Equality, Efficiency, and Market Fundamentals: The Dynamics of International Medical Care Reform.

Public opinion surveys uniformly show low support for medical-care systems in developed countries. The longstanding conflict between equal access to care and efficient service provision partly explains this dissatisfaction. But the trade-off is particularly acute in medical care, as new technologies developed over time have increased the cost of care and made the equity commitment even more expensive. Countries first dealt with rising costs by maintaining equal access and restricting total spending. Efficiency suffered, however. As a result, many countries are considering a move away from spending controls and toward incentive-based medical-care reform.

1. Introduction I s NO COUNTHV are people particularly happy with their health-care system.In a survey ol people in ficveial countries, Karen Donelan and colleagnes asked respondents to react to the statement: "On the whole, the heLilth system works prett\' well, and only minor changes are necessary to make it work better" (Donelan et al. 1999).In the United States, only 17 percent of people agree with the statement; 83 percent perceive the need for "fimdamonta!change" or "complete rebuilding.""In Canada, with universal insurance coverage and medical spending as a share of CIDP at 70 percent ofthe U.S. level, only 20 percent ol people agree with the statement.In the United Kingdom, with .spendingus a share of (iDP at hall the U.S. le\ el and overall health nieitsures just as good, only 25 percent agree with the statement.In AustnUia and New Zieiiland, support is 19 percent and 9 percent respectively.Granted, health care is a diffieult issne for societies.Bnt why such great discontent?
Medical care is snch a problem, I argue, beeanse it is fundamentally a setting of conilicting goals.Methca!systems developed with clear equity considerations.In most developed conntries, universal insurance coverage was designed to guarantee equal access to medical care for all.Solidarity in health care dictated no rationing by price.
The classic tradeoff in economics is between equity and efficiency, and this shows up in medical care.Efficiency was not a great concern when health systems were established; conntries were content to have inefficient medical-care systems provided they treated all equally.
But the equality-efficiency balance has been thrown into conflict by the fnndamentals of the medical-care market.Medical eosts have increased rapidly over time, as technological change has expanded the eapability of medicine.Since 1960, medical care has more tban doubled as a share of GDP.The result of this technological change is that govennnents face increasingly severe finaneing erises.Many countries can no Joutiial of Economic Literature, Vol.XL (September 2002) lotiger afford the coiiimitnient to complete equalit)' tliat they once coiikl.
The Rrst response of most countries to the problem of rising costs was the enactment of regtilatoty limits on costs.These hmits were incrcLisingly pre\a]ent in the 1970s and 1980s.Govemtnents reduced provider fees and rationed access to medical technolog)'.Cost growth slowed quite a bit.Compared to the United States, which never had widescale limitations, medical costs in other developed countries teli from about 90 percent of the U.S. level as a share of GDP to about 65 percent of tliat level.Further, health outcomes did not seetn to stiffer.A decade into these rationing systems, overall satisfaction with medical eare was higli.
But recent years have made the regulatoiy solution increasingly less attractive.The mareli o( tcchnologv' has continued, even in regulated systems.Mediciil-care cost growth resnmed when expenditure controls were not actively being tightened.Further, cost controls have nuide the lack of efficiency more noticeable.Waiting lines and access testrictions have become increasitigly important issues as the constraints increase in intensity.
As a result, the regulatoiy solution to medical care is cotiiing tmder disfavor.In many countries, there is an incipient movement away from regnlation arid towards marketbased solntions to medical-care problems.Countries are introducing competition into medical care and in some cases providing incentives for people to ttse less medical care and to choose less-expensive health insurance plans.Providers are being asked to consider costs in their care decisiotis as well.
Iiieentive reforms bring the potential for cost savings withottt painful pulilie cuts.That is why countries are attracted to them.But incentive reforms bring conflict as well, Priee-govenied systems are not as equitable as the older systems they replace.The poor do not have the sauie access to medical care as the rich when prices are used to ration care.Fttrther, less-healthy people may suffer compared to the healthy.And incentive sys-tetns may restrict access to some pro\'iders.One way that money is saved in medieul eare is to limit the providers one contracts with, and n.se the exclusiou power to negotiate lower prices.But abandoning the commitment to equality" of access is not easy.As a result, there has been great reluctance to tise competitive measures to their fullest extent.Many countties are tiow exploring the proper scope and applieation of incetitives in medical care.How couutries resolve this debate will have major implications for the worlds niedical-eare systems for decades to come.
In this paper, I trace tlic histoiy oi international niedical-eate reform and lay out the issnes today.I begin in section 2 with a discussion ofthe birth ol ttuiversal systetns.Section 3 then examines the move to controls and rationing.Seetion 4 considers the effects of expenditure constraints on the prtnision of medical eare.Section 5 discusses the problems with rationed systems, and section 6 examines the new wave of reform.

The First Wave of Hectlth-Care Refomi: Universal Coverage and Ecfual Access
The late nineteenth centurv' saw the beginnings of health insurance in most developed countries.'^Mutual aid societies, or sickness fimds, were formed for some workers, In some coutitries, snch as Geriuauy,"* the development was spurred by the central government.More frequently it took place ill the pri\ate sector.
But even ;is late as \\'orld War 2, access to medical care was not a paitieularly high prior-it\' for tJie public sector.In part, the desire for medical insurance was limited because tliere wiLs little that medical care could do foi" sick people.Panl Starr (1982) traces the tnuisformation ol the American medical system.Starr notes that medical knowledge was poor in this period, ancl it was not until the mid twentietli centnrv' that tlie medicii!profession was seen as a significant laetor in helping to cure disease.In atldition, medicid insiiranct^ was rarely high in families' economic priorities beeanse niftlic'ul costs were not particularly variable.With little to do lor sick people, the financiiil risk dissociated with being siek was low.Tbe insurance that was available for sickness freqnently covered lost wages, not medical costs.
World War 2 cbanged the situation substantially (Starr 1982).Advances in penicillin and other antibiotics convinced people that medical care was Viilnable.Conntries wanted to reward tbemselves for years of strnggle.Qnasi-socialist governments elected after the war wanted to expand tbe role oftlie state in tbe provision of basic needs.The result was a major emphasis on expanding health insurance coverage.Table 1 sbows tbe creation ol national medical-care systems in G7 eoimtries.'^Tbe lirst post-war reform was in tbe United Kingdom.The Beveridge Report of 1942 delineated tbe inadequacies of tbe prior system aud recommended a goal of national bealtb insurance coverage.*""Tbis goal was met in 1946, witb the passage of tbe National Ilealtb Sendee (NHS) Act; tbe NIIS began formal operation iu 1948.Tbe Britisb system was followed, some years later, by tbe Japanese system •' Nniricrons cl;itu soiirfes were n.sed in creating tbe tiilile.sol"conntr\' features.Tbe OECD (19^51 and Roenier (I9i)l) ]i;i\e nnit-li inlornmtion, along witb cite.s to otlu'r articles witb more lietail.s.'' In 1941, an iiiterdcpartTnciitjil connnittce on Social Iiisinance and Allied Senices w;i.s appointed to .survey the existing schenie.sol social in.suranee.Tbe eoniinissioii made a "diagnosis ot want" by snr\ eying conditions ol liie in a nnnihcr ol towTis in Britain.Tbe Beveridge Report ci)nclnded that abolition of want required a redistribntion ot ineonic tbrougb scK'ial insuraiicc and by taniily Tieeds.Tlif report rfconiuiended a (.•oniprchcnsive national health service to provide lull ai'ces.s to all beneficial treatments without economic barriers (Beveridge t942>., tbe Canadian .system (1966-71), andtbe Frencb system (1967).In all of tbese conntries, universal coverage was tbe cnlmination of many years of partial coverage and government subsidies for insurance.More recent universal systems were finally acbieved in Italy (1978) and Germany, altbougb in both cases coverage rates were veiy bigb just prior to tbe universal legislation (Milton Roemer 1991).Tbe only C7 conntnwitbout a universtil bealtb insurance program is tbe United States, altbough it does bave a program for tbe elderly (Medicare) aud a program for the poor (Medicaid).Indeed, within the OECD (prior to its recent expansion), only Turkey and tbe United States were witboiit nniversal insurance coverage.
Beyond jnst desiring universal coverage, countries wanted to ensure that tbe poor bad tbe same access to medical care as tbe ricb.Medical care was perceived as a rigbt, not a good, and markets were not looked upon witb favor.For example, tbe Beveridge Commission stated."From tbe standpoint of Social Security, a bealtb serviee providing fnll preventive and cnrative treatment of eveiy kind to every citizen without exceptions, witbout remuneration limits and witbont an econoune barrier at any poiut to delay recourse to it, is the ideal plan" (Part VI, see.437).Similarly, tbe Canada Healtb Act of 1984 stated."It is bereby declared tbat tbe priniar) objective of (^ianadian healtb care policy is to protect, promote, and restore tbe pbysical and mental well-being of residents of Canada and to facilitate reasonable access to healtb services witbout fniancial or otber barriers" (cb.6, sec.3).
Tbese sentiments bad fundamental implications for tbe design of medical-care systems.Tbey led medical systems to be extremely generous iu covered services aud low in required cost-sbaring.At least tbis was true for acnte medical care, wbicb tvpieally aeeounts for about 70 percent of total spending.Table 2 sbows tbe cbaracteristics of universal bealtb insurance systems around the 198()s and early 199()s."In almost all countries, covered benefits were very generotts.It is easier to report the services excluded from coverage than the services incktded.Excluded services iueluded dental care in some countries, visiou and bearing aids, and oceasionally otttpatient prescription dnigs.In countries where these services were excluded as a genera!rule, sueh as Canada and the United States, the pnblic set-tor sometimes covered the costs for the poor.Indeed, some countries had stieh generous services that they covered spa betiefits (Germany).
Furthet; there were few restrictious on using covered seiMces.In all countries, patients had free choiee of primary care providers, and cost-sharing for covered services was minimal.In many countries (Canada, Cermany, Italy, and the United Kingdom), access to some or ail physieians and hospitals is nearly free.Among these coimtries, only Italy has any cost-sharing for specialist physicians, but the cost-sharing is quite modest.Prescription drugs are somewhat less well insured but often still covered.
Three of the conntries (France.Japan, and the United States) have more extensive cost-sharing.In France, coinsnrauce rates ior pliysician senices are 25 pereent, althongh 80 percent oi the population has stipplemental itisttrance to cover the pliysieian  Still, many elderly have supplemental insurance to pay for these costs.
The result was tliat in most countries, there were very few demand-side constraints on utilization.As one indicator of this, the government paid for 72 percent of medical care in the average G7 country, and 76 percent outside the United States.Social and not individual financing was the goal.Economists might worry about moral hazard, hut governments were more worried about equity.
On the supply side, the systems were encouraging as well.In five of the coimtries-Canada, France (outside the public sector), Germany (ambulatory care), Japan (ambnlatory care), and the United States-physicians were predominantly paid on a fee-forservice (piece-rate) basis.In such a system, doctors that do more collect additional income; provided price is above marginal cost, the incentives to limit care to situations where it is truly appropriate arc weak.Â second alternative is to salary physicians.This was the route chosen in the United Kingdom and for selected specialties in France (public sector), Cermany {Jiospital-based physicians), Italy (hospital-based physicians), and Japan (specialists).Even salaried physicians had weak incentives to hmit spending; doctors did not suffer fhiancially if unnecessary care was provided.** George Bernard Shaw (1911) laiiiented tbis almost a eentury ago: "That any sane nation, having ohse'rveti tiiat yon eoiild provide for tbe supply ol bread b)' giving haker.s;t pee^uliar^• interest in baking for you, should go on ti) give a surgeon a pecuniary interest in eiittinj^ oH' your leg, is enough to make one despair of political nil inanity." A third alternative chosen in some countries is to pay physicians on a capitation basis-doctors receive a fixed amount of money per patient in return for providing medical services as needed, at no additional charge for each service.Included in the capitation amount is the physicians' own time and sometimes other services such as hospital care or prescription drug costs.Italy and the United Kingdom used this payment system for some doctors.Capitated systems can provide strong incentives to limit medical costs.But the capitated systems countries put in place were generally not very stringent.Many nouphysician services were not included in the capitation amount; surplus ftmds could not be retained by the doctor; and the ability to negotiate with downstream providers was limited.No country had a system like what is evolving in the United States today, where doctors are at risk for most of the care their patients receive and have large income fluctuations on that basis.As a result, access to physician services was largely unconstrained.
Hospital payment was also relatively generous, although more diverse.In many countries, the public sector ran at least some of the hospitals.The extent of public ownership ranged from very little (Canada, the United States, and Japan-20 pereent or less) to more moderate levels (France-35 percent; Germany-over 50 percent) to nearly all of the hospital system (Italy, the United Kingdom).In countries where the publie sector ran the hospitals, hospitals were topically financed as line items in the budget.When hospitals were private, they were often paid through global budgets-ĥ ospitals were given a fixed auiount of funds for the year to cover the costs they incurred in caring for tlieir patients.Global budgets were often coupled with public review of hospital investment decisions, for example for bed expansion or purchase of capital eqviipnient.
Global payment systems can be constraining--and later were-but frequently were Cutler: Intemationai Medical-Care Reform SS7 not at tlif time tlir systems were set up.Oltcn, the puhlic sector paid for hospital deficits at the end of the year, as in Italy.In other cases, the global budgets were adjusted for costs in previous years or were set at deliberately high levels.Still other eountries had fee-for-service payment for hospitals (Japan, the United States), creating the same iucentives on the inpatient side as they did on the outpatient side.
Arthur Okiui (1975) highhghted oue of the classic tradeoffs in eeouoniics, between efficiency and distributiou.Achieving a more equal tlistribntion ot resources than markets naturally provide geuerally requires inakiug markets less effieient.Iu the medieal-care context, distributiou was a clear priority over effieieuey.Eveu the rhetoric of universal insurauce coverage stressed distributional equity over economie effieietiey.By 1980, countries in the GT, and in the developed world as a whole, had set up niediealeare systems designed to encourage access to Tuedieal eare but not t(j promote the officieut deliveiy oi care.

The Seeotid Wave: Controls, }â nd ExfX'tuliftin-Caps
\\ itii few constraints on inedieai-care deinaiul or supply, medical speuding was bound to be inefficient.Countries were willing to accept speudiug ahove efficient levels to uieet distributional goals.But spending was also growing more rapidly tliau eouutries eould easily afford.While tax revenues were inereasiug at the rate of payroll or consumption growth, roughly the rate of GDP growth, medical care speuding was increasing twice as rapidly.In the average OECD country, medical care rose from 3.8 pereent of GDP iu 1960 to 7.2 perctMit in 1980.
Medieal-care costs were increasing so rapidly in large part because of tlie growing teehnieal sophistication ol medical practice.The development of new teehnoiogies aud the diffii.siou of those technologies to more patients led to large increases in spending.Empirical estimates by Joseph Newhouse (1992) suggest that techuological change accounts for at least half of overall cost growth, with the remainder resulting from inereased prices of services and increased use of existing teehnoiogies because ot the spread of insurance.Sinee the expansion in technology was worldwide, or perhaps U.S. driven, no other country could control its rate.
Figure 1 shows the inerease in the percentage of GDP claimed by health care over various decades; tahle 3 shows summary data.I examine spending relative to GDP to account for the income effect of medical care.As countries get richer, they naturally spend more on medical care.International eomparisons of medical spending geiierally suggest an income elasticity at or ahove 1 (Getzeu 2001), so it is natural to look at spending as a share of GDP.
The tirst bars in figure 1 are for the 1960s.In the CT7 countries, with the exeeption of the Uuited Kingdom (discussed helow), medieal spending as a share of GDP inereased substantially in this decade.Even iueludiug the Uuited Kingdom, the increase averaged 1.6 percentage points.The inerease wiis not limited to the G7 countries, as table 3 shows.Spending in the rest of the OEGD increased hy 1.3 percentage points.The pattern was repeated in the 1970s.Essentially all countries, with the exceptiou of Ganada, experienced rapid increases in medical costs.The average growtb was again 1.6 percent of GDP.
As medical spending increased, the burden on the puhlic sector grew.Tiix increases to pay for rising medical costs were increasingly painful.Eurther, the mid-1970s saw a reduction of income growth and the beginning ol structural puhlic deficits.This CUTIvinced countries that they needed to manage their medieal spendiug more actively.
The result was a gradiuil shift away from concentrating on the generosity of coverage and toward containing the eost of care.Where "iiceess to medieal eare" liatl oiiee been the pritnan' focits of public policy, "cost containment" became equally itTiportant.There are tvvo potential strategies for cost containment.First, incentives can be usetl to indnce people to spend less.Some services miglit no longer he covered hy the public progiain, or cost-sharing might he increased.At the level ofinsurance purchases, people can he given financial incentives to choose less-generous insurance plans, with lower overall cost.s.Doctors can also be given stronger financial incentives tt> provide less care.Alternati\ely, .spendingcan he constrahied through regulation.Access to technologies can he restricted, or expenditure caps imposed.Spending will thus be lowered by fiat.
Regttlation was the soltition chosen in essentially all couutries.The details ot the implementation varied somewhat across countries and types of services, hut the principles were similar.Tahle 4 shows the nature of these conttol measttres.In the hospitiil sectot; countries with giohal btidgets tightened the budget limits.Hospitals had to live vvith less-rapidl\ increasing funds.Tliis was often coupled with tighter constraints on hospital investment antl capacity expansion, to ensure that aggregate utilization would not exceed what was affordahle.
Budgeting physicians is more difficult than htidgeting hospitals, since it is not feasible to make individtial physicians hear all of the costs for services ttsed by their patients (hiring tile year.But alternative mechanisms are available.Many governments, for example Germany, imposed an overall spending limit for physician services as a whole or hy specialty.Different activities were then assigned relative weights, reflecting the resources involved in providing the services.The equilihrium price per tinit weight is set ex post to meet the overall budget.In other  cases, fee schedules were tightened, but without an explicit spending target in mind. These types ot expenditure controls require suhstantial government involvement in tlie medical system, and countries covild not implement them fully until the pnblic sector had enough authority over the system to carry out the task.Thus, the evolution of regvilation necessarily followed universal insurance coverage.The earliest country with universal coverage, and the first to implement tight constraints, was the United Kingdom.British doctors were paid on a salary or capitation hasis, and the puhlic sector ran the hospitals, sinee the estahlishment of the NHS.Thus, the British government was in a nattiral position to implement controls.Dnring the 1960s, governinent constraints led cost growth in the United Kingdom to he suhstantially helow the rate in other countries (figure 1).
C^anada adtjpted spending limits in the 1970s, following the estahlishment of universal coverage in the late 1960s.Giohal hndgets on hospitals were put in place when universal coverage was enactetl, and provincial governments used this payment authority to limit cost increases.As figure 1 shows, spending growth in (Canada was veiy low in the 1970s, when controls were enacted. Other countries followed in later years: France in 1984-85 (giohal hudgets for puhlic hospitals) and again in 1993 (giohal budgets for private hospitals); Germany in 1977 (expenditure caps on amhtilatorv' physicians) and again in 1984-86 (global bndgets for hospitals); and Japan in the 1980s (lower I'cvs for physicians and hospitals, and caps on hospital heds).Figure 1 shows these conntries generally having less-rapid cost growth in the 1980s.
Governments in the United States never controlled enough oftlie medical .S)stem to adopt wide-scale regulatory controls, but they did use less-systemic versions of these techui(jties iu pnblic programs.In 1983, the federal govenunent adopted the Prospective Payment System (PPS) lor hospital admissions paid for hy Medicare.Hospitals are paid a fixed amount per admission conditional on the diagnosis, regardless oi the services provided.^A relative weight is estahlished for each diagnosis, and reimbursement is determined on the hasis of the relative weight and the price per unit weight.Once the system was estahlished, it was straightforward for the government to lower the price paid per unit weight, which it did in the 1980s and 199()s.Prospective payment systems spread to Medicaid programs and many private insurers after it proved workable in Medicare."*Physician payments were similarly controlled.In 1992, the federal government introduced the Resource Based Relative Value Seale (RBRVS) in Medicare, which assigned relative weights to ^ Tlie .sy.steni is not (jnite as "Inird" as this.Hospitals reeeive outlier payments for patient.^ in the lio.spital a particularly lon^ time, and snrt^ic'ul lulTnissions are weighted niorc highly than medieal aLlniis.sions. The Prospective Paviiient System shares sonie similarities with incenti\'e4]ased reforms, in that it gives hospitals incentivt'S to economize on nse of marginal .ser\ices.Physician pa)'ment reforms did not huiulle services as highly, and were thus not as high powered.each physician activity.Payment is made on the basis of the relative weight for the ser-v ices provided, tnultipHed by a price per unit weight.Again, the fee schedule has been tightened o\ er time.
Still, tiie limitation of expenditure controls to the public sector made snch controls less effective in the United States thim in other countries.In I960, the average (J7 country spent 12 percent less on medical care as a sliare of GDP tlian did the United States.By the early 1990s, spending was about 35 percent less as a share ot GDP. Figure 2 shows medical spending as a share of GDP in 1990.The United States spent far more than all otJier G7 countries on medical care.

Evaluating^ the Re^tdated System
To understand the consequences of regulated systems, one must look at more than Japan just their immediate spending impact.Spending less money miglit affect the deliver)' of medical care, and this needs to be taken into account.Further, the long-rnn impact of these systems must he judged.To a first approximation, most countries outside the United States have highly regulated medical systems, while tlie United States does not.I thus evaluate regulated systems by comparing the United States to other advanced countries.
Kf'gulated systems can save money in two ways: by lowering prices paid for services or reducing the qnantit\' of ser\1ces provided.The two are obviously related-price changes may lead to quantity changes-but are couceptualK iuiportant to differentiate.When there are (juusi-rents in medical-care provision, as there are iii the return to past investments by physicians, price reductions need not be accompanied by reduced supply.
Empiriciil evidence shows that regulation atlected both prices and qnantities of care provided, Providers earn less in regulated systems than they do in unconstrained systems.Real earnings of physicians in the United States increased by 35 percent between 1970 and 1990.In countries with expenditure liuiits, by contrast, real earniugs were flat.As a result, doctors in the United States now earn twice what their counterparts cam in other countries.''Physician income is about 20 percent of total medical spending.Thus, the 50-percent-lower physician earnings in other G7 countries cau explain spending differences of about 10 percent.This is close to one-third of the 35-pcrcent overall difference in spending between the United States and the average G7 country.
There may be savings in other factor payments from cost controls, but these will be smaller.The opportunities for nurses, orderlies, aud other personnel to move to other industrie.sprevents a large income reduction for these groups.Physicians, in contrast, have much more profession-specific human capital.Pharmaceutical prices are lower in regulated systems, altliough the evidence is not clear on exactly how much.Retail prices in dnig stores are higher in tlie United States, but prices to managed-care insurers and other large purchasers may be lower (Danzon and Chao 2000).Still, pharmaceuticals Sonif (i! this is ht-fansc m erall wa^e.s are higher in tlie United Stiitcs, hut not ninch.Per capita GDP is ahoiit 10 to 20 percent hijflier in the United States tlian in other C.7 coLiiitries, Journal of Economic Literature, Vol.XL (September 2002) are only about 10 percent of overall medical spending, so price differentials in pharmaceuticals cannot account for much of the overall spending difference.All told, the difference in prices paid to factors probably explains a bit under half of total differences in medical spending between the United States and other countries.
Price rednctions are, in the short term, a transfer from medical-care providers to consumers.There may be long-term effects on the quantity or quality of medical-care per-soTmel, but tliese effects will not occur for some time.
Ofmore immediate importance for shortterm outcomes is the impact of regtihition on service provision.Abundant evidence shows that service availability is highly correlated with the stringency of regulatory constraints.The United States, with the leastconstrained medical system, has the most technologically sophisticated treatment patterns of any countr); in most dimensions.As other countries have imposed tighter constraints on spending, the rehiti\'e availability of services has declined.
A rigorous comparison ol how regulation affects the provision of medical sei^vices re-(jnires comparing medical care received by patients with similar conditions in different countries.This is difficult to do on a wide scale but is possible in some settings.The most convincing evidence comes from comparisons of the United States and Canada, where detailed data on patient diagnoses and treatments are available.The comparison between the United States and Canada has obvious appeal: the demographic characteristics of the two countries are similar, and physicians are frequently trained at the same institutions.
Table 5 summarizes evidence on treatment differences in the United States and Canada.Treatment of cardiovascular disease, and heart attacks in particular, has received the most attention.A heart attack is an acute event, resulting from a blockage of arteries supplying blood to the heart.If not diagnosed early and treated effectively, death is common.Among elderly people with a heart attack, one-year mortality is about 30 percent.Because of this high mortality, essentially everyone with a heart attack will receive medical treatment.Thus, differential selection into treatment is not believed to be important.
There are several potential treatments for a heart attack.Cardiac eatheterization is a diagnostic procedtire to detect the extent of arterial blockage.If the blockage is particularly extensive, physicians may decide to perform either of two revascularization procethires: b)q;)ass snrgery (creating a blood How around occluded arteries) or angioplasty (inserting a balloon catheter into the arteries to remove the blockage and restore blood flow).Most of the studies of heartattack treatments across countries analyze the use of these surgical procedures.
The first four rows of table 5 present results of studies comparing heart-attack treatments in the United States and Canada.All of the studies show that the United States uses the sophisticated technology much more commonly than does Canada.Cardiac catheterization is nsed two to five times as often in the United States, and revascularization is two to eight times more common.In addition to these studies, two other studies, reported in the next rows of the table, examine the share of the overall elderly popnlation receiving byjiass snrgery.Rates of procedure use are 40 to 80 percent higher in the United States than in Ganada.
The final row of table 5 shows rates of outpatient psychiatric use, The share of people receiving outpatient treatment lor iiu'iital illness is 70 percent liiglier in the United States.'"The e\1dence in table 5 is therefore generally clear: the United States has much higher rates of intensive proeednre nse than does Canada.^The link between limits and less intensive care raises the specter of worse health onteomes in res^iilated systems.Four of the studies presented in table 5 examine health differences in the United States and Canada.'^These resnlts are detailed in the last column of the table.Mortality is the easiest measure of outcomes.Mortality rates are suqjrisingly similar in the United States and Canada.One-to two-year mortality ranges from equal in the two countries to a few percent lower in the United States, to a couple of percent higher.Some of the estimates finding lower mortality' in the United States arc statistically significant, but tliefr maguitude is uowhere near the two-to eightfold difference in treatment rates across countries.Some studies snggest that morbidity is lower in the United States.For example, rates of reinfarction, chest pain, acti\'ityliniitiiig augiua, and other complications are H) to 40 percent lower in the United States than in Canada.One study estimates funetional outcomes to be better as well, lint morbidity is very hard to measure.It is not known how sensitive these results are to dil-Tliis was not offset by i^rt'atcr use o\ iij services in Canada.
'•'Tile limited (hitii avaihthlc outside the United States and Canada support tins inlerenee.An analysis ot tefiniol()y\' use in heart ;ittaek eare tor niauv European eonntries.presented hy Dan Kessler. Mark McClellaii. and C'ulleai;ues (1999), found use nites in most eonntnes at or beknv tlie (Canadian ievel, and far below the U.S. level.Steven Sclinteder (1984) slums that iuispital stafling patterns are also related to spendiiifr.More anecdotally, Herny Aaron and William Sfhwartz (1984) surveyed physieians in the United States and the Unitei!Kingdom about the treatments they pnnide.British physicians were far less oriented to high-tech eare than their American connteqiaris.' •* For obvions reasons, comparisons ol heahh ontconies as awhoie are not particniarly revealing.Indeed, in a cross-section medical spending and lile expectancy are negatively, bnt n{)t statistically significantly, related.fcrent smvey instrnments or factors such as when in the year the snr\'ey took place.Overall, the surest outcome measnre is mortality, which is very similar in the two countries.The Canadian system does much less, but does not seem to suffer that much.
There are several potential explanations for the finding of" few mortality differences despite very large treatment differences.One explanation is that Canada provides low-tech care more generously than the United States, thus offsetting the reduced use of high-tech care.Rates of medication usage to prevent long-term heart damage are notoriously low in the United States.Further, patients see physicians for evaluative and management care more freqnently in Canada than in the United States.Alternatively, wfiat is rationed in Canada may be care that is not essential for survival.People who benefit the most from intensive medical procedures may be equally likely to receive these procedures in the United States and Canada, bnt more marginal patients miglit be disproportionately treated in the United States.
Whether the lack of adverse outcomes is a re.suit of l)etter provision of low-teeh care or efficient rationing is not known.It is an important topic for research.Bnt regardless of the explanation, regnlated systems did not fail becanse of poor heiilth outcomes.

The New Dilcnuiuis
In the late 198()s, it seemed as if the solution to the world's health-care problems was at hand.Regulatory constraints lowered costs but still afforded high-quality care.The poor were not discriminated against, at least openly.'''Indeed, people expressed satisfaction with these svstems.Fifty-six percent of Canadians thonght that their medical-care system was fnndanientally sonnd, compared to just 10 percent of Americans.Advocates of health-care reform in the United States followed this evidence and urged the creation of a Eiiropean-st)'le systein.The Clinton Administration's national health reform proposal reflected this view.The Clinton plati imposed a limit on the growth o( health insurance premiums, modeled on the global budgets enacted in other couutries.As Cutler (1994) discusses, international evidence showing the po.ssibility of lower costs without worse outcomes was a key factor in tbis proposal.
But enthusiasm for the rationed model iias waned considerably in the decade since that time, to the point vvliere even countries with an ingrained commitment to that tv^je of system are considering major changes in medical care.There are three reasons why.
Limited Supphj and Unlimited Demand.The first problem is standard from economic theory: when limited siipply is matched with generons demand-side incentives, people arc uuhappy that they cannot get the services they want.Rationing is easier in medical care than in other industries because people are not fully informed about what seivices they uiiglit obtain.Thtis, some rationing takes place without notice.But as rationing becomes more severe, patients leam abotit the limits.Waiting lines are observable, and the lack of care is more pronounced.
As systems became increasingly constrained, these problems surfaced.Survey data shows the extent of this.Robert Blendon et al. (1990) present results from common surveys in 1990, asking people in nine eotmtries to rate their health system using the qttestion presented in the iutroduction.Figtire 3 shows the relation between the sbare of people who were satisfied with the healtl) system and per-capita spending ou medical care.People in the United States are the least satisfied with tbeir medical-care system, despite the highest spettding.In probing responses, the reason for this dissatisfaction is that people are concerned they will lose tbeir coverage or bave to pay more if they become sick.Outside ofthe United States, this is not a concern; insurance is uuiversal and gtiaranteed.Indeed, omitting the United States, tbe dominant isstie explaining satisfaction with the medical-care system is spending on medical care.^^Countries that spend more on medical care have greater satisfaction with the medical-care .system.In most countries, tbe eliief complaint when people are nnluippy is waiting lists and concern about too icw resources.
Inefficieneij.Tbe second problem with regulator)' cuu.straiuts is that tbey do not provide incentives for service provision to be efficient.A variety of evidence suggests tliat productive eflicieucy in medical care is very low.For example, as countries bave reduced the fees they pay physicians, one way tbat physicians have responded is to see patieuts more freqtieutly-have two visits wben one might do; prescribe drugs for a shorter Ieugtb of time so tbat additional visits are needed, et cetera.Tbis type of scheduling allows physicians to collect multiple payuicnt for treatment courses that formerly brought in one.Stories about "inibtindling" physician services abound in (>anada, Italy,'^ and Japan,^^ among other "'The regression line omits the United Stiites hecatise thf reiLson for disple;tsiire with thf system is so different. '' This is (}ftei] called "practice cliiirning."It is discussed in more detail hy t^ Da\id Na\ lor (1999).
' *^ \\'itne.ssthis quote alioiit the Italian liealth-care system IroiTi a Tiiemher oi Parliament.Carol Beehe Tanintelli: "There are absolutely no incentives for efficiency.Since hospitals get ]>aid h\-iia\in^ their beds filled, they are eneoiirageii to keei) ptvjple there for nnich longer than necessarv.And if (hose beds are occupied by people who are not so siek.then the hospital inanajfeiTient figures it is so much the hetter hecanse they require less care and atteiition.Bnt meanwhile, people who really need it are .strandedwithout help.People often have to wait np to twenty davs in the hospital (or tests to fie done, because the labs are only open fonr hours a day.So they stay there occupying a bed that could fie nse<! by soinehody wlio is serionsly ill-all because the hospital is happy to get money (rom tlie state for a patient who iloes not recjnire tiiem to lift a finger" {The Washington Post. April 3. 1994).
"^ A joke abont the Japanese liealtli-care system is telling: Several elderly people are waitinij in the physicians office, as they do every day.One ol' them asks."Wliere is Mrs. Morita?She is not here today."Another person answers."She conkln't make it; she is sick."eonntries.^"The consequence of unbundling is that patients wait longer for care, but this waiting is not counted in aggregate spending totals.
Other examples concern nse of prescription dnigs atid acquisition of high-tech therapies.In Japan, the fee schedule for routine visits is so low that many physicians earn money by buving prescription dnigs wholesale and selhng them to patients retail.-'They also earn money by investing in hightech scanning equipment and referring patients to those scanners.The result is that [apan has extremely high rates of pharmaceutical consumption and substantiall) more scanning technology' than other countries.
^" This phenomenon is not unheard of in the United States.It is coninioii, for example, for goveninient actuaries to assume that even' $1 cut in physician payments made by Medicare will only translate into $.50 in public sa\in';.s.The remaining H.5ii will be reconped in increasci!physician volume (the "volnme oi"tset").
^' Low fees in Japan actually date back Imndreds oi years, to a time when Japanese doctors eanied all tlicir income from the sale of drug.s.
Althougli tbere are no data on the eificieney of different systems, efficiency concenis have become predominant in essentially all of tlie G7 eountries (OECD 1998).
As ratiouing and system efficieney become problems, people look for ways around the limits.Tahle 6 .showsiiumifestations of this.One way to evade limits is to ptirchase supplemental insurance.Tn the United Kingdom.10-20 percent of the population has some forui ot supplemental insurance.Timothy Besley, John Hall, and Ian Preston (1998) note that this insurance is fre(juently used to jump the NlIS qneues; doctors see patients with supplemental insurance outside of tlieir normal clinic hours and vdtli shorter waits.In Italy, supplemental insurance pays ior doctors not in the national system.
--In addition to the n.ses here, snpplernental insnratice in many countrie.scovers scr\iccs uot included in tlie basic benefits package.This nse is mnch less controversial.Direct private payments serve the same [Toal in Japan, Frauee, Italy, and the United Kingtioiu.In Japan, there are frequent reports of people paying bribes to physicians to iiwass waiting lines (Naoki Ikegami 1991).-*In France, some physicians are allowed to •'baianee-hill" patients above the fee seliedule.This ha.s become a wa\' for rich people to e\ade the spending limits.And in Italy; people often pay physicians privately for more rapid service; physicians are allowed to Jiiaiutain a private practice separate from their publie practice.-"The economics of supplemental insurance is complicated, For a fixed distrihti--'^ Especially to get into the large pnhlic hospitals which liave higher perceived (jualit\.
-' Scandals are common, imoKing doctors who pnrchasc cijnipnicntou die pnblic budget but then use the equipment in tlieir private practice.tion of medical resources, queue-jumping by some reduces services for others.Thus, some countries, such as Canada, have banned any payment outside the national system.But allowing sttpplemeiital payments could also increase the overall amount of medical-care provision by supporting more providers or by encouraging existing providers to work additional hours.This might free up more resources for those without siippleuientai instirance.No empirical studies have analyzed these two effects.
Mobility is another way to evade limits.In Italy, some areas (particularly in the north) are well funded while other areas (southeru Italy) have far lewer medical resources.Intra-countr\' mobility of tbe sick is common.Some Canadians come to the United States for medical care, although the extent to which this occurs is hotly contested.^'The cominon theme in all of thest" CLises is that people antl providers attempt to evade onerous regulatory hmits.As systems became more constrained and less efficient, the attempts to evade the limits became more prominent.
One-Time Savings and Lon^-Hiin Cost Growth.The third problem with regulated .systems is stnictural: even regulated systems grew more rapidly than governments could afford.Figure I shows this situation.Countries that imposed expenditure constriiints generally experienced al)out a decade of lower cost growth.But after tliat time, spending growth increased.This happened in the United Kingdom in the 197()s (after slow growth in the 196()s), in Canada in the 19805 (after slow growth in the 1970s), and in Germany and Japan in the 1990s (after slow growth in the 1980s).
The reason for this rebound is the underlying dynamic of medical technolog)'.Expenditure caps did not eliminate technological change; they just suppressed some of tlieir manifestations.But ultimately, the technology was adopted and led to increased spending.This is clearest in the case of price reductions.As noted above, a large share of the savings from expenditure caps was in lower prices paid to doctors.But quantity growth is a far more important driver of longrun cost increase than is price growth.Thus, priee cuts are an inherently limited way to rednce spending increases.When prices are falling, spending growth will slow, but growtli will then resume when price cuts cease.That is what happened in many countries.
Even hmits OTI the diffusion of technology were reliixed over time.Countries allowed more investment after periods of tightness, and thus teehnologieal change resurfaced.Invasive treatments for heart attacks, for example, are used more in Canada today than ^^The (Ifhatf is tiiken up liy David Azcvt'do (199,3), Chuck Jones U993).and Michael Walker and Greg Wilson (2()01), in the past, though still below their use in the United States.
Teehnologieal change is taken here as exogenous to each country, but that is not necessarily true in aggregate.At the world level, technological change results at least partly from the incentives in the medical-care system.The traditional medical-care system, with low cost-sharing and generous physician payment, ereated substantial incentives encouraging technological innovation (John Goddeeris 1984a,b;James Baumgardner 1991;Burtou Weisbrod 1991;Peter Zweifel and Willard Manning 2000).But the distinction between one country and the world inechcal system is important.The large.st market for medical innovation is the United States.Innovation is thns geared towards the U.S. market.At a practical level, therefore, countries outside ofthe United States could not control the rate of technical change they faced.Eveii as these eountries imposed cost controls, incentives in the United States were still verv' generous, and tiie pressures to adopt technology mounted.Most of the world, therefore, views technical cbange in medicine, correetly, as independent of national goveninient actions.
Overall,, cost controls in other countries lowered the level, but not the long-run growth, of medical spending.A decade or so after implementation, the reduction in cost growth slowed, and costs again began to increase.
Given the importance of technology in long-run cost growtb, tbe inability' of regulations to reduce the long-run growth of costs is not surprising.Since technological change generally comes from the United States, other countries bad very little ability to iiffect the rate of fundamental cost increase.Nor is the cost inerejise necessarily a cause of concern.The key issue in evaluating medicalcare cost increases is wliether tbe services bouglit by the increiise in cost are wortb tbe money that is spent on them.There is no definitive answer to this question, but much recent research suggests that cost increases may be worth the expeii.se(Cutler imd McClellan 2001).If thi.s is true generally, eouiitries ma)-he better oil" socially with cost increases and financing pressures than without them.Whether cost increases are good or bad is ultimately an empirical question.
In practice, the failure of constrained systems to limit cost growth led to tremendous nnea.se,aboutthese systems.In the 1980s, Canada iacetl a resurgence ot cost growth, alter a period of slow growth in the 1970s.Hy the early 1990s, governments in Canada were imposing new limits on fees, closing hospitals, and tightening resource limits.The (Canadian popniation, which thought tli(^ medical-care linancing system was stable, was upset by the required actions.Confidence in the medical system dropped dramatically.Whereas 56 percent of C'auadians were satisfied with the system in the late 1980s, only 29 percent were satisfied in 1994, and only 20 percent were satisfied in 1998.Cutbacks have provoked similar controversy in other countries as wel!.-*^As the 199()s progressed, unease about the medical system grew.
tem has become more prominent, healthcare reform bas again coine to dominate the agendas of OECD governments.The issues are remarkably similar in different countries-the rising costs of care and the perceived inefficiency of the .system(Colleen Flood 2000).^But in the face of these recent difficulties, governments have turned away from strict reliance on rationing and controls, to a more pluralistic view of ways to limit mediea!spending, The resnlt is the beginning of a third wave of health-care reform, focusing on incentives, and in particular competition, as central elements in the medical .system.Table 7 shows some characteristics of these reforms.
To date, incentive-based reforms bave been only partial.(Joveniincnts have a natural inclination to go slowly in any situation as important as medical care.Further, tbere are substantive reasons for caution.Incentives in medical care are not always aligned in socially optimal ways, and relying on incentives may lead to poor outcomes.The most important misalignment is the potential for sick people to suffer in a competitive medical-care system, sinee payments for them do not always cover their costs.Finally, the commitment to cquit\' is a major stumbling block.A significant sliare ofthe popniation still sees medical care as a right, not a good.Social solidarity is a unifying factor in many countries, and the idea of using incentives to allocate medical care risks violating this solidarity (Uwe Reiiihardt 1997).For all ol these reasons, competitive reforms have been tentative.Whether to employ these mechanisms more is a major question facing many countries, Incentive-based reforms are of three t\pes.At the patient level, some reforms increase eosts paid when nsing services.Increiising cost-sharing reduces government spending directly, by redirecting some costs to individuals and away from taxpayers.It also reduces demand for seivices, saving additional resources.
foumal of Econotnie Literature, Vol XL (September 2002) Increased cost-sharing has been pursued in Canada, France, Germany, Italy, and Japan.In Canada and France, the increa.se in cost-sharing has been relatively minor.For example, in Canada cost-.sharing has inereased marginally and some marginal .serviceshave heen "dc-insured" (dental ser-\'ices for children; vision exams).^'In other countries, the increase in cost-sharing has been greater.In Japan, 1997 reforms increased coinsurance rates from 10 to 20 percent of total medical hills for those insured through the uatioiuil health program.-*''In Germany, there were large increases in copayments for preseripHon drugs in the 1990s.
Debate aljout iurther increases in costsharing continues.In Japan, proposals were recently made to increase cost-sharing an additional 10 pereent, hut these were dropped after provoking opposition.Even in Canada, there is discussiou of a "two-tiered" medical-care system, where the government would insure basic services, and private insurance would cover the rest.Iu one recent survey, over 40 percent of Canadians supported such a systeui.Support drops, though, when potential inequities between rich and poor are highlighted (Gallup 1996).
Other countries have introduced competition at the level of insiu'anee purchase rather than at the time services are used.Allowing insurers to compete with one another, it is hoped, will increase the efficiency of service provision while saving nioney.^^Însurance market eompetition has been '^' Tlierf is currently a coutrnvcrsy in Ciinada ii.s to which st-mces are "medically nece.ssaiy."ami provinces are iiegiiiniug to (liF(eren(i;ilK (If-in.siircvarious ser\ices a.s a nietlitxl ol eost eontiiiiiiiieiit.For exaninlc.Pritne Minister Jean Chrftieji stated in 199.5: "Notiody lose.shis iuinie beciiusc somebody lias a prohleni witli his teeth or eyes, noniially.But it you have major sur^eiy.it's something else, so Vieclicarc wiis intendc'd tor that." -'' The iTicrease in c()pa\nnents was meant to be conditinnal on more fundamental relorms to the system, incluilin^ elimination ot" priee differentials between wholesale and retiul drugs, hut tlu'sc retoriiis were postponed due to vigorous protests Irom physician associrttions and pharmaeeutieal compiinies.
-•' Proposals for insuruiice market reform often gu niider the rnhric ot "niiiiiiigetl competition" (Alain most proniineut in the United States.While the United States never had a heavily regulated system, it did have a very generous systeui historically, with few eonstridnts on either the demand or supply side.Prior to the 198()s, most employers in the United States contracted with jnst one insurance company, which ran the traditional ])lan.
In response to decades of rising h^iilth insurance preminms, companies hegiui to shop around tor less-exjiensive insurance options.Some companies switelied their insurance entirely to lower-cost insurers, geneniUy to managed-care insurers.Others created competition hetween plans for enrollment.The traditional iudenmity plan was still offered, hnt so were lower-priced managed-care plans.Being lower cost, managed-eare plan.swere at a snlxstantial advantage.The cost advaTitage was compounded by adverse selection, which led to premium differences far larger thau differences in benefits alone would suggest {Cutler and Sarah Reher 1998).The result was an enorinoiLs change in plan enrollment.Between 1980 aud 2000, insurance for the privately insured population went from o\'er 90 pereent in unmanaged indemnity insurance to over 90 percent iu managed eare.
These incentives at the insurer level led to substantial emphasis on eost savings.'^"Insurers first reduced tlie prices the)' paid to providers.No systematic evidence ou prices paid hy mauaged-care insurers is available, but tlie studies that have been coudueted suggest price reductions of up to 30 percent for some high-teeh senices (Cutler, McClellau  tmikiterally.Maiiaged-care insurers, in contiiist, had to hiirgiiiii for price reductions.Tlie leverage manuged-care insurers used was tlie ability to exclude providers from their network.As a result, lower prices came aloTig\\1tli constrained access to providers in tlie United States, where it did not in other countries.Managed-care insurers also cut back ou the receipt of some care.Insurers intensively uionitorcd what providers ditl, and enacted finaucial incentives for providers to do less.Many priinar)' care physicians, for e.Kamplc, are paid on a capitation basis-they are reimbursed a lee covering their own costs and part or iUl of hospital and physiciiin costs.Spending less on one's own practice or purchased services increases die physicians earnings, while sj^endiiig inore cuts income.When hospitals are paid hy the plan, they are no longer pmd on a fee-for-servicc basis; instead, they receive payments per day in the liospital or per admission.The result has been a large redttction in the number aud length of hospitalizations, and less ordering of tests and procedures.
Managed care was successful iu reducing cost growth.Between 1993 and 1999, medical spending in the United States was constant as a share of GDP, after rising continuously since World War 2. Figure 1 shows the much lower growth rate of costs in the Uuited States in the 199()s than in previous decades.Compared to historical growth rates, medical costs hy 1999 were ahout 10 percent below expected levels.
As in other countries that have cut back, the evidence surveyed by Robert Miller and Harold Luft (1997) suggests that most managed-care savings have come without adverse health consequences.There seeuis to have been enough waste aud excessive fees to allow significant squet^zing without worsening patient outcomes.
Still, even with a far less egalitarian starting point in the United States than in other coimtries, incentive-based reforms like the managetl-care revolution have encountered enormous hostility.Many of the issues are familiar from other countries.With generous demand-side incentives but constraints on the supply side, people are uuhappy that they cannot get all the care they want.Maiiagcd-care rationing is a major concern.Further, adverse sclectiou has beeu a major side effect of managed care (Cutler and Richard Zeckhauser 2000).And where the Uuited States lias a comuiitment to equitv^-, as in tbe Medicare program, there is a fear about managed care eroding that.One of the arguments against creating more insurance market competition in Medicare is tliat it would reduce the universality ofthe system.As a result, policy is now facing a "managedcare backlash" more tlum a desire to increase medical-eare competition.
Steps to increase insurance market competition have not been confined to the United States.In Italy, legislation taking effect in 1995 allowed people to opt out ofthe social .securitysystem to obtain private health insurance.In (ierniany, legi.shitionenacted in 1993 and expanded in 1997 allows all eitizen.s the right to choose among competing "sickness" funds, us only higherincome workers were previously able to do.'^'At the time the law was enacted, there were about 1,000 sickness fnnds, so tliere was the potential for significant eompetition. Plan choiee in Germany has had some salutary effects on the medical system (Lawrence Brown and Volker Amelung 1999).Sickness fnnds are catering more to individual demands than they tised to, including emphasis on health promotion and expansion of benefits to itidude cotnplementary and alternative medicine.Consumers face tradeoffs between plan generosity and priee where previously there were none.
But the degree of competition is limited.While people can choose in.surunce companies, the companies are reqtiired to pay doctors on the same negotiated pay scale.Ftirther.plans cannot selectively contract with particular providers.Insurance is still seen as separate from medical-care provision.Thus, there is not mueh that plans can do to really change the nature of care provision.
To some extent, this was the desired outcome.At the time of the German reforms, there was conceni that competition would lead to sick people being denied coverage or discouraged from enrolling in particular plans.While payments to plans are partly adjusted for the health risk of enrollees, the •' ^ The {icririaii leyi shit ion followed rcrorms proposed in the Nt^thcrlands in 1987 (Frededk Schut 1995), adjustment factors used are not particularly extensive (age, sex, family status, and income).Were eompetition more intense, plans eould nse more sophisticated selection tnechanisms to limit their exposure to siek people,'^-In addition, tliere was concern that competition wotild come at the expense of social solidarity.Differential access to medical-care providers violates the longshared goal that the medical .systemshould treat everyone equally.
As a result, the overall impact of the German reforms has been modest.Germans periodically debate whether to extend the reforms or baek off from them.Competition has a natural constituency because the alternatives are not very appealing.The traditional way that Germany controlled medical costs was to rethice fees to hospitals and physicians.Decades of acrimonious negotiations hetween insurers, governments, and providers led to immense frustration with that system.On the other hand, if eompetition has little effect on costs-and the partial nature of reform, combined with the fundamentals of technology diffusion, virtually guarantee this ontcome-the competitive system may be seen as a failure as well.The deeision between enacting even more competitive reforms or reverting to the old system would be a diffienlt one (Brown and Amelung 1999).
A third set of eotmtries has attempted to introduce incentives within tlie provider community.The leading country taking this approach is die United Kingdom,'^^ The major British reform was the National Hedth Service and Gomniunity Care Act of 1990, passed by the Thatcher Administration (Julian LeGrand 1999), The thrust of the Thatcher reforms was to separate tlie pnrehase of medical care from the provision of ^~ One health insurer wrote a letter to its diabetic enroliees, pointing out that another plan was known to treat tliat fondition particularly well, •' •' Internal markets have also been created in Italy, Under a 1992 rflorni. larger hospitals are allowed to become independent public agencies, and can charge patients extra for services.
jottmal of Economic Literature, Vol.XL (September 2002) services.Most importantly, the Act allowed prinuiA' eare physicians to beeonie "fimdliolders."F"undlu)klers received a budget to purchase services such as elective surgery and o\itpatient phannaceiitieals.'^"'Funderholders could negotiate pavment rates with hospitals, and, to some extent, control outj^iatient phiirmaeeutical use.Fundhoklers could keep the surplus if spendiug was below the budget, provided that the money was reinvested iu the practice.Most general practitioners became fuudholders.In addition, uiost hospitals were inoved out of the public sector aud made into "tnists"-souiewhat similar to the uot-for-profit hospital in tbe Uuited States.Hospital tnists have iudependent governing bodies, which make decisions about technological iivailabilitv ;md pricing.
Separating tbe purchasing of care from the provision of care had some real effects.Studies by Howard Glennerster, Mauos Miitsaganis, andPat Oweus (1994) andCarol Propper andNeil Soderlund (1998) show that prices paid bv fundbolders ibr iLOspital seivices tell relative to prices paid by non-fundholders.Conrad Harris and Cileu Scriveuer (1996) show that fundhoklers had lower prescriptiou drug speuding tban did uon-fnndbolders.Bernard Dowling (1998) demoustrates that waitiug times for patients of fuudholders fell relative to waiting times for nou-fundholder physicians.
But iu other ways, the reforiu was teutative.Hospital trusts were still significantly regulated.Public approval was required for investment and capital decisions, aud tnists could uot keep auy surpluses they generated.They also had an iuiplieit claim on the public sector if they rau a deficit.As a result, the financial incentives tnists operated under did not encourage much change in care delivery'.Tbis was reinforced by the general desire of regional health authorities to avoid closing hospitals.Overall, while empirical evidence shows some salutary effects of the •^ Not iill of the pavments for these services come out of the same capitation amount.In atkfitioii, there Wits reinsurance for ver}' higli-cost cases.creation of primary care fundhoklers, the creation of hospital trusts did not seem to have major impacts on care (LeGrand 1999).
But even diese tentative reforms bave provoked considerable controversy, (>)uceru about adverse selectkni limited the extent to whicb incentives were tried at the start.Further, the emphasis on competition ratber than cooperation upset many.Indeed, Britain's Labor government elected iu the late 1990s explicitly opposed the emphasis on competition.It reformed tbe system to a "cooperative" model, away from the "competitive" model.General practitioners have been grouped into primar\' care groups, to jointly manage capitation payments for tbe people tbey serve.
But the Labor government did not completely backtrack, and incentives have actually been strengthened in some dimensions.For example, hospital tnists arc now allowed to keep surpluses they generate, providing more incentive to limit the care they provide.Similarly, primaiy care gronps will be allowed to retain surpluses as well, provided they are reinvested in the practice.The United Kingdom is thus experimenting with A balance between incentives and regulation, as are many other countries.

7, Sttitunanj
The slowness and zigzagging of reform in virtually all countries reflects many factors.Fear of making too sudden a cbange is an important brake.So too is concern uboiit unwanted ontcomes suob as adverse selection.And equity concerns liave played u large role as well.Since World War 2, countries have taken pride in kecpiug the market out of mediciue.Reversing this commitment is veiy difficult.
Arthur Okun formalized tiie tradeoff between equality und efficiency a quartercentury ago.Medical-care reform has that tradeoff aud more.Not only are etjuidit) and efficiency in conflict, but both are hostage to CD Italy ^ Japan Fif!,tiir 1. Increase in Medical Care a.s a Sliare of CIDP Stnirre: Data ATV from OECD (2(HI(n.Duta f(ir Ceniianv itre iiiissiiiK [or lUfiOs.Last VL^r of data is UK US and countries that joined the OECD in the Source: OECD i'lOlX)).
\KF OHT.IIINED OuTsiDK NATIO.NAL SYSTEMS . and Newhouse 2(K)0).These price rednetions are similar to what happened in other countries when medical spending WILS rednced.But the meclianism used in the United States wtLs a hit different.Coverumeuts in other couutries eould simply lower prices for all providers Eiithoveu 1993).Managed eompetition was one basis for the Clinton Health Plan iu tiie United States, along with heallh reform [)lans in other cnnntries snch as the Netherlands.'"' Slieriy Glied (2000) reviews the literature on managed care.Ctttler: International Medical-Care Reform 901 MEDICM,

TABLE 7 E
REFORM IN THK 199t)s