Monday, March 12, 2007

Global Warming: A Taxonomy of Questions

CafĂ© Hayek has a couple of good posts on global warming, one by Don Boudreaux, the other by Russ Roberts. I especially liked this sentence from Boudreaux: “I am not so much a skeptic of global warming. But I am indeed a skeptic of combating global warming.” A crucial distinction. The global warming debate is too often cast as a simple dichotomy: either you believe global warming is real, human-caused, hugely destructive, and cause for massive political action… or you don’t, in which case you’re a skeptic. But the debate has many more facets than that.

I’ve been trying to create a taxonomy of all the questions that ought to affect our choices about dealing with global warming. The first three questions are all best directed toward atmospheric scientists. Since I’m not one, I’m inclined to defer to the experts on these, provided I don’t think they have institutional incentives to reach certain preconceived conclusions. My perception is that the debate has revolved almost entirely around the first two questions, with some (but insufficient) emphasis on the third. But the other questions are just as relevant and will become more important as we ponder policy responses.

Existence of global warming. Is there in fact a trend of rising global temperatures? My impression is there’s a great deal of agreement on this. Call it 95%+ confidence.

Human contribution to global warming. Has human activity contributed to that trend? My impression is that support for this proposition is somewhat lower; call it 90% confidence. But I’m still not sure how much of the effect scientists think is attributable to human activity, though it’s pretty clearly less than 100%. In any case, this question may not be terribly relevant; as David Friedman points out, even if warming were entirely caused by non-human factors, we might still want to try abating it if the harms were large enough and our responses efficacious enough.

Magnitude of the warming effect. How much warming will actually occur? According to the IPCCC report, the consensus seems to be that global average temperature will rise by a couple of degrees Centigrade, with a rise in sea levels of maybe 1.5 feet (with a confidence interval around both estimates, of course).

Net harms or benefits due to warming. The key word here is net. There will assuredly be some benefits from global warming, including longer growing seasons in some areas, lower heating bills, perhaps fewer deaths from cold winters, etc. It makes absolutely no sense to focus on harms only while ignoring benefits, yet I almost never hear anyone make this point in mainstream forums. Even if the harms are larger than the benefits – as they may well be – we still have to subtract out the benefits to determine the desirability of any proposed policy responses. After all, any policy that reduces global warming will reduce its benefits as well as its harms (though not necessarily to the same extent).

Extent of decentralized response. Given that the changes wrought by global warming will occur over the course of many decades, there is every reason to believe that individuals and markets will have time to respond to changing land values and resource uses. They won’t just wake up one morning to find that Florida’s coast line has disappeared and still be looking at the same old maps. Real estate prices will adjust, people will move, land uses will change. (Unless, of course, ill-advised government insurance programs give people an incentive to stay put in flood-prone areas.) Decentralized responses mean that harms of global warming are likely to be smaller than predicted by non-economists (or lousy economists). It also means that benefits will probably be larger, as entrepreneurial efforts lead to the exploitation of new opportunities created by global warming.

Marginal impact of collective abatement efforts. Supposing we could get all the nations of the world to cooperate, how much could we expect to reduce global warming? Humans have been burning fossil fuels at a high rate for at least a century, and climate change responds to the stock of greenhouse gases in the atmosphere, not just the current flow. So to a great extent, the damage has essentially already been done. If concerted global efforts could only reduce the expected rise in sea level by a couple of inches (a number I’m pulling out of thin air, since I have no idea what the real marginal effect would be), that should reduce the perceived desirability of policy responses.

Marginal impact of unilateral abatement efforts. Realistically, we cannot get all the nations of the world to cooperate. Some rapidly developing nations like China and India will at least be reticent to adopt growth-restricting policies. Given the political reality, how much effect will unilateral efforts – or efforts adopted by only some nations – have on global warming? Although this consideration would probably reduce the perceived desirability of policy responses, it could point the other way if additional response by some nations could make up for the inaction of others.

There are many more questions than these, particularly with regard to the range of policy responses, their differing marginal efficacy, and their differing costs. So my taxonomy is incomplete, but this list is at least a start. Contributions welcome.


Mike Lutz said...

You hint at this in your last three questions, but I would have a specific one on the relative effects of abatement vs. adaptation.

Joel said...

How about the degree to which these problems are self-correcting?

Cheap dirty energy encourages economic growth, spurring technological advancement. Technological advancements make cleaner sources, and more efficient uses, of energy available.

Colin Keese said...

The certain and possible losses associated with unilateral and multi lateral political actions need to be examined. It seems as if we will lose a lot in terms of liberty and productivity if even the moderate steps are taken, let alone "bold" action like the EU's.

Lindsey said...

I think that your image of environmentalists and other global warming activists is rather limiting. Is your view of environmentalists as "luddites" not based around a generalization? If anything, global warming activists and specialists advocate technological innovations that will incorporate energy efficiency, in the form of buildings' energy needs, increasing the density of cities, decreasing driving distances and the distances of shipping products for consumption.

I don't understand a necessary association with environmentalism and loss of liberty/productivity. If anything, facing environmental issues requires innovation and technological advancement - two qualities that a free market and American individualism hold in high regard. In fact, many of the suggested approaches towards combatting global warming are inherently free market, ie. carbon trading and purchasing carbon emission offests. I suggest taking a look at some of the more exciting advancements that environmental challenges are promoting. I like and

Furthermore, there are a lot of other reasons why we should change our approach to energy consumption. There might be unanswered questions about global warming, but LA's smog still looks like some kind of green swamp gas, and we are still seeing our lakes and rivers polluted at alarming rates. The list of worries goes on and on. We cannot simply sit around and wait stubbornly for some sort of scientific answer to your taxonomy of questions. While valuable, these answers are not prerequisites for action. There is plenty evidence of environmental degradation elsewhere.

Brian Moore said...

"Is your view of environmentalists as "luddites" not based around a generalization?"

I don't think he made that claim. In the cases you describe -- advanced technology makes things cleaner -- reinforces precisely what he's saying. That through the process of optimizing technology, we will help reduce the negative effects described.

"I don't understand a necessary association with environmentalism and loss of liberty/productivity."

It isn't necessary, but many of the policies advocated by environmentalists do indeed encroach on liberty and productivity. He would propose ones that do not.

"In fact, many of the suggested approaches towards combatting global warming are inherently free market,"

And I think those are ones we all agree on already.

"We cannot simply sit around and wait stubbornly for some sort of scientific answer to your taxonomy of questions. While valuable, these answers are not prerequisites for action. There is plenty evidence of environmental degradation elsewhere."

But he's specifically talking about global warming. If someone is dumping toxic waste in a river, that's outside the scope of his post.

Benoit said...

I think we should look at global warming through the lens of Decision Theory. After all, the goal of the whole global warming debate is to make a decision about the best actions to take with regard to this phenomenon.

Decision theory has two main steps. The first step is to determine the probability of the different outcomes for each possible action including reducing greenhouse gases to different levels vs doing nothing. The debate seems to largely stay on this level. Mathematically we could write probability equations like:

p(no_warming| reducing carbon emissions)
p(warming | reducing carbon emissions)
p(no_warming |doing nothing)
p(warming| doing nothing)

or using a more specific versions with probability densities.

p(warming=x degrees| reducing carbon emissions)
p(warming=x degrees| doing nothing)
p(warming=x degrees| actions3)

The second step which seems to be largely ignored in the debate, (although I think, some people use it intuitively which might explain why they are alarmed by GW) is to determine the cost/benefit or utility of each of the outcomes. Then we can multiply the probability densities of outcomes by the cost of these outcomes (or cost probability distributions, adding more dimensions to the problem) to get a utility distribution. We can use this to compute an expected utility for each action or an expected cost distribution over all the possible actions. A rational agent should pick the action with the highest expected utility or lower cost.

I think the reason people are intuitively alarmed by global warming is not that they necessarily thing that it is that much probable, but they perceive the potential cost as being very large and when you multiplying even a low, say 5% probability of warming by the huge potential cost, it can lead to rationally choosing an action that will protect us from it.

In the case of global warming, because of the global nature of the problem, the range of magnitude of the potential costs, like the costs of the effects of warming _and_ the cost of mitigating it, somewhat overshadows the range of the probability of their actually being global warming. The debate should reflect that. I think we should focus more on the costs of mitigation vs effect of warming rather than the probability of their actually be warming (unless we think it is very low).

What I am trying to say here is that, _even_ if global warming is not THAT likely (like over 1%), doing something to mitigate it might still be rational. This is the same rational we use when we protect ourselves from unlikely events by using safety equipment, seatbelts etc.

For more about the use of decision theory over probability theory read the fascinating Bayesian book by E.T. Jaynes: Probability Theory: The Logic of Science

Glen Whitman said...

Benoit -- I agree with your basic decision theoretic approach. It's true that we shouldn't focus only the point estimate, but also on the range of possibilities. However, a couple of caveats:

1. For those "outer limits" cases, we have to discount them by their low probability. In addition, I suspect those are the cases in which our avoidance efforts will have the least marginal impact, because the bad effects are largely the result of actions already taken.

2. If we're going to consider the whole range of possibilities, we also have to consider the other end as well -- the possibility that global warming hardly has any effect at all, or that our carbon emissions are the only thing staving off global cooling (another ice age). I'm not saying those are likely cases, only that are possibilities with as much justification for being included in our calculations as do the unlikely catastrophe scenarios.

Lindsey said...

Back to Brian:
True, I rather sloppily compiled a list of complaints garnered from the post, the link the post referred to, and the comments on the post - not very good journalism work! Sorry to Glen who posted a quite rational, albeit guarded, op-ed. I suppose my opinions on the matter get ahead of me...

But tell me if I'm completely wrong when I say that there certainly seems to be a sense of market vs. environment in these discussions, as though what is best for both is mutually exclusive - or at the very least, a sense that embracing a future fight against climate change is antagonistic towards symbols of economic progress like individual liberty, etc. If this is not what is said verbatim, I read it between the lines. First, these discussions exhibit an inherent resistance towards theories of human-caused global warming and the need to institute lifestyle adjustments aimed at limiting (or preparing for) its effects. Why do we need this taxonomy of questions? What hysteria does a list of rational questions like these protect us from? I do not see this hysteria. From a pro-environmental standpoint, reactions to global warming predictions have been agonizingly slow and foot-dragging.

Most importantly, I would like to emphsize that global warming is very much related to every other dire environmental consequence resulting from human lifestyles of overconsumption, pollution, and waste. If the sh** were to hit the fan, so to speak, it would certainly help our chances of survival and adaptability if our rivers and soil weren't polluted, our oceans still had fish in them, and we weren't so oil-dependent. That's why I bring up other environmental issues - because they are all vastly interconnected in a web of life that is far more complicated than human minds can possibly understand, quantify, or articulate. We can only value life and try to treat the earth with respect - it's much bigger than us.

Bay said...

It's notable that scientists have shifted to using the term "Global Climate Change" rather than "Global Warming". In terms of fewer heating bills, etc. melting ice caps may bring colder water into the oceans thereby cooling the globe over all. Some NASA scientists state that global climate change could bring about an ice change.

For related data sets: