David R. Henderson  

Jonathan Gruber's Analysis of "ObamaCare"

PRINT
End the IMF?... Words of Wisdom on Signaling...

Last month, MIT economist Jonathan Gruber published an NBER Working Paper, "The Impacts of the Affordable Care Act: How Reasonable are the Projections?" WP No. 17168. Gruber points out in a footnote that he was a paid advisor to Governor Mitt Romney when Romney and the Massachusetts legislature imposed "RomneyCare" and that he was a paid consultant to President Obama when Obama and Congress imposed "ObamaCare." This is contrast to his failure to disclose, in a WaPost op/ed during the health care debate, substantial amounts of money he received from the Obama administration.

It's actually a fairly nice summary of the major provisions of the Patient Protection and Affordable Care Act (ACA). Beyond that, though, given that the paper is supposed to be about how reasonable the projections are, it's a disappointment because he never really answers the question he poses in the title. One gets the feeling that he thinks the Congressional Budget Office's projections of spending and taxes under the ACA are reasonable, but it's just an impression. He doesn't ever take a stand. The paper seems hastily done, and that impression is reinforced by the fact that on pages 2 and 24, the editor's queries to the author remain.

Some highlights and omissions.

1. He claims that there are "enormous disparities in access to health care" in the United States and his Exhibit A to support that statement is, "For example, the infant mortality rate for whites in the United States is 0.57 percent, while for blacks it is more than twice as high, at 1.35 percent." So note that rather than giving evidence about access per se, he gives evidence of outcomes, which are, of course, dependent not just on access but also on other things.

2. In listing the ACA's new regulations of insurance--he claims that "these reforms are viewed by most as long overdue." This is the problem with the passive voice. It doesn't specify who views them that way. Most economists? Most people? Most people he talks to? He doesn't say and so his claim is difficult to evaluate.

3. He points out that the government, with guaranteed issue combined with not allowing insurance companies to price for risk, has set up an adverse selection problem. In an op/ed in the St. Louis Post-Dispatch in 1994, I called this "adverse selection on purpose." That's why, writes Gruber, an individual mandate is required. Otherwise, he argues, some people would free-ride, going without health insurance until they get sick and then getting health insurance. He points out that by 2016, when the law is fully phased in, most people will have to have health insurance or pay a fine equal to the larger of 2.5% of income or $695.
One might think that most low-income people would pay the fine because doing so is way cheaper than buying health insurance. But Gruber points out that that's why ACA subsidizes health insurance so heavily for low-income people.
What I found interesting, though, is simply his admission that regulation creates adverse selection. He notes that in the 1990s, five state governments regulated their non-group insurance markets with guaranteed issue and restrictions on pricing for risk and by 2006 those five states were among the states with the most-expensive non-group insurance.

4. In discussing budgetary implications, Gruber never mentions that the phaseout of the subsidies for health insurance will raise implicit marginal tax rates for modest-income people by over 20 percentage points. I would expect an effect on labor supply to the above-ground economy. This implies less federal income tax revenue and payroll tax revenue than otherwise. He doesn't mention it.

5. Gruber hedges at the end by saying that the ACA "may, however, provide a first step towards controlling costs." That cautious tone is justified. It's hard to reconcile that, though, with this letter that he signed on January 26, 2011. Just 5 months earlier, Gruber was so sure that the ACA was a good idea that he signed a letter that stated, among other things:

[T]he Affordable Care Act contains essentially every cost-containment provision policy analysts have considered effective in reducing the rate of medical spending.
. . .
Taken together, these provisions are likely to reduce employer spending on health insurance. Estimates suggest spending reductions ranging from tens of billions of dollars to hundreds of billions of dollars.
. . .
The budgetary impact of repeal also would be severe. The Congressional Budget Office concludes that repealing the Affordable Care Act would increase the cumulative federal deficit by $230 billion over the next decade, and would further increase the deficit in later years. Other studies suggest that the budgetary impact of repeal is even greater. State and local governments would face even more serious fiscal challenges if the Affordable Care Act were repealed, as they would lose substantial resources provided under the new law while facing the burdens of caring for 32 million more uninsured people. Repeal, in short, would thus make a difficult budget situation even worse.

What changed in Gruber's thinking in just 5 months?


Comments and Sharing





COMMENTS (11 to date)
Jonathan writes:

What are the editor's queries?

The version currently at http://www.nber.org/papers/w17168.pdf doesn't appear to contain anything on page two, and page 24 is simply Table 1.

David R. Henderson writes:

@Jonathan,
On p. 2: **AUTHOR: PLEASE PROVIDE A CITATION AND REFERENCE**
On p. 24: **AUTHOR: IS THIS THE CORRECT REFERENCE?**
This is the hard copy. Perhaps you have the electronic copy.

Yancey Ward writes:

What has been changing his mind? Reality.

Daniel writes:

Is there any word you can use to describe the enactment of healthcare legislation other than "impose"?

Daniel writes:

Beyond the "impose" issue, which is stylistic...

You have a good point in general about his non-disclosure earlier on, though it hardly seems relevant here.

Who is to say what his paper was "supposed to be about" except the author? Some writing with questions in their titles is really just about the question and the evidences for and again with no final position taken. As to your other points:

1) Good point.

2) The problem isn't with the passive voice. "Most view these reforms..." would be active voice and just as confusing. The problem is the lack of the word being modified by "most". And it's not that big a problem. "Most" by itself doesn't mean "most economists" any more than it means "most giraffes". It means most people. Follow pollingreport.com on how people feel about various parts of Obamacare, and I think you'll see that things like mandatory issue and community rating were, for better or for worse, popular.

3) Your point about people paying the fine rather than taking out the insurance doesn't follow from even basic cost/benefit analysis. The cost of paying the fine is lower, but so is the benefit. All the fine has to do is cover the difference between the benefit of insurance and its cost. Also, this really isn't an issue for most truly low income people, as they won't receive insurance through the exchanges but will instead get it for free via Medicaid. Now, how good Medicaid insurance is is still open to debate. The recent Oregon study shows that it's better than nothing, but that's a low bar to get over.

4) Figured you'd quote Mankiw on this one. This is really the point he's been developing and, potentially at least, it is a good one.

5) You're straining at a gnat on this one. There just isn't much difference between "may" (in the latest source" and "likely" in the first one. Nor is there anything in the former to show that he's backed away from the CBO estimates in the first.

Chris Koresko writes:

1. He claims that there are "enormous disparities in access to health care" in the United States and his Exhibit A to support that statement is, "For example, the infant mortality rate for whites in the United States is 0.57 percent, while for blacks it is more than twice as high, at 1.35 percent."

It's worse than that, if my memory serves: Don't Hispanics have a lower infant mortality rate than whites? If that's true, this `evidence' pretty much collapses, doesn't it?

Shayne Cook writes:

For Daniel ...

The word, "inflict" came immediately to my mind when ObamaCare/ACA was being enacted.

Daniel writes:

For Shayne....

I'm sure it did.

Roger Sweeny writes:

Taken together, these provisions are likely to reduce employer spending on health insurance. Estimates suggest spending reductions ranging from tens of billions of dollars to hundreds of billions of dollars.

Long ago, when Congress was considering funding the Space Shuttle program, NASA assured them that the Shuttle would be a "space truck": cheap, reliable, and safe. Anyone with any sense knew that was ridiculous. But so many wanted to believe.

Three decades later ...

Floccina writes:
1. He claims that there are "enormous disparities in access to health care" in the United States and his Exhibit A to support that statement is, "For example, the infant mortality rate for whites in the United States is 0.57 percent, while for blacks it is more than twice as high, at 1.35 percent." So note that rather than giving evidence about access per se, he gives evidence of outcomes, which are, of course, dependent not just on access but also on other things.

He forgot to mention that the infant mortality rate for Hispanics in the United States is lower than for white even thought they have less access to healthcare.

And coincidentally Hispanics have a lower rate of multiple births than whites and Blacks a higher rate.

Victor writes:

Is there a non-gated copy?

Does he ever address how reasonable the projections are by region of the country? The impact of reform will be dramatically different between currently-underwritten states versus those that have already pursued "anti-selection on purpose" (a good phrase).

Also, does he explain why the projections didn't change significantly for the change in grandfathered timing from 12/31/13 (PPACA, November 2009 version) to 3/23/10 (PPACA as we know it today)? The issue is huge because of guaranteed-renewability of individual insurance; in the first PPACA draft, people would have had the option to go through underwriting in the fall of 2013 and receive select rates for the remainder of their lifetimes. This would remove their contributions to the broader Exchange pool vis a vis the risk pooling and risk adjustment mechanisms. In the enacted draft, individual sales of insurance will dry up in 2013, and the 2014+ risk pools are presumably supported by healthy individuals who have no other legal choice.

Despite all this, the CBO continues to advertise that their premium estimates performed on November 30, 2009, are the best available. Truly bizarre. Unless that sort of oddity is addressed, I don't see how one can discuss "reasonableness" with a straight face; they are seemingly admitting up front that their model doesn't reflect current risk-selection arrangements or provide for the proper implementation of risk pooling under the law.

Comments for this entry have been closed
Return to top