COMMITTEE ON WAYS AND MEANS BILL THOMAS, California, Chairman
PHILIP M. CRANE, Illinois E. CLAY SHAW, Jr., Florida NANCY L. JOHNSON, Connecticut AMO HOUGHTON, New York WALLY HERGER, California JIM MCCRERY, Louisiana DAVE CAMP, Michigan JIM RAMSTAD, Minnesota JIM NUSSLE, Iowa SAM JOHNSON, Texas JENNIFER DUNN, Washington MAC COLLINS, Georgia ROB PORTMAN, Ohio PHIL ENGLISH, Pennsylvania WES WATKINS, Oklahoma J. D. HAYWORTH, Arizona JERRY WELLER, Illinois KENNY C. HULSHOF, Missouri SCOTT MCINNIS, Colorado RON LEWIS, Kentucky MARK FOLEY, Florida KEVIN BRADY, Texas PAUL RYAN, Wisconsin	CHARLES B. RANGEL, New York FORTNEY PETE STARK, California ROBERT T. MATSUI, California WILLIAM J. COYNE, Pennsylvania SANDER M. LEVIN, Michigan BENJAMIN L. CARDIN, Maryland JIM MCDERMOTT, Washington GERALD D. KLECZKA, Wisconsin JOHN LEWIS, Georgia RICHARD E. NEAL, Massachusetts MICHAEL R. MCNULTY, New York WILLIAM J. JEFFERSON, Louisiana JOHN S. TANNER, Tennessee XAVIER BECERRA, California KAREN L. THURMAN, Florida LLOYD DOGGETT, Texas EARL POMEROY, North Dakota
Allison Giles, Chief of Staff Janice Mays, Minority Chief Counsel SUBCOMMITTEE ON OVERSIGHT AMO HOUGHTON, New York, Chairman
ROB PORTMAN, Ohio JERRY WELLER, Illinois KENNY C. HULSHOF, Missouri SCOTT MCINNIS, Colorado MARK FOLEY, Florida SAM JOHNSON, Texas JENNIFER DUNN, Washington	WILLIAM J. COYNE, Pennsylvania MICHAEL R. MCNULTY, New York JOHN LEWIS, Georgia KAREN L. THURMAN, Florida EARL POMEROY, North Dakota
Pursuant to clause 2(e)(4) of Rule XI of the Rules of the House, public hearing records of the Committee on Ways and Means are also published in electronic form. The printed hearing record remains the official version. Because electronic submissions are used to prepare both printed and electronic versions of the hearing record, the process of converting between various electronic formats may introduce unintentional errors or omissions. Such occurrences are inherent in the current publication process and should diminish as the process is further refined.

Advisory of April 30, 2002, announcing the hearing

Council of Economic Advisers, Hon. R. Glenn Hubbard, Chairman

Joint Committee on Taxation, Lindy L. Paull

CONSAD Research Corporation, Pittsburgh, PA, Wilbur A. Steger, and Frederick H. Reuter, statement and attachment

MODELING THE ECONOMIC EFFECT OF CHANGES IN TAX POLICY

House of Representatives,
Committee on Ways and Means,
Subcommittee on Oversight,
Washington, DC.

The Subcommittee met, pursuant to notice, at 2:02 p.m., in room 1100 Longworth House Office Building, Hon. Amo Houghton (Chairman of the Subcommittee) presiding.

[The advisory announcing the hearing follows:]

Chairman HOUGHTON. Good afternoon, ladies and gentlemen.

We are delighted to have you here, Ms. Paull, Mr. Hubbard.

I would like to make a few comments, and then my associate, Mr. Coyne, will make his comments.

As we all know, despite the fact that I used to be in the glass business, looking into a crystal ball is not the easiest thing in the world. However, there are trends and there are economic road maps we can use to more scientifically tell us the probabilities. So to have an understanding of legislation, you got to use all sorts of different methods. There is no alternative other than to just fly blind, and, of course, we don't want to do that.

Certainly, it is not realistic to expect that tax revenue estimates will faithfully predict the precise outcome of every tax proposal. That is just not going to happen. It is understood that your estimates represent the best judgment of experienced tax professionals and economists on your staffs, and they are not crystal ball predictions of the future. Nevertheless, the methods that are used to arrive at these estimates fundamentally affect tax policy for all of us. It makes sense to use the best tools available to make the best possible predictions.

So because of the importance of your predictions in forming tax policy, it is also important that the methods behind them are publicly disclosed to the fullest extent possible. I understand you have disclosed a great deal of information about the estimating process, and I want to specifically acknowledge your willingness to work with congressional staff to understand individual estimates where the need arises.

So today we will hear from Ms. Paull -- we are delighted to have you here -- the Chief of Staff of the Joint Committee on Taxation. I don't think there is anyone in the city who understands these issues better than you, Ms. Paull. You may disagree with me, but I feel that very strongly.

In addition, we are honored by the presence of Glenn Hubbard, the Chairman of the President's Council of Economic Advisers (CEA). Mr. Hubbard was the Chief Economist in charge of estimating tax proposals for the U.S. Department of the Treasury during the first Bush Administration and is wonderfully qualified to address this issue.

So, we are delighted that you are here, and I would like now to yield to Mr. Coyne.

[The opening statement of Chairman Houghton follows:]

Mr. COYNE. Well, thank you, Mr. Chairman, and thank you for holding these hearings today.

Revenue estimates are prepared by the Joint Committee for tax bills approved by the Committee on Ways and Means. These revenue estimates measure the anticipated changes in Federal receipts that result from proposed legislative changes in the Internal Revenue Service (IRS) Code. This information is very important to the development of tax policy, as we all know.

It has been suggested that in making revenue estimates the Joint Committee should take into account the projected macroeconomic effects that would result from a particular tax proposal. As we explore this issue today, I think we will find that the dynamic scoring may sound good in theory but that it would be extremely problematic in practice.

I look forward to the testimony of the Chief of Staff of the Joint Committee on Taxation on this issue and also the views of the Administration. I know that we all benefit from the in-depth understanding of the methodology used by the Joint Committee in creating the revenue estimates that the Committee uses in marking up tax bills.

Thank you, Mr. Chairman.

[The opening statement of Mr. Coyne follows:]

Chairman HOUGHTON. Thank you very much, Mr. Coyne. Well, now we will call our panel, Ms. Paull and Mr. Hubbard. Will Mr. Hubbard begin?

STATEMENT OF THE HON. R. GLENN HUBBARD, CHAIRMAN, COUNCIL OF ECONOMIC ADVISERS

Mr. HUBBARD. Sure. Thank you very much, Mr. Chairman.

I will be relatively brief and make a few observations about dynamic scoring in my experience as a Treasury Department official and then also the Administration's view on dynamic scoring.

As you know, the staff of the Joint Committee and of the Treasury Department has for many years looked at behavioral effects of tax policy. That is not in and of itself easy. Having been the chief stargazer on that at the Treasury Department, I can attest to that.

What I would like to do today is spend the time on the issue that you teed up in your remarks, and Mr. Coyne did as well, which is the issue of macroeconomic or aggregate effects, what I think popularly goes by the term dynamic scoring. Here, I think what economists usually have in mind is, can we somehow adjust revenue estimates that incorporate changes in the level of output in the economy and how that filters through the tax system, through the tax basis for earned income, for corporate profits, dividends, and so on?

I would really just like to make five simple observations with you, and let me start with the simplest but perhaps most important.

At a conceptual level, it is very hard to argue against having dynamic scoring. The idea is, in and of itself, correct, that if we are trying to do the best possible job of providing information to you as decision makers, we need to give you all of the information that we can. That would mean evaluating the economic growth effects of tax proposals and trying to suggest to you how we think those tax proposals would affect receipts through economic growth.

Give you an example. If we were to tear up the current Tax Code and move to a broad-based consumption tax, I think most economists would suggest that would improve economic performance. A consensus estimate might be a level increase once and for all in Gross Domestic Product (GDP) of about 4 percent. That would mean that every year we would have more wages, we would have more dividend and interest income and so on, and generate additional revenue. It is difficult to estimate these things, as Mr. Coyne himself teed up, but that doesn't invalidate the basic point.

A second point, dynamic scoring, I think of as simply representing additional information about the tax policy process. In all of your minds as you think about tax policy, you are interested in no small part because you believe it will affect economic activity. You believe it will affect incentives to work, to save, to invest, to start a business, and so on. To be specific, I think it is straightforward to conduct a revenue estimate using existing methods, i.e., what the Joint Committee does now or the Treasury Department would do for us in the Administration, but to supplement those estimates with what I call, for lack of a better term, an impact statement that would give to you our views as economists of what the aggregate consequences are for the economy and for revenue.

So, point one, it is conceptually correct. Point two, we need an impact statement. Now, what is point three? Dynamic scoring simply doesn't make sense for every proposal. I don't know how many proposals the Joint Committee estimates for revenue in a year, maybe 1,000, over 4,000. I am sure the Treasury Department would have similarly large numbers. Many policy changes, while they may be very important, arguably have a trivial impact on the economy. So if you were to think about dynamic scoring as something useful, you would want to restrict it to quite major exercises like, for example, the tax cut the President proposed and you enacted. I think of this as nothing more than a benefit cost test: Do the dynamic scoring to provide you information when you think the benefits of that information exceed the cost.

The fourth simple observation I would like to offer you really just builds on the previous two that I gave you. Because not every proposal you are going to consider actually requires or would benefit from dynamic scoring and because macro-consequences are really a supplement to, not a substitute for, the revenue estimates you get from the Joint Committee. I don't believe there is a need to imbed dynamic scoring in the budget process itself.

Now let me be specific there. If we had pay-go rules, which we don't at the moment, but if we did I would suggest you don't want dynamic scoring for pay-go per se. However, what you wanted is information for you in deciding which proposals to accept.

The final observation I would offer you is obvious, although many things economists say fall into that category perhaps. Dynamic scoring is actually quite hard. When you think about the methodologies that economists use, there are many different approaches to modeling tax policy, and while each of these models may have some strong attributes, they may have to give different answers. I would think that it is quite important for you to give a flavor for that and consider a range of estimates.

To give you an example, suppose we were to imagine a proposal that would eliminate the double taxation of dividend incomes and suppose that we were to decide that this proposal were to be enacted this year but not become effective for two more years. We would want a model that somehow took into effect the announcement effect of that policy on asset prices, on saving and investment, and GDP. Quite frankly, many models used for commercial purposes don't do that. I think it is very important to acknowledge that different modeling strategies would give you different answers, but I still see great value for you in having a range of estimates that inform your decisions.

So just revisiting these five simple observations, I think dynamic scoring is something you should take very seriously as a concept. I think you have to realize that it can't and should not be done all the time. I think you need to take into account the fact that there are different models that might give slightly different answers. What you really need is an impact statement that helps you make the decisions that you do.  I would refer you to work, that you are already I am sure well aware of, the Joint Committee's exercise in the mid-1990s that led to a 1997 Joint Committee pamphlet, and the current work being done by the Joint Committee and by this Committee on modeling.

Thank you very much, Mr. Chairman.

[The prepared statement of Mr. Hubbard follows:]

Chairman HOUGHTON. Thank you, Mr. Hubbard. Ms. Paull?

STATEMENT OF LINDY L. PAULL, CHIEF OF STAFF, JOINT COMMITTEE ON TAXATION

Ms. PAULL. Thank you, Chairman Houghton, Mr. Coyne, Members of the Committee. Thank you for inviting us to visit with you today on the subject of dynamic scoring.

I have fairly lengthy written testimony that I would like to summarize for you and have the entire testimony submitted for the record.

This is one of those unusually obscure subjects, kind of an inside-the-Beltway type topic which involves the methodology we use to generate revenue estimates of proposed changes in the tax law. The Joint Committee staff is responsible for making those estimates under the Budget Act, and in recent years we have been estimating over 4,000 proposals a year.

The staff is continuously striving to improve the revenue-estimating process, and in doing that we tend to be guided by three principles: one, to produce a process that is going to provide consistently accurate estimates, a process that is viewed as fair and impartial, and the challenge of our staff is to keep abreast of the latest economic work and improve our methodology based on the consensus view of that economic work.

An area of our work that has produced much confusion is the extent to which our estimates incorporate behavioral effects. We have always incorporated behavioral effects into our microeconomic estimates. That is to say, our estimates, our revenue estimates take into account numerous behavioral effects, and they are not static from that standpoint. For example, if we were to estimate a proposal to increase excise taxes, we would incorporate into that estimate the behavioral effect that sales of the product would diminish. So from that standpoint, our estimates are not static.

It is worth repeating because we often get labeled as having static estimates, but the further step of incorporating macroeconomic effects into our estimates is not something that we have been doing, and that is the kind of effects that Mr. Hubbard was describing. That is, for example, the possible impact of a major tax proposal on the overall economy or on investment and savings and whether or not somebody would work more.

Since 1997, the Joint Committee staff has been working on developing a macroeconomic growth model for this purpose. We have gained access to several other models, macroeconomic models, which we have described more fully in my written testimony.

It is just worthy to note, however, that there are significant uncertainties that remain with respect to our ability to provide the type of information that Mr. Hubbard was describing that we would like to provide to you as a supplement to our revenue estimates, but we still have a number of uncertainties that need to be addressed further in order to be able to provide that information.

Our goal is to be able to comply with the House rule and to be capable of providing a supplemental analysis of what the macroeconomic effects would be for a major -- not a, you know, run-of-the-mill kind of tax law change, one that would have significant change in revenues. At the request of Chairman Thomas, we have invited a wide spectrum of economists with significant macroeconomic estimating and modeling experience to review our work, evaluate our model, and to make recommendations for its use and to make recommendations also on the kind of supplemental information that might be generated by this model. This effort will take the bulk of this year to accomplish.

We are presently planning two meetings, one in June and one in September, and we will ultimately publish the results of the work of this panel of advisers.

In summary, I would like to reiterate a couple of points.

The revenue-estimating process should provide Members with consistently accurate estimates of their proposals. Difficult issues are presented in developing the ability to incorporate macroeconomic effects into revenue estimates, and these difficulties should not be minimized. While the staff remains committed to improving the revenue-estimating process by assessing the potential macroeconomic effects of major tax legislation, these issues must be addressed in a manner that is accepted by expert economists. To do otherwise would undermine the integrity of the revenue-estimating process and could reduce rather than enhance the accuracy of our staff's revenue estimates.

With that, that ends the summary of my testimony.

[The prepared statement of Ms. Paull follows:]

Chairman HOUGHTON. Thank you very much, Ms. Paull. Mr. Coyne?

Mr. COYNE. Thank you, Mr. Chairman.

Ms. Paull, the Congressional Budget Office (CBO) considers the dynamic effect of tax legislation enacted into law when it recalculates its baseline each year, but your testimony indicates that the Joint Committee includes some behavioral responses -- is that your testimony here -- when you provide the Committee with your estimates as well?

Ms. PAULL. Yes, we do. For example, if you were -- I gave the example of increasing the excise tax would reduce consumption and therefore reduce the amount of purchases, sales of those products. Another example would be we would incorporate the changes in investment strategies that individuals might undertake with respect to, let us say, a change in the capital gains tax rate or in expansion of interest of tax-free bonds, things like that.

Mr. COYNE. Why did the Joint Committee on Taxation's analysis use a 10-year period rather than a year-by-year approach in estimating?

Ms. PAULL. Well, we have not -- our estimates today do not include the macroeconomic feedback effects or these dynamic effects. What we are attempting to become capable of but we have not provided this yet -- that is what this panel of advisers is going to help us on -- is to look over the overall 10-year period and provide some supplemental information to our traditional estimates. That information will be discussed with these advisers, and we will try to formulate a set of information which we think would rise to the level of accuracy that we think would be useful for Members to consider along with any proposed changes in the law.

Today, we do our estimates on a year-by-year basis, but the reason why you would want to do the supplemental information over a longer time frame, it is because these macroeconomic effects would take a long time -- there is a lot of disagreement within the economists as to the timing and the magnitude of these effects, but it would take a long time frame to realize some of the effects.

Mr. COYNE. Why did your analysis of the entire bill, the tax bill, the most recent tax bill, why did you evaluate the entire bill rather than on a provision-by-provision basis of the bill?

Ms. PAULL. Well, we did provide our traditional revenue estimates on the entire bill on a provision-by-provision basis over 10 years because, while the House has a 5-year Budget Act, the Senate has a 10-year budget requirement. In addition to that, at the request of -- under the House rule of -- I believe it was the Budget Committee Chairman, but I am not positive of that -- we included a footnote in last year's revenue table that indicated, not quantifying in any way, but that the marginal rate cuts that were in the bill would have some positive long-term economic effects on the economy investment savings.

Mr. COYNE. Could you describe for us a macroeconomic feedback note? Just in your words how do you describe what that is?

Ms. PAULL. Well, as I said, that is something that -- the kind of analytical information that we would want to provide to the Congress would be broken down into sectors of the economy as to -- depending on the type of tax proposal that we are analyzing, it would be broken down into, for example, would the proposal encourage people to work more or less? Would it encourage them to save more or less, businesses invest more or less? What the overall impact -- well, perhaps what the impact on interest rates might be, a number of factors, depending on the proposal, and give you some detailed analytical information about that.

Again, as I said, we would like to utilize this panel of advisers that we were just in the process of setting up a meeting with to help us formulate that kind of analysis.

Mr. COYNE. Would that be more than the concept of the 2001 footnote? Would it go beyond that?

Ms. PAULL. Yes, it would.

Mr. COYNE. You would. Thank you.

Chairman HOUGHTON. Ms. Dunn.

Ms. DUNN. Thank you, Mr. Chairman.

I think this is such an interesting topic. I think we have got to be nuts to be so fascinated by the scoring system, and yet I think if we hear anything discussed behind the scenes it is how frustrating it is to try to get an accurate score on a bill.

I have been working with death tax repeal for years now, and I still don't believe that we do enough to take into consideration the positive aspects of what would happen if we truly had predictability in scoring on the death tax. In fact, recently I have heard a number of groups in our debate on tax permanence on the Floor a couple -- few weeks ago, talk in terms of how much the Treasury Department would lose between 2011 and 2020 if death tax were made permanent. They say $4 trillion.

What is your view of long-term predictability of our economy, and how it would be affected by some of these tax relief proposals? Is there any validity to them, or are we still at the point where we think truly a 5-year estimate is much more realistic than anything beyond that time?

I would ask you both to answer that.

Mr. HUBBARD. I think you have raised a very important question, and the reason that I suggested using an impact statement is to get at exactly that. If you take the President's tax bill that became the tax law, our estimates are that, over the long term, say over the next 20 years, that probably adds to GDP growth about 15 basis points a year. Now that may not sound like much, but over a 10-year period it is very, very large indeed, and it is very important I think for all of you to have that information.

The same, frankly, would be true for the death tax which, as you know very well, is a capital tax. It discourages saving and investment. The kind of information you get in these long-term impact statements would be very useful, I would think.

Ms. PAULL. Well, I think one of the hurdles -- we have done extensive review of the economic research in this area, and one of the kind of challenges for us to be able to provide more information to you is there is probably agreement about a number of kinds of tax proposals, what the effect would be over the long term. The magnitude of those effects, there are wide ranges, and kind of the timing of those effects are -- you know, for example, Mr. Hubbard just mentioned 20 years. We are operating within a 10-year budgetary regime as all we are capable of right now, but some of these effects really go on outside 10, 20, 30 years, and that is what a lot of the economists are able to work on -- you know, predict at this point.

So that is what our challenge is, to find a comfort zone where there is enough economic work that you can find some consensus for a reasonable range here and the magnitude and the timing, and that is what we are trying to work towards.

Ms. DUNN. I think the other end of that is important, too. In the example I cited, the permanent repeal of the death tax, there are impacts that occur early on that are positive because of predictability, phasing out, making permanent the repeal of that tax.

For example, there would not be a need to plan beyond January 1, 2010, if we knew that it were going to be permanent, to purchase expensive life insurance policies or to spend great deals of time with certified public accountants, CPAs, lawyers, and so forth. So where we seem often to take into consideration the negative impact of these proposals, we don't always add in what can happen at the front end.

I simply wanted to say, Ms. Paull, I am very happy to hear you talk about that committee of macroeconomists who I hope are going to be looking at this whole area, and we will be interested in their report after the end of the year.

I would say specifically, Mr. Hubbard, that the impact statement used to supplant the scoring of a bill would be very interesting to me. I am wondering, how you would see this work? Would you do it automatically or would you take a look at whether an impact statement, a behavioral look at a piece of legislation would be worth it, or would you simply be directed by the wish of some Member of Congress or a committee that wanted something scored that way?

Mr. HUBBARD. Well, of course, that decision would be yours. My advice would be that you do it only for large proposals and where you think it would make a difference.

Let me give you a concrete example. Suppose you were trying to decide between a proposal to give everyone rebates every year in perpetuity or to cut marginal tax rates. The cash flow implications of those two proposals on revenue may look very similar, but an impact statement would reveal to you that the growth effects of cutting marginal tax rates are substantially greater, so that might be information you would want to have in making your decision. So, I would do it for big proposals and where you think it can highlight a contrast for you.

Ms. DUNN. Mr. Chairman, I am sorry to push into my overtime, but I guess the point I would make, it is very tough to get scores out within months. I have a letter in, I think, Ms. Paull, to your organization, Joint Tax, of last February requesting a score that we still haven't gotten. So I am sure that it is seasonal and it has to do with whether the models are prepared or not, but I worry that we are asking for something that is simply going to increase the length of time for us to get the score. What do you think about that, Ms. Paull?

Ms. PAULL. Well, many of the kind of proposals that we score from the standpoint of our normal estimating process have a very narrow scope to them, and they are really not the kind of proposals -- they are very narrowly focused, they are small -- that would have any significant macroeconomic feedback effects. You really need to focus the macroeconomic feedback scoring or analysis on major proposals, and so I don't think it should hold up.

I regret that you have a proposal in that hasn't been estimated for that length of time, but it should not hold up our normal estimates in any way.

Ms. DUNN. Thank you, Mr. Chairman.

Chairman HOUGHTON. Thanks, Ms. Dunn. Mrs. Thurman?

Mrs. THURMAN. Thank you, Mr. Chairman and thank the two witnesses for being here today.

Ms. Paull, I was looking at your testimony, and it is my understanding that I guess it is House rule 13 that actually allows us to include a dynamic estimate of the changes in the Federal revenue after the enactment of a piece of legislation. Has that been used before?

Ms. PAULL. On occasion, we have had requests. I would say my experience is, I think, it has only been once since I have been in this position. It is --

Mrs. THURMAN. Do you know when that was?

Ms. PAULL. For last year.

Mrs. THURMAN. Okay. For the tax --

Ms. PAULL. The way the rule is written, it is a request of the Majority Leader who then is to request the Chairman of the Joint Committee, I think, because the House Member -- it is Chairman Thomas -- to ask for that estimate. It is supplemental information, and, as I said, we have not provided any sort of detailed, quantified kind of estimate pursuant to that rule. We have been doing all this work to become capable pursuant to that rule, and we have provided the kind of thing that was reflected in the footnotes to the major tax bill last year on occasion to Members who have asked for an estimate of a significant type of a tax proposal.

Mrs. THURMAN. I appreciate that, because I noticed in your conclusion in your testimony that there really are a lot of factors that it sounds to me that have not really been settled as to why these macroeconomic issues may be a problem in some cases. Because you mentioned what still needs to be resolved in any of this is predicting the effects of business cycle, fluctuations, predicting the potential response of the Federal Reserve Board, predicting how the macroeconomic effects of tax policy changes would influence different types of individual and corporate income, sensitivity of results to behavioral assumptions, and then the consistency of treatment between revenue and spending proposals.

So, there are a lot of outstanding issues in using this particular way of looking at a tax proposal, and so I appreciate the fact that you have brought that to our attention that there are still some problems in it. It sounds like to me that there are a lot of people still having a rather lengthy discussion about these issues.

Ms. PAULL. Yes, this is an area that is fraught with some problems and uncertainties, and that is why I think even Mr. Hubbard is not suggesting that these estimates -- this kind of estimate be incorporated in our year-by-year estimates but be provided in a supplemental manner. When you provide it in a supplemental manner -- that is what we are going to explore with this panel of advisers -- it is possible that the range of views could be quite broad and so --

Mrs. THURMAN. Right. I think Mr. Hubbard --

Ms. PAULL. That would be supplemental information for you, but it would be difficult to, you know, have a pinpoint saying this is what we think the answer is.

Mrs. THURMAN. I think what Mr. Hubbard said was it is of great value, but you need the range of estimates so that we, as policymakers, would have the opportunity to have that discussion.

Mr. HUBBARD. I think that is right.

I would like to caution you, though, that because something is difficult doesn't mean you shouldn't do it. Surely the microeconomic estimates are subject to the same uncertainty and disagreement. So while there is a range, I wouldn't want to leave you with the impression that dynamic scoring is, in and of itself, harder to do than the traditional.

Mrs. THURMAN. I am not suggesting that. What I am suggesting, though, that what I do believe you are saying, and you can tell me yes or no, is that it is better for us to have a range of these estimates, to have what is currently being done and others and not just for the politics of making it sound better. Is there really good policy and understanding of what action we are taking and the consequences of that action? I mean, it would seem to me that for us, being kind of the people supposed to look at the taxpayers, that it is important for us to have a wide range of these analyses in front of us, versus just one. Is that --

Mr. HUBBARD. I would certainly agree with that, absolutely.

Mrs. THURMAN. Okay. The second thing I would ask very quickly, and Ms. Paull knows this because I brought it up in the last markup, an issue on some bond issues. I am just curious as to why we don't use this on some spending bills, because certainly there would be economists that would argue that you stimulate the economy through providing jobs and that jobs may be through building roads, infrastructure, water and sewerage, those kinds of things. In your consideration are you looking at this as far as spending bills as well?

Ms. PAULL. Well, that would be a question for the CBO that does the scoring for the spending. So, you know, I think that is a legitimate question. If you are going to do one part of the budget, why not do the other part of the budget? Again, I think that you would have to be looking at something that is of major significance, as opposed to just every single kind of smaller spending proposal. Otherwise, there would be such a de minimis impact on the economy, but that would be something I would ask the CBO.

Mrs. THURMAN. Mr. Hubbard, do you have a comment on that?

Mr. HUBBARD. I would certainly agree with what Lindy just said, but add something as well.

I think in the tax area there is much more agreement on effects of major tax policies. There has to be a range, to be sure. Economists can never really agree on any one thing. If you were to say, what is the effect of moving to a consumption tax, that is a question that is easier to answer than what is the effect of, say, a large increase in early childhood education. So I would caution you, it is a much more difficult exercise, although I certainly agree with you in principle.

Chairman HOUGHTON. Mr. Portman.

Mr. PORTMAN. Thank you, Mr. Chairman.

Mr. Hubbard, I appreciate your being here, and, Lindy, thank you for all the help you give us. I have another meeting I am late for. I just wanted to express quickly a couple thoughts and then yield to Mr. Crane.

First is, we should also be talking about revenue raisers, because I think the macroeconomic effects, sometimes we don't consider, and historically, when you look back, some of our revenue raisers have not raised revenue that they were intended to.

The second point is, I love the idea of the impact statements. I appreciate what the CEA provided for us in relationship to the tax relief but also the most recent stimulus package, but for those of us who are legislators, we need more than that.

A good example would be tax simplification. Many of us believe the alternative minimum tax, AMT, should be repealed, for instance, but it is difficult for us to move with simplification because of the enormous costs. So having an impact statement is wonderful. It gives us something to talk about on the Floor. Actually to get it through the process and to deal with the budgetary impact, I think we do need to figure out a way to get the macroeconomic impact incorporated into the estimate.

With that, I yield to Mr. Crane.

Mr. CRANE. Thank you very much. I appreciate the opportunity to be here, even though I am not a Member of the Subcommittee. I have been very fascinated with this entire subject, and I have had communications, as Lindy knows, with her office with a view to trying to effect some potential reforms.

I remember vividly when we had witnesses testifying back in 1978 on the impact of cutting the capital gains rate at that time and all of the experts came in and told us what a revenue loser it would be, and it turned out to be a big revenue raiser. Then I remember in 1981 when we cut the capital gains rate yet again, and this time even Reagan represented those anguished when the question was put: what might the revenue impact be? Well, it might be a loser, but still we think it is a good thing. It was an even bigger revenue raiser than the cut in 1978 was.

This is the sort of thing that, from our standpoint here in trying to contemplate what might be moves in the right direction and most beneficial to our economy, we all on the Committee on Ways and Means are very much involved in. That is why we really need the input from you experts, and I don't want to try to dictate guidelines. You are the ones that have better insights certainly than I do on this subject.

I was wondering if there is anything, Lindy, that you are working on in the way of new macroeconomic models with a view to giving us estimates in the future.

Ms. PAULL. Mr. Crane, I only summarized my testimony -- the written testimony. We do describe in our written testimony the model we have been working on for macroeconomic estimation. This model is one that we hired an outside consultant to work carefully with us, and we have devoted two to three of our Ph.D. economists to work on this model since 1997 when we held the symposium. So we are in the process now of -- we have recently invited a number of outside experts on macroeconomic modeling to come in and review our work, the work that we have been doing in-house, and to help us make any improvements that need to be made to this and also help us design the kind of output, the kind of information that we could provide to the Committee as a supplement to our normal revenue estimates.

We also have access through subscriptions to three other models that we would use to supplement the information, but this model is designed to work with our big other macro -- micro-Internal Revenue Servicemodels so that it would provide a really -- I guess it is really well-suited towards the kinds of needs we would have in the future for being able to do some of this kind of work.

Mr. CRANE. Well, I hope that you might let Members currently review some of the methodology using the score bills, but to tell you and our distinguished guest here today, Mr. Hubbard, that -- what resources you folks need in your efforts please communicate with us, because what you do is critically important to the job we have to do.

I want to thank the Chairman for letting me come here not as a Member on the Subcommittee and participate in this. Thank you, and I yield back the balance of my time.

Ms. PAULL. Thank you, Mr. Crane. I know we have had continuous discussions about this over the years.

Chairman HOUGHTON. Thanks very much, Phil. Mr. Johnson?

Mr. JOHNSON. Thank you, Mr. Chairman. Hi, Lindy.

You know, the IRS decided to issue regulations regarding employee stock purchase plans and that would overturn 30 years of tax policy by requiring payroll taxes be paid on the plans. If the IRS simply decides to reverse itself and return to previous policy, there is no revenue impact. If the court subsequently overturned the rates, there is no revenue impact. Yet, if the Congress is forced to legislate to overturn the new IRS position, the score is costing $23 billion.

When the House passed the Retirement Security Act, there was no real objection to the House spending $23 billion because no one seems to believe the fairy-tale revenue estimating that accompanies it from you guys. Remember, these are regulations that have never gone into effect, yet we are being told it will cost $23 billion to overturn them. I wonder if you could explain that, either one of you or both.

Ms. PAULL. Well, Mr. Johnson, as I think I indicated at the markup, these are unusual situations, but whenever we are faced with them, I don't think anybody likes the result.  

Mr. JOHNSON. That isn't the question. Why are you doing it that way, if nobody else thinks it costs anything?

Ms. PAULL. Well, the reason we are doing it that way is that the starting point for our estimates, our revenue estimates, is whatever present law is. While the IRS may not have enforced the law or maybe they could interpret it a different way, they now have assessed the law to be that an employer should be collecting the payroll taxes because there is no specific exemption from the payroll tax for this item. Even in the absence of their regulations, because we did face this issue in the Senate about a year ago, we would have scored this the same way.

In the absence of a pronouncement by the IRS saying there is no tax, we will never collect any tax on the stock options. Basically, the way the law works is, unless there is a specific exemption from the payroll tax, employers are supposed to pay the payroll tax. That is the state of present law today.

Mr. JOHNSON. I understand, but you don't consider whether the revenue is coming in or not in your estimates.

Ms. PAULL. Well, the revenues are not coming in right now because the IRS said they won't collect any revenue until beginning next year. There was a somewhat de minimis but a mixed bag of revenue that was being collected before the IRS made that announcement. So there has been some tax -- I would have to say that it was erratic -- but some tax that has been collected on stock options in the past.

Mr. JOHNSON. Do you have anything to add, Mr. Hubbard?

Mr. HUBBARD. No, I will leave it to Lindy to --

Mr. JOHNSON. Well, she is smarter, I think, on that issue.

Let me ask you another one. There are times when the tax consequences of current law preclude companies from engaging in certain business transactions. These same transactions, however, can be performed without any tax consequence by nonprofit organizations in government entities. By not providing a level playing field, I think in the long run the government loses revenue because only nonprofits are doing the work, rather than tax-paying businesses, and I will give you an example.

A good example is a bill regarding Contributions in Aid of Construction, or CIAC, when privatizing to the U.S. Department of Defense (DoD) water treatment facilities which is House bill 2130. Right now, the DoD can donate existing water treatment facilities to municipal governments and other nonprofits that renovate or replace the equipment and then run the new facility for the DoD. That is privatizing. However, if DoD tries to privatize a water treatment plant using a for-profit business, in other words, if they go out and offer this property to everybody, the IRS says the contribution of the rundown facility is a taxable transaction.

For-profit businesses believe they can outperform traditional water treatment facilities in the long term, even with the burden of paying income taxes which they don't object to. The problem is, they can't start the process with a huge tax burden being taxable of the gift. If this initial layer of taxes is not lifted, I don't think businesses can compete. Would you like to comment on that, and tell me what the tax consequences are for us for private enterprise?

Ms. PAULL. Yes. Mr. Johnson, this is a consequence of, actually, the law. I don't think it is an IRS ruling.

When you make a contribution, it was really designed towards developers. When you make a contribution of the infrastructure to a private facility, the private facility is to include in income the amount of the infrastructure. In this case a water treatment facility, they would depreciate it, and then get depreciation deductions over the life of the property as they were earning the income. That is a --

Mr. JOHNSON. Well, most of these are --

Ms. PAULL. Consequence of the tax law.

Mr. JOHNSON. Most of these are 20 or 40 years old, so they have been depreciated out probably already.

Ms. PAULL. Well, but they were owned, you are saying, by the DoD, so there --

Mr. JOHNSON. Yes.

Ms. PAULL. They wasted away, is what you are saying.

Mr. JOHNSON. Yes.

Ms. PAULL. So then if they are not valued very high, then there wouldn't be much different income derived from --

Mr. JOHNSON. Are you saying it is okay to tax for-profit but not --

Ms. PAULL. No. I am saying that is a feature of the tax law.

Mr. JOHNSON. Can we change that law, and what is --

Ms. PAULL. So you need to change the law --

Mr. JOHNSON. What is it going to cost us to do it?

Ms. PAULL. I don't --

Mr. JOHNSON. Would there be an impact --

Ms. PAULL. Remember that estimate off the top of my head, but I believe we had a --

Mr. JOHNSON. Is there an impact?

Ms. PAULL. Yes, there is.

Mr. JOHNSON. So, you are going to charge us to change the law, even though nobody is making any money off of it right now. It is another case like we just discussed. There is no money coming in from that right now --

Ms. PAULL. The --

Mr. JOHNSON. Because we give it to somebody, you won't --

Ms. PAULL. The facilities have some value, I assume. The facilities have some value, or there would be no tax consequence.

Mr. JOHNSON. Well, Mr. Chairman, I hope we will look at that. You understand what I am saying. Two instances here where we are costing us to make the law change when there is no cost at all, and it is scored wrong. Maybe we need to write some rules on how to score, I don't know.

Thank you very much. I appreciate your time.

Chairman HOUGHTON. Thanks, Mr. Johnson. Mr. Pomeroy?

Mr. POMEROY. Thank you very much, Mr. Chairman.

As you know, there is no one in Congress that I hold in higher regard than you, and so I am glad you are having this hearing.

That said, the hearing -- the subject matter of the hearing and the conclusion announced by our Joint Tax Committee expert does cause me some concern. One year ago, we passed a tax plan, and now we are hearing that there is going to be a vote on debt-limitation increase within 13 months of enactment of that tax plan. At the time, no one was talking about any need for debt increases. The forecasts  -- you were in this room, Ms. Paull, you heard them as well as I did -- were quite different than what has happened.

To me, this discussion is like getting lost using a road map, and so you want to get a different road map. You are still lost, and I think we are heading the wrong way, and I don't care how you figure it. We are still heading the wrong way fiscally, and I think that the truth is in the numbers.

You indicate in your testimony there have been some improvements made since 1995 in our ability to do this, but I do want to cite two individuals that testified at a January 10, 1995, hearing. I hold them both in high regard and want us to reflect briefly on their testimony at that time.

Ken Kies, former Joint Tax Committee Chief of Staff, testified problems on dynamic scoring including macroeconomics effects of revenue proposals but not spending proposals would cause a serious inconsistency. Most revenue proposals would have little or no macroeconomic consequence. Complexity and lack of consensus on macroeconomic effects would undermine credibility of estimating process and, finally, macroeconomic analysis would reduce pressure to address the Federal deficit problem.

Now, even though we may have gone some ways in terms of modeling proficiencies since 1995, are those issues still with us today, Ms. Paull?

Ms. PAULL. Sure. Those issues are still present. They are still described in our testimony.

Mr. POMEROY. They are. I note six that you outline. I will talk about them in a minute.

Alan Greenspan is quoted from the same hearing as saying, and I quote: "Unfortunately, the analytical tools required to achieve dynamic scoring are deficient. In fact, the goal ultimately may be unreachable. Accordingly, we should be especially cautious about adopting technical scoring procedures that may be susceptible to overly optimistic assessments of the budgetary consequences of fiscal actions. In summary, the current relatively straightforward scoring system has served us well in many respects."

It almost undeniably produces a more conservative number than dynamic scoring, and, therefore, if we are to try to plan for downside scenarios, that result may not be altogether a bad one from a policy standpoint.

Ms. Paull, do you think so?

Ms. PAULL. Well, I think what was suggested in my testimony was to provide supplemental information that supplements our estimates. Our estimates are still based on the baseline that is provided to us by the CBO, and there is uncertainty in that baseline, especially nobody can predict the downturns in business cycles.

Mr. POMEROY. In fact, I note -- I guess I said six. I think it is five. You indicate predicting the effects of business cycle fluctuation, predicting the potential response of the Federal Reserve Board to fiscal policy changes, predicting how the macroeconomic effects of tax policy changes would influence different types of individual and corporate income and international capital flows, sensitivity of results to behavioral assumption and model structure, and, five, consistency of treatment between revenue and spending proposals.

You indicate all of those are very real problems, still, in terms of building a dynamic scoring model, all of which I agree with, and then you come up with a conclusion I don't agree with. So let us build a dynamic scoring model.

You have seen this Committee grab whatever, a number, you will be so well aware, because the number is the creation of you and your staff. It is basically a theoretical abstraction based upon modeling that may or may not have accurate assumptions built into it, but once the dollar is out there, it is as though we have a dollar in hand, an absolutely concrete figure. We commit based upon that figure. That is certainly what happened in the tax vote last year.

Do you really think having one number and, yet, an additional number -- one number based on already difficult modeling to project and another number based upon even more speculative, subjectively based dynamic scoring -- is going to help our process?

Ms. PAULL. Well, to the extent -- I think you came in after I testified. These are issues that we are going to be exploring with a broad spectrum of macroeconomic experts, modeling experts, and we will see if you can come up with a consensus to try to deal with each of these difficult issues.

These difficult issues are not going away. That is the work we are going to be doing this year, and we would invite you to participate, if you wish, in that. You know, this is an issue that I think, to the extent we can find useful information that is -- that meets a standard of accuracy, that we would like to provide it to the Congress as a supplement to the estimates that we already provide. That is what we will be working on this year with this panel of advisers, to see what kind of information could be developed that would produce a range of accuracy based on the economic literature and based on the work of, really, experts in the field.

Mr. POMEROY. Thank you, Ms. Paull.

Mr. Chairman, thank you. I believe it is a start down a very slippery slope indeed. I will be happy to be involved in -- not in coming up with a new modeling formula, I guess that is a little beyond my abilities, but I will want to keep an eye on what you are doing.

Ms. PAULL. Sure.

Chairman HOUGHTON. Thanks, Mr. Pomeroy. Mr. Ryan?

Mr. RYAN. Thank you, Mr. Chairman. I, too, am not on the Subcommittee, so I want to thank you for allowing me to participate in this conversation. I just flew in from Wisconsin.

So, Ms. Paull, I missed your opening testimony, and you and I have talked about this quite a bit in the past, and to me it all goes down to whether or not we have the truth. When we legislate here, hopefully we are legislating in pursuit of the truth. If we don't know what the consequences, the real-life consequences, are of the policy we pass through this Committee and into law, we end up exacerbating misinformed public policy.

So the idea of dynamic scoring is really a well -- well-regarded, well-known, well-honed skill. It is no longer voodoo scoring, voodoo economics. It actually is more appropriately called reality-based scoring. It seems to me that the importance is in pursuit of the truth and in pursuit of making sure that when we measure not just tax relief or tax cuts, but maybe revenue-generating things that don't end up generating as much revenue because they adversely affect personal behavior, it seems that we want to find the truth.

We want to make sure that we, as accurately as, possible measure all of the policies coming out of this Committee. Because there is so much academic literature available on this, there are so many in the field of economics who have worked on these issues, who work on this, who run very well -- sophisticated models that are continually updated, that at the very least, it would be important to have transparency in our scoring system so that we can engage in a dialogue. We can continue to improve, we can continue to make progress on pursuing the truth with respect to our economic measuring sticks.

So, by involving and incorporating those who are also participating in this process, those who in the private sector -- whether it be the Wharton people or whoever gets involved in this process -- at the very least transparency, I think, is important.

Lindy, you and I have talked about this, and last summer I raised a number of questions about capital gains tax cuts, capital gains policy, the scoring of it, and about the change in stock prices that result from tax changes and how the change in stock prices affects the change in economic growth, and the change in GDP growth ends up changing revenues.

The fact that we hadn't been incorporating that process -- a very, very well-documented process -- has led to very, very off scores, very, very misleading scores which impact public policy. You asked me to put to you in a memo -- I am not trying to play "gotcha" here, Lindy, but you asked me to give you my concerns in a memo, to which you would respond.

I sent the memo last August; we followed up a number of times. We haven't heard from Joint Tax yet on our concerns on some of those fundamental scoring issues. I understand you talked about your Blue Ribbon panel. Maybe that is where these issues are going to be discussed.

I have two questions. One, I don't know if you want me to go through all of the concerns I raised in that memo on capital gains scoring. I would like to get your -- I would just like to get a commitment that you will respond to that.

Ms. PAULL. Yes. You have my commitment.

Mr. RYAN. Okay. Soon, please.

Ms. PAULL. Okay.

Mr. RYAN. It has been a while.

Also, what do you hope to get out of this Blue Ribbon panel? Is this Blue Ribbon panel -- is it going to be like the 1997 JCT Symposium where we got a lot of good testimony, where we heard about the new macroeconomic modeling techniques, where we saw how much more accurate they were; or is it going to be a panel that actually produces results?

Is there a time line? Is there going to be more transparency coming from this? Will there be specific recommendations that will be implemented from this, or what?

Ms. PAULL. Those are great questions. Let me tell you -- let me describe to you what we have in mind and also invite you to participate, if you wish.

We will be holding two sessions, at the moment, with these advisers, one in June and one in September. The first session will involve a detailed description of the model that we -- the framework that we purchased, and then all of the modeling work we did basically in-house since then, so that there is a discussion about -- kind of critiquing the work that we have done. At that same session we planned to have done a simulation and will present the simulation to the advisers as food for thought also.

Mr. RYAN. From your current model?

Ms. PAULL. Yes. Then we would be requesting some input on the kind of information that would be useful to be provided as a supplement in a lot of detail as opposed to, yes, this would have a positive or negative effect on various factors, but in a lot more detail.

Then we would be doing a second simulation during the summer -- many of these are academics where it is difficult to get them to meet in the summer, so early September is when we would be following up -- and we would be, at the first meeting, asking for input on what additional simulations they would like to see us do. We would be presenting that information at the follow-up session so that -- and, again, having a full discussion and critique about the results or the ranges of results and the different, alternative kinds of assumptions you might be making. Then we would be publishing the results of that. We would then be making improvements.

I think that the staff that has been working on this project, all but one participated in 1996-1997, were benefited greatly from the discussion of the modeling, having experts in the room discussing modeling issues, and bringing to their attention the various types of research and concerns that we would be raising with them. So, it is a nuts-and-bolts kind of a session, not just reviewing other people's work.

Mr. RYAN. It seems like it is a very constructive thing to do, but would you agree that it would be even more constructive to do it on an ongoing basis, meaning --

Ms. PAULL. Sure.

Mr. RYAN. Releasing the data, releasing your R-squares, releasing all of the supporting data you have when you release a score, and releasing all of the supporting data within the model on an ongoing basis so that the public -- so that those who are engaged in econometrics can send you suggestions, can evaluate the modeling.

Maybe then you will be able to mark and put your model up against other models to see from now on whether or not they are closer in reaching the truth than other models. If not, maybe you can learn some from the models that have been more accurate.

If you go back historically, I think you will find some other models more accurately predicted revenue effects from different changes in tax policy. So, I think that it would be helpful to have more transparency at Joint Tax, and at the Treasury Department, of course, and maybe we can learn something from these other models that have historically been shown to be a little bit more on the mark on revenue changes than the models we use here at Office of Tax Analysis, OTA, and at Joint Tax.

Ms. PAULL. Well, Mr. Ryan, we have provided detailed descriptions of our models periodically, and we haven't updated that recently, and we plan to. I am not sure if you want something different from -- I mean, those are quite detailed --

Mr. RYAN. Something that is accurate.

Ms. PAULL. Models. The problem here is, of course, we are available to discuss with anybody the assumptions that underlie our estimates, but we have a small staff.

Mr. RYAN. I know.

Ms. PAULL. We produce over 4,000 estimates a year in recent years.

Mr. RYAN. I know you do. You guys --

Ms. PAULL. For us it is very labor-intensive work for people who don't care -- all they want is their number -- for us to do that kind of detailed analysis with respect to every estimate that we produce.

Mr. RYAN. I realize that. That is why I have been patient since my August memo to get a response.

Ms. PAULL. Yeah. This estimating -- building this model is really oriented towards large proposals as opposed to a lot of the kind of day-to-day stuff that we do.

Mr. RYAN. That's right, but these large proposals impact the day-to-day numbers, the day-to-day estimates, and they impact whether or not we are getting closer to reality or the truth. We will not. Then we will make better decisions.

That is more of a statement than a question. I see my time has run up.

Lindy, I appreciate what you guys do. You guys work very hard down there. Every time I leave the building late at night, you guys are still there. So you have great staff, and you have hard-working, intelligent people. It is just the concern that every time we conduct tax policy, I fear that there is a bias against incorporating the real effects, and we miss the truth sometimes.

We all can improve, and I just hope that we can have open minds to improve our models and to have transparency, so we can have a good exchange to try and better perfect what we do here. That is all.

So thank you. Appreciate it, Mr. Chairman.

Chairman HOUGHTON. Thank you, Mr. Ryan.

Well, I guess I am the last one on the panel. I would like to ask you a couple of questions.

Mr. Hubbard, when you were a Professor of Economics at Columbia University -- I think it was 1995 we began discussing this dynamic scoring system -- what has changed since those days?

Mr. HUBBARD. Well, I think the topic remains very, very important. There are always new developments in economics, but to go back to Mr. Ryan's question, I think that what has changed is an increasing indication of interest and willingness on the Committee for this sort of information.

Chairman HOUGHTON. Ms. Paull, you were what in those days?

Ms. PAULL. I was working for the Senate Committee on Finance.

Chairman HOUGHTON. Did you get into this issue at all?

Ms. PAULL. Yeah. I think that interest in this issue is shared among certain Members of the Senate Finance Committee, as well, sure.

I met Mr. Hubbard -- the first time I met him was over the capital gains tax back in the early 1990s, when he was serving at the Treasury Department -- or maybe it was the early -- late 1980s, serving at the Treasury Department, and the Senate Finance Committee was considering whether or not the capital gains tax was too high. He went and did a thorough briefing on all of the economic literature in that area, which, you know, there is -- there are selected topics in revenues where there is significant economic research, but it does not cover all of the areas that we have to cover when we are doing revenue estimating. There are some very significant areas, and we keep abreast of that research, and the research is ongoing.

Chairman HOUGHTON. Are you possibly saying that the difficulty with dynamic scoring is that the science, or the number of people required to do the analysis, is not there?

Mr. HUBBARD. I don't think, Mr. Chairman, that dynamic scoring, in and of itself, is harder or subject to greater uncertainty than the estimates that you see. I think, though, that what you wanted is just additional information. I think you have a right and, I would humbly suggest, a responsibility to know the economic-growth consequences of the policies that you consider. So, I think it is something you should well pursue.

Chairman HOUGHTON. One of the things that always intrigued me when I was first on the Committee on Budget is that, as contrasted to business where no business, I think, can exist without the static scoring system because you couldn't get a return on investment, and payback you wouldn't know, but you would spend X amount of money, and yet, you wouldn't really record if that was fruitful money invested.

That was always very difficult for me to understand, because our job as Members of this Committee, or as Members of the Budget Committee, is to see whether there is a proper return for the citizens of this country. One of the things that I am always interested in is trends in impact and effect, making sure that we are on the right track, and I am not sure that we coordinate properly.

For example, I would be very happy, whether it is static or dynamic scoring, to be able to have regular reports in terms of, you know, you are doing an appropriation on the XYZ project. This is what happened last year, this is what happened 5 years ago, this is what the result was. I don't think we mesh our information together.

Do you have any comments to make on that?

Ms. PAULL. Well, I think those are very important oversight functions, and typically what happens as opposed to having a comprehensive oversight plan, we end up picking and choosing which things we are going to look at by sending the GAO out to look at it or whatever.

I mean, our function is principally oriented towards the legislative process and not looking back at what is already achieved. So, your Subcommittee is an oversight Subcommittee and often needs to use other resources to be able to do that, because we are so focused and busy all the time with the legislative process.

I think you have a valid point and not enough attention is paid sometimes to whether or not tax proposals are achieving the purpose for which they were designed.

Chairman HOUGHTON. I am comfortable operating on 1 year budgets, but if you have got to go 20 years, you take a look at a 1-year budget, and you can say there is 90-percent probability. For a 5-year budget, maybe it is 50 percent. For 10 year budgets, maybe it is 20 percent; 20 year budgets is 20 percent.

I don't think we evaluate those things. You give us the figures, and we are not interacting with you, and I think that is a real mistake.

Ms. PAULL. Well, it is hard to go backwards and evaluate, especially in the tax arena, as to -- because the tax law, there are so many things that can affect particular -- sometimes it is a very narrowly crafted provision, and you can look and see what has happened. Often there is interaction with a lot of other things, and it is hard to pinpoint what the effect was of any one particular item.

So, it is difficult for us to go back and look and see if we did a good job estimating a particular item, especially when you are talking about the individual -- overall individual income tax, where a lot of changes have been occurring recently.

Chairman HOUGHTON. I always remember the story of -- I think it was John Gardner -- in 1965, when Medicare went in, and they estimated 25 years out, 1990, that Medicare would cost something like $12 billion for the country. I don't know how many hundreds of billions of dollars -- if we had had dynamic scoring, would we have done a better job?

Ms. PAULL. No.

Chairman HOUGHTON. Why not?

Ms. PAULL. Well, I think that one is one of the legendary -- there are a number of legendary items -- that is, we just -- the estimations. This was not out of our office, so I can't really give you a full flavor of why. This is something that was estimated by the CBO, I believe, or maybe it predates them even.

I would just note to you that on the revenue side, every year, we update our big models, the individual income tax model, the corporate income tax model, and with the latest information and with the latest macroeconomic estimates of the forecast of the CBO. I think every year our modeling does improve, and that is, I think, the most you can expect from this kind of a process that is not precise. The consistencies and the improvements all lead you in a direction that provides -- that is providing you with more accuracy.

I think that is what is happening on the spending side also at the CBO. They are much better at estimating the proposals now than they were then, but brand-new programs, just like in the tax law, brand-new tax credits -- let us say, for something brand new, there is a high level of uncertainty because you have no experience in estimating those kind of things. It is good to put a sunset on them and reevaluate over a period of time.

Chairman HOUGHTON. Mr. Hubbard, are there other countries that do a better job of forecasting or estimating than we do?

Mr. HUBBARD. Well, the truth is, it is hard everywhere. Economists here are just as two-handed as they are anywhere else in the world. It is very hard to forecast receipts generally, not so much because of dynamic scoring issues, macro-issues, as shifts in relationships.

For example, what is the relationship between receipts and taxable income because of executive compensation or other issues, all of these are factors? So, I think it is very difficult around the world.

Chairman HOUGHTON. If you were setting up a scoring system and your projection was for 10 years, wouldn't you tell those people who would appropriate the money --  let's say, 3 years out into the 10 years -- if there had been some major disruption, that there was a disruption and those figures were going to be dramatically impacted?

I don't see the give-and-take.

Mr. HUBBARD. Well, I think, to go back to Lindy's answer, that would be an appropriate oversight function. It is not clear that it is so much a revenue function, as going back and back-casting, if you will. It is difficult, because many other things have also changed in the economy, but I agree that both for spending and for receipts, you would like some sense that the forecasts you have been given at the beginning were true.

Chairman HOUGHTON. Is on track or not. Right.

Do you have any questions? No?

Do you? Any other comments you would like to make?

Well, I thank you very much for your testimony. There being no further business before the Subcommittee, the hearing is adjourned.

[Whereupon, at 3:20 p.m., the hearing was adjourned.]
[A submission for the record follows:]

CONSAD Research Corporation, Pittsburgh, PA, Wilbur A. Steger, and Frederick H. Reuter, statement and attachment