The Precision Benefit

This chapter offers the basic justification for personalized law: the precision benefit. Personalized rules could accomplish the underlying goals of any law more effectively. Any goals, of any law. The reason is almost trivial: personalized law accounts for more relevant circumstances and differences in tailoring individual commands. This is the same reason that custom-made shoes fit better than single size, or that medicine based on personalized diagnostics cures better than one-size-fits-all treatments. Uniform rules and commands, even if optimal on average, are a poor fit for people with diverse preferences, characteristics, histories, and means. The chapter begins by demonstrating the value of personalization in other sectors of human activity, like medicine and education, and then uses an example of data protection law to display the benefits of personalized rules. It also discusses the production costs of personalized law, primarily the costs associated with the use of more information in the promulgation of legal commands.

Personalizing Rules by Age

To complete the demonstration of personalized law in action, this chapter focuses on the inputs used to tailor the individualized commands. One input that is likely to feature in the personalization of many rules is age. Age is informative because it is often correlated with personal attributes that matter to achieving the goals of a law. Preferences, cognition, judgment, experience, and physical ability—all vary with age. Young age is a factor in the denial of legal capacity and in the conferral of various paternalistic protections, whereas old age represents changing needs, capacities, and entitlements. Under personalized law, age would be an input affecting legal commands that are currently age-invariant, such as the intestate succession default rule or speed limit laws. In addition, age of capacity laws, which are currently used to regulate entry into various activities, would use different age cutoffs for different people.

Personalized Legal Areas

This chapter illustrates personalized law “in action” by examining it in three areas of the law: standards of care under the common law tort doctrine of negligence, mandated consumer protections in contract law, and criminal sanctions. In each area, the chapter examines personalization of commands along several dimensions. In tort law, standards of care could vary according to each injurer’s riskiness and skill, to reduce the costs of accidents. In contract law, mandatory protections could vary according to the value they provide each consumer and differential cost they impose on firms, to allocate protections where, and only where, they are justified. And in criminal law, sanctions would be set based on what it takes to deter criminals, accounting for how perpetrators differ in their motives and likelihood of being apprehended, with the potential to reduce unnecessary harsh penalties.

Governing Through Data

Personalized law requires massive information, and this chapter examines some of the problems relating to the accumulation of personal data in the hands of the government. It first surveys what kinds of information would be needed and how lawmakers might hope to acquire that necessary data. While much information is already available in government databases, is it realistic to expect commercial databases to share the data with the government? The chapter then shifts to asking how the personalized commands would be communicated to actors. It argues, counterintuitively, that in important areas, private actors may often find it easier to know their personalized command than figure out the uniform command. Finally, the chapter examines problems of privacy and data protection, arising from the accumulation of data in the hands of governments. It argues that privacy interests vary across people, and thus privacy protection—like other aspects of personalized law—could itself be personalized, allowing people to opt out of some privacy-sensitive personalized treatments.

What Is Personalized Law

This chapter introduces the paradigm of personalized law as a distinctive jurisprudential method. It is a particular version of contextual law characterized by individualization: legal commands depend not only on external circumstances, but also on interpersonal differences between people. To identify with precision the relevant differences, and to use these features in a properly weighed manner, personalized law relies on machine-sorted information. For example, algorithms would be trained to identify personal attributes correlated with riskiness, so as to tailor personalized standards of care. The chapter identifies embryonic versions of personalized rules in existing and in old legal systems, to demonstrate that even when legal rules are formally uniform, personalized commands sometimes emerge in their shadow. It also shows the prevalence of private personalized regulation, whereby non-governmental entities develop personalized norms to regulate commercial, religious, and household domains.

Personalized Regulatory Techniques

This chapter expands the demonstration how personalized law would transform existing legal institutions. The chapter shifts the focus from specific doctrines to regulatory techniques. These are generic approaches to the design of legal interventions, used in every area of law. The chapter examines the personalization of several techniques: default rules, mandated disclosures, compensatory damages, and bundles of rights. With each of these tools, the law presently prescribes one-size-fits-all rules, designed to either best fit the average person, or to promote the interests of a specific subgroup of the population. By shifting to personalized rules, the law could simultaneously advance the interests of different groups and individuals. The chapter shows that designing personalized default rules, disclosures, damages, or bundles of rights would promote the goals underlying these interventions. Personalized default would mimic peoples’ preferences more successfully and reduce the incidence of opt-out. Personalized disclosures stand a chance of being more useful to people. Personalized compensation would come closer to making victims of wrongs whole. And personalized bundles of rights would recognize the diversity of people’s interests and aspirations.

Personalization and Distributive Justice

This chapter examines how equality in the eyes of the law would survive if legal commands are personalized and result in different rules for different people. It argues that nothing in the framework of personalized law violates equality before the law. On the contrary, personalized treatment provides tools to distribute rights and burdens in a manner that conforms to egalitarian views and to notions of desert and need. If desert and need are determined by relevant attributes in a proportional manner, a just system should treat people differently. The chapter examines how personalized law, when designed to promote goals other than equality, could be bolstered (or constrained) by various notions of distributive justice. It recognizes that the use of Big Data and artificial intelligence could itself be a source of injustice, perpetuating historical biases. The chapter discusses ways to resolve this concern. Finally, it compares the deliberate differentiation of commands under personalized law with unintended forms of differential treatment pervasive under uniform laws. It concludes that the use of a multitude of relevant factors to personalize commands, derived from transparent statistical methods, offers novel opportunities to promote distributive justice goals under the law.

Introduction

This chapter introduces the main themes of the book. Personalized law is a new paradigm of legal order, where uniform rules are replaced by commands that vary across people. Like other personalized schemes, personalized law would rely on algorithms that sort through Big Data to identify personal features relevant to the optimal design of personalized legal treatments. The chapter illustrates this novel regime by imagining a day in the life of a typical household. It then outlines the benefits, but also the challenges and problems of personalized law, and maps out how the book addresses them. Part I of the book examines the benefits of personalized law. Part II demonstrates personalized legal rules in a variety of areas. Part III addresses the problem of equality under the law. Part IV explores additional problems of coordination, manipulation, and the power of data in the hands of governments. The concluding chapter discusses the role of human lawmakers in personalized law.

Personalized Law and Equal Protection

This chapter examines personalized law from the perspective of the Equal Protection Clause in the United States Constitution. Some classifications of people, when made for the purpose of differentiated treatment, are subject to stifling doctrinal constraints. Could such classifications be made under personalized law? The chapter argues that personalized law mitigates the constitutional concerns relating to suspect classifications. Treating people as individuals, using multi-attribute data-weighed tailoring, and not as identical members in a certain class, is permissible because members of the class are not singled out for class-specific uniform treatment. The chapter examines landmark Supreme Court cases on sex and race classifications, showing that the limits set by the Court and the narrow permission it granted for some uses of classifications, all fit well within a scheme of personalized law. In addition, the chapter examines problems of unintended disparate impact that could arise under personalized law, and demonstrates the unique advantage of the algorithmic methods fueling personalized law in reducing and eliminating such effects.

Conclusion: Legal Robotics

This chapter concludes the book by offering some preliminary reflections on the robotic aspects of personalized law. It begins by identifying some early experiments with the use of algorithms and machine learning in law, noting the immense potential they unveil. It confronts the “see” versus “scan” methodologies for individualized treatment—judges “looking people in the eye” versus algorithms analyzing the numerous personal aspects they are permitted to scan. The chapter highlights the critical roles of humans in algorithmic personalized law, primarily in setting the goals that the algorithms will be coded to optimize, in choosing the data by which algorithms are trained and people are subsequently screened, and in scrutinizing and repairing undesirable patterns. The chapter argues that the need to set specific goals and priorities for each law would transform the common law method of legal refinements, and would offer greater transparency for legislative accords. The book ends by pointing to areas of the law most ripe for phase-one personalized law.