Yesterday I blogged about Paul Krugman’s recent piece on climate control policy. The bottom line: After recovering from a shaky start, Krugman does a good job of laying out the issues and posing many (though not quite all) of the right questions. But I’m not sure he gets the answers right.
A few years ago, writing in Slate, I listed the key questions that the Al Gores of the world mostly fail to address or even acknowledge. (See also the discussion on pages 186-190 of The Big Questions.) Krugman (thankfully) is no Al Gore, and he does address most of these questions. Let’s see how he does with them.
- How much does human activity affect the climate? This is of course a hard question and one where economists have little in the way of special expertise. (We do, however, have some experience interpreting data and some sense of how models work.) Krugman gives substantial credence to estimates along the lines of a 9 degree Fahrenheit increase over the next century. I have absolutely no idea what to believe on this issue, so I’ll go along with Krugman for the sake of argument.
- How much harm is likely to come from that climate change? This is the big one. I think it’s fair to say that virtually all economists believe that the harm from a given temperature increase is likely to be less than most climatologists believe. That’s because economists understand that—in Krugman’s words—“we have an advanced economy, that kind that has historically shown great ability to adapt to changed circumstances”. We will find ways to limit the damage, and those ways will be less expensive than you think.
Suppose for example that the sea level rises to the point where we have to move New York City. Non-economists envision abandoning the existing city and starting over from scratch somewhere else. Economists recognize that most of New York City is going to be rebuilt over the next century anyway (because few buildings last a hundred years) and that it can gradually move inland one building at a time. Non-economists envision massive crop failures and a mad scramble to find new food sources. Economists envision farmers—who are constantly experimenting with new crop varieties anyway—gradually shifting to varieties adapted to an increasingly warmer climate.
Krugman sides with his fellow economists on this (it would be hard to take him seriously if he didn’t) but rests much of his case for climate control policies on his belief that there remains some significant chance of utter catastrophe a century or so down the line. I have no idea how plausible that scenario is, but even if we accept it at face value it means that Krugman’s case for climate control depends on some fairly distant future effects. This magnifies the importance of the next question.
- How much do we—or should we—care about future generations? The famous Stern Report to the British government based its policy recommendations on the assumption that we should care about future generations exactly as much as we care about ourselves. Of course we do no such thing. The fact that we tax capital income (thereby discouraging saving and impoverishing future generations) is pretty good evidence that compared to ourselves, we hardly care about our great-grandchildren at all. If you’re going to make climate policy on the basis of your deep concern for future generations, then consistency requires you to make tax policy on the same basis. But I don’t see the Sierra Club lobbying to eliminate the corporate income tax.
Krugman disavows the extreme position of the Stern Report, but still argues for weighing future generations’ welfare more heavily than private markets do. There’s a good theoretical case to be made for that position, though Krugman does not digress into making that case (and neither will I).
Of course in our concern for future generations, we should remember that they will bear not only the benefits of our climate control polices but also the costs. Krugman suggests that the Waxman-Markey climate control bill might trim the annual growth rate by about .09 percentage points, say from 2.4 to 2.31, and he emphasizes how small that change is. What he doesn’t tell you is that if you trim the growth rate from 2.4 to 2.31, then you’ve cut the incomes of our descendants a century from now by about 10%. That’s not a catastrophe, but it’s not small either.
- How likely are those future generations to be around, anyway? As I said in Slate, if you think the earth is going to be destroyed by an asteroid in the next 200 years, you don’t want to spend money improving the climate 300 years from now. For that matter, if you think the earth is going to be too hot to inhabit 100 years from now no matter what we do, then you don’t want to spend money reducing that future temperature down from 300 degrees to 250. Krugman is quick to entertain doomsday scenarios for man-made climate change, but he never addresses the likelihood of alternative doomsday scenarios.
- Just how rich are those future generations likely to be? If you accept Krugman’s 2.4 (or 2.31) percent growth rate—which I think is probably low—then a century from now, our descendants’ incomes will be about ten times ours. (That’s ten times after adjusting for inflation; in other words they’ll be able to afford ten times as much stuff as we can.) Do we really want to spend money improving the quality of life for our fabulously wealthy descendants when we could spend that same money improving the quality of life for our contemporaries living in rural Africa on $400 a year? Krugman fails to address this.
- How risk averse are we? Here, I think, Krugman has the economics wrong—or at least incomplete. He says that the more risk-averse we are, the more we should spend on climate control policies, because risk-averse people like to prevent catastrophes. But risk aversion cuts two ways. One of the biggest risks we all face is the risk of being born (relatively) poor. The way to alleviate that risk is to make poor people richer. Compared to our fabulously wealthy descendants, we are the poor—and one way to make us richer is to stop asking us to spend so much on climate control.
Krugman is right that risk-aversion creates an argument for more climate control spending. What he doesn’t tell you is that at the same time it creates a counter-argument for less spending. He should have acknowledged this. On the other hand, I’ll cut him slack since I realize a single magazine column can’t go on forever. Neither can a blog post. I have just a little more to add (watch this space later in the week) but I’ll stop here for now.
The fact that we tax capital income (thereby discouraging saving and impoverishing future generations) is pretty good evidence that compared to ourselves, we hardly care about our great-grandchildren at all
Really? Then what does the fact that we tax labor income tell us?
Great post. I know this is probably ridiculously naive, but is there any way to get Paul Krugman and Steve Landsburg in a room discussing climate policy? It would be extremely refreshing to hear honest people with different positions actually talking to each other.
I promise to promote the youtube link if it happens…
Economists are nearly ignorants on economics. So what do I have to say about economists in climatology?
Regarding question 5: Comparing generations by measuring their GDP per capita makes sense only if the things that matter for their welfare but are not included in the GDP “stay the same” across time: for example, the quality of air and water, how blue the sky is etc. This may be roughly true for a 10 year difference but not for a 100 year difference, especially if the climate might change dramatically. Maybe the future generations will be able to buy 10 times more cars, refrigerators and plane tickets but if the sky is red, how rich are they really?
@Steve, your post highlights the difference between economists and non-economists. Overly simplifying your argument, you appear to be saying, “if there’s a 5% chance that climate change will end life on earth [whatever that means], we should be prepared to spend up to 5% of GWP to prevent this from happening.” Most non-economists (other than apocalyptic evangelicals) would say that if there is a 5% chance of ending life on earth, we should make 100% sure this doesn’t come to pass, no matter what the cost.
Good thoughts, but I have a few nit picks:
The biggest impact predicted by climate change isn’t a temperature increase, but rather larger temperature extremes. It is hard to account for this in farming by just using warmer temperature crops since those crops general have trouble weather the low temperatures as the weather swings.
Don’t forget about black swan principles and our inability to make these kinds of future projections.
To address #6, that is like saying buying insurance isn’t risk adverse because it makes you poorer due not only to the cost of the insurance but the additional premium that serves as the insurance companies profit margin.
You can’t count “being poor” as a risk if you can predict it exactly and than choose how “poor” you wish to make yourself by choosing how much to spend on climate controls.
AI V: what if there was a 0.5% chance? or 0.05% chance?
And when you say “no matter what the cost”, do you include “obliterate all human life on the planet” in that equation?
I have a hard time taking carbon reduction arguments (pro or con) seriously because they are not realistic.
That is, until somebody puts forth a credible scenario way that China, India, and the developing world can be convinced to stop emitting carbon (and thus keep their citizens poor) I really do not care what rich countries do about carbon reduction.
I am much more interested in geo-engineering as that is at least potentially realistic.
Is there not something contradictory about believing that future generations will be 10 times richer than we are AND that global warming is a looming disaster?
“Most non-economists (other than apocalyptic evangelicals) would say that if there is a 5% chance of ending life on earth, we should make 100% sure this doesn’t come to pass, no matter what the cost”
I would disagree with this statement. Firstly, “most non-economists” probably would NOT make that statement. Secondly, we make those kinds of risk assessments on a daily basis and never react so risk adversely.
Chance of getting killed driving to work.
Chance of getting food poisoning and dying, by eating food purchased at a local school fundraising fair.
Chance of getting cancer after a lifetime of diagnostic xrays.
Chance of getting killed flying to exotic locations for vacations.
Chance of slamming into a tree sking.
etc.
If we should make 100% sure that these things never happen, then the only “logical” response would be refrain from any of these actions. Life is not RISK FREE.
I’ve always wondered about this ‘annual growth rate’ thing and how it will make my great-great grandchildren incredibly wealthy. So… does it mean that everyone in the future is going to have a Mercedes and a solid-gold toilet? I’m really not sure how this works. Would people work more? Invest wisely? It seems (to me) like my wealth at a given time is related to everyone else’s wealth. How would the day-to-day life of people be affected?
(I’m not necessarily doubting it, I just don’t get it)
@Super-Fly
Your wealth is the amount of stuff you have. Over time, humans get more creative and knowledgeable and figure out new efficient ways to provide goods and services. Capital reaches a steady-state level, but there is no upper bound to technological progress or improved rules and institutions.
Super-Fly, just think about how people lived 100 years ago. No air travel, little auto traffic, no television, primitive dentistry and medical care….
“I have absolutely no idea what to believe on this issue, so I’ll go along with Krugman for the sake of argument.”
Why go along with Krugman on this issue for any reason?
Can Krugman – or anyone – point to a statistical global temperature model, covering, say, the last forty or fifty years, which employs CO2, solar activity, and other factors as independent variables and which shows a significant r-square for the CO2 variable? Without doubting the theoretical physics of the greenhouse effect (of CO2), the most important question is clearly its measured impact.
“Most non-economists (other than apocalyptic evangelicals) would say that if there is a 5% chance of ending life on earth, we should make 100% sure this doesn’t come to pass, no matter what the cost”
I realise that has already been done but if that statement is true, it’s pretty strong evidence that we should listen to what economists have to say about this question, and not non-economist. Economists would stop and ask themselves the relevant question: what is an appropriate cost to pay for an extra 5% chance of survival (since clearly we can’t avert all possible risks)?
As it happens, I think most people who’ve thought hard about existential risks (eg, Nick Bostrom and especially Eliezer Yudkowsky) tend to think that the average person isn’t willing to spend *enough* to avoid them – whatever the reality, it’s pretty clear that there is such a thing as enough, and such a thing as too much.
1. Since all scientific societies of the major developed countries accept that man-made global warming is virtually certain, it would be absurd to offer a differing view as a non-specialist. The IPCC reports set out a range of possible outcomes, the most likely is most widely advertised. It also includes less-likely outcomes, such as less warming, and at the other end, much greater warming. There is a possibility of collapse of Greenland icesheets in the not too distant future – which would cause several meters of sea level rise. This is not very likely over the next hundred years. There is also a realistic possibility that positive feedback systems such as methane emmissions from melting permafrost could cause rapid acceleration of the warming.
2. That there will be harm is almost certain. There is also a significant (much more than 5%) chance that the harm will be catastrophic, at least to some. Moving a building at a time may not be viable, as they will all become untenable at the same time. However, there will be different methods not yet thought of to ameliorate the damage which the wealthy nations will implement. The Bangladeshi’s do not have such an advanced economy. They can only just cope as things are now. As the sea level rises, there will be more floods. It seems you are saying carry on and cope may be better than to stop the temperature rise. Bangladesh will only cope with huge human suffering and great many deaths. Other areas will suffer droughts and floods as weather patterns change. They will “cope” in the sense that people living in unviable areas will die, thus solving the problem. There will be many deaths. We know that the human costs will be borne initially mostly by the developing world.
3. If the future generations are going to be so fabulously wealthy, then reducing that fabulous wealth by 10% should not matter very much to us. The greatest risk to that wealth is probably global warming.
4. There are some very small risks such as asteroid impact which could cause human extinction. These are very unlikely in the next few centuries, but possible. We should put a bit more money into near-Earth object surveys to reduce these risks. Compared to global warming, the costs are tiny. There is no scientific consensus that the Earth will be 300 degrees (unlesss you mean Kelvin) . I don’t think there is any projection that indicates Earth will be uninhabitable by humans. However, it is possible that our society will not survive.
5. Humans have been around for a few hundred thousand years. Economic growth has been around for 300 years. To extrapolate from current data is always risky, and the further you extrapolate the riskier it becomes. To attempt to extrapolate 100 years into the future based on 300 years of anomalous data is not very rigorous. There is one thing pretty certain – exponential growth cannot continue forever in a finite system – a limit must be reached somewhere. Our growth so far has partly depended on cheap energy. We seem to be getting to the limit of our ability to continue to grow the oil supply, so the oil price will probably be high from now on. (Predicting global warming is not the same as extrapolating growth rates as it is based on theory, not just extrapolation.)
There are two types of cost to reducing greenhouse gases – damage limitation and investment. Carbon Capture and Storage (CCS) is an example of the former – it increases the cost of coal powered energy and provides no benefit other than CO2 reduction. However, CCS and similar will be needed in the short term. Investment in solar power is the latter – it will provide cheap energy for decades to come. Investment in these schemes will eventually enrich our descendents – and even our future selves. Currently it is impossible because the energy produced is slightly more expensive than current energy production costs, which do not include the undeniable costs of CO2 emmissions. There are also political problems as the energy must be transported across large distances and borders. If we put serious effort into these schemes we could soon be enjoying cheap energy from these sources.
But let us not suppose we know all the answers. Let the market decide. It is almost undeniable that there are costs to CO2 emmissions. Let the emitters pay these costs, whether through cap and trade or Pigovian tax doesn’t matter. I think we will then see solar power cheaper than coal. Or possibly tide, or wind, or wood, or waves. Who knows? At the moment these do not get a look in because fossil fuels are subsidised by everyone on Earth. There is no political will to implement either a tax or cap and trade at a level that would be meaningful.
The third option of geo-engineering is really the stupidest. If we can’t manage to achieve a simple technological fix with low costs, known technology and predictable outcomes, why should we put our faith in uncertain, untried speculation with unknown (but enormous) costs, and unknown effects?
We know as certainly as we know anything that our current actions will cause harm to people in the future. We are not absolutely sure if this harm will be bad, very bad or catastrophic. We know that the costs, financial and suffering, will rise rapidly the longer we leave it to take action. We know the costs will fall most heavily on the poorest and least able to adapt. We could address this with relatively little cost to ourselves. Your argument not to do this is to a) Assume that the continuation of current growth trends will be greater than the rise in the costs and say “leave it to the future as they will be wealthier than us”. b) Assume there will be a technological fix or innovation to reduce the cost. I don’t buy it. I do, however, hope you are right, as the chance of us actually doing anything meaningful is remote.
@Dave – I agree, I’m not arguing that we should spend billions to mitigate an infintesimal risk, only that there is a difference between hiw economists and the general public view the question.
This reminds me of some of the concerns raised when the LHC was turned on. A physicist stated that there is an infintesimal risk that the LHC could create a black hole that would destroy the earth. Many people were horrified, not understanding the magnitude of the risk.