Edwin S. Rubenstein is an economist and an 4unct fellow of the Manhattan Institute. He also writes an economics column for National Review. The research for this article was supported by the Brunie Fund for New York Journalism.

Medicaid, New York State’s single fastest growing major spending program during the 1980s, is a major cause of the state’s fiscal collapse and of New York City’s budget woes as well. From 1980 to 1990 New York State’s share of the Medicaid bill increased by almost 150 percent, in real terms, rising from $1.4 billion to $3.5 billion in 1990 dollars. In 1975 Medicaid accounted for 5.6 percent of New York’s general fund expenditures. By 1990 that figure had more than doubled to 12 percent. (See Figures 1 and 3.) The Governor’s initial fiscal 1991 budget had Medicaid soaring to $4.2 billion, a 17.8 percent increase over 1990. The revised budget, drawn up in response to the fiscal emergency, proposes to keep Medicaid spending roughly flat, through reduced reimbursement rates and service cuts. But more fundamental reform has been elusive.

Moreover, the figures for state spending actually understate the problem. New York’s local governments face a $2 billion annual Medicaid bill of their own, over which they have little control. Medicaid, meant to provide basic medical protection for the poor, is a federally initiated program for which roughly half the costs are borne by Washington. But the most crucial decisions about the size, scope, and management of the program are made by the states. Albany has mandated an expensive, minutely regulated statewide version of Medicaid, for which county governments must pick up roughly 20 percent of the bill.

Medicaid now dominates most county budgets, taking as much as half of all county tax revenues in some areas. New York City, with about two-thirds of the state’s Medicaid population, has spent over $1 billion in city funds on Medicaid each year since 1982, and more than $1.5 billion in fiscal 1990.

Nevertheless, this bleak picture conceals considerable hope for the state’s fiscal future. We know that Medicaid costs can be controlled without compromising the quality of care, because other states have done it. If New York can too, we will have taken a large and relatively painless step toward fiscal health.

Why is Medicaid so expensive in New York? Part of the reason is that this state extends more types of aid to a broader segment of the population than most states. Federal Medicaid rules require states to extend basic coverage to recipients of AFDC (welfare) and recipients of Supplemental Social Security Income (SSI, an income program for the disabled poor). But states can opt to extend full or partial coverage to certain other groups, or more extensive coverage to the whole Medicaid population. New York extends more benefits to more optional groups than any other state and has a reputation for liberality in determining eligibility. (See box on page 51.)

Broad eligibility rules can be defended as a conscious, democratic decision to help more people. But broader coverage alone does not explain the explosion in Medicaid costs during the 1980s.

The number of New Yorkers receiving Medicaid benefits has not changed much in a decade. Most of the increased spending of the 1980s reflects higher costs per patient. From 1980 to 1989, total per patient costs rose from $1,989 to $4,523, an increase of 128 percent, or 51 percent adjusted for inflation. (These numbers include federal, state, and local shares, but spending parameters are largely set by the state.) With only 9.6 percent of all Medicaid recipients, New York State accounts for 19 percent of all Medicaid spending. New York now spends twice as much per Medicaid recipient as the rest of the country. (See Figure 2.)

Though New York’s general population is older, sicker, and has more poor people than the national average, New York certainly does not have the oldest or poorest population in the country. As for the Medicaid population, it is, by definition, poor in every state. Nor is there any evidence to suggest that New York’s Medicaid population is sufficiently sicker (because of AIDS, for example) than the average to account for New York’s extraordinary Medicaid costs.

In an attempt to discover why New York Medicaid is so expensive, the office of the State Comptroller Edward Regan conducted a ten-month study. The study, released in March 1989, compared New York to eight other highly urbanized, industrial states that provide more than the minimum Medicaid benefits: Massachusetts, Wisconsin, New Jersey, Ohio, Pennsylvania, Illinois, Michigan, and California. Among the findings:

* New York’s Medicaid program is more ambitious than those of other states, but its underlying costs are also higher, absolutely and relatively.

* The average New York Medicaid hospital patient spends more days (11.3) in the hospital than those of the other eight states, except Massachusetts, and more days than the general New York population (9.7 days).

* New York was one of only two states in the study for which the average hospital stay was longer for Medicaid patients than for the general population. In the other states studied, hospital stays for Medicaid patients averaged 7.2 days, compared with 7.5 days for the general population.

* New York’s Medicaid patients stayed more than twice as long in the hospital as California’s.

* New York spends far more on skilled nursing home care than the rest of the nation, devoting 23 percent of all its Medicaid funds to such care, compared with less than 10 percent for the rest of the nation. Length of stay was again a prime difficulty: 265.5 days for New Yorkers versus 199.5 days for the other states. Average daily costs were $86 in New York, compared with $51 for the other states.

* A major component of higher nursing home costs is higher labor costs. New York has more staff (103) per 100 occupied beds than any of the other eight states-except California. Nevertheless, professional staff costs were lower in California, because New York has a more expensive skill mix: more registered nurses, fewer nurse practitioners, nurses’ aides, and orderlies.

* Nursing aides are paid twice as much in New York City as in Los Angeles, though the two cities have comparable living costs.

* New York’s home health-care program provided nearly 140 visits per patient during 1987. In Massachusetts and Michigan (the only states that provided the Comptroller’s Office with sufficient data) the average figures were 41 and 13 visits per year, respectively.

* Home health-care costs grew at 17 percent per year from 1983 to 1988 until they consumed 11 percent of state Medicaid spending. Only one-tenth of this money is spent on actual medical services. The rest goes to personal care services, which often substitute for care that was previously provided by family or friends.

* Medicaid recipients are limited to two outpatient mental health-care visits per month in California and four per month in Ohio. In 1987, New York Medicaid paid for an average of more than four visits per week for 251 patients, and from two to four visits per week for 1,213 other patients.

* California’s Medicaid program, called “Medi-Cal,” serves 1.45 million more recipients and costs $4 billion less than New York’s program. California does use other funds to help pay for a few medical relief programs that New York funds exclusively through Medicaid; even so, New York’s Medicaid program is still several times more expensive per patient than California’s.

* Of the 50 states, New York’s Medicaid program spends the second most per recipient, Medi-Cal the fourth least, though at least one national study (by the Public Citizen Health Research Group) rates the two state programs as essentially equivalent in quality.

* In 1989 New York spent $1.2 billion more on skilled nursing home care than did California, while taking care of 17,000 fewer patients. New York’s per-recipient nursing home costs were 2.6 times California’s.

* In New York, inpatient hospital care, which accounts for about 30 percent of New York’s Medicaid costs, is about 50 percent more expensive per patient than in California.

As these last figures suggest, and as the comptroller’s report forcefully argued, New York could learn a good deal from California. For California once faced the same dilemma New York faces today. In 1982 the Golden State faced recession, a looming $1.5 billion budget deficit, and Medi-Cal costs that consumed a greater proportion of California’s budget than New York’s Medicaid program does today. Medi-Cal costs seemed out of control and were expected to double within six years.

Yet California got its Medicaid costs under control. It did this not by minor cost-control measures, or juggling of eligibility formulas, or one-time blows of the budgetary ax, but by fundamental reform that cut costs, apparently without compromising the quality or scope of care.

Essentially, California created a more competitive market in health-care services. Medi-Cal then exploited that competition to find providers who would take care of Medicaid patients for less.

There is no such thing as a “free market” in health care, either in California or anywhere in the United States. Nor is there any evidence that many citizens would want such a thing. But regulated markets can be more or less competitive, more or less efficient. The government of California surrendered some of its control over the health care market and found that both the state budget and the poor benefited.

The centerpiece of the California reforms was “selective contracting.” Under selective contracting only hospitals that contract with the state to provide services at specified prices participate in the Medi-Cal program. Medi-Cal patients can go to any hospital for emergency services, but otherwise they must use approved hospitals. Even emergency patients are usually transferred to hospitals with Medi-Cal contracts after they are stabilized.

Contract prices are based on past experience. Knowing how much it had paid a given hospital, or hospitals in a given area, to care for Medicaid patients in past years, the state negotiates for a reduction, or at least a low rate of increase. Only hospitals that submit attractive bids win contracts.

For California contracting has not been simply a way of squeezing more money from hospitals. Instead, it has forced hospitals to eliminate existing inefficiencies: unnecessary surgery, extra days of recovery, overpaid non-professional staff, suboptimal professional staff mixes, etc. It gave them a stake in holding costs down and a chance to make money by finding innovative ways to cut costs. It encouraged hospitals to specialize in services they could provide efficiently and to drop services that other hospitals were beating them on.

A formal evaluation by the U.S. Health Care Financing Administration estimated that Medi-Cal’s switch to a competitive system cut per-diem hospital rates by an average of 16 percent. California’s own review of the program reports savings of $1.5 billion over the program’s first seven years. Comptroller Regan’s study found that between 1982 and 1987 the cost to the state of inpatient hospital care actually declined 11 percent in real terms in California. During the same period those costs rose 40 percent in New York, adjusted for inflation.

Within three years after California began contracting, the average length of stay for a heart-attack patient admitted to a northern California hospital fell from about 13 days to six, for a savings (in 1985 dollars) of about $1,700 per patient. In 1982 a patient having a hernia operation would on average spend a day in the hospital at a cost of about $1,800; in 1985 that person was more likely to have the operation in an outpatient clinic, at a cost of about $ 1,000.

While other reforms helped keep costs down too, evidence from within California itself confirms that the key was increased competition. A study by Glenn Melnick of the UCLA School of Public Health shows that after 1982 the annual rate of increase in costs per discharged patient was 3.5 percent lower for hospitals in highly competitive areas than for those in less competitive areas.

This last datum points up the most important feature of California’s system. Selective contracting itself is not a revolutionary idea. It worked for Medi-Cal only because the California health-care market had enough competitors, whose facilities had enough excess capacity, to give the state a broad choice of hospitals and a lot of leverage in the marketplace. This, in turn was true only because California had wisely avoided some of the most powerful health-care regulatory schemes popularized in the 1970s.

Chief among these was the “certificate of need” strategy adopted by many states and still fiercely defended in New York. Under this approach, even private or voluntary hospitals are effectively prevented from buying expensive new equipment or adding beds unless state authorities certify that there is a genuine need for such facilities. The theory was that, since the health-care market is not really competitive, all costs incurred by hospitals, justified or not, are eventually passed along to either insurance companies or the taxpayer. Thus downtown may need only one high-tech trauma center, but if both downtown hospitals build one, the costs will be borne by the public. The only way to cut costs, it was argued, is to forbid one of the hospitals from building its trauma center.

In a completely uncompetitive market, in which health-care providers could simply pass along all their costs, plus a bit more for profits, without the consumers having any say-so, this theory might have made sense. But the U.S. health-care system has never been entirely such a “cost-plus” system. And the certificate of need strategy became popular just when this country was beginning to move even further away from cost-plus and (slowly) toward real prices and competition.

In California the certificate of need process had not been as constricting as in some states; with the 1982 reforms it was eliminated. Lucien Wulsin, chief consultant for the California Assembly’s Special Committee on Medi-Cal Oversight, explains that California abandoned the system in part because it did not accomplish even its basic objective: to rationalize the provision of services. Hospitals got around the restrictions by bureaucratic and political maneuvering.

In any event, the repeal of California’s certificate of need rules brought an explosion of hospital capital investment. By one measure it might be called overinvestment: As hospitals competed to fill more beds or pay the overhead on expensive new technology, some did go out of business. More often they dropped or deemphasized certain services for which the market was glutted in order to concentrate on others.

Overall, this surge of capital investment great increased the availability or “excess capacity,” of medical care in California. That gave big clients such as Medi-Cal the leverage to negotiate cost reductions for most services. In a direct repudiation of the theory underlying certificate of need and other central planning strategies, California has been cutting costs precisely because it has more beds than it needs. New York’s hospitals and nursing homes, under strict certificate of need regulation, are filled to overflowing, with no overbuilding allowed. Yet Medicaid (and other health-care costs) have soared, even as more and more New York hospitals sink into financial distress.

New York State officials implicitly confirm this critique. Responding to a draft of this article, Terrance McGrath, assistant director of the Public Information Office of the State Department of Social Services, commented:

By way of background, California has always had a loosely regulated Certificate of Need (CON) program, which has now been phased out. As a consequence, California’s hospital industry is over-bedded, or more than one-third empty. When such excess capacity exists, conditions are ripe for negotiating to increase market share.

By contrast, New York was the first state to enact such a CON law and it has maintained tight control over the supply of beds. Currently, the occupancy rate for general hospitals is 82 percent statewide and 86 percent in New York City. The optimum rate is approximately 85 percent. However, as recently as last year, some of New York City’s hospital occupancy rates approached 100 percent, with patients held for several days in emergency rooms and corridors (in violation of the state hospital code). Under such conditions, there is no incentive for hospitals to negotiate a better price for the state Medicaid program. Moreover, where would they put the additional volume of patients?

As Mr. McGrath argues, the overcrowding in New York hospitals might well make selective contracting less effective here. But that overcrowding is a result of conscious state policies to avoid “excess capacity.”

There is now a powerful consensus among health-care economists that certificates of need and other ambitious central planning techniques should be abandoned. But New York’s regulators, for the most part, disagree. Stephen Goodman, the state health program administrator, still believes certificate of need laws allow the state to spend “limited health-care dollars in a rational way. . . . Why would you build a hospital with 100 beds, if you can only fill 80 of them?”

The answer, surely, is that on average, the medical professionals closest to the action are most likely to know whether 80 or 100 beds are called for. What central planners gain in perspective, they lose in detail. Their rules tie the hands of the people on the scene, yet their decisions about equipment, staff, and hospital beds will often be wrong, no matter how sophisticated the decision-making process. Indeed, the more sophisticated the system, the slower and less flexible it will be. We would not think of having an Albany committee set rules on which patients must be treated first in an emergency room, or exactly what sort of surgery a certain patient needs. But Albany does effectively control how long that patient will have to wait for surgery and what equipment will be in the operating room.

Moreover, even if the regulators were right, and the builders of the 100-bed hospital were wrong, the state would stand to gain from deregulation, as overbuilt hospital s in need of cash flow would likely cut prices to compete for state contracts. Thus the more burdens the bureaucracy accepts, the less likely it is to achieve its primary goal. If the bureaucracy accepts the burden of protecting hospitals (or nursing homes or other providers) from overinvestment, it thereby denies itself not only the benefits of the hospitals’ expertise in managing expansion, but also the benefits of hospital mistakes.”

As David Swoap, the former secretary for the California health and welfare agency, argued in a 1986 article in the New England Journal of Human Services:

Competition cannot reach its full potential until the chains of excessive regulation are unclamped from hospitals and other health-care providers so that they can make management adjustments quickly enough to compete in a rapidly changing environment. . . . [This] is expected to save hospitals and thus, ultimately, consumers, approximately $400 million over the next five years. Now the marketplace, not government regulation, is expected to decide where new beds and new equipment are needed.

It would be fanciful to imagine anything like a fully free-market health care system in this country. But the California reforms, presented here in very simplified terms, moved that state in the direction of a more competitive market, whereas New York, if anything, has been moving the other way.

New York State’s health care bureaucracy has long been one of the country’s most ambitious; its regulations on nursing homes and hospital care have long been among the most stringent and complex in the nation. Such excess regulation seems to be one of the main reasons New York hospitals tend to be unusually large: Bureaucratic control favors big players, who have the economies of scale to pay for full-time liaison with regulators. Yet the comptroller’s report found a strong positive correlation between hospital size and the cost of inpatient care.

Despite the failures of central planning, New York shows no sign of abandoning the approach. In just the past two years the State Health Department has established for New York hospitals extensive new minimum operating standards in administration, medical staff, nursing services, and patients’ rights. That bundle of reforms is expected to cost the state a total of $270 million per year in additional Medicaid reimbursements and may cost hospitals tens of millions more in unreimbursed expenses. The Hospital Association of New York State reacted strongly to what it regarded as the arbitrariness of the new rules:

Policy priorities involving substantial public expenditures have been set unilaterally by the Department of Health through the regulatory process. The incremental ... requirements, i.e., those exceeding federal standards, added hundreds of millions of dollars of cost at a time when the healthcare system is struggling to meet the most fundamental needs in health care, such as nursing salaries, access, and care for the indigent.

Another aspect of the California reforms was to rely more heavily on Health Maintenance Organizations (HMOs) as providers of Medicaid services. HMOs are membership organizations that provide all needed medical services for a flat monthly fee. The fee is essentially an insurance payment, but unlike insurance companies, HMOs provide, or manage, patient care themselves. By managing patient care more closely they are usually able to reduce costs. The HMO member is secure, but disciplined: His primary physician must authorize all nonemergency procedures. As many HMO members are discovering, HMOs also provide pleasant relief from the usual double-barreled hospital and insurance company paperwork.

HMOs can be especially effective with poor patients, who are less likely than other patients to have a regular relationship with a family doctor, and who often pursue health services in a disorganized and dangerous way. Using the emergency room, or several different emergency rooms, as other people might use a doctor’s office, they drift from doctor to doctor. With no one to manage their care or take custody of their medical histories, they require frequent retesting and make it difficult for doctors who are ignorant of changes in their condition over time to diagnose their ailments. Such random care is expensive as well as dangerous. HMOs, by definition, manage care more comprehensively. A good HMO can thus be a particularly effective tool for improving health care for the poor while actually cutting costs. In California, Medi-Cal services provided by HMOs cost about 5 percent less than services offered by more traditional providers.

About 10 percent of Medi-Cal recipients have chosen HMOs, and California is seeking to increase that percentage. New York, however, keeps HMOs alive only as an eternal pilot program, with just 1 percent of Medicaid recipients enrolled.

The dramatic savings achieved by the California reforms immediately raise the question: Is it possible to save all that money without compromising care? The answer seems to be yes.

It is notoriously difficult to compare the quality of health care even across different hospitals, not to mention different states. The most apparently objective measure, the outcomes of similar cases, turns out to be overly subjective, since it would require impossibly difficult adjustments for the seriousness of illnesses and the general health of patients. Measuring patient satisfaction is another way to go at the problem, but good patient satisfaction surveys are still rare. A simpler but less subtle approach is to assess the formal outlines of health-care programs: what services are offered and to whom, and what assurances or mechanisms are built into the system to make sure those services are provided well.

A 1987 study by the Public Citizen Health Research Group, an advocacy group associated with Ralph Nader, assessed state Medicaid programs along roughly these lines. Such an assessment will tend to favor high-spending states because it does not attempt a detailed analysis of the quality of the services delivered. Nevertheless, on its overall evaluation Public Citizen ranked both New York’s program and California’s much cheaper program among the top six, and in the subcategory for “quality” ranked California ahead of New York. Public Citizen praised California’s nursing home licensing standards and enforcement mechanisms, and said New York’s quality control in nursing homes could be “improved substantially.”

The Public Citizen survey suggests that though there is a relationship between cost and quality, it is a loose one. Figure 5 lists the 20 highest-ranking state Medicaid programs, according to Public Citizen. It also shows the average cost per patient for those states. Many relatively low-cost states rank near the top, while many relatively high-cost states rank lower down.

Lucien Wulsin, chief consultant for the California Assembly’s Special Committee on Medi-Cal Oversight, says there has been no apparent decline in the quality of care since the 1982 reforms, though there have only been a few studies of the issue. In December 1987 the executive director of the state association of public hospitals reported that contracting had not brought the declines in quality of care that critics had predicted. Several other studies, by Lucy Johns of Lewis-ICF for the National Governors’ Association, and by health systems agencies in California in conjunction with the State’s Office of Statewide Health Planning and Development, tend to confirm that quality has not declined in California.

Mr. Wulsin reports that studies have indicated a fall-off in Medi-Cal recipients’ access to prenatal and dental care. He also believes the solution to that problem is contained in the system: The state should insist on increased prenatal and dental care when next it negotiates rates with the hospitals. New York lags behind California in prenatal care.

The California reforms, as briefly outlined here, would not address every inefficiency of New York’s system. Another aspect of the system that cries out for review is New York’s decision to spend more of its Medicaid dollars on long-term care than other states. We spend less than 20 percent of Medicaid funds on such short-term items as outpatient hospital services, clinic services, prescriptions, and physician fees; the average for the rest of the country is nearly 30 percent. These are the sort of largely preventive services that the very poor need most.

A recent state crackdown on Medicaid fraud resulted in highly publicized arrests of private physicians accused of padding their Medicaid bills. But New York Medicaid typically pays physicians only $7 to $11 per visit, considerably less than in most states. As a result, only 15 percent of the state’s doctors will see Medicaid patients, and private practitioners are hard to find in poor neighborhoods. Patients instead flock to already overcrowded and underfunded clinics, or use expensive emergency rooms for routine matters. Worse still, they neglect small problems until they become big ones, requiring long hospital stays.

These mistakes too might be rectified by a health-care policy that takes more seriously the disciplines of the market and the difficulties of central planning. Surely a system inclined to take market signals seriously would long ago have concluded that the shortage of doctors in poor neighborhoods meant that Medicaid was pricing their services incorrectly, and that the abuse of emergency rooms suggested that Medicaid’s primary constituency wished more funds shifted to short-term preventive care.

The full range and complexity of the California reforms cannot be addressed in an article. Nor is California the only state that has been successful in reforming Medicaid and getting costs under control. Other states have taken instructive and successful steps as well. But most successful reforms hum along to California’s theme: Reduce the responsibilities of the central bureaucracy, allow providers and patients to make more decisions, and take full advantage of competition and feedback from the market wherever possible. By doing the same, we in New York could tame our most fearsome fiscal threat.


One reason New York State’s Medicaid bill is so high is that it extends generous coverage to the “medically needy,” a particularly expensive-to-cover “optional group” not included in the basic federally defined Medicaid population.

The medically needy are people who would not ordinarily be considered poor but whose incomes are squeezed by extraordinary medical bills. That is what makes them so expensive to cover: By definition these are expensive cases.

New York extends more services to a larger segment of the potentially medically needy population than any other state. As a result New York spends half of all its Medicaid funds on the medically needy, though they make up only a quarter of the Medicaid population.

The medically needy are supposed to be people who would be driven into poverty without Medicaid coverage, but the program is an obvious target for manipulation. Recipients’ income and assets cannot exceed specified amounts. But most essential assets—homes and furnishings, automobiles, and personal belongings—are exempt. Even income-producing assets such as stocks, bonds, and rental property may be exempt if they do not push the recipient over the acceptable income level.

With careful estate management the loopholes can be opened wider still, especially for couples in which one partner is healthy. Some of the more obvious moves include transferring assets to a trust or to the healthy spouse, or having the spouse transfer assets to children.

The laws governing asset protection are meant to prevent fictitious transfers, aimed solely at making someone eligible for Medicaid. Those laws are complex, but the specialists with whom we spoke believe New York’s laws allow more asset protection than those of other states. Moreover, New York is home to an aggressive corps of estate managers and tax lawyers, partly because New York City is the capital of the country’s legal community and partly because very high state and local taxes encourage aggressive estate management.

New Yorkers may simply be exceptionally savvy loophole hunters. In Vermont, one of the handful of states whose income limit for the medically needy is higher than New York’s, only 9 percent of the Medicaid population consists of the medically needy, compared to New York’s 25 percent.


City Journal is a publication of the Manhattan Institute for Policy Research (MI), a leading free-market think tank. Are you interested in supporting the magazine? As a 501(c)(3) nonprofit, donations in support of MI and City Journal are fully tax-deductible as provided by law (EIN #13-2912529).

Further Reading

Up Next