Tax Analysts®Tax Analysts®

My Subscriptions:

Events

TRANSCRIPT
ROUNDTABLE DICUSSION:
TAX REFORM AND SIMPLICITY —
MUST A GOOD TAX CODE BE A SIMPLE ONE?



TAX ANALYSTS
Washington, D.C.
Friday, July 8, 2005

DISCUSSANTS:

CHRISTOPHER BERGIN
President, Tax Analysts

NINA OLSON
National Taxpayer Advocate

BILL DIEFENDERFER
Former staff director for the Senate Finance Committee

GENE STEUERLE
Senior fellow at the Urban Institute

ALSO PRESENT:

DON ALEXANDER

DAVID BRUNORI

SHELDON COHEN

HOWARD GLECKMAN

MARTIN LOBEL

JOE MINARIK

BILL NISKANEN

STEVE ROSENTHAL

ERIC TODER


PROCEEDINGS

(9:02 a.m.)

MR. BERGIN: Good morning on a rainy day. Nice to see all of you. I'm Chris Bergin, the president of Tax Analysts, the nonprofit publisher of Tax Notes magazine, Tax Notes Today, State Tax Notes, Tax Notes International, and many other print and online publications. Welcome to the third in Tax Analysts' series of roundtable discussions on tax reform, and thank you for coming.


President Bush said he would make tax reform a high priority in his second term, and then he appointed a panel to study the issue. We at Tax Analysts thought that we could provide an important public service by highlighting the key issues that any serious reform effort should address. We decided to air these issues in a series of roundtable discussions with people like you, leaders in the tax policy community.

The first roundtable in April was a stimulating discussion of the lessons we can learn from the effort that produced our last major reform, the Tax Reform Act of 1986. In our second roundtable, we asked, What Role Can Tax Reform Play in Promoting Economic Growth? That question, too, prompted a stimulating discussion. Today, we will discuss whether a simple tax system necessarily makes the best tax system, or to put it another way, what role should simplicity play in tax reform.

For those of you who are new to our process, here's how it works: I will open with some brief remarks on our subject, then I'll introduce our panel of three distinguished speakers who are around this table. Each of them will address aspects of our topic for about 10 to 15 minutes. After that we'll open up the discussion, and you are all encouraged to participate, whether you're sitting around the table or elsewhere in the room. Just wave, and I will find you. We're recording this event, and we will post the transcript to our Web site, as we did with our previous roundtable discussions. Also, for media purposes — and there is media here — we're all on the record, so when I recognize you, please tell us who you are. Also speak into a microphone; for those in the audience, we have hand-held mikes that we will get quickly to you. I will moderate the discussion and we will end no later than 11:00. So, let's get started.

I'd like to begin with a confession. Probably because I spent so many years with the tax code, I've developed the habit of dealing with complex issues by trying to break them down into simple pieces; it works pretty well for me, so I like to think of myself as simpleminded. My wife keeps begging me to stop telling people that; she thinks that simpleminded means, well, not too bright. But whether people tend to accept my meaning of simpleminded or hers, our difference on this issue helps illustrate my point: Simple can mean different things to different people. At first glance, one could argue that the United States has a relatively simple tax system. That would probably be someone who thinks that a truly simple tax system must have few marginal rates, no more than two or three. We've got six, with rates of 10, 15, 25, 28, 33, and 35 percent; that's not too bad. And then there's the Form 1040. Personally, I think the Form 1040 is a marvel of simplicity — really, just look at it. It boils down all the individual income tax rules to one piece of paper, two sides. If you prepare your own return, you soon get used to the 1040 drill. Basically, you use the same lines year after year, with some detours along the way, and you barely glance at the lines that don't apply to you.

But dig a little deeper into our tax system, dig below the official marginal rates and beyond Form 1040, and you learn that perhaps our tax system isn't so relatively simple at all. For example, if you itemize deductions, you fill out a Schedule A, relatively straightforward effort for those who have done it a couple of times. But wait, you may have to go to a worksheet to find out if the tax code limits all those deductions that you thought you were entitled to. There's nothing simple about that worksheet, and nothing fun about finding out you're going to get less than you thought. Next, you may have to go to another worksheet to see if you can claim all your personal exemptions — you know, the spouse and the kids. Well, at least there's the child credit that lowers your tax dollar-for-dollar. Not so fast. You may need to go to another worksheet to find out if you get the full credit, only some of the credit, or none of it at all. By this time, you've probably developed your own word for the term "worksheet." And by the way, you'll have to check whether you'll have to pay anything under the alternative minimum tax. You've guessed it: There's another worksheet for that.

So how many new and hidden marginal tax brackets have we discovered, and how many more lay hidden inside our tax system? Many more. And if hidden effective marginal tax rates are tough on the upper and middle classes, the ones that can apply to low- and moderate-income working Americans are just plain brutal. Before we start talking about how many marginal tax rates we should have, maybe my government should start talking to me about how many tax rates we do have. That's something that worries me about the folks promoting a flat tax or a national sales tax. I am supposed to believe that the crowd that's never been straight with me on taxes — that would be Congress — is going to create a simple tax system that has no gimmicks, that are designed to take more of my money without telling me about it up front. Even the great Tax Reform Act of 1986 — which I do think was great — had smoke and mirrors, and I'm not just talking about the AMT.

When President Ronald Reagan signed the Tax Reform Act into law on October 22, 1986, the American people were told that the individual tax system was simpler because it had only two brackets: 15 percent and 28 percent. Actually, it had three. The politicians put in something called the bubble, a 33-percent bracket that was designed to take away the benefit of the 15-percent bracket for taxpayers in certain income categories. So just because we call something simple doesn't make it so. Theoretically, of course, we could devise a tax system that most taxpayers would consider simple. It might have just one or two rates and no deductions or credits. It might require no more than a postcard-sized tax return. But a tax system like that would not do all the things that many, including President Bush, think a tax system should do, things like a mortgage interest deduction to encourage home buying, a charitable deduction to encourage giving, and tax incentives for families to save for retirement or pay for college. Which brings me to the point of our discussion today: Notwithstanding the appeal of simplicity, is the best tax code really a simple one?

Now, let me introduce our panel. Nina Olson is the National Taxpayer Advocate. She's right there. Bill Diefenderfer is former staff director for the Senate Finance Committee during the last successful round of tax reform. And Gene Steuerle is a senior fellow at the Urban Institute, a long-time columnist for Tax Notes magazine, and was in the leadership role at the Treasury during the last tax reform. Nina, may we start with you?

MS. OLSON: Oh, I guess so. I often — I was wondering why I was here at this panel because I don't do tax policy, as I keep telling people. And then someone reminded me that we had identified as the most serious problem for taxpayers this past year the complexity of the code, and that sort of then explained to me why I was here, that we talk about the problems that taxpayers experience because the code is not — or this tax system, which is broader than the code, is not simple.

And I thought that I would start out by just simply saying how my office looks at these issues, which is, in many ways, taxpayer-centric and, secondarily, IRS-centric actually, because I am of the firm belief that whatever Congress passes or whatever the Executive Branch proposes in the form of tax simplification, has to be something that makes sense to taxpayers, is not just some — no offense to economists, but some economist's view of what might be simple for a taxpayer to comply with but actually that taxpayers can comply with it, or at least most taxpayers can. And then secondly, that the IRS can actually implement the darn thing without imposing an enormous burden on itself or on taxpayers, that it makes them tie themselves up in pretzels just to comply with the so-called simple provision. And the third area that I look at tax law about — and I haven't really heard a lot of conversation about this as we start talking about tax reform — is the concept of taxpayer rights. And I honestly don't know how you implement a tax code unless you actually articulate an underlying framework for what rights taxpayers have within that tax system so that when you do come up with disputes or disagreements, that there is no question how the taxpayer can seek clarification and avail themselves of those rights.

So I — what my first point would be that in any conversation about tax reform, that you have to think about the taxpayer rights aspect of that and the due process aspects of provisions, how taxpayers get their day in court, how — what kind of administrative review they get, et cetera. Then from that point — and I think that if you took those three issues, you know, looking at the law from taxpayer perspective, the burden that it puts on taxpayers and taxpayer rights, you can basically map back any of the things that I have written about it over the last four years in my annual report to Congress, and the proposals that I, at least, personally have made to Congress, both legislatively and then to the IRS, administratively.

In other conversations with people, I have since come up with a mantra for myself in thinking about tax reform. And some of you in this room may have heard me tell this story, but for the last two or three years, I've been in ongoing conversations with the United Kingdom, with what used to be called Inland Revenue and is now called Her Majesty's Revenue and Customs — because they just merged revenue and customs - - and the folks at HMRC have been wrestling for the last four or five years with some reform of their child benefit payments and moving them into the tax system, and they certainly have some struggles with that issue, and they've come over and consulted with many of us in the IRS, and just this past May I spent a week over there meeting with them about their provisions and what I could learn from them about the Earned Income Tax Credit.

But in one of my conversations, I was trying to explain to my counterpart some of the more complicated provisions of the Earned Income Credit, in particular, the provision that we require the proof that the child live with the parent for more than half the year, and they have to have the right relationship and things like that. And I was also talking about the dependents exemption, how we have the support test, where the taxpayer claiming the child has to have paid more than half the support of the child for the year. And that person looked at me and said, why do you care? And I just thought, I don't know. Why do I care about that particular proposal or that aspect of it? And that really stuck in my head, and I just thought, one of the tests that I'm going to ask about any kind of provision that comes across my desk is, why do I care about this particular aspect?

In terms of the EITC or child benefit or child tax credit, if I'm only — if I want to provide some kind of support, financial support, for persons who are working and who have children — well, maybe, one, I need to separate out the support that I'm giving to persons who are working and then support for persons who have children. And if I only see that child coming in once to the system, and I've decided that X is the right amount of support for someone associated with that child, then why am I doing all of these really complicated and difficult-to-comply-with rules if I only see that child showing up once in my system for one payment?

Why do I care about all this other stuff? Now, we may have lots of reasons why we care, but we should look at every single one of those reasons and say, does that warrant imposing a burden on 144 million taxpayers who are claiming dependency exemptions. Does it warrant that? And I don't know whether it does, but I thought, that's a really important question to ask. And of course, with the AMT, you can ask that question as well. Why do I care that people are escaping taxes because they claim the standard deduction or the dependency exemptions, you know? And so if you go through — if you keep that as your mantra, I think it really helps you answer the question and then weigh those things that matter to you compared to other things. I also set out for the tax reform panel a few things that we suggested that they think about, about any tax system, and I think it's worth - - when you're trying to come up with tax reform, it's worth thinking about.

And again, some of this is taxpayer-centric, some of this is IRS-centric, some of this sort of thinks about taxpayer rights. And the first one is — and maybe this goes to some of Chris's points — you don't design provisions that entrap taxpayers. And of course, some of what I'm thinking there is the AMT, that, you know — most people do not think that because they have dependents exemptions or the standard deduction that they would be trapped — would be pulled into the AMT, and they file their returns normally, and then voila, they get the notice from the IRS saying, you think you paid your taxes, but you haven't. I don't think that bodes well for the tax system. I don't think it makes people feel like they have a tax system that understands how they live their lives.

And if we're asking people to comply with the tax system, then we have to help them comply, and one way is to not have things trap them. I do think that for the majority of Americans, they need to be able to comply with the tax system, fill out their form on a simple, you know, two- page form, and that the rest can be document-matched. I think there's some opportunities there for simplification. And I think that you have to have a system that even if taxpayers choose not to fill out their own returns, it's designed so that both taxpayers and many small businesses can complete their own returns, and that that should be a question that is asked regularly throughout the reform process. Does it drive people to preparers, or is it something that the average person can understand and fill out. I think that it has to be something that tax administrators can explain, that you don't — you know, when you think about the job of the poor folks in IRS forms and pubs trying to put into lay language some of the provisions that we've got, if we literally, in the process of tax reform, cannot explain something in lay language about a provision that is going to affect many, many taxpayers — as opposed to those provisions that affect a few — then we should probably go back to the drawing board on that provision because it's just going to lead to errors, which causes taxpayers problems and the IRS problems.

We have to anticipate the largest areas of noncompliance. If we have a population of taxpayers — the low-income taxpayers who are going to be pulled into this system through the Earned Income Credit — then we darn well better design a program that they can navigate because otherwise we're setting them up for noncompliance, and the last 20 years of hearings on EITC noncompliance have demonstrated that. We have a tax system that creates a whole army of new products and industries: Refund anticipation loans, tax preparers, people selling cars because of electronic filing — you know, those sorts of things. Well, the tax shelters are another whole industry in and of itself. I think we really have to think about those consequences too because all of those industries make different players and different pressures on a system to deviate from what most simple model. And finally — this is a really important point — I think you have to have a system that allows for choice, but not too many choices. And in my annual report this past year, you know, we identified the education provisions and the retirement provisions as prime examples for — and prime areas for — simplification.

When you have nine education provisions and all these different definitions of what constitute higher education - - or qualified education expenses and different income thresholds for different provisions, you know, a taxpayer has to sit down and wade through those materials and really leaves that process thinking, did I get the right answer for me? Is there something else out there that I missed? And that leads to, "they're doing this to torment me. Why should I take the time to get it right?" And I think all of those considerations are things that we should think about.

The last point that I really will make is, the IRS has a burden model, and one of the things that I find most frustrating about the burden model — and I haven't been successful in getting them to change it — there are two aspects, actually of it. One is that the burden model measures the number of lines as additional burden. You know, like if you have a provision that causes this many lines on the return or requires two notices rather than one — that the taxpayer's getting two notices — there's an assumption or presumption that that is more burdensome. I give you as the example a procedure that the IRS used, which is combination letters.

In an examination, what they will do for some of what they think of as their easier exams, namely EITC exams, which we all know are not easy, they used to send the taxpayer one notice that would say, we've looked at your information, we don't agree with you, if you want us to see more, send it into us, and oh, by the way, you have 30 days to request an appeal. What that notice was doing was telling them two things, and they were contradictory things: Continue to talk with exam, but also if you want to have an appeal with the Office of Appeals, send in that to us. Well, what does the taxpayer do? Very confusing. And it took me three years to get them to agree to separate out those notices. And part of the argument was, well, you're sending the taxpayer two notices. Well, that's not burdensome if those two notices achieve clarity, if those two notices help the taxpayer understand what it is that they're supposed to do. Same thing with number of lines on the return. You have to really understand what those lines do. I think I'll just leave it there.

MR. BERGIN: Thank you, Nina. Bill?

MR. DIEFENDERFER: Thank you, Chris. I hope I don't offend anybody by more or less reading this. When I was involved in tax policy day-to-day, I could do this extemporaneously and be relatively coherent. I'm no longer involved in tax policy day-to-day other than my own tax forms. I once held a job as majority counsel of the Senate Committee on Commerce, Science, and Transportation. It was a wonderful job. The jurisdiction of the Committee spanned the globe and the universe: Telecommunications, NASA, fishing, insurance, science, railroads, trucking, maritime commerce, and a myriad of regulatory agencies: The FCC, the ICC, and the Federal Maritime Commission, and yes, the Federal Trade Commission. I learned from the Federal Trade Commission about the importance of full disclosure, so here I go.

Well, I was trained in the law generally and international economic law more specifically. My training in tax law was filling out my own tax return, and later of the businesses that I either ran or owned, and I do that to this day. I became chief counsel of the Senate Finance Committee because the incoming chairman, Bob Packwood, had been chairman of the Senate Commerce Committee and liked my work, so he invited me to give up my law practice and come back to run the finance committee. A liberal arts education made me feel I could learn and do almost anything, given a bit of time. As time passed, I've narrowed that list to disinclude things like brain surgery, and maybe it should have been tax policy, too. Besides, Chairman Packwood had been quoted in The Washington Post as saying he liked the tax code the way it was, so how hard could this reform thing be? I was about to find out.

We had to go through the motions, at least, so we held about 14 months of hearings, and I suspect I'm probably one of only three human beings on earth who attended each and every one of those hearings. It was fascinating. The lead tax theorist in the nation testified, Nobel Laureate economists, practicing tax professionals at the zenith of their career, and we weren't holding hearings, they were just hearings. They were at our beck and call for tutelage, and we called on them. Think tanks fought to educate us. It was a glorious educational opportunity, and I took total advantage.

I learned many specific things, most now are forgotten, and many general things, which most I think have been retained. Let me give you brief examples of the latter. The road to hell is paved with good intentions. Social Security was enacted because we needed to take care of our elderly, so we taxed the working poor and the not-so-poor to do so. Good intentions. As time passed, that burden increased, especially on the working poor. We were essentially asking parents of young children to set aside money for their retirement when they could not meet the needs of their children. Imagine a mother robin with three babies catching three worms and setting one aside for her future. Very good intention, but we were distorting nature at its most fundamental point.

Not to worry, we invented the Earned Income Tax Credits to give back some of that money to the working poor. Of course, this involved complicating the code, but that seemed the least we could do. When a recession loomed or was, in fact, underway, we learned Congress would routinely enact investment tax credits to stimulate the economy. Sounded good, but the economists were almost unanimous in saying that the stimulus was too little too late, and in fact, usually ended up overheating the economy at a later day. This was further compounded when one panel of economists, which I think included a Nobel Laureate, was asked how much of the investment tax credit stimulus actually sticks to the wall, goes to stimulus. No more than 40 percent was the answer. Well, our intentions were good.

Final example, I met with some of the most influential men and women in corporate America, who wanted to help us understand what good tax policy was. I am not being sarcastic; they were good people. However, I did learn to ask each one of them one question after hearing their proposals for tax code changes: Please tell me how much, in dollars, this proposal is going to increase your company's tax bill. The law of large numbers failed. Not one of them would have increased their tax bill, and all would have decreased their tax bill. Imagine that. This may sound cynical, but it is not intended to be. We must understand that we ask the code to do too much. It is bound to become complicated over time, and in some instances, actually hurtful to economic progress in the United States. Reform is about paring back these problems, not eliminating them. Human nature and good intentions and, in some cases, judgment, will not allow us to design and implement the perfect tax code.

That said, I want to be directly responsive to the question this session is meant to address: Tax reform and simplicity — must a good tax code be a simple one. If this were "Law & Order," and I was directed to give a yes-or-no answer, my answer would be no. Having a Republican pedigree, I hate to use the word "nuance," but this issue is more nuanced than yes or no conveys. First, let me briefly set the context for successful tax reform in our lifetime — what I believe to be successful tax reform in our time. All proposals must be evaluated with an admission that this is a political process and that though some proposals may be simple and arguably good, they will not prevail, at least in the foreseeable future.

For example, scrapping the present tax system for a sales tax, regardless of the merits or lack thereof of this proposal, is too radical to be a serious contender, although I personally do not believe such a proposal would be simple when implemented or good by measures I deem important. Second, to markedly improve our present code, the reform must be principle-based with all changes tested against those principles. And all the participants need to agree to the principles. The key principles being discussed, in my estimation, are fairness, efficiency, and simplicity. Of those, I judge fairness and efficiency to be the most important. Simplicity is important, especially when it is not at war with the principles of fairness and efficiency.

Let me try to explain by example. Reality, political and otherwise, often puts principles like fairness and efficiency and simplicity at war with one another; at other times, two or more of these principles may reinforce one another. My professional experience in the tax arena was during 1985-86; during that tax reform effort, all deductible items on Schedule A were reviewed with an eye to simplifying them out of existence — simplifying. In fact, the elimination of the deduction for state sales tax seemed fair, as sales taxes were viewed as regressive — many of the states taxed food, medicine, and clothing.

The elimination of interest deductions other than on home mortgages was seen as encouraging consumption financed by debt over that financed by cash. This offended efficiency; they seemed reasonable, and we did adopt them. Both changes made the tax process simpler. We even tilted against the windmill of second-home mortgage deduction, to no avail. Some non- politicians argued that any mortgage subsidy violated efficiency, fairness, and simplicity. While this argument was a non-starter with all committee members, it was hoped that eliminating the deduction for the mortgages on second homes would find more favor supported by the principle of fairness and efficiency. Some people didn't have first homes, let alone second homes.

This proposal did not receive a single expression of support. We recognized political reality, which we will have to do going forward, and we moved on. Fairness versus simplicity was highlighted — for me, personally, highlighted for simplicity — by a proposal to eliminate the medical expense deduction.

The example that was most often cited and stuck in my mind was a family with a sick child, which had exhausted their health insurance benefits, if they had benefits. No one felt comfortable pulling that trigger in the name of any overarching principle. In some cases, fairness and simplicity were allies. Six million working poor — people who were below the federal established poverty level — were filing income tax forms and paying income tax — or in other words, the federal government was taking money from the taxpayers so it could be given back to them, minus the cost of handling, in the form of lunch subsidies for their children or student loans. Only one member raised an objection to the elimination of the tax burden on those below the poverty level, and his reasoning was that all people should pay federal taxes so they would oppose tax increases. Thank God that didn't prevail.

The relationship of simplicity and fairness and efficiency needs to be worked out on an almost one-to-one basis as we move forward in reform. Perhaps this is a good place to give a brief description of the terms "fairness," "efficiency," and "simplicity" as I am using them and why I deem them to be most important. Efficiency, which I believe to be the most important for economic reasons — not being an economist, but having an opinion, like most people — medicine has as its motto, "do no harm." Of course, this refers to the body of the patient.

In the world of tax policy, the body in question is our economic system. Our economic system is efficient because we largely allow markets to assign capital, and not the tax code. If, in fact, our economists could agree on which sectors of the economy warranted capital infusion at any point in time, and the politicians could enact their crystal ball ruminations into law with dispatch, we would still fail, because markets are in constant movement, and the tax committees are rarely in constant movement. No one can credibly argue that our tax code does not assign capital to a certain degree — the deduction for mortgage interest is a prime example — but the matter of degree is becoming more important by the day. Our economy was the engine, I believe, that brought the Soviet-led Communism to its knees. Of course, the politicians allocate all the capital in Communist countries. Our economy's ability to adapt to changing conditions and needs was a key element of that victory.

To move from the topic of Communism to managed economies, I can distinctly remember when the Japanese economy was going to dominate the world, and I was personally lectured, while serving in the first Bush White House, by the founder of Sony on the superior Japanese economic model. Japan essentially had a managed economy where businessmen and politicians collaborate on how capital would be allocated. Their fall was not dramatic as Communism, thank goodness, but we did sell them Rockefeller Center at a premium and bought it back at a discount.

(Laughter.)

MR. DIEFENDERFER: But our greatest challenge in the area of efficiency — and the reason tax reform needs to go forward now, in my belief — the greatest challenge to date is Communist China. They are introducing a few market elements into their economy, but the real advantage is the number of people they have and the government's ability — for the moment anyhow — to contain their expectations and aspirations. That will change, but until it does, our economy must operate at peak efficiency. Our tax code must be reviewed and amended from this point of view immediately if we are to improve our chances of maintaining economic superiority and all that goes with that.

The emerging challenge of China should be an organizing principle for tax reform that both parties should be able to unite behind. We know from experience that trade barriers, tariffs, and the like are not the answer; we must double our bet on what has made us great to date.

I'm going to interrupt here for a story which is a true story. One night we were waiting for some amendment on the Senate floor, which wasn't during tax reform but affected the jurisdiction of the Finance Committee. And the ranking member was present, Russell Long, a total genius, and my chairman, Bob Packwood. They were sitting together, waiting for this amendment to come up on some bill, and we were going to object to it. And as we were waiting — the bill being considered was a defense bill — Chairman Packwood leaned to Chairman Long and said, "You know, Russell," he said, "the way we run our defense program," he said, "I don't have a clue how the hell we ever win any wars." And Russ Long said, without missing a beat, said to him, "Bob, you got to remember," he said, "we're fighting other government-run armies. We just got to be better than they are."

(Laughter.)

MR. DIEFENDERFER: And it's the same way with — I believe, with the economy. Fairness — this is the most difficult of the concepts. Economic theory enters the picture, these theories are used to divide and exploit, I believe, but it does not seem that reasonable people can't come to consensus on the issue of how much progressivity will satisfy all the camps, the meaning of income, and double taxation of income. It will not be a clean-cut decision, but hopefully, a substantial majority of taxpayers will be satisfied. Congress is a substantial part of the fairness challenge, and this has already been mentioned.

In the 1986 tax reform, we eliminated many of the deductions and shelters that were used to reduce actual tax burdens and allowed a dramatic lowering of the rates to compensate. Once the rates were reduced and the shelters blown up, the rates were raised twice, and to be fair, once by a Republican president and a Democrat Congress, and once by a Democrat president and a Democrat Congress, and I participated, kicking and screaming, on the Bush tax increase, because I was in the White House at that time. These actions of Congress are not fair, and they are not perceived as being fair, and Congress is rightfully under suspicion when anybody mentions reform.

Finally, simplicity. Simplicity can substantially contribute to both efficiency and fairness. For example, in the areas of fairness, a more complex code usually means more ways for the rich to avoid paying taxes, and the average American knows this. I gave — I can't count the number of speeches — when we were doing tax reform, and when I would talk about tax shelters — not to the Washington crowd, but out in the hinterlands of average taxpayers — none of them had tax shelters, but they had all read about somebody who'd bought a tax shelter for bull sperm, or greyhound dogs, or something like that, and sheltered most or all of their income. It made them suspicious, it made them resentful, and complication does that.

So simplicity plays a major role in making people feel — it can play a major role in making people feel that their tax system is on the level. A simpler tax code should mean much less money is spent on administrative and compliance costs — and I have no expertise in this area whatsoever; I read as much as I could preparing for this, and the estimates are anywhere from $75 to $130 or $150 billion a year is spent on compliance. I don't know whether those numbers are right or wrong. We can't take them down to zero, but it does seem to me simplicity could help bring those numbers down. Bringing those numbers down does two things: It gives business more money to invest, and it gives families more money to invest in themselves.

Finally, I believe a rational process would address these issues in two distinct but linked parts. I believe reform on the individual side should be done separately with one level of simplicity addressed because of sort of average people. On the corporate side, it needs simplicity also, but generally there are more sophisticated people dealing with more sophisticated issues, so there — by delinking them in a sense, I think we could make more progress. Thank you, Chris.

MR. BERGIN: Thank you, Bill. Gene?

MR. STEUERLE: Thank you. It's an honor for me to be here with some of the best minds in tax policy and administration, and it's a special honor for me, actually, to be asked by Tax Analysts, which is one of my favorite of all organizations. I realize that may sound a little self-serving, since they have me write a column for them, but it's one of these organizations that I think puts integrity at the highest level — among its first principles, and I think if we had a lot more of that in society, we wouldn't be facing a lot of the problems we have, which go far beyond complexity in the tax system.

Ever the bridesmaid, simplification seems never to get the attention it deserves no matter which political party's in power, mainly because broader agendas always seem to be being pursued first. It would be a mistake, however, I think, to believe or even fault elected officials for pursuing these broader agendas; that's their job. Think about it: Government doesn't exist to simplify itself. It is entirely appropriate for tax policy to be the handmaiden to broader budgetary and economic policy. Whether the issue is something like rate reduction — if you think that makes the economy more efficient, as we did in 2001, or whether you think it's something like deficit reduction in 1993. Also simplification is merely one principle among many, and Bill did a very good job outlining what many of those were and also outlining what some of the tradeoffs were, and when you pursue things like equity or progressivity, it sometimes does add to complexity. Still, in pursuing these broader objectives, in recognizing that we have to balance principles, I think almost everyone would agree that we've gone overboard.

Simplicity has just simply given too little weight in the legislative process, and what we have is far more complex than we need to really meet almost any of these principles. Many items in the tax law add significant complexity without improving anything with regard to progressivity or horizontal equity or any other — efficiency — or other principle that you want to note. And almost no one would introduce, I think, many of the items we now have in the tax law if we were starting with a code from scratch.

As an economist I want to make clear that complexity adds waste, not merely cost. It's one thing if the government taxes you $100 and gives it to your neighbor — that involves a transfer of $100 from one person to the other — but that doesn't necessarily mean that there's at that point a net loss in the economy except to the extent we then distort behavior. But waste, which includes both the fact that we may change people's behavior but also the fact that we require so many people to spend so much time filling out tax returns, to deal with the IRS, to deal with the various aspects of government, that is a real cost to the economy. And I think among those costs and the ones we don't measure well — and I think Bill, again, outlined this as well as Nina — is I think taxpayers come to resent what is happening to them, and that resentment sort of has a cost which is sort of beyond just, say, the time value of their cost. It's not just that it — $20 an hour, they spend an extra hour filling out their tax returns; I think they come to resent what government does. And if you don't believe that, imagine — think of some bill you got from somebody that really wasn't even that high, but it was just outrageously wrong, and how offended you were and how much time you might even spend fighting it, even though it makes no economic sense.

So what do we do in this or why? What I'd like to do in my brief time is give some examples of how complexity arises. I want to talk a little bit about some process reforms. And I want to give some examples, again, in the system that I think we would all agree need to be fixed. So some very quick examples on how complexity arises. The first one I might give you as an example is the way that we have developed the taxation of children, our children's investment income — and I may be partly responsible for that. I was sitting at Treasury as coordinator of the Tax Reform Project, trying to get any idea I could from anybody on the staff, and Victor Thuronyi, who now works at the IMF, came to me and said, "You know, we've got this problem with all sorts of taxpayers trying to get their money into trusts because if they get their money into trusts, they can actually get their children's income — capital income taxed at a lower rate than if they're taxed as an individual." So we decided to examine the issue of trying to put forward to the adult taxpayer the income of the children. But we did it — we did propose this, and Congress did enact it, but the way we proposed it to Treasury was to say, well let's keep a substantial personal exemption. So we were willing to admit that maybe you'd want to tax all the — if you think you ought to tax households equally, maybe you want to tax all the child's income at the adult's tax rate, but we decided let's just not go there because that was going to add a lot of complexity. It got up to Congress, and it wasn't clear to me that it was the members of Congress as much as it was the Joint Committee or someone else who says, you know we can get a little bit more money if we back off of that exemption, and we really tax most of these kids' income at the adult's tax rate. And all of a sudden, you've got this enormously complex system, and people who've gotten into the system were actually people who mowed the lawn for too many people in the summer and could actually have to pay tax because they couldn't use their capital gains and count against their standard deduction — all this other stuff that resulted. And again, I'm not even sure that in this case, I would blame the legislators. It may have been the staff perhaps pressed to do it because it was a way to get a tiny bit more money in the system.

Another example of complexity — and all these examples, by the way, tell different stories — and so in some sense, it reveals how complex it is to simplify, because there's sort of no one particular problem at hand. A second story has to do with the refundable child credit. The refundable child credit came along in 1997 and then was doubled in 2001, mainly because the legislators wanted to indicate they were really doing something different. Now, between the Earned Income Credit — which we already had to give credits to low-income people - - and the dependent exemption — which we already had to give breaks for children to higher-income people — we already had enough mechanisms in place if we wanted to beef them up. And actually, I think there was a case — not everybody agrees with this — I think there was a case for expanding a child credit or a dependent exemption or making more adjustments for household size. Some people would disagree with that, but I think there was a legitimate case. But the legislators wanted to do it in a dramatic fashion, so they had to create something new, and then they had to deal with the fact that, well, they also wanted to make part of it refundable, and then it was going to be refundable, they ended up deciding, well, what about households that have three or more taxpayers? Let's create a special little credit here.

Then we went to 2001, and they said, well, you know, we can't really get rid of that special little credit when we double our other credit. And all of a sudden we got this very complex set of systems that now, by the way, is very hard to fix up, among other reasons, because each of them apply to a slightly different group of taxpayers. For instance, one applies to only kids under 17, and one applies to kids who include children in college, and so if you decide you're going to integrate the two, you're either going to spend a lot more money because you're going to expand the one to include kids in college, or you're going to basically cut the other one back to maybe kids who aren't in college and so all of a sudden, taxpayers will be aroused. It was totally, totally unnecessary, I think, from all standards.

A third example that people have raised is the alternative minimum tax, which by the way, Treasury did not propose in its tax reform proposals to Congress in '86. And again, Congress felt that it had to add the item. Now, among the biggest items there, by the way, is — as Nina mentions — is the dependent exemption. I'm still not sure why the personal exemption just got treated as a tax shelter other than it was a nice revenue raiser. There may have been — and I'll blame my profession a little bit — there may have been a few — there are a few economists who like to argue that children are consumption, and so you shouldn't probably have any dependent exemption at all in the first place, and so that probably helps sell some people that — well, maybe we can include that one, too, but you know — who knows why we got children in there.

And then we have other items, such as state and local private purpose bonds, which we argued should basically either be capped or eliminated out of Treasury, and Congress wasn't sure it wanted to do that, so it decided, well, let's put private purpose bonds in an alternative minimum tax and deal with it there and it got more — it just got more and more complex.

And if we go to the current period, you know, we could give all sorts of examples. The great example to me is this energy bill. I've yet to see anybody put out some analysis saying what we're getting either in terms of efficiency or equity or simplicity or anything in terms of this energy bill. It's a bill, as have been many of the bills lately, basically designed by the lobbyists with some slight staff comments on how to make sure the costs don't rise to several hundred billion dollars. But beyond that, God only knows what's in this bill or how it is — and actually, it's interesting enough that all the time we spend worrying about, you know, fixing up the income tax and all these other things, how these things like these energy bills just go through with, you know, maybe Tax Analysts trying to do an article or two here or there — I'm not even sure you guys have got a bunch of energy studies that you've put out. You know, who knows how we deal with this.

Well, I'd like to switch to another topic now, which is, has the world gotten a little worse in the terms of complexity, and I'd argue yes, it has, and I can't prove it, but I'll point to four areas where I think there has been a historical tendency to add to complexity. One of them is that, I think, as the tax and transfer systems have started to merge, it's really changed the world. We used to debate complexity of the tax law, mainly with respect to investment and business types of issues.

Basically, if you look at the social spending, or the spending outside of just raising revenues, in the tax code, it's mainly on the social side now, it's more middle-class relief, it's less, really, high-income and business relief, even though there's an extraordinary amount of the latter. But if you actually look at the numbers, it's become more and more this middle-class social relief, and I think one reason for that is the tax system has more and more gotten into the transfer area, but the transfer area has gotten more and more into the tax area. It's not clear to me that our classic division of government, even, between taxes and spending works so well. It probably made sense to think about taxes separately when mainly they were funding highways and defense and justice, but once they're mainly funding Social Security and Medicare and everything else, it's not clear, you know - - it's not clear that you need — you want to even think about taxes and transfers separately. And I don't think, in a broader sense as a society, we've dealt with this issue, and I think part of what's going on, in fact, is that what's happening is — what we're seeing in the tax code is exactly what we're seeing on the spending side of the budget. That is, if you think about all the ways new items and new provisions get added there, that sort of milieu, or that ethic, has sort of now carried over to the tax system, and it's not clear to me that the solution is merely going to be a nice clean tax policy solution over here.

So we'll take the Earned Income Credit out of the tax code, but we'll still have IRS send W-2s to HHS so they can figure out who really has wages and then HHS will some way or another deal with all these things. I don't know that that's going to be a simpler world, even though I know IRS might like it better. A second issue — Bill Niskanen can comment on — more on this than I could — but I do think lobbyist activities have increased. The recent numbers I saw — I think it was in Tax Notes, wasn't it, that they've doubled? The lobbyist expenditures have doubled in the last five years. Bill has written several articles on why there are so few lobbyists. Given the extraordinary amount of money at stake, but the fact that they are doubling — I do think that they've become more important, and I would also argue if — again, if you look at bills that have been designed recently, more and more they're designed by the lobbyists. And by the way, I include — by the lobbyists — for what they think to be principled items, whether it's, you know, supply-side lobbyists or people who are arguing for more progressive tax system. The lobby side of the effort has really started to dominate more and more, and, you know, the design of a tax bill by a joint committee or Treasury or some other place has really waned, and sometimes they don't even get a seat at that table. I do think that's become worse.

One of the other items that I think has increased is this extraordinary need of legislators to legislate. You know, it's almost like — you know, if we don't do at least 20 tax bills a year, of some sort or another, then we're not really doing our job. And maybe we've gotten more efficient at it at some level and that's why we're doing it, but I do think that's changed over time. So, given this new world, given the complexity we have, whether it's a worse world or not, what do we do about that?

And I think you've got to start turning to some of the things that Nina, and I think Bill in more detail, got to, and that is I think we really need to start having to think a lot about the process itself. Bill talked about trying to figure out some way to put principles back as the first consideration when we do reforms. I think we could think very seriously about ways we could force the process to do this a little more. I think, for instance, we could — if Congress would be willing to do it — we could adopt rules that required some period of time where the complexity of a bill was presented up front and given more attention. There are rules that have come out of the — was it '97, '98 legislation? Whatever the IRS reform legislation was that sort of requires the joint committee to do a complexity analysis, but it sort of comes along at the last minute and it's almost worthless. I think if we're going to do the type of process reform that Bill and others are talking about, I think we've got to figure out ways to make sure the process deals with them. I also fault — and I've done this for a number of years — I also fault the IRS significantly here. I think the IRS's claim that it doesn't do policy, it does administration, is a little bit of a sham. In a lot of areas —

MS. OLSON: What?

MR. STEUERLE: I didn't say — Nina. I didn't say —

MR. BERGIN: Well, we'll have no problem with the next phase of this meeting.

MR. STEUERLE: I wasn't — talking about the taxpayer advocate. —

MR. BERGIN: Needless to say. Well, there's a couple of former commissioners here, too.

MR. STEUERLE: But there's a Government Performance and Result Act of 1993, and I've always thought that a commissioner with a little bit of — sort of a nerve behind him or her — have we ever had a her?

SPEAKER: Yes.

MR. BERGIN: Yes.

MR. STEUERLE: Oh, that's right, we've had two, that's right — would actually take something like this and say, you know, the law requires me, under DIPRA, to report on the efficiency and administrability of every major provision of the tax system. And in fact, OMB and others have actually tried to push IRS and Treasury to do this, and they keep saying, well, we can't do it because we don't have the data, so therefore we're not going to be able to do it. So we can't really report — unless we're really pressured to report — on the efficiency of the Earned Income Credit. Well, Congress is finally up in arms, and maybe we'll try it, we'll finally set up this extra division to look at the Earned Income Credit. But we're not going to do anything about the charitable deduction or the charitable sector. Well, whoops, all of a sudden Congress is worried about it now; you know, with six months to prepare, we'll see what data's lying around, and maybe we'll finally get some data on in-kind benefits, something we didn't even do a statistical sample on for 20 years. Maybe we'll look at that.

To me, the IRS has a responsibility in every one of these program areas to at least have somebody in charge of reporting on it, even if it's some little amount, even if it's a half-staff person or one staff person says, these are the data I have, or reporting that we don't even know — we have no idea what's going on with empowerment zones. We have no idea whether they're administrable, we have no idea who's getting the benefits, and that's our report to Congress. That report should go out to Congress because it says something about the program, and I think something that's done and then announced like that could actually have a difference. And by the way, a number of their studies that they do on audits actually do reveal error rates, and those error rates could be highlighted and bigger reports — and actually, Nina's tried to do this, but — here are the five biggest error rates in IRS this year, here's what they are. I think it sort of forces attention to things, and that's another way. So we could talk more about this, but I just want to throw out, I think there are process ways that you could force more attention to simplification issues. And one that I tried to push on the IRS for ages — and it would require a substantial amount of resources — but when one of these bills comes near to being final — is to actually send Congress a copy of the tax forms that they've just created by enacting the bill. It would require a substantial amount of resources very quickly at a point in time — we did that briefly when I was deputy assistant secretary for the Reagan catastrophic health bill that led to Rostenkowski's being —- and everything, and they actually did amend the bill because of those tax forms that we got put forward; they didn't cure it, but they actually did make some improvements.

So again, I'd like to talk about — with you about some of these possible process reforms. So the last thing I have — because I think I've taken up my time — I'm just going to list quickly some of the items that I say fit into this category, items that I think nobody would argue for under any other principle; they're just problems. One of them is this phase-out after phase-out after phase-out of different things in the tax system that create these hidden tax rates, and Chris, you started off by mentioning one of them right at the start, which is this extra 33 percent bracket that then became the phase-out of the dependent exemption, combined with the phase-out of the itemized deductions, which is not really a phase-out of the itemized deductions for most people; it's really a tax rate increase. And I think we could make that explicit. I don't know, by the way, that either the Democrats are willing to do it because in many cases, these phase-outs do make the system more progressive. And I don't know if the Republicans are willing to do it because they often like to argue that what the statutory rate looks like is what they care about.

So, you know, I've yet to hear either side say they'd be willing to trade the AMT for an equivalent rate structure for this — for the same people in the same brackets. Another one is the pension and saving incentives for which we have, you know, got this chart that makes the Clinton health care plan look simple. I mean, it's hundreds of different types of pension and saving incentives that are just — they said they'd be in about 12 or 15 categories, but it's just an enormously complex area. And that complexity it's quite clear now, from the analysis, from behavioral economics, that complexity is clearly decreasing the amount of saving we're actually doing.

Another one is the tax treatment of dependent children; I mentioned that one. All of the capital gains tax rates are absolutely silly. I don't know why we can't come up with one rule — I've said this — I've actually written this in Tax Notes, and it actually did lead to some activity and some legislation, but I think we can simplify it — we should have one rule for — if you have a mutual fund — we could have a lot of areas like that where you could have a 1099 that would actually report the income you have and not leave options up to people. We have all the multiple educational tax breaks that probably don't belong in the tax system in the first place, but could be made into one. I think there's a lot we could do in the charitable area. I would actually push for more 1099 reporting from charities to the IRS, something I think the charities would not like, but I think the IRS does not administer a good deal of this area.

We can simplify the child credits and dependent exemptions and fold them into one, and also the estimated tax rules are still a bit archaic, and we could make life a lot simpler for taxpayers with an estimated tax rule and not worry about that being a source of revenue raising. So these are just among the items — let's say — it doesn't matter if you're for a consumption tax or an income tax or anything else. These are items that seem to me that should be on the agenda and hopefully will be listed by the tax reform commission. And I've talked long enough, so I'm going to stop.

MR. BERGIN: That's great. Thank you, Gene. I've got already three simple questions that we should send to politicians when they're thinking about changing the tax law. Politicians to themselves: Why do we care? Politicians to the lobbyists: Please tell me how much this proposal you favor will increase your tax bill? And politicians to the IRS: Can you make a form for this? So I'm going to open this up now, the three great presentations, and — I saw Mr. Alexander's hand first. I'm not going to call anybody by name, okay? I'm going to try real hard not to do that, so you state your name.

MR. ALEXANDER: I'm Don Alexander, former tax collector, and, Dief, I agree with nearly everything you've said, except your answer to the question. The answer to the question was, I think, no. Must a good tax code be a simple one? I think the answer ought to be yes because the IRS has to administer this thing to try to make it work, to try to be fair to people, those who pay their taxes, by making other people pay them, and to try to make our tax system, which is unbelievably complicated — as has been spelled out this morning very well — one that people want to comply with, and can comply with.

What we're trying to do now in our tax system is put at least half, if not more, of our discretionary expenditures into the Internal Revenue Code, but guess what? They're administered by the IRS. Poor little IRS has to try to do it. Our distinguished taxpayer advocate is doing great job in being somewhat of a gadfly, I would suggest.

MS. OLSON: I don't do tax policy.

MR. ALEXANDER: Let the record clear —

DR. STEUERLE: Socrates defined himself as a gadfly, and you know what happened to him.

(Laughter.)

MR. ALEXANDER: Yeah, well, hang in there, Nina. But the IRS simply cannot handle half of our discretionary expenditures in our budget, $900 billion worth of annual payments, benefits, that have nothing to do with a sound tax system, and it's all my fault. It's all my fault because the Earned Income Tax Credit was a brainchild of a guy named Milton Friedman — talk about brilliant people, all right? He wanted to have something called a negative income tax. I thought it was a lousy idea. He came down from Chicago, and we had a debate in Secretary Schultz's little office, and the only thing I succeeded in — because we were sitting on high chairs and Friedman was even shorter than I was — I succeeded in making him move his legs at an extraordinary fast pace, like that, which he had a habit of doing when he was upset at something, and clearly he was upset at my trying to say that the Earned Income Tax Credit belonged with HEW at that point, did not belong with IRS, that the IRS could not satisfactorily administer it, and it would create huge problems both in itself and in showing that baubles — maybe it's a — it's a great idea.

Gee, moving from welfare to work — that's good — but let's have the proper department of our government administer whatever thing we want to do to encourage people to do what is in their interest and what is in the national interest. Well, lost that one because I think President Nixon, of all people, wanted to do what has been done many times later, and that is instead of giving authority to a department to pay out money to people who deserve it — because they have numerous children and can't afford them and other good things like that — handle it through the Internal Revenue Code because when Congress thinks they get some benefit out of it, as Gene pointed out, another Congress wants to do it. An even worse trend is coming in now, and that is that politicians are putting their names in the Internal Revenue Code. Just think how much the Code would be shortened if we could remove all those names.

MR. BERGIN: That's a good point. Like getting a street named after you.

MR. ALEXANDER: It ought to be named after you, sir. So we now have IRS attempting — and not succeeding, I think, to administer these wonderful things that are embodied largely in credits. Why? Because a deduction doesn't give you enough. A deduction simply removes something from the tax you pay; a credit gives you something from the federal government. And if it is a credit, then, one argues, it has to be a refundable credit because those who don't pay taxes are deprived of that great benefit that we've just decided should go to these people who otherwise meet the test.

If it's a refundable credit, it has no business in the tax system. It belongs somewhere else where the department of our government — HHS, HUD, whatever, Energy — has the right and the obligation to administer it and the right and the obligation to see whether this thing is worthwhile renewing and whether changes ought to be made in it. It would be a good idea, Gene, if the IRS were to report more — I'm not sure anybody would pay any attention to those reports, but it's nice to have them — about new baubles that are constantly being introduced, or repeat baubles, like the extra child credit that another — that one Congress wanted to get credit for so they wouldn't do what they should have done, which is simply improve or expand what another Congress had done before it.

But if you had genuine simplification — and I dearly hope we can have it — to me that means making the tax law administrable, making the tax law a way of raising the revenue necessary for our country's needs and leaving to the departments that have the obligation to review other goals of our country, the duty to review those goals, and the authority to review them. Now, people love to have authority without responsibility — it's great, bureaucrat's dream, have all the authority and none of the responsibility — so the IRS gets the responsibility and others can claim the authority. Apart from that, I don't have strong feelings.

MR. BERGIN: You do make a great point, though, Don. I mean, when they start naming tax laws after themselves, these politicians, we are definitely not on the road to simplicity.

MR. ALEXANDER: No, we're not.

MR. BERGIN: It crosses some kind of line. I think that's a great metaphor, sir.

MR. NISKANEN: Bill Niskanen from the Cato Institute. The implicit assumption of those who favor a simpler tax code is that the allocative and distributional objectives of the government are better addressed on the spending side of the budget than in the tax code. I share that view, but the case for this position is nowhere near as evident nor as strong as one might think. Let's look first at the set of conditions that affect the nature of political decisions.

One is transparency. In general I think you get more detailed information about the level of resources and where they are going on the spending side of the budget than on the tax side of the budget, but we do have now a complex listing of tax expenditures — I think that that substantially offsets that, even though that list, I think, is biased by an assumption of what the proper base is, and it's incomplete in a number of ways. But transparency is one of those conditions that one might think is important in deciding whether the issue ought to be addressed at the spending side or the tax side.

The second is the default rule. For most of our tax code, the provisions continue unless they're repealed. For at least most of the spending side of the budget at the moment, although not 10 years from now, the spending is repealed unless continued. And so I think that we're better off with an automatic sunset provision on these issues than with the rule that it's going to continue forever unless repealed. Now, that is not — again, that doesn't necessarily apply everywhere because it may be that you want a provision that leads people to make investment decisions over a period of time, for example, that contribute to economic growth, and you don't want those to be subject to periodic — or at least short period — political action. So the default rule, I think, is quite important. Now, we're eroding the difference in those default rules, of course, because now we have sunset provisions on important elements of the tax code — the tax legislation in the Bush Administration, and a lot of an increasing share of the spending side of the budget continues unless repealed because it is — entitlement programs are not the discretionary part of it.

There's another consideration that I think we ought to pay attention to, and that is the implication of changing from a tax code that is complex with high rates to a tax code that is simple with low rates for the same revenue collected, which is the characteristic direction of tax reform measures; by broadening the base, you get the rates down. There's a lot of reason to believe that that move invites an increase in the size of government because it reduces the cost of government spending. By reducing the misallocative effects of the tax code, you invite an increase in the size of government. And I think it is not an accident that the 1986 Tax Reform was followed in each of the two subsequent administrations by tax increases without changing the base. In other words, a broad-based tax system invites higher tax rates over time. So we have to be careful about the effects of that.

My view on this matter — on this third matter is that if we ever achieve a really broad-based tax system, we ought to accompany that with a supermajority rule on rates so that you don't have the effect of — that we're having now, that we've have had in the past 20 years from — of a broad base inviting higher tax rates. So it isn't absolutely clear in many cases that the tax code is less transparent than the budget side. I think the budget side is increasingly less transparent in part because of all the entitlement programs. The default rule — I think the difference between the tax code and the spending side of the budget is not as much under the default rule as it used to be because of the particular provisions of the Bush tax cuts and because of the rapid increase in entitlements spending. And on this third effect, I think we have to take seriously that our political decisions will not be independent of whether the tax code is perceived to be having lower social costs — misallocative costs — than it does now in the sense that I think our big decisions on the relative size of government are dependent upon that.

MR. COHEN: I'll only comment on that last thing by saying that when I arrived at the Internal Revenue Service in 1952, the —

MR. BERGIN: And you are who, sir?

MR. COHEN: I am Sheldon Cohen, and I used to do what Don did. When I arrived at the Internal Revenue Service in 1952, the top rate for individuals was 91 percent, and the top rate for corporations was 80 percent. I'll let the rest of that stand. Simplicity, I'm afraid, has a constituency of one: The commissioner of Internal Revenue, whoever he or she might be at the time is concerned with administrability, as Don indicates, and no one else cares — no one else cares enough. They care in the public sense, but they don't care in their viscera, so that if it comes to complication or political feasibility, they go for political feasibility.

We've not had a chairman of Ways and Means or Finance for a number of years — a long number of years — who cared. Wilbur Mills used to have Larry Woodworth and me in his office the day before he was going to take a bill up, and he would ask all these questions about what do you do in this and what do you do in that and what complication does this — you know, what do you recommend. So the commissioner was a player, never out loud, but in the back room. I would think it would be proper now, that is, the commissioner ought not be a player. He's not a tax policy maker, but he ought to have a voice in feasibility: Is this workable, what kinds of complications will it create.

Choices — Nina said you ought to have few choices; I'd say you ought to have almost none. Choices are the death knell of simplicity. Michael Graetz has this wonderful idea that he's going to tax rich people — people over $100,000 — on an income tax and everybody else on a VAT. The problem is, that that's not a standard set pool; that is a moving craps game. Suppose I hit the lottery this week — I made $50,000 — never made more than $50,000 in my life, this year I hit the lottery. How do I know that I file a tax return? Where is the discipline?

You see, you've got to worry about training. You train those taxpayers — hopefully — so that they have an expectation and they know how it works. And if they aren't trained, they won't know how it works. So administrability is a problem, and we don't take it into account. The Treasury really takes it into account; when it becomes a choice of administrability or political feasibility, they adopt political feasibility each time, and they won't — I mean, I used to have an awful time with Stanley. I could — surely I could make him, on occasion, stand up for me, but most of the time, the Treasury will go with the political feasibility. That's — I mean, that — they've got the broad economic agenda and this little provision that's going to cause the commissioner of Internal Revenue nightmares is just a small part of their pie; it may be a large part of his pie. And I agree. The commissioner has to be a player; the question is where and how, and it's as much he has to be a player at the Treasury. Stanley and I had lunch once a week to talk about issues that our staffs were fighting about to try to iron them out. I doubt if that happens much anymore.

MR. BERGIN: Nina.

MS. OLSON: Just to that point, I think — what's interesting to me about the Office of the Taxpayer Advocate and the position in the National Taxpayer is that Congress has, after the fact, you know, specifically said, comment on and make legislative proposals about how we've messed things up. You know, that they've — they've given that office that authority rather than — if you really wanted to avoid complexity in the code, you would say — you would create the mechanism whereby the commissioner and others dealing with administrability are coming in and saying to the tax-writing committees, you want to do this, this is what's going to happen. That's an odd thing because you know how long it takes to get corrections to things that are already in the revenue stream and things like that; it's almost impossible. So your talking about that, that's just very interesting to me.

The only other point I'd make about some of the talking about the research, I've obviously been a really big backer of additional research, and my office right now is doing a number of studies because we've just sort of said, okay, well, we're going to do the research, you know. On the downstream consequences of certain provisions — because even — when you talk about, you know, something being complex, if you don't take into consideration the entire administration of that provision — which would include examination, appeals, litigation, collection, you know, the education provisions that we — you know, that we — the taxpayer service side, the outreach and education — if you don't have an understanding of all those costs, you really can't say what it takes to administer a proposal. And if you can come up with some models that actually look at that, working off of some existing proposals, then you can maybe get that to Congress and say, look, you've been down that road before; here's what we know. We're not anywhere near that, but we are taking particular provisions and particular existing provisions and mapping out what it does to taxpayers and, conversely, the IRS.

MR. BERGIN: And they need to listen to you, right?

MS. OLSON: Well, I'm a gadfly.

SPEAKER: Do you do tax policy?

(Laughter.)

MR. BERGIN: Go ahead, David.

MR. BRUNORI: David Brunori, Tax Analysts. I wanted to go back to something the gentleman from Cato said, which stunned me because it runs counter to everything I ever learned about tax policy, and that — he called into question the idea of a broad base and low rates, which I have always thought — and everywhere I've read going back to, I don't know, Adam Smith — not that I read a lot of Adam Smith, but — say that the hallmark of good tax policy is broadening the base and lowering the rates in terms of efficiency, you know, and lacks a little bit in fairness at times, I suppose, but certainly in terms of simplicity, you know, the broader the base, the lower the rates, the more simple everything is, which — so I was wondering if — I don't know if we have lots of very seasoned tax policy experts in this room, but I was wondering if anybody could comment on that statement because I — if it's not true, then I need to be rethinking my tax policy — sound — principles of sound tax policy.

MR. ROSENTHAL: Steve Rosenthal with Miller & Chevalier. Broad base, low rates — I think we've skipped over the definitional steps of what is simplicity because you could think of it — we've blurred the notion here, talked about administration of tax laws, we've talked about transparency, and then we've also talked about implicitly computational problems, like rates, phase-outs, AMT. And I would suggest that the last category — the computational issues — are actually the rare glimmer of hope because it's quite easy from a computational standpoint to calculate your taxes.

I taught my son and daughter to file their first tax return last April — they both have income from part-time jobs, high-school teenagers — and I walked them over to TurboTax. Put in their names, we put in their numbers — you know, the funniest thing was after we had finished, and the correct number popped out, my son asked me, well, what does this mean. And I couldn't explain it. I'm a tax lawyer, having done my own tax returns from the time I was, you know, a teenager myself, and always could understand where we were headed when I was filling out a box on a return and the like. But a couple years back, I stopped understanding the numbers TurboTax gave me. Last year, I had no AMT. The year before, I did. I can't explain that to you, but from a computational standpoint, I'm fairly confident I have the right number on my tax return, and I also finished my taxes quite quickly.

So when we want to think about simplicity and complexity, I think we have to focus in part on what the standard is that we're trying to apply. And as I say, the rare glimmer of hope, if you're thinking about computational complexity or simplicity, is that the one thing that has developed in huge leaps in the 20 years since we last looked at fundamental tax reform is the strength and power of information technology. And if you look to what California's been doing with some of its return-free filing — Professor Bankman testified that credit card companies can calculate your credit card liability and send you a bill, and you can check against it, why can't the Internal Revenue system as well — the Service provide you a calculation of your taxes and send you a bill and do the information matching and computations necessary and even knock out TurboTax from business? So at least at that level — of course there's trade-offs, but at that level, you have a realm of simplicity.

So to answer your question most directly, I don't think broad-base low rates matter, particularly from a computational standpoint, but there may be other aspects of simplicity for which they do matter.

MR. COHEN: This is Sheldon Cohen again. The Service tried a number of years ago that — by the way, you can send in a 1040A below whatever the number is, and the Service will compute it for you; you don't have to compute it yourself.

MR. ROSENTHAL: Right.

MR. COHEN: Nobody does because they don't trust the Service. An illustration: President Johnson once told me — he said, design a simple tax form; that's your first — I mean, that's the day I'm appointed. Yes, sir. So I went out and I hired a private consulting group that did forms for corporations. I said, our people need help. So they send in a committee, and they started designing the form, and they would come to me every once in a while, and they would say do this or do that, and I would say, can't, the statute doesn't allow you to this or do that. So that's the first problem.

Second is, we finally designed a form called the 1040X, which was a questionnaire; you made no calculations. You filled it out like you did an informational form. Do you have dividends? If yes, fill in the amounts, if no, skip to the next section. And that was seven or eight pages long for a normal 1040A. But no computation — not a single computation had to be made. We tested it on 200 secretaries in the national office, reasonably intelligent people, but no technical skills. And about 75 percent of them loved it — and they used last year's tax information so we had real information. We tried it on a larger sample, we got similar results. And the staff said to me, okay, let's try it. I said, wait a minute, are we going to send this thing to 100 million people? No, we've got to pick a sample. So we picked a sample, 1 percent of the people who filed in the Southwest Service Center — which happened to be the best service center we had at the time because they had to process them — and we sent it out to the 1 percent — it was about — I think it was 170,000 people. And we got it back, and they all — by the way, we had to send them a form because we realized that some people would want to compute it, so we sent them a form — here's a separate schedule. If you want to fill it in, you can fill — 75 or 80 percent or more filled in; that is, they wanted to — because they didn't trust the computer to do it for them.

MR. BERGIN: Yeah, but that's changed, though I think you make a great point. If — and was it 6251, the AMT calculation form, which is the computational equivalent of a root canal? If you fill out that form — and I do — I did once, just for fun — I'm a little weird that way — you ask yourself at the end of the process, why don't I just pick up my pitchfork and head for Capitol Hill? I mean, this is nuts. But if you push a button — if you have the fairest system in the world, so to speak, and the Internal Revenue Service provided you with a button — and you trusted the Internal Revenue Service — do you solve the simplicity problem? Is technology the answer? It's a great point.

MR. BRUNORI: The answer is no.

MR. BERGIN: Why?

MR. BRUNORI: I hate to be disagreeing with my boss and all, but — no, I'll tell you why. The broad base — I'll tell you what, it goes back to my question of does broad base matter — broad base, low rates matter. If you look at the income tax right now — and we have how many provisions dealing with higher education credits, deductions, exemptions, et cetera, et cetera, right? Seven, eight, ten, twenty — I don't know, a hundred thou— who knows? There's a complexity there that costs the country money in terms of figuring that out.

Now, you may be able to go and plug this stuff into TurboTax in April at the end of the day, but right now, the vast majority of people in this room have no idea what the higher educational benefits through the tax system are. It is just far too complex because we have — we're narrowing the base in that sense. If you look at the state and local sales tax or the sales tax, the fact that when you go in Virginia and buy shampoo, you buy — if you buy Head & Shoulders shampoo, it is exempt from tax, but if you buy Prell shampoo, it is taxable — it's true in Virginia, not true in Maryland by the way. There are complexities in terms of not broadening the base that I think that there's all kinds of costs involved and that complicate the system. So I think it matters not so much when I'm filling out my TurboTax — because I also use TurboTax; I plug in the numbers, and I find it painless —

MR. BERGIN: I'm sorry. Joe — I promised Joe.

MR. MINARIK: And you keep your promises.

MR. BERGIN: I try.

MR. MINARIK: Joe Minarik from CED. I guess a definition of a good discussion is that there are a lot of threads of things going on, and therefore, I feel compelled to say a lot of things. So you can all take a —

(Laughter.)

MR. MINARIK: In answer to the question about Bill Niskanen's point earlier, I think — I understand that it's not just — I think that for Bill, a good tax is a bad tax, which is, of course, a very old rule. The reason being that — Bill, I think if I understand where you're coming from, you don't like spending, and what will make it easier for the federal government to finance spending is therefore not to be favored. And if a tax that reduces the apparent inefficiency of raising revenue induces the federal government to raise more revenue and spend it, that's a bad thing. So it depends on where you're coming from, whether you think, number one, if we have an awful budget deficit that is running us in a very dangerous direction, and therefore we've go to raise more revenue; or if you believe there are useful things for the federal government to do, a broad-based, low-rates — hey that's the way to go.

Someone from Bill's perspective would feel differently for that reason, so I think I've got that explained, but in any event — I do want to say that Don Alexander has just given me two more reasons why he is one of my very favorite people on this earth. One is he argued with Milton Friedman, and the second is that if anybody ever comes to me and talks about how complex the tax system is and how bad that is, I can say that Don Alexander said it's his problem and his fault. Don said that you ought to have programs assigned to an appropriate government agency that ought to have responsibility, and actually, I think Eric Toder suggested that solution a while ago. If I recall it, he was the one who did it, which was the proposal that we create a new Cabinet- level department of refundable tax credits where we could put all these things, and that would be an appropriate department.

I do want to make one serious comment, and that is, it seems to me that we could segment this problem of complexity possibly to some useful effect, and I think that Gene Steuerle gave us a possible key to do that. It seems to me that we might have really three different problems aligning to some degree with the affluence of taxpayers. For the bulk of middle-income taxpayers who earn their income from employment, we have a finite set of self-inflicted problems, which I think essentially corresponds to the close of Gene's presentation, talking about multiple choices for retirement savings, multiple choices for sheltering — for achieving tax savings from the cost of education, and so on and so forth. If we would take care of that finite list of self-inflicted problems, probably for the vast majority of taxpayers, we would be in a pretty good place in terms of complexity.

We have the low-income problem where there is, as Don pointed out, the natural difficulty of trying to get cash to people. We've decided to do it through the tax system to get cash to people with low incomes who have the toughest jobs in the world, they're trying to be parents, they're working all the time, they don't make much money. If we decide to do it through the tax system, we've created some problems in terms of definitions of dependency and so forth — I thought Nina had some good ideas in that area. It's probably never going to be easy, but it's a reasonable thing to do.

Then when you get to the upper end of the affluence scale, we have the self-employed, who have all the problems of accounting if they have to satisfy their bankers, magnified to some degree because they have to take care of the tax responsibilities that come from the definition of income, the problems of hiding expenses, and so forth. And then we have extremely affluent people with a lot of income from capital who are trying to move it to the Cayman Islands so they don't have to pay U.S. income taxes on it, and also trying to invest in greyhounds and bull sperm and whatever else. But — correct me if I'm wrong, but I think that those are our relatively identifiable problems, and for the vast bulk of the American people, we did it to ourselves. And we can fix it.

MR. BERGIN: Bill, I've got to let you get back in here.

MR. NISKANEN: Let me be clear: I favor a tax system with a broad base and low rates, accompanied by a supermajority rule on rates. I mentioned this whole issue to make sure that you were aware that the decisions on the relative size of government are not independent of the nature of the tax code. My own preferences, again, are for broad base with low rates, accompanied by a supermajority rule on rates.

MR. LOBEL: I'm Martin Lobel, Lobel, Novins & Lamont in Washington. Two things occurred to me as I'm listening to this discussion. Number one, everybody seems to have assumed, except for Don, that we ought to use the tax code for anything — for social good and not just simply as a revenue raiser. I think if we go back to basics and say the tax code is only to raise revenue; social welfare ought to be someplace else — as Don suggests — we would have a much better system and less likely to be abused by special interests. Now, secondly, and even more interesting, is all the discussion I've heard today is about individuals, and nothing about corporations. And there is where the real abuses and complexities are.

If you take a look at the effective tax rate of the American corporation, if they pay anywhere close to what they're supposed to under the code, they ought to fire their accountants. The most dramatic thing was in Tax Notes a couple of weeks ago when someone revealed that the after-tax disparity between rich and poor was greater after taxes than before. That's devastating. The middle class has been stagnating for years, it's faced with these incredibly complex things that by and large don't work very well, and the corporations are escaping taxes. Nobody seems to want to go back to the basics. You want us to tax corporations? Under Sarbanes-Oxley, the CEO is required to certify that those figures are accurate — why don't we just eliminate that and say, all right, that's your taxable income and apply the tax code to that? Very simple, easily enforced, you don't have — and if you combine that with unitary tax accounting, you wouldn't have to care about all these complex provisions, you wouldn't have to carry two and three and four sets of books; corporations would be free from all these onerous accounting standards, they could carry on their business.

The idea that we can control the economy by using the tax code really is an open invitation — and I hate to be a tad cynical about this — to the lobbyists and members of the Ways and Means and Finance Committees who seem to be intrigued by the tax code, not because of the intellectual premises that they have to do, but by the amount of money they can raise or the amount of subsidies they can get. We really have to go back to fundamentals. You know, most of these ideas are excellent ideas, but they don't get to the fundamental problem; the incentives are all wrong, and until you change your incentives, you're never going to have a simple tax code because right now, any businessman who doesn't try to buy a tax subsidy is a fool. His rate of return is going to — if I'm getting a tax subsidy — is going to far exceed anything he can get in the marketplace for most businesses. So if you're going to talk about tax simplification, you really have to go back to basics and say, our purpose in a tax code is to raise revenue; it's not to do social welfare. I realize that makes me sound somewhat to the right of Bill Niskanen, but I think it's the right position.

MR. BERGIN: Bill.

MR. DIEFENDERFER: A couple of quick thoughts. I think that's theoretically a very good suggestion, but I would guess, if you looked at the income tax code when it was enacted compared to today, you'd say it was relatively simple. Over time, government did what Bill Niskanen said, they wanted to take more and more — and maybe it should have taken — I'm not arguing whether it should or shouldn't have. That caused a natural reaction among people to protect themselves, and the code got more complex. Charities came in, they wanted X, and somebody else wanted to invest in tax credits, and the oil people wanted X, Y, or Z. It gets more and more complex. You combined that with the issue of transfer — raising and transferring within the code, transferring income — another influence for that is the jurisdiction of committees.

If you go to the Finance Committee or the Ways and Means Committee and say, boy, we have a neat idea, like Milton Friedman said, you can be the taxer, and you're going to control the spending, and it's going to be all done at the IRS and is a Treasury problem that you control. It's like, stop me before I kill again; they'll do any of those. They're just people. The incentives are there for them to want to do that sort of thing, and if you believe that transferring activities of those — that such from the Treasury Department to other agencies is a panacea — or at least a step up — think again. The Treasury Department people are so much better than other companies; they are like companies — departments. My first job in Washington was on the Education and Labor Committee. You want to deal with a place that's out of control, move to the Department of Education. I'll give you a specific example.

In the area of higher education, there's something called 9-1/2 percent tax- free bonds which allows states and others to borrow tax-free, lend to students, and get 9-1/2 percent in the market where the rate's, like, 2 or 3 or 4 percent max. The Education and Labor Committee, they're never going to deal with that. If that were in the responsibility of the Treasury Department, they might not get it done, but you would have responsible comment and an attempt for responsible action. So I wouldn't be the first person to want to move things out of Treasury because of the competence — they're burdened, I understand, IRS, but they're much more competent than, say, the Department of Education.

MR. BERGIN: We'll go Nina and then Gene. Just move in the circle.

MS. OLSON: I guess I want to respond to Don's point about the EITC and your points. You know, I have gone back and forth over the years, depending on how irritated I am with the way the IRS is administering the EITC, about keeping it in the IRS or moving it out. And I think where I've really come down is something that actually Gene had made a point to me about in a private conversation, which was, you know, where — when you have a provision that is income- based — what agency in the government has that data?

Well, the Internal Revenue Service has that data. And if you were to move it into HHS or then HEW, you would have the IRS giving that data or giving verification of that data to another government agency. And we have — Mr. Alexander having protected — you know, been there for 6103, and the whole point of 6103 is to prevent that wholesale moving of our most protected and most valuable asset, which is taxpayer information, anywhere but the IRS unless you have a very, very, very good reason. And so I have really come back to, in my own mind, saying, like it or love it, or hate it, we should be administering at least this program of this size. I mean, there are 22 million, isn't it — 32 — I can't remember how many people there are now — that are involved in the EITC, and to see that kind of data going out through our doors to somewhere else just makes me really nervous.

MR. BERGIN: Frightening.

MS. OLSON: The second point I wanted to make was to Steve's point about the tax preparation software. You know, tax preparation software is very seductive, you know, it's painless, and yet it doesn't help in any way to simplify the code; in fact, it's actually the thing that allows the code to become more complex without any pain. And I think that it has the side effects for taxpayers that can be actually — and we are only beginning to see this — it's actually very harmful to the tax system when people get a return that comes out and — as happened to me two or three years ago when the AMT form, you know, 6251 spewed out and I thought, you know, wait a minute, I only have my son, I'm only filing head of household, why am I under — you know, why do I have the AMT? And I had to go back and go by hand to figure out why that was and understand it. Now, you know, I at least knew what the AMT was. I knew what was happening. But what happens to the average taxpayer who gets that form spewed out, and how do they feel about what's being done to them, and they have no explanation except the machine did it, and the IRS did it. And I think that is one of our biggest problems for erosion of confidence in the tax system and willingness to comply, and it increases the arbitrariness of the tax system to the average taxpayer, and no good comes from that. No good.

MR. BERGIN: Or you fill out a whole portion of the form to find out you don't have to pay it.

MS. OLSON: Right, well, you —-

MR. STEUERLE: I think a lot of the discussion we've had lately serves to point out the problem with taking a general rule which probably is right and applying it in an extreme where we get cases that are probably pretty wrong. To give some examples, I hear Bill Niskanen sort of having a rule that the more efficient the government is, the more we invite it to become inefficient. In some sense, he's saying, well, if we have a really messed up tax system, then maybe people will want smaller government. And from now on, Bill, whenever I drive down 23rd Street, which always has potholes — although I was thinking that Bill Niskanen's arguing, if we only had more potholes in that road, the government would only hire really bad contractors, then the D.C. government might be smaller. And there's some truth to that. Actually, there's some level of truth to it; the problem is you don't want to take it to an extreme and say — because the next step is, well, let's add inefficiency, let's make the tax code even more inefficient. I think you would probably agree with that.

Another one where I would sort of probably get to the point of having problems of having a general rule is now your — your response is, okay, I agree, but let's still have a supermajority with tax rates. And you know as well as I do — and this gets back to the spending versus tax thing - - we can build tax rates in anything. We don't — statutory rates are only one of a zillion places where we hide tax rates. So we can put them in alternative minimum tax and phase-outs and in the biggest tax rates, which drives me crazy — and Nina and I talk about this a lot — is the biggest tax rates are all these phase-out of these welfare programs. You know, we all worry about tax expenditures — I once wrote a little piece for — I guess it was for you guys, which — no, this was the National Tax Association, which I called — deemed something I called expenditure taxes, which is all the tax systems that are an expenditure system.

So you could argue — they could be having their little conference and saying, you know, we shouldn't have any of our tax systems over in the expenditure system unless we put them all over in IRS. We shouldn't phase out any of these programs. So there's a danger in taking any of these rules too far. Does social policy belong in the tax system? Well, in many cases, it's probably not — doesn't do well there. Probably the best case — and the tax expenditure budget offsets that a little — is it still turns out that politicians count social spending, or business tax breaks, or whatever they are, as being negative taxes rather than positive expenditures, so it does appear to be smaller government. So that's sort of relatively unambiguous, the bad about having spending in the tax system, whether it's social or any other type. However, you can go too far here — and Bill gave a great number of examples, and Nina, too — there is just a really messed up system. You know, you really want — you think the tax code's messed up, you know, you really ought to look at some of these other systems.

We just had a conference, the American Tax Policy Institute yesterday where somebody was going through and talking about changes in involvement in various social programs in the state of Wisconsin, and he says, well, you know, what's really going on in that state, what seems to be the principal policy lever was they had a light — what they called a light touch policy, which meant that they essentially were trying not to get people in systems by not making them aware of them, or they made things like a homesteader's credit so complex that nobody applied for them, and so — you know, is that what we really want? We really want a system that lacks transparency and —

MR. MINARIK: Gene, that's called a light touch system?

MR. STEUERLE: The light touch system in Wisconsin was, don't let people know about programs; they might not apply for them.

MR. MINARIK: These are incentives, right?

MR. STEUERLE: These are incentives for the bureaucracy, not — because they don't know how to deal with them.

MR. MINARIK: But you don't let people know about the incentives that you've created for them.

MR. STEUERLE: Right. Which is what — for instance the question why so few people even on food stamps in Wisconsin was among the issues. So that's a problem with taking that. Another one, which you raised, which is the broader base, lower rates — well, as a general rule, that probably works, but don't forget that what happened by the time we got done with tax reform in '86 is that in some cases, broadening the base meant doubling the base — you know, we'll tax it twice or three times, and that was a broader base, and we had a lower rate. And we get this all the time at state levels where we get this push towards gross receipts tax, which are turnover taxes; boy, you can get a really low rate of tax if you just — if you tax every transaction, which is a horrible type of tax for efficiency reasons. So again, it's a general rule, but I think we have to be a little careful about applying this treatment.

And finally, back to my first thing about you don't always want simplicity. The biggest tax spending — social spending tax expenditure — measured in the tax code is this employee exclusion for employer-provided health care, and I've written on it in Tax Notes. The way it's designed is every - - the additional money we're spending there is buying more uninsured people every year. It's that bad of a subsidy from an efficiency standpoint. There are a lot of us who say, well, let's cap it. Or if you want to spend the money somewhere else — that, by the way, would not be simple, but it would be a heck of a lot more efficient.

So I think all this says is that — when you deal with issues like simplicity, you really have to get some people with basically good will around a table and come to a balanced approach to these issues. And one rule is not just going to solve the problem, whether that rule is broader-based lower rates, whether it gets social policy out of the tax code, whether it's let's have simplicity. It's just not enough to have one rule; it just doesn't work by itself.

MR. BERGIN: I agree with you. Howard, you wanted to go?

MR. GLECKMAN: I'm Howard Gleckman at Business Week. I'd like to ask a question actually of some people on the panel. Maybe Bill Diefenderfer could talk about this and maybe Gene and anybody else wants to jump in. And that is, do people think that the solution to this is a very dramatic change in the code, something like Mike Graetz's add-on VAT? I know what Sheldon Cohen thinks — he's told us — but the idea of simply — of dealing with simplicity by getting nearly everybody out of the income tax system. Bill, what do you think?

MR. DIEFENDERFER: Bill Diefenderfer. Theoretically, like broad- based low rates, that would work, and it makes sense. In the real world, it's not going to happen. It took probably — I just happened to hold a job when 20 years of screaming and hollering about the tax system fomented the '86 reform, as imperfect as it was. If you're looking at, like, going to a VAT system or a sales tax system, we're looking at 20 more years of debate before anything happens. That's what I believe, knowing how politicians work. So I think you're stuck with the present system, and it's a matter of paring it back, just like bringing a ship into dock — get the barnacles off the hull, you know, put new oil in the motor, whatnot, and get it moving forward. Otherwise, I don't think anything will be done, other than in theory.

MR. BERGIN: Bill.

MR. NISKANEN: In response to your question, you should look very carefully at the experience of Europe, where the size of government in Europe is 10 to 20 percentage points of GDP higher than it is in the United States, financed by a big VAT. So be careful about that. Now, for those people who casually assume that we make better decisions on the spending side of the budget than on the tax side, I encourage you to look at the most recent energy bill or transportation bill —

SPEAKER: Good point.

MR. NISKANEN: — both of which are better — best described as "leave no lobbyist behind" bills.

MR. BERGIN: Sir.

MR. ALEXANDER: I'd like to hear from Eric Toder, but first I think I need to respond to a couple of things. Nina, 6103 — I agree with you completely that taxpayer information should be held private. I agree with you completely that 6103's reversal of prior law — and not prior regulations — in the '76 act is — was a good idea then and remains a good idea today. But I don't agree that protecting taxpayer information necessarily means that the Earned Income Tax Credit, if it were possible to move it now — and I'm not sure it is — probably isn't — to where it should have been in the first place — would improve the tax system immensely, get rid of some of the major problems that overworked Internal Revenue Service has to face today, and be administered by people who might be free of being called jack-booted thugs, as the IRS was called only a few years ago by the ill-advised 1998 Act. So I would hope we could do that. Now, Eric, because we're running out of time, I'm going to shut up.

MR. BERGIN: Eric, I'm sorry, I didn't see your hand.

MR. TODER: I don't know if you — I don't think I had my hand up, but I'll make a few comments.

MR. BERGIN: Well, you've been nominated.

MR. TODER: I'll make a few comments anyway.

MR. BERGIN: He says you talk, you have to talk.

MR. TODER: All right, I've been enjoying a lot of insightful comments, so I will respond to the handful of comments I disagreed with. First of all, Joe, that idea of a refundable credit department was not my idea, but I do think it's a great idea. You probably got the idea from a paper I wrote in which I cited something David Bradford had written in which he advocated eliminating the Department of Defense and instead giving refundable tax credits to all suppliers of defense, and that would be — cause a substantial reduction in the federal spending with no impact on anything.

Second, I wanted to respond to one comment Bill Niskanen made about sunsets. I didn't want to let that — I think that's actually a terrible idea for the tax code, the proliferation of sunsets. And what you have is not only — not only do you have the uncertainty for business, but you really have this budget shell game going on where we think that we're raising X amount of revenue in future years, and we're not because we're going to extend all these tax cuts. And my favorite example is the R&D credit, which I think was first introduced in 1981, and it's always extended for three years or one year, and of course, it's still here.

SPEAKER: Great fund raiser, though.

MR. TODER: It's good for lobbyists. Nina, can't leave you alone. Nina said people don't like it when their AMT is spewed out by a calculator. Believe me, they like it a lot less when they get a notice from the IRS. So I do think, from some work I've done, that software and preparers are providing a service for taxpayers in helping them cope better with the tax system and that they're a net benefit. I do want to say, though, we — at IRS, when I was there, we did develop a new burden model which is different from what Nina has described. And one of the things that was found from the burden studies was — a couple of things are worth mentioning.

One is the largest component of burden is — for individual taxpayers — are borne by those who are self-employed; even though they are a minority of taxpayers, they have a majority of the compliance costs. And many of the types of things that they have to deal with are things that no reform can really fix; for example, something like the home office deduction, I think most people would agree that's legitimate if you don't have a place of business to have a home office deduction. Or it's legitimate to deduct your gasoline for business travel, but not for personal travel, and trying to figure that out is a mess.

The other thing is the largest single component of burden is recordkeeping, and again, the software in the tax calculations — software doesn't do anything about that. A simpler tax system might.

Finally, I do want to endorse Gene's list of things that we could get rid of. I think he mentioned education credits — not get rid of, but simplify — retirement saving credits, a whole list of things. I really think a reform could focus on those items without changing the distribution of the tax burden, without addressing the issue of whether subsidies should be in the tax code or not, without addressing a whole lot of hot-button issues — the size of government — could simply make the system simpler and work better for people. And so I think anybody who had a red tie and a blue shirt should be listened to at this meeting.

MR. BRUNORI: Eric, did you say that the preparer industry was a net benefit to the taxpayers?

MR. TODER: Yes, I — no, I believe that, you know, this is basically the private sector arising to fill a need. It would be better if the system were simpler and you didn't need such a big sector, but that's —

MR. BRUNORI: But under the circumstances —

MR. TODER: Yeah. Yeah.

SPEAKER: You're talking about the software.

MR. TODER: I'm talking about both software and preparers.

MS. OLSON: And you know, I mean, I wasn't saying that the, you know, the software was bad, I was just saying that there are consequences and repercussions from the software that go to the system that's broader, that if you think that — if you — it might lead us — and I think it has led us — to say, well, we can implement these things that have different phase-outs and all those sorts of things because people can go through a questionnaire, and the machine will decide which is the best education credit or provision that they should take. I mean, I did that this year, and, you know, I'm still not confident that I chose that. Maybe I can tell you — I mean, my personal experience of filling out my return this year — I was actually filling out my son's, and I hope I don't get him audited — but, you know, he was self-employed, and I was, you know, being a good mom and doing his return, and he had a bunch of equipment that needed to be depreciated. Well, there are tons of different ways to depreciate equipment now, including the fact that he lives in the Liberty Zone in New York City, so there's now a new provision.

So I'm sitting there, you know, trying to figure this out — and he's also made so little money that, you know, he's very low income, and in one of the little upper corners there is the thing that says how much money you owe in taxes as you work through the machine, right, the process. And he comes out owing a little bit of self-employment tax. And I thought, well, let me see if I take this class of equipment and do section 179 with this one as opposed to 179 with this one, and all of a sudden, whammo, he got $25 of the earned income credit, you know? It went from a negative — you know, he owes $379 — and I thought, well, I'll take that one. Now, I have no idea whether that was the right answer or I just violated the law for my son, you know. I mean, I went back and tried to read the return, I went through the questionnaires over and over and over again. I don't know what happened really. I don't think that's good. But it was easier, and it let us — you know, it — I think it feeds into letting us pass all these different provisions because a machine will take care of it.

MR. DIEFENDERFER: It used to be a lawyer can take care of it, and they're more expensive than the machine, so it's not a bad tradeoff. You know, so —

MS. OLSON: Right, well, that's true.

MR. BERGIN: It goes back to Sheldon's point, though, what I was talking about in the beginning. I really do like the Form 1040. I mean, there's a whole bunch of stuff on it I don't have to read, but every time I get driven to a worksheet, I'm getting angrier and angrier at the IRS. Now, I do this for a living; I know who screwed up the tax code. It wasn't the IRS, but it doesn't matter because every form has their name on it, and those are the guys I'm mad at. So I wouldn't trust them if they sent me a computer disk either. I think the filling out the forms is a good exercise. I worry about technology because I don't think it keeps us in touch with our tax system.

MR. COHEN: My predecessor caused the formation of the tax preparation industry before — when I was a kid, the IRS would fill out returns, and there were long lines around all IRS offices. First floor of the IRS building had benches along it; people would sit and wait for an IRS person to help them fill out their return. Now, Mort said this is a terrible waste, and he stopped it. Luckily, I didn't have anything to do with that decision; I was chief counsel at the time. He stopped it, and that was the spawning of H&R Block. And then it proliferated after that.

MR. BERGIN: I would say Joe Minarik again, but it's actually Joe Minarik still.

MR. MINARIK: We've been dancing around two related points that nobody — I don't think anybody has said yet, but I think they're worth mentioning. Number one, at the outset, part of the — part of an implicit hierarchy of good things and bad things for a tax system in terms of simplicity was the number of tax brackets. I think most of the people around this table would agree that the number of tax brackets has absolutely no implication for the degree of difficulty of — for the taxpayer filling out the tax return.

The other thing that has been implied, but maybe not yet said in bright red grease pencil, is that the subordination of the complexity of the tax system creating a mass of phase-outs so that we can keep the stated marginal tax rates lower as some measure of the economic efficiency of the tax system is terribly misguided, and we could do a lot better if we were willing to go up a percentage point or two on marginal tax rates and get rid of a lot of those phase-outs. Just to put that in explicitly for purposes of posterity, and those of you recording this, thank you.

MR. BERGIN: We're right up against 11:00. Anybody out here have anything quickly? No? Let me then thank the panelists. This was great. You guys were marvelous. We once again picked well and thank all of you around the table and out there who helped us. This was a great panel discussion. Thank you.
(Whereupon, at 11:01 a.m., the PROCEEDINGS were adjourned.)

* * * * *