Tax Analysts®Tax Analysts®

My Subscriptions:

Events

TAX ANALYSTS

THE 2006 TAX ANALYSTS CONFERENCE SERIES

TAX INCENTIVES —
DO THEY WORK, AND ARE THEY WORTH

THE COST?


Washington, D.C.

Friday, April 7, 2006



PARTICIPANTS:

KEN KIES
Former Chief of Staff, Joint Committee on
Taxation, and Managing Director, Federal Policy
Group of Clark Consulting

JANE GRAVELLE
Senior Specialist in Economic Policy,
Congressional Research Service

MIKE BROSTEK
Director Strategic Issues, Government
Accountability Office

TIMOTHY MCCORMALLY
Executive Director, Tax Executives Institute

CHRISTOPHER BERGIN
President, Tax Analysts

OTHER SPEAKERS:
VAN DOORN OOOMS
SHELDON COHEN
DON ALEXANDER
JOE MINARIK
GENE STEUERLE
BILL NISKANEN
NORMAN KURLAND
JAMES S. HALPERN
PAMELA OLSON
DIANE LIM ROGERS
JAMES WHITE
MARTY LOBEL
MARTY SULLIVAN
HOWARD GLECKMAN
ERIC TODER
CHERYL BLOCK
MONICA McGUIRE

* * * * *

PROCEEDINGS

MR. BERGIN: Good morning. I'm Chris Bergin, the president of Tax Analysts, which is the non-profit publisher of Tax Notes magazine, Tax Notes Today, State Tax Notes, Tax Notes International, and many other print and online tax publications.

Welcome to the first in a series of roundtable discussions we have scheduled for the 2006 season on key issues in tax policy. The topic for today is tax incentives. I'll get back to that in a couple of minutes.

As many of you know, we conducted a series of roundtable discussions last year, and as I look around, I'm glad to see many familiar faces. But for those of you who are new to our process, let me explain how it works.

I will open things up with some brief remarks to introduce our topic. I will then introduce our distinguished panel of four speakers. Each of them will address aspects of our topic for about 10 minutes.

After that, we will open up the discussion, and you are all encouraged to participate, whether you are sitting at the not-so- round table or elsewhere in the room. Just wave and I will find you.

We are recording this event and will post the transcript to our Web site, as we do for all our discussions.

Also, for media purposes, we are on the record. So, when you are recognized, please tell us who you are. Even if I know you, and I know a lot of people here, I am still going to ask your name, so, don't be insulted.

Also, please speak into a microphone. For those of you in the audience, we have hand-held mikes, and we will get them quickly to you.

I will moderate the discussion, and we will end at 11:00.

One more procedural point. Now that we are starting our second year of roundtable discussions, we want to make sure they are as good as they can be. You can help us. We left a short survey at each place at the table and on each chair in the audience. We have extra copies at the table outside if you need one.

If you have a moment after we finish this morning, please fill out the survey. We very much want to know what you thought of this session, and we would welcome any suggestions on how to improve our roundtables.

As I said, the topic for discussion today is tax incentives. More specifically, we ask the following questions: do they work, and if so, are they worth the cost?

Perhaps by the end of this discussion, we may conclude that the term "incentive" is as fuzzy and non-descript as the term reform when placed anywhere near the word tax. But we'll at least start with a definition.

According the Webster's New World College Dictionary, an incentive is something that "stimulates one to take action, work harder, etc."

I suppose that could be a gun to the head. But when I think of incentives, I think of good things like lowering taxes or pepperoni to make my pizza more enticing.

Still, I think we can all agree that too much of a good thing, even pepperoni pizza, is eventually no good for you. Just the incentives in the code dedicated to saving are unduly complex. It makes me yearn for the world described by Marty Sullivan in this week's Tax Notes in which some powerful, almost unmoved, almost divine force, like Alan Greenspan, oversees our tax system to ensure that it is solely — used solely for raising revenue and nothing else.

Marty calls his dream social science fiction. And for those of you who haven't seen his article or several other good articles we have this week on tax incentives, we have magazines at the table outside. Please grab one on your way out.

I agree with Marty that his utopia is fictional because based in part on an unscientific study I've been conducting the last couple of weeks, nobody that I can find thinks we can run the current tax system, or anything like it, without some incentives.

One need look no further than the president's panel on tax reform. For me, a key aspect of the president's executive order creating that panel was that it recognized the reality, if not the necessity, of incentives in our tax system. Indeed, it specifically recognized the sanctity of at least two. The deduction for charitable contributions and the deduction for mortgage interest on a house. The president gets no argument from me here. I've met people who wouldn't give a dime to charity unless they thought they were getting something back. And that something is a tax deduction.

As for homeownership, you merely need to watch that famous Jimmy Stewart movie "It's a Wonderful Life." Who among us would disagree with George Bailey when he tells Mr. Potter that owning a house makes us all better citizens? But the good comes with the bad.

Tax incentives make the code enormously complicated and they are unfair because they favor some people over others and some corporations over others. And like pepperoni pizza, the lobbyist can't get enough of them. You know things are getting really bad when the chairman of the Ways and Means Committee scolds the corporate tax community for abandoning sweeping tax reform and opting instead for the instant gratification of tax avoidance measures, such as the research credit. The research credit.

The chairman's complaint raises the larger question, at what point can we no longer afford all the good things we're trying to do through the tax code? At what point do incentives overload the code, making it too complex for too many people?

To answer these and other questions about tax incentives, we have assembled a group of talented and provocative speakers today. Let me introduce them to you in the order they will speak.

Ken Kies is a former chief of staff of the Joint Committee on Taxation. And he is now the managing director of the Federal Policy Group at Clark Consulting.

Jane Gravelle is a senior specialist in economic policy at the Congressional Research Service.

Mike Brostek is director of Strategic Issues at the Government Accountability Office.

And Timothy McCormally is the executive director of the Tax Executive Institute.

So, I will now get out of the way and Ken, I'd like to start with you.

MR. KIES: Okay. Thank you, Chris. I'd like to start with just a little test of the audience in the room. How many people here if you could vote would vote to not have any tax incentives in the code, just raise your hand?

How many of you would vote to have incentives? Okay, one last question —

MR. ALEXANDER: How do you define "incentives," Ken?

MR. KIES: Well, it doesn't — it doesn't matter because it's not my point. There's my point. There's —

MR. ALEXANDER: That wasn't your point, either.

MR. KIES: No, here's my point, Don.

MR. ALEXANDER: Yeah.

MR. KIES: How many here are members of Congress? That's —

MR. ALEXANDER: Well, keep trying.

MR. KIES: That's what I thought. I'm not like Gene Steuerle, who has deeply pondered this, as he says in his article, but I wanted to start with a basic thought and then go through sort of a number of sub-thoughts, and that is we can have the greatest wisdom in the world here about this issue, but the reality is that the decision- makers about this are going to be at the other end of this avenue, and the best that we can do is hopefully impact their thinking at the margin, but we have to face the reality that we operate in a system where like it or not, there will always be tax incentives in the code, and part of our challenge should be how do we make them the least intrusive and most efficient, and that's a very difficult balance to strike.

We also don't live in isolation. The R&D credit, which Chris talked upon, touched on as an interesting point. China is advertising for companies, U.S. companies, to bring their R&D to China. Canada, right across the border, is doing the same thing. And there are consequences beyond revenue effect to having for example, U.S. R&D done outside the United States. And it's not just impact on jobs, it's a lot of other things beyond that where our intellectual property rights are going to be held. So, we have to realize we don't operate in an isolated environment, we operate in a world that is quite global. I don't want to overdo that, but the truth is things are much more mobile today, not just where capital can be invested, but where people can be placed and where activities can occur.

Incentives in the tax code, and I disagree with one of the articles — I think it was by Calvin Johnson in this week's edition. He stated as though it were a fact that a government spending program would be more efficient than a tax incentive as a means of delivering some sort of government subsidy. I think there's a strong argument that can be made that government spending programs are very inefficient, they involve — they ask bureaucracies that are dramatically greater than the size of the Internal Revenue Service, which administers the entire tax system for I think around $7 billion a year, if I recall correctly, assuming — is that number around right?

MR. ALEXANDER: Ten.

MR. KIES: Ten?

MR. ALEXANDER: Ten, eleven.

MR. KIES: Okay. It's up since Don was there. Don probably got by on about half a billion or so.

MR. ALEXANDER: But we didn't collect much money.

SPEAKER: You get what you pay for.

MR. KIES: But — but the notion that a good substitute to a tax incentive is a government spending program, I think is one that one should think very carefully about, and frankly, I think is very challengeable.

Tax incentives, this goes to part of your question, Don, what's a tax incentive? I think in some cases, Congress puts tax incentives in the code not necessarily to incent people to do things, but to reward them for things that they would do. I don't necessarily share Chris's view that nobody makes a charitable contribution other than because they'd have a tax deduction. Maybe I'm in a more charitable crowd than you are, Chris — no, I'm just kidding.

MR. BERGIN: I might just be in the wrong crowd.

MR. KIES: Yeah, right. But —

SPEAKER: Non-itemizing is due.

MR. KIES: Yeah, well, I think the one thing we have to realize is in some cases is Congress puts provisions in there because they actually want to reward behavior, not necessarily create it, and you can decide that's a good thing or a bad thing, but the point is, every tax incentive is not necessarily created because of the expectation that it will cause the behavior to occur. In the case of charitable giving, I think it causes behavior to occur at the margin, but I think it also rewards behavior, and I think that's exactly what Congress wants to do.

If you're going to have tax incentives in a corporate system, I think the greatest challenge and one that Congress has wrestled with both successfully and unsuccessfully is how do you create a system in which you do not create a significant competitive imbalance because of the fact that tax incentives that are not in the form of refundable credits or can otherwise be monetized create a system in which profitable companies are significant beneficiaries of those incentives, and nonprofitable or marginally profitable companies get very little out of them.

For those of us that are old enough to remember safe harbor leasing, which was included in the '81 act, it was — they put there with a very explicit purpose in mind, and that is the Treasury Department, as they were crafting very generous depreciation provisions that in combination with incentive — or the investment tax credit, was going to create a system in which highly-profitable companies would get very substantial tax benefits and marginally profitable companies would see very little, and the concern was about what impact that would have on the competitive balance — particularly in the circumstances where those two companies were in the same industry. And safe harbor leasing was actually a very efficient response to it, because unlike traditional leasing, which leaves with the investor or owner a lot of the tax benefit, safe harbor leasing was designed so that it would deliver to the user of the equipment a high percentage of the benefits, and actually a system was set up that required participants in those transactions to report to the Treasury Department to demonstrate what was the amount of the incentive that was going to the user. And those reports, which are pretty interesting to go back and read, demonstrated, and I'm remembering this roughly, that the amount of incentive going to the user was in the 90 percent range, which was substantially greater than what you saw in a traditional leverage lease.

Safe harbor leasing met its demise when it was reported that companies were zeroing out their tax liability. In other words, they were actually using it the way Congress intended them to use it. And had it actually been structured from the very beginning so that, for example, you couldn't have reduced your tax liability by more than 50 percent, we might still have safe harbor leasing in the code today. But it did make a very important point, and that is when you have corporate incentives, you have to think carefully if you're not going to make them refundable or allow them to be monetized about what you're doing to competitive balance.

I want to thank Jane in her article for praising the '97 provision on the $500,000 exclusion for capital gain on housing. Actually, the way we came to do that in '97 was we realized very few people were actually paying the capital gains tax on a house. And it wasn't because they didn't have capital gain; it was because we concluded they couldn't compute it. And nobody actually knew how to keep track of the basis in their house, and they were keeping shoebox-fulls of receipts, but they really didn't know what to with them, so, it's an example of — it's an incentive, I guess, in the code, but it was put there for a practical reason, which was compliance in that area was abysmal and the burden that we were putting on people was kind of absurd.

And, as Jane points out in her article, it was causing seniors, for example, to stay in houses that were bigger than they needed because they felt locked in to avoiding the payment of a capital gains tax on what would otherwise have been a wise economic decision, which would have been to sell the house and buy a smaller one.

So, it's actually a good example of an incentive, if you want to call it that, that addresses a very practical situation and makes complete sense.

Another thing to keep in mind is the focus on how do you make incentives work, and interestingly, there seems to be some debate about this. For example, I'll just use as an example, the section 29 credit, which has gotten a lot of criticism by, for example, Lloyd Doggett, a member of the Ways and Means Committee, mercifully on the other side of the aisle, but he has pointed to the section 29 credit and said, see, look, this is a terrible credit because they've admitted that they wouldn't do this but for the credit. And I just stop and think, well, isn't that the definition of a very effective incentive. Why would we want to create an incentive and give it to people who are going to do it anyway? And that's the real challenge of tax incentives, and that is, how do you design them so you actually create a maximum or significant amount of behavioral response in the direction that the incentive was supposed to incent, while not spending money on people that are otherwise going to do the same thing?

Another example of an incentive that's had an interest effect is the enactment of HSAs, Health Savings Accounts. It's an interesting experiment. If in the absence of HSA legislation, HSAs wouldn't exist, people wouldn't be buying high-deductible policies. We can have a whole debate from a healthcare perspective of whether that's a good thing or a bad thing, does it create adverse selection, but clearly, the enactment of HSA legislation has resulted in a million HSA accounts in the space of a couple of years that would not exist otherwise.

The same thing can be said in a way about the wind credit. The wind credit's an interesting example of no one would have commercially produced electricity but for the wind credit when it was originally enacted. The cost per kilowatt hour today for wind- generated industry — energy is I believe something like 25 percent of what it costs when the wind credit was originally enacted. So, the wind credit actually had the intended effect, which was to get people to go out, use a new form of energy generation, and then improve it scientifically so that we've reached a point where we're probably within reach of the wind energy industry being economically viable without any incentive effect.

The challenge for Congress in the case of the wind credit is when are they going to actually say we've succeeded, and it's now time to let the industry go on its own? But it is an example of a credit carefully structured, actually creating the result that people intended.

I guess the final point, and I'll conclude because I think I'm close to my 10 minutes, is we actually have to be very careful about how we put parameters on incentives, too.

One of the studies that I did at the Joint Committee, actually just as I was leaving in January of '98, was a study of how many taxpayers are subject to a higher marginal rate than the nominal marginal rate because of various phase-outs in the code. A number of the phase-outs relate to phasing out the benefit of incentives, and I was asked to come back and testify about that report two or three weeks after I was gone just because there was nobody left there to still talk about it, and I'll never forget this, I was sitting at the witness table while some other witness was testifying, and I was kind of just reviewing my testimony and we had come up with an example of an individual who was 62 years old, had kids in college, maybe he was 65, was getting Social Security benefits, was subject to the phase- out of the Hope College Credit, was subject to the earnings limit on Social Security, and we actually had computed that the marginal effective tax rate of that person was 105 percent. And I was reading this, and I was thinking to myself, well, yeah, I guess theoretically you could have that situation, you know, some being college kids, and so on and so on, and I thought wait a minute, that could be me. But the point being, it's actually an interesting study to go back and take a look at. It was published in January or February 1998. And it showed that something like 50 to 60 million individual taxpayers have a marginal rate that's higher than the nominal marginal rate because they're subject to one or more phase-outs in the code. And there's a lesson there in terms of the incentives that we think about creating in the code, and that is, do you — if you're going to structure them so that you're only going to make them available to people in certain income classes — you are necessarily going to create these high marginal rate situations and I think there's a lot that we should be concerned about in that kind of system because I think it creates odd behavior. So those are, Chris, my opening comments, and I'll be happy to talk about any of them when we get to the discussion.

MR. BERGIN: Great. Thank you, Ken. If you do your own returns, it also makes your evening of doing your taxes very uncomfortable when you're taking all this stuff away. Jane?

MS. GRAVELLE: Thank you. First, let me say that anything that I say should not be — is not the view of the Congressional Research Service.

I have to admit the first thing that flowed to my mind when I was going to address the question of "do tax incentives work," is I hope not, because the vast majority of them are really not consistent with what, you know, an economist would suggest.

The vast majority of tax incentives are artifacts of very early tax law or regulation. They reflect times when legal concepts dominated when things like constructive receipt were very important, and they remain important today. In some cases, they responded to transitory issues, but once the issue had passed, the provision remained.

An example of that is the predecessor of percentage depletion, called discovery of value depletion, which was enacted in 1918 to help with the war effort.

Another example is the DISC/FSC/ETI, which we are finally being forced to quit giving money away to other countries by other countries' actions. That was first proposed to deal with fixed exchange rates in the late 1960s, and by the time it passed, shortly after, we went to flexible exchange rates, but the provision remained. Some of it is in response to interest, clearly interest group pressures. Maybe even about a transitory issue that — that again, once you get a tax provision in the law, it's extremely, extremely hard to get it out.

Now, since the tax expenditure list or the tax incentives don't provide to me a consistent notion of what we should be doing, I have to look through it through the lens of an economist, so, I'm going to be looking mostly at issues of economic efficiency. There's distributional issues, as well, but mostly economic efficiency.

But let me stop and entertain you for a moment. Sometime back I worked on a book with some other people in the room for the Century Foundation called Bad Breaks All Around. It never got lots of sales, I don't think, but I would highly recommend it. But I'm not getting any royalties, so, I don't have a personal interest in that. But in the process, we made a list, a consensus list of what we call the dirty dozen. These were sort of the worst tax expenditures, not because of size, because they were so inconsistent with economic efficiency issues, or rationales, so, I'll read them to you very quickly.

The first is export incentives. That includes the disappearing of FSC/DISC/ETI, but also the title passage rule, which is still in existence. The excessive percentage over [off microphone], original economic development incentives like empowerment zones, graduated corporate income taxes rights, exclusion interest on private activity bonds, tax credits for non-conventional fuels, I guess Ken will disagree with me on that one, but — medical savings accounts, the exception of credit union income, timber subsidies, special rules for employee stock ownership plans, the small life insurance company deduction, and the exclusion of fringe benefits through cafeteria plans. So, just as a little menu.

So, there's three ways to look at this question of, you know, do tax incentives work in the sense that the economists want them to accomplish a more efficient economy. And the first is what we would call target efficiency. We have chosen some objective that we agree upon, so, the question is, does a tax system work better than alternatives to achieve this objective?

Now, one example where I actually think you can make a case that it's not targeted, efficient is actually charitable giving. And by the way, I think I run with a better crowd than you do too, Chris. I actually think for many, many people the kinds of motives that — that encourage charitable giving are very divorced from the rational economic man. I think it's also true when you vote. I mean, there are many aspects of your behavior where money is not paramount. Because, remember, you're always — unless it's some kind of little scam — you're always giving away more than you get back in taxes.

For some time, there were studies that indicated that the price of elasticity of charitable giving is greater than one. And when it is greater than one, you can actually get more aggregate giving through a tax incentive than through just taking the revenue loss and giving it to charity. But the more recent evidence, Bill Randolph has done a paper on this, indicates that those with strong relationships were because they were studying cross-sectional data and people — high-income people tended to time their charitable deductions when the tax rates were high. So, you get this kind of artificial transitory relationship, and when you try to control it for that, you find actually pretty low price elasticities of charitable giving, which I think would be consistent with my, you know, feelings about what the charitable motive is. In that case, you'd be better off just giving money directly to charity.

Now, the recipients of charity are different from the recipients of government transfers in very dramatic ways.

For one thing, very little charitable giving helps poor people, very little. And there's also a significant fraction of probably 40 percent or so that goes for sacramental services of a religious and less-taxed organizations which the government would not do.

You may look at the distribution as good or bad. But in the aggregate, there's actually a case against your favorite incentive, charitable giving.

And certainly, I think there's even more of a case against the giving of appreciated assets where you get bigger tax incentives. And also, giving when you don't really give the gift right away, like through foundations and supporting organizations and donor [off microphone] funds.

Other examples that might not be target efficient are [off microphone] and bond interest. There's evidence that high-income people get higher rates of return adjusted for risk from taxes and bonds so that the subsidy, which really should be intended to go to the states, is actually diverted off to high-income people. You probably could make some kind of argument about other things that are funneled through third parties because the ultimate beneficiary doesn't have tax liability. It is true somewhat of safe harbor leasing. It may be true of the low-income housing credit. It may be true of these original incentives whose objective, supposedly, is to help low-income workers in poor areas. The original incentives and one of the reasons they're on the bad dirty dozen list is not only are they funneled through third parties, but many of them are capital subsidies, and if I were an economist from Mars, I would say "why are you trying to help labor through a capital subsidy?" You could actually end up making labor worth [off microphone] if you're having a lot of substitute ability between capital and labor.

The second sort of issue is allocated efficiency. I mean, are these provisions affecting allocated efficiency in a good way, and let's say mostly they probably are not.

I wouldn't divide these provisions by the functional categories of the budget; I would divide them by the kinds of activities. So, the subsidies that affect labor supply subsidies or allocations, subsidies that affect human capital investment, subsidies that affect savings and investment, and subsidies that alter consumption choices.

By far, the biggest group of these is probably savings and investments, clearly in terms of physical goods favors housing. And in terms of types of savings, favors retirement accounts.

I can't see any evidence — the obvious evidence to violate the notion that all income from capital should be taxed at the same rate. So, I don't think there's a reason to favor housing. And I'm not sure there's really a reason to favor retirement savings. I mean, if you think people don't have enough good sense to save for their retirement, maybe you should just do it through an expanded Social Security system instead of a tax system that only favors high-income people who will likely take care of their situation anyway. And some of these benefits like IRAs and 401-K plans are not marginal for everyone.

I think one provision that may be justified is a lower rate on capital gains, at least for corporate stock and dividends as a way to deal with the double taxation of capital source income, but I don't think this one's entirely clear, given that capital gains rates also play a role in a lot of tax sheltering activity.

To me, the least justifiable subsidy in the tax system is export incentives. And not only are you possibly distorting the allocation of resources, but you are giving money away to people who don't even need it. You know, like Europe. So, why should you be doing that? I mean, it's just like I'm growing apples and trading them for oranges and I'm going to cut my apple price. I mean, why would you want to do that?

For human capital investment, I think the tuition credits and the deductions also are kind of on weak ground. There is an argument that you need to subsidize human capital investment because you don't have slavery. And so, you can't pledge your indentured servant, you can't pledge your future services to repay loans. But, you know, we've dealt with that with the student loan program. We had huge subsidies of all sorts to public skills. We have an existing transfer program for low-income students. And we had a system set up already in place if we wanted to expand those subsidies and credits. I think the reason we did it was because we can do things through the tax system that we can't do through spending because it makes government look smaller. I think that's probably the major reason.

But one of the problems with doing it through the tax system and not making it refundable is that you miss the very group that you probably would most like to target that might be most influenced by a subsidy to go to school, which is lower-income people.

Most of the labor incentives, I think, are questionable, but small. I think the earned income tax credit clearly compared to transfer programs is more likely to not disturb work effort. There's a question about which I know some people here have strong views about whether you should be running it through the tax system. I think one desirable — Ken mentioned one desirable provision, that happened — that I really think has to do with labor is the capital gains, not taxing capital gains on unoccupied housing. I think we would never want a system where you taxed those gains because it would be just one more barrier to labor mobility, aside from the other sort of unfairness and distortions it creates. And the system we had was just the epitome and they [off microphone] taxed, it raised virtually no revenue and caused nothing but distortion and an administrative burden.

The only unfortunate thing I think about the 1997 change is that the cap was not indexed. And it's becoming smaller and smaller as time goes on. I would just —

SPEAKER: And you live in Washington.

MS. GRAVELLE: And I live in Washington. Well, I got past the cap a long time ago. I'm going to hang on to my house until I die.

SPEAKER: Only if you could move your book.

MS. GRAVELLE: But if it were me, it's so little money and as long as you have a cap, and people are uncertain about it, they're still going to have to keep records. I just hate recordkeeping — I just hate having that, so I'd get rid of the cap if I were dictator of the tax system.

And consumption, you know, one of the consumption items would be charitable giving. I've already said something about that. Housing you can see is a consumption, I've said something about that.

The other big consumption item is really health, and I actually don't think that the subsidies for health care are that bad an idea. I think they have really helped with a lot to consider, perhaps, the most serious problem in health insurance, which is adverse selection.

On the other hand, I think you could say that by having that system, if we had not had that system, our health system would have crashed and burned a lot earlier, and maybe we'd had done something else, national health insurance, and that might have been bad or might have been good.

But in the absence of anything else, I think employer — if I were, again, tax dictator — I would shape those incentives to shape the kind of plans I want. In other words, I wouldn't just let people do anything with them, I'd say, you know, here's a kind of co- payment, deductible, and stuff like that — that I want in my system.

So, I think we also have to remember that there are all sorts of things that we can not easily eliminate because of administrative problems, and that's the last reason to have these.

I think that assigning — doing the correct thing in theory, which is assigning the benefits of employer-based healthcare to individuals is impossible. How are you going to assign it? I mean, is it going to be different for men and women? Is it going to be different because of race? Is it going to be different because of age? How do you determine what particular employees are getting a benefit of health? I think that would be a nightmare. Similarly, I think that'd be a big problem with defined benefit pension plans. What do you do if you assign a benefit of a pension plan and then these companies do what they've been doing, which is, you know, converting them to cash plans and people lose some things. So, there are some things that I think we have to do just because they're too hard to deal with. But, my overall evaluation is — aside from a handful of provisions — that most tax incentives don't "work" because their objectives are not consistent with efficient markets. And so, I think perhaps we should be grateful that at least in my view, most elasticities are also small.

MR. BERGIN: Thank you, Jane. Mike?

MR. BROSTEK: Good morning.

MR. BERGIN: Good morning.

MR. BROSTEK: I'm with the Government Accountability Office. I appreciate you inviting us today to participate.

It's obviously a very big, very important topic. I want to talk a little bit about why I think it's an important topic and talk about a couple of the challenges that are associated with trying to determine if tax expenditures work and whether they're worth the revenue costs and make GAO's pitch that we really think this kind of discussion and the assessment of whether they work is important to do, and it's a contribution and what the Congress needs to have in order to make better decisions about public policy. Any errors I make are mine, of course.

One of the reasons why it's important to have this kind of discussion is just the number of tax expenditures, it's broad in tax incentives, but the number of tax expenditures has increased, and it's more than double now than when they were first started to be tracked in '74 in a systematic way.

The handout that we gave you earlier shows that there's been a rise and most recently in the number that are being created. So, just the number of tax expenditures is going up, and that creates complexity for the taxpayer, obviously, in planning their financial lives, and planning their lives, period, and as well, as in doing their taxes, and it creates complexity and broadens the scope of responsibility of our revenue agency in doing its job. So, it's clearly important from that perspective.

The flip side of the paper that we passed out has this chart with all the lines on it and is another reason to think about tax incentives and how important they are. This chart shows the duration of tax expenditures. And one of the things that you see quickly is that there's a lot of churning. It is true that there are a number that have been on the books a long time. Forty-six, according to our count, have been on the books since at least '74, and many of those go back to when the income tax was created. But, as you can see on these lines that start and stop, there are a lot of tax incentives that are created, last a little while, and go away. So, this gives you a different perspective on complexity.

And there may be 163, or whatever it is, tax incentives in the code now, but there have been a lot that have been created and people had to cope with over time. And that doesn't tell you anything about the changes that are made to the rules of the given tax incentive over time.

So, all of those things to me suggest that there's just a great deal of complexity that tax incentives create both for the citizens of the country, as well as the revenue agency in administering the code.

Another point that I think has probably been alluded to is these have very significant effects on people's lives. The mortgage interest deduction has led a lot of people probably to make the major asset in their lives their homes, and now that we have that kind of provision in the code, making a change to it has significant effects on people's lives, and it makes it very difficult for us to modify the tax expenditures. Or tax incentives.

And the last thing, one other motivation for looking at this topic and trying to get a better understanding of it is simply the long-term fiscal challenges of the country. Our boss, and I'm sure, has said in forums that you've all heard, that this is one of the major things that the country faces. We've got very bleak long-term fiscal outlook, and we're going to have to do something about it. Answering this question about whether tax incentives work and are they worth the revenue costs, we would hope would contribute to sound policy decisions in dealing with those long-term fiscal challenges — hopefully before we're forced to deal with them and while we can take actions that are a little painful for the country.

And one of the articles that appeared in Tax Notes magazine in preparation for this meeting today talked about some of the challenges in trying to understand whether a tax expenditure works, whether it's worth the revenue costs. And I just wanted to talk about a couple of those a little bit.

One is it's challenging to come up with the quantification of how much an incentive actually changed some kind of behavior. So there are lots of challenges in that, but we also have a challenge in deciding whether that behavior is worthwhile, and whether that change is worthwhile.

And I think one of the best examples for me is the R&D credit. GAO did some work; I know others have done work on the extent to which the R&D credit actually creates new original research as opposed to subsidizing research that would have occurred otherwise.

I don't remember the specific figures, in part because I didn't do the work and it's been quite awhile, but what GAO found and others have is there's not necessarily a lot of incremental additional research that's created by that credit. And on that ground, you might wonder, is this worth the revenue cost?

But then there's the important question of, okay, was that additional research important research? And if it created the cure for cancer, we'd probably all say, yeah, there may have been some significant revenue costs, but it was worthwhile. Valuing that research is a major problem in terms of answering the question of "are these worthwhile?"

Another issue that one of our colleagues has mentioned many times is there's not a lot of data to use in evaluating whether these incentives are worthwhile. In many cases, anyhow. Even when Congress is interested in answering that question, it can be difficult for those who are charged with doing so to answer it because there's not data that's collected.

The GAO is mandated by law to take a look at the enterprise communities and enterprise zones to give Congress some sense of are these actually having some effect on the targeted communities? Well, from the tax incentives standpoint, we've been unable to do much at all because IRS collects no data at all on what taxpayers are using the benefits that are available for the communities, they're affected.

We tried to do a survey of what we would figure to be the most likely taxpayers who would use those benefits and got a very low response rate. So, we have basically no understanding at all about whether the tax incentives that are embedded in those programs are accomplishing anything because we don't have any data to use to do that.

Another point that I think has probably been brought up already, but I think is an important one, is we're very interested, obviously, about the tax system. People in this room probably focus a lot on that in their professional lives. But there's also the question of if the federal government is going to do something, if the federal government is going to try to ensure that everyone has adequate health services of some kind?

All right, if the tax provision is working well, or not well, or poorly, how is it working in comparison to any other tool that we have to accomplish the same end? So, if you have a poorly working tax incentive, but it's working better than anything else that we can come up with, then it may be the best choice to continue with the poorly working tax provision, assuming that we really want to address that problem.

So, with that kind context, the GAO, being an organization that's all about trying to provide policymakers facts and analyses to help them make policy decisions, it's been our long standing position that one of the things we do need is more effort to study the effect of the tax incentives to see whether they work, to see whether they're worth their revenue cost. And we do that in a number of areas where it's also very difficult.

For instance, when a regulation is being created or modified in some substantial way in the federal government, OMB circulars require regulatory agencies to do some kind of a cost benefit analysis. They face the same kind of major problems that occur for looking at tax incentives when they do those analyses — they often don't have reliable data, there are a lot of conceptual issues that they have to face, but we think as a government that it's worthwhile to do that, to make the effort, if for no other reason, that it helps us think through what the issues are and put them on the table so that policymakers consider them.

So, GAO has pushed for a long time, and most recently, in a report that we issued last year, to try to get more analyses done of the tax incentives that are in the system. And then for transparency purposes, we pushed to try to get them into documents that policymakers would normally use.

So, if you have a policymaker whose major understanding or a major document for understanding federal interventions is maybe the budget of the U.S., we think it would be good to try to get more information in there about all of the tools that the government is trying to use to accomplish some end, and so, we would like to see more information that would get into the budget that would tell us what are all the things we're trying to do to encourage or to help people with the costs of post-secondary education. And we have the grant programs, we have the loan programs, but there are a lot of tax provisions that are intended to help people with those costs. Those aren't in the federal budget. Would it help policymakers to see them together, and to hopefully get a little information about the relative strengths and weaknesses of those different tools?

We did have a request from the Senate Finance Committee to do that kind of thing ourselves, and we issued a report last year and provided them some information on the grants and loans and tax incentives to help our tax subsidies to help people cope with their post secondary education expenses. And we think that the committee found that useful and hopefully will help make a more informed decision about what, if anything, to do with the education provisions that they have jurisdiction over.

So, those are some opening comments.

MR. BERGIN: Thank you, Mike. Timothy?

MR. MCCORMALLY: Thank you Chris, and good morning to everyone.

For background purposes, the group that I work for, Tax Executives Institute, is a group of in-house tax professionals from the Fortune 2000, Fortune 3000 companies, and I won't have the same disclaimer as the government folks have here in terms that these are my own views. But I'm sure that some of the folks won't necessarily agree with what I say about some things. I'll begin with this, which is where you sit determines where you stand. And I mean that's a very hackney phrase, but I think that we need to keep remembering again, of course, where we sit is not on a clean slate.

We didn't sit with a clean slate when we put in the ETI regime, so that caused some, was it distortions, was it accommodations, or what? And so, we need to think about it. We also don't live in a closed system. TEI members come from throughout the world. We now have companies represented by our membership overall, and although probably most of the members overall, the vast majority work for U.S.-based companies, that is not universal, and therefore, to say that a system that provides an incentive that works in that it affects behavior as Ken pointed out, is necessarily distorted, it depends not only on your definition of distortion, but also whether or not you think that's a good or a bad thing.

With respect to do they work or how do they work, just take for a second — and I think it was Marty Sullivan's piece this week that talked about the location of a plant in one state versus another, or one country or another, is a distortive effect. And somehow, that's bad because you should determine those things based on labor costs.

Well, if labor is a cost, and tax is a cost, why all of a sudden are we differentiating between those two? But more than that, there are reasons why states — and why countries — decide to have incentives that may not fit any pure economic model.

One of my colleagues in our state and local committee analogized incentives at the state level to free agency. And it may well be that depending on your view of the purity of the baseball model to describe everything, then you think free agency is the worst thing in the world or the best thing in the world. But if you think about the NBA for a minute rather than baseball, and you could always go to Italy to play basketball if you didn't like who you were drafted by, you can see that maybe the analogy has at least some relevance. So, the closed system or the non-closed system to me is something we always have to keep in mind.

I think also, we try too hard to make these things work properly. The R&D credit is on a lot of people's lists as as an incentive that does not work as well as it should, although, I believe it helps locate research in the United States. But I believe that one reason it doesn't work as well is because the tax administrator has demonstrated a hostility to it from before its enactment, and that hostility led to rule upon rule to stop abuses. And that made it more complicated, and therefore, that put an incentive on the people who quite candidly in the aggregate are probably always going to be smarter than the people across the table from them, the private sector versus the public sector, and that's with no disrespect to the people who work for government. It's just the amount of resources you can throw at a problem on an isolated per-company or per-industry basis is always going to be more than that in the aggregate that the government can control at trying to solve a problem that has not yet or maybe has not yet manifested itself. So, I think that the issue of are we trying too hard, are we to trying to slice the pie too accurately, and how does that affect how we put the incentive together.

Finally, a note on permanence. Actually, Mike's comments and this chart are very telling in a couple regards. That one, once it's there, it's very hard to get away. And yet, we never tell people it's going to be there for a long time. I happen to think one of the challenges, at least in the economic efficiency of the R&D credit, is the fact that for budget or other reasons, we always say only for a year, only for 18 months, and to think that something that is supposed to encourage a particular type of activity, especially in the research area where the time and planning horizons can be 5, 10, even longer years into the future, it's just silly. So, to the extent that it doesn't make sense, that's one of the reasons. It doesn't make sense.

I mentioned an aspect of one of Marty Sullivan's pieces that I happen to disagree with, at least the comment about location being an issue. One that I found very provocative was the piece he wrote in late January looking at the world, looking at where we're going on tax rates. What does that point to? It points to a low tax rate, therefore, fairly broad-based, few incentive type system, and then the question always is, well, who gets to pick those few incentives? I heard from more of my members about "did you see that piece? That was really great." And I said, well, what did you think about that provision about stripping out the incentives and it was always kind of silence on the other side of the phone, and that just goes back to my very first point about where you sit is where you stand.

I'll make one other point. Ken stole all of the good lines, which is why I'm kind of doing it ad-hoc here. I agree completely with his comment about refundability of credits, whether that's R&D or whether that's something else that when companies can't see as far as the eye can see, ever receiving a benefit from it, it's kind of hard to expect it to work as efficiently as it might.

MR. BERGIN: Thank you, Timothy. Okay, we're going to open this up for discussion now. Again, I'll remind everybody, please state your name. If you have a question, I'll recognize you, if you're in the audience, just wait for a second so we can get a mike to you. And let's begin.

MR. OOOMS: Thank you, Chris. I'm Van Oooms from the Committee for Economic Development.

I wanted to go back to one of Ken's opening points, which it seems to me is really critical to the discussion, and that is the distinction between whether tax incentives are here to change behavior, or are here as rewards. And it wasn't clear to me from Ken's comments whether he took the sort of positive or normative view here, whether they should be here for rewards or whether they simply are here as rewards in many cases.

But I find the notion that we should have a system of tax incentives of the extent and cost that we have now that is essentially a way of rewarding behavior that would take place otherwise to be really extraordinary in terms of other criteria that we think about when we think about how we ought to govern.

And in particular, because certainly if you're the government — if our federal government is going to be in the business of simply handing out rewards to various special interest groups, the one thing that we should require of that would be very high transparency.

And yet, that's of course what the tax code does not provide. What the tax code does is to disguise the transparency and to to sort of bury it, and then we cover it with a lot of rhetoric about incentives, and no body out there in the public is aware of what the actual transfer for benefits and resources is if we're just out there rewarding people. And I would ask everybody just to do a little mental experiment, you know, take the eight score and some tax incentives that we have and think about all of those being converted into a system of government grants and subsidies that were quite transparent in the budget, and ask yourself how many of them would survive more than a few Congresses? Well, maybe quite a few the way Congress works. But I would think a lot less than are actually there. I would have thought that the public would rise up in outrage at the notion that we were simply handing out rewards through the tax system on the kind of arbitrary basis that a lot of these tax incentives are set up to do, so, I mean, my own view is that I commend Jane on her excellent presentation and I agree with most of it. I think the incentive component here — it certainly differs from one provision to another — but they're in general quite small. The rewards are enormous. The rewards are upside-down in many senses of that word that we can discuss. And so, I just come back to the original point - - I just find the notion that we should have a reward system generated through the tax system, you know, to be an extraordinary sort of breach of the way we normally think we ought to operate the government. Thanks.

MR. BERGIN: Howard?

MR. GLECKMAN: Howard Gleckman with Business Week. I can't provide any economic analysis, but I can provide lots of dreaded anecdotes.

In the 20 years I've been at Business Week, we often have corporate executives who come through the office to talk about one thing or another. And they make their pitch for whatever it is, and then I ask them, usually off the record, about provisions like the R&D credit. And I can tell you that in 20 years, I've never had a single corporate executive from the pharmaceutical industry or the high tech industry, or anyplace else tell me that they have done a dime's worth of research that they otherwise wouldn't have done as a result of the R&D credit. They spend lots of time and effort reallocating costs so they can take advantage of the credit, but they don't actually do any more research.

One other anecdote: a couple of weeks ago we had a senior executive of a large multi-national energy company come in, and I asked him if he had drilled a single well that he otherwise wouldn't have drilled as a result of the recent energy bill, and he said no. I asked him if he had any plans to, and he said nope. But, he said, "we're going to take the money."

MR. BERGIN: Gene?

MR. STEUERLE: I have two comments I want to make. One is a lot of the accounts people are making —

MR. BERGIN: Say who you are. We all know, but...

MR. STEUERLE: I'm sorry. Gene Steuerle from the Urban Institute.

MR. BERGIN: Thank you.

MR. STEUERLE: The first one is that a lot of the discussions that have taken place now could be applied or the comments could be applied just as well to the expenditure side of the budget as the tax side. And in fact, one of the dilemmas we have as is more and more of what goes on in taxwriting more and more resembles what goes on on what I'll call the expenditure committees of Congress. And I think for a lot of us, that's been a loss; we used to think of the tax committees as having higher standards in a way — the way they did things, but it's not surprising that the two would sort of merge over time.

And the second comment I want to make, which is sort of one of these, you know, great "on the other hand" comments is, is I think design is really crucial, and I'll give a bunch of examples. You can't come to clear-cut decisions on tax incentives being good or bad or exactly what you're going to do, at least in a world of second best.

So, with respect to the first one, that the tax issues we're raising are not much different than the expenditure issues, let me give as the primary case, this issue of permanence versus temporary examples that Mike gives in his paper. I mean, on the expenditure side of the budget, we're arguing that one of our biggest problems is we have entitlements, the entitlements are permanent, they're permanently growing, they push aside other choices we make as a society. So, they actually prevent us from making democratic decisions. When you go up to people who work in tax [off microphone] and say, well, gee, we really want these things to be permanent because we want simplification of filling out our tax forms, those are really not consistent positions, especially if you're talking about a subsidy that's intended to do exactly the same things. There is a good reason for reviewing the R&D credit. Now, we shouldn't review it the way we're reviewing it now — it shouldn't be every year or year-and-a-half. But if you're going to have it, you should have a review every three or four or five years. You should not make something like that permanent on the tax side, just as you would not make it permanent on the expenditure side, so I think we have to be a little careful about when we make these arguments. We want to make clear that in some ways we really would apply — especially when we're talking about incentives — we want to apply roughly some of their standards.

Now, in some cases we don't, which is the point that Van and Jane make, among other things, and Mike did also. We don't really — well, I would say on average — we don't study tax subsidies as much as we do expenditure subsidies, but I'm sure that everybody in this room could actually figure out exceptions to that, as well. I mean, we don't actually study Medicare very well either, and we spend huge doses of money there. We do know that budget policy is biased towards tax breaks just in the way the politics plays out because they get recorded as a tax cut rather than a spending increase, and so, they get recorded as smaller government and there is a clear bias for the tax subsidy.

But I just want to make clear that one of the dilemmas we're facing as a society is more and more of these tax subsidies are for middle class tax breaks through very similar to — to spending type things, and we really have got to deal with the broader issue of what we do as a government with all our programs. Having to switch a thousand tax programs over to the spending side of the budget may not necessarily be any great gain for the society.

In the sixth set of issues, I don't want to take up too much time about design issues, so I'll put down a summary without going into the detail.

Jane says that, for instance, we can't deal with the health exclusion because it's too complex. In fact, we had a system in place at one time that was perfectly administrable, it was called the Federal Employee Health Benefit, which had a capped benefit and when you bought any insurance above that cap, you paid out of after-tax income. It was totally administrable, and we now have a system where at the margin we're spending — in another five years, we'll spend another $100 billion per year on this health subsidy, and it — at the margin, it's increasing the number of uninsured. Now, I don't care how hard it is to administer, you don't want a subsidy system that increases the number. You spend $100 billion extra to increase the number of uninsured.

With housing, which economists hate ... they always rail against housing. The tax break in housing is, as far as the exclusion of the gross rental value of the house, and that is something we're not going to repute to people, we're not going to tax it for practical reasons. And just going after the interest deduction (the interest deduction is not really the tax break), when you go after the interest deduction, you're coming in the backdoor and you're zapping the people who have mortgages, but you're not zapping the people who own a house outright. So, I can easily switch and pay off my mortgage and I still get the full tax break, and so, you've got to deal with that.

With the R&D credit and tuition-type things, there is this issue. I would not put any of those things in the tax system. But I will say one thing about them being there right now, is they operate a bit more like vouchers where the government's less deciding what R&D it's going to subsidize, less deciding what schools it's going to subsidize, and a lot of what we do on the direct spending side, so, it's not clear just throwing the money over to the direct spending side, where the government says okay, this is the research I'm going to subsidize, at least traditionally, that's how they've done it, is a better system. Now, I would put a voucher system on the direct side if I was going to do it, but it's not clear a voucher-like system in the government side, which is open to some sort of choice of what gets done versus a direct spending program on the other side is better.

With respect to saving incentives, the simple story there is that they're not saving incentives, they're incentives — incentives for deposits, and that's one of the great dilemmas. So, whether you have them or not, they just are badly designed, they are not incentives for saving. We don't measure saving, they're incentives for deposits and any of us can go out and borrow the money and put it in savings accounts. So, these design issues, it seems to me, are all crucial — we need to start to decide do you or do you not like various incentives and what you want to do about them.

MR. BERGIN: I agree with what you said about the research credit, every year, every other year coming around, but is what we do really reviewing it? Do we review it and did anybody here doubt that it won't be reinstated? Or my question more is should we?

MR. STEUERLE: I think the history I see, and there's people in this room that probably know more about it, that I see is these things get renewed and renewed and renewed until you hit certain points of time when things are just really, really tough for government to operate.

I'll give you an example, this isn't a tax example. I use this on the Social Security side where people say well (tape drop) and all of a sudden we got this unexpected expansion of the economy, in most parts of the budget, revenues grew and spending was stable, and we went into surplus. In Social Security we had this wage indexing, so, all of us sitting in this room, including people in their 50s and 60s got all of a sudden another — or late 50s, got this boost to 5 or 10 percent of our Social Security payment, and I think Congress would not have acted then. Every — almost every other year they would have, but when it was tight, they would not have acted. And I think the same thing happens with things like R&D credits — they'll get renewed 5, 10, 15 years, things get really tight because it comes up, and it automatically comes up, that's when they tighten on it, and they did tighten on it when — I mean, I know that TEI had some problems with it, but originally people were taking in their R&D credit for things like advertising. And the fact they had to come back and —

MR. BERGIN: Yeah, that's a good point.

MR. STEUERLE: — and review it actually allowed them to come in and do some worthwhile tightening, maybe not all them worthwhile, but — but it did help.

MR. BERGIN: Thanks, Gene.

MR. STEUERLE: Just to answer your question.

MR. BERGIN: Sure.

MR. STEUERLE: The practical side of this, though, is that the one-year extension of time is exactly the reason there is no serious review because look at the current existing R&D credit — it's based on a base period that's back in the 80s.

MR. BERGIN: Right.

MR. STEUERLE: And the explanation that is repeatedly given when people say maybe we need to revisit all of that is we're only doing it for a year. We'll revisit it when we actually have a serious look. And if you recall, Treasury was assigned to produce a study about how to fix the R&D credit, which never ever appeared. And Thomas used the fact that Treasury was supposed to be doing it as an excuse for not doing anything to the change, so the one year at a time is actually a culprit in not getting a serious review. I mean I think the only time —

MR. BERGIN: I'm not defending the current system.

MR. STEURLE: Well, no, but I'm saying — I'm just saying that's the reason it's not happening. In theory, a sunset — in fact, back in the late 70s, there was this concept that all tax provisions ought to sunset at some point, is not a bad thing if there was going to be a serious effort to examine those provisions when that sunset occurred. It's just practically speaking the way it's been working for the past decade is one year at a time or 18 months at a time and somebody goes well, it's only going for another year, so, you know, let's just let it go.

MR. BERGIN: It won't hurt anything. Let me fulfill some promises here. I'm going to go here, here, and here. Start.

MR. WHITE: Jim White, GAO. I just wanted to follow-up briefly on Van's point about transparency that in addition to the point that he made, that there's another aspect of transparency — it makes the government look smaller.

I think Jane alluded to this, as well. But people in the press measure the size of government by direct spending by the federal budget, and so, tax expenditures don't get picked up there. The recent discussions about the growth and the size of the government and so on, it's all based on direct federal spending, tax expenditures. Tax expenditures aren't the only item that gets ignored. Regulation mandates are ignored, as well, but tax expenditures get ignored in that discussion, there's less transparency about the size of government.

MR. BERGIN: Yeah, I think that's a great point. If I remember correctly, you take just health, housing, and charity, and it's $250 billion predicted for 2007. It's a lot of money. Sheldon? Even though I called your name, please tell us who you are.

MR. COHEN: Thank you. Thank you. Lo and behold, back in World War II, in Korea — I'm — excuse me, I'm Sheldon Cohen, and I teach at George Washington University.

We issued certificates, the War Emergency Group, or whatever it was called, Globalization Board, whatever, issued certificates and that entitled you to fast write-offs. We didn't just give them to everybody. I've said, and I'm sure other commissioners, former commissioners have said before, any number of the iterations of the Ways and Means and Finance Committees, that if you keep moving at this pace, you can abolish the Agriculture Department, the Interior Department, you could just abolish it and administer everything through the IRS.

Now, of course, none of it will work because it will become so complex and it has, that no one can administer it. And that's one of the problems that the administration of the service, which was very good, and is now still reasonably good, but not as good, is sliding off and will slide off because we keep doing this. We keep putting in an agricultural subsidy or an industrial subsidy into the Internal Revenue Code where — which has no expertise. And no skill and say administer it. And we forget that.

I mean, I'm going off the high plane now into the trenches, but in the trenches is where the damn thing is administered and we're talking about a code that — look, I taught a class last night in depreciation, and I showed them the 1954 provision, which I drafted, which was about two pages long, a page and a half long. Ninety pages of code plus commentary since 1986, and that's, you know, just changes we keep making in the depreciation rules. We have to leave them there because if I'm going to look at what last year's rule was, I've got to, you know, it's — it's all there, I mean, it isn't 90 pages of active live statute for the future that's dealing with the last 20 years. Can we do that? I mean, we just can't administer that and we're proving it everyday.

MR. ALEXANDER: Is it my turn?

MR. BERGIN: Yeah, I promise.

MR. ALEXANDER: I'm Don Alexander, and I agree with the point that — the points that Sheldon made. IRS is called on to do too many things in too many ways for too many people that have no relation whatever to the tax system, except for the fact that they look, as Gene Steuerle pointed out, like tax reductions. You get it twofold — if you put something for the benefit of your particular clients into the tax law, as opposed to direct expenditure. Number one, gee, you're cutting taxes. And number two, of course, you're avoiding these people who have presumably rewarded you by doing what the R&D lobby does and that the faux pas will be — one of the demise, if you will, almost every year of the R&D credit is that those who live in this town of lobbying for the credit get what they want, and that is increased revenues for the professional firms, and of course, the members of Congress get what they want, which is political contributions of the next election. Great, wonderful system, but the tax system right now is vastly overburdened by these ornaments.

Gee, I remember World War II, but I'm not going to into that. But I was there. But I think the only one where I was ever successful and I should have lost on earned income credit, which should have been administered by then HEW, and of course, now you got to leave it in the tax law, I reckon that's that.

A congressman named Jimmy Burke from Boston, okay, had the garden tool credit. The garden tool credit gave you up to $25 for a tool that you could use to plant something in a window box in Boston, Massachusetts, provided of course, you were willing to eat what came out of the window box.

That, I was happily able to ridicule with the great help of Wilbur Mills, chairman of the Ways and Means Committee and the garden tool credit passed the House, but died in Congress.

Everything else, just about every other little tax expenditure that has a great relation to what may be good for our economy, what may be good for the Department of Education, what may be good for the Department of Energy is stuck in the Internal Revenue Code because it's easier to get it there, and once there, Gene, it's harder to check to see whether it's doing its job. It is harder.

I was rarely asked, and I'm sure Sheldon was rarely asked to evaluate how something is working, even the earned income credit. IRS is regularly blasted, and properly so, I guess, because it can not administer all these ornaments as effectively as well as fairly as it should. IRS and Treasury did have to issue some pretty tough regulations on what is research because some people, and I'm sure they weren't members of that fine organization that you have been, but some people were cheating like mad on it, and possibly, a few still are.

MR. BERGIN: I knew that word "incentive" wasn't going to survive. So, to continue our discussion of ornaments, let's go to the audience. It's a better word.

MS. ROGERS: Hi, I'm trying to see if I can talk. I am Diane Lim Rogers, and I'm with The Brookings Institution now, but formerly of Ways and Means Committee Democrats, and I wanted to make a few points or tell a few stories, and I think it's all piling on things other people have said.

First, there is a lopsidedness, certainly, in budgeting favoring tax cuts over spending programs, and just in case none of you have noticed this, when the president's budget was unveiled this year, we learned about a new Web site called expectmore.gov. I don't know if you've been there. And, you know, you can go to that Web site and type in keywords for government programs and hit return and get a report card on various spending programs, so, it's basically rating or grading. I think it's color-coded, probably. But what I noticed early on is that tax expenditures aren't on there. You know, so, right away, you're making the public very aware of wasteful spending programs, but have no mention of any tax expenditures.

Second, I was going to comment on the R&D credit because I learned something while I was working at Ways and Means about the R&D credit, and that was that it doesn't — it didn't make sense to me that we have to come back revisit this thing every year to renew one that's obviously in the Ways and Means Committee bribed by partisan support for it. You know, not a contentious issue at all.

And the one night when we were late into a markup, one of the members sat down back with me bored and I asked him, why do we keep doing this every, you know, year or two, why don't we just make it permanent, it's so annoying, it's not in the baseline of — I was explaining all of the economists' reasons for hating it, and he looked at me and said, are you crazy? We can't make it permanent. You know, I wouldn't get any friends visiting me anymore and — so, I mean, I learned right there that there's a big difference between what sort of makes sense from a budget and economics standpoint, and what makes sense from a political standpoint, and I'd never seen it that way.

And a third point I was going to make is the whole issue of doing things through the tax system versus doing things on a spending side.

A few years ago when I was at the Urban Institute, I worked with a welfare reform expert, Alan Wilde, and we wrote a paper for the National Tax Journal that sort of looked at doing welfare reform through the tax code, because it seemed like more and more things were possible or were getting in through the tax code. And, you know, we really realized doing that exercise that there's a difference, you know, sometimes you might want to do it through the tax code because it's generally administratively, you know, cheaper to do it through the tax code, and it's less intrusive in a way because you don't have to, you know, find out all this personal information.

But those are precisely the reasons why in some cases you may not want that because spending programs easier to target, you know, easier to get specific at particular industries, particular types of households where you have to collect personal information. And so, sort of a personalization issue.

You know, sometimes we economists would rather that it not get so targeted because we view that as distortionary. Other times, there may be other reasons other than economic efficiency that we want to target some benefit toward a specific — a very specific narrower beneficiary.

MR. MINARIK: Joe Minarik, the Committee for Economic Development. A few scattered thoughts with maybe one partially unifying theme. I think Jane's presentation really had things exactly right in the sense that if you're looking at existing tax incentives, you really have to look at them one at a time, or at least in small groups, and figure out how do they work and are they justified. And that fact, I think, makes it hard to generalize about tax incentives broadly. But there is maybe one question that is worth asking sometimes, and that is what are you trying to do with your tax incentive in terms of scale, and that thought came to mind in — here in Ken's — I think in one sense very reasonable remarks about safe harbor leasing, which, as I recall, was also referred to at the time as nationwide consolidated filing for the reason that it was intending to do something rational with incentives that were in terms of scale so enormous that they were intended to take an underlying tax system whose purpose was ostensibly to tax income from capital and totally to eliminate that taxation of income from capital. And it was in a sense that scale which brought safe harbor leasing down because it spread these enormous incentives so more widely, but still so deeply that many people found them to be offensive.

I think Ken made a very good point with respect to the wind incentive, which I think brings up another question, which maybe reflects on charity, as well. And that is sometimes we have provisions that we might think of as incentives that to some extent are alternatively viewed as measurement of income provisions.

Some people believe that a charitable contributions deduction is appropriate because money that is given to a public purpose should not be considered an income for purposes of taxation. Maybe in some sense that's a reward, but if you look at it in that sense, you might justify the charitable contributions deduction on that ground, and then you get into conflict with some of the things that others have suggested about making it a marginal incentive over, you know, some base amount of contributions as a percentage of income if you think that what you're trying to do is measure income in the first place, then a charitable contributions deduction can make sense, but that's a very different view of the provision.

The wind incentive, I think, might be thought of in some view as a correction for measurement of income if you believe that the industry producing wind energy and its consumers are in some — to some extent taking on externality costs of alternative production of — of energy which pollutes. And in some sense, you might think of that incentive as a correction for a externality and indirectly as a measurement of income.

The higher education incentives that we have are a real mess in terms of complexity. And many would say are misdirected. I wonder if we are at a kind of a social inflection point in the sense that if you go back far enough historically, there was no such thing as public education and there was no sense that society should pay to educate a child because there was a deep economic analysis at the time, but if you think about it, a child who is educated is going to earn a much higher income because he will be distinguished from everybody else who is not educated, and that individual appropriate the return, so why should society pay for it? You got to a stage where society advanced to the point where people thought that an elementary school education is such a contribution to citizenship that the returns to it are social returns, they're not appropriated to the individual, and therefore we ought to have free elementary education. You know, move forward a few decades and people started thinking the same thing about high school education, which prior to that, had been thought to be totally appropriated.

Well, when I was your age, and I was studying taxation for the first time, one of the things that I learned was that higher education is distinguished from elementary and secondary education because the returns to it were fully appropriable to the individual, and there's nothing you learn in college that makes you a better citizen, you learn how to be a citizen in elementary school, so, it's all very private.

I wonder if we're getting to the stage where we're starting to think that people need a college education on the ground that if you don't get a college education, you are a burden to society because your earning power is less. And if that's true, maybe we're trying reflexively through the tax system to find a way to turn higher education into a public — publicly provided service because it is a public — it provides a public good. We're doing it in a lousy way, and maybe we need to think about it more carefully, but perhaps that's one explanation.

Just one comment about capital gains taxation for a moment. I'm a broken record on this subject. Just the notion that we might talk about a capital gains tax break on corporate stock because it provides some kind of relief from double taxation, well, when you think about the taxation of capital gains, you want to keep in mind that there's a deferral benefit which individuals are able — to which individuals are able to reduce their taxes, time their realizations, etc. There are all sorts of aspects of capital taxation that are rough justice agreements between the taxpayer and the tax system, and it's not clear cut where you get the balance in that process, and you know, short of accrual taxation, we're probably going to be struggling with that one for a long, long time until we decide that we flat out don't want to tax income from capital at all, which personally, I think would be a mistake.

MR. KURLAND: I'm Norm Kurland. The Center for Economic and Social Justice, and I'm co- founder of the American Revolutionary party.

I start out with agreeing with Jane Gravelle in that I don't think most tax incentives work. And I also want to start from the framework President Bush pointed out that we should move to an ownership society; I don't happen to agree with his approach in the kinds of incentives that he's providing to Wall Street in order to bring that about, but I think the idea of moving from a wage system economy to an ownership culture is a good one, and I think it goes right back to the founding of this country. So, I do agree that the tax system should not be used to reward. We have government agencies that should be doing that, and we ought to look at the purpose of taxation is to pay the cost of limited government and to make sure that we safeguard private property in the means of production, and that we restore free and open markets in terms of the efficiency of capital and the efficiency of production that we set up a tax system that encourages the accumulation by every individual, say of $1 million tax-free, take a glorified IRA that we would call a capital homestead account, taken off from Lincoln's notion of the 160 acres, but tying it into a bigger frontier. An unlimited frontier, the technological frontier. Let's imagine a tax system very simple, a flat-rate tax of income from moral sources above a poverty level. And then you would in all forms of income, no matter what source derived, would be be taxed at a rate necessary to balance the budget. So that whatever legitimate things the Congress thinks should be spent, then Congress can do that through the appropriations process, but I think the complexity that we've added to the tax system is one that forces people to lose faith in everything that goes on within the beltway and including sessions of this kind which are not really addressing the kinds of issues that the founders were talking about. How can you create a society where people are not dependent, you're not rewarded by government, but government is dependent on the people?

Now, if we had — we pointed this out, we had a congressman write Alan Greenspan. I'm glad that Alan Greenspan happened to agree with him that the tax system should not be used other than to pay the cost of government. But then he said he agrees with the idea of broad-based capital ownership. What he says is you must do it through the tax system. In other words, he was pushing it off to you people. Whereas, we're saying that the tax system should be neutral, should allow people to accumulate in a tax-free way, say on a $1 million per person. Inheritance laws encouraged the widespread diffusion simply by allowing people a tax above a certain accumulation rather than of the recipient, so that I think we really should be thinking in a different box and I would say that a paper on this was published thanks to Tax Notes on October 24, 2005, to go into more details.

MR. BERGIN: Thanks. Bill? And I'll get to you, Pam.

MR. NISKANEN: Bill Niskanen of the Cato Institute.

First, I'm concerned that none of the four speakers identified the basis for estimating the costs of these incentives. Let me suggest that the appropriate definition of that cost is the benefits that are foregone but not being able to maintain the same amount of revenue as is the existing situation, the benefits that are foregone by not reducing the effective tax rate on the base from which these incentives are granted.

In other words, the effective cost of the incentive is the benefits that are foregone by not reducing the tax rate on the particular base from which the incentive is granted.

Now, that would — that would — the beginning estimate of that is how much revenue you lose. But it is not — there is no obvious way of saying what the benefits are of increased revenue. What you have to do is hold a revenue constant. If you eliminate the base, if you eliminate the incentive, what are the benefits of reducing the tax rate on the — the base from which this incentive is deducted — to — to the general population?

In addition to the reduction of the complexities of the tax system, now our recent estimate of the complexity in the federal tax system is about $250 billion. Just in terms of the breadth of the complexity of it. So, I think we have to have a very much more precise definition of the nature of the costs of these incentives because an incentive or a deduction from a tax that already has significant social costs, it costs more than a deduction from a tax that has a relatively little social cost.

Second, I think we should recognize that a lot of the deductions are — the incentives in a tax system are partial accommodations toward a better tax base. We have instead of moving toward a consumption tax, we do it in dozens of different ways. Fine-grained ways.

Instead of reducing the general taxation of the income from capital from double or triple taxation of the income from capital, we do it in dozens of different little ways, some of which are probably not very effective.

So, in many cases, the political case that is made for these deductions is that the existing tax base is the wrong base. And so, they make the case that it is better to move selectively from a income-base to a consumption-base or better to move selectively from a double or triple taxation of capital income to the smaller taxation of capital income. We ought to compare that effort of dozens of little moves toward a consumption-base with the consequence of a general move toward a consumption-base, and we ought to compare the costs of dozens of moves toward reducing the double or triple taxation of capital income to a lower — to a better base — for the taxation of capital.

A third point, I agree with Gene Steuerle that the deduction for mortgage interest is not an incentive for homeownership. The incentive for homeownership — because the landlords take a deduction for homeowner — for mortgage interest payments, too. The incentive for homeownership is the homeowners don't need to pay taxes on the implicit rental income in their home. Now, if they did, then you would have a neutral tax system between renting and borrowing. Only then if you have neutral between revenue and borrowing, it is the lack of putting the implicit rental income on owner-occupied homes in the tax base, rather than the deduction for mortgage interest that is the effective incentive for homeownership.

I think it's worthwhile, all of you have made the point that a lot of these deductions and incentives are very badly designed in the sense that there's a good bit of revenue loss and very little change in behavior. And that's an important point to make. But I think you have to compare whatever benefits there are to these incentives to the benefits of collecting the same amount of revenue by getting rid of the incentive and reducing the tax rate on the base from which this is a deduction, and that is — that's [off microphone] through - - likely to do the analysis.

MR. BERGIN: Thanks, Bill. Pam? And then Ken.

MS. OLSON: Pam Olson, Skadden Arps. I want to go back to something that Joe mentioned when he talked about the fact that we really haven't defined what we're talking about here. I guess it seems to me that if we look back to the foundations of the income tax from which a whole lot of the things that are embedded in income tax derive, and what we find is that the analysis was very different. When we started this whole thing, we were measuring income. And as we measured income, we thought that the money that you gave to a charity, as Joe said, shouldn't be something that was included in income [off microphone] tax, I think you can say the same thing about interest. We looked at interest as — as merely something that reduced income, and we allowed a deduction for all interests.

As time has passed, our analysis of all of these issues has changed. I don't know if we've introduced more as a — as certainly in quantity when we've introduced more economists into the mix on a percentage-basis, I'm not sure whether that's true. But certainly, economic thinking has gone a long ways further than it was back then, and we understand things a whole lot differently.

That said, I think we have a very different understanding today of the tax code, and that affects the analysis that we hear in this room. But I also think that at some point when we started looking at things differently and we started looking at applying incentives to the tax code, that the wheels came off the train on the legislative side, as well because all of a sudden, it was okay to put things into the code that had nothing to do with measuring income, and instead, turned into all sorts of — of ways for people to hand things out in a disguised fashion.

And if, you know, we look at what Stanley Surrey starting doing with a tax expenditures analysis, I think when he kicked that off back in the 60s, that a whole lot of folks would have looked at that tax expenditure analysis and — and a lot of it would have been foreign to them, I mean the notion that a mortgage interest deduction was a tax expenditure would have been a foreign idea, and to a certain extent, I wish that — to go to I guess Bill's point, we had done a better job of defining exactly what our base was that we could do a better job of measuring deviations from it. We'd have a whole lot better understanding of it, and maybe we'd have a better tax code than we do today, as well.

MR. BERGIN: Ken?

MR. KIES: I wanted to respond to one of Bill's basic points, which was what I think is an intriguing idea but where the practicality suggests it's not doable.

The intriguing idea is that the proper way to approach these incentives is to say how much lower would our marginal rate be if we didn't have them?

One of the things learned in 1997 —

SPEAKER: And the consequence for it.

MR. KIES: No, no, I understand. That — that gets even more complicated. But in '97, the Joint Committee was under a lot of criticism from the Wall Street Journal editorial page for the alleged fact that we did static revenue analysis. And —

SPEAKER: Only in '97?

MR. KIES: No, no, no, it's continued but I can assure you that when your picture appears on the editorial page, this is a bad sign.

MR. BERGIN: Yeah.

MR. KIES: And so, we conducted a year-long project —

SPEAKER: That's known as a targeted attack.

MR. KIES: Yeah. Attack, yes.

SPEAKER: But at least the Wall Street Journal said you guys were lawyers and you didn't know any better.

MR. KIES: I believe it was "obscure beltway lawyer" to be precise. No, excuse me, "otherwise obscure beltway lawyer."

But we conducted a year-long project, which Jane participated in, and I think Gene may have, and others around here, to try and determine a slightly different thing, which is how would you actually measure macroeconomic effects of changes in the tax system? Because that was the real criticism, and the Wall Street Journal has always — I don't want to use the term lied, but they've intentionally —

SPEAKER: And why not?

MR. KIES: Well, they know —

MR. BERGIN: You're among friends.

MR. KIES: They know — they knew full well that the Joint Committee has always taken into account behavioral response. The real issue is not that, it's whether or not you measure macroeconomic effects like change in interest rates, labor supply, GDP growth rate.

Well, we published the results of this year-long study in December of '97, but one of the fundamental things that it found, which was not surprising to anyone, but it did, I think, a good job for the first time of documenting in a fairly thoughtful way, was the fact that the vast bulk of tax changes, even things that you may think of as big, do not have any measurable macroeconomic effect. And I would submit to you, Bill, that the same thing probably would be said of the vast bulk of tax expenditures. If you tried to individually measure what would be the consequences of what would be the modestly lower marginal tax rates, you would have a very difficult time quantifying it.

And so, I think that — while I think what you posed is a very intriguing prospect or concept, I think it's practicality in terms of implementation would be very difficult to produce something that would be meaningful to decision-makers.

MR. BERGIN: Bill?

MR. NISKANEN: It's not clear that it would be marginally lower. I mean, the total of tax expenditures is something on the order of, I don't know, between 60 and 80 percent of the revenue from the individual income tax.

MR. KIES: But what Bill was posing was we should be analyzing individually the benefits or usefulness of tax expenditures, and when you do that, I will assure you that the vast bulk that is 90 to 95 percent would produce a marginal tax difference, which would be almost difficult to calculate unless you went out to four or five places in terms of decimals.

MR. NISKANEN: But you — Bill Niskanen, again. You can always at least make a crude estimate of so what's the consequence of reducing the marginal tax rate by one percentage point? And use that as the base for estimating how much you could reduce the tax rate given you — elimination of a particular incentive. And so, I think what you ought to do is address the correct problems even if they're more difficult.

MR. KIES: Okay, but just one follow-up and I know you want to go on, is when we did this project in '97, and I'll be the first to admit I'm not an economist, and I'm not saying that's a good thing or a bad thing, you can judge that on your own. But I actually spent a lot of time understanding and trying to learn how the models that these economists were using to evaluate these things, and one of the interesting aspects of this project was it went through a series of sort of iterations where each of the teams produced their first product, and they all criticized each other, which was really fun to watch because they were pointing out what idiots they were to having a — this project where —

MS. GRAVELLE: I was the only one who was right.

MR. KIES: Right, no, no, exactly. But — but one of the things I learned from that was how much economics is an art as opposed to a science, and therefore, I would suggest that if — when you're measuring these — trying to determine the effects of these relatively small marginal tax rates, it would really depend largely on what kind of economic model you're using, and one of the things you then learn when you get into it is what's fundamental to these economic models are the elasticities that are assumed inside them, and then when you ask, well, what's that doing there, the answer becomes very, very fuzzy. It's based on a study that some guy did 10 years ago of what Indians would do if they were handed rubies, and I mean, it's so, and I'm just saying that we have a long way to go, and I think our project in '97 proved this, and I don't think we've made a lot of progress since. Other than in major, major tax changes. Like, for example, taking a fundamental income tax system and going to a fundamental consumption tax system. Other than those kind of major changes, any of these measured effects really I think are highly at least unreliable in terms of them being the basis for decision-making.

MR. BERGIN: Jane, I'll let you get in here.

MS. GRAVELLE: I just wanted to make three quick comments about things I have heard here.

First is what you call a free agency, we economists call a race to the bottom.

The second is there's been a lot of talk about the R&D tax credit, and I agree there's horrendous problems with identifying R&D and administering that provision. But actually, among economists, the R&D tax credit is probably one of the ones that's a little more palatable because we do have evidence that the social returns to innovation are much higher than the private returns, so if there was something you wanted to subsidize, that would probably be it, and now there's sort of an offsetting argument that people rush too fast to discover things before other people do, so, you're not sure about even the theory.

But basically, that stands head and shoulders in terms of desirability if you can do it right compared to most other tax incentives.

The third is that the Haig-Simon definition of income says you should deduct interest. A deduction of interest is entirely appropriate to income tax. The reason that you would criticize a reduction of interest is because interest is often incurred to purchase consumer durables, and that's not just houses, it's cars, it's furniture, and whatever. But if you borrowed money to, you know, go on a vacation or eat an ice cream cone or something that's truly consumption, it is perfectly proper to deduct that. And — which is kind of ironic, because when we cut back on the interest deduction. Of course, we disallowed all the interest except the one that was less obviously questionable I have in there, which is the mortgage interest for unoccupied housing, and also it created in a little industry in equity — home equity loans in the process.

So, I just wanted to kind of clarify that issue about interest deductibility.

MS. BLOCK: Cheryl Block, George Washington University.

I wanted to raise an issue of concern to me and put it on the not-so-round table. As much as I sympathize with the victims on the Gulf Coast and the disasters, I'm curious and concerned about the increasing use of the tax code for disaster relief and rebuilding. In particular, with the rebuilding effort, wondering as to whether we should be using the tax code to support one region at perhaps the expense of another. And yet, we use this after 9/11, we're using it again, it's become a very popular technique.

Curious, too, about Jane's comments earlier that empowerment zones are among the least effective, and here what we have is use of the empowerment zones. So, sort of just raising a concern and interest, and if anyone else has thoughts on this as to whether or not there's a better way to go, and if it's appropriate to be using the tax code this way.

MR. BERGIN: Eric?

ERIC TODER: Yeah, I guess I had a thought on that question, which is really related to other things that have been said in Ken's point earlier about there's no reason why we should think a tax code is worse — provision is worse than a spending provision.

I think the threshold question is not whether it's in the tax code, but whether we should be doing this at all through the government. And then the question of whether it should be in the tax code or not is really a second order question, and sometimes the answer is on one side of the line and sometimes on the other side of the line. I do agree with people who have spoken before who said there's a bias toward putting things in the tax code, and so, probably you get more stuff in the tax code that shouldn't be done than other places, but the threshold question we really — unfortunately, at Treasury, we have often said this doesn't belong in the tax code, when the really right answer should have been this doesn't belong anyplace, and we didn't feel it was our business to say that.

MR. STEUERLE: I've got an anecdote on that from the '86 reform. I remember when we were drafting all the reforms we were going through, which was a pleasure to start with base broadening because we just did the rate schedule to keep the same progressivity, and the lawyers would start coming with these drafts, and they'd say this thing is really bad, it's a tax expenditure, it doesn't belong in the tax code, and I said, you know, we actually can't say this, which is an issue we haven't gotten to, I said because we're stuck in dealing with the tax code only, and the only reform we can offer is to get rid of it and offer rate reduction, we can't offer the reform of putting it in the spending code, so, the only economist choice we can offer the people is fixing it in the tax system or getting it out off the tax system, but we can't switch it to the expenditure system.

So, I think part of what your — I agree perfectly, Eric, but it also raises the whole question of, and this goes to this issue on what's going on with [off microphone]. A lot of the stuff is jurisdictional, but taxwriters want to be able to do something that other people can do, and they've got the power to do it.

MR. BERGIN: I'm sorry, go ahead.

MR. TODER: I just had a couple of other general points.

I mean, I think Don's original question, what is an incentive, has been answered by a lot of people at the table, but I guess to go through what we went through with Don and Jane, and I guess I'm willing to identify that the two of us worked with Jane on that project she talked about. When we developed this dirty dozen list, part of the reason we didn't pick the biggest tax expenditure is because many of these provisions were provisions like the housing exemptions, like the capital gains exemptions, which were kind of fundamental tax structure issues, which we may or may or not have agreed with the way the tax code dealt with them, but they were controversial within our group, and if we sit down and try to design a tax system that's administrable and addresses income measurement issues, you're going to have a lot of things in there that look like things that are in the tax expenditure budget.

And so, I think that the point about those, at some level, they are incentives because they do have effects on behavior. But they're not in there deliberately like the issue you raised to create an incentive which grayed in exception. And there's lots of provisions in there that would have no justification on any tax policy grounds, but are there for some spending purpose.

I want to say a couple of things about — broad things about trends, I guess. One is one of the largest areas of tax expenditures in the code is in the areas of health and pensions, as you all know, the United States developed a welfare system after World War II, which was very different than that developed in Europe and Japan and elsewhere, where instead of doing some things where other countries do it through their government, we did it through encouraging private employers or large employers to provide pensions in health insurance plans for their employees.

And so, the basis of a lot of the big tax expenditures was that broad social policy; I think looking to the future, that system is unraveling. And so, the question of whether the tax system — incentives, some of these provisions should be kept in the law to keep them from unraveling so fast that you leave a lot of people out in the cold, or remove from the law because we're substituting something else, it's something that's a very critical issue that we're going to have to face up to.

The second issue I wanted to raise is the issue of refundable credits, and Ken raised that with regard to businesses, but obviously, it's an issue with respect to low-income individuals, and we've reached the system. I did some calculations yesterday that for a family of four, your income is — got to be over $40,000 in 2005 before you owe any income tax at all. For a family of six, it's $60,000, and I realize that's a little unusual these days.

But what we're getting to, when you have — when you lay on all these credits that they are going on for middle-income and lower- income individuals, is you keep running into situations where people can not use the credits if they're not refundable. And I think it's very hard to justify capping the benefits of credits in that sort of way.

So, I think we really need to think about either making them refundable credits and controlling their costs in some other way, or pushing them over to the expenditure side of the budget. But I think we're getting into almost an increasingly untenable situation with regard to these provisions.

SPEAKER: But you understand there's a requirement that goes with that, that the IRS increased the audit of a refundable credit because somebody has delivered the money.

MR. TODER: Well that's —

SPEAKER: And that's —

MR. TODER: — one of the issues.

SPEAKER: — something they've never mentioned.

MR. HALPERN: Jim Halpern of the United States Tax Court.

A lot of what Eric said I agree with, and he said some of what I wanted to say, but I'll take a minute to say it anyway.

I thought it was a bit interesting that Pam Olson took us back to Stanley Surrey and the origin into the tax expenditure budget. Maybe people will remember that there was a rebuttal to that approach by Boris Bittker, who cogently wrote that this notion that there is an ideal tax system, and that you can define what it is, whether it's anyone in particular is kind of difficult, and Boris gave a list of things that he said, well, is this a tax expenditure, is the rate system a tax expenditure? Is the standard deduction a tax expenditure? With a little bit more thinking, you could ask, is anything other than sinking fund depreciation a tax expenditure? It does come down, I think, to you set a standard, you set it. If you look at the consumption-based tax, then incentives for savings from an income tax perspective or a tax expenditure, but they're certainly not if you have a consumption-base view point. The Haig-Simon is not serviced well in this kind of analysis.

And I come back to Eric and indeed to Don in the sense of saying it doesn't really matter to me what's an incentive. The question is can it be done well through the tax system? Is it — I would apply an efficiency criterion, and I would also apply a political criterion in the sense of is this what we want? Do incentives, if you want to call them that, work better outside the tax system? I think you only have to look at the government's expenditures on trailers to relive the Katrina victims and compare that to anything that was passed by the Congress and the tax system and give somewhat of a lie to the notion that you necessarily do it better than a tax system.

So, I hope that I've sat here for an hour and a half and been able to summarize a little bit of what I've heard and those are my views. Thank you.

MR. BERGIN: Thank you. Marty, you want a microphone?

MR. SULLIVAN: I'm Marty Sullivan with Tax Notes.

One thing that always makes me uncomfortable at conferences like this is how there's a disconnect between the academic standards and the reality on the Hill. That many of us are familiar with. And so, we go through this conference, and it's a very good thing, but then we are dissatisfied because that's as far as it goes because maybe it'll get in the press if we're lucky.

We — so, I started thinking in terms of —

MR. BERGIN: Hopefully.

MR. SULLIVAN: We think there's a dirty dozen. Maybe that'll get some press

Two years back, I wrote an article called "Joe Six-Pack." In there I calculated the tax expenditures of a per-dollar amount per cap basis. So, the R&D credit, three bucks a year.

So, that, you know, it's one way of communicating it to — so, another I — I'd like to see is just a member of Congress every year propose — introduce a bill and have it estimated by the Joint Committee what if we got rid of all tax expenditures and [off microphone] rates be. Is always have that on the table as an option.

So, these are all gimmicks, those are all ways of getting people to think about. And so, it's not going to come from within the room, but it's going to come — it's going to be facilitated by what happened in the room, but it's outside forces.

And the outside force, which is I think [off microphone] in tax reform right now is — is international tax competition. And I just was so surprised when you look at Europe, they're lowering their rates, they're lowering their rates because of tremendous competition, but they're not lowering their taxes. That's what they're broadening their base, and so, I'm most hopeful.

First, I think we need to look for gimmicks, and then I'm hopeful that tax reform will take place through tax competition.

MR. BERGIN: Sir?

MR. ALEXANDER: We've heard several times today that IRS can administer a tax expenditure somehow more efficiently at less cost than the agency that has the proper responsibility of that particular expenditure, the Department of Energy, Education, what have you. I'm not sure that's been tested. I'm not sure that that works, despite the trailer example that a distinguished journalist raised. Not all of the other agencies are super efficient. Surely. But IRS isn't either. And if IRS is given more than it can handle, and it's surely been given that lately, IRS is not doing the job that it should do for the American public in trying to make our tax system work effectively. I think we have reached the tipping point, and I think that we go over pretty soon if we decide that more than $900 billion of what should be in our outlaid budget has to be administered by the Internal Revenue Service. They can't do it.

MR. BERGIN: Good point. We'll go to the audience here.

MS. MCGUIRE: Thank you. Monica McGuire, National Association of Manufacturers.

I have a question going back to the R&D tax credit, and this is for anyone who would like to answer it. Moving from the academic arena to the real world, recent analysis has shown that you can go anywhere in the globe, any sector of the globe and there is a competitor nation to the United States that has a permanent and a more generous R&D tax credit. And we know based on recent statistics that the growth in R&D, the international or non-domestic R&D growth is growing faster than domestic R&D. And so, I'm wondering if what's happening in the global market is any indication of the effectiveness of the U.S. R&D credit?

MR. BERGIN: That's a great question. Anybody want to jump on it? I'm going to close with this. Marty, go ahead.

MR. LOBEL: You know, that question —

MR. BERGIN: Tell everybody who you are.

MR. LOBEL: Oh, I'm sorry. I'm —

MR. BERGIN: That's my boss.

MR. LOBEL: — Marty Lobel, Lobel, Novins, and Lamont.

That question raised an issue that's been present in this room and nobody's raised it. The fault isn't in the tax code; the fault is in the system of how the tax code gets written. So long as we have a tax code which is written by auction system, we're going to have these problems and you're not going to behave rationally.

You know, I remember when Ken was up there, he'd do all these wonderful things and, you know, members would look at him and say that's real interesting, and go then go vote their pocketbook interest.

What the Joint Committee on Taxation has done is essentially set the outer parameters of where grand larceny begins. What you've really got to do is change the incentive system. In the system right now, if I'm a businessman, I'd be a damn fool not to try and buy a tax revision because the rate of return is going to far exceed anything I can get in the marketplace. And that's going to continue so long as this system continues.

All the economic analyses and at one time, I have to confess, I'm not an economist, but I did supervise a Joint Economic Committee for Senator Proxmire And the one thing I learned is ask what the assumptions are. And then, you know, these wonderful econometric models sort of float off into the thin air. The issue it has, we've got to change the congressional system to eliminate this kind of behavior so that rational tax policy can begin.

You know, in '86 we did it. I remember the '86 act. It was a big battle. We scrambled enough eggs so the lobbyists couldn't figure out where their dollars were coming from.

R&D tax credit, I have a very good friend who's in Ways and Means, and he looks at it as his best fundraising device of all. I mean, why do these guys get on Ways and Means and Finance, because they're interested in the intricacies of the tax code? Come on. I mean, what we're really talking about is we — we, as tax professionals — or economists and so forth, you're dealing with the tax code and what the political system is doing to it. And until we reform the political system, we're always going to be operating at the margin.

So, this idea of the R&D tax — why did the Wall Street Journal, who is — the news part is really good, the comic pages are really amusing, but nobody takes them seriously, but the R&D tax credit is a classic example. Why did Microsoft move its "office" in Ireland into a solicitor's desk and save $300 million? Glenn Simpson reported it in the Wall Street Journal. Why did he do it? Because they could. They should. If they are rational businessmen to save their taxes.

And, you know, we are a big market, we have a vested interest in an efficient economy. We're not going to have an efficient economy if we keep doing what we're doing now, which is salami-slicing the tax code into something that is totally unenforceable and unrecognizable.

Why do you think the enforcement budget of the IRS is about a third of what it should be? It's been cut and cut and cut. Why did Congress go out and say, oh, we've got to enforce the tax code, so, for those earned income credits, we're going to, you know, audit, you know, someone under $25,000 a year income has four times the chance of being audited as someone who over $100,000 a year income.

Is that rational? No, but it's political. And you know, quite frankly, everything we've said here is important, but until the political system changes, it ain't going change.

MR. BERGIN: I'm going to let those be the last words because I promised 11 o'clock. I apologize to all of my friends who I called cheapskates in my opening remarks. It's not true. I made it up just for dramatic effect. And more importantly, would like to thank our panelists, Ken, Jane, Mike, and my friend, Timothy. And all of you. Thank you very much. We'll do this again soon. Have a good day.


(Whereupon, the PROCEEDINGS were adjourned.)

* * * * *