Spitting In The Ocean
Perceived Barriers to Marginal Cost Pricing
In a paper that appeared in the American Economic Review1 , the Harvard economist Robert Stavins sets up a modeling framework to address the question, how can we most cheaply achieve reductions in atmospheric carbon. In the essay, he looks at the competitive prospects for sequestering carbon in forests, as a substitute for actually reducing emissions of fossil carbon into the atmosphere. After presenting his argument for quantitatively framing this question, he notes this:
It might be argued that since the policy intervention I model is a tax/subsidy on land use, not carbon emissions and sequestration, it does not lead to the true (minimum) carbon sequestration marginal cost function. This criticism is not valid in a realistic policy context. It would be virtually impossible to levy a tax on carbon emissions or a subsidy on sequestration, because the costs of administering such a policy would be prohibitive.
As important an issue as atmospheric carbon may be, I do not really have a dog in that fight. So the veracity of Professor Stavins' assertion about the costs of administering a unit pricing scheme for carbon markets is not what interests me about this quote. What interests me about the assertion that unit pricing could raise total costs instead of reducing them is that it might be generalized to other pollution problems. Like non-point source nutrient pollution reduction in the Chesapeake Bay drainage.
Let's be clear on what we are talking about. If we buy pollution reduction on the basis of an easily measured proxy such as acres on which pollution reduction practices are employed, then we are accepting an average price for all of the acres available for those pollution reduction practices. Realistically, we know that some of those acres will reduce more pollution than others, even given the same pollution reduction practice. But by pricing pollution reduction by the acre, we are accepting an average return of pollution reduction for our payment.
The alternative to buying pollution reduction by the acre is buying it by the pound. That is, using what we think we know about the reduction efficiency of a practice on various sorts of acres to figure out how many pounds of pollutant were reduced on any particular acre and to price that acre accordingly. Such a payment system would induce suppliers of pollution reduction to line up from most efficient to least efficient, and we would expect to achieve greater pollution reduction with any budget, short of buying it all.
This alternative, unit-based pricing approach is what Professor Stavins asserts to be prohibitively expensive in carbon markets. Would it be prohibitively expensive in nutrient pollution mitigation programs in the Chesapeake? Well, let's look at cover crops. What do we need to know to make payments per acre dependent on the number of pounds of nitrogen that are expected to be reduced by planting cover crops on that acre?
We need to know whether the acre is on the coastal plain or the non-coastal plain. We need to know how and when the cover crop was planted and what cover crop seed was used. And, we need to know what was planted on the acre the previous growing season and how it was grown. This may seem like a lot of information to collect, but it is all being collected by Maryland's current Cover Crop Program. So collecting the information that is needed to change to a unit-based payment system would not impose additional costs. Perhaps using it would.
To use the information described above, which is thought to be sufficient for predicting loads and load reduction for the TMDL regulation, one would need a way to collate it so that suppliers could easily find out what they might be paid under the available alternatives. We have devised a way to do that and posted it here: www.mainstreeteconomics-b.com. In our schematic pricing tools, we have pulled "per-pound" prices out of our hats. Real prices would need to be set by policy-makers figuring out how to get the most nutrient reduction for the least money.
The only way that this pricing tool would be costly is if we obtained a patent for the process and charged a toll every time it was used. This is not a bad idea, but even if we did that, I doubt that we could make it as costly as it would be cost-reducing. So I don't believe that Professor Stavins' assertion about unit pricing applies to programs targeting nutrient pollution reduction in the Chesapeake Bay drainage. Not for cover crops or riparian buffers, anyway.
1 Robert Stavins, The Costs of Carbon Sequestration: A Revealed Preference Approach. The American Economic Review; Sept. 1999, 89,4, pp. 994 -1009.