Below a small article I submitted to 'Planners' Newsletter' as an item for discussion,
and a small discussion which followed.
'Planners' Newsletter' was an e-mail forum within Shell up until 2004.
| Economics and Climate Policies the fallacy of cost benefit assessments The economics of climate policies are quite often characterized by the following notions:
To be able to make a cost benefit assessment a project (or a scenario) needs to be defined, which starts from an assumed emission reduction (of greenhouse gases), a climate change or a combination of both. As the benefits of a climate policy include the costs avoided due to climate change, it is required to know the potential climate changes irrespective of the policy to be implemented. Dealing with uncertainties However, there is a great uncertainty in each step in the following cause-consequence chain:
This is already very difficult or next to impossible for ‘normal’ climate change scenarios. In such studies the consequences of a ‘drastic’ scenario (e.g. a temporary shut-down of the North Atlantic drift, putting Europe in the cold) are not included. And my feeling is that they cannot be included. It would end up in wild speculations. Another source of uncertainty is the issue of local developments all over the world. It is on a timescale of say 50 years very difficult to forecast what will happen where (from an economic point of view). The same holds for climate change projections. For the World as a whole, forecasts can be made (with quite a margin of uncertainty: a temperature rise of 1.5-6 degrees C by 2100), but as soon as one tries to make more localized forecasts the margins of uncertainty become even bigger. Secondly, damaging weather incidents tend to behave very erratically. When in 1992 Andrew made its $30 billion hit onto Florida, the Caribbean enjoyed a quiet hurricane season! Since 1995 (with exception of the El Niño year 1997) there has been much more hurricane activity in the North Atlantic and the Caribbean. In spite of this there haven’t been any real straight hits like Andrew. This may be plain luck, but makes the point quite clearly. There is no ‘optimum’ A serious and solid climate policy will include structural adjustments in the energy infrastructure and transportation systems. This will require increased fundamental knowledge, technical innovation and a drastic change in the way infrastructures are run. It will also imply the unlocking of dominant technologies, especially in the fields of energy, materials and transport. However, necessary changes in the economic structures are uncertain and may well develop more slowly than anticipated. Therefore it is in my opinion wise to stimulate the developments mentioned above at an early stage. The notion of an ‘optimal’ investment strategy with respect to these new developments, e.g. looking for a good timing, is in my opinion leaning too much on the assumption that the future is predictable and can be managed into a desired direction. (In Dutch: "maakbaarheid") Dealing with complexity We are dealing with a very complex issue which is characterized by uncertain facts, values in dispute, high stakes and potentially urgent decisions. As typified above, it is extremely doubtful whether it is possible to make a model of a climate-biosphere-economy system. The report "Limits to Growth" of the Club of Rome contained a model which was slashed down as too simplistic by many economists. But are the present models on the economics of climate change much better? As an example, one may look at the following: as a consequence of climate change, food production collapses in a part of the world. If this is a (highly) developed country it will mean a decrease of its Gross National Product by 1-2%. However, needless to say, the consequences can be much more far reaching. It may initiate a total economic collapse. If one compares the risks of climate change with the ‘tiny chance - catastrophic effects’ of nuclear energy, one can observe that cost benefit assessments didn’t play a role in the debate on the use of nuclear energy. One could take a similar stance with respect to climate change. A precautionary approach Adaptation under stable conditions is safer than adaptation under unstable conditions. Waiting longer in order to get more information is not useful. The biosphere is so complex that the position of critical thresholds remains an unknown. Therefore the chance of a discontinuous catastrophic change can never be eliminated. Against this background, a climate policy based on a strategy of precaution is a rational one. The use of cost benefit assessments is in my opinion useless, even harmful. They are often only used to maintain the status quo by opponents of climate policies, in spite of the fact that the value of these assessments is extremely nebulous. What finally remains is the debate on how far one should go in the use of the precautionary principle, and how it should be handled. This is not always done in a proper way, as was shown in the Waddenzee (a fragile coastal wetland in the Netherlands) discussion in the Dutch parliament. Drilling (for natural gas) was forbidden, whilst the effects on the environment would be very small, but fishing for cockles and mussels was not forbidden in spite of proven adverse effects (it contributed to mass starvation of Eider Ducks).
The Precautionary Principle is a Double-Edged Sword
Evert has spectacularly demonstrated an important point - although perhaps not the one he set out to demonstrate. He makes a telling and eloquent case for the use of cost-benefit analysis. Why? Because without cost-benefit analysis, I can turn Evert’s arguments against him. Evert deploys the standard case: that climate change is highly uncertain, it is a complex system, radical damage cannot be ruled out: "The biosphere is so complex that the position of critical thresholds remains an unknown. Therefore the chance of a discontinuous catastrophic change can never be eliminated. Against this background, a climate policy based on a strategy of precaution is a rational one." But this is also true of the economic system. If I were so minded I could use the ‘precautionary principle’ to argue that we should do nothing about climate change. An attempt to move to a low-carbon economy could lead to economic collapse, mass starvation and war. This is a low-probability, high-impact event and subject to the ‘precautionary principle’. Therefore we should under no circumstances attempt to move to a low-carbon economy: "The economy is so complex that the position of critical thresholds remains an unknown. Therefore the chance of a discontinuous catastrophic change can never be eliminated. Against this background, an economic policy based on a strategy of precaution is a rational one." When the same slogans can be used to argue for whatever prejudice takes our fancy, we need to look elsewhere. And cost-benefit analysis remains the perfect tool for the job. From analysis to policy Cost-benefit analysis is not the straw man Evert makes out. He implies that it is not possible to accommodate uncertainty, nor to build in information about the benefits of learning by waiting (be they large or tiny). But of course these considerations can be built into cost-benefit analysis. When uncertainty is high, cost-benefit analysis will give a wide range of answers - but properly done it gives us a much better guide to action than the empty sentiments of the ‘precautionary principle’. For instance, the economic models DICE and RICE suggest that modest reductions of carbon dioxide should not be costly at all if the reductions are structured to pick up the easiest gains - which could be done through an auction. An auction for carbon permits, initially with a large number of permits available, would quickly reveal the truth: how much are societies willing to pay to avoid reducing emissions? And who finds it easiest to reduce emissions? And at the same time, the auction turns this information into action. I suspect that such an auction would quickly reveal that carbon emissions are easy to peg back - when auctions were used to reduce sulphur dioxide emissions in California, the low cost of abatement caught everybody by surprise. [There is more information about auctions for carbon permits both in the Economic Environment Review (forthcoming on the PXG website) and in some detail in this article by auction theorists Peter Cramton and Suzi Kerr:
But I have strayed from analysis into policy - and there are many obstacles to implementing such a policy. But that is perhaps the point. How are we to build consensus for action on climate change? While environmentalists try to argue that the possibility of disaster is an overwhelming argument for ‘a climate policy based on a strategy of precaution’, economists are busy designing policies to minimise the pain of adjustment. Taking the first few steps towards a low-carbon economy would probably be quite painless - and that would be a more telling argument for further action than the ‘precautionary principle’ will ever be.
Use Economic Analysis Tools Where They Lead to Useful Results
Tim tries to invert the arguments I used to arrive at a precautionary approach in climate policies (F1551): "An attempt to move to a low-carbon economy could lead to economic collapse, mass starvation and war. This is a low-probability, high-impact event and subject to the ‘precautionary principle’. Therefore we should under no circumstances attempt to move to a low-carbon economy: ‘The economy is so complex that the position of critical thresholds remains an unknown. Therefore the chance of a discontinuous catastrophic change can never be eliminated. Against this background, an economic policy based on a strategy of precaution is a rational one.’" In being able to invert an argument, one has to deal with subjects/issues which are to a sufficient degree symmetrical. Factors like the irreversibility of system changes have to be sufficiently similar. Also the ability (of humans) to redress (catastrophic) changes has to be of the same order of magnitude. Is this so when comparing "an economic policy based on a strategy of precaution" with "a climate policy based on a strategy of precaution". I don’t think so. First of all, one can observe that the economy is a highly complex system, but a climate-biosphere-economy system is a cascade of complex systems. The latter is much less well understood. I even doubt whether it is possible to build models which produce useful forecasts. Economic catastrophes versus climatic ones When looking at economic catastrophes (e.g. The Great Depression of the 1930s), humans - with some difficulty - have been able to clean up the mess. This also shows that the ill-founded policies which created this mess (the social/economic policies of the 1920s, especially in the US) were of a completely different nature than policies on climate issues which might be considered at this moment. So, precaution against what? Turning to the risk of climatic step changes, are we able to redress the mess, e.g. a shift of precipitation zones on Earth which turns cereal belts into (semi)deserts? I think that our ability to repair this sort of damage is of an entirely different order than in the case of economic mischief like The Great Depression. Cost estimates can be made with a very reasonable accuracy - also in the case of climate policies. In that case, economic tools are useful. However, making benefit assessments is next to impossible. If one reads the IPCC report carefully, one can find out that there is an overlap in global average temperature forecasts when comparing the following cases:
How turn this into policy and actions? However, and here Tim indirectly makes a good point, the question then arises of how to turn this into a concrete policy. Empty words are of no use. In my opinion, the things that should be done now are to make sustainable energy and transportation options available; start developing them now. Which ones are the best is still to a large degree an open question. How fast they will (have to be) implemented is currently also an open question. However, doing nothing, against the background of a policy of precaution, is no option. When developing policies, all kind of scenario studies (e.g. "Global Energy Perspectives" by the IIASA and the World Energy Council) are extremely useful. They help to get a broad picture of what is possible. In order to be able to implement a climate policy, a minimal degree of consensus has to be built. If economic tools like auctions of carbon emission permits are useful to find out what people are willing to pay, use them. If carbon emission permit trading systems lead to a better allocation of resources to decrease the carbon intensity of energy generation and transportation, use them. A disaster scenario? Finally, imagine the following: A category IV-V hurricane makes a direct hit onto Miami. There are few casualties (people were warned in time and drove off), but it is a ‘$80 billion whammy’. No doubt some people will start yelling "climate change". (If the Republicans in the US change their minds due to such an event - politicians with homeless voters tend to get a bit nervous - it may very well be doing the right thing for the wrong reason, because the relation between global warming and hurricane formation is still unclear.) But is it in this case possible to make a cost benefit assessment with respect to climate policies? No. I very often do not agree with doomsayers. However, in my opinion there is enough reason to call for some caution on the issue of climate change. We are talking of risks which are currently non-quantifiable. Cost assessments are useful. Scenario studies are often very useful. Other economic tools are often so too. But I don’t want to fool myself (that is how I feel about it) with benefit assessments of climate policies. |