The E-Taxman Cometh

David G. Post
American Lawyer, "Plugging In," March 1997
[Return to D.Post home page]
You may freely redistribute this column; please retain author and publication attribution. If you are not currently receiving these columns directly and would like to do so, please send me an e-mail at and let me know -- D. Post

As commerce moves to the Internet, the taxman, surely, will not be far behind. What he will find there, however, is an environment that poses some fairly fundamental challenges to long-held principles of tax administration and tax policy. And while that may require adjusting these principles to apply them to Internet transactions, the special problems of taxation on the Internet may have important spillover effects for tax policy in the real world -- so even if you (or your clients) have no interest in the Net, you should pay attention to developments in that domain.

That, at least, is the conclusion I draw from the Treasury Department's long-awaited report on "Selected Tax Policy Implications of Global Electronic Commerce," issued this past December (and available at, the latest in the Administration's efforts to come to terms with the policy implications of the global network.

First, a note of congratulations. As someone who has in the past been critical of the federal government's clumsy policy-making in regard to Internet activities -- most notably, the passage of the Communications Decency Act and the Administration's White Paper on Intellectual Property and the National Information Infrastructure, both of which illustrated how little about this medium policymakers understand and how little they will let that stop them from bulling forward with ill-conceived "solutions" to perceived Internet problems -- I should in fairness be equally quick to credit good work when I see it. And good work this is.

The Report recognizes that some serious surgery may be required to fit the unusual transactional capabilities of the Internet into our current system of tax administration. But it also recognizes that the Internet holds enormous (but as yet unfulfilled) promise for the development of new kinds of transactions and payment systems, and that we would probably be doing more harm than good to try now to design ambitious new regulatory initiatives. There appears to be a genuine effort to stimulate public dialogue here, on the theory that the goal -- to develop a framework that will allow electronic commerce to flourish -- will not be best served at this point by precipitous action.

Before considering the Report's findings, a brief sojourn into the world of international taxation is in order. The US, like most other nations, asserts the right to tax income that has its source in the US, as well as income that is earned by persons or entities residing in the US. That is, "source income" earned within the US is taxable by the US even when earned by foreign persons, and the worldwide income of US citizens/residents is taxable by the US wherever it is earned. So the US asserts the right to tax the income that Volkswagen earns from sales in the US (as US source income), as well as the income that a US resident (like GM) earns from sales to overseas customers.

Because Germany and the US each may assert the right to tax income earned within Germany by a US resident (Germany on the "source income" theory, the US on the "residence income" theory), one of these principles must yield to the other if double taxation is to be avoided. Much of international tax law can be seen as attempts to reconcile these conflicting demands, through a complex network of bilateral tax treaties (the US is a signatory to over 45, including one with Germany) that limit, or eliminate, each country's right to tax source income unless it comes from a 'permanent establishment' in a jurisdiction. This, in effect, converts what might be thought of as "source income" into a kind of "residence" income, so that a foreign company (like Volkswagen) is safe from US tax unless it maintains a permanent establishment in the US (which, of course, it does).

But certain features of the Internet -- familiar to regular readers of this column -- play havoc with this scheme. As the Treasury Dept Report itself notes, the Net is "radically decentralized"; it "has no physical location," users "have no control and in general no knowledge of the path traveled by the information they seek or publish," and "it makes no difference whether the information or electronic money sought to be transmitted are within one jurisdiction or between several, as the Internet pays little or no attention to national boundaries." Transactions on the Net therefore have no location: "Even if an e-mail address is clearly associated with a certain person and computer, that person and her computer could be located anywhere in the world. This makes it difficult to determine a person's location and identity, which is often important for tax purposes."

The bottom line? By "dissolv[ing] the link between an income-producing activity and a specific location," the Internet renders the very notion of "source income" problematic, if not downright incomprehensible. Source-based taxation of Internet transactions, in the Report's words, "could lose its rationale and be rendered obsolete by electronic commerce."

This is not an insignificant point -- and it may affect far more than just commerce on Internet. Throughout, the Report stresses a single, fundamental guiding principle for tax policy: neutrality, the principle that economically similar income should be treated in similar ways, regardless of whether it is earned through electronic means or through more conventional channels of commerce. The tax system should not play favorites between selling widgets from a Web site and selling widgets from a New York storefront, so that market forces, and not artificial rules of taxation, determine the success or failure of alternative commercial forms.

Combine these two principles -- the incoherence of source based taxation rules for Internet transactions and the principle of neutrality -- and the entire superstructure of international taxation may need to be rebuilt. If our goal is neutrality between electronic and non-electronic transactions, and if source-based taxation cannot be applied to electronic commerce, then it cannot be applied to real world transactions, either. If it makes no sense to impose a source-based tax on the widget seller when his sales are conducted on the Net, then the principle of neutrality would seem to require eliminating this basis for taxing those sales that arise from more traditional modes of commerce. The Net, in other words, may generate a unifying policy that covers the entire range of taxable transactions, off-Net as well as on.

But, sad to say, all is not simplicity in the world of electronic taxation. The Net may lead the government ultimately to simplify the rules regarding multi jurisdictional taxation, but there is the small problem of enforcing compliance with whatever rules are in place. The compliance problem has two facets. First, the Net is an engine for "disintermediation" [See "The Net Squeeze on the Middleman," December 1996] eliminating intermediary institutions -- everything from traditional "publishers" to traditional "banks." This causes severe headaches for a tax system in which compliance is facilitated by identifying key 'taxing points' -- such as the banks -- on whom reporting and record-keeping requirements are usually imposed.

And even more troublesome is the specter of electronic money. All electronic money systems permit users to move funds stored in the form of digital bits across the global network. In what's known as "accounted systems," the e-money issuer maintains a trail of transactions, and can identify the persons to whom the money is initially issued and those to whom it flows as payment for goods or services. In "unaccounted systems," on the other hand, e-money is issued and moves through the economy without a transaction trail -- just like real cash. And, like cash, it opens up new vistas for those determined to avoid their tax obligations.

Indeed, the use of unaccounted e-money for Internet transactions may present a far more difficult problem than does cash itself in the real world. First, it can be handled easily even in large denominations (no more satchels full of heavy paper to lug around), and second, while the movement of ordinary cash can be discerned by observing the movement of physical inventory, on the Internet many transactions will involve intangibles that cannot be so easily traced.

No need to panic -- yet. We don't know whether consumers will use unaccounted e-money systems at all, or whether they will do so for anything other than the small transactions for which cash is already the preferred medium of exchange. But it is surely conceivable that they will, especially if that offers a way to avoid tax obligations.

This is not going to be an easy one to solve. Two years ago in this column I proposed that we start thinking about an answer, and now that the US Treasury has thought about it, we don't seem any closer to a resolution. The Report ends with a request for comments on "how the tax system can be adapted to deal with such disintermediated micro-transactions" -- which, appropriately enough, can be sent by e-mail to Perhaps, then, it might be a good time for those of you with clients interested in tax issues -- and are there any clients not interested in tax issues? -- to chime in.