Diane Archer has a post at the Health Affairs blog arguing that Medicare is more efficient than private insurance. One can only reach such a conclusion through such sleights of hand as conflating spending with cost, and by ignoring most of Medicare's administrative costs.
As a pre-buttal, I offer this excerpt from a paper I wrote about a "public option" (emphases generally added and citations omitted):
Is Government More Efficient?
Supporters of a new government program note that private insurers spend resources on a wide range of administrative costs that government programs do not. These include marketing, underwriting, reviewing claims for legitimacy, and profits. The fact that government avoids these expenditures, however, does not necessarily make it more efficient. Many of the administrative activities that private insurers undertake serve to increase the insurers’ efficiency. Avoiding those activities would therefore make a health plan less efficient. Existing government health programs also incur administrative costs that are purely wasteful. In the final analysis, private insurance is more efficient than government insurance.
Time magazine’s Joe Klein argues that “the profits made by insurance companies are a good part of what makes health care so expensive in the U.S.and that a public option is needed to keep the insurers honest.” All else being equal, the fact that a government program would not need to turn a profit suggests that it might enjoy a price advantage over for-profit insurers. If so, that price advantage would be slight. According to the Congressional Budget Office, profits account for less than 3 percent of private health insurance premiums. Furthermore, government’s lack of a profit motive may not be an advantage at all. Profits are an important market signal that increase efficiency by encouraging producers to find lower-cost ways of meeting consumers’ needs. The lack of a profit motive could lead a government program to be less efficient than private insurance, not more.
Moreover, all else is not equal. Government programs typically keep administrative expenditures low by avoiding activities like utilization or claims review. Yet avoiding those activities increases overall costs. The CBO writes, “The traditional fee-for-service Medicare program does relatively little to manage benefits, which tends to reduce its administrative costs but may raise its overall spending relative to a more tightly managed approach.”7 Similarly, the Medicare Payment Advisory Commission writes:
[The Centers for Medicare & Medicaid Services] estimates that about $9.8 billion in erroneous payments were made in the fee-for-service program in 2007, a figure more than double what CMS spent for claims processing and review activities. In Medicare Advantage, CMS estimates that erroneous payments equaled $6.8 billion in 2006, or approximately 10.6 percent of payments. . . . The significant size of Medicare’s erroneous payments suggests that the program’s low administrative costs may come at a price.
CMS further estimates that it made $10.4 billion in improper payments in the fee-for-service Medicare program in 2008.
Medicare keeps its measured administrative-cost ratio relatively low by avoiding important administrative activities (which shrinks the numerator) and tolerating vast amounts of wasteful and fraudulent claims (which inflates the denominator). That is a vice, yet advocates of a new government program praise it as a virtue.
Medicare also keeps its administrative expenditures down by conducting almost no quality-improvement activities. Journalist Shannon Brownlee and Obama adviser Ezekiel Emanuel write:
[S]ome administrative costs are not only necessary but beneficial. Following heart-attack or cancer patients to see which interventions work best is an administrative cost, but it’s also invaluable if you want to improve care. Tracking the rate of heart attacks from drugs such as Avandia is key to ensuring safe pharmaceuticals.
According to the CBO, private insurers spend nearly 1 percent of premiums on “medical management.” The fact that Medicare keeps administrative expenditures low by avoiding such quality-improvement activities may likewise result in higher overall costs—in this case by suppressing the quality of care.
Supporters who praise Medicare’s apparently low administrative costs often fail to note that some of those costs are hidden costs that are borne by other federal agencies, and thus fail to appear in the standard 3-percent estimate. These include “parts of salaries for legislators, staff and others working on Medicare, building costs, marketing costs, collection of premiums and taxes, accounting including auditing and fraud issues, etc.”
Also, Medicare’s administrative costs should be understood to include the deadweight loss from the taxes that fund the program. Economists estimate that it can easily cost society $1.30 to raise just $1 in tax revenue, and it may sometimes cost as much as $2.36 That “excess burden” of taxation is a very real cost of administering (i.e., collecting the taxes for) compulsory health insurance programs like Medicare, even though it appears in no government budgets.
Comparing administrative expenditures in the traditional “fee-for-service” Medicare program to private Medicare Advantage plans can somewhat control for these factors. Hacker cites a CBO estimate that administrative costs are 2 percent of expenditures in traditional Medicare versus 11 percent for Medicare Advantage plans. He writes further: “A recent General Accounting Office report found that in 2006, Medicare Advantage plans spent 83.3 percent of their revenue on medical expenses, with 10.1 percent going to nonmedical expenses and 6.6 percent to profits—a 16.7 percent administrative share.”
Yet such comparisons still do not establish that government programs are more efficient than private insurers. The CBO writes of its own estimate: “The higher administrative costs of private plans do not imply that those plans are less efficient than the traditional FFS program. Some of the plans’ administrative expenses are for functions such as utilization management and quality improvement that are designed to increase the efficiency of care delivery.” Moreover, a portion of the Medicare Advantage plans’ administrative costs could reflect factors inherent to government programs rather than private insurance. For example, Congress uses price controls to determine how much to pay Medicare Advantage plans. If Congress sets those prices at supracompetitive levels, as many experts believe is the case, then that may boost Medicare Advantage plans’ profitability beyond what they would earn in a competitive market. Those supracompetitive profits would be a product of the forces that would guide a new government program—that is, Congress, the political system, and price controls—rather than any inherent feature of private insurance.
Economists who have tallied the full administrative burden of government health insurance programs conclude that administrative costs are far higher in government programs than in private insurance. In 1992,University of Pennsylvania economist Patricia Danzon estimated that total administrative costs were more than 45 percent of claims in Canada’s Medicare system, compared to less than 8 percent of claims for private insurance in the United States. Pacific Research Institute economist Ben Zycher writes that a “realistic assumption” about the size of the deadweight burden puts “the true cost of delivering Medicare benefits [at] about 52 percent of Medicare outlays, or between four and five times the net cost of private health insurance.”
Administrative costs can appear quite low if you only count some of them. Medicare hides its higher administrative costs from enrollees and taxpayers, and public-plan supporters rely on the hidden nature of those costs when they argue in favor of a new government program.
Cost Containment vs. Spending Containment
Advocates of a new government health care program also claim that government contains overall costs better than private insurance. Jacob Hacker writes, “public insurance has a better track record than private insurance when it comes to reining in costs while preserving access. By way of illustration, between 1997 and 2006, health spending per enrollee (for comparable benefits) grew at 4.6 percent a year under Medicare, compared with 7.3 percent a year under private health insurance.” In fact, looking at a broader period, from 1970 to 2006, shows that per-enrollee spending by private insurance grew just 1 percentage point faster per year than Medicare spending, rather than 2.7 percentage points. That still omits the 1966–1969 period, which saw rapid growth in Medicare spending.
More importantly, Hacker’s comparison commits the fallacy of conflating spending and costs. Even if government contains health care spending better than private insurance (which is not at all clear), it could still impose greater overall costs on enrollees and society than private insurance. For example, if a government program refused to pay for lifesaving medical procedures, it would incur considerable nonmonetary costs (i.e., needless suffering and death). Yet it would look better in Hacker’s comparison than a private health plan that saved lives by spending money on those services. Medicare’s inflexibility also imposes costs on enrollees. Medicare took 30 years longer than private insurance to incorporate prescription drug coverage into its basic benefits package. The taxes that finance Medicare impose costs on society in the range of 30 percent of Medicare spending. In contrast, there is no deadweight loss associated with the voluntary purchase of private health insurance.
Hacker nods in the direction of non-spending costs when he writes, “Medicare has maintained high levels of . . . patient access to care.” Yet there are many dimensions of quality other than access to care. It is in those areas that government programs impose their greatest hidden costs, on both publicly and privately insured patients.
The paper goes on to discuss how private insurance bests Medicare on quality, but this excerpt is long enough. For more on the comparison between private health insurance premiums and per-enrollee Medicare spending, see this blog post, where I conclude, "If [this comparison] were a farm animal, and social scientists farmers, they would have to take it behind the barn and put a bullet in its head."
In addition to committing the same errors and Hacker and others, Archer fails to note that Medicare Advantage reduces spending in traditional Medicare -- thereby treating us to the spectacle of an opponent of competition taking credit for one of competition's many benefits.