Ron Evans/RTP/USEPA/US

10/11/2007 10:05 AM

	

To

Stefan.Osborne@mail.doc.gov

cc

Darryl Weatherhead/RTP/USEPA/US@EPA, David Misenheimer/RTP/USEPA/US@EPA,
"Johansson, Robert" <Robert_C._Johansson@omb.eop.gov>, Tom
Walton/RTP/USEPA/US@EPA

Subject

Re: FW: RIA NAAQS Workshop









Stefan, these questions are very helpful in setting up the presentation
for next week.  I appreciate you taking the time to put this together.

I will leave the technical questions and considerations for the meeting
next week.  However, I did want to take a few lines here to put the
EMPAX analysis in the RIA into the context of the NAAQS setting process.

First, it is important to recall that the setting of the NAAQS is health
based.  The Clean Air Act specifically says that cost is not to be taken
into account in the standard setting process.   We estimate the costs
and benefits of a changed air quality standard in compliance with E.O.
12866.

We did not include the EMPAX results for the known control strategy in
the ozone proposal RIA because of the concern that we could not properly
estimate the social costs for the extrapolated portion as well.   In my
discussion with Rob during the interagency calls on the proposal, we
jointly determined it was better to present a balanced treatment
(engineering cost for known controls plus engineering cost for the
extrapolated portion).  We agreed that this was an issue which needed to
be discussed/come to a better resolution for the final RIA.  This issue
was not raised as a significant concern for the PM NAAQS RIA, that is
why you see EMPAX results there.

I am committed to getting an appropriate characterization of the costs,
including the social costs, of the hypothetical control strategies we
present in the RIA.  I think the characterization of the extrapolated
costs is the most significant analytical issue for this RIA which we
need to discuss with the Interagency review committee.  

I am looking forward to meeting you in person next week.

Ron Evans

Leader, Air Benefit & Cost Group

HEID/OAQPS/OAR/EPA

Mail Drop C-439-02

919-541-5488

919-541-0839 fax

Stefan.Osborne@mail.doc.gov 

10/10/2007 11:05 AM

	

To

Ron Evans/RTP/USEPA/US@EPA

cc

David Misenheimer/RTP/USEPA/US@EPA, "Johansson, Robert"
<Robert_C._Johansson@omb.eop.gov>, Tom Walton/RTP/USEPA/US@EPA, Darryl
Weatherhead/RTP/USEPA/US@EPA

Subject

Re: FW: RIA NAAQS Workshop









Thanks Ron.

If I understand correctly the EMPAX analysis has been takne out of the
current version of the RIA.  The fact that we are discussing it means
that we intend to put it into (or are considering putting it into) the
final RIA, yes?  If so, would the EMPAX analysis require its own
separate comment period since it was removed from the NPRM?  For these
comments I'm going to assume the EIA will be in the RIA somewhere.  If
not they are still relevant, I think, for when the EMPAX model should be
used.  Although I can't imagine a more obvious need for CGE modeling
than in this instance.

Given that we are going to discuss why EPA does not include extrapolated
costs in the economic impact analysis, let me first opine on including
the EIA in the RIA and then I'll give my thoughts on extrapolated costs:


Including EMPAX analysis in the RIA:  Given the interest from industry
in the NAAQS rules it is important to maintain the quality of analysis
from rule to rule and, where possible, to improve it.  The PM 2.5 RIA
had EMPAX analysis in it.  Removing it from the ozone RIA would be a
large step backwards in the economic rigor of the RIA, and I am opposed
to removing it.  

Extrapolated costs:  I think removing the extrapolated costs from the
EMPAX analysis invalidates the whole point of doing it, if you are
trying to calculate the social costs of the rule and compare them to the
engineering costs.  If the costs are focused on capital intensive
industries there will be an adverse effect on investment, which will
cause the social costs to be somewhat higher than the engineering costs.
 If it is technically impossible to include the extrapolated costs in
the EMPAX model, I am fairly confident we could do it in the USAGE
model.  If EPA is for whatever reason unwilling to include the
extrapolated costs we should discuss it.

Assigning extrapolated costs to industries:  I understand there may be
some concerns that the extrapolated costs cannot accurately be assigned
to the appropriate industry.  If that is the case there may be a number
of solutions to that.  First of all, the costs are extrapolated from
specific control strategies, so the costs can be assigned to the
industries where those controls are implemented.  That may, of course,
make the extrapolated costs hit hardest in industries that already bear
a large burden of the costs.  Also it may make some of the control
strategies economically infeasible (although from my perspective this is
why the exercise would be interesting).  A more neutral assumption would
be to assign the extrapolated costs evenly across the economy, according
the the relative size of the industry.  In that instance the social
costs and the engineering costs would likely be about the same, not very
interesting, or likely.  Or, if we know the controls will all be
associated with capital, we can assign the costs neutrally across
capital -- that will still have interesting macroeconomic consequences. 
That would make the impact on services industry fairly small, for
example.

Engineering/social costs ratio:  Characterizing the social costs
according to the EMPAX model as "small" is misleading -- if you have
done the modeling correctly, the size of the shock to GDP will be of the
same order of magnitude as the engineering costs input into the model. 
If the input is 7 billion dollars, a small fraction of GDP, then the GDP
shock will also be small.  That is missing the point, however.  Compared
to other microeconomic policy shocks, 7 billion dollars is quite large. 
The World Bank estimates the benefit to the United States from
liberalizing manufacturing tariffs to be about $15 billion in 2015.  The
NAAQS ozone rule alone would undo half of that benefit - and that does
not include extrapolated costs!  At any rate, the important comparison
is the size of the social costs (i.e. changes in GDP or relevant
components thereof) compared to the engineering costs input into the
scenario.  In the PM 2.5 RIA this ratio isn't even cited.

Sensitivity analysis:  I also feel that more can be done with the EMPAX
modeling than was done in the PM 2.5 RIA.  One of the main advantages of
having a CGE model is that you can perform sensitivity analysis, and
there was none in the PM 2.5 RIA.  We can improve the analysis by doing
sensitivity analyses of parameters of interest to industry, like the
Armington elasticities.  The elasticities between domestic production
and imports is lower than elasticities between regions.  How much lower?
Can we summarize the armington elasticity assumptions in some clean way?
 Can we proportionally increase the armington elasticities and show it
does/does not have an effect on the overall results?

Hope that helps, 

Stefan 

Stefan Osborne

Office of Competition and Economic Analysis

International Trade Administration

U.S. Department of Commerce

1401 Constitution Ave., NW

Room 2224

Washington, D.C. 20230

(T) 202.482.1607

(F) 202.482.4614 

Evans.Ron@epamail.epa.gov 

10/04/2007 09:25 AM 	





To

"Johansson, Robert" <Robert_C._Johansson@omb.eop.gov> 

cc

stefan.osborne@mail.doc.gov, walton.tom@epa.gov,
Weatherhead.Darryl@epamail.epa.gov, Misenheimer.David@epamail.epa.gov 

Subject

Re: FW: RIA NAAQS Workshop









Thanks.

Stefan, EMPAX is on the 18th so we are good to go.  If you have any

specific questions about EMPAX which you can send us in advance it will

help us plan our presentation.  Thanks.

                                                                       

            "Johansson,                                                

            Robert"                                                    

            <Robert_C._Johan                                        To 

            sson@omb.eop.gov         Ron Evans/RTP/USEPA/US@EPA        

            >                                                       cc 

                                     <stefan.osborne@mail.doc.gov>     

            10/04/2007 09:17                                   Subject 

            AM                       FW: RIA NAAQS Workshop            

                                                                       

                                                                       

                                                                       

                                                                       

                                                                       

                                                                       

Hi Ron,

I believe commerce is interested in your model and how it might compare

to their model.

Rob

From: Stefan.Osborne@mail.doc.gov [mailto:Stefan.Osborne@mail.doc.gov]

Sent: Monday, October 01, 2007 10:31 AM

To: Johansson, Robert

Subject: RE: RIA NAAQS Workshop

I'm interested in the EMPAX related material.

By the way, the Cellulosic Ethanol Summit is on October 16th and 17th.

If the EMPAX material is on the morning of the 18th that would be great.

Stefan Osborne

Office of Competition and Economic Analysis

International Trade Administration

U.S. Department of Commerce

1401 Constitution Ave., NW

Room 2224

Washington, D.C. 20230

(T) 202.482.1607

(F) 202.482.4614

