SYMPOSIUM ON THE INTERNET AND LEGAL THEORY: THE INTERNET, SECURITIES REGULATION, AND THEORY OF LAW
Tamar Frankel *
* Professor of Law, Boston University School of Law. I am indebted to Professors Hugh Baxter and Wendy Gordon of Boston University School of Law for their insightful comments on this article. Many thanks to my assistants Dan Pierce and William Hecker for their meticulous research and editorial comments.
... Rarely has a change in the environment affected society as dramatically as the Internet. ... Further, the Internet environment is particularly conducive to the creation and enforcement of customary law. ... B. The Commission's Response to the Internet Environment: Substance, Enforcement, Mechanisms for Change, and Methods of Change ... The law was adapted to the changing environment of the Internet by allowing its use (and its advantages) while maintaining the values and policies underlying the securities laws. ... Thus, securities law enforcement has used the new technology; for example, the Commission has established areas on its Internet site where investors can inform the Commission about violations of the law and ask questions from the staff. ... Because the Internet can be used to exchange information, it can be used as a forum for securities trading, similar to securities exchanges. ... If traditional intermediaries disappear and no mechanism fills the void, it is likely that trading-site markets will not continue to function, as investors find their trade agreements ineffective and too costly to enforce. ... B. Securities Lawmakers' Response to the Internet Environment: Substance, Enforcement, Mechanisms for Change, and Methods of Change ... First, nothing concerning trading sites suggests that the fundamental policies of ensuring investor confidence and facilitating capital should be changed. ...
Rarely has a change in the environment affected society as dramatically as the Internet. It has transformed the way we retain, transfer, and exchange information. At minimal cost, the Internet offers far more information at a faster pace than ever before. It enables us to interact around the globe with more people than at any time in the past. When such dramatic environmental changes occur, drastic changes in the law often follow. n1
The Internet affects the environment in which securities markets operate and the laws that govern them. n2 The use of the Internet has already begun to change the way information about securities is disseminated and the way securities are traded, n3 two activities regulated by the securities laws. n4
The purpose of this article is to begin an inquiry on a number of questions: Should the securities laws be adapted to the use of the Internet? If so, how? What path of inquiry should be taken to answer the questions, and how should we think about adapting law to a changing environment of actors and actions subject to law? These questions are limited to the securities acts regulating securities markets. The inquiry, however, is broader. It touches on the way law should change generally in a broader context of legal theory. The article starts by defining law, then examines two particular situations in which securities markets have used the Internet, and the lawmakers' response to these uses. The article concludes with a critical analysis of this response in the context of legal theory.
Part I of this article proposes a view of law as an adaptive, self-replicating system of coercive communications, consisting of three parts: substance of coercive communications, mechanisms for enforcement, and mechanisms and methods for changing law. Parts II and III of the article show how the three parts of the legal system react to behavioral changes of those who are regulated by law as their environment changes.
To examine the response of law in such cases, I chose two situations under the securities acts. The first, discussed in Part II of the article, deals with the use of the Internet to deliver prospectuses. The Securities and Exchange Commission ("Commission") issued an interpretive release in connection with transfers by the Internet in a way that maintained the status quo. n5 In this manner, the Commission allowed issuers and some investors to capture the benefits of the Internet, while reducing adverse effects of this use on other investors, in furtherance of the policies of the securities acts. n6 The three-part model of law is applied to analyze the adaptation of the securities laws on this subject.
Part III of the article describes the regulation of securities exchanges - a species of securities trading markets - and the use of the Internet to establish on-line trading sites. In some cases, the staff of the Commission allowed issuers to provide such trading sites to their shareholders. n7 Applying the model of law to this example, the far-reaching implications of such permission are highlighted and lead to a proposed method of adapting the securities laws to Internet trading sites.
Here, we must reexamine fundamental policies underlying the law and adopt a new format of changing and enforcing law. This Part examines an emerging cooperative effort among the adaptive mechanisms of law: the Commission and custom-creating markets. I believe that in order to cope with the fundamental changes that the Internet may introduce into the securities markets, two parts of law's adaptive structure - the Commission and the markets - must closely interact to allow for experiments and controls almost simultaneously. These two actors are already moving in this direction.
The proposed three-part model of law is designed to provide a framework for explaining, and especially guiding, the way we think about adapting law to changes in the environment of actors. Hopefully, a developed model will help predict the law's response to changes in the environment. I recognize that technology and markets are not the only drivers of legal change; law is also affected by the existing legal infrastructure, which may provide actors both incentives and disincentives to certain behavior and changed environments. The focus of this article, however, is on adaptation of law to changes in people's behavior as their environment changes.
The inquiry here transcends the particular issues under the securities laws. It may help develop a broader generalization - a model of law's adaptation to the changed environments of those whom it regulates. As we are experiencing an ever-increasing pace of change in all aspects of our lives, a systematic inquiry into the response of law to change is critical.
I. A View of Law: A Three-part Adaptive, Self-replicating System of Coercive Communications Used to Help Organize Society
I view law as an adaptive, self-replicating system of coercive communications. n8 While not the only view of law, this view provides a useful framework for examining law's response to a changing environment. "Changing environment" means changing beliefs, activities, and behavior of actors in the society. "Response of law" means response to this changing environment - changes in the organizational rules under which society functions. n9 Further, there is no consensus on the concept of coercion. Some would view the law as coercive only if it is backed by the force of a political unit, such as the state. n10 I take a more expansive view of coercion to include the coercive force of custom. This, of course, means that there is no clear dividing line between law and social rules that people may feel constrained to follow without threat of state-enforced sanctions. We may distinguish between customary law that is sanctioned by "legal-like" sanctions and methods, and law that is not. In any event, the problem is not unique to our context here. The law concerning the definition of contracts as opposed to non-contract obligations poses similar issues and can provide a guide to the distinction as well. n11
Law is a structured system. The structure consists of substance, enforcement mechanisms, and mechanisms and methods of adaptation. The term "system" denotes a holistic view of a complex arrangement composed of types of items and individual items related in a predictable pattern - a repetitious rhythm. n12 "Structure" denotes a stable relationship among the parts of the system: "The mutual relation of the constituent parts or elements of a whole as determining its peculiar nature or character ...." n13 The securities laws, with which I deal later, also constitute a system. We have a seamless web of systems within systems, from the atoms to the cosmos, each with somewhat different fundamental rules and yet each relating to the whole and to each other. Our organization of chaos into patterns of orders and systems is not a reality but one which we need in order to be able to think about issues and sort out solutions. This article deals with a specific legal system - the securities laws - and relates it to law generally, a broader system to which the securities laws belong.
A. Law Is a Structured System n14
Law's structure contains three main parts: (1) the substance of coercive communications to members of a society; (2) mechanisms and methods of enforcing these communications; and (3) mechanisms and methods for adapting the structure and the substance of its components to changing environments.
I recognize that the very act of enforcement can change law and that many enforcement and adaptive mechanisms can perform both functions. Some mechanisms and their impact differ in degree. By enforcing the law, the police introduce some change, but not as much as legislatures or courts. It seems that the more "hands-on" enforcement mechanisms perform, the less immediate general impact they may have. Perhaps the police change law in a way similar to the way market actors change the law, gradually and incrementally. Enforcement and change may be distinguished by their purpose, focus, and intended results. This distinction helps the analysis and justifies treating them separately.
The main part of the law's structure consists of the substance of coercive communications to members of a society. This part is organized according to the generality of the communications: (1) specific decisions, defining the relationship among individuals and entities; (2) rules that apply generally to types of individuals and entities; n15 and (3) foundational norms, values, and policies that the rules are designed to implement. n16 Rules subsume and generalize specific decisions; values and policies subsume and generalize both latter types of items. n17 This format of communication is not surprising. It mirrors the format in which humans retain information and is a usual format (though not the only one) of communicating, organizing, and retaining detailed and complex data. n18 To be sure, the features of legal substance are not as clear-cut. The characteristics of "legal values" can be attributable to non-legal disciplines as well. n19 However, taken together with law's other features there emerges a unique system that is distinguishable from other systems.
2. Mechanisms and Methods of Enforcement
Law's structure contains mechanisms of enforcing law's communications. The mechanisms include designated actors that implement these coercive measures in law. The actors themselves are not part of law; their designation, however, is. Some measures are used by governments to enforce their coercive communications; others - usually similar mechanisms - are used by the public to enforce its customs. In addition, regulators and the courts recognize accepted customs and use their enforcement powers to coerce compliance with the dictates of these customs. n20 While law prescribes the identity of law enforcers and their methods, enforcers, such as administrative agencies, can develop "soft" indirect enforcement mechanisms that produce compliance. These methods are effective when applied to regulated indus 1326 tries whose business depends on an ongoing relationship with regulators. n21
3. Mechanisms and Methods for Change
Mechanisms for change are part of an adaptive system and the structure of law. The existence of such mechanisms per se does not destabilize the structure; rather, the mechanisms are crucial to the continuous existence of the structure. The structure must be stable, as well as resilient to change. Without adaptability, the structure will be too brittle and break with even minor changes in the environment.
Adaptive mechanisms can cause changes in various parts of the structure, mostly in the number and sometimes nature of the individual items, of which the structure is the organizing pattern. Even if the types of items change, the structure may remain intact. However, the structure may lose its identity and become another type of system, or perish, if fundamental changes occur in the nature of the relationships among the items and types of items within the structure or in the main components of the structure. n22
The same principles apply to law. Most adaptive changes in law do not affect the foundational components of law's structure nor the norms, values, policies, or types of items in the substantive part of the structure; they affect the content and the number of items in the parts of the structure, leaving the structure intact. However, if norms, values, or policies within the substance of law's coercive communications change in a fundamental way, or if the mechanisms for enforcement or mechanisms for change are eliminated or significantly altered, the particular legal system may be doomed to lose its identity; it may becme another kind of system or perish. n23
In law, the mechanisms of change consist of four designated types of actors, each of which can create and adapt law: (1) legislatures that enact statutes or rules; (2) agencies that implement statutes, rules, and their own decisions, such as the Commission; (3) courts that adapt legislative statutes, rules, and their own decisions; and (4) markets that create and adapt customary laws.
Arguably, markets are part of the actors' environment rather than part of law. Clearly, markets are unique and distinguished from the three other lawmakers. First, in the markets, the actors that change the law and the actors that change the environment may be the same or closely connected. Business persons and lawyers create new customs or more likely adjust existing customs in reaction to an invention. An invention that changed the environment and customary law are both products of the market. The Internet demonstrates this point.
Second, in market law, it is harder to identify the individuals or institutions that change law. To be sure, there are leaders that trigger a new or different path. Each follower leaves its imprint on and strengthens the coercive force of new customs. Moreover, the resulting customary laws can be attributed to preferences and decisions of a multitude of individuals. Thus, market customary laws are the product of actions and ideas of numerous individuals. Their hand may be visible, but not identifiable.
The difficulty of identifying the law changers has a number of benefits and disadvantages. On the minus side, among other things: (1) it is more difficult to attach responsibility to particular actors for changes in the law; (2) there is no requirement that the changes will be rationalized or justified; (3) there are no review mechanisms regarding the substance of changed custom; (4) at the initial stage of change, it is difficult to predict where the change in customary law will lead and when it will end; and (5) market activities are difficult to monitor because few market actors have the duty to report.
Some of these disadvantages are reduced by the unique process of change: (1) review and legitimacy of an adaptive new custom depends on the degree of its following; (2) because custom must be publicized (or else it will have no followers), the rationalization of the changes in the custom may be aired in the public domain, similar to the public domain discussion of political ideas; (3) the itemized justification may not be as important when numerous actors follow the new custom. Their following is a "black box" that proves the custom's value, like the price of a product that determines its value, even though the buyers may have different reasons for buying the product; (4) market law does not become coercive immediately; law's coercive weight depends on the degree and intensity of its following. Thus, as adoption and following of market law - custom - grows, so does its coercive backing and its predictability. Hence, it is also less necessary to identify the persons who started breaking the new path; (5) whenever custom requires a review, amendment, or adoption, the other three mechanisms of change (the legislatures, courts, and administrative agencies) are available to perform the task; and (6) market activities and complaints about inadequacies of market law can be brought to the attention of the three other mechanisms of change by competitors, consumers, and other affected parties through hearings and by other means.
In addition, other lawmakers also change the environment in which law operates. The traditional lawmakers are increasingly drawing on the markets for the substance of law by incorporating "best practices" of industries. n24 Lawmakers also use the markets to enforce law through self-regulatory organizations, trade organizations, and other means. n25 Further, the Internet environment is particularly conducive to the creation and enforcement of customary law. Many of the characteristics of customary law are immediately recognizable in the world of cyberspace. Markets have become a lawmaker in too many ways to be relegated to a mere environmental factor. Therefore, notwithstanding the differences between markets as lawmakers and other lawmakers, I consider markets to be lawmakers rather than the mere environment in which law operates.
Adaptive methods differ in different structures. Karl Popper has observed such differences in genetic, behavioral, and theoretical structures. In reaction to environmental changes, genetic systems mutate in a non-cognitive and non-intentional manner. n26 Those unlucky genes that do not mutate "correctly" die, and those lucky genes that happen to mutate in the "right" way live. n27 Behavioral systems usually adapt by intentional, cognitive actions using a trial and error method. n28 Unlucky actors that do not learn from the errors suffer the consequences and die, and those who tried the right way or learned from the errors live and prosper.
Scientific theories are changed cognitively as well, but unlike behavioral systems, changes are guided by reasoning (and partially by intuition). n29 The changes occur after rigorous testing by reasoning and experiments. n30 In addition, scientific theories must pass the testing of expert colleagues. n31 In some respects the process is similar to market adaptation. A leader breaks a new path and, unless he convinces the followers, the theory dies or lies dormant until such a following is formed. A similar process applies to other lawmakers. The process, however, is more institutionalized, legitimizing change and its support only through the proper channels.
Adaptation of law is generally effected in three ways: (1) textual interpretation, by exploring meaning and applying analogy; (2) policy considerations, more remote from text: identifying the problem, explaining the problem, choosing the criteria for solutions, listing the options, and choosing the solutions in light of the criteria; and (3) culmination of decisions and actions of numerous individuals and units in the markets. However, all adaptation methods depend to a certain degree on followers and consensus, including desires of powerful organized groups. Adaptations are also affected by the lawmakers' own interests (e.g., to be re-elected or avoid being overruled by higher courts). These pressures rarely form part of the formal methods of adaptation; their existence, however, is recognized (and sometimes criticized).
Methods of adaptation depend to some extent on which type of lawmakers engage in adapting law. Custom adapts through a gradually increasing following. n32 The other three kinds of lawmakers - legislatures, courts, and administrative agencies - use methods similar to scientific theories' adaptations, based on cognitive decisions. However, courts and administrative agencies use both textual and policy methods, while legislatures use mostly policy methods.
Arguably, the methods of courts and administrative agencies differ: courts use a textual approach, while agencies use a policy analysis approach. I conclude, however, that both courts and agencies use both approaches. Administrative agencies must interpret the text of the governing legislation and follow its dictates to determine their authority and the guiding principles for their actions. However, where agencies, such as the Commission, have powers that extend beyond judicial decisions, such as ex ante advisory powers or exemptive powers, n33 they may have to use a policy analysis approach. Some judicial decisions have taken the policy approach. n34
B. Relation of Adapting Law to Existing Law
Karl Popper said that while all new scientific theories change at least parts of former theories, showing that the former theories are either wrong or incomplete, new theories encompass the (partial) truth of the theories they contradict. n35 Thus, Einstein's theory of relativity differs from Newton's theory by showing that it is not true in certain environments, yet subsumes Newton's theory for the limited environment as to which it is true. n36
In law, the relationships between new and traditional policies and rules seem different than in science. Law can be tested by norms of right and wrong as well as by truth and falsity. Theoretically, norm setting seems to allow lawmakers more discretion to change existing laws than scientific theorizers would have; lawmakers can introduce new fundamental policies and values that fully trump and deviate from their predecessors rather than subsume them.
Yet, in reality, the way in which the law changes is astoundingly similar to the way in which new theories in science are fashioned. n37 Most new legal rules and underlying policies conflict with parts of their predecessors but contain and reaffirm part of their predecessors. n38 Generally, like most new scientific theories, new adaptive laws subsume most of prior laws and only "tweak" them in certain areas.
There can be a number of reasons for the conservative attitude of lawmakers to adapting and modifying law. In fact, these reasons are similar (and some are identical) to the reasons for the doctrine of stare decisis.
First, new laws are risky and costly for lawmakers and the regulated. Existing laws have been tested, and their strengths and weaknesses in a normally changing environment are known. New laws that have not been tested may bring unanticipated results. n39 Unlike scientific theories, n40 it is almost impossible to test new laws in advance and limit these results. n41
Second, new laws impose learning costs on the legal profession, lawmakers, the regulated, and the public. Even something as minor as the introduction of new section numbers to a body of law imposes such costs. n42 In addition, while old laws have been examined and interpreted, new laws must acquire their interpretative gloss. Developing this gloss is costly.
Third, we have a limited capacity for attention. That is why we create habits that put ourselves on "automatic pilot." Doing so enables us to pay attention to new and unexpected (and, therefore, more risky) events. Old laws are habitually followed by most people (or habitually not followed by some people). That is why, while they are excited by new ideas, people cannot easily break old habits. New laws require law-abiding citizens to break old habits and create habits of compliance. n43 That is costly.
Therefore, even when logic may require a complete break from the past, experience builds on the past. People rarely adopt entirely new values and policies or rules that do not relate to existing legal and social structures and behavioral norms. n44
Finally, laws are not enacted in a vacuum. They do not start with a clean slate but against the backdrop of the other existing laws. Fundamental changes in the structure of law affect these other laws and may require drastic changes of all existing laws addressing the same area or subject matter. That is costly. Consequently, when changes in the laws are necessary, the bulk of existing law is left intact.
C. Relationship of Environmental Changes to Legal Change
The impact of a changed environment on law is not uniform. The impact depends on which part of the law's structure the changed environment affects. If the effect is on specific cases or rules, the impact of the rules will change. Therefore, rules must be amended in order to maintain the status quo: the fundamental policies which the rules implement. If, however, the changed environment affects the fundamental policies and the values on which the legal system is based, when the old order that is subject to regulation is changed, when different actors, driven by different incentives take the place of the old actors, then a far more complex response by the law may be necessary.
A new environment may eliminate the problems that existing laws, regulating the old order, were designed to reduce or eliminate; it may raise new problems that existing laws do not address. A new environment can lead to different problem definitions and values underlying the definition. In such cases the underlying policies and perhaps the very structure of the laws must be reexamined. n45
II. Regulation of Prospectus Delivery
The emerging cyberspace is posing significant challenges to legal adaptation. In the securities area, cyberspace has a serious impact in two senses. First, it creates an entirely different mode of communication. The question here is: How should legal rules change to accommodate the new communication technology? If cyberspace does not alter our fundamental values, how should legal rules adapt and change in order to maintain our current values? What should the substance of our rules be in light of the changing environment for the actors in the securities markets?
Second, cyberspace allows communication at an unheard of speed. The issue here is: Which mechanism and method for legal change is more suited to respond quickly in this new environment, recognizing that some existing mechanisms and methods of legal adaptation simply cannot operate at such speed? Police using a horse and buggy cannot match nor catch a speeding car, let alone a plane.
Based on the framework developed in Part I, these two issues relate to the substance of the legal system and to the mechanisms and methods of change of the legal system.
This Part of the article inquires into the impact of the Internet on the delivery of prospectuses. The securities acts regulate disclosure concerning securities offerings in a number of ways. Some information is required, and some information is prohibited. n46 The discussion here relates only to the requirement of delivering a prospectus to potential investors in a public offering. n47
There are two main policies underlying the securities acts. One is to maintain investors' confidence in the integrity of the markets mainly through disclosure: reducing investors' information costs and shifting these costs to the issuers and others, such as market intermediaries, for whom the costs are lower. n48 The second policy underlying the securities acts is somewhat subservient to the first but has become increasingly important in recent years. This policy is to reduce the costs of and encourage capital formation. n49 The Commission is sensitive to the costs imposed on issuers and intermediaries and is sympathetic to means of reducing the regulatory costs. n50
A. Need to Adapt Securities Regulation to Internet Communications: Impact of the Internet on the Environment of the Actors that Are Subject to the Securities Laws
As compared to traditional modes of information transfer, the Internet can reduce the cost of transmitting information for issuers, intermediaries, and some investors. Hence, these parties sought to deliver and receive prospectuses through the Internet, and they asked the Commission to clarify the status of this form of information delivery. n51
B. The Commission's Response to the Internet Environment: Substance, Enforcement, Mechanisms for Change, and Methods of Change
1. Impact on the Substance of the Securities Laws
As a new mode of communication, the Internet does not affect the desirability of information transfer, nor does it impose a different institutional arrangement concerning delivery of prospectuses. The new mode of information transfer affects only the manner in which the information is transferred. Thus, the use of the Internet does not change the values underlying the fundamental policies on which our securities laws are based, the criteria of what is good - which law should either leave alone or support - and bad - which the law should restrict or prohibit altogether. Further, prospectus delivery by the Internet does not seem to affect the fundamental policies of the securities acts. If the status quo for investors remains the same, the underlying policies of the securities acts are not adversely affected by the use of the Internet to deliver prospectuses. In fact, by reducing the costs of capital formation, the securities laws' policies are advanced.
To maintain the status quo, the rules must be changed. Investors' costs of receiving prospectuses through the Internet must be examined. First, not all investors have access to the Internet or wish to receive prospectuses through the Internet. n52 Some of those who do, however, prefer to receive information through this new medium. n53 Clearly, if investors can choose between the new mode of information transfer and the traditional one, they can evaluate their costs and make the decision accordingly. This is precisely the first condition that the Commission imposed on the use of the Internet for prospectus delivery. n54 However, once investors make the choice, they bear the burden of notifying the senders if they change their mind. After their initial choice has been made, the mode of information transfer remains the same until they change the choice.
This, however, is not all. Prospectuses are written in English. n55 Information transfer through the Internet involves an electronic language as well. Therefore, regardless of the investors' consent, the Commission requires that the electronic language used to transfer the prospectuses not be unduly complex or unavailable. n56
Professor Lawrence Lessig suggested that the first question we should ask about the regulation of the Internet is: "Should this new space, cyberspace, be regulated by analogy to the regulation of other space, not quite cyber, or should we give up analogy and start anew?" n57 In the area of prospectus delivery, the Commission chose the path of regulation by analogy. n58 Under existing laws, there are a number of acceptable ways to deliver a prospectus, such as the mails or physical handing over. n59 These traditional delivery forms have proven highly reliable. n60 There is less certainty today that Internet delivery will be as reliable, and investors bear a greater risk that prospectuses will fail to reach them. Therefore, the Commission required that senders of prospectuses receive some indication of receipt: for example, that the receiver downloaded or copied the information, or actively responded in some other way to the information. n61
In addition, materials sent by the Internet may be intercepted and changed before they reach the recipient. n62 The Commission imposed conditions on the senders to reduce the recipient's risks to this possibility, subject to a reasonableness standard. n63 Presumably, the Commission determined that by agreeing to receive prospectuses through the Internet the investors do not agree to bear this risk, except where reasonable precautions by the issuers are insufficient. Thus, so long as the costs to the information sender of protecting recipients from these risks are not high, the sender is liable.
In sum, the Commission allowed the use of the Internet and at the same time imposed conditions that equate information transfer through the new medium with information transfer through the mails and other traditional media. The law was adapted to the changing environment of the Internet by allowing its use (and its advantages) while maintaining the values and policies underlying the securities laws. The Internet rules have the same impact as the pre-Internet rules. n64
2. Response of Securities Laws' Enforcement Mechanisms
The enforcement mechanisms of the requirement for prospectus delivery are affected by the Internet. Arguably, the real world has lost control over the Internet - the cyberspace and the virtual world it has created. n65 Hence, some proponents of this view say: cyberspace should be left alone. n66 I beg to differ.
To be sure, a number of the traditional enforcement mechanisms of this "real" world cannot be effectively applied to the virtual world because the costs of such enforcement have greatly risen. For example, the Internet has increased the cost of enforcing the prohibition on misleading statements and offers of securities without registration because it is more difficult to identify and apprehend the senders and because senders can reach many parties. n67 However, the loss of control is not complete so long as some connections exist, as they must, between the real world and the virtual world of the Internet.
First and foremost are the actors. They may exist in the two worlds, but none of them occupies only the virtual world. Law applies to these actors and their actions. The Internet allows actors to do things they could not do before, in ways they could not do before, or in less time that was required before, but that is all. It may be harder to locate some actors than to locate others, but most can still be located and disciplined. In principle, those in the real world can control the virtual world.
However, in the virtual world, governments' ability to enforce some of their rules has weakened. n68 The "earthbound" actors, such as the telephone companies and other technical staff or the National Science Foundation, are not necessarily those who communicate. These conduits may be justified in rejecting responsibility for the substance of the information they transmit. Therefore, there may be a need to adopt new enforcement mechanisms and perhaps different norms to effectively regulate the virtual world. n69 Further, the focus of enforcement may change. For example, rather than regulate securities offerings on which some securities acts are based, it may be more effective to regulate securities purchases. That would bring into play the issuers' acceptance of payments rather than their communications, and the payment mechanisms that may be more identifiable and subject to regulation. In addition, the securities issuers may be regulated to require them to disseminate information more frequently and fully.
Second, new enforcement mechanisms may be created. For example, a possible enforcement avenue is to provide incentives for a new profession to police the communications or the communicators on the Internet. Cyberspace gatekeepers could be required to receive approvals from, and pay for, the monitoring of such professional actors, similar to those who must receive and pay for audited financial statements or legal opinions in order to make a public offering of securities. Such incentives to professionals are effective because the profession obtains a government monopoly and can make a good living. The cost to the gatekeepers must, of course, be evaluated, but competition among the new professionals may reduce their fees. These fees may be far lower than the losses incurred by investors short-term and by the markets long-term when prohibition of harmful practices is not strictly enforced by government.
Third, technology is now being developed to limit and control access to the Internet or access to particular audiences. For example, offers to sophisticated investors that need not be accompanied by an effective registration statement and a statutory prospectus can be made either by providing investors with special keys to the particular offering sites or by announcing that the offerings are not available to all readers. n70 Alternatively, cautionary language on offerings that certain regulatory systems do not apply to them may prove effective. n71
Fourth, the costs of enforcing law may have changed. But not all costs have increased; the Internet has helped reduce some of the costs. n72 Thus, securities law enforcement has used the new technology; for example, the Commission has established areas on its Internet site where investors can inform the Commission about violations of the law and ask questions from the staff. n73
The Commission has used the Internet to warn investors against certain promises that are "too good to be true" because "they are" and to caution investors against fraudulent practices. n74 The National Association of Securities Dealers ("NASD") has a statutory power and duty of regulating the broker-dealer community, subject to the supervision of the Commission. n75 Among its duties, the NASD supervises the advertising of its members. n76 Since many broker-dealers use the Internet to advertise, the NASD has been using software that scans the Internet automatically and picks certain words, such as "assure," "secure," "guarantee," "20 percent and more," and similar words that denote a promise of high return and low risk. n77 These advertisements are then evaluated by examiners. n78 In short, fire can sometimes be fought with fire.
3. The Choice of Adaptive Mechanisms and Method of Adaptation
In the case of prospectus delivery, the Commission is the legitimate traditional adaptive mechanism and there is no reason to substitute another mechanism for it. The Commission's relationship with the industry is ongoing, as issuers file their registration statements with the Commission. The response of the agency to developments was quite prompt. In early 1995, the Commission guided the industry by publishing an Interpretive Release with respect to the delivery of prospectuses through the Internet. n79 This adaptation of the law to the changed circumstances of the industry and the market actors seems to have worked smoothly and quite well. n80
III. Regulation of Securities Exchanges
A second change in the environment of securities markets and its actors is the introduction of trading sites. In this case, sponsors or issuers offer the holders of securities a meeting place where buyers and sellers of the securities may trade. The Securities Exchange Act of 1934 regulates the securities markets, including exchanges, where investors trade their shares. n81 Since markets in the United States are mainly conducted by market intermediaries - broker-dealers, market makers, underwriters, and other institutions that have joined the markets, such as subsidiaries of bank holding companies and insurance companies - the primary regulation of these marketplaces or market systems is performed by a self-regulatory organization of the intermediaries, subject to the oversight of the Commission. n82
Based on the framework developed in Part I, this Part inquires into the lawmakers' reaction to this new format of trading and into the need for adaptation of the law's substance, enforcement, and change mechanisms and methods.
The policies underlying the regulation of the exchanges can be summarized as: (1) maintaining investor confidence in the markets (treating investors, especially small investors, fairly, and avoiding overreaching and conflict of interest treatment of investors by market intermediaries); n83 (2) maintaining efficient markets from which both issuers and investors benefit (orderly markets, assuring instantaneous or timely public price information, effective enforcement of trades); n84 and (3) reducing the costs of trading. n85
A. Need to Adapt Securities Regulation to Internet Communications: Impact of the Internet on the Environment of Actors that Maintain Exchanges and Trade on Exchanges
Because the Internet can be used to exchange information, it can be used as a forum for securities trading, similar to securities exchanges. Entrepreneurs, issuers, or market intermediaries (e.g., brokers and dealers) can set up websites, inviting shareholders for free or for a fee (or other form of compensation) to enter the sites and trade with other shareholders. For holders of shares that are not traded on exchanges and have illiquid or no markets, such websites are likely to increase liquidity. Shareholders value liquidity and are willing to forego returns on liquid shares. Therefore, for the issuers generally and issuers of illiquid securities especially, such trading sites can reduce the cost of capital.
Trading sites established by actors other than groups of intermediaries are a new phenomenon. n86 The current securities laws regulating exchanges do not fit these trading sites. The Commission was asked to determine whether trading websites fall within the definition of "exchange" in the Securities Exchange Act of 1934, whether they should be regulated, and if so, how. In response to a request for clarification of the status of trading sites, the staff issued several no-action letters that allow the opening of such sites under certain conditions. n87 This discussion is limited to sites established by issuers of securities, providing a forum for trading by the holders of their securities.
To evaluate the impact of the need for changing the securities laws, we first examine the impact of the Internet on the actors that maintain and trade on exchanges and trading sites by exploring the differences and similarities between trading sites and traditional exchanges. Internet trading sites and exchanges are similar in that they offer trading forums to investors and the benefit of liquidity. n88 However, this similarity is not as close as it seems at first blush because Internet trading sites offer a trading forum to investors who "do it themselves," n89 while the exchanges are forums for intermediaries who trade on behalf of investors or to some extent for their own accounts. n90 Thus, both the promoters and users of the sites are different from those involved in exchanges.
In essence, trading sites eliminate the broker-dealers and other securities intermediaries, resulting in benefits and costs to investors. n91 On the benefits side, sites eliminate the cost of intermediaries and the risks from intermediaries' overreaching. However, Internet trading sites pose for investors a number of dangers - dangers against which they are protected by the intermediaries in traditional exchanges and other securities markets. When the intermediaries leave, these dangers appear.
Intermediaries serve two crucial functions in the markets. First, they ensure that the parties will not renege on the trades. n92 More often than not, securities trades cannot be executed simultaneously. n93 Because securities markets are volatile, either sellers or buyers would usually have incentive to renege on the trades before execution. n94 If that happened often and investors learn about this risk, they would either demand high returns or cease trading; no markets would develop.
Intermediaries ensure the execution of trades by acting as escrow agents, holding both sellers' securities and buyers' money. n95 Since these intermediaries obtain their commissions upon execution of the trades, they have strong incentives to ensure the execution and bear the costs. If traditional intermediaries disappear and no mechanism fills the void, it is likely that trading-site markets will not continue to function, as investors find their trade agreements ineffective and too costly to enforce. Further, if investors do not understand the working of the markets, this experience may also reduce their trust in the traditional markets.
Second, intermediaries publish the prices of the trades. n96 Price information substantially reduces the trading costs for other investors because it offers shorthand information about the value of the traded securities. The duty to publish the prices is not imposed on the parties to the trades but on those who service traders. n97 If these service providers disappear, someone else must provide the services as a condition to maintenance of efficient markets. Further, even though price information benefits investors generally, not all traders are interested in publishing the price of their bid or ask price. n98 Individual investors may wish to shield their offer or bid prices because this information may signal their trading position or give the wrong signal about their evaluation of the securities. n99
Third, trading sites need not be connected to other trading exchanges. If they develop in isolation, and if different promoters offer sites for trades in the same securities, inefficient segmented markets may develop, with different prices for the same securities.
Fourth, issuers' control of trading sites poses unique threats to the integrity of the sites. Issuers may be tempted to affect the price of their securities. n100 Moreover, trading sites arguably pose a competitive threat to established exchanges if the sites provide the same services to investors at lower costs. Such competition may be unfair if trading sites remain unregulated while the exchanges bear regulatory costs. Although competition can enhance the efficiency of the exchanges, an unequal playing field can threaten their existence in the future. Since the United States' exchanges, especially the New York Stock Exchange, are among the most, if not the most, efficient exchanges in the world, the risk of their lost hegemony by action or inaction of the Commission can have serious consequences. Thus, before trading sites are allowed to proceed, the continued viability of the existing securities markets should be ensured, at least until it is clear that trading sites can provide a viable alternative to existing markets. At the same time, because I believe that competition per se is desirable as a matter of policy, attempts should be made to allow trading sites to develop under certain conditions. In fact, the Commission has adopted a new regulatory framework to allow alternative electronic trading systems, as it attempts to adapt the law to the new environment. The new framework offers persons who wish to conduct these electronic trading systems a choice to register as national securities exchanges or as broker-dealers. n101
B. Securities Lawmakers' Response to the Internet Environment: Substance, Enforcement, Mechanisms for Change, and Methods of Change
1. Impact on the Substance of the Securities Laws
Internet trading sites change not only the mode of trading among investors but also the nature and institutional structure of traditional securities markets. Trading sites are fundamentally different from securities exchanges. They are operated by different actors with different incentives and different rewards. Thus, Internet trading sites provide different institutional infrastructures for securities trading. This raises the question of how the new type of market affects the policies underlying securities markets regulation and the rules codifying those policies.
A textual analysis of the definition of "exchange" in the Securities Exchange Act of 1934 n102 is not helpful. There are no guides in the existing legislation to interpret the Internet trading "exchanges" as the new trading sites. They were not anticipated by Congress; they involve not only a new and different technology but also a very different structure and participating actors.
The underlying policies and values of the Act may be affected by the trading sites. The results of their regulation are hard to predict and may lead to new policies and values or to reinforced current policies and values underlying the law. In sum, trading sites can put in question the current fundamental policies of the law and require a different regulatory regime altogether.
2. Impact on Enforcement Mechanisms
Presumably, trading sites do not pose enforcement problems for the Commission. However, they pose serious enforcement problems for investors. As discussed above, investors may need new and other guarantors of the trading contracts among them and some new mechanisms to ensure the performance of the trades regardless of price fluctuations. It is difficult, however, to envision which mechanism would effectively enforce the traders' contracts. Presumably we can create mechanisms for guaranteeing the execution of the trades among individual traders. Such guarantees may be offered by the promoters of the sites or by third parties, and many other possibilities come to mind. We can create a link to a national depository that would confirm trades upon proof of payment. The costs of the new mechanisms must be evaluated as experience about the sites is gathered.
3. The Choice of Adaptive Mechanisms and Method of Adaptation
The first question that trading sites raise is: Which of the mechanisms for change is best suited to determine the adaptation of the securities laws to the new Internet trading sites: Congress, the courts, the Commission, or the markets?
Congress should not legislate new regulation for the trading sites. There is little experience on how these sites function and the consequences of their operations. Congress cannot experiment in allowing piecemeal limited use of the sites nor monitor the problems they raise and the problems they solve. n103
The courts are also not the appropriate mechanism for adapting the securities laws to the trading sites. Courts cannot act unless they are requested to do so in the case of conflict among parties. Besides, courts are not equipped to conduct the study necessary to design a regulatory system nor to enforce such a system. Presumably, if courts were asked to determine whether trading sites are exchanges, they would use the policy analysis adopted in the VALIC case n104 and impose the current securities laws on such sites, awaiting the Commission's adaptation of the law to these new "exchanges."
This situation is precisely one that requires experimentation by an administrative agency. n105 First, the values and policies underlying the securities acts support implementation of new technology and trading sites. Second, the Commission has authorized trading sites under certain conditions. n106 These conditions might be strict at the outset and then relaxed as more experience is gathered respecting any problems that these sites pose. n107 Moreover, the Commission has expressed that it is "mindful of the benefits of increasing use of new technologies for investors and the markets, and has encouraged experimentation and innovation by adopting flexible interpretations of the federal securities laws." n108
If the Commission approached the issue by imposing on trading sites the regulations applicable to exchanges as they exist today, it is likely to freeze the development of trading sites. These sites cannot operate under the current exchange regulation and to change the regulation would require congressional action, which, as we noted above, is not the appropriate adaptive mechanism in this case.
This brings us to the markets. Could the markets be left to adapt and shape market customs to regulate trading sites? Leaving the markets to develop best practices for trading sites is a very attractive suggestion. The hand of the multitudes of investors and the promoters' behavioral adaptive mode of trial and error might offer regulators a proven, optimal model of regulation. If markets also provide effective sanctions to those who violate adapted customs, perhaps no government regulation would be necessary. Further, market "creeping" regulation may avoid unanticipated consequences, which government regulations tend to bring as side effects, regardless of how well-intentioned and how focused regulators try to be.
I reject markets as a sole mechanism for adapting the securities laws to trading sites for a number of reasons. First, nothing concerning trading sites suggests that the fundamental policies of ensuring investor confidence and facilitating capital should be changed. These remain the main guides to the institutional structures of the markets. Inexpensive enforcement of trades must be secured. Prices must be published. Segmentation should be avoided. Yet it is not clear that markets would heed these policies and ensure compliance with them. If markets would have heeded these policies in the past, we would not have government securities regulation today.
Second, the stakes are too high to allow promoters to break, on their own and for their own benefit, new grounds in shaping the new institutions for markets, especially when these new markets are still small. Third, and most importantly, failure to ensure the integrity of the new markets may taint existing markets. Investors do not always make refined distinctions when they are defrauded or find they made decisions without sufficient information. Paradoxically, "free" markets require a regulatory infrastructure. n109 This infrastructure can be developed by the actors, such as broker-dealers and other intermediaries. However, infrastructure is less likely when markets develop by suppliers of sites and actions of investors. Neither the suppliers nor the investors are likely to have sufficient identity of interest to combine in self-regulatory organizations and provide the infrastructure. n110
Leaving the markets alone to design the trading sites is not the answer. Leaving the Commission alone to design a regulatory scheme for the trading sites is also not the full answer. It seems that we should develop a close interactive adaptive system between the markets and the Commission. In fact, such an interactive approach is to some extent already being practiced by the Commission. The Commission publishes many company proposals for comments; Commissioners and staff meet regularly with industry representatives, lawyers, and consumer representative groups. The industry may make presentations to the staff to educate and provide information. Lawyers seek and receive the staff's interpretation of the law ex ante, while offering the staff information about events in the markets. This does not mean that the parties bare their hearts to each other. Neither the staff nor the industry are fully informative, but a substantial amount of information is exchanged between them in a search for a solution satisfactory to both parties.
It seems that the trading sites require a period of experimentation, with controlled limits, under close monitoring of the Commission. It may well be that this method is beginning to take shape in the form of no-action letters that the staff has issued for issuer trading sites. The method of change in this case is therefore complex, slow, experimental, and public. This method is, of course, remindful of the method by which markets adapt their customary law. It may well be that we have reached a stage where law's adaptation requires interaction between markets and agencies. Markets represent an amalgamation of perspectives of different actors driven by differing interests, understandings, and levels of information. In the case of securities markets and other markets, this amalgamation is translated into the price in dollars. The dollars become the common denominator.
In the case of market custom, however, the amalgamation of perspectives of different actors, driven by various interests and understandings and levels of information, does not always reflect a common denominator. Although custom represents a pattern of behavior that is followed by an increasing number of actors, this pattern is not transparent, nor clearly predictable. When regulatory agencies adopt a custom or "best practices" of an industry, the agencies can meld together the various aspects of the custom-producing actors; in some respect the agencies provide the missing common denominator that money would have provided in the form of price, to create better rules while subsuming the substance of the customs.
How do agencies collate and meld together items that do not emanate from similar sources or for similar reasons (sometimes from conflicting reasons)? Put differently, how do agencies generalize disparate items that are generated by disparate driving forces and for different reasons? The answer is: not by mathematical addition. Rather, a "soft process" of judgment based on information, context, and experience brings about a coherent rule based on the substance of the custom. This process is demonstrated by the restatements of the laws and codification of uniform laws - drafting, articulation of norms, and rationalization of market customs.
One form of interaction between the staff of the Commission and industry is worth noting; this form is not unique to the agency and is practiced, in different ways, by other agencies as well. The Commission's staff has been offering the public informal views on proposed activities that may raise issues under the federal securities laws. n111 These letters offer a number of advantages.
The letters inform the staff about proposed transactions in the market, possible legal barriers to novel transactions, and problems they might pose for investors. The letters help avoid or reduce enforcement by litigation and leave some room for restructuring and legitimizing proposed activities. No-action letters benefit the public and strengthen the rule of law by publicizing the staff's interpretation and application of the securities laws. n112 The business community has come to depend on the consistent application of the letters' rulings in making business decisions. The process facilitates business transactions, especially novel transactions, that may not exactly fit within the regulatory framework when the interpretation of the law applicable to them is uncertain.
The main source of the staff's authority is its discretion to recommend that the Commission prosecute violations of the statutes under its jurisdiction. n113 The weight given to the staff's letters depends mostly on the type of reasoning on which the decision is based. When the letter represents the opinion of the staff on a particular legal issue, n114 the letter is likely to be viewed as a precedent on the legal issues in the particular fact pattern, at least until it is reversed by the staff, the Commission, or the courts. n115 Courts, however, often defer to the staff's legal interpretation and expertise. n116 When the staff grants a no-action position even "without necessarily agreeing with" the requestor's legal position, n117 such a letter has less weight than an interpretative letter. n118 It might still be used by third parties if the facts at hand closely resemble those cited in the letter. n119
Formally and officially, no-action letters have no value as precedents. n120 Neither the Commission nor the staff are bound by these no-action letters. n121 Yet, practicing attorneys and academics view no-action letters as a source of law, and they are considered precedents by parties other than the recipients, providing a partial safe harbor and guidance to practitioners. n122 The letters augment the limited number of court cases and Commission interpretations and are sometimes the only authoritative interpretations of the Act. Perhaps the paucity of judicial decisions may be attributed in part to no-action letters that meet the needs of the parties and the industry. Moreover, the no-action process is generally less costly than a formal exemptive application or Commission administrative action, and far less costly than litigation.
So long as the staff and the Commission value no-action letters, they will accord them precedential weight. If people could not rely on these letters people would cease to ask for them, and the benefits from such letters to the staff, the Commission, the industry, and the parties would be lost. That may be the reason why the staff is concerned with uniformity and predictability of no-action letters. The frequency with which the staff reverses prior letters differs depending on changes in the business environment and the industry, the problems that the letters addressed, and the unintended consequences that such letters brought about. n123 Thus, the letters provide substantial comfort to their recipients even in private litigation. n124
In some respects no-action letters are similar to judicial decisions because they are based on specific fact situations, rely on precedents, and constitute precedents, especially when they provide analysis and reasoning. The letters differ from a judicial decision because they are given with little express legal authority and are granted ex ante, and not ex post, as a result of conflict. The letters are similar to customary law. Their age and extent of following adds to their precedential weight. Their flexibility allows for adjusting law to changing circumstances. For law's adaptation to the Internet environment, they may be a most appropriate tool.
This article offers a first step towards an examination of adaptation of law to a changing environment of the actors that are regulated by law. I suggest a model of law as a structured, adaptive, self-replicating system of coercive communications regulating relationships among types of actors (individuals or groups). The three-part structure of the law consists of substantive communications that differ in their degree of generality (specific cases, rules, policies, and values), mechanisms for enforcing the law, and mechanisms for adapting law when the actors that are regulated experience a new environment (legislatures, administrative agencies, courts, the markets), using methods of adaptation (textual analysis, analogy, and policy analysis).
When the environment of regulated actors changes rapidly, as it changes today and is likely to change in the future, a new combined mechanism of markets and agencies is being developed to adapt the law accordingly. This new mechanism and its methods of adaptation invite a close examination and empirical study, for I believe this is the future mechanism for legal change.