Additions and Alterations by Richard Tol

Up to this point, when discussing Richard Tol and his role in the recent IPCC report, I've only focused on section 10.9.2 of the Final Draft. That's not the only section added after the last round of reviews had already been completed. Section 10.9.3 was also added.

That section's topic, the social cost of carbon (SCC), was covered in the same section of chapter 19 as aggregate impacts were. We can confirm it was moved from there by comparing it's first sentence:

The social cost of carbon (SCC) monetizes the expected welfare impacts of a marginal increase in carbon dioxide emissions in a given year (i.e., the welfare loss associated with an additional tonne of CO2 emitted), aggregated across space, time, and probability (Tol, 2011).

To a nearly identical sentence in Chapter 19's SOD:

The social cost of carbon (SCC) is an alternative index of aggregate damages that monetizes the expected welfare 53 impacts of a marginal increase in carbon dioxide emissions in a given year (i.e., the welfare loss associated with an 54 additional tonne of CO2 emitted), aggregated across space, time, and probability (e.g., Newbold et al., 2010; Nordhaus, 2011a; Tol, 2011; Kopp and Mignone, 2012).

There are only two differences. First, the new version of the sentence is missing the clause which defines SCC. I don't know why anyone would wish to remove the definition of the subject of the section.

The other difference is the original version had four sources, but the new version has only one. I don't know why three sources were removed, but it's troubling the only one left was Tol, 2011, a paper published by Richard Tol, a Coordinating Lead Author of the chapter.

There are more similarities between the two sections, but the next few sentences are new. They discuss an entirely new table and figure, saying:

Figure 10-2 shows estimates published before AR4 and since, using the kernel density estimator by (Tol, 2013), extending the data with new estimates by (Anthoff and Tol, 2013b; Hope and Hope, 2013; Hope, 2013; Interagency Working Group on the Social Cost of Carbon, 2013). Central estimates of the social cost of carbon have fallen slightly for all pure rates of time preference and the uncertainty has tightened, particularly for studies that use a pure rate of time preference of zero. See Table 10-9. For comparison, the EU ETS price in July 2013 was about $21/tC.

Tol's name pops up again. This time he's the sole reference for the methodology used to generate a new table and figure. Another oddity is the text refers to "the EU ETS price," but that was never defined. All other acronyms in the chapter are written out the first time they're used.

More interesting, however, is this text claims estimates of one measure of the cost of greenhouse gas emissions (SCC) have decreased and uncertainties regarding them have tightened. That is the key conclusion of this section. It basically says, "Things are better than we thought and we are more certain of this." That is quite different from what the original version said:

Thus the risk for aggregate damages is similar to that expressed in AR4 and Smith et al., (2009) as indicated in 27 Figure 19-5, with confidence in the assessment unchanged.

The original version said estimates of risks were similar to before, as is confidence in the results. In other words, "Nothing's changed." That's quite different from the new version which said things are better than we thought. And of course, this is based on a figure and table created via Richard Tol's methodology.

Similarly, the old version said:

A further source of uncertainty is whether and how the possibility of catastrophic damages is accounted for (Dietz, 2010; Nordhaus, 2011b; Weitzman, 2009), which requires bounding potential losses with a parameter akin to the value of a statistical life (representing, essentially, willingness to pay to avoid human extinction) (Dietz, 2010; Kopp et al., 2012, p.2012). Without such a parameter, SCC estimates incorporating risk aversion and potential catastrophic impacts can be unboundedly high.

The new version undercuts this by saying:

Concerns have been raised that the uncertainty about climate change is so large that the SCC would be unbounded (Weitzman, 2009), but this result is sensitive to assumptions about the utility function (Buchholz and Schymura, 2012; Millner, 2013; Nordhaus, 2011) and disappears when climate policy is formulated as balancing the risks of climate change against those of mitigation policy (Anthoff and Tol, 2013a; Hwang et al., 2013).

This is a direct repudiation of what was said in the earlier version. It basically lists the earlier draft's position then says that position is wrong. Of course, one of the two references for this argument is Richard Tol's own work.

There are two more short paragraphs in this section, but there's nothing of interest in them. The entire section consists of only four paragraphs, two of which are little more than statements that Richard Tol's work shows earlier versions of the section are wrong, thus things are better than we thought.

Plus there's the table and graph derived from his work:



Chapter 10 has only eleven tables and two graphs (three if you count a cross-chapter box). Combined with our previous examination of of Section 10.9.2, we've now seen two of those tables and both graphs are derived entirely from Tol's published work.

Moreover, all four of those were added after the last round of reviews. They were all created for sections added to the chapter after the last round of reviews. Both of those sections ran contrary to pre-existing text, text which had been reviewed during the normal IPCC process. All the differences favored Richard Tol's views and work. Richard Tol was one of two people in charge of the chapter.

Does anything more need to be said?

Leave a Reply

Your email address will not be published. Required fields are marked *