Over at Econolog, TCS writer and right-wing hack Arnold Kling reports, under the headline “The Stern Swindle”:http://econlog.econlib.org/archives/2006/11/the_stern_swind.html , that Cambridge economist Partha Dasgupta “criticizes the Stern report”:http://www.econ.cam.ac.uk/faculty/dasgupta/Stern.pdf for applying a very low discount rate to the interests of future generations. Kling writes:
bq. What Dasgupta is saying is that the approach Stern uses to evaluate intertemporal trade-offs would, if applied generally, suggest that our consumption should drop from over 80 percent of GDP to 2.5 percent, in order to leave the target legacy to our children. What Dasgupta’s comment does is crystallize for me the magnitude of the intellectual swindle that Stern is attempting to pull off. Any time you assign a far-from-plausible interest rate to a long-term intertemporal problem, you get distorted results.
What Dasgupta _actually says_ :
bq. I have little problem with the figure of 0.1% a year the authors have chosen for the rate of pure time/risk discount….(p. 6)
Dasgupta’s real argument is that Stern shouldn’t adopt the egalitarian approach it does to intergenerational well-being whilst being _at the same time_ indifferent to inequality among members of the present generation. Dasgupta thinks the well-being of the actual poor should take priority over climate change abatement. Of course, we’ve heard arguments along these lines before, but Dasgupta, as someone with a record of concern for development and the well-being of the global poor, is someone who should be taken seriously when voices them and might be expected to devise and support policies that benefit the worst off. Right-wing hacks, are, needless to say, a different matter.
(Dasgupta’s critique seems to me to support the idea that economies like China and India shouldn’t be pressured into climate change abatement because the value of the benefits their growth brings to the poorest outweighs the harms to future generations. It doesn’t look anything like so plausible to claim that the least advantaged would be similarly harmed by the wealthier countries cutting back their carbon emissions.)
{ 5 trackbacks }
{ 57 comments }
David Kane 11.30.06 at 10:14 am
Just the sort of fair-minded commentary that we have come to expect from the Little Green Footballs of the academic left! Since Greg Mankiw (Harvard) and Bill Nordhous (Yale) seem to come to the same conclusion as Kling (that the Stern Report’s use of discount rates is highly suspect), perhaps the “right-wing hack” takeover of Ivy League universities is almost complete. Congratulations to Karl Rove!
aaron 11.30.06 at 10:19 am
Ummm. Chris this post is incomprehesible. WTF is the point you are trying to make.
Chris Bertram 11.30.06 at 10:33 am
Mankiw and Nordhaus may have come to that conclusion about Stern’s use of the discount rate, David. My point is that that isn’t Dasgupta’s argument, as you’d discover if you read his piece. Kling misrepresents him.
John Thacker 11.30.06 at 10:37 am
Dasgupta’s real argument is that Stern shouldn’t adopt the egalitarian approach it does to intergenerational well-being whilst being at the same time indifferent to inequality among members of the present generation.
No, it is not. You have it wrong. From the very next sentence to what you excerpted from Dasgupta’s paper:
But the figure they have adopted for eta – the ethical parameter reflecting equity in the distribution of human well-being – is deeply unsatisfactory. To assume that eta equals 1 is to say that the distribution of well-being among people doesn’t matter much, that we should spend huge amounts for later generations even if, adjusting for risk, they were expected to be much better off than us.
To give you an example of what I mean, suppose, following the Review, we set delta equal to 0.1% per year and eta equal to 1 in a deterministic economy where the social rate of return on investment is, say, 4% a year. It is an easy calculation to show that the current generation in that model economy ought to save a full 97.5% of its GDP for the future! You should know that the aggregate savings ratio in the UK is currently about 15% of GDP. A 97.5% saving rate is so patently absurd that we must reject it out of hand. To accept it would be to claim that the current generation in the model economy ought literally to starve itself so that future generations are able to enjoy ever increasing consumption levels.
He is NOT saying what you claim at all. He is claiming that Stern IGNORES the egalitarian approach to intergenerational well-being. He is saying that setting eta to 1 completely ignores the differences in intergenerational wealth.
You’re confused about the difference between eta and the discount rate.
Chris Bertram 11.30.06 at 10:42 am
No, John, I don’t think so, my understanding was drawn from what Dasgupta says on p.4:
bq. The Review assumes that delta ought to be set equal to 0.1% per year, which is a very low figure if we are to compare it with the values advocated by other climate economists (see below). This is to adopt *a very egalitarian attitude across the time dimension* . But curiously, the Review adopts *a very inegalitarian attitude with regard to the distribution of well-being across people when futurity is not the issue – for example, when comparing the well-beings of the poor and rich in the contemporary world*. The Review’s central case is based on the assumption that eta ought to be unity, which, as I show below, reflects a fairly indifferent attitude toward equity over the distribution of well-being among people, qua people.
John Thacker 11.30.06 at 10:45 am
What Dasgupta argues is that eta, while also a measure of how we treat inequality among persons in the present, produces a different result with regards to intergenerational effects because the people in the future are not the same as the people in the present.
The discount rate, by itself, makes sense for one person comparing their present wealth and future wealth– assuming that they don’t die in the meantime. If they do, then the rewards in the future go to their heirs rather than themselves. If people were immortal, then Dasgupta would have little trouble with the Stern Report. (Though, conversely, he also has little trouble with Nordhaus’s assumption of a higher discount rate as well; he is simply willing to accept either assumption as a reasonable one.)
However, since people are not immortal, when we look at events whose effects will happen over multiple generations, eta affects the intergenerational transfers.
It is all very well to say that if I know I will be alive in 50 years, then I am indifferent between $100,000 now and $200,000 then. But the question is different if I know that I will be dead in 50 years. In that case, it is a choice between $100,000 now and $200,000 divided amongst my heirs. In the latter case, it matters to me how rich my heirs are expected to be in the future. Will they be wealthy, and the $200,000 mean little to them? Or will they be poorer than me, and the $200,000 be much more useful to them than to me?
aaron 11.30.06 at 10:52 am
All Kling is writing about is what can be inferred from Dasgupta’s analysis. He’s not putting words in his mouth.
John Thacker 11.30.06 at 10:54 am
Dasgupta’s essential critique, then, is that setting aside inequality concerns biases the time concerns as well in a manner not considered by Stern. The Stern Report wanted to answer the question “ignoring the effects of inequality.” However, that leads to its own problems. The people who will inherit the savings in the future are not the people who do the saving in the past.
If you think, for example, that exhausted supplies of fossil fuels will mean that our descendents will be poorer, then it makes sense to save more now. On the other hand, if you believe that advances in technology will occur making us all tremendously wealthier, then it makes sense to save less now than otherwise.
As Dasgupta notes, it certainly could go either way, depending on how wealthy one believes that the future will be. But the Stern Report ignores this. Now, Dasgupta does note in passing that in general wealth has increased with civilization, and thus there is reason to believe that our heirs will indeed be wealthier, all things being equal. (And Professor Kling certainly agrees.) However, he does leave it open for the other possibility.
ponte 11.30.06 at 10:56 am
Arnold Kling appears to be following the model of the humanitarian appeal for the invasion of Iraq.
John Thacker 11.30.06 at 10:56 am
Eta set to one ignores intergenerational wealth because different generations are different people, and eta of one completely ignores that fact by ignoring inequality. The discount rate affects how time is considered, but considering that man is mortal, time alone is not a sufficient consideration.
Functional 11.30.06 at 11:01 am
Dasgupta’s point was this:
It is an easy calculation to show that the current generation in that model economy ought to save a full 97.5% of its GDP for the future! You should know that the aggregate savings ratio in the UK is currently about 15% of GDP. A 97.5% saving rate is so patently absurd that we must reject it out of hand. To accept it would be to claim that the current generation in the model economy ought literally to starve itself so that future generations are able to enjoy ever increasing consumption levels.
Kling interprets this remark as being about the fact that Stern’s analysis would require current generations to starve themselves for the sake of future generations.
Bertram interprets this remark as somehow being about “inequality among members of the present generation.”
Kling isn’t the hack here.
aaron 11.30.06 at 11:02 am
Kling’s description of Dasgupta is correct. From what you excerpted “It is an easy calculation to show that the current generation in that model economy ought to save a full 97.5% of its GDP for the future! You should know that the aggregate savings ratio in the UK is currently about 15% of GDP. A 97.5% saving rate is so patently absurd that we must reject it out of hand.
That is what Kling says.
Functional 11.30.06 at 11:03 am
Anyway, Mr. Bertram shouldn’t be worried what people say about the Stern report. Bill Nordhaus, Greg Mankiw, and Partha Dasgupta may think its conclusion is rubbish, but John Quiggin likes it. End of story.
Alex Gregory 11.30.06 at 11:10 am
Surely someone could email Dasgupta and resolve this very swiftly?
John Thacker 11.30.06 at 11:11 am
Stern, in preparing the report, clearly wanted to do something like “We’ll ignore inequality, because the important thing is to maximize wealth first. Once we do that, we can agree how best to divide it up later.” Thus, he sets eta to one, ignoring inequality among any two people. This ignores inequality among the rich and poor in the present generation (NB, functional), but is also has an unexpected effective of ignoring intergenerational differences because people are mortal.
The choice of discount rate affects whether the $X spent at time 0 by society to prevent global warming is worth more than the $Y of benefits to society at time t. By setting eta to one, Stern wants to compare the value to “society” as a whole. However, society is made of individuals. (Cue the Baroness Thatcher here, perhaps, on the existence of it?) Many of those forgoing consumption in the present will not be around to reap the benefits in the future; instead, their descendents will.
There will be other effects on the wealth of society besides that of spending to prevent global warming. If society ends up being much wealthier in the future, then the descendents will be better off on average, and thus the rewards in the future may be better directed towards people in the present– so long as we consider inequality at all. If society ends up being poorer, then the reverse holds.
Chris Bertram 11.30.06 at 11:14 am
John, I take your point about the way considering inequality _also_ impacts on intergenerational assessment. But I hardly think that the fact of _mortality_ is what matters here, it is just the fact that there are different people (whether different people in the present, or the future) that makes eta relevant.
Setting eta to one ignores both inter- and intra- generational inequalities in wealth. I’m led to believe that the poverty of the present poor looms large in Dasgupta’s thinking here because he uses their plight to exemplify his critique on p. 4.
abb1 11.30.06 at 11:17 am
It sounds like the “It is an easy calculation to show” thing is an attempt at reductio ad absurdum, not an actual critique of assumptions used in the study.
aaron 11.30.06 at 11:18 am
I’m not sure about this, but I think it’s Stern that implies the argument that Chris claims Kling is making. The conflict is in the Stern Report and why doesn’t make sense. Isn’t that what Kling and Dasgupta are drawing attention to?
ponte 11.30.06 at 11:18 am
Because it isn’t possible to find any difference between Cline’s book and the Stern Review if we look at the figures taken to be appropriate for delta and eta, I turned to the work of William Nordhaus, who has been studying the economics of climate change for over three decades. The most remarkable conclusion of his studies – conducted on his Dynamic Integrated Model of Climate and the Economy (DICE) – has been that, despite the serious threats to the global economy posed by climate change, little should be done to reduce carbon emissions in the near future; that controls on carbon should be put into effect in an increasing, but gradual manner, tarting several decades from now. This conclusion has withstood the many modifications Nordhaus and others have made to the climate science embodied in DICE. Their idea is not that climate change shouldn’t be taken seriously, but that it would be more equitable (and efficient) to invest in physical and human capital now, so as to build up the productive base of economies (including, espcialy, poor countries), and divert funds to meet the problems of climate change at a later year.
Their idea also neglects the costs of infrastructural overhaul and engrained institutional habits. It’s one thing to change from gasoline to ethanol. It’s another thing entirely to switch infrastructure and networks to a carbon-free economy. Even President Bush had to acknowledge that we are “addicted to oil”.
John Thacker 11.30.06 at 11:18 am
Surely someone could email Dasgupta and resolve this very swiftly?
We could, but you can also just read the paper. Nordhaus’s critique is indeed of a different nature, as Chris Bertram claims, and has to do with the size of the discount rate. (Note that Dasgupta has “no problems” with Nordhaus’s assumptions about the discount rate either.)
Chris is correct that eta affects inequality. It does indeed affect inequality among between in the present, but the hidden part, as Dasgupta clearly states, is that since man is mortal, people in the present and people in the future are different. Thus if we care about inequality, the existence of mortality biases changes across time in a different way than the discount rate.
The discount rate alone would be fine for a race of immortals, Dasgupta says. But since generations are made up of different people, it is important to select a realistic eta in order to get a good result. He doesn’t believe that eta of one is realistic; most people don’t.
The choice of eta equals one was made in an attempt to simplify the mathematics. A noble goal, I think, but Dasgupta’s paper points to the hidden flaw in this simplification. We cannot simply let eta be one and get the correct choice for society and work out redistribution later. It ignores that we cannot in the future send money back to the people in the past who are dead now.
On the other hand, if we invent either a time machine or become completely immortal, then the simplification is reasonable.
John Thacker 11.30.06 at 11:28 am
But I hardly think that the fact of mortality is what matters here, it is just the fact that there are different people (whether different people in the present, or the future) that makes eta relevant.
I’m led to believe that the poverty of the present poor looms large in Dasgupta’s thinking here because he uses their plight to exemplify his critique on p. 4.
Well, he also uses the plight of starving the present to make a rich future wealther in p. 7, so I’d say he uses both examples. They’re both reducto ad absurda, anyway.
The first example is used to demonstrate that eta equals one is clearly wrong. The second example is used to demonstrate that the choice of eta has strong affects across time; that the discount rate alone is not enough to decide the correct decision.
The Stern Report attempts to simplify the calculation by ignoring inequality and looking just at time. The idea is to assume that we want to make the total wealth of society, and then we worry about inequality separately. The idea is that if the total wealth of society is maximized at each time, then how to divide it is a totally separate question. Thus, the thinking goes, eta can be ignored when trying to maximize total societal wealth at each time.
Dasgupta goes along with this to a certain degree, at least as far as the objective. However, his critique is pointing out that the two problems are not truly orthogonal. We cannot separate out the problem neatly into a “maximize societal wealth at each time” problem and a “minimize inequality at each time by redistributing the wealth” problem. Well, we can, but only by being logically inconsistent and treating (when doing the second problem) inequality between people at one given time as important, but totally ignoring inequality among people who live in different times.
Now, I concede that there is a view of inequality that does totally ignore inequality among people who live in different times, and only considers it among people living in one time. If that is your philosophical view, then the Stern Report is okay as far as that goes. However, Dasgupta clearly disagrees. He believes that people in different times should be treated just like people in different places. Once you do that you can no longer separate the problem of “what to do about global warming” into two independent problems and address only the first, as the Stern Report has done.
Chris Bertram 11.30.06 at 11:28 am
John, your clarifications have been very helpful. But let me quibble with you when you write:
The discount rate alone would be fine for a race of immortals, Dasgupta says.
He doesn’t say exactly that, anywhere, but even if he did, what he _should say_ is surely:
The discount rate alone wold be fine for a race of immortals _who were indifferent about the extent of inequality among themselves_
aaron 11.30.06 at 11:34 am
Which brings us to…The Stern Report is absurd.
Functional 11.30.06 at 11:41 am
It sounds like the “It is an easy calculation to show†thing is an attempt at reductio ad absurdum, not an actual critique of assumptions used in the study.
That’s got to be one of the silliest things ever written on CT, beating out some tough competition from other comments. A reductio ad absurdum precisely IS an “actual critique of assumptions used in the study.” That’s what a reductio ad absurdum argument does: It looks at the assumptions, and then observes that an absurd result would occur if you took the assumptions seriously. The reason that this argument works is because (if done properly) it forces the other person (i.e., Stern) either to: 1) disavow the conclusion while keeping the assumptions, thereby admitting that he is illogical or inconsistent; 2) embrace the absurd conclusion, thereby undermining his own credibility; or 3) disavow his original assumptions.
John Thacker 11.30.06 at 11:41 am
Chris,
You’re right that I should clarify my comment, and I believe I did so in the two recent comments.
To summarize, Dasgupta argues that for a race of immortals, we could cleanly separate out the problem of “what to do about global warming” (a time question) from “what to do about inequality.” (a distribution in the present question) The Stern Report attempts to do so, and considers the first question alone.
However, since we are not a race of immortals, it is impossible to cleanly divide the problem into two separate questions– any decision which affects a time far in the future must consider inequality since the population changes over time. That is, so long as you consider inequality among people living in different times to be important like inequality among people living in the same time.
So more precisely, he says, “The Stern Report’s approach of dividing the problem in two would be appropriate for either a race of immortals or people who are totally indifferent about inequality among people in different times (but who may care about inequality among people living in the same time.)”
It is philsophically consistent to argue that since we don’t know the people in the future, certain negative effects of inequality like envy and “keeping up with the Joneses” don’t come into play and that eta should be reduced. I might believe part of that, but I would still not argue that, morally, eta should be one. Dasgupta, his paper reveals, believes that eta should be treated the same, and that people in the future are just like people elsewhere in the world at the same time.
John Thacker 11.30.06 at 11:47 am
Another philosophically consistent position that might agree with the Stern Report and disagree with Dasgupta is one that views people as indifferent between their own consumption and their heirs’, where those with no heirs are equally happy between their own consumption and leaving it to the government/rest of society. In this way, the population now would be equivalent to the population in the future in the proper way.
I suspect that this does not agree with reality either.
John Thacker 11.30.06 at 12:06 pm
So, in short, I think that your charge of hackery against Professor Kling is inaccurate.
There are different reasonable assumptions and different consistent philosophical positions about eta and whether we want to treat time and space identically (and thus different people at different times the way we do different people at the same time).
Dasgupta makes a strong logical argument that if we care about inequality among different people living in different times, then Stern’s approach of designating the global warming problem as a discount rate problem of time only and separateing it from the inequality problem is invalid. He then makes a normative appeal for caring about such inequality. Professor Kling accurately summarizes the important point about Dasgupta’s paper, and clearly agrees with him that we should care about inequality among people living in different times.
I believe that that is the salient point of the article, not the discussion of inequality among people living in the same time. Indeed, if that were all we cared about (and not inequality across generations/people in different times), then we could separate the problems into two independent ones and behave as Stern suggests. It is precisely because Dasgupta believes that we should consider inequality among people living in different times that he considers Stern’s paper flawed; thus that, not inequality in the present, is the relevant point of Dasgupta’s paper. (Not to say that he does not care about present inequality, but, as noted, it could be dealt with in a separate way.)
The criticisms of Nordhaus are of a different nature, about the discount rate. The discount rate is very difficult to determine, and Dasgupta point out that he could accept several different values, despite those values leading to very different policy recommendations. That is, however, a different issue.
ponte 11.30.06 at 12:06 pm
If you think, for example, that exhausted supplies of fossil fuels will mean that our descendents will be poorer, then it makes sense to save more now. On the other hand, if you believe that advances in technology will occur making us all tremendously wealthier, then it makes sense to save less now than otherwise.
Isn’t it really a question of what KIND of wealth we accumulate for the future? You assume that our descendants will inherit an energy infrastructure that will somehow be a fungible good in the future. It won’t be. They won’t be able to exchange their overabundant oil platforms, refineries, tankers and service stations for solar and wind power. Instead, they’ll be stuck with devalued junk. Presuming that there won’t be any gullible alien junk merchants breezing by the planet, of course.
rwoqiw 11.30.06 at 12:21 pm
functional: you are being unfair to abb1’s position. A reductio absurdum works like this.
1. You claim A1.
2. It would be inconsistent to claim A1, without also claiming A2.
3. Claiming A2 would be absurd.
There is no easy sense in which A1 is necessarily absurd; in fact, it may be quite reasonable, but inconsistent with one’s other beliefs. I believe that abb1 thinks Dasgupta is saying the following:
1. Stern makes certain claims about contemporary interpersonal welfare comparisons.
2. It would be inconsistent to make these claims, without also making parallel claims about intertempporal interpersonal welfare comparisons.
3. Those parallel claims are absurd.
I would hardly call this a “silly” interpretation.
rwoqiw 11.30.06 at 12:24 pm
*ad absurdum
Aaron_M 11.30.06 at 12:26 pm
John:
When you talk about inequality “between people in different times†you mean inequality between individuals within a specified future generation right? In other words, you mean inequality between people in future generation A,B,C…. as opposed to the inequality between the total welfare found in various generations (i.e. ours and future generations A,B,C…)
Daniel 11.30.06 at 12:27 pm
If it actually were possible to invest 97.5% of our GDP at 4% rate of return, then this might actually be the correct thing to do. However, it isn’t (this is the shell game that drives Dasgupta’s argument). The percentage of GDP that it is possible to invest at any particular rate of return is dependent on the state of technology and the availability of resources.
abb1 11.30.06 at 12:30 pm
Actually, Functional, yes, he does criticize their ‘eta’ value (whatever it is), but your description of ‘reductio ad absurdum’ is incorrect. It only demonstrates that there is at least one case where an absurd result would occur. So, he had to create a specific scenario. And this Stern Report could, perhaps, still defend their approach by arguing that the scenario is unrealistic.
abb1 11.30.06 at 12:35 pm
Rwoqiw, I didn’t ignore you comment, I just didn’t see it – got distracted typing and watching a movie at the same time.
Walt 11.30.06 at 1:36 pm
Aaron has got to be a paid troll.
John Thacker 11.30.06 at 1:53 pm
When you talk about inequality “between people in different times†you mean inequality between individuals within a specified future generation right? In other words, you mean inequality between people in future generation A,B,C…. as opposed to the inequality between the total welfare found in various generations (i.e. ours and future generations A,B,C…)
No. I mean the inequality between person A in the present time and person B living in the future who has not been born yet. That is the crux of Dasgupta’s complaint.
If people two hundred years from now will be so wealthy (due to other technological advancements and such) that even the poorest will be fantastically wealthy and have access to Star Trek-style replication machines, then making us in the present poorer to make them better off doesn’t make as much sense as if people two hundred years from now will be poorer than now thanks to the exhaustion of fossil fuels. Dasgupta argues that we can and should compare inequality across time, that we care if person A now is poorer or richer than person B in the future.
They won’t be able to exchange their overabundant oil platforms, refineries, tankers and service stations for solar and wind power. Instead, they’ll be stuck with devalued junk. Presuming that there won’t be any gullible alien junk merchants breezing by the planet, of course.
This, of course, could similarly be an argument for not investing in, say, conversion of infrastructure to ethanol in the short term, but rather concentrating on long term solutions. However, it’s rather irrelevant to Dasgupta’s point. He clearly said (and I agreed) that it might make sense to save more now or less, once you take into account inequality among person A living now and person B living in the future. It all depends on whether the future will be wealthier or not, and to what extent. However, if you do think that it is possible to compare inequality of people living in different times, what you can’t do is ignore it in the way that Stern report did and try to split it into different problems.
John Thacker 11.30.06 at 2:01 pm
If it actually were possible to invest 97.5% of our GDP at 4% rate of return, then this might actually be the correct thing to do.
For a period of hundreds of years? Only if you think that people living now are willing to live in poverty in order to make their descendents fabulously wealthy. You have to make philosophical assumptions that people view their heirs as exactly equal to themselves to justify that.
The percentage of GDP that it is possible to invest at any particular rate of return is dependent on the state of technology and the availability of resources.
True, but irrelevant to the dispute and not “the shell game that drives Dasgupta’s argument.” He was simply arguing that the model is inappropriate because it does not consider inequality between a person living now and a person living in the future.
your description of ‘reductio ad absurdum’ is incorrect. It only demonstrates that there is at least one case where an absurd result would occur. So, he had to create a specific scenario.
It is a reducto ad absurdum. He takes the model to its logical conclusion. The Stern report argues inequality of people living in different times is irrelevant. It’s an absolutely necessary assumption in order to justify separating the problem of how much to invest in global warming prevention over from how to redistribute wealth and fight inequality at one particular time. Dasgupta argues that we should take it into account, and uses a standard reducto ad absurdum to justify his point.
There are consistent logical positions that reject his point of view, of course. And he takes no absolute position on whether people in the future will be wealthier– he explicitly states that considering this problem may call for either more or less investment now. What you can’t do is ignore it, at least if you consider the difference in the standard of living of person A now and person B one hundred years from now to be a real source of inequality worth equalizing.
aaron 11.30.06 at 2:02 pm
Just bored Walt.
Aaron_M 11.30.06 at 2:27 pm
“Aaron has got to be a paid troll”
SORRY, next time I will click on the links and read the stuff before asking a silly question.
I thought there was something different being brought up here, but it is essentially the same “you have to account for people being richer in the future” argument that always comes up with discount rates and climate change.
Matt Kuzma 11.30.06 at 2:41 pm
While Arnold Kling clearly doesn’t understand the difference between delta and eta, he draws a more-or-less salient conclusion from Dasgupta. It would be nice if he didn’t confuse intertemporal with intemporal, but I don’t think he’s grossly misrepresenting the extent of the Stern report’s fallacy. He is certainly, however, spinning the whole thing awfully hard.
I think you also misinterpret the Dasgupta review to some degree. Specifically, your statement:
Dasgupta thinks the well-being of the actual poor should take priority over climate change abatement.
In fact, the only place this is actually said in the review, it’s attributed to someone else (specifically the DICE report). The conclusion of Dasgupta’s review is actually that Stern should have shown how his conclusions are sensitive to variations in eta, and that a higher value is probably more reasonable.
It’s certainly wrong to conclude, as Kling has, that the Stern report is exaggerating the cost of doing nothing, but I think it’s worth noting that the assumed ethical values of the Stern report are being legitimately questioned.
ponte 11.30.06 at 2:49 pm
aaron_m:
SORRY, next time I will click on the links…
I’m pretty sure walt was talking about the other aaron.
Walt 11.30.06 at 2:52 pm
aaron_m: Notice the distinct lack of an underscore and an m in “Aaron”.
Barry 11.30.06 at 3:34 pm
david kane: “Since Greg Mankiw (Harvard) and Bill Nordhous (Yale) seem to come to the same conclusion as Kling (that the Stern Report’s use of discount rates is highly suspect), perhaps the “right-wing hack†takeover of Ivy League universities is almost complete. ”
This would be Greg ‘Whored for Bush’ Mankiw? He’s not a right-wing hack, of course; he’s a right-wing prostitute. Which, in terms of ‘can his word be trusted?’, amounts to the same thing.
John Quiggin 11.30.06 at 4:31 pm
The underlying problem here is that expected utility theory (which everyone here is using), along with the plausible technical assumption of constant relative risk aversion means that a single parameter (eta in the discussion above) simultaneously determines attitudes to interemporal wealth transfers, interpersonal redistribution and risk reduction (transfers of RSNincome between states of nature).
The problem is that no single choice of eta produces results consistent with actual observations on all three. One manifestation of this is the equity premium puzzle. Given plausible estimates of risk attitudes, either the real bond rate is much too low (it’s typically between 1 and 2 per cent) or the real rate of return to equity is much too high (it’s typically 6 to 8 per cent). The model suggests that the two should differ by no more than half a percentage point.
So it’s easy to find contradictions. Taking Dasgupta’s example, if there really existed unlimited risk-free investment opportunities available to the UK public sector realising 4 per cent, it would be possible to undertake an unlimited arbitrage by borrowing from foreigners at 1 or 2 per cent.
I’ll try to post more on this Real Soon Now.
Arnold Kling 11.30.06 at 5:52 pm
The topic here is economics, and although my handful of publications in peer-reviewed economics journals is nothing to get excited about, I do not think Mr. Bertram is in a position to use the ad hominem “hack” epithet.
As to substance, my reading of the Dasgupta paper is that it has absolutely nothing to do with “the well-being of the global poor.” As I see it, he is saying that the ethically appropriate discount rate combines: how you weigh future generations relative to current generations as people; and how you deal with a presumption of diminishing marginal utility of consumption in the context of economic growth.
My view is that however one chooses an “ethically appropriate” discount rate, if it is far from the market interest rate, then that in and of itself is going to lead to very dramatic conclusions. Dasgupta’s calculation illustrates the extent to which a below-market discount rate can affect the results, and I thought it brought a relatively esoteric point home in a relatively clear way.
Although I disagree with Mr. Bertram’s interpretation of what Dasgupta was trying to do, I don’t think that Mr. Bertram was being intentionally dishonest. I would appreciate it if he would grant me the same acknowledgement.
Chris Bertram 11.30.06 at 6:18 pm
I’m quite happy to acknowledge that your misreading of Dasgupta was not deliberate or dishonest. But person spinning accusations of dishonesty here isn’t really me Mr Kling. After all you write about “the magnitude of the intellectual swindle that Stern is attempting to pull off.”
Swindling, as far as I recall, is a form of intentional dishonesty, is it not? Perhaps you should grant Stern a similar acknowledgment.
ponte 11.30.06 at 6:41 pm
Perhaps you should grant Stern a similar acknowledgment.
Not to mention the many climate scientists who have advised against continued “business as usual” CO2 emissions, i.e. the “anti-global-warming crusade”.
Arnold Kling 11.30.06 at 7:10 pm
I think that Stern himself may have made an honest mistake, in that perhaps he did not realize what a large order of magnitude was involved in a seemingly trivial assumption about interest rates. But that does not excuse the team as a whole. Either (a) someone on the team knew what they were doing and tried to pull a fast one or (b) no one on the team understood the impact of the interest rate assumption. In the former case, it was a swindle; in the latter case they were not really competent to issue the report.
John Quiggin 11.30.06 at 8:59 pm
As I pointed out above, the long-term real riskless bond rate is now, and has historically averaged, between 1 and 2 per cent (current 10-year Treasury is 4.5 per cent, core inflation about 3 per cent). This is the same as the discount rate you get with an eta of 1, and a long run growth rate for per capita income of between 1 and 2 (IIRC, Stern uses 1.75 per cent). So, if your criterion is matching the market interest rate, Stern is spot-on.
rwoqiw 11.30.06 at 9:36 pm
Brad DeLong has an . Dasgupta assumes all growth in consumption comes from previous savings. If you assume that some growth comes from technological advances that are not controlled by investment decisions, the equation changes a great deal. DeLong throws in some numbers and gets 22% savings, rather than 97%.
T.R. Elliott 11.30.06 at 11:10 pm
I have to say, my gut reaction to Arnold Kling has always been that he’s a right wing hack (supporter of Iraq war, wants us to bomb Iran, and other right wing fiascos).
And if you take a look at his language:
“intellectual swindle that Stern is attempting to pull off”
and
“The Stern Review is arguably the policy equivalent of biweekly mortgage conversion. I hope that is not true for the anti-global-warming crusade as a whole”
Notice that, according to Kling, (a) Stern is consciously trying to pull off a swindle and (b) Kling maligns the entire global warming community as crusaders.
His language is always like this. It’s rather annoying, making anything he writes that might be intelligent indigestible to anyone with half a sense of dignity and etiquette.
I first realized his hackdom when I stumbled upon his completely erroneous cheerleading for shale oil–which he later corrected when James Hamilton published some real analysis on the topic.
What is really ironic is the way they threw me off the EconLog board when I incorporated Kling (and Caplan’s) style into my contributions. They told me something about my disruptiveness and that then provided some garbled argument about how they needed to incentivize my behavior by throwing me off.
Daniel 12.01.06 at 6:33 am
No. I mean the inequality between person A in the present time and person B living in the future who has not been born yet. That is the crux of Dasgupta’s complaint.
If people two hundred years from now will be so wealthy (due to other technological advancements and such)
This can’t be Dasgupta’s argument. His model is one with a single technology which generates a deterministic 4% return forever. There is no technological change at all in his model (and no possibility of the economy being devastated by climate change either).
Daniel 12.01.06 at 6:37 am
I think that Stern himself may have made an honest mistake, in that perhaps he did not realize what a large order of magnitude was involved in a seemingly trivial assumption about interest rates. But that does not excuse the team as a whole. Either (a) someone on the team knew what they were doing and tried to pull a fast one or (b) no one on the team understood the impact of the interest rate assumption. In the former case, it was a swindle; in the latter case they were not really competent to issue the report.
This is bullshit, Arnold, and if Chris is withdrawing the charge of “hack”, I will happily now take ownership of it based on this comment alone. As you know, the Stern team spent a lot of time and effort justifying the choice of the discount rate precisely because they were aware of its importance for the overall numbers. Dasgupta’s argument is that the criterion which Stern uses to judge intergenerational equity is insufficiently inegalitarian (which is precisely Chris’s point). It certainly isn’t a claim that the discount rate methodology is technically wrong. Give over.
Robin 12.01.06 at 10:47 am
On the issue, Delong’s comments and Partha Dasgupta’s response in the comments over at BDL’s Semi-Daily Journal are worth a look.
http://delong.typepad.com/sdj/2006/11/partha_dasgaptu.html
engels 12.03.06 at 9:16 am
In the kingdom of the blind, the one-eyed man is Kling.
asg 12.04.06 at 4:38 pm
Aaaaaaaaaaand, with this post, Crooked Timber officially jumps the shark. Maybe next season will be better. See you next year!
Jonathan Goldberg 12.05.06 at 9:09 pm
This whole analysis makes me queazy. I’m not comfortable with the continuity assumptions being made. As Daniel noted, parenthetically but astutely: “(and no possibility of the economy being devastated by climate change either).” If I understand correctly (not having the patience to wade through the whole report), taking out insurance against catastrophic change is part of Stern’s rationale.
I tend to be a zero-discount person myself, because I disagree with the assumptions about fungibility needed to justify a positive rate.
As for Arnold Kling, I gave up on him when he wrote that he wasn’t sure that economists should let themselves be “bullied” into accepting the reality of global warming.
Comments on this entry are closed.