Earlier today, DDay pointed out that the NYT, after issuing a fairly pointed correction revealing Jonathan Gruber’s ties with the Administration, then used a quote from Gruber without disclosing his role in the Administration. DDay was focusing on the NYT’s actions. But I would like to focus on the quote.

Jonathan Gruber, a Massachusetts Institute of Technology economist, predicted the excise tax would raise workers’ wages from 2010 to 2019. “There are many academic studies showing that when health costs rise, wages fall,” he said. “In the mid- and late 1990s, when we got health costs under control, wages rose nicely.” But he added that other factors could have also lifted wages during that period.

According to Stephen Greenhouse, Gruber repeated the claim that the excise tax would result in a pay increase for the little guys, and then did the following:

  • Noted that “many” academic studies (though he doesn’t say it, some of them are Gruber’s own studies) show that “when health costs rise, wages fall”
  • Pointed out that during the late 1990s, the slowing rise in health care costs coincided with wage increases
  • Admitted that “other factors could also have lifted wages during that period”

With this formulation, Gruber dramatically backs off one of the key claims excise tax supporters make about the tax–that it will result in a pay increase for those affected. Indeed, he seems to suggest (though I’d need to see a direct quote to be sure) that he doesn’t actually know whether decreasing health care costs would increase wages. He certainly doesn’t appear to say he’s got a study to prove that.

This is one of the reasons why I believe Gruber’s now-revealed ties to the Administration are so important.

The Recursive Claims To Support the Excise Tax Raises

As you might recall, one of my biggest gripes about the excise tax have to do with a bunch of seemingly unexamined assumptions that go into it. For example, the claim that employers would have actual savings from cutting back expensive insurance plans, rather than managing simply to keep health care spending constant; or the claim that increasing out-of-pocket claims will decrease costs without affecting outcomes. Both are dubious and, if they’re wrong, then not only will the tax not generate the revenue promised, but it will make people less healthy.

But one of my biggest gripes was precisely this claim, that employers would pass on presumed savings to employees. I first questioned the claim when Ezra made it based on an uncritical demonstration of the relationship in the 1990s that Gruber points to. Then, I spent an entire post both trying to trace how the White House had justified that claim in a blog post, and pointing out the evidence that contradicted it. I also spent the better part of a day asking economists to point to proof that employers would pass on savings to employees; one normally friendly economist who has elsewhere backed the claim ignored my question; several more pointed to studies showing, again, the inverse (that when health care costs rise, wages fall). More recently, the Economic Policy Institute did analysis–partly based on White House aide Jared Bernstein’s work–showing that wages didn’t go up in the 1990s because health care cost increases slowed.

In other words, no one has been able to point to a study that supports the case. And a lot of data from the real world suggests just the opposite would occur–that employers would pocket the savings as profit rather than passing them onto employees.

The Joint Committee on Taxation and the Excise Tax Raise Claim

Now, the one place that people do point to (aside from things like Gruber’s papers showing the inverse relationship) to defend their Excise Tax Raise claim is a report from the Joint Committee on Taxation, a non-partisan Congressional Committee that provides the same kind of reviews as CBO (but, because it is managed by Committee chairs, may be more exposed to political pressure). I’ll come back to this, but here’s what Center for Budget and Policy Priorities says JCT said about the Excise Tax Raise:

Similarly, the JCT writes, “We expect that consumers will seek less costly policies that will reduce their exposure to the excise tax. Cost reductions could be achieved through several strategies, ranging from managed care plans and limited provider networks to more out-of-pocket cost sharing by consumers. When employers offer employees less costly plans, the employees will have less compensation in the form of non-taxable health care benefits and more in the form of [taxable] cash compensation.”

JCT projects that only 20 percent of the revenues from the proposal in 2014 will come from the excise tax itself, with the remaining 80 percent coming from additional income and payroll taxes on the increased cash compensation that workers will receive. By 2019, fully 83 percent of the additional revenues will come from taxes on higher wages and salaries, not the excise tax.[11]

And here’s what the Chief of Staff of JCT, Thomas Barthold, said to Congressman Joe Courtney, one of the biggest skeptics of the excise tax.

As you can see in the table, other than the first year, the percentage owing to excise taxes is declining over the period, as consumers shift away from higher cost health coverage towards increased wage benefits.

That is, JCT appears to simply assume that workers, not their employers, get to choose whether they want higher cost health care or raises. But I have not found JCT actually citing a study that supports that claim.

Gruber and the Excise Tax Raise Claim

As it happens, Gruber has written three papers (none of these are peer-reviewed; these papers seem to be hybrids that place him squarely in the policy debate about these issues, but not–Gruber says–part of his work-product under the contract with HHS) that include defenses of the Excise Tax Raise claim.

These papers–particularly the November 5 one–are another source cited to defend the claims about the excise tax.

Let me start with the November 17 paper. I don’t dispute some of Gruber’s argument about the AFSCME complaints (note, Gruber hasn’t done a similar response to the CWA’s much stronger report illustrating the problems with the Excise Tax, nor to a recent EPI report that seriously challenged this claim). Here’s what Gruber says about the Excise Tax Raise:

Claim: The argument that reduced employer-sponsored insurance spending will lead to higher employee wages is “speculative”

Reality: The available evidence clearly illustrates that there is essentially a one to one offset between employer insurance spending and wages. There are a number of economics studies that support this conclusion. But it is perhaps most vividly illustrated by simply comparing the growth rate of health insurance costs to the growth rate of wages, a task recently undertaken by Ezra Klein:

do_lower_health-care_costs_mean_higher_wages_(2)

It is readily apparent in this graph that when health care costs moderate, wages rise – but as health care costs increase, wages fall.

Moreover, the ultimate authority on this topic is the Joint Committee on Taxation, and they have clearly spoken: the shift away from high cost insurance raises wages. As I have illustrated in another analysis, the JCT estimates imply that net worker wages will increase by over $300 billion over the next decade under the Senate Finance Committee’s proposed excise tax, after taking out the payments on this high cost insurance tax. The shift out of excessively generous health insurance plans and into wages is a major boon to U.S. workers.

Now, Ezra, to his credit, twice notes this is a correlation (and even calculates it), not proof of causation. But not the economist we’re paying $400,000–for him, “it is readily apparent” is strong enough proof. In my post on this, JTMinIA raised some problem with this correlation. And the EPI does an even more thorough job explaining why this graph can’t prove what Gruber claims it does.

The other value of Ezra’s post is this classic:

Earlier in the day, I’d been talking to MIT economist Jon Gruber about this issue. “There are a few things economists believe in our souls so strongly that we have a hard time actually explaining them,” he said. “One is that free trade is good and another is that health-care costs come out of wages.”

Ahem.

These are, Gruber makes clear, articles of faith, not proven facts. And the absence of any study actually proving this claim (which, I’m sure, most economists would consider less foundational than free trade, for what that’s worth) is instead pitched as a simple “hard time actually explaining.”

Which leaves Gruber, in this paper, with the JCT paper. The “another analysis” Gruber refers to is his own November 5 paper, in which he refers to the JCT’s apparently uncritical assumption that employers will pass on savings in the form of wages.

This [JCT] memo shows the year-by-year revenues raised by the High-cost insurance tax. Importantly, the memo highlights the two different ways the High-cost insurance tax raises revenues. The first is through actual excise tax receipts paid by those high cost plans that remain above the High-cost insurance threshold. The second is through the fact that firms will spend less on health insurance – and this reduced spending will be shifted to workers in the form of higher wages. This conclusion of wage shifting is supported by both economic theory and evidence, and is assumed in modeling by both the JCT and the CBO. This division is very informative: the JCT estimates that about 80% of the revenues raised by the High-cost insurance tax will come from revenue from higher wages, not from the excise tax itself.

Gruber–the guy who has been distinctly unforthcoming about the fact that he had been doing simulations to find out the outcomes of various policy outcomes for more than seven months by the time he wrote this paper–then makes a show of calculating the increased wages based on the JCT numbers.

The JCT estimates can be used infer the impact of the High-cost insurance tax on wages.

[snip]

Key findings[...] are:

  • Worker wages rise by $74 billion by 2019,
  • Worker wages rise by $313 billion in aggregate over this time period, or more than one-third of the estimated price tag of the entire health reform bill

In the November 20 paper, Gruber repeats the same exercise using the revised Senate bill.

Estimates from the Joint Tax Committee (JCT) can be used to demonstrate the important effect of the High-cost insurance tax in terms of increasing worker wages. Using data from the JCT, I show in this memo that the high-cost insurance tax will

  • Raise net worker wages from 2013 through 2019 by $234 billion
  • By 2019, net wages per insured household will be $700 higher because of this excise tax

Do you begin to see why Gruber’s failure to disclose the fact that he has been doing simulations on this stuff is so problematic?

Now, before I move on, let me emphasize what these papers don’t do. They don’t cite any study that proves that a decrease in health care costs bring about an increase in wages. Gruber says they exist,

There are a number of economics studies that support this conclusion.

But as his primary proof, Gruber links to Ezra’s pretty picture (which has since been debunked) and to the JCT analysis (which itself doesn’t seem to cite any evidence). And, apparently, in the conversation supporting Greenhouse’s article, Gruber didn’t cite one either. Gruber, supposedly the expert on precisely this issue, twice stops well short of providing proof for this assertion that wages will increase if benefits are taxed.

Who Is Doing the Simulations?

But I’m more interested in where that leaves us. The popular press often cites Gruber for this claim (without, of course, any disclosure that he’s working for the Administration), and Gruber cites JCT. And neither of them, apparently, cite any study proving this claim.

More interesting to me is that a guy who had, by the time he wrote his November 5 paper, received something in the neighborhood of $250,000 doing simulations showing what would happen if “the President’s plan” were implemented, seeming to work backwards off of the public JCT data, all the while proclaiming “the ultimate authority on this topic is the Joint Committee on Taxation.”

In one explanation to Ben Smith, Gruber said,

Gruber said his work for the administration was running the sort of cost simulations that the Congressional Budget Office does, based on a model that he’d spent 10 years developing. “I’m the numbers guy,” he said.

Part of its value, Cohn writes, is that it helped the administration predict CBO scores.

In an earlier one, Gruber said,

Throughout this year I have provided technical assistance to the administration and to Congress with my micro-simulation model, as well as based on my experience as a member of the Massachusetts health connector board.

Now, to be fair, Gruber once told Jonathan Cohn that he wasn’t doing the same analysis as the CBO on a different topic,

He hasn’t formally modeled the impacts of the reforms on premiums; for this analysis, he has relied simply on available data from the Congressional Budget Office.

But Gruber’s admission that his consulting includes working with Congress doing the same kind of analysis as CBO, and his repeated production of papers that apparently replicate analysis done by CBO and JCT, apparently does just that sort of analysis before the CBO and JCT do them.

Only, in papers he says had nothing to do with his contract, to then work backwards off the analysis of CBO and JCT to prove points that then get used in the Administration’s support of its preferred version of health care reform.

I’m sure Gruber is not doing the CBO’s work for him. I doubt he’s doing JCT’s work either. (Though, in both cases, I’m mindful of the delay Reid had every time he submitted something to the CBO.)

But the way in which Gruber repeatedly does this kind of analysis, all the while suggesting he hasn’t been doing precisely this kind of analysis for the Administration, raises more questions about his role.