Veterans Health Administration: A Preview of Single-Payer Health Care

Executive Summary

In recent years, many health-reform advocates have cited the Veterans Health Administration (VHA) as an example of how a single-payer, government-run health care system would work. They claim that the VHA's integrated system provides excellent access to high-quality care that a private-sector system cannot match, at a lower average per-patient cost.

Not only have recent revelation of off-the-books waiting lists shown that the “excellent access to care” claim is the result of fraudulent data reporting; claims that the VHA has lower costs are also incorrect. While the VHA does spend less per enrollee than the private sector or Medicare, because many veterans have other sources of health coverage, the VHA provides only about one-third of the average enrollee's health care. Adjusting for this factor, we find that the VHA actually spends more than Medicare, and as much as 80 percent more per patient than the private sector.

The VHA advocates are right about one thing, however. The VHA is a model of what a government-run, single-payer system would be like. It would provide worse access to health care and dishonest performance figures, while costing more than other health care systems.

The VHA and the Single Payer Health System

The VHA operates an integrated health system providing care for eligible beneficiaries. The physicians, nurses, and other health professionals – as well as administrators – are all employees of the federal government, which also owns the hospitals and other facilities. It is essentially a single-payer, government-run health care system similar to Britain's National Health Service, although the VHA does contract out a small percentage of its care (for example, primary care providers in areas without a nearby VHA facility).

In 2005, Phillip Longman wrote an article on the VHA entitled, “The Best Care Anywhere,”[1] which prompted Paul Krugman to describe the VHA as “a health care system that … provides a helpful corrective to anti-government ideology. For the government doesn't just pay the bills in this system — it runs the hospitals and clinics.”[2] Longman, Krugman, and others claimed that the VHA delivered better health care, at a lower cost, than both private sector health insurance and government-funded programs providing access to privately-provided health care, such as Medicare. Longman followed up with a book-length treatment two years later, subtitled, “Why VA Health Care Is Better Than Yours”[3] and another article (with the provocative title, “Best Care Everywhere”[4]) proposing a VA-like system for all Americans as the “obvious” health care reform solution.

However, when we look at the whole picture, we find that the VHA’s costs are not demonstrably lower than those of the private sector; they appear lower only because many veterans have other sources of care, and the VHA provides, on average, only 36 percent of the health care received by VHA-enrolled beneficiaries. When that is taken into account, the VHA is, in fact, more expensive than both the private sector and government programs such as Medicare.

The recent revelations of off-the-books waiting lists[5], with veterans waiting 4 to 6 months or longer for appointments when the official books show waits of less than one month[6] also give an insight into how single-payer government systems works. The VA health system is a model for what happens when a health system is run by bureaucrats, using bureaucratic “performance” metrics tied to something other than actual results. Similar phenomena have been observed in other single-payer, government-run systems in other countries as well.

The VHA boosters were right about one thing: The VHA is a model for a single-payer, government-run health care system. It is a model for how easy it is to manipulate performance measures for years without being detected. The system costs more than other health care options but data are selectively reported to make it appear that it costs less; bureaucratic manipulation of performance data make the system appear to be working well; and administrators collect their bonuses – while veterans die awaiting care.

There is a tragic human cost to this bureaucratic manipulation, but that cost does not show up in official figures. The deaths – or extended illnesses, disabilities, and unnecessary discomforts – of veterans on the unofficial VHA waiting lists didn't show up until whistle-blowers started to leak the truth. It is these human costs which are the true price of a single-payer, government-run health care system.

The Hype

Nowadays, the VHA is widely admired for an innovative electronic medical records system known as VistA (“Veterans Health Information Systems and Technology Architecture”). VistA is viewed by many single-payer advocates as an example of how a single-payer, fully-integrated, government-run health care system can reduce costs. It is claimed that the VHA is more cost-effective because it is a fully integrated health care system, which can choose the optimal mix of care from different sources – physicians, hospitals, clinical pharmacists, drugs, and devices – in a way that the highly fragmented private system cannot.

In theory, the system can provide reminders when patients are due for screening tests, flu shots, follow-up visits, or prescription renewals. Preventive care can become something that is proactively practiced by the health care system, rather than merely a covered service dispensed at the patient’s initiative. In addition, VHA claims that one of the benefits of the system is that if they get a good deal on an alternative to a particular drug, they could switch 95 percent of their patients taking that drug within 90 days.[7]

The Truth

Now, we know that the VA health system demonstrated how easy it is for a single-payer, government-run system to manipulate performance measures for years without being detected, Most famously, they did this by “shortening” time spent on waiting lists by creating an off-the-books waiting list to get on the official waiting lists – which patients would be placed on only when an appointment was available within the target time frame.

In other words, the VA health system is a model for what happens when a health system is run by bureaucrats using bureaucratic “performance” metrics tied to something other than actual results. In this sense, it is not the least bit unique. A few years ago, the hospitals in the British National Health Service (NHS) were found to be meeting a requirement to treat 95 percent of emergency patients within 4 hours of arrival – by keeping patients waiting in ambulances for over an hour (sometimes much longer), until the wait times in the hospital itself dropped below the threshold. This made the hospital appear to be meeting the target, since the clock did not start on the “4 hours” until the patient was actually inside the hospital building. They were, in effect, using the ambulance parking area as an unofficial, off-the-books “waiting list.”

There’s a tragic human cost to this behavior that extends beyond the one patient waiting in the ambulance for hours. Keeping patients waiting in ambulances prevents those ambulances from rescuing other patients.

Likewise, the deaths, extended illnesses, disabilities, and discomforts of veterans on the unofficial VHA waiting lists did not show up until the truth was leaked by whistleblowers.

The VHA has long been praised for its state-of-the-art electronic medical records (EMR) system, its (apparent) focus on preventive medicine and coordinated care, and its (apparent) ability to take advantage of its near-lifetime relationship with patients to focus on keeping patients healthy rather than delivering treatments when they are sick. Now, it turns out that the state-of-the-art EMR system is also very useful for “documenting” performance achievements that are not actually occurring.

This will necessarily be the case with any single-payer, government-run system. When dissatisfied patients cannot take their business – and their money – elsewhere, there is no check on internal fraud. While individual fraudsters might occasionally be caught by whistleblowers or investigators, the system will not change and the incentive for fraud will not disappear, because the organization's budget is not dependent on satisfying patients.

The Cost to Veterans

Did the VHA at least provide care at a lower cost than other health care systems, like private insurance and Medicare? The claim is, in part, that the VHA is more cost-effective because it is a fully integrated health care system, which can choose the optimal mix of care from different sources – physicians, hospitals, clinical pharmacists, drugs, and devices – in a way that the highly fragmented private system cannot. Also, because VHA has a near-lifetime relationship with its patients, it has incentives to make appropriate investments in preventive care and that the VHA can achieve economies of scale; for example, by negotiating lower prices for prescription drugs.

When we look at the whole picture, we find that the VHA’s costs are not demonstrably lower than those of the private sector, and in fact appear to be higher. The data required to demonstrate this have not been released to the public since 2004 (though some numbers from 2006 were released by the Congressional Budget Office).

In Fiscal Year 2006, the VHA had 7.9 million enrollees, and a total budget authority of $36 billion.[8] This works out to average per-capita expenditures of $4,557. For the entire population of the United States, the Center for Medicare and Medicaid Services (CMS) estimates the total 2006 U.S. health care expenditures to be $2.1 trillion, for a population of 299.7 million.[9] This implies per-capita expenditures of $7,025 – a considerably higher figure. This is even more stunning when one considers that the VHA enrollee population is older, more disabled, and otherwise more likely to consume health care than the average American.

However, this neglects a crucial fact: most VHA enrollees have other sources of care. According to VHA’s own figures,[10] the average VHA enrollee receives only 36 percent of his or her health care from the VHA. In VHA parlance, this figure is referred to as “reliance” and is defined as the percentage of an individual VHA enrollee’s health care services that the enrollee received from the VHA, instead of from other sources.[11]

When we adjust for 36 percent reliance, we find that if the VHA provided 100 percent of its enrollees’ care (at the same average cost[12]), VHA would have to spend $12,658. Now, instead of appearing to spend 35 percent less per patient than the national average, VHA spends 80 percent more per patient.

Of course, this does not necessarily imply that the VHA is cost-inefficient. There are many reasons why the cost per enrollee might be higher for VHA patients than for the general public that are completely unrelated to the VHA’s efficiency. For example, compared to the general public, enrollees are on average older (median age 63), and are therefore more likely to have chronic and age-related diseases. A substantial percentage have service-connected disabilities (including, but not limited to, those from war wounds). This is only partially mitigated by the fact that, having served in the military, they are less likely to have chronic conditions that would have appeared at an early age and disqualified them from service.

Due to the age profile of VHA enrollees, it might be instructive to compare the adjusted VHA expenditures to the comparable figure for Medicare beneficiaries, 84 percent of whom are over age 65 and the remainder disabled.[13] In 2006, Medicare spent[14] $401.3 billion to care for a total of 43.1 million beneficiaries for an average government expenditure of $9,311; in addition, Medicare beneficiaries spent an average of $3,103 out-of-pocket for a total per-beneficiary expenditure of $12,414 – slightly less than the “more efficient” VA, and with an older (and possibly sicker) population.

These comparisons are summarized in Table 1.

Table 1:  Table 1: 

Comparison of Per-Capita Expenditures for

VHA Enrollees, U.S. Population, and Medicare Beneficiaries


National Health Expenditures

Veterans Health Administration

VHA Reliance

VHA Adjusted to 100% Reliance

Medicare Expenditures (Incl. OOP)

Total Expenditures ($Billion)






Population (millions)






Per-capita expenditures (dollars)








The VHA not only delivers worse access to care than its official figures indicate; it also spends more money providing that substandard care.

Unfortunately, this should not be surprising to anyone. When VHA officials are given incentives to make performance metrics come out a certain way, they will find a way to make the performance metrics come out that way, even if it does not reflect reality or requires outright fraud. Because the VHA is a government entity, it does not face competition so it cannot lose money by dissatisfied patients deciding to take their business elsewhere. As a result, the metrics become a substitute for reality. A private integrated health system facing a competitive environment could not long survive even if its managers resorted to fraudulent performance metrics. Unsatisfied patients would take their business elsewhere, and the reality would trump the fraudulent figures.

So while Longman, Krugman, and other champions of the VHA were advocating it as the model for government-run health care, they were right about one thing: the VHA is, indeed, a preview of what single-payer system would be like. It would provide poor access to health care and likely hide performance failures, all while costing more than other alternative health care systems.

[1]                      Phillip Longman, “The Best Care Anywhere,” Washington Monthly, January/February 2005, at

[2]                      Paul Krugman, “Health Care Confidential.” The New York Times, January 27, 2006, at

[3]                      Phillip Longman, Best Care Anywhere: Why VA Health Care Is Better Than Yours, PoliPointPress, 2007 (subsequently issues in second and third editions).

[4]                      Phillip Longman, “Best Care Everywhere,” Washington Monthly, October 2007, at

[5]              For example, see Scott Bronstein and Drew Griffin, “A fatal wait: Veterans languish and die on a VA hospital's secret list,” at

[6]              VA Office of Inspector General, “Veterans Health Administration Interim Report: Review of Patient Wait Times, Scheduling Practices, and Alleged Patient Deaths at the Phoenix Health Care System,” May 28, 2014, Report # 14-02603-178, at

[7]              Allen Dobson, Joan DaVanzo, Lane Koenig, and Robert Book, “Medication Therapy Management Services: A Critical Review,” The Lewin Group, February 27, 2005.

[8]                      Allison Percy, “The Health Care System for Veterans:  An Interim Report,” CBO Pub No. 3016, Dec 2007.

[9]              Aaron Catlin, Cathy Cowan, Micah Hartman, Stephen Heffler, “National Health Spending In 2006: A Year Of Change For Prescription Drugs” Health Affairs 27:1, Jan/Feb 2008,  pp. 14-29.

[10]           Department of Veterans Affairs, “FY04 VA Enrollee Health Care Projection Model: Final Model Run Reports,” January 2004. The CBO used a more recent, unpublished, figure of 36 percent in its 2007 report. When I request more recent data from VHA, I was informed that post-2004 reports are deemed “sensitive” to the Department and are only shared with “good reason.”

[11]                    The reliance calculation by VHA depends on information about what health care services were delivered by other providers. This might seem to be difficult for VHA to obtain. For VHA enrollees who are also enrolled in Medicare, it is possible to match VHA data to Medicare data on a patient-by-patient basis, and calculate reliance figures which are probably quite accurate. For enrollees whose alternative is private-sector health insurance (for example, from an employer), VHA conducted a Survey of VA Enrollees (SOE) in 2002.

[12]                    See the “Final FY04 VA Enrollee Health Care Projection Model,” Section III-3. This document reports reliance figures on an unweighted, per-patient basis. That is, reliance is calculated for each individual enrollee, and then averaged across enrollees without taking into account the fact that some enrollees use more health care serviced than others. As a result, the above calculation can only be an approximation of the true reliance adjustment. If patients who use more or less services than average have systematically different levels of reliance, then the above calculation will bias statistically biased. Unfortunately the published figures by the VHA do not permit a statistically unbiased calculation.

[13]                    This includes a small percentage of patients who qualify due to end-stage renal disease (ESRD), regardless of age or disability status.

[14]                    Aaron Catlin, Cathy Cowan, Micah Hartman, Stephen Heffler the National Health Expenditure Accounts Team, “National Health Spending In 2006: A Year Of Change For Prescription Drugs,” Health Affairs, 27:14-29 (January/February 2008).