New Health Dialogue - logo

COSTS: The Price of Pessimism -- What the CMS Actuaries Missed

Last week, Medicare's chief actuary (formally known as the Office of the Actuary, or OACT) released an analysis of the financial impact of the health reform legislation recently approved by the House of Representatives (H.R. 3962).  Here are a few thoughts:

Get familiar with the source.  Remember their history.  The Medicare and Medicaid actuarial team's job is to track and understand Medicare spending patterns.  By design, it is also their job -- and their historical pattern -- to be skeptical about proposals for change. Just for context, the office's estimate of the cost of the Medicare Modernization Act in 2003 (the bill that created the Medicare prescription drug program, Part D) was $100 billion, or 25 percent, more than the Congressional Budget Office's (and CBO is also conservative by nature and design).  Last year CMS's Chief Actuary testified to Congress that the 10-year cost of the Medicare drug benefit is 37 percent lower than originally projected in 2003, and 17 percent lower than the previous year's updated projections.   Don't get me wrong. We need conservative estimators to prevent Pollyanna policy from being enacted into law.  But we should take that conservatism for what it is: a useful check on the naturally optimistic expectations of reformers.

Incentives can change behavior.  The actuary's office loathes predicting behavioral changes. It therefore underestimates the private sector's ability to adapt to new incentives.  This is why the only real savings they score are direct and unambiguous price changes, like the House bill's reductions to yearly market basket updates.  They discount and ignore the impact of the Center for Medicare and Medicaid Innovation, which is charged with implementing every reasonable payment reform pilot imaginable, including: accountable care organizations, medical homes, and bundled payments that give clinicians across organizations incentives to coordinate and improve patient care.

Further, the CMS actuaries express worry that the profitability of hospitals, skilled nursing facilities, and home health agencies would be so grievously harmed by the proposed payment cuts that they  would cease to accept Medicare's business.  This reflects a pessimistic belief that providers could not become more efficient.  A recent McKinsey memo, described by Jonathan Cohn in The New Republic (subscription required), concludes the opposite.

"McKinsey seems convinced that this entire package of reforms will influence behavior," Cohn says.  "McKinsey's analysis suggests that -- as long as they adjust to the new incentives -- doctors, hospitals, and insurers will be just fine."

The McKinsey memo also suggests that both CBO and the CMS actuary underestimate the ability and self-interested drive of providers to respond to incentive changes. That leads to these overly pessimistic estimates about the effectiveness (or ineffectiveness) of reform legislation.  Now I ask you, dear reader:  Who knows the health industry and their clients best -- McKinsey or the well-intentioned but relatively cloistered actuaries for Medicare?

Finally, when compared to the CBO, the CMS actuaries predict 5 million fewer people will get coverage and twice as much revenue will be raised from the individual mandate penalty under the House bill. This is another example of the CMS skepticism about behavioral effects.  I think it is fair to say that CBO analysts spend a lot more time than do Medicare's actuaries thinking about people under 65 and their potential responses to changing insurance prices.  I have to give more weight to CBO's estimate here.

Reducing overpayments will not deprive Medicare beneficiaries. The CMS actuary's memo states that the Medicare Advantage (MA) overpayment reductions in the House bill would mean fewer benefits in most MA plans.  It neglects to mention, however, that it would not reduce benefits below the statutory benefit package guaranteed to our nation's seniors.  In addition, the CMS actuary memo fails to mention that current Medicare Advantage payment levels overcompensate plans by quite a bit (14 to 18 percent, according to most independent analysts).  Only some of this extra compensation translates into additional benefits for some of the one-fourth of Medicare beneficiaries enrolled in Medicare Advantage plans.  In short, this analysis is overblown, especially since it does not mention that the competitive bidding approach to changing Medicare Advantage plan payment in the Senate Finance bill would preserve the incentive to provide extra benefits made possible by the efficiencies of some plans.

Small growth in National Health Expenditures is a good deal.  The CMS actuaries estimate that after all is said and done, national health spending will claim 0.3 percent more of GDP in 2019 than under current baseline trajectories.  They come to this conclusion despite being pessimistic about behavioral responses to delivery system reforms and despite assuming we will be covering 34 million more Americans by that time.  In other words, even analysts who are pessimistic about savings assert that we can just about pay for covering most of the uninsured out of savings from elsewhere in the health system.  Even a little more success than they project (far less than the percentage they were wrong about the Medicare Drug Benefit) and we will be covering the uninsured at lower cost than we would have spent without reform.   

In sum, there are some fine analysts at OACT, and Rick Foster and his team serve our nation well as actuaries of the Medicare program.  When it comes to knowledge of the delivery system and the under-65 population, however, they are perhaps less up to speed than McKinsey or CBO.  People should not get hung up on their judgments about the potential for delivery system change.  People should notice, however, that they estimate the savings provisions in the House legislation will generate $20 billion more over 10 years than CBO estimates.  This is their area of relative expertise.  Predicting behavioral responses, of hospital CEOs or of uninsured individuals, is not.