STATE OF FLORIDA et al v. UNITED STATES DEPARTMENT OF HEALTH AND HUMAN SERVICES et al

Filing 83

NOTICE Errata re Exhibits in support of 82 Defendants' Motion for Summary Judgment by TIMOTHY F GEITHNER, KATHLEEN SEBELIUS, HILDA L SOLIS, UNITED STATES DEPARTMENT OF HEALTH AND HUMAN SERVICES, UNITED STATES DEPARTMENT OF LABOR, UNITED STATES DEPARTMENT OF THE TREASURY (Attachments: # 1 Table of Exhibits, # 2 Exhibit 1, # 3 Exhibit 2, # 4 Exhibit 3, # 5 Exhibit 4, # 6 Exhibit 5, # 7 Exhibit 6, # 8 Exhibit 7, # 9 Exhibit 8, # 10 Exhibit 9, # 11 Exhibit 10, # 12 Exhibit 11, # 13 Exhibit 12, # 14 Exhibit 13, # 15 Exhibit 14, # 16 Exhibit 15, # 17 Exhibit 16, # 18 Exhibit 17, # 19 Exhibit 18, # 20 Exhibit 19, # 21 Exhibit 20, # 22 Exhibit 21, # 23 Exhibit 22, # 24 Exhibit 23, # 25 Exhibit 24, # 26 Exhibit 25, # 27 Exhibit 26, # 28 Exhibit 27, # 29 Exhibit 28, # 30 Exhibit 29, # 31 Exhibit 30, # 32 Exhibit 31, # 33 Exhibit 32, # 34 Exhibit 33, # 35 Exhibit 34, # 36 Exhibit 35, # 37 Exhibit 36, # 38 Exhibit 37, # 39 Exhibit 38, # 40 Exhibit 39, # 41 Exhibit 40, # 42 Exhibit 41, # 43 Exhibit 42, # 44 Exhibit 43) (BECKENHAUER, ERIC)

STATE OF FLORIDA et al v. UNITED STATES DEPARTMENT OF HEALTH AND HUMAN SERVICES et al

Doc. 83 Att. 26

Exhibit 25

Dockets.Justia.com

M a r k e t Wat c h

M a r k e t Wa t c h
Community Rating And Sustainable Individual Health Insurance Markets In New Jersey
Trends in New Jersey's Individual Health Coverage Program reveal troubled times for the program.
by Alan C. Monheit, Joel C. Cantor, Margaret Koller, and Kimberley S. Fox
ABSTRACT: The New Jersey Individual Health Coverage Program (IHCP) was implemented in 1993; key provisions included pure community rating and guaranteed issue/renewal of coverage. Despite positive early evaluations, the IHCP appears to be heading for collapse. Using unique administrative and survey data, we examined trends in IHCP enrollment and premiums. We found the stability of the IHCP to be fragile in light of improving opportunities for job-related health insurance. We also found that it is retaining high-risk enrollees. Institutional realities and the difficulty of identifying a control group preclude attributing causality to the plan's pure community rating and open enrollment provisions.
n a n e f f o r t to stabilize a financially p r e c a r i o u s i n d iv i d u a l h e a l t h i n s u r a n c e market, assure access to affordable coverage regardless of health risk, and stimulate premium competition among insurers, New Jersey implemented the Ind iv idual Health Coverage Program (IHCP) in August 1993.1 The IHCP adopted a number of sweeping reg ulatory provisions, most prominently guaranteed issue and renewal of health plans, pure community rating within specific plans, restrictions on waiting periods for preexisting health conditions, and the requirement that carriers maintain a minimum loss ratio of 75 percent. In addition, all carriers selling health insurance in New Jersey were required to participate in the IHCP, either by selling policies to meet an enrollment target or by not selling and paying a share of the losses incurred by

I

other carriers.2 Finally, IHCP enrollees were given a broad choice of health plans with standardized benefits, including traditional indemnity plans with varying deductible and c o i n s u r a n c e p r ov i s i o n s a n d h e a l th ma i n t e n a n c e o r ga n i z a t i o n ( H M O) c o v e r a g e w i t h differing copayments. n Previous research and the current context. In 1999/2000 evaluations, Katherine Swartz and Deborah Garnick concluded that initially the IHCP achieved its stated goals, and they found no evidence of adverse selection by enrollees.3 While they noted that a decline in IHCP enrollment began in 1996 and was accompanied by rising premiums, they attributed the latter to perverse incentives inherent in the IHCP loss assessment system rather than adverse selection. In contrast, the IHCP's current situation

Alan Monheit ([e-mail]) is a professor in the Health Systems and Policy Division, School of Public Health, University of Medicine and Dentistry of Ne w Jersey, in Piscataway. Joel Cantor directs the Center for State Health Policy, Rutgers University, in Ne w Brunswick, where Margaret Koller is associate director for planning and program initiatives and Kimberley Fox, a senior policy analyst.

H E A LT H A F F A I R S ~ Vo l u m e 2 3 , N u m b e r 4
DOI 10.1377/hlthaff.23.4.167 ©2004 Project HOPE­The People-to-People Health Foundation, Inc.

167

Health

Tr a c k i n g

points to a market that is heading for collapse. Enrollment has declined from a peak of 186,130 lives at the end of 1995 to 84,968 at the end of 2001. In addition, premiums have increased t wo - to th r e e f o l d ab ove th e i r e a r ly l eve l s . These changes have raised concer ns as to whether a comprehensive reg ulator y effort such as the IHCP can yield a sustainable health insurance market. In this paper we examine trends in IHCP e nrol lme nt and premiums and consider whether the IHCP regulatory provisions may be associated with some unintended consequences for insurance market stability and access to coverage. We also consider whether the institutional and economic realities that confronted the IHCP contributed to these trends. We find evidence suggesting that the stability of the IHCP market may be sensitive to changing economic circumstances and opportunities to obtain employment-related health ins u r a n c e . Th e t r e n d s t h a t we o b s e r v e a l s o suggest that the IHCP may be retaining adverse health risks. n Community rating, adverse selection, and market stability. IHCP provisions such as pure community rating and guaranteed issue were intended to expand access to coverage by limiting insurers' risk selection and med ical under w riting practices. However, s u c h r e q u i r e m e n t s c a n a l s o h ave a d e s t ab i l i z i n g e f f e c t o n s p e c i f i c h e a l th p l a n s within an insurance market and, ultimately, upon the entire market. Since pure community rating imposes the same premium on low- and high-risk people, th e p r e m i u ms o f l o w r i sks e x c e e d th e i r actuarially fair level, while those of high risks are lower than their fair level. A sustainable market equilibrium may be tenuous under such a requirement.4 In a market with choice among community-rated health plans, low risks will seek entry to cheaper, more restrictive health plans that are unattractive to high risks, leaving the latter in the more generous and expensive plans. As low-risk people leave the more generous health plans, higher risks will dominate such plans, and their premiums will rise.

Such a "separating market equilibrium" can be sustained only if low risks find the more restrictive plans to be of value and remain in the market. Should low-risk people defect from the market, total plan enrollment will decline. The market risk profile will become increasingly dominated by high risks, and a marketwide adverse-selection death spiral may ensue as cycles of rising premiums spur further defections of lower-risk enrollees. As we discuss below, the decline in IHCP enrollment and rise in premiums has been accompanied by retention of potentially adverse health risks. However, establishing a causal link to the presence of pure community rating remains difficult.

Data And Methods
Data on IHCP aggregate enrollment, planspecific enrollment, and premiums were obt a i n e d f r o m a d m i n i s t r a t iv e r e c o r d s o f t h e IHCP board.5 For our analysis, we use premiums for single coverage, the predominant form of IHCP contracts. Our tabulations are based on enrollment data from the first quarter of 1994 to the fourth quarter of 2001 for representative IHCP plans: Plans B, C, and D (all indemnity) and all HMO plans. Trends in IHCP plan premiums are based upon data from March 1996 to Decembe r 2001. For each of the fol low ing plan types--Plans B and C (40 percent and 30 percent coinsurance, respectively, and both with $1,000 deductibles); Plan D (20 percent coinsurance and $500 deductible, the most generous plan offered based on these provisions); and an HMO plan ($15 copayment)--we constructed a plan-specific composite premium. This measure is based upon premiums for each plan from the top four carriers (based on enr o l l m e n t) w e i g h t e d b y t h e i r e n r o l l m e n t shares.6 We also compare trends in the composite premiums to those for selected small carriers. We supplement these tabulations with household data from the Current Population Survey (CPS) on the health insurance status of New Jersey residents, published data on residents' employment, data on premiums for employment-based coverage in New Jersey from

168

J u l y /A u g u s t 2 0 0 4

M a r k e t Wat c h

the Medical Expenditure Panel Survey--Insurance Component (MEPS-IC), and data on premiums in the New Jersey small-employer group market. Finally, we include data from the 2001 New Jersey Family Household Survey (NJFHS), a statewide probability sample of 2,265 families conducted by the Center for State Health Policy, Rutgers University, and its 2002 supplement of 601 families with IHCP subscribers.

Findings
n Trends in IHCP enrollment and premiums. The IHCP initially experienced a sharp increase in enrollment, from 51,648 lives in the first quarter of 1994 to 186,130 lives in the fourth quarter of 1995. However, since then IHCP enrollment has declined dramatically (Exhibit 1). Enrollment declined among all three indemnity plans that we track, with the sharpest decline in Plan D, the most generous indemnity plan (Exhibit 2). In contrast, after an initial increase and subsequent decline, enrollm e n t i n m o r e r e s t r i c t iv e H M O c o v e r a g e remained relatively stable for much of 2000

and 2001.7 HMO enrollment also increased from roughly a third to more than ten times Plan D enrollment over our study period. The shift in enrollment was also accompanied by rising premiums for all IHCP plans sampled (Exhibit 3). Plan D displayed the most pronounced increase in composite monthly premiums, rising by more than 3.5 times its initial level. Although at any point in our series, composite premiums for the other indemnity plans were considerably lower than those for Plan D, the other plans displayed simi l a r g r o w th . In c o n t r a s t , th e g r o w th i n m o n t h l y p r e m i u m s f o r t h e r e p r e s e n t a t iv e HMO plan ($15 copay) was comparatively smaller. In fact, HMO premiums fell from 70 percent to 38 percent of Plan D premiums. Th e s e t r e n d s a p p e a r c o n s i s t e n t w i t h a ma r k e t w i d e ad ve r s e - s e l e c t i o n d e a th s p i r a l spurred by open enrollment and pure community rating. However, such causality is difficult to identify, because of the presence of a number of additional factors, including pricing inc e n t iv e s i n h e r e n t i n t h e I H C P r e g u l a t o r y structure, institutional changes in New Jersey's regulated health insurance products, and

EXHIBIT 1 Enrollment In The New Jersey Individual Health Coverage Program (IHCP), First Quarter 1994 Through Fourth Quarter 2001

H E A LT H A F F A I R S ~ Vo l u m e 2 3 , N u m b e r 4

169

Health

Tr a c k i n g

EXHIBIT 2 Enrollment In The New Jersey Individual Health Coverage Program (IHCP), By Plan Type, First Quarter 1994 Through Fourth Quarter 2001

improved economic circumstances. n Potential factors contributing to IHCP market instability. Carrier loss assessment mechanism. As Swartz and Garnick noted, the loss assessment mechanism (which allowed reimbursement for payouts in excess of 75 percent

of net premium income) encouraged small carriers to sell coverage in the IHCP and to charge very low premiums during their initial years, to expand their market share.8 However, this strategy was not successful. By charging low premiums, these small carriers incurred larger-

EXHIBIT 3 Trends In Premiums For Selected New Jersey Individual Health Coverage Program (IHCP) Plans, By Plan Type, First Quarter 1996 Through Fourth Quarter 2001

170

J u l y /A u g u s t 2 0 0 4

M a r k e t Wat c h

than-expected losses, which led them to raise premiums and, later, led several small carriers to defect from the market. This contributed a certain degree of " market chaos," as many enrollees shifted carriers or dropped coverage. Swartz and Garnick have argued that the rise in IHCP premiums between 1995 and early 1998 ref lected the behavior of these small carriers rather than adverse selection against certain carriers or the IHCP.9 We find evidence that is consistent with this assertion. Compared with the trend in the composite premium over this period, the premiums of such small carriers increased markedly. For example, between the first quarters of 1996 and 1998, Plan B premiums for Manhattan National Life Insurance increased by 415 percent; for Metropolitan Life Insurance, 86 percent; and for Time Insurance Company, 110 percent. In contrast, the composite premium for Plan B increased by only 20 percent. As small carriers raised premiums, their market shares declined precipitously from 27 percent of insured lives in 1996 to less than 1 percent by 1998. Eight small carriers accounted for roughly half of enrollment losses over this period. By 2000 only one of the carriers (Metropolitan) remained in the market. Subsequently, the state legislature changed the loss assessment mechanism to a two-year retrospective period effective 1 January 1998 and required reimbursable losses to exceed 115 percent of carriers' income. State officials believe that the latter provision greatly reduced incentives for carriers to "game" the system.10 Small carriers continued to raise premiums and lose enrollment; after 1998 several additional small carriers withdrew from the market. Access to e mployer-sponsored health insurance. Over our study period, a lower-cost substitute for IHCP coverage--employer-sponsored insurance--became more widely available. This was the result of two factors. First, New Jersey experienced sizable increases in economic activity and employment opportunities, which provided increased access to employer coverage. For example, between January 1994 and January 2000 seasonally adjusted resident emp l o y m e n t i n c r e a s e d f r o m 3 . 70 2 m i l l i o n t o

4.035 million.11 Correspondingly, CPS data for New Jersey reveal that enrollment in employer coverage increased from 4.7 million in 1996 to 5.3 million in 2000.12 Next, coincident with the implementation of the IHCP, New Jersey created the Small Employer Health Benefits Program (SEHBP) in January 1994, which provided small employers (those with 2­50 full-time employees) with o p e n e n r o l l m e n t i n to s t a n d a r d i z e d h e a l th plans at modified community rates (adjusted for employees' age, sex, and family status and for business location). The SEHBP ex perie nced a rapid increase in e nrol lme nt, f rom 694,312 in the fourth quarter of 1994 to a peak of 937,784 in the third quarter of 1999 (declining to 884,104 in the third quarter of 2001). Finally, IHCP data also indicate a decline in contracts issued to employed relative to nonemployed subscribers, from just over 2:1 in June 1994 to just over 1:1 in January 2001, consistent with the shift to employer-sponsored coverage. Exhibit 4 suggests that differential premiums provided ample incentives for IHCP enrollees to obtain employer-sponsored coverage. The data indicate that by 2000, IHCP premiums for our sampled individual coverage plans exceeded and (with the exception of HMO coverage) rose faster than employer coverage premiums. Moreover, if people perceive the cost of employer coverage as their own contribution rather than the full premium, then the differential in out-of-pocket premium costs between employer and IHCP plans is substantial. As well, modified community rating in the SEHBP may have exacerbated the impact of growing availability of employer coverage on IHCP enrollment and premiums, as the cost advantage of employer coverage would be greatest for lower-risk (that is, younger) workers.13 Finally, to the extent that insurers were unable to vigorously enforce the requirement that self-employed "groups of one" obtain coverage from the IHCP rather than the SEHBP, th e p r e m i u m d i f f e r e n t i a l s b e t we e n th e s e sources of coverage provided such people with clear incentives to defect from the IHCP.

H E A LT H A F F A I R S ~ Vo l u m e 2 3 , N u m b e r 4

171

Health

Tr a c k i n g

EXHIBIT 4 Annual Premiums For Single Coverage, Employer-Sponsored Insurance And New Jersey IHCP Plans, 1996 And 2000
Plan
Employer-sponsored insurance Total premium Employee contribution IHCP plans Indemnity Plan B Indemnity Plan C Indemnity Plan D HMO plan

1996 ($)
2,354 263 1,792 2,063 4,245 2,702

2000 ($)
2,911 486 3,797 5,254 10,231 4,001

Percent change
23.7 84.8 111.9 154.7 141.0 48.1

SOURCES: Premiums for employer-sponsored insurance were obtained from the Medical Expenditure Panel Survey--Insurance Component (MEPS-IC); premiums for Individual Health Coverage Program (IHCP) plans were obtained from administrative records of the New Jersey IHCP Board. See text for descriptions of payment provisions of the specific IHCP plans used in the comparisions. NOTE: HMO is health maintenance organization.

Elimination of the Health Access subsidy program. After the beginning of 1996, New Jersey eliminated Health Access, the state's subsidy program for IHCP enrollees with incomes below 250 percent of the federal poverty level. The program began in May 1995; by March 1996 it subsidized as many as 20,000 enrollees. Fiscal constraints ended new enrollments in the program after 31 December 1995. This precluded replacement of low-income enrollees whose circumstances no longer warranted use of the access program.14 Implementation of the Ne w Jersey SCHIP program. As part of the State Children's Health Insurance Program (SCHIP), the state enacted NJ KidCare in December 1997 for children under age nineteen in families earning less than 350 perce nt of poverty. The prog ram was expanded to adults with incomes less than 200 percent of poverty through NJ FamilyCare in July 2000. It is unlikely that these programs resulted in sizable crowding out of IHCP enrollees. Enrollment in KidCare lagged initially, and much of the subseque nt e nrol lme nt growth was in plans for children with family incomes below 150 percent of poverty (who were very unlikely to have enrolled in IHCP coverage). In addition, FamilyCare was implemented toward the very end of our study period and was not a factor for most of the period

of declining IHCP enrollment.15 n Further considerations. Carrier loss assessment, growth in employer coverage, and elimination of the Health Access program may explain a substantial part of the decline in IHCP enrollment and rise in premiums before 1998. Although the robustness of the employer coverage market through 2000 may have continued to fuel IHCP trends, its effect was diminishing as economic activity slowed down. Indeed, since 1998 IHCP enrollment has continued to decline at a rate of about 3 percent per quarter, and premiums have continued to rise. Thus, it is reasonable to ask whether these continuing trends are unique to the IHCP, ref lecting an adverse-selection death spiral, or simply consistent with national trends in enrollment and premiums. Changing age composition of the IHCP. Comparisons between data from the NJFHS and published 1996 tabulations provide evidence that the IHCP has shifted markedly toward older and thus potentially higher-cost enrollees.16 In 1996, 44.6 percent of new adult enrollees to the IHCP (those ages 18­64 enrolled less than a year) were ages 45­64, with a mean age of 41.9 years; by contrast, in 2002, 66.3 percent of new adult enrollees were between those ages, and their average age rose to 48.4 years (both figures significantly different from 1996, p

172

J u l y /A u g u s t 2 0 0 4

M a r k e t Wat c h

.01). As noted below, the change in these data tional trends. for IHCP enrollees during this period also exA caveat regarding interpretation. While the ceeded that for people with employer cover- trends in IHCP enrollment and age composiage. Finally, we also find some evidence that tion may ref lect necessary conditions for an the percentage of new adult enrollees report- adverse-selection death spiral in a communitying fair or poor health rose between 1996 and rated insurance market, we cannot establish 2002 (4.3 percent compared with 8.1 percent, p that they are sufficient. This ref lects at least .10), although this increase was not statisti- two important limitations. First, the unique cally different from that for people with em- nature of our administrative and household ployer coverage. data on IHCP enrollment limits comparisons Following Swartz and Garnick, we also with potential control states that lack the compared the age composiIHCP regulatory provisions. tion of new adult IHCP enThus, we cannot in fer "Trends for New whether community rating in rollees with all New Jersey Jersey may reflect the IHCP played a causal role adults having employer covthe unique i n th e t r e n d s we o b s e r ve . erage. We found that the forcircumstances of the Next, while household data mer were older (48.4 years suggest that the age composicompared with 40.7 years, p IHCP rather than tion of new IHCP enrollees .01) and more likely to be ages national trends." may have changed over time, 45 ­ 6 4 y e a r s ( 6 6 . 3 p e r c e n t further analysis is required to compared with 39.5 percent, p .01). These differences have widened consid- discern whether such health-related characerably compared with those reported for 1996 teristics have also changed across IHCP plan (data not shown). When we compare all adult types. Thus, while the IHCP trends that we IHCP enrollees with adults having employer observe may be consistent with adverse seleccoverage, we observe that the former are older, tion and the retention of poor health risks, we more likely to be female, and more likely to be cannot definitively assign causality to pure in fair/poor health (data not shown). Thus, the community rating. IHCP may be attracting and retaining more Conclusions potentially costly enrollees. Our analysis strongly suggests that the Comparison with national trends. The decline in IHCP is in the midst of an enrollment crisis IHCP enrollment appears to be far more severe than national declines in individual coverage. that threatens its market stability and ability While IHCP enrollment declined by 41 per- to fulfill its stated goals. In assessing the New cent between 1996 and 2000, national CPS es- Jersey experience, it is important to note that timates of individual coverage indicate only a 4 enrollment in individual health insurance is percent decline.17 MEPS data between 1996 fragile to begin with, representing a small miand 1999 reveal an 18 percent decline in non- nority of all nonelderly, privately insured peogroup coverage--well below the 34 percent ple. A certain amount of churning and disendecline in IHCP enrollment during this period. rollment is also expected in this market, since Comparisons with national sur vey data on enrollment spells are typically short and freemployer-sponsored coverage also suggest quently serve to bridge spells without emthat the IHCP sustained a more pronounced ployer coverage.19 One must further recognize shift in enrollment from indemnity to HMO that the IHCP is a voluntary and unsubsidized coverage.18 Correspondingly, IHCP premiums insurance market. Critics will be quick to attribute problems rose more than employer coverage premiums in the IHCP to community rating and open end i d . Th e s e c o m p a r i s o n s s ug g e s t t h a t t h e trends for New Jersey may ref lect the unique rollment. However, other factors we have idencircumstances of the IHCP rather than na- tified may have played a key role in the decline

H E A LT H A F F A I R S ~ Vo l u m e 2 3 , N u m b e r 4

173

Health

Tr a c k i n g

in covered lives and rise in premiums. These trends exceed those observed nationally and have been accompanied by a deterioration in t h e I H C P 's r i s k c o m p o s i t i o n . W h i l e t h e IHCP's market rules may have contributed to an adverse-selection death spiral, we cannot d e f i n i t iv e l y i d e n t i f y t h i s m a r k e t d y n a m i c . More importantly, it is not clear whether the decline of this market would have continued absent the buoyant employer coverage market. In this regard, the New Jersey ex perience raises the critical policy issue of whether the individual market can serve as " private insurer of last resort" when it is inherently unstable and encounters alternative private coverage with considerably lower out-of-pocket premium costs. Market stability may be further threatened should pure community rating exacerbate disparities in coverage costs between these markets for low-risk enrollees. In this context, existing research points to the difficulty of sustaining the individual market, especially when small-group market reform is present and robust economic circumstances provide better access to lower-cost employer coverage.20 The interrelationship between the individual and employment-based markets, therefore, is an unavoidable reality confronting policymakers in the design of individual market reform. Finally, the behavior of the individual insurance market in New Jersey raises some hard choices for policymakers. Given the state's current fiscal difficulties and empirical evidence suggesting that large subsidies would be required in the face of low demand responsiveness, efforts to stabilize the IHCP through subsidized premiums would seem prohibitive.21 Alternatives such as moving to modified community rating might make individual coverage more attractive to low-risk enrollees. However, such a strategy might further segment the market to the disadvantage of higher risks, the very group whose access to coverage the IHCP is seeking to ensure. While pooling of IHCP together with the SEHBP was earlier evaluated and dismissed because of the potential for a sizable increase in SEHBP premiums, such pooling may have less of an impact now,

as IHCP enrollees now represent less than a tenth of total enrollment in the SEHBP market (compared with nearly a quarter when blending was first evaluated).22 In sum, should IHCP enrollment continue to decline and premiums continue to increase, New Jersey policymakers may face the dilemma of seeing the goals of the IHCP vitiated, plan choice diminished, and the market segmented between plans enrolling high- and low-risk people. Should this be the case, history may repeat itself as insurers experience unsustainable losses and the availability of even high-cost individual coverage diminishes. Financial support from the Commonwealth Fund and the Changes in Health Care Financing and Organization (HCFO) initiative of the Robert Wood Johnson Foundation is gratefully acknowledged. The authors are indebted to Ellen F. DeRosa, Ne w Jersey Department of Banking and Insurance ( NJDOBI), for the data used in this analysis. They are grateful to DeRosa, Wardell Sanders, and Vicki Mangiaracina of the NJDOBI for helpful discussions regarding the interpretation of trends in IHCP enrollment and premiums. They also thank Kathy Swartz and two anonymous referees for useful and insightful comments, and Piu Banerjee for excellent research assistance. NOTES
1. For a discussion of circumstances leading to the IHCP's formation, see K. Swartz and D. Garnick, "Lessons from New Jersey," Journal of Health Politics, Policy and Law 25, no. 1 (2000): 45­70. 2. The loss assessment system served to reduce the costs to carriers seeking to enter the market and to provide subsidies to sustain a market with potential adverse selection. See ibid.; and K. Swartz and D. Garnick, "Regulating Markets: Lessons from New Jersey's Individual Health Insurance Coverage Program" (Unpublished manuscript, Harvard University, October 1998). 3. Swartz and Garnick, "Lessons from New Jersey"; and K. Swartz and D. Garnick, "Can Adverse Selection Be Avoided in a Market for Individual Insurance?" Medical Care Research and Review 56, no. 3 (1999): 373­388. 4. M. Rothschild and J. Stiglitz, "Equilibrium in Competitive Insurance Markets: An Essay on the Economics of Imperfect Information," Quarterly Journal of Economics 90, no. 4 (1976): 630­649; and T. Buchmueller and J. DiNardo, "Did Community

1 74

J u l y /A u g u s t 2 0 0 4

M a r k e t Wat c h

5.

6.

7.

8.

9.

10.

11.

12.

13.

Rating Induce an Adverse Selection Death Spiral? Evidence from New York, Pennsylvania, and Connecticut," American Economic Review 92, no. 1 (2002): 280­293. For the most recent IHCP plan characteristics and monthly premiums, see "New Jersey Individual Health Coverage Program Rates," 4 March 2004, www.nj.gov/dobi/ihcrates.htm (13 April 2004). Carriers' market share changed over time, so the composite premium will include different carriers. We expect that network and referral restrictions would lead most enrollees to consider HMOs to b e l e s s ge n e r o us th a n i n d e m n i t y p l a n s . S e e Buchmueller and DiNardo, "Did Community Rating Induce an Adverse Selection Death Spiral?"; and R. Feldman and B. Dowd, "Risk Segmentation: Goal or Problem?" Journal of Health Economics 19, no. 4 (2000): 499­512. The discussion follows Swartz and Garnick, "Lessons from New Jersey"; Swartz and Garnick, "Regulating Markets"; and K. Swartz and D. Garnick, "Unintended but Predictable Outcomes of Regulations: The Case of New Jersey's Individual Health Care Program" (Unpublished manuscript, Har vard University, 12 Januar y 1998). Swartz and Garnick, "Lessons from New Jersey"; and Swartz and Garnick, "Regulating Markets." The small carriers were Manhattan National Life Insurance, Metropolitan Life Insurance, National Casualty, Protective Life Insurance, Time Insurance, TMG Life Insurance, Travelers Insurance, and Washington National Insurance. Based on discussions with representatives from the New Jersey Department of Banking and Insurance, Fall 2003. New Jersey Department of Labor, "New Jersey Economic Indicators: Historical Data Series, 1976­2001, Resident Employment, Series 2," www.wnjpin.net/OneStopCareerCenter/Labor MarketInformation/lmi07/series2.htm (13 April 2004). Data are from P. Fronstin, Sources of Health Insurance and Characteristics of the Uninsured: An Analysis of the March 2001 Current Population Survey, EBRI Issue Brief no. 240 (Washington: Employee Benefit Research Institute, December 2001). Following Fronstin, we have adjusted the estimates to ref lect the inclusion of an insurance status verification question in the CPS in 1998. See New Jersey Small Employer Health Benefits Program Board, "The Effects on the Individual and Small Employer Health Coverage Markets of Permitting Individuals to Purchase Small Employer Health Benefits Plan: A Report to the New

14. 15.

16. 17. 18.

19.

20.

21.

22.

Jersey Legislature" (Unpublished report, 25 September 1996), for a comparison of monthly prem i u ms f o r a s ma l l e m p l o y e r g r o u p i n s u r e d through specific IHCP and SEHBP plans. Swartz and Garnick, "Regulating Markets." S. Silow-Carroll et al., Assessing State Strategies for Health Coverage Expansions: Case Studies of Oregon, Rhode Island, New Jersey, and Georgia (New York: Commonwealth Fund, November 2002). Swartz and Garnick, "Lessons from New Jersey." Fronstin, Sources of Health Insurance. Data are from Henry J. Kaiser Family Foundation/Health Research and Educational Trust, Employer Health Benefits: 2002 Annual Survey (Washingt o n : K a i s e r Fa m i l y F o u n d a t i o n , S e p t e m b e r 2002). T. McBride et al., "The Dynamics of Individual Insurance Coverage in the U.S." (Paper presented at the Twentieth Annual Research Meeting of AcademyHealth, Nashville, Tennessee, 27­29 June 2003). Median spell length for people with individual coverage is five months. On average, older people, especially those nearing retirement, remain in the market longer than younger people. See M. Hall, "An Evaluation of New York's Reform Law," Journal of Health Politics, Policy and Law 24 no. 1 (2000): 71­100; K. Thorpe, "Who Purchases Individual Insurance? A Comparision of New York State, Regional, and National Patterns, 1994­1997" (Unpublished manuscript, Tulane University Medical Center, March 1999); and D. Chollet, "Changes in the Individual Insurance Market" (Paper presented at the Annual Research Meeting of AcademyHealth, Washington D.C., 23 June 2002). M.S. Marquis and S.H. Long, "Worker Demand for Health Insurance in the Non-Group Market," Journal of Health Economics 14, no. 1 (1994): 47­63. See New Jersey SEHBP Board, "The Effects on the Individual and Small Employer Health Coverage Markets." Figures in the text are the authors' computations; details are available on request by e-mailing Alan Monheit, monheiac@ umdnj.edu.

H E A LT H A F F A I R S ~ Vo l u m e 2 3 , N u m b e r 4

175