Role and Efficacy of International Bodies and Agreements

U. S. Perspectives on Consumer Protection in the Global Electronic Marketplace

Federal Trade Commission Public Workshop

Discussion Paper

June 9, 1999

Henry H. Perritt, Jr.[1]

Dean, Chicago-Kent College of Law

Illinois Institute of Technology

Introduction and summary

An explosion of electronic commerce in the Internet gives rise to concerns about the efficacy of existing legal systems. The Internet does not change human nature, abolish greed or ensure ethical behavior. When large enterprises are involved, existing domestic and international legal rules and institutions -- governmental and private -- can police commercial conduct. But the Internet's low barriers to entry and low transaction costs facilitate transactions in which small enterprises and individuals deal directly with each other or through new kinds of electronic markets. For these types of transactions, new types of private regulatory systems, backed up by innovative public international law institutions, may be necessary.

This paper explains why electronic commerce on the Internet may necessitate the development of new legal systems of an international character, argues that consumer protection is an especially appropriate subject for legal innovation, inventories models of international legal regulation, evaluates political factors likely to shape legal development, and concludes by identifying particular issues for special attention.

If legal and public policy communities accurately identify the unique characteristics of Internet-based electronic commerce, while acknowledging the inevitable -- and appropriate -- role of government, they can make use of public and private tools to create a framework for new self regulatory mechanisms, which can reduce uncertainty and preserve low transaction costs, dealing with jurisdictional uncertainties, thus permitting the benefits of the new characteristics of the Internet to be enjoyed within the inevitable political and legal judgments animated by traditional concerns.

Why should we care about the Internet and international legal frameworks? 

Transnational merchant and consumer transactions have been occurring for thousands of years. International legal systems, including the Law Merchant, bodies of substantive commercial law, and private international law, are well accepted frameworks for such transactions. The Internet provides a new kind of market for electronic commerce ("e-commerce") transactions that cross national borders, but it is not necessarily true that a new international legal system is necessary for international e-commerce.

So two threshold questions must be answered in thinking about new roles for international institutions in regulating the Internet. The first is, “Why should we expect the Internet to require regulation at the international, rather than the domestic[2] level; i.e. why cannot domestic law handle whatever problems may arise from growing electronic commerce in the Internet?”

The second question is “Why do we need new institutions; i.e. why cannot traditional international and domestic institutions handle the problems of electronic commerce, domestic or international?”  These two questions structure analysis of the broader issue “Does existing law need to be modified for application on the Net or do we need a new set of principles and adjudicatory procedures for international electronic commerce?”

Seven propositions frame responses to the question.

First, some things have not changed.

Human behavior is the same.  Crooks still exist, and they are using new Internet media just as they used older technologies.  Malcontent consumers still exist, and some of them are making unreasonable complaints about products and services delivered through the Net just as they always have about products and services delivered in other ways.

Certain concerns that shape legal rules have not changed.  The need to strike an appropriate balance between private ordering on the one hand and public regulation on the other has not changed. Underlying concerns about inequality of bargaining power are still appropriate to limit purely private regulation.  Also, it remains true that private self-regulatory mechanisms usually involve agreements among competitors that raise questions of antitrust law.  These issues determine where the boundaries should be drawn between deference to private regulation and insistence on public regulation.  And, concerns about certain core matters of public order and protection of vulnerable groups will continue to drive the political process in democratic societies and ultimately determine the law’s rules.

Second, some things are new with the Internet.

The Internet offers sharply lower transaction costs, and this makes it possible for low value transactions to take place on the Net. It presents low barriers to entry, and this makes it possible for small enterprises and individuals to participate directly rather than through traditional intermediaries. More profoundly, the Internet is an inherently global phenomenon and it is difficult to localize conduct.  This is so even though all of the world’s legal systems are premised on localization because sovereign states are defined in terms of geographic boundaries.

Third, the difficulty in localizing conduct creates jurisdictional problems. 

Thomas L. Friedman observes that The Internet is going to be like a huge vise that takes the globalization system[3] . . . and keeps tightening and tightening that system around everyone, in ways that will only make the world smaller and smaller and faster and faster with each passing day. . . . [T]hanks to the Internet, we now have a common, global postal system,  . . . we now have a common global shopping center . . . we now have a common global library, and we now have a common global university."[4]

David R. Johnson and David Post say:

“Cyberspace[5] radically undermines the relationship between legally significant (online) phenomena and physical location.  The rise of the global computer network is destroying the link between geographical location and: (1) the power of local governments to assert control over online behavior; (2) the effects of online behavior on individuals or things; (3) the legitimacy of a local sovereign's efforts to regulate global phenomena; and (4) the ability of physical location to give notice of which sets of rules apply.  The Net thus radically subverts the system of rule‑making based on borders between physical spaces, at least with respect to the claim that Cyberspace should naturally be governed by territorially defined rules.”[6]

Statistics presented by the National Consumers League for 1997 show that Internet fraud complaints involving non-U.S. consumers ranked 8th, just after Illinois consumers and just ahead of Virginia consumers.[7]

Of course “Information technology has been thought to erode the power of sovereigns since at least the time of James I of England, when the spread of the printing press alarmed the Church and the Crown.”[8] So one might view expressions of alarm with some skepticism.[9] Nevertheless, certain features of the Internet—its global reach, and its low barriers to entry -- distinguish it from printing, radio and television, suggesting that traditional sovereigns may have more difficulty regulating it than earlier information technologies.[10]

When electronic commerce occurs through the Internet, barriers to entry and transaction costs do not vary with distance. In this respect, the Internet is different from print, radio, television and telephone channels. The irrelevance of distance makes the Internet inherently global, although barriers of language and culture still remain.

It also is true, however, that the Internet may make traditional regulators more effective.[11] The Federal Trade Commission and the National Association of Attorneys General offer more visible complaint resolution mechanisms through the Internet than any private dispute resolver.

Jurisdictional issues exist in two different forms, with respect to prescriptive jurisdiction—what American lawyers call choice of law[12]—and adjudicative jurisdiction—what American lawyers call personal jurisdiction. Both types of jurisdiction are thrown into question when commerce occurs in Internet-based markets. If I put up a Web[13] site in Georgia, will I be bound by a judgment entered by a court in Guanxi, China?  Will the laws of Bainbridge, Georgia, or of Bulgaria govern my site?

Fourth, these jurisdictional issues create uncertainty and the potential for conflicting regulation.  

Both uncertainty and conflicting regulation increase transaction costs,[14] reducing the potential for the Web to fulfill its potential as a marketplace for small transactions.

It is possible, of course for traditional sovereigns to regulate the Internet. Professor Jack Goldsmith regularly reminds his academic colleagues that plenty of effective targets – users, and Internet service providers – are within the reach of geographically based power.[15] Goldsmith argues that those concerned about Internet jurisdiction make three basic errors: they overstate the differences between cyberspace transactions and other transnational transactions; they fail to distinguish between default laws and mandatory laws; and they underestimate the potential of traditional legal tools and technology to resolve the multijurisdictional regulatory problems implicated by cyberspace.[16]

The argument between Goldsmith and others is not as intractable as it might seem. Peter Swire suggests why not by distinguishing between “elephants” and “mice.”[17]

"Elephants are powerful and have a thick skin, but are impossible to hide.  They are undoubtedly subject to a country's jurisdiction.  Once legislation is enacted, they likely will have to comply.  By contrast, mice are small and mobile actors, such as pornography sites or copyright violators, that can reopen immediately after being kicked off of a server or can move offshore.  Mice breed annoyingly quickly‑new sites can open at any time.  Where harm over the Internet is caused by mice, hidden in crannies in the network, traditional legal enforcement is more difficult.  In such instances legal enforcement, to be successful, will focus on someone other than the mice themselves.  Candidates for enforcement include the individual users, the Internet service providers, the financial intermediaries that transfer money to the mice, and the offshore countries that provide the mice a cozy nest."[18]

Mice are more numerous in the Internet than in other global marketplaces because of the Internet's low barriers to entry.

As I have observed “The problem is not the adaptability of International Shoe-obtaining jurisdiction in a theoretical sense. The problem is obtaining meaningful relief. The real problem is turning a judgment supported by jurisdiction into meaningful economic relief. . . . The transaction costs dwarf the value of the claim.”[19]

As the following section explains, considering mice and elephants separately helps reconcile the Goldsmith/Johnson debate. This is the underlying logic: Because actors and consequences have physical locations, traditional jurisdictional rules can be applied. The resulting judgment is of real value, however, only if it is enforceable, and enforceability involves high transaction costs.  Historically, the value of the judgment has made such costs reasonable, so that either the defendant defends initially or the plaintiff is willing to go to the defendant’s forum to enforce the judgment.  When transactions occur through the Internet the value of the dispute often will make the relative level of such costs prohibitive. Therefore, there is a need for domestic sovereigns to agree about which judgments they will enforce or for some new mechanism to replace the traditional lawsuit.

Fifth, private regulation offers some answers to jurisdictional uncertainty, because private regulatory regimes are not inherently limited by geographic boundaries of sovereign states.

Private regulation reduces uncertainty because of the capacity of private persons to specify choice of law and to select forums (including arbitration) in contractual arrangements.[20] A relatively complete treaty framework makes international arbitration awards enforceable,[21] while there is no such treaty with respect to civil judgments.[22]

Johnson and Post suggest a new legal regime, considering cyberspace to be a new kind of sovereign.[23] I have reviewed the arguments in favor of self-governance of cyberspace and concluded that “considerable autonomy can be achieved through contractual arrangements featuring arbitration accompanied by choice of customary substantive law. However, there may be difficulties in defining community boundaries, in implementing effective enforcement mechanisms, and in avoiding antitrust problems in the electronic network context.”[24]

David Oedel studied self-regulation in the banking industry over the last two centuries.[25] He identified six prerequisites for effective self regulation: absence of formal public regulation, pooling of resources and sharing of risks and rewards among those engaging in self regulation,  provision of discrete, useful service by the regulating entity, regulation by quasi independent hybrid association rather than by self interested member of regulated group, effective monopoly by regulatory entity, independent operational role by regulating entity.[26] He also concluded that self regulation is unlikely to be effective to control the behavior of atomized individuals such as consumers.[27] The transaction costs of organizing self regulation of consumers are so high that regulation through the government is more efficient[28]

Goldsmith is skeptical of proposals for private self-regulation, arguing among other things, that they fail to recognize the difference between default and mandatory rules. Default rules can be changed by private agreement; mandatory rules cannot.

There are some interesting new developments in creating credible private mechanisms for self-regulation of electronic commerce.  Such mechanisms must have teeth or they will not be credible either with consumers or with political actors.  The Department of Commerce and the information industry have been working hard in making progress in negotiating with the European Union on parameters for such a new self-regulatory regime.[29]

Sixth, however credible and complete private self-regulatory mechanisms are, they cannot function without some sort of foundation and framework derived from traditional law. 

All private, self-regulatory regimes ultimately depend on support from traditional law articulated and applied by state-based institutions. Arbitration awards must be enforced through legal machinery that can execute against property[30] and incarcerate for contempt, although the creation of a fund such as through iEscrow, discussed later in this paper eliminates the need to rely on state-based institutions for enforcement. Exclusion from portions of cyberspace as a penalty for misconduct will give rise to breach of contract or antitrust lawsuits, and the courts hearing those lawsuits must decide whether to respect or to punish the private rulemakers and enforcers. Persons complaining to private enforcers may be subject to liability for defamation or intentional interference with contractual relations.

Working out the rules for deferral and immunity are difficult enough in the purely domestic context. Because of the Internet’s global context such rules also must be worked out internationally. This means some sort of international agreement, to define the scope of permissible self-regulation and to immunize the private regulators and those who make complaints to them.

Seventh, new institutional approaches are most necessary in the international context, when transactions involve buyers and sellers located in different states.

The introduction posed the question, “Why should we expect the Internet to require regulation at the international, rather than the domestic level; i.e. why cannot domestic law handle whatever problems may arise from growing electronic commerce in the Internet?” The answer results from the interaction of two realities. First, low barriers to entry and low transaction costs result in a greater portion of Internet based transactions crossing state borders, compared with other marketing channels. Second, traditional state-based mechanisms are likely to be less effective in the international context, because of diffusion of the effects of conduct and dilution of power, legitimacy, and notice.

Why should we focus on consumer transactions?

The rapid growth of electronic commerce on the Internet is beginning to focus attention on consumer protection. According to statistics maintained by the National Consumers League, reports of consumer fraud on the Internet grew from fewer than 1,000 in 1996 to nearly 8,000 in 1998.[31] Consumer protection is a broad category, encompassing securities fraud, deceptive advertising, failure to deliver promised products or services and personal data privacy protection.  It is useful to deal with privacy separately, because it already has received so much attention.  While ultimately, it may be desirable to focus separately on distinct aspects of consumer protection, initially it is useful to seek to identify characteristics that may distinguish the broad category of consumer protection issues from other Internet legal issues such as taxation, hate speech, pornography, payment systems, and intellectual property.

Peter Swire's elephants and mice metaphor[32] explains why new international[33] regulatory institutions may be most necessary when elephants deal with mice,[34] and when mice deal with mice. Adequate international mechanisms exist for elephant-to-elephant transactions: private negotiation of rules in the form of contractual terms, and private arbitration within the framework of the Federal Arbitration Act and the New York Convention. Examples of successful private elephant-to-elephant regulatory frameworks are international funds transfer networks, and the Internet domain naming and address assignment system under Jon Postel. When mice are involved, however, these mechanisms are less effective and/or less acceptable.

Mouse-to-elephant transactions

One distinguishing characteristic of consumer protection issues is the disparity of bargaining power between seller and purchaser.  The bargained-for-exchange model of contract is conspicuously absent from the vast majority of consumer transactions.  Instead, sellers unilaterally specify the terms of the sale, offering them to consumers on a take it or leave it basis.  This disparity of bargaining power leads legislatures and administrative agencies to prescribe special terms for consumer contracts, terms, which supply default legal rules, and in some cases which are non-waivable by contract -- so-called "mandatory rules." [35]  The power disparities also lead law reform initiatives such as UNCITRAL to limit their drafting activities to business-to-business contracts[36] and to special treatment of consumer transactions under the UCC.[37]

On the other hand, it is not correct that consumers have no choices in electronic marketplaces.  Presently, the Internet intensifies competition, offering consumers a wider array of products and services from different sellers than they would have in geographically defined markets.  While consumers may not be able to bargain with their voices across the negotiating table, they can bargain with their computers, rejecting offers from sellers specifying less attractive terms.

But bargaining with a computer is a reality only while consumers are shopping, and not necessarily after they have agreed to purchase, and only when the information costs of knowing differing terms offered by different sellers are low.  Once a consumer has purchased a continuing service, switching costs may be high enough to reduce significantly the range of real consumer choices after the purchase decision is made.

Disparity of bargaining power often has invited governmental intervention, and less reliance on private ordering in consumer transactions, if only by requiring certain levels of disclosure of the terms of the bargain offered, much as the UCC requires that disclaimers of certain warranties be prominently disclosed.

A second characteristic of consumer transactions is relatively low transaction value.  Indeed, the possibility of small transactions is what makes the Internet such an interesting medium for electronic commerce.  Its inherently lower transaction costs make it economical for buyers and sellers to purchase and sell units of smaller value than they could do economically in physical markets with their higher inventory, rental and utility, labor, and transportation costs.

But low transaction value means that less is at stake when something goes wrong.  Accordingly, consumers victimized by unscrupulous or incompetent sellers are less likely to devote necessary resources to investigating and prosecuting violations of legal rights.  This creates an economic environment in which the costs of consumer abuse are diffuse while all the benefits are concentrated. Such asymmetry of costs and benefits is a traditional justification for governmental intervention in the form of publicly funded investigation and prosecution resources, for example by appropriating public funds for a governmental consumer protection agency, or by allowing class action lawsuits and attorneys fees provisions to encourage private representation of diffuse consumer interests.

A third distinguishing characteristic of consumer protection is the existence of consumer protection agencies at almost level of government.  Such existing agencies naturally seek to extend their jurisdiction into new areas of commerce to preserve their reason for existence.  Most such agencies have considerable autonomy in making rules and in allocating resources for investigation and prosecution.  The result is heightened potential for conflicting geographic demands on actors in an inherently global marketplace. 

Private ordering is a traditional response to transnational jurisdictional and regulatory conflicts, but private ordering techniques such as rule making through self-regulation and dispute resolution through private arbitration are less appropriate when disparity of bargaining power exists.  Absent public law intervention, self-regulation may simply enhance the position of sellers who are members of the self-regulation “club” at the expense of consumers who are not.  Trying to deal with this problem by giving consumers a place at the self regulation table encounters all of the difficulties of representation mechanisms encountered by the architects of domain name administration reform.

Dispute resolution through arbitration similarly may provide consumers far less due process—real or perceived—than regular courts because of relatively less knowledge on the part of consumers about arbitration procedures, and the characteristics of potential arbitrators.  This information deficit has been widely noted in connection with arbitration of individual employment disputes.  The low transaction value makes it less likely that consumers will avail themselves of private counsel.

But there is still room for private regulation of elephant-to-mouse transactions and development of effective private systems serve the interests of elephants.  There is every reason to expect that mouse-to-elephant transactions will occur to a significant degree within the framework of existing credit card payment and dispute resolution systems.[38]  Credit card payment systems include both rulemaking and quasi-adjudicatory elements.  The rules are embodied in merchant agreements, which define the relationship among merchant sellers, their banks, and the credit card issuer, and credit card consumer agreements, which define the relationship between the card issuer and the purchaser/card holder.  Cardholders may be relieved of the obligation to pay for merchandise or services if they complain to the card issuer and an investigation reveals violation of the terms of the merchant agreement.  Typical merchant agreements obligate merchants to make timely delivery of conforming goods or services, and to respond to complaints.  These credit card systems typically do not provide for “hard” dispute resolution in the form of third party arbitration.  Nor do they provide for referral of complaints to public authorities.  They are backed up by the economic threat of terminating service to a merchant or terminating the extension of credit to cardholders.   The framework provided by dimensional law for these private regulatory mechanisms is relatively thin.  There are only modest requirements under federal law, [39] and there is always the possibility of a breach of contract lawsuit by buyers or sellers against each other or against the credit card issuer.  Such lawsuits involve application of choice of law rules under private international law and choice of law provisions in the merchant and cardholder agreements, reinforced by the European contractual choice of law convention.  Personal jurisdiction would be decided under private international law rules, the Brussels and Lugano conventions and under the proposed Hague convention. 

Mouse-to-mouse transactions

The Internet's low barriers to entry and low transactions costs make it easier for mice to be involved on both sides of transactions. A mouse who wants to be a seller need not set up a large purchasing and distribution network. According to statistics gathered by the National Consumers League,[40] more than two-thirds of the Internet fraud complaints involved auctions, where it is reasonable to infer that mice were involved as sellers as well as buyers. Only about a quarter involved "general merchandise" and "computer hardware and software" and "Internet related services," where it is reasonable to infer that an elephant was the seller. Most of the remainder involved "work-at-home," business opportunity/franchises," "multi-level marketing and pyramids," "credit card issuing," "advance fee loans," and "job offers/overseas work," where it also is reasonable to infer that mice were the sellers.

When mice deal with mice, their relationships are unlikely to be sufficiently persistent to give rise to private rule making or establishment of arbitration agreements.  Some disputes involving mice on both sides will be resolved in the traditional way, by lawsuits in national courts and the claims made through administrative agencies.

Should all this be worked out through the application of private international law concepts of comity, or should it be worked out within a framework developed by international treaty or contractually provided by new kinds of private intermediaries such as Ebay? 

Market forces well may stimulate creation of fairly complete systems of private dispute resolution for mouse-to-mouse disputes.  Ebay is a good example.  Ebay [41] is a Web-based auction service popular with individuals and small businesses wishing to buy and sell merchandise and services. Apparently because Ebay fears that people will stop using its service because of threats of fraud and non-performance of agreements, it offers several mechanisms for avoiding or resolving disputes.  One mechanism is iEscrow, which allows purchasers to escrow their payments until they accept delivered merchandise.  The function of iEscrow is very much like a documentary letter of credit.[42]  Both the letter of credit and the iEscrow are intended to protect the interests of both buyer and seller in international transactions where there is a risk of non-performance on one side or the other.  Ebay also offers a complaint mechanism and feedback and evaluation of buyers and sellers.  The combination of these services permit a mouse to assess, in advance, whether it wishes to deal with another mouse based on that mouse’s reputation in the Ebay community, and then, if a complaint arises, to enlist Ebay in attempting to resolve the dispute.  Ultimately, Ebay commits only to refer serious complaints to public authorities.  This is an obvious example of traditional law backing up a private dispute resolution mechanism.

The National Consumer League provides a somewhat similar service,[43] although it, unlike Ebay, is not involved in facilitating electronic commerce directly.  NCL accepts complaints of e-commerce fraud and forwards them to the Federal Trade Commission and to state attorneys general.  This is another example of a new type of intermediary that helps mice mobilize the resources of public authorities.

Some of this is not very international yet. The NCL service only provides for reference to U. S. authority, and there is no suggestion that Ebay will refer complaints to consumer protection authorities in other countries.  On the other hand, there is nothing about the iEscrow, the feedback and evaluation system, or the internal Ebay complaint resolution mechanisms that limit them to national boundaries; they function just the same in transnational transactions as in local ones.  Moreover, it is entirely conceivable that including arbitration within the scope of the New York convention could strengthen the Ebay complaint resolution machinery.  Agreement to arbitrate could be either a condition of doing business on Ebay or an option much as iEscrow is an option.

Politics of consumer protection

The strongest political forces will focus on elephant-to-mouse transactions, because it is those transactions that engender broad sympathy for overreaching and abuse of private power. It is such political forces that stimulated so many states to set up consumer protection units in their attorneys general offices in the first place.[44]

Three political realities will shape the evolution of international – and national – consumer protection mechanisms. First, the usual mechanisms of issue identification and priority setting in domestic politics are likely to prefer government-based to private self-regulatory mechanisms. The aphorism that foxes should not guard chicken coops has not lost its power merely because of the Internet. Second, private enterprises always will prefer unilateral decisionmaking and other attributes of autonomy to any form of third-party dispute resolution. That is why non-union employment dispute arbitration has been so slow to take root.  That is why the Virtual Magistrate project did not, in the end, produce the expected flood of referrals from America On Line and other private providers. Third, much of the ethos of Silicon Valley is undergirded by a remarkable ignorance of politics and government, and ignorance breeds contempt. Some (but not all) of the demand for the Internet being its own sovereign simply reflects lack of awareness of what sovereign governments are. These attitudes of elephants will discourage development of effective private regimes for protecting mice when they deal with elephants.

The combination of the second and third factor are making it extremely difficult to negotiate an agreement with the EU regarding privacy; a significant faction within the interested industry rejects the legitimacy of any government regulating conduct affecting its citizens when they deal on the Internet.

A taxonomy of international mechanisms

Several types of international legal frameworks could regulate -- or back up private regulation of -- consumer transactions on the Internet.  The most traditional would be a treaty framework, giving rise to state-based institutions resembling the World Trade Organization, the International Civil Aviation Organization, or the World Intellectual Property Organization, or supporting private regulation as in the case of UNCITRAL's model laws for electronic commerce and for international credit transfers.

Too many people interested in the Internet’s development make the mistake of assuming that treaties cannot be part of the answer.  They mistakenly assume that treaties take decades to negotiate.  They do not.  The Department of Justice often negotiates bilateral mutual assistance treaties relatively quickly and, when the relevant interests where aligned and sufficiently motivated, new copyright treaties were negotiated with breathtaking speed in Geneva.  Whether or not one likes their content, they did not take decades to negotiate. International treaties do represent a potentially useful tool in shaping the legal climate for the Internet.  In any event, negotiations between governments, like the Department of Commerce/EU negotiations over privacy, will be an inevitable feature of law making for this new medium. The Hague Conference on Private International Law has underway two relevant projects that may provide helpful treaty frameworks. The first is a “Special Commission on international jurisdiction and the effects of foreign judgments in civil and commercial matters,”[45] which aims at negotiation of an international judgment-recognition treaty. The second is a working group on Internet jurisdiction, scheduled to meet in September, 1999.

There several difficulties with this approach, however.  First, private-sector forces are hostile to treaty based organizations because of unsatisfactory experience with many of the long-established frameworks.  The frameworks are widely perceived as being slow, being dominated by states that have little immediate connection with Internet commerce, and having a tendency toward over bureaucratization.  It also is widely perceived that it takes too long to negotiate the requisite treaties, and that technology and the problems purportedly justifying legal regulation are likely to change dramatically by the time a treaty can be negotiated.

Some of the shortcomings of the traditional treaty based approach might be alleviated through the same mechanisms employed in structuring domestic U.S. legal institutions as the regulatory state developed in the 20th century.  This approach involves delegation of rule making and adjudicatory functions to administrative agencies within a relatively broad statutory (treaty based) framework.

An international administrative agency approach also suffers from some deficiencies.  First, it may be politically unpalatable because it would involve ceding more sovereignty to international administrative agencies that suffer the same flaws as existing treaty-based organizations. If states A and B cannot agree on substantive rules and judgment enforcement, it is not intuitively obvious that they could agree on a new body to craft rules and to enforce the rules thus crafted.

The traditional treaty-based framework limits participation to states rather than to natural or artificial persons and rather than delegating rulemaking adjudicatory powers to international institutions, only provides frameworks for further negotiations over the rules by states and provides a some or framework for adjustment of disputes by state representatives rather than providing for "hard" adjudication.  Taking the next logical step by delegating rulemaking and adjudicatory powers to international administrative agencies encounters legal as well as political problems.  It is far from clear that such delegation could pass constitutional muster in either the United States[46] or Germany.[47]

Private self-regulatory mechanisms avoid the difficulties with both conventional treaties and treaties establishing international administrative agencies.  Building upon the experience with international banking networks, and drawing foundation also from the New York convention for adjudicatory dispute resolution through arbitration, they may be the best approach for elephant-to-elephant transactions.  Even here, however, local law such as article 4A of the Uniform Commercial Code and the Model Law on transnational credit transfers, anchor the private self-regulatory networks.

The likelihood of hybrid private/public arrangements is greatest for elephant-to-mouse transactions.  Purely private regulation of elephant-to-mouse transactions confronts several potentially insuperable problems. First, purely private regulation engenders the accusation that the "fox is regulating the chicken coop," giving rise to pressure for more government involvement. Second, disparity of bargaining power and asymmetry of information makes private arbitration arrangements suspect. Third, the difficulty of formulating fair, acceptable and efficient representation arrangements for both elephants and mice makes it difficult to design rulemaking institutions.  Given the likely resistance to purely private regulation on the one hand, and the advantages of private decision making on the other, it would be fruitful to focus governmental energy on negotiating “safe-harbors” for private regulation, as in the EU/DOC privacy discussions.

More complete private self-regulatory mechanisms may be most appropriate for mouse-to-mouse transactions, where the efficacy of government-based regulation at the state level is lowest.  New kinds of intermediaries who organize markets for such transactions in the Internet (eBay is a good example) will be motivated to enhance consumer security by providing for private dispute resolution mechanisms and other transactional security mechanisms such as escrow and letter of credit arrangements, under the framework of privately prescribed rules unilaterally determined by the intermediaries.  There is no obvious need here for government involvement, although designers of such private systems must recognize and design for the potential that unhappy participants may will jump out of the private mechanisms into local courts seeking resolution of certain disputes.  The advantages of private intermediary-provided mechanisms for mouse-to-mouse transactions include the avoidance of interstate jurisdictional and choice of law controversies. These private mechanisms are better than national lawmaking and dispute resolution procedures when the resources of claimants are limited and the ability of respondents to pay judgments similarly limited, compounded with the difficulty of finding small respondents.  The deficiencies of national legal systems are compensated for by the availability of resources available to pay claims through escrow or letter of credit arrangements and the effectiveness of the denying further participation to buyers and sellers through the intermediaries.

Moving toward private standing

None of the traditional mechanisms for dealing with transnational trade provide directly for private recourse to dispute resolution mechanisms.[48]  It is conceivable that direct private access could be allowed under new treaty based arrangements, but political opposition and constitutional difficulties with respect to any mechanisms that look a lot more like the adjudication of private claims represent large barriers to such a course of action.  It may be better to conceive of hybrid arrangements that allow private parties recourse to private adjudicatory, and participation and private rulemaking procedures, backed up by a treaty based framework that obligates and permits state representatives to work out mechanisms for granting immunity to and recognizing the decisions of the private bodies.

What needs to be done now?

 All of this suggests that Internet consumer protection law initiatives fruitfully can focus on the following issues:

·         How different are the basic ground rules imposed by public agencies on consumer transactions?  If they are not very different, it may be possible to harmonize basic rules through an international treaty or through UNCITRAL or OECD guidelines that can be applied by existing institutions. Progress toward cooperation among the FTC and other national consumer protection agencies,[49] and cooperation in the antitrust arena[50] are examples of the viability of this approach.

·         Is it feasible to develop representation arrangements that would ensure a reasonable balance of consumer and seller viewpoints in self-regulatory mechanisms, without swamping the self-regulatory process with disputes over representation?  If such mechanisms are feasible, they could be made part of the criteria for accreditation of self-regulatory entities.

·         What attributes of an arbitration systems are necessary to make if fair for consumer disputes, including the burden of participating, accessibility, selection and identity of arbitrators, and reviewability of awards?  A starting point for such an inventory might be found in the EEOC guidelines for arbitration of individual employment disputes.  If an appropriate inventory of characteristics can be developed, it could be imposed as one of the criteria for accreditation of private dispute resolution mechanisms.  Accreditation might result in a higher degree of deference by and preclusive effect in public judicial forums.

·         What terms would be appropriate for an international convention that provides a safe harbor for private regulation of mouse-to-mouse transactions, granting immunity from tort and antitrust liability, requiring deferral to private decisions, and allowing for their enforcement through state-based coercion?

It is especially appropriate to concentrate on developing hybrid systems for elephant-to-mouse commerce, while waiting for further spontaneous evolution of mouse-to-mouse intermediaries.

This workshop is one important arena for considering these questions. Another important arena is the American Bar Association Internet Jurisdiction Project.[51]


Nationally and internationally a strong trend toward privatization of rule making and adjudication is likely.  It is implausible, however, that governments will wash their hands altogether of disputes over electronic commerce.  Political pressures for governmental involvement will be especially strong to protect consumers, especially respect to elephant-to-mouse transactions.  The interaction between privatization and pressures for some governmental involvement will produce some interesting new hybrid international institutions – combinations of contracts and treaties, and combinations of contracts and regulatory agency approvals and enforcement. Purely private regulation of mouse-to-mouse and elephant-to-elephant transactions will prove more viable than purely private regulation of elephant-to-mouse transactions.

[1] I appreciate thoughtful comments and suggestions from my colleague, Margaret G. Stewart, reporter for the ABA Internet Jurisdiction Project, and from my friend, Stuart P. Ingis, Esq., of Piper & Marbury.

[2] "Domestic," "municipal," and "national" refer to state-based, as opposed to international, law, legal institutions, and enforcement machinery, as the context requires.

[3] Friedman argues that globalization changes the traditional balance between nation states, that it changes the relationship between nation states and global markets, and -- most profoundly -- that it changes the balance between individuals and nation states. Friedman at 11-12.

[4] Thomas L. Friedman, The Lexus and the Olive Tree 118 (1999).

[5] "Cyberspace" refers to the virtual space in which personal, political and commercial relationships can be established. It is a superset of the Internet, including also private electronic networks using other protocols.

[6] David R. Johnson & David Post, Law and Borders – The Rise of Law in Cyberspace, 48 Stan. L. Rev. 1367, 1370 (1996)

[7] <>

[8] Henry H. Perritt, Jr., The Internet as a Threat to Sovereignty? Thoughts on the Internet's Role in Strengthening National And Global Governance, 5 Ind. J. Global Legal Stud. 423, 425 (1998).

[9]  See Henry H. Perritt, Jr. & Margaret G. Stewart, False Alarm: European Privacy Law and International Jurisdiction, 51 Fed. Comm. L. J. 811 (1999) (suggesting that it is hard to find jurisdictional problems with European regulation of transnational personal data flows)

[10] Perritt, 5 Ind. J. Global Legal Stud. at 428-29.

[11]  See Perritt, 5 Ind. J. Global Legal Stud. At (explaining how the Internet can strengthen both national and international legal institutions).

[12]  “[Prescriptive jurisdiction] refers to ‘the authority of a state to make its law applicable to persons or activities,’ and is quite a separate matter from "jurisdiction to adjudicate," Hartford Fire Ins. Co. v. California, 509 U.S. 764, 813 (1993) (quoting Restatement (Third of Foreign Relations Law of the United States).  The concept of prescriptive jurisdiction is incorporated into municipal (national) conflict of laws rules. Banco Nacional de Cuba v. Sabbatino, 376 U.S. 398, 445 (1964).  Prescriptive jurisdiction enters into conflict of laws analysis in the following way. When a court must make a choice of law decision, applying its own choice of law rules, it considers whether its own lawmakers (legislators or common law decisionmakers) have acted within the limits of prescriptive jurisdiction to extend their substantive law to a transaction that has some extraterritorial features. It also considers whether another sovereign has acted within its prescriptive jurisdiction to apply its substantive law to the same transaction. When only one sovereign has prescriptive jurisdiction, that sovereign’s law applies. When both sovereigns have prescriptive jurisdiction over the transaction, other choice-of-law rules must be applied to resolve the conflict. Similar analysis applies when a second sovereign is called upon to recognize a judgment from another sovereign.

[13] The World Wide Web is the dominant Internet space. The Internet also includes electronic mail, newsgroups, and facilities for remote control of computers. Most of this paper focuses on the Web.

[14] "[A] single [Internet] actor might be subject to haphazard, uncoordinated, and even outright inconsistent regulation by states that the actor never intended to reach and possibly was unaware were being accessed." Swire, 32 Int'l Law. at 991.

[15] See Jack Goldsmith, What Internet Gambling Legislation Teaches About Internet Regulation, 32 Int'l Law. 1115 (1998)(rebutting conventional wisdom that federal legislation prohibiting Internet gambling cannot work; states can punish users who reside there, and can force domestic Internet providers to exclude gambling); Jack L. Goldsmith, The Internet and the Abiding Significance of Territorial Sovereignty, 5 Ind. J. Global Legal Stud. 475, 475 (1998) (“from the perspective of jurisdiction and choice of law, territorial regulation of the Internet is no less feasible and no less legitimate than territorial regulation of non‑Internet transactions.”)

[16] Jack L. Goldsmith, Against Cyberanarchy, 65 U. Chi. L. Rev. 1199, 1200-1201 (1998).

[17] Peter P. Swire, Of Elephants, Mice, and Privacy: International Choice of Law and the Internet, 32 Int'l Law. 991 (1998), Elephants are large enterprises, such as AOL. Mice are small enterprises and individuals.

[18] Swire, 32 Int'l Law. at 993.

[19] Henry H.  Perritt, Jr., Will the  Judgment-Proof Own  Cyberspace? 32 Int'l Law. 1121, 1123 (1998).

[20] See generally (International Chamber of Commerce information on international commercial arbitration).

[21] New York Convention.

[22] The Brussels Convention provides for judgment recognition within the European Union, but attempts to negotiate such a treaty with global scope so far have proven unsuccessful.

[23]  Johnson & Post, Stan. L. Rev. at ___; Perritt, 12 Berkeley Tech. L.J.  at 419 (“self-governance may be more efficient; the rules and/or the adjudicatory techniques for applying the rules may need to be different from those of the surrounding community; it may be impracticable to apply the rules of the surrounding community; or compliance with basic norms of the community may be higher when members of the subcommunity participate in self- governance).

[24] Henry H. Perritt, Jr., Cyberspace Self-Government: Town Hall Democracy Or Rediscovered Royalism?, 12 Berkeley Tech. L.J. 413, 417 (1997) (also suggesting points of tangency between cyberspace and other legal systems and rules of thumb for sovereign deference to cyberspace “sovereignty”).

[25] David G. Oedel, Private Interbank  Discipline, 16 Harv. J. L. & Pub. Pol’y 327 (1993).

[26]  16 Harv. J. L. & Pub. Pol’y at 402-404.

[27] Id. at 406-407.

[28] Id. at 407.

[29] National Telecommunications and Information Administration, Department of Commerce, Notice and request for public comment, Elements of Effective Self Regulation for the Protection of Privacy and Questions Related to Online Privacy.

[30] Suggestions to recognize intangible Internet interests as property subject to execution, see Perritt, 32 Int’l Law. at 1138-39, are beginning to be accepted by the courts. See UMBRO Int’l Inc. v. 3263851 Canada, Inc., No. 174388 1999 WL 117760 (Va. Cir. Ct. March 12, 1999) (allowing garnishment of Internet domain name).

[31] <>

[32] Peter P. Swire, Of Elephants, Mice, and Privacy: International Choice of Law and the Internet, Elephants are large enterprises, such as AOL. Mice are small enterprises and individuals.

[33] The Swire metaphors also explain why new domestic mechanisms may be necessary. As explained earlier, however, the stress on traditional approaches is greatest when transborder transactions are involved.

[34] A consumer is one type of mouse. (A small, individual seller is another type.)

[35] Article 5 of the European Economic Community Convention On The Law Applicable To Contractual Obligations provides in material part as follows: “Notwithstanding the provisions of Article 3 [providing that a contract shall be governed by the law chosen by the parties.”] a choice of law made by the parties shall not have the result of depriving the consumer of the protection afforded him by the mandatory rules of the law of the country in which he has his habitual residence.


“ - If in that country the conclusion of the contract was preceded by a specific invitation addressed to him or by advertising, and he had taken in that country all the steps necessary on his part for the conclusion of the contract, or


“ -  If the other party or his agent received the consumer’s order in that country, or


“ - If the contract is for the sale of goods and the consumer traveled from that country to another country and their goods was ordered, provided that the consumer’s journey was arranged by the seller for the purpose of inducing the consumer to buy.”

Article 5 (2) of the New York Convention allows signatories to refuse enforcement of arbitration awards when “(a) the subject matter of the difference is not capable of settlement by arbitration in the law of that country, or

 “(b) the recognition or enforcing of the award would be contrary to the public policy of that country.”  This language leaves it open to enforcing states to provide by domestic law that consumer contracts are not arbitrable.

[36] A footnote to Article 1 “Sphere of application” of the UNCITRAL Model Law on Electronic Commerce (1996), http//, provides, “this law does not override any rule of law intended for the protection of consumers.”

[37] Article 2-102 of the Uniform Commercial Code provides, “nor does this Article impair or repeal any statute regulating sales to consumers, farmers or other specified classes of buyers.”

[38] See generally David G. Oedel, Private Interbank Dicipline, 16 Harv. J. L. & Pub. Pol’y 327, nn.139, 142 (1993) (summarizing credit card clearing process and dispute resolution procedures).

[39] 15 U.S.C. §§ 1601-1666j (1997); and 12 C.F.R. § 226.13 (1999) (Federal Reserve Regulation Z, imposing requirements on credit card billing dispute procedures).

[40] <>



[42] A letter of credit is a written commitment by a bank to make payment of a defined amount of money (usually from an account set up by a buyer) to a beneficiary (usually the seller) according to the terms and conditions specified in the contract between the buyer and seller, typically the presentation of certain documents, such as a bill of lading or an inspection certificate. Examples of documentary letter of credit forms can be found at <> (visited 29 May 1999).

[43] See Fraudulent Schemes on the Internet, Remarks to the Senate Permanent Committee on Investigations, by Susan Grant, Director of the National Consumers League's National Fraud Information Center/Internet Fraud Watch Programs

February 10, 1998,

The NFIC was originally established in 1992 by the National Consumers League, “Consumers can .. . report [Internet] fraud through our Web site,, or by calling the hotline at 1-800-876-7060. Though the Web site was launched only two years ago, we have had more than 5 million visitors to date. Every week, the National Fraud Information Center and Internet Fraud Watch programs receive an average of 1,500 calls and an equal number of e-mails, plus dozens of letters . . . .We also take reports from consumers about possible telemarketing or Internet fraud and relay them to a variety of federal, state and local law enforcement agencies in the United States and Canada.

Our data system uploads new reports daily to an electronic database maintained by the Federal Trade Commission and the National Association of Attorneys General,”


Charts showing types of fraud and geographic location for 1997 are available at Comparison charts for 1998, showing types of fraud, age of victim, and payment methods are available at

[44] Commissioners Starek, speaking in 1995, identified some of the problems confronting consumer protection agencies in dealing with Internet fraud and consumer disputes, while also illustrating examples of international governmental operations in investigating and remedying such problems. See generally Prepared Remarks of Roscoe B. Starek, III, Commissioner, U.S. Federal Trade Commission, on the topic of "Consumer Protection in the Age of Borderless Markets and the Information Revolution" Prepared Remarks of Roscoe B. Starek, III Commissioner U.S. Federal Trade Commission before the Conference on Transborder Consumer Regulation and Enforcement University House Balmain Crescent, Australian National University Canberra, Australia June 7, 1995,

[45] Hague Conference On Private International Law, Work. Doc. No. 144 E.

[46] Henry H. Perritt, Jr. International Administrative Law for the Internet: Mechanisms of Accountability [forthcoming in Admin. L. Rev.]

[47] Brunner v. European Union Treaty, [1994] 1 C.M.L.R. 57 (BVerfG 1993) ("Accordingly the Federal Constitutional Court will review legal instruments of European institutions and agencies to see whether they remain within the limits of the sovereign rights conferred on them or transgress them.")

[48] For example, Article 20 of NAFTA allows “parties” (the three state signatories) to invoke dispute settlement procedures. Under ANNEX 2 of the WTO Agreement, “Understanding On Rules And Procedures Governing The Settlement Of Disputes”, only “parties” to disputes may use the dispute resolution procedures. The language makes it clear that only “members” can be “parties.”

[49] See generally Prepared Remarks of Roscoe B. Starek, III, Commissioner, U.S. Federal Trade Commission, on the topic of "Consumer Protection in the Age of Borderless Markets and the Information Revolution" Prepared Remarks of Roscoe B. Starek, III Commissioner U.S. Federal Trade Commission before the Conference on Transborder Consumer Regulation and Enforcement University House Balmain Crescent, Australian National University Canberra, Australia June 7, 1995, (giving examples of interstate cooperation among consumer protection agencies).

[50] See Department of Justice and Federal Trade Commission, Antitrust Enforcement Guidelines for International Operations, 59 Fed. Reg. 52,810 (Oct. 19, 1994).

[51] See