What follows is a compelling explanation of a complicated set of issues that seem likely to be with us for years to come.

HAL VARIAN: First, it’s important to note that consumers are required to pay state and local taxes on items they purchase from out-of-state vendors, whether they buy over the Internet or by mail order. That law is on the books in every state. It’s just that there is no effective mechanism for enforcement. So almost no one pays those taxes. Most consumers don’t even know that they’re required to pay those taxes.

That misperception colors the debate we’re having about how to handle taxes on retail purchases made over the Internet. (It’s important to keep in mind that the relevant issue is business-to-consumer sales, not business-to-business sales, where most of the growth will be, because B-to-B sales are already taxed. States have the power to require businesses to pay taxes on out-of-state purchases, and businesses are audited periodically, so compliance is reasonably high.)

I think that what is most likely to happen in the next few years is that we will keep the status quo, despite all the discussion about changes in taxation on a national scale.

That isn’t such a terrible thing because on-line retail shopping is really not a very big deal. It was only about 0.5% of total retail sales last year and is dwarfed by traditional catalog sales, which are at least six times as large. Even optimistic forecasts for 2002 predict that on-line sales will be only 1% or 1.5% of total retail sales.

The highest-value item sold to consumers on-line is airline tickets, which aren’t subject to local sales tax. And some of the growth in on-line sales will just be people placing their orders to L.L. Bean on-line rather than via the telephone. People already generally don’t pay taxes on catalog purchases, so there is no net loss in tax revenue.

The shift to on-line sales will have minimal effect on tax revenue. A recent study from the University of Tennessee economists Donald Bruce and William Fox estimates that lost state tax revenue will be on the order of $11 billion a year by 2002. That seems like a lot of money, but it’s less than 2% of total local and state tax revenue, so the sky isn’t falling.

We have a lot of time to come up with a sensible plan. The status quo is quite viable, at least for the next five years or so.

KARL FRIEDEN: I agree that we’ll probably have the status quo for a while. However, the status quo will not be a conflict-free zone, but an increasingly contentious area for disputes between government and business over issues such as jurisdiction, sourcing of income, and tax simplification.

The problem is complicated because our 19th-century system of state and local federalism (and a weaker central government) is in conflict with the 21st-century new economy. We may not like some of the consequences of giving so much autonomy to state and local governments—such as the 7,600 different sales tax rates and how many different tax bases they’ve created—but Americans certainly seem to like the balance of power between state and federal government. Therein lies the source of many of our tax complexities and controversies.

The problem is exacerbated because the cluttered political landscape inhibits consensus over any one particular solution. For instance, a lot of established businesses want tax simplification. But a lot of dot-coms don’t really care about the simplification because they aren’t generally collecting state and local taxes right now. You also have Republican and Democratic party politics, presidential politics, and a lot of other cross currents.

That’s why, even as government and businesses try to build support for significant tax reform, it’s more likely to be a stalemate, at least for some time.

VARIAN: Under the Constitution, Congress could pass legislation granting the states more authority allowing them to collect taxes on out-of-state sales. And this is one of the few times when the states would welcome federal intervention. But Congress would essentially be imposing a tax where the federal government doesn’t collect any of the revenue. Congress would take the heat, without getting any of the benefits by having revenue to spend. It would be difficult for Congress to take such action.

By the way, there is a lot of evidence that consumers are quite sensitive to the tax issue. A recent survey found that 46% of on-line buyers said they’d never paid sales tax on Internet purchases and that 75% said they’d buy less on the Internet if a sales tax were imposed. There was a very nice study by economist Austan Goolsbee at the University of Chicago, where he estimated that on-line sales could drop as much as 25% if remote purchases were taxed at the same rate as local purchases.

FRIEDEN: If there is some national compromise that would help enforce sales tax collections on on-line sales, the wild card remains whether consumers will weigh in as they did when the mail-order industry tried to make a compromise on taxes with states. The public might insist: No, we don’t want to be paying those taxes.

While it is unlikely, it is certainly possible that Congress could enact a moratorium on sales taxes on anything sold on the Internet, which could lead to some pretty creative behavior. Companies might set up Internet stalls or kiosks within stores and have purchases by consumers made over the Internet in the retail stores themselves.

VARIAN: This is also an extremely complex issue for two other reasons. One is jurisdictional. There are myriad different local taxing authorities. It’s hard to even map their boundaries, because they don’t generally coincide with ZIP Codes. The second, and bigger, problem is the lack of standards for how the tax base is defined. Magazines are taxed in California but not in Massachusetts. Large marshmallows are taxed in New York, but small marshmallows aren’t. Are shoelaces an item of clothing? Are handkerchiefs an item of clothing? These metaphysical issues go on and on, and local tax authorities answer them in a wide variety of ways, so simplifying the tax system will be hard.

FRIEDEN: There are really five trends within the Internet that are creating problems for tax systems in the U.S. and abroad. It’s important to mention them, at least briefly, to show just how complex the tax issue is.

First is borderless commerce. The innately global nature of the Internet makes geography irrelevant. Obviously, cross-border transactions present huge challenges to tax authorities, and the volume of these transactions is increasing enormously.

Second is digital commerce. When products are purchased electronically, it’s harder to determine where they’re sold, to trace where they are used, and so forth. Moreover, digital products are frequently treated differently than tangible goods for purposes of income and transactional taxes. This will create problems as tangible products—such as software, books, movies, music, and educational and training information—are increasingly delivered electronically and as services, such as medical diagnoses, are more often provided on-line.

Third is the hollowing out of corporations. The Internet facilitates partnerships as well as a movement away from vertical integration. Companies can have smaller core competencies and just reach out, get a customer base, and serve lots of people without having to do the manufacturing, the research and development, the customer service, or the distribution or fulfillment. This doesn’t just create issues related to sales taxes, but also to income taxes, which get short shrift at the moment, because many of the dot-coms aren’t making any money and, thus, don’t have to pay any.

The issue is where profits will be taxed. A lot of these Internet companies, the dot-coms and the separate subsidiaries set up by bricks-and-mortar companies, are going to have significant locational flexibility. They are going to look for jurisdictions where tax laws are more beneficial to them. That could put lots of pressure on state and national income-tax systems.

Fourth is real-time commerce. Because products and services are often delivered instantaneously, once they are sold, tax-compliance systems need to be automated. Otherwise, it’s too hard to manage the process.

Fifth is the change in business models. You can’t just think about e-retailing. You have a variety of developing business models such as auctions, reverse auctions, new types of intermediaries, and virtual communities. All these raise tax issues, some of which have never come up before. Sometimes, it’s hard to even tell who the seller is, and who is just acting as an agent.

VARIAN: To your list of complexities, I’d add the 100-year-old trend toward services accounting for a larger percentage of consumption. Because services generally aren’t taxed, the sales tax base will continue to cover a smaller and smaller percentage of the economy.

There are international complications, too. Europe, from what I can tell, is tending to rely more and more on the value-added tax, and we’re moving, if anything, in the other direction, so there is a big problem with international rationalization. Even if we manage to get our own house in order, we’re going to have problems making our house compatible with all the other houses on the block.

FRIEDEN: It will take years really, two or three at a minimum, for significant tax reforms to be implemented. Consensus—or a majority compromise—is one thing, and that is going to be hard enough to reach. But even once you reach that, there is going to be a huge amount of fighting over details.

In the meantime, there are many existing tax rules that cannot be ignored. There are some major tradeoffs for businesses. If you want to limit your tax profile in a customer state, you may have to limit your marketing there or not hire third parties to help you in that state. You have to be careful about attending trade shows or visiting your suppliers there. You have to figure out who is doing your shipping in the state, to make sure you won’t be considered to have nexus there.

Some of the rules could get tougher, too. If the state and local governments get frustrated that there is no national solution, they’ll probably do two things. They’ll start simplifying on their own to show that they’re headed in the right direction. They’ll also be more aggressive about enforcing existing rules that might let them tax companies with minimal presence or with agents in their jurisdiction. The level of conflict is going to increase substantially.

VARIAN: I think that’s one scenario. But it’s not a foregone conclusion that we’ll see any dramatic change. States may not lose much tax revenue. It may be that, after two years, the whole issue just recedes into the background.

You have to stay alert to possibilities. It could be that you see simplification based on some national agreement. Another scenario is that the status quo just continues in its own messy way. Just as we lived with the tax losses because of catalog sales and home shopping, we absorb the relatively small Internet purchases. Another scenario, which personally I would find the most attractive, is that we use this as an opportunity to make some fundamental reforms in the way sales taxes are collected in the U.S. What we’ve got now is a mess compared with any system that exists elsewhere in the world. It would be hard to think of a poorer way to finance state and local operations.

It may be a bit utopian to say this, but I think we should just get rid of the sales tax for both on-line and off-line transactions and substitute an income-tax surcharge.

One problem with the current sales taxes is that businesses pay about 40% of the total, on intermediate products they buy from other businesses, when really they shouldn’t be paying any at all. The taxes can pyramid, if there are several intermediate products, so the actual tax paid on products when they get into the hands of the ultimate consumer is much larger than the 5%, 6%, or 8% nominal sales tax.

The second problem is that the sales tax applies to such a small base. These taxes were imposed back in the ’30s as temporary, emergency measures to make up for property taxes lost because of the Great Depression. Sales taxes are mostly on physical products, which make up about 35% to 40% of consumption these days. Once you take away food, which isn’t generally taxed, and once you take away services, which aren’t taxed in most states, you have a relatively high tax on a very narrow base. It’s much better from the viewpoint of economics to have a low tax on a broad base than a high tax on a narrow base.

A 2% surcharge on income tax would raise about the same amount of revenue as a 6% sales tax, and you’d end up with a much lower compliance cost and must less hassle. Of course, I recognize that this is a big change in how we’re thinking, so it is somewhat utopian.

FRIEDEN: I will say that, while we can’t predict with confidence the outcome of the Internet tax debate, there is a lot that business, and for that matter state and local governments, can gain from the current debate.

Some of the things that are wrong with our tax systems are hard to change, because of inertia. But, we are going to find ourselves under more intense pressure to simplify, clarify, and modernize our tax laws than at any time in recent decades. Something is now going to happen, and that’s potentially beneficial, even if it is enacted on a piecemeal or experimental basis.

VARIAN: I think it’s also good that the digital economy is, in effect, making companies and customers more mobile. We want states to have to compete in providing high-quality services at low tax cost. Having companies and consumers who can choose where they do business is a great way to get states to compete. We get a lot of benefits from competition in the private sector. Having some at the state level will help rein in the excesses of government.

Varian can be reached at hal@sims.berkeley.edu. Frieden can be reached at karl.a.frieden@us.arthurandersen.com.