Arguments
Climate dollars and sense – preventing global warming is the cheap option
Posted on 22 April 2014 by dana1981
The IPCC has now released all three of the reports that comprise its 2014 Fifth Assessment of climate science. The first report tackled the physical changes in the global climate, while the second addressed climate impacts and adaptation, and the third looked at climate change mitigation. Ironically, after the second report was published, many media outlets argued that the IPCC was shifting its focus from global warming prevention to adaptation, seemingly unaware that its report on mitigation was scheduled to be published just a few weeks later.
Other media outlets have incorrectly argued that the IPCC reports conclude it's cheaper to adapt than avoid climate change. This error stems from the fact that the second report says about the costs of climate damages,
"the incomplete estimates of global annual economic losses for additional temperature increases of ~2°C are between 0.2 and 2.0% of income ... Losses are more likely than not to be greater, rather than smaller, than this range ... Losses accelerate with greater warming, but few quantitative estimates have been completed for additional warming around 3°C or above."
The third report then said about the costs of avoiding global warming,
"mitigation scenarios that reach atmospheric concentrations of about 450ppm CO2eq by 2100 entail losses in global consumption—not including benefits of reduced climate change as well as cobenefits and adverse side?effects of mitigation ... [that] correspond to an annualized reduction of consumption growth by 0.04 to 0.14 (median: 0.06) percentage points over the century relative to annualized consumption growth in the baseline that is between 1.6% and 3% per year."
The challenge is that these two numbers aren't directly comparable. One deals with annual global economic losses, while the other is expressed as a slightly slowed global consumption growth. While these numbers can be put in terms of their net impact on economic growth, the next problem is that the first is not a proper estimate of the costs of climate damages. The IPCC was only able to estimate the costs of climate damages for another 2°C warming, but limiting global warming to another 2°C will require substantial mitigation efforts.
Thus this estimate only tells us the costs of global warming in a scenario where we also act to significantly reduce greenhouse gas emissions. As the second report notes, economists can't even accurately estimate the costs of climate damages in a business-as-usual scenario with global warming well above an additional 3°C. So how do we determine the economically optimal path?
Sorting Out the Numbers with Chris Hope
To answer this question, I spoke with Cambridge climate economist Chris Hope, who told me that if the goal is to figure out the economically optimal amount of global warming mitigation, the IPCC reports "don't take us far down this road." To do this comparison properly, the benefits of reduced climate damages and the costs of reduced greenhouse gas emissions need to be compared in terms of "net present value." That's the sort of estimate Integrated Assessment Models like Hope's PAGE were set up to make.
According to Hope's model, the economically optimal peak atmospheric carbon dioxide concentration is around 500 ppm, with a peak global surface warming of about 3°C above pre-industrial temperatures (about 2°C warmer than present). In his book The Climate Casino, Yale economist William Nordhaus notes that he has arrived at a similar conclusion in his modeling research.
To limit global warming to that level would require major efforts to reduce greenhouse gas emissions, but as the IPCC report on mitigation noted, that would only slow the global economic growth rate from about 2.3% per year to about 2.24% per year. According to these economic models, this slowed economic growth rate would be more than offset by the savings from avoiding climate damages above 3°C global warming.
Although the IPCC didn't make this comparison, these economic modeling results are consistent with its reports. As shown in the quote above, the second report was only able to estimate the costs of climate damages for an additional 2°C of global warming, and noted that beyond that point, the costs accelerate to a point where they become very difficult to estimate. Nordhaus has similarly noted,
"In reality, estimates of damage functions are virtually non-existent for temperature increases above 3°C."
Australian and Turkish Fruit Salad
Author and analyst Bjorn Lomborg of the Copenhagen Consensus Center has been the most prominent voice in incorrectly claiming the IPCC concluded that climate adaptation would be cheaper than mitigation. For example, he was interviewed in Rupert Murdoch's The Australian, and authored a piece in the Turkish Today's Zaman.
Both pieces are lemons for the same reasons. Lomborg argued,
"If we don't do anything, the damages caused by climate change will cost less than 2 per cent of GDP in about 2070. Yet the cost of doing something will likely be higher than 6 per cent of GDP, according to the IPCC report"
This compares the annual global economic losses figure for 2°C additional warming in the second report with the slowed global consumption growth figures to limit the warming to another 1°C in the third report. The problem is that this is an apples and oranges comparison. The former tells us the cost of climate damages in a scenario where we also take significant steps to slow global warming. It's not the cost of adaptation if we continue with business as usual, which would result in another 4°C warming by 2100 and incalculable damage costs.
Without the modeling tools used by economists like Hope and Nordhaus, these figures can't properly be put into an apples to apples comparison. Both the costs of mitigating and the costs of adapting to climate damages must be taken into account. I discussed this point with Lomborg and he agreed,
"I agree that the right way to look at the climate issue is to run integrated models and finding where the costs and the benefits are equal (so we don't underinvest in climate but don't over-invest either) ... However, the UN Climate Panel actively decided in 1998 to *not* do cost-benefit of climate."
So Lomborg and Hope agree that the IPCC reports don't allow for a simple comparison between the costs of global warming prevention and adaptation. Lomborg used the two figures discussed above to make the only comparison possible from the reports, but this looks to be an incorrect comparison, inconsistent with the results from economic models.
Lomborg also cherry picked the year 2070 to make his economic comparison between the costs of adaptation and mitigation. Why 2070? By that point, in a business-as-usual scenario the planet probably won't have warmed much more than 2°C compared to current temperatures. The problem with this cherry pick is that the world won't end in 2070; in fact, most of today's children will still be alive in 2070. If we continue on that business-as-usual path, global warming will continue to accelerate after 2070, past the point where economists can't even accurately estimate its accelerating costs. That's bananas.
IPCC AR5 projected global average surface temperature changes in a business as usual scenario (RCP8.5; red) and low emissions scenario (RCP2.6; blue).
Another problem in this argument is that as shown in the second quote above, the IPCC estimates of the cost of reducing greenhouse gas emissions are "not including benefits of reduced climate change as well as cobenefits and adverse side?effects of mitigation." For example, the cleaner air and water, and associated health benefits that come with transitioning away from dirty high-carbon energy sources save money that the IPCC doesn't take into account. So the costs of avoiding global warming would in reality likely be even less than the estimated 0.06% per year slowing in the rate at which the global economy continues to grow.
Meanwhile, the IPCC noted that the costs of climate damages for just another 2°C warming "are more likely than not to be greater, rather than smaller" than its estimates. And if we don't take serious steps to reduce greenhouse gas emissions, we'll blow past 2°C warming into uncharted economic damage territory.
Avoiding Global Warming is Cheaper than Adapting
The bottom line is that economists can't even accurately estimate how much climate damages will cost if we fail to take serious steps to slow global warming. On the other hand, taking those steps can have a negligible impact on global economic growth. The IPCC report also makes the point that the longer we wait to reduce our emissions, the more expensive it will become. In determining that mitigating global warming is affordable, the IPCC used the following scenarios.
"Scenarios in which all countries of the world begin mitigation immediately, there is a single global carbon price, and all key technologies are available, have been used as a cost?effective benchmark for estimating macroeconomic mitigation costs"
This post has been incorporated into the rebuttal to the myth - adapting to global warming is cheaper than preventing it
I've had this very discussion on a few of the pseudoskeptic blogs (such as here), pointing out that mitigation is far less expensive than adaptation to climate change under a Business as Usual (BAU) economic strategy. And that if the pseudoskeptic is concerned about economic consequences, BAU is probably the worst choice possible.
The usual responses are sputtering (often accompanied by links to something by Lomborg, whose work has issues such as discussed above) or changing the subject. Sigh.
Dana – Lucid as ever. That’s quite a fruit salad you’ve got there, apples, oranges and bananas.
And talking of bananas, I have never been able to understand Lomborgs’ persistent preoccupation with the strange notion that its not only OK to do nothing when it comes to mitigation but desirable. Rather like his misplaced advocacy for global warming and claim that increased CO2 emissions are beneficial for us - at least until 2070.
Clearly, both contentions are nonsense, unsupported by climate science and unsupportable by economics, or for that matter common sense.
Riduna@2,
As is the case in long term policies, the sense of Lomborgs' advocacy depends on the point of view. Lomborgs takes a point of view that only the economy until 2070 does matter anything beyond is not worth considering. Considering that most of us will not be alive at the date in question (I certaqinly won't), the issue becomes more of inter-generation ethics rathrer than economics. It was nicely (although rather vulgarily) described in Australia's Coal Policy to which Lomborgs undoubtedly subscribes.
OP: The challenge is that these two numbers aren't directly comparable. One deals with annual global economic losses, while the other is expressed as a slightly slowed global consumption growth.
Even more basically: The WG3 estimate is essentially about the mitigation cost of limiting warming to roughly 2 °C. Shouldn't it be compared to the losses from the warming that mitigation avoids, then? That is, shouldn't it be compared to the losses that would be incurred due to warming beyond 2 °C, and not at about 2 °C, which is what WG2 estimates?
Christian Moe @4, strictly it should be compared to the costs of warming at BAU (RCP 6.0 or 8.5) minus the costs of warming to 1.4 C. 1.4 C because temperatures are referenced to 2000 levels (strictly, 1986-2005) rather than to preindustrial levels (taken as 1850-1900) in the report, so that the 2 C target relative to pre-industrial levels often quoted becomes a 1.4 C target relative to 2000. That then becomes complicated because the costs sited are strictly for 2.5 C relative to 2000. That compares to projected temperatures of 2.2 +/- 0.5 C (1 SD range) in 2080-2100 for RCP 6.0; and 3.7 +/- 0.7 C for RCP 8.5.
So, for RCP 6.0, estimated costs are slightly less than those stated in the report. For RCP 8.5, they are greater than the estimate by an amount larger than the estimated costs are greater than the costs at the 2 C above pre-industrial mitigation target - so that overall the costs cited are an underestimate of the cost of not mitigating. That is because costs increase more than linearly with increased temperatures. Worse, the costs estimates are more likely to underestimate costs than to over estimate them (according to the IPCC), which I consider a considerable understatement. Further, temperature projection risks are greater on the high side than on the low side. Therefore the expected value of the risks is substantially understated in any event.
So, you are correct in your intuition, but because of the poor composition of the reports on these relevant points, direct comparison is simply impossible.
There is discussion elsewhere (http://tinyurl.com/oxthlrc) that a significant comment in the IPCC report was not included by Mr Nuccitelli. The section in question gives assessments of median values of annual economic loss in 2030 0f 1.7%, 3.4% in 2050 and 4.8% in 2100. In his comments Mr Nuccitelli noted that the IPCC report did not mention annual economic losses when clearly it does. Can Mr Nuccitelli explain why the relevant data that disagrees with his comment was not mentioned? Is he being unfairly maligned? My comment is not off topic, is not political, is not ad hominem and gives the relevant reference so seems to meet the standards set by Skeptical Science.
Poster @7 - the criticism is ridiculous because I had included the figures in question in the Lomborg quote immediately preceeding the second Lomborg quote in question.
That said, I didn't explain the apples and oranges point well - Christian @4 and Tom @5 have hit on the key point, and I've revised the post to clarify it and also included it in the myth rebuttal now linked at the bottom of the post. That being, limiting global warming to a further 2°C is not 'no action', it requires substantial mitigation efforts. Lomborg tries to sweep this under the rug by only looking at costs in 2070, ignoring the accelerating costs thereafter if we 'do nothing'. Not to mention that the 2% is likely an underestimate and the 6% likely an overestimate, as the IPCC explicitly stated.
This is why you need to go beyond the numbers in the IPCC reports and look at models like PAGE to determine the economically optimal path, and they conclude it's around 450-500 ppm CO2, which is a scenario that requires substantial mitigation.
Poster @7, the table Lomborg refers to is this Table SPM.2:
You will notice (and I suspect, already knew) that Dana quoted the compounding cost for restricting CO2eq rises to 450 ppmv, for which you quote the values at 2030, 2050, and 2100. For comparison, the compounding cost at 2030 is 1.2%, at 2050 it is 2.4%, and in 2100 it is 5.5%. That is, it underestimates the cost for the earlier years, but over estimates the cost in the later periods. In all years, however, it is not statistically significantly different from the specific values.
Your charge is that Dana incorrectly claims that WG3 does not mention annual losses, but Dana is correct. It does not give the annual values for any other than the three years stated. That only allows a direct comparison if the year in which the global temperature increases to 2.5 C above recent values is one of those three years. In order to make a direct comparison, the IPCC would need to report the costs for each level of temperature increase, correlate that to the specific years to give a cost in each year, and provide a cost in each year for mitigation, and then integrate the two. Indeed, done properly they would repeat that several thousand times in a monte carlo procedure allowing for potential error in estimates of temperature increase, cost at a given temperture and costs of mitigation to generate expected costs of the options - a procedure which would show higher expected costs from global warming than simply comparing mean values.
Finally, the cost due to an increased temperature is simply a raw cost. That cost, "the equivalent of less than one year of recession" according to the Lomborg article you linked to, can be expected to have its own impact on economic growth - but does not include any such impacts. Logically it cannot include such impacts as it is a cost at a particular temperature without refference to the year in which it occurs, or temperature trajectory over time to that year. Consequently the actual cost of not mitigating will be that 0.2-2% plus any impacts on economic growth from from the raw impacts. As, by Lomborg's own admission, the impact at 2.5 C is comparable to a years recession, and that impact will be felt every year, that means with BAU by the end of this century the impact of global warming will be equivalent to being in permanent recession - ie, a complete stagnation of economic growth.
As Lomborg's own words show, he is simply playing the old game of comparing incomparables, and hiding the actual costs of global warming.
TonyW @6, in fact costs of mitigation will be greater if it is unexpectedly difficult to integrate renewable energy into the future energy equation. The middle section of table SPM.2 (see my post @9) deals with that issue, and shows that the cost of "limitted solar/wind" penetration in the market will increase the cost by 6%. Far more concerning is the 138% cost if Carbon Capture and Storage proves untenable (as is widely believe by many at SkS). However, even with that increase, fully calculated the cost of not mitigating will exceed the cost of mitigating to keep temperatures below 2-3 C. Further, there is a real risk of catastrophic effects from warming that raise the expected cost of unmitigated warming well above any reasonable estimate of the cost of mitigation.
Thanks Dana for your rapidreply, it is appreciated. Tom Curtis@9, no I didn't already know and furthermore I'm not charging Mr Nuccitelli with anything at all. Others have done that and it is that I was asking about. I very much appreciate that Mr Nuccitelli had the courtesy to answer my questions both civilly and rapidly. His attitude is markedly and refreshingly different from the attitudes of others in the blogosphere dealing with Climate Change.
Poster @11, I based my comment on the link which you provided in support of your claim that "There is discussion elsewhere ... that a significant comment in the IPCC report was not included by Mr Nuccitelli".
As it happens, the article to which you linked did not discuss Dana's article (not even in comments). Nor did it mention the figures you quoted, instead stating misleadingly that "..strong climate policies would be more expensive than claimed as well – costing upwards of 4% of GDP in 2030, 6% in 2050, and 11% by 2100". The first of those figures lies outside the 95% confidence interval of the costs of mitigation. All are the rounded upper bounds of the 95% confidence interval so that Lomborg in effect argues that the IPCC states the costs will be equal to or higher than (upwards of) the upper bound of the 95% confidence interval of the IPCC stated values. Given that he quotes the 84% (+/- 1 stdev) range for costs, this biased presentation looks like straight forward, and intentional deception. (Apparently, however, something you do not, or at least did not have a problem with, while having a problem with Dana's correct figure.)
In any event, as you did not draw the figures you quoted from your cited source, it was a reasonable assumption that you drew them from your sources cited source, ie, table SPM.2, or the text which states:
So, to summarize, based on the information you gave it was a reasonable assumption that your knowledge of the figures came directly from the IPCC report, and hence came with a direct statement of the near equivalence of Dana's figure.