On February 14, 2002, President Bush announced a Clear Skies Initiative that
proposes, among other things, to reduce the existing cap on total SO2 emissions
from approximately 8.9 million tons under the existing provisions of Title IV of the
1990 Clean Air Act Amendments to 4.5 million tons starting in 2010 and to 3.0
million tons starting in 2018. The proposed reductions in the SO2 cap are similar
to that facing the 263 generating units that were mandated to be subject to
Phase I under Title IV and which occasioned a significant amount of early overcontrol
and consequent banking of a llowances for later use. If enacted, there is
every reason to believe that electric utilities would similarly engage in banking
behavior prior to the reductions in the cap. Accordingly, any evaluation of the
costs and economic effects of this proposal must make some assumption about
banking.

The first section of this paper briefly describes banking and summarizes the
grounds for concluding that banking behavior under Title IV has been largely
rational, and therefore nearly optimal. This conclusion is the subject of another
paper now being written by Juan Pablo Montero and myself and the most that
can be done here is to adumbrate the argument. In the following section, the
simple model that closely tracks observed banking behavior under Title IV is
used to simulate the response to the proposed further reductions in the SO2 cap.
The results reported concern marginal and total costs of abatement, emission
levels, allowance prices, and the value of the existing endowment of allowances.
This section is then following by one presenting a sensitivity analysis in which the
three principal uncertainties—the timing and levels of the reduced caps, the
discount rate, and the predicted rate of growth in counterfactual emissions—are
varied; and a final section concludes.