Richard Tol challenges assertion by Ottmar Edenhofer in ZDF

Richard Tol:

Ottmar Edenhofer claimed in „ZDF umwelt“ on September 5, 2010 “Die Aussage, der IPCC hätte bewusst Dinge herausgehalten, die ihm unbequem waren, die nicht gewissermassen in eine Gesamtstory gepasst hätten, kann ich beim besten Willen nicht sehen”. (I cannot understand, even if I try hard, the assertion that the IPCC would deliberately have omitted things, which would have been inconvenient, which would not have been consistent with the overall story.)


This assertion of the co-chair of Working Group III of the IPCC is at best peculiar if not outright false. In the following, I will back this statement in some detail, by demonstrating how specific conclusions from white publications, known to the IPCC lead authors, have been filtered out in support of a (false) claim of consensus in the Summary for Policymakers. At the time of his interview, Dr. Edenhofer was aware of these inconsistencies.


The Synthesis Report of the Fourth Assessment Report says: “studies […] that take into account induced technological change may lower the[se] price” of emission reduction; “[g]lobal modelled costs will decrease […] if induced technological learning is included”; and “[i]n the models that consider induced technological change, projected costs […] level are reduced”. The Summary for Policy Makers of Working Group III says: “In the models that adopt [induced technological change], projected costs for a given stabilization level are reduced”; “[s]tudies that assume the possibility that climate change policy induces enhanced technological change […] give lower costs”; “some [models] show GDP gains because […] they assume that more technological change may be induced by mitigation policies”; and “studies […] that take into account induced technological change lower the[se] price” of emission reduction.

Thus, AR4 argues that induced technological change reduces abatement costs. It says so three times in the Synthesis Report, and four times in the Summary for Policy Makers. In one case, the word “may” is used, but the remaining six cases express the finding without a doubt.

This is strong conclusion, but unfounded.

The IPCC puts forward its evidence in Chapter 11. Ottmar Edenhofer was a lead author. He and one of the convening lead authors, Terry Barker, have regularly argued in favor of the hypothesis that induced technological change reduces the costs of greenhouse gas emission reduction – and Chapter 11 indeed cites papers by Barker and Edenhofer in support of this result.

The evidence rests on the comparison of two sets of models – with and without induced technological change. There is no experimental evidence, and no observational evidence to support (or reject) the hypothesis. The models have not been validated. So, the hypothesis so confidently expressed by the IPCC has no empirical content.

The models used are relatively simple, conceptual models. A mathematician can understand the models’ behavior without resorting to computer simulations. In fact, the main mechanism is so simple that it can be described in a few sentences:Greenhouse gas emissions are reduced by improved energy efficiency and switching to alternative energy. Those technologies would be used more and experience would improve them. Companies would invest in R&D to further advance these technologies. That is, climate policy would reduce the costs of climate policy. This is the mechanism used in the models that support the IPCC conclusion.

However, this is not the end of the story. If smart people focus on improving clean energy technology, they will not research other issues. Climate policy will not make people smarter, so progress slows down for non-energy technologies. The cost of this would at least partly offset the gains of better energy technology, and the costs may indeed be greater than the gains.

The IPCC therefore overstated its confidence. Induced technological change may increase or decrease costs. This is an empirical question. No empirical work was done. We therefore do not know the sign with great confidence.

We can, however, make an informed guess. Energy is less than 5% of the cost of living and the cost of doing business. Accelerating technological progress in energy at the expense of decelerating technological progress elsewhere can hardly be a winning proposition. In fact, technological progress in energy initially decelerates too because of the switch to less well-developed and more expensive but clean energy.

One could therefore expect that induced technological change would increase the costs of emission reduction. Nordhaus (2002), for example, finds that induced technological change increases costs by $347 bln.

Chapter 11 says: “[w]hile some models find a large reduction in mitigation costs (e.g. Popp, 2006a), some find small impacts (e.g. Nordhaus, 2002)” and “Nordhaus finds very modest mitigation cost savings.” Although Nordhaus (2002) reports an increase in costs, the IPCC claims he finds a decrease!

Is this an honest mistake? Many of the models used in Chapter 11 are incomplete. They do not account for the negative impact of climate policy on overall technological change. On February 23, 2005, in Berlin, Peter Dixon of Monash University told the modelers that their models are incomplete – with three IPCC authors in the audience. On January 24, 2005, at an IPCC workshop in Washington DC, Sjak Smulders of Tilburg University told his audience that incomplete models tend to get the sign wrong.

Two referees of the first order draft argued that Chapter 11 should make reference to Smulders’ papers. The response was: “A very few authors (e.g. Smulders) have found that allowing for ETC in top-down models increases costs, and many have found that it reduces them. This is not a consensus, but it does suggest that the balance of findings is that inclusion of ETC in the modelling reduces the cost estimates.” One referee raised the issue again in response to the second order draft. The response was: “REJ[ect] […] The text is describing the literature. ITC through LBD reduces the costs in the model applications reviewed.”

Writing the first-order draft, the authors of Chapter 11 were aware that there is no consensus (their words) on the effect of induced technological change on the costs of emission reduction. Writing the second-order draft, the authors claimed that there is a consensus. In the published chapter, there is not a sliver of doubt on the sign. It says “those modelling studies that allow for induced technological change involve lower costs”; “induced technological change […] tend to reduce costs”; “induced technological change […] has its own opportunity costs, which may reduce the potential for cost reduction […] substantially”; and “at the upper end of the range, including [induced technological change] […] reduces mitigation costs by about 90%, but at the lower end it makes no difference”. According to Chapter 11, induced technological change may have a small or a large positive effect on costs – but it cannot have a negative effect.

Chapter 11 cites two papers (Nordhaus, 2002; Smulders, 2005) that show the opposite. The authors were told at two meetings that their hypothesis does not stand. Review comments on both drafts pointed out that Chapter 11 misrepresents the literature.

In that light, Edenhofer’s “[d]ie Aussage, der IPCC hätte bewusst Dinge herausgehalten, die ihm unbequem waren, die nicht gewissermassen in eine Gesamtstory gepasst hätten, kann ich beim besten Willen nicht sehen” is most peculiar.

---------------------------------------------------
Comment by Hans von Storch: Ottmar Edenhofer as well as the Renate Christ from the IPCC secretariat have been informed about this challenge