The Tyranny of Tautology

A response to A conversation with Dr. Roger Pielke Jr.

Guest essay by Scott Bennett

Willis Eschenbach described the Kaya Identity as being “trivially true”, his opinion is uncontested by Dr Pielke Jr., whose only retort in its defence was, ‘the math is simple’.

The Kaya is a simple Identity, used as a tautological instrument. To deny this, would be to deny the very heart of its utility. The algebraic cancellation and isolation of its terms is de rigueur for its use.

clip_image002

Fig. 1. The “Kaya Identity” as depicted in the lecture by Dr Roger Pielke Jr. : Climate Policy for a High Energy Planet4

I really wanted to understand how the Identity was actually applied, both mathematically and as a “tool” of policy discourse. To that end, I spent several days grappling with Kaya, as demonstrated by Dr Pielke Jr. .

When I felt I fully understood its application, I turned to the real world, from whence the model was presumably derived.

It doesn’t take very long to see why the Kaya is being used as an instrument of policy. Examining the real world, makes it abundantly obvious, just what a stake-to-the-heart, reality is, for policy wonks!

The Kaya’s real value is in its use, as a claim to authority. It is a construct, designed to frame the debate and thus isolate and compartmentalise contradiction.

Everywhere I looked, the terms as factors of total emissions where erroneous. But how could this be, I wondered? It seemed reasonable to suppose that the factors as given in the Kaya, according to Dr Pielke Jr., are the ‘only levers available in the tool box’.

I spent some time gathering data and comparing real places. More and more I began to see, that there was a fundamental factor missing. How is it possible that emissions weren’t a direct measure of the energy intensity of GDP and the efficiency of its energy production? Clearly there was a missing factor that was making the proportionality of the Kaya’s terms aberrant. Some hidden input was providing efficiencies that oddly, reduced the size of real world terms, making their ratios, counter intuitive!

But before I reveal what it is, I will tell you why it was left out! It was censored because it exposes the fact that the relationships of the Kaya are not universally applicable (Across the countries of the world). The inclusion of this important term renders the Kaya impotent as a tool of national policy.

Truly, the phrase “one size does not fit all” could never be ascribed more applicably than to the Kaya Identity!

Land area1 is the missing term and including it makes it very difficult to compare economies directly, and at the same time keep a straight face!

Ratios like, population density and emissions per km, would seem to be, essential aspects of any genuine and realistic analysis. Without this quantity it is irrational to compare national emissions and their individual contribution to the global total.

Singapore, with the world’s highest population density, is 11,000 times smaller than Australia. Australia’s land area represents 5% of the Earth’s surface, while its emissions are just 1% of the global total. The entirety of Europe2 fits inside Australia with room to spare.

Singapore’s population is 4 times smaller than Australia, its GDP is 5 times smaller, its emissions are 3 times smaller and its total energy usage is 45 times smaller. Yet, using the ratio of Emissions/GDP3, we find that Singapore produces 1.7 times more CO2 emissions for every dollar of GDP than Australia. This isn’t a real mystery, when you realise that not all GDPs are equal, of course!

It is probably safe to say that the resources in Australia’s vast land area, something Singapore lacks, is the missing factor in this case. The numbers are also strongly at odds with the assumptions spruiked by Kaya devotees, because Singapore produces all its electricity from natural gas while Australia is coal fired!

It is also probably not a surprise, that with such a small land area, Singapore produces 3,500 times the CO2 per km compared to Australia’s tiny contribution of just 5.5 kt/km.

This is the weakness of the Kaya. It can’t be universally applied. As soon as you compare figures across countries you discover the logical fallacies inherent in it.

Australia’s ratio of, emissions to GDP, is just double that of France. If emissions per square kilometre are compared however, France emits 12 times that of Australia.

It is clear why governments around the world aren’t rushing to embrace the logic of the Kaya. They understand, that they would be ill advised to do so. The Kaya is a tool of the global minded, useless for national policy, that reveals with perfect clarity, the hubris of groupthink and the latent stupidity of collectivist ambitions.

=============================================================

Notes:

1. Absolute values are given here, rather than “Real Land Area” which is of less relevance to the geography of climate.

2. Western Eurasia excluding Asia and Russia. The West or Western Europe.

3. This ratio is demonstrated in Dr Pielke’s lecture! The intent here, is to highlight that its “usefulness” also extends to invalidating the relationships between all four terms of the Kaya itself 😉

4. https://www.youtube.com/watch?v=QTUE5Ue6Z38

UPDATE: Dr. Roy Spencer has a nice simplification of the terms cancelling issue here in The Kaya Identity Crisis

UPDATE2: Elevated from a comment.

The problem with the Kaya identity is in its application, not in its arithmetic or ability to produce a bit of understanding about the real world. It is being used to help generate policy; long term policy that will be around for decades. It is being used to generate a meme; a way of thinking that will influence decision makers for many years to come.

The Kaya identity begins with the assumption that CO2 emissions MUST be reduced. RPjr stated in his video that it wasn’t even worth talking about the science of climate change anymore. He implied that there was absolutely no point in even discussing climate sensitivity to CO2 emissions and that such discussions are actually harmful. (I was gobsmacked!) The Kaya identity is part of the meme that proclaims “The science is settled!” He argues that it doesn’t matter what the science says about CO2′s impact. The Kaya identity is valid regardless. While that may be true for the identity, it is just stupid to carry that thinking over to the process of making policy. There is nothing more important than the science in making good policy decisions.

The Kaya identity ends with disaster. It is inherently linear in every aspect. The world is inherently non-linear in every aspect. The Kaya identity gives an illusion of knowledge and wisdom to decision makers; convincing them that they will be making good choices. In reality, there is a near zero chance that policies resulting from the use of the Kaya identity will be positive. The outcomes from such policies will range from bad to disastrous.

The Kaya identity gives decision makers the idea that they actually have a control knob. A half turn to the right gives a certain result every time. A half turn to the left gives another result, but just as predictable and dependable as the half turn to the right. This is a complete illusion!

Using the Kaya identity to make policy is like deciding to paddle your raft with two strokes on the right, followed by two strokes on the left, for the entire duration of your trip down the Colorado river. Such a strategy will not get you very far and may actually kill you. They way to paddle your raft down the Colorado river is by constantly assessing your current situation and deciding the best possible paddle strokes for that moment.

The same is true for climate change policy. There is no need to implement solutions today that will solve all climate change problems for the next 100 years. In fact, that would be impossible, and any attempt to do it would almost certainly cause more harm than good. In order to make good decisions, those decisions should be focused on the short term, and the main objective should be the strengthening of the position of future decision makers. That means the current policies should promote adaptability in all areas while enhancing the financial strength of future generations to deal with their issues; issues that they will certainly understand far better than we do today. It means the science is constantly assessed, along with the current state of the population and their needs. It means the UN should be concentrating on potable water for all of humanity today and not on the average global temperature 100 years from now.

The use of the Kaya identity rationalizes the bad decision making process. It allows decision makers to ignore the vital importance of adaptability and weaken the financial strength of future generations. It is the height of hubris and the antithesis of wisdom to use the Kaya identity in the manner it is being used by the United Nations and other bureau-crazies; and apparently promoted by Roger Pielke, Jr; a man I admire and respect, but strongly disagree with on this topic.

The climate data they don't want you to find — free, to your inbox.
Join readers who get 5–8 new articles daily — no algorithms, no shadow bans.
4 1 vote
Article Rating
207 Comments
Inline Feedbacks
View all comments
July 24, 2014 10:57 am

Anthony says: I’m prepared to leave with the identity being trivially true for science to true for political purposes.
True for political purposes? I really think you do need to (re)read what Jim Clarke said in this thread on July 24, 2014 at 6:46 am
It’s one of the most intelligent responses in this whole discussion, and too bad it gets drowned into a lot of _____
REPLY: hadn’t seen it. BTW true for political purposes doesn’t mean I agree with the trueness, only that they think it is true, and as we know there is very little truth in politics. So “true” for politics really doesn’t mean much. – Anthony

Ted Clayton
July 24, 2014 11:03 am

sinewave said July 24, 2014 at 10:09 am

Has anyone reached out to Yoichi Kaya for his comments on all this?

Kaya, in the 1993 book Environment, Energy, and Economy: strategies for sustainability, was looking at the economics. Japan has continued to suffer from some kind of major, historic economic sustainability syndrome.
Since the early ’90s, several others have revamped, repurposed, refactored the Kaya Identity. It is now IPCC ‘canon’, but within their own “framework”. No less than PNAS have published [2002] A framework for sustainability science: A renovated IPAT identity, which although obviously the identity in question, references “Kaya” exactly one time (and then using the “so-called” qualifier). Again, with the “framework” buzz.
Kaya appears to be emphatically & purposefully/intentionally no longer involved. To the point where efforts are being made to ‘scrub’ the name. Which is a bit interesting, in its own right.

July 24, 2014 11:14 am

Shawnhet (9:51 am):

… I do think that land area will have an impact on CO2 emissions. I just think that this relationship is going to prove to be highly non-linear so it will probably not be easy to quantify. That said, without quantifying the relationship somehow there is no way to test this idea and it is only handwaving. It is one thing to claim that Kaya + land area does a better job of explaining CO2 emissions that Kaya alone. It is another thing to *show* that.

You (and Scott and others) are still not quite getting the point: Land area is already in the Kaya Identity. A country’s land area affects its economic structure, and its economic structure affects the value of Energy/GDP — one of the ratios in the Identity.
As you point out, land area affects energy usage for transport, among other things. Singapore only needs to move goods a short distance from port (or, often, within the port). Australia needs to move them hundreds of miles.
The Kaya Identity is only an analytical framework that needs to be fleshed out with real-world facts — such as how land area affects the ratios in the Identity.

July 24, 2014 11:16 am

Ted Clayton says: July 24, 2014 at 11:03 am
Kaya appears to be emphatically & purposefully/intentionally no longer involved. To the point where efforts are being made to ‘scrub’ the name. Which is a bit interesting, in its own right.
IPAT emerged out of a debate between Barry Commoner, on the one hand, and Paul Ehrlich and John Holdren on the other. For that reason, it’s sometimes called the Ehrlich-Holdren identity / equation.
Prof. Kaya and collaborators independently “(re)discovered” the Kaya identity /equation, which i basically the same as IPAT.
Neither Ehrlich-Holdren nor Kaya seem very interested in having their names attached to the identities/equations.

July 24, 2014 11:17 am

Shawnhet:
By the way, you do know that different countries have different values for ratios in the Identity, right?

July 24, 2014 11:25 am

Dr. Doug says: July 24, 2014 at 11:14 am
You (and Scott and others) are still not quite getting the point: Land area is already in the Kaya Identity.
Of course GDP covers a lot of activities. Just take a look at the number of sectors in e.g. the North American Industry Classification System or NAICS.
That doesn’t mean one isn’t allowed to further decompose GDP (per capita), into as many terms as one likes. In fact, that is what economists do all the time, except of course they don’t call it KAYA.

July 24, 2014 11:27 am

Duster (10:21 am):

How again is the Kaya identity informative?

The Kaya Identity is informative to the extent that one can plug in informed numbers for the ratios in the identity. Pielke does so, showing, for example, that Britain’s energy policy cannot plausibly achieve its carbon-reduction goals.

JJ
July 24, 2014 11:28 am

Bart says:
I get 1/2 emissions. Why is that strange?

Equivocation parade from steveta_uk in 3…2…1…

Resourceguy
July 24, 2014 11:28 am

Kaya sounds like the kaka used by Jeffrey Sachs to gen up papers with students in place of real research. It also has a lot of similarities to the more familiar version of ranking cities, states, and countries with a handfull of general variables to concoct a new kaka measure and research-sounding headline.

Shawnhet
July 24, 2014 11:33 am

Dr. Doug, I understand both of your points to me. My point, was that it may be possible to make our understanding *even better* that the KI itself does (perhaps by helping us understand why some countries E/GDP are higher than others).
GDP is a one dimensional number and does nothing on its own to explain why E/GDP varies – figuring that out is a valuable exercise IMO.
Cheers, 🙂

July 24, 2014 11:45 am

steveta_uk says:
July 24, 2014 at 10:09 am
“Bart, please try it with some real numbers.
Fill in the entire Kaya identity with real information from any where.
Then 1/2 the population and keep everything else the same.”
steve, one of the troubles with the arithmetic is that if you half the pop, you more than half the GDP. GDP grows with population, at least in industrial countries.

July 24, 2014 11:50 am

Jim Clark (6:46 am):

The Kaya identity ends with disaster. It is inherently linear in every aspect. The world is inherently non-linear in every aspect. The Kaya identity gives an illusion of knowledge and wisdom to decision makers; convincing them that they will be making good choices. In reality, there is a near zero chance that policies resulting from the use of the Kaya identity will be positive. The outcomes from such policies will range from bad to disastrous.

Jim, any policy tool can be used either well or poorly. Don’t you see value in Pielke’s conclusions about British energy policy?
Yes, much of the world is non-linear. Marginal changes in, say, the CO2 intensity of energy might well differ from the current average CO2 intensity of energy. Newly added power plants might be natural gas fueled, for example, while legacy power plants are coal fueled. Still, the Kaya Identity can be applied well enough to the marginal changes.
So, yes, do oppose any stupidity that abuses the Kaya Identity. But don’t judge the matter without facts.

Joseph Murphy
July 24, 2014 11:53 am

Jim Clarke says:
July 24, 2014 at 6:46 am
—————————-
Jim nailed why the Kaya identity is not overly useful. It gives the illusion of having control knobs, one of which is population, yikes. (People) x (production per person) x (energy use per production) x (carbon emmissions per energy use) = industrial carbon emmission, not overly complicated stuff. If you want to get nit-picky here is food for thought. It is not hard to imagine a completely automated industrial construct still chugging along with no people left to complain about it. Also, if we removed all industry, mankind would still have an affect on the carbon cycle.
Aren’t we lucky that most of our energy production’s waste product is completely harmless to all animals and beneficial to all plant life on this planet? Should we not be celebrating this fact? I am all for double checking to make sure we are not overdueing it but, there is really no reason to believe we are. If This debate were about the science maybe that would be a reasonable position. Oh hell, I misplaced my soapbox.

Ted Clayton
July 24, 2014 11:54 am

Johan says: July 24, 2014 at 11:16 am

IPAT emerged out of a debate between Barry Commoner, on the one hand, and Paul Ehrlich and John Holdren on the other. For that reason, it’s sometimes called the Ehrlich-Holdren identity / equation.
Prof. Kaya and collaborators independently “(re)discovered” the Kaya identity /equation, which i basically the same as IPAT.

Ahhh. IPAT preceded KAYA. That’s why PNAS don’t have to cite Kaya. And I really did wonder why Paul Ehrlich and John Holdren weren’t in this discussion. Back to the stacks. Thanks!

dp
July 24, 2014 12:24 pm

I’m prepared to leave with the identity being trivially true for science to true for political purposes.

That has been a common view through all the threads even by people who know how the identity is used. The evidence it is principally a political tool is observed in the literature where it appears. PNAS and Pielke, Jr being good examples, but also because the author of it is a strong advocate of renewables, a topic that requires political intervention because it isn’t supported by market forces.

Joseph Murphy
July 24, 2014 12:24 pm

Dr. Doug says:
July 24, 2014 at 11:50 am
Jim, any policy tool can be used either well or poorly. Don’t you see value in Pielke’s conclusions about British energy policy?
———————-
I don’t disagree but some tools are better suited for certain purposes. Let’s rewright it so that at least we are not offing people. CO2 emmisions = GDP x (Energy use per production) x (Carbon emmissions per energy use). I feel safer already.

richardscourtney
July 24, 2014 12:26 pm

Dr. Doug:
You conclude your post at July 24, 2014 at 6:38 am by saying

I hope that we can wind up this thread quickly without wading into the swamps of incomprehension that made the last three threads on this topic so frustrating.

I agree. And for three threads I have been saying why; i.e.
The Kaya Identity is a propaganda tool that permits malign policy-makers to pretend a logical basis for their desires because –in reality – the Kaya Identity is merely an unjustifiable collection of prejudices presented as an equation with the intention that the prejudices displace scientific evidence and argument.

Proponents of the propaganda attempt to support the equation which has been devised as a tool to present the propaganda. And their support has generated the “swamps of incomprehension” about the egregious propaganda tool and, thus, has hindered outright rejection the propaganda tool. as being the illogical nonsense which it so clearly is. But several people have seen through this support for the unreason presented as the Kaya Identity. For example, Jim Clarke says the same with more detail in his post at July 24, 2014 at 6:46 am, and his post concludes

The use of the Kaya identity rationalizes the bad decision making process. It allows decision makers to ignore the vital importance of adaptability and weaken the financial strength of future generations. It is the height of hubris and the antithesis of wisdom to use the Kaya identity in the manner it is being used by the United Nations and other bureau-crazies; and apparently promoted by Roger Pielke, Jr; a man I admire and respect, but strongly disagree with on this topic.

Richard

TimC
July 24, 2014 12:29 pm

I’m interested in Dr Spencer’s posting as to “miles = [hours] x [miles/hour]” – that (on the one hand) this says that distance travelled = time elapsed x speed (a useful construct) but (if simplified) just that “miles = miles”.
The test seems to be that all of the “contributing quantities” (distance, time elapsed, speed) are familiar concepts frequently used separately and independently, in various contexts, and the formula just reminds us of the intrinsic relationship between them. I see plenty of sense in this (that it is all essentially down to context in which used) – but accept others may not …

July 24, 2014 12:30 pm

Shawnhet (11:33 am):

Dr. Doug, I understand both of your points to me. My point, was that it may be possible to make our understanding *even better* that the KI itself does (perhaps by helping us understand why some countries E/GDP are higher than others).
GDP is a one dimensional number and does nothing on its own to explain why E/GDP varies – figuring that out is a valuable exercise IMO.

In agreement with your earlier comment, I’m sure that land area affects E/GDP in ways that are both non-linear and difficult to quantify. And you’re quite right that GDP is an aggregate that hides many material differences between countries — differences that are relevant for many purposes.
Nonetheless, I’m not sure how much would be gained by quantifying the effect of land area on E/GDP. Land area is not itself a policy variable. Perhaps it might be useful to look at something more concrete like land-freight ton-miles (or tonne-kilometres), which may depend on land area (or more generally on geography), along with energy usage per ton-mile. The latter is potentially a policy variable, for example in promoting railways over trucks (lorries).
For many purposes it may be sufficient to simply note that countries with different economic structures have different E/GDP ratios, and to have a rough idea as to the reasons why.

Dave
July 24, 2014 12:32 pm

JJ says:
July 24, 2014 at 5:19 am
OYG WUWT’s experiment in communal embarrassment continues. Each post inexplicably more asinine than the last.
####
Amen to that. Meanwhile real work in fighting the stupidity goes on, but here:
http://stevengoddard.wordpress.com/2014/07/24/more-from-bridgehampton-ny/
instead of here on this blog.
REPLY: So I didn’t do enough? Gosh, next time I won’t even bring stations like this to anyone’s attention for fear of your judgement because I didn’t do enough in the first post about it and someone built upon what I started. That’s called science BTW. Since we are in the judgment business today, what if anything have you done, to fight such “stupidity”, other than complain that I’m not doing enough, or that I’m allowing discussion of concepts you consider unfavorable?
Show your work.
Anthony

brians356
Reply to  Dave
July 24, 2014 12:51 pm

>> OYG WUWT’s experiment in communal embarrassment continues. Each post inexplicably more asinine than the last.
Thanks for joining in!

July 24, 2014 12:40 pm

Update #2 said: “The use of the Kaya identity rationalizes the bad decision making process. ”
In actuality, the use of the Kaya identity rationalizes the bad decision making process upon which Progressives subsist. It allows them to hide their true intentions behind a smokescreen of “green energy” policies.

richardscourtney
July 24, 2014 12:48 pm

John West:
re your post at July 24, 2014 at 9:30 am.
Contrary to your assertion,
There is NO EVIDENCE for anthropogenic global warming (AGW); n.b. none zilch, nada.
Three decades of research conducted world-wide at a cost of more than US$5 billion per annum have failed to find any such evidence. If you think you have some then publish it because the discoverer of such evidence will be awarded at least one Novel Prize and probably several Nobel Prizes.
Evading the fact that there is no evidence for AGW is one of the purposes of propaganda tricks such as the Kaya Identity.
Richard

July 24, 2014 1:01 pm

Joseph Murphy (12:24 pm):

I don’t disagree [that any policy tool can be used well or poorly] but some tools are better suited for certain purposes. Let’s rewright it so that at least we are not offing people. CO2 emmisions = GDP x (Energy use per production) x (Carbon emmissions per energy use). I feel safer already.

For what it’s worth, I also oppose treating population as a policy variable. For that matter, I also oppose restricting GDP for the sake of reducing CO2. However, in applying the Kaya Identity to the future, it’s necessary to make projections of future GDP, and that requires making projections of population.
By the way, in case anyone wonders, I play no role in environmental or energy policy.

Ted Clayton
July 24, 2014 1:04 pm

see also Wikipedia entry: I = PAT
(This is prominently linked, in Wiki’s Kaya identity entry, which several of us have used for a cheat-sheet.)

The equation [IPAT] was developed in the 1970s during the course of a debate between Barry Commoner, Paul R. Ehrlich and John Holdren. Commoner argued that environmental impacts in the United States were caused primarily by changes in its production technology following World War II, while Ehrlich and Holdren argued that all three factors were important and emphasized in particular the role of human population growth.[1][2][3]

The Kaya identity is closely related to the I = PAT equation. [also mentioned in the reverse relationship, in the Kaya entry]

So the Kaya-thread actually begins in the 1970s, under the name IPAT, originally with 4 well-known ‘authors’. Johan has mentioned that Ehrlich/Holdren no longer want to be associated with it; perhaps Commoner with be a better search-lead.
But, the 2002 PNAS article I linked earlier will have at least some history on IPAT, the paper being titled as a “renovated IPAT”. I’ll revisit it. (This paper is a lead citation in the Wiki Kaya entry.)
Is this all worthwhile? Well, we have a lot of Kaya-discussion, without noting (or seeming to know) that it had quite an extensive previous AND parallel history & intellectual context, under another name and by different authors.
‘Serious players’ don’t want to talk ‘Kaya’. They talk about the same thing, when they want to, under the IPAT moniker, sometimes seen as I=PAT.
There is a lot of Mathus (200 years back) in this, particularly when we consider the involvement of Ehrlich & Holdren, who would rather we not … which tells me ‘dig here’. 😉
P.S. I like mathematical ecology. It’s basically hardcore science. Ehrlich did some good science (in that field), very classical, and at the time his Doomsday warnings did not seem remiss to many of us. His science-example seemed refreshing, and his politics well-intentioned … once upon a time. His, and especially Holdren’s affinity for, uh, population-adjustment solutions, cooled my affection in the longer run.

July 24, 2014 1:06 pm

To both Shawnet and Dr. Doug (who will probably not read this).
It would be nice if you actually read what both the original blog poster and yours truly are saying.
“Land area” is not what matters, but the “availability of natural resources”. And that could be a relevant “policy variable”. In the same sense that a country could try to improve its energy intensity (by means of introducing more energy efficient technologies), that same country could also try to improve the efficiency with which it “extracts and uses” it natural resources (including both replenishable and non-replenishable resources).
I once more introduce the SCOTT BENNETT IDENTITY
CO2/GDP = (L/GDP) * (P/L) * (GDP/P) * (E/GDP) * (CO2/E)
Where L would be “availability of natural resources”, not “land area”. To lower CO2/GDP, one could – ceteris paribus – lower (L/GDP). And lowering (L/GDP) simply means, for the same amount of GDP, you use less of your “natural resources” (or alternavely, for the same amount of natural resources, you could produce more GDP).
Granted, (P/L) would now mean “population per unit of available resources”, and not “population density” (population per unit land area). It’s not obvious how that could be a “policy variable”. You certainly don’t want to lower population P, keeping L constant. And one can hardly increase non-replenishable resources (add iron ore?). But if L includes replenishable natural resources, there might be a way to increase L (at least the replenishable part), while holding P constant.
And of course, in (macro) economics one always has to be aware of the danger of “double counting”. The primary sector is evidently part of GDP.
Anyway, although it could be considered some “refinement” of KAYA, it would not solve the main problem with KAYA, as clearly demonstrated by Jim Clarke.
Yours truly.

1 3 4 5 6 7 9