Dear Deena, Rosanne, Gerry, Jon, Bob, Mihir, Rob, Nada, Dan, Jon, Susan,
Brad, Jim, Hilary, David, Jay, Gib, Arturo, Monica, and Todd -
I hope that all is well, and that as the busy fall season
unfolds, you are managing to find time to make headway on your tax
expenditure research. I know that in my own case the approach of our
next pre-conference meeting is an important prod to action!
When we met in July, there was broad agreement that it would be
useful to meet on November 1, the morning before the NBER Public
Economics program meeting. I have been able to secure space at NBER for
that meeting, and I hope that even though I have not been in touch to
confirm this plan, you will still be able to attend. Everyone involved
in the pre-conference is welcome to stay for the Public Economics
meeting, which begins with lunch at noon on Thursday and concludes by
3:00 on Friday afternoon November 2. A copy of the Public Economics
program schedule is pasted to the bottom of this message. We will plan
to start the pre-conference with a continental breakfast at 7:30; the
meeting will begin at 8.
I am happy to report that Jim Hines and Susan Guthrie have formally
agreed to join our project team, with a research paper that will explore
the impact of tax expenditures on government outlays. This is a novel
aspect of tax expenditure research and I am delighted that they are
moving ahead on this project.
I would like to use our November 1 meeting to review progress on
individual papers and to discuss some broader issues that cut across all
of the studies. I have been keeping a list of such issues: let me
summarize what I have, and encourage everyone to suggest other issues
between now and November 1 so that we can discuss them as a group.
1. Calendar year / fiscal year. Although budget projections and
revenue analyses are usually done on a fiscal year basis, working with
calendar years that are coincident with reporting periods on income tax
returns and also in most household surveys seems appropriate for our
study. This unfortunately means that our estimates may not be directly
comparable to numbers that are presented in budget documents.
2. Base Year and Baseline Assumptions. Because we want the results of
the project to be useful to policymakers, it would seem that we should
focus our attention on estimates for 2006 or 2007, and also include some
projections for a year like 2011. Dan reports that the NBER state tax
calculators are updated through 2006. The question is how to move
forward to 2007 -- and what to assume about the AMT as we do that. I am
hoping that by the time we get into the early part of the new year,
there will be some clarity about what the 2007 tax code looks like with
regard to AMT. We will need to make assumptions in any case to do 2007
calculations -- we'll need to "age" variables from tax returns for
earlier years to obtain 2007 values. I hope that Dan, David, Jay, and
others will be able to talk about assumptions for such "aging" when we
meet, and that we will be able to agree on a set of baseline aging
assumptions that can be used to transform data from tax files, the2004
SCF, and other data sets in to a 2007 baseline. If there is a strong
sentiment that the 2007 baseline is a lot more difficult than 2006, we
could probably use that instead.
3. Projections. We should try to make estimates for 2011, both with
and without the extension of the tax cuts of the early 2000s. For
something like the state and local tax deduction the treatment of AMT
under the two scenarios will make a large difference. But one thing I
hope our studies will be able to shed light on is how the different
paths of tax policy in the years ahead may affect tax expenditure
analysis.
4. Categorizing Households. We will need to agree on both the variable
or variables that we will use to divide households into categories, and
on the way we want to make distribution tables. This should be done in
the same way across studies, even when the underlying data are different
(ie SCF vs. SOI vs. CPS) so we should make sure that everyone has a
template for constructing an estimate of the ranking variable from their
data set. I am inclined to think AGI should be our target variable, in
part because it is complicated to construct something like CBO's "full
income" from some of the data sets we are likely to work with. Perhaps
everyone could explore how hard it would be to construct an AGI estimate
from whatever data you are using as your baseline?
With regard to categories, I think we would want to break out
households with very low or negative income, and also get some picture
at the very top. Thus I would suggest income categories like:
AGI < 0
0-25
25-50
50-75
75-100
100-250
250-500
500-1M
1M+
For some data sets (CPS?) it may be very difficult to fill in the upper
strata of this table, and that is something we should be ready to
discuss on November 1.
I hope that you will be able to attend the November 1 meeting --
NBER will of course cover travel expenses and hotel costs. You will be
receiving an email with logistics from either Carl Beck or Brett
Maranjian at NBER early next week. If you will NOT be able to join the
meeting on November 1, please let me know right away so that I can plan
our schedule accordingly. I look forward to seeing you on November 1.
Jim
PS. If you are interested in studying the chapter on tax expenditures
in the president's latest budget, it can be found at
http://www.whitehouse.gov/omb/budget/fy2008/pdf/apers/receipts.pdf
PPS Here is the program for the public economics meeting that starts on
the afternoon of November 1:
THURSDAY, NOVEMBER 1:
12:00 PM Lunch
1:00 PM
B. DOUGLAS BERNHEIM, Stanford University and NBER
Beyond Revealed Preference: Choice Theoretic Foundations for
Behavioral Welfare Economics (Joint with Antonio Rangel)
Discussant: DANIEL SILVERMAN, University of Michigan and NBER
2:45 PM
EDUARDO M. ENGEL, Yale University and NBER
RONALD FISCHER, CEA
The Basic Public Finance of Public-Private Partnerships (Joint with
Ronald Fischer and Alexander Galetovic)
Discussant: RAJ CHETTY, UC, Berkeley and NBER
4:15 PM
ALAN J. AUERBACH, UC, Berkeley and NBER
Welfare and Generational Equity in Sustainable Unfunded Pension Systems
(Joint with Ronald Lee)
Discussant: PETER A. DIAMOND, MIT and NBER
6:15 PM
Reception/Dinner
Harvard Faculty Club
20 Quincy Street
Cambridge, MA
Dinner Speaker: EDWARD P. LAZEAR, Council of Economic Advisors
FRIDAY, NOVEMBER 2:
8:45 AM
WILLIAM G. GALE, Brookings Institution
Effects of Individual Development Accounts on Asset Purchase and Saving
Behavior:
Evidence from a Controlled Experiment
(Joint with Emil Apostolov, Gary V. Engelhardt, Michael Eriksen,Gregory
Mills, Rhiannon Patterson)
Discussant: KATHERINE BAICKER, Harvard University and NBER
10:00 AM
LOUIS KAPLOW, Harvard University and NBER
Optimal Policy with Heterogeneous Preferences
Discussant: IVAN WERNING, MIT and NBER
11:30 AM
HANMING FANG, Duke University and NBER
Dynamic Inefficiencies in Employment-Based Health Insurance
Systems:Theory and Evidence (Joint with Alessandro Gavazza)
Discussant: AMY FINKELSTEIN, MIT and NBER
12:45 PM
Lunch
1:30 PM
ANDREI SHLEIFER, Harvard University and NBER
The Effect of Corporate Taxation on Investment and Entrepreneurship
Discussant: JOEL SLEMROD, University of Michigan and NBER
Received on Mon Nov 05 2007 - 19:11:44 EST
