Implementing Personalized Negligence Law

Jared I. Mayer1Jared I. Mayer is a law clerk at the Supreme Court of New Jersey. He thanks Omri Ben-Shahar, Ariel Porat, and the Coase-Sandor Institute for Law and Economics for inviting him to make this contribution. He thanks as well the participants in the Personalized Law Conference, the participants in the Junior Scholars Colloquium, and the Canonical Ideas seminar for their helpful feedback on earlier drafts, and gives special thanks to Ryan Sakoda for extremely helpful conversations on the ideas contained within this Essay. Lastly, he thanks the University of Chicago Law Review editorial team for their excellent editorial work. Nothing herein represents the views of the Supreme Court of New Jersey or the New Jersey Judiciary.

* * *

A part of the series, Personalized Law.

Negligence law seldom accounts for a person’s idiosyncrasies. The objective standard of care simply demands that people take reasonable care under the circumstances.2In this Essay, I use “standard,” “standard of care,” and “negligence law” interchangeably, even when prefixed with “objective” or “personalized.” One’s greater risk profile or ignorance of what the duty of care requires does not matter. A defendant breaches his duty of care when he fails to take reasonable care—nothing more and nothing less.

Not so under personalized standards of care. In their new book, Personalized Law: Different Rules for Different People, Professors Omri Ben-Shahar and Ariel Porat imagine a world where legal systems can create standards of care that are tailored to each person’s risk profile, allowing parties to take optimal precautions and reach optimal activity levels. Ben-Shahar and Porat provide numerous ways in which each person can learn about his or her own standard of care and in which the legal system can learn about each person’s standard of care. Their proposals are exciting precisely because they challenge the old negligence law’s assumption that determining each person’s unique standard of care is not feasible.

In this Essay, I want to highlight the best path forward for implementing a system of personalized negligence law. In Part I, I recount Ben-Shahar and Porat’s shift away from the objective standard of care toward personalized standards of care. In Part II, I argue that relying on individuals to know their respective standards of care based on the fact that they know themselves is a mistake. People are relatively unreliable when it comes to knowing particular facts about themselves and knowing how those facts correlate with personalized standards of care. Part III will then examine Ben-Shahar and Porat’s proposed avenues for implementing a system of personalized negligence law. Methods of informing people of those standards of care, I show, create a tradeoff between ex ante knowledge of one’s standard of care and the degree of applicability of that standard of care. In Part IV, I propose resolving this dilemma by equipping people with feedback mechanisms—ways of measuring and easily correcting their activities—to help overcome these limitations and implement personalized negligence law. With this framework of personalized negligence law in mind, Part V explores three normative implications of personalized negligence law.

I.  From the Objective Standard of Care to Personalized Standards of Care

Before identifying Ben-Shahar and Porat’s account of personalized negligence law, it is worth reviewing the objective standard of care and the orthodox account of why we have it. The objective standard of care requires each person to behave and adopt precautions as a person of “ordinary intelligence and prudence” would do.3Oliver Wendell Holmes, The Common Law 108 (1923). The orthodox, “law and economics” account of why legal systems have adopted the objective of care is that standards of care require taking into account a slew of details about each person. These include physical, cognitive, and emotional characteristics such as age, height, weight, spatial awareness, hand-eye coordination, and impulsivity. Since legal systems do not have reliable access to each person’s characteristics, it would not be cost efficient for them to create personalized standards of care.

This would not be the case under a personalized negligence law regime, according to Ben-Shahar and Porat. The legal system would be able to capture individuals’ distinct features and therefore be able to create a standard of care for each person. By having greater and reliable access to each person’s idiosyncratic features, the legal system can thus move away from using the crude objective standard of care and migrate towards personalized standards of care.

But how can each person know his or her own standard of care? Ben-Shahar and Porat suggest that each person can know his or her standard of care either through self-knowledge or through some kind of external mechanism. To see how each of these proposals may play out, I will examine each of them in turn.

II.  Self-Knowledge and Knowing One’s Standard of Care

Ben-Shahar and Porat’s first proposal is that each person can know his or her own standard of care based on knowing about his or her unique characteristics:

People know their own weaknesses. They know if they are clumsy, slow-reacting, narrow-visioned, distractable, or shaky. They also know their strengths—whether they are skilled and experienced. They are aware of their state of mind, when they are tipsy, tired, or bored. And while people are known to view themselves charitably and through biased cognitive lenses, the fear of unforgiving liability would reduce the impact of such self-serving bias.4Ben-Shahar and Porat remind us that “similar distorted beliefs about legal standards could also arise under a uniform regime.”

Thus, for personalized negligence law to work, both the legal system and each person must know his or her standard of care.5Strictly speaking, each person is likely going to have more than one standard of care because each person’s unique characteristics may require that they adopt higher or lower standards of care depending on the context. An elderly person may have a higher standard of care when it comes to driving, but a lower standard of care when it comes to riding a bike through a quiet park. Be that as it may, each person would still have to know his or her standards of care for different contexts. For simplicity, I will refer to each person’s standard of care in the singular. Since those standards of care would be constructed using data about each person’s physical and psychological characteristics, it is plausible to think that people would have access to at least some of that information. People may well know their age, sex, height, and current mental state (such as “I am currently tired” or “I am currently perceiving a computer in front of me”) and, armed with that information, discern their own standards of care.

There are two reasons to be skeptical of that prospect. First, people may not have reliable access to their own characteristics. Philosophers—including, more recently, those armed with empirical findings—have cast doubt on the notion that people have reliable access to their own mental states, including their beliefs. Biases and other cognitive distortions can further warp people’s judgments about their own characteristics. The Dunning-Kruger effect, for example, demonstrates that people who lack a particular skill will consistently overestimate their ability to exercise that skill. Other research has suggested that people underestimate the risk that they impose upon themselves and others. Simply put, this research suggests that people may have inaccurate beliefs about their unique characteristics.

Second, each person’s standard of care would not directly correlate with any singular characteristic or a particular set of characteristics. This is by design: basing standards of care on innumerable factors, including one’s characteristics, ensures that people take the most efficient precautions and engage in activities at the most efficient levels. And since many factors would be at play in creating one’s standard of care, it would be difficult to manipulate that standard of care by changing one factor. Yet if innumerable factors coalesce to create a person’s standard of care, there’s good reason to think that that person would not know if and how her characteristics correlate with her standard of care. And because each person’s standard of care would be personalized, looking to what people do on average would not help. Without the ability to look to oneself or anyone else, each person may be at a loss in determining just what his or her standard of care is.

III.  Methods of Conveying Personalized Standards of Care and Their Shortcomings

Ben-Shahar and Porat recognize “the enormity of the informational challenge for actors in a personalized law regime” and therefore propose three models for how to provide people with information. Each of these proposals differs significantly in how much information each person would have at the outset about his or her standard of care.

One proposal is for courts to use presumptions in litigation. A court, for example, would presume that a defendant accused of negligently harming the plaintiff should be held to the highest standard of care. Under those circumstances, the defendant could show the court that he or she should be held to a lower standard of care.  A second proposal is “crude” personalization, in which each person is assigned a high, medium, or low “level of care.” When driving on a three-lane highway, for instance, people must drive in only one of those lanes—fast, medium, or slow—depending on their personalized standard of care. The last and most ambitious proposal is to have communication technology convey to each person in real time not just what their personalized standard of care is but also what precisely he or she should do in order to meet that standard of care. Speeding examples once again come to mind; each vehicle’s dashboard can present the driver’s personalized speed limit.

These proposals differ in the extent to which people are informed of their standards of care ex ante. Under the first proposal, people may know their standards of care only if they are sued—and, even then, only if they reach discovery. Under the second proposal, people can know their standards of care only after they are sued or otherwise informed of their standard of care.6With speeding examples in mind, it is easy to think of nonlitigious ways in which people could be informed of their respective standards of care. Receiving a speeding ticket can convey to a driver that he or she was driving in the wrong lane. And under the third proposal, people would be informed in real time of what they must do to comply with their personalized standard of care.

Much like relying on one’s self-knowledge, the first two proposals would fail to adequately inform each person of his or her standard of care ex ante. Waiting until litigation—indeed, waiting until discovery—to inform someone of his or her standard of care fails to give them guidance as to which precautions they should take and to what extent they should engage in any particular activity. Assigning people crude levels of care under a “gradual” personalized negligence law system would not provide people with guidance as to what “high,” “medium,” and “low” levels of care would mean. This, too, would not inform people as to which precautions they should take and the extent to which they should engage in any given activity.

Communication technology provides a more promising avenue for informing people of their standards of care. By providing people with more precise guidance in real time, communication technology could afford individuals the opportunity to know ex ante what their standard of care is. The benefits of doing so are straightforward: individuals would be able to know which precautions are worth taking and to what extent they should engage in a particular activity in the first instance.

Yet while communication technology provides an elegant solution to the problem of informing people of their standard of care ex ante, one may question how widely people can use communication technology in the various domains of life in which negligence standards usually apply. To see this, contrast communication technology with the ex post presumption model that Ben-Shahar and Porat proposed. That model is the most flexible; a court could establish benchmarks for what it would mean to comply with a high standard of care for a particular kind of activity. The defendant could then introduce evidence showing that he or she should be held to a lower standard of care. The court could do this for the wide range of activities to which negligence law ordinarily applies.

Not so with communication technology, which assumes that there will be discrete courses of action that one can take (or not take) in order to meet one’s standard of care. Obviously, a three-lane highway gives drivers only three lanes from which to choose; driving between lanes is not an option. Similarly, individuals can be informed of their particular speed limit and be expected not to exceed it. But negligence law has always applied to “lumpier” situations, where there is no precise line delineating negligent from non-negligent conduct.

Let me spell out the applicability problem more clearly. Gradual personalization and communication technology systems assume that there will be discrete courses of action that individuals can take in order to fulfill their standard of care. Speeding examples illuminate these proposals precisely because speeding is largely, although not entirely, structured by limited courses of action. Drivers are told the speed limit for each road; each car has a speedometer, which keeps track of the driver’s speed; and the gas and brake pedals provide clear ways for drivers to comply with the speed limit.

These strictures are not present in other domains of life where people do not have discrete courses of action from which to choose. For example, people can be injured in slip-and-fall accidents under ordinary circumstances, yet it would be hard to translate one’s gradual personalized standard of care into a particular course of action and to distinguish between exercising “high,” “medium,” and “low” standards of care. It is also not clear how one could be given a discrete course of action with respect to walking on the sidewalk that does not collapse into general platitudes about safety (e.g., “Walk carefully”).

If people do not have clear, actionable directions for how to comply with their standards of care, people will have incentives to reduce their activity levels. These outcomes stand in tension with personalized negligence law’s ambitious goals: promoting (among other things) optimal deterrence by tailoring precautions and activity levels to each person. If personalized negligence law cannot meet these goals, then one can question whether personalized negligence law will be able to achieve more efficient (and just) outcomes relative to ordinary negligence law.

In order to successfully implement personalized negligence law, then, we need to (a) promote ex ante knowledge of one’s standard of care while (b) not relying on self-knowledge and (c) not sacrificing personalized negligence law’s wide applicability. I want to point out one aspect of Ben-Shahar and Porat’s analysis—an aspect that we currently utilize—that can guide the way toward achieving these goals.

IV.  Feedback Mechanisms and Personalized Negligence Law

Increasing the presence of feedback mechanisms can help make implementing personalized negligence law a reality. For my purposes, feedback mechanisms are technologies that provide people with (a) clear information to guide their actions and (b) limited, simple courses of action that they can use to adjust their conduct. These are important because, along with communication technology, they could provide people with both a personalized standard of care as well as a way in which they could adjust their conduct to meet that standard of care.

To illustrate the importance of feedback mechanisms, consider Ben-Shahar and Porat’s example of personalized speed limits, where a vehicle would automatically post a driver’s personalized speed limit, taking a bevy of factors into account. What makes this example particularly intuitive is how familiar it is. The elements at play in these futuristic vehicles have long been at play in contemporary vehicles: gas pedals, brake pedals, speed limits, and speedometers. These deceptively simple features together constitute a feedback mechanism. The speedometer provides the driver with a clear statement of how fast she is driving the car. The speed limit directs the driver how fast she can go. And if the driver is going too fast or too slow, there is no mystery as to how to correct course; the driver simply needs to use the gas and brake pedals to adjust her speed. This combination of factors makes having numeric speed limits—as opposed to a “drive reasonably” standard—worthwhile in the first instance.7See also Louis Kaplow, Rules Versus Standards: An Economic Analysis, 42 Duke L. J. 557, 585 (1992) (“When the government promulgates a rule, it gathers information before individuals act and announces its findings. As a result, the information is available to individuals when they act; individuals then may be guided by it and spared the expense of producing such information themselves.”); Isaac Ehrlich & Richard A. Posner, An Economic Analysis of Legal Rulemaking, 3 J. Legal Stud. 257, 262 (1974) (“[A]n increase in specificity, by reducing the variance in outcomes associated with engaging in a particular activity, would tend to have a disproportionately deterrent effect on undesirable activity and a disproportionately encouraging effect on desirable activity.”).

Feedback mechanisms, then, solve the problems we touched upon earlier. They would provide people with clear guidance not only on what to do but also how to do it. And notice that these problems would be defused not because individuals have any greater insight into their personal characteristics or any greater motor control. Rather, feedback mechanisms would empower people to know and act upon their personalized standards of care even without that self-mastery.

Vehicular speeding, however, seems relatively unique in the world of negligence law. Since negligence law applies to such a wide array of activities, it is no surprise that feedback mechanisms have not cropped up in most—much less all—of them. Nevertheless, there is reason to think that we will see feedback mechanisms become more prevalent throughout many domains of life, making personalized negligence law all the more plausible.

Take, for example, gun ownership. Gun owners are governed by the objective standard of care: they must exhibit due care in safeguarding and using their weapons, including in storing and regulating who may access them. Third parties—especially children—can harm themselves and others if they gain access to those guns. Gun owners can consequently face tort liability for negligently entrusting those third parties with those guns.

Recent developments in firearm technology have presented a way to mitigate these risks. So-called “smart guns” utilize identification mechanisms, such as fingerprint activation and radio-frequency identification (RFID), to allow only registered users to fire them. Smart guns incorporate a feedback mechanism into the firearm by (a) restricting who can use the firearm, (b) informing people who can and cannot use the firearm, and (c) providing a clear way to add or subtract the people who can use the firearm (i.e., add or remove them as registered users). This technology narrows the possibility of negligently entrusting a firearm to a third party, since merely giving it to them does not mean that they will be able to fire it. It is easy to imagine that, depending upon a person’s personal characteristics—riskiness, gullibility, and so on—people may be held liable to differing degrees based on who they register as the firearm’s user. Those who are riskier and more gullible, for example, may be held liable for gun-related negligence if they had registered any other person as a user with their smart gun. Less risky and more skeptical persons may be able to register more people as their smart guns’ users before incurring negligence liability.

This example also illustrates how personalized standards of care and feedback mechanisms could work symbiotically. As feedback mechanisms proliferate and simplify more domains of ordinary life, each person’s personalized standard of care may also change. A person who is somewhat forgetful or careless could conceivably be held to a relatively high personalized standard of care if he had no feedback mechanisms to help him. Yet with respect to the negligent entrustment of a smart gun, for example, the legal system could reduce his personalized standard of care, since the weapon itself precludes unregistered persons from using it. By the same token, personalized standards of care—especially those that are quite unforgiving—may induce actors to create feedback mechanisms that would enable them to meet those standards of care. Assigning someone a demanding standard of care for fender benders, for example, incentivizes them to outfit their vehicle with cameras, sensors, and other technologies that notify them when they come too close to a nearby vehicle. Feedback mechanisms and personalized standards of care are not merely complimentary; each one influences the development of the other.

In short, by incorporating a feedback mechanism into their design, smart guns allow for another domain of negligence law—negligent entrustment of a dangerous instrument—to fall within the scope of personalized negligence law. As other feedback mechanisms appear across various domains of ordinary life, we can increasingly see the potential for individuals to be held to, and to adjust their conduct with the knowledge of, their personalized standards of care. What is more, one’s ability to use a feedback mechanism may change his or her personalized standard of care. By the same token, enforcing personalized standards of care may incentivize persons to invest in feedback mechanisms to comply with those standards of care. Communicating those standards of care to each person across the many domains of life—and marrying feedback mechanisms to those domains of life—can help make personalized negligence law a reality.

V. Implications of Personalized Negligence Law

Before concluding, I want to highlight three normative implications of shifting towards a personalized negligence law system.

First, with the rise of personalized standards of care, we can expect to see a parallel rise in the doctrine of negligence per se. Courts use negligence per se to incorporate a statute or regulation’s command into a uniform standard of care. The Second Restatement notes that courts do not mechanically incorporate these statutory and regulatory requirements into the standard of care. They rather do so under narrow conditions, such as when the statute or regulation’s purpose is to protect the class of persons of which the plaintiff is a member and the kind of interests that the defendant supposedly infringed.8See Restatement (Second) of Torts § 286 (Am. L. Inst. 1965). The concern, it seems, is that incorporating these statutory and regulatory requirements carte blanche will constrain a court’s ability to determine the reasonableness of a defendant’s actions (or lack of action) under the circumstances at hand. Put differently, negligence per se entices courts to implement legislative and regulatory policies, even if it would be inappropriate to do so given the factors at play in the case.

This risk would seemingly not arise when marrying negligence per se to personalized standards of care. Personalized standards of care already would incorporate all the contextual information needed for each person in each circumstance. If the legal system can capture that information accurately, then negligence per se in a personalized negligence law regime would incorporate context-sensitive judgments contained within each person’s standard of care while also promoting stability and consistency in adjudicating negligence law cases. In fact, as Ben-Shahar and Porat discuss, there may be a greater risk of judges undermining that algorithmic determination if they use their own judgments about reasonableness.

Second, given the anticipated rise of negligence per se, we may expect to see more doctrinal shifts in other areas of negligence law, such as causation and damages, that are critical for establishing a defendant’s ultimate liability. As standards of care become increasingly tailored and clearly communicated to each person, there may be less room for litigating what a defendant’s standard of care is and whether the defendant adhered to that standard of care. Doctrines relating to causation, apportioning liability between the parties, and determining the extent of a plaintiff’s damages would take center stage, both as a matter of litigation strategy and doctrinal development.9See Restatement (Third) of Torts: Apportionment of Liability § 8 (Am. L. Inst. 2000).

This may be for good or for ill. Developing causation and damages doctrines may enable judges to avoid inequitable or undesirable outcomes that are partly the product of personalized standards of care. The corresponding risk, of course, is that judges may use causation and damages doctrines to effectuate their preferred policies and fashion their preferred outcomes at the expense of enforcing the normative values that the personalized standards of care are supposed to embody.

Moreover, negligence law sometimes features highly unusual cases.10See, e.g., Palsgraf v. Long Island R.R. Co., 248 N.Y. 339 (1928). The communicated standards of care and feedback mechanisms may be ill-suited to pick up on those unusual cases, even if they could jointly enable people to take cost-efficient precautions and engage in activities at optimal levels under usual circumstances. Thus, there would remain the question of how to allocate losses when all the parties had adhered to their personalized standards of care. Since each person’s standard of care would be a product of his or her idiosyncratic features—features of which he or she may not even know—allocating losses to the least cost avoider may be difficult to do.11See Guido Calabresi, Ideals, Beliefs, Attitudes, and the Law: Private Law Perspectives on a Public Law Problem 74–77 (1985) (discussing the “least cost avoider” approach to tort liability). Instead, legal systems may adopt doctrines that aim to spread liability out as thinly as possible when all the involved parties had adhered to their personalized standards of care.12See Guido Calabresi, The Costs of Accidents 39–45 (1970) (discussing loss spreading as a subgoal of tort law).

Lastly, legal systems would have to consider the distributional effects that the rise of personalized negligence law and feedback mechanisms may bring about. Recall that for personalized negligence law to work, individuals must have access to both communication technology and feedback mechanisms. Some groups may have widespread access to communication technology, such as smartphones, as well as certain feedback mechanisms, such as vehicles equipped with speedometers. But there is reason to think that access to new communication technologies and feedback mechanisms would take time to disseminate to the least well-off members of society. This is worrying because, absent communication technology and feedback mechanisms by their sides, people from under-resourced and marginalized communities would be unable to know which precautions to take and the extent to which they should engage in various activities. And if communication technologies and feedback mechanisms are the keys to navigating a personalized negligence law without incurring liability, not having those resources would leave members of those under-resourced and marginalized communities unwittingly vulnerable to tort liability. To effectively implement a personalized negligence law system, then, lawmakers need to think broadly about introducing a robust system of redistribution to support a personalized negligence law regime—and a more expansive personalized law regime.


Communication technologies and feedback mechanisms are simple and powerful devices because of what they allow individuals not to think about. Given the limitations on self-knowledge and the need to impose clear, workable, and personalized standards of care, increasing the prevalence of feedback mechanisms and communication technologies is essential to creating a system of personalized negligence law. While personalized negligence law carries with it multiple doctrinal and practical implementation challenges, Ben-Shahar and Porat’s book serves as an engine for future thought about making—and whether to make—personalized negligence law a reality.

* * *

Jared I. Mayer is a law clerk for the Supreme Court of New Jersey and a 2021 graduate of the University of Chicago Law School. He thanks Omri Ben-Shahar, Ariel Porat, and the Coase-Sandor Institute for Law and Economics for inviting him to make this contribution. He thanks as well the participants in the Personalized Law Conference, the Junior Scholars Colloquium, and the Canonical Ideas seminar for their helpful feedback on earlier drafts, and gives special thanks to Ryan Sakoda for extremely helpful conversations on the ideas contained within this Essay. Lastly, he thanks the University of Chicago Law Review editorial team for their excellent editorial work. Nothing herein represents the views of the Supreme Court of New Jersey or the New Jersey Judiciary.

* * *

Click here to return to the Personalized Law series main page.

Leave a Reply