Lomborg-errors: "Cool it!"

Cutting carbon and what it costs
    Home                                                                                                                                   Cool it

 "Cool it!", chapter 2: It´s getting hotter: The short story
On cutting carbon and the Kyoto protocol, pages 21-38.


    Lomborg tries to present the Kyoto protocol as an inefficient treaty that will do practically no good, and at a very high cost to society. According to one of his sources (Wigley 1998), the global yearly CO2 emissions will rise by 2100 to 20.6 GtC with business-as-usual. If all annex I countries, including the USA, would stick to the limits set by the protocol not only until 2012, but also in all the following years, it will rise to slightly less, namely 18.4 GtC. The main reason that even af full implementation of Kyoto would reduce CO2 so little, is that there are no restrictions on CO2 emissions from China, India and other developing countries. Data from IPCC (Third assessment report 2001, WG 3, p. 154) show that in most scenarios of the future, if the annex I countries stick to the set limits, the rest of the world - the non-annex I countries - will have higher CO2 emissions per capita from some time in the middle of the 21st century, probably from about 2040 or 2050 onwards. The situation described by Lomborg is therefore an absurd situation where the former poor countries have an unrestricted growth in emissions and emit more per capita, even when their total population is much larger than the population in the annex I countries. Such a situation would of course never be accepted. The developing countries would of course have to meet the same standards as the developed countries.
    The usual answer to this dilemma is that Kyoto is only a first step, and that the developing countries will have to contribute later on. Lomborg rejects this possibility. We can now see how absurd this rejection is. It has nothing to do with anything realistic.
    Lomborg also presents his punch-line sentence that Kyoto will postpone temperature rise by just seven days. It must be added that this is so only if every nation abandons Kyoto when the treaty expires in 2010; in that case, the effect ninety years later will be a postponement of seven days. If the nations stick to the promised reductions, the postponement will be seven years.
    Lomborg makes us feel that it will be very costly to keep CO2 emissions constant in the industrialized countries. One should not just take his word for it, but check what is said about this in the IPCC reports, for instance here. From these reports, you get a very different impression. To stabilize the atmosphere at a level below 590 ppm CO2-equivalents would reduce the global average annual GDP growth rate by less than 0.1 percentage points, and by 2050, the global GDP would have become about 1.3 % lower (range: less than 0 up to 4 %) than with business-as-usual. This is much less scary than the impression that Lomborg gives us.


    Page 26 in Cool It is used to explode the thesis that one can cut emissions and actually make money through the process. Economists talk about whether there exist so-called no-regret options. Lomborg postulates that economists are wary of such claims about no-regret options, and to document this, he brings a reference to his own rejection of them in The Skeptical Environmentalist (which is discussed on Lomborg-errors here). Lomborg´s text here demonstrates that he adheres to the ideology of market fundamentalism, according to which there is a perfect market that always ensures the best solution under the given conditions. According to this ideology, there cannot exist unutilised options for saving money. When there is a lot of evidence contradicting this view, Lomborg chooses deliberately to leave out this evidence, and instead supports his view by bringing flawed examples of anecdotal character. He thus breaks his own principle (TSE p. 7) that global problems can only be elucidated with global figures.The truth is that only those economists who have a fundamentalistic belief in the existence of perfect markets reject the no-regret options.
    No-regret options are studied by so-called bottom-up analyses (which are rejected by market fundamentalists). The latest IPCC report says about the result of bottom-up analyses that a global reduction of 5 to 7 GtCO2-equivalents per year is possible at no net cost. This corresponds to a reduction of the projected emissions by 7 to 14 % (summary for policymakers, p. 9). As an example of such free reductions, IPCC says that by 2030, about 30% of the projected greenhouse gas emissions in the building sector can be avoided with net economic benefit (summary for policymakers, p. 13).
    There have been produced a number of aggregate carbon abatement cost curves, where technologies arer ranked in order of the costs per t of carbon avoided, with the cheapest solutions put first. One example is a carbon abatement cost curve for UK, which is presented as figure 9.2. in the Stern Review (here). Another example is the cost curves produced by McKinsey & Company, e.g. a cost curve for the whole world (here). According to these curves, there are a lot of options for reducing carbon emissions and making money through the process. Worldwide, about 5 Gt of CO2 annually may be avoided at a cost less than zero by 2030.
    IPCC makes the following claim which is stated with "high agreement" and "much evidence": "Both bottom-up and top-down studies indicate that there is substantial economic potential for the mitigation of global GHG emissions over the coming decades, that could offset the projected growth of global emissions or reduce emissions below current levels." So whereas Lomborg says that there is no potential to do so, consensus among experts in the field is that there is "substantial potential".
    Lomborg is known for his talk about getting the most environment for the money. When it comes to reducing CO2 emissions, the most economic approach is probably to increase energy use efficiency (insulating houses etc.). Why, then, does Lomborg make so radical claims that there is nothing to win by increasing energy use efficiency ? Maybe he is not really concerned about getting the most environment for the money. Maybe his main concern is to make us all do what is to the greatest benefit of the large oil companies.


    Lomborg discusses what is called "the social cost of carbon", or SCC, by climate economists. SCC is usually estimated as the net present value of the impact over the next 100 years (or longer) of one additional ton of carbon emitted to the atmosphere today. This should not be confused with the average impact of climate change (the total impact divided by the otal emissions of carbon) (reference).
    A key argument in Cool It is that if a carbon tax is introduced, it must not surpass the social cost of carbon. On page 31 he cites ONE economist, Richard Tol, for his "best guess" about the costs of emitting CO2, and this guess is $2 per ton of CO2 (that is $7 per ton of carbon). Consequently, according to Lomborg, it will reduce the prosperity of the world society to no avail if carbon taxes are set higher than $7 per ton of C. However, Tol himself does not recommend an estimate of $7/tC, but rather cites an average estimate of about $16/tC, and probably recommends to use a value of $23-25/tC. Lomborg usually leans on W. Nordhaus, but also Nordhaus has higher estimates, viz. $16 - $35/tC. And a cooperative study by fifteen economists recommend to use a value of $50/tC. So Lomborg´s estimate, which he advances with great confidence, is much lower than estimates advanced by any other professional authority. Lomborg´s low level will, of course, mean that he can recommend only very modest reductions in consumption of fossil fuels.


Lomborg presents figures for global costs and benefits in the endeavour against climate change. These figures are not taken from the published literature, and they do not span the range of values that various scientists have arrived at. Instead, the figures are based on just one source, indicated as "Nordhaus 2006d", which in the bibliography reads: "RICE model. Retrieved 27-11-06." The RICE model is an integrated assessment model developed by the climate economist W. Nordhaus. Lomborg does not explain how the reference to this model is to be understood. Has he had access to computer runs performed by Nordhaus? Or has he made model runs himself? With what parameters? We do not know. So his figures on costs and benefits originate from one single source, which cannot be checked.
    The figures cannot be compared directly with figures produced by others, because different people do not assess the issue in precisely the same way. But we may make informed guesses as to what the figures should approximately be if they were to represent mainstream assertions. Let us look at the situation where the global temperature rise is reduced from the projected value of c. 5.5°F in 2100 (Lomborg says 4.5° + 0.86° = 5.48°) to only 4.5°F. The mitigation cost for this - the costs of all efforts during the century to reduce greenhouse gas emissions to obtain this reduced temperature rise - is given by Lomborg to be $15.8 trillion ( page 36 top). Is this a reliable figure?
     To evaluate this, we first consider that a temperature rise of 5.5°F by 2100 is approximately what we see in one of IPCC´s scenarios, viz. the A1B scenario. In this scenario, the CO2 concentration by 2100 is about 750 ppm, and with this concentration, there is a so-called `radiative forcing´ of about 7 W/m² (as judged from IPCC assessment report 4, working group 2, chapters 2 and 3). With this CO2 concentration and this forcing, temperatures will be bound to rise, until they eventually, many decades later, stabilise at an increase of about 7.7°F. But what we study here, is just what happens up to the year 2100. Next, we must know how much smaller is the forcing if the temperature rise should only be 4.5°F rather than about 5.4°F. Judging from available data in the IPCC chapters referred to, and setting the climate sensitivity for doubling of CO2 at 4.5 - 5.4° F, a good estimate will be that the mitigation corresponds to a reduction of the forcing to about 6 W/m². Then, in Figure 3.25 in the IPCC report, we may see a range of estimates of how much is the cost of mitigation if we keep the forcing down at 6 W/m². This Figure presents data from various model runs, performed by various research teams. When we look at the year 2100 and 6 W/m², we see that the various estimates range from about 0 to about 2 trillion dollars, and certainly not more than that. These are figures for the net present value of cumulative abatement costs, based on a discount rate of 5%. This is a rather high discount rate, which will tend to make the amounts small, but this rate is certainly not lower than that used by Lomborg, so it will not make the amounts smaller than with Lomborg´s discounting.
    So, the result of these considerations is that the mitigation costs to keep temperature rise at maximally 4.5° F by 2100, are somewhere between 0 and $2 trillion. In other words, it seems that Lomborg has greatly overestimated the costs. We cannot know the reason for this, but some parameters that would contribute to an overestimation of mitigation costs would be for instance an overoptimistic projection of future costs of fossil fuels, and underestimation of the scope for replacing fossil fuels with renewables.
    Next, we look at the benefits. How much damage is avoided by keeping the temperature at 4.5° rather than nearly 5.4° F? Damages due to future warming are of course very difficult to predict with any degree of certainty, but in order to arrive at figures comparable to Lomborg´s, we may use figures from W. Nordhaus, because all Lomborg´s figures are based on Nordhaus´ model. Nordhaus has presented estimates of global climate damages as a function of temperature increases. The curves describing the relationship between temperature and damage are presented by IPCC (4th assessment report, WG 2, Figure 20.3, taken from Nordhaus & Boyer (2000)). Here, we may read the damages at temperature increases of 4.5° F and 5.4° F respectively, and we find that the difference between them corresponds to a reduction of the damages by about one fourth. Lomborg tells us that the total damages, summed over the century, are $14.5 trillion. Parts of this damage (probably about one fourth, i.e. $7.5 trillion) is then avoided by mitigation.  However, as damages occur mainly during the latter part of the century, discounting may cause the present value of the damage to appear smaller than 1 % of the present value of the GDP, so it may probably be just an artifact of the calculation method that the damage estimate becomes less than 1 % of the discounted GDP. So the avoided damages may probably be $7.5 trillion, or probably less, with Lomborg´s rate of discount.
    So Lomborg tips the balance. In his presentation, when we keep the temperature increase by 2100 at no more than 4.5° , the costs of doing so are $15.8 trillion, and the benefits around $5 trillion. Obviously no favourable deal. But if we try to repeat his calculations, with data presented from IPCC, including data from Nordhaus, we arrive at costs of 0 to 2 trillion dollars, and benefits of about 7.5 trillion dollars or somewhat less. Apparently a fair deal.
     Will it also pay off to reduce the temperature rise even further, to 2.7° F? We may proceed in the same way as above. This temperature of 2.7° above present would probably correspond to a radiative forcing of little more than 4 W/m². In Figure 3.25 in the IPCC report referred to above, we will read for this level of mitigation that the costs will be somewhere between $1 trillion and $12 trillion, or probably even somewhat more. From Nordhaus´ curves describing the relationship between temperature and damage , we may roughly estimate that the spared damages will be more than 3 times larger than in the previous case, i.e. it might amount to about $22 trillion. So this too would pay off.

Flaws on particular pages in Lomborgs text:


Page 22: " . . . the industrial nations should reduce their overall CO2 emissions in the period from 2008 to 2012 by about 20 percent below what they would otherwise have been."
Error: Lomborg´s source is Wigley (1998), page 2286. Here, the emission cuts by 2010 are 16.4% of the so-called "natural" emissions. The nearly 20% cited by Lomborg are the reductions relative to the reduced emissions in  2010.

Page 22: " . . would be postponed just five years, from 2100 to 2105."
This figure is calculated by Lomborg, not by his source. Inspection of the source gives a result of about seven years.

Page 22 bottom: "Even its staunchest backers admit that Kyoto is only a small first step."
Let us cite Lomborg´s central reference here, Wigley (1998): ". . reductions in temperature and sea level rise under the Protocol and the extensions considered here are relatively small, but nonetheless important as a first step towards stabilizing the climate system." So, Lomborg uses Wigley to tell us that Kyoto is mostly a symbolic treaty, whereas Wigley himself says that it is an important first step. Wigley is representative for the general conception, which is that Kyoto was always meant as only a first step. This conception is not restricted to some alleged minority of staunch backers that have to "admit" that the treaty is insufficient.

Page 26: "DOESN´T CUTTING EMISSIONS SAVE MONEY ?" Flaw: Lomborg tries to persuade us that if it actually pays to reduce consumption of fossil fuels, then people will already have done so, and therefore, he claims, there are few opportunities left to cut emissions and making money. To support this claim, he brings a few examples of anecdotal character. He leaves out the vast amount of information from experts stating that large amounts of money can be saved worldwide by increasing energy efficiency, e.g. by insulating houses. For instance, the report IEA(2006): "World energy outlook", which Lomborg has read, formulates an "alternative policy scenario", with emphasis on more efficient electrical equipment, appliances and buildings. Globally, this alternative scenario implies that up to 2030, investments in equipment and buildings will be $2.4 trillion higher than in the reference scenario, but at the same time it avoids investments of $3 trillion to improve supplies of fossil fuels, and it saves $8.1 trillion by consuming less fuel. Thus, up to 2030, there can be a net saving of $8.7 trillion, without any loss of welfare, and without Japanese bureaucrats having to bundle up with sweaters. Lomborg also fails to mention that according to the latest IPCC report, which he has also studied, about 30% of the projected greenhouse gas emissions in the building sector can be avoided with net economic benefit. Or he could have referred to carbon abatement cost curves according to which the world  could save emissions of about 5Gt CO2 annually by 2030 at a net economic benefit.
    There is thus a lot of evidence contradicting Lomborg´s view that there cannot exist unutilised options for saving money. Lomborg chooses deliberately to leave out this evidence, and instead supports his view by bringing examples of anecdotal character. He thus breaks his own principle (TSE p. 7) that global problems can only be elucidated with global figures.

Page 26: ". . . insulating your house will reduce the energy bill . . . ". Comment: The whole text is an attempt to tell us that there is little money to save by insulating houses better. Lomborg has found one American paper that says that although money was saved by insulating houses, the savings were not as large as is often promised. By citing only that, he neglects the vast amount of data from many countries telling that there is a lot to be obtained from better insulation. He cites the book "Heat" by George Monbiot, which has a fine chapter on what can be obtained by insulating houses in Britain. But instead of citing that chapter, which is relevant in the context, he cites an anecdote from the preface which serves to warn us that not all promises of energy saving are realistic. Actually, Monbiot writes that 10% of British homes have no insulation at all - giving an obvious scope for improvements. Monbiot writes (p. 65): "Given that injecting mineral fibres between the bricks is so cheap that it pays for itself within two to five years, the 65 percent of homeowners who choose not to use it must either be so poor they have no capital to spend, so poorly informed that they have never head of the process, aware that someone else (the tenant) is picking up the heating bill, or perversely attached to burning money."  

Page 26 bottom: ". . . a good deal. However, an independent analysis. .  .". Comment: Lomborg cites an anecdote from the preface of Monbiot´s book "Heat". In that book, the anecdote serves to warn us that not all promises of energy saving are realistic. Lomborg places the source reference in such a way that you believe that Monbiot believes in the promised saving of 50%, whereas an "independent analysis" finds otherwise. Actually, it is Monbiot who has the role of the "independent analysis". Monbiot also tells that if a wind turbine should actually produce 50% of the household electricity, it would produce a thrust so strong that it would tear the house apart.

Page 28: "The question then is, Which ton should we cut first?". Comment: This is a misleading question. Lomborg tries to give the impression that cutting carbon means lower welfare. For instance, you will have to let the children walk to school, or stand in the hot shower for a shorter time. By presenting only such possibilities for cutting carbon (and by talking about Japanese bureaucrats wearing sweaters in unheated offices), Lomborg tries to make climate mitigation unpopular. However, there are lots and lots of possibilities to reduce carbon emissions without a loss of welfare, for instance by letting alternative energy sources cover a larger part of our energy budget, especially where this is practical.

Page 29: "A one-dollar tax on CO2 will actually lead to an overall drop in emissions of a bit more than 2 percent.". Flaw: Lomborg refers to these figures as if ithey represents facts. Actually, however, the source is just a reference to the DICE model produced by W. Nordhaus. There is no indication at all as to who performed the particular model run, and what were the conditions and parameters fed to the model. It seems that this may be a model run performed by Lomborg himself, in which case there is no control or check whatsoever on the results. Being completely undocumented, the figures are worthless.

Page 29 bottom: "In a global macroeconomic model, the total present-day cost . . . ". Flaw: The whole text here is fully undocumented. We are not told what model, and we are not told what parameters were used, e.g. what was the discount rate (which is crucial for the results). A good guess is that we are dealing with a run of one of Nordhaus´ models, performed by Lomborg himself. You can then only have trust in the figures presented if you trust the way that Lomborg treats figures in general. There is very little basis for having such a trust.

Page 30 bottom: "First, the really scary, high estimates typically have been neither subjected to peer review nor published."
Flaw:  This is not a fair representation of the data. The "really scary, high estimates" are only 2 out of the 103. These two assume no economic growth. But there were many high estimates both among the peer reviewed studies, and those not peer reviewed, including some very high estimates in peer-reviewed studies by Tol himself. Only, there were slightly fewer of such high estimates among the peer reviewed studies.

Page 31 top: ". . . the cost is very unlikely to be higher than fourteen dollars per ton of CO2 and likely to be much smaller."
Flaw:  This is a slightly distorted quote. The original (p. 2073) says: " . . . climate change impacts may be very uncertain but [it] is unlikely that the marginal damage costs of carbon dioxide emissions exceed $50/t C and are likely to be susbstantially smaller than that." (note: $50/t C = $14/t CO2). We see that Lomborg leaves out the sentence about uncertainty, replaces `marginal costs´ with `costs´, `likely´ with `very unlikely´ and `substantially smaller´ with `much smaller´. Actually, Figure 3 in Tol´s paper shows that if we look at peer-reviewed studies, the probability of a value in excess of $50/t C is twenty percent. This is not what is usually understood by `very unlikely´.

Page 31 top: " . . . true researchers invariably are this way . . . ". Comment: Notice how Lomborg takes care to paint a positive image of the researcher that substantiates Lomborg´s claim. If the researcher had been in opposition to Lomborg, he would probably have been called `vague´ (the term used p. 133 for a person opposing Lomborg).

Page 31 top: " . .  but gave a best estimate of two dollars per ton." (note: $2/tCO2 = $7.3/tC). Error: The figure of $7/tC appears in Table 3 in Tol (2005). It is the very lowest figure one can find in that table. If instead of the median we take the mean, the figure grows to $16/tC, and this is for a rate of discount of 4-5 %, which is very high (3 % pure rate of time preference + economic growth added). With a somewhat lower discount rate, the mean rises to $51/tC, and with a much lower disocunt rate, it rises to $261/tC. The average for all peer-reviewed studies, with differing discount rates, is $50/tC. It must be stressed that there is no "correct" discount rate, and the choice of rate is subjective. In addition, one may or may not  use equity weighting (without equity weighting, you disregard that lsoing one dollar is much worse for a poor man in Africa than for a rich man in Europe; including equity weighting will raise the estimate further (and increase the uncertainty)). So the "best estimate" referred to by Lomborg is a subjective choice, based on personal preferences, not an attempt at estimating something that exists in the real world. The explanation given by Lomborg in his note is probably flawed; a good gues is that Lomborg, not Tol, has demanded to have the median, not the mean, and to have the estimate for a high discount rate.
It is interesting to read a text by Michael Pawlyn (link): "I phoned Lomborg´s main source for the $2 figure, Professor Richard Tol. He told me that `The number 2 ($2) comes about when you ignore all the uncertainties and you just go for a high discount rate, but if you start including the fact that things could go dramatically wrong then you would come up with a much higher number.´ Professor Tol´s guess for the right figure is that it should be around $23-25 per tonne and he has assessed the probability of it being much higher."
A level of $23-25 per tonne is clearly above the average of the estimates presented by Tol. The reason why he, in the phone, referred to these figures, may probably be that above this level, the probability curve falls off rather steeply. So using these figures, you are relatively sure that you do not underestimate the costs of carbon emissions. Another reason could be that the lower estimates are for studies that do not include uncertain climate effects; if uncertain climate effects are considered, the costs will be higher.

Page 31 top: " This means that the damage we will cause by putting out one more ton of CO2 is likely two dollars and very unlikely to be higher than fourteen dollars" (note: $14/tCO2 = $51/tC). Error: Here, Lomborg makes the shortcut that the estimate presented by one single economist with `a gun put to his head´ is an authoritative figure. One should rather consult a joint paper by 15 economists, including Tol, namely Downing et al. 2005. Their paper is based on four different approaches, one of which is the metastudy made by Tol. This metastudy, however, is based nearly exclusively on studies that do not include uncertain climate effects - for instance studies may consider projected changes in temperature, which are relatively certain, but may omit the changes in regional precipitation, which may be economically very important, but which can be projected with much less certainty. They demonstrate that the estimated costs depend heavily on how much global temperatures will rise with a given greenhouse gas emission. They also demonstrate that the choice of discount rate, and the choice of equity weighting, has very large influence on the estimates, which therefore in any case remain subjective. They present data from runs with two different integrated assessment models in order to better include the more uncertain effects. The average of many model runs give £38 and £46, respectively, for the two models, but with considerable probability that the costs could be much lower or much higher. They conclude that a lower benchmark of  £35/tC is reasonable for a global decision context committed to reducing the threat of dangerous climate change and includes a modest level of aversion to extreme risks, relatively low discount rates and equity weighting.
    The benchmark of  £35/tC corresponds roughly to $50/tC, or $17/t CO2. So this is the value for the social cost of carbon recommended by these fifteen economists. This benchmark means that many more technologies are profitable than when you use the estimate of $7/tC advanced by Lomborg.
    Even one of Lomborg´s preferred sources - Nordhaus (2006e) - which he uses in his next section, has a different estimate than Lomborg. Nordhaus writes that  the latest run of the RICE model suggests a carbon price of $16/tC in 2005, rising rapidly over time. Even this low estimate, which is stated to omit certain important aspects of climate change, is higher than Lomborg´s own preferred figure. And in a later article by Nordhaus (2007, J of Economic literature 45(3):698), he writes:
"Run 1 calculates the optimal carbon price tin 2015 to be $35 per ton C, rising over time to $85 in 2050 and to $206 in 2100 . . . the social cost of carbon without emissions restraints in 2015 is also $35 per ton C." This paper is approximately the same as Lomborg´s source "Nordhaus (2006d)". So Lomborg must have seen these figures.
To leave out these estimates, even though they are produced by the same model that Lomborg relies upon, and focus so strongly on his own extremely weakly founded figure, is to mislead deliberately.

Page 31: " . . . as proposed in one radical report . . . ". Comment: It is unnecessarily derogative to call the Stern Review, commissioned by the British government and headed by a respected climate economist, a `radical report´.

Page 31: "If we tax it at $85 . . . we lose up to $83 of social benefits." Flaw:  This is not a fair representation of the figures. Of course the authors of the Stern Review do not suggest to put a tax on CO2 emissions which is higher than the social costs of the emissions. Lomborg fails to mention how the Stern Review treats this matter. It refers to Tol´s metastudy of estimates of the social costs of carbon, and it refers to the paper by Downing et al. (2005), which I also refer to above. But then it goes further and makes a new estimate, based on the PAGE integrated assessment model, and this new estimate lies around $85/tCO2 .(= $310/tC).  It is in the upper range of the estimates dealt with by Tol, but not in the extreme upper end. For studies using a low discount rate - as the Stern Review does - Tol has an average estimate of $261/tC, i.e. only slightly lower than the estimate in the Stern Review.  So this is hardly a `radical report´. The Review explains its high estimate thus: "It should be remembered that this model is different from its predecessors, in that it incorporates both explicit modeling of the role of risk, using standard approaches to the economics of risk, and makes some allowance for catastrophe risk and non-market costs, albeit in an oversimplified way. In our view, these are very imprtant aspects of the social cost of carbon, which should be included in its calculation even though they are very difficult to assess."
Instead of admitting that others put the costs of carbon emissions higher than he does himself, and have valid reasons to do so, Lomborg distorts the meaning of the Stern estimate and gives his audience the impression that the authors of the Stern Review (some of which, unlike Lomborg, are economists), advocate for stupid losses of benefits.

Page 31 bottom: " . . . twenty-three dollars per ton of CO2 - between two and eleven times higher than the likely cost of climate change." (in the notes: this is equivalent to £45/tC). Flaw:  The UK estimate is very close to later mainstream estimates in Downing et al.  (2005), which are £38 and £46, respectively. Lomborg´s much lower estimate, on the other hand, is very weakly founded. It is a regrettably typical Lomborg attitude to state with great confidence that his own extremely weakly founded figure is the right one, and that the mainstream estimate is `too high´.

Page 32: " . . . and according to the IPCC they all produce more or less the same substantial results. "  Flaw:  Lomborg´s source is the IPCC report from 1996. There have been two reports since then, both of which Lomborg has studied. If there had been any support for his statement in the later reports, he would of course have cited them, not the old one. When the old report was written in the years before 1996, there had been only very few years of experince with the DICE model. Since then, many more integrated assessment reports have been constructed, and they often give results that differ widely from Lomborg´s preferred figures (depending on what conditions are put into the models). Typically, those models which give the kind of figures that Lomborg favours, are those that more or less disregard or underestimate the possibilities of replacing fossil fuels with renewables.

Page 35: " Cutting more carbon is likely to cost ever more while doing ever less extra good." Flaw:  This is based on Lomborg´s own calculations and figures which he presents on page 36, but with no source. An independent check on the costs and benefits of the 4.5°F scenario and the 2.7°F scenario (see the general text above) yields very different figures, which indicate that it does pay to reduce temperature rise even to the low threshold of 2.7° F. So the impression given here is not in accordance with mainstream estimates.

Page 36 top: " . . . but at a rather high cost of $15.8 trillion. "  Flaw:  The figure is based on just one source, indicated as "Nordhaus 2006d", which in the bibliography reads: "RICE model. Retrieved 27-11-06." The RICE model is an integrated assessment model developed by the climate economist W. Nordhaus. Lomborg does not explain if he has had access to computer runs performed by Nordhaus, or if he has made model runs himself, and he does not state any crucial parameters applied for the runs. So his figures on costs and benefits originate from one single source, which cannot be checked.
    However, as explained above under the heading "Estimates of global costs and benefits", one may arrive at estimates for the figures given by Lomborg. The costs of mitigation may be taken from the latest IPCC report, which gives a range of estimates from model runs performed by various research groups. Let us as an example consider the scenario where the temperature rise is kept at 4.5° F by 2100, and study the data for the level of mitigation that is likely to result in this temperature rise. The result is then that with 5 % discounting, the summed mitigation costs over the 21st century lie between 0 and $2 trillion. This is much lower than Lomborg´s figure of $15.8 trillion.

Page 36 top: " Thus, the cost of global warming can be estimated at $14.5 trillion. "  Comment:  This is a rather low figure - only about 0.5% of the value of global GDP discounted over a hundred years with Lomborg´s rate of discount. The average mainstream estimate is a total damage cost of about 1% of global GDP. Lomborg writes on page 196: "In the peer-reviewed studies, damages run at about 1 percent of GDP . . . ". This would correspond to about $30 trillion. However, as damages occur mainly during the latter part of the century, discounting may cause the present value of the damage to appear smaller than 1 % of the present value of the GDP, so it may probably be just an artifact of the calculation method that the damage estimate becomes less than 1 % of the discounted GDP.  

Page 36: "One central model shows it is possible to achieve . . . cost of $84 trillion." Error: The model referred to is Nordhaus´ RICE model, but the origin of the figure of $84 trillion is obscure; maybe it has been calculated by Lomborg himself. As explained in the general text above, the latest IPCC report estimates that the mitigation costs for this scenario will be somewhere between $1 trillion and $12 trillion. Compared to this, Lomborg´s $84 trillion is way off the mark. It gives a completely misleading impression to state that this is the output of a `central model´, which would indicate that the output were a central estimate.

Page 36 top: " For every dollar spent, it will do thirteen cents´ worth of good. "  Flaw:  The postulated benefit/cost ratio of 0.13 is flawed not only because the costs are estimated at a very high figure, but also because the benefits are estimated at a low figure (a fraction of $14.5 trillion, rather than a fraction of $30 trillion, see above).

Page 37: " In a review from 2006. . . "  Flaw: The review referred to is the Stern Review. At the middle of page 31 the Stern Review is called a `radical report´, which according to the text advances so stupid suggestions that we should of course not take it seriously. Here, on the other hand, Lomborg presents the Stern Review as an authoritative source which we should rely upon. So we should rely on the source when it supports Lomborg´s agenda, and not take it seriously when it goes against Lomborg´s agenda.

Page 37: "These studies recommend that . . . This is such a robust result because . . "  Flaw: The quote from the Stern Review is taken out of context. Stern refers to three studies, by Nordhaus & Boyer (1999), by Tol (1997) and by Manne et al. (1995). With reference to these three as an example, the text reads: "In some cases, the models have been used to estimate the `optimal´ amount of mitigation that optimises benefits less costs. These studies recommend that . . ". The text then goes on to say: "However, the optimal amount of mitigation may in fact be greater than these studies have suggested. Above all, they carry out cost-benefit analysis appropriate for the appraisal of small projects, but we have argued in Chapter 2 that this method is not suitable for the appraisal of global climate change policy, because of the very large uncertainties faced." Lomborg´s phrase `such a robust result´ is hardly compatible with the wording in the original text: `very large uncertainties´.