Global Warming and the Ghost of Frank Knight

Frank H. Knight (1885-1972), a University of Chicago economist recognized as one of the deepest thinkers in 20th century US social science, is famous for his distinction between risk and uncertainty. Although he could never have anticipated all the ways it could be applied, Knight’'s 1921 distinction helps explain why it is confused to put any faith in a market for emissions credits generated by carbon-saving projects.

Risk, in Knight’s sense, refers to situations in which the probability of something going wrong is well-known. An example is the flip of a coin. There is a 50-50 chance of its being either heads or tails. If you gamble on heads, you risk losing your money if it turns out to be tails. But you know exactly what the odds are.

Uncertainty is different. Here, you know all the things that can go wrong, but can’t calculate the probability of a harmful result. For example, scientists know that the use of antibiotics in animal feed induces resistance to antibiotics in humans, but can’t be sure what the probabilities are that any particular antibiotic will become useless over the next 10 years.

Still worse, as Knight’s successors such as Poul Harremoes and colleagues have pointed out (The Precautionary Principle in the 20th Century, Earthscan, London, 2000), are situations of ignorance. Here you don’t even know all the things that might go wrong, much less the probability of their causing harm. For example, before 1974 no one knew that CFCs could cause ozone layer damage.

Obviously, this ignorance would have invalidated any attempt, at the time, to calculate of the probability of ozone depletion. Similarly, before 2000, it was not known that the albedo of trees could change a forest’s effect on global warming; before 2005, how much carbon recently sequestered by land plants is being moved by the Amazon to the oceans and the atmosphere; and before the 1990s, that certain factors including release of methane from ocean floors or the switching off of the Gulf Stream were capable of “flipping” the earth’'s climate rapidly from one state to the other.

In situations of indeterminacy, finally, the probability of a result cannot be calculated because it is not a matter of prediction, but of decision. For example, it might be “implausible” for subsidies for fossil fuel extraction to be removed within five years, but you can’t assign a numerical probability to this result, because whether it happens or not depends on politics. In fact, trying to assign a probability to this outcome can itself affect the likelihood of the outcome. In such contexts, the exercise of prediction can undermine itself. Risk, uncertainty, ignorance and indeterminacy each call for different kinds of precaution.

Risk fits easily into economic thinking, because it can be measured. For instance, as Knight pointed out in 1921, “the bursting of bottles does not introduce an uncertainty or hazard into the business of producing champagne”:
“
"[S]ince in the operations of any producer a practically constant and known proportion of the bottles burst, it does not especially matter even whether the proportion is large or small. The loss becomes a fixed cost in the industry and is passed on to the consumer, like the outlays for labor or materials or any other. . . This, of course, is the principle of insurance, as familiarly illustrated by the chance of fire loss"” (Risk, Uncertainty and Profit, Houghton Mifflin, Boston, 1921).

Uncertainty, ignorance and indeterminacy, however, call for a more precautionary and flexible, and less numerical, approach.

Take the carbon credits to be generated by tree plantations. If these credits were threatened by nothing more than risk, insurance-type calculating techniques would be enough to handle the problem. You could insure carbon credits from a plantation just as you take out fire insurance for a building.

But such credits are characterized not only by risk, but by uncertainty, ignorance, and indeterminacy as well. For example:

*How long will plantations last before they release the carbon they have stored to the atmosphere again, through being burned down or cut down to make paper or lumber, which themselves ultimately decay? This is not simply a risk, in Knight'’s sense, but involves uncertainties and ignorance that can’t be captured in numbers. For example, it is still not known what precise effects different degrees of global warming will have on the cycling of carbon between different kinds of trees and the atmosphere.

*How will plantations affect the carbon production associated with neighbouring ecosystems, communities, and trade patterns? Again, uncertainty and ignorance, not just risk, stands in the way of answering such questions.

*How many credits should be subtracted from the total generated by plantations to account for the activities that they displace that are more beneficial for the atmosphere in the long term, for example, investment in energy efficiency or ecological farming? No single number can be given in answer to this question, since “it is inherently impossible to verify what would have happened in the absence of the project”. That is, the answer is indeterminate.

By mixing up “the analytically distinct concepts” of risk, uncertainty, ignorance and indeterminacy, schemes such as the Clean Development Mechanism and Joint Implementation have blundered into what Knight called a “"fatal ambiguity"”.

In this case, it is an ambiguity that undermines the effectiveness of the entire Kyoto Protocol and one that can only be remedied by the suspension of such projects.

---------------------------------------
For more information: http://www.thecornerhouse.org.uk/.