Memo to Media: The MIT study being used by opponents of the Waxman-Markey bill was published in April 2007 (see here). Needless to say, it doesn’t model the bill or any of its key provisions.
The author, MIT Professor John Reilly, explained to me today in an exclusive interview that “the Republican approach to estimating the cost of cap-and-trade is just wrong.” He said even “apart from the misrepresentation of the costs” by the GOP, “it is inappropriate to draw conclusions on the costs of Waxman-Markey” from a study published two years ago that doesn’t model key cost-containment provisions, such as the use of offsets! ["Inappropriate" is an academic term. A better word might be "fraudulent."]
Prof. Reilly said that the Weekly Standard reporter “feigned stupidity” in an effort to elicit answers that could be taken out of context and misrepresented. Reilly called the analysis in the resulting article, “Fuzzy Math: According to an MIT study, cap and trade could cost the average household more than $3,900 per year” — now being cited by conservative blogs and politicians — “just silly” for reasons discussed below.
Reilly said that his study found the cost to a typical family in 2015 of $80 (in terms of reduced economic welfare). And, as we’ll see, he overestimates the cost of CO2 prices by a factor of 2.5 times compared to what it is likely to be under the Waxman-Markey bill aka “The American Clean Energy and Security Act of 2009.”
Reilly confirmed that in his study, even with an 80% reduction in greenhouse gas emissions, GDP roughly quadruples by 2050 and that a reasonable estimate for the slight reduction in GDP growth from even that strong cap is 0.1% of GDP a year — one tenth of a penny on the dollar — which as I’ve noted is pretty much what every major study finds (see “Intro to climate economics“). Since the MIT study is “inappropriate” to use for estimating costs from Waxman-Markey, the only credible analysis of the bill out there is EPA’s, which finds it “could make the median household … better off than they would be without the program.”
FRAUDULENT USE OF MIT STUDY TO PROJECT COSTS OF WAXMAN-MARKEY
No matter how many times MIT Professor Reilly tries to explain to conservatives and the right-wing media that they are misusing his work, they just won’t stop. Three weeks ago he told the GOP to stop ‘misrepresenting’ his work and inflating the cost to families of cap-and-trade by a factor of 10.
That should have put an end to things, as even “mainstream publications” like Congressional Quarterly, The Hill, Politico, McClatchy, and the Wall Street Journal reported on “the falsity of the $3,100 per household cap-and-trade estimate,” as Weekly Standard deputy online editor John McCormack put it in his silly article.
But then conservatives pounced on an irrelevant error Reilly made, namely that the annual cost of the cap-and-trade bill that he modeled — which, again, was not Waxman-Markey or anything like it — would be $800 per family (if you were looking at an averaged, net present value of the costs from 2015 through 2050, not the actual cost to a family in 2015). And then the Weekly Standard with its “feigned stupidity” strategy, as Reilly described it to me, extracted an email response from the MIT Professor that it misrepresented as follows:
The $800 paid annually per household is merely the “cost to the economy [that] involves all those actions people have to take to reduce their use of fossil fuels or find ways to use them without releasing [Green House Gases],” Reilly wrote. “So that might involve spending money on insulating your home, or buying a more expensive hybrid vehicle to drive, or electric utilities substituting gas (or wind, nuclear, or solar) instead of coal in power generation, or industry investing in more efficient motors or production processes, etc. with all of these things ending up reflected in the costs of good and services in the economy.”
In other words, Reilly estimates that “the amount of tax collected” through companies would equal $3,128 per household–and “Those costs do get passed to consumers and income earners in one way or another”–but those costs have “nothing to do with the real cost” to the economy. Reilly assumes that the $3,128 will be “returned” to each household. Without that assumption, Reilly wrote, “the cost would then be the Republican estimate [$3,128] plus the cost I estimate [$800].”
And this led to the Weekly Standard‘s headline: “Fuzzy Math: According to an MIT study, cap and trade could cost the average household more than $3,900 per year.”
When I read that headline to Reilly, he said “That’s just silly,” which again is academic-speak for either unfeigned stupidity and an outright lie.
The McCormack and GOP projections of Waxman-Markey costs using the MIT study are both fatally flawed — because they utterly misrepresent the MIT analysis — and wildly inappropriate — because the study doesn’t model Waxman-Markey.
Reilly explained to me that what he told the Weekly Standard in one of the emails was that the only way the cost to a household would be “the Republican estimate plus the cost I estimate” is if the government literally took the revenues from the cap-and-trade and either “burned them” or “flushed them down the toilet” — literally, those were the two specific cases Reilly offered. Needless to say, he views that as “highly unlikely” and regrets that the reporter who “feigned stupidity” was able to extract even that statement from him. In any case, it is not what Obama or Waxman-Markey will actually do with the revenues.
And remember, this is just a series of emails — not the study itself. It is an outright lie to say “According to an MIT study, cap and trade could cost the average household more than $3,900 per year.” A more accurate headline would be “According to a gross misrepresentation of an MIT study, cap and trade could cost the average household more than $3,900 per year” or perhaps “MIT study ‘inappropriate’ to use for estimating cost of Waxman-Markey bill.”
Indeed, it wasn’t until I interviewed Reilly that I realized how inappropriate (i.e. fraudulent) it is to use his study in the debate over Waxman-Markey. Reilly explained that he just was modeling generic cap-and-trade bills at the time. He doesn’t model any of the important details of the Waxman-Markey bill, and as he confirmed to me “the details matter a lot.”
Here are some things that the MIT study doesn’t model or that it models poorly:
- No offsets. Now, I’m not the world’s biggest fan of rip-offsets, but they are in the bill and likely to stay there. They are specifically in the bill for cost containment, so an economic analysis that doesn’t include them, tells you nothing about the costs of a bill that does. That is no doubt the primary reason that the MIT study projects a price for CO2-equivalent allowances of $50 a ton in 2020, when the EPA gives a much more likely price of $17-$22/tCO2e.
- No government efforts to directly boost energy efficiency or lower market barriers to efficiency for consumers and businesses. Of course, Waxman-Markey and the stimulus bill have major efficiency-boosting provisions (which even EPA didn’t get around to modeling). Reilly told me “if there are barriers that could be corrected through government policies, we have not looked at those.”
- Serious undermodeling of renewables. The MIT model projected that under even the strongest emissions targets, new renewables would rise from 1.5% of electricity production in 2005 to a mere 2.1% in 2035 (!) to only 2.5% in 2050 (!!) — a gross underestimate of a key low-carbon solution.
- Absurd lowballing of future gasoline prices. The MIT model projects a real (inflation-adjusted) price for gasoline in 2020 of $2.40 and in 2050 of $2.10. Hard to believe, but true. Needless to say, for gasoline to be significantly below $5 a gallon in 2020 would take a miracle — or rather 6 miracles see “Science/IEA: World oil crunch looming? Not if we can find six Saudi Arabias!” and “IEA says oil will peak in 2020“). See also “Merrill: Non-OPEC production has likely peaked, oil output could fall by 30 million bpd by 2015“). The higher the future price of oil and gasoline, the lower the net cost of strong energy and climate action, since such action saves so much oil, which is, indeed, a key point of the entire effort.
Obviously, Waxman-Markey, properly modeled, would have far, far lower net costs than what the MIT study found for some generic climate bill two years ago using the above assumptions.
It is time for conservative media and politicians to stop misrepresenting the conclusions of the MIT study and to stop using the study as a source for cost estimates of Waxman-Markey.