Leading economists speak to RTCC about the row over how to account for global climate impacts
Frank Ackerman (L) and Richard Tol (R)
By Ed King
US economist Frank Ackerman has accused Sussex University professor Richard Tol of distorting his work, and of producing misleading reports on the costs of addressing climate change.
The charges, which Tol denies, are significant given his position as a contributing lead author on the last UN climate science study compiled by the IPCC.
Used by governments to work out their long term climate strategies, the IPCC reports set the tone for international talks at the UN aimed at developing a global emissions cutting plan.
Last week RTCC spoke to both economists – first Tol, and then Ackerman. Here’s what they had to say.
Ackerman’s is a little shorter, as he was off work with a cold – any spelling mistakes are mine.
Ed King: Can I get your wider views on this [Ackerman] paper, because it’s fairly strong stuff…
Richard Tol: You should start reading at the end, where Ackerman gives a skewed account of history… you should read it all in that light. He raises essentially three points. He starts with an introduction, then a paragraph ‘only sixteen’, which suggests that I would have overlooked stuff, that there’s more then 16 , but he doesn’t actually find any. I mean, he says, why are there only 16 estimates? The suggestion is that I would have overlooked 24 other studies.
But he cannot list a single one. Which is a bit misleading I find. Then, well, Ackerman nowadays is a consultant, so he doesn’t have time to keep up with the literature. He raises a specific issue about dependence between studies, that would affect statistical estimates. Unfortunately he overlooked a later study that I did that is now forthcoming in Computational Economics, where I explicitly test for this , and it doesn’t affect the results at all. But yes, this is just Ackerman not keeping up with the literature.
Then. In the second section, shrinking the SCCs [sic]. He sort of makes a similar mistake. He complains about dependence between studies, but overlooks this has been tested for, and doesn’t affect the results. This is older studies that did this, so Ackerman has less of an excuse for not having studied the literature properly that he seeks to criticise. The other point that he raises is that some of these estimates are sensitivity studies and some are proper estimates. But again he overlooks that the method actually corrects are that, and has done so since the first paper that goes back to 2005. So again it’s just poor reading skills on the part of Mr Ackerman.
The third paragraph or section – making up missing estimates – um – again – um – all this points have been raised before. I mean, there is a big issue here with selection bias…in the model runs. So the IPCC 4th assessment report and 5th assessment report make a basic mistake we try to beat out of our undergraduates, and that is that if you compare estimates of the impacts of climate policy across models you should be very careful, because models that find emission reduction is very expensive tend not to report results for very stringent climate policy.
Whereas models that find climate policy is cheap, actually go for very deep emission cuts. So, if you’re taking an average across results for lenient climate policy you find one result, but then you can’t compare that to the same uncorrected efforts across very stringent climate policies. And that is exactly the mistake the IPCC has made twice in a row. And that is what the [inaudible] paper is about for correcting for selection bias.
Now, I’m not quite sure why Ackerman wants to re-introduce selection bias, because as I said it’s something we try and beat out of our undergraduates. The other point that he raises here is that EMF22 is not the only study that is out there…which is true, there’s no disagreement there… [inaudible] but the problem is the results are incomparable – EMF22 included I believe 11 different models that all ran the same scenarios, so we can compare and contrast different models.
Whereas the studies he refers to – well, he alleges but doesn’t give a proper reference, but I know what he’s talking about, does not do that same control. So if you look at their database it’s actually a big mess, and you have to introduce a whole lot of assumptions to actually include these estimates.
That is why, the paper he refers to, explicitly makes the point that these studies did not deliver data in the same high quality as EMF22 did. So I’m working on that and going back to the IPCC database for the 5th assessment report…where the same problem occurs that things are just not comparable across models, there has been no vetting of model results, for instance, some of the models, some of the data, suggests that we give subsidies that are greater than our total income. Which is impossible.
You simply can’t spend more money than you have. There are also models which have carbon taxes which are greater than GDP. So we have an income tax that exceeds 100%, which is also tough on people. There are all sorts of reasons why this [inaudible] is not as good as he claims it is…and used for these purposes. And that is why my 2013 paper still uses EMF22. There is simply nothing of comparable quality later.
Ackerman is throwing out allegations and insinuations without having checked whether what he is saying is feasible or correct or anything. It’s just a low quality piece of work.
EK: Just so I’m clear – generally – do you refute all the allegations there?
RT: There are no allegations – there are just insinuations…
EK: Well…he says things that are quite serious
RT: No they’re not
EK: He says you have seriously misrepresented some of his work…on page 4 for instance – do you completely reject that allegation? Is that completely wrong?
RT: Where does he say ‘seriously misrepresents’?
EK: On page 4…5. That’s serious isn’t it?
[silence/inaudible]
RT: Oh – no, no, no…this …ok. No. He again just misread what was done, ok. So yes he reports these estimates and therefore they are in the database. That is how databases work. If you report something it is stored. Now after that, there’s nothing incorrect, but he sort of starts the sentence with ‘incorrectly blames’ – that’s the first sentence – and then the second sentence is…or the third, fourth, is ‘we did report this’. What sort of nonsense is this? At the start of the paragraph it says no we did not do this, and at the end he says yes we did do this. I mean he’s contradicting himself there.
Now, I did not fall into the trap he suggests I fell into here, and that is yes – these estimates were reported, and therefore they are in the database, but they are given a very low weight. I mean nowhere did I claim this is Ackerman’s best guess. Nowhere. And if he actually had taken trouble to look at the data, it’s freely available on the web and everyone can look at them, he could see I had heavily discounted the estimates he reports.
The fact that he chooses to re-report certain estimates rather than other estimates, because there are by now almost 1000 of these estimates out there, the fact he chooses to re-report one, contrary to what he says here, actually does increase the degree of that particular estimate and therefore it should be re-included.
That is the reasoning behind it, as if fully explained in the paper that uses these methods and these data… Ackerman simply – I mean the man simply has a problem of reading comprehension.
EK: Ok… I mean – does this point to a wider problem in that climate economics…looking at all the people quoted there – people like Chris Hope who we come into contact with infrequently – climate economics given how important it is for governments working out what their strategy is going to be – is a pretty small world…
RT: No – there’s a fair number of people
EK: But compared to climate science – it’s not comparable
RT: Yeah yeah, it’s much much smaller. Now the name you mentioned Chris Hope is actually a bad example. Chris is not an economist, he’s not trained in economics he’s not employed in an economics department
EK: Sure… well he does seem to do quite a lot in this field
RT: Which is a bit peculiar given he’s not an economist… right. There are mistakes in Hope’s work that is perhaps for another time. No there’s a good few people working in this field, so I don’t think this is a problem. The issue that is going on here right is that somebody who is not at a university, but works for a consultancy, writes what he claims to be an academic, writes it, puts it on the web, it gets publicity apparently. It’s not peer reviewed, it’s not from a credible source, and yet we are talking about it.
If you focus on what is in the literature, stuff published by academics, rather than consultants, you find much less controversy than what is suggested here. I’m not saying we all agree – that would be untrue. There is big fights going on in the academic world around the economics of climate change, but Ackerman is not part of that discussion, Hope is not part of that discussion.
EK: Ok – well – just so I’m clear in my head – the allegations made here refer to a number of calculations and work you have done in the past few years. Do you stand by all those calculations…? Is it fair to say you stand by what you have published already and you are not going to back down in the face of what Ackerman is saying?
RT: No, because Ackerman is…who is he? What does he do? He puts a self-published document on the web. He’s not an academic. He actually in this paper, he doesn’t do anything. As I said, all the data are freely available to anybody, actually the code to process the data is free to anybody, including Ackerman, so what he could have done is take the data, take the code, change what he doesn’t like about the data or the code, re-run the thing and show it makes a substantial difference to the conclusions.
Does he do that? No he doesn’t. He comes up with insinuations and allegations but he doesn’t actually exercise through what the implications of what I’m saying. It’s that easy. The code is there – the data is there. This is not a serious piece of work.
He also refers to another piece of work by Backer and Crawford Brown [phonetic spelling] which is the same stuff [inaudible] but if you look at their paper they’re just bitching, they don’t actually take our data that freely available, take our code that is freely available – re-run it and show, yeah, this assumption we didn’t like and we changed it and therefore the conclusions are different. That is not what they do, they just sit there bitching about what people have done, without re-running the exercise.
EK: I mean…it does strike me that there does seem to be an intense amount of competition between academics – between various academics. Is there a problem here…in terms of people exchanging ideas and data?
RT: Not from our point. Our stuff is freely available. The issue here is that somebody puts up a document on the web, doesn’t send it to a journal, does not actually make a contribution. Makes all sorts of insinuations, actually doesn’t follow up on what are the implications of the insinuations, and then we are forced to talk about it with a journalist.
I mean, what is this, why doesn’t Ackerman just do his own study, and rather than insinuate that I or we did things wrong, actually show what are the implications if I change the assumptions he doesn’t like. Why doesn’t he do that, finish that and come back.
EK: Ok – just to be clear – are you planning to release an official response…will you fisk what he says
RT: No. There is nothing here to respond to…
EK: Surely if this is complete bollocks… if someone wrote this about me – I’d want to take account with them. Surely you have some legal recourse if someone slags you like this?
RT: No, you see that wrong. For the following reason. This is essentially a baiting exercise…Ackerman has a history of libelling me. He also has a history of claiming that I would take him to court if he didn’t stop libelling me, which is untrue, as I’ve never threatened to take him to court, and he actually can’t prove I’ve threatened to take him to court.
That’s another insinuation he makes. I think what he’s trying to do is provoke me to suing him…and he’s not a rich man [inaudible] but it would immediately destroy my reputation, because law suits between academics are… in the case of Ackerman and hangers on… is completely not done in academics circles. So if I were to sue him my academic reputation would be completely and utterly destroyed, and that is also why he keeps on insinuating that I threatened to sue him, even though I didn’t, because it’s a blemish on my reputation – to make him the good guy and the poor victim. This is just a baiting operation.
EK: And just to clarify… there are some claims – I think Bob Ward has written about it – that you are a member of the Global Warming Policy Foundation – is Sussex University your main role…are you a member of any other climate sceptic organisations?
RT: Well…I’m a professor at the University of Sussex, I’m a professor at the Free University of Amsterdam, I’m a research fellow of the [inaudible] institute, and a research fellow at FEO in Munich. So those are my official affiliations. Bob Ward as you probably know is a failed PhD student who is now the PR man for Nick Stern
EK: Ok – great thanks. Good to get your views…you’re not taking part in the New Climate Economy study are you…ahead of the Ban Ki-moon summit – that Felipe Calderon is pulling together?
RT: No – that’s run by Nick Stern. And Nick doesn’t like me.
-ends-
Ed King: Can you synthesise your concerns – is there a particular issue you have with this paper?
Frank Ackerman: He adopts a magisterial authoritative tone in saying ‘I will tell you what economics knows’ and ‘what good economics is’ and there are exactly 16 studies of this and so many studies of that, and then I will tell you what is the correct way to analyse those.
He, like many people has his own vision of what it all means and how it fits together. But he adopts in this a tone of speaking for the economics profession and this absolutely does not represent the whole of the economics profession. It is not a reasonable methodology for summarising the findings of different and inconsistent studies and it, I believe, slips in a personal leaning toward finding the problem not very severe and finding that the optimal policy is to allow an extraordinary level of emissions and warming, and then presents that as what the economics profession believes. So as a lifelong economist who is interested in these things, that sure isn’t what I believe, and when I look to see how my work is represented, I find it is misrepresented.
EK: Professor Tol suggested you haven’t looked at what he has written, and that this is more personal as opposed to focusing on the content of his findings. From your perspective, did you read all of his paper and the surrounding literature?
FA: Do you think he had read the eight pages I wrote? There are a few paragraphs at the end that are personal, but seven and a half of the pages are entirely about the content. I talked about the specific studies, he says there are 16 studies about climate damages. I talked about one of those – is an expert survey done more than 20 years ago, and I suggested that might not count as one of the current data points on what we now know about climate change.
I talked about the age of those studies – he continues to rely on very old studies. Six of the 16 studies he believes are the only valid studies are all from 1996 or earlier. That’s not really personal. It’s hard to see what more one will do in terms of engaging with the content. I talked about the very strange use he made of the EMF22 models. That’s not something I have personal feelings about – that’s about the content of his paper.
I talked about the ways in which a different survey of the same kind of models done by European researchers came to very different conclusion, and that he made a particular use of a particular survey which was not well suited to his purpose, as Terry Barker and others have commented on in the literature. Take a good look at that paper – I have a lot of citations, specifics about numbers and documents.
He would like to present himself as the voice of the economics profession – and anyone who disagrees with him has personal feelings. But I think it’s time to reject the idea that he represents the economics profession. He represents one opinion way over to one extreme and has been willing to write endlessly about it, bully people into silence when they disagree with him. As the one person who seems to have withstood that I feel like I have some responsibility to stand up and say – look – the emperor has no clothes.
EK: You talk about a small number of economists… is it an issue the wider focus on climate economics is not there?
FA: Well – there could certainly be more, but the issue is there is an excessive focus on a handful of people who have published frequently in the past. Yes it could use more attention but there is starting to be more attention. There is very interesting work by a young economist called Melissa Dell, who’s now at Harvard University that suggests ways in which temperature affects growth rates much more seriously that we previously expected – that’s one of the very interesting things you look at.
Nicholas Stern has done a tremendous amount to open up the meanings of economics of climate change, and to suggest ways in which the traditional research is limited. Tol has a personal animus about Stern… everything he seems to write manages to work in his personal feelings about what’s wrong with Stern.
Stern never appears on his list of the leading climate economists – and Martin Weitzman, who’s at Harvard, has written interesting things about catastrophic risk, and how do you deal with that in a situation of great uncertainty… as a motivation for more active climate policy. So I think there is a world of interesting writing, and that Tol’s manner of summarising what he sees as the field serves to help keep the focus narrow on the same handful of sources.
And I think even within his world of sources, one of the things is William Nordhaus, who is everyone’s number one name at the top of the climate economics field, developed the DICE models and has been writing about this for 20 years. If you look closely at Nordhaus… he’s changing his tune. His analysis says it is a more serious problem than his earlier analyses used to be.
If you look at his newest book which came out last year – the Climate Casino – it says much more about the nature of economic models and the nature of catastrophic risk. So to continue to cite Nordhaus’s older work, which is what Tol does, without looking at how he is evolving is trying to freeze the field on a narrow and dated vision of it.
EK: The allegations in this paper are quite serious and represent a challenge to the IPCC process, because this paper, as you mention in your writing, Chapter 10 drew on it quite closely. Are you going to submit this higher up? Does this need to be sent to a journal or the IPCC?
FA: I think if people want to raise it that’s all for the good. I think the IPCC represents a valuable consensus view on a lot of things. I think it’s clear that WG2 CH10 with Tol as one of the coordinating lead authors leaned much too heavily to Tol’s work, and does not deserve the same kind of recognition as a consensus view.
I’m not aware of other cases where chapters have followed so precisely the work of one of the coordinating lead authors – they’re usually more judiciously multi-sourced. It’s fine with me if someone who’s in the IPCC process wants to raise that. I’m not personally.