Fresh from Krugman’s blog. As usual, it can’t be put much better.
Economics of catastrophe
Away from the headlines, theres a really important discussion going on about how to think about the economics of climate change. The key player is Marty Weitzman, who has made a simple point (albeit using very, very difficult math) thats nicely summarized at Env-Econ:
Climate change is fundamentally a problem about uncertainty. We are conducting an experiment with our planet by doubling CO2 levels in the atmosphere from pre-industrial levels. Concentrations have not been this high in hundreds of thousands of years. By and large, we dont know much about the implications. Tackling this uncertainty is crucial. Extreme outcomes fat tails matter and should be at the heart of much of research.
You can see how important this point is by looking at the latest from Bjorn Lomborg, who says that climate change will reduce world GDP by less than 0.5%, so its not worth spending a lot on mitigation.
Weitzmans point is, first, that we dont actually know that: a small loss may be the most likely outcome given what we know now, but theres some chance that things will be much worse. (Marty surveys the existing climate models, and suggests that they give about a 1% probability to truly catastrophic change, say a 20-degree centigrade rise in average temperature.)
And heres the thing: on any sort of expected-welfare calculation, the small probability of catastrophe dominates the expected loss. Suppose that theres a 99% chance that Lomborg is right, but a 1% chance that catastrophic climate change will reduce world GDP by 90%. You might be tempted to disregard that small chance but if youre even moderately risk averse (say, relative risk aversion of 2 econowonks know what I mean), you quickly find that the expected loss of welfare isnt 0.5% of GDP, its 10% or more of GDP.
The question is, can we mobilize people to make modest sacrifices to protect against low-probability catastrophes in the distant future?
The first thing we do, let’s kill all the wingnuts.
Of course, that would only be very modest sacrifice.
Good post Nic.
Unless Im mistaken he gives a higher probability of a catastrophic climate change event occurring in two centuries time than
1. we do some bad shit with genetically modified foods and cause a catastrophe
2. or a meteor hitting the earth.
Thats a buyable proposition and he gives good reasons why we should think that.
A few issues with the paper (its very good by the way).
Why act now and not say 50 years time? He doesnt seem to say. This is important, as (possibly) it may be better allowing global GDP to accumulate unmolested for a safe period of time rather than starting mitigation efforts in the present time. What are those odds and what is the relative payout ratio for the scenario of simply waiting it out and acting then. Acting now if we don’t need to will lose us unmolested GDP growth potential which over a span of decades is compounds into a big number.
Just as importantly why do we have a moral duty to somehow insure the well being of people 7 generations down the road? Were being asked to make a sacrifice for species preservation now when he doesnt sound convincing or even suggests why we have such a duty (or that we should need to even think about making that decision).
Moreover hes making and awfully big assumption. Hes assuming that global GDP composition will resemble something like the present. I would bet that GDP composition in 30 years time will look nothing the present days let alone 100 or 200 years into the future.
There are plausible theories that GDP acceleration could reach critical mass in 50 years time when we could actually see a doubling over weeks rather than years made possible with massive computing. Trends suggest that Global GDP is actually accelerating so that unlikely event shouldn’t be discounted if we are dealing in small odds large payout ratios.
What hes doing is essentially calculating a payout ratio offering what looks like a smallish possibility for a very large payout
1% chance loses you 90% of GDP in 200 years time =.9 (under what I perceive to be relatively static scenarios).
Unless he gives us other information such as what GDP composition will be in the distant future and therefore our technology capabilities he cant ascribe odds of a major clusterfuck occurring in 200 years time.
Thats how I see it anyway.
Bjorn Lomborgs arguemnt is that its not worth spending on mitigation because its not worth the benefit. We’ll delay 2 years in 100 by current suggested, planned mitigation. This doesnt help any extreme outlier scenarios happening. It is political handwaving. I expect nothing less from Krugman.
Lomborg was on Lateline tonight. Probably available on the ABC website.
Ah, I love SJ. Objectively, that approach would suggest that one starts with anyone possibly susceptible of supporting or facilitating or not opposing communism, since that is objectively the greatest calamity ever visited on humans.
That is good writing from Krugman, though. It is also quite convincing support for the argument that we are never going to do anything meaningful on climate change, because the preferences revealed in everyday life suggest that answer to his question is no.
Consider the last time you shared the roads with the people you are asking to make those modest sacrifices.
Lomborg the famous economist? or the not so famous statistician?
Australia spends 2.4% of its GDP on defence, much of it to guard against low-probability catastrophes (an attempted invasion of Australia).
It makes the case for spending a similar amount on preventing a global catastrophe pretty much a no-brainer.
Patrick, I don’t follow what you’re saying when you say “Consider the last time you shared the roads with the people you are asking to make those modest sacrifices.”
Well, there are a lot of things we could do to increase the safety of our driving, like be slightly slower in certain areas, or allow someone to enter the lane, or rabat to the left, or not make/take a mobile call. All of these appear to me to be
, so what hope of getting people to make such sacrifices against distant catastrophes?
Maybe I am barking up the wrong tree, or just plain barking mad?
Well we do all these things – to some degree.
I watched Lateline last evening and found myself wondering why a program like that would bother with giving the lunatic fringe – Bolt and Marohasy – any airtime at all, save for highlighting the idiocy of the non-arguments offered. The deniers seem to have sidestepped the fact that the major issue is climate CHANGE. Not global warming because that description isn’t accurate. Science doesn’t regard the problem as global warming, but climate CHANGE. Whether temperatures are increasing or decreasing is irrelevant. That climates across the globe are changing is irrefutable. That human industrial development is the most probable cause is also at the head of the short list of probable instigators. It’s also the only probable which humanity can have any mitigating effect upon. Claiming that industrial pollution, habitat destruction and species extinction aren’t our problem is real ostrich territory.
Then there was Bjorn Lomborg. How disingenuous is that man?!! Spending time, effort and money on mitigating climate change input from humanity isn’t worth undertaking because benefits-per-dollar-spent don’t match with combating malaria, poverty and food shortages. He adroitly side-stepped the challenge from Tony Jones that perhaps spending on military technology and weapons systems ought to be re-directed on a benefit-per-dollar-spent basis. Back to malaria instead. Fools! I suppose when you’re being funded by private monies to tell people what dupes they are to spend their money on a non-profit, green-tainted pursuits, you keep right on doing so in order to salve self-interest.
Let’s not do squat about climate change. It’ll only cost everyone able to read this today money which could be better spent on plasma teles and flash cars while the oil lasts. Besides, everyone able to read this today will be long dead by the time impacts from climate change can be categorically nailed down by science. No-one will hear the “told-you-so’s” when they’re fifty years dead.
Expenditure on defence underpins defence, our expenditure on climate change will not change the climate.
There is a strong argument on spending on R+D for malaria HIV and TB as these seriously impact on developing countries economies, the effect can be measured on a global scale. However, most countries do not spend on on these diseases as they are unconvinced that they are directly harmful and are only of a concern to others.
Like it or not, most governments are concerned only with the welfare of their own citizens.
Governments are concerned only with their own political survival
In defence of Bjorn Lomborg, he is not denying the reality of climate change, although he has been strident about people he thinks have exaggerated the risk, but he is simply pointing out that there is a whole bunch of ills that affect humankind that could be seriously addressed sooner with less money that is claimed to be required for climate change.
It is not an either/or proposition. But he wants to know why initiatives with modest cost and well-quantified global benefits should be ignored in favour of the climate change glamour crusade.
Lomborg’s Copenhagen Consensus 2008 for instance has highlighted:
We have two different world views at work here. They are both real, but I think that Lomborg’s complaint is that the climate change view is distracting people from the Copenhagen Consensus view of helping people who are in desperate trouble now.
I’ll try a different allusion.
Reading my last comment back, I realise it’s too ambiguous. It needs explanation (and hence is a failure). My intent was that the “me” in the final sentence be interpreted as an environment us humans can live in, rather than a literal Christ.
I’ll shut up now.
Several points arising from this discussion.
Julian says that “Bjorn Lomborgs arguemnt is that its not worth spending on mitigation because its not worth the benefit. Well delay 2 years in 100 by current suggested, planned mitigation.”
In fact the mere 2 year delay point in my understanding relates to the current Kyoto targets, which just about everyone agrees were minimalist and mostly just symbolic starters. The reductions now being urged/negotiated (i.e. 60-80% reductions in carbon emissions by 2050) are hugely more substantial and will have hugely more beneficial effects than the piddling Kyoto targets. Most economic modelling shows, contrary to Julian’s assertions, that the mitigation costs of achieving those sorts of reduction are nevertheless considerably lower than the costs of doing nothing and “adapting” as Lomborg urges.
The next point worth making relates to Joe Cambria’s rhetorical question: Why act now and not say 50 years time? Because the carbon we’re emitting now gets largely sequestered in the ocean and released in 50 years time. There’s a 50 year lag between any mitigation measures we take and their beginning to have a significant measurable effect. The next 50 years of warming is already locked in and it’s too late to do anything about it. It’s the 50 years after that which we’re currently in a position to affect, and that will continue to be the case ad infinitum.
Next point: in advocating doing nothing but adapting, Lomborg’s advocacy also implicitly assumes that warming will somehow miraculously stop at the year 2100 even if we do nothing to stop it. Otherwise he couldn’t sensibly suggest that we can rationally pursue a mere adaptation strategy. But if we don’t take mitigation action, atmospheric CO2 levels will keep getting higher and higher and the earth will keep getting hotter and hotter and hotter. We might be able to adapt OK to the extent of warming that will have occurred by 2100 (though not on the worse case scenarios), but it will just keep getting hotter after that on an ongoing basis. Lomborg seems to want us to emulate a lobster in a cookpot. The lobster keeps getting gradually hotter and hotter but there’s never a moment when it gets a short sharp shock that signals to it that jumping out of the pot might be rather a good idea, until it eventually finds itself cooked and dead. It keeps adapting until it reaches its limit of adaptation. It might not happen by 2100, but it will surely happen if we follow Lomborg’s advice. Bjorn can cook in his own pot, thank you very much, on his very own planet.
Ken
My issue isn’t about the science as enough scientists have spoken to make it uncontested. The science is actually relatively the easy bit as i think the really hard part is the economics… how do we organize ourselves with a high(ish) probability scientific projection of a problem in the most optimum way that maximizes human welfare and with the least burdens.
For anyone to say we should act now they need to quantify the cumulative cost of damage into the future (everyone seems to be using 2100 as the base year) and compare that to the benefits of unmolested GDP growth.
Stern is one guy who has tried that arguing ( from memory) that we would lose 20% of global GDP by 2100.
Let’s do the numbers>
Current world GDP is around $47 trillion. Assuming a 3.5% compound growth rate (which could be low as GDP really does seem to be accelerating) world GDP in today’s dollars will be around US$ 1,110 trillion.
Stern suggests that AGW problems would take away 20% of the figure in 2100 so according to him unmitigated GDP would be US$890 trillion after damages.
The cost of mitigation lowers the growth from 3.5 to 2.5% and global GDP becomes US$456 trillion by 2100.
Now I am only going from memory about what Stern said and these numbers aren’t finessed, but it doesn’t look to me that lopping 1% off the growth rate and putting it to mitigation helps us.
Again I want to repeat, this is nothing to do with the science as( although i think serious sceptics should be listened to) the science is pretty robust in this area at the moment.
Stern didn’t consider the catastrophic end of the probability distribution. Several reputable sorts are suggesting that there’s a more than infinitesimal probability that the impacts of climate change could be more than 50% or higher of GWP.
This doesn’t take into account the vast non-utility losses of biodiversity. I’d prefer a world ten times as rich as now but with remnant biodiversity than a world twenty times as rich but where we all live underground.
Of course, it’s a complete failure of economics if you think we can have an economy without an environment. There’s a fair bit more ecosystem pricing to be done yet.
Wilful
there’s no failure in economics. Economics assumes that human welfare and the environment as locked in. Human welfare could never improve otherwise.
this is why I maintain that it is wrong (impossible even) to assume a static GDP composition/tech change 30-100-200 years out while making dynamic climate change forecasts.
Google has a market cap of $150 billion while GM has a market cap of $8 billion.
Global GDP composition is changing and will be unrecognizable.
This is no fault of the climate models as they seem set on static economic assumptions.
Krugman is to the economics of climate change what Fisk is to accurate reporting in the Middle East.
People. Please stop conflating the profoundly unscientific rantings of a few global warming alarmists with genuine environmental concerns…which is basically the argument that Lomborg is making.
As has already been pointed out, the trouble with multiplying a small-probability event with a large resulting outcome is that there are so many small-probability events out there.
The plots of human population over history (the one’s I’ve seen) have a clear notch caused by the “Black Death” hitting Europe, but communism can claim no similar notch in world population (despite larger total kills, there were more people available at the time). Our main protection against disease is antibiotics and Charles Darwin is about to be proven right once again as our antibiotics become ineffective. Thanks to the nature of positive feedback systems and the magic effect when a disease achieves a short-term loop gain bigger than 1, we will see the effect suddenly rather than gradually. I’m plugging for disease to be the next big nasty.
Hmmm, mangled the quotes on that one, should have used the PREVIEW button
I’d also point out that while humans as a species will likely be able to adapt in someway to even the most extreme outcomes, there’s at least something of a moral question as to those who won’t be able to adapt – poorer and weaker humans, and of course, other species. While I’m comfortable with the concept that as a species we have as much right to put our own needs ahead of other species as other species do themselves, I can’t extend that to the point that we can claim we have a right to transform the planet into something upon which most other species will not survive.
On that basis, even if it cost us (i.e., relatively strong and wealthy nations) more to mitigate than to adapt, there seems to me a fairly strong moral imperative to follow the course of mitigation. Especially because as wealthy nations we’ve generally contributed the most to the problem in the first place (though I accept from a moral point of view, we can only really take into account the actions of individuals that are still alive today).
How could we ever survive if that happened? Why do you make that assumption?
We could survive without *most* other species surely, and with sufficient technological advancements, in principle, all of them.
to large extent species endangerment has basically occurred because we have fenced these critters off in one way or another. However you’re talking about wholesale extinctions. If that happened we wouldn’t be ordering caviar.
Not caviar produced by live fish no – but again, in principle, all current food could be artificially produced. Indeed, in principle, all the current enjoyment we get as humans from experiencing other species could be simulated without their actual existence at all. But even if we had such technology, I would have trouble with the idea that it would perfectly OK for us to knowingly obliterate them.