EBRI Some analysts argue that mandating community rating or eliminating demographic adjustments would raise EM PI.t)YI_ I.: employ Furthermore ed worke , rs increases and 22 pin ercethe nt number of workers of individuals iPr n firms incipal wiTabl th without e fewer P 3oint health ts han 10insurance employees were were greatest covered among through those a gro whose up Table 5 ployment-based coverage Americans , particular With ly and among Without individua H lealth s (and Insur thei an r ce: families Implications ) working infor smaReform ll firms. The number BENEFIT Nonelderly Population with Selected Sources of Health Insurance Coverage, by Family Income as a rates for many groups Simulated and create Single incentives Adult Premiu for m adverse s Under selection. Various Assu Adverse mptions selection on Size occurs of Emp when loyeindividuals r with family health plan head sponsored worked for bya their small own firm employer rather than in for 1992, a la compared rge firm. Among with 70 the percent additional of workers 4.2 million in firms T-92 nonelderly with 1,000 or of nonelderly Americans with employment-based coverage in 1992 was 138.0 million (62.5 percent of the nonelderly Percentage of inPoverty, Risk Pool 1992, andEmploy Whethe ee r M Ben edicaid efit RR es ee cipi arch ents Institut are Includ e Analysis ed in th ofeth Pool e March 1993 CPS RENEARCil • Eighty-three percent of nonelderly Americans and 99 percent of elderly Americans Mr. Chairman, I am pleased to submit this statement for the record on health insurance reform. The Employee greater health risks are disproportionately enrolled in a particular plan or group. Community rating limits insurers' more Americans employees. without health insurance coverage between 1989 and 1992, 19 percent were in families in which the population), a decrease from 139.8 million (64.1 percent of the total nonelderly population) in 1991 (table 1). IN'TIT! TE Family Income EmployerCoverage No Health Employer Size Cap (aged 65 and over) were covered by either public or private health insurance in 1992. BenefitasResearch a PercentaInstitute ge (EBRI), aTotal nonprofit, nonpartisan public policy research Other organization, Total is dedicated Insurance to Composition ability to charge different premiums to groups on the basis of risk. As a result, premiums for groups that represent family head worked for a firm with fewer than 25 employees; 21 percent were ill families in which the family head of Poverty Total Private Total Direct Indirect Private Public Medicaid Coverage of Risk Pool 100 500 1,000 Workers' family members were also more likely to be covered by an employment-based plan if the family head providing These est objective imates analysis and mostofof health those care presented and other below work are force derived issues. from the March 1993 supplement to the Census good health risks would rise with the implementation of community rating, while premiums for groups represent- worked for a firm with 25-99 employees; 25 percent were from families headed by a nonworker; 14 percent were (millions) • The number of nonelderly Americans without health insurance increased to 38.5 Medicaid Out $2,181 $ 2,097 $2,068 worked for a large firm. Among workers (and their families) in firms with fewer than 10 employees, 38.8 percent Bureau's Current Population Survey (CPS). Most researchers familiar with this survey agree that the numbers ing poorer Total risks would 220.8 fall. Some 156.6 of the healthie 138r .0 individuals 68,9 would 69.1choose 18.8 not to purchase 33.4 health 25.6insurance 38.5as a All Under Cap In $2,336 $2,236 $2,202 from families in which the family head worked for a firm with 100-499 employees; and 21 percent were from million in 1992 (17.4 percent of the nonelderly population), from 36.3 million in 1991 While the employment-based system for financing health insurance coverage continues to provide coverage for were covered 0-99% by an employment-based 33.0 6.1 plan, compared 3.6 1,5 with 81.4 pe 2.1 rcent of 2.5 workers (and 17.2 their 16.4 family members) 10.8 in presented provide a snapshot of insurance coverage at a given point in time during 1992. Another Census survey, result of the premium Source: Employee increase, Benefit while Research more Institute of those simulations individuals using the who March are 1993 poorsupplement er health to risks the Current would purchase health fanglies in which the family head worked for a firm with 500 or more employees. The increase in noncoverage 100%-124% 9,8 4.0 3.1 1,2 1.9 0.9 2.9 2.5 3.5 Population Survey and the National Medical Expenditure Survey. (16.6 percent), 35.7 million in 1990 (16.5 percent), and 34.4 million in 1989 (16.1 per- 63 percent of Americans under age 65, that source of coverage is eroding for many, especially those who are em- firms with 1,000 or more employees (table 2). 125°/,=-149% 9.6 4.7 3.7 1.5 2.2 0.9 2.0 1.6 3.4 the Survey of Income and Program Participation (SIPP), follows a smaller group of individuals over a two-and-one- insurance. The result would be an increase in the pool's average risk, increasing premiums and potentially creating among those in small firms was even more pronounced between 1991 and 1992. Forty-two percent of the additional cent). 150%--199% 20,2 12.6 10.7 4.5 6.2 1.8 3.0 2.0 5.6 ployed by smaller employers. Eighty-three percent of nonelderly Americans and 99 percent of elderly Americans half-year 200°period. /_399% The latest 74.7available 61.8information 55.6 from 25 SIPP .5 indicates 30.1 that, in 6.3 1987, 32 million 5.2 (132.5 percent) 10.4 Ameri- a vicious cycle that could end in an unviable health insurance market. The likelihood of this scenario actually 2.2 million individuals without coverage between 1991 and 1992 were in families in which the family head worked Income400% andorHealth More Insur73.5 ance Coverage 67.5 61.3 34.7 26.6 6.3 3,1 0.7 4.9 (aged 65 and over) were covered by either public or private health insurance in 1992 (table 1). Although some of the cans were uninsured at any given moment, 50 million (21 percent) were uninsured for some portion of 1987, and occurring depends on the sensitivity of the demand for health insurance to changes in premiums among individuals for an employer with fewer than 25 employees. An additional 15 percent were in families in which the family head Mana • These ged com espet timates ition m pr ode ov lside often a snaps requirehothat t oemployers f insuranceunder cova era certain ge at size a g must iven purchase point incoverage time (percentage within coverage categories) nonelderly had public health insurance (15 percent), the most common source of coverage was private insurance Statement 16 who million represent (7 percent) good and were bad uninsured risks and for on all theof ability 1987. of Adjusting individualsthose to dete numbers rmine for thei population r own risk growth status. and increases Total 100.0% 100.0% 100.0% 100.0% 100.0% 100.0% 100.0% 100.0% 100.0% worked Income du forrian in sgalso employer 1992. relatedAdj with touh sealth tbetween ing insurance estim 25 ates and coverage. 99fr employees. om sur Invgeneral, eys that individuals follow iwith ndivid higher uals levels over oftiincome me for are through a risk pool formed by a regional H1PC or alliance. As table 5 indicates, increasing the size of the employer usually obtained through an employment-based plan. 0-99% 150 3.9 2.6 2.1 3.0 13.4 51.5 63,9 28.1 in the number of uninsured to make them comparable to the 1992 estimates from the Current Population Survey more likely to be covered by private health insurance, while those with lower levels of income are more likely to be required popula to purchase tion coverage growth th an rough d incthe reases HIPC in or talliance he numb decreases er of un the insp ure remium d tocha mra ged ke th forem cove compa- rage within 100%-124% 4.4 2.5 2.2 1,7 2.7 4.8 8.6 9.6 9.0 Generally, in the managed competition models that create regional health insurance purchasing cooperatives The uninsured also tend to be young. About 25 percent of the uninsured are children under the age of 18. Even yields 1 estimates 25%-149% of 24 million 4.4 Americans 3.0 uninsured 2.7 for the 2,2 entire year, 3.2 38 million 5.0 uninsured 6.0 at any 6,4given moment 8.8 covered rable by a publicly with the sponsored 1992 esti plan. matIn es 1992, from onlythe 16C percent urrent ofPindividuals opulation inSurv families ey ywit ields h income estimates below of $5,000 that pool. Including those presently covered by Medicaid in the pool increases the average premium by an average The number of nonelderly Americans without health insurance increased to 38.5 million in 1992 (17.4 percent of 150%.199% 9.1 8.0 7.8 6,5 9.0 9.7 9.0 8.0 14.5 Before the (HIPCs), low income families, individuals not connected with a group, and employees of small employers would among adults, the uninsured tend to be younger than those with coverage. Twenty-seven percent of those aged during the year, and 58 million uninsured for some part of 1992. 200%-399% 33.8 39.5 40.3 37.0 43.5 33.6 15.4 9.7 26.9 were covered by private health insurance, compared with 92 percent of those in families with income of $50,000 or of about 24 7 percent. million However, American simulating s uninsurthe ed perfocapita r the costs entireof y providing ear, 38 m coverage illion uninsure to nonelderly d atMedicaid any give recipi- n the nonelderly population), from 36.3 million in 1991 (16.6 percent), 35.7 million in 1990 (16.5 percent), and 34.4 400% or More 33.3 43.1 44.4 504 38.5 33.4 9.4 2.5 12.7 purchase coverage through the cooperative. These cooperatives or alliances would community rate over a risk pool 18 to 29 are without health insurance, and that group comprised 40 percent of all uninsured adults. moment during the year, and 58 Committee million uninsured on Financefor some part of 1992. ents after reform indicates these costs to be $3,309. Thus, including them in the risk pools lowers the per capita costs million in 1989 (16.1 percent). A primary reason for the increase in the number of uninsured was a decline in em- (percentagewithin poverty status categories) Employment-Based Coverage formed on the basis of employer size and individual work status. The composition of these cooperatives or alliances U.S. Senate Total 100.0% 70.9% 62.5% 31,2% 31.3% 8.5% 15.1% 11.6% 17.4% to federal and state governments by over 29 percent. Insurance Reform would determine the costs of health insurance and the distribution of these costs. Even if employees of small firms Table 2 • The uninsured tend to be young, live in families that are generally low income and 0-99% 100.0 18,4 10.8 4,5 6.4 7.6 52,0 49,5 32,7 The most important source of health insurance coverage is employment-based coverage. In 1992, 62.5 percent of Non 100 eld %e --124% rly Population 100.0 with Sele40.4 cted Sourc31.3 es of Health 12,3 Insuranc19.0 e, by Firm 9.3 Size of Family 29.3 Head's 25.1 Employer, 35,6 1992 are, on average, healthier than most Americans, the other two groups are likely to be less healthy. For example, Tables headed 4 and 5by assume eitherthat nonwo all individuals rkers or purchase workers health employed insurancebycove small rage. employers. If some individuals choose not 125%-149% 100.0 48.5 38.8 15.5 23.3 9.9 20.9 16.9 35.3 Employee Benefit Research Institut Table e Analysis 1 of the March 1993 CPS The characteristics of the uninsured will be an important determinant of the impact of reforms on the health the nonelderly were covered by employment-based insurance (table 1). This is a reduction from 1988, when 150%--199% 1O0.0 62.2 53.2 Hearing 22.3 on 30.9 9.1 14.8 10.1 27.7 table 4 provides results of simulations on the premiums likely to be charged to single adult individuals in purchas- Nonelderly and Elderly Americans with Selected Sources of Health Insurance Coverage, 1989-1992 to purchase health insurance, and if these individuals are healthier, on average, than those who elect to purchase Firm Size o| EmployerCoverage No Health 200%-399% 1O0.0 82.7 74.4 34,1 40.2 8.5 6.9 3.3 13.9 insurance market. Most of the health insurance reform proposals include a focus on the difficulty individuals and 66.8 percent of the nonelderly were covered through an employment-based insurance plan. Employee Benefit Research Institute Analysis of the March 1990, 1991, 1992, and 1993 CPS Family Head's Total Other Total Insurance •400% Some or More proposals 100.0 move 91.9 the insurance 83.4 market 47.2 toward 36.2 community 8.5 rating 4.3 so that 0.9 insur-6.7 ing cooperatives under different assumptions about their composition. coverage, the premiums will be higher. Health Insurance Reform Employer Total Private Total Direct Indirect Private Public Medicaid Coverage small groups have in obtaining health insurance at the same cost as larger groups. Small groups often face higher Total Population Nonelderly Elderly ance would be offered to all small groups at common rates. Community rating results Note: Details may not add to totals because individuals may receive coverage from more than one source. Declines in employment-based health insurance coverage were somewhat offset by an increase in the number of (millions) costs per participant because of their higher per capita administrative costs and insurance companies' limited ability Source of Coverage 1989 1990 1991 1992 1989 1990 1991 1992 1989 1990 1991 1992 Changing in increased the way risks premiums are pooled for will groups have irnportant that represent consequences good health in the health risks,insu while rance premiums market. Many by Total 220.8 156.6 138.0 68.9 69.1 18.8 33.4 25.6 38.5 Americans with coverage from a public source. The number of nonelderly Americans receiving public coverage (millions) to pool risks. Insurers currently price their policies on the basis of the expected risk of the individual group. If an small groups for groups and individuals representing will seepoorer the costsrisks of health would insurfall. ance Some fall as of a result, the healthier while others individuals will see an increase. Nonworker 25.7 6.6 4.1 Tab 2.7 le 4 1.5 2.5 14.3 12.2 6.0 more. Although many individuals in poor families are covered by public health plans, that coverage is far from steadily increased between 1989 and 1992--33.4 million nonelderly Americans received public coverage in 1992 Total Population 243.3 246.0 246.7 251.7 213.7 215.9 218.1 220.8 29.6 30.1 30.6 30.9 Fewer than 10 32.3 19.5 William 12.5 S. Custer, 5.8 Ph.D. 6.7 7,0 3.6 2.8 9.8 insurer pools all the groups it insures together and charges a premium based on that total pool, some of the groups Simulation of Single Adult Premiums under Various Assumptions of Composition of Risk Pool. would choose not to purchase health insurance as a result of the premium increase, Even 10--24 those groups whose 16.4 premiums 10.4 will increase 8.8 under some 4.3 form 4.5 of community 1.6 rating 1.9 may be better 1.6 off if 4.4 the Total with Private Health universal. In 1992, less than 50 percent of the nonelderly with income below the poverty line were covered by Director of Research (15.1 percent Assum of the es All total Employ nonelderly ees of Employ population), ers with compared Fewer Than with500 31.7 Employ million, ees orPurchas 14.5 percent, e Coveof ragthe e through nonelderly 25-99 26.8 19.2 17.3 8.8 8.5 1.9 28 21 5.5 in theInsur pool ance will pay higher premiums 180.4 178 than .9 they 178.4 would 177.5 if the 160.4 premiums 158.3 we 157.7 re set156.6 on their200 risk alone, 20.6 while 20.7 others 20.9 will Health Insurance Purchasing Cooperative (HIPC), while more of those individuals who are poorer health risks would purchase health reforms stabilize premiums. Small employers who currently have good risk profiles may still not offer health Medicaid. 100-499 29.7 23.6 22.2 11.1 11.1 1_4 2.8 2.1 4.2 Employee Benefit Research Institute Employer coverage 150.2 148.7 150.0 148.0 140.8 138.7 139.8 138.0 9.4 10.0 10.1 10.1 population in 1991; 29.2 million, or 13.5 percent, in 1990; and 26.2 million, or 12.3 percent, in 1989 (table 1). The All Other Employees Purchase Coverage Outside HIPC 500--999 11.7 9.8 9.3 4.5 4.7 0.6 1.0 0.7 1.3 pay lower premiums. Other private coverage 30.3 30.3 28.6 29.6 19.7 19.7 18.0 18.8 10.6 10.6 10.6 10.8 insurance. The result would be an increase in the pool's average risk, increasing benefits because one catastrophic illness could make these health benefits unaffordable. Some form of community 1,000 or More 78.3 67.4 63.7 31.6 32 1 3.8 6.9 4.0 7.2 Total with Public Health increase in public coverage is, at least in part, due to the impact of the recent recession and to changes in Medicaid Composition Premiums Premiums Characteristics of the Uninsured Insurance 54.5 58.1 61.2 63.2 26.2 29.2 31.7 33.4 28.3 28.9 29.5 29.8 rating reduces premiums the variability and potentially in health insurance creating premiums. a vicious cycle that could end in an unviable (percentage within coverage categories) of RiskPool OutsideHIPC InsideHIPC In the current health insurance market, insurers who attempt to pool risk across groups will find the lower risk coverageMed for icare children and pregnant 31.5 women. 32.3 32.9 33.7 3.2 3.5 3.5 4.0 28.2 28.8 29.4 29.7 Washington, D.C. Total health insurance 100.0% market. 100.0% 100.0% 100.0% 100.0% 100.0% 100.0% 100.0% 100.0% Medicaid 21.1 24.2 26.8 28.5 18.5 21.6 23.9 25.6 2.6 2.6 2.9 2.9 WorkersOnly $2,188 $1,979 groups will choose another insurer whose premiums reflect only their own risks and are therefore lower. By chang- CHAMPUS/CHAMPVA a 7.0 7.0 7.1 6.9 5.9 5.9 5.9 5.7 1.1 1.1 1.2 1.2 In 1992, 17.4 percent of the nonelderly population--or 38.5 million people--were not covered by private health Nonworker andNonworkers 11.6 4.2 3.0 $2,018 3.9 2.1 13,5 42.9$2,097 47.7 15.6 Reforming the health insurance market by itself is unlikely to significantly increase health insurance coverage. Workers No Health Insurance 34.7 36.0 36.6 389 34.4 35.7 36.3 38.5 0.3 0.3 0.3 0.4 Fewer than 10 146 125 9.1 8.5 9.7 37.5 10.9 11.0 25.5 andMedicaid $1,683 $2,236 ing the incentives that keep insurers from pooling small groups, employment-based coverage may expand to insurance and did not receive publicly financed health assistance, up from 36.3 million (16.6 percent) in 1991. 10--24 7.4 6.6 6.4 6.2 6.5 8.5 5.8 6.1 11.5 Although • An some Employee groups may Benefit see lowe Research r premiums Institute as a result simulation of insurancefound reforms, thatothers pooling will face working premium in- (percentage) 1 February 1994 25-99 12.1 12.2 12.6 12.8 12.3 9.9 8.4 8.4 14.4 Source:EmployeeBenefitResearchInstitutesimulations usingthe March1993supplement to the CurrentPopulation include many of the employed uninsured in small firms and their dependents. Among the 38.5 million nonelderly Americans who did not have health insurance coverage in 1992, most were Total Population 100.0% 100.0% 100.0% 100.0% 100.0% 100.0% 100.0% 100.0% 100.0% 100.0% 100.0% 100.0% creases, single and the adults stabilization with of nonworkers health premiums and may Medicaid not be enough eligible to offset individuals these incrincreases eases. Health worker insurance 100-499 Not surprisingly, workers 13.4 were 15.1 much more16.1 likely to 16.2 be covered 16.1 by employment-based 7.6 8.4 health 8.2 plans than10.9 non- Surveyandthe NationalMedicalExpenditure Survey. 500-999 5.3 6.3 6.7 6.6 6.9 3.0 3.0 2.9 3.3 Total with Private Health working adults (56.7 percent), while the remainder were children (25.4 percent) and nonworking adults premiums by almost 13 percent. reform workers.will Seventy redistribute percenttheofcosts workers of health were care covered services by an from employment-based the poorer risks plan, who may compared currently withbeonly excluded 37 percent from of 1,000 orMore 35.5 43.1 46.2 45.9 46.4 20.1 20.6 15.8 18.8 Insurance 74.2 72.7 71.7 70.5 75.1 73.3 72.3 70.9 67.7 68.4 67.7 67.7 Most proposals include some means for guaranteeing that all small groups have access to insurance and are not Employer coverage 61.3 60.4 60.3 58.8 65.4 64.2 64.1 62.5 32.0 33.2 33.1 32.6 (17.8 percent). The total number of uninsured under age 65 has increased from 33.6 million in 1988 to 38.5 million in the nonworkers. market to In theaddition, better risks. 77 percent The burdens of individuals imposed inby (percentage families this redistribution headed within firm bysize awill full-year, categ depend ories)full-time on the number worker were of good covered risks denied Other coverage privatebased coverage on individual 12.4 cha 12.3racteristics. 11.5 11.8 However 9.2 , proponents 9.1 8.of 2 insu8.5 rance 35.7 market 35.2 reform34.7 recognize 35.0 Total 100.0% 70.9% 62.5% 31.2% 31.3% 85% 15.1% 11.6% 17.4% Total with Public Health 1992. Although some of this increase can be attributed to population growth, the percentage reporting no health who remain The views in the expressed pool. in this statement are solely those of the author and should not be attributed to the by group health plans, compared with only 37 percent of those in families headed by other workers and 16 percent Table 4 estimates single adult premiums that would arise under different risk pools, in the workers only pool, that guaranteed Insurance availability alone 22.4 accomplishe 23.6 s24.6 little25.1 unless 12.3 premium 13.5 rates 14.5 for small 15.1 group 95.8 s are96.0 stabilized. 96.3 Some 96.6 Nonworker 100.0 25.8 16.0 10.3 5.7 9.9 55.7 47.6 23.4 Employee Benefit Research Institute, its officers, trustees, sponsors, or other staff. The Employee Benefit insurance Medicare coverage has also increased 12.9 13.1 from 15.9 13.2 percent 13.4 to 17. 1.5 4 percent. 1.6 1.6 1.8 95.6 95.7 96.0 96.2 of individuals Fewer than 10 in families 100.0 headed by 60 a .nonworker. 4 38.8 18.0 20.7 21,8 11.3 8.7 30.4 all employees of firms with fewer than 500 empk)yees are included in the pool. The estimated premium is $1,979. Research Institute is a nonprofit, nonpartisan, public policy research organization. proposals move the insurance market toward community rating so that insurance would be offered to all small Medicaid 8.7 9.8 10.8 11.3 8.7 100 11.0 11.6 8.7 8.6 9.5 9.4 10-24 100.0 63.1 53.4 26.1 27.3 9.8 11.9 9.5 27.0 CHAMPUS/CHAMPVA a 2.9 2.8 2.9 2.7 2.8 2.7 27 2.6 3.7 3.7 3.8 3.9 When 25-99 nonworking individuals 100.0 are added 71.7 to the 64.8 pool, the32.9 annual premium 31.9 rises 7.0by $118,10.4 and when 8.0 individuals 20.7now groups at common rates. Others would allow insurers to adjust community rates for factors such as age, sex, geo- No The Health uninsured Insurance live in families 14.3 that 14.6 are generally 14.7 1low 5.4 income 161 and 16.5employed 16.6 by 17.4 small 1.0 employers. 1.0 Just 0.9over 1.2 100-499 100.0 79.6 74.9 37.5 37.4 4.8 9.5 7.0 14.2 Workers were also more likely to be covered by an employment-based health plan if they worked for an em- Suite 600 receiving Medicaid are added, the premium rises by another $139 to a total of $2,236. Pooling working single adults 500-999 100.0 84.3 796 38.9 40 7 4.8 8.6 6.3 11.0 graphic location, and industry type (class rating). 2121 K Street, 60 percent Note: NWDetails of the mayuninsu not add redto totals live because in families individuals with may totalreceive family coverage incomefromofmore less than thanone 200 source. percent of the federal poverty ployer 1,000 with or More a larger number 100.0of employees. 86.1 Insurers 81.4 may 40.4 charge less 41.0 per capita 4.8 for large 8.8 employer 5.2 plans because 9.2 with alncludes nonworkers only theand retired Medicaid military and eligible members individuals of their families increases provided wo coverage rker pthrough remiums the Civilian by almost Health13 and percent, Medical or Program an esti- for the Washington, DC level. Fifty-one percent of the uninsured live in families whose family head works for an employer with fewer than Uniformed Services and the Civilian Health and Medical Program of the Veterans Administration. Excludes active duty military personnel Note: Details may not add to totals because individuals may receive coverage from more than one source. they are able to spread both risk and administrative costs over a greater number of people. Only 23 percent of self- 20037-1896 matedand $21 members per month. of their families. 100 employees. 202-659-0670 Fax 202-775-6312