Stern versus Tol on climate change

the BBC website alerted me today to the linked paper by my ex – Free University colleague Richard Tol, who is still an environmental economist but has become somewhat famous since. The paper and the interview makes fascinating and sobering reading. Let me give you some highlights of how he makes mincemeat out of the Stern report which, after the spin put on it by Blair and his team, managed to change public perceptions in Australia:

The Stern Review cites his [Tol’s] work 63 times; but that does not mean he agrees with it.
“If a student of mine were to hand in this report as a Masters thesis, perhaps if I were in a good mood I would give him a ‘D’ for diligence; but more likely I would give him an ‘F’ for fail.”
There is a whole range of very basic economics mistakes that somebody who claims to be a Professor of Economics simply should not make,” he told The Investigation on BBC Radio 4.
Professor Tol believes the figures for damage [of climate change] are exaggerated.
“Stern consistently picks the most pessimistic for every choice that one can make. He overestimates through cherry-picking, he double counts particularly the risks and he underestimates what development and adaptation will do to impacts,” he said.


And the main conclusion of Tol’s paper on Stern is even more damming:

“In sum, the Stern Review is very selective in the studies it quotes on the impacts of climate change. The selection bias is not random, but emphasises the most pessimistic studies. In this sense, the Stern Reveiw reminds one of Lomborg (2001) [skeptical environmentalist]. The discount rate used is lower than the official recommendations by HM Treasury. Results are occasionally misinterpreted. The report claims that a cost-benefit analysis was done, but none was carried out. The Stern Review can therefore be dismissed as alarmist and incompetent.”

Richard also nicely points out that the scare campaign on climate is often based on nonsense. He for instance points out that global food security is not threatened by climate change and that there are obvious adaptations to be made to the very slow moving process of climate change. More particularly, since the change takes centuries, options like migration and changes in the spacials patterns of food and industrial production are easy to imagine and indeed, standard economics. His own estimated total climate cost of carbon emissions after adapations are taken into account is $29 per ton of CO2 which is pretty small when you consider it.

The basic picture Richard paints of the debate is that of a woefully ignorant media and general public who get taken in by selected interpreters of the hard-core science which in turn is much less pessimistic and alarmist than the Sterns and the Gores of this world.

Let me point out here that Richard is by no means a climate change sceptic and never was one. In his linked paper he advocates carbon trading and carbon taxes, and I can personally testify that he believed the climate was changing at least 12 years ago when I had a conversation in an Amsterdam pub with him on the subject. He had the same position as a student then as he now has as a prof, i.e. climate change is happening and there’s hope in carbon trading reversing the trend in fossil fuel use. I too had the same position then as now, i.e. climate change is happening, but we’re simply going to burn up the fossil fuel as fast as we can and live with the consequences. In terms of the outcomes so far, they seem to favour my predictions rather than his but he’d probably argue we should give the idea of trading and taxing more time. The wider public then didnt believe the climate was changing and went about their lives happily maximising their material well-being, whilst the wider public now believes unrealistic scenarios about climate change, puts its faith in a modern version of a raindance to stop it, but is in fact not doing anything of much climate consequence apart from planning more consumption to speed up climate change.

This entry was posted in Economics and public policy, Science. Bookmark the permalink.
Subscribe
Notify of
guest
49 Comments
Oldest
Newest Most Voted
Inline Feedbacks
View all comments
observa
observa
14 years ago

“climate change is happening, but were simply going to burn up the fossil fuel as fast as we can and live with the consequences.”
Watching the ABC show on Sunday night about the world’s most crappy jobs and the hard physical slog of our Victorian ancestors, sure convinced me about that, particularly when I work in the building and construction industry and know what physical labour means and how hard it has become to get Gen Y off their arse to do any modicum of physical work, their ancestors and the third world can only dream about. Rain dancing sure sums up these fairies at the bottom of the garden greenies nowadays.

Bannerman
14 years ago

This business of economically modelling climate change and it’s impacts is as accurate as picking the lotto numbers on Saturday night. Lotto probably has better odds. However, that doesn’t mean modelling shouldn’t be done because it’s outcomes will be inherently inaccurate. If modelling is never done, we’ll never know what we might or might not be up against. At least if modelling is done, assuming a worst case scenario allows us a maximum margin for error. Something perhaps Professor Tol doesn’t take into account with his critique.

SJ
SJ
14 years ago

The Stern Review cites his [Tols] work 63 times; but that does not mean he agrees with it.

The fact that he’s cited 63 times is an acknoweldgement the he’s published a lot and has some kind of expertise, I guess. That doesn’t make him the ultimate authority, though. Tol writes comments over at JQ’s site constantly, and they tend to be of the “Stern’s wrong, I’m right, because I’ve been publishing and teaching this stuff for years”. I might have missed it, but I haven’t seen any post where Tol gives any sensible reasons for his beliefs.

James Farrell
James Farrell
14 years ago

Paul, are you taking Tol’s critique on trust, or have you actually satisfied your self that his arguments are right? There are two kinds of criticisms in the paper you cite: the first, that Stern has a bias towards pessimistic estimates of climate trends and their economic impact; the second, that he makes mistakes in basic accounting. I’m not qualified to judge the former, but as far as the latter is concerned I certainly wouldn’t trust Tol’s judgement without taking a close look. This encounter with John Quiggin was very telling: on the discounting question, he turned out to be all bluff and bluster (not to mention rude).

paul frijters
paul frijters
14 years ago

James,
I am taking some of his arguments on trust, though not all and particularly not the discount rate argument, which is what you are referring to. I agree Richard got a little flustered in his exchange with JQ. He shouldnt have argued on textbook definitions of the revealed market rate of discounting (which is where he was out of depth), but should have stuck to the discount rates used in applied literatures, for instance, in macro models or as internal discount rates for government and business projects.

Richard had a fair point though, which was that the revealed rate according to JQ would lead to a zero percent discount rate, which is so out of whack with mainstream modelling that you’ve got to wonder whether it makes sense to look at the real bond rate. JQ was hence benig a bit silly by truly making a 0 percent discount rate sound like the textbook definition, even though he’d be fully aware that he’s out of line with that assessment. There simply were no competent (and confident!) economists in that blog thread to take JQ on on that point, but this doesnt mean that the real bond rate is a sensible starting point in the debate.
Especially when looking at the long-run it makes more sense to look at the real rate of return on investments because their higher year-on-year risk averages out over the centuries. For the short-run it makes more sense to think of extreme risk aversion of a few as pinning down the bond rate to such a low level, but the yearly real bond rate is is not a fair measure of the risk-free rate in the long-run. Taken from this point of view, once more you’d have to agree that the Stern report made a very selective assumption leading to a ridiculously low discount rate, further strengthening Richard’s critique. I agree this reasoning was beyond Richard’s knowledge of financial economic modelling, but, to be fair, it would be beyond most economists I know. Yet intuitively Richard was more right than JQ who managed to wrong-foot Richard by drawing him into his own area of expertise. You may think that reflects poorly on Richard, but I thought it reflected somewhat shabbily on JQ who temporarily bought into a ridiculous discount rate simply to appear to win an argument.

John Quiggin
John Quiggin
14 years ago

You seem to be a bit out of your depth also, Paul. I’ve mostly referred to the 10-year bond rate, not the 1-year rate (I assume this is what you mean by the yearly rate). The differences between the two are not large enough to change anything greatly.

The fact that much higher rates are commonly used to discount government projects reflects the general tendency to include in the discount rate terms that should really be in the cash flow (underestimates of the probability of project failure). On top of this, there’s the equity premium puzzle, where I’ve been quite clear in saying that no-one knows why the observed equity premium is so much larger than theory would suggest. But there’s no reason to use the equity rate in discounting future climate damage.

As far as I can see, your analysis simply amounts to using the term “ridiculous” to describe a conclusion you don’t like.

Jc
Jc
14 years ago

Tols’ right. Stern had an opportunity to present a decenct report and he screwed it up by applying to low a discount rate for generational accounting.

AGW is real from a risk managment perspective. But Stern didn’t do anyone any favours by not admitting to the error and correcting it.

Not to apply a low time value of money (not the inflation rate) for inter generational accounting is a disgrace. Even if Stern was technically unable to pick up the mistake there should have been others on the panel to warn him.

Now we have thge spectacle of people trying to cover up for this and making it worse.

Jc
Jc
14 years ago

Bannerman:

Economic modelling needs the best possible parameters in order to ascertain the Present value to future value. If you can’t offer a best guess estimate then we shouldn’t model at all.

“At least if modelling is done, assuming a worst case scenario allows us a maximum margin for error. Something perhaps Professor Tol doesnt take into account with his critique.”

Really? So you get up every morning and estimate that the worst possible thing that could happen to you is getting killed in a car crash and therefore don’t head to work?

This is silly isn’t it.

We have ok estimates that we can work with. We continue to upgrade as new information comes in. However to work on worst case scenario is the same as not getting up in the morning because you fear a car crash.

paul frijters
paul frijters
14 years ago

JQ,

:-) so you actually maintain a zero discount rate is a reasonable assumption?

Ok, let’s talk particulars:

– the theoretically best way to find the tmie preference is to deduce it experimentally from individuals so as to exclude all other possible contiminants. The most authoritative general review I know about time discounting is the piece ‘Time Discounting and Time Preference: A Critical Review’ by Frederick, Loewenstein and ODohonue in the JEL, 2002. They report that individuals implicit discount rate when they make inter-termperal tradeoffs in purchasnig decisions can be in a large range, but 10% is the average result. Experimental studies vary in the extreme, but even amongst those 0 percent would be an outlier to the bottom.

– I’ve tried estimating the rate of time preference myself (which is a notoriously hard thing to do because likelihood functions are usually very flat in the discount rate, as Adrian Pagan also showed holds for macro models in his 2006 review (http://cama.anu.edu.au/Working%20Papers/Papers/2006/Fukac_Pagan_102006.pdf)) in structural models for German individuals in my 2006 EJ piece with Bas vd Klaauw. Our estimate was 27% percent per year. A little high I admit, but not unusual at the individual level.

– As Richard already said, most government departments use a much higher discount rate to judge the benefits of a project. You dismiss this by saying these departments do that because they make the mistake of including the probability of failure into it. For one, failure is one-off event with a certain probability and hence its usually different to a rate, but I think more importantly you’re making a pretty big unsubstantiated claim here. It makes no sense to speak of ‘failure’ when discounting the benefits of, say, preservnig a forest as a national park. Were you trying to bluff again?

– You do not at all address the simple point I made that the best market discount rate to look for is not the 1-year or even 10-year bond yields (which dont differ much indeed), but rather the rate of return on very long-run investments, i.e. buying stocks and holding on to them indefinitely. Since the risk surrounding that average return also goes to zero over the decades, the rate of return to that IS the risk-free long-run rate of return. Since you cant buy bonds forever, theyre not the appropriate investment to look at for the time horizons involved in climate change.

– As to the equity premium puzzle, since you admit being out of your depth on that one, let me simply refer you to the nicest explanation I know of, i.e. Campbell, J.Y. and Cochrane, J.H. (1999). By force of habit: A consumption-based explanation of aggregate stock market behaviour. Journal of Political Economy, vol. 107, pp. 205-251. Their stylised story is that extreme short-run risk-aversion causes the low short-run risk-free rate of return, but that this does not translate into long-run time preference because of habit effects. That explanation has since been adopted and expanded by many, though the basic idea originated with Constantinides in 1991.

paul frijters
paul frijters
14 years ago

Jc,

I am in full agreement with you on the issue of how wrong it is to assume the worst. I dont like the spectacle of people covering up for using such a low discount rate either though I will try and JQ some credit for presuming he’s not trying to be the cheerleader of the climate scare campaign by covering up for bad practise.

Jc
Jc
14 years ago

Thanks Paul

Oh my… Should read:
To apply a low time value of money (not the inflation rate) for inter generational…….

observa
observa
14 years ago

Pardon me for interrupting this enthralling discussion on the finer points of inputs into long term mathematical modelling on costs, but the trogs are knocking on the door again
http://www.telegraph.co.uk/news/main.jhtml?xml=/news/2007/07/22/nbook122.xml

John Quiggin
John Quiggin
14 years ago

There are so many errors here that I really don’t know where to begin. To start with, the discount rate is not the pure rate of time preference, since discounting is primarily driven by income growth and diminishing marginal utility. Furthermore, any direct translation from individual to social time preference is highly problematic as pointed out, by, among others Guo, Tol et al who note the existence of a strong case for using either zero or a very low rate of time preference in social decisions.

On the equity premium, no one who’s familiar with this literature would claim, as you do, any one of the dozens of proposed resolutions that have been put forward is a definitive answer. All have big problems, and none fully accounts for the observed data. An obvious problem with your interpretation of Campbell and Cochrane is that the problem should go away if you look at 10-year or 30-year bonds instead of short-dated bonds. As you know, it doesn’t. The critical question is not the term of the investment but whether you should look at bonds or equity. You haven’t given any reason for suggesting that equity is the right choice.

On discounting procedures for government (and private) projects, and I suggest that you read the large literature on this, starting with Megaprojects and Risk by Flyjvberg et al.

Jc
Jc
14 years ago

Ummmmmm
One of the most renowned experts on the economics of climate change, Richard Tol, produced a detailed, four-page critique of the Stern report , saying that it uses only the most pessimistic impact studies, starts from a too-low discount rate and has no real cost-benefit analysis. Tol therefore called the report “alarmist and incompetent”. “This is not to say that climate change is not a problem, nor that greenhouse-gas emissions should not be reduced. There are sound arguments for emission reduction. However, unsound analyses like the Stern Report only provide fodder for those sceptical of climate change and climate policy,” Tol concludes.

http://www.euractiv.com/en/sustainability/stern-report-alarmist-incompetent/article
-159346

Let’s what are some of the reasons:

1.Let us first examine the Stern Review conclusion that climate change will cause economic disruption now and forever. The now and forever is preposterous. The world economy is growing briskly; immediate threats to economic growth are imbalances in the US, overheating in China, and lack of reform in the EU. But the forever part is also problematic. It assumes that society will never get used to higher temperatures, changed rainfall patterns, or higher sea levels. This is a rather dim view of human ingenuity. It contradicts what we know about technological progress, adaptation, and evolution.

2. Tol then mentions that no cost benefit analysis was done despite Stern suggesting so.

3.The Stern Review highlights several impacts of climate change. One is water. The work here is based on Arnell (2004). The Stern Review correctly that Arnell (2004) does not include adaptation and is therefore severely biased.

In other words Stern goes pretty close to assuming an almost stand still rate of tech development which is a startling consclusion.

4. It over estimates the adverse effects of AGW while underestimating the benefits of emissions reductions.

5. Conclusion
In sum, the Stern Review is very selective in the studies it quotes on the impacts of climate change. The selection bias is not random, but emphasizes the most pessimistic studies. The discount rate used is lower than the official recommendations by HM Treasury. Results are occasionally misinterpreted. The report claims that a cost-benefit analysis was done, but none was carried out. The Stern Review can therefore be dismissed as alarmist and incompetent.

http://www.euractiv.com/29/images/sternreview_tcm29-159365.doc

Stern sees increasing hurricane damage in the U.S. as a powerful argument for carbon controls.

Stern may want to discuss his concerns with Chris Landsea who resigned from the IPCC because of his concerns that the IPCC was offering conclusions about Altlantic weather patterns when Landsea sees no case to be made with the evidence on hand.Landsea is the world authority on Atlantic storm activity. In fact more than one study has suggested that the US would actually benefit from AGW due to the mitigation of severe winters.

Re the discount rate. Stern is inferring through his low discount rare that present day goods have equal value to future goods. Can anyone borrow money at close to zero for 100 years. If they can, I’ll take all you can offer. Just dump it on the front law if you have to as I don’t care. Seriously I’ll take 100 year money at zero… all you can get.

Jc
Jc
14 years ago

Observer

Very interesting piece……

The UK’s current wheat production is 11 million tons (against our consumption of 10 million). To meet the 10 per cent target by 2020 from wheat alone would require us to grow 14 million tons of wheat a year, 3 million more than we currently grow. World demand for wheat is rising so fast that, in the past two years, a global surplus has become a deficit.

———————-

The piece also mentions that the world wheat surplus is turning into a deficit. Meanwhile across the pond Bush’s “smart” policy (due to the notion of killing two birds with one stone-national security and greenie pressure) of increased ethenol blend is causing the price of corn to tighten.

Are our state governments still banning the use of GM crops? Yep!

Global Deforrestation reversal anyone? Consequences?

I’m now starting to think with the famous bet that guy had with Bill Simon- at least about food shortages -was right but wrong timing. There could be food shortages because we’re burning it all up in our cars as a result of government policy. It’s almost too sad to laugh.

Yep. That road to hell is well traversed. Let’s have more of this policy without thinking two steps foward.

paul frijters
paul frijters
14 years ago

JQ,

I see you’re not interested in debating any point I raised seriously, but persist at sly snides like ‘so many mistakes, I dont know where to begin’, which is followed by a complete lack of substantive argumentation. You did not nail your own claim that the much higher discount rates advocated for this kind of thing, for instance by the UK Treasury as referred to by Richard, came from misunderstanding economics. You did not bother to seriously debate the quite sensible solution to the equity premium puzzly by Cambell and Cochrane but simply point out that their simple model doesnt fit everything perfectly (simple deviations from it can fit the longer-term bond rate). You didnt argue for bond rates as the natural starting place for very long-run discount rates (which they shouldnt be: many of them are bought by governments for political reasons, so to pin climate discounting on them is silly. Furthermore, they are subject to default and hence not risk-free). In fact, I could have written your reply in 5 seconds.
I must say, JQ, you disappoint me by being so irresponsible in the climate change debate. Cheap point scoring should never take precedent over serious reflection and reasoned assessment. A 0 percent discount rate is ridiculous.

observa
observa
14 years ago

Yes JC, the concerns expressed in that article, particularly wrt biofuels and food have to be of major concern. WRT the other technical discussion here I was reminded of a time many years ago, when studying economics, about the dangers of becoming so esoteric as to become almost meaningless. I was studying a BEc at Flinders and scored a distinction for a macroeconomic essay on some topic, which was rare indeed from that faculty at the time. So much so that a number of other students wanted to read it to see what a distinction eco essay was really all about. After half a dozen had read it and looked at me rather strangely as some kind of alien, I reread it myself. Now there was some time lapse since I’d wrote it, but I was gobsmacked at not understanding what I’d written so much so that I didn’t even recognise my own work. I had been right into the subject at the time and obviously it had provoked some new minor nuance of understanding to cause a stir at cuppa time for the eco faculty and hence the I deduced a distinction was awarded. However it was now gobbledygook and meaningless to both myself and my peers. It was then I realised that you can get so wound up in esoteric nonsense as to become meaningless, albeit that is often the way blinding new truths can become apparent.

That’s my problem with this discussion here, (although I have only a cursory knowledge of the thread of it) and with JQ’s insistence that the 60% reductions in fossil fuel use will cost only a nett 1% of world GDP growth, all things considered, discount rates, yada, yada. It just flies in the face of any reasonable, practical observation of the world around us, as the article I alluded to does. The grain to biofuel figures problem, simply mugs you with practical reality of the true costs of what is being proposed.

observa
observa
14 years ago

I should add that the overall conclusion I reached at university was this- Economics has some basic and very important analytical truths to offer, largely tied up in the marginal analysis of those supply and demand curves, which I note many a successful, small businessman who left school early, has an innate and intuitive understanding of. Apart from that, none of its collected wisdom could predict the stagflation of the 70s, nor can it accurately predict the next overdue recession, nor the weather next week.

Majorajam
Majorajam
14 years ago

paul frijters,

I’ll give you an argument why the equity rate is inappropriate, or at least borrow one: there is little reason to believe that returns to climate change spending are perfectly or even highly correlated with income, given the types of damages threatened, for example to land usage. That fairly self-evident proposition calls for a money discount rate at the very least in the direction of real bond yields, and, given the uncertainty around the estimated correlation coefficient, more still.

Another major issue with what you have written concerns equity returns. The argument their return premium over equities relates solely to ‘extreme short term risk aversion’ that is beyond investor’s wee brains to arbitrage is not coincidentally, (neatly having been published at the tail end of a near 20 year bull market in equities), very much akin to the argument that underpinned “Dow 36,000”, the prognosticators of such an arbitrage. So much for the equity return puzzle. One would think the fact that the assertion is manifestly at odds with investment behavior as practitioners and actuarial consultants know it, to say nothing of the experience of a generation of Japanese investors, would be relevant. Then again, I’m not an economist (to prove it, I occasionally post over at Professor Quiggin’s blog).

More generally, the idea that asset return data, which is to say 100 years of US asset return data, is not heavily afflicted with selection bias is laughable and, if you believe some of the scarier prognostications of observers of credit dynamics, ripe for a paradigm shift.

PS Also fyi, the tenor of the most commonly issued government bonds are far less relevant than you appear to believe, to the point of irrelevance. A quick analysis of the interest rate sensitivity of a 30 year issue and a perpetuity should demonstrate that sufficiently.

Jc
Jc
14 years ago

“More generally, the idea that asset return data, which is to say 100 years of US asset return data, is not heavily afflicted with selection bias is laughable and, if you believe some of the scarier prognostications of observers of credit dynamics, ripe for a paradigm shift.”

Why is that, majorjam. As an asset class bonds have been the worst performers in the US since the 29 crash while stocks have been the best performers.

Patrick
Patrick
14 years ago

if you believe some of the scarier prognostications of observers of credit dynamics, ripe for a paradigm shift.

Which ones? Do you know what a paradigm shift is? It sounds like you are talking about Lyndon LaRouche.

James Farrell
James Farrell
14 years ago

Paul, John has posted on discount rates four or five times since Stern came out, and argued his case in many a long thread, so I’m not surprised that he wouldn’t want to pursue it at length here.

I am not any authority on this topic, but I must say I haven’t encountered any good reason why it would be appropriate to apply a (significantly) positive pure rate of time preference in greenhouse policy analysis. The Frederick et al. paper is very interesting, especially the way it calls into question the very notion of a single psychological parameter that captures time preference. But I can’t see the obvious relevance of the empirical estimates of the rate, based on observations of decisions by itinerants and frenzied shoppers, to what is largely a question of intergenerational equity.

As for the equity premium, you evidently have your own preferred theory as we all do, but John is quite right to say there is no consensus position on it.

My mind is not made up, so, if you could cite me one paper or text that best argues the theoretical basis for your preferred rate of discount for cost-benefit analysis of public projects and regulations, I’d be very happy to consider it.

Finally, I must say it would be helpful if you were more precise in your terminology. You say “A 0 percent discount rate is ridiculous”, though you’re perfectly aware that it’s the pure rate of time preference that Stern and Quiggin want to set at zero, not the discount rate. As you’ve noticed there are plenty of opinionated participants in the discussion who don’t understand these concepts, and it’s not helpful encourage them in their misconceptions.

Majorajam
Majorajam
14 years ago

Patrick, I’ve used the term paradigm shift not in a Kuhnian sense but in a marketing mumbo jumbo one, hoping, forlornly as it turned out, that I would be given enough credit by the reader for that to be obvious in context. The biases inherent in the US series I think can be rather easily guessed at. I’ll give you a hint: in the past 40 years, government bonds have outperformed stocks in Brazil.

JC- talk about selection bias! You’re even revising out the greatest stock market failure of the survived series. Anyway, I take it this is beside your point. To that, I think you will find it difficult in the historical record to find periods of free falling bond prices and sanguine equity markets. In other words, there are inferences there that, while IMO not particularly cryptic, have been left out.

Majorajam
Majorajam
14 years ago

James,

The PRTP that Stern uses is not 0, just close to 0. The larger error that Paul makes is conflating an individual’s PRTP with a social one, which might account for the word ridiculous.

Jc
Jc
14 years ago

James

Maybe you would like to explain to us mortals why… let’s use and example….

Why the present model car you wish to purchase this weekend has equal vlaue to you than a car that is four model cyles away. You’re valuing them with equal weighting aren’t you under your thesis if time your preference is zero.

Maybe I’m wrong but I would be interestd to see your answer. it would be great if you could refer to this example so others understand.

I’ve seen the professors argument and he tries to get around this by arguing that each generation should be separated from the others in a sort of way. I think i’m right here. However the hole in the argument is that earlier generations are actually going to spend the money to make life far more comfortable for later generations (100 years for now) who will be much wealthier than we are both from a monetary perspective and acceleration of technological change.

James Farrell
James Farrell
14 years ago

Majorajam,

I think we’ve all implicitly agreed to call 0.1 percent zero. As for Paul’s error, that’s what I was trying to say in my second paragraph (though I think these are tricky issues, and I’m not an expert). However, my point in quoting that sentence is that I wish Paul had said PRTP, not ‘discount rate’, which only confuses people.

Patrick
Patrick
14 years ago

This debate never really seems to be about economics. I think JC is far closer to what almost always seems to me to be the underlying point of difference – some people such as he, and I, are convinced that Julian Simon was right – broadly speaking everything gets better everywhere all the time – whereas others are convinced that, broadly speaking, our ‘current’ ways are unsustainable/sure people are getting richer, healthier, longer-lived etc but ‘really’ they are just making themselves unhappy/bref, the end is nigh.

Majorjam, it was a forlorn hope as far as I a, concerned. I am a pretty literal person, though, so don’t take it personally. I am glad you meant it that way, but I do wish you could have simply said what you meant :)

James Farrell
James Farrell
14 years ago

“….broadly speaking everything gets better everywhere all the time…”

Yes, and Stern’s argument (broadly speaking) is that this improvement is the only significant consideration in choosing a discount rate. There shouldn’t be an addition of some extra percentage points reflecting the fact that we are (broadly speaking) impatient.

James Farrell
James Farrell
14 years ago

Joe, I appreciate the fact that you’ve been polite and pleasant on this blog, but I haven’t forgotten all the insults and ridicule you’ve directed at me and other people elsewhere. Until you announce a change in your style, here and anywhere else you comment, I’d rather not interact with you.

Jc
Jc
14 years ago

James

Horse for courses. I will behave exactly as the comments policy demands. Get over it and don’t hold grudges. You shouldn’t hold grudges from one site to another as it’s bad form. I always wear my tux here anyways. Look and see how well behaved I am here. I even amaze myself. Think of it as a Jeckell and hide personality disorder and try and deal with it.

By the way it’s not as though i have been on the receiving end too you know.

Majorajam
Majorajam
14 years ago

JC,

The PRTP discounts utility. It’s being non-negative or zero implies that all that great stuff 100 years from now is worth less than the same great stuff would be to the current set of air breathers. In other words, there is a hole in the hole you speak of and it’s opaque.

Patrick,

Fair enough. I did say what I meant, given that both usages are in the lexicon, but admit to forgetting that amongst educated types how they got there matters. Not to mention, the sentence isn’t very well written.

This debate is indeed not about economics, but rather a never ending string paradoxes- we should have no problem adapting to climate change, but adapting to life with less fossil fuel is a bridge too far. Society can be myopic and by consequence act in ways not always consistent with its long-term betterment. Because of that ‘revealed preference’, the hurdle on public works projects that benefit society in the long-term ought to be so high that they not be undertaken. The list goes on. In my view, the optimists in this debate are on the side of action, so I don’t agree with your framing of the issues.

John Quiggin
John Quiggin
14 years ago

Paul, apologies for losing my temper on this one. My interactions with Richard Tol on this have probably made me oversensitive. James Farrell does a much better job of explaining my position at #22. As well as the blog posts, I’ve written quite a few papers on this topic and on discounting for public projects (I support the same view here as in the Stern debate, and explain the point about cash flows and uncertainty). If you’re interested I can send you copies.

Richard Tol
Richard Tol
14 years ago

#3: SJ — there seems to be a gap in your knowledge — Gary Yohe and I published six detailed critiques of the Stern Review, none of which use the argument that he and I have 40 years of experience between us, compared to Stern’s 18 months — Dasgupta, Maddison, Mendelsohn, Nordhaus, Pielke and Weitzman also published critiques that were more or less detailed, and raised issues that Gary and I did not — and yes, all of these people have a track record in the economics of energy, environment and climate that Stern does not have, but none used that as an argument

on the utility discount rate, I just reiterate my position: if John Quiggin can provide empirical evidence for his outrageous claim, I will nominate him for the Nobel Prize — I am sure he has not, and I don’t think he ever will, but I always like to be surprised

Patrick
Patrick
14 years ago

Majorjam, it seems to me that you just open up another two related schisms – the extent to which one thinks that society is or is not myopic and the extent to which one thinks that this should be ‘cured’ by an exercise of some third party’s judgement.

I remain convinced that the technical arguments are, in this case, merely skirting fundamental ‘prejudices’. I remain hopeful that, if those prejudices were more clearly acknowledged, we would be much clearer on the technical positions adopted.

For example, in the case at hand, JQ and you clearly believe that JC and PF are overdiscounting and v-v. Primarily, this appears to me to be not a technical disagreement about the appropriate discount or time value for these purposes, but your different ideological/intellectual conceptions of discounting. Ie, I think you all understand discounting as technically achieving similar goals, but on different rationales, which underscore your respective opinions as to the appropriate rates.

Of course, perhaps I am wrong, and should leave the debate to those who actually understand the technical stuff sufficiently :)

paul frijters
paul frijters
14 years ago

JQ,
same here. I too lost my temper.

James,
I agree that many terms are often used interchangeably and that, in order to further this debate here, it would be better to more clearly separate them. What we are interested in at the end of the day is whether we can improve time-preference-discounted utility.
There are two ways of looking at paying for climate change prevention:
. by paying something with a loan that bears a certain real rate of interest to be paid back at the future date. Since the economy grows between now and the day we have to pay back the loan, the marginal utility of the reduction in future consumption may be less than the marginal utility now of less consumption.

paul frijters
paul frijters
14 years ago

James, (something went wrong with my server whilst typing, so here goes again…)

I agree that many terms are often used interchangeably and that, in order to further this debate here, it would be better to more clearly separate them. What we are interested in at the end of the day is whether we can improve time-preference-discounted utility.

There are two ways of looking at paying for climate change prevention:
1. by paying something with a loan that bears a certain real rate of interest to be paid back at the future date. Since the economy grows between now and the day we have to pay back the loan, the marginal utility of the reduction in future consumption is less than the marginal utility now of less consumption. Hence the ‘composite discount rate’ to be applied on the future loan pay-back is less than the pure rate of time-preference by a non-linear factor depending on risk-aversion (only by approximation will it be linear, but in such long time-frames this approximation is not very good). If you add other-regarding preferences to the mix (such as in Cambel and Cochrane) the issue becomes even tougher. This approach leads one to have to first find the pure rate of time preference, which consisted of the majority of my response. The ‘composite discount rate’ is much trickier.

2. We simply give up consumption now in order to prevent climate change in the future, with no loans involved. This is a more world-view because it recognises that loans have to come out of intended consumption by someone. Then the issue of discounting becomes much simpler: costs now are in dollars today and benefits are in utility-equivalents discounted by the Social rate of time preference.

Yes, we can argue about the social rate of time-preference versus personal rate of time-preference. We can potentially measure the personal rate, but the social rate is at the end of the day a social choice. Hence the relevant social rate is the consensus rate used for such wide projects by our chosen representatives and their departments, like the UK Treasury. Much, much, bigger than 0 (give or take minimal rounding).

Majorajam,
I fully agree with much of what you say about the complexities of investment behaviour, but they are largely irrelevant for my argument: when looking for the risk-free rate of return in the long-run relevant for global events, one tries to answer the question ‘what investment pattern could the world make to get the highest risk-free return in 2 centuries time’. My smilpe contentiion is that that wuold be to spread investments over global equities, which do not run the risk of government default (like the Russian Tran-Siberian bonds!) and whose large year-on-year risks average out to zero over the long run. Sure, there is still risk but the higher average returns dominate everything: the odds of average returns on the equity spread going below the bond return for one country becomes 1 in the long run if the only risks one allow for is government default and imperfectly correlated equity risks.

paul frijters
paul frijters
14 years ago

Richard,
nice to have you visit :-) Do these other economists you mention (whom I’ve not read on this issue, except for Maddison who is not on your list but is in your references) also make the same point about discount rates?

paul frijters
paul frijters
14 years ago

Sure, there is still risk but the higher average returns dominate everything: the odds of average returns on the equity spread going ABOVE the bond return for one country becomes 1 in the long run if the only risks one allow for is government default and imperfectly correlated equity risks.
oops.

Jc
Jc
14 years ago

“To that, I think you will find it difficult in the historical record to find periods of free falling bond prices and sanguine equity markets.”

Really? Try Zimbabwe the last 12 months. Bonds are worth less than the paper their printed on and the stock market has risen about 1.2 million%. May not be sanguine, but on a relative basis a little more cheerful than holding bonds in that part of the world.

Majorajam
Majorajam
14 years ago

Patrick,

I’m just trying to get around the unhelpful jargon. It’s more meaningful in my view to think of the PRTP as the pure rate of societal myopia or impetuousness than ‘time preference’. In fact, I agree it is difficult to argue from empirical data that such a thing does not exist. This means that Stern’s choice of a PRTP of more or less zero left him exposed to reasonable criticism, (although, IMO, the critics who’ve picked up on this tend to be entirely unreasonable in their high-handedness; even one who, in an astonishing feat of cognitive dissonance, acquiesced to good arguments for Stern’s chosen parameter value in his published work prior to his professed outrage, i.e. he voted for it before he voted against it.). So, while, according to the rules of this game, I certainly don’t see how one could justify using an equity risk premium to discount climate change returns, I wouldn’t count myself as backing Stern’s formulation or anyone else’s for that matter. The only thing I can say with any semblance of conviction is that prohibitive values of the PRTP are paradoxical and that the degree with which economists take leave of these issues is farcical, (as are some of their other proclivities, e.g. unquestioning acceptance of financial return data).

But I take your point- to some degree, the climate change debate is a rhetorical inkblot. I also feel though there is plenty of room left over for the evidence to color or even overwhelm one’s preordained position. For my own prejudices, I’ll come clean: I believe that in spite of the abounding sea of data and analysis, there are only a tiny few pieces of information and conclusions that actually matter in this debate. The rest are white noise. The best example of the latter are estimates of the cost of climate change, even abstracting from the uncertainty in accounting for a distribution of temperature inputs, its interactions with cost drivers, with temperature inputs, with interactions, etc. etc. This to me is about as valuable as pinning a tail on a donkey, only far more counterproductive- imagine the donkey is set against an electrified fence- and, frankly, corresponding with Richard Tol in the blogosphere has only served to harden this prejudice. I believe further that this overwhelming uncertainty extends beyond the economics of climate change to the science as well but weakens substantially as it arrives.

Which brings me to the pertinent evidence: the climate is warming as a result of increased concentrations of Green House Gases. C02 is a greenhouse gas with a demonstrable effect on the climate, its concentrations have risen and are rising in an unprecedented fashion as are global temperatures in ways consistent with the postulated effect on climate of the increased concentration of C02. I think this much is indisputable and Occam’s razor requires reasonable people to conclude that this quacking waddling feathered thing with a flat beak and webbed feet is, indeed, a duck. This is a serious cause for concern as we are familiar with the tragic ramifications that can result from a changing climate without attributing a cause. That said, the largely uncontroversial, (outside the counterrevolutionary community), evidence required for this conclusion is insufficient to inform policy, so we’re stuck in the realm of the useless, i.e. economics. But there remains one other piece of evidence of a sort that points to a way out, not to mention, accounts for my position: the uncertainty itself (or my prejudicial interpretation of it). The uncertainty of climate sensitivity and feedbacks, of damages to the habitability of the only planet we have, even of discount rates and the utility of insurance. It is this uncertainty, (as Weitzman puts it, according to IPCC forecasts a two standard deviation event would imply more temperature change than currently separates us from past ice ages), together with the instinct to survive instinct that demands action, even if there exists absolutely no way of having any good estimate of the effect on utility of that action.

Perhaps all of that also means that Stern’s philosopher king amateurism was more enlightened than we know.

Richard Tol
Richard Tol
14 years ago

#37

Paul,

John Q is the only economist of name that I know who argues that the observed pure rate of time preference is near zero.

Most (but by no means all) economists that I talk to appreciate the ethical argument that the social pure rate of time preference should be zero or thereabouts, but they also tend to think that ethical arguments are largely irrelevant.

Majorajam
Majorajam
14 years ago

Paul,

To be honest, I’m not sure I follow. For many investors, notably ones with very short term liabilities like property and casualty insurers, banks, etc. equities are the ultimate in riskiness. When we talk about ‘global investors’, are we excepting these intermediaries? Pension funds that have little to no surplus might consider investing substantially in TIPS or index-linked bonds- is this decision unwise? I should also point out, default happens with equities just as it does with bonds- it just goes by another name: bankruptcy. What’s more, deflations are devastating of equities, while government bonds are largely immune (and actually benefit), while fraud and or Yukos style government malfeasance most often afflicts equity investment. The point is, or should be, that bonds exist because they are efficient for both buyer and seller in the real world, and not only for dupes. Equities may outperform in the majority of the historical record, but as history has a knack of pointing out, that record is incomplete.

JC,

Does that mean you are forecasting hyperinflation striking the US anytime soon? Btw, we have another example of the lack of correlation in stock and bond markets in post WWI Germany. In that case, the market was strong even in US dollar terms though it ended badly for shareholders circa 1945. All of which goes to show, never underestimate the ability of a pyramid scheme to look like an equity market even as the world falls apart around it.

Majorajam
Majorajam
14 years ago

Richard,

I would assume the dearth of economists making arguments on this subject reflects the few number of immortals about from whom to glean data. Be that as it may, IMO, this remark is a cop out, even if it represents the consensus of the league of economists. Using an ’empirical’ which is to say average prtp across time and space implicitly endorses the ethical ramifications of said, not least in the real world where people use bottom line summarys to make policy decisions (or, as is more often the case, justify policy decisions made).

John Quiggin
John Quiggin
14 years ago

To restate my position on the empirical evidence, taken as a whole it shows that people don’t act as if they have any consistent rate of time preference. You can find situations where the apparent rate of time preference is very high (and also where implied levels of risk aversion are very high). You can also find market evidence, namely that of bond markets, which implies that the pure rate of time preference must be very low. All of this is based on the auxiliary assumption that we can apply expected utility theory with time-separable preferences. Once you drop this auxiliary assumption, it gets very tricky to derive social judgements from observed individual preferences.

As regards the conduct of the discussion, readers might wish to check James Farrell at #4 and then note the tenor of Richard Tol’s comments, such as #33.

jc
jc
14 years ago

Majorj

Don’t be silly. You asked my to show an example of a falling bond prices and a rising stcok market and I gave it to you, albeit an extreme example. There is really no correlation between bond pirces and stocks in terms of directional plays in modern economies.

John:

Tol’s behavior is fine. You’re just being supersenistive to being wrong and getting found out. Hey, we all make mistakes and I always admit to them if I’m wrong.

To be perfectly honest, I may not be the brightest spark that ever hit the universe, but I’m also not the dumbest. I read , re-read and read again your comment 44 and cannot make head or tail of it. It would be helpful if you could try to use less jargon next time so that mere plebs like moi are able to understand.

In sum to argue there is no time prefeence over a 100 year gap is to deny the obvious…. there is, huge amounts of it.

Let me suggest this. Tol seems to be serious about you going for a Nobel prize.
If you’re too shy about it send me all your stuff and I’ll present it as my work. We then split the fee (prize) 50/50….. Okay 60/40 if we score the jackpot. Don’t worry I’m great at making speeches in public so I won’t embarrass you any.

Then again you could go for it yourself and possiblly share the podium with ALgore when he gets his. I think you would scub up well in a tux anyway and if you really want I’ll come over to Stockholm to offer moral support.

Richard Tol
Richard Tol
14 years ago

#45

John Q: Thanks. Your current position, that the data allow for a range of interpretations, is a very sensible one. That the range includes a few extremes on either side is as would be expected.

paul frijters
paul frijters
14 years ago

time to wrap it up. As indicated by the 2002 JEL article on the subject, rates of time preference are indeed noticed to have a wide spread. This should be read as that article reads it: the model of individuals ratinoally discounting all future events at the same rate is not a completely valid descriptino of how people make choices. Presuming a fixed rate of time-preference abstracts from greater underlying complexities. Once one makes that abstraction though, one is after the ‘usual’ rate of time preference in these debates. That most certainly does not hover around zero, making JQ’s position untenable if he wishes to draw on experiments as his support.
If we’re thinking of revealed preference (such as the rate my own estimate referred to), then probably the best ‘open’ indication one can have is the time at which individuals stop investing in their education even though an additional year of education still raises total life-time income. Typical rates of return on education are around 8-10%. If we substract the growth rate of income from that (which is a generous correction factor for risk-aversion), then 6-8% is the ‘typical’ rate of time preference which is indeed close to common practise.

Majorajam,
sure, risks on equities are high in the short-run but as I said before, all these risks wash out in the very long-run if one bets on a world-spread.

Majorajam
Majorajam
14 years ago

JC,

I do try to keep the responses earnest, but going from an assertion about US data to Zimbabwe proved difficult to take seriously. Also btw, your assertion about the relationship between stocks and bonds runs counter to the basic tenets of the capital asset pricing model, the view of the investment community and academia, and for that matter, current events.

Majorajam
Majorajam
14 years ago

Paul Frijters,

Until you differentiate between individual and societal time preference, (the education analogy applying only to the former), your assertions on this subject will continue to be less than persuasive. Also, risk may wash out in time series data, but if you’ve gone bust and/or lost your job in the meantime, the technicality will not put bread on the table.