-pocket expenditures have increased 9.7 percent per year during the 1980s, decidedly faster than during the 1970s (Exhibit 4). Although other factors are reflected in these statistics, changes in employee health benefits apparently have had a major influence on this trend. The faster growth of out-of-pocket expenditures in the 1980s is not due to a retrenchment in government health expenditures. The share of both personal health expenditures and hospital expenditures financed by public sources has remained constant through the 1980s, although increases in public expenditures helped to slow the Exhibit 4 Growth In Per Capita Out-Of-Pocket Expenditures And In The Consumer Price Index (CPI), 1970-1986 Out-of-pocket expenditures Personal health care Hospital Physician Dental Nursing home Level 1970 1980 1986 $123 15 30 11 $268 $466 1 0 5 6 7 43 77 38 78 Annual percentage change 1970-1980 1980-1986 8.1% 8.5 7.4 Consumer Price Index b All items Medical care 116.3 120.6 246.8 265.9 328.4 433.5 7.8% 8.2 4.9% 8.5 Hospital room 145.4 418.9 753.1 11.2 10.3 13.2 9.7% 12.0 9.5 10.2 12.7 Sources: Out-of-pocket expenditures data for 1970 and 1980 are from D.R. Waldo, K.R. Levit, and H. Lazenby, National Health Expenditures, 1986, Health Care Financing Review (Fall 1986): 1-21; out-of-pocket expenditures data for 1986 are from the Health Care Financing Administration, Division of Cost Estimates, National Health Expenditures 1986-2000, Health Care Financing Review (Summer 1987): 1-36. a b Not availale from publishered figures Annual averages of monthly figures.
DATAWATCH 191 growth of out-of-pocket expenditures in the 1970s. 8 Nor has the proportion of the population that is privately insured or insured by employers changed markedly since the late 1970s. 9 Therefore, the main factor in the growth of out-of-pocket expenditures seems to be a decline in health insurance payments to the insured, not a decline in the proportion of the population with private insurance. The upward shift in out-of-pocket expenditures is even more significant when measured against changes in the general cost of living as measured by the Consumer Price Index (CPI). In real terms, the current rate of increase is 4.8 percent per year the difference between the growth in out-of-pocket expenditures and the CPI compared to 0.3 percent per year in the previous decade. This is due largely to the fact that general inflation tapered off considerably (from 7.8 percent to 4.9 percent), while medical care prices continued to increase at about the same rate (8.2 percent in the 1970s and 8.5 percent in the 1980s). In addition, while out-of-pocket expenditures roughly kept pace with the rate of medical care inflation in the 1970s, they have increased even faster than medical care prices in the 1980s. Cost Sharing For Health Insurance In the BLS survey, premium contributions by employees also have increased faster than the rate of medical care inflation (Exhibit 5). 10 Most significant is the increasing number of employees who are enrolled in plans requiring an employee contribution. Historically, employers have Exhibit 5 Premium Contributions Of Active, Full-Time Workers In Medium And Large Firms, 1982-1986 For individual coverage Percent with contribution required 1982 2 % 51% 1983 35 1984 38 54 1985 39 1986 46 59 1982 1983 1984 1985 1986 Average monthly employee premium for contributory plans $ 9 10 12 12 13 For family coverage Source: U.S. Bureau of Labor Statistics, Employee Benefits in Medium and Large Firms, Annual Bulletins. $28 33 36 38 41
1 9 2 HEALTH AFFAIRS Summer 1988 provided free individual coverage for employees more often than free family coverage, but even family coverage was free to 49 percent of employees in 1982. Since then, contributory family coverage has increased from 51 percent of employees in 1982 to 65 percent in 1986. Forty-six percent of employees are now required to contribute toward individual coverage, compared to 29 percent in 1982. In addition, monthly employee premiums for contributory plans have increased 44 percent since 1982 for individual coverage and 46 percent for family coverage, compared to a 32 percent increase in medical care prices. Taking into account both the increasing prevalence of contributory plans and the increase in the premiums required from 1982 to 1986, employee contributions increased per capita (multiplying the numbers from the top and bottom of Exhibit 5 to average over all participating employees) from $31 per year to $72 for individual coverage and from $171 to $320 for family coverage. Catastrophic Protection Although the average out-of-pocket expenditures of employees have increased, the burden of cost sharing has been redistributed, away from the small number of employees who incur very large medical bills to the much larger number with normal illness experiences. In other words, although out-of-pocket expenditures have increased on average, the risk of financial catastrophe has decreased. This has been accomplished mainly through the spread of stop-loss provisions from 55 percent to 81 percent of workers with major medical insurance (Exhibit 6). Although a stop-loss provision is more relevant to catastrophic protection in most situations, raising major medical maximums also means improved protection. The proportion of employees with lifetime maximums of $250,000 or less has been nearly halved since 1983. On the surface, because major medical enrollment has declined from 84 percent of employees with hospital benefits to 76 percent since 1980, it may seem that catastrophic coverage has not improved. However, this decline is more than offset by the growth of HMO enrollment, from 2 percent to 13 percent between 1980 and 1986 in the BLS survey. Because HMO benefits are comprehensive and unlimited, they also provide catastrophic protection. Implications The changes in employee health benefits since 1980 have been substantial. Their effect has been to increase the average level of employee
DATAWATCH 193 Exhibit 6 Full-Time, Active Workers With Major Medical Insurance By Stop-Loss Provisions And Maxim um Benefit, 1980-1986 Stop-loss None With stop-loss $2,000 or less Percent distribution 1980 1983 1986 45% 29% 19% 55 71 81 7 12 10 15 17 22 20 25 30 13 17 19 Maximum benefit Lifetime maximum $250,000 or less _b 47% 25% $250,001-$500,000 -b 12 18 More than $500,000 -b 14 25 Other maximum -b 11 12 No maximum -b 16 20 Source: U.S. Bureau of Labor Statistics, Employee Benefit in Medium and Large Firms, Annual Bulletins. a Dollar amounts refer to covered expenses. The corresponding level of out-of-pocket expenses would be roughly 20 percent of the amount shown. b Not available from published statistics. expenses for health care and health insurance, while providing improved protection against the risk of real financial catastrophe. Given the nature of these changes, employees also are more at risk for the effects of medical care inflation, currently almost twice the general inflation rate. In particular, because hospital care weighs so heavily as a percentage of total medical expenses and has gone up in price so much faster than other medical services, the widespread introduction of hospital cost sharing means that out-of-pocket expenses will track the overall rate of medical care inflation more closely-even without further modification of benefit provisions. Delivery systems that hold down costs with incentives and controls aimed at providers, as an alternative to patient cost sharing, also have taken hold. HMO enrollment in the BLS survey, currently 13 percent, is more than three times the level in 1980. Nationally, HMO enrollment has increased from about 9 million in 1980 to nearly 28 million in 1987, a growth rate of nearly 18 percent per year. 11 The number of employees and dependents in preferred provider organizations (PPOs) also continues to accelerate, from nearly 6 million in 1985 to 16.5 million a year later. 12 Should employer premiums be taxed? These changes have taken place without major new legislation. In particular, despite repeated calls to include employer-paid premiums in the taxable income of employees, this large subsidy for employee health benefits remains. The introduction
194 HEALTH AFFAIRS Summer 1988 of tax-exempt flexible spending accounts to finance the premiums paid by employees is further encouraging health insurance purchases. 13 Nevertheless, despite these incentives to the contrary, employee health benefits are moving in the direction envisioned by reformers with increased HMO enrollment and a shift away from comprehensive front-end benefits and toward better catastrophic protection. Under the circumstances, should the taxation of premiums still be pursued as a matter of federal health policy? 14 The argument in favor is that there is still plenty of room for additional progress in transforming employee benefit practices. While the changes since 1980 have been significant, it is all too easy to exaggerate them. A slight majority of employees (54 percent) still had first-dollar coverage for hospital expenses in 1986. More than half, had access to free health insurance for themselves; a third were offered free family coverage. Major medical deductibles have increased, but on average they have lagged behind the rate of medical care inflation. Stop-loss provisions are now widespread, but one out of five employees is still without this catastrophic protection. Cost sharing for covered services especially hospital care has increased, but a wider range of services now is covered. On the other side of the coin, the growing financial burden on employees must be weighed against the hope of further transforming health insurance benefits. The real price of health care is escalating rapidly for employees, even without the additional price effect of a tax on premiums. From the experience of the early 1980s, it appears that the existing financial pressures may be sufficient to induce continuing changes in health benefits without cutting further into the real disposable income of employees. Nor is the reduction of employee health benefits the only, unambiguous objective of national health policy. Most notably, there also is great interest in reducing the number of uninsured Americans. By subsidizing employer-sponsored health insurance, the tax exemption for employerpaid premiums is one way of attacking this problem without instituting or expanding a public insurance program. At least, it is a way of trying to keep the problem from getting worse when employment in the United States is shifting away from heavy industries where health benefits historically have been generous, to service industries where they have not. Indeed, the 1986 Tax Reform Act moved to extend rather than curtail the tax subsidy, by temporarily allowing tax exemptions for premiums paid by the self-employed and small employers. Finally, as public authorities increasingly require private health insurance arrangements to serve public purposes, public subsidies to compensate for the cost of meeting these objectives may be warranted. For
DATAWATCH 195 example, state laws currently mandate certain employee benefits such as mental health coverage; employer-sponsored plans have been designated the primary payer for elderly workers instead of Medicare; and proposals are before Congress to make health insurance a mandatory employee benefit. To the extent that such legislation forces private health insurance to meet the social objectives of a public insurance program, both efficiency and equity require public financing of the costs. This paper was presented at the annual meeting of the American Public Health Association in New Orleans, October 1987. The views expressed herein are those of the author. No official endorsement by the Department of Health and Human Services or the National Center for Health Services Research and Health Care Technology Assessment is intended or should be inferred. NOTES 1. U.S. Bureau of Labor Statistics (BLS), Employee Benefits in Medium and Large Firms, 1980-1986 editions (Washington, D.C.: U.S. Government Printing Office, 1981-1987). 2. The last exhaustive survey of the health insurance benefits and premiums of all American households, including the employees of small firms and nongroup enrollees, was the 1977 National Medical-Care Expenditure Survey (NMCES). Amy K. Taylor and Walter R. Lawson, Employer and Employee Expenditures for Private Health Insurance, National Health Care Expenditures Study Data Preview 7 and DHHS Pub. no. (PHS)81-3297 (Rockville, Md,: National Center for Health Services Research, 1981); Pamela J. Farley and Gail R. Wilensky, Private Health Insurance: What Benefits Do Employees and their Families Have? Health Affairs (Spring 1983): 92-101; Gail L. Cafferata, Private Health Insurance: Premium Expenditures and Sources of Payment, National Health Care Expenditures Study Data Preview 17, DHHS Pub. no. (PHS)84-3364 (Rockville, Md.: NCHSR, 1984); and Pamela J. Farley, Who Are the Underinsured? Mibank Memorial Fund Quarterly (Summer 1985): 476-503. The Small Business Administration (SBA) recently completed a survey of employee benefits that encompassed firms of all sizes, but interpretation of the SBA estimates is clouded by the 20 percent survey response rate. ICF, Inc., Health Care Coverage and Costs in Small and Large Businesses, report prepared for the Office of Advocacy, US. Small Business Administration (1987). Thus, the trend in employee health benefits in small firms remains a mystery. From NMCES we do know that some of the innovations in larger firms comprehensive major medical insurance with a stop-loss provision instead of basic hospital insurance, major medical deductibles larger than $100 were more common in small firms at the start of the decade, but we do not know if employee cost sharing has increased further in small firms. 3. BLS, Employee Benefits in Medium and large Firms, 1986 (1987); and Farley and Wilensky, Private Health Insurance. 4. Pamela J. Farley, Private Health insurance in the United States, National Health Care Expenditures Study Data Preview 23, DHHS Pub. no. (PHS)86-3406 (Rockville, Md: NCHSR, 1986). These modified comprehensive major medical plans are categorized as basic hospital plans in the BLS statistics. 5. BLS, Employee Benefits in Medium and Large Firms, 1986 (1987). 6. Farley and Wilensky, Private Health Insurance.
196 HEALTH AFFAIRS Summer 1988 7. The share of consumer expenditures paid out of pocket is affected by the size of the uninsured population, as well as the proportion of expenses that are paid by persons who are insured. The elderly are almost always covered by Medicare, so they are rarely uninsured, and as an increasing number are retiring with health benefits from their former employers, private insurance is increasing among the elderly. Pamela Farley Short and Alan C. Monheit, Employers and Medicare as Partners in Financing Health Care for the Elderly, paper presented at a conference sponsored by the Leonard Davis Institute, What We Have Learned from Twenty Years of Medicare: Implications for Public and Private Sector Policy, October 1986. The two data sources that are available to examine the recent trend in the percentage of the population under sixtyfive that is uninsured disagree on whether that figure is increasing. The March supplement of the Current Population Survey shows an increasing proportion of the nonelderly population without insurance (15.6 percent in 1982 and 17.4 percent in 1985). Employee Benefits Research Institute, A Profile of the Nonelderly Population without Health Insurance, EBRI Issue Brief, no. 66 (May 1987). The Survey of Income and Program Participation (SIPP) shows a peak in late 1983, with the relative size of the uninsured population returning to a level similar to the late 1970s. U.S. Bureau of the Census, Disability, Functional Limitation, and Health Insurance Coverage: 1984-1985, Current Population Reports, Series P-70, no. 8 (Washington, DC.: U.S. Government Printing Office, 1986). Although private health insurance is not entirely employment-related, 82 percent of private enrollees (87 percent excluding the elderly) are insured through employers (U.S. Bureau of the Census, 1986). 8. Daniel R. Waldo, Katharine R. Levit, and Helen Lazenby, National Health Expenditures 1986, Health Care Financing Review (Fall 1986): 1-21; and Health Care Financing Administration, National Health Expenditures 1986-2000, Health Care Financing Review (Summer 1987): 1-36. 9. During the first quarter of 1977, 77 percent of the population under sixty-five had private insurance, including 69 percent from their own or a family member s current or former employer (unpublished estimates from the 1977 National Medical Care Expenditure Survey). SIPP yields estimates of the comparable figures for the last quarter of 1985, 77 percent and 66 percent, respectively (U.S. Census, 1986). The estimates from the Current Population Survey for early 1986 are 74 percent privately insured and 66 percent with employment-related insurance (EBRI, 1987). 10. How this picture would change if employees in small firms were added is not clear. In 1977, small firms, although less likely to offer insurance, were more likely to pay the entire premium for insurance that was offered than large firms (Taylor and Lawson, Employer and Employee Expenditures). The SBA s recent survey indicates that this is still true (see note 2.) 11. InterStudy, National HMO Census, 30 June 1983 (Excelsior, Minn.: 1984); and Inter Study, The InterStudy Edge (Summer 1987). 12. Gregory de Lissovoy, Thomas Rice, Jon Gabel, and Heidi Gelzer, Preferred Provider Organizations: One Year Later, Inquiry (Summer 1987): 12-135. 13. By lowering marginal rates and flattening the tax brackets, the 1986 tax reform reduces the incentive to choose employer-paid premiums over the same amount of income, when only the latter is taxed. 14. From a more general public finance perspective, there are other arguments for eliminating the tax exemption. Namely, the revenue loss is significant, and, because the exemption is only for employer-paid premiums, it discriminates against those who must purchase insurance on their own.