American Enterprise Institute

April 16, 2007

[Edited transcript from audio tapes]

8:45 a.m.	Registration and Breakfast
9:00	Speaker:	Leonard E. Burman, Urban-Brookings Tax Policy Center
	Panelists:	Daniel Shaviro, AEI and New York University
		Alan D. Viard, AEI
	Moderator:	Alex Brill, AEI
10:30	Adjournment

Proceedings:

Alex Brill:  Good morning.  Thank you, everyone, for joining us here at the American Enterprise Institute for a conference entitled, Options to Fix the Alternative Minimum Tax.  My name is Alex Brill and I’m a research fellow here at AEI and the moderator for this morning’s event.

Before we begin, I would like to start with one quick announcement, which is that today’s event is being co-sponsored with the Brookings-Urban Tax Policy Center.  AEI is pleased to be working together with Brookings and Urban highlighting and researching this important issue and we look forward to future events together.

Let me make a brief introduction, and then we can dive into the presentation, the discussions, and then we will have time to open the floor for Q and A.  The alternative minimum tax is particularly concerning to me because it’s only a few years older than I am, but already showing clear and growing signs of mid-life crisis.  If the AMT could talk, it might say, “My life wasn’t supposed to be like this.”  As you likely have read before and will hear again today, if left untouched, the number of taxpayers affected by the AMT will rise from 4 million in 2006 to about 23 million in 2007.  Because the AMT is not indexed for inflation, it will eventually start to affect true middle income taxpayers in significant numbers.  While I wrote recently that the AMT bite is concentrated among just a few coastal states, it does remain true that it will become a serious and complex problem for millions of taxpayers to deal with if not addressed by Congress.

With that, let me turn the floor over to Len Burman, our presenter this morning, who is the director of the Brookings Urban Tax Policy Center and Senior Fellow at Urban Institute.  Len will be presenting some ideas about various ways Congress and policymakers could deal with the AMT. 

After Len, we will hear from Dan Shaviro from the New York University Law School and a visiting scholar here at AEI and Alan Viard, a resident scholar at AEI.  In short, we will have the benefit of both a lawyer’s perspective and an economist’s perspective on Len’s presentation.

Complete bios are available on the packets on your chairs and on the AEI website and with that, I turn it over to you, Len, for 30 minutes.  The floor is yours.

Len Burman:  Alex, thanks very much.  I want to thank the American Enterprise Institute and Alan Viard for doing all of the work on this first of what I hope will be a series of co-sponsored events with the American Enterprise Institute.  It’s a great pleasure for me to be here.  On your screen, you have a movie poster from the 1950s for a movie called The Blob.  Indescribable, indestructible, nothing can stop it. 

The really scary thing is the AMT, which for all of those adjectives apply in spades.   I guess we are all hoping that maybe it isn’t indestructible, that in fact there are ways to fix it.  In fact, people in this room could come up with 100 different ways to fix the alternative minimum tax - since we never have to run for office, it’s a piece of cake. 

I want to start with a little bit of background, and then I’m going to talk about ways in which things you could do to either lessen the bite of the AMT, retarget it so it could meet its original objectives or to repeal it, and how to finance those options.

Just some background, the AMT started in 1969.  There was testimony by Treasury Secretary Joseph Barr that 155 high income taxpayers – back then, that was people earning over $200,000 a year – had paid no income tax in 1966.  And this actually unleashed a firestorm of protest.  According to lawyer Michael Graetz, there were more letters to Congress about this, 155 high income non-taxpayers in 1969 than there were about the Vietnam War.  Alex, you don’t remember that, but the Vietnam War was a really big deal in 1969.

So Congress could have done one of three things.  They could have explained to the public that these people weren’t paying tax because they were taking advantage of tax incentives that Congress had designed and this was a good thing.  That would have been a hard sell; some of those tax incentives really made no sense at all.  They could have eliminated the egregious tax shelters so that people would no longer be avoiding all income tax.

Or they could do what they did, which is put a band-aid on the income tax.  Basically there was an additional tax that higher income people had to pay so that a future treasury secretary wouldn’t be able to come back to Congress and say that 155 people didn’t pay tax anymore.

Well, the original tax was significantly different from the current one and I think that Dan is going to talk about that some.  But the AMT, as it has evolved, has not been very well thought-out or very well designed.  It’s not going after high income non-taxpayers so much as it is encroaching on the middle class, as Alex had pointed out.  By the year 2010, if there is no change in law, 32 million households will be subject to the AMT. 

I want to just give a brief overview of how the AMT is calculated.  I am not going to go into the details because we’d all fall asleep, but the basic idea is that you go through your regular tax calculations, you think you are done.  And if you pay attention, you see this little note in the instructions that said under certain circumstances, which apply to almost everybody, you should fill out form 6251 and see if you owe extra tax. 

You add back in a bunch of things that are called preferences or adjustments to your taxable income to create something called alternative minimum taxable income.  You subtract from that a special exemption that applies to the AMT and then you calculate your tax under a different rate schedule, which has got two statutory rates, 26 and 28 percent.  And you pay whichever one is higher.  If your tentative AMT is higher than your regular tax, you pay the difference, and that is actually your AMT. 

For the last few years, Congress has been trying to keep too many people from being subject to the AMT.  Since 2001, they have been temporarily increasing the AMT exemption level basically, the base level above which nobody could owe AMT.  In 2006, the exemption was $62,550 for a couple.  In 2007, it reverts back to its pre-2001 levels and for a couple, it goes back down to $45,000.  What that means is that everybody who is on the AMT, couples who are on the AMT are going to have almost $18,000 of additional taxable income if Congress doesn’t change the law.  And people who weren’t on the AMT are going to be much more likely to be subject to the AMT because they get to subtract less.

None of the AMT’s parameters is indexed for inflation, and that is a serious problem.  The other thing is that the rate brackets don’t vary depending on your marital status and as you will see in a couple of minutes, the AMT has pretty onerous marriage penalties built into it. 

There is also a hidden tax built into the AMT, which arises from the fact that the AMT exemption, that $62,550 or $45,000, phases out with income over a broad range and that creates higher phantom tax rates, as high as 35 percent.  That has important implications, both for the coverage of the AMT and for our evaluation of this tax, whether it makes sense as an alternative. 

This chart shows what the AMT tax rate schedule looks like compared with the regular income tax schedule in 2006.  One of the things that is kind of surprising about it is that a lot of people have referred to the AMT as basically a flat tax.  It applies to a broad base and it only has two statutory tax rate, a 26 percent and 28 percent.  But as you will recall, because of the phase-out of the AMT exemption, the effective tax rates under the AMT go as high as 35 percent.  Basically in the phase-out range, every dollar of additional income translates into a dollar and a quarter of additional alternative minimum taxable income.

Kevin Hassett of AEI has actually written a lot about phantom taxes in the tax code.  He had a great article years ago called, “The Skyline Tax System,” where he showed that if you plotted out the real way taxes affect people, these affective tax rates by income, the tax schedule doesn’t look like the stair-step thing that it was designed to do, but it looks like the skyline of New York and some of the skyscrapers are created by the AMT.  So it’s not a flat tax. 

The other thing you will notice on the AMT line is that the highest income people actually face lower tax rates than people who are moderately high people, people with incomes between $150,000 and $400,000.  That is one of the reasons why the AMT is much more likely to hit upper middle income people and moderately well off people than it is to hit millionaires.  It is not a flat tax, it’s kind of a goofy design and that’s what it is. 

What kinds of things get people on to the AMT?  You’d think that, if this thing was designed to go after people with tax shelters, it would mostly be about oil and gas tax breaks and other kinds, accelerated depreciation and things like that.  And, those are in the tax base for the AMT.  But the biggest items are things that actually affect middle income families.  The state and local tax deduction, you add back in your state income taxes, your sales tax deduction, if you take that, your property taxes into taxable income for purposes of the AMT, and more than two-thirds of AMT preferences, these add-backs for the AMT, are state and local taxes. 

Now, nobody chooses to live in a high tax state because it’s a tax shelter, but that’s the way it’s treated for purposes of the AMT.  The second thing is personal exemptions.  One way you can get lots of personal exemptions is to have lots of kids.  There are many wonderful things about children, but they are not a tax shelter - they are not actually a great financial investment. 

Miscellaneous itemized deductions – that sounds like that maybe that has something to do with tax sheltering, but in fact a lot of the things that are called miscellaneous itemized deductions are legitimate adjustments to ability to pay.  If you win a lawsuit, under certain circumstances, your lawyer’s fees are considered miscellaneous deductions.  That means that the entire award you get from the lawsuit is included in your income, but the lawyer’s fees are not deductible if you are under the AMT. 

New York Times reporter David Cay Johnston wrote in a book that there were people who actually were worse off after winning a lawsuit with contingent legal fees than they would have been if they had lost, because the legal fees were so high that after tax, it actually cost them money.  That is not good.  Some of the other things, net operating losses, incentive stock options, passive activity losses, those are more technical things and those might be related to tax sheltering, but they are a tiny fraction of the total. 

Why has the AMT been growing so much over time?  What this chart shows is what would have happened – the bottom sliver that is almost invisible there shows that if we had not enacted the tax cuts of the last six years and if the AMT had been indexed for inflation before 2001, there would be about 300,000 people on the AMT and it would stay constant over the entire ten year budget period, from 2007 to 2017. 

There are two main factors in the growth of the AMT.  One of the factors is the failure to index the AMT for inflation.  Why does that matter?  Well, the regular income tax back since 1982 has been indexed for inflation.  That means if your income just keeps up with inflation, your tax doesn’t grow any faster than your income.  You don’t get pushed up into the higher tax brackets.  That is not true for the AMT.  Now remember the AMT exemption in 2007 is exactly the same in nominal terms as it was in 2001.  If it had been adjusted for inflation, it would be much higher than $45,000 now.

The second factor is the Bush tax cuts.  Basically, the 2001 to 2006 tax cuts lowered regular income tax rates, but except for a temporary fix, which has been extended a few times, it didn’t do anything to the AMT.  You pay the higher of your regular tax calculated under the regular tax rates and the AMT.  The AMT staying the same, your regular tax goes down and all of a sudden a whole lot of people have to pay the alternative minimum tax.  It’s basically doubled the size of the problem.

Why is the AMT so hard to fix?  Well, the main reason is that we’re counting on it for an enormous amount of revenue.  The top line shows the revenue expected from the AMT over the next ten years – when President Bush says that he is going to balance the budget in 2012, he is counting on raising about $150 billion from the AMT in 2012.  Now he said in other context that he doesn’t want to do that, but that’s not built into the budget, any fix to the alternative minimum tax.  So basically we all know that the AMT doesn’t make any sense as policy, but we’re counting on it for the revenue or at least we are pretending that we are.

One of the most shocking graphs, I think, is this one.  This is the cost of repealing the AMT compared to the cost of repealing the regular income tax.  There are a couple of different ways you can measure this.  We define repealing the regular income tax as getting rid of everything, including all of the credits, and just leaving the alternative minimum tax in place.  By that metric, it would actually cost more to repeal the AMT right now than it would cost to repeal the regular income tax and leave the AMT in place.  That gives some indication of the magnitude of the problem.

Over the next ten years, we are counting on the alternative minimum tax for $850 billion of revenue, and that assumes that the tax cuts expire as scheduled at the end of 2010.  If they are extended, as President Bush would like, the cost of repealing the AMT basically doubles.  Only 4 percent of taxpayers were affected by the AMT in 2006, as Alex pointed out, but that share is scheduled to rise to 26 percent in 2007, more than a quarter of taxpayers.  And by 2010, 34 percent, more than a third, would be subject to the alternative minimum tax.  If it ever was a tax on rich tax shelters, that clearly is changing. 

Let me talk just a little bit about the AMT demographics, who it is going to be affecting.  48 percent of couples will be subject to the AMT in 2010 – that is sort of the worst year for the AMT over the next few - compared with only 4 percent of singles.  In other words, you are twelve times as likely to be subject to the AMT if you are a married couple than if you are a single person. And, that is for a couple of reasons. 

One is that couples have higher incomes than singles do, but a big reason is that they also have children and children are considered AMT preference items.  89 percent of couples with two or more kids and adjusted gross income between $75,000 and $100,000 will be on the AMT in 2010.

Residents in high tax states are seven percentage points more likely to be subject to the AMT than residents in low tax states and that is because the AMT does not allow state and local tax deduction. 

So how does the AMT play out by income?  This tax, remember, was originally intended to hit people who, in current dollars, would be millionaires.  Well, in the year 2007, about a third of people with incomes over $1 million are subject to the AMT.  By comparison, 36 percent, more than a third, of people with incomes between $75,000 and $100,000 are subject to the AMT and 71 percent of people with incomes between $100,000 and $200,000 are subject to the AMT.  These aren’t rich people – these are middle class, upper middle class people who aren’t necessarily doing anything special to avoid tax.

The other thing about the AMT is that it creates pointless complexity.  Many middle income taxpayers have to file the AMT even though they don’t owe the tax.  A lot of people don’t find out they are subject to the AMT until they get a note from the IRS saying you should have paid attention to that little line in the instructions, and then they can owe interest and penalties, additional tax. 

Most of the deferral preferences (the thing that I didn’t really talk about, the timing preferences) don’t actually change overall revenues; they just change when the government collects the revenue.  And small businesses, people who are subject who are affected by the deferral preferences, often have to keep two sets of books.  They’ve got to keep track of their AMT liability from year to year and it vastly complicates tax planning.

The other thing is that it really confuses people.  There was an article in Consumer Reports, I think last month, saying that the AMT is going to get you from all sorts of things that affect middle class people, including if you own your home, your mortgage interest deduction can throw you on the AMT.  That’s not true, but Consumer Reports is better than average in checking these things and if they get confused about it, you can just imagine how it affects regular people.  So this tax could be entailing burdens on people that aren’t even related to the way it actually collects tax revenue.

This slide shows the percentage of AMT taxpayers that face higher marginal tax rates under the AMT.  Now one of the things that sometimes people say about the AMT is that it is a broad-based, low rate tax.  Well, you do add things back into your income, but then you subtract this AMT exemption.  And because of the phase-out of the AMT exemption, marginal tax rates under the AMT can go up to 35 percent for people whose statutory tax rates are often lower. 

This chart shows that in 2006, 71 percent of people on the AMT faced higher marginal tax rates.  That means that they were discouraged from working and saving, they were encouraged to engage in tax shelters, kind of ironically, and that’s a bad feature of the tax.  By 2010, when it has consumed the entire middle class, almost 90 percent of households will face higher marginal tax rates under the AMT. 

One of the rationales for the AMT is that it improves the progressivity of the tax system, and that is certainly true.  I mean, the AMT, despite the fact that it’s encroaching on the middle class, collects most of its revenue from people with relatively high incomes, people with incomes over $200,000.  But that is also changing over time.  You see, in 2006, 7 percent of revenues collected from people with incomes between $100,000 and $200,000.  In 2010, 35 percent of the revenue, almost a third of the revenue, is collected from that group and relatively less of it is collected from people with higher incomes.

So what should we do about the AMT?  This is the most complicated slide here.  But what it shows is sort of a progression of options that you could adopt that would lessen the impact of the AMT.  It also shows the effects of revenue if the tax cuts expire at the end of 2010 and if the tax cuts are extended.  And you can see basically that the revenue costs are significantly higher if the tax law is extended.  The solid red line at the top shows the number of people subject to the AMT in 2010.  The solid red line shows the number of people subject to the AMT – 32 million people subject to the AMT in the year 2010 if current law is extended. 

If all you do is index the AMT exemption level for inflation over the next five years, that would cut the number of people on the AMT to about 4 million in 2010.  It would be about a million who would be subject to the AMT with incomes under $100,000.  So that is indexing the AMT is the second bar there.  If you also allow dependent exemptions, the number of people subject to the AMT declines further.  If you allow standard deductions and itemized deductions for state and local taxes and the miscellaneous itemized deductions, basically no middle income people will be subject to the AMT and the total number of people on the AMT would fall to 1 million in the year 2010.

But the other thing that you can do, in terms of options, are to repeal the AMT exemption phase-out, the thing that creates those phantom higher tax rates.  That would basically eliminate almost everybody from the AMT to have less than a million people subject to the AMT in 2010, but it would also cost almost as much as repeal. 

Repeal of the AMT would cost, over the ten years, about $1.3 trillion.  All of those incremental options would only save about $100 billion over the ten years.

So how do you finance repeal?  I guess one question is whether you even should try to finance repeal of the AMT.  Arguments have been made, actually by Senator Grassley when he was Chairman of the Finance Committee, that we shouldn’t have to pay for repealing for AMT.  We never should have counted on this revenue and we keep on patching it, so therefore we ought not to have to make up the revenues that are lost. 

Well, it is probably true that nobody was counting on 32 million people being subject to the AMT in 2010, but it’s also true that we were counting on that revenue when we passed each of the tax cuts over the last six years.  If we had fixed the AMT, if the AMT had not been in place, the cost of the 2001 tax cuts would have been much, much higher.  The AMT takes back about 30 percent of the Bush tax cuts by the year 2010.  Put differently, the cost of those tax cuts would be one and a half times as large, so the program would have had to be scaled back.

The other thing is that we have enormous revenue challenges facing the country, which I think everybody acknowledges, from the growth of entitlements and from discretionary spending as well.  Congress has shown no particular ability for dealing with those revenue challenges by cutting spending.  So if you can’t cut spending, you need to come up with a revenue source.  Basically, we are counting on the AMT revenue in that sense and if we don’t get it from the AMT, we need to get it somewhere else.

So my paper has 23 different options to pay for repeal or modification of the alternative minimum tax.  This is just a few of the repeal options. 

One option would be to just raise tax rates – repeal the alternative minimum tax and raise tax rates across the board.  In 2007, that means the top rate would go up from 35 percent to 37 percent.  This option would be regressive, because it would be raising taxes on a number of people who aren’t subject to the AMT right now and it would be cutting taxes on those people in the $100,000 to $500,000 category especially who face enormous AMT liabilities. 

To deal with the regressivity, one thing you could do would be to target the rate increases on the top three rate brackets.  Those are the people who are subject to the AMT now.  The rate increases would have to be higher – the top rate would go from 35 percent to over 40 percent in that option.

One option that was proposed by the President’s Tax Reform Panel, and actually I think has a lot of merit as policy is to just repeal the state and local tax deduction.  If you did that, you could raise more than enough revenue to pay for repealing the alternative minimum tax.  You could actually cut regular income tax rates by 2 percent and the top rate would go from 35 percent to 34.3 percent.

That would actually be a very progressive change.  The people who benefit from state and local tax deduction, if the AMT is in place, are mostly or disproportionately very high income people.  Upper middle income people don’t benefit from state and local tax deduction because the AMT is already taking back a substantial portion of that deduction from them.  Lower income people don’t benefit from the deduction because they don’t itemize deductions.  So, increasingly, the only people who benefit from the state and local tax deduction are high income people.  The people who get the most benefit are very high income people because they have a lot of state and local taxes.  You could also just roll back the top rates and that would target the relief among the people who potentially could lose from repealing the state and local tax deduction.  That would allow you to lower rates, the top rate from 35 percent to 33 percent.

Another option is that you could actually roll back the capital gains rate cuts, roll back the income tax tax rate cuts for the top three brackets and just keep those the same over the next ten years.  The top rate would go from 35 percent to 42.8 percent and that compares with a top rate of 39.6 percent in 2011 under current law.  If the intent of the AMT is to rein in tax shelters, that has some merit, because almost any tax shelter that you can imagine involves converting ordinary income into capital gains.

But all of these proposals are politically challenging and would offend one group or another.  There are also incremental options which could be financed with tax rate increases.  They could also be financed by various other kinds of base broadening measures, such as rolling back the state and local tax deduction. 

This chart shows four of the options that we discussed in our paper.  The first is indexing the exemption, and you would have to raise the top rate to 39.1 percent from 35 percent to pay for the revenue loss from doing that.  That would reduce the number of AMT taxpayers in 2007 to 2.4 million from 23 million under current law and only 100,000 with incomes under $100,000 would be subject to the tax. 

You could add to that, allowing dependent exemptions, allow people to take deductions for their children; the rate would have to go up a little bit higher.  The additional cost is not all that great, it would cut the number of people on the AMT from 2.4 million to 1.7 million.

You could allow, in addition, deductions for state and local taxes, miscellaneous itemized deductions, standard deductions.  The top rate would have to go up above where it was before the Bush tax cuts, up to 39.9 percent.  The number of people on the AMT would fall to 300,000 and the number of people with incomes under $100,000 would fall to zero.

And the fourth option, which isn’t really described very well, this is retargeting.  This actually returns the AMT more to its original purpose, which is going after tax shelters.  Under this option, capital gains and dividends would be taxed in full under the AMT.  Currently, capital gains and dividends are subject to a maximum tax rate of 15 percent and that applies not only under the regular income tax, but it also applies under the AMT.  It is very complex to implement the lower tax rate, but the other objection to providing the preference under the AMT is that if the purpose of the AMT is to rein in tax shelters, capital gains ought to be taxed. 

It’s hard to think of a tax shelter that doesn’t involve converting ordinary income from high taxed wags and salaries or other compensation into capital gains.  In fact, capital gains was the biggest preference item in the AMT before the Tax Reform Act of 1986.  If you did that, instead of flat rate on the AMT of 28 percent, got rid of the phase-out of the exemption, then the top rate on an ordinary income could be 37.7 percent on a revenue neutral basis.  There would be slightly over 1 million people subject to the AMT. 

Alternatively, instead of raising ordinary income tax rates, you could raise the AMT rates to a flat 33.9 percent, but that would increase the number of people on the AMT because the higher the AMT rate is, the more people would be subject.  But still, there would be nobody with income under $100,000 and in total, about 5 million people subject to the AMT. 

So conclusions – the AMT has no policy merit aside from the fact that it raises revenue.  There is pointless complexity, there’s a bizarre pattern of taxes, it’s inefficient and in a lot of ways, it is unfair - it is increasingly a tax on the upper middle class.  And anybody who doesn’t have to run for office can come up a thousand ways to make up the lost revenue, but there is no silver bullet. 

I testified before Congress and one of the Democrats, who had actually invited me to testify on this issue asked me what my favorite option was and I said to repeal the state and local tax deduction.  He said that he would learn to find out the answer to my question before I ask it the next time.  I said that there are always reasons why you should like it, but just politically, it’s very difficult. 

It might be that the best way to fix the AMT, the time when the AMT is finally fixed, will be as part of an overall tax reform when we deal with a lot of the problems in the tax system and not just this one egregious, glaring problem. 

Alex Brill:  Thank you very much, Len. 

Dan Shaviro:  Well, thanks to all of the sponsors for inviting me.  I am glad we are having a discussion of this very important, if not very pleasant, issue, which more and more Americans are going to be finding out about in the next couple of years.

Well, the AMT is a part of broader instability.  I thought that in a conference at AEI, there is nothing better than starting with Adam Smith who said that old tax is a good tax – when the rules are in place and you know what they are, it’s easier to plan and you’re not facing needless uncertainty, and so forth. 

Now sometimes we want the tax laws to change.  For example, there is new information, new political preferences, we decide that 50 or 70 percent rates like we had a long time ago are too high, we want lower rates.  We have a costly foreign war – oh, no, I guess we don’t raise taxes to pay for that.  Sometimes you want to change because you want to do something differently, but not to have built-in instability, creating uncertainty for no good reason, because you have rules in place that can’t continue. 

Well, the word for that, it’s kind of a technical word that tax lawyers use, is moronic.  It doesn’t make sense to have instability simply because you have rules that on their face can’t possibly stay in place.  Now there are several causes of built-in instability in the U.S. tax system today, three in particular.  And I think there are all gratuitous in the sense that it’s really just a problem of political will – you really don’t need them at all. 

First is the overall U.S. fiscal gap.  We have an unsustainable relationship between taxes and outlays and we just know that can’t continue, but no one has wanted to fix it.  The second is these tax-law sunsets.  No one really believes that you can have an estate tax of zero in 2010 and then it’s going to be back at the pre-2001 level in 2011.  It would be a fascinating research topic for economists who want to look into the tax responsiveness of death, but it’s not really a feasible arrangement.  And then finally there is the AMT with its projected growth path where everyone is going to be on it increasingly.

So these are three things.  You just look at the law and you know that these can’t continue, but absolutely no one has, in the political system, been able to do anything to solve them.  So I’m certainly glad that we are discussing one of these things today. 

Unfortunately, actually, it’s the relationship between the three makes them much harder to solve.  You have overall fiscal gaps and you can’t just, responsibly at least, get rid of the AMT.  You have people talking about tax reform due to the sunsets, the two sides can’t agree on what the baseline is.  The problems all, I think, make each other worse politically.  So, it’s one reason why three problems are more than three times as bad as one. 

Anyway, Len described some of the options in his paper that he and some co-authors wrote laying out options for getting rid of the AMT or fixing it one way or another.  It is a really useful exercise.  Not pleasant reading, because none of the things look that enjoyable, but useful to have it because you need to know what the tradeoffs are.

I do think that it’s just wildly irresponsible, given the U.S. fiscal gap, to simply get rid of the AMT.  Congressman Thomas says we shouldn’t count on these revenues and he has a point.  But the fact is, if you have a bad fiscal gap situation, you don’t really want to make it worse.  It’s going to be hard enough to fix as it is.  It’s desirable to know what the tradeoffs look like.

In the paper the authors admit that they don’t have behavioral estimates, they just kind of assume current behavior the way the revenues are raised.  Obviously, that’s a detail that at some point you want to fix, but there is plenty of time to do that later if there is ever a serious inquiry.

Now we are not going to pick an option today here in this room.  Actually, the sad thing about this is, you have like full revenue information and you could take an economist from AEI on the more conservative side, an economist from Urban, Brookings and the more liberal side, put them in a room and give them an hour, and they could solve the problem.  It wouldn’t be pretty, because none of the options are nice, but they could solve the problem.  Unfortunately, you are taking Democratic and Republican politicians are running for office, it doesn’t quite work that way.  So we’re not going to pick an option today, and if we did, no one else would really care.  So I wanted to kind of say how should we think about this problem, and there are two scenarios, there are two kinds of things to think about.

One is the short term – what would be a feasible scenario in the short run for a permanent fix?  The long-term issue that I have in mind is where do we really want to go, where do we want the tax system to be heading?  And the point here is that a fully financed AMT fix would be a major change with significant effects.  There is just enough revenue involved here that you can’t just do a trivial change and get rid of it – it has to be something big.  So you would kind of like to be heading in the right direction.  Even if you don’t solve everything once and for all, you would really like the fix you came up with to be sending you in the direction that you want to be going in the long run.

Now the President’s Tax Reform Panel a couple of years ago thought that this could be traded in for fundamental tax reform and that is certainly an appealing idea.  It didn’t fly very well when they tried it.  Of course, one reason was that they were telling people that they were going to take away all of these tax goodies you like in exchange for this problem that hasn’t hit you yet, which didn’t make it very politically exciting to the general public.  So I don’t know that fundamental tax reform is going to be part of the fix – it would be nice if it was, but we have to assume that it might not be. 

Well, what are the kind of realistic long-term tax policy goals we should have in mind when we try to get rid of the AMT problem?  The first is that the fiscal gap is going to have to be addressed.  We have really an unsustainable relationship between revenues and outlays.  It is just not going to be possible to stay on this path.

I think that rates will rise at some point.  Would I like them not to rise?  Yes, I would prefer them not to rise, but I think it’s inevitable that they are going to rise at some point.  But I think basically, on the left and on the right, all responsible tax policy people and budget people would agree, you would rather really rather broaden the base than raise the rates.  You would certainly want to start there.

Now the second question about where to go is should we get rid of the AMT or just scale it back?  I think that a greatly scaled back AMT is probably something that we could live with, we did for 20 years.  But really if you are thinking what you want the system to look like, what is a better system, no AMT would be preferable.  And that may seem obvious, but I would like to say that actually 20 years ago it wasn’t completely obvious that no AMT is better than a small AMT. It has become obvious now, and here is the mea culpa part. 

I have to admit, this is my Nuremberg Defense moment that I, as a staffer, worked on the AMT in ’86, was involved in designing it.  I would like to emphasize that our job was to make it work technically.  And I think we did, maybe a little too well.  By technically, I mean, for example, handling people going back and forth between the two systems from year to year. 

So at the time, it wasn’t my decision to have an AMT, but I felt at the time that there was a reasonable argument for one and it was based on politics or political economy.  Now I actually had to work on writing the committee report that explained it and we said it’s because it is very bad when taxpayers make too much use of tax preferences.  Like if the rate were 40 percent, it’s okay if you cut our rate from 40 to 20 by using preferences, but 20 to zero, that is a naughty little no-no. 

I had to write that up in the committee report, but I thought it was nonsense.  I didn’t think it really makes any sense.  If the preferences are good, then there is nothing wrong with using more of them.  And if the preferences are bad, then all uses of them are equally bad.  So I didn’t think that was a very good rationale. 

The argument I believed in at the time – and when I say believed in, I thought it was plausible, it wasn’t my decision of whether to go with it or not – there was kind of a reasonable political economy that you get more overall base broadening with the AMT than without it.  In other words, you have these preferences.  Congress was unwilling to repeal them and was reluctant to even scale them back in an across the board way.  But they were willing to deny “too much” use by high income tax payers per taxpayer. 

So the notion was, it would actually be a kind of base broadening. It would expand the amount of base broadening overall that you could really have.  The basic idea here was that, my thought was that if one had he opportunity to wipe out the AMT and simply put all of the base broadening that was there, put it in the regular tax, that overall preferences were being used just the same amount, you would be crazy not to like that deal.  But that deal might not be available.

It might be the case that getting rid of the AMT would simply mean the preferences, which would distort economic decision making, would be bigger than otherwise.  So I felt that this supported a decent case for the AMT.  You have to assume the preferences are bad, but Congress is unwilling to repeal them.  Then by having the AMT, you have less of the preferences, but you kind of have a new level of preferences or dis-preferences which is that these arbitrary factors of how your overall tax return plays out would affect who gets the preferences.  So it was like two wrongs trying to make a right, having distortionary limitation of a distortionary item.  But I thought it was plausible that the AMT could improve the political system overall given the impossibility of trading it in straightforwardly for one better system.

But I don’t think that is the way the world is operating today.  I don’t see it happening today that we have preferences that can’t be repealed or narrowed through the regular tax, instead they put in the AMT.  Even when new ones are created, they are carefully kept out of the AMT.  So I simply think that the scenario I had in mind where you could make an argument that it was possible having an AMT would make the system a little better given political limitations, I don’t think that’s really the way that it’s played out over 20 years.

So, Len had The Blob, which actually is an amusing grade B 1950s movie and I have a zombie in my slide.  It’s kind of just staggering forward.  It’s not like Congress has decided – well, we are unwilling to wipe out state and local tax deductions in the regular tax, but let’s put it in the AMT.  No, it’s really just the force of inertia – the thing is there.

Actually, as one who was involved in drafting it, why didn’t we put those things in the AMT in 1986?  Well, state and local tax deductions were not in the prior AMT and there was enough political will to think – well, let’s not get it out of the AMT.  The other two items that Len mentioned that are really important, personal exemptions and miscellaneous itemized deductions, they just had not been there before, so they stayed not there, really for no good reason.  And those two items are not even preferences. 

The example I like to give for family size is, do you think that if you have two single parents earning $40,000, one with no kids and the other with 17 kids, do you really think they should pay the same amount of tax?  Once you concede that those two shouldn’t, you really are just quarreling over the details.  So these are not even preferences.

Then there is the state and local tax, the one item that is important in the AMT that is, at least arguably, a tax preference.  Even though I live in New York, and it’s painful for me in the pocketbook at least, I have to admit that I think the view that it is a preference has the better of it.

Anyway, so the thing isn’t really, on a stable basis, doing indirect base broadening.  Only one of the big items has anything to do with it.  The thing to do with the state and local taxes would be to trade it in one time. 

Now if we have to keep the AMT, if it turns out the best solution, getting rid of it, is too costly or something, what should we do?  And the answer, I think, is keep it from growing again.  So the structural features we would need with a restored AMT that again was tolerable, if not ideal for many years, higher exemption amount, index for inflation, I would say lower rates. 

Now when I was involved in drafting the AMT – again, I would like to take the credit for this, but it was just the politicians, I was just one of the hatchet men – lower rates.  We enacted in 1986 with rates of 20 and 21 percent – not 26 and 28 percent.  When you make the rates too high, you create more of these problems, so the rates should really be lower compared to the regular tax and no phase-outs.  I never thought phase-outs made sense, they just lead to bubble rates, temporarily high rates. 

And it would be nice to have the thing focused on actual base broadening rather than what it does now.  Interestingly, when I worked on the AMT, actually long before the 1986 Act, my boss at the Joint Committee on Taxation had me combing the tax code finding all to the preferences that we should possibly put in the thing.  That is just not the way the system works today.

Okay, back to the long-term goals.  Well, the only meritorious base broadening that would naturally accompany trade-in of the current AMT would be the state and local tax deduction.  Even if one would like 100 percent repeal, given how various people who do have to face the voters feel about that, one could well imagine a compromise where it’s partially repealed, and this is viewed as a cease fire in place.  So you have Democratic politicians from New York, for example, saying yes, I will agree to X percent to repeal the AMT because you’re really not making anything worse.  They can’t say that on 100 percent repeal.

But lots of other base broadening one way or another would be desirable.  The two biggest items that I think need to be addressed in the long run when we think about what’s wrong with the tax base are the home mortgage interest deduction and the exclusion for employer provided health insurance, which does a lot to mess up the market for healthcare.  Now of course, I say those two things as someone who is not facing the voters either. 

One idea that is worth thinking about that has come up from time to time, including the Tax Reform Panel, is to convert various preferential items into percentage credits.  For example, the Tax Reform Panel was going to make various items like the mortgage deduction 15 percent credits, let’s say, would even make these refundable credits. 

Okay, now a few other of the big issues in the AMT.  In an alternative political economy universe, when they did the changes in 2003 to capital gains and dividends, you could have imagined them saying that the AMT will be a backup, you don’t get these things on the AMT.  Whether that is better or whether that would be a good idea or a bad one, that is kind of the way I think it would have happened if the scenario I saw in the mid-‘80s had been operating.  You enact something that is arguably a preference, you put it in the AMT. 

Now these things are going to have to be addressed as part of the overall thing at some point.  I would say, I think that one of the good arguments for a capital gains preference is the lock-in problem that people won’t sell – there is actually a revenue maximizing rate that you can easily get above.  But I would say that a 15 percent capital gains rate strikes me as a bit low, especially when people are, in effect, borrowing against and deducting the interest that they are leveraging against the capital gains at a full rate. 

In regard to the sort of dividends part, the 15 percent dividend rate, I personally think that corporate integration should be more thoroughgoing than that.  Also, I think a big mistake of the way it was done in 2003 is that you have to have a fix to the corporate tax (obviously too complex a topic for today) that doesn’t distinguish between debt and equity.  Because tax lawyers can make any instrument you like and call it anything you like debt or equity, just make it deductible or else not.  So you really need corporate integration that doesn’t distinguish between debt and equity, and there are various ways of doing that.

Of course, these items would only be on the table in terms of a global tax reform effort, but I think they are going to be part of addressing these overall tax base problems.  Okay, all of that was really by way of thinking of the long-term agenda, where do we want to get to. 

What about the short-term agenda for a fiscally responsible AMT fix?  One thing is how would this happen, namely a fiscally responsible fix, if it did happen.  Well, I thought of a couple of similar questions, one from Monty Python is what if Queen Victoria could fly.  The other from Catch 22 is when some atheists argue about what God would be like.  Those are similar questions in the sense that we’re talking about something that I think isn’t going to happen. 

The problem is it requires politically painful regular tax base broadening and raising some people’s taxes.  If you have a revenue-neutral fix to the AMT, obviously every single person won’t be paying the same amount of tax as before – we still need to have the AMT for that.  So, if it’s revenue neutral overall, some will get tax cuts and others will get tax increases. 

That is not a good formula in Washington, because the people who get the tax increases mind it more than the ones who get the tax cuts like it.  Now the only way to do it is with Reagan-era-style bipartisanship.  Reagan and the Democrats, it’s well known, worked together a number of times in 1982 and 1984 to address the fiscal gap, 1983 to address Social Security, 1986 to reform the tax code, 1988, although it only lasted for a year or two, to do something important to Medicare.  In those days, the two parties worked together and would take painful poison together, so that neither party would get all the blame.  We’re not in that era today.  It takes two to tango and I’m not convinced that we even have one – we certainly don’t have two.

Well, what’s the prognosis?  I would say, not to sound like a Leninist revolutionary or something, but I don’t share their optimism, things have to get worse before they get better.  And unfortunately, that’s not to say that they will get better. 

The real question that we face, and I don’t know the answer to this, is whether responsible problem solving, instead of just focus groups and trying to get the other side to be blamed for it, is still feasible in the U.S. political system today?  It almost always has been – I don’t know that it is now. 

The final sobering note is that I am not wildly optimistic about this and the AMT is not even the biggest or the hardest test that we will face, when you think about the fiscal gap and all of that.  Well, thank you.

Alex Brill:  Thank you very much, Dan.  Alan Viard?

Alan Viard:  Thanks, Alex.  Well, Len’s paper, I think, has provided an excellent overview of the background of the situation and the 23 options that he has outlined provide a pretty good starting point for thinking about this.  What I am going to do today is address five broad questions, which arise when we think about the AMT.

The first question is whether we should repeal it or merely reform it.  The second is whether the cost of addressing the AMT problem should be paid for today.  The third is whether it would be better to broaden the base or raise the rates as a way of financing the relief.  The fourth question is whether there are some fixes that would be worse than the status quo.  And then finally, how should we proceed if no long-term solution is found in 2007? 

So, many of these questions have been addressed by Len and by Dan.  For the most part, my answers are going to be similar to theirs, although not in every respect.

My view is that we should repeal rather than reform the AMT, so I think we are probably in accord on that.  I view the existence of the AMT as a parallel tax system, as being the underlying problem.

Some people have formulated the problem differently, that you’ve got these upper middle-income taxpayers who are paying the AMT and as a result, their total income tax burden is higher than it otherwise would be.  And so they view the problem as being that the total income tax burden on those income groups is excessive.  Well, it that were the problem, then of course reforming the AMT would be a sensible way to address that.  Although you wouldn’t even need to do that, because you could lower their total tax burden in other ways by tinkering with the regular income tax. 

But I view the underlying problem as being the existence of the AMT as a parallel tax system.  It’s my own view, since I support consumption taxation, that no country actually needs an income tax.  But regardless whether you agree with me on that or not, we could probably all be in accord on the proposition that no country needs to have two income taxes.  And that, of course, is what we now have.  The second income tax adds complexity.  It adds distortions, because you have different people facing different incentives, which can distort the allocation of investment and other activities.

Furthermore, and Len has made this point, and Dan as well, the AMT design is flawed.  Even if you thought that having a second income tax system could be good under some circumstances because the regular system is bad and maybe you could have a second one that is better – and Dan was just mentioning his thinking back in 1986 that that could work – I think we are all in agreement that the current AMT design is deeply flawed.  It doesn’t measure up to that type of standard.  It does do some base broadening, but it is very ill-chosen, it is selective, it is misdirected and the marginal rates are comparable to the regular income tax.

I do see media story after media story which refers to the AMT as having two tax rates, 26 percent and 28 percent.  And, of course, there is a sense where that is technically correct – those are the two official statutory rates.  But as Len keeps pointing out and as I try to point out in all of my discussions of this topic, there are these 32.5 and 35 percent rates, so it’s not a really low flat rate tax.

So it’s my own view then that reform is not the way to go.  If you look at Len’s paper, which is in your packet, options 9 through 23 all deal with various types of reform, and so I would judge all of these as falling short.  They don’t address most of the AMT’s flaws, although some of them do address a few of them.  The AMT is really about as bad a system for those it applies to under those reform options as it is today.  In some of them, I would argue, even worse, where the capital gains and dividend rates are put back at full rates.

So the only virtue of those reform options is that it does reduce the number of people who are on this flawed system.  The system is still bad, but there are fewer people on it.  Well, that does seem like a step forward and of course, in some ways, it is.  It greatly reduces the complexity.  If you have 3 million, 4 million people facing a complicated system, that is better than having 23 million face it.

But the reduction in distortions is probably much more modest because the people who remain on the AMT under these reform options tend to be high income people, and therefore we have a large volume of economic activity being done by people who are still subject to the AMT.  You haven’t really taken off of the AMT the people who have the most economic activity that is being affected by the distortions that the AMT is causing.

So you are solving the complexity problem to a large extent, but you are solving the distortions to a much more modest extent.  And, in fact, of course, by taking most of the people off the AMT who would be on it, you are actually weakening the political support for actually repealing the thing and solving the problem. 

So I really have a not-too-favorable view of trying to pursue those reform options.  Obviously, one argument in favor of these reform options is that the revenue loss is smaller than if you try to repeal the tax outright.  And, of course, that is true. But if you look at the numbers, and again, Len’s paper lays this out very well, the revenue loss is still very large under these options.  Even if you go with the bare-bones form of relief of just inflation indexing the AMT exemptions, that costs two-thirds as much as full-fledged repeal.  And that is just really, I think, the minimal step you can do.  As Len adds in the other types of relief, the cost gets still larger and begins to approach the cost of full repeal.  So I think while we’re at it, we ought to just try to repeal it. 

Now there is a controversy, as the other speakers have mentioned, of whether we need to pay for AMT relief today.  Senator Grassley, as Len mentioned, has expressed one view on that, so I am going to join my fellow panelists in saying that we should pay for AMT relief today.

I phrase it that way because some people have posed the question should we pay for the AMT relief.  Well, that is really not the issue at hand.  The government’s long run budget constraint mandates, as a matter of economics, that any tax reduction or any spending increase has to be paid for.  It has to end up resulting in tax increases or spending cuts, either today or in the future.  So we don’t have a choice of doing this and not paying for it.  We do have a choice of doing it and not paying for it today, but that of course simply means that it will be paid for tomorrow by future generations. 

Yes, the argument is made that the spread of the AMT and the revenues it is producing was not intended.  Well, that is right, and that is why we want to get rid of the AMT.  But it is not true that the fiscal burden that we are putting on future generations under current law is smaller than intended.  On the contrary, we’ve got a very large fiscal burden waiting for future generations, and if we were to provide AMT relief today while putting the costs off to the future, we would just make that burden bigger.  So the fact that the AMT may not have been intended is really irrelevant to the question of paying for it today versus tomorrow.

So if we are going to repeal the AMT and we are going to pay for it today, we need $800 billion of offsets through 2016.  Of course, another implication that if we do make the tax cuts permanent, that there would be an additional $500 billion or so through 2016 – actually, it’s a bigger number if you go out another year – cost there in terms of how expensive the tax cut would be.

So, there are some options that have been ruled out in this discussion and I think for understandable reasons.  Some of them just can’t yield the money.  There is a lot of talk on Capitol Hill, for example, about closing the tax gap, going after all of the people who are not paying their taxes.  It’s a worthy goal, of course, to try to collect taxes from people who aren’t paying them, people who are illegally cheating on their taxes.  There hopefully are some steps that can be taken to achieve that goal, hopefully without imposing too many burdens on taxpayers who are complying.  But of course the amount of revenue that can realistically be raised from that isn’t remotely enough to deal with these issues.

There are some other approaches that are just politically infeasible, I think, in this context – maybe politically infeasible all together, but I think certainly just wouldn’t fly in the context of AMT reform.  There is a strong case to be made for cutting entitlements, but I can’t see that emerging as part of an AMT fix. 

There is a strong case to be made for a carbon tax or for increasing gasoline taxes.  But again, I don’t think that would be a viable topic to bring up within the AMT framework – the distributional tables wouldn’t look right, and so on.  So Len confines his attention to the changes within the income tax system and I think that’s a reasonable focus.

So the tradeoff is, do you raise rates or do you broaden the base?  And I guess again, for the most part, I will agree with my fellow panelists here that we want to avoid rate increases.  Some people have argued that rate increases are not that harmful because they think of the problem with high rates as primarily being that they discourage people from working.  And then they argue – well, you know, those effects just aren’t that big.  People wouldn’t really change their work effort all that much if you raised their rates.  Well, that may be right and it may not, it depends upon the people involved and there are different elements of labor supply that could be affected. 

But of course, raising a rate does far, far more than just affect the decision to work or not to work.  Right now, our federal income tax applies to a relatively narrow item called taxable income.  If you tax that taxable income at a higher rate, you give people an incentive to do more of anything that would lower their taxable income.  It gives them an incentive to work less, absolutely. But, it also gives them an incentive to increase fringe benefits instead of cash wages.  It is an incentive for additional municipal bond issuance, an incentive for more owner occupied housing, an incentive for tax sheltering.  You are magnifying all of the current system’s distortions.

The rate increases at the top tend to be more harmful than rate increases across the board.  That is partly because people at the top have more options on changing their behavior, but it would be true if that wasn’t the case.  If you have a rate increase that only applies to income in excess of $200,000, it has the same disincentive effects for people who are subject to it as if it applied to all of their income.  Yet, the revenue gain from it is smaller because their first $200,000 isn’t affected by it.  So you get a much worse tradeoff if you target the rate increases at the top. 

Now I think I am going to take a different tone than my fellow panelists here on the capital gains and dividend question.  I think that it would be a mistake to simply scrap the preferential rates for dividend and capital gains, either under the AMT or under the regular tax, because this is a way of avoiding or ameliorating the double tax on corporate income.  It is not the best way, I think that’s true – Dan made some excellent points on that – but to simply scrap this type of relief without some type of replacement, I think, would be a mistake.  So I would avoid that type of rate increase.

Base broadening is better, employer provided health insurance, a very big tempting target, owner occupied housing.  I think municipal bond interest should be put into the mix as well.  This is a smaller item, but I think the case for including it is very clear-cut.  To be fair, though, the transition relief may limit the revenue pickup because I don’t think you could apply it to existing municipal bonds, at least not in full.

And then, of course, the state and local tax deduction, I agree with Dan and Len, I think this is kind of an obvious or natural offset for the AMT since the AMT is taking that away anyway.  There are some tricky issues of really getting this right as to how you treat firm-level taxes to be consistent with how you are treating taxes collected from households, but I do think this is an option to pursue.

So when I look through the options that Len has presented in this paper, the information there is really detailed and thorough, there are some complicating factors.  The first one is just what Dan has called the instability built into the system, the odd timing that our current-law baseline now has because of these sources of instability.  And then, of course, the odd distribution of the AMT, the fact that it tapers off at the very top and hits more at the upper middle-income range also makes the distributional tables a little odd.

You know, I think that Dan is absolutely right that picking an option today is kind of pointless because who is going to listen to us, who is going to listen to me?  Well, probably no one, but I do think the best options in Len’s paper are four and five, which take on the state and local tax deduction.  Again, you have the details of all of those in Len’s paper, which is in your packet. 

The AMT is a big problem, but I think there are some fixes that are worse than the status quo.  I’ve argued that we don’t want a fix that isn’t paid for today, we don’t want to finance it by raising by marginal rates and we want to scrap the AMT rather than leaving it in place.  I do think that if you have fixes that really fail along some of these dimensions, you have to think at some point that the status quo would be better.  And then, of course, hopefully at some later date, a better reform option would emerge. 

I don’t want to say, of course, that we should reject any fix just because it doesn’t satisfy all three of these criteria, but I do think, as you weigh them against each other, there are certainly some combination of these failings that would give you something worse than the status quo.  Certainly not paying for the relief would aggravate an already dire fiscal situation. 

If there is no reform in 2007, no long-term reform, and of course there probably won’t be, should we extend the patch as people have conventionally assumed?  Well, maybe so.  I certainly think if there is an extension, it should be paid for, and that should be done without raising marginal rates.  I mean, the logic that I’ve been applying to a long-term fix would apply equally well here.  If we do keep patching it every year without paying for the cost today, obviously the permanent sequence of doing that is the same as having a permanent fix that you didn’t pay for today.

But even if it’s paid for without raising marginal rates, even at best, the patch still leaves the AMT in place and therefore avoids dealing with what I consider the real problem.  I hesitate to say this, but I do think at some point that what we may need to do is to have a year without the patch in order to actually bring home what the problem really is.  Thank you.

Alex Brill:  Thank you very much.  We’re going to float a Q and A and a mike is coming around.  I just wanted to ask one question, if I could, my prerogative as moderator, to Len.  Len, you had some good charts illustrating the effects of the Bush tax cut on the AMT, and I wonder if you could drill down at all to identify particular provisions in the Bush tax cut that seem to be the driving force in the growth.

Len Burman:  The main factors are lowering marginal tax rates, particularly at the top.  I’m not really getting into the merits of what the tax rates ought to be, it’s just a matter of arithmetic.  If you lower the tax bill for people who would be subject to the AMT or close to it and you don’t do anything about the AMT, you’re going to push them on the AMT and the people on the AMT are going to pay more. 

Capital gains and dividend provisions don’t make any difference because they are preference items for both.  Although one of the peculiar features about the AMT is that it actually raises effective capital gains and dividend rates in that bubble range, although it still preserves the difference between capital gains and dividends and other income. 

Alex Brill:  Thanks.  And if I could just ask that you identify yourself and the organization that you are with.

Kevin Hassett:  I am Kevin Hassett of AEI.  I saw, maybe in The Wall Street Journal, an analysis that suggested that the whole AMT problem was caused by the Clinton AMT tax hikes, rather than the Bush tax cuts.  It seems like a partisan point one way or the other.

But it did raise the question in my mind, weren’t you at Treasury when they increased the AMT and what the heck were they thinking?  I mean, because it does seem like one of the most imprudent tax policy maneuvers of the last few decades, right up there with establishing it in the first place.  So what is the political history, why did the Clinton people decide to increase the tax rate on the AMT?  And couldn’t we also, rather than thinking about fixing it on the income side if we can’t get repeal, fix it by just say reversing those increases?

Len Burman:  First of all, I wasn’t there when they did that – that was 1993.  Second of all, just as a technical matter, since there has been so much rhetoric on this, we looked at what would happen if you restored the rates and the exemption to what they were before the 1993 change. And the answer is that more people would pay the AMT because the exemption was a lot lower, although the amount of AMT paid would be significantly less. 

So what the 1993 Act did was it sort of followed a long tradition of having ordinary income tax and AMT rates move in lock step. So at the same time that they raised the AMT rates, they also raised the ordinary income tax rates, so it didn’t actually by itself precipitate a huge increase in the number of people on the AMT. 

You know, the debate about who should have fixed it and when – I mean, everybody back to 1969, that was President Nixon, right?  I mean, everybody should have fixed this.  If at any point we had either eliminated the AMT, not enacted it, or indexed it for inflation, the problem would be much smaller.  But the fact is that right now, we are left with this blob that makes no sense as a matter of tax policy.  We certainly agree with Alan that – you can argue about whether you want an income tax or a consumption tax, but one is really enough. 

Dan Shaviro:  One thing that I just want to mention.  Being on the staff in 1986, I have often wondered in recent years, why didn’t we index the AMT exemption amount for inflation?  And I think the answer is absolutely clear that if anyone on the staff – of course everyone on the staff knew about indexing, because they had indexed the regular rates brackets in 1981 – had said, let’s index the brackets, the answer would have been, you can’t do that within the five year period we are revenue estimating, because it would have made it too hard for the 1986 act to meet its parameters.

So the question is, what if someone on the staff had been – me for example – had been far sighted enough to say let’s index it starting in year six, because we are not estimating that.  It’s interesting to speculate if history would have been different.  Basically someone of a higher pay grade would have had to decide whether to let us do it.

I think actually the benign reason for not doing it was that in those days, we didn’t really think about changing tax laws six years in the future.  Now there are all of these absurd sunset types of rules.  In general, I think it’s a healthy instinct to say we’re not going to pass tax changes that don’t take effect for six years.  If we mean to do it, we are going to do it a lot sooner.  So I think it was a healthy instinct we had not only to think of it, although of course in retrospect, it’s really too bad that it wasn’t indexed starting in year six.

Alan Viard:  I want to echo what Len and Dan have said about how both parties certainly have responsibility for the situation that has developed.  I think that is definitely the case and agreed upon by economists.

Len had mentioned the pattern of changing the regular rates and the AMT rates in lock step, and that is a very sensible thing to do.  I’ve said that the AMT should be scrapped, we shouldn’t keep it.  But if we do keep it, one reform that might be considered is actually putting in some type of automatic linkage where the AMT rates would be a specified fraction of the regular rates.

We actually have a precedent for something sort of like that in a couple of places in the tax laws, where certain types of tax rates, like those on excess accumulated earnings of corporations, which only tax lawyers really know about, that rate is automatically set equal to the top individual tax rate.  And there are other places in the Internal Revenue Code where certain rates are set automatically equal to the top individual tax rate.  So any time the individual tax rate schedule is changed, that rate changes automatically with it unless, of course, Congress were to write something into the change specifically overriding that and saying don’t let it happen.

So I think you could do something similar with the AMT.  You wouldn’t set the rate to be equal to the individual rate, but you would set it to be some specified fraction of it.  Of course, Congress would always have the option to override that whenever they change the regular rates, but it would set up the default presumption that we want these two things to move together, which might be a pretty sensible approach.

Alex Brill:  Over here, and also in the front?  As we move to the next question, I will just add one additional historical note.  Last year when we enacted a patch on the AMT, it was a patch designed to hold harmless the number of AMT filers, while in previous years the patch had been to hold constant the dollar amount of the exemption.  We sort of lumped those together as the one-year patch approach, the band-aid approach.  There is a little bit, I think, of an important shift in policy where the slow growth of the number of AMT filers that had occurred in 2002, 2003, 2004, was ended in 2006.  The new standard has become the more expensive one-year patch. 

Suzanne:  Suzanne from the Capitol Insights group.  I was going to ask about the patch, because we talk about all of the criticisms of the AMT, but then you get the feeling that Congress does a patch each year, so it doesn’t really hurt anybody. But in fact, there must be money raised by the AMT each year, not everybody escapes it.  Or, have they managed to do a patch each year where after all of this analysis of who gets hit, they really don’t get hit?  And if there is money raised by it, is there a way, that in a Democratic controlled Congress, they can patch it differently than they might have if they were all Republicans?  So that, in fact, they get at really rich people or something.  Can you talk about the patches in general, in addition to what you were just saying?

Len Burman:  They’ve taken a pretty minimalist approach.  They have raised the exemption level – and that is actually the most efficient way to cut down on the number of people subject to the AMT in a particular year, because that is the entry threshold.  If they were going to do something like allow state and local tax deductions for one year or miscellaneous itemized deductions or other things, I think that it would raise the cost of the fix and I think it would be as politically difficult as – I guess it wouldn’t be as difficult as on a permanent basis – Alex probably has a better sense of the politics.

Alex Brill:  I mean certainly it’s the simple approach that has been favored.  It’s often been patched at the last minute and in conjunction with some other policy changes that have been going on.  I suppose that if you thought hard, though, you could create other patches.  You could sort of go after the preferences, either in a cap or a limited way, to sort of hold down the number, if that was your goal.

The higher threshold of keeping constant the number was an issue that came out of the Senate last year.  Of course, the irony is that it’s not keeping constant the same AMT payers; it’s just keeping constant the numbers.  But one could sort of experiment with other types of patches.

Len Burman:  I would just like to make one more point about the patches as an ongoing matter of policy.  And this idea of sort of retroactively fixing the AMT every year – Dan had talked about the instability in a tax system.  Right now everybody in this room probably ought to have $4,000 of additional tax withheld from their paychecks, and you probably haven’t done it.  But right now, the exemption for couples is $45,000 for 2007. 

If Congress doesn’t change the law, you are going to owe – if you are on the AMT already, you are going to owe $4,000 plus in additional tax.  And that is a really goofy way to set a tax system.  I mean, there is this bit of uncertainty that maybe somehow the game of chicken will go on too long and we’ll actually get to tax filing time.  Last year, the IRS actually wasn’t able to write into the instructions the laws that actually applied to taxpayers, because they had been enacted after the date when they had to send their instructions to the printers.  From my perspective, it’s a horrible way to make policy.  If you are going to extend it year by year, you should at least do it a year ahead so people can plan ahead knowing what their tax situation is.

Tom Mullins:  Hi, I am Tom Mullins at Sullivan and Cromwell.  Alan touched on this a bit, and I know it’s slightly off topic because we’re rousing the slumbering masses here today.  But a lot of people in the corporate community are feeling a bit left out as this debate goes on.  There is, of course, a corporate AMT as well.  So the question comes, is it necessary or do you see it from an economist’s perspective or even – Alex can answer from the political side – a way in which these could be linked?  Or does it just make the problem insurmountable in terms of the cost and complexity?

Alan Viard:  Yes, there is a corporate AMT.  It is interesting that the two issues are dealt with almost completely separately.  You could imagine that a different framework of discussion could have developed in which the two would be handled and thought about together, but that is simply not how people have looked at them.  I don’t know that it would be feasible to try to link the two issues in people’s thinking.  I’m also not sure that it would be desirable.  I don’t know that it would make either of them easier to address.

The corporate AMT is an ill-designed tax for many of the same reasons that the individual AMT is.  It’s a second tax system and, again, it just doesn’t make sense.  Yes, one corporate income tax probably is one too many, but two is certainly too many.  So, why have this parallel system?  You know, it’s especially odd. 

Len commented that, under the individual AMT, you have some timing items that are handled differently under the two taxes and those add a lot of complexity without really raising very much revenue in the end.  The corporate AMT consists almost completely of timing items of that type.  Enormous complexity is created and very little net revenue is raised.  In that sense, the policy argument for repealing the corporate AMT is actually stronger. 

Dan Shaviro:  The corporate AMT was thought 20 years ago to be a bigger part of the mix than it turned out to be.  They more or less gutted the corporate AMT by taking out the depreciation.  And again, it really was all about timing.  I think that anyone who thinks that there should be corporate base broadening, which normally means that you want to have it more like income-tax accounting and less like consumption-tax accounting, I don’t think that any of those people are really counting on the AMT to do the work for them.  So I think it is kind of a residual item that is just not very important – obviously if the taxpayers are on it, they will care – no one is particularly anxious to bring back into importance.

Alex Brill:  I would just add, from a political perspective, I think there is a little bit more of a risk than opportunity from the corporate side.  And the risk being that the pay-for for any AMT change, whether it be a patch or a more fundamental change, will be a tax increase on the corporate side, not the individual side.

Joann Weiner:  Joann Weiner – I’m now with Tax Analysts.  When I left Treasury back in 1999 for another planet – I moved to Belgium – we had a symposium for the NTA called Tax Surpluses from now Until Infinity.  I come back last summer and suddenly we’ve got this huge problem.  So, I sort of have a new perspective on this. 

But I am wondering, and Len sort of answered this question already, which was President Bush has given two of the three biggest tax cuts in history and yet you said that the AMT takes back 30 percent of those tax cuts.  My two-part question is, one are these the same people who got the tax cut and now it’s coming back?  And then the second question is there are a lot more people on the AMT than ever expected, but are these people paying more tax than they should be, or is this just a way of collecting the revenue the country needs?

Len Burman:  I mean, it is true that the AMT, were it not for the patch anyway, would take back a significant share of the 2001 to 2006 tax cuts.  With the patch, not many people are paying the AMT.  So the numbers that I was referring to in 2010, if the AMT just played out and was taxing 32 million people, that almost 30 percent of the tax cuts would be taken back.  About half of the people in the $200,000 to $500,000 income category I think would actually – I think it’s half of them would actually get no tax cut.

The question about whether the AMT is taxing the people it should be taxing, it increases the progressivity of the tax system.  It is mostly collecting tax revenues from people with incomes over $100,000.  And if you were just to repeal the AMT without any offsets, it would be another large tax cut mostly for high-income people.  But it doesn’t look like a progressive tax in the sense that most of us think of it, which is a higher share of tax as income goes up.  Very high-income people are significantly less likely to pay the AMT than people with just moderately high incomes.

It is obviously up to the political establishment to determine what exactly the right distribution of tax burdens is, but it’s a little bit hard to figure out exactly how you would design a tax system where the AMT would be an important part of your progressive tax mix.

Alex Brill:  In the back on the other side?

Jim Klumpner:  Hi, Jim Klumpner, Senate Budget Committee.  There is actually a lot of agreement amongst you guys about the analytics of this and so there have been a lot of comments on the side about political feasibility.  I wanted to invite each of the panelists to comment on what would be politically the most fruitful way to package this. 

Should we just be looking at fixing the AMT and paying for it by adjusting some parameter in other taxes, which are largely unaffected?  Should instead it be part of comprehensive tax reform, or should it be part of a global deal where you fix the AMT and entitlements and fundamental tax reform all at once?  What is the best way to put it together?

Len Burman:  I actually wrote an article for Tax Notes called “The Tax Reform Act of 2010,” in which we imagine political candidates accidentally getting into a position where they have to fix the tax system.  This is actually how we got the Tax Reform Act of 1986 – Ronald Reagan miscalculated and thought that Walter Mondale was going to propose a tax reform plan and then it actually played out and it was a significant tax reform. 

The thing is, you can’t replace – Dan made this point – you can’t replace an irrational tax with a rational tax and collect the same tax revenues from everybody.  Some people are going to pay higher taxes, some people less.

The AMT is only of the many significant structural problems in the tax system.  We’ve got a budget mess coming pretty soon.  The baby boomers are going to start retiring, we know that entitlement spending is out of control.  Some combination of meaningful entitlement reform and fundamental tax reform that would bolster up the income tax system, make it a system that more people would perceive as fair and that would be adequate to meeting the enormous revenue needs we are going to have in the coming decades, I think would be a good option. 

But it would require extraordinary leadership, as Dan pointed out.  It would require bipartisanship of the kind we haven’t seen in twenty years.  From my perspective and probably everybody here, it’s the obvious solution, but the hero who is going to lead this charge is not quite apparent at this point.

Dan Shaviro:  I am not in the business of giving political advice because obviously there are people who are more qualified than me to do so.  But if I were sitting on the Senate Budget Committee or advising people who are, I would certainly say it’s simply, obviously a waste of time to think about this year doing something like a bipartisan grand deal.  It’s just not going to happen - 2009, after the 2008 election, at the soonest.  I think the problem the Democratic leadership in the Congress faces is, is it worth doing a short-term patch of some kind and is there some way they can either make a bipartisan deal? 

A few years ago, under pay-go, there used to be something called the revenue predators.  Whenever someone had a nice giveaway, they would look for something that you can pick money up from that is politically and policy feasible.  Now I think it’s been pointed out today that the tax gap, although worth addressing, isn’t really a big enough pot of gold.  But is there some way of doing – even though not really in the long-term the most responsible to do, but doing nothing isn’t great either – a short-term patch with something that makes it kind of paid for.  Something that is politically feasible and it is not too awful substantively.  It’s that or nothing, I think really, because the long-term, the big deal is not on the horizon yet.

Alan Viard:  I think I agree with Len and Dan.  It’s hard to imagine anything long-term being decided upon in 2007 or 2008.  Maybe the political landscape will look different in 2009.  Our country will be facing some big decisions in 2010, since we have large components of our tax system essentially expiring.  That may be an opportunity for, as Len says, a Tax Reform Act of 2010.  At least we can wait and hope.

Alex Brill:  Any other questions?

Richard Rubin:  Richard Rubin with Congressional Quarterly.  If the pay-for were the getting rid of the state and local tax deduction, how much effect, or what effect, would that have on decisions by state and local governments in setting taxes?

Len Burman:  That’s a really good question, and I don’t know this evidence well.  But when I asked about the same question, the evidence is pretty ambiguous on the effect of deductibility on the ability of state and local governments to raise tax revenues.

I mean, the one thing that is clear is that you wouldn’t design a system of aid to state and local governments that lookd like the state and local tax deduction.  The states that receive the largest benefits are the ones with the big tax bases – high-income people and they can support higher tax rates – New York, California, Washington, D.C.  States like Louisiana and Alabama and Mississippi get next to nothing because they’ve got a very low tax base – hardly anybody, a very small percentage of their taxpayers even itemize deductions. 

The largest benefit goes to people in high marginal rates and those people who pay a lot of taxes, which is high income people.  I think it’s an issue.  I think that most people who look at state and local finances would say that for the amount of revenue we give up through state and local tax deduction, you would be much better off with targeted grants to state governments to provide needed services.  They wouldn’t mostly go to New York and California, but instead to the states that have really great needs.

Alex Brill:  One final question.  Anyone have a final question?  Just give a chance for the speakers, if they have any closing comments.

Dan Shaviro:  I guess I will just say, and will say this briefly in my remarks, just the one kind of political virtue of doing some of the state and local tax deduction is that if the regular tax repeal isn’t 100 percent, there is kind of a way for both sides to sell it to their constituencies.  For example, the Democrats obviously couldn’t tell their people we protected you if the state and local state deduction were 100 percent repealed.  But if it were partly repealed and the AMT problem were taken away, they could say, see, we just made a trade to keep it from getting any worse.  I think maybe that is one of the more hopeful ways to proceed, but that would require a bipartisan willingness to do business on this issue, at least.

Len Burman:  We actually did have a couple of options where we looked at replacing the state and local deduction with a tax credit, a refundable one.

I wanted to respond to a couple of the things that Alan said.  I mostly agreed with what he said, but one issue is when you look at the effect of repealing the AMT and raising marginal tax rates, Alan said that tax rates go up.  But basically you are replacing implicit high marginal tax rates, high effective tax rates with explicit higher tax rates.  It is probably true that at the very top, since millionaires don’t pay the AMT so much, that they would face higher marginal tax rates.  But it’s not quite so – I think if the Joint Committee on Taxation were trying to score the dynamic effects, that they wouldn’t find anywhere near as large an effect as say, just from an increase in marginal tax rates by itself. 

One other point is just on the issue of capital gains and dividends.  I mean, I think that most economists think that an integrated corporate tax system would make a lot of sense, but the current system we have is very, very erratic.  There are a lot of companies that don’t pay any tax and the relief they get from taxation under the income tax results in less than full taxation of their income.  And also a lot of capital gains do not reflect corporate income, but other kinds of profits.  Lawyers have spent a lot of time trying to find ways to channel ordinary income compensation that would be taxed at rates up to 35 percent into income that is taxed at capital gains rates of 15 percent.

That is why – there clearly is a valid issue having to do with the double taxation of corporate income.  But on the efficiency effects of having a lower capital gains tax rate, I really think it’s a mixed bag.  If anybody is interested, I wrote a book about this in 1999 called The Labyrinth of Capital Gains Tax Policy, and the rates are lower than they were then, but the story is pretty much the same.

Alan Viard:  Just to briefly respond, I think that actually Len’s views and mine are not very far apart.  It is absolutely right that the preferential rate for dividends and capital gains is not a perfect way to correct the double taxation of corporate income.  Len pinpointed one big flaw – the capital gains preference extends to capital gains on all types of assets, not just corporate stock.  And so to justify all of the preference, you would have to turn to a different argument about lock in or something like that.  So it is a highly imperfect fix for the double tax on corporate income.  I still maintain the view, though, that it would be a mistake to simply scrap that without bringing in some other replacement to address the double tax.  But, it would be better to do it in some different fashion. 

On the marginal rates, that is right – you would get some marginal rate reduction if you repealed the AMT.  So it’s not the same as if you were raising marginal rates to like pay for new spending or something.  But you would still have a net increase in marginal rates at the very top, which is where the rate increases are the most harmful.  And certainly it would be less economically beneficial than if you did do it through base broadening.  So I think again, we are probably pretty much in agreement on that – base broadening is better.  Rate increases of some sort are not necessarily fatal, but it would be better to avoid them.

Alex Brill:  Thank you very much for coming this morning and thank you to those speakers and presenters. 

[End Conference]