by Steve Keen, Steve Keen’s Debtwatch
Enumeration: It’s what you don’t count that counts
This post continues my series on the appallingly bad work on climate change done by mainstream economists, especially 2018 “Nobel Prize in Economics” winner William Nordhaus. The previous post analysed the absurd assumption they make that today’s GDP: Temperature relationships across the globe can be used to predict the impact of increasing global temperatures. This post looks at the so-called “Enumeration” approach they used to concoct a different set of estimates of the global warming:change in GDP relationship.
Please share this article – Go to very top of page, right hand side, for social media buttons.
I honestly don’t know which of these two methods is worse.
Editor’s note: This article is adapted from a post on Patreon. Following his retirement from university life, Steve Keen is conducting a Patreon campaign to enlist financial support that will enable him to continue his unique research outside of the academic world heretofore his home. Econintersect strongly endorses this campaign.
The most frequently used technique to generate the numbers – I refuse to call them “data” – to which economists fitted their “damage functions” was “enumeration”: ten studies out of the 18 shown in Figure 6 on page 7 use this method. Ostensibly, it’s a reasonable approach, especially since – unlike the “statistical approach”, where the numbers were invented by economists entirely independently of the research by climate scientists – it appears to put into monetary form research into damage estimates by climate scientists. In Tol’s words:
In this approach, estimates of the “physical effects” of climate change are obtained one by one from natural science papers, which in turn may be based on some combination of climate models, impact models, and laboratory experiments. The physical impacts must then each be given a price and added up. For agricultural products, an example of a traded good or service, agronomy papers are used to predict the effect of climate on crop yield, and then market prices or economic models are used to value the change in output. (Tol 2009, pp. 31-32)
However, a little digging finds that this approach suffers from the same fundamental flaw as the “statistical” approach: of regarding today’s GDP and temperature relationship as indicative of what will happen with climate change. The ancestry of this belief can, as happens so often, be traced back to William Nordhaus, in a 1991 paper and 1994 book (Nordhaus 1991; Nordhaus 1994). The section “Estimating the Damages from Greenhouse Warming” in the 1991 paper starts with the caveat that:
First, it must be recognised that human societies thrive in a wide variety of climatic zones. For the bulk of economic activity, non-climate variables like labour skills, access to markets, or technology swamp climatic considerations in determining economic efficiency. (Nordhaus 1991, p. 930)
The first sentence draws the same false equivalence between variations in climate across the globe today, and the alteration of the entire global climate, which afflicted the “statistical method”. But the key clue to why this method is delusional is the second: “For the bulk of economic activity, non-climate variables … swamp climatic considerations”. What does he mean by “the bulk of economic activity”? He means… anything that is done indoors.
You think I’m joking? Here is Nordhaus’s explanation of how he selected industries that he believed would be affected by climate change, and those that would not:
Table 5 shows a sectoral breakdown of United States national income, where the economy is subdivided by the sectoral sensitivity to greenhouse warming. The most sensitive sectors are likely to be those, such as agriculture and forestry, in which output depends in a significant way upon climatic variables. At the other extreme are activities, such as cardiovascular surgery or microprocessor fabrication in ‘clean rooms’, which are undertaken in carefully controlled environments that will not be directly affected by climate change. Our estimate is that approximately 3% of United States national output is produced in highly sensitive sectors, another 10% in moderately sensitive sectors, and about 87% in sectors that are negligibly affected by climate change. (Nordhaus 1991, p. 930. Emphasis added)
On the basis of this assumption that work done in controlled environments today (i.e, indoors) would be unaffected by climate change, Nordhaus chose to virtually exclude from consideration all manufacturing, all mining(!), all non-water-based transportation, all finance and real estate (apart from land rents), all retail trade, and all government services.
“false equivalence between variations in climate across the globe today, and the alteration of the entire global climate”
He thus excluded, by assumption, 87% of the US economy from facing any consequences from climate change. Construction and energy itself would only face “Moderate potential impact”. Only “Farms, Forestry, fisheries” were going to be “Potentially severely impacted” (see Figure 16).
Figure 16: Nordhaus’s self-constructed table of industries and their exposure to climate change
Having exempted 87% of the US economy from any effects from Global Warming, Nordhaus then went to task on the remaining 13%. Throughout, his perspective was of “swings and roundabouts”: yes, there were some potential costs from increasing the global average temperature by 3°C by doubling the level of CO2 in the atmosphere, but there were potential benefits as well:
Greenhouse warming will increase the demand for space cooling and decrease the demand for space heating, with but a small net impact on the energy sector. The forest products industry may benefit from CO2 fertilisation. Water systems (such as runoff in rivers or the length of ice-free periods) may be significantly affected, but the costs are likely to be determined more by the rate of climate change than the new equilibrium climate. Construction in temperate climates will be favourably affected because of a longer period of warm weather. For recreation and water transportation, the outlook is mixed depending upon the initial climate. Cold regions may gain while hot regions may lose; investments in water skiing will appreciate while those in snow skiing will depreciate. But for the bulk of the economy – manufacturing, mining, utilities, finance, trade, and most service industries – it is difficult to find major direct impacts of the projected climate changes over the next 50 to 75 years. (Nordhaus 1991, pp. 931-932. Emphasis added)
In the end, he quantified the estimated damages from a 3°C increase on only three aspects of the US economy: agriculture (with a range estimate), Energy, with two point estimates (one for electricity demand and the other for heating), and sea level rise damage, with three components. In sum, he estimated the impact of a global temperature increase of 3°C, would be to reduce the US GDP in 2050 by 0.26%, relative to what it would have been in the total absence of global warming.
“He thus excluded, by assumption, 87% of the US economy”
Farming, which was in his “Potentially severely impacted” category in his Table 5 (Figure 16 here), came out as a line-ball call, with anything from $10.6 billion in damages to maybe a $9.7 billion gain. Forestry, also “Potentially severely impacted”, ended up line-ball. “Moderate potential impact” for construction” ended up as an overall unquantified positive (see Figure 17). The only numbers with a definitive sign and magnitude were for Energy and sea level rise.
How did he arrive at these numbers? The source is given at the bottom of his Table 6: the 1988 draft of the 1989 EPA (Environmental Protection Agency) report to the United States Congress, The Potential Effects of Global Climate Change on the United States
Figure 17: Nordhaus’s personal estimates in 1991 of damages from a doubling of CO2 (Nordhaus 1991, p. 932)
If you trusted Nordhaus, it would appear that he had shown that climate change would have a trivial impact on the economy. But of course, for good reason, I don’t trust him. So I went searching for this EPA Report , and thanks to the Internet, I found it.
Nordhaus’s estimate for Farms accurately reproduces the range given in this document (Table 6-4, p. 104 of the final report), but his other numbers are significant and puzzling transformations of its estimates.
The most glaring discrepancy is over the estimated impact of Electricity Demand. Nordhaus’s Table 6 states the cost of 3°C warming as $1.65 billion (see Figure 17), whereas the Final Report puts the annual “Electricity Demand”
costs at “$33-$73 billion”:
Annual electricity generation in 2055 was estimated under the transient scenarios to be 4 to 6% greater than without climate change. The annual costs of meeting the increase due to global warming, assuming no change in technology or efficiency, was estimated to be $33-$73 billion (in 1986 dollars). (Smith and Tirpak 1989, p. xl)
How on earth did $33-$73 billion in the EPA’s Report become $1.65 billion in Nordhaus’s paper? Nordhaus’s figure is 1/20th of the minimum figure given by the EPA. Differences in base and terminal years simply don’t cut it as an explanation. 1988 prices were 34% higher than 1981 prices – not 2000% higher. The 5-year difference in terminal years also can’t explain a 20-fold difference in the numbers. So what magic did he use? I can’t know of course, but – unless he simply made that $1.65 billion number up – I expect that he simply applied the magic sauce of “technology”.
“Nordhaus … states the cost of 3°C warming as $1.65 billion (see Figure 17), whereas the [EPA] Final Report puts the annual “Electricity Demand” costs at “$33-$73 billion”
The EPA said that it assumed “no change in technology or efficiency”, right? Well that’s gotta be wrong, right? So, let’s assume technological change happens fast enough that $33-$73 billion in roughly 65 years (from 1989 till 2055) gets reduced to something insignificant, like, say $1.65 billion? What’s that require? Oh, a rate of technical change and efficiency improvements of 4.5%-5.8% per year. That’ll do it. Yeah, let’s use those numbers. Bingo!: a worrying range of $33-73 billion (between 1.3% and 3% of 1981 GDP) becomes a trivial $1.65 billion, or just 0.07% of GDP.
I doubt that I’m using Nordhaus’s voice here, but I’m pretty certain I’m channeling his mind. Though he does not deny that Global Warming is happening, he does deny that it will have a significant impact on the economy – as I’ll detail later. Faced with numbers that implied it would be significant – even after dismissing its impact on 87% of the economy – he looked for a way to trivialise it. The fact that the EPA did not factor in technological change is probably the way he went about doing so.
An adjustment for technical progress is warranted, but a more conventional estimate of the rate of technological progress in industry in general is of the order of 1-2% p.a. Given what was known about electricity generation costs and technology at the time of Nordhaus’s paper, a reasonable single point estimate would have been the average of the range given by the EPS ($53 billion) divided by a rate of technical progress of 1.5% a year for 65 years, which is $20 billion – more than 10 times the figure used by Nordhaus, and equivalent on its own to 0.8% of GDP.
His estimates of the total annual costs from sea level rise of $5.29 billion are equally magical. No such figures were evident in the EPA Report itself, which instead gave costs in terms of cumulative capital expenditure, ranging between $32 billlion for a 50cm rise by 2100, to $309 billion for a 200cm rise.
Figure 18: Sea level cost estimates from EPA-230-05-89-050 (Smith and Tirpak 1989, p. 141).
How did Nordhaus reduce this $32 billion to $309 billion range for capital costs to annual costs of just $5.29 billion? I can’t know, but again, the reduction is ridiculous. For obvious reasons, seawalls require constant maintenance, particularly in the context of an expected rise is sea levels. A modest estimate of depreciation and maintenance costs would be in the range of 5-10% of capital costs. That implies $15-$30 billion in annual costs, not $5 billion. A reasonable point estimate from the EPA report, using the average of the two extremes for capital costs ($170 billion) and a 7.5% maintenance cost, is $25 billion, or another 1% of GDP.
So even with 87% of the US economy ruled out of contention, using the EPA report, the realistic estimate of damages from a 3°C temperature rise, in terms of electricity and sea level rise, is of the order of a 2% fall in GDP. Though this is still far lower than can be deduced from estimates by scientists of the likely impact of 3°C warming on human society, it is substantially more than the estimates Nordhaus derived from the “space=time” assumption, without attempting – as Nordhaus then did – to increase the number by considering “unknown unknowns”.
“How did Nordhaus reduce this $32 billion to $309 billion range for capital costs to annual costs of just $5.29 billion?”
Nordhaus instead reported that 3°C would cause a mere 0.25% fall in GDP. This is similar to the statistical error on an estimate of a single year’s GDP today. Over the 65 years till Nordhaus’s target year of 2055, this is utterly insignificant – and also utterly unjustified:
We estimate that the net economic damage from a 3°C warming is likely to be around ¼% of national income for the United States in terms of those variables we have been able to quantify. This figure is clearly incomplete, for it neglects a number of areas that are either inadequately studied or inherently unquantifiable. We might raise the number to around 1% of total global income to allow for these unmeasured and unquantifiable factors, although such an adjustment is purely ad hoc. It is not possible to give precise error bounds around this figure, but my hunch is that the overall impact upon human activity is unlikely to be larger than 2% of total output…
These remarks lead to a surprising conclusion. Climate change is likely to produce a combination of gains and losses with no strong presumption of substantial net economic damages. This is not an argument in favour of climate change or a laissez-faire attitude to the greenhouse effect. Rather, it suggests that a careful weighing of costs and damages will be necessary if a sensible strategy is to be devised. (Nordhaus 1991, pp. 932-933. Emphasis added)
I expect that I’m the first person ever to check on the veracity of Nordhaus’s numbers here. Having done so, it leads me to a surprising conclusion: Nordhaus is a charlatan.
References
Nordhaus, W. D. (1991). “To Slow or Not to Slow: The Economics of The Greenhouse Effect.” The Economic Journal
101(407): 920-937.
Nordhaus, W. D. (1994). Managing the global commons : the economics of climate change / William D. Nordhaus. Cambridge, Mass., Cambridge, Mass.
Smith, J. and D. Tirpak (1989). The Potential Effects of Global Climate Change on the United States. E. P. Agency. Washington, Environmental Protection Agency.
Tol, R. S. J. (2009). “The Economic Effects of Climate Change.” The Journal of Economic Perspectives
23(2): 29 – 51.
.
