Why Do Estimates of Healthcare Costs Vary So Much?

Everyone involved in the health care debate is waiting expectantly to see the Congressional Budget Office’s analysis of the newly-revised Baucus health bill. That estimate may arrive on Wednesday, which could allow a vote in the Senate Finance Committee as early as Thursday (but quite possibly later).

In preparation for that release, I have one simple request: Could everyone please take more care in characterizing the cost of the revised bill? And the cost of the other health bills?

To date, there has been much unnecessary confusion about the cost of the health bills now being discussed in Congress. Using the same budget estimates, observers often report very different figures for the same bill. When Senator Baucus unveiled the initial draft of his bill, for example, he described it as costing $856 billion over ten years, but many observers looked at the official cost estimate and concluded that the bill would actually cost $774 billion. The House Tri-Committee bill has generated an even larger range of claims. Some observers have characterized the bill as costing about $1 trillion over ten years, while others have pegged the cost at almost $1.3 trillion or more than $1.5 trillion.

Why do these figures vary so much? Because the health bills are trying to do many things:

  1. Increase coverage through higher Medicaid spending and “carrots” such as subsidies for purchasing insurance through an exchange or tax incentives to get coverage.
  2. Increase coverage through “sticks” such as penalties for individuals who don’t have coverage.
  3. Prevent Medicare payment rates for doctors from being cut by more than 20% at the end of the year, as would happen under existing law.
  4. Increase spending on other federal health programs (e.g., prescription drugs in Medicare)
  5. Increase revenues by raising taxes in ways related to health insurance coverage (e.g., on insurers).
  6. Increase revenues by raising other taxes (e.g., new taxes on health providers, taxes on high earners, or reduced income tax deductions).
  7. Reduce spending on federal health programs to pay for the other expansions (e.g., reduce provider payment rates and roll back Medicare Advantage).

When observers characterize the overall cost of the health bills, they must choose (whether they know it or not) which of these items to include.

That decision is easy for the last two items on the list since they are offsets that are otherwise unrelated to health insurance coverage. The confusion thus arises with the first five items on the list.

If observers want to characterize the total cost of the bills, they should include not only the cost of expanding coverage (category 1), but also any provisions that would increase other spending. In other words, the total cost of a bill includes items that address the Medicare doctor “fix” (category 3) or raise other federal health spending (category 4). Those are often overlooked, however, because the official cost estimates group those provisions together with items from category 7, the provisions that reduce health spending. To get to the right figures, observers need to dig into the details of the cost estimates.

If observers want to characterize the total cost of expanding coverage, they should focus solely on category 1, the new spending and subsidies that would expand the number of people who have health insurance.

If observers want to characterize the net cost of expanding coverage, they should combine categories 1, 2, and 5. The “stick” measures in category 2 should be included because they encourage individuals to purchase health insurance or encourage employers to provide insurance to their workers. However, they do so in a way that reduces the net cost to the government. The tax increases in category 5 should be included because they reduce coverage.

Participants in the health care debate should take much greater care in distinguishing those three measures of cost. In principle, cost estimates from the Congressional Budget Office include enough information to calculate each of them. In practice, however, exact figures may sometimes be difficult to calculate. To determine increases in other health spending, for example, one can identify each provision that increases spending and add those together to get a gross amount of new spending. But it may not be possible to determine how interactions among policies (which CBO often scores separately) should be betted against those provisions. Thus, there will be uncertainty about estimates of the total bill cost (as Senator Baucus discovered).

The following table illustrates these calculations for the Baucus bill and the House Tri-Committee bill:

As you can see, Chairman Baucus chose to emphasize the total cost of his proposal—$856 billion over ten years—when he unveiled it. That’s a laudable choice. The $774 billion figure, in contrast, refers only to the cost of expanding coverage.

Most commentary about the House health bill has focused on the net cost of expanding coverage, slightly more than $1 trillion. That’s fine if commentators are careful to distinguish that concept from the overall cost of the bill. That distinction is often lost, however, and the public may therefore have missed that the House bill, in its entirety, would cost more than $1.5 trillion over the next ten years for coverage expansions and the doctor fix. And the price tag rises to close to $1.6 trillion when other spending provisions are included as well.

As the health debate moves to its next stage, it is essential that observers and participants start to use apples-to-apples figures when comparing the bills.

Source of the numbers: Congressional Budget Office (here and here) and author’s calculations. Changes to employer insurance have indirect effects on tax revenues. If the indirect effect causes revenues to rise, those revenues are included in category 5; if those tax increase cause revenues to decline, they are included in category 1. In some cases, it is difficult to determine whether particular budget impacts are due to coverage expansions or other policies. The estimate of gross other spending for the Baucus bill is based on his original estimate of the total cost; the estimate for the House bill is based on individual health provisions that increase spending. Because of interactions and other factors, those figures are the most difficult for an outside analyst to calculate (in short, category 4 is an estimate based on the CBO reports).

About Donald Marron 294 Articles

Donald Marron is an economist in the Washington, DC area. He currently speaks, writes, and consults about economic, budget, and financial issues.

From 2002 to early 2009, he served in various senior positions in the White House and Congress including: * Member of the President’s Council of Economic Advisers (CEA) * Acting Director of the Congressional Budget Office (CBO) * Executive Director of Congress’s Joint Economic Committee (JEC)

Before his government service, Donald had a varied career as a professor, consultant, and entrepreneur. In the mid-1990s, he taught economics and finance at the University of Chicago Graduate School of Business. He then spent about a year-and-a-half managing large antitrust cases (e.g., Pepsi vs. Coke) at Charles River Associates in Washington, DC. After that, he took the plunge into the world of new ventures, serving as Chief Financial Officer of a health care software start-up in Austin, TX. After that fascinating experience, he started his career in public service.

Donald received his Ph.D. in Economics from the Massachusetts Institute of Technology and his B.A. in Mathematics a couple miles down the road at Harvard.

Visit: Donald Marron

1 Comment on Why Do Estimates of Healthcare Costs Vary So Much?

  1. Compromise doesn’t mean compromising the essence of policy, and all know it !!

    1. As regards a Baucus scheme, the source of funding coming from a middle class is utterly against the commitment of Democratic party while it leaves 25 million Americans without health insurance . And the bill would require people to buy insurance they can hardly afford.

    In response to the scheme, the letter from 154 House Democrats to Speaker Nancy Pelosi urges her “to reject proposals to enact an excise tax on high-cost insurance plans that could be potentially passed on to middle-class families.”

    “This is not an obscure detail of health care reform,” said Connecticut Rep. Joe Courtney, who drafted the letter. “Taxing health benefits was explicitly debated in the campaign by presidential candidates and people running for Congress.”

    Furthermore, looking closely at the new CBO report, it won’t be until “2014 or 2015” that folks start seeing a serious reduction in the number of uninsured.

    2. No cost-competitive advantage of the insurer-friendly scheme does not clear the grave concern about the unsustainable growth in cost of overall health care program in the long run. Baucus scheme Doesn’t Bend Cost Curve Enough, Experts Say.

    And the scheme proposes a “fake” alternative, nonprofit insurance cooperatives — and it places so many “restrictions” on these cooperatives that, according to the Congressional Budget Office, they “seem unlikely to establish a significant market presence in many areas of the country.”

    Beyond, the bill would lock many workers into health plans selected by their employers, without allowing them to shop for better, cheaper plans, an alternative that could help drive down costs for everyone.

    Senator Rockefeller is also upset that the scheme would even turn nearly a half-trillion dollars over to insurance companies, whose profits he says are “out of sight.”

    3. Even with some benefit for primary practitioners, the baseless scheme does not come with fundamental payment reform, or a pay for value reimbursement formula. It means that the insurer-friendly scheme is not cleaning up the concerns over quality, regional imbalance issues and $9trillion of deficit over the next ten years.

    ((Here is some of CBO analysis : While the costs of the financial bailouts and economic stimulus bills are staggering, they are only a fraction of the coming costs from Social Security, Medicare, and Medicaid. Over the next decade, the Congressional Budget Office (CBO) projects that each year Medicaid will expand by 7 percent, Medicare by 6 percent, and Social Security by 5 percent. These programs face a 75-year shortfall of $43 trillion–60 times greater than the gross cost of the $700 billion TARP financial bailout)).

    4. For Medicare & Medicaid system to survive from the most wasteful structure on earth, enough savings via fundamental changes need to be ensured, in return, the savings thereof suffice to meet the goal of well-planned public option.

    ((Even with far less visits to docs, which average a half or a third of them in any other free states, Americans pay roughly twice as much per person right now)).

    5. For the record, prior to nation-wide deployment of reform, The State Of “Yes We Can”, Minnesota influenced by Mayo clinic spends “20 percent” less per patient than the national average and 31 percent less than in the highest cost state. It highlights that no substantial tax raise is needed at least for sure.

    ((The $583 billion of revenue package, and the astronomical savings of public option aside, “20%” of $923.5bn (the combined Medicare and Medicaid cost per year, as of July) represents around $184.7bn per year and 1.847trillion over the next decade, and this patient-centered value alone could be sufficient to meet the goal of public option)).

    6. In principle, the long-awaited and most hopeful health care plan is to meet these criterias : Affordability, Quality, and A Check function against runaway premiums thereof.

    Clearly enough, due largely to its lower overhead cost, purchasing power and fundamental payment reform, the well-planned public option would be doing more than the baseless scheme by THE INDUSTRY in these aforementioned aspects.

    Now is the moment to open the page of contemporary energy and financial upgrades glossed over in 8 years.

    Thank You !

Leave a Reply

Your email address will not be published.