Economic Viewpoint - January 2006

Introducing Market Forces into Health Care: A Tale of Two Countries

Market forces have been introduced into two health care systems in an attempt to create incentives for providers to innovate in ways that improve outcomes, including patient satisfaction, and reduced costs.   By market forces, Enthoven refers to significant responsible consumer choice among providers, and providers who gain their incomes from serving the consumers who chose them.   This Economic Viewpoint summarizes what happened, and offers some reflections on the experience.

• Related Event
• Featured Reading

(First Presented at the Fourth European Conference on Health Economics Paris, July 10, 2002; Revision for Post-Conference Publication, July 11, 2002 )

The British National Health Service

In 1985, I wrote: “The NHS is caught in the grip of forces that make change exceedingly difficult to bring about, a ‘gridlock’ of its own… Managers have no powerful incentive to make efficiency-improving changes. In fact there are many disincentives to ‘rocking the boat.’”  I went on to suggest “an internal market model for the NHS” in which District Health Authorities would be re-cast as purchasers of services on behalf of their populations, and not mere conduits of directives and money from above to providers. Purchasers would be free to buy what they considered to be the best services for their patients, from wherever supplied within the NHS. They would be paid for services they provide to others’ patients. Managers would be freed from massive amounts of central directives, and held responsible for the overall results they achieve.ⁱ

     Criticism of the organization of the NHS is no longer viewed as a partisan matter. In 2002, Alan Milburn, Secretary of State for Health, said of the NHS: “While its values are right, its structures are wrong. …  Patients are disempowered with little if any choice. The system seems to work for its own convenience not the patient’s—a frustration that is shared by patients and staff alike.  The whole thing is monolithic and bureaucratic…. For fifty years, the structure of the NHS meant that governments—both Labour and Conservative—defended the interests of the NHS as a producer of services when they should have been focused on the interests of patients as the consumers of services.”ⁱⁱ

     Fundamental change in this structure began in January 1989, when the Thatcher government published Working for Patients, a strategy for reforming the NHS based on increased local responsibility, patient choice, and competition. Key elements included “the purchaser-provider split” in which District Health Authorities would be re-cast as purchasers, not providers; self-governing NHS Hospital Trusts that would earn revenue from the services they provide and would operate with greater independence; “General Practitioner (G.P.) Fundholders,” large G.P. practices that could take responsibility for an increased budget with which to buy elective hospital and other services. G.P. practices would “be encouraged to compete for patients by offering them better services.” ⁱⁱⁱ “The practices and hospitals which attract the most custom will receive the most money.” 

     With government encouragement, some G.P. Fundholder practices teamed up to accept larger budgets accounting for most of the per person costs of their patients in “Total Purchasing Pilots.”   In fact, a great variety of cooperative G.P. purchasing innovations grew up all over the country. ^iv

      In 1998, I could find some evidence of improved economic performance, but not much, and not very strong.  Important measures of outcomes, service and satisfaction were lacking.^v

     Creating a market model that works in such circumstances takes a great deal more than merely re-casting the players and assigning them budgets.  Institutional foundations need to be in place, but were not to an adequate degree in the early 1990s.  Purchasers and providers need reliable information on outcomes, quality and cost on which to base decisions.  The government was just beginning to develop costing of services.  Information on outcomes and quality was virtually nonexistent, even years later.  There must be purchasers who are motivated to make the efforts needed to demand improvement and who have the freedom to purchase selectively.  Districts found their authority very constrained, and they did not face powerful incentives to use their purchasing power to demand improved performance.  There need to be providers capable of responding to market forces, something hospitals could barely do and with great difficulty. Mrs. Thatcher said: “Money follows patients,” but in fact, in the design implemented in the early 1990s, patients went where their district had contracted for services.  The government needs some distance from providers so that it can deal with them at arm’s length.  There needs to be a pro-competitive regulatory framework, one that might have, for example, prevented anti-competitive mergers.  And more. My advice in my Rock Carling Lecture in 1999 was that “…the most practical way forward now is to build on the strengths of the internal market and to try to correct the factors that held it back.”

     During the 1990s, the Labour Party denounced the Thatcher reforms, promised to abolish them, claimed they had failed, and then, after being elected, said they had abolished them.  In fact, while claiming to be doing something different, they built on and extended the Thatcher reforms and took advantage of the lessons learned in the 1990s.   They kept the Purchaser-Provider split.  Rather than abolishing G.P. Fundholding, they built on the Total Purchasing Pilots, and ordered all G.P.s to participate in Primary Care Trusts (PCTs), fundholders, which would eventually control 75% of the total budget. ^vi  They kept the NHS Trust Hospital idea. Now the Secretary of State proposes to take the idea a step further: with government encouragement, NHS hospitals may become independent not-for-profit institutions, called “NHS Foundation Trusts,” more like hospitals in Europe, the USA and Canada. Per the Secretary of State:  “…NHS Foundation Trusts will be freed up from having to respond to an excessive number of prescriptive central demands, guidance and reporting arrangements. As free-standing organizations they will be held to account through the commissioning process rather than through day to day line management from Whitehall.”^vii  (That is what the NHS Trusts were supposed to be.) The NHS Plan includes contracting with non-government hospitals, “the most appropriate provider—be they public, private or voluntary”, and even European hospitals. ^viii 

     The government proposes to pay hospitals for services based on “Health Resource Groups,” fixed prospective payments per case, similar to the practice of the US Government paying hospitals in the Medicare program. (I presume that, like Medicare, these prospective payments will be adjusted for local factor costs and for the additional costs of teaching programs.) Some have criticized this as “price controls.”  But I think it is sensible. This method spares the PCTs the costs of negotiating and contracting, and focuses their decisions on quality and access. And this approach prevents hospitals from exploiting monopoly power in local markets. This is the government acting as a purchaser, not a third party regulator.

     The present government is making substantial progress on quality-related information, available to the public, again building on previous efforts. And it is developing and enforcing national standards of care, something clearly lacking in the 1990s. 

     In other words, the Milburn redefinition of the National Health Service is a very promising and significant extension of the Thatcher reforms, including the improvements one would expect from learning from experience, and the incorporation of new developments from outside the NHS, such as evidence-based medicine and continuous quality improvement.  It emphasizes patient choice, money following patients, Primary Care Trusts in buyer-seller relationships with NHS Hospital Trusts or free standing hospitals.

     What this design might eventually accomplish is to distance the government from specific providers and to recast the NHS as a purchaser on behalf of patients, fully committed to the interests of patients, rather than being captive to NHS provider interests.

     Some major recent developments are shedding light on possible future directions for the British National Health Service.^ix  This includes:

• PCTs are being merged (from 300 to 100) and becoming realigned with local government social services and Strategic Health Authorities (SHA) are becoming realigned with regional government authorities.^x Although it is hoped that this will improve horizontal coordination between the PCTs and local government social services and between SHAs and regional government services, it has two potentially negative implications.  First, it could make vertical integration among different levels of care more difficult. And second, it is a return to central planning rather than market based incentives, thereby undermining internal market control mechanisms. 
• The capital investment budget is expected to drop from 7 to 3 percent of public expenditure on the NHS by 2008.  Although this will help control overall rising costs in the NHS, there is a risk that it signals a return to past traditions of under-investing in capital stock.  It would not take long before recent gains in this area were lost.

     There is also increasing concern about the emerging levels of overspend by NHS providers. Large increases in real income have been offset by generous, centrally agreed pay increases for doctors and nurses, and a one off provision to pass the future risk of NHS Pensions from the Treasury to Department of Health (DH). This policy is creating pressure for greater co-ordination of quality, financial and competition regulation of NHS activities.   A significant overspend in 2005 has led to pressures to postpone non essential surgery and other activities designed to improve quality and reduce waiting times. Perhaps, growth in NHS expenditure during the past five years has been so fast that some slow down is inevitable.

     These observations led me to conclude with some cautionary notes. First, if “money follows patients” in a fixed global budget, there will be losers as well as winners.  Will the government let the losers lose revenue and be forced to shrink, or will it bail them out  with “extra contractual payments,” inevitably at  the expense of winners, thus destroying the incentive to be a winner. The effectiveness of the Internal Market in the l990s was considerably attenuated by this practice.

     Second, the NHS has been under central control for over 50 years.  Recall Aneurin Bevan’s bedpans, whose falls were to be heard in Parliament.  These are deeply ingrained habits.  Will the government be able to let go? Will ministers be able to resist responding to every problem with a blizzard of new directives?  Will politicians be able to resist making the details of health care into political issues?

     Will the public employee unions allow genuine and fair competition from hospitals in the private sector?  It will take great courage for this government to persevere in this course. It remains to be seen whether this is a genuine policy or mere rhetoric.

     And fourth, in the present design, the PCTs will be territorial monopolies. Mergers of PCTs may make this problem worse.  I doubt they will face effective incentives to innovate to improve service and care.  I think the possibility of competition among PCTs should be considered seriously. A pilot project in which people could change PCT in pursuit of better service would be compatible with Mr. Milburn’s commitment to consumer choice.

The USA

     The modern era of health insurance in the USA began during and after World War II when employers began offering health insurance to their employees.  Virtually all of this insurance fit with the model demanded by the medical profession: fee for service payment, free choice of provider for the patient, free choice of prescription for the doctor, direct doctor-patient negotiation over fees, and no discrimination among providers by the insurers, like France’s Charte Medicale.^xi Most doctors were in solo or small single specialty group practice.  In 1965, the Congress enacted the tax-funded Medicare and Medicaid  Programs to cover all the aged and disabled and the “deserving poor.” These followed the traditional fee for service model.  Not surprisingly, the inflationary incentives in this model, combined with government sponsored expansion of medical schools, building of hospitals and biomedical research drove up the share of GDP spent on health care.

     In the 1970s, the government tried several regulatory interventions to limit spending growth, including price controls on doctors and hospitals, regulation of capacity expansion called Comprehensive Health Planning, and regulation of medical practice, called Professional Standards Review Organizations.  None of it worked.^xii  The doctors protected their real incomes by increasing volumes of services.  Comprehensive Health Planning couldn’t stand up to the pressures from the many who would benefit from hospital building. Public policy ignored economic incentives.

     There was alternative approach to medical organization and finance known as Prepaid Group Practice, in which large multi-specialty group practices contracted to provide comprehensive health services for a fixed per capita payment, set in advance, instead of the usual fee for service method.  Their approach included practicing in teams, good patient access to primary care, matching resources used to the needs of the population served, early diagnosis and treatment, and disease prevention and management. Competition for members and per capita prepayment gave them an incentive to use resources wisely.  Studies showed that they could provide high quality care for some 25% less than fee for service care, and they did this in the absence of real economic competition. ^xiii

     In 1976, Presidential candidate Carter promised that he would bring the American people universal, comprehensive health insurance.  I was retained by the Carter Administration to develop a plan for redeeming that promise. I called it “Consumer Choice Health Plan,” a plan for universal health insurance based on regulated competition in the private sector.^xiv This idea later became known as “managed competition.”^xv The plan would resemble the Federal Employees’ Health Benefits Program (FEHBP), the health insurance scheme that covered about 10 million federal government employees, retirees and dependents.  This program offered beneficiaries a wide choice of health insurance plan, including traditional fee for service, Prepaid Group Practice, and other arrangements.  The government contributed a fixed dollar amount based on a formula related to the premiums of the largest plans. The employee made his choice and paid the difference. Thus employees could benefit personally and financially by choosing economical health care.  One striking feature of this rational economic model of consumer choice is that Prepaid Group Practices had large shares in that market  in the regions in which they operated. (An important feature of managed competition that is not, but ought to be, a part of the FEHBP is a design to prevent uncompensated adverse risk selection, that is “risk adjustment” of premiums.)

     The key idea of my plan was to expose traditional free choice fee for service to economic competition from Prepaid Group Practices and other innovative arrangements that we now call Individual Practice Associations, Preferred Provider Organizations, Point of Service Plans and the like, let people make their choices, and see the health care system transformed, gradually and voluntarily, from inflationary fee for service to economical plans that reflected people’s preferences and willingness to pay.   The government’s financial contribution could be keyed to the lowest cost plans in each region, the least inflationary component of the health care system.

      The proposal attracted discussion and some prominent supporters, but President Carter did not accept it.

     From the point of view of public policy to improve the health care system, the 1980s were mostly wasted.  President Reagan liked the rhetoric of markets and competition, but he mistook that for merely abolishing some existing regulation.  Apparently his advisors did not understand that to create market forces that would motivate improvement of quality and value for money, the employment-based health insurance system would have to be restructured fundamentally so that every individual employee would have a responsible choice from among a variety of health insurance plans.

     The 1980s did see rapid growth of entities sometimes called “managed care” or “Health Maintenance Organizations” (HMOs), really insurance companies without their own doctors or medical groups, who tried to implement at least some of the cost saving features of Prepaid Group Practice, while contracting with independent doctors.  I call them “Carrier HMOs” to distinguish them from “Delivery System HMOs” or Prepaid Group Practice.   Going into the 1990s, these entities grew rapidly and were able to slow the growth of health expenditures substantially, but temporarily by taking advantage of the surplus of providers and driving hard bargains for prices. But they did not bring about fundamental structural change in the delivery of care.

     In the presidential election year of 1992, both candidates endorsed “managed competition” as their preferred solution to America’s problems of cost and access to health care.  Neither understood it, or if they did, they didn’t mean what they said. Candidate Clinton used the rhetoric of managed competition, but when it came to an actual plan, he presented a very complex scheme for a complete government takeover of all health care finance.^xvi It went nowhere in the Congress.

     Some employers adopted managed competition. It was working well for them on their limited scale, but it was not transforming the health care delivery system because they covered too few people.  The main managed competition employers were the federal and state governments and universities, in part because they had large enough concentrations of employees to make offering multiple choice of health insurance plan practical.  Also, the State of California created a pooled purchasing arrangement that offered a wide range of choice of health insurance plan for employees in small public sector employer groups. A few other such “exchanges” came into existence.

     With a few notable exceptions, private sector employers did not adopt managed competition.  In 2002, Maxwell and Temin published a survey of the 500 largest corporations in America.  They found that less than 10 per cent of these employers offered choice and a fixed dollar contribution to their employees.^xvii  Smaller employers were even less likely to do so. Another survey indicated that about 17% of all workers, public and private, are offered a choice and a fixed dollar contribution.^xviii

     Most private sector employers followed three themes in purchasing health insurance: a single source of health insurance, rather than multiple choice at the employee level, a policy that the employer would pay all of the insurance or, if there were choices, a flat 80 to 100% of the premium of the employee’s choice, and self-insurance, i.e. the employer itself would bear the insurance risk and hire an insurance company to process and pay claims from the employer’s own bank account.

     Many employers resist offering choices for several reasons. They see greater administrative cost and complexity. Many are too small to make multiple health insurance contracts practical.  Insurance carriers don’t like to compete at the employee level within a group.  They would prefer to have a monopoly within each employment  group and the inelastic demand that goes with it. They offer employers a better price if they can insure all the employees. Some fear adverse selection.  And often, good alternatives in the form of more economical health plans do not exist. (Of course, it should not be surprising that they do not exist if employers do not create a market for  them.)

     In failing to offer choices, employers fail to take advantage of effective managed care.  The single source policy generally locks everybody into the costly fee for service system. In particular, it denies choice to those who would prefer less costly care if they could keep the savings.  Employers say they want less costly care, but the market environment they create does not accept cost-reducing innovation.

     Effective managed care is not a good candidate to be a single source of health insurance for a whole group. An efficient managed care organization must carefully select providers for quality and economy and willingness to cooperate with utilization management and quality improvement programs.  Not everybody’s favorite doctor can participate in such a cohesive organized system.  Such a system cannot be a single source because patients must be members by choice, and in a frame of mind to cooperate.   Many may want to keep their own doctor who is not in a particular delivery system. The facilities may not be convenient to where they live.  And many may prefer the traditional style of care and be willing to pay its extra cost.

     Efficient health care plans link insurers and providers.  People understandably want to choose their doctors. So people must be able to choose the managed care plan that includes their doctors.  Patient satisfaction is powerfully related to whether the patient had a choice of plan.^xix

     About 20% of employees were put into managed care without a choice.^xx  To ease the pain, employers often demanded that the managed care organization include in its network all of the doctors and hospitals their employees wanted to use, which became practically everybody.   Since provider selection is a key to quality and economy, employers damaged the ability of managed care to improve both.  Moreover, this employer practice destroyed the bargaining power of managed care: once a provider knew it had to be included in a network, it could drive a very hard bargain. Moreover, managed care organizations were positioned by employers as the agents of the employers’ policy of taking away their previously unlimited benefits. Employee anger at the “takeaway” was directed toward managed care rather than the employers who were responsible for it.   Research showed that people put into a managed care organization without a choice were several times as likely to be dissatisfied as people who were there by choice.^xxi

     Employers got into the habit of paying 100% of the employee’s health insurance because that was excluded from the employee’s taxable income and because health benefits were an attractive source of bargaining prizes for trade unions.  For those who did convert to fixed dollar contributions set at the price of the lowest priced plan, there was a difficult transition because it meant taking something away from those employees accustomed to the most expensive fee for service plans. Complex transitions and compensations had to be worked out.  But the prevalent employer policy of paying all of the premium or a uniform high percentage of it creates complete insensitivity to price on the part of the employee.^xxii

     Employers had several reasons for preferring to bear the risk and pay for health care out of their own bank accounts, including exemption from regulations and taxes on the insurance industry.  But this almost always means inflationary fee for service payment, with no reward to providers for reducing people’s need for medical care.

     Private sector employers, for the most part, did not convert to managed competition for several reasons.  First, they face a collective action problem. One employer, acting alone, will not be rewarded with a efficient transformed delivery system. That would take the similar action of many employers in a market.   Second, the promise of an efficient transformed delivery system would take years to be realized, a decade or more. Private sector employers focus on the short term. There is no short term solution to this problem. This is definitely a case of short-term costs for uncertain long-term gains.

     Collective action is needed. I think our government needs to mandate and support the formation of neutral exchanges, that is institutions that bring together many health insurance plans and consumers into a large market in which each person has a wide choice of plan, and in which uncompensated adverse selection is minimized.  We have examples of successful exchanges.  And second, our government needs to limit the amount of employer contribution to health insurance that can be excluded from each employee’s taxable income so that, above that limit, employees are using their own net after tax dollars and will therefore be sensitive to price. Today there seems to be little prospect of either happening.

     So today in 2002, we in the USA are back to “double digit inflation” in health insurance premiums. In fact, this year, many employers are experiencing 25% or 30%  increases, devastating amounts. That will bring the average price in employer health plans to well over $9000 per family per year.  The private sector employers appear immovable in their policies, and the inflationary impact of their policies is spilling over onto the “managed competition sector.” That is, the most efficient forms of organization and finance are being limited in their growth or even destroyed by the employer policies that do not reward efficiency. 

     Much of the increase in premiums is happening because of the diffusion of effective technologies that are extending and benefiting people’s lives. Per capita consumption of joint replacements and angioplasties is rising rapidly.  But at the same time, these increases strain public finances, suppress real wages, price health insurance out of reach for families of moderate means or for taxpayers who might help them, thus driving up the number of people without health insurance. 

     Only someone who does not understand economics would suggest that the USA today has a rational functioning market economy in health services.

Some Lessons from the Experience

Reflection on these experiences leads me to the following lessons:

1. Markets for health insurance and health care services are beset by the well-known causes of market failure—uncertainty, moral hazard, adverse selection, asymmetry of information, plus the implications of a moral judgment that nobody should go without needed medical care for lack of ability to pay. This makes creation of a  market in health care that drives improvement a particularly complex undertaking.  The details are important and must be got right.  Not any thing that sounds like “competition” or “markets” or “private sector” will necessarily improve economic performance.
2. All the main elements must be in place: information on quality and cost; purchasers with incentives to seek value for money and freedom to buy selectively; providers capable of responding to market forces; a pro-competitive regulatory framework (i.e. a competition or anti-monopoly policy); a capital market; and a careful sorting out of what can and cannot be left to the market.  Partial implementation is not likely to work.  Many American employers thought managed care would save them money, which it did temporarily in the mid-1990s, but they failed to create a market for managed care with price elastic demand that could motivate continuing improvement.  And they failed to create the incentives that would lead employees to choose and accept the limitations of managed care.
3. The interests of employers, workers and providers are powerful and focused on resistance to any change that would disadvantage them.  None of them are interested in an efficient market. They are interested in preserving their positions.  To overcome them requires a great deal of understanding and commitment to change by the rest of society.
4. People diversify in consumption, but specialize in production.  Therefore, they tend to vote their provider interests.  As a consequence, government systematically favors provider interests over consumer interests, as Secretary Milburn observed of the NHS.^xxiii Markets favor consumers with purchasing power.  A consumer-centered, government controlled, health care system is a contradiction in terms, or at least a very difficult challenge. 
5. People respond to many incentives other than economic ones, especially in health care. Yet it is important to get the economic incentives right so that resources will flow to those people and institutions that do the best job.  Engineering economic incentives is crucial to a good long-term outcome. In health care, it is not possible to make them perfect, but they can be made roughly right, and that, in turn, can be improved upon.  Economists have valuable expertise in this domain.
6. The fact that something is done in the private sector does not mean that rational economic incentives necessarily apply.  The fact that something is in the public sector does not necessarily mean that they do not. In the USA, it is the public sector employers who have done the best job of implementing rational economic structures for employee health care.
7. I find it a great irony that a British Labour government appears to be well ahead of private-sector employer-purchasers in the USA when it comes to bringing market forces to bear on health care services. That is the way it appears to me today. How can this be?  In Britain, a government with a majority in Parliament is  held responsible for outcomes by the whole population, and it has the power to enact change. In the UK, the performance of the NHS got bad enough that it became a salient issue and the government had to act. The political imperative of re-election apparently overcame Labour’s aversion to market or consumer choice models.

 In the USA, power is very diffused. For example, I am not sure which rascals to blame for the recently enacted farm subsidies or other "pork barrel" spending in the USA.  There is little party discipline.  Proposed legislative changes are held hostage to a great variety of parochial interests. Incrementalism is one of the first laws of our politics.  Things have to get very bad before a strong consensus for fundamental change can  develop.  As President Clinton’s experience with health reform illustrates, the President is not in charge, even if his party controls the Congress. In the USA, the cost, quality and access to health care will have to get a lot worse before government will be able to act to make them better.