This reminds me of my piece on climate and the Precautionary Principle.
7 thoughts on “Cost/Benefit Analyses For Climate”
There is something to the Dismal Theorem. For example, suppose we have a costly event occur which comes from a selection of countably infinite choices with positive probability where the cost of a possible event is inversely proportional to its likelihood, let’s say in dollars. So if an event is 1% likely to occur, its associated cost is $100. If it’s 0.1% likely to occur, then its cost is $1,000.
For any given outcome, the expected cost of that outcome is $1 because it is the cost times probability and the product happens to always be $1. But there are an infinite number of such outcomes and hence, the total expected cost happens to be infinite. Another way to look at it is that there would be some high probability events that are low cost, say one 50% likely event that is only $2 in cost. But then there’s an infinite number of events that are 1 in a billion or less likely and hence, would cost a billion USD or more. There are an infinite number of really small probabilities that increase in cost without bound. Needless to say, you’re not getting insurance for this.
One of the criticisms was that there was no demonstration that the tail of climate change events was broad enough where this sort of phenomenon of arbitrarily high costs would occur. It’s one thing to say that infinite cost is possible, it’s another to have it in climate outcomes.
There’s obviously more problems with this analysis that just that, but in a sense the theorem is true, it’s just a very long way from being relevant.
Moving on, for some reason I was of the impression that this would be a more general discussion of cost/benefit failings in climate analysis. There are many such problems. For example, greatly exaggerating the cost of losing land to sea level rise by counting only the cost of the land predicted to be submerged, but not the increase in value of land that becomes inhabited in its place. A similar thing goes on for calculating loss of arable land, but not its gain.
Another is greatly dropping time value rates which just so happens to exaggerate future costs, often by huge amounts. I think the dishonesty of the climate change propaganda effort shines most clearly with these bogus cost/benefit analyses which consistently exaggerate the costs of AGW while greatly downplaying the near term costs of AGW mitigation.
So you’re saying it’s just more crap. ^_^
I would note that the land and climate we find at the rise of civilization is, from the standpoint of geology and climate, random. The sea shore is found where the sea shore happened to be when we found it. If it wasn’t where it was it would’ve been somewhere else. The main benefit we had was that we were no longer in an ice age, which reduces available productive land by huge amounts due to the way the continents are configured.
The small differences between these random states during an interglacial shouldn’t have an inherent, measurable difference in cost, and a shift between them should only involve the sunk costs of assets that have to be moved or abandoned. But those are few, mainly consisting of beach umbrellas, and probably none equal to the loss of the Aral sea due to changes in drainage – which in theory irrigated land somewhere else and increased its value by more than the Aral sea was worth.
The only climate change we should really worry about is another glacial period, which would bury Canada and northern Europe, shorten the growing season, and cause all kinds of havoc. We’re trying to mitigate that risk by driving SUV’s.
Thanks, Karl. I haven’t read a detailed explanation of Weitzman’s Dismal Theorem, but from what little appeared in the linked discussion I wonder, does he consider the impact of doing a constrained optimization? In the real world, nobody says “let’s buy insurance for every possible negative outcome” for the very good reason that no one has the money to purchase a policy for an effectively infinite number of low possibility risks. To illustrate with a positive example, every company or individual has a finite amount of money to invest. So you can’t buy every investment that has a positive return. You invest your limited capital in those with a high probability of a high return.
They’re not afraid of rain in Seattle. They hardly notice it without an umbrella.
The world is a ball. Raise or lower temps a few degrees and parts of the ball get better weather while others get worse.
Any global decision is wrong. Local decisions rule.
Decisions that have more to do with power should not even be considered.
Nordhaus also notes that catastrophic climate change is not the only thing we might worry about. Other low-probability civilization-destroying risks include “biotechnology, strangelets, runaway computer systems, nuclear proliferation, rogue weeds and bugs, nanotechnology, emerging tropical diseases, alien invaders, asteroids, enslavement by advanced robots, and so on.”
It’s hard to sleep at night, ya know?
Climate Hustle – Are they trying to control the climate or you?
There is something to the Dismal Theorem. For example, suppose we have a costly event occur which comes from a selection of countably infinite choices with positive probability where the cost of a possible event is inversely proportional to its likelihood, let’s say in dollars. So if an event is 1% likely to occur, its associated cost is $100. If it’s 0.1% likely to occur, then its cost is $1,000.
For any given outcome, the expected cost of that outcome is $1 because it is the cost times probability and the product happens to always be $1. But there are an infinite number of such outcomes and hence, the total expected cost happens to be infinite. Another way to look at it is that there would be some high probability events that are low cost, say one 50% likely event that is only $2 in cost. But then there’s an infinite number of events that are 1 in a billion or less likely and hence, would cost a billion USD or more. There are an infinite number of really small probabilities that increase in cost without bound. Needless to say, you’re not getting insurance for this.
One of the criticisms was that there was no demonstration that the tail of climate change events was broad enough where this sort of phenomenon of arbitrarily high costs would occur. It’s one thing to say that infinite cost is possible, it’s another to have it in climate outcomes.
There’s obviously more problems with this analysis that just that, but in a sense the theorem is true, it’s just a very long way from being relevant.
Moving on, for some reason I was of the impression that this would be a more general discussion of cost/benefit failings in climate analysis. There are many such problems. For example, greatly exaggerating the cost of losing land to sea level rise by counting only the cost of the land predicted to be submerged, but not the increase in value of land that becomes inhabited in its place. A similar thing goes on for calculating loss of arable land, but not its gain.
Another is greatly dropping time value rates which just so happens to exaggerate future costs, often by huge amounts. I think the dishonesty of the climate change propaganda effort shines most clearly with these bogus cost/benefit analyses which consistently exaggerate the costs of AGW while greatly downplaying the near term costs of AGW mitigation.
So you’re saying it’s just more crap. ^_^
I would note that the land and climate we find at the rise of civilization is, from the standpoint of geology and climate, random. The sea shore is found where the sea shore happened to be when we found it. If it wasn’t where it was it would’ve been somewhere else. The main benefit we had was that we were no longer in an ice age, which reduces available productive land by huge amounts due to the way the continents are configured.
The small differences between these random states during an interglacial shouldn’t have an inherent, measurable difference in cost, and a shift between them should only involve the sunk costs of assets that have to be moved or abandoned. But those are few, mainly consisting of beach umbrellas, and probably none equal to the loss of the Aral sea due to changes in drainage – which in theory irrigated land somewhere else and increased its value by more than the Aral sea was worth.
The only climate change we should really worry about is another glacial period, which would bury Canada and northern Europe, shorten the growing season, and cause all kinds of havoc. We’re trying to mitigate that risk by driving SUV’s.
Thanks, Karl. I haven’t read a detailed explanation of Weitzman’s Dismal Theorem, but from what little appeared in the linked discussion I wonder, does he consider the impact of doing a constrained optimization? In the real world, nobody says “let’s buy insurance for every possible negative outcome” for the very good reason that no one has the money to purchase a policy for an effectively infinite number of low possibility risks. To illustrate with a positive example, every company or individual has a finite amount of money to invest. So you can’t buy every investment that has a positive return. You invest your limited capital in those with a high probability of a high return.
They’re not afraid of rain in Seattle. They hardly notice it without an umbrella.
The world is a ball. Raise or lower temps a few degrees and parts of the ball get better weather while others get worse.
Any global decision is wrong. Local decisions rule.
Decisions that have more to do with power should not even be considered.
Nordhaus also notes that catastrophic climate change is not the only thing we might worry about. Other low-probability civilization-destroying risks include “biotechnology, strangelets, runaway computer systems, nuclear proliferation, rogue weeds and bugs, nanotechnology, emerging tropical diseases, alien invaders, asteroids, enslavement by advanced robots, and so on.”
It’s hard to sleep at night, ya know?
Climate Hustle – Are they trying to control the climate or you?
Destroying climate fraud.