Carpenter Should Replace Katz in Fourth Amendment Law

For more than 50 years, legal scholars, judges, attorneys, and law students have centered their analyses of the Fourth Amendment on the famous Katz test. It determines in most cases what a Fourth Amendment “search” is and, thereby, whether the amendment applies at all. The Katz test essentially asks whether an act of government surveillance has violated a person’s “reasonable expectation of privacy.” In practice, this standard is notoriously ambiguous and unhelpful. The Katz test has been widely criticized for decades, and its lack of coherence has made Fourth Amendment search law particularly confusing.

In 2018, the Supreme Court decided Carpenter v. United States, which held that government agents had to obtain a warrant before collecting cell phone location data showing virtually everywhere a suspect had traveled over a seven-day period. In the course of its decision, the Court also determined that individuals can retain Fourth Amendment rights in personal information even if they disclose it to third parties. Commentators hailed Carpenter as an enormously important, paradigm-shifting Fourth Amendment decision. But exactly what Carpenter meant going forward was unclear, and scholars were divided on its scope and power. 

One of us argued soon after Carpenter was decided that it had established a three-factor test that should come to replace the Katz test. The other engaged in a large-scale empirical study in 2021 examining how lower courts had applied Carpenter—and identified a similar collection of factors that courts had begun to adopt. A forthcoming essay examines the interplay of the Katz and Carpenter tests and concludes that they compete and overlap in the lower courts in a haphazard, confusing manner. 

In all of these works, we have advocated for a broader adoption of the “Carpenter test.” Here, we set the argument out directly: The Carpenter test should replace the Katz test as the primary test for Fourth Amendment searches. It is more coherent, predictable, and specific than Katz’s vague reasonableness standard. It is better at addressing new and changing technologies. Courts are capable of effectively assessing its factors, and many courts have already done so. And it is based on sound theoretical claims and grounded in decades of Supreme Court precedent. Courts should adopt it and, at last, abandon the Katz paradigm that has caused so much confusion over the past several decades.

The Carpenter Test

The Carpenter opinion did not expressly set out an official standard to guide future decisions. But it did identify several factors that compelled its decision that cell phone location data collected and held by a service provider is protected by the Fourth Amendment. First, the Court discussed the deeply revealing nature of location data, which could “provide[] an intimate window into a person’s life, revealing not only his particular movements, but through them his familial, political, professional, religious, and sexual associations.” Second, it addressed the large amount of location data collected by the government, including 101 data points per day that marked the defendant’s movements. These massive quantities of data substantially increased the potential for intrusion on an individual’s privacy. Third, the Court mentioned that cell phone data is not voluntarily transmitted by a cell phone user but is automatically transmitted by the cell phone. Likewise, using a cell phone is a largely inescapable part of modern life, so users have little choice but to use one. Accordingly, any disclosure to a third party was essentially involuntary. The Court mentioned other factors and considerations as well, such as the large number of people potentially affected by cell phone surveillance and the low cost of tracking people via their cell phone records.

The Court identified these factors as motivating its decision, which in turn indicated that they might be used as a test in future cases. Indeed, many lower courts have done just that, applying the factors to help resolve numerous novel Fourth Amendment cases in the years since Carpenter was decided. Over time, three of the Carpenter factors have appeared most frequently in substantive post-Carpenter cases: the revealing nature of the data collected, the amount of data collected, and whether the suspect voluntarily disclosed their information to others. The relative importance of these factors largely remains to be determined, but early indications are that the revealing nature of the data is the most influential of the factors, while the amount of data is slightly more influential than its voluntary disclosure. In any event, all three factors appear to matter substantially to case outcomes.

These factors have appeared in numerous recent cases that substantively apply Carpenter. Of the 129 federal and state cases discussing at least one Carpenter factor through March 31, 2021, 93 of them mentioned the revealing nature of the data collected; 116 of them mentioned the amount of data collected; and 61 of them mentioned the voluntary or involuntary nature of the data disclosure at issue. Moreover, statistical analyses confirm the influence of these three factors. Together, they make up an emerging test with powerful influence in current case law and the potential to guide courts in a wide variety of future Fourth Amendment cases. 

The Overlapping Frameworks of Katz and Carpenter

Current Fourth Amendment law is in a sort of limbo, stuck between two paradigms. The Katz test is well established in the case law and continues to apply to Fourth Amendment search questions as a doctrinal matter. But it is vague and unhelpful, a generic reasonableness test that fails to guide courts’ analyses of novel Fourth Amendment questions. The Carpenter test is far clearer. It is based on three relatively concrete factors that courts can effectively assess and have effectively assessed in numerous cases over the past several years. 

But it remains unclear when the Carpenter test applies. Currently the Katz and Carpenter tests operate simultaneously in Fourth Amendment search law. In the absence of a framework for choosing among these approaches, lower courts can choose to emphasize one or the other, or both, in a variety of configurations, creating profound uncertainty.

Some courts continue to apply primarily the Katz test, not mentioning Carpenter or doing so only in passing. For example, in United States v. Fanning, a federal district court held that the government could warrantlessly install a telephone pole camera near a warehouse where Fanning worked. Applying classic Katz analysis, the court reasoned that Fanning possessed no legitimate expectation of privacy in a public area and had taken no steps to protect himself from view or otherwise manifest his expectation of privacy.

Other courts have largely ignored the Katz test and focused primarily on the Carpenter factors in addressing new Fourth Amendment questions. In United States v. Tolbert, for instance, the court held that subscriber information, IP connection logs, and a list of friends associated with two AOL accounts were not protected by the Fourth Amendment. The opinion evaluated the Carpenter factors in some detail, concluding that the information at issue was unprotected because it was not intimate or revealing, was limited in amount, and was generated by Tolbert’s affirmative, voluntary acts. The court did not cite Katz or apply its test.

Finally, many courts use a blend of Katz and Carpenter concepts, conceptualizing and configuring these tests in various ways in the course of evaluating new Fourth Amendment issues. Some courts emphasize Katz and apply Carpenter as a small refinement of its concepts. Others emphasize Carpenter and occasionally incorporate classic Katz concepts into its multifactor analysis. Still others use different tests to address different aspects of government surveillance, applying the Katz test to publicly exposed data and the Carpenter test to data exposed only to a single third party. The potential permutations are virtually limitless. 

In all of these cases, the relative influence of the Katz and Carpenter tests seems to be largely a matter of judicial whim. No principles have arisen to govern when each test should be used, and no guidance can be found in the Supreme Court’s cases. Nor have any clear patterns emerged in the broader case law. Both tests are used to address simple surveillance practices and advanced digital technologies; both tests are used to address third-party data and direct government surveillance. In the face of this uncertainty, litigants are also likely motivated to argue under both the Katz and Carpenter tests, possibly to the detriment of the clarity of their arguments. It is, in short, a free-for-all.

The Carpenter Test as a Replacement for the Katz Test

We believe that the best way to address the current muddle of Fourth Amendment search law is for the Carpenter test to replace the Katz test as the primary standard for Fourth Amendment searches. The Carpenter test presents a unique opportunity for courts to substantially clarify Fourth Amendment search law, something they have been unable to achieve since Katz was decided.

Carpenter’s multifactor test will lead to more predictability in Fourth Amendment law. The Katz test has always been open textured and ambiguous. In contrast, the Carpenter test is relatively concrete and precise. To be sure, judges will inevitably disagree over how to assess one or another of the factors, or about how to weigh the various factors against each other. But the spectrum of disagreement will be narrow and cabined compared to the almost limitless range of disagreement possible under Katz. 

The Carpenter test also resonates more directly with the Fourth Amendment’s history. It treats the Fourth Amendment as a restriction on the government’s power to obtain information on its citizens, and not solely as a protector of privacy. For example, a government program that collects large amounts of personal data may be subject to Fourth Amendment restrictions under Carpenter even if the data was shared with some other people and therefore not “private.” The test is focused more on the government’s power over those it surveils than on the privacy expectations of the targets. 

The test also impels courts to engage in a deep consideration of the specific features of technology and society’s embrace of technology that was usually lacking from the conventional Katz test. This can prevent the kind of inadequate responsiveness to progress that plagued the Katz approach from its birth. For example, Carpenter recognized the way government power tends to get amplified by the purely private acts of corporations, such as the decision by cell phone providers to collect and store years of location data about its customers. Before Carpenter, Katz had often been interpreted to ignore the way in which private data practices feed government power, especially in the reviled third-party doctrine, but Carpenter avoids this misstep, better keeping the Fourth Amendment relevant in the technological age.

More broadly, Carpenter provides a standard that cabins and shapes courts’ reasoning. It requires at least an attempt at consistent analysis, in a way that Katz does not. Judges faced with high volumes of revealing data not voluntarily disclosed to others will be hard pressed to withhold Fourth Amendment protection. And judges considering small quantities of nonrevealing information voluntarily disclosed to others will find it difficult to extend Fourth Amendment protection. The Carpenter test, unlike Katz, is a real standard.

Further, Carpenter’s standard can be applied in a wide variety of cases involving potential Fourth Amendment searches. It is useful for both third-party disclosure cases and direct government surveillance cases. It is equally effective at addressing digital and nondigital data. It may even, in slightly modified form, be able to apply to classic physical intrusion scenarios. That is, courts could assess the intimacy of the place or thing examined (rather than the revealing nature of digital data), as well as the amount of information collected and any voluntary disclosure to another party, to address a case of physical intrusion. In any event, most physical searches of a person’s property are now per se searches under the Supreme Court’s recently revived trespass test. The types of surveillance to which Carpenter is ideally suited are those most likely to arise in modern Fourth Amendment cases: nonphysical captures of data via new surveillance technologies. 

Given Carpenter’s workability and conceptual reach, there is little reason to continue applying Katz’s vague reasonable expectation of privacy standard in most cases, except perhaps as a throat-clearing exercise at the beginning of a judicial opinion. Carpenter itself cited Katz and did not suggest it was replacing the Katz test. Until the Supreme Court makes such an announcement, which we urge it to do at the earliest opportunity, lower courts might feel compelled to at least cite Katz. To be clear, it matters little whether courts ignore Katz altogether or recite the Katz test before using Carpenter to determine whether an individual has a “reasonable expectation of privacy.” The important thing is that Carpenter governs the analysis of whether a government action is a Fourth Amendment search, unencumbered by any Katz-related assessments of actual expectations, or public exposure, or positive law, or social norms, or any of the other conflicting conceptual models of the Katz test. 

This is not to say that the Supreme Court’s myriad Katz precedents are valueless or need to be discarded when the Carpenter test is adopted. Indeed, another argument in favor of Carpenter is that its factors were not invented from whole cloth. Rather, they arose from decades of Supreme Court precedents grappling with difficult Fourth Amendment questions under the Katz framework. In this body of law, several principles gradually emerged that helped the Court identify when a government action required constitutional scrutiny. Those same principles make up the Carpenter test. As detailed here, the Supreme Court has looked to the revealing, intimate nature of government surveillance in numerous cases over the past several decades. It has often discussed the amount of data gathered by government officials in Fourth Amendment cases and constitutional privacy cases. And it has repeatedly addressed voluntary disclosures to third parties, although that factor is only one of several under Carpenter, rather than the definitive consideration it was in the Court’s early third-party doctrine cases. 

A Fourth Amendment Sea Change

The Fourth Amendment is in a period of flux, and we cannot assume that courts will inevitably choose the optimal Fourth Amendment test. Those who wish to make Fourth Amendment law more coherent must advocate in favor of a more intelligible approach. We begin that effort here. Carpenter creates a specific, concrete test that lower courts have successfully applied in numerous cases. It directly addresses government power and enables courts to effectively adapt to new technologies and social contexts. And its factors are not experimental or wholly novel; each one is grounded in Supreme Court precedents decided over the past several decades. It is time for courts to adopt Carpenter overtly, as a test, turning away from the Katz reasonable expectation of privacy test and taking an important step toward clarifying Fourth Amendment search law.