Blog Home


Health Spending Projection Spin Cycle: Rinse And Repeat, Or Reset?

August 11th, 2011

One of the annual rituals of Washington’s health policy calendar involves the release of projections for the next ten years of national health spending by actuaries at the Centers for Medicare and Medicaid Services (CMS). It then is followed immediately by desperate efforts by various interest groups and advocacy “analysts” to spin the new numbers to their advantage.

Last month’s latest set of predictive statistical entrails published in Health Affairs could be pointed in several directions, depending on which way the political wind was blowing.  Actual national health spending in 2010 was lower than previously predicted (growing only 3.9 percent after a previous historic low rate of 4.0 percent in 2009).  Just one year ago, CMS had predicted annual growth rates of 5.1 percent for 2010 and 5.8 percent for 2009 (close enough for government work?).  And the usual two-percent-or-more historical spread by which health spending grows faster than Gross Domestic Product (GDP) compressed surprisingly in 2010, when GDP growth was almost the same (3.8 percent), and the health spending share of the economy remained at 17.6 percent of GDP.

In some quarters, happy days are here again under the new Affordable Care Act (ACA).  According to Karen Davis of The Commonwealth Fund, the slowdown indicates that “the original federal budget costs of health reform were overestimated.” White House deputy chief of staff Nancy-Ann DeParle chimed in by pointing to CMS projections that national health spending would grow only 5.8 percent annually from 2010-2020 (just 0.1 percent higher than projected before passage of the ACA) — even while 30 million more Americans are expected to gain insurance coverage. And Drew Altman of the Kaiser Family Foundation, noting that the CMS actuaries estimated that the government share of national health spending would only grow from 45 percent in 2010 to a projected 49 percent in 2020, concluded, “there is, in fact, no imminent danger of a government takeover” of health care.  (In “Pulling It Together,” he did not comment on whether he considered it nevertheless “a good start…”)

However, a more sober assessment would observe:

  • A substantial factor behind the recent health spending slowdown was a deep recession and its slow-growth aftermath.
  • There may be persistent signs of a “new normal” in medical utilization trend, as health care consumers are delaying or passing up use of more discretionary health care services.
  • Lower projected health spending costs for those to be newly covered under the ACA simply reflect assumptions that focusing on insuring primarily younger and healthier individuals is much cheaper.
  • The overall cost effects of the new health law are more complex and conjectural, but its feasibility and sustainability remain highly doubtful.  The ACA aims to put the troubled Medicaid program on budgetary steroids, beginning in 2014, while the financially unstable Medicare program is milked like an aging cash cow — with unprecedented cuts in provider reimbursements and a rollback in private plan options — to help finance all the other insurance coverage expansions proposed in the new health law.
  • The overall government share of national health spending will continue to increase even under the newly revised CMS figures, which understate its full scope and scale.
  • The actual out-of-pocket share of national health spending continues to be low and declining, despite flawed efforts to paint a different picture. ACA provisions aim to accelerate its relative decline.
  • Even short-term health spending projections are subject to unexpected swings. Longer-term projections may suggest the general direction of future trends, but they remain highly uncertain and incomplete.

The Recession’s Dampening Effect On Health Spending

Looking behind the headlines and political talking points triggered by the spending projection article, published in Health Affairs on July 27, one finds that even though the health sector of the economy has traditionally grown faster than annual GDP, wages, and non-health areas of government spending, the deep recession of 2007-2009 and its lingering aftermath slowed the absolute dollar growth of health spending (and health care utilization), if not its relative share of a slower-growing economy.  As a similar team of CMS health spending actuaries explained in another Health Affairs article last January, “Although health care spending has grown at a slower rate every year since 2002, the deceleration, or slowdown in the rate of growth, was more pronounced in 2008 and 2009 because of the severe economic recession. In contrast to prior recessions, when there was usually a lag before health care spending growth slowed, the recession that lasted from December 2007 to June 2009 had a more immediate impact on the health care sector.”

CMS actuaries report that the continuing impact of losses in employment and health insurance coverage associated with the recession in 2010 (the number of people enrolled in private insurance plans fell by roughly 5 million) limited growth in private health spending to 2.6 percent last year. Growth in spending for physician services slowed even more, to a historic low of 2.4 percent, due to recession-related declines in office visits (along with a milder than expected flu season).

So the illusory good news of the last few years might be that the Obama administration can claim at least superficially that it already helped slow down somewhat the growth rate for health care spending, and related health insurance premiums – by slowing the overall economy.  Whether or not it continues to “succeed” in this approach, the bad news is that we neither can afford, nor should we expect, to rely on a prolonged recession or sluggish economy to keep doing so.

A New Normal For Health Spending?

Consider a different preliminary hypothesis: that a longer-term trend toward declining growth rates for health spending – beyond just the effects of the recent recession – is underway. Evidence that levels of patient visits, drug prescriptions, and other medical procedures are down over the last year or two (“Americans Cut Back on Visits to Doctor”) suggests that what cannot go on forever will not. Will assumptions about ever-rising years of health spending relative to the rest of the economy last longer than past assumptions about ever-rising housing prices?

The latest CMS projections do suggest an uneven path to some moderation in the medium-term growth rate for health spending from 2010 to 2020. For example, Medicare spending grew less rapidly in 2010, due to a lower rate of growth in Medicare Advantage private plan payments. Deeper reductions in those reimbursement levels, as well as formula-driven rollbacks in fees paid to other medical providers, are ahead under the ACA. However, CMS projects that faster growth in Medicare enrollment  (as millions more baby boomers  reach retirement age) will offset slower growth in per-person spending  due to those ACA payment policies, and raise Medicare’s average growth rate for spending to 6.3 percent from 2013 to 2020.

Overall national health spending is supposed to pick up modestly from 2011 to 2013, at a 4.9 percent annual average rate, due primarily to gains in employer-sponsored private insurance coverage and increases in personal income as the economy recovers.  The full effects of the new health law do not begin to appear until 2014, when Medicaid spending increases 20.3 percent over the previous year, and private insurance spending will increase 9.4 percent (4.4 percent above pre-ACA projections).  The combination of expanded Medicaid coverage – one-third higher than the year before — and more “private” coverage in highly regulated (and highly subsidized) exchange plans eventually will boost overall insurance coverage by about 30 million people (with 23 million newly insured in 2014 alone). This will more than offset cuts in Medicare payments in the ACA.  Hence, the overall increase in national health spending would be 8.3 percent in 2014, compared to 5.5 percent in 2013.

By 2020, Medicaid is projected to account for nearly 20 percent of all national health spending, a more-than-healthy increase from its 15-percent share in 2009.  This change reflects greater quantity, but an overstretched program will be challenged to avoid delivering lower quality. The primary economic assumption by CMS behind projecting lower figures for spending per insured person in those later years is that the newly insured will be younger and healthier, on average, than current Medicaid and private insurance populations.  “The kids are all right” or at least cheaper to cover than seniors enrolled in Medicare and Medicaid. Medicare’s overall share national health spending is projected to remain at 20 percent (unchanged from 2009), as its beneficiary enrollment growth is countered by reduced levels of provider reimbursement under the ACA.

The newly insured also are optimistically assumed to use physician services and prescription drugs to a greater extent (and more efficiently) than higher-cost hospital care, despite past evidence of surges in pent-up demand and overall medical price inflation for such populations. This key assumption, along with another assumption that reductions in Medicare reimbursement fees further below the break-even level for many providers can be sustainable (and/or not push private insurance premiums higher as an offset), is behind the purportedly lower per-capita health spending projections under the ACA. That’s the health policy equivalent of drawing an inside straight!

The latest national health spending projections also show that the government share of the health pie will continue to rise, from 45 percent in 2010 to 49 percent in 2020. Last year’s set of 10-year projections had estimated the government share of health spending already at 50 percent in 2011, rising slightly to 51 percent by 2019, but CMS actuaries subsequently revised their past methodology for categorizing government health spending. However, all of those calculations substantially understate the true level of government control over health care spending, by overlooking the additional role of its tax expenditures, increased regulation, and dominant payer status. In any case, the CMS actuaries conclude that when the main components of the health law are implemented in 2014, “health care financing is anticipated to further shift toward governments.”

The new health law also will continue the longstanding trend of reduced out-of-pocket (OOP) health spending.  Two related effects are at work here: increased levels of private insurance coverage and public health program coverage lower the relative level of OOP health spending in the overall health economy, and consumers also tend to restrain more their use of health care goods and services that are subject to greater cost sharing.  In 2010, such direct spending by consumers rose just 1.8 percent (and 0.4 percent in 2009).  In 2014, the rate of OOP health spending actually will decline 1.3 percent as ACA-based insurance coverage (subsidized by taxpayers) increases.

The current low OOP share of national health spending – 11.8 percent in 2010 – is projected to fall to 9.6 percent by 2020.  It would have reached 10.5 percent without enactment of the new health law.

Some recent analysis of cost-sharing trends confuses changes in absolute dollar levels versus relative percentage levels and miscategorizes increases in the employee-paid share of health insurance as true “out-of-pocket” spending that influences levels of health care use. For example, the 2010 version of the annual employer benefits survey by the Kaiser Family Foundation and the Health Research Educational Trust noted several insurance enrollment shifts, such as an increase in the share of employees enrolled in high-deductible health plans with a savings option (from 8 percent in 2009 to 13 percent in 2010) and increased enrollment in employer plans with deductibles of $1000 or more for single coverage (from 22 percent of employees in 2009 to 27 percent in 2010).  However, it failed first to disclose what the previous year’s average deductible level was compared to the 2010 average, and then compare that percentage change to the percentage increase in employer coverage premiums.  The survey’s primary analysts nevertheless acknowledged that for workers in plans with such deductibles, “there were no significant changes in the average deductibles for each plan type for single or family coverage from 2009 to 2010.”

The broader conclusions from the latest CMS spending projection analysis are that national health spending will grow at an average annual rate of 5.8 percent between 2010 and 2020, or about 1.1 percent per year higher than average annual GDP growth.  In the years following fuller implementation of the new health law (2015-2020), national health spending will grow faster, at 6.2 percent per year. Accordingly, the health sector’s share of GDP will rise from 17.6 percent to 19.8 percent.

Uncertainty And The Wisdom Of Duane Thomas

But those projections rely on some unnatural assumptions (required under current law, but not compatible with economic and political reality). For example, they assume that a massive “sustainable growth rate” (SGR) reduction in Medicare’s  fees for physicians – amounting to 29.4 percent effective in January 2012 – will be executed as officially required rather than be overridden again by Congress. Under an alternative scenario calculated by the CMS actuaries, in which Medicare pays doctors in line with increases in the Medical Expense Index, Medicare spending would grow 6.6 percent in 2012, as opposed to the 1.7 percent level used in their primary analysis described above.  Overall physician and clinical services spending would grow 4.5 percent under this alternative scenario in 2012, as opposed to 0.8 percent under “current law.”   (Due to other long-term assumptions, the effects of an alternative scenario SGR “fix” are greatest in 2012 – increasing growth in national health spending that year by about one percent more – and the health spending share of GDP in 2020 would only rise to 20.1 percent).

To be fair, the CMS actuaries have been transparent in discussing the limits of current-law assumptions and ten-year projections, acknowledging: “These projections remain subject to substantial uncertainty given the variable nature of future economic trends and a lack of historical experience for many Affordable Care Act health system reforms.” Moreover, the supply-side effects of the ACA (changes in provider behavior with an influx of newly insured individuals) “remain highly uncertain and are not estimated at this time.”

Oh, and we really don’t know how much, if any, of the new health law will be left standing several years from now or can even be implemented in practice!

So, the latest set of ten-year spending projections is interesting and informative, but it provides at best a temporary directional guide to what’s ahead, rather than much precision as to its magnitude.  Then again, there will be another set of such projections in another Health Affairs article NEXT YEAR.  Just like long-term congressional budget resolutions that are revised annually (if they are ever approved, in recent years…).

Former Dallas Cowboys star running back Duane Thomas summarized this phenomenon best when asked in 1971 how he felt about playing in Super Bowl VI: “If it’s the ultimate game why do they play one every year?

Email This Post Email This Post Print This Post Print This Post

Leave a Reply

Comment moderation is in use. Please do not submit your comment twice -- it will appear shortly.

Authors: Click here to submit a post.