Rising Health Care Costs: The New Role for Consumer Empowerment
, Greater Cost Medical Savings Accounts, and Two‐​Tiered
Defined Contribution Health Plans


My name is Tom Miller. I am director of health policy studies atthe Cato Institute, a nonpartisan public policy researchinstitution headquartered in Washington, D.C. I would like to thankChairman Underheim and Rep. Weickert for inviting me to testifytoday about what can be done about rising health care costs and, inparticular, the possible role of medical savings accounts inrestraining the growth of health care costs and making health caremore affordable and available.

This committee has already heard quite a bit of testimonyregarding the various possible factors behind the return of annualdouble-digit percentage increases in health care spending andhealth insurance premiums. Some of those factors may remain outsidethe immediate reach of public policy, such as increased consumerdemand for health services in an aging and relatively wealthiersociety. Other factors include a mix of both higher costs and evenmore highly valued benefits, such as more effective prescriptiondrugs and innovative technological advances in medical diagnosticsand treatment.

Indeed, our society may well decide to spend even higher andhigher shares of our nation's resources on health care in futureyears - as long as someone, somewhere can be found to foot thebills - but American consumers will receive more value for eachdollar they spend only if the distorting effects of government'smultiple role as a regulator, purchaser, and subsidizer of healthcare are reduced. Our objective should be neither to artificiallykeep spending levels higher, nor lower, than theirmarket-determined costs, but instead to allow individual consumersto seek the best value that balances their spending preferences andpriorities with the resources that they can command. I'll discuss alittle later how best to sort out the respective roles of efficientmarket-based mechanisms for delivering health care and societalgoals of safety net care, income security, and welfareassistance.

A number of the cost-drivers behind higher health care bills arerelated in part, if not entirely, to current public policies thatare outside the primary scope of my remarks today. Mandated healthbenefits and excessive health services regulation at the state andfederal levels unquestionably boost health care costs - estimatesmight range from a conservative 10 percent to as much as 25percent, on average - and they contribute significantly to thelevel of Americans who cannot afford to purchase private healthinsurance. Medical malpractice costs, including defensive medicine,cannot be eliminated but they can be made better or worse byparticular government policies.

Although adoption and dissemination of advanced health caretechnologies is frequently blamed as the key culprit in thelong-term secular trend toward rising health care costs, that sortof oversimplified analysis often neglects to ask whether (1) ourlives are better off, on balance, even after paying those highercosts, and (2) how the pattern and pace of such technologicaladvancement might change if we relied less on third-party payers tofund it and increased the share of health spending that is paidout-of-pocket by individual consumers.

Another related set of governmental and non-governmental factorshave combined to diminish the former role of managed care insurersin holding down the costs, if not improving the value, of privatesector health care. The managed care backlash, in conjunction withprovider pushback on prices, the pendulum swing of the insuranceunderwriting cycle, and employers' competition for scarce labor,may have been inevitable to some degree - but public policy effortsto regulate, if not outlaw, many managed care practices and toencourage court challenges to third-party restrictions on access tocare certainly contributed to a loosening of referral andauthorization rules, as well as more inclusive provider networks.As a result, today we have more choice and access to care in manyhealth insurance plans, as long as we (or our employers) still canafford to pay the higher premiums.

Medical Savings Accounts & Cost-SharingClones

In the current environment of rising health care costs, thefading away of managed care's third-party controls on the supply ofhealth care, and (apparently) continued political resistance toeven higher levels of government involvement in the regulation andfinancing of health care, more employers and their employees areturning to less comprehensive insurance coverage, greaterindividual cost-sharing, and reduced "insured" benefits as the mostpromising and, until now, relatively less-explored means to controlhealth care costs.

Consumer-driven health care ranges across a wide continuum ofhealth financing vehicles - from modest increases in cost-sharingunder more conventional employer-sponsored health plans (higherdeductibles, multi-tiered copayments for types of coveredprescription drug purchases, higher out-of-network coinsurancepercentages) to various types of two-tiered, defined contributionhealth plans (combining a higher-deductible group insurance policywith an individual health savings account) to the "real thing" -medical savings accounts (MSAs).

In a health care world where "nobody else" seems to managingeither the cost or the quality of health care very effectively,MSAs and their cost-sharing cousins are picking up the baton andleading employers and their workers toward a consumer-driven modelof health care purchasing.

Note that a recent Watson Wyatt Worldwide survey found that only43 percent of workers were satisfied with the overall performanceof their health plan. Less than half (48 percent) trust theiremployer to design a plan that will provide the coverage they need.Approximately the same percentage (47 percent) think that betterhealth plans are available for the same cost. And almost four outof ten employees want their employer to contribute a fixed dollaramount toward the premium for any health plan - even if it meansfinding their own health plan.

The evidence is overwhelming that increased cost sharing reduceshealth insurance premiums substantially. For example, Jason Lee andLaura Tollen recently noted in a June 2002 article in HealthAffairs that increasing cost sharing from a plan with $ 15 copaysand no deductible to one with 20 percent coinsurance and a $ 250deductible reduces premiums by about 22 percent; and a combinationof 30 percent coinsurance and a $ 1000 deductible would reducepremiums by 44 percent. Offering less comprehensive insurance planswith greater enrollee financial responsibility is designed toencourage enrollees to be smarter consumers of health careservices, limit demand for less beneficial "discretionary" care,seek out higher-value options, and save money for more criticalmedical needs in the future.

The recent revival of greater cost-sharing and so-calledconsumer-driven health plan options may provide a partialtransition vehicle for employers who are rethinking their healthbenefits strategies but remain ambivalent about relinquishing mostof their role in structuring employees' choices and monitoringhealth plan vendors.

The "full-strength" version of consumer empowerment, of course,remains an MSA. MSAs provide a health care savings account incombination with a high-deductible health insurance policy. Thesavings account is controlled by the insured person and used to payroutine health care expenses. The accompanying catastrophicinsurance policy covers more substantial health care costs. Becausethe cost of such a policy is usually significantly less than thecost of a low-deductible policy, the money saved may be used toincrease contributions by an individual (or his employer) to an MSAadministered by a designated trustee or custodian.

Unspent MSA funds, including any interest or investmentearnings, accumulate from year to year, providing money to coverpossible medical expenses in the future.

MSAs help control costs, improve access to health care, expandconsumers' choice in and control of health care, and increasesavings.

By putting individuals back in control of more purchasingdecisions, MSAs create incentives for individuals to purchasehealth care more prudently and reduce their overall health carespending in a given year.

Whereas out-of-pocket payments by individual consumers accountedfor about 50 percent of total health care spending in 1960, theshare of third-party payments (by private health insurers,employers, and government agencies) for health care has grown toabout 80 percent. Third-party payment of health bills insulatesindividual consumers from the real cost of their health caredecisions and treatment. Consumers have less reason to avoidunnecessary care, question costs, or shop around for the besttreatment available at a reasonable price, but they have everyincentive to demand more services.

Excessive third-party coverage with low deductibles increasesadministrative costs, because every small bill must be submittedfor review and checking for accuracy.

Instead of limiting the supply of desired medical services, MSAslower the demand for those services by requiring individuals to paydirectly and up front for their discretionary health carechoices.

The RAND Health Insurance Experiment, conducted from 1974 to1982, demonstrated that the more people had to pay for medical arewithout insurance reimbursement, the less they would spend on totalmedical care.

Because MSA plans are linked to high-deductible insurance thatcovers health claims that are more catastrophic in nature, theymake the cost of insurance coverage more affordable for mostAmericans. Less -comprehensive coverage will mean lower premiumsfor a larger fraction of people with low incomes. The majority ofstandardized insurance policies currently available are generousand expensive - making them unaffordable to low-income people. Onthe other hand, catastrophic insurance for very large,less-predictable health care expenses forces consumers to bear thefull marginal costs of health care up to the point where their useof health care exceeds the deductible.

Under many third-party health benefit arrangements, consumershave little incentive or ability to become more knowledgeable abouthealth care. MSAs stimulate consumer demand for information aboutthe quality and price of health care.

A number of studies have illustrated that MSAs improve healthplan options not just for affluent and health individuals, but forall Americans.

In April 2000, RAND Corporation researchers examined the effectsof making MSA options available to small businesses. RAND rejectedthe assumption that MSAs appeal most to the wealthiest andhealthiest workers. It found that HMOs would remain more attractiveto higher-income workers, primarily for tax reasons, andexceptionally good health risks would be more likely to decline anyinsurance at all than to select the MSA option.

A 1996 study by National Bureau of Economic Research analystsconcluded that most workers would end up retaining a substantialportion of the contributions they made to MSAs by the time theyretired. Approximately 80 percent of employees would have retainedover 50 percent of their MSA contributions by the time ofretirement, and only 5 percent of workers would have saved lessthan 20 percent of their contributions. Although workers with highhealth care expenses in one year tend to have lower but stillhigher than average expenses in the next few years, theconcentration of annual expenditures declines continuously as moreand more years of expenditures are cumulated. High expenditurelevels typically do not last for many years.

Another 1996 study of Ohio-based firms that offered MSAs thatdid not qualify for tax advantages under the Health InsurancePortability and Accountability Act (HIPAA) determined that theemployer's total cost for family coverage under those MSA plansaveraged 23 percent less than traditional family coverage, yet theaverage employee with family coverage also would be $ 1355 betteroff under the worst-case (maximum out-of-pocket liability)scenario.

So, if MSAs are so great, why don't we see more of them in themarketplace? Primarily because the federal MSA program authorizedunder HIPAA in 1996 has been unnecessarily handicapped, if notpermanently crippled, by unreasonable restrictions on what stillremains a "demonstration project." Congress still needs topermanently authorize federally qualified MSAs; lift the enrollmentcap and allow an unlimited number of people to have MSAs; expandMSA eligibility to include employees in businesses of all sizes, aswell as employees without employer-sponsored insurance; allow MSAplans to offer a much wider range of deductibles; allow MSA holdersto fund fully their MSAs each year, up to 100 percent of theinsurance policy deductible; allow employers and employees tocombine their contributions to MSAs at any time within a givenyear; and either preempt first-dollar state-mandated benefits orprovide the flexibility for MSA plans to adjust to comply withthose conflicting mandates.

In the case of MSAs, most of the problems have been caused atthe federal policy making level. Aside from making equivalent stateincome tax benefits available to MSA owners and avoiding oreliminating restrictions on high-deductible insurance policiessubject to state regulation, state policy makers should continue topress Congress to make MSAs universally available. They also mightconsider offering MSA-like health plan options to state employeesas part of their benefits package, in order to boost demand in thiscurrently thin market.

Remove Barriers to Growth of Defined Contribution HealthPlans

Despite the potential benefits of two-tiered definedcontribution (DC) health plans, as well as the recent tax guidanceissued by the Internal Revenue Service clarifying how accumulatedbalances in an individual employee's health reimbursement accountsmay be treated when rolled over at the end of a year, severalregulatory barriers to the future growth of DC plans still need tobe removed.

First, "pure" DC plans for fully insured employer groups, inwhich an employer distributes defined health benefits contributionsto each eligible employee and allows them to purchase their ownindividual or non-employer-group insurance coverage, run the riskof being regulated inconsistently. They might be treated both asemployee welfare benefit "group" plans and as "individual" healthplans under state law.

To clarify the regulatory treatment of any plan or fund underwhich medical care is offered to employees by an employer solelythrough provision of a monetary payment or contribution to aparticipant or beneficiary and that is used exclusively to purchaseindividual health insurance coverage, state policy makers and otherinterested parties should press CMS and members of Congress toclarify that such plans or funds should not be considered an"employee welfare benefit plan" for regulatory purposes under theEmployee Retirement Income Security Act (ERISA). However, suchplans or funds would retain their "group" tax exclusion benefitsunder the Internal Revenue Code. Such hybrid treatment (group fortax purposes, individual for regulatory purposes) would be premisedon the conditions that (1) only the employer, rather thanindividual employees, may decide to provide health benefits throughdefined contribution payments, and (2) such defined contributionsmust be provided to all employees or all members of a class ofemployees based on work-related distinctions.

Second, the defined contributions employers make to individualemployees in pure DC plans, to be used to purchase individualhealth insurance coverage, should be allowed to vary on the basisof health status in the event the employer uses an approvedrisk-adjustment mechanism. That is, employers would be allowed tomake larger contributions to workers with poorer health status tooffset the higher premiums they would face when they seek topurchase individual coverage. State insurance regulators would needto approve this exemption from HIPAA non-discrimination rules for"group" plans, or press CMS to update and revise its pastregulatory interpretations of this provision.

Lower Costs Trump Insurance Subsidies

To achieve better health outcomes, we need to provide individualhealth care consumers with stronger incentives to be cost-consciousin using scarce medical resources. Making the market-based cost ofcare more transparent to all parties involved in health spendingdecisions will encourage its more efficient consumption anddelivery. Reducing the long-term rate of growth in the cost ofhealth care remains more important than (and, beyond a certainpoint, operates at cross-purposes to) expanding the scope and scaleof subsidized health insurance coverage. Health insurance subsidiesincrease medical costs and the demand for health insurance,creating net welfare losses estimated at 20 percent to 30 percentof total insurance spending. In the opposite direction, access to"free" care dampens the demand for private health insurance. Instriking the necessary balance, the effects of comprehensivethird-party insurance on raising costs and limiting access tohealth care substantially outweigh any disincentives to obtaininsurance protection that may be caused by direct provision ofcharity care. When rising health care expenditures outpace wageincreases, their strongest effect is to reduce health insurancecoverage for low-income workers. Hence, at the margin, increasingincentives to purchase less-comprehensive health insurance andfilling in urgent gaps in direct delivery of health care throughsafety net mechanisms may produce more affordable and accessiblehealth care.

Market-Driven Deregulation via CompetitiveFederalism

Empowering consumers with a greater diversity of affordablehealth benefits choices will require exposing exclusive statehealth care regulation based on geography to competition frommarket-friendly regulation across state lines.

Lower-income workers in small firms bear the brunt of excessivestate health insurance regulation, because their employersgenerally are unable to self-insure and, thereby, gain ERISAprotection from state benefit mandates, restrictions on rating andunderwriting, and other regulatory burdens. In general, increasedstate regulation has raised the cost of health insurance andlimited the range of benefits package design. A wide assortment ofsmall-group regulatory measures imposed by many states during the1990s failed to improve levels of insurance coverage and, in somecases, priced low-risk consumers out of the small-group market.Various state government regulatory attempts to force low-riskinsureds to subsidize high-cost insureds through devices likemodified community rating and guaranteed issue often werecounterproductive, because they triggered premium spirals thatdrove younger, healthier, and lower-income workers out of thevoluntary insurance market. In other words, state health insuranceregulation has been part of the problem, not part of thesolution.

Rather than try to solve state-based regulatory failure with anew round of heavy-handed federal rule making or preemption, thebetter route to restoring a market-friendly, consumer-empoweringenvironment at the state level is to facilitate competitivefederalism-revitalized state competition in health insuranceregulation that reaches across geographic boundary lines. (Theclosest successful model for such competitive federalism involvescorporate law and the business of corporate charters, in whichDelaware has specialized and excelled by consistently producingbenefits to its "customers"-investors.) Such regulatory competitionwould limit the excesses of geographically based monopolyregulation. Currently, insurance consumers (at least in thenon-self-insured market) are subject to a single state government's"brand" of insurance product regulation. Solely by virtue of wherethey live, they are stuck with the entire bundle of their homestate's rules. Short of physically moving to another state, theyare unable to choose ex ante the type of health insuranceregulatory regime they might prefer and need as part of theinsurance package they purchase.

Competitive federalism could facilitate diversity andexperimentation in health insurance regulatory approaches. It woulddiscipline the tendency of insurance regulation to promoteinefficient wealth transfers and promote individual choice overcollective decisions driven by interest group politics. In short,it would improve the quality of health insurance regulation,thereby enhancing the availability and affordability of healthinsurance products.

Insurers facing market competition across state lines would havestrong incentives to disclose and adhere to policies thatencouraged consumers to deal with them. Employers and individualspurchasing insurance would migrate to state regulatory regimes thatdid not impose unwanted mandates but, instead, fit the needs oftheir consumers. State lawmakers would become more sensitive to thepotential for insurer exit. At a minimum, interstate regulatorycompetition would provide an escape valve from arbitrary ordiscriminatory regulatory policies imposed at either state orfederal levels.

Key design requirements for regulatory competition in healthinsurance would include:

  1. Only one sovereign has jurisdiction over a particular set ofhealth insurance transactions, and its law controls the primaryregulatory components of the regime governing them. Other statesprovide regulatory reciprocity (also known as the "principle ofmutual recognition" in the European Union), by respecting andenforcing that state's insurance charter and its accompanyingrules. Such reciprocity works through private arbitrage ofjurisdictional competition, rather than politically mandatedharmonization that suppresses competition.
  2. Health insurers can choose their statutory domicile, orotherwise determine the applicable forum and applicable law, andmake it part of the purchasing option they present to consumers.Insurers and their consumers can exercise the right of free exit:they can vote with their feet and their pocketbooks. Insurers canchoose their domiciles, the markets where they prefer to operate,and the bundle of laws and regulations attached to the productsthey sell. They can relocate to alternate jurisdictions atrelatively low cost. Consumers may choose not only the state inwhich they live but also the legal rules attached to the insuranceproducts they buy.
  3. States must receive some benefits, such as tax revenues, fromcompeting in the production of specific laws and regulations thatreduce insurers' business costs and increase the value of insuranceproducts. Conversely, states also must feel within their ownborders a sufficient number of any negative consequences of theregulatory regimes they choose to adopt and "export" to consumersin other states.
  4. Competition for the marginally informed consumer must operateto protect other consumers who are not aware or informed of theparticular regulatory regime.
  5. Rather than present a single set of contract terms on anall-or-nothing basis, insurers can offer consumers a menu ofalternative policies that are priced to reflect differentregulatory approaches.
  6. Solvency regulation should remain decentralized and kept at thestate level, to avoid federal domination over other regulation inthe name of protecting consumers and taxpayers. Regulatorycompetition for insurance product design, pricing, and poolingcould be accommodated within the current state-based guaranty fundsystem in a manner that limits an individual state's opportunitiesto impose costs on other jurisdictions.

Several mechanisms or paths could lead to vigorous interstatecompetition in health insurance regulation. A more indirect, butsustainable, approach would involve strategic use of choice offorum clauses, and perhaps choice of law clauses, in healthinsurance contracts. Insurers would condition sales of a particularpolicy on a consumer's consent to the designated litigation forum.That forum would be matched to the state whose regulatory law wasselected. This choice of forum would need to be adequatelydisclosed and executed at the beginning of the contractual period,not just at the time of litigation. Insurers could increase thelikelihood that the agreement would be enforced and regulatorycompetition enhanced by linking the designated forum to theircompany's domicile-rather than to the site of the salestransaction.

Federal law could provide some shortcuts-such as a statutemandating enforcement of choice of forum contracts under thecommerce or full faith and credit clauses of the Constitution.Congress also could provide uniform disclosure requirements forchoice-of-forum and the insurer's domicile in insurancecontracts.

A more direct federal statutory approach might set an "insurerdomicile" rule, in place of a "site of transaction" rule, fordetermining applicable state law and regulatory authority-at leastas a default rule for multi-state transactions where the respectiveparties do not otherwise designate operative law. For example, Rep.Ernest Fletcher (R-KY) recently introduced the "State CooperativeHealth Care Access Plan Act of 2002" (H.R. 4170), which wouldauthorize a health insurer offering an insurance policy in oneprimary state (the primary location for the insurer's business) tooffer the same policy type in another secondary state. The product,rate, and form filing laws of the primary state would apply to thesame health insurance policy offered in the secondary state.

Another route to interstate competition in insurance regulationmight be built on decisions by individual states to grantregulatory "due deference" to determinations by out-of-stateinsurance regulators that a particular insurance company isqualified to conduct such business. Once an insurer submittedevidence of good standing in its domestic jurisdiction and (ifdifferent) in the jurisdiction where it conducts the largest shareof its health insurance business, it would qualify for licensure inthe state granting such regulatory deference. Regulators insecondary states would be most likely to treat proof of licensureand good standing in the primary state as prima facie evidence ofqualification for licensure in the secondary state, while stillrequiring additional routine documents and fees and compliance ofthe primary states' insurance department with broadly acceptedaccreditation standards, such as those maintained by the NAIC.Initially, an individual state's decision to grant regulatory duedeference would be similar to a declaration of unilateral freetrade in health insurance products. The state would be eliminatingor reducing its own regulatory restrictions on out-of-stateinsurance to benefit its citizens and to provide a model for otherstates to emulate.

A Real Safety Net for the MedicallyUninsurable

Medically uninsurable individuals represent a small percentageof the uninsured population (roughly no more than 1 percent to 2percent of the uninsured have ever been denied health coverage formedical reasons). But they present the strongest case for publicassistance. To some degree or another, at least 29 states currentlyoperate high-risk pools that make insurance coverage available tothe uninsured and subsidize their premiums. States withwell-structured and adequately financed high-risk pools are moresuccessful in keeping their individual health insurance marketscompetitive and insurance rates affordable. Such pools allow theindividual insurance market to operate efficiently, while carvingout for special treatment high-cost individuals who are beyond thecapacity of the individual market to handle on an unsubsidizedbasis.

However, not all state high-risk pools are adequately financed(ideally, the funding should come from general revenues rather thanthrough taxes on insurers within the state), and many states do notprovide such subsidized coverage at all. Using the rationale thatthe "medically uninsurable" (at least to the extent that theunsubsidized price to insure them privately far outstrips theirability to pay) should be considered "medically needy," mandatoryMedicaid coverage and matching federal assistance should beextended to this class of beneficiaries, provided that the federalfunds are channeled through state-operated high-risk pool programsthat meet certain minimum criteria (for example, premium ceilings,waiting periods, rejection by at least one insurer, catastrophicconditions allowing automatic pool acceptance without prior carrierrejection) already in practice, but not "new" ones. The scope andscale of this Medicaid-financed high-risk pool coverage for themedically uninsurable would be capped at an upper ceiling thatequals the higher amount of all individuals in a state facingprivate insurance premiums that are at least 200 percent ofstandard rates (plus those who cannot obtain any coverage at all,for medical reasons) or 2 percent of all people covered in astate's individual insurance market.

Citizens' Appropriations for Charity Care

To bolster financing for charitable safety net care and ensurethat it is delivered with private-sector efficiency, a new 100percent, dollar-for-dollar federal income tax credit (above theline) should be provided for certain charitable contributions toprovide health care services to the low-income uninsured. These"citizen appropriations" would be modeled in part on the Arizonatax credit for education "scholarships." The maximum individualcredit amount allowed would be no greater than 10 percent of anindividual's federal income tax liability in a given tax year.Eligible donations would have to be made to approved organizationsthat provide health insurance coverage, health care services, orpayment of medical bills to uninsured individuals who are noteligible for optional federal health tax credits or Medicaidassistance. Organizations eligible to receive the donations musteither be a non-profit, in accordance with section 501(c)(3) of theInternal Revenue Code, or, in the case of health care providers andthat who wish to receive direct donations, they must create aseparate non-profit subsidiary to receive and distribute suchfunding. Eligible organizations could spend only as much of theirdonations as they could document were directed toward paying thehealth care expenses of qualified uninsured individuals. Taxpayerscould designate the institution to which their donation would bedirected, but they could not pinpoint the individualbeneficiary.

States could play a role in jump-starting this process, byproviding their own state income tax credits along similar linesfor such "citizen appropriations," even in the absence of federalpolicy action.

"Getting Over" Adverse Selection

Despite hypothetical concerns about adverse selection and risksegmentation in a more competitive, market-based private healthinsurance system, there actually is little evidence thatindividuals and families can identify and anticipate most of theirfuture medical expenses in ways their potential insurers cannot. Arecent study by Cardon and Hendel in The Rand Journal of Economicsfinds little empirical evidence of information asymmetries, marketfailure, and adverse selection in health insurance markets.Differences in health expenditures between the insured anduninsured are mostly due to observable differences in demographics(age, gender) and price sensitivities (higher-income workerscapture more tax subsidies for insurance coverage), rather thanunobservable factors related to health status.

Long, Marquis, and Rodgers also find little support for thehypothesis that people anticipate changes in their insurance statusand arrange their health care consumption accordingly. The authorsalso find no evidence that people choose to purchase or dropinsurance coverage in anticipation of change in their overallhealth care needs and conclude that insurer selection is anunlikely explanation for this failure to find quantitativelyimportant transitory demand. However, they observe that recentstate reforms aimed at eliminating or limiting some insurerrestrictions on coverage of pre-existing conditions ironicallymight increase the ability of patients to adjust their treatmentpatterns for chronic conditions in anticipation of insurancechanges.

Private insurers don't need to remain helpless and cluelessregarding potential adverse selection problems. In competitivemarkets, they may use a number of tools: set periodic limits onplan switching, vary premiums according to the amount of insurancepurchased, underwrite and rate based on risk categories, createmore homogeneous risk pools, or rely on the law of large numbers todiversify risks in large pools. Consumer inertia and individualdifferences in aversion to risk further limit the applicability ofadverse selection theory to the real world.

Many difficulties we observe in health care insurance marketsare due to government intervention rather than adverse selection orother market failures. If insurers are not allowed to chargedifferent premiums to different risks, price predicted riskappropriately, and match their policy configurations to marketdemands, they will be more likely to resort to higher uniformprices, less savory practices like excluding or discouragingcoverage of high risks, and, ultimately, market exit. Creamskimming (selecting only the best risks) becomes the insurers'mirror image of adverse selection by insurance customers. Politicalinterventions don't alleviate underlying differences in risk acrosscustomers or eliminate insurers' knowledge of such differences.They only force insurance companies to cope in inefficient ways andcreate new problems.

It is preferable to allow private insurers to do what they dobest-evaluate risk and price it accordingly-and then deal withremaining outlier problems (for example, the medically uninsurable)through explicit, transparent public subsidies rather than morecamouflaged regulatory cross-subsidies. We should separate supportfor societal objectives of income redistribution and protectionagainst prohibitively expensive, but predictable, health risks fromthe competitive operations of commercial insurance markets.

Health status information is most likely to be asymmetric whenit is scarce and costly. While government mechanisms prefer toignore, hide, or shift those information costs, markets createproper incentives to discover efficient ways to signal relevantprivate information and put it to use.

Deregulating insurance choices and providing greater tax parityfor all insurance purchasers can fill the real gaps in privateinsurance coverage, by providing breathing room for further marketinnovations, such as new forms of voluntary risk pooling andlong-term insurance contracts. The growing availability of onlinehealth information and insurance products further strengthens thecase for empowered consumers.


Market mechanisms can't eliminate every unfortunate humanexperience in health care access, affordability, and quality.Private charity and a backup safety net of transparent, directsubsidies have necessary roles to play. Unlike centralizedgovernment "solutions," markets don't promise perfect outcomes,just better ones.

The current climate of annual double-digit percentage increasesin health care costs, dissatisfaction with the mature version ofmanaged care, and remaining political resistance to centralizedcommand-and-control mechanisms points to greater acceptance of thelast remaining, relatively unexplored health care reformoption-putting choices back in the hands of individual consumersand competitive free markets.

In a more market-based health care system, you would, to a largedegree, get what you pay for, unless someone else wanted to pay forit voluntarily on your behalf. Income redistribution issues shouldbe debated separately and resolved in the larger political arena,while we finally allow health insurance markets to operate moreefficiently for the purposes for which they are best suited.