Money is Not the Measure of All Things
I think there is a compelling case that if we were to use the best estimates from the IPCC and similar technical bodies for the most likely impacts of Anthropogenic Global Warming (AGW) on average global GDP over the next century, then proposed programs of emissions mitigation are not economically justified. Many very smart bloggers have made the point that that using average global GDP as our only metric to evaluate the relative attractiveness of potential future outcomes misses a lot of what should be important to us.
I believe that there at least two intertwined strands to these objections:
1. Average GDP misses a lot. We could wipe out the GDP of many poor countries, and still only have a small impact on global GDP, and it doesn’t seem fair to consider lowering U.S. GDP by a fraction of 1% on one hand, and entirely eliminating the country of, say, Bangladesh on the other, as equally bad in some important moral sense.
2. GDP misses a lot. There are many things that we care about that are not captured in GDP statistics, such as human health or suffering, maintaining traditional ways of life, aesthetic beauty and so on.
I’ll try to address these one at a time. I’ll rely heavily on papers by Indur Goklany in which he integrates multiple analyses, predominantly from the IPCC and the UK government.
1. Relative economic impacts on the developing vs. the developed world
There is some trade-off between economic growth and mitigation of AGW damages, at least in the short-term. Mitigation advocates often correctly point out that the global poor will be disproportionately affected by AGW damages, but it is also the case that they will be disproportionately affected by reductions in global economic growth. An empirical question is the relative size of these two effects.
Consider Goklany’s review of research that compares the change in climate and wealth under various UN IPCC scenarios for development over this century. I’ll show the two extreme scenarios to make a point: A1F1 (the IPCC scenario for global development that is most heavily dependent on fossil fuels, and has a projected increase in global temperature of about 4C by the end of the century), and B1 (the scenario that assumes greatest deployment of alternative technologies, and has a projected increase in global temperature of about 2C by the end of the century). Here are the projections for each scenario for the developed then the developing worlds:
Developed Countries Projected GDP / Capita in 2100:
A1F1: $107,300 B1: $72,800
Developing Countries Projected GDP / Capita in 2100:
A1F1: $66,500 B1: $40,200
In other words, at least through the next hundred years, the average person living in the developing world is better off in money terms with more economic development and more AGW damage, on net. A lot better off in fact: $66,500 is more than 65% higher than $40,200.
I’ll note in passing that by 2100 the average person in the developing world is projected be at a level of income comparable to the U.S. in 2009.
2. Impacts not directly captured by GDP
I’ll focus first on two items which any reasonable analyst would consider to be important, and for which we have some projections: hunger and water.
Goklany has collated detailed projections for the projected change in various metrics between a baseline year of 1990 and a projection year of 2085 from the UK Government’s Fast-Track Assessment of global climate change (FTA). This is a 95-year projection, and as there should be some acceleration of warming effects this should be a tolerable estimate for impacts for the highly-overlapped 91-year period from 2009 to 2100.
Here are the results for projected humans at risk from hunger under the A1F1 versus B1 scenarios:
Under either scenario, the world should be able to push those at risk for hunger down to 1% – 2% of the world’s population by the end of this century, at any projected level of warming. The realistic risks to this are war, other political action, or threats to a world of interdependent trade and economic growth. The impact of global temperature change is rounding error in comparison.
Here are the results for humans at risk of water stress:
Because wealth allows us to insulate ourselves from environmental risks, a warmer but richer world is projected to be better off on this metric.
Here are some other metrics. The percentage of the world’s population that is at risk for coastal flooding is well under 1% in the baseline, and is not projected to rise close to 1% in any scenario within the 95-year forecast. Malaria deaths have historically been in effect eliminated by societies that achieve several thousand dollars per year of per capita income – the key risk here is once again slower economic growth that keeps parts of the developing world poorer longer.
Again and again, we see the same pattern: at least for the next century, changes in human welfare, even on metrics that are not purely economic, are fundamentally driven by changes in economic development, not AGW damages. This is why it makes sense to be focused acutely on risks to economic growth when considering the overall effects of any emissions mitigation program.
For point 1, what happens to the median inhabitant of the developing world under the two scenarios?
— Chris · Jun 30, 12:29 PM · #
Hopefully James will show up and tell us why the both the Sex Vote and our ultra-modern Moral Aesthetes “feel strongly” about “doing something” to battle global warming.
One thing your economic argument hasn’t addressed (or maybe it has, I don’t know): the Free Range Chicken phenomenon. Higher prices, higher risk of salmonella, no nutritional advantage, moral high ground? Yes please.
— Kristoffer V. Sargent · Jun 30, 01:00 PM · #
Chris:
Great question. I don’t know. I don’t think there are gini coefficient projections that vary by scenario, so I don’t know how the mean / median difference would vary between A1F1 and B1. Without having reviewed it myself, however, I’m pretty sure there aren;t any examples of two large societies with an average difference of inceome of 65% in which the median person in the lower-income society has higher income than the median person in the higher-income society.
Kristoffer:
I was hoping nobody was going to ask that question!
— Jim Manzi · Jun 30, 01:21 PM · #
Sargent: You forgot “tastes better” and “involves the infliction of less unthinkable cruelty”. But I’ll be the first to tell you that the Free Range label means very little.
— John Schwenkler · Jun 30, 02:03 PM · #
How robust is this to other climate scenarios? If we add a few more degrees C to the A1F1 scenario, would GDP still be quite so important?
— Consumatopia · Jun 30, 02:05 PM · #
I was under the impression that many models predict rapidly accelerating ice-melt (it becomes a self-reinforcing process, is the idea), which would lead to far more than 1% of the population being displaced, right?
— paul h. · Jun 30, 03:07 PM · #
John, I can’t tell the difference. I also find the ‘cruelty’ inflicted very thinkable, and very easy to wash down. With milk, perhaps.
— Kristoffer V. Sargent · Jun 30, 05:47 PM · #
Irreversible loss of half the world’s species? Guess Sargent can probably wash that one down too.
— David Roberts · Jun 30, 06:52 PM · #
Perhaps we’re getting off topic, but the difference between free range chicken and Cap and Trade is that the former is the concern of private consumers, whereas the latter will determine how our government (and other governments around the world) will use our tax dollars.
And Sargent: Despite your sneering, I don’t think there’s anything wrong with products whose main appeal is their moral high ground. We wouldn’t dream of scoffing at Kosher food, which in the case of meat is mostly a question of how the animal was treated and butchered. So long as Obama doesn’t order me to shop at Whole Foods, I support our having the option to choose between different types of meat, even if much of the aforementioned movement is a marketing scheme disguised as a moral principle.
— John Slefinger · Jun 30, 08:43 PM · #
Jim: How big a component of Goklany’s analysis is China? I’ve skimmed his paper, but didn’t have time to go to his sources. I’m assuming that China “fits” into the definition of “developing world,” and given their high average rate of economic growth and development in the past decade and anticipated progress going forward, they would be significantly impacted by the differences between the A1F1 and B1 scenario.
The major issue I’m trying to get at here is that a focus on global averages still doesn’t address objection #1 from your lead-in. Namely a disproportionate impact across countries and regions. There is no “average person in the developing world,” there are only economic resources averaged across the total population of the developing world. You can make one person much better off in China, and three people worse off in Bangladesh and still get the averages that you’ve sited. The “developing world” is signficantly less homogeneous than the developed world. I won’t be convinced until I see a more granular analysis (perhaps a comparative of potential outcomes for China vs. Bangladesh under A1F1 and B1?).
— Michael Torrens · Jun 30, 11:00 PM · #
Mr. Manzi, it seems to me your arguments suffer from a sort of black swan fallacy in that you’re using a “lack of proof” as “proof” for your position. As I see it you are using your view that since we have no definitive proof that humans will suffer unbearable calamities from climate change as proof that we will not suffer. That is not true. At best you can say the question is unresolved.
Neither you nor any economist nor any climatologist has any idea what will happen if our climate warms on average by 5 degrees. It may be helpful to run models, but in the end they are all just exercises in futility. Have you ever seen even a ten year economic model prove accurate? Not likelly. And you’re using 100 year models to make definitive assertions that we should do nothing about a problem humanity has never encountered through all of recorded history.
The climate issue is much like the asteroid issue that many astronomers are fearful of. Even if there is only a 2% chance of climate change having disastrous effects, the impacts of those effects are so great that they must be prepared for: the potential for massive disruptions to our way of life. If you want an example of the drawbacks of ignoring the potential for rare, high impact events, see world economy, 2009.
— Noseeum · Jul 1, 09:47 AM · #
Slefinger, I don’t mean to sneer. I grew up working on farms, I am very familiar with chickens and their plight, and, regardless, I maintain a low-radius circle that doesn’t encompass fowl and some humans.
In this medium it’s hard to come across in a precise fashion, but I was going for ‘playful needling’. With maybe an undercurrent of reformed Dr. Evil.
— Kristoffer V. Sargent · Jul 1, 04:37 PM · #
Noseeum:
Would you advocate devoting 1% of global GDP for the next 200 years to asteroid detectiona dn intervention?
What do you think the maximum temperature change and associated impacts are that have 2% probability, and what is the source of this estimate?
— Jim Manzi · Jul 2, 12:54 AM · #
Sorry to be long winded and thanks for responding.
Mr. Manzi, I have no source for the 2% assertion. If I did have a source, I would consider it bunk. All of the predictions I could cite go out for decades and then are completely revised only one or two years later when new data comes in. And yet every time a new prediction is published it’s taken as gospel from whoever agrees.
Take sea level rises for example. As of 2007, it was believed that sea levels rose about 1.7 mm per year since the 60s. So that’s what the models used. Then lo and behold, satellite imagery confirmed that since the 80s it had been rising 3.35 mm per year and the “predictions” were completely revised and republished.
The predictions you are using, and anyone else for that matter, are filled with assumptions like this rise in sea level. Many of them are wrong. Some will turn out to be too aggressive, and some will turn out woefully conservative. And the hope is that all the errors will balance themselves out to the prediction being for the most part right. How ridiculous is that?
Too often we are lulled into a sense of certainty when we mathematically quantify dozens of assumptions for long term projections.
Again, see the financial services industry, from stone ages until now. 98% certainty is still potential for 2% wrong. A financial advisor will tell you with a straight face that you should expect a 7-9% annual return in the stock market, despite the ebbs and flows. Well, tell that to the people that lost half their portfolio in the last two years. The advisor will shake his head and say, “Sorry, that was a rare, random, and unpredictable event that only happens once in a lifetime.”
Gee, thanks. That and a starbucks card will buy me a cup of coffee, but I still lost all of my money when you told me to expect 7-9% per year.
And his response will be, “Well, 98 years out of 100, what I said will be right.” And he will still expect that he should be paid.
In a system with no limits, like the possible different climates, sea levels, fights over water rights, hurricanes, etc. etc. we may see 100 years from now, 98% likelihood of a certain prediction coming true is meaningless. Because the 2% of the time that prediction is wrong is devastating. One cannot reliably use statistical analysis on a system where the range of possible outcomes is not closed, despite our constant attempts to do so.
It comes down to this: if you agree that there is even the possibility that climate change can have devastating impacts on our livelihoods, no matter how remote, and you also decide it’s not worth attempting to address because it’s too expensive, you are wagering the prosperity of the entire human race on your bet. What if you’re wrong?
Sorry is not going to cut it when we’re looking for another planet to live on.
“Would you advocate devoting 1% of global GDP for the next 200 years to asteroid detection and intervention?”
This is a bit of a red herring, since even the astronomers worried about this are asking for a mere pittance. Some tens of millions of dollars. Not even billions.
But yes, hypothetically speaking, if there’s a 2% chance of our planet being completely uninhabitable with no notice due to an asteroid impact (and hence we all DIE), and 1% of GDP will completely eliminate that threat, yes, I would support it.
If you’d be willing to risk our entire existence for a 1% improvement in our daily lives, I will have to disagree.
— noseeum · Jul 2, 05:47 AM · #
noseeum, once you have come up with 50 events that have a 2% probability of wiping us out, we will be using all the world’s resources to take actions of uncertain effectiveness to stave off consequences of uncertain severity. You cannot spend limitlessly to protect yourself from every disaster you imagine might occur. You may not like Manzi’s analysis, but in a world of finite resources and infinite risks, we have no alternative to doing the best we can to estimate the costs and benefits of various paths we might choose. If you take action without the analysis, you are still making an implicity cost/benefit calculation, but you are just not admitting to yourself that you are doing it.
— andrew · Jul 6, 06:59 PM · #
Andrew,
Another red herring. We’re not faced with 50 things that legitimately could happen within a reasonable probability that will cost us 1% of world GDP to solve. If that were the case, you’d be correct.
There might be people who would argue we need to build a missile defense system at a cost of 3% of world GDP in order to stop an alien invasion, but there wouldn’t be a 2% chance of that happening.
Or there might be a 10% chance that North Korea will launch a nuclear bomb. But North Korea only has 1 bomb or 2 at most. So even if North Korea did this, the world would go on.
But climate change has a very legitimate chance of destroying our ability to survive on this planet. So do asteroids, since they’ve already caused mass extinctions. Luckily, only one of these costs 1% of GDP by Mr. Manzi’s standards to solve. Searching for risky asteroids costs about .0000001% of GDP to deal with at this point.
So if we find 50 asteroid like terrors to deal with, that have a legitimate risk and a low cost, yes, we should fund them.
As far as I know, global warming is the only major threat to our survival that is expensive and yet fairly straightforward to deal with. Sure, if we had 50 of them, we’d have to rank them. But list me the 50 first, and then we can start ranking. My guess is global warming will be at the top by a wide stretch.
— noseeum · Jul 7, 05:39 AM · #
noseeum, You keep saying that global warming is going to destroy entirely our ability to survive on the planet. I at first thought you were just using hyperbole, but now I see you really think that. Hey, if you are going to just imagine any degree of severity of any problem you choose, then if you can’t come up with 50 the problem is your imagination, not the paucity of risky events.
The scientific community, however, is talking about 5 degrees of warming over a century or so. Even if the warming is more than that, humanity will survive. I’ll take 5 degrees of global warming over 100 yrs over 2 nuked cities courtesy of North Korea. Of course, China has more nukes than North Korea, and Pakistan has some, and we don’t know what Russia and its cousins have. And what about disease? Smallpox! Swine flu! The next AIDS!
I don’t know how you can bear to leave the house with your apparent level of risk aversion. The point remains, whatever you spend your resources on deprives you of being able to spend those resources elsewhere. People might die from global warming in 100 years, but people are dying right now from lots of things that could be addressed with the resources you’d have us devote to an uncertain effort to contain an uncertain risk that will unfold over an uncertain period of time. A child dying today of starvation deserves your attention just as much as or more than your progeny 100 years hence.
— andrew · Jul 8, 05:59 PM · #