Tropics even more sad if the Earth heats up

Last year, La Stampa published a rather skeptical article on climate change under my name. It was not written by me, as discussed here and here. Today, the record is set straight (original). The piece calls for a carbon tax in Italy.

42 replies on “Tropics even more sad if the Earth heats up”

Rationality as defined in economics is a much narrower concept than any dictionary would define it. Economists do not insist people are ‘rational’, it’s just that their calculations become intractable without this radically simplifying assumption.

By this definition, if a rational person can steal a wallet without risk then he will steal it. If he can profit from murder then he will commit murder. If the extinction of the great apes were to profit him one cent, he’d wipe them all away without hesitation.

Discounting, another concept in economics, measures the value of a euro — or a euro of damage — a year in the future relative to the value of a euro today. Economists measure the discount rate by observing people’s behaviour in the real world.

Here’s what you told an interviewer about the application of discounting to environmental problems:

Stern’s stark predictions about the impact of climate change would vanish had he used a standard discount rate. The discount rate reflects, partly, how much you care about the future. You should make up your own mind about the discount rate that is right for you. People do, and economists and psychologists have measured people’s discount rates. People use much higher discount rates than the one preferred by Nicholas Stern. Stern essentially tried to impose his views on others — like a colonial master would tell the savages what to think.

Let’s break that down sentence by sentence, setting aside the first for a moment.

The discount rate reflects, partly, how much you care about the future. You should make up your own mind about the discount rate that is right for you. People do, and economists and psychologists have measured people’s discount rates.

Fair enough so far…

People use much higher discount rates than the one preferred by Nicholas Stern. Stern essentially tried to impose his views on others — like a colonial master would tell the savages what to think.

Leaving aside for a moment the ‘colonial master’ reference, the sort of personal attack your defenders find so inoffensive when directed at others but so outrageous when directed at you, let’s look at the second last sentence.

“People use much higher discount rates than…Stern”, but they are making personal decisions. How do you know how they would behave if they were instead directing the affairs of nations?

You elide the transition from the individual to the social without even mentioning it, and it is this sleight of hand that enables you to come up with any answers you like with your projections.

A rational person would always organise the state to benefit himself. Real people frequently act otherwise. Real people die, nations do not. Real people discount future losses and profits heavily because they may be dead before the day of reckoning ever comes. How do nations behave?

Let’s imagine a world where everyone is rational and nations discount as if they were mortal.

The Parthenon would never have been built, nor would the Taj Mahal or CERN. The dome of St Peter’s would be a practical brown. Shakespeare would have gone into the maritime insurance business, Rembrandt plied his trade as an interior decorator and Galileo would have become an accountant.

Instead of promising to fight them on the beaches, fight them on the landing grounds, fight them in the streets and in the hills, Churchill would have had a black uniform fitted and laid out a red carpet for the Nazis. Never in the history of human conflict would anybody have owed anything to anybody.

Neither Pythagoras nor Euclid would have wasted his time on pointless matters. As their utility builds only incrementally and is initially negligible, there wouldn’t be a single university in the world.

There would be no civilisation.

Your transposition of the personal discount rate to civilisation as a whole has no foundation in economics yet you present your results as if they had the imprimatur of science.

You told the interviewer that “Stern’s stark predictions about the impact of climate change would vanish had he used a standard discount rate”. Standard, implying conventional or agreed upon, is deceptive. There is no mechanism in economics for weighing the interests of people not yet born against those alive today. Whether your discount rates are even ‘standard’ is more than debatable.

Stern’s discount is every bit as defensible as yours economically. The difference is that Stern represents an ethically defensible choice whereas yours represents self-conscious theft from future generations. Manipulation of the magical discount factor makes the benefits of GHG abatement vanish. This isn’t science and stop presenting it as if it were.

There is no single discount rate, personal or state. The discount rate is largely a function of uncertainty, or risk, over time.

Investment in building something that will certainly be useful for a long time, or in avoiding something that would certainly be bad could use a low discount rate.

Additionally, but differently, risk is different if it is small and dispersed or systematic. We treat nuclear power risk differently from pedestrian fatality risk. Climate change risk is more like nuclear power risk than pedestrian risk.

As you may have guessed, climate economists are acutely aware of this. I’d say that 20% of the literature is on these topics exactly — with contributions by some of the great names: Arrow, Chichilnisky, Dasgupta, Gollier, Heal, Lind, Nordhaus, Pearce, Pindyck, Schelling, Stiglitz, Weitzman.

@Hugh Sheehy

I’m pretty much in agreement.

Using a 3% discount rate, a €1000 gold bar placed in a time capsule to be opened in a century becomes worth only €48. Absurdly, more than 95% of the value is held to have vanished while it was in the ground. This is the magic of Professor Tol.


If you are acutely aware of the giant hole in your work then my suggestion is that transfer your time and energy from trying to convince people that €0.02/l of petrol will make everything okay to communicating the fact that your plans involve impoverishing the future for the sake of the present.

@Richard Tol

You said,

“Climate change is unlikely to affect the Gulf Stream in any substantial way. That scenario is supported only by models that oversimplify the physics of ocean circulation. It is a mathematical artifact in an inappropriate model.”

It strikes me that the whole thing is a political artifice based on inappropriate models.


As per usual, only so-called ‘sceptics’ call in to offer their endorsement. You seem to find their support agreeable, never question their claims and denounce your opponents as wooly-headed greens. This is strange for someone who claims to endorse scientific orthodoxy.

It doesn’t take an environmentalist to oppose trashing the world for personal advantage, or to oppose the smoke-and-mirrors of your sophistry.

Matthew 25, 13-30

The Parable of the Talents.

The evil and lazy slave was the only Socialist among the group.

The other two reflect what is Capitalist in mans nature.

The “Master” would be considered “the consumer” in Austrian Economics.

The question might be: how long slaves? In the Capitalist pair – not so long. In the case of the Socialist; forever.


Your gibberish discredits the Professor more eloquently than I could ever hope to do.

@ Adrian

Re your discount rate comment. You misunderstand the concept.

Btw your maths is also off – €1000 discounted @ 3% for 100yrs has a PV €52.03.

Using your gold bar example: the calculation of PV (present value) does not tell you what it will be worth in the future, but rather what having it in the future is worth to you now. Let’s say you actually had the €52.03 cash today; and you see a long-run investment promising 3% per annum with minimal risk (e.g. bonds). That investment will be worth €1000 in 100 years – so using you €52.03 today, you can ensure that you may have €1000 with which to buy your gold in the future.

Now if you’re talking about gold bar worth €1000 today, the discounted present value is €1000, since n=0. You discount future benefits and costs, not current ones.


Your figure of $52 is correct rather than $48. Both figures are a long way from $1000, however, and I don’t misunderstand the concept.

The intention was to illustrate the distortions implicit in Prof Tol’s cost/benefit analyses. All costs are accounted for 100% in his calculations, but the benefits are disappeared via discounting. Arguing in terms of net present value, you could as easily claim the planet in 100 years has a net present value of 0 as the hypothetical gold bar.

The benefit of $1000 of damage averted a century hence nearly vanishes in Tol’s calculations and the effort taken to avoid it becomes not worth while — except it’s an illusion derived from an accounting trick wholly inapplicable in the circumstances. The Professor has not attempted to justify this, and his attack on Stern’s choice are not supportable by economics.

What discount rate would you suggest and why?

@ Adrian

I think choosing any one discount rate (as Stern did) and not including any sensitivity analysis is wrong. Getting a range from say 0% up to 5% allows the reader to make their own decision about it.

Re discounting more generally; the theory is sound. Climate change policy costs a lot today and delivers a lot of benefits far in the future. Therefore some discount rate is necessary to account for opportunity cost of capital and time preference. Economists often go for 3.5% because it is the green book discount rate; used for most public investments. The only compelling (to me) reason not to use the green book rate (or something like it) for climate policy is the inter-generational split of benefits and costs. The lag between benefits and costs is especially long relative to most public investments.

So the payers are not the main beneficiaries (and that’s without adding geographical considerations); do we make them value other people’s benefits as highly as their own? It all gets a bit philosophical here. The closest thing to a scientific approach is to try to gauge people’s long-term discount rates and people’s tendency to discount benefits to other people. The results are a bit disappointing.

Still, if I was to choose only one discount rate I’d go with 3.5%, though I wouldn’t feel overly confident about it.


‘Compensators’ are not the main recipients of the ‘compensation’ is more correct. What’s asked is that those using fossil fuels pay the entire cost instead of paying only for extraction and refining, as is broadly the case globally now, while foisting the rest onto everybody else.

While a full response must wait until tomorrow, I’ll just note for now that Tol has A) attacked Stern’s discount rate as an imposition B) claimed it would require us to invest our entire income to benefit future generations.

A) Assumes individual preference in the private sphere can be translated to the public sphere without complications.

B) Conflates foregoing opportunities to do good with avoiding inflicting harm, a confusion I can’t understand as being accidental.

Jesus – and I thought that the BizarroWorld of climate change denialism couldn’t get any odder. Now we have people assuming the identity of Richard Tol to publish this kind of stuff.

So, how do we tell in future which is the ‘real’ Tol, and which are the fake-pseudonym Tols?

Why but life insurance? If you need it, you’d be dead !
Why wear a seatbelt? If you need it, you’d be dead !

Why defer Jam Today ? Save the (fruit) trees?

I had indeed suspected that discount rates would be a major discussion in the field.

The other area where I have not guessed, but merely wonder, is how the possibility of future development of cheap technologies to undo the problem is incorporated into the calculation. NPV calculations are notoriously difficult beasts and technological change is such a wildcard that it’s practically impossible to build in – into a business plan at least.

Actually, the financial modelling of climate change might actually be simple enough that you could quickly incorporate a calculation of the possibility that someone will invent a technological solution that costs something trivial like a few trillion to deploy and which could solve the problem. A sort of default risk calculation, but backways. If the probability of a technology solution only needs to be very low then doing anything now might be a bad bet in the first place. People are a clever bunch, sometimes.

Again, I’m probably parrotting a very old idea.

Indeed. It’s referred to as “backstop” in this literature, and the notion was introduced, I believe, by Nordhaus (1977).

The possibility of a backstop qualitatively changes the policy recommendations, as it takes away a major reason for immediate & drastic emission reduction.

Sorry, I disagree! Solar variations exist in the invisible parts of the EM spectrum and are clearly associated with sunspot activity for the ease of reference by economists who might otherwise find it difficult. Fewer sunspots =less energy transfer.

CO2 does not cause what we have seen so far, that is down to dodgy measurements and selective media presentations. We will all be experiencing the weather and conversing gently about it, so all this heat is rather funny except that some are trying to make money out of it, knowing as they did, that banking was “entering a bit of a trough”! It is a social welfare scheme for kleptocrats. Until the money machine picks up!

We agree on this one. Over the years, La Stampa has taken a rather brown view on climate change — and last year they took it a step further by publishing a forged article.

@ Adrian

Re your first paragraph, I actually agree. That’s what a carbon tax should ideally do (though as jc pointed out to me last week, carbon taxation may not be politically viable). In economics it’s called internalisation of externalities – forcing people to bare (or at least compensate for) the costs they impose on others.

I actually agree that we shouldn’t be discounting benefits because they are realised far away. It’s disappointing that we do this with all other policy areas. I think what Prof Tol frequently tries to illustrate is that if there is €Xmillion on the table to help poor people in the tropics, then helping them combat the problems they face now might be more beneficial. It may even mitigate some of the public health effects of climate change.

Re your two points:

A) Fair enough, but just making up your own discount rate (and refusing to publish a sensitivity analysis) is no solution to that issue; it’s even worse.

B) That’s a moral question, not an economic one – for mathematical/econometric treatment this distinction is not crucial. You’re (I think) treating climate policy as a moral imperative. I think most people would agree to some degree. But how much exactly do we spend? Do we forgo economic development that would also benefit future generations? Cost-benefit analysis isn’t perfect at all, but it does give us an idea.


Indeed. It’s referred to as “backstop” in this literature, and the notion was introduced, I believe, by Nordhaus (1977).

The possibility of a backstop qualitatively changes the policy recommendations, as it takes away a major reason for immediate & drastic emission reduction.

While not impossible, there’s no way of knowing if this is possible or not and good reason to think it may never be achieved. CO2 is the lowest energy state of carbon. Nobody has any idea how it might be recovered from the atmosphere once emitted. Possibly it will never be economical.

Chemical conversion is not possible without putting back in the energy the extraction of which was the entire purpose of burning the carbon in the first place.

It would be as reasonable to rely on aliens providing us with a magic wand as to rely on future technologies of which nothing is known. Did Nordhaus provide any drawings of his mysterious device?

@Pat Donnelly

The basic physics of warming is very simple. Only the details are complex. Rapid atmospheric warming was the obvious and first outcome scientists suspected when they looked at the problem.

Nature may come up with some deus-ex-machina but nobody has plausibly indicated what that might be. Conversely, there are a large number of suggested positive feedbacks in the system and the geological record is replete with rapid warmings — all of which occurred at much lower CO2 levels than are projected for later this century.


We have every right to let future generations look after their own problems; we’re not obligated to sacrifice our economic wellbeing for theirs.

We are obligated not to inflict avoidable damage on them, however. Taking this fairly standard perspective to moral issues, Prof Tol’s requirement that we invest everything we have to benefit the future disappears. More after lunch.

Removing CO2 is only one possible solution to the problem of warming. It’s the most attractive one since it’d remove the cause of the warming rather than compensating for warming, but there are still other possibilities.

@Hugh Sheehy

Tol writes that “the possibility of a backstop qualitatively changes the policy recommendations, as it takes away a major reason for immediate & drastic emission reduction”.

…only this ‘backstop’ is not a possibility. It’s not known if it will ever be a possibility, yet many of the models we’re asked to base policy on assume it will be even though no such deus-ex-machina has even been proposed. I’ve asked Tol to provide details of this plan, but am not holding my breath…

From the Bishop Hill blog, Dec 23 2009:

RichieRich (Quoting Caney): “Martin Weitzman criticises Stern, and also William Cline, for their affirmation of a zero pure time discount rate…

“The exact nature of the complaint is unclear but the key point seems to be that it is inappropriate for an economic analysis to rely on the ethical judgements of the analyst. Instead one should rely on the revealed preferences of the people. Ethical judgements (including judgements about discounting) are the subject of disagreement and so it is inappropriate for an economist to invoke their own moral assumptions.

“The argument as stated, however, is multiply flawed. First, and most crucially, it rests on an illusion. Nordhaus and Weitzman object to a climate analysis that rests in any way on the author’s ‘ethical vantage point’ (as Nordhaus puts it). To do this is to impose ‘their own value judgements on the rest of the world’ (as Weitzman puts it). But this is inescapable. It is simply impossible for economists to avoid such moral judgements. For example, Weitzman’s claim that the rate of pure time preference across generations should be settled by determining people’s revealed preferences is itself a moral doctrine. It holds that it is wrong, morally wrong, for decisions to proceed otherwise. Furthermore Weitzman’s assumption that policies should seek to satisfy preferences is again based on his moral judgement that this is the morally correct way to determine policy. There are very many other moral approaches and he and Nordhaus are adopting one moral perspective (a
preference-satisfying consequentialism)…

“To see the peculiarity of Nordhaus and Weitzman’s claims (and the worries attendant on following the revealed preferences of the members of society) consider issues not related to climate change. Would they say that a book on the welfare economics of the abolition of slavery may not rely on the analyst’s (moral) judgement that blacks and whites have equal moral standing when making utility judgements? Presumably not. In doing so they are implicitly relying on their own (perfectly sound) moral judgement. Such a case shows not only that appealing to moral analyses is inescapable but also that it is not troubling.”

Tol: “All true. But Caney does not answer the question: If not the preferences of the people, then whose preferences? The Ayatollah Khomeini, who was a great scholar and a wise man, did answer that question: Mine!

“The problem with a low pure rate of time preference is that it would not only apply to climate change, but to all intertemporal decision problems: education, pensions, R&D, road safety, and so on.

“If you follow Caney and Stern, then you should drastically cut consumption. If Dasgupta is right, then Stern’s assumptions imply that we should save 99% of our income, instead of 20%. That is, we would slash consumption to 1 in 80 from what it is today.”

All tosh. The absurd 99% figure assumes we should not overlook any opportunity to benefit future generations instead of merely avoiding consciously harming them. An oversight? I don’t think so. Does this follow from Dasgupta? Who knows? Would any sane person trust Tol to judge? What’s certain is that Stern neither stated nor implied anything even remotely resembling this travesty.

Wait to see what ultimately happens to the Ayatollah reference.

RichieRich: “Surely the answer is that, for better or worse, we must rely on the moral judgement of our elected government. You said in an earlier post that you favoured a democratic solution and it would seem that the three main parties all buy into Stern’s view of discount rates. Isn’t Caney’s key point that economists cannot rightly claim that theirs is a superior because non-moral view. If everyone has a moral view and we are ruled by our elected government, then surely it’s their moral view that counts.”

Tol: “The claim that the economic viewpoint is a-moral is just nonsense. If one takes the pure rate of time preference of the median voter, then one adopts the moral position of one person, one vote.”

This is nonsense. The ‘median voter’ Tol refers to was looking after his own affairs. S/he was never asked to express an opinion on national policy. ‘One person, one vote’ is an utter distortion when the electorate never even knew an ‘election’ was taking place. Of course Tol glosses over this.

Richie Rich: “My previous comment re non-moral arguments was simply harking back to what Caney took to be the Philosopher Kings argument, namely that “it is inappropriate for an economic analysis to rely on the ethical judgements of the analyst”. In my view, Caney effectively rebuts this argument. You appear to agree with his rebuttal and with the claim that for better or worse we must rely on the moral judgements of our elected governments.”

Tol: “We seem to agree”

Swiveling abruptly round to a position diametrically opposed to everything he told the press about Stern in the first place. Lovely… and what happened to the Ayatollah? The no-quarter, ding-dong Tol v Tol duel goes on without pause.

The point of the backstop is exactly that it IS a possibility, but that it is ONLY a possibility until it is actually invented.

The problem is to guess how possible, or probable, it might be.

@Hugh Sheehy

Many economic models include this ‘possibility’ in their projections as if it were fact, however — a hypothetical technology for which no designs exist or have even been proposed.

It’s just a name; nobody’s suggested how it might be achieved and we can be certain chemical conversion is not a possibility. Current output of carbon is 8 GtC (equivalent to 30 billion tonnes CO2) annually, btw, so add this to the non-existence of even a design concept and the irresponsibility of adding yet more black magic into the models becomes apparent.

The design concept for the steam engine dated back to ancient Egypt and the design concept for the helicopter to Da Vinci. Nordhaus, on the other hand, isn’t a scientist and he’s given no indication of how his device might operate, let alone operate economically on the vast scale that would be needed.

I think you’re missing my point.
I’ll try again. The possibility of such a technology is a fact. It need not be a device. It need not be aimed at CO2. Several potential solutions have already been suggested, and I make no claims for any of them.
Crikey, one solution is to depopulate several major countries and plant them with forest, but the cost of that solution might be a bit high.
The possibility of a solution can easily be included in the financial modelling.
I’m not making any pro or con argument, not even a technology argument.

@Hugh Sheehy

You’ll need to explain to me how

I think you vastly underestimate the difficulty of the task. A great many things are possible in theory but present insuperable difficulties. For instance Kelvin postulated a little demon that would operate a little door between two gas-filled compartments. It would open the door whenever a fast molecule from one compartment, or a slow one from the other, approached. Thus one compartment would gradually become hotter and the other cooler. Eventually free energy might be generated in this way, in violation of the second law of thermodynamics. Now there are a great many reasons, too involved to get into, why this will never happen: practical problems that make it impossible.

It’s possible to construct a machine that would disassemble a person atom by atom and reassemble him at some distant time or place. Cold fusion reactors are also possible but certainly won’t be a practical possiblity in this millenium if ever. It’s even possible for the entire human race to construct inter-stellar spacecraft, take off on a trajectory that loops round a distant star, and travel to some point arbitrarily distant in time via the phenomenon of time dilation thus circumventing the whole climate change bother.

What does the word ‘possible’ mean? You point to the possible — and completely uneconomical — planting of vast forests. This does not satisfy the definition of the word ‘device’. The requirement is to filter tens of billions of tonnes of CO2 — at very low concentrations — from the atmosphere annually without relying on the logical nonsense of chemical conversion which would require the energy be returned that was the whole point of its emission in the first place.

I would claim application of the word possible to imaginary devices of which nothing is known is an abuse of language. You cannot shift the burden of proof to your opponents and claim everything that isn’t provably impossible can be listed as a plausible solution. It falls to those who suggest such a device to clarify their plan until the orthodox dictionary definition of ‘possible’ is satisfied. An undefined machine for which no description exists does not achieve this.

@Hugh Sheehy

PS Please explain how the probability of the machine’s construction is to be quantified so that it can be included in the financial modelling.

First, it was Maxwell’s demon. Kelvin simply gave it the cool name. There are some great jokes with the demon in.

Second, you include the possibility of a technological solution into the financial modelling as a notional cost and a notional probability and then you play scenarios.

Again, I make no claims or predictions about any technology or about global warming.

However, it is easy enough to make the point about technological possibility. A device that can cause a party’s hostess’s undergarments to leap one meter to the right is so unlikely that it is not worth considering, and seems likely to remain so. However, the invention of ultra broadband through-wall imaging systems would have seemed merely unlikely a few decades ago. There was plausible physics.

There are various potential solutions where there is plausible physics for global warming solutions. They’re possibilities. They may never be actually implementable, but that possibility is not zero. That’s all.

@Hugh Sheehy

Notional costs and probabilities are basically guessed-at costs and probabilities. These do exist in some (all?) models and illustrate the many deviations from scientific rigour involved.

Assumptions about trend growth etc. are also made in the models — guesswork without scientific evidence. Per-capita growth in the rich economies cannot be usefully estimated as it can for poor countries. We can not invent new electricity, powered transportation, flight, refrigeration etc. The wave of globalisation, the principle engine for growth in the recent past, is already passing. Scope for further liberalisation does not exist in the USA, UK, Ireland etc.

Given the roll-out time for new technologies today, it’s difficult to imagine radical departures outside of the life sciences. Nonetheless, the models include estimates for growth despite their lack of any empirical or theoretical basis. Prof Tol etc. invite us to just take a stab in the dark at what trend growth might be, even though his own record in this department includes the truly abysmal Medium Term Review 2008-2015.

At least some (AFAIK all) models omit biodiversity losses, known to be certain on the current trajectory. Extinctions will impoverish our world and it will remain impoverished in this way for a thousand or ten thousand or a million years or however long we survive on this planet.

We’re increasingly asked to believe that climate change is a problem that can’t be reliably modeled while climate change plus much more complex economic consequences can be; oh and by the way everything’s fine. This is the essence of Prof Tol’s position and meanwhile he associates with Lindzen whose ideas cannot be reconciled with Plimer’s, whose ideas in turn cannot be reconciled with common sense. Unquantified and unquantifiable positives are included whereas negative certainties are ignored.

Taken together, this obscene distortion of science amounts to denial. The argument can be summarised as follows: Climate change isn’t real unless it is real in which case it isn’t a problem unless it is a problem in which case it isn’t our fault unless it is our fault in which case there’s nothing we can do about it unless there’s something we can do about it in which case it would cost too much money. This is the GWPF position in a nutshell.

Professor Tol is a politician instead of a scientist and no matter what the question is the answer is always that fossil fuels are fine. The sole purpose of his work is the negation of scientific evidence, and every trick in the book is employed to this end.

Financial models are never scientifically rigorous. Never.
They always include guessed at costs and probabilities. We still need such models. Get used to the idea.

@Hugh Sheehy

Financial models are never scientifically rigorous. Never.
They always include guessed at costs and probabilities. We still need such models.

Establishing the lack of scientific rigour was my main purpose apart from examining the validity of the attacks on Stern’s choice of discount rate, so fine.

Get used to the idea.

I’ve no idea what you felt the point of this remark might be.

There was no point to the remark other than that we cannot do without financial models, whether they’re scientific or not.

Comments are closed.