Monday, June 02, 2008

Should California's ARB Trust Computable General Equilibrium Models for Judging AB 32's Likely Economic Effects?

Magic is in the air. The California Air Resources Board (ARB) faces a hard economic and political problem. It is mandated to implement climate change mitigation regulation. AB32 mandates a 80% cut below 1990 Greenhouse gas emissions levels by 2050! A reasonable question is what will be the costs and benefits of this regulation for the California economy and for consumers and firms?

To answer these important questions (or at least to take the heat off of politicians who endorse AB32), computable general equilibrium models (CGE) are being used to predict scenarios and likely impacts of regulation.

"ARB is starting to develop the Scoping Plan mandated by the Global Warming Solutions Act of 2006 (AB 32). Economic analysis is being developed and used to support the development of the Scoping Plan. The following contains information on three economic models, including links to more information on the models, their applications, and previous peer reviews."

California's AB32 and CGE Economic Models

While I appreciate the importance of forming predictions of the future, I also believe in confidence intervals! To be honest, these CGE models are crap. These black box models must take a strong stand on dozens of behavioral responses where in each of these cases we don't really know what the elasticity is. Feed in 20 bad estimates and you won't get back a great tasting meal.

Let me give you one example, suppose that a cap and trade system for carbon dioxide raises the price of water in Los Angeles by 2 cents per gallon --- how will this price increase change consumption of firms, households, and people who flush the toliet in Los Angeles? These CGE proponents don't know the answer but they make assumptions to generate a number and politicians then believe that the wise economists know the answer. (Why would a carbon tax raise the price of water? Pushing water around California is highly energy intensive and much of this electricity will come from natural gas fired power plants and they create greenhouse gases).

The CGE crew's model isn't science, it is magic and it makes economists look bad.

Here is my counter-proposal. The Air Resources Board should conduct a series of field experiments to actually estimates the structural demand and supply elasticities that it needs to judge how consumers and firms will adapt to the new post-AB32 pricing. Once these estimates exist, then these estimates could be plugged into the CGE models.

Even then I wouldn't be happy. If you sit down and read the technical documention to these CGE models, you will see that they are static models. In English, the models make no assumptions about technological change. I would bet my remaining hair on the top of my head that if energy prices go up in California that renewable energy technological advance will take place and this will lower the cost of complying with AB32 but this type of "endogenous technological advance" is ignored by the CGE crew.

Unfortunately, the politicians appear to be in such a rush they have gotten ahead of the basic research. I hope they are not shopping around for CGE models that generate results that they like.

I realize that my tone is sharp here but AB32 is multi-billion dollar regulation. Academic economists, rather than consulting firms, should be asked to do the heavy lifting here to figure out the likely impacts of the program. Now, if our best guesses don't actually affect any real policy decisions then you can delegate these model estimates to consulting firms. But, I've always hoped that economists are useful people whose ex-ante predictions do affect policy maker behavior by improving their information rather than simply "rubber stamping" what the politicians wanted to do.