Talking to Arthur Laffer about taxes, taxes, taxes and Barack Obama

My column this week is about the persistence of the Republican canard that tax cuts raise revenues. Some tax cuts do raise revenues, of course, and many others deliver economic benefits that offset some of their cost. But it has apparently become required of Republican politicians at the national level that they speak as if tax cuts always and everywhere pay for themselves.

Denouncing this nonsense has become a standby for wonky mainstream journalists, and with the presidential campaign heating up there's been a noticeable increase in such screeds. The Washington Post editorial page has been on a tear lately about the "tax fairy." The New Yorker's normally mild-mannered Jim Surowiecki weighed in on "the great lie of supply-side economics." And Jonathan Chait of The New Republic has a whole book out that's devoted in large part to debunking the tax myth.

An excerpt from Chait's book ran in TNR in September, and caught the attention of Arthur Laffer, the economist who got this whole tax-cuts-pay-for-themselves thing going back in 1974. He wrote a 12-page response (warning: pdf), which inspired me to give him a call. A few quotes from our conversation are in my column--and the dead-tree version is illustrated with a brand-new drawing of the Laffer Curve that the great man rendered at my request, on a napkin. But I figured I should share more.

What follows is an edited and much-abridged transcript of our conversation. It's still really long. But there's entertainment galore. Reading through it, I see lots of points that I should have pressed him much harder on. He's just so effervescently nice. And I'm clearly too well-mannered to ever make it in this journalism business.

I noticed that there's been a lot of chatter in places like The New Republic and the New Yorker about supply-side economics. And I noticed that you had noticed too.

I did I did I did. It's fun, isn't it?

I understand their frustration with every Republican candidate saying tax cuts increase revenue, and the current administration saying that all of its tax cuts have been long-term revenue increasers. I'm not convinced that on every form of the tax code we're on the wrong side of the Laffer curve these days.

You shouldn't be, so you're correct so far.

But I am sympathetic with the idea that in the 1980s, the Reagan tax cuts, at least the ones on the top marginal tax rates, did have ...

That's unambiguous, that really is unambiguous. I mean, the amount of revenues in the top 5% that went up, it was just huge.

Do you think the Bush tax cuts ...

When you talk about the Bush tax cuts, which were tax cuts across the board, all the inframarginal tax cuts were dead-weight revenue losers, period.

Inframarginal?

Yeah, the nonmarginal tax rate. Let's say you make $100,000 and let's say we're talking just about you, Justin Fox, and let's say there's no chance on earth your income would fall as low as $50,000. All those tax cuts we do on Justin Fox up to $50,000 for sure are dead-weight revenue losers. You just lose the money from now until the cows come home. That's clearly true. I don't think you'll find many people who disagree with that. So the lowest end of the rate cuts, what I call inframarginal, those things do lose the money.

So you talk about a Bush tax cut, he had two sets, the first one I wrote not only wasn't going to pay for itself, but probably wasn't going to stimulate anything. The second one was a lot better. The second one was when he got panicked, as he probably should have.

Now there are parts of the tax cut he did, for example child-care tax credits, those are dead-weight revenue losses. There is no feedback effect on those at all, because they don't effect the marginal rate at all.

The ones that have the feedback effect the most are those at the highest bracket. I can give you an example of this. Let me use Kennedy, so I can get away from current politics.

Kennedy cut the highest federal marginal income tax rate from 91% to 70%, and he cut the lowest tax rate from 20% to 14%. He cut all the rates in the middle, too, but I'm just going to take the two extremes for you.

Before the tax cut, one second before the tax cut, a guy would earn a buck in that highest tax bracket, he would have to pay 91 cents in taxes, and he would be allowed to keep 9 cents after tax. That's his incentive for working. Now after the tax cut, the guy would earn that same pretax dollar, he'd now pay 70 cents in taxes and he'd now be able to keep 30 cents after taxes. The increase in that incentive for doing the exact same job for the exact same gross pay would be 233% .

Now let's take the lowest tax bracket guy. Let's assume that it's the guy that's on the margin. That guy makes a buck before tax, paid 20 cents in taxes, he got to keep 80 cents. That was his incentive on the margin for doing that work. Now Kennedy cut that from 20% to 14%. The guy went from 80 cents after tax to 86 cents for that exact same job. That's a 7.5% increase in incentives.

The top tax bracket he cut from 91% to 70%. That's a 23% cut in tax rates with a 233% increase in incentives. That's a 10-to-1 benefit-to-revenue-loss number.

Now you take the guy in the bottom bracket. The guy in the bottom bracket had a 30% cut in tax rates, from 20% to 14%, and he had a 7.5% increase in incentives. So that's a 1-to-4 benefit-to-revenue-loss calculation.

You can see why it's far more efficacious, why you'd far more expect revenue feedback to be positive in the upper brackets than in the lower ones.

And that's leaving aside the ability to avoid taxes.

The guy in the upper bracket intuitively has a lot more options open to him or her than the guy in the lowest bracket. So you're probably dealing with a far more elastic supply curve than in the lowest bracket. The guy could take a job in London. He can probably do a tax shelter with a lawyer or an accountant.

That's part of the argument on capital gains taxation, it's more flexible.

Exactly, that's why it's even more effective with capital gains.

But obviously there are diminishing returns as you get down that scale.

Of course there are. Now there are a lot of taxes I haven't included. For example, there are sales taxes, there are state taxes, there are auto taxes, blah blah blah blah. So the 91% is really not the marginal tax rate after all taxes. It's even more exaggerated.

I'm not going to push a lot harder on this thing, but that's why you find these enormous responses in the upper brackets. These guys fire their lawyers and accountants and actually pay their taxes. Yay! Isn't that what we want them to do?

But do you think that going from 39% to 29% [actually, it's 35%] is enough to have that much of a feedback effect?

Is it across the board, or is it just on the highest income earners?

That part of the Bush tax cut.

You're getting close in there, and let me talk to you a little bit about getting close. And I don't know the answer to these things obviously, so I'm just talking with you. Let's imagine we cut tax rates just on the upper income groups, how do you go about calculating what the revenue feedback is?

No. 1, the cost to collect a dollar from a guy is a far more than the dollar you collect from him. When you look at the government, when the government collects a buck, it's not free. They have to spend resources, the IRS, audits, all this sort of crap, to collect the dollar. I'm not assuming any Laffer curve effect here at all. There are just transactions costs of collecting that money. None of these guys look at that.

No. 2, in addition that, there are all sorts of costs of evasion, avoidance, underground economy, going out of work, moving to another location, or producing more or less. All of that goes in the calculation as well.

I'm just raising these as thoughts to you, Justin, not to give you answers. You've got to make up the answers youself. But if you cause that guy to work another hour by cutting his tax rate by 1%, what happens to all the people he employs, to the people he works with, all this other sort of stuff? You will collect more payroll taxes. You will collect more capital gains taxes. You'll collect more tariffs. Sales taxes, all these other taxes in the system.

You want to look at the overall long-term dynamic impact here. Let's imagine that I have a profits tax rate of 10%, and, based on that presumed tax rate of 10%, I build a factory. I finish the factory and on exactly the day that I finish the factory you bop that tax rate up from 10% to 90%. What do I do? Do I tear that factory down? Of course not. I operate with it. You've got me in an inframarginal position. But when the factory wears out, I don't replace it. You're going to collect a lot more revenues in the first year and as the years go on you're going to collect less and less and less. You follow me?

Let's take the Bush tax cuts, although I'd much prefer to stay with Kennedy because it's so far away no one gets emotionally involved. The question is, did the Bush tax cuts change output growth? Did they change retail sales tax receipts? Did they change state and local taxes? Did they change property taxes? If they did, how much? And how long did it take to materialize?

Now I'm going to tell you what I believe to be true. Bush put in his tax cuts in 2001 and he phased his tax cuts in, the tax cuts actually came into full effect on Jan. 1 2003. If you know they're going to cut tax rates next year, what do you do this year? You defer all the income you can, don't you? So what happened to GDP growth in the first quarter of 2003. It went from almost zero to 6.5%. [I just checked; it really went from 0.2% to 1.2%.] And we've just seen the last quarter, the third quarter of 2007, at 4.9%. Do you believe those tax rate reductions influenced the growth path? Then if so by how much.

I'd say yeah but I have no idea on the how much.

Me neither.

I've got to think most of it is just the economy had been down and thanks to low interest rates and other things it was coming back.

That's right. But I don't know what that number is.

The longer you're willing to wait, the more you'll get that feedback effect. I don't know whether it pays for itself or not fully. But I do know it pays for itself partially if not fully.

Let's say it doesn't pay for itself at all. Does that mean you should raise taxes to 100% of GDP? No.

The Laffer curve is an interesting concept, and it is a true concept. There are offsets that do take place in the system that make the cost of a tax cut less or much less or maybe even positive. And there are costs to tax increases that are much worse than people think, maybe they even lose money.

But when you look at the whole operation, the Laffer curve should not be the reason you raise or lower taxes. It is a consideration, but it's not the reason.

But Jude Wanniski really built it into something at one point. [Wanniski was the Wall Street Journal editorial writer who popularized the Laffer Curve and supply-side economics in the 1970s.]

I'm sorry, you know, you journalists have to control yourselves! I had been lecturing this stuff forever, and all of a sudden they liked that argument. It's not a wrong argument, it's a correct argument.

I would say every single economist I've ever talked to agrees with you on that.

They didn't beforehand, let me tell you.

Really?

You raised tax rates by 10%, you increased revenues by 10%. Everyone did it back then. That was totally true. You go back and look at any of the textbooks and you find that Laffer curve in them before me. Now, if you go back 100 years you'll find it everywhere. I just brought it back. I didn't invent it or anything.

There are so many considerations here. Raising taxes is not a frivolous venture that you do on the editorial page of The New Republic for god sakes. It's something that you really have to think about and go through carefully.

No one says that taxes are everything, but you know they're a pretty big part of everything. Taxes are not trivial, they're a huge portion of this overall economy. And that's why I focused on them. Some people may talk about lemon laws and reselling used cars. I don't think they're quite as relevant as I am. [This would seem to be a dig at Nobelist George Akerlof, America's nicest economist™.]

You moved to Tennessee because you were sick of California taxes?

Yeah, plus I also thought California was going straight to hell in a handbasket, which it has.

Tax rates aren't everything with regard to incentives to work. I would probably work at a 100% tax rate next to a nude modeling studio. I'm joking, but you know what I'm saying. There's a lot more to it than just tax rates. It's economics that I do, I don't do nude modeling studio economics. People do respond to taxes. They may respond to lots of other stuff too, but they respond to taxes, and that's what my work is.

The studies over the last 10, 15, 20 years have moved very much in my favor.

Within academic economics I think of Marty Feldstein ...

He's much better now than he was when I first met him.

But even he, starting in the 70s, was pushing these ideas of deadweight loss from taxation.

Yes, that's true. He did that and he was very good and I don't want to take anything away from him. He's much better now than he was back then, let me tell you. With the president in 1981 and '82 [Feldstein was chairman of Reagan's Council of Economic Advisers] he wasn't that good. He and Stockman were right together. Larry Lindsey was good, though.

I think of somebody like Greg Mankiw, who generally wants lower taxes but at the same time in his textbook he goes off on the Laffer Curve.

Why would he do that? Because I don't get it. I've never said all tax cuts pay for themselves. I never even said Reagan's tax cuts would pay for themselves.

The evidence I cited there on Kennedy, it's clear. Kennedy went from a deficit to a surplus with the tax cuts and the pro-growth policies. I do not believe that was luck. I don't believe that Reagan was luck. I don't believe that Bill Clinton was luck. I think Clinton did a great job as president.

But Clinton had, very early in his first term ...

One income tax hike cut that almost cost him almost everything. Then he became more Reagan than Reagan the day afterwards. He lost the House, he lost the Senate, he lost the governorships, he lost the state legislatures. And then he became more Reagan than Reagan: He got Nafta through Congress, against the unions, against his own party. He reappointed Reagan's Fed chairman twice. He signed welfare reform, that you actually have to look for a job to get welfare. He cut government spending as a share of GDP by 3.5 percentage points. No president ever has come anywhere near him on that. He had the biggest capital gains tax cut in our nation's history in '97. He got rid of the retirement test on Social Security. This guy was a great president and I voted for him twice.

You like him better than the current president?

I like Clinton and I like Bush as economic presidents.

What does your firm do now?

We have two firms. One is a research firm, a fairly large research firm on economics for institutions, pension funds, banks, trust departments, money management firms. And then we have a money management firm which is co-owned with General Electric.

There are all sorts of other issues in life besides economics. I mean abortion, pornography, Panama Canal, all that stuff. Iraq. Those aren't issues I'm into. I'm really into making this economy fly. I'm really into trying to relieve the poverty levels here in the U.S. I wrote enterprise zones god knows when, 100 years ago. I did that one. I have tried to remove the marginal tax rates be it at the lowest end or the highest end. My original proposals on enterprise zones were because the marginal tax rates on inner city dwellers are horrendously high. Every dollar more they make they lose all their welfare benefits. Which is stupid.

What Chait did in his article was he said the income disparity has increased dramatically. I'll stipulate that, counselor. There's no question that income disparity has risen. I don't mind rich people making more money. That doesn't bother me. What bothers me is poor people making less money. That bothers me a lot.

Now the question is, how do you raise the income levels of the lowest group? If you wanted to reduce income distribution discrepancies, let me go to the extreme, you would reduce it to zero. Everyone who made above the average wage, you would tax them 100% of the excess. And everyone who made below the average wage, you'd subsidize them up to the average wage.

That will reduce income discrepancies. That will. But you'll have everyone making the same at zero. And that's intolerable.

I don't believe that Chait and these other people are aware or care about that by reducing the incomes of the upper incomes you're going to lower the incomes of the lower incomes. That I really believe is true. That to me is far more important than any goddamned Laffer curve. I don't mind running deficits, if you make people better off. Do you?

It's what you're investing the money in. With Reagan it ended up looking like a good investment to spend all that money on the military.

And with Clinton he did it perfectly correctly paying down the debt. He didn't need the money. What Clinton did was he gave Bush the fiscal flexibility to do what was right, fiscally. By the time Bush took office on Jan. 20, 2001, we were in the midst of a real problem. The market had been crashing for a year. And then we get bombed 8 months later. What was the guy supposed to do? Raise taxes on the last three people working? He needed to stimulate the economy and spend for defense spending, and Clinton gave him the ability to do that. I mean, my hat's off to Clinton. As I told you I voted for him twice.

Was there any president in the last 30 years you didn't like?

Bush Senior I did not. But the last 30 years have been very wonderful years.

Did you work in the Nixon or the Ford administration?

I worked Nixon and Reagan. The only reason I worked in Nixon was my boss went there. I was George Shultz's right hand person. That's why I went. I wasn't a Nixon person.

I was the first chief economist at the OMB. I told my mom, "I just wrote a speech for Nixon, mom. He used it, verbatim. Well, he did make two changes: Everywhere I said 'is' he said 'is not' and everywhere I said 'is not' he said 'is.' But other than that, mom, it was exactly the same speech." Everything Nixon did was the antithesis of what I believe in.

[After this we headed off in other directions for a while, finally returning to current politics.]

My guess is you're going to have a Democratic president. My guess is you're going to have a filibuster-free Democratic senate majority. My guess is you're going to have very strong Democratic House control. You're going to have a lot of Democratic control of the state legislatures and also the governors. They're not the old-line Democrats, they're not my Democrats. They're not Wilbur Millses, they're not Jack Kennedys, they're not Lloyd Bentsens. They are going to put through a series of policies. What I would like to do is make a conditional bet with you as to what happens with the budget and the economy when they do it.

Whoever takes over is going do so with an economy coming out of a recession or at least a slowdown.

The economy has done so well. It's the best performing economy ever on earth. It's really hard to make it better. And yes, it'll be coming slower. But now, when you and I look at the future, and these guys do what they're planning on doing, I want to make a conditional forecast with you, and make a bet.

But there are a lot of people--I don't know about the Democratic candidates, but I do know about [Obama adviser] Austan Goolsbee ...

I know him too. And he's crazy on this thing. What's he talking about is raising the highest rates because it's a consumption tax cut. That's silly. Consumptions don't work.

I'm on Larry's show [Kudlow & Company] with him last Wednesday, and I love Austan, he's a great guy, great papers. And he's with Obama and he says, "Oh, Art, I'm doing a middle class tax cut by raising up rates to stimulate consumption." I said, "Austan what are you telling me guy? Consumptions don't work." You've got to make it more worthwhile for someone to give up leisure time to work, or to give up consumption to invest. You've got to make the supply of work effort more attractive, and low-end tax cuts don't do that.

He knows that. The question is why did he do the job with Obama? Because that was the girl that said yes.

And because he saw Obama as pretty moderate ...

Not from what I read. I love Obama and I think he's a hell of a neat guy and a smart guy. But he'll destroy the economy if he puts in what he says. Then I'll make you the bet.

I don't want to speak for Chait and others, but I don't think they're saying they want to go back to pre-Reagan tax levels. They're just saying that clearly it didn't kill the economy in the 90s to have a 39.6% top tax rate.

You've got to put the whole package together. There are things other than tax rates. There is monetary policy, there's trade policy, there's incomes policies, there's all sorts of things. I did not dislike Clinton even though he did raise the highest marginal rates. I didn't support that part of it, don't get me wrong. But what he did with monetary policy, what he did with trade policy and all these other things more than offset what he did with tax policy. And then after the first two years he was great on tax policy. And he was terrific on government spending. He was great all around. I think the only real mistake he made on the economy was that first-year tax increase.