2022-03-05

The flower of poor thinking is to lack influence

Maggie Or, The First Step Toward Saving the Planet Is Ignoring the Economists by Andrew Dessler in Rolling Stone. You may well say that being published in Rolling Stone gives you more influence than a humble blogger and I would be forced to agree. So it goes.

Speaking of how it goes, my picture shows LMBC rowing over comfortably ahead of Caius to retain the Lents headship. For connoisseurs of mud, the towpath was horrible, since it rained, lightly, almost all week. But despite the Covid-induced break, everyone seemed to remember how to bump.

The nub of AD's analysis is "The U.N.’s latest climate report shows that we don’t know how expensive the climate crisis will be, which means cost-benefit analyses weighing how to combat it are pointless". The new report is really quite long though, and AD doesn't provide any quote from it demonstrating that his assertion is true, so I think that very basic step can be regarded as Unproven (do feel free to provide a suitable quote in the comments). Furthermore, if we genuinely had no idea how expensive it would be, we would have no basis for action at all (a quote below shows this isn't true). However, let's assume that the IPCC (why has AD switched from "IPCC" to "UN"? Odd; perhaps he is trying to put off the right wing nutters) does indeed say that; why would it demonstrate that economic thinking, including CBA, is pointless? His end point is revealing in this regard: As the latest IPCC report says, “The cumulative scientific evidence is unequivocal: Climate change is a threat to human well-being and planetary health. Any further delay in concerted anticipatory global action on adaptation and mitigation will miss a brief and rapidly closing window of opportunity to secure a liveable and sustainable future for all.” Do we really need a cost-benefit analysis to convince ourselves to address this threat? Taken all that at face value, no, we don't need a CBA to decide to do something. But we do need a CBA to decide what to do. Because a large number of solutions from different people are proffered, and we cannot do them all; and many of them, whilst appearing brilliant and obvious to their proponents, appear stupid to others. How shall we decide what to do? Shall we just accept arbitrary diktak? Or should we do... that which involves least use of resources?

I have said this before - but cannot now find it, so will have to repeat myself - that avoiding economic thought is often a way of saying "do what I want". Which is generally bad.

AD's prime example of why CBA is bad is During the Obama administration, the social cost of carbon... was estimated to be $35.  The Trump administration altered some of the assumptions that led to his estimate, particularly how much they valued future generations versus ours, and how much they valued people outside the U.S. versus those who live in America. They estimated the social cost of carbon to be as low as $1. But this was politics, not economics. Calling CBA bad because of Trump is just dumb.

Is there any virtue to AD's piece at all? If there is, it isn't the quoting Franta bit. His discussion of "Obama’s climate bill" isn't great either: although, as he notes, some made hyperbolic claims about the costs, he is firstly wrong to attribute those claims to economists: they were made by pols. Secondly, he is wrong to assert that reality disproved those claims: since the bill didn't come into action, we'll never know what costs it would have had. Thirdly, as he mentions but doesn't think about, even without the bill the aims of the bill were achieved. So all those making hyperbolic claims about how necessary the bill were, have been proved wrong by reality. My guess is that AD would have been one of those people, but I can't be bothered to find out.

Refs

* Make Desertion Fast by Bryan Caplan

* A Populist Attack on Big Tech by David Henderson

* Not So SWIFT by David Henderson; A SWIFT and sure way to punish Russia by Scott Sumner

"It could be said of democracy that all theory was against it and all experience for it." - TF

* What Should Economists Do Now? by Mikayla Novak

The Age of the Judicial Thoroughbred aka Benjamin Barton’s The Credentialed Court.

Should we care about the world after 2100?

The ETS is stupid, part n

Waging the War of Ideas, via What the EA community can learn from the rise of the neoliberals, via ACX.

The Actual Food Problem - via Timmy. Not sure it is, but "fertiliser used to be 55 per cent of the cost of his business, but will now be near to 80 per cent..." is interesting.

Oxford’s Sustainable Law Programme establishes the Climate Litigation Lab

18 comments:

tidal said...

I am always reluctant to comment on these things, because I am somewhat sympathetic to the critiques... but I've kept waiting for coming on twenty years and it always seems to revert to "whatever the economists are saying, it's wrong. And I don't have my own estimates, or at least in a structured/supported way, but the costs are going to be YUGE!!!"

Everyone can cite Martin Weitzman as saying that fat tails and deep uncertainty challenge standard cost-benefit approaches. But hardly anyone points out how he said how profoundly unhelpful it was to just look at the tail, say" YUGE! Precautionary principle! Spend!", because the public and policymakers literally can't deal with such claims.

Example:

Here's atmospheric physicist Scott Denning making a cost estimate:

≪It's all about opportunity cost.

Our kids lose $1000's of trillions if we keep burning carbon.

And of course that's not counting the sociopolitical costs of famine, billions of stateless refugees, wars over food and land, etc≫

https://twitter.com/airscottdenning/status/1392206442271100931?t=93b9Mk3sbdSiWLjUd7W73g&s=19

*Thousands* of $trillions? (and that just "our kids"?) *Not counting* all those other issues?

First off, clearly he's making an appeal that the benefits (avoided damages, etc.) clearly outweigh any costs, even if the numbers ($1000's of trillions!) seem to just appear out of thin air.

But, given that the global economy is only about $90 trillion, what does that mean we should spend/invest?

As Weitzman noted, it seems to suggest we should be willing to spend/invest ≥100% of current GDP just mitigating climate.

But this is nonsensical, so you can't even get to talking about what can/should be done.

Which circles back to the need for credible, supported cost and benefit estimates.

I'll leave with this quote from a *thoughtful* IAM critique - from an IAM modeller, go figure!

≪... If IAMs were scrapped and we started from a blank slate, it is inevitable that to understand the potential pathways to a low-carbon economy, energy and climate change analysts would construct models relating the extraction, transformation, distribution and use of energy resources in the world economy. Such models would be specified to represent the technologies involved in these processes with their associated emissions, and account for potential levels of energy demand in the future. Following demands from policy-makers, businesses and others concerned with the costs of the transition, the models would also be developed to account for the costs of different technology and energy resource combinations, by specifying costs and performance characteristics of technologies and extraction costs of primary energy fuels, as well as the "rules" under which technologies and fuels are selected (i.e., least-cost, welfare-maximising). WE WOULD THEN HAVE REINVENTED IAM's AND RIGHTLY SO.≫

He (they) doesn't mention damages because he was looking more at energy-systems IAM's rather than CBA IAM's, but the point about what would happen if we were to literally just "first, ignore all the economists" is spot on! We'd be demanding to know answers to very same questions!

https://www.mdpi.com/1996-1073/12/9/1747

Tom said...

Perhaps Mr. Dessler understands too well that a CBA would point to Fast Mitigation measures as appropriate first steps. As that is opposed by the most concerned of climate activists, economics is not useful. To them. I think your prior conversations with Michael Tobis should have signaled that to you a decade ago...

William M. Connolley said...

Tidal: thanks. That's largely what I think too, so I don't have much to comment. For my ref, your link is A Review of Criticisms of Integrated Assessment Models and Proposed Approaches to Address These, through the Lens of BECCS by Ajay Gambhir, Isabela Butnar, Pei-Hao Li, Pete Smith and Neil Strachan.

I have however found one of the things I wanted to link to earlier: Architecture and morality: I think this view (the morality view) is shared by, say, Greenpeace, whatever their public pronouncements about costs might be. They think GW is Bad and should be forbidden. Not taxed; not slowed; not ameliorated; not adapted to: forbidden. This is the moral approach: it is a bad thing, so don’t do it.

Tom: what do you mean by "Fast Mitigation measures"?

Unknown said...

I agree with Tom (and tidal, if I understand them correctly). If you're even mildly hysterical about climate impacts - and I'd include ATTP along with the Desslers and MT's of the world - then it will be necessary to ignore climate economists and any sane CBA because, of course, they usually have some connection to the sane world, not the hysterical one.

The insistence on rushing towards 'net zero' or 1.5 degrees also comes from the same hysterical thinking. If something provides 500 dollars of benefits along with 50 dollars of negative (but unpaid) costs, you don't want to stop the something, just include the negative costs. Guess what - a carbon tax will do the job! At that point, as Timmy remorselessly says, you can step back and let things take their course.

It's true that we don't know very accurately the costs of climate change but surely a best estimate (a CBA estimate of course) will lead to the most sensible course of action.

To reiterate, you're probably only going to want to ignore the consensus of climate economics if you're an outlier. There's a whole section of the the English language which has been co-opted for those far beyond these consensus walls. There's some delicious irony that it is currently applicable to those who used to use it against the crowd who couldn't, or didn't want to, believe in the consensus of climate science.

Anteros said...

Previous comment by me.

...and Then There's Physics said...

Since Anteros mentions the "consensus of climate econommics" I'd be interested to know what they think it is. It's not obvious to me that there really is one (at least not one that is particularly strong). Also, if this consensus involves the words "good", or "best", or "optimal", so something similar, then why should people be expected to accept this particular judgement? It's possible to accept a consensus position (if it exists) but still disagree with what some regard this as implying.

William M. Connolley said...

I don't think there's anything at all IPCC-like for economists. But every time there's a survey, there's widespread agreement on "use a carbon tax". Since the thing they pretty well agree on is so widely ignored, there is perhaps little point in fiddling around with the details.

I'd say that most econ would indeed agree that a CT is best or optimal, and perhaps even good; and I certainly accept that judgment; do you disagree? You have heard, I hope, the argument that setting direction and letting people get on with it is superior to govt micro-managment.

...and Then There's Physics said...

Yes, I agree that many economists are in favour of using a carbon tax. However, as I understand it, you could use a carbon tax to try and follow an optimal pathway, or to try and achieve some target. Also, I think many think we should combine a carbon tax, regulations, and incentives, so I'm not convinced the economic consensus is "just use a carbon tax".

Anteros said...

ATTP - fair point. I think I had two things in mind when I mentioned a consensus within climate economics. One is the fundamental idea of a pigouvian tax on the negative externality of CO2 damages. It is rumoured to be the first and only thing that economists agree upon. The other thing was a half-remembered comment from Nordhaus that the central estimate for the social cost of carbon was 50 quid a tonne.

Of course it will always be the case that there's less certainty in that estimate than,say, the "about 3 degrees" for ECS, but that is true whatever your approach.

You focus on the risk of climate change impacts possibly being much worse than expected but a) taxes can be adjusted if and when estimates change, and b) spending a trillion too much is just as bad as a trillion too little.

Mark B said...

Is "just use a carbon tax" really consensus absent some useful bound for what that carbon tax might be? Or is it just consensus that taxing externalities is sound economic policy while sidestepping the real problem of quantifying the costs and benefits?

Anteros said...

An additional point about the carbon tax is that for the purist Pigouvian it isn't something that is used to try to reduce carbon emisions - although that would be an expected result. It is simply to include a hitherto unpaid cost of emitting carbon in the price. That's it. If the marginal social cost of a ton of carbon was known to be 50 pounds sterling, you'd simply add exactly that to the cost of emitting the carbon and go home, job done.

To see carbon taxes as primarily an attempt to reduce emissions is to miss the fundamental simplicity (and beauty!) of carbon taxation.



...and Then There's Physics said...

Anteros,
Yes, I do get that a Pigouvian tax is primarily to internalise externalities. However, this does assume that it's possible to reliably estimate these external costs. Also, there are - IMO - some ethical considerations. For example, the developed world could decide to impose a carbon tax that internalised the external costs, and the damages could predominantly occur in parts of the developing world that have not benefitted much from the emissions. Is this fair? Maybe there can be compensations, but I'm not sure we're all that well known for doing so.

You focus on the risk of climate change impacts possibly being much worse than expected but a) taxes can be adjusted if and when estimates change, and b) spending a trillion too much is just as bad as a trillion too little.

Something to bear in mind is that the current evidence is that global warming, and the resulting climate change, depends primarily on cumulative emissions and is probably irreversible on human timescales (unless we can implement large-scale negative emission technologies). So, if we do decide that the best policy is to follow some cost-benefit optimal pathway and then discover that the damages have been under-estimated, or the impacts are more severe than expected, the changes that will already have occured will probably be locked in, as will the further changes that occur as we implement stronger policies to bring emissions down faster.

My own view is that this implies as slightly more risk-averse approach. Others can, obviously, have different views.

Anteros said...

ATTP,
I can't say the 'irreversibility' of climate change concerns me greatly.

I also don't think that assessing climate change impacts as likely to be not very significant is necessarily less risk-averse. I start with the baseline that compared to other impacts (demographics, technology, politics etc) climate change's will be "small" (IPCC 5). I agree that as Tol says, negative surprises are more likely than positive ones but that, to me, implies heading towards moderate, from small. Just a different perception, I suppose.

A good criticism of carbon taxes, which I haven't found an answer to, is that in the real world there isn't a snowballs hope in hell of them ever being implemented. Especially because to be even moderately effective they would need to encompass a goodly proportion of emitted carbon world-wide. Simply not going to happen.

So, for CT enthusiasts such as Tol, WMC, Timmy, myself and the great preponderance of economists, what do we suggest? Some might say that the second best option is 'do nothing', but that's probably one step too far, even for me. So, I don't know. Probably just accepting the messy, inefficient, top down, piecemeal approach we have now with aspirational temperature targets and rather large sums of money splurged on ineffectively supporting renewable electricity generation.

Anteros said...

Here's a decent expression of the case for a carbon tax by someone who has been advocating them for a couple of decades

https://scholar.harvard.edu/files/mankiw/files/smart_taxes.pdf

William M. Connolley said...

Note for self: Ignoring the Economists? - by ATTP. I gave up when I realised I couldn't be bothered to work out whether table 1 of Nordhaus was actually relevant or not. I suspect not. I still don't really believe ATTP's “How can a recent CBA suggest that the optimal pathway is one that would probably lead to about 3.5oC of warming, when we’re already probably heading for less than 3oC of warming?”.

William M. Connolley said...

A's link - https://scholar.harvard.edu/files/mankiw/files/smart_taxes.pdf - is nice because it reminds us of something even more basic: "As an economics educator, I have always been fascinated by topics about which
there is a large gap between the beliefs of economists and those of the general public.
For example, economists are generally supportive of free trade among nations, while
the public is more skeptical". Free Trade is an area where the economic consensus is clear, and yet the general public reliably supports idiot pols who propose protectionism.

So I think it is clear that "we" aren't going to do "the right thing". I have grown rather hopeful of the "do nothing" option - by which I mean the pols pretend to do useful things, but actually the really useful things - cheaper solar panels, etc - just happen due to tech. Of course, as I said somewhere, the current Ukraine situation is delivering a very useful increase in fossil fuel prices for the moment; strangely, I haven't noted any of the usual suspects daring to celebrate this.

Michael Tobis said...

While I am a bit disengaged on climate right now, largely because I think we have more immediate existential threats than climate, I feel that I am being misinterpreted here and want to say a few things in case we don't succumb to our present dysfunctions.

People perceive me as the opposite pole from Tidal, but in fact we speak often and agree on many things. How is this possible when I find IAMs to be worthless?

In brief, I am NOT saying we don't need cost-benefit analysis, nor am I saying that money isn't important. What I AM saying is, hmm, five things.

===
mt's position summarized:

1) Money tells you nothing on century time scales.

2) Even if money did tell you something on long time scales, IAMs tell you nothing about money on those timescales.

3) If we take climate change seriously, we are entering a HIGH growth period, not a low growth one! People get this backwards.

4) Every decarbonization policy is a carbon tax. We're only debating details.

5) Economics as currently constituted can help us reach our objective but can't help us set it. We should just just stick to 2 C.

===

1) Money tells you nothing on century time scales.

Money is a metric that is valuable in the short term but meaningless in the long. We don't know what the cost function we ought to be that we're optimizing for. Total dollar equivalent wealth or total dollar equivalent economic throughput are signs of thriving in the short run but there is no reason to believe that across scenarios the world with the greatest gross economic product or the greatest constant-dollar denominated wealth will be the one we prefer.

2) Even if money did tell you something on long time scales, IAMs tell you nothing about money on those timescales.

We need economists to think about these things, and we need ways to evaluate tradeoffs and outcomes over the long run, but I see very little sign that economics (or the IAM community if regarded as separate from economics) as currently constituted has any meaningful skill at doing so. The fact that we WISH IAMs were useful, and that IAMs are indeed an inevitable exercise, in no way indicates that their results are useful. I have looked under the hood of DICE. Its approach is risible, and anyone familiar with conceptual modeling and simulation of physical systems would (or at least should) dismiss it as worthless.

Look. I read Asimov's Foundation series as a child. I love the idea that the future could somehow be predictable using statistical methods. But Salvor Hardin's dream is obviously impossible. And even if in some universe it were possible, it wouldn't be possible in an Excel spreadsheet taking ten year time steps.

IAMs are toys. You can't be basing policy on them. Just because you wish they were useful doesn't make them so.

Michael Tobis said...


3) If we take climate change seriously, we are entering a HIGH growth period, not a low growth one! People get this backwards.

Following on point 1, there's a bit of a conundrum in the observation that short term policy (to me, obviously) optimizes for increasing economic throughput aka "growth" and long term policy (to me, obviously) does not. This requires a direction of serious thinking from specialists of which I have seen precisely none; instead there's some rather simple-minded polarization between "pro" and "anti" growth. I despair for the academy sometimes. It may have once been a bit less stupid; I suspect so.

Fortunately, as Tidal has convinced me in our conversations, there is no rush in resolving this if we take decarbonization seriously. The amount of economic activity required to replace our energy infrastructure in a reasonable time is immense. If we take this on, it will be an enormous, sustained, multi-decadal stimulus to growth, whether we should or do wish it were so or not.

This is a recent change in my perspective.

4) Every decarbonization policy is a carbon tax. We're only debating details.

I'm opposed to relying on a direct proportional carbon tax as the sole method to achieve carbon neutrality; it puts the burden on small users rather than large ones. It will exacerbate already fierce competition between rural conservative/reactionary impulses and urban liber/radical ones. It basically amounts to a wealth transfer from rural areas to urban ones. It's only going to make the politics of the situation that much harder.

But that said, some people think it's a bad thing when the price of fossil energy goes up and at the same time are concerned about climate. This doesn't work. Any adequate policy of any sort *will* create an artificial scarcity which *will* drive prices up. Any policy which keeps prices down will encourage continued use. This isn't Nobel-worth economics! (I'm not sure, though, that anything is.)

5) Economics as currently constituted can help us reach our objective but can't help us set it. We should just just stick to 2 C.

One might wish that it were otherwise, but (see points 1 and 2) we don't have enough skill in economics to set a target. Consequently we need to set a target (GMST, CO2 concentration, emissions pathway) more or less intuitively. For a number of reasons I think 2 C is a good one. I would suggest that one reason for using it as a stretch goal is that twenty years ago it was being talked about as a worst case.