Mobile Menu - OpenMobile Menu - Closed

Henry J. Aaron, Ph.D.

Henry J. Aaron, Ph.D., Brookings Institution, Bruce and Virginia MacLaury Senior Fellow

Mr. Chairman:

Thank you for the invitation to testify today on the proposal to convert funding for the Veterans Health Administration (VHA) from a discretionary to mandatory basis.  In the course of my remarks, I should like to stress four points:

First, the VHA faces an unusually difficult challenge—it must deliver an extraordinarily wide range of services to highly diverse populations. The VHA provides ordinary primary, secondary, and tertiary somatic medicine, as well as mental health services.  One of its most important responsibilities is to offer a subtle combination of physical therapy, mental health services, and somatic treatment to victims of spinal cord and traumatic brain injury.

Second, the VHA has performed remarkably well of late.  Inspired management has transformed the VHA from being the poster-child for low-quality medical care  into a model  organization that delivers higher quality health care than the average of private health care providers and does so at a comparatively reasonable price.

Third, the budget of the VHA is part of the long roster of federally financed health care services.  The cost of federal health care obligations is projected under current law to increase enormously.  In fact, growth of these programs accounts for more than all of the long-term deficits recently to which the Congressional Budget Office and various private analysts have recently drawn attention.  Put more positively, if the nation deals with the imbalance between projected revenues and spending for health care, revenues at current levels are projected to be sufficient to pay for all other anticipated government commitments, including all Social Security benefits promised under current law.

Fourth, proposals to boost federal health care spending abound.  Not all can be funded without unduly raising federal spending.  Different groups would benefit from each of these proposed increases. Sensible budgeting requires a comparison of these competing claims. Unfortunately, Congressional committee structure inhibits such comparisons.  To illustrate this problem, I list three such candidates for increased spending.  For what it is worth, my judgment is that the priority of converting VHA spending into mandatory funding ranks below the other two possible uses of federal funds.


The VHA administers more than 1,200 hospitals, outpatient clinics, nursing homes, and rehabilitation facilities.[1]  These facilities comprise one of the largest health care delivery networks in the United States, with revenues approximating those of the largest private domestic health care system, Kaiser Permanente.

The statutory clientele of the VHA, currently more than 23 million veterans, is enormously varied in its needs.  It includes veterans who have crippling service-connected spinal-cord and brain injuries that prevent them from earning a living or taking care of themselves.  It includes other veterans who, despite serious service connected disabilities, support themselves and their families.  It includes veterans with comparatively minor service connected disabilities that have no bearing on their current activities.  It includes veterans with no service related disability whatsoever who currently have low incomes.  And, finally, it includes, millions of veterans who have no service connected disability and have what is normally regarded as an adequate income.

America owes its thanks to all military veterans for their service to this nation.  All took time from civilian lives to help protect the rest of us.  But it in no way diminishes the contribution made by those veterans who came home healthy and uninjured and have prospered to say that the nation owes a special debt to those who suffer daily physical reminders of their service.  This sense of priority is reflected in the VA health care priority groups. 

It is also manifest in the use that each of these groups makes of the VA health system.  More than half of veterans in priority groups 1, 2, and 3 are enrolled in the VA health system, just under half of priority groups 4 and 5, about one-third of groups 6 and 7, and less than 20 percent of group 8.  The VHA is particularly good at treating those conditions that peculiarly affect veterans, and veterans turn disproportionately to the VHA for care of these conditions.  This pattern reflects a match of need and expertise.  Other veterans choose health care providers from the private sector.


The Veterans Health Administration has undergone a remarkable transformation since 1995.  At the time, critics charged the VHA with high cost, low quality, providing the wrong mix of services for its clientele, and poor accessibility.  The key reforms included reorganizing numerous separate providers into veterans integrated service networks (VISNs) that received budgets from which responsible officials had to manage variety of service providers.  Budget authority was shifted to where veterans were most numerous.  These reforms gave the VHA authority to bargain over the prices of pharmaceutical products that, linked to the VHA’s size, gives it more clout than virtually any other single purchaser.  The performance of the VISNs is measured and advertised around the VHA and VISN managers receive bonuses for good performance.[2]  These quite business-like incentives illustrate an important proposition: government can achieve the efficiencies normally associated with private businesses if its managers are given the flexibility and incentives to operate effectively.

Unfortunately, Congress has interfered with the VHA’s administrative freedoms in various ways and has made efficient administration more difficult than it needs to be.  Congress has prevented the VHA from contracting with one or a few suppliers of some products whose prices are lowest.  The late completion of work on budgets and the all-too-frequent use of continuing resolutions has hampered efficient hiring and other planning.

On a more positive note, the VHA has gone further and faster in introducing electronic medical records (EMR) than have most private health care providers.  EMR could be introduced expeditiously because VHA management had centralized control, something that is lacking in nearly all of the private U.S. health care system.  And it proceeded as fast as it did also because the VHA also had adequate financial backing—an estimated $300 million for wiring, $450 million for computers, and $485 million a year (an average of $90 per patient) in upkeep.[3]  The VHA experience illustrates why all the talk about electronic health records for the private sector has produced so few results.  In contrast to the money that the VHA had to back its EMR ‘play,’ the legislation that created Office of the National Coordinator for Health Information Technology stipulated that no additional funds would be appropriated to support its activities.  One would be hard pressed to find a better example of ‘you get what you pay for.’

Objective measures indicate that the quality of care provided by the VHA at least equals that of private sector health services.  One study that found that two-thirds of VA patients but only 51 percent of privately served patients receive all indicated care when they see a doctor or visit a hospital.[4]  Another study reported that the VHA provided better care  in twelve of thirteen categories than private providers rendered to Medicare patients.[5]

These comparisons clearly indicate that the VHA has come a long way since the days when the quality of its care was almost universally criticized.  They are also consistent with a  view that, at least in the case of health care, a well-managed public agency, authorized by Congress to operate in a business-like manner, can deliver care as good as or better than that rendered by the private sector as currently organized.  No doubt improvements in efficiency similar to those of the VHA could have been achieved in the private sector if current administrative arrangements were altered.  Alas, the currently fragmented organization of private providers and payers alike deprives most of them of the capacity to execute the reforms that centralized management made possible in the VHA.


Governments—federal, state, and local—now directly account for 47 percent of national health care spending and an even larger share—56 percent—of hospital spending.  The full role of governments is even larger than those numbers suggest, because premiums paid by  employers for their employees are partially offset by the revenues forgone as a result of the exclusion of this portion of consumption from all tax, corporate or individual.

Although it is already large, the public share in the cost of health care is certain to increase.  Growth of health care spending has outpaced the growth of income by an average of 2.7 percentage points a year for more than four decades.  A gap of similar size is likely to persist.  The rate at which the menu of beneficial medical interventions increases is not expected to slow as the genomic revolution, nanotechnology, and personalized medicine proceed.  Furthermore, the population is aging.  The financial burden of supporting health care for the elderly disproportionately falls on the public.[6]  The proportion of the population covered by public programs will increase.  Furthermore, the value of the exclusion from tax of privately financed health insurance premiums will continue to grow faster than income does.

What is insufficiently understood, in my opinion, is that meeting this challenge will require a transformation of both publicly and privately financed health care.  Measures to slow the growth of both public and private health care spending and to boost taxes will be necessary.  This nation has come to a national consensus that Americans—old and young, with and without disabilities, rich and poor—should enjoy similar—not identical, but similar—access to health care.  Hospitals and physicians treat all patients similarly; indeed, if they do not, they are—and should be—open to successful suit for malpractice.  That the polity would long tolerate cuts in either Medicare or Medicaid sufficient to significantly lower the rate of growth of spending is, in my view, an insult to the generosity and compassion of the American people.

What is also inadequately understood is that successfully balancing public spending for and revenues dedicated to health care would eliminate any long-term budget problem, based on the best current projections.  The following figure shows the long-term budget projections of the Congressional Budget Office, adjusted for health care spending statistics available early this year.

Bar Chart showing the Projected Budget Deficit or Surplus (Excluding interest)

The dotted bars show projected primary budget deficits, as a percentage of gross domestic product.  The primary budget deficit is the difference between total spending other than interest on the debt and total revenues.  If the primary budget deficit were to balloon, as shown here, the actual situation would be far worse than indicated.  As the ratio of debt to GDP increased, interest payments would grow for two reasons, which interact multiplicatively.  First, there would be more debt on which to pay interest.  Second, the interest rate at which the government can borrow would rise as lenders become apprehensive that the government will be able to meet future debt service obligations. The result would be explosive increases in interest payments.

The figure also shows projections of the primary budget if one subtracts projected government spending on Medicare and Medicaid, earmarked taxes, and a share of general revenues equal to the current support level.  This difference is shown in solid black bars.  As is apparent, if the impact of rising spending on Medicare and Medicaid in excess of revenues is eliminated, there is no projected deficit.  Current projections indicate that a small surplus would emerge.  In other words, if the nation deals with its health care financing problem, the remaining revenues under current law would be adequate to cover all projected private spending, including all Social Security benefits promised under current law.

Let me hasten to emphasize that long-term projections are never exactly right.  They simply extrapolate the implications of current assumptions.  Small differences in those assumptions compounded over enough years can produce large differences, but mean little. Projections are useful when they show large imbalances and broad trends.  This figure clearly indicates two realities that should shape current fiscal debate. 

  • The nation faces large long-term deficits under current policy
  • The nation does not face an overall fiscal crisis or an entitlement crisis: it faces a big health care financing problem that should lead to a vigorous national debate on much health care we want and how to pay for it.


Against this background, should funding for the VHA be converted from a discretionary to mandatory account?  The answer, I believe, is that it should not, despite the genuine claim that veterans have on public support for their health care and the excellent record in delivery high quality health care at a reasonable price that the VHA has established in recent years. 

  • This switch would create incentives for undue expansion of the VHA.
  • This expansion would very likely not be consistent with the longer term objectives of reforming the overall health care system. 
  • Finally, the conversion would likely boost federal spending at a time when other increases in federal health care spending would yield greater benefits.

At present, the VHA annually receives a fixed appropriation set by Congress based on the president’s budget.  If VHA funding were mandatory, it would presumably be based on actual enrollment multiplied by a sum set to approximate the per person cost of providing care to enrollees.  This is the system proposed in H.R. 2514.  Baseline spending is to be set at a percentage of past spending—130 percent in H.R. 2514.  Future funding would be based on baseline, per capita spending multiplied by enrollment in the VHA system in the preceding year multiplied by an index—the CPI in H.R. 2514.[7]

Such a system would likely create powerful incentives for the VHA to enroll as many veterans as possible, whether or not new enrollees use VHA services as much as current enrollees or even whether they use them at all.  In fact, incentives would be strongest to enroll those expected not to use VHA services, as the resulting addition to budget would encourage VHA administrators to enhance services to entice others to join.  As funding increases, the VHA would be able to enrich services, encouraging both current and new enrollees to increase the proportion of health care they seek from the VHA.  The VHA now enrolls only about one-third and annually serves just over one-fifth of all veterans.  Furthermore, nearly 80 percent have non-VHA health insurance coverage from public or private sources. 

These facts mean that the potential for increasing VHA service levels is vast.  The Congressional Budget Office, using similar reasoning, has estimated that converting the VHA to mandatory funding on the lines of the 2005 proposal would roughly double total spending.  Some drop in spending under other government programs would occur, but, according to the Congressional  Budget Office estimates of H.R. 515 submitted in 2005, which resembles H.R. 2514 submitted this year, the offsets would be modest.[8]

To be sure, converting the VHA to mandatory funding would not entirely insulate it from budgetary pressures.  Congress could cut the  per person funding amount or exclude certain groups of veterans from the formula used for computing annual funding.  The funding formula contained in H.R. 2514 could be modified to hold down spending.  But I think such modifications are unlikely to gain much traction.

Is an increase in VHA funding the best way to increase health care spending?  Is it likely to move health care delivery in a direction that the nation is likely to follow?  If the answer to these questions is ‘yes,’ then this budgetary commitment is justified.  Each of us will have views on the answers to these questions.  Mine is that the answers are ‘no.’

The VHA is the nearest approximation in the United States to the British National Health Service, a publicly funded entity that directly employs most health care providers.  That form of organization differs from the U.S. norm—third party payment to private hospitals, physicians, and other providers.  It is unlikely to be widely adopted in the United States.  Little support exists anywhere on the political spectrum for turning health care providers into public employees.  All strategies for extending coverage—tax incentives, state initiatives, single-payer, employer mandate, individual mandate—call for payments to private hospitals, physicians, and other providers.  To encourage an increased fraction of the U.S. population to receive an increased proportion of its care from a system based on publicly employed and managed providers would be a step away from any future national system.

Furthermore, Congress is duty bound to weigh the relative merits of various proposals to boost public spending on health care.  Such comparisons are difficult given the prevailing committee structure of the U.S. Congress, but it is right to make them.

  • At present, Congress has just sent to the president a proposal to extend the State Child Health Insurance Program, at an annual cost of approximately $7 billion over the next five years.
  • Congress will likely prevent the full scheduled cut in physician reimbursement under Medicare from taking effect.  CBO has estimated that the cost of raising physician reimbursements 1 percentage point instead of cutting them, as required under current law, would boost spending by an annual average of $4.8-$6 billion over the next five years.
  • I have not seen a cost estimate for H.R. 2514.  Adjusting the estimate for H.R. 515 for the passage of time and the change of index for per person costs, leads to an estimate of increased annual outlay of $45-50 billion over the next five years.

The current budgetary climate will not readily accommodate spending increases that boost the budget deficit.  Indeed, part of the controversy surrounding SCHIP is its cost, despite the fact that it would be offset by increased tobacco taxes.  Furthermore, making VHA funding mandatory would not be offset by reduced spending elsewhere or by increased revenues.  It is not yet clear whether any added spending to avoid reductions in physician reimbursements under Medicare will be offset.  But what is clear is that the deficit increasing effect of H.R. 2514 is vastly larger than that of either of the other two bills. 

On substantive grounds, and contrary to allegations of some of its critics, the SCHIP bill builds on and reinforces the private provision of health care.  SCHIP has enjoyed bi-partisan support since its enactment in 1997.  Avoiding the full reduction in physician reimbursements under Medicare is necessary in order to discourage significant and possibly catastrophic defections by physicians from being participating providers or even participating at all in the Medicare program at all.  Both uses of public funds reinforce established ways of providing health care to dependent populations, building on a public private partnership.  Both these measures should enjoy far higher priority than does H.R. 2514.

**The views expressed here are my own and do not necessarily represent those of the trustees, officers, or other staff of the Brookings Institution.

[1] Sidath Piranga Panangala, Veterans’ Medical Care: FY 2008 Appropriations,” CRS Report for Congress, 25 June 2007, p.3.

[2] Adam Oliver, “The Veterans Health Administration: An American Success Story?” The Millbank Quarterly, vol. 85, no. 1, 2007, pp. 5-35.

[3] Oliver, p. 20.

[4] Stephen Asch et al., “Comparison of Quality of Care for Patients in the Veterans Health Administration and Patients in a National Sample,” Annals of Internal Medicine, Vol. 141, no. 12 (21 December 2004), pp. 938-945.

[5] A. K. Jha, et al., “Effect of the Transformation of the Veterans Affairs Health Care System on the Quality of Care,” New England Journal of Medicine, vol. 348, no. 22, pp. 2218-2227.  Patients covered by Medicare received better care than those covered by commercial insurance in eight of twelve categories where comparisons were possible.  Patients covered by Medicaid consistently received poor services.

[6] The statement in the text is a factual, not normative.  Even if the nation embarked on a policy to require the elderly, or most of them, to save enough to pay for the health care they will use in retirement, the transition to such a system would take decades.  Whether such a policy shift would be desirable is irrelevant to the text statement.

[7] The choice of index has an important bearing on how rapidly funding grows in the long term.  H.R. 515, introduced in 2005, proposed to use a health care index, which would have resulted in more rapid growth in spending than would occur under H.R. 2514.

[8] Based on H.R. 515 introduced in 2005, spending on Medicare, Medicaid, and the Federal Employees Health Benefits Program would fall initially by about 5 percent of the increase in VHA spending, growing to about 7 percent after ten years.  Congressional Budget Office, Cost Estimate, H.R. 515, ‘Assured Funding for Veterans Health Care Act of 2005, July 25, 2005.