Re: [Pen-l] Computable General Equilibrium Models and Climate Change

[Gernot Koehler <gko15@xxxxxxxxxxx>]

Hans Ehrbar asked (26mar09): [quote] Can one also argue that CGE models are being misused ? Is it possible to use them in the right way? [end quote]

COMMENT:
I have an interest in forecasting and simulation but no specific expertise in CGE models (Computable General Equilibrium Models). My answer to the two questions raised by Hans Ehrbar tends to be a double affirmative: CGE models can be misused and they can be used in the right way. That is parallel to the old thing with statistics - yes, statistics can be misused and statistics can be used the right way. Therefore, CGE models (like statistics) can (and should) always be subjected to ideology critique.

I read the example of a CGE model mentioned by Hans, namely:

http://www.beaconhill.org/BHIStudies/WCI-2009/WCIReportFinal090323.pdf

That report of a CGE modeling exercise concludes that jobs will be lost by the year 2020 due to cap and trade. The report has very much a flavour of propaganda for the energy and transportation industries.

One possible line of attack is a general one - namely, critiquing CGE models in general, for example, questioning the forecasting capability of such models (as does Dorman in the blog mentioned by Ehrbar). That can be done, just like one can criticize statistics in general; but that does not bring much in a specific practical case. In this case, you want to win a public debate about cap and trade. In the context of such a debate it would be very effective to (a) expose all the biases and limitations of the presented model and its results, and (b)develop a counter-model. (Just like in another situation where you reject somebody's statistics, you would expose the flaws in the given statistics and present counter statistics.)

Here are some critical comments on the above-mentioned Beaconhill study:

(a) The report does not reveal all the assumptions used in the modeling exercise. For a full critique, one requires a more comprehensive report.

(b) The report treats cap and trade as a taxation problem and nothing but, i.e., the effect of cap and trade on the energy and transportation industries is seen as an additional tax on those industries. That tax will destroy (X) number of jobs in those industries. The government collects (Y) number of dollars as additional revenue and can hire (Z) number of additional public servants for that money. The number (X) of private sector jobs destroyed will be greater than the number (Z) of public sector jobs created (they say). What are their assumptions about public sector jobs created? Did they consider that the government could use those additional revenues for supporting alternative green industries? Do they simply plug the average wage of all public servants into their model in order to estimate the number of public sector jobs expected?

(c) Other detailed critical questions can be raised about this particular study, including questions regarding their assumptions about petroleum prices and their development; technological response within businesses to a cap and trade regime; and others.

(d) A counter-model would be nice to have. Change some of their assumptions, plug in some other variables of interest, insert some other logical connections (algorithms, formulae, etc.)- and voila - you _ prove _ the opposite of their conclusion. From a scientific point of view, such an exercise may not be of much interest, but it would a valuable tool in public debate.

Some thirty years ago, Herrera et al. did a magnificent job of developing a complex leftist model in the context of the limits-to-growth debate (the so-called Bariloche model).

Gernot

---

Make your Messenger window look the way you want. Express Yourself!

_______________________________________________
pen-l mailing list
pen-l@xxxxxxxxxxxxxxxxxx
https://lists.csuchico.edu/mailman/listinfo/pen-l