The New World of E-Commerce Taxation
First Fridays: Seminar Series in Economic Policy

September 5, 2003

Unedited transcript prepared from a tape recording

9:15 a.m.	Registration
9:30	Introduction:	Kevin A. Hassett, AEI
	Presenter:	Michael S. Greve, AEI
	Discussants:	Daniel Shaviro, NYU School of Law
		Peter Merrill, PricewaterhouseCoopers LLP
	Moderator:	Kevin A. Hassett, AEI
11:30	Adjournment

Proceedings:
MR. HASSETT: [Inaudible] policy studies here at the American Enterprise Institute. And you are at a First Fridays Seminar Series on Economic Policy. I believe it is a first Friday of the month. That doesn't always happen with our First Friday Seminar Series. But it seems we've accomplished it this time.

We are very pleased to have a distinguished panel gathered here to discuss a very interesting new paper that's in your materials that's been written by Michael Greve, my colleague here at AEI. And the discussants will be Dan Shaviro of the New York University School of Law, who will also be spending a great deal of time visiting with us at AEI for the next year or so--we're very pleased to be able to announce that--and Peter Merrill of PricewaterhouseCoopers.

We're going to proceed by allowing Michael a half hour, or so, to present his work. And then we're going to give each of the discussants 10 to 15 minutes to comment on that. And then after, perhaps a little give and take to the extent that it seems like it's merited by the conversation, we'll then go to the floor.

And so if you have points of information during the presentation, feel free to interrupt. But if you have questions or objections, save them for the last half hour or so.

And with that, I will hand it off to Michael.

MR. GREVE: Thanks, Kevin.

Thank you all for coming. And I'm particularly grateful to Dan and to Peter for agreeing to comment on this paper because they're somewhat in contrast to me. And I'll get to that point. They're actual experts on this.

As everybody knows, the term sales and VAP [ph.] tax system is principally based on destination. That is to say goods and in the case of your services and tangible goods are subject to taxation at the rate and on the base of the customer's home sale. What this paper that's in these materials does is it tries to make a case for reversing that arrangement. We're saying let's tax goods or services, as the case may be, at their origin. That is to say the seller's home state or home jurisdiction. And let's do that for all sales through all channels, internet, catalogues, spot sales, anything.

The paper basically makes two points. One is negative and the other one is slightly more positive. The negative point is this. That I think the destination-based tax system is broken. It was broken long, long before the internet. And I believe it cannot be made to work. And the positive point is that I think origin-based taxation is quite attractive in its own right. And it certainly beats any kind of destination-based system that you might be able to practically construe.

And some of the arguments I, some of my arguments in this paper are sort of economic in nature and efficiency oriented. And others have to do with political institutions. And what I'll do this morning is just briefly describe these arguments, and then pipe down.

The first point, the destination-based tax system is broken. I think that point is actually quite widely acknowledged, in fact, almost universally acknowledge. The first criticism that is generally leveled at the existing systems. And that goes for the United States and for the international context alike is that the system imposes extravagant compliance and administrative costs. And that is because, of course, sellers have to comply with a tax with conflicting and varying tax regulations in multiple jurisdictions, thousands, in fact. And that's just very, very hard to do. And it's particularly difficult in an electronic context where you have no, sometimes have no reliable way of ascertaining where your customer, your buyer actually lives, and is taxable.

The second broad set of arguments that is leveled against the existing systems has to do with the inequities and the distortions that are inherent in the system. The system is not neutral for sellers or buyers. It variously, depending on the rules, favors one sales channel or industry over another.

In the United States, of course, the famous hang up or the famous inequity is the Quill Rule and the Amazon book sale that's not taxable because it comes over the internet, but the local book sale that is taxable. That argument, of course, has been with us since the beginning of this debate. And we can't seem to get rid of it. And the reason I think why we can't get rid of it is that there's a great deal of merit to it. We should not artificially favor one sales channel over another.

And for buyers the system is not neutral. Even if you have a fairly rigorous destination-based regime, there will always be physical cross border shoppings. People will travel to Delaware, no matter, and buy tax-free goods, no matter what you do. They'll get on the bus. They'll get in the car. And that invariably introduces an element of origin based taxation into the system because, of course, those kinds of purchases are taxable or not taxable, as it were, in the state where they are made and not in the state where the goods end up, at least for practical purposes unless we start arresting people in the Holland Tunnel.

And the same set of problems, and I won't detail them, but the paper talks a little bit about it, the same sets of problems you get in the EU. There, of course, the debate is over the taxation of services and intangibles. And it's now found sort of its, its--or taken a big step forward, so to speak, with July VAP directive issued by the EU. But the problems you get, and which I'll discuss for purposes of exposition here only in the American context are structurally very similar.

And after those arguments, the one about the administrative and compliance cost and the other one about the distortions I'll have two more arguments that aren't economic in nature, but that gravely concern me. Because what I actually do here at AEI is I study institutions. I don't study taxation, per se. And my concerns are institutional in nature. One is that I believe that the internet tax debate, both and in the international context, is meant to set a precedent for a broader range of issues, both on internet governance and on taxation.

I'll give you one example. Privacy regulation, especially on the net, there, I think the only principle that makes sense, the only principle that makes sense is origin regulation. If you start subjecting electronic transactions to destination regulation in that context you're basically killing the market. That's sort of the information stone age. That hasn't deterred the European Union, but I think it should deter us.

Now, of course you could say that, well, okay, fine. I'll give you that. There the origin principle makes sense, but that doesn't mean it necessarily makes complete sense in the taxation context. That argument, of course, may have a certain theoretical force. I just despair of the ability of our political institutions and especially international institutions to make those kinds of fine distinctions. And that being so, I'd rather stamp my foot and say, let's, let's take a stand on the general principle and see how far we can push it.

And similarly in the domestic context, it's quite clear that the states are the mobilizing forces behind the SSTP want more than simply full destination taxation for remote purchases. They want to have the same sourcing rules for business, income taxation. For example, I think that will be a very great mistake. But there, and there again, you could make a theoretical distinction and insist that, no, no, no, we should have different rules for these kinds of systems. But I think it's just very, very hard to make that sort of thing stick. And therefore, there again, I would rather stamp my foot and say, take the principle, see how far you can push it. Unless something really bad happens as you look at it, let's stick with it.

My second institutional concern is that destination-based taxation invariably reaches very, very deep into other states or countries affairs. You have to, in fact, enforce tax collection in other states, in other countries. That's not taxation without representation, but it is tax collection obligations without representation. And it gets no better because the politicians collude. These deals are becoming more and more frequent, common in American politics. The States say to one another, you regulate my guys, and I regulate yours. And the outcome of those kinds of intergovernmental arrangements is invariably that all the politicians are better off and all the citizens are worse. And all those, and, and I'm nervous about all of those things.

I'm not saying, I'm not saying all of those arrangements are necessarily bad and evil. Some may be necessary. But what I am saying is we ought to start in those kinds of situations with a very strong presumption of nervousness, which is what I'm doing.

Now, for the first two problems, the compliance costs and the distortions, the proposed, the usually proposed fix is this. First, reduce the administrative and compliance cost through some, some combination of simplification and application of new fangled technologies that are taxed software. And second, we should abolish the inequities by capturing all transactions through all sales channels. That's, of course, what the overruled Quill assertion is about.

And that's the basic agenda of the OECD, of the EU, and of the SSDP. And in my mind that is a fool's errand, because it will make things worse. The paper lists a whole, rattles off a whole slew of reasons why that might be so. I'll just give you the most important line of argument. The only equilibrium point on our destination-based tax system is perfect collusion among all governments, because otherwise you get free-running governments that don't play the game. And then you get the distortions again.

Now, what does that mean, what do you actually need to--you have to enforce perfect collusion. You can't just rely on spontaneous cooperation. Internationally that means that you need something like the United Nations with teeth for the tax context. There are actually proposals out there to do just that. And to that, to those proposals I have a very differentiating answer, which is no, under no circumstances.

In the U.S. it would mean you have to mandate state participation. That is ultimately what the SSTP wants from Congress. And you'd have to harmonize the tax base, even if not necessarily the tax breaks. Now, is that an option? It's effectively what are you then talking about is a national sales tax and give the money back to the states. Is that an option? Yeah. It may even be a serious option theoretically speaking. But it's not politically a serious option because the states don't want it. And that's the end of the debate on that front.

So if you don't have enforced cooperation by Congress you have to figure out some other way to do it. That, of course, is the SSTP, that is the states themselves get together outside Congress and try to establish some regime that guarantees cooperation. But now you've just bought yourself a whole set of new problems. First, the system is not universal. And, in fact, the SSTP is not universal. There are some states that haven't joined and there are some states that will never join. And that means you lose half of the efficiencies that you hope to gain in the first place.

And the second thing is that the political factors, the political forces that account for all these variations and all these complexities and local and state tax systems aren't going to go away just because somebody shouts SSTP in a crowded legislature. They'll continue to be operating. So even if you could harmonize the system once, it will deharmonize itself later on unless you entrust this to the SSTP with the authority to make binding decisions on an ongoing basis, and you don't know in advance what they might like. And, in fact, the [inaudible] SSTP, which isn't even in fullblown operation yet, has already done so. Because unbeknownst to most of its members at the front end it has since said that, oh, the sourcing rules that we want here for the purposes of sales and use taxation should also govern business income taxation.

That is a very, very dramatic step and I think one that raises gave problems, not just, I mean factual problems about what, what should we do about business income taxation, but institutional problems. It's not very good, I think, to have this junior elite congress out there making decisions on behalf of some states and for other states.

And if you don't give the SST--now, there are two problems then. First, the authority to make binding decisions operates at exactly for the states as a deterrent for joining this beast, because you have to be extremely nervous about what it might do and you have to be extremely nervous about surrendering your tax authority. And the second problem is if you don't entrust that body with that kind of decision, you're not going to get lasting simplification. What you're going to get is more complexity because the sales tax system, as agreed upon by the SSTP, now operates over a broader range of transactions and goods.

And in light of all of that my conclusion is scrap the whole system. See how that far that gets you. And reverse the principles and say, let's tax goods delivered over any sales channels at the point of origin.

We already have that system, of course, for a large number of transactions. We have it for local sales. We have it also for interstate flower sales. If you go to FTD or do one of those transactions, you tax the base at the local sales tax rate and not where the flowers end up. We even have it under the SSTP. The Great State of Texas has joined the SSTP for all of the state and said that we agree with the destination-based rules OF the SSTP except for the City of Round Rock where the origin principle shall prevail. You may not know where Round Rock, Texas is, but I can tell you who is there, Dell. And those kinds of pressures will operate, no matter--excuse me--what you do.

So you have all these origin-based elements in the system. What happens if you extend the system? Well, one obvious advantage I think you get is the system all of the sudden becomes extremely simple, relative, that is, to a destination-based tax system, because every transaction is taxed only once or maybe twice if you add local taxation, namely the seller's home jurisdiction. I bet it's not quite so simple when outfits have sales locations, as many of them do, in other states, then the tax to them--I think those kinds of local sales have to be taxed at those tax rates. So, in other words, if you're Walmart, you're going to have to file 50 returns no matter what you do.

But that's not where the problems lie. That's generally manageable. The problems really lie in, in the taxation of internet, of interstate sales and there the system would greatly simplify what we have now. And, in fact, I mean, if this council is proved, you can't prove something that doesn't exist, so I've just talk to trade representatives, and they all say, well, of course, I mean, it might be complex, but it's far, far better. If we could get there, we would take this bargain in an instant. And I think that's a pretty good indication.

And the second thing that you would get, the second advantage is that governments would stick to regulating their own firms and their own citizens, as opposed to firms in other states. And I think that sort of reduction on intergovernmentalism is worth having for its own sake.

Now, what are the objections to this regime? I think there are two. One strikes me as fairly unpersuasive and the other one is actually very serious. And I can't say that I've sort of fully come to grips with it, but I think I've for the most part come to grips with it. The unpersuasive objection is the states and the cities of these mayors who run around with sort of little fire trucks and say, unless we tax the internet or internet sales, there will be no more fire protection at the local level. If you roll this out and confront them with the idea they'll say, there will be a race to the bottom. There will be a revenue loss that will make your teeth rattle. And we simply can't have this.

I think that would be true probably if you had a true choice, true free choice of law regime. That is to say if the tax regime worked corporation law where you can pick a Delaware charter no matter where you actually operate. In that case, I think if you make certain additional assumptions, the tax equilibrium is, in fact, zero. But that's not true for the rule that I've proposed, which depends on the seller's actual physical home state as defined by the uniform code by the UCC. And the reason for that is that firms choose their jurisdictions and their locations and states compete on a zillion margins. This is just one more margin on which they compete.

And under those kinds of bundling conditions, can you actually tell what the effect will be exactly. Well, you can guess, but you can't know. And I don't profess to know. But I think what that means, but I think what can safely be dismissed is this notion that it's an evolution of the sales tax.

The serious argument is derived, I think, from efficient tax theories. And Dan, I suspect, will talk a lot about it. And that's the argument that, look, taxes should be locationally neutral. Whereas, of course, under an origin-based taxation system the local tax rate operates as a kind of factor endowment for the firm. It becomes a decision-making factor in deciding where to locate, and so forth. And there are certain inefficiencies and losses associated with those kinds of regimes.

As I said, I think that's a serious objection, but I think it's ultimately not compelling. And there are two reasons why, I think, it--why it fails to dissuade me. One is that it may be that concerns over tax efficiency and locational neutrality are a bit overrated. It's--other words in a lot of ways for tax competition and states now compete over taxes in other ways for citizens. And it's hard to see for me why competition for firms is necessarily bad or destructive, or why it should be bad or destructive on this margin but not on any other margin. We generally suppose that state competition for citizens assets and talents is by and large efficient, by and large preferable to centralization and state cartels, to say nothing of a regime that would let states choose their own citizens.

You have to be careful in what you claim for that. It's no more than a presumption. But I want to start with a presumption and not abandon it.

And the second reason and the more, to my mind more forceful reason why I'm a little--why I'm not fully persuaded by efficiency arguments in this context is that destination rules won't produce locational neutrality either. As I said, there will always be the mix. And what that means in sort of theoretical terms is that you can't decide the decision, I mean, you can't decide the balance between destination-based and origin-based rules on purely theoretical grounds. And what means in turn is you better take, I mean, the sensible step is, to me, to say, look, let's pick the system that overall minimizes the administrative and transaction and enforcement cost because that is reasonable and rational no matter what else on any grounds.

So that's the sum and substance. The paper concludes. And I'll conclude here on a pragmatic note. The deal I propose is, of course, politically impossible certainly at the national level. But what I hope one can imagine is something like an origin-based sales tax project among a few brave states that have misgivings about the SSTP, Colorado, Georgia, places like that. Let them implement those things and then you can study what actually happens in the real world. I think the internet debate, internet tax debate has for a long, long time been rather sterile. And maybe what we need is not more theoretical arguments. Maybe what we need is an actual experiment to break through the sterility. Thank you.

MR. HASSETT: Before I pass off to a discussant, I have a couple of clarifications I'd like. In a tax competition model usually a lot of the set up, when economists are doing this, depends on where you start. And so very often it's sort of we start where everybody's got a tax rate of, say, 10 percent, and then we let competition go. And then maybe competition drives it down to 0 or you get an national equilibrium where everybody stays at 10 percent, or, you know, varying around 10 and then they focus in on one and you get multiple equilibria where they all agree to be 10 percent or they all, or they compete down to 0, those are the two things that generally happen.

But I've not seen any interesting results in any of these models when you start where everybody's 0 and then you introduce competition. And, and, so that's sort of the set up. So right now I don't think there are any states that, say, with internet sales apply an origin based tax. Is that correct?

MR. : That's correct.

MR. HASSETT: And so, now, like so how would it start? How would it--so, so, because, you know, any state right now says okay, well, we're going to, you know, have internet sales tax at 5 percent, then it's probably not very likely that that could sustain just cause no one would try to sell from the internet from that state. And, and so they're sort of stuck at 0. So you'd have to start the thing by having some kind of central authority dictate to everybody that, okay, have this origin based tax. But then the whole point of having the competition was not to have the central, central authority.

MR. GREVE: Now, well, I mean, what we have right now is, is basically an export subsidy. That is to say before any exports from the state, the sales tax is 0 even though your sales, the same corporation sales in that state may well be subject to a sales tax. Now, if you have a destination-based system, that's, of course, a rational part of the system. That's the way it has to be.

Once you move to an origin based tax system by, I mean, through, as I suggested at the end, reciprocity, I'm not sure what would happen. It could be that states continue to say, okay, we'll have that. We'll continue to tax exports effectively at a 0 rate for all sorts of industrial policy reasons or for other reasons. But it may also be that, to my mind at least, I mean it may be that they rethink the system because it's no longer a part of a larger sort of back drop of destination-based taxation. And because it may act as a spur to sort of take a broader look at the sales tax system in general.

The general trend in sales taxation over the past decades has been to raise the rates and to narrow the base. And a lot of states know that that makes no sense. It's just politically difficult to deal with. So if you have some exogenous event, like, you know, okay, fine, now we'll agree with some other state to try this origin-based sales taxation, maybe the political process changes. Maybe the look at it and say, okay, and maybe they'll sit up and say, okay, what we ought to do is broaden our base. And that includes now exports and lower the rates.

MR. HASSETT: Could we have--would you support, just again trying to focus what we're debating later on.

MR. GREVE: Right.

MR. HASSETT: Would you support a federal law that said you can tax internet sales, but only on an origin basis? That would start this game, right? That's one way I could think of starting this game. Say okay, go ahead and tax the internet sales, but it's got to be origin based.

MR. GREVE: That would start it in a real hurry. What you, the vehicle, of course--or the vehicle there, I suspect, I should not say of course--the vehicle I suspect that would lend itself to that is to start against the theoretical assumptions of destination-based taxation and then say let's have a very, very, very tight nexus standard. Which is the same, right? And if you say, okay, it will be destination-based, but states may tax only those outfits that have an actual, physical sales venue in their state. Warehouses don't count. Goods ending up there don't count. None of that counts. That's the essential equivalent of an origin based taxation under grever [ph.] rules.

And that vehicle exists. And if Congress could bring itself to that thing, I'd say hallelujah, more power to them. I just don't see it. Which is why I have this sort of little experiment about sort of six brave states.

MR. HASSETT: Okay. Thanks, Michael.

And I guess we'll go in the order of the program. And Dan Shaviro is up next.

MR. SHAVIRO: Thanks, Kevin. And thanks, Michael, a very interesting paper. And let's see if I can get this launched here. Okay. Here we go.

I'll start my comments, let's see if this works--um, I think we're already--this is an interstate import from New York by the internet. Ah, there we go. Okay.

Starting with some important points that the paper makes. One is the significance of compliance complexity. I remember when I wrote a paper for AEI on state and local taxes 10 years ago, there were like 50,000 sales tax districts, including mosquito abatement districts and the like--that's certainly an important point--and the difficulty of reducing it through voluntary harmonization. And one problem I talked about in that paper way back when was that there are all these little government units that the people there just enjoy having the power to decide things, you know, whether it's a crass story or they just like being powerful or important.

I remember it's a well known point that different jurisdictions with different political preferences would want to set their tax rates differently, for example. But the sort of proliferation of tax bases, it was hard to see really where the value, it's hard to tie this to consumer preferences, the voters, and think something really meaningfully important. You know, New York, they don't like S corporation rules in their corporate income tax and in Oregon they do. So I, I had--I think the fact that just these, all these places people just enjoy exercising political power and it imposes administrative costs nationwide, that is a real problem. Makes it hard to coordinate.

An important caveat is that the locational distortion matters too. Michael does say that, but there's a, well, I guess I'll pester him a little bit about a thing about how--in the paper, about how this is like a thing from MIT professors blackboards. You know, these MIT professors have blackboards and they think this matter is--but isn't this just too theoretical.

On the contrary, it's a very practical concern. You know, if Amazon and Kramer Books have different costs of production, you buy from the "wrong one," the higher cost one because of a tax difference, obviously you're getting a classic inefficiency. And I realize the end of this story is more complicated, compliance causes part of it too. But the contrary issue is certainly an important one.

The second point Michael notes that harmonization is kind of, creates cartelization. The government is now able to get everything. And do we want local power projected outwards this way? He talks a little bit about the regulatory example of origin versus destination. And obviously you can imagine cases where we'd be uneasy even about the enforcement cost that one jurisdiction wants to impose being sent to other jurisdictions. So that's a good point too.

But it is true to provide public goods, by definition you can't charge the beneficiaries for it. You have to have cartelization of some kind. You have to have some sort of market power to impose taxes on them or else public goods can't be provided. Also an important thing not mentioned in the paper, but it's certainly part of the overall story with something like retail sales taxes including those that are destination-based is the Teabow [ph.] story where local jurisdictions compete for residents by offering tax benefits packages. New York City if they want to charge me a 95 percent sales tax is probably going to have a hard time getting me to live there unless they provide really amazing benefits.

So jurisdictions that run destination basis sales taxes or property taxes and other things, if they, if they're taxing their residents and providing public goods to their residents, they are to some extent in competition because you can choose to live somewhere else. And I think that's an important part of the cartelization and competition story that has to be considered.

The next question I asked myself is what are origin basis and destination-basis retail sales taxes? I've recently, you know, acquired some interest in the international story in the VAT income tax things, aspects of it. And so I wanted to compare it to VATs, which generally are destination-basis. One reason why the VAT case matters in the U.S., one reason why it might matter in a tax reform setting is the flat tax, as is well known, is basically a VAT in which they've thrown in a deduction for payments to workers and the workers included, the reason for that simply being to apply two rates, a zero rate, and then a positive tax rate at the business rate typically to workers so you can have some rate graduation. The David Bradford's X tax is basically, apart from the way he's developed it a little differently, is basically a flat tax with more rates. You could have, for example, a flat tax. You'd start with a VAT and you'd turn it into a system with 0, 15, 25, and 35 percent rates, let's say. And you could have something with a very familiar looking rate structure that could achieve familiar looking distributional results. You'd have a lot of, it would be a conception tax and have a lot of the administrative advantages of simplification.

Anyway, now when people talk about those things, the main focus is on interbusiness transaction. It's not the exclusive focus because sales to consumers matter too. But interbusiness transactions are always what people are talking about. And the rule is that, you know, origin basis businesses include the money that they get from exports and they deduct the dollars that they pay for imports. And destination basis you do neither. And it's often said, you know, in sort of a VAT or consumption tax choice, the two are equivalent apart from administrative differences in transition.

Why is this? Basically, it's--well, you could put it several different ways. But one way to put it is that you have people living in the United States why do they pay for imports or why do they pay for--how do they pay for--why do they sell exports? To get money they can use to pay for imports. In the long run the cash flows in theory should be equal. The money that flows out comes back in again, or the money that comes in flows out again. So in the long run the exports equal imports, a sort of well known international trade principle. And the two are equivalent apart from administrative differences, which I'll get to in a moment, which are quite important, and transition issues, which I think I'll completely duck for today because they're somewhat complicated.

A further point, income taxes are always origin basis. One way of putting the reason for that is that in a consumption tax you can say, well, we're either going to include and expense--we're either going to include the money we get from when we export and expense the money we pay for imports or neither. But in income tax you start having to, on the expense side, you start having to capitalize inventory or amounts paid for equipment, and so forth. So in income tax you have to be origin based. You have to include the cash flow both ways so you can decide which outlays are capitalized rather than being expensed. So then, sort of a footnote, yeah, well, they're equivalent, but you can't only do one of them for an income tax.

Now, when you're thinking about this in the VAT world, the big administrative difference is the origin basis--and this, of course, applies to income tax as well--requires transfer pricing. If Toyota/ Japan sells a car to Toyota/U.S., which then sells it to a consumer, you have to figure out the price that Toyota/U.S. paid Toyota/Japan. Of course, as an economic matter Toyota, the Toyota umbrella doesn't care at all. So the nightmare of origin basis taxes is transfer pricing. And anyone who practices in international tax knows what a problem it is, which consumes, certainly, a lot of resources. Although to the benefit, I guess, of economists who get to be expert witnesses.

Destination basis requires policing the border. You have to observe the imports. You have to figure out what really are exports. And there's the so-called tourism problem. Because in theory destination-based tax, really destination is not the U.S., but it's American consumers. So if i go to Europe and I start buying consumer goods, in theory destination-basis tax rationally defined we'd probably want to reach that, that's going to be pretty hard to do, just like as it might be hard if I download something to my computer while in the U.S. So you have kind of the offsetting administrative problems. And either one is bad, but you have to choose your poison.

Then you have the economic equivalence points. And, again, we're under these VATs, and the like, including the flat tax and the ed tax tax. And it's often said the origin basis taxes national production. We--you, like the current income tax, if you sell--if you produce something here and sell it to France, you're still taxed on what the French paid you for it. Destination basis taxes national consumption. So you can tax either one. Who cares which one it is? These are equivalent in the long run. Defined in terms of residents being mobil individuals. And the idea is that in the long run what you get consume is equal to the value of what you get to produce. Hence, the idea they're sort of equivalent under various other assumptions, such as tax rates remain the same.

This is why consumption tax can use either one as in a wage tax if the wages are defined broadly enough, or sales tax, or retail sales tax. Again, a consumption tax, even if you think of it as taxing national consumption, since in the long run national production is equivalent, you can kind of do either one. But in income tax you kind of care about whether production comes first before consumption. And you hold wealth for a while. You kind of want to have the tax depend on that. So that's why an income tax has to use one or the other.

So I put this out here because you talk about origin basis and destination basis and you're talking with people who are kind of familiar with these like VAT type issues. You have these--or with tax reform, like, let's put in a new consumption tax to replace the income tax. How should we do it for cross border transactions? This is the set of things that will always come up as what you think about.

Getting back to the RST, because we're talking about retail sales tax here. Well, all of the sudden we have a pretty different picture. The origin basis no longer requires transfer pricing. If, even if Amazon opened an outlet store in, in Washington, D.C., you wouldn't have to figure out the price that Amazon out west charged Amazon, D.C. to open this store for the goods they put in the store. Rather, you would tax the full value of the retail sale made in the hypothetical Amazon outlet store here. So basically what happens is the sale to the resident consumer is taxed in full or not, or it's not taxed at all, depending on a nexus issue. There's no--what transfer pricing does is it isolates the in jurisdiction production, the basically added value of the store running its services in the VAT context. But here you're going to tax the whole thing, or none of it, depending on whether there is nexus. So it's a little different.

Destination basis still requires monitoring sales by outside businesses to residents. And so this is really--underlies Michael's point that, gee, it looks like it's a heck of a lot, you know, in VATs no one would say that it's just obvious off the bat that origin basis is simpler than destination basis. Here, Michael is saying that really because you wiped out the transfer pricing issue. Now, so there's a reason for having a transfer pricing issue. And there may be other problems that emerge instead. But that's kind of how the administrative comparison starts looking differently.

Then we start asking, well, are all these things still economically equivalent now that we're talking about a retail sales tax? Destination basis still is taxing local consumption as defined by individuals' residence. And, again, in theory you'd be getting, you know, if you're a tourist and you go to France, in theory we'd still be getting VAT. Under the destination basis retail sales tax obviously no way. Of course, in the U.S. context states try to do it through use taxes.

But what exactly does the origin basis tax? And with a VAT and a transfer pricing we show it kind of taxed local production or production by ignoring problems like corporate residence, which are kind of messy and don't make a lot of sense in some ways anyway. You, you're taxing the production by individuals under an origin basis tax. But here, what you're taxing is consumption by anyone at a store that's within the jurisdiction. Or, I should say they don't have to consume it at the store. Obviously anyone buying, anyone at all who buys a consumer good at the store within the jurisdiction. And also apparently you're taxing sales that are, come from the production side within the jurisdiction if they are remote sales. So that's your tax base.

And I'd say, maybe this can be rationalized administratively. This is the argument Michael's making. But it's hard to, it's hard to see it as a coherent and rational tax base otherwise. So as a starting point, sort of a pure MIT blackboard, professor's blackboard quibble, it doesn't seem coherent. It doesn't seem to make sense intellectually. In fact, there is no way to make sense of it really except administratively.

[Inaudible] to say, well, this doesn't matter. You know, let's, let's not be MIT professors, even if a number of them have deservedly won the Nobel Prize in Economics. And does this really matter? And the first point I'd say, well, you know, suppose, you know, there's fixed revenues. Obviously not, in fact, the case here. Just a question of which jurisdiction gets it. Everyone has the same tax system, it's just a question of which jurisdiction gets it.

Well, of course, there would be no tax competition in that scenario. So it would be a very different story. It's often hard to say normatively which jurisdiction "deserves" to get the money. Most arguments about that I've seen in tax literature often lack the real normative groundings. So you'd say, ah, so what? It's some way to divide the pie. But that's not all that's going on here.

Also, if your idea is to make life hard for local jurisdictions because you think they're, you know, basically cartels trying to grab money and waste it, if you want to shut them down or maybe just shut down the retail sales tax because it is, in fact, a very bad--in practice it's a very bad tax instrument, then maybe that's great too. If you just want to, kind of--the reason we're doing this, we're destroying the village in order to save it, as the saying goes. We kind of want to make the RSD, make it hard to collect. That would be another rationale.

But there are some countervailing points. And the first thing that occurs to me is the difficulty of defining nexus. It's not a coherent economic idea. And it would tend to distort economic decisions that create nexus, that affect nexus. You could imagine someone wanting that nexus to get into a low tax, if for some reason you're producing in New York, you can imagine you'd want to have nexus in Colorado. You'd want to try to create it. It's a difficult idea to figure out how it works.

And what if the states disagree? Here, I wonder if you're going to get back into the problem of needing national coordination. California says that there's a, they can tax all the sales into New York because there's no nexus in New York. And New York says, oh, boy, it's kind of to know how the states are going to coordinate these things.

I think Michael is presupposing a kind of an accepted decision as to whether there's nexus in every case.

Now, another issue here, and I guess it relates more to the economic decisions, is--well, certainly we're outside my area of expertise if we start talking about the UCC definition of residents being used to define where something is going on. So maybe there's something I'm missing about how that works. But I'm naturally thinking of a corporate group, they're production outlets in many states. Maybe they're separately incorporated or resident companies in many states. And it's kind of hard to see how you decide even if there's no, even if it's an internet sale, it's kind of hard to see how you decide where the production occurred.

If one state is suppose to be taxing it all, you know, there are a bunch of different corporate entities and there are a bunch of different activities in different states. I'm not quite sure how you do that. It strikes me, it could be very complicated. It could also be very possibly responsive to tax planning in the form of you have an Oregon corporation that's there just to kind of handle the goods for two seconds. Again, in a retail sales tax, because you don't have transfer pricing, you may be thank goodness you don't have transfer pricing, but you're not actually basing it on where value is created. So it's hard for me to see how it's going to be very administratively elegant, or anything like that.

Then the second point, the paper, Michael does concede locational, you know, that locational distortions may be unfortunate like you do or don't put a sales outlet in D.C. because of the D.C. sales tax, or alternatively the consumer buys from Kramer Books, not Amazon even though Kramer Books has higher production costs because--I'm sorry. I have it backwards. Because you're avoiding the sale.

Now, I can understand a tax base that tries to tax all consumption by residents or all production by residents. Now, again, as we saw under VAT, you can kind of tax all production by residents or all consumption by residents under destination, origin basis, in some way they're the same. But when you think of what an origin basis RST means, you no longer really have that going on. You are just taxing local, you know, sales by local stores to anyone or by, or from local production. So you're no longer really taking your residents and having a tax base that's based on their production or their consumption.

It's fairly reasonable to say that governments may and should consider distributional objectives in allocating the costs of providing public goods and in deciding their transfers. This is why few people would support uniform head taxes. And most people would kind of agree. You know, Bill Gates should probably pay more than a homeless person, who either should probably get something.

Why don't I go back for a second?

And when you, you kind of don't have a principal coherent base, which an origin basis RST, not having--use the transfer [inaudible] really is not. You don't really have a coherent way of dividing the course to govern it, based on some measure of how well off people are, which would be their production or their consumption typically.

So the first thing I would say is origin basis RSTs appear to be inconsistent with use of the tax to pursue, you know, the jurisdiction's distributional objectives with regard to residents. Again, that is not necessarily true of origin basis income taxes, not true of origin basis or jurisdiction basis VATs, flat taxes or X taxes. But it seems to be true here, because of how the origin basis works.

Now, I'd say, well, maybe it's no big deal given the limited use of retail sales tax in our fiscal system, also the fact that they're, really they are kind of a bad instrument. I mean, they have, as Michael noted, they tend to have high rates, a narrow basis. They very often--

[Tape 1, Side 2.]

MR. SHAVIRO: So they're really bad instruments. And most people, by the way, that have looked at on a national level having an RST versus a VAT have concluded a VAT is much better basically because the paper trail for when businesses have to submit credits or vouchers, there's a better paper trail for policing evasion. So, you know, if you kind of screw up the RST, maybe that's a good thing in some ways.

And then I'd say, well, it is true the paper concerns also do matter. That, you know, we do care about administrative and compliance cost and all the other things that Michael emphasizes. Again, but this distributional points against the locational neutrality is not the only offsetting factor.

Let's just one more time mention location neutrality. It really does matter. It really is important and the Teabow story that jurisdictions that can reach anywhere around the world to tax their residents' consumption, no matter where the residents get the consumption good, they're still in competition based on their tax benefit packages with all other jurisdiction. And I think that is an important part of the story. And I'll stop there. Thanks.

MR. HASSETT: Thanks a lot, Dan. And we'll go straight to Peter, and then give some time to responding.

MR. MERRILL: I think you may need technical assistance in our e commerce world here to get--

MR. SHAVIRO: Yeah. I'm sorry. I didn't mean to--

MR. MERRILL: --the next Power Point presentation lined up.

MR. SHAVIRO: I don't mean to make Peter repeat my thoughts.

MR. MERRILL: You did it far better than I could. Thanks very much.

MR. SHAVIRO: You probably--

MR. MERRILL: Let me see if I can run it from here.

I think it was Ronald Reagan that called the U.S. income tax a national disgrace. When he was governor, I don't know if he ever looked at the California sales tax in any detail, but one would think that any observer of the retail sales tax system would have to characterize that as a national abomination. It is hard to imagine a worse tax system than the way that the current retail sales tax operates in the United States. And in that regard I absolutely agree with Michael and Dan.

The question here though, I think, is whether an origin tax would be better. And it is remarkable how simple taxes are that don't exist. It's just extraordinary. They have no problems at all. So I thought maybe what we'd do a little bit today is try to think about how an origin based tax might work. And my initial conclusion is almost exactly opposite Michael's. I think to get an origin-based tax system to work would require far more government cooperation than the destination-based tax system. So that's really the punch line.

I didn't go to MIT so I'm not going to actually do the theoretical blackboard talk. I'm actually approaching this more as a tax administrator. Dan actually could have talked more about--or talked more about economics than I'm going to talk. So I'm going to really focus on how would this work.

Okay. So the first thing I think we need to do is what is the problem that's addressed or created really by the internet that we're trying to address by going to an origin system. The second is, does e commerce really totally change everything causing the destination principle to no longer be feasible in the modern world? The third is, would the origin-based sales tax be better? Fourth, can we fix the destination-based tax to address some of the issues that Michael raised? And some conclusions.

I'm not very good with animations on Power Point, so I sprinkled in some aphorisms. The first one is a problem stated is [inaudible] the solution. I just want to isolate what is the problem that was introduced by e commerce. The internet is in the title of Michael's paper. I know it's not the only motivation. But it seems to be a motivation by many proponents of fairly major changes in tax, both sales tax and income tax, that the internet, you know, has created some very novel problems.

So the way I would state the problem created by new business technologies, new business models created by internet technology is that it is difficult to collect destination-based sales taxes on intangible goods and services. Okay. This is important. These are things that can be digitized that are intangible. This would include movies and records and, you know, books, and so forth, that can all be sent in digital form, all sorts of data, information, as well as many types of services that can now be performed at a distance. You still need to get your haircut obviously locally. But in today's world, there are many, many services, financial services, airline industry, you name it, services that can be provided at a distance to the end consumer. And that is quite new. And I think that is really the issue.

So we're talking about intangible goods and services. And then we're talking about remote sellers. So this is a seller that's not in the same taxing jurisdiction as the customer. And last, we're talking about non business customers. Okay. The issue of a non business customer, certainly in a retail sales tax shouldn't be taxed, but it will turn out it's important in any event. But non business customers can be addressed quite effectively in destination-based systems even for intangible goods and services. And that--I can tell you how about how that's done in the VAT.

So, you know, I think that is the definition of the problem. Intangibles sold remotely to end consumers. And, of course, I'm talking about the sales tax here. There are issues raised in the income tax, the papers about the sales tax.

Okay. The issue of business customers can be addressed through the reverse charge mechanism in a VAT system. In a VAT system if a U.S. company exports a digital good to a business customer in Europe, what happens is that a customer simply assesses themselves VAT on that purchase and claims a credit. It's an offsetting transaction. You might say, what's the point of assessing the tax and claiming the credit because the end result is that purchase, that value added, that purchase ends up in the sales price of the European business. And so when they on sell, pay back on their sales, you pick up the tax that was due. So it doesn't matter that you self-assessed it and claimed a credit.

The reason they do that is there's some business purchases that don't pay tax on their sales, financial intermediaries are one, and so it's important to have the self-assessment.

The way it works in a sales tax is through enforcement of the use tax. The sales tax in the United States has two components. There is the tax, the standard sales tax that is collected when the seller has nexus and the state under the Constitution and the Supreme Court decisions can force the seller to collect the tax. There's companion use tax which deals with situations where the customer is in the jurisdiction, the seller is outside the jurisdiction. You can't force a seller to collect the tax. So you go after the customer. Obviously the ability to enforce that tax against households is extremely difficult. No state really makes a serious attempt to collect the use tax. When you and I, you know, buy something over Amazon. I don't know about you, but I haven't recently paid any sales tax in my jurisdiction, I confess, on Amazon purchases. And 99.999 percent of Americans don't. And there is a use tax there. You're suppose to pay it. It's not enforced.

For business customers, quite the contrary. Business customers pay the use tax. In fact, about 35 percent of all retail sales tax revenues are from business customers. Okay. So this is a very serious amount of tax. It also results in a lot of cascading of tax, because they don't get to review the credits. So you have tax on top of tax, which is one of the defects. But it's quite possible to audit companies on their use tax. So this is not an issue about business customers. Okay.

Well, what about tangible goods? Why isn't there a problem there? Well, you know, obviously in the United States, you know, with Quill, you know tangible goods do represent an issue. It is not an issue in the VAT system. So it's not an inherent defect of a destination system. Clearly for transactions that cross international borders, unless you're within a common market, import of a tangible good is going to be assessed tax by the customs officials, you know, along with tariffs. If it's mail ordered to a direct consumer, what has been done is to use the package delivery services to require the UPS or the FedEx, or whatever, to make sure that the tax was paid. It doesn't work perfectly, but it is a system that is in place.

There is also a requirement within the EU that if German companies is doing mail order to French consumers, if they do more than sort of the diminimus amount of sale, like $25,000 a year type amount, they have to register. Okay. And that raises some issues about cooperation. But this is an example where cooperation is taking place to enforce the law between the various jurisdictions in the EU. Obviously it's easier for them to cooperate because they're all under the same directives that guide that VAT system. Okay.

So does e commerce change everything here? I'd say not, not at all. The issue of remote sales is certainly one that goes back a very long way. The Sears and Roebuck Catalogue I think popularized remote sales back in the 1890s. What the internet is largely doing is actually displacing the remote sales mail order sales that we executed through the mail by fax and telephone. And now we're doing those transactions over the internet. It's certainly expanding the market. It's also displacing the market that was there.

The transactions that are accomplished over the internet are a tiny percentage of all direct market sales. The numbers, you know, are somewhat controversial, but certainly under 10 percent of all direct market sales are actually executed through the internet. The real big business in the internet is B to B. Most studies indicate that that's at least 85 or 90 percent of the volume of internet and it's a fast growing part. And the problem area, as I said, is really the internet transactions involving intangibles. Okay.

And, of course, the sales tax, retail sales tax in the United States doesn't even attempt to impose tax on supplies or intangibles even when they're provided locally. They're just outside the base. Retail sales tax is a tax on tangible goods. Okay. So it doesn't really represent a huge erosion for the states since they never attempted to tax the stuff in the first place. So I'm not at all convinced that the internet changes a lot.

One could say that mail order is a huge problem. It's been with us a long time. But it's not a new problem. Okay.

Well, would an origin tax sales tax be better? The aphorism for this slide is all is not gold that glitters, from [inaudible]. I was very puzzled as I read the paper how an origin sales tax would actually work in practice. It isn't laid out in any detail. And I don't think that was your objective, Michael, to lay out a blueprint but more to raise it at a conceptual level.

But, you know, in order to really talk about this seriously one needs to have some sort of, you know, concept of how it would work in practice. And in thinking about it, I came to the conclusion that for this to work in practice would require far more intergovernmental cooperation than exists today for our destination system.

Why is that? Well, the destination jurisdiction, and the example I think that Dan was putting forward where you've got a company in one state, say, New York, okay, that makes something and ships it for sale in California. Okay. If you're going to have an origin-based system, okay, New York is going to impose the sales tax on that shipment presumably. If it's a final good, that would be the end of the story. But, of course, most shipments are really intermediate goods. They're not final goods. And so, you know, the question is--and even a final good, there's going to be some value added in the marketing end of it. So, you know, I guess the first, you know, question is, if we're going to have an export from New York of a good that's going through another business, okay, it's not going to the final consumer, presumably there has to be some way to prevent a second tax when the business, say, in California that bought this good sells it, you know, at retail. Or it ends up being sold ultimately at retail or it maybe ships that good, you know, for further processing, you know, to Pennsylvania. Okay.

There has to be some way to prevent, you know, multiple levels of tax from taking place. Well, how would that take place? Well, obviously if this doesn't agree, if this doesn't take place, without this type of agreement or mechanism it would be very hard for any individual state to impose origin tax because their exporters, you know, are going to be in a situation where there may be two 5 percent sales tax imposed on their goods. So you're clearly going to have a very great reluctance in any jurisdiction to do this.

So how would you prevent, you know, doubled, tripled taxation from occurring? Well, let's see, I think I may have discussed this on another slide. But one mechanism, you know, clearly would be that you would--the origin state would not enforce the tax when there is a sale to a company that has an exemption certificate. All right. Well, exemption certificates exist today for businesses that buy for on sale. Okay. But this would require a new system where you'd have to be looking--the New York company that's exporting has to look at the exemption certificate for the California company. Okay. If it's a California exemption certificate, it has to verify that certificate.

So, you know, all the sudden you have to have some cooperation, I think, between New York and California on, you know, what certificates count.

The other way to do it is California could give a credit to the California company that bought the product with sales tax that was imposed by New York and then sells it again with California sales tax. So it could be a crediting mechanism. But, again, this would require, you know, some cooperation between the states to prevent fraud.

I guess another just sort of basic point is to get this type of intergovernment cooperation is extremely difficult. Both is a mechanical problem, but there are some revenue consequences about this. And that is in the origin system, you know, a net exporting jurisdiction is going to like an origin system and it's not going to like the destination and vice-a-versa. We're starting, as Kevin points out, in a world with generally a destination system. If you move to origin, states that have imbalances are going to want to be compensated. Also, there are differences in sales tax rates between jurisdictions that make cooperation quite difficult because of the movement of companies to those jurisdictions that don't have sales tax.

I think the contrast is, you know, a destination-based tax actually generally doesn't require this type of coordination. Simply because in the retail sales tax the tax is collected on retail sales in that jurisdiction. You do not need to cooperate with other jurisdictions to prevent the multiple tax on intermediate goods or business to business transactions. So I think it's actually much simpler. Obviously digital commerce is the exception. I don't think it's the rule.

Well, would an origin-based sales tax be the better tax? I guess just sort of a basic question. And it's not proof. But if an origin-based sales tax is such a good idea, why aren't there more success stories out there? We have 125 countries with VAT. Certainly none of them at an international level exempt imports and tax exports. In fact, all the excise taxes that I'm aware of in the United States are on a destination basis, at least internationally. I remember that was changed in super fund taxes in '86. That exports are exempt, imports are taxed under virtually every excise tax the United States has. As was pointed out for tariffs, which are not sales taxes, but with tariffs every jurisdiction operates on destination. There is no country that taxes the exports and allows the imports to come in free.

And, you know, you might say, well, why is that? And my thesis is, is that the level of cooperation that would be required to go to origin basis is completely unrealistic and way more than is required under a destination system. So this cartelization would actually have to take place far more, I think, if we switched to origin.

I think it's instructive to look at the EU experience. It is sort of recounted in the paper that's in your package. In 1977 the EU adopted the VAT. It had the sort of hybrid place of supply rule, goods are destination. Some supplies are actually origin. And that reflects that back in '77 that almost all services, you know, were consumed and supplied in the same place. It is true the internet is now creating a distance between the places supplying consumption for services. So, you know, in '77 place of supply or origin was a pretty good proxy for destination when you went to services. Okay. That has changed.

In the early '90s actually the European Commission in sort of thinking about how to simply the VAT, you know, which, you know, 15 jurisdictions they went to eliminate common borders within Europe. And so it became, you know, a little bit more complicated. When you have goods moving across borders and you don't have any customs control, how are you going to make sure that a destination-based sales tax actually works? And so they were facing, you know, the type of mail order issues that we faced in the United States.

And their bold plan actually was not to dissimilar from what Michael was saying. Within the EU the proposal was to go to the definitive system, which is to tax goods and services on an origin basis. However, in order to keep the states whole on a revenue basis, the states in Europe, to keep the member states whole, the exporting/importing states had to be neutralized. The way it was going to work was they were going to look at GEP flows between the various states. And net exporting states would be compensated, you know, and net importing states would be compensated. The payments made, okay, you know, between the net exporting states that would be taxing on an origin basis, and they would have to make net payments to the importing states. And they worked for about five years to try to refine that system.

It kept getting postponed. There actually were implementation dates for that. I think, you know, around 2000 was the original deadline for it. It was about a 10 year plan. And those dates got pushed off several times.

In 1996 they noticed an interesting problem with telecommunications. Telecommunication is a classic service where the place of supply and the place of consumption, you know, were the same for many, many years. Every telecommunications company in Europe was a national monopoly. And so, you know, if you were in England, you bought from British Telecom. And, you know, British Telecom was a supplier. They were taxed at origin. But that was the same thing as taxing at destination because all their customers were in the UK. And they slap them, you know, with huge taxes and they have extremely high rates on telecommunications, and so forth. It's a national monopoly.

And some clever U.S. companies started offering long distance service directly to European customers through call back service. You know, where you pick up your phone, you call a number locally, and you get a call back. It's a dial tone. The dial tone is from a U.S. long distance carrier. And you make your call. And they bill you through your credit. Okay. So this completely bypassed the national monopolies, which got the governments really made. And, you know, it also avoided VAT because VAT wasn't collected on the U.S. company because it, you know, collected the place of supply. The place of supply actually was in the U.S.

So it was an origin system, but it didn't last very long. Okay. And it didn't last very long because of the competitive problems that the telecom companies face. And, of course, the fact that they were government owned made the competitive problems a government problem as well. And in '96 retroactively, six months retroactively they changed the place of supply in telecoms and said, wait, it's where the customer is located. Okay.

This gives you a feel for, you know, how origin taxation is vulnerable to this type of intense competition and the interest of the home jurisdiction to protect its domestic producers.

In 2003, actually on July 1 of this year, electronically supplied services, which is a pretty broad category from radio and broadcasting to all the digital goods, you know, the movies and CDs and so forth, as well as web posting, internet service providers, and so forth. Most actual services that are sort of mostly provided only because we've got the internet technology were swept into electronically supplied services. And they were switched to the telecom basis. They were switched to destination, whereas before they had been origin.

Finally, this year the European Commission completely scrapped the definitive system. Where they're going to move everything on origin. They're going to put all goods on an origin basis. And they said, we've got a new idea. We're going to make the whole thing destination-based. Okay. So this is the new definitive system. And so they're going to push, actually, to put everything on a destination basis. Which basically means those services now which remain on origin basis would be moved to destination.

It's a little political economy. It suggests, you know, to me that there are a variety of reasons to not think that an origin basis system is one that is politically, you know, likely to occur.

Well, let's talk about how the origin-based sales tax would work in the U.S. context. And I sort of jumped ahead a little bit and talked about this. As I say, you know, I think theory is fine. But we do have to have some blueprint how it would work. I think the first question is, are we talking about would this origin system apply to all goods and services or just the digital intangible variety? If they only apply to the intangible variety, there are obvious competitive distortions that go beyond, you know, the locational issues that have been talked about. But they get to the issue of do you buy your book in hard copy or do you buy it in soft copy, and then maybe print out a copy on your home printer? Do you buy your software shrink wrapped or do you buy it as a download? Obviously if you've got origin for one and destination for the other, you create a huge competitive disparity between the two products. If you go to the origin system for the electronically supplied, the destination, you know, for the physical good, people will clearly buy the channel that gives them the lower tax rate.

The second issue is one that I've already discussed, which is how do you prevent, you know, multiple levels of taxation. This is not tax on tax, the cascading problem in the VAT. This is literally each state imposes their 5, 6 percent tax, you know, one on one, plus the other. Not 5 percent of 5 percent, but 5 percent, plus 5 percent, plus 5 percent, whatever, obviously getting to be enormously high amounts. And, you know, to make this work would require some type of, you know, verification procedures of exemption certificates or credits, or whatever. But, you know, all the same problems that you face when you try to collect internet taxes on remote sellers today, you know, exactly the same issues.

The ones that are confronted on this July 1, 2003, VAT directive where, you know, U.S. seller now through an e commerce channel to a consumer is, in fact, suppose to register in Europe. And they're suppose to collect tax at the rate that applies in that country to that consumer. And if it's a business, they're not suppose to collect the tax, because the business reverse charges. They have to know where does the customer live? Are they a business or not? Is it a valid registration? You've got all these issues in the destination system. And they are in full force in the origin system if you care about avoiding multiple imposition of tax on intermediate goods.

And I think then just the last practical problem, and this has been talked about to death is, you know, how do you prevent a massive shift in mail order to those five states in the United States that don't have a sales tax at all? And I think, Mike, your point is that, well, you know, we're really looking at collecting tax, you know, where the production takes place. It's not so easy to move production. And that will be true for a lot of things. But, you know, there are certainly, you know, in the area that you're talking about, the internet type area where there are digital goods and services, you know, obviously it's really, you know, quite easy to put the place of production when you're selling, you know, software electronically anywhere. I mean, it's not, you know, so hard to move that.

So in the exact area of the problem that we're talking about, sort of the intangibles and the services, and so forth, that stuff is pretty mobile. The other big area I think would be imports. I mean, you know, think about diamond imports. I think everybody would buy their diamonds mail order from Delaware, right? You know, in an origin system. I mean, you know, it's sort of obvious that that would happen. Whole, you know, high value low shipping cost industries would, you know, sort of disappear in origin states with high tax rates. So, you know, it's been mentioned a lot. But, I mean, from a tax administrator standpoint, this is a big problem.

I think one, one more point. It's just a political point. Thinking about how do we get from where we are to there? This is Kevin's point. If you're an elected government official, okay. And you now have a destination based tax, would you vote for a bill that amends it, that amends it in the following way. Okay. Instead of exempting exports, we're going to tax all our exports. Okay. And instead of taxing imports, we're going to exempt all the imports. I mean, you can imagine that all the exporters in your state are going to be lobbying saying, wait a minute, you're putting me at a competitive disadvantage compared to exporters in other states that are not origin. Plus, you know, the importing state is probably going to impose another sales tax. So it might be doubled, you know, taxed unless we could work that out. Plus, the complexity of it. And then every import competing company is saying, wait, you're going to put me out of business.

I mean, this is why the European Union changed electronically supplied services. It wasn't the national monopoly telecoms, it was, you know, the Britlesmans [ph.], and all those guys that were publishing records and newspapers, whatever, saying, wait a minute. The Americans are coming in here. They're shipping the stuff in not paying VAT. We're paying VAT. Okay. That's what really moved the electronically supplied services. It's a political economy point. It's just hard to see how you get there given that type of opposition. Okay.

Yes. Can we fix the destination-based sales tax? My sense of the retail sales tax has far more fundamental problems than the e commerce issue. Many of them were mentioned by Dan. But just ticking them off, there's a very small base because services are exempt. Services are a huge part of our economy today. It's a cascading tax because a lot of business purchases are taxed. All of the direct market sales, cross border sales, not just internet, but the whole shebang is not subject to tax because of Quill. There are over 8,000 jurisdictions. There's no harmonization. It's a complete mess. It is a national abomination. It should be fixed. But e commerce is the littlest of the problem.

Well, what do we do to address e commerce in a destination system? There's a great report actually by the technology TAG, this is a technical advisory group created by the OECD, to look into it that discusses the kind of technologies that ultimately can be used. These include harmonized product definitions as applies in the customs areas. There is a harmonized tariff code. All the, God knows, 200, whatever countries that are part of the WTO use it. Digital certificates, anybody that uses Lotus Notes, you use a digital certificate today that identifies who you are. It could be used for tax purposes to say if you're a business or not, and include tax information. Certainly certified software systems, Vertex [ph.] taxware have software solutions for dealing with the 8,000 jurisdictions in the United States. It would help if there were uniform product categories.

And last, there is a competitive model for dealing with this, which is trusted third parties. Which would be businesses that would be in the business of tax compliance. And this would spring up overnight if the government simply allowed a collection allowance. Okay. So if we move to a world, like in Europe, where you can impose tax on remote sellers, if the core problem were overcome, as long as the states provided some collection allowance, third party business would spring up overnight, financial intermediaries, IBM, who knows, that would jump into the void and offer to provide the tax compliance services, you know, for the collection allowance.

So conclusion, I don't believe the sky is falling here. Clearly, the retail sales tax is badly designed. It has been for a long time. It should be fixed. E commerce issues are a relatively minor problem given how badly designed the retail sales tax is today. I do think destination sales taxes can be modernized to address e commerce issues. Certainly if enlightened use of technology and competitive trusted third party models are used.

And finally, you know, there are so few success stories for origin-based taxes because they require an unrealistically high level of government cooperation to prevent various competitive distortions complexity, double taxation, and shifts in the revenue base. Thanks.

MR. HASSETT: Thanks a lot, Pete and Dan for great comments.

I have one clarification question for Peter, just quickly. So the retail sales tax is to some extent under pressure because of the internet sales that are not taxed. And am I correct in saying that a conclusion that follows from your remarks is that what we should do is just, you know, use technology to allow states to spread their collection to internet sales. So what they could do is tell Star Power, well, you provide internet service to Kevin. If Kevin makes a purchase from anywhere, then, you know, you, Star Power, you know, can hire a trusted third party, Pricewaterhouse, to make sure that Kevin pays the sales tax on that.

So what you want to do is spread the destination sales tax to the places where it doesn't exist right now in addition to other reforms?

MR. MERRILL: Right. Just to go right back to the basics here. Right now in the retail sales tax, whether it's remote or locally supplied intangible goods and services generally are not subject to tax at all. Okay. So that would be the first change I think we'd have to make before we can even answer your question whether it should be taxed, you know, in a remote transaction.

The second question is, if we're going to tax intangible goods and services, then do we want to tax them on a collection basis where we're going to require the seller to collect it even when the seller doesn't have nexus? Okay. That's what the Constitution doesn't allow. Well, if the Constitution changed the rule so that you could impose tax, you know, the responsibility for tax on the remote seller, okay, they could hire a trusted third party or not. That's their choice. But the legal liability to make sure it's collected is on the third party. Then in that world, which we're very far from, I think, you know, trusted third party models are a great model to use. Give the choice to use a third party who can specialize in the business, you know, of doing all the requirements of identifying what's the proper rate, you know, should it apply or not. You know, making sure it's collected on the transaction, filing the tax returns, all those things could be, you know, centralized, you know, with software.

So redefining the question to a world where I think it's a relevant question, I think a trusted third party model would be a very good idea. But it requires the states to allow a collection allowance. I mean, if there's not a collection allowance, I don't think that model would spring up very easily.

MR. HASSETT: So wait. So you're maintaining that tangible goods are not in any way pressuring retail sales, tangible good activity over the internet?

MR. MERRILL: I'm saying the internet is a tiny percentage, a tiny percentage.

MR. HASSETT: So that's a yes, right? It's a yes?

MR. MERRILL: We've had these mail order sales which are gigantic, I mean, you know, they're hundreds of billions of dollars, okay, of direct market sales in the United States. The internet might be 3 percent of that. So I think there is a huge problem with the fact that all direct market sales are out of the system, okay. And I think that is something that, you know, ideally should be fixed. You'd have a far more neutral system. But the internet is, you know, not the issue. Okay. The issue is, you know, Quill, the Constitution, in my mind.

MR. HASSETT: All right. Okay. Well, thanks.

Well, Michael, I'll give you a couple of minutes for you to respond. And then we'll go to the floor.

MR. GREVE: Just, just a few brief remarks. And, and, again, thanks for what I think are really very forceful criticisms that I'll take into due consideration, albeit, not now.

I just want to clarify my position on a few points. First, a lot of the objections to the system one way or the other arise from general problems with the retail sales tax. And I want to identify myself with those criticisms. There are all sorts of problems with the way that the system is arranged quite apart from the sourcing issues. Business to business sales is one problem. The wholesale exemption de facto essential for services is another. If you're worried about neutrality, that's where you ought to start.

If somebody wanted to say, look, that system is so broken that there's nothing we can do to fix it within the certain parameter just by fooling around with the sourcing rules, I'd be the first one to say, be my guest. Except that's not this, but the way this debate has gone in the United States, and to a certain extent I mean even in Europe, is to say, well, within the existing tax structure, how should we deal with one aspect of the problem? And so the paper just moves on that trajectory of the debate.

Again, if somebody were to say, that's what's wrong with the debate, we should take a broader view and put everything on the table, I'd be the first to say, hallelujah. Go, go ahead.

Peter's point about the political economy is, I think, almost entirely right. I just want to--the story about the European Union was very fascinating. The same story, of course, was true in the United States when. When the current sales and use tax system was created, the National Tax Administrators Association, or whatever it was called then, started on the basis of saying, look, we ought to have an origin system. And the reason why that didn't go over was that the market states, that is to say the import states, said, no, no. Look, we are the poor suckers in the system to begin with. And if you stick with an origin basis, we'll be poorer still. And all the money will be made by the big producer states.

I think that is the real reason why origin- based taxation hasn't been tried on the broader level. It's not that there are some inherent complexities. I, I don't believe that part of the argument. But I do believe the part of the argument that the, the trouble in adopting something like this is the non neutrality. You are, if you move from one to the other, favoring export states over import states.

I should add one additional thing. There are proposals that I have seen. They were made to the ACC, the Advisory Commission, eon ago, to say let's go to origin-based taxation on the grounds of simplicity. Evidently those authors shared sort of my gut that it's slightly easier to do. And I don't want to go into those weeds. And they then said, and in order to help that over the hurdle, let's neutralize the system. That is to say, figure out some administrative and political regime to even out the flows, the revenue flows. And that I'm totally against. If that is part of the system, then I'm no longer in favor of origin-based taxation.

I want to say one quick word about Dan's really terrific remarks. It may be true that the tax base on an origin-based system, I mean, first I really don't want to go into transfer pricing. Dan is entirely right about that. That then makes the tax base arbitrary in some sense. I grant that. The question is, compared to what? You have the same distortions under any system, including the existing system. It's a fair enough observation that, no, this is, it's a somewhat arbitrary base. But what I think you cannot do and, and what I think is not right, is to compare that regime or that proposal to a highly stylized ideal type destination-based system which we don't have now and which we never will have.

MR. HASSETT: Thank you.

MR. : Just a quick comment to Michael. Are you, in a way, overselling your point? Your point might be a more limited one that given where we are the case for getting rid of Quill and having more aggressive collection isn't as strong, isn't as strong as people say because, for example, the enforcement costs are greater, you know--aren't you perhaps overselling? Maybe a more limited point would be, we're in a really lousy place now, but it's not obvious that by correcting one genuine distortion we make it any better?

MR. GREVE: That's a big, bit part of the paper. I mean, in fact, if you sort of look through it. Sort of the critique of the existing system and the possibility or perceived impossibility of fixing it within the certain--within the existing parameters is the much larger part of the argument, yes. And I grant I maybe overselling.

I certainly, after all I've heard from, from Dan and from Peter, I certainly in this paper oversold the elegance and simplicity of this model. And that's one thing I'll fix.

MR. HASSETT: One thing that you maybe undersold that occurred to me, the last observation before we go to the floor, and it's a short one because I know that the audience has been very patient, is that Dan makes the excellent point that what you're really saying in your paper is that if we have an origin-based tax, then that really makes the Teabow model, the sort of idea that tax competition makes sure that the politicians are spending the money on the good things. It makes that really work. So, you know, if you're going to tax, then you got to create a good business environment that offsets the tax. So the businesses still want to be there. And that that kind of discipline will be good. And as an expert on institutions and stuff, that's the angle you're coming from.

Dan says, well, wait a minute. But they, they have Teabow competition for consumers too. That if they have a retail sales tax, then they've got to make, you know, have a nice park or good schools, or something. And if they don't, then people won't want to go there.

Well, it occurs to me, given what Peter said, though, about, you know, what might happen if, if you had your way, is suggested that we probably expect that the Teabow forces are much more powerful with businesses and location than they are with individuals and location.

And I think that, I mean, it follows from Peter's remarks and it is somewhat of a rejoinder to Dan, and I think it's just an interesting observation that really what your paper is saying is partly that. That if we think we want to use Teabow to make sure that the government officials are paying attention to what people are doing, then we need to recognize that those forces, perhaps, are more powerful for businesses. And so, you know, I think that that's a key point in your paper, perhaps not bullet pointed. And I think it's something that's been sustained by the conversation here.

And with that, unless you want to say I'm wrong, we'll go out here.

So, please, we'd love to have your questions. Stick your hand up if you have one. Wait for the microphone. Please state your name.

Let's start with this fellow right here, and then we'll go to Robert. Name and affiliation. And please make it a question, if you could.

MR. KATZ: Thanks. My name is [inaudible] Katz [ph.]. I'm a reporter for Tax Notes Magazine. It's a question for Michael.

Regarding your, your experiment that you're discussing, working with states that were sort of resisters to the SSTP. Have you had conversations with officials or law makers in any of those states regarding your ideas about origin based taxation?

MR. GREVE: I've--

MR.. KATZ: I have a follow-up as well, Peter.

MR. GREVE: Oh. I've spoken to some of them, not many.

MR. KATZ: My first follow-up, I guess, is do you have a sense that the idea is gaining some traction on the state level? And secondly, do you have a sense that the hypothetical federal legislation that you discussed is gaining traction as well? Or, or is maybe something that could be considered as an alternative to possible authorizing legislation for internet sales taxation, legislation that could come up this fall?

MR. GREVE: I don't, I don't follow the sort of ins and outs and daily battles over the stuff very, very carefully. My general impression, and if people here do follow this as lobbyists, or handle this, they should correct me, is that once upon a time there were, in fact, in the last round of ITFA discussions, proposals to codify some sort of nexus standard including some very demanding or restrictive bills, if you will.

My sense is that the momentum since has been in the other direction. That the SSTP has gained some traction. That--so, so, and--right. And that industry coalition has become broader. So that's, and that drift would go in the, in the opposite direction.

I don't think this has ever been formally presented to any collection of potentially friendly state officials. So I don't really know what their response would be. Right. Frankly, I mean, I started thinking about this some time ago and then dropped it for other stuff. Way back then one of my candidates for a state that might actually want to join this was the Commonwealth of Virginia, my home State. That was because Governor Gillmore was sort of on the side of the angels, I thought, and the ACC fight, except with some of the wrong arguments. I thought, hey, if someone were to roll this out in full regalia, you know, what would Governor Gillmore say? Of course, Governor Gillmore is no longer Governor Gillmore.

MR. : I am Robert Hershey. I'm a consultant. To what extent has the administration of contract law been brought into this? For instance, if you sign a contract it always says, this is administered under the laws of a certain state. And it would seem this would fit better with the origin state because the origin state, if they are recording the transaction, and if there's fraud or disputes, that they would be the ones who could go to the factory and confiscate things or put the owners in jail. Has this been brought in as something that would be a responsibility of the collecting state as opposed to just getting tribute?

MR. GREVE: I hope I'm not misunderstanding the question. But I've never thought much of this in terms of sort of choice of law principles, which is the technical term for this, simply because people can't agree by contract not to pay taxes. Right? So they can't just say, okay, I live in California. You live in New York. We'll now do this particular sale or transaction under the sales tax laws of the State of Delaware. That's just not done. Although you could specify in other settings, in truly contractual settings you could specify the laws of any given state. So I'm not sure exactly how, how this would apply here. Does that make sense?

MR. : [Inaudible.]

MR. GREVE: Next question. Sir? Please wait for the mike. Yes.

MR. EHRENHAFT: Thank you. I'm Peter Ehrenhaft. I'm a lawyer at Miller and Chevalier. I have a question about the relationship of this issue to income tax which in my experience, anyway, representing companies involved in trans national transactions is a much more significant aspect of their tax planning than are sales taxes.

And in the income tax area, I think that they are very well defined or apparently well defined concepts, and so on, that essentially are similar to the issues that you're talking about. And where we have essentially adopted the rule that where the income of the company is derived, normally at its place of production and so on, is the place where it's going to pay tax. It seems to me that it's not a sensible idea to have a different system for this tax division than the income tax. And I know that this is one of the great--

[Tape 2, Side 1.]

MR. EHRENHAFT: --in the end it's the taxpayer that pays it, whether it's on the good or on the entity. And so isn't it a more desirable goal of the taxation system one that would not differentiate between these kinds of systems, but would harmonize income tax with sales taxes so that there would be a universal kind of symbol single tax system applicable to all transactions?

MR. GREVE: Dan should answer. I mean, I, I, he's Mr. International Transactions. He actually knows this. I think I'll know what he'll say. And I think that I'll agree.

MR. SHAVIRO: Well, one, we're in a very different environment in the international thing. For example, if you assume Americans are going to be Americans ignoring, you know, people migrating people in or out, destination and origin basis are in some ways the same. So I don't think that having, for example, an origin basis RST and a destination basis income tax--I'm sorry, I have that backwards--is really a problem.

In the state picture it's a lot more complicated because, you know, you tend to think Americans will produce and consume the same amounts. But I live in New York and I produce and I go to Florida to retire and I consume there. All the equivalences fail because you've wiped out the assumption that people always live in the same place.

But I think people who know a lot more than I do about the international picture don't think it's a problem to have, say, for example, origin basis income taxes and destination basis VATs.

MR. : [Inaudible.]

MR. SHAVIRO: Well--

MR. : [Inaudible.]

MR. SHAVIRO: Well, and it means you have duplicative--

MR. : [Inaudible.]

MR. SHAVIRO: Yeah.

MR. : [Inaudible.]

MR. SHAVIRO: Yeah. Well, it has to be because you can't get from consumption to a measure of production is basically what income tax is about.

MR. : [Inaudible.]

MR. SHAVIRO: But, for example, in Europe you have a European country. They have an origin-based income tax and an destination-based VAT. Apart from the problem the fact that you get both sets of problems, but at least at a lower rate than if you had one system, a lower rate for each one, I don't know that it's especially worse. You have an European destination-based system. Now their transfer pricing issues are only at the income tax rate. So they have both types of problems, but the transfer pricing problems are at a lower rate, for example.

MR. : I would put the question maybe slightly differently. Which is if you wanted to increase the share of revenues from origin-based taxes, I think the most logical thing from a tax administrator standpoint is raise the income tax and get rid of the VAT. Become like the United States, okay, rather than like a European country.

I think a European tax administrator or politician would say, you know, if we do that, though, our income tax rate is going to be way too high. And we'll have trouble collecting it. It will make us noncompetitive. And I think tax administrators like to have multiple different bases. And the income tax and the destination-based VAT have a different base. They definitely have some differences in incidents. And having multiple broad-based low rate taxes from an administrative standpoint is preferable.

I don't see an argument, though, for going to an origin-based sales tax and an origin-based income tax. I think you end up, and this is discussed in Mike's paper, but particularly for the immobile, the more immobile industries, that sales tax then becomes in effect borne by the producer. It becomes like an income tax. But it's not measured by income. It's measured by the gross value of output. And so for immobile producers using origin based sales tax has the effect of imposing multiple different effective tax rates depending on what their margins are. So it seems to me the worst of all worlds. If you want more origin-based tax, raise the income tax. Get rid of the VAT. If you want to spread it out among multiple bases, use a destination-based VAT I would say in an origin-based income tax.

MR. HASSETT: Okay. We've got time for a couple more questions. I guess we'll start with you and then let Harry finish us off.

MR. GOOD: Hi! I'm Larry Good with Electronic Commerce Association. I've been following the streamlined sales tax project very closely since its inception. But there are a couple of things that you said, Michael Greve, that either I misunderstood or maybe I missed something. One of them was that the SSTP has said that the sourcing rules that it has for sales should also govern income taxation. And the second was that the SSTP ultimately wants Congress to mandate state participation. Is that what you said?

MR. GREVE: I think that is, that has to be their ultimate move. I believe that has to be their ultimate move, whether they say it now or not, yes.

MR. GOOD: And that they want the sourcing rules to apply to income taxation?

MR. GREVE: Business income taxation, let's be precise. I mean, you have in the business income tax context, the problem has to do with the allocation. Okay. Whether you can sort of allocate business income, yes or no. And there the SSTP wants to make the same move, yes, as it has or the analogous move that it has made in the sales tax and the use tax context, yeah.

MR. GOOD: Is that documented?

MR. GREVE: Yeah. I can find you those articles, yes.

MR. HASSETT: Okay. And thanks so much.

And, Harry, do you still have something you want to ask?

MR. : [Inaudible.]

MR. HASSETT: Please wait for the mike, please. Harry Gruber [ph.], United States Treasury, not speaking for the Secretary.

MR. GRUBER: Six or seven years ago Scott and you and I did a paper which, an AEI pamphlet which this is [inaudible] comment, but a consumption tax in the world economy which goes into a lot of these origin/destination principles. I try to publicize it at AEI on a couple of occasions.

But in any case, if you accept the basic equivalence of the origin and destination principle taxes which has been outlined, I don't understand some of these comments like--I don't understand this thing about Teabow. Why is Teabow--if they're the same basis, why does Teabow matter? You're taxing consumption in both cases. And similarly, then, so I didn't quite understand Kevin's discussion of--

MR. HASSETT: Let me explain the Teabow point just because if you asked three things, we'll drop two of them and not talk about them.

MR. GRUBER: If the destination principle are the same, except for the [inaudible].

MR. HASSETT: No. No. I mean, if you have a benevolent government, it's not an issue. But if you've got governments where they might be taxing things and redistributing, it was something that was mentioned. And something, you know, and the redistribution harms businesses--

MR. GRUBER: But it's the same tax. It's the same tax.

MR. HASSETT: No. No. But then if you have multiple states, then the extent to which they're in competition with each other where their total package of government and taxes depends on mobility. If you have high transaction costs so people can't move across states, then there's not going to be a Teabow competition that's meaningful. And alls I'm saying is that it appears that Peter's opinion was that it's very easy for businesses, they have very low transaction costs, to move across states for some transactions. And then those types of transactions--

MR. : [Inaudible.]

MR. HASSETT: --wouldn't be subject to Teabow competition. I don't see what the hard part about that point is.

MR. GRUBER: But they're rejecting the equivalence of the two bases. So there must be, you can reject the equivalence of the two bases in some cases because of transfer pricing, and things like that and initial consumption, initial assets. Us oldtimers, we have initial assets we can, under an origin basis system, we could consume tax free by importing from Delaware. But I'm not sure. Are you appealing to those non equivalences? If they're basically a simple system of importing and exporting, they're the same base. So how, how could things differ potentially?

MR. HASSETT: I think it is the state. Harry's point is--

MR. GRUBER: [Inaudible] include the distribution.

MR. HASSETT: Yeah. If you had really like the VAT model where you really are taxing production of the consumption of a set of people, he's saying you can't have more Teabow competition on one system than the other because the two are in the long run equivalent. I think this whole discussion, though, is in the state environment where I produced in New York and I consume in Florida because of the pool of residents. And that certainly puts the Teabow resident competition, gives it a meaning it won't have in the international context when you assume Americans are Americans.

MR. : Right.

MR. SHAVIRO: I think there are some differences between states and international. One is that the adjustment process is much quicker to get to the equilibrium in the international context where you have floating exchange rates. If everyone's using the same currency, as in the United States, to get an equilibrium requires ratings to drop in one jurisdiction versus another jurisdiction which obviously is a very different process to get to the equilibrium than in a floating exchange rate environment.

I think that would be one, one difference. And then I think all of this theoretical equivalence does depend on immobility of consumers and producers. I think once you're out of that context, I don't know that these equivalences hold up. I don't know. But I wonder.

MR. : [Inaudible.]

MR. HASSETT: Well, thank you so much, everyone for your comments. And it was a wonderful session.

Thanks again.

[Whereupon, the seminar was concluded.]