D AW -C H C S

DESIGNING A WORLD-CLASS HEALTH CARE SYSTEMHoward J. Bolnick*

ABSTRACT

Public health systems are finding it increasingly difficult to fund all the health care that their citizenswant and need. Fiscal constraint is causing many nations to reconsider the respective roles of theirpublic programs and private health insurance. The author examines the potential for and perfor-mance of health systems around the world and the advantages and disadvantages of public andprivate health financing. This analysis leads to the development of a framework for improvementsin today’s mix of health care financing and to high-level principles for a better-coordinatedrelationship between public and private programs. The role of health actuaries in moving towardmore effective health systems is then explored.

1. INTRODUCTION

Since the dawn of civilization health care hasbeen a focus of public and government interest.As early as 1700 B.C.E., Hammurabi, ruler ofBabylon, developed laws on health care mattersthat included access to services, payment forcare, and quality control. Hammurabi’s Code ofLaw (King 1910) declares:

● Cost: “If a physician make a large incision withan operating knife and cure it, or if he open atumor (over the eye) with an operating knife,and saves the eye, he shall receive ten shekelsin money” (law 215).

● Access: “If the patient be a freed man, he (thesurgeon) receives five shekels. If he be the slaveof some other, his owner shall give the physi-cian two shekels” (laws 217–218).

● Quality: “If a physician make a large incisionwith the operating knife, and kill him, or open atumor with the operating knife, and cut out theeye, his hands shall be cut off” (law 218).

While our sensibilities about acceptable ap-proaches to stewardship of our systems certainlydiffer from Hammurabi’s, these same three issuesof cost, quality, and access continue to shapeongoing debates over health care systems aroundthe world.

Nations throughout the world are faced withgrowing demands on their health care systems,often accompanied by diminishing abilities to sat-isfy and pay for all the health care needs andwants of their citizens. The purpose of this paperis to develop an international perspective onhealth care systems and a preferred general di-rection of change that allows policymakers (in-cluding actuaries) and politicians to adapt to uni-versal supply and demand pressures.

The health care sector of the global economy ishuge: Its issues are many and they are exceed-ingly complex. Arguments over trade-offs be-tween social solidarity versus personal autonomy,public versus private health care financing, publicversus private provision of health care services,and the need for high-quality medical care versusmore basic levels characterize health care de-bates in virtually every nation. Yet, to move for-ward, we must understand these enormouslycomplicated problems and find a path that movesus toward executable health care policies.

To help in this quest, this paper explores recentevolution in international thinking about whatmight constitute an “ideal” world-class healthcare system—a public system offering basichealth care to all citizens, complemented by aprivate system providing for the nonessentialhealth services that many citizens want, but donot need, to attain world-class health outcomes.The path to understanding what constitutes an“ideal” system covers a wide range of consider-

* Howard J. Bolnick, F.S.A., Hon.F.I.A., M.A.A.A., is Chairman andCEO of InFocus Financial Group Inc., 30 N. LaSalle St., Suite 3440,Chicago, IL 60602, e-mail: [email protected].

1

ations. A brief overview of this paper’s structurefollows.

● The paper begins by describing important per-sonal, social, economic, and political drivers ofhealth care and health care systems. Thesenonmedical factors greatly affect the shape ofnations’ health care systems and account forthe wide diversity in how they are organized.

● It then discusses health care systems’ goals andpresents a generalized structure for how systemsare organized to fulfill their societies’ goals.

● Next, I use comparative data from the WorldHealth Organization’s (WHO 2000) WorldHealth Report 2000 to look at the world’s 191health care systems. This analysis leads to iden-tifying 24 “world-class” health care systems andkey characteristics that separate these systemsfrom others.

● The worldwide comparison then is extended toa more thorough analysis of the U.S. and U.K.health care systems. Based on differing non-medical drivers, these two “world-class” sys-tems adopted very different structures to satisfytheir citizens’ health care goals.

● Next, I explore the economic characteristics ofprivate and public insurance markets: Howchoice of market structure affects health caresystems’ performances and the ability of sys-tems to fulfill societies’ goals.

● Lessons learned from the analysis are then usedto describe an ideal health care system, with“ideal” being a very general statement of pre-ferred design principles. These principles leaveroom for differences among health care sys-tems. This is necessary, since each nation’s per-sonal and social sensitivities, its economic sys-tem, and its politics are all hugely important inshaping the particular structure of its healthcare system. There is no one-size-fits-all healthcare system. Thus, the “ideal” system pre-sented is not a specific design; nor does it pro-vide a uniform prescription for change or en-dorse any particular political health care reformproposal in the United States or other coun-tries.

● Lastly, I discuss how actuaries can play impor-tant roles as national health care systems con-tinue to evolve.

The World Health Report 2000 (WHO 2000) is amajor source of conceptual information and most

of the data used in this paper. This report pre-sents a structure and data to measure and com-pare the worlds’ 191 health care systems. Thiswas an enormous undertaking by WHO research-ers, and not one without controversy. Interestedreaders will want to refer the report itself for adetailed description of the data, data collectionmethods, and limitations on it’s reliability anduse. References are also provided in Section 6 tosources critical of the validity of these data.

2. HEALTH CARE SYSTEMS POTENTIALS

2.1 Health IdealsAll modern, well-developed health care systemsshare a common goal for those who depend ontheir services: hope for full and healthy lives. Thepersonal importance of this goal inevitably meansthat health care is much more than an ordinaryeconomic good or service. Good health is a stateof being that is necessary for each of us to havethe opportunity to fully express our human selvesand to be able to reach our human potential. Ournational health care system, whether public orprivate or mixed, is the social and economic ex-pression of support for our individual good health.

The personal importance of good health andhealth care systems is well expressed by healthethicists Roberto Mordacci and Richard Sobel (1998).

“Health can be seen as a means, a foundation forachievement, as a first achievement itself, and anecessary premise for further achievement . . . .The sick individual suffers isolation, loss of free-dom to act, loss of the familiar world; the future isin doubt and all attention is concentrated on thepresent . . . . When ill, we no longer trust our bod-ies and . . . we no longer trust life” (p. 34).

No wonder health care financing and provision forhealth care services are such sensitive public is-sues.

Personal importance of health drives societalexpectations for health care systems. The sensi-tive nature of health and health care led WHO toadopt an expansive and controversial definition ofhealth in the preamble to its 1946 constitution.This “ideal” definition of “health” as a goal forWHO and its member nations’ health care sys-tems is: “a state of complete physical, mental andsocial well-being and not merely the absence of

2 NORTH AMERICAN ACTUARIAL JOURNAL, VOLUME 7, NUMBER 2

disease or infirmity” (WHO 2000). This definitionclearly moves beyond medical care and can easilybe read to include a wide range of nonmedicalcare outcomes (e.g., housing, nutrition, educa-tion, adequate income).

As an “ideal,” this definition of health is hard tofault. However, as a practical definition drivinggoals and objectives for the world community’shealth care systems, it is far too expansive and,therefore, wildly unrealistic as a guide to policy.As unrealistic as it is as a goal for national healthsystems, a state of complete well-being does seemto drive personal expectations for health, partic-ularly in higher-income nations with well-devel-oped health care systems. Ethicist Daniel Calla-han (Callahan 1998) of the Hastings Center hascriticized the U.S. health care system as actuallyreaching for an implicit goal of providing “all thecare we ‘want’ and ‘need,’ when we want it!” Heargues that fulfilling this expansive expectation isimpossibly expensive and that this is a majorreason that the U.S. health care system is somuch more expensive than those of other na-tions. A key aspect of Callahan’s criticism to noteis his distinction between care citizens “need”that is less than the care and amenities they“want.” Separating “wants” from “needs” is animportant part of this analysis.

2.2 Ethics/Ethos and Politics Shape aSystem’s Potential

How these medical care “wants,” “needs,” andpersonal expectations are fulfilled in a nation’shealth care system is driven by a number of non-medical considerations. Ethics and politics aretwo considerations that have profound affects onhow a nation organizes the financing and provi-sion of medical care for its citizens.

The body politic of nations differ in their soci-etal ethic. The distinction between social solidar-ity and personal autonomy is an important driverunderlying a nation’s choice between public andprivate health care systems. At the extremes ofthis dimension are the U.K. health care systemand the U.S. system.

The United Kingdom adopted its public healthcare financing and delivery system as a “reward”to itself following the turmoil and pain of WorldWar II. Social Insurance and Allied Services(Beveridge 1942), also known as the “Beveridge

Report,” laid the foundation for creation in theUnited Kingdom of a National Health Insurance(NHI) program. Reading the “Beveridge Report,” Iwas struck by the almost total absence of argu-ment for a scheme based on social solidarityrather than personal autonomy. The report im-plicitly assumes that the British body politic so-cial ethic accepts a strong sense of equality andpursuit of common well-being among U.K. citi-zens as an essential ethic that should underlietheir health insurance scheme.

At the other extreme, periodic health care re-form debates in the United States show strongbias on the part of the American body politic forpersonal autonomy over social solidarity.

Medical decision-making ethics also stronglyaffect the choice between public and private pro-vision of health care services. Whether an ethicprevails that favors state paternalism, patient au-tonomy, or professional paternalism makes astrong difference. Using as our extremes the U.S.and U.K. health systems, we can better under-stand this ethics-driven choice.

Since early in the development of modern med-icine, physicians in the United States have cre-ated and strongly defended a private health caredelivery system that gives them a great deal ofautonomy and professional paternalism (Starr1982). In contrast, in the United Kingdom, federalgovernment tightly controls NHI. State paternal-ism reins in the U.K. health care system, in largepart because it is ethically acceptable to U.K.physicians. It should be noted that, as citizensbecome more educated on health and health carematters, both of these systems are adopting moreof a patient autonomy ethic.

Family ethic also drives the shape of a nation’shealth care system. For example, in nations withstrong intergenerational family bonds, much ofthe care support for patients, particularly thosewith long-term chronic conditions and infirma-ries of aging, are taken care of by families. Innations with looser intergenerational familybonds, patients often demand nonmedical caresupport from the medical care system. A loosen-ing of family ties is characteristic of most devel-oped nations, including the United States and theUnited Kingdom.

Political-economic ideology, political powercenters, and the political decision-making pro-cess all also have strong affect on a nation’s health

3DESIGNING A WORLD-CLASS HEALTH CARE SYSTEM

care system. It does not take strong argument tosee that a collective-socialist society will favorpublic health care programs and that private pro-grams will more likely find favor in free-marketeconomies. Political decisions will also be influ-enced by the relative political power and ideologyof health care providers, consumers, and bureau-crats. The United States has a deserved reputa-tion for favoring free-market economics, whereasthe U.K. system has a more socialist character toits free-market economic system.

Lastly, the decision-making process itself willaffect the potential for change. The U.S. tripartitepower sharing among the President, Congress,and the courts makes change more difficult thanin a parliamentary government (e.g., as practicedin the United Kingdom), or, even more clearly,than in an authoritarian government.

As I argue here, health care systems are shapedby important external nonmedical factors. Healthis fundamentally important to human existence.Designers of health systems can expect almostconstant public pressure to “do more.” What canbe done, and what needs to be done, then, isshaped by the social ethics, economics, and pol-itics of each nation. These differences among na-tions make it certain that no “one-size-fits-all”health care system solution exists. External fac-tors are too important and differ too much amongnations for a detailed ideal health care templateto exist.

3. GOALS FOR HEALTH CARE SYSTEMS

As means to an end of good health, those respon-sible for stewardship of health care systems adoptgoals for their work. These goals almost invariablycontain elements related to the same three crite-ria promulgated on by Hammurabi: cost, quality,and access. These three goals are “ideals” thatcitizens and those who steward health care sys-tems debate and strive to reach. Actual healthcare system performance quite often falls short offulfilling these goals.1 One major reason for aperformance gap is that assessing and measuring

cost, quality, and access as well as their interre-lationships and trade-offs are quite difficult.

In 1978, in its “Declaration of Alma-Ata,”2

WHO developed a goal of “Health for All” by theyear 2000: “delivery to all of high-quality essen-tial care, defined by criteria of: effectiveness, cost,and social acceptability.” While this goal has notyet been reached, it has been integrated intoWHO’s work in developing nations and continuesto serve as its vision. And, very important to ourpurposes, this goal serves as the basis for WHO’sresearch into better understanding and measur-ing the performance of health care systemsaround the world.

Two other examples drawn from developed na-tions with very different health care systems arehelpful to this exploration. The two nations arethe United States, which is characterized by itslarge private health care sector, and the UnitedKingdom, with its almost exclusive public healthcare sector.

There is no “official” U.S. goal. However, areading of various government and private re-ports from all parts of the political spectrum re-veals a generally shared goal for “universal accessto high-quality, comprehensive, cost-effectivehealth care” (Bolnick 1995). The U.K. NHI doeshave a stated goal. It is to provide “comprehen-sive, high-quality medical care to all citizens on abasis of meeting professionally judged medicalneeds and without financial barriers to access”(Institute of Directors 2000). These are two vastlydifferent systems in terms of structure and per-formance and, yet, their goals consist of quitesimilar criteria for cost, quality, and access.

Our discussion to this point has uncovered anumber of important facts.

● Health is of vital importance to each of us;therefore, health care is important to the “bodypolitic” and national government.

● Stewardship of the health care system is a vitalgovernment function.

● All health care systems have cost, quality, andaccess goals.

● Specific characteristics of each national health

1 There is a particularly large gap in the United States between goalsand performance. This gap reflects a particular social, economic, andpolitical environment. See (Bolnick 1995) for a discussion of thedifferences between U.S. goals and performance and reasons for thegap.

2 The Declaration of Alma-Ata was adopted at the Joint WHO/UNICEFInternational Conference on Primary Health Care held in Alma-Ata,U.S.S.R. (now Almaty, Kazakhstan) in 1978.


care system are shaped by its personal andsocial ethics, economics, and political drivers.

4. A CHANGING MIX OF PUBLIC-PRIVATE

FINANCING

In the 1990s, and most notably under the leader-ship of Director-General Gro Harlem Brundtland,M.D. (1998–present), WHO has shifted its objec-tives for realizing the ambitious Declaration ofAlma-Ata toward a “new universalism.” Figure 1is an overview of how goals for health care sys-tems have evolved from expansive “classical uni-versalism” of the Declaration of Alma-Alta intothe new universalism.

New universalism adopts two important themesfor national health care systems. First, it recog-nizes that it is not possible or realistic to have asa goal provision of all possible health care foreveryone. As the potential array of health careservices grows, and the cost of providing theseservices to everyone increases faster than eco-nomic growth, governments increasingly reachfiscal limits on services they can afford to deliver.

Second, WHO recognizes that, as governmentfunded health care systems are forced to contractrelative to a goal of providing all services for ev-eryone, the private sector increasingly plays animportant role in providing for citizens’ healthcare demands. If the first priority is to provideservices to all, then not all health care servicescan be provided. An acceptable, less comprehen-sive standard must be found.

WHO is not alone in incorporating a solid doseof health care realism into its goals and objec-tives. For example, European Union (EU) mem-ber nations have historically depended on publichealth care systems funded by the state (taxes) orsocial insurance premiums and made available to

all citizens regardless of their ability to pay (clas-sical universalism). Fiscal constraint is making itmore difficult to continue this policy. While vol-untary (private) health insurance (VHI) has notplayed a big role in most EU nations, policymak-ers are increasingly looking at VHI as a possiblemeans to take pressure off public programs (Mos-sialos and Thomson 20013).

The EU’s slow move towards health systemswith more private financing (and often privateprovision of health services) is aided by the 1994third nonlife directive from the EU Commission.This directive establishes a single market for VHIin the EU and abolishes national controls on pre-miums and policy conditions in order to increasecompetition and consumer acceptance. More re-cently, the EU Commission is moving toward anotion that the EU is a single health care marketwith need for freedom of movement across bor-ders and, therefore, ultimately a single healthcare system encompassing the entire community.

5. HEALTH SYSTEMS STRUCTURE AND

MEASURABLE OBJECTIVES

All health care systems must perform an array offunctions designed to measure objectives, the ul-timate objective of a health care system beinghealth itself. This section describes a frameworkand measurable objectives developed by WHOand presented in its report (WHO 2000). Figure 2summarizes these functions and objectives andthe interactions among them.

To work effectively, a health care system mustperform four high level functions. Delivering ser-vices is the most visible of these functions. How-ever, in order to deliver services, a health systemmust first be effective at creating resourcesthrough investment and training. Funds for re-source creation and service delivery are an essen-tial lubricant; therefore, health systems must alsoarrange financing. Funds must be collected,pooled, and ultimately used to purchase neededgoods and services.

The concept of stewardship occupies a distinc-tive place in this functional overview. Steward-ship is defined by WHO (2000) as a “function of

3 This paper provides a valuable comprehensive study of VHI in theEuropean Union.

Figure 1WHO’s Changing Notions of Coverage and

Scope of a Health Care System

Source: WHO (2000), p. 15.


government responsible for the welfare of thepopulation, and concerned about the trust andlegitimacy with which its activities are viewed bythe citizenry” (p. 119). Economic activities offinancing, creating resources and delivering ser-vices each might take place in either the public orprivate sectors.

Stewardship is best understood as a broad con-cept. A national health care system might chooseto centralize stewardship in a single governmentagency, such as its health ministry. Alternatively,stewardship might be shared among variousbranches or levels of government. Governmentalso might delegate some stewardship functionsto the private sector. At an extreme, a govern-ment decision to leave health care solely to theprivate sector, with no government involvementor oversight, can be understood as “stewardship”based on faith in private-market solutions to so-cial issues. Even in this extreme case, govern-ment is ultimately responsible to its citizenry forthe performance of its health care system throughthe nation’s political system.

Health care systems function to provide forpopulation health needs. Fulfilling these needs isthe system’s objective. Health is clearly the mostbasic objective. WHO (2000) notes:

“. . . while improving health is clearly the mainobjective of a health system, it is not the only one.The objective of good health itself is really two-fold: the best attainable average level—good-ness—and the smallest feasible differences amongindividuals and groups—fairness. Goodness meansa health system responding to what people expectof it; fairness means it responds equally well toeveryone, without discrimination” (p. xi).

Based on this notion of how health care systemsperform, WHO adopts three fundamental objec-tives. These are:

● Improvement of the health of the populationserved.

● Responsiveness to citizen’s expectations oftheir “needs” and “wants.”

● Financial fairness in providing protectionagainst the costs of poor health.

WHO researchers then developed measures ofgoodness (level) and fairness (variability) foreach of these three objectives and gathered datato create a ranking for each of its 191 membernations. Levels attained on these measures andnational rankings are reported in the StatisticalAnnex to the report (WHO 2000, Tables 1, 5, 6,and 7).

WHO ranks nations on two overall measuresand an alternative measure of health level:

● Overall Attainment of Objectives is a weightedaverage measure of attainment of goodness andfairness objectives for each nation.

● Overall Health System Performance is aweighted average measure of attainment versuswhat each system should be able to accomplishgiven their health resources.

● Health Performance is a measure of health at-tainment—disability adjusted life expectancy(DALE)—versus what each system should beable to accomplish given its health resources.

Detailed information about these measures canbe found in the report, which is available fordownload on the WHO Web site at http://www.who.org.

The following analysis uses WHO’s objectivemeasures and rankings to analyze existing healthcare systems and help us to develop a frameworkfor an ideal health care system. Given the impor-tance of these data to our exploration, we mustrecognize that WHO has been both praised andcriticized for its work. Some of its measures arewell developed and well known to health policyanalysts, DALE being the most important amongthem (Murray and Lopez 1996). Others, particu-larly measures of responsiveness and financialfairness, are new, and data underlying specificmeasures are sometimes scarce or virtually non-existent.

WHO has made a commitment to continual

Figure 2Health Care Systems Overview

Source: WHO (2000).


improvements in its measurements and datagathering techniques. This commitment is beingwidely praised. However, WHO has also been crit-icized for the lack of scientific, totally objectivedata for some of its new measures, and the valueof its various national rankings have been ques-tioned (Williams 2001 and Almeida et al. 2001).4

6. WORLD HEALTH CARE SYSTEMS

PERFORMANCE

WHO data can be used to demonstrate usefullessons about spending and performance ofworldwide health care systems. It should come asno surprise that health spending varies enor-mously across the world. Using as our measurehealth spending as a percentage of gross domesticproduct (GDP), Figure 3 demonstrates this point.

The United States is the world champion healthcare spender, devoting 13.7% of its GDP. This isfully 3.2% of GDP higher than the second-placespending habits of Germany (10.5% of GDP) and5.5% of GDP higher than the 8.2% average of all 15EU member nations. For comparison purposes, U.K.health care spending is near the bottom of EU na-tions at 6.8% of GDP. Spending by the 29 Organiza-tion for Economic Cooperation and Development(OECD) member nations, which ranges from 6% to10% of GDP, is an even broader indicator of devel-oped nations’ financial commitment to health care.

At the other extreme of health care spending arepoor nations, particularly those in Sub-Sahara Af-rica where tropical diseases ravage the population.The two nations that spend the least as a percentageof their GDP towards health care are Nigeria (3.1%of GDP) and Niger (3.5% of GDP). Their spendingdeficiency is even more obvious when translatedinto U.S. Dollars at the official exchange rate. Usingthis measure, Nigeria spends $30 U.S. per personper year and Niger, only $5 U.S. while the UnitedStates spends $4,187 U.S., U.K. spends $1,303 U.S.,and the E.U. averages $1,868 U.S.

WHO researchers use DALE as their most basicmeasure of health. DALE is a variation on themore familiar measure of overall population life

expectancy (Murray and Lopez 1996). Ratherthan simply measuring survival from birth todeath, DALE reduces life expectancy by a mea-sure of years lost to disability and infirmary. Thisgives an overall population measure of averageyears of disability-free life from birth.

DALE is a useful and valid means of measuringboth the absolute and relative health of nationalpopulations. As a measure of population health,DALE captures the effectiveness of a nation’shealth care system as well as its public healthprograms, social and economic policies, and pop-ulation’s personal health habits. Figure 4 presentsDALEs for nations with a DALE of 70.0 years ormore and the three EU member nations that falljust short of this level of health.

The world’s healthiest overall population is Ja-pan’s, with 74.5 years DALE. Twenty-four nationshave a DALE of 70.0 years,5 which can be con-sidered as a “world-class” health outcome. TheUnited States ranks No. 24 at 70.0 years, theUnited Kingdom ranks No. 14 at 71.7 years, andthe 15 EU member nations average 71.4 yearsDALE. At the other extreme, the world’s leasthealthy populations are found in Sub-Sahara Af-rica: Sierra Leone ranks No. 191 at 25.9 years andNiger ranks No. 190 at 29.1 years DALE.

There are vast differences in DALE between thehealthiest and least healthy nations of the worldcommunity. Most of the difference is explained bysurvival rates and not by years alive with disabil-ity. In the healthiest nations, disability-adjust-ments reduce life expectancy by about 10%. Forexample, life expectancy at birth in Japan is 80.0years, compared to 74.5 years of DALE. In theleast healthy nations, this adjustment amounts to20% of life expectancy. Sierra Leone has 34.3years life expectancy versus 25.9 DALE.

The health profiles of populations in the healthi-est nations also are very different from those of theleast healthy nations. WHO measures this burden ofdisease using the number of disability-adjusted lifeyears (DALYs) lost to various diseases (Murray andLopez 1996). Figure 5 compares DALYs by categoryfor high-income versus low- and middle-income na-tions. Developed nation’s burden of disease is

4 Based on a review the WHO data and discussions of their accuracy andusefulness, I am comfortable that any problems do not materially affectthe conclusions drawn by their use in this paper. However, carefulreaders will want to draw their own conclusions based on their ownstudy of the WHO report and published criticism and commentary.

5 The choice of 70.0 years DALE as “world-class” is an arbitrary cutoffused for this paper. The U.S. rank as No. 24 at exactly 70.0 yearsDALE is a fortuitous consequence of this cutoff.


heavily caused by chronic diseases. Low- and mid-dle-income nations bear a severe burden of infec-tious disease and diseases of childbirth. Most of

these infectious diseases are endemic solely in trop-ical areas, or they are public health problems thateither do not exist or have been well managed in

Figure 3World-Class Health Care Systems’ Costs

Source: WHO (2000).Note: National rankings (in parentheses) are by health attainment, which WHO researchers measure by disability adjusted life expectancy (DALE).


Figure 4World-Class Health Attainment


high-income nations. In addition, low-income na-tions suffer from a very high burden due to injuriesrelative to the burden in middle- and high-incomenations. Developing nations’ burden of disease is

heavily weighted toward infectious disease, diseasesof childbirth, and accidents. There has been a greatdeal of thinking done about how to addressthis morbidity profile (Sachs 2001). Researchers

Figure 5Disease Burden


note that much of this disease burden is amenableto basic public health intervention. Clean water,sewage treatment, basic prenatal and post-natal care, accident prevention, and other simpleinterventions that we in developed nations take forgranted can greatly reduce these health problems.

Almost all developed nations have built theirhealth care systems on heavily public-fundedmodels. The major exceptions to this are theUnited States and South Africa. Figure 6 displaysdata on the size of public health care expendi-tures (as a percentage of total health care costs)compared to health outcomes (measured byDALE). If there were inherent advantages to pub-lic versus private health care systems, the datashould reveal a clear correlation between DALEand size of nations’ public sectors. A glance at thefigure, though, shows no such correlation.

While the mean for public health care spendingis greater than 50% of total spending, higher pub-lic percentages do not result in greater numbersof years spent free from illness. This is an impor-tant observation, since the serious political andpolicy question of public versus private healthcare systems cannot be answered solely by a pos-itive correlation with health outcomes.

The segment of data from the 24 “world-class”national health care systems, those with 70.0 ormore years of DALE, demonstrates this point. Atthe extremes are the United States, with 70.0 yearsDALE and 44.1% of total health care spending com-ing through the public sector; the United Kingdom,

with 71.7 DALE and 82.6% public spending; andLuxembourg, with 71.1 DALE and 91.4% publicspending (see Table 1 for a complete set of statisticson these systems).

This same wide spread in public versus privatespending is characteristic of the larger number ofdeveloping nations, generally those with DALEs be-tween 55.0 years and 70.0 years. In this group, therange is from a public-spending high of 99.9% (Kiri-bati) to lows of 20.3% (Uruguay) and 24.9% (China).The last segment, poor nations, shows the samewide range of public versus private spending for asimilarly wide range of outcomes.

Perhaps the most revealing WHO data is pre-sented in Figure 7, which compares the averagetotal per capita health expenditure (in U.S. dol-lars) to DALEs. A clear pattern emerges from thiscomparison: Regardless of the “publicness” or“privateness” of a nation’s health care system,total spending of about $1,000 per capita (1997)is enough to provide a “world-class” health out-come of 70.0 years DALE.

In addition, by focusing on the portion of Figure 7representing nations with 70.0 or more years ofDALE, we see that spending greater than $1,000 percapita does not continue to improve a population’shealth. Our world champion spender—the UnitedStates at $4,187 per capita per year—actually hasslightly worse population health than the UnitedKingdom ($1,528 per capita) does or than the EUaverage ($1,868 per capita). This begs the questionsof why big-spender nations invest extra money onhealth care and on what it is being spent (see Table1 for a complete set of statistics).

This overview of data from the WHO (2000)report and our previous discussion of externalfactors affecting characteristics of national healthcare systems demonstrate support for a numberof important facts:

● Basic health care and public health spending cancontrol infectious disease, problems of childbirth,and injuries that create a disproportionately largeburden of disease in developing nations.

● Developed nations’ burden of disease is heavilycaused by chronic diseases.

● All effective health care systems are, by politi-cal choice, mixed public-private systems;

● Spending about $1,000 (1997) on health care isneeded to produce a “world-class” health out-come (at least 70.0 years DALE).

Figure 6Public Expenditure as % of Total Health

Care Spending


● Greater public health care spending, by itself, isnot correlated with better health (higher DALE).

● Spending more does not necessarily continue toimprove health outcomes.

7. WHO COMPARISON OF HEALTH CARE

SYSTEMS

The WHO (2000) report contains three differentrankings of the world’s 191 national health caresystems. For each objective of health, responsive-ness and financial fairness, WHO has developedmeasures of population average (goodness) anddifferences among individuals and groups (fair-ness). These individual measures are thenweighted and ranked into three overall measures.

One ranking is based on measures of what na-tions have actually been able to achieve withrespect to the three objectives (overall attain-ment). Two other rankings are based on whateach system is capable of achieving with availableresources (performance)—one measures the sin-gle dimension of health performance and theother measures overall performance on all threedimensions of achievement. Table 2 summarizesthese three WHO rankings (WHO 2000, Statisti-cal Annex Table 1).

Overall attainment ranks nations on the basis

Table 124 World-Class Health Care Systems and Other EU Member Nations, Ranked by DALE

Rank/Country DALEDisabled

Years

Total HealthSpending as% of GDP

(1997)

Total HealthSpendingPer Capita(US$ 1997)

Public Spendingas % of Total

Health Spending(1997)

1. Japan 74.5 6.4 7.1% $2,373 80.2%2. Australia 73.2 6.4 7.8 1,730 72.03. France 73.1 6.2 9.8 2,369 76.94. Sweden 73.0 6.5 9.2 2,456 78.05. Spain 72.8 6.0 8.0 1,071 70.66. Italy 72.7 6.1 9.3 1,855 57.17. Greece 72.5 5.5 8.0 905 65.88. Switzerland 72.5 6.8 10.1 3,564 69.39. Monaco 72.4 6.8 8.0 1,264 62.510. Andorra 72.3 6.6 7.5 1,368 86.711. San Marino 72.3 6.9 7.5 2,257 73.512. Canada 72.0 7.0 8.6 1,783 72.013. Netherlands 72.0 6.6 8.8 2,041 70.714. U.K. 71.7 5.5 5.8 1,303 96.615. Norway 71.7 6.9 6.5 2,283 82.016. Belgium 71.6 6.3 8.0 1,918 83.217. Austria 71.6 5.8 9.0 2,277 67.318. Luxembourg 71.1 6.9 6.6 2,580 91.419. Iceland 70.8 7.5 7.9 2,149 83.820. Finland 70.5 6.6 7.6 1,789 73.721. Malta 70.5 7.8 6.3 551 58.922. Germany 70.4 6.5 10.5 2,713 77.523. Israel 70.4 6.7 8.2 1,385 75.024. U.S.A. 70.0 6.7 13.7 4,187 44.127. Ireland 69.6 6.7 6.2 1,326 77.328. Denmark 69.4 6.7 8.0 2,574 84.329. Portugal 69.3 7.0 8.2 845 57.5

Source: WHO (2000, Statistical Annex Tables 5 and 8).

Figure 7Per Capita Health Expenditure (US$), 1997


of what they have been able to achieve with re-spect to health, responsiveness, and financial fair-ness. Weightings are 50% health, 25% responsive-ness, and 25% financial fairness. The health andresponsiveness measures are each based on equalweighting of measures of goodness (average) andfairness (distribution).

Overall performance uses the same measuresand weighting as overall attainment: But, ratherthan measuring what each nation has actuallyaccomplished, this ranking measures what na-tional health systems have attained versus theirtheoretical efficient frontier given available re-sources.

WHO’s last overall system measure is healthperformance. This ranking focuses solely onhealth. It measures actual health attainment ver-sus what a system is capable of attaining based onits available resources. Health performance dif-fers from health attainment, which uses DALE torank nations’ health. Health attainment, pre-sented in Figure 4, was discussed previously.

To compare health care systems, WHO devel-ops and uses measures of health, responsiveness,and financial fairness. These measures do notcorrespond directly to the widely accepted cost,quality, and access goals discussed previously.WHO’s approach recognizes the difficulty of ade-quately defining and directly measuring cost,quality, and access. Health, responsiveness, andfinancial fairness are definable and measurable;however, using measures that do not directly re-flect goals of cost, quality, and access is one rea-son for criticism of WHO’s measures and rank-ings.

There is, though, a strong, but not always di-rect, relationship between WHO’s measures anduniversally pursued goals:

● Cost is indirectly captured in the two perfor-mance measures of overall performance andhealth performance, which compare systems’attainments with their theoretical efficientfrontiers. Performance is a measure of how wella system does considering the nation’s healthcare resources. These resources are a functionof spending. Nations with higher performancerankings can be understood as receiving morebenefits for their spending than do lower rank-ing nations.

● Quality is more directly measured. Health(DALE) and responsiveness directly measuregoodness and fairness of fulfilling health careneeds and wants.

● Access is measured indirectly, mainly by health(DALE) and fairness (variability) of health, re-sponsiveness and financial fairness: More equalaccess to health care is reflected in lower vari-ance in these three measures; and, low variancecoupled with higher DALE measures better ac-cess to higher quality health care.

Thus, data from the report could be combined tomeasure more directly nations’ relative perfor-mance in reaching cost, quality, and access goalsas measured by WHO standards.

8. COMPARISON OF HIGHLY EFFECTIVE

HEALTH CARE SYSTEMS

Detailed measures of health, responsiveness, andfinancial fairness are very helpful to better under-stand differences among national health care sys-tems. Our search for a path toward an ideal healthcare system uses this data to focus on differencesamong world-class health care systems, definedfor this paper as systems in nations with DALE of70.0 or more years. This provides us with up to 24national systems to study.

Fortunately, what we need to understand canbe demonstrated by looking more closely at twoof these 24 national health care systems: those ofthe United Kingdom and the United States. Thesetwo systems closely represent the extremes. TheUnited States has the largest private-sector sys-tem and the United Kingdom one of the largestpublic-sector systems.

Table 2Health Care Systems Rankings

HealthPerformance

OverallPerformance

OverallAttainment

1. Oman 1. France 1. Japan2. Malta 2. Italy 2. Switzerland3. Italy 3. San Marino 3. Norway4. France 4. Andorra 4. Sweden5. San Marino 5. Malta 5. Luxembourg6. Spain 6. Singapore 6. France7. Andorra 7. Spain 7. Canada8. Jamaica 8. Oman 8. Netherlands9. Japan 9. Austria 9. United Kingdom10. Saudi Arabia 10. Japan 10. Austria24. United Kingdom 18. United Kingdom 15. United States72. United States 37. United States

Source: WHO (2000).


The following discussion is descriptive in na-ture; it is not intended as political commentary orvalue judgment on either the U.S. or U.K. healthcare systems. As I mentioned previously, the re-sources a country devotes to health care and theshape of its system are driven by personal impor-tance of health, its social ethic, and its economicsand politics. Differences between the U.S. andU.K. systems stem largely from differences inthese external factors. My purpose is solely tounderstand these differences and not to judge therelative merits of each system.

Figure 8 summarizes rankings6 and detailedmeasures of health, responsiveness, and financialfairness in the United States and the United King-dom and compares them with an average of all EUmember nations. The EU average is a helpfulpoint of comparison, since all EU members ex-cept Ireland (69.6), Denmark (69.4), and Portugal(69.3) qualify as world-class health care systems.

A very short summary of the U.S. and U.K.systems will suffice for our purposes. Readers notfamiliar with either or both of these health caresystems are referred to the excellent summariesof a wide range of world health care systemsprepared by the European Observatory on HealthCare Systems. These summaries can be accessedon its Web site at www.observatory.dk.

Health care in the United States is deliveredalmost exclusively by private sector providers.Many hospitals are owned and operated by for-profit companies; others are not-for-profit orcharitable institutions. Physicians and otherhealth care workers are either independent prac-titioners or work for private-sector health careinstitutions.

For most Americans, access to health care pro-viders is gained through employment-based pri-vate health insurance. Health care for the twolargest groups of nonworkers—the nonworkingpoor and retired citizens—is financed throughtwo large government social insurance programs:Medicaid for the poor (jointly funded by state andfederal general revenue) and Medicare for the

retired population (funded by a federal payroll taxand general revenue).

Through the combination of private health in-surance, Medicaid, and Medicare programs, about84% of the population is covered. The remaining16% of the population, composed mainly of theworking poor and their families, must depend ontheir own (usually inadequate) financial re-sources or charity care to pay for needed medicalcare. Needless to say, the level of medical care foruninsured Americans is often substandard andtheir health is often compromised by lack of ad-equate care (Institute of Medicine 2002).

The U.S. mix of public and private sector re-sponsibilities has its characteristic strengths andweaknesses. At 70.0 years DALE, the UnitedStates is one of the nations with “world-class”health outcomes. The other outstanding charac-teristic of the U.S. system is its responsiveness: Itranks No. 1 in the WHO survey.

Understanding responsiveness is very importantto understanding differences among world-classhealth care systems. The WHO measurement ofresponsiveness is based on surveys designed tomeasure the following (WHO 2000, p. 32):

● Respect for persons: respect for dignity, confi-dentiality, and autonomy.

● Client orientation: prompt attention, quality ofamenities, access to social support networks,and choice of provider.

Survey questions used by WHO researchers to mea-sure responsiveness clearly point to attributes of

6 The rank for each measure in Figure 8 is the numerical rank of theUnited States and the United Kingdom out of 191 nations in theWHO database. See the Statistical Annex of the report for nationalrankings.

Figure 8Health Care Systems Performance


the system that make people feel good about theirencounters with providers and, yet, do not seemto affect their health.

Health Attainment (DALE) captures the effec-tiveness of a national health care system in satis-fying its population’s health care “needs”: HigherDALE is clearly associated with a healthier pop-ulation. Responsiveness measures capture the in-tangible health care “wants” that arise from theimportance of health care to individuals’ sense ofwell-being: For a given level of health attainment(DALE), a health care system that is more client-oriented and patient-centered will rank higher inresponsiveness. This point is demonstrated by acomparison of the strengths and weaknesses ofthe U.S. and U.K. health care systems in the fol-lowing sections.

Distinguishing “wants” from “needs” implies ahierarchy that values national health care sys-tems’ addressing their citizens’ health care“needs” as more important than health care“wants.” This distinction does not imply that ful-filling “wants” is not an important goal for manynational health care systems. Countries that suc-cessfully fulfill their citizens’ health care “needs”uniformly find themselves facing demands for ad-dressing health care “wants.”

While the U.S. system has world-class healthresults, it actually ranks low in overall attainment(at No. 15). There are three main reasons for thisrelatively poor performance.

● First, private health care financing, combinedwith Americans’ high expectations for theirhealth care system’s responsiveness, is an ex-pensive mix of incentives. The U.S. health caresystem is by far the most expensive in theworld, ranking No. 1 in cost in the WHO data-base. And, it costs almost twice as much as theU.K. system (measured as a percentage of GDP).While I know of no studies that directly addressthis observation, anecdotal evidence and datapresented in this paper leads me to believe thatthe U.S. systems’ very high responsiveness toAmericans’ health care “wants” adds significantcost to the system.

● Second, the U.S. health care system is charac-terized by an incredible variation in qualityacross its population. At its best, the U.S. healthcare system is arguably No. 1 in the world—thehighest possible quality providers and facilities

offering state-of-the-art care with extremelyhigh responsiveness to patients’ needs and theirnonmedical wants. At its worst, the system isinaccessible to those without health insuranceor without easy access to other than low-qualityproviders and facilities. There is actually aThird-World health care system buried deepwithin the United States.

● Third, the U.S. system ranks very low in finan-cial fairness. This results directly from theWHO (2000) definition of “financial fairness”emphasizing that “. . . the risks each householdfaces due to the cost of the health care systemare distributed according to ability to payrather than to the risk of illness” (p. 35). With asystem heavily funded by risk-based privatehealth insurance premiums, and with a largenumber of uninsured Americans who are ex-tremely vulnerable to the high cost of medicalcare, a low standing for the United States onWHO’s measure of financial fairness is no sur-prise.

In contrast to the United States, the U.K.’s NHIsystem is a federal tax funded health care system.All U.K. citizens have the right to obtain theirhealth care, at no direct cost to them, throughphysicians that are salaried employees of the NHIand hospitals owned and operated by the govern-ment. The amount of money spent on health isbudgeted by Parliament.

NHI and it’s constituent pieces determine howtheir budget is to be spent. In practice, this meansthat there are waiting lists for certain nonemer-gency services. While all U.K. citizens have accessto NHI physicians and hospitals, 11.5% of thepopulation bought voluntary health insurance(VHI) in 1998 (Mossialos and Thomson 2001).VHI actually duplicates coverage provided byNHI; however, it provides almost immediate ac-cess to physicians and hospitals, thereby bypass-ing NHI waiting lists (queue skipping).

Not at all surprising, the U.K. health care sys-tem has very different strengths and weaknessesfrom those of the U.S. system (see Figure 8). TheU.K. system has far less variation in health out-comes across its population that does the UnitedStates, and it ranks much higher than the U.S.system with respect to financial fairness. Thisoutcome is a direct result of the U.K. federal


tax-based system versus the private risk-basedfinancing in the U.S. system. In addition, the U.K.system is very low cost when compared both toother developed nations’ systems and, most con-spicuously, to the extremely high-cost U.S. sys-tem.

WHO’s measures demonstrate that the U.K.health care system has world-class health out-comes, with little variation across its population,and high marks for financial fairness. There is,though, one large negative to this system: It re-ceives relatively low marks for its responsiveness.While the financially unconstrained U.S. systemis quite responsive to its citizens “needs” and“wants,” the financially constrained U.K. systemis much less responsive. With a population DALEof 71.7 years, it is hard to argue that the UnitedKingdom does not meet it’s citizens health“needs” as well as any health system in the world.However, tight control of funding and health re-sources means that it’s citizens more intangible“wants” are constrained, resulting in waiting listsfor nonessential medical care and low marks fromWHO for responsiveness.

Table 3 provides us with an interesting viewfrom within three world-class health care sys-tems: the United States, the United Kingdom, andCanada. The Chart summarizes opinion surveyson characteristics of the three countries’ healthcare systems obtained from a multinational pop-ulation survey (Donelan 1999). Significant differ-ences among systems are highlighted in bold.

In all three systems, people share very similar

opinions that their own overall medical care isvery good to excellent and express confidence inreceiving the most advanced medical care. How-ever, while U.K. citizens express a relativelystrong support for their system, Canadians andAmericans are far more likely to be critical oftheir systems.

When asked about direct experiences withhealth care providers and financing, there aredistinct differences in responses across surveyedcountries that are consistent with what we havelearned from the WHO rankings discussed above.U.K. survey respondents report virtually no worryabout paying medical bills. They do, though, re-port long waits for surgery and short physicianoffice visits. These results refer to responsivenessissues and not to truly needed medical care. Incontrast, U.S. respondents report no wait for careand longer office visits while worrying much moreabout paying for their care.

By delving more deeply into the U.S. and U.K.health care systems, we gain perspective on thecharacteristics of world-class systems. This anal-ysis demonstrates support for a number of impor-tant facts and observations:

● Developed nations with world-class health out-comes (DALE � 70.0 years), which includeboth U.S. and U.K. health care systems, allspend “enough” on health care to satisfy theirpopulations’ health care “needs.”

● Additional spending on health care tends toimprove responsiveness (i.e., “wants”) morethan health outcomes (i.e., “needs”).

● Health is so essential and sensitive to individu-als that even the best health care systems havedifficulty managing their public’s concerns andexpectations.

● Public programs like those in the United King-dom are increasingly having financial con-straints that make it difficult to provide for theircitizens’ health care “wants.”

A growing demand for fulfilling health care“wants,” which can be quite expensive, points toa developing role for private health insurance as a“safety valve,” providing for those health care“wants” that exceed real health care “needs.”This last point will be more fully explored in theremainder of this paper.

Table 3Public Feedback

U.K. U.S. Canada

Health care system:Works well 25% 17% 20%Needs complete rebuilding 14 33 23

Worried about receiving mostadvanced medical care 46 47 63

Worried about paying for neededmedical care 37 47 45

Report wait of four� months forelective surgery 33 1 12

Report having problems payingmedical bills 3 18 5

Report last M.D. visit as 10minutes or less 65 30 33

Believe overall medical care wasexcellent/very good 50 49 54

Source: Donelan et al. (1999).


9. PUBLIC OR PRIVATE HEALTH FINANCING:WHICH APPROACH WORKS BEST?

Designing a world-class health care system inevita-bly requires choosing between or mixing public andprivate approaches, most particularly in financing.Which choice or mix makes most sense depends inlarge part on the characteristics of public and pri-vate health care financing. Understanding the sig-nificant differences in the performance of publicand private markets requires a digression into theeconomic of insurance markets.

Private insurance markets have been widelystudied by some of this era’s prominent econo-mists, including Nobel Laureates Kenneth Arrow,Milton Friedman, Joseph Stiglitz, and Daniel Kah-nemann. Economists’ interest in insurance mar-kets stem from three main concerns:

● First, private insurance markets behave in avery strange economic manner. They are char-acterized by information asymmetry, that is, ameaningful difference in information betweenbuyers and sellers. Stiglitz won the 2001 NobelPrize in Economics for his work in describingmarkets with asymmetric information (seeRothschild and Stiglitz 1976). In the case ofinsurance, asymmetric information has to dowith risk characteristics. Obviously, buyers(consumers) know much more about their risksthan do sellers (insurance companies). How in-formation asymmetries affect markets is ofgreat interest not only to economists, but alsoto actuaries who must understand and managethe consequences of market anomalies.

● Second, economists are interested in under-standing if a good or service is best produced inthe public sector or private sector. Insurance(protection) is one service that is commonlyfound in both sectors. So, economists turn theirattention to understanding the relative meritsof public versus private insurance. Financinghealth care is clearly one of the main areas oftheir exploration. Arrow (1963), another NobelPrize winner, is known for his work in this area.

● Third, economists have been interested in howindividuals and businesses make risky deci-sions. Friedman and Kahneman, more recently,explored this subject using insurance purchaseas one of their considerations (see Friedmanand Savage 1948 and Kahneman and Tversky1979).

As Arrow and Stiglitz both inform us, private-voluntary insurance markets are characterizedby “market failure”7: incomplete insurance dueto the inability of these markets to shift all therisk of any economically relevant event (Arrow1971). This idea should sit well with actuaries,since we are all aware of it from our daily expe-riences of uninsurable risks and the need for costsharing in insurance contracts.

In their seminal work on markets with asym-metric information, economists Michael Roths-child and Joseph Stiglitz (1976) describe marketfailure in even more disturbing terms:

“Economists generally prescribe competition as asolution for markets that do not work well . . . .Insurance markets differ from most other marketsbecause in insurance markets competition candestroy the market rather than make it work bet-ter” (p. 73).

Understanding the market dynamic that leadRothschild and Stiglitz to make this very strongand truthful statement is beyond the scope of thispaper.

Economists describe two underlying reasonsfor market failure: (1) adverse selection, arisingfrom asymmetric information, and (2) moral haz-ard. Actuaries are quite familiar with both ofthese problems and work daily on managing theirconsequences.

Adverse selection occurs when bad risks buyinsurance and good risks choose not to purchaseinsurance. Adverse selection can occur if the in-surance company cannot distinguish good risksfrom bad risks. The consequences depend on de-gree. Some well-known consequences of adverseselection range from higher premiums for insuredrisks when a product or market is subject to amild degree of adverse selection to complete fail-ure of a product, insurer, or market when a highdegree of adverse selection develops. Actuarieshave developed sound risk management tools to

7 Market failure is used by economists to describe a characteristic thatprivate-voluntary insurance markets often do not shift all risks thatconsumers want to insure. For example, while not usually thought ofas a problem, deductibles, copayments, limitations, exclusions, andbenefit maximums all are examples of economic market failures.Examples of economic market failure that fit better with non-eco-nomic understanding are the non-existence of a market for insur-ance, or the collapse of an existing market.


cope with adverse selection. These include riskselection (underwriting), risk rating, and variousrisk renewal techniques (policy terms and rates).

Moral hazard is the other underlying marketmalady. Moral hazard occurs because “. . . (T)heinsurance policy itself might change the incen-tives and therefore the probabilities upon whichthe insurance company has relied” (Arrow 1971,p. 142). This too is a very familiar problem toactuaries. We have developed risk managementtools such as underwriting benefit limits in rela-tion to risk, deductibles and copayments, andmanaged care to control the escalating frequencyand severity of health insurance claims.

Without effective risk management tools, theRothschild and Stiglitz (1976) warning of compe-tition destroying insurance markets would be farmore pervasive than we actually experience.However, even with effective risk managementtools, we will see that adverse selection and moralhazard characterize and constrain performanceof private insurance markets when compared topublic insurance systems.

9.1 Private-Voluntary InsuranceMarkets

Private-voluntary insurance markets are charac-terized by choice. Consumers have completefreedom to buy or not to buy insurance. Multiplecompeting insurers, too, have their own choicesof whether or not to offer insurance and muchfreedom over their terms of coverage. Private-voluntary markets can be illustrated by describ-ing a well-behaved market situation with no neg-ative consequences of adverse selection or moralhazard. In this well-behaved state, a subset of theuniverse of potential consumers chooses to buyinsurance from all insurers competing for its busi-ness, and, a subset of those consumers choosingto buy insurance chooses coverage from insurerA. Under these ideal conditions, average loss fre-quency and severity is the same for each of ourthree sets: Insurer A has the same average lossexperience as all insured customers (all insurers)who, in turn, have the same average losses as theentire universe of potential insureds.

However, actuaries will quickly recognize theideal nature of this depiction. It is quite likely thatthe consequences of adverse selection and moralhazard will cause the losses for all insureds to be

higher than those of the universe of potentialcustomers. And, even more likely, a single insurercompeting in the market may have its own lossdiffer, perhaps significantly, from overall marketlosses and from those of other insurers. Thus, afamiliar dynamic of private-voluntary marketscan be summarized as follows:

● As a consequence of choice, adverse selectionand moral hazard naturally arise in private-voluntary markets to lesser or greater degree.

● Choice causes uncertainty about the level ofrisk borne by all insurers.

● Choice causes uncertainty about the level ofrisk borne by each insurer that competes in themarket.

● Competition among insurers, adverse selection,moral hazard, and buyers’ pursuit of their ownbest interests in making their insurance pur-chasing decisions shape market behavior.

To compensate for potential negative conse-quences of adverse selection and moral hazard,insurance companies must develop and use effec-tive risk management tools. These tools includerisk selection, risk rating, and renewal risk man-agement. However, risk management tools oftencause problems for bad risks:

● Bad risks may not be able to obtain insurance,making universal coverage impossible to attain(risk selection).

● Bad risks may pay more for their insurancethan good risks do, making the sick pay morefor their insurance than the well (risk rating).

● Bad risks may be treated differently at renewalor have restrictions on moving from one insurerto another, leading to charges of heavy-handedrating practices by insurers and problems suchas “job lock,” where employees cannot movefrom one job to another without the threat oflosing their health insurance (renewal riskmanagement).

Given strong individual and societal sensitivitiesto health and health care, the negative conse-quences of these necessary private-market riskmanagement tools often are viewed as unaccept-able behavior on the part of insurers. The public,press, and politicians focus their critical com-ments on the natural negative consequences ofprivate-voluntary market behavior and not on the


underlying market structure that causes thissymptomatic behavior.

Growing concerns can lead to a loss of publicconfidence in health insurers, spirited media andpolitical charges against insurance company be-havior, and, finally, to regulation of health insur-ers and health insurance products. Practices thatare tolerated in other less sensitive insurancemarkets (e.g., risk selection and risk rating for lifeinsurance) are not long tolerated in health insur-ance markets due to the particular sensitivity of“health” as an insured risk.8

9.2 Public-Mandatory InsuranceMarkets

In stark contrast to the nature of private-volun-tary insurance markets, public-mandatory mar-kets have quite different characteristics. Public-mandatory markets are characterized by a lack ofchoice. This type of market arises when a govern-ment uses its power to mandate that the entireuniverse of potential consumers be covered forinsurance and arranges financing and insurancein publicly acceptable ways to accomplish thisobjective. The U.K.’s NHI and the U.S.’s Medicareprogram are two examples of successful public-mandatory markets.

The key difference between private-voluntarymarkets and public-mandatory markets is that byeliminating choice, public-mandatory social in-surance programs eliminate adverse selection.When the entire universe of risks is covered in asingle risk pool, uncertainty about the level of riskis caused by choice is totally eliminated. Moralhazard, though, is not eliminated in public pro-grams. It continues to exist as a cause of marketfailure.

The consequence of public-mandatory marketsis that programs using this structure do not needto develop the wide range of risk managementtools that characterize private-voluntary markets.

Risk selection is not necessary due to the man-datory nature of these programs. Risk rating isunnecessary since government can use its taxingpowers and other tools to raise funds and does notneed to be concerned with the adverse selectionconsequences of actuarial rating systems. And, bysetting up a single risk pool that eliminates move-ment among competing risk pools, public pro-grams do not need renewal risk managementtools to compensate for adverse selection.

Insurance market structure predictably drivesconsumer and insurance company behavior,which in turn results in the use by insurers ofcharacteristic risk management tools. Anotherway to look at this fundamental insurance marketdynamic is that actors in insurance markets arebound by the role dictated to them by marketstructure. To change the behavior of consumersor insurance companies fundamentally, govern-ment needs to create proper market structures.This is precisely how government exercises itsstewardship of health insurance markets.

Table 4 summarizes the strengths and weak-nesses of private-voluntary markets versus pub-lic-mandatory markets. In addition to the pointsdiscussed above, the table also captures informa-tion on the types of benefits (private markets aremore responsive), administrative costs (publicmarkets are less costly—there are no marketing,underwriting, or renewal costs), and quality ofadministration (public programs are bureaucratic).

8 Examples of how critically these risk management practices areviewed by the public are common in the U.S. press, particularly whenhealth care reform becomes a national political issue. Three particu-larly critical articles from the past decade are: “Unsurance,” by Con-stance Mathiessen (Hippocrates, November/December 1989, pp. 36–46); “Insurance: The Death Spiral,” by Jane Bryant Quinn,(Newsweek, Feb. 22, 1993, p. 47); and “Insurance Tactic: If You GetSick, The Premium Rises,” by Chad Terhune (Wall Street Journal, April9, 2002, p. 1).

Table 4Comparative Performance

Private-Voluntary System Public-Mandatory System

Strengths• Choice (responsiveness)

—Expansive “wants”benefits

• Available to mostcustomers willing to pay anequitable cost for coverage

• Private control ofdecisions over provisionof health care

Weaknesses• Universal coverage

impossible• Fragmentation of risk pool

—“Cherry-picking”—High sales and

administrative costs• Risk-rated premiums• Challenge to limit costs

Strengths• Universal coverage• Direct cost control

through governmentbudgets

• No market fragmentation• Tax-salary based financing

(financial fairness)• Low overhead costs

Weaknesses• Bureaucracy

(unresponsiveness)• Not likely to provide all

health care “wants”demanded by public

• Strong public involvementin provision of medicalcare services (may beconsidered a strength)


The private-voluntary and public-mandatorymarket structures as described in this sectiontake their simplest form. In reality, governmenthas the ability to regulate private-voluntary mar-kets so that they function more like public-man-datory markets and vice versa. Examples of theseregulated markets exist throughout the world.The value to us of discussing simple models is tohelp clearly understand the design and function-ing of more complicated market structures.

From this discussion, we glean a number ofimportant facts.

● Private-voluntary markets are not structured tobe capable of providing universal coverage ortax-based financing.

● In response to adverse selection and moral haz-ard, insurers and customers in private insur-ance markets will naturally exhibit a range ofunattractive market behavior.

● Public markets, or their structural equivalent,are needed to provide universal coverage and toallow for use of tax-based or other nonrisk-based funding.

● Private markets, subject to adequate steward-ship, are the most appropriate means to fundvoluntary health care benefits that satisfy“wants” rather than “needs.”

10. AN IDEAL HEALTH CARE SYSTEM

To this point we have discussed nonmedical sen-sitivities toward health care systems; comparedthe effectiveness of the world’s 191 nationalhealth care systems; delved in more detail intothe characteristics of two very different large,world-class systems (U.S. and U.K.); and exploredthe economics of how public and private insur-ance markets differ in their risk bearing capaci-ties and characteristic risk management tools.

These considerations were explored with theintention of providing information needed to ad-dress the primary goals of this paper: creating ahigh-level template for turning the goals dis-cussed in Section 3 into operating structures forthe stewardship and financing of an ideal world-class health care system.

Our structure is necessarily a high-level one.We must leave sufficient room for nations to in-corporate their own human sensitivities, politics,and ethics. These nonmedical human and soci-etal differences are incorporated, directly by gov-

ernment intervention or indirectly through func-tioning private markets, into the design of anyand all nations’ health care systems. There, then,cannot be a “one-size-fits-all” design for a world-class system. Despite these important restric-tions, there is much that we can say about de-signing world-class health care systems.

10.1 Principle 1Health care systems need enlightened steward-ship. The most basic lesson from our explorationis that government stewardship of health caresystems is necessary. We have discussed how im-portant nonmedical considerations and marketstructure are to a health care system. It is virtu-ally impossible to conceive how these matters canbe incorporated into a health care system withoutgovernment stewardship through its legislativeand executive functions.

The need for stewardship of public health careplans is clear. Stewardship is not, though, limitedto public programs. We have seen in our discus-sion of private-voluntary markets that, withoutenlightened regulation of health insurance (i.e.,government stewardship), universal access tomedical care is virtually unattainable. This is truebecause risk management tools needed to providemarket stability restrict access to insurance forbad risks and raise their risk-based premiums.Whether totally public, totally private or mixedpublic-private health care systems, nations needenlightened stewardship.

10.2 Principle 2The base of an ideal system is a universal healthcare program providing for citizens’ medicalcare “needs.” Universal access is consistent witha social ethic of social solidarity in providing atleast basic medical care services to all citizens.9

Given the market failure problems that charac-terize private-voluntary markets, there are onlytwo potential market structures that can provideuniversal coverage:

9 Debate over universal access to health insurance has occurredperiodically in U.S. health care reform debates. Americans espousesupport for universal access, but is unwilling to support changesneeded to reach this goal. Congress has consistently failed in itsattempts to enact reforms.


● A public-mandatory market structure with tax-based and/or employment based funding, or

● A government regulated private-mandatorymarket structure in which the purchase of pri-vate insurance coverage is made legally manda-tory for all citizens.

In the case of a government-created private-man-datory market, law and regulation must be gearedto minimize and properly compensate for remain-ing aspects of risk selection, risk rating, and re-newal risk management that run counter tosmooth functioning of a universal health insur-ance system. This can be done, but not withoutcareful thought and effective government stew-ardship of private-sector health insurers and theiruse of risk management tools.

In the discussion of goals for health care sys-tems, I noted how there is a growing consensus indeveloped nations that funding all health carethat citizens both “need” and “want” is increas-ingly beyond governments’ resources. This fiscalconstraint means that health care systems willbenefit from the difficult task of clearly differen-tiating medical care that is “needed” for citizensto achieve world-class outcomes from medicaland nonmedical “wants” that, while valued bypatients, do not add to population health.

10.3 Principle 3The needs-based universal core program is sup-plemented by private health insurance coveringhealth care “wants.” Having defined differencesbetween health care “needs” and “wants,” anideal public universal health care system shouldcover population “needs” and leave funding forand coverage of population “wants” to an ade-quately regulated private-voluntary health insur-ance market.

A social ethic of liberty and autonomy drivesthe need for an ideal system to make available theopportunity for citizens to voluntarily purchase ahigher level of benefits than can be afforded by anation’s universal health care system. While thistwo-tier health care system makes sense at a con-ceptual level, we must clearly note that it is verydifficult to define “needs” versus “wants.”

10.4 Principle 4There needs to be a seamless, nonduplicativeinterface between universal needs-based and

voluntary wants-based parts of the health caresystem. Those charged with stewardship of thesystem need to design this interface carefully,which also is not an easy task.

European countries already have a variety ofexamples of universal coverage and supplementalprivate health insurance (Mossialos and Thomson2001). To my knowledge, though, none of thesecountries has found a good solution to how atwo-tier health care system with both universaland voluntary parts can work to fulfill and fundtheir citizens’ complete health care “needs” and“wants.”

10.5 Principle 5The health care system must have adequatemedical care resources, and it must be flexibleand able to adapt to new health care technolo-gies and new medical care needs. Even if wewere to be able to describe our ideal health caresystem in detail, we would have to admit that anycurrent solution will eventually need to change.As medical research continues to drive new tech-nology, and as countries grow richer and theircitizens value more and more medical care, so toowill the boundary between “needs” and “wants”change.

An artificial heart is an example of this dy-namic. As artificial heart technology improves,will this extremely expensive medical interven-tion be made available and, as it becomes moreeffective and common, will it ultimately changefrom a “want” to a “need?” This single exampleshould amply demonstrate to the reader the eth-ical, political, technical, medical, financial, andemotional issues that will continue to challengehealth care systems. An ideal system must havestrong stewardship and be flexible enough to al-low it to adapt to continuous change in medicalcare technology and its implications for citizensand the system itself.

10.6 Principle 6Our ideal system must allow each nation to meetits health care goals for cost, quality, and accessin ways consistent with its financial capacityand its personal, social, economic, and politicalhealth care drivers. Our ideal health care systemsupports national goals for cost, quality, and ac-cess, but does so leaving a great deal of flexibility


for nations to adapt systems to their environ-ment.

● Cost is addressed by focusing a universal pro-gram on population health care “needs” whileallowing citizens’ “wants” to be satisfied by pur-chasing private supplemental health insurance.While this split funding does not solve problemscaused by escalating health care costs, it doesallow government latitude to control publiclyfinanced care costs by shifting some costs to theprivate sector.

● Quality is more directly addressed. A universalprogram covering citizens’ health care “needs”will maximize population health (DALE) andminimize variations. Quality “demands” arisingfrom citizens’ health care “wants” are mostly,but not completely, satisfied through privatesupplemental health insurance. Satisfaction of“wants” will depend on the breadth of privatecoverage, and on the popularity and affordabil-ity of its premiums.

● Access to needed medical care is assured by auniversal health care system. Access to moreresponsive care is available through the pur-chase of private supplemental insurance.

10.7 Principle 7Developing countries need to consider today theevolution of their health care systems. Designingan ideal world-class health care system is not onlydifficult, but also expensive. Our discussion of anideal system implicitly speaks only to those coun-tries that can afford roughly $1,000 (1997) percapita in health care expenditures. Many coun-tries simply cannot afford this financial burdenright now, and some likely will not be able toafford anywhere near this amount for the foresee-able future. These countries clearly are not at apoint to consider many of the issues concerningdesign and implementation of an ideal health caresystem.

Our analysis does contain two important les-sons for developing countries, though. First, it isnecessary to think clearly now about the futureand how a developing country’s health care sys-tem will be organized as it becomes more fullyfunded. The reason for this is that not doing somay create markets, usually private-voluntaryhealth insurance, that may currently work well

but also may create entrenched political and so-cietal barriers to later change.

Second, looking back to Figure 5, note the vastdifference in the causes of morbidity betweendeveloped nations and developing nations. Devel-oping nations’ burden of disease is heavilyweighted toward infectious disease, diseases ofchildbirth, and accidents. There has been a greatdeal of thinking done about how to address thismorbidity profile (Sachs 2001). Researchers notethat much of this disease burden is amenable tobasic public health intervention.

11. ACTUARIES CAN HELP

Implementing an ideal health care system re-quires a great deal of additional thought and plan-ning. How can actuaries help? Actuaries aretrained in understanding and modeling publicand private system’s responses to varying incen-tives, policy and administrative options, and incommunicating costs and consequences to rele-vant audiences. Actuaries are trained in gatheringdata and using it to refine health care systems’policies and operations (the actuary control cy-cle). Actuaries are trained in the consequences ofprivate-sector insurance regulations and in howto price benefits and policy options for both pub-lic- and private-sector health plans.

Figure 9 is an adaptation of the actuarial con-trol cycle to an overview of stewardship of healthcare systems. Actuaries have particular expertisein the technical effectiveness and efficiency ofhow systems operate, measuring outcomes, andproviding feedback to those responsible for stew-

Figure 9Toward a Better Health Care System


ardship of the system. However, actuaries are nottrained in ethics, health policy, or politics. Weserve the public best in those areas by expressingour informed opinions solely as concerned citi-zens.

In performing these functions, actuaries oftenencounter other trained professionals, notablyhealth economists, health statisticians, and healthpolicy analysts. Their skill sets and ours oftenoverlap. This makes it difficult to argue that ac-tuaries are in a unique position to make contri-butions to nations’ health care systems. However,while we must recognize this overlap in skills, ourtraining in the practical consequences of policyoptions and in gathering and analyzing data doput us in a position to be of great value to thoseresponsible for the stewardship of health caresystems and to the private health insurance com-panies that choose to participate in the market.

National actuarial organizations can help topromote the role of actuaries in national healthcare systems by becoming known to health poli-cymakers and to those responsible for steward-ship. National organizations can provide informa-tion on the actuarial aspects of health policyissues and also promote the hiring of actuaries innational health systems and in insurance regula-tory bodies.

The International Actuarial Association (IAA)also has a role to play. International organizationssuch as WHO, OEDC, and the World Bank areengaged in thinking about health care systems.The IAA can play a role by promoting actuarialinput to these organizations’ health policy staff,conferences, and journals. By creating more op-portunities for skilled health actuaries to provideinput, actuaries can make a difference.

REFERENCES

ALMEIDA, C., ET AL. 2001. “Methodological Concerns and Recom-mendations on Policy Consequences of the World HealthReport 2000,” The Lancet 357, (May 26): 1692–7.

ARROW, K. J. 1963. “Uncertainty and the Welfare Economics ofMedical Care,” American Economic Review 53: 941–73.

———. 1971. “Insurance, Risk and Resource Allocation,” inEssays in the Theory of Risk-Bearing, Chapter 5, pp. 134–143, Chicago, Markham.

BEVERIDGE, W. 1942. Social Insurance and Allied Services. Lon-don: HMSO. Reprinted in Bulletin of the World HealthOrganization, 2000, 78 (6).

BOLNICK, H. J. 1995. “The U.S. Health Care Reform Odyssey,”Contingencies (July/August): 28–33.

BRITISH MEDICAL JOURNAL. 2001. Report on presentations at theFourth European Health Forum in Bad Gastein, Austria323 (Oct. 6): 772.

CALLAHAN, D. 1998. False Hopes: Why America’s Quest for Per-fect Health Is a Recipe for Failure. New York: Simon &Schuster.

DONELAN, K., ET AL. 1999. “The Cost of Health System Change:Public Discontent in Five Nations,” Health Affairs (May/June): 206–216.

FRIEDMAN, M., AND L. J. SAVAGE. 1948. “The Utility Analysis ofChoices Involving Risk,” The Journal of Political Economy56(4): 279–304.

INSTITUTE OF DIRECTORS. 2000. Healthcare in the UK: The Need forReform. London: Institute of Directors.

INSTITUTE OF MEDICINE. 2002. Care Without Coverage: Too Little,Too Late. Washington, DC: National Academy Press.

KAHNEMAN, D., AND A. TVERSKY. 1979. “Prospect Theory: An Anal-ysis of Decision under Uncertainty,” Econometrica 47(2):263–292.

KING, L. W. TRANS. 1910. Hammurabi’s Code of Law. Online athttp://www.wsu.edu/�dee/MESO/CODE.HTM.

MORDACCI, R., AND R. SOBEL. 1998. “Health: A ComprehensiveConcept,” The Hastings Center Report (January/Febru-ary): 34.

MOSSIALOS, E., AND S. M. S. THOMSON. 2001. Voluntary HealthInsurance in the European Union, London School of Eco-nomics Health and Social Care Discussion Paper No. 19,May.

MURRAY, C. J. L., AND A. D. LOPEZ. EDS. 1996. The Global Burdenof Disease and Injury. Cambridge, MA: Harvard School ofPublic Health on behalf of the World Health Organizationand the World Bank. Summary online at http://www.hsph.harvard.edu/organizations/bdu/summary.html.

ROTHSCHILD, M., AND J. STIGLITZ. 1976. “Equilibrium in Competi-tive Insurance Markets: An Essay on the Economics ofImperfect Information,” Quarterly Journal of Economics90(November): .

ROTHSCHILD, M., AND J. STIGLITZ. 1997. “Competition and Insur-ance Twenty Years Later”, The Geneva Papers on Risk andInsurance Theory, 22: 73–79.

SACHS, J. D. 2001. “Macro Economic and Health: Investing inHealth for Economic Development.” Report of the Com-mission on Macroeconomics and Health.

STARR, P. 1982. The Social Transformation of American Medi-cine. New York: Basic Books.

WILLIAMS, A. 2001. “Science or Marketing at WHO?: A Commen-tary of ‘World Health 2000,’ ” Health Economics 10: 93–100.

WORLD HEALTH ORGANIZATION (WHO). 2000. The World HealthReport 2000. Geneva: WHO Office of Publications.

Discussions on this paper can be submitted untilOctober 1, 2003. The author reserves the right to replyto any discussion. Please see the Submission Guide-lines for Authors on the inside back cover for instruc-tions on the submission of discussions.


D AW -C H C S

Documents

Transcript of D AW -C H C S