7

Periodically, the tensions and contradictionsemanating from the big, marvelously innovative, highly inequitable, and hugelyexpensive U.S. health care system force ageneral reassessment of the way this country finances and delivers health care forits citizens. One of these periods appears tobe approaching—although, as Ted Marmorpointed out over a decade ago, coalitionspreferring the status quo almost always prevent these reassessments from resultingin more than incremental change.1 Today,over 46 million people are uninsured, families with health insurance fear that theymay lose it, firms with household namesseek ways to extricate themselves from providing health insurance for their employees, and the new Deficit ReductionAct of 2005 permits doctors and hospitalsto deny services to Medicaid recipients whocannot meet required co-payments anddeductibles. In an early 2006 article, TheEconomist asserts that the “world’s biggest

and most expensive health-care system isbeginning to fall apart;” it also suggests that health reform is “one of the most complicated challenges facing America’seconomy.”2 Why has health care become amajor challenge to the U.S. economy and toeconomic policy makers? At least threedevelopments explain the growing impor-tance of health reform as an economic issue.

Clearly, the health care sector is now verylarge and touches most aspects of the U.S.and New England economies. In 2004,spending on medical care amounted to 16percent of U.S. nominal GDP—more thanconsumers spent on food, clothing, andenergy in total and about equal to all busi-ness investment in plant and equipment.Further, health care’s share of non-farmemployment is now 9 percent and grow-ing—that’s roughly akin to manufacturing’sshrinking share of the workforce. In NewEngland, health care looms even larger,

Jane S. LittleVice President and Economist

Federal Reserve Bank of Boston

Teresa M. FoyPolicy Analyst

Federal Reserve Bank of Boston

Where There’s A Will,

There Could Be A Way

Reforming the

U.S. Health

Care System:

8

accounting for almost 12 percent of regionalemployment. In the future, this sector is almostcertain to absorb an even greater share of GDP,for, as OECD data suggest, as national incomesrise, countries generally choose to spend a grow-ing share of their income on health and health care(Figure 1).3

With health care spending projected to reach 22percent of GDP by 2025,4 it becomes increasinglyimportant that U.S. policy makers be able tomeasure health care output, prices, and productiv-ity accurately—no easy task. Currently, the mostfamiliar measure of health care costs is probablythe medical care CPI, which measures inflation inconsumers’ out-of-pocket costs for medical care, afraction of total health care spending. For a vari-ety of reasons, the medical CPI has been increas-ing a lot faster than the core CPI, helping to boostbroad measures of inflation and labor costs aswell. In addition, rapid medical cost inflation hascontributed to a widespread impression that pro-ductivity in the U.S. health care sector may berather low. By contrast, a growing body of recentresearch provides evidence of significant produc-tivity gains in health care for patients sufferingfrom specific widespread problems like cataracts,depression, and heart attacks. But do these find-ings apply to the whole health care sector? Indeed,international data indicate that the United Statesspends far more per person on health care thanwould be expected given its per capita income(Figure 1),5 while data on expenditures and out-comes suggest that this country’s extra spendingmay not be particularly productive (Figure 2).6

A second reason for economists’ concern aboutthe health care system reflects its possibly distorting effect on the operation of the U.S. labormarket. Compared with other OECD countries,employment-based insurance plays an unusuallylarge role in the U.S. health care system, where itfinances about 40 percent of U.S. health carespending. But of course, not all employers offerhealth insurance. And over the decade to 2003,the share of private-sector workers actually participating in employer-provided medical plansfell from 63 percent to 45 percent, in part reflecting

10,000 20,000 30,000 40,000 50,000 60,0000

1000

2000

3000

4000

5000

6000

USA

AUT

LUX

FIN JPN

IRL

CZE

KOR SVK

MEX POL

HUN

NLD

BEL

PRT

DNK

ITA

NZI ESP GBR

SWE

GRC

DEU ISL

CHE

CAN

FR

NOR

Health Care Spending Per Capita

vs. GDP Per Capita, 2002

Health care spendingper capita

GDP per capita

TRK

AUS

Source: Organisation for Economic Co-operation and Development. Data are expressed in purchasing-power-parity U.S. dollars.

Figure 1

International Comparisons

0

Public and private health care spendingas a percent of GDP

Healthy life expectancy at birth (years )

Source: Organisation for Economic Co-operation and Development.

60 65 70 75 804

6

8

10

12

14

16

USA

AUT

LUX FIN

JPN IRL CZE

KOR SVK MEX

POL

HUN

NLD

BEL

PRT DNK

ITANZI

ESP GBR

SWE

GR

DEU

ISL

CHE

CAN FR

NOR

AUS

Health Care Spending

vs. Healthy Life Expectancy, 2002

Figure 2

International Comparisons

9

workforce shifts from full- to part-time and unionto non-union status. In addition, a smaller share ofworkers who are offered health insurance nowchoose to take it—most likely because a growingfraction of employers are requiring workers elect-ing this benefit to contribute more toward itscost.7 Another factor may be the increase in two-worker households.

Are these employment-based financing arrange-ments affecting the supply or demand for labor inthis country? Are they influencing the structure ofemployment, encouraging a shift toward the useof temporary or contract labor? Does our healthcare system distort our labor market and reduceits flexibility? The answers to these questions concern policy makers.

Turning, finally, to fiscal issues, the tax-financedshare of health care is estimated to have reachedabout 60 percent in 1999,8 up from 55 percent in1990 and a higher percentage than most peopleexpect. The large and rising share of publicly fund-ed health care puts pressure on federal and statebudgets, limiting those governments’ non-healthpolicy options. According to the Social Security andMedicare Trustees Report of 2005, total Medicareexpenditures will rise as a share of GDP from 2.6percent currently to 13.6 percent in 2079. If so,Medicare expenditures will exceed those for SocialSecurity in 2024 and represent twice the cost ofSocial Security in 2079 (Figure 3). Moreover, at thestate level, many governments have taken steps toexpand the scope of Medicaid in order to extendhealth insurance coverage to particularly vulnera-ble groups, such as children. This trend has placedan increased burden on state budgets (Figure 4).How the nation and individual states address theseimbalances—through increased taxes, reducedbenefits, or increased borrowing—will affect U.S.interest rates, private savings and investment, andinternational capital flows.

Prompted by its interest in these issues, theFederal Reserve Bank of Boston brought togethereconomists, health practitioners, and policy makers to examine the topic “Wanting It All: TheChallenge of Reforming the U.S. Health Care

System” in June 2005. This essay summarizes thethemes and the consensus-based prescriptions foraction that emerged from that conference.(Please see the box on page 14 for a list of conference presenters.)

Social Security and Medicare Costs

as a Share of GDP

Percent of GDP

19702030

20502010

20702080

16

14

12

10

8

6

4

2

0

19901980

20002020

20402060

Medicare

Social Security

Figure 3

Source: Medicare Trustees Report, 2005; Social Security Trustees Report, 2005.

forecast

Figure 4

Medicaid as a Share

of State Expenditures

Percent of state expenditures

1990 1996 19981994 2000

24

22

20

18

16

14

12

101992 20042002

Source: National Association of State Budget Officers, State Expenditure Report, 1990 - 2004.

10

Rising health care costs are the result of advancements inmedical technology, not population growth or aging.

11

Defining the Health

Care Challenge—

“The Problem with

No Obvious Solution”

This country’s health care goals include broad,secure access to “appropriate,” high-quality carebased on active discovery and innovation at an“acceptable” (aye, there’s the rub) cost to the ulti-mate payer. All industrial countries share thesegoals, although, as Kieke Okma points out, notnecessarily the weights they assign to them. Forexample, Europeans tend to put more weight onaccess to care than do Americans, who seem toput consumer choice at the top of the list andaccess toward the bottom. But in the end, in“wanting it all,” every country struggles with theinherent conflicts between these goals. In particu-lar, since all countries adopt new medical technologies as they become available, all strug-gle to contain the rapid pace of growth in healthcare costs. And most could put more emphasis onprevention and achieving good health.9

These inherent conflicts reflect the essential valueof health care to many patients/consumers. Theyalso reflect, as William Nordhaus points out, society’s embrace of “specific egalitarianism”10 aswell as society’s reluctance to ration health careby price or even by regulation. Obviously, these attitudes do not accord well with an equally widespread lack of political will to pay for otherpeople’s health care. And these inconsistenciesare only exacerbated by information asymme-tries; by the absence of cost consciousness amongconsumers; and by limited competition amongproviders and health plans. Finally, Richard Frankand others raise a host of behavioral issues thatfurther compound the situation, issues thatinclude patient-doctor inertia, rules of thumb,excessive optimism, and myopia regarding theneed to save for medical emergencies. Theseinherent conflicts lead David Cutler to call healthreform “a hard problem;” Nordhaus to call it “avery hard problem;” and Henry Aaron to call it“the problem that won’t go away.”

Measuring and Valuing

Health Care

David Cutler and William Nordhaus both demonstrate that improvements in public healthand medical care have added enormously to ourstandard of living over the past one hundredyears. Nordhaus even concludes that the value ofthe gains stemming from improvements in healthstatus equals the value of all other gains in consumption over the past quarter century. Notsurprisingly, then, as physicians have becomemore effective and societies have grown wealthi-er, people have chosen to spend a higher share oftheir incomes on health care—they value whatdoctors can do for them. In addition, as Cutlerpoints out, health care turns out to be highly priceelastic; properly measured, some quality-adjustedhealth care prices are actually falling, and peoplespend more in response. Further, as Cutler alsodemonstrates, cost-benefit analysis of specificinterventions, like treatment for heart attack,finds that such interventions are clearly “worth”their cost, based on common assumptions regard-ing the economic value of the additional years oflife resulting from the intervention. For example,$30,000 in expenditures for a 45-year-old cardiacpatient leads on average to three years’ longerlife. Since three years’ longer life has a discountedpresent value of $120,000 by common estimates,the return on the investment is 4 to 1.

But, as Cutler also notes, the fact that much oftoday’s health care is highly valued (particularlyby individual doctors, patients, and their familiesconfronting specific medical crises) does not necessarily make it affordable (particularly to tax-payers, to whom hypothetical patients are merestatistics). Nor does this high valuation mean thatall health care dollars are well spent. Cutler suggests that at least 20 percent of health carespending is wasted, while Wennberg, Skinner, andFisher (who find that Medicare spends half asmuch per patient in Minnesota as in Miami withequally good results) conclude that the waste inMedicare is closer to 30 percent.11 But underspending also contributes to the inefficiency ofthe U.S. health care system. For example, too little

12

is spent on prevention and chronic disease man-agement—for the insured as well as for the uninsured. And the system often does a poor jobof coordinating different aspects or phases of apatient’s care, such as the transition from acute tochronic care, or the transfer of records from onehospital or doctor to another.

Improving Efficiency:

Consumer Incentives,

Provider Incentives,

and Technology

Prescriptions for reducing the inefficiencies plaguingthe U.S. health care system include making con-sumers more sensitive to the costs of their medicalcare, making providers more responsible for healthcare outcomes, and encouraging better use of infor-mation and communication technology throughoutthe health care system. To start with consumerawareness, most analysts, including those at theBoston Fed conference, agree that the tax subsidyfor employer-provided health insurance, which cur-rently cuts federal tax revenues by about $200 bil-lion per year,12 reduces cost consciousness andshould be eliminated for the non-poor.13

A second, newly popular approach to encouragingpatients to be more cost conscious involvesincreasing the availability of low-cost insurancewith high deductibles and high co-payments,combined with Health Savings Accounts (HSAs)or Health Reimbursement Arrangements.Together, these elements make up “consumerdriven health care” (CDHC), which, to be effective,requires that health care cost information bewidely available and of significance to patientsmaking health care decisions. While several conference participants, including Alain Enthoven,Mark Pauly, Gene Steuerle, and Stuart Altman,see some merit in aspects of consumer drivenhealth care,14 many attendees are concerned thatCDHC will encourage underutilization of preven-tive care, particularly by low-income individualsunable to afford the high co-payments anddeductibles. And such concerns appear to be warranted, judging by a recent study that finds

that, for reasons of cost, 35 percent of individualswith CDHC plans skipped or delayed health care, compared with 17 percent of persons with comprehensive health plans.15 In addition, confer-ence participants including Richard Frank, RobertGalvin, Sherry Glied, and David Meltzer point tothe general absence of the information regardinghealth care costs that would be required to makeCDHC work; the reluctance of doctors andpatients to discuss matters of cost; the impor-tance of advice from family and friends; and theprominence of inertia in determining patientchoice of health care providers.

As for motivating providers to improve efficiency,many conference participants see considerablepromise in “pay for performance,” a reimburse-ment system that rewards providers for good outcomes and for following prescribed protocolsfor vaccinations and other preventive care—that is, for doing what they ought to do. A smallergroup, led by Alain Enthoven, advocates combin-ing “pay for performance” with support for integrated delivery systems like Kaiser Permanentein California and Harvard Vanguard inMassachusetts. Such systems are built around acore multi-specialty group practice that has a sig-nificant share of its revenues based on per capitapre-payment. Further, members of the practice areencouraged to adhere to up-to-date clinical stan-dards developed by the team.16 According toEnthoven, integrated delivery systems, alsoknown as “delivery system HMOs,” should besharply distinguished from “carrier HMOs,” rather inclusive networks of unaffiliated physi-cians generally working under fee-for-servicearrangements. In choosing to receive care from anintegrated delivery system, an individual is optingto hire a general contractor, to use a Karen Davismetaphor, rather than to deal with the plumber,the roofer, the painter, and the candlestick makerindividually. Obviously, the individual’s care is likely to be better coordinated, and, between capitation and patient inertia regarding choice of doctor, the system’s managers have consider-able incentive to provide good preventive care and disease management, using non-physicianproviders whenever appropriate.

13

Many conference participants see considerable promise in “pay for performance,” a reimbursement

system that rewards providers for good outcomes and for following prescribed protocols.

But while Kaiser, Mayo, and Harvard Vanguardare widely acknowledged to provide great care,integrated delivery systems are not popular out-side of California and, to a lesser extent,Massachusetts and Connecticut. Why not?Chernew and Glied suggest that people fear pre-committing to a narrow set of doctors beforeknowing what their medical needs may be andthat such systems may require too much travel.But in their eyes, the major deterrent is likely to beresistance to switching doctors, a resistance thathas fostered the spread of preferred providerorganizations (PPOs) and other almost universal-ly inclusive networks of independent providers.Richard Frank and David Meltzer also raise somebehavioral concerns about the efficacy of practiceguidelines and “pay for performance,” noting thatphysicians tend to be over optimistic, over confi-

dent, and reluctant (oruncertain how) to changetheir ways. In the end, whilemost observers view inte-grated delivery systems and“pay for performance” aslikely to improve the efficien-cy of the U.S. health caresystem, no one claims thatthese options will keephealth care expendituresfrom rising as a share ofincome. And, as Chernewpoints out, the more efficientthe system becomes, theharder it is to avoid thepainful trade-offs betweenquality and access.

Turning to technology, whilealmost everyone agrees thatadvancing medical technolo-gy is the primary driver of ris-ing health care costs—“it’sthe technology, stupid,” toquote Mark Pauly—manyconference participants re-main convinced that betteruse of information and communications technology

holds great promise for improving the efficiency ofthe complex, disjointed U.S. health care system.According to Jim Mongan and David Brailer, forexample, electronic medical records will do farmore than cut paperwork and reduce error; moreimportant, they will also drive medicine towardevidence-based practice. Galvin, Brailer, Davis,and Mongan all see huge potential in a nationaleffort to identify and spread best practices and to develop and publicize quality measures.Nevertheless, Pauly and others suspect that evenwith better consumer and provider incentives andimproved information and communications tech-nology, U.S. policy makers will likely need to find agraceful, politically acceptable way to slow theadoption of new or unneeded medical technologyfor the insured middle class.

14

Henry AaronSenior Fellow The Brookings Institution

Stuart AltmanProfessor Brandeis University

David Brailer, M.D.National Coordinator for Health Information Technology U.S. Department of Health and Human Services

Michael ChernewProfessorUniversity of Michigan

David CutlerProfessorHarvard University

Karen DavisPresidentThe Commonwealth Fund

Alain Enthoven Professor Stanford University

Judith FederDean, Georgetown Public Policy Institute Georgetown University

Richard FrankProfessorHarvard Medical School

Henry FarberProfessorPrinceton University

Robert Galvin, M.D.Director Global HealthCare General Electric Company

Sherry GliedProfessorColumbia University

Brigitte MadrianAssociate ProfessorUniversity of Pennsylvania

Theodore MarmorProfessorYale University

David Meltzer, M.D. Associate Professor University of Chicago

James Mongan, M.D.President and CEO Partners Healthcare

Joseph NewhouseProfessorHarvard Medical School

William NordhausProfessorYale University

Kieke OkmaVisiting Professor Catholic University of Leuven

Mark PaulyProfessorUniversity of Pennsylvania

C. Eugene SteuerleSenior Fellow

Urban Institute

Alan WeilExecutive Director National Academy for State Health Policy

This essay provides a summary of the views presented at the Federal Reserve Bank ofBoston’s 50th Economic Conference: “Wanting It All: The Challenge of Reforming theU.S. Health Care System, ” which was held in June of 2005. We thank all of the presenters, who are listed below, for contributing to the success of the conference.

The conference agenda and the presenters’ papers and biographies can be found at www.bos.frb.org/economic/conf/conf50/index.htm

15

Employer-Based Health

Insurance: Pros and Cons

In the United States, members of the middle classgenerally obtain their health insurance throughemployer-provided health benefits. Althoughemployment-based insurance crops up in manycountries, this arrangement has played an unusu-ally dominant role in the United States. In the1940s, U.S. employers constrained by wartimeprice controls were encouraged to compete forworkers by offering tax-subsidized health benefitsin place of higher wages; today, employer-provided benefits are the primary source of healthinsurance for the non-elderly. These employment-based arrangements cover 63 percent of the non-elderly population; by contrast, public programs like Medicaid and Medicare cover just17 percent (Figure 5). As Brigitte Madrian pointsout, the result is a highly fragmented systemwhere thousands of employers define the healthinsurance options available to their workers andwhere even Medicaid comprises 50 different stateprograms. Does this employment-based systemserve the country well?

Many conference participants, including AlainEnthoven and Henry Farber, answer no. Theydescribe the system as “hopelessly flawed” and a“terrible idea” because it leaves millions of peoplewithout access to affordable health care, bearsmost heavily on low-wage workers, and makesthe U.S. labor market less flexible and dynamic. Tostart with this last point, just 60 percent of U.S.employers offer health insurance to any part oftheir workforce, and that share has been decliningin recent years as health benefits have grownmore costly. As a result, Madrian and others findthat worker demand for affordable health insur-ance and employer efforts to minimize the cost ofoffering this benefit distort labor market deci-sions, reducing labor market flexibility and workerproductivity. On the supply side, the availability of affordable health insurance significantlyaffects individual decisions regarding where towork or whether to work at all. Further, becauseemployer-provided health insurance is notportable, insurance contracts exclude pre-existing

conditions, and people hate changing their doctors, the employer-based system tends to discourage labor mobility, producing a phenome-non known as “job lock”17—even “wedlock” onoccasion. More importantly perhaps, on thedemand side, employers face an incentive to sub-stitute part-time or temporary workers for full-time workers in order to avoid health insur-ance costs. Similarly, firms may ask existing full-time staff, who already have health benefits,to work more hours instead of hiring more full-time workers, who will add to insurance costs.Given the evidence that workers do in fact pay fortheir health benefits through lower wages as economic theory would suggest, such employerefforts to minimize health insurance costs mayseem puzzling. But it is not clear that the wage-benefit trade-off is either immediate orone-for-one. For example, as Joseph Newhousepoints out, minimum wage laws limit employers’practical ability to shift big increases in insurancecosts to low-wage workers. Nor is it easy to askcurrent workers to pay for big increases in the costof retiree insurance, especially since, as Farbernotes, mature firms like GM now have more pen-sioners than active employees.

Figure 5

No health insurance

Source: Employee Benefit Research Institute, 2003.

Purchase ownhealth insurance

Covered by insurance offered by employerof a family member 30.8%

32.2%

Covered by insurance offered by employer

Health Insurance Coverage

of the Non-Elderly

Medicaid12.8%

17.7%

Other public4%

6.6%

16

Just 60 percent of U.S. employers offer health insurance to anypart of their workforce, and that share has been declining inrecent years as health benefits have grown more costly.

17

In addition, Enthoven, Farber, and Bob Galvinagree that many employers are ill-equipped topurchase health insurance for their workers. Few small employers have a good understandingof health care issues, and employer/worker interests may not coincide. For example, whileemployers clearly have an interest in attractinghealthy, productive workers, management’s interest in their workers’ long-term health mayhave declined in recent years as average jobtenures have fallen and lifetime employment hasvirtually disappeared.

On the other hand, as Altman, Pauly, and Galvinargue, large firms with good benefits departmentsdeliver very responsive health care to their work-ers in a very efficient manner. These firms havetaken the lead in promoting fitness and wellnessprograms, in encouraging “pay for performance,”and in developing accessible information onprovider quality and costs. Further, as Galvinemphasizes, in an employer-linked system, deci-sions regarding the use of new technologies aremarket-based. Without these market signals, howwould the nation determine how much to invest indesirable medical innovation? Would a single-payer system with a “politically acceptable” global budget do as well?

Fiscal Pressures

Even now, the federal government’s existingresponsibilities for health care are projected tocreate extraordinary fiscal—and political—pres-sures in the decades ahead. Although political andmedia attention has so far focused primarily onthe need to address the Social Security “crisis”approaching with the retirement of the BabyBoom generation, the government’s future com-mitments under the Medicare and Medicaid programs loom considerably larger, as HenryAaron, Stuart Altman, and others emphasize.

To draw the comparison more precisely, the baseline, or intermediate, estimate from theCongressional Budget Office (CBO) projects thatfederal spending for Social Security will rise from4.2 percent of GDP in 2005 to 6.4 percent in 2050.

By contrast, in the intermediate case, federalspending for Medicare and Medicaid, also 4.2 percent of GDP today, is projected to reach 12.6percent of national output by mid century (Figure6). Unfortunately, however, the CBO’s intermedi-ate projection assumes, as do the Medicaretrustees, that Medicare and Medicaid spendingper enrollee will exceed per capita GDP growth byjust 1 percentage point per year—an unrealisticassumption judging by U.S. history and by international trends. As the CBO points out,Medicare-Medicaid spending (and health carespending more generally) has in fact grown anaverage of 2.5 percentage points faster than per capita GDP since 1970. Again, this gap largely reflects technological improvements, notpopulation aging. If these trends continue,Medicare-Medicaid spending will account for 22percent of GDP in 2050—almost 18 percentagepoints more than currently.18 Further, as HenryAaron points out, because the private and publicsectors share responsibility for health care spending in this country, at current trends, healthcare will claim about half of all U.S. income and allof the increase in economic output by mid century. Valuable as health care is, is this outcome realistic?

Projections of

Federal Spending

Percent of GDP

19702030

20502010

40

35

30

25

20

15

10

5

0

19901980

20002020

2040

Medicare and Medicaid

Total

high spendingintermediate spending

Figure 6

forecast

Source: Congressional Budget Office, Long-Term Budget Outlook, 2005.

18

Confronted with these prospects, what will theU.S. electorate do? Among the alternatives Aaronposits, one course might be to continue, bydefault, along the current path and simply pay thebill. This option would allow increasing our non-health standard of living for a while, but, as healthcare came to claim all of the growth in economicoutput and then more, the situation could turnunsustainable—if the share of economic outputdevoted to education, research and development,and crucial infrastructure began to shrink, eco-nomic growth itself would slow. As an obvious,desirable alternative, U.S. policy makers couldredouble their efforts to make the health care system more efficient, but, as already discussed, abetter targeted system requires more spending insome areas and less in others, making the net savings likely not very large. To curb Medicarespending specifically, Congress could pass restric-tive legislation, increasing the Medicare eligibilityage to 67, for example. While this change mightencourage people to work longer, it would notsave much money because the young elderly are reasonably healthy. Congress could alsoincrease Medicare deductibles, co-payments, andpremiums,19 but, as Aaron notes, these changeswould simply shift costs to the private sector orreduce the elderly population’s access to medicalcare. While Medicare administrators could, forexample, conceivably slow the pace at which theyapprove Medicare coverage for new technologies,the Boomer Generation, as Stuart Altmanobserves, has always been a demanding, spendinglot, even in their 30s and 40s; he doubts they willpermit substandard care for the elderly (andpoor?) to reemerge as they age.

How then is the nation going to pay this medicalbill? Assuming that the current gap between thegrowth in health care costs and the growth in GDPcontinues, meeting current Medicare-Medicaidcommitments, Henry Aaron calculates, willrequire doubling payroll and income tax revenuesas a share of GDP by 2040. Even slowing theincrease in health care spending to 1 percentagepoint above per capita GDP growth would meanraising tax revenues by 6 percent of GDP by 2040.But, according to Stuart Altman, the United

States is a “tax-phobic” nation with an EleventhCommandment proscribing tax rates above 18percent to 19 percent of GDP, while JosephNewhouse notes that U.S. tax revenues haveexceeded 20 percent of GDP on just one occasionin the post World War II era.

The options are limited—to us collectively as asociety or individually. The more we choose toemphasize individual responsibility, the more costconscious the system will be, but the more accessfor the poor and the seriously ill becomes problem-atic. In the end, U.S. voters will have to decidewhat they are willing to spend for other people’shealth care, for, as Alan Weil points out, while people are willing to spend a lot for their ownhealth care, it is less clear what they are willing tospend on the care of others. In Henry Aaron’s view,resolving these issues will impose major stresseson the democratic polity of this country in coming decades.

Wanting It All,

Getting Much of It—

Areas of Agreement

Most of the health care experts attending theBoston Fed’s June 2005 conference appear toagree with Karen Davis, whose remarks, entitled“Getting It All,” argue that we actually do knowhow to achieve much of what we want for the U.S.health care system, including even broader access,and we should “just go ahead and do it.” Withinthis group of analysts, all tend to cite the same listof ways to increase the efficiency of the U.S.health care system and move it toward the production possibility frontier. In their view, goodsteps to take include encouraging increased use of“pay for performance” and integrated deliverysystems—with ongoing efforts to understand thebehavioral issues that might undermine theirspread and effectiveness. They also advocateadded emphasis on primary and preventive careand disease management as well as broader useof communications and information technologyto identify what works. Less obviously, perhaps,most experts also support renewed efforts toimprove consumer cost consciousness by eliminating

19

If current spending trends continue, health care will claim half of U.S.income and all of the increase in output by 2050, squeezing other

important areas such as education and infrastructure.

20

The last several years have seen privatehealth insurance premiums rise and theranks of the uninsured swell, while statebudgets have come under increased fiscal pressure, limiting expansion orcompelling cuts in existing programs.Nevertheless, some states have man-aged to summon the political will toimplement health reform strategies thatstretch health care dollars by using a portion of state money to leverage private, federal, and additional statefunds in order to expand coverage andimprove program efficiency. Initiatives ofthe New England states include usingfederal Medicaid waivers and StateChildren’s Health Insurance Program(SCHIP) waivers to expand coverage tonontraditional beneficiaries; enacting“pay or play” laws; and creating grouppurchasing arrangements.1 The pro-grams of three states are explored here.

Rhode Island. In 1993, Rhode Islandapplied for a Medicaid 1115 waiver, permitting it to conduct a demonstrationproject, RIte Care. RIte Care providescomprehensive coverage to families onthe Family Independence Program (formerly AFDC) and eligible uninsuredpregnant women, parents of children 18and younger, and children up to age 19.The program experienced a higher-than-expected take-up rate, resulting in fiscalpressure. In 2001, in an effort to reducethe cost burden without cutting eligibili-ty, the state obtained a SCHIP 1115 waiver, converting the parents of children eligible for public health cover-age from Medicaid to SCHIP and, in sodoing, receiving a higher SCHIP federalmatch for these enrollees. Additionally,Rhode Island created RIte Share, a pre-mium-assistance program for RIte Careeligible families with access to approvedemployer-sponsored health insurance.RIte Share leverages employer dollars,resulting in savings to the state for every

family enrolled in this plan instead of RIteCare, which has a full public subsidy.Under RIte Share, the state pays theemployee’s share of work-based insur-ance premiums (families above 150 percent of the federal poverty level makecontributions according to a slidingscale), the employee’s co-payments, andwraparound coverage for Medicaid benefitsnot in the employer’s health plan.

The results of RIte Share are encouraging.The Rhode Island Department of HumanServices (DHS) has determined that subsidizing a family in RIte Share plusproviding wraparound services costs thestate slightly more than half the expenseof covering the family through the RIteCare managed-care plan. Thus far, DHShas transitioned 4 percent of the RIteCare population into RIte Share, resulting in a savings of about 2 percentof the program.

Maine. Maine’s Dirigo Health Plan, created in 2003, aims to increase accessto affordable health insurance coverage,slow the growth of health care costs, andimprove the quality of care. One compo-nent, DirigoChoice, offers affordablehealth care insurance, through privatecarriers, to small-business employees,the self-employed, individuals withoutaccess to employer coverage, anddependents of these eligibles. The pro-gram pools employee, employer, state,and federal funding sources to be able todeliver reduced-cost health insurance.

To increase coverage for its low-incomepopulation, Maine obtained a federalwaiver to extend its state Medicaid program, MaineCare, to parents withincomes under 200 percent of the feder-al poverty level and to childless adultswith incomes up to 125 percent of thefederal poverty level. For working persons who are ineligible for MaineCare

and whose income is below 300 percentof the federal poverty level, the state pro-vides assistance in purchasing DirigoChoicecoverage on a sliding scale. Both the slid-ing scale and the MaineCare expansionare financed by redirecting a portion ofthe disproportionate share hospital(DSH) allocation.

In an effort to contain health care costs,the Governor’s Office of Health Policyand Finance now sets explicit targets forquality, cost, and access to health care,and establishes a budget to assist inresource allocation. In a move toincrease transparency, Maine requiresthat average charges and paymentsaccepted for commonly performed pro-cedures be posted at each provider site.In addition, Maine has expanded thereach of its certificate-of-need programto cover functions and expendituresregardless of the site of care and has putvoluntary limits on the growth of insurance premiums and health carecosts. Mandatory provider use of healthcare information technology has alsobeen proposed.

In its first nine months, DirigoChoice has enrolled over 7,000 residents andachieved $43.7 million in savings for theMaine health care system. However,enrollment has been lower than expect-ed, and a survey of enrollees finds thatonly one in four was uninsured at thetime they purchased state-subsidizedinsurance. The majority of DirigoChoiceenrollees simply switched from other private insurance.

Massachusetts. In Massachusetts, April2006 saw a bipartisan bill break politicalgridlock and extend health care coverageto the state’s 500,000 uninsured. Thenew legislation combines the individualmandate championed by conserva-tives—that all individuals should have

Health Insurance Reform in Three New England States

21

health insurance—with liberal measuressuch as large subsidies to help low-income individuals buy insurance and aproposed employer mandate—that allfirms with 11 or more employees shouldprovide health insurance. Under the legislation, the approximately 200,000uninsured Bay State residents who canafford to buy health insurance will berequired to purchase it or face tax penal-ties. To help these individuals acquirecoverage, the state will create a grouppurchasing arrangement, allowing indi-viduals and small businesses to buyinsurance as one entity.

The state’s additional uninsured comprise two groups: (1) 100,000individuals who qualify for Medicaid butare not signed up for it, and (2) 200,000individuals who do not qualify forMedicaid but are too poor to buy healthinsurance on their own. Those who qualify for Medicaid will be enrolled in it,with the cost split between the state andthe federal government. For the secondgroup, those earning up to 100 percentof the federal poverty level will receivecoverage at no cost, while those withincomes between 100 percent and 300percent of the federal poverty level willpay a portion of the premium, based on asliding scale. Funding for both groupswill come from (1) state funds set asideto pay hospitals and other providers fortreating the uninsured, as well as (2)$385 million pledged by the federal gov-ernment if the state can show it is on apath to reducing its number of unin-sured. Funding would also come fromthe proposed “pay or play” provision ofthe new law, which requires all employ-ers with 11 or more employees to providehealth care insurance or to pay an annu-al penalty of $295 per worker.

Rhode Island, Maine, and Massachusettshave implemented innovative policies to

address the rising ranks of the uninsuredand control health care costs. Whilenone of these plans to date has provideda solution to all of the challenges thehealth care system currently faces, theydo offer innovative ideas and reinvigo-rate the ongoing national debate.

– Teresa M. Foy

1. The strategies employed by states include reinsurance, high risk pools, and limited benefit plans. This section only covers a subset of the New England states’ utilization of federal waivers and other state health system reforms.

22

While Americans are willing to spend a lot for their own health care, it is less clear what they are willing to spend on the care of others.

23

tax subsidies for employer-provided health bene-fits and, to a lesser extent, by additional provision of consumer-directed health plans.While the conference attendees admit that individually these measures will not save a lot ofmoney, 10 percent here and 15 percent therebegins to add up.

Importantly, moreover, these experts broadlyagree that insuring the uninsured would requirerelatively modest amounts of additional money:less than $100 billion a year—less than 5 per-cent of current health care spending, roughlythe money returned to taxpayer pockets byrecent below-average tax rates,20 and moneythat could prevent 18,000 premature deaths ayear among the under 65’s, according to JimMongan. On net, the extra cost is likely to bemodest because the uninsured already get somemedical care, often in emergency settings, andbecause providing preventive care and diseasemanagement for these people would actually bemore efficient over time.

Thus, once again, these analysts concur, thenation should “just do it”21 and move to provideuniversal coverage without waiting until we figureout how to control health care costs, for, as JudyFeder argues, the uninsured minority have beenheld hostage to our unwillingness to slow thegrowth of health care spending for the well-insured majority for 50 years. Henry Aaron concludes that universal coverage may be a necessary precondition for controlling overallhealth care spending; others argue that universalcoverage must come first because cost controlwithout coverage would mean squeezing low-income people out of the system.

As a result, the conference participants gener-ally advocate using any cost savings reapedfrom the reforms discussed above to fundbroader health insurance coverage. As oneexample, Alan Weil suggests making employerpayments for health insurance benefits taxableand using the resulting revenue gains to funduniversal coverage.

Where Achieving

Consensus Becomes

a Challenge

Beyond the large areas of agreement justreviewed, two issues—the role of employer-basedinsurance and the most appropriate way to con-trol the growth of U.S. health care costs—defyconsensus. To start with the first issue, conferenceattendees clearly have differing views on the mer-its of this country’s employment-based system,with some viewing it as a disaster and others find-ing it an efficient organizing mechanism and aprogressive force. But whatever their views on itsmerits, many analysts, including Altman,Newhouse, and Feder, are convinced that theemployment-based system is crumbling badly,because, as Bob Galvin notes, many employersare seeking to escape from providing health insurance.That explains why employers are responding toconsumer driven health care (CDHC) with enthu-siasm; they really do believe that consumers mustbecome more cost conscious, but they are alsolooking for an exit strategy. Thus, Bob Galvin pre-dicts, 20 to 30 percent of all workers will soonhave health savings accounts (HSAs), which willdrive out traditional health insurance just as401Ks drove out defined benefit pensions.Employers don’t want to abandon their employ-ees, but CDHC provides them with an acceptableway out.

Unfortunately, however, CDHC and HSAs maynot work well for low-income workers, who mayopt to buy low-premium insurance but be unableto pay the required deductibles, co-payments, andother large, but less than “catastrophic”22 expens-es, or who may opt out of buying health insurancealtogether. These people will swell the ranks of theuninsured or the Medicaid population because, asnoted above, many states are making imaginativeefforts to redefine their Medicaid programs to letthem cover nontraditional beneficiaries. (See thebox on page 20 for a description of recent stateinitiatives in New England.)

24

But, as Alan Weil points out, the fiscal stresses atthe state level are becoming enormous. As aresult, the U.S. Congress passed the DeficitReduction Act of 2005 to give the states new lee-way to charge premiums and raise co-paymentsfor Medicaid benefits. Further, this law, for the firsttime ever, allows states to end Medicaid coveragefor people who fail to pay these new premiumsand permits doctors, hospitals, and pharmacies todeny services to Medicaid recipients who cannotmake required co-payments. To judge from current trends, the end result of employer effortsto avoid health care costs may be a de facto single-payer (or largely single-payer) system, butone in which impoverished people can be deniedneeded health care. For analysts who favoremployer-based insurance, the only way to stemthis tide may be to get “pay or play” laws thatrequire all employers to provide health benefits orto contribute to a state insurance pool back on thelist of live policy options.

The conference attendees also fail to reach con-sensus on further ways to curb the growth inhealth care costs beyond those that would posi-tion the U.S. health care system to operate atmaximum efficiency, although most agree thatsuch efforts would have to include limiting insuredmiddle-class access to valuable new technologies.At one extreme, a de facto single-payer systemwould require a global budget. Would such abudget fund optimum investment in new tech-nologies, Bob Galvin wonders, or would a market-based system do a better job? Also envisioning anongoing role for private insurance, Mark Paulysuggests that insurers develop low-cost insurancewith limited access to new interventions and technology and tout these products as “prudentcare” in order to slow the adoption of possiblydangerous (and clearly expensive) new technolo-gies. By contrast, Gene Steuerle would focus onfinding ways to encourage cost-saving rather thancost-increasing new technologies. But privatelyfunded health care would set the standards for all,because, as Jim Mongan points out, while we findprice rationing acceptable in the case of hotels, wenaturally find it far less palatable in the case ofhealth care. Still, non-price rationing through

government or private-payer limits leads to unac-ceptable queues and shortages. In the same vein,Nordhaus sees some attractions in Oregon’s sys-tem of ranking medical interventions by usingcost-benefit analysis; and good sense would sug-gest then drawing a line where the health carebudget is totally absorbed. Although the Oregonsystem has many problems and critics, and, afterall, only applies to Medicaid patients, Nordhausargues that it is logical and flexible, responding toboth technological and fiscal developments.

In the end, conference participants conclude, themajor challenge posed by the U.S. health care sys-tem remains summoning the political will to makethese difficult allocational decisions in a responsi-ble and equitable way. Failure to meet this chal-lenge would have serious consequences for theU.S. macro economy and polity—as well as forevery individual family’s well-being.

Endnotes1. See Theodore Marmor. 1994. “The Politics of Universal Health

Insurance: Lessons from Past Administrations?” Political Science and Politics, vol. 27, 194-198.

2. “Special Report: America’s Health-Care Crisis,” The Economist,January 28, 2006, 24-26.

3. Population aging will contribute modestly to this trend as well. 4. See Council of Economic Advisers. 2006 Economic Report of the

President.5. Many health economists argue that it is foolish to expect the

income elasticity of health care spending to be similar across coun-tries and particularly foolish to expect the relationship to be linear. Further, this country’s “outlier” status largely reflects the fact that the United States pays its health professionals relatively well, not that the U.S. system is inefficient. However, GDP does provide one constraint on health care spending, and one might ask why U.S. health professionals earn relatively high wages.

6. Looking beyond the healthy life expectancy data shown in the chart, the United States also uses more cardiovascular procedures per capita than Australia and Canada with less effect in terms of reduced mortality from heart disease. The United States also ranks near the bottom of OECD countries in terms of infant mortality and years lost to premature death, in part reflecting the uneven distribu-tion of health care resources in this country.

7. See William J. Wiatrowski. 2004. “Medical and Retirement Plan Coverage: Explaining the Decrease in Recent Years,” Monthly Labor Review, vol. 127, 29-36.

8. Tax-financed” includes Medicare and Medicaid, health care spend-ing for the military and their dependents, health benefits for govern-ment employees, and the value of tax subsidies for employer-provided health benefits. Steffie Woolhandler and David U. Himmelstein. “Paying for National Health Insurance—And Not Getting It,” Health Affairs, July/August 2002, 88-98.

9. However, Okma argues that some single-payer systems are quite good at prevention. She notes that the Germans are good at diseasemanagement—for instance, by sending cardiac patients to spas to learn how to change their lifestyle by exercising and losing weight.

25

10. Specific egalitarianism is the belief that a program or service should be distributed equally across all people, as with voting, the wartime draft, and primary and secondary education.

11. Even worse, a study by the Institute of Medicine finds that medical error is the eighth largest cause of death in the United States. John E. Wennberg, Elliott S. Fisher, and Jonathan S. Skinner. “Geography and the Debate over Medicare Reform,” Health Affairs, Project Hope, January/February 2002.

12. See Council of Economic Advisers. 2006 Economic Report of the President.

13. This subsidized system also places low-wage workers at a comparative disadvantage because health insurance premiums loom larger relative to their wages than they do for highly compen-sated workers.

14. Gene Steuerle points out that most people, including health economists, have no idea that total health care spending per household averaged $16,000 in 2003.

15. Further, 42 percent of those with high-deductible plans spent 5 percent or more of their income on health care (premi-ums and out-of-pocket items) compared with 12 percent of those with morecomprehensive plans. Paul Frontsin and Sara R. Collins. “Early Experience with High-Deductible and Consumer-Driven Health Plans: Findings from the EBRI/Commonwealth Fund Consumerism inHealth Care Survey,” EBRI Issue Brief 288, December 2005.

16. In a somewhat narrower setting, David Meltzer also notes how the development of “hospitalists,” physicians who specialize in providing inpatient care, has cut costs and improved the quality of hospital care delivered both by the hospitalists and by other physicians who work with them.

17. However, because most workers are relatively healthy, Mark Pauly suspectsthat job lock is unlikely to be a major concern. The growing number of two-worker households also helps alleviate this problem.

18. According to the CBO long-term out-look, the intermediate path would result in primary spending (defense, Social Security, Medicare and Medicaid, and other non-interest expenditures) rising from 17.5 percent of GDP in 2005 to 25 percent in 2050. The higher path would see primary spending soar from 17.5 percent of GDP to 34 percent. (U.S. Congressional Budget Office, “CBO Study: The Long-Term Budget Outlook,” December 2005.)

19. Making a dent would require some really big changes. According to Aaron, just to keep Medicare costs from rising faster than GDP would require boosting the eligibility age for Medicare to 83 in 2040 or reducing Medicare’s share of health care spending by the elderly from 60 percent now to 23 percent in 2040.

20. Federal tax revenues have averaged 18.3 percent of GDP over the past 30 years but were just 17.5 percent of GDP in 2005.

21. The mechanisms for doing so vary and could include broadening eligibility criteria for Medicare, Medicaid, the Federal Employees Health Benefits Plan, and other health plan purchasing organiza-tions, instituting an employer or individual mandate, or shifting to a single-payer system.

22. Low-premium, high-deductible health insurance plans do tend to cover catastrophic medical expenses.