Robert S. Pindyck is a professor of economics and finance at MIT, with several decades’ experience publishing articles and books dealing with energy. Moreover, as he explains in this interview, Pindyck believes that man-made emissions of CO2 and other greenhouse gases will impose climate change damages on future generations, and is an advocate of a carbon tax agreement among the major world governments (though he is doubtful such a tax is politically feasible). With a pedigree like that, you might expect Pindyck to be very complimentary about the computer models that the Obama Administration and other policymakers are using to justify the economics of anti-carbon measures. But as it turns out, Pindyck has written a new, peer-reviewed paper (forthcoming in the Journal of Economic Literature) that is absolutely scathing in its critique of such models. In this post I’ll highlight some of his points.
Two Types of Computer Models
In the climate change policy debate, there are two types of computer models. One type refers to models of the Earth’s climate that are created as simplified simulations of the atmosphere, ocean, sun’s radiation, etc. that rely just on the natural sciences such as physics, chemistry, and biology. These are the computer models that people have in mind when they say things like, “Global temperatures have been basically flat for years, and yet the official models predicted more warming than has actually occurred.”
But there are another set of models—called Integrated Assessment Models or IAMs—that have been created by economists, not climate scientists. The IAMs rely on condensed versions of the full-blown climate models as part of their structure, but they also rely on (crude) simulations of the global economy to try and assess the interaction between the economic and climate systems. In addition to all of the uncertainty stemming just from the physical science itself—such as asking how much global temperatures will increase in the long run, in response to a doubling of atmospheric CO2 concentrations—the IAMs have another layer of guesswork. For example, they have to make projections of “business as usual” growth in carbon dioxide emissions, in order to understand the full economic impact of emitting one more ton of CO2 today. These computer simulations are then used to gauge the likely results of various types of government policies to restrict emissions, which will affect both the economy and the climate.
Current Crop of Computer Models “Close to Useless”
It is this second class of models, the economic/climate hybrids called Integrated Assessment Models, that Pindyck discusses. Pindyck’s paper is titled, “Climate Change Policy: What Do the Models Tell Us?” Here is his shocking answer, contained in the abstract:
Very little. A plethora of integrated assessment models (IAMs) have been constructed and used to estimate the social cost of carbon (SCC) and evaluate alternative abatement policies. These models have crucial flaws that make them close to useless as tools for policy analysis: certain inputs (e.g. the discount rate) are arbitrary, but have huge effects on the SCC estimates the models produce; the models’ descriptions of the impact of climate change are completely ad hoc, with no theoretical or empirical foundation; and the models can tell us nothing about the most important driver of the SCC, the possibility of a catastrophic climate outcome. IAM-based analyses of climate policy create a perception of knowledge and precision, but that perception is illusory and misleading. [Bold added.]
For those unfamiliar with academic prose, such inflammatory language is almost unheard-of, particularly for a politically sensitive topic such as climate change economics. Pindyck is here reaching the exact same conclusion that I gave in my recent testimony before Senator Barbara Boxer and other members of the Senate Environmental and Public Works Committee: The computer models used by the Obama Administration’s Working Group to estimate the so-called “social cost of carbon” should not be the basis of federal policy.
After my prepared remarks during the hearing, Boxer and others dismissed my testimony as the product of willful ignorance of “the science,” yet I pointed out that it was she and her colleagues who were misinformed. The professional economists who specialize in climate change would know that every point of my testimony was accurate; indeed I was merely explaining to Senator Boxer et al. what the Obama’s Administration’s own Working Group was saying in their official report.
In his paper, Pindyck goes over a few crucial problems with the economic/climate computer models, the so-called Integrated Assessment Models (IAMs). One major point is the arbitrariness of the discount rate used to compare economic current costs of curbing emissions with the projected benefits (sometimes centuries in the future) of mitigating climate change. However, I have already covered that point extensively for IER’s readers, so in this post I’ll focus on another of Pindyck’s arguments, which is the arbitrariness of the “damage function” in these models
“Any Result One Desires”
In my testimony, I said the “economist can produce just about any estimate of the social cost of carbon desired.” Pindyck reaches the same conclusion in his paper when he writes:
And here we see a major problem with IAM-based climate policy analysis: The modeler has a great deal of freedom in choosing functional forms, parameter values, and other inputs, and diﬀerent choices can give wildly diﬀerent estimates of the SCC and the optimal amount of abatement. You might think that some input choices are more reasonable or defensible than others, but no, “reasonable” is very much in the eye of the modeler. Thus these models can be used to obtain almost any result one desires. [Pindyck p. 5, bold added.]
As Pindyck is here underscoring, our claim isn’t that the economists generating large values for the social cost of carbon are “lying.” The point is, these models are so open-ended—they’re trying to model the entire climate system and global economy through the year 2300, for crying out loud—that the analyst has to pick and choose what items to include, and which to omit. If the economist wants the computer to spit out a big scary number, that’s not hard to accomplish.
Robert Pindyck is exactly the sort of expert we are being told should have the floor in the climate change debate: He is not from industry, but instead is an academic at a prestigious post at MIT. He has been publishing on energy issues (including entire books) since the 1970s. And yet, in his recent paper assessing the computer models currently driving federal policy, he concludes: “I have argued that IAMs are of little or no value for evaluating alternative climate change policies and estimating the SCC. On the contrary, an IAM-based analysis suggests a level of knowledge and precision that is nonexistent.”
Now it’s true, Pindyck still thinks there is a strong case for federal intervention to curb emissions of greenhouse gases, and on that score he and I part ways. Yet when it comes to the Obama Administration’s official rationale for its anti-carbon policies, even Pindyck the MIT expert agrees with me: these computer models are close to useless. I wonder if Senator Boxer and others will have a change of heart, since they claim to follow the peer-reviewed literature?