Big Data and Predictive Reasonable Suspicion
The Fourth Amendment requires “reasonable suspicion” to stop a suspect. As a general matter, police officers develop this suspicion based on information they know or activities they observe. Suspicion is individualized to a particular person at a particular place. Most reasonable suspicion cases involve police confront-ing unknown suspects engaged in observable suspicious activities. Essentially, the reasonable suspicion doctrine is based on “small data”—discrete facts, limited information, and little knowledge about the suspect.
But what happens if this small data suspicion is replaced by “big data” suspicion? What if police can “know” personal information about the suspect by searching vast networked information sources? The rise of big data technologies offers a challenge to the traditional paradigm of Fourth Amendment law. With little effort, officers can now identify most unknown suspects, not through their observations, but by accessing a web of information containing extensive personal data about suspects. New data sources, including law enforcement databases, third-party records, and predictive analytics, combined with biometric or facial recognition software, allow officers access to information with just a few search queries. At some point, inferences from this personal data (independent of the observation) may become sufficiently individualized and predictive to justify the seizure of a suspect. The question this Article poses is whether a Fourth Amendment stop can be predicated on the aggregation of specific and individualized, but otherwise noncriminal, factors.
For example, suppose police are investigating a series of robberies in a particular neighborhood. Arrest photos from a computerized database are uploaded in patrol cars. Facial recognition software scans people on the street. Suddenly there is a match—police recognize a known robber in the targeted neighborhood. The suspect’s personal information scrolls across the patrol car’s computer screen—prior robbery arrests, prior robbery convictions, and a list of criminal associates also involved in robberies. The officer then searches additional sources of third-party data, including the suspect’s GPS location information for the last six hours or license plate records which tie the suspect to pawn shop trades close in time to prior robberies. The police now have particularized, individualized suspicion about a man who is not doing anything overtly criminal. Or perhaps predictive software has already identified the man as a potential reoffender for this particular type of crime. Or perhaps software has flagged the suspect’s social media comments or other Internet postings that suggest planned criminal or gang activity. Can this aggregation of individualized information be sufficient to justify interfering with a person’s constitutional liberty?
This Article traces the consequences of a shift from “small data” reasonable suspicion, focused on specific, observable actions of unknown suspects, to a “big data” reality of an interconnected, information rich world of known suspects. With more specific information, police officers on the streets may have a stronger predictive sense about the likelihood that they are observing criminal activity. This evolution, however, only hints at the promise of big data policing. The next phase will use existing predictive analytics to target suspects without any firsthand observation of criminal activity, relying instead on the accumulation of various data points. Unknown suspects will become known to police because of the data left behind. Software will use pattern-matching techniques to identify individuals by sorting through information about millions of people contained in networked data-bases. This new reality simultaneously undermines the protection that reasonable suspicion provides against police stops and potentially transforms reasonable suspicion into a means of justifying those same stops.
This Article seeks to offer three contributions to the development of Fourth Amendment theory. First, it demonstrates that reasonable suspicion—as a small data doctrine—may become practically irrelevant in an era of big data policing. Second, it examines the distortions of big data on police observation, investigation, and prediction, concluding that big data information will impact all major aspects of traditional policing. Third, it seeks to offer a solution to potential problems using the insights and value of big data itself to strengthen the existing reasonable suspicion standard.
Choice of Law in Fraudulent Joinder Litigation
The grant of subject matter jurisdiction to federal courts based on diversity of citizenship has, for centuries, required complete diversity between parties to the litigation. If a case is brought in state court and complete diversity exists between the parties, the defendant has the statutory right to remove the case to federal court. Absent another basis for federal jurisdiction, the lack of complete diversity strips the federal court of subject matter jurisdiction, and the court is required to remand the case to state court. Therefore, the presence of a single nondiverse defendant is sufficient to defeat diversity jurisdiction.
The doctrine of fraudulent joinder has arisen in response to plaintiffs’ efforts to take advantage of this complete diversity requirement and thereby control whether a state or federal court hears their case. Fraudulent joinder refers to a plaintiff’s attempt to defeat complete diversity and generally occurs in one of two ways: (1) a plaintiff commits actual fraud by inaccurately pleading the citizenship of the parties to the lawsuit or (2) a plaintiff sues a nondiverse defendant against whom the plaintiff cannot establish a cause of action. This Comment focuses only on the second method, which is litigated with much greater frequency than the first. In this context, the diverse defendant will generally remove the case to federal court and argue that the judge should ignore the citizenship of the nondiverse co-defendant because the plain-tiff has no chance of stating a claim against that defendant.
If the federal court determines that the plaintiff has no cause of action against the nondiverse defendant, the court is permitted to ignore that defendant’s citizenship and thereby establish complete diversity. However, if the court determines that the nondiverse defendant is properly joined, then complete diversity is absent and the case must be remanded to state court. The result of a fraudulent joinder dispute therefore determines whether a suit will be heard in state or federal court—an important consideration for litigants.
To resolve the fraudulent joinder question, the federal judge must look to a source of law to determine whether the plaintiff has established a cause of action against the nondiverse defendant. In many cases, all parties agree on which law applies to the case; in those circumstances, the judge will conduct the fraudulent joinder inquiry using the agreed-upon legal standards. In some cases, however, the events giving rise to the litigation have connections to multiple jurisdictions and therefore multiple laws could potentially apply. In these cases, the judge must decide which state’s law to use to determine whether the plaintiff has stated a claim against the nondiverse defendant.
In such a case, it is not clear how a judge should approach this choice of law determination or whether the judge is even permitted to make a choice of law determination at all. On the one hand, if the nondiverse defendant is properly joined, the federal court does not have subject matter jurisdiction and arguably cannot make a choice of law determination. On the other hand, there may be cases in which the plaintiff can state a claim against the non-diverse defendant under one state’s substantive standard, but not under another state’s standard. In these cases, the court cannot resolve the fraudulent joinder dispute without making a choice of law determination.
This Comment takes the position that a court must perform a choice of law analysis as part of its fraudulent joinder inquiry. This determination is necessary to decide whether a law could apply to the case that would sustain the action against the nondiverse defendant. Further, the court should give the same deference to the plaintiff’s choice of law as it gives to the plaintiff’s choice of forum under the jurisdiction’s fraudulent joinder standard. This conclusion draws upon the idea that choice of law is fundamentally a merits-based inquiry and that a decision on the merits re-quires reference to an applicable law. Therefore, a court cannot properly make a determination about whether the suit against the nondiverse defendant has merit, or is merely “fraudulent,” without selecting and applying an appropriate law.
The Future of Government-Mandated Health Warnings After R.J. Reynolds and American Meat Institute
Government-mandated disclosures and warnings aimed at promoting public health are ubiquitous. Alcoholic beverage labels bear government warnings against alcohol consumption during pregnancy. Both prescription and over-the-counter drugs must comply with extensive Food and Drug Administration (FDA) labeling requirements. Automobiles carry mandatory safety rating labels. Cigarette packages have included warnings about the dangers of smoking since 1965. Even chain restaurants must now follow the federal nutrition labeling requirements that have applied to food packaging for two decades. Warnings and disclosure requirements are likely to become even more widespread given President Obama’s 2011 executive order encouraging administrative agencies to use these “[f]lexible [a]pproaches” wherever “relevant, feasible, . . . consistent with regulatory objectives, and . . . permitted by law.”
Despite their widespread use as a regulatory tool, government-mandated warnings and disclosures are not immune from legal challenge. In the 2012 case of R.J. Reynolds Tobacco Co. v. FDA, the U.S. Court of Appeals for the D.C. Circuit invalidated FDA’s graphic cigarette warnings on First Amendment grounds. The tobacco manufacturers’ challenge forced the D.C. Circuit to wade into unchartered waters. Although there is a long line of Supreme Court cases addressing First Amendment challenges to commercial speech restrictions (e.g., advertising bans), the Court has heard only two challenges to commercial speech disclosure requirements, neither involving government-mandated warnings. Further, while the Court has been clear that it reviews commercial speech restrictions under the Central Hudson intermediate scrutiny test, it has applied a standard akin to rational basis review when examining purely factual disclosure requirements targeting consumer deception, without explaining in what other circumstances rational basis review would apply. Thus, faced with a novel question of law, the R.J. Reynolds court concluded that the graphic cigarette warning requirements did not merit rational basis review protection, because (1) they did not seek to cure consumer deception and (2) they were not purely factual and uncontroversial warnings, but rather “admonitions: ‘[D]on’t buy or use this product.’” After deciding that Central Hudson intermediate scrutiny was the correct standard of review, the court held the warnings unconstitutional because FDA failed to produce sufficient evidence—indeed, “failed to present any data”—that the warnings would directly and materially advance its goal of reducing smoking rates. Although most commentators expected the case to go to the Supreme Court, FDA instead withdrew the proposed images and said it would issue revised graphic warnings.
To the extent that R.J. Reynolds could be read as holding that only commercial speech mandates that are both purely factual and designed to correct consumer deception receive rational basis review, it was overruled by the 2014 en banc decision of the D.C. Circuit, American Meat Institute v. USDA. Aligning the court’s position with that of other circuits, the D.C. Circuit held in American Meat Institute that Zauderer v. Office of Disciplinary Counsel, the first Supreme Court case to apply rational basis review to a government-mandated disclosure requirement, extended “beyond problems of deception”—and thus applied to the U.S. Department of Agriculture (USDA) country-of-origin disclosures at issue in the case.
Given that the D.C. Circuit is responsible for reviewing many federal agency regulations, American Meat Institute marks a significant victory for regulators. A contrary holding—one limiting the protection of Zauderer rational basis review to compelled speech aimed at curing deception—would have threatened to unsettle the current regulatory regime, and would have particularly threatened mandates aimed at promoting public health. These disclosure requirements often do not target potentially deceptive commercial speech, and they rarely are supported by the level of evidence R.J. Reynolds deemed necessary to satisfy Central Hudson intermediate scrutiny. (These evidentiary difficulties arise in part because many public health problems are complex and cannot be eradicated by a disclosure requirement alone.) While R.J. Reynolds raised important questions about the effectiveness of disclosure requirements, the First Amendment should not be an insurmountable obstacle when the commercial speaker’s constitutionally protected interest is, as the Court has said, “minimal” and the government interest is substantial.
Although American Meat Institute lessened the blow R.J. Reynolds dealt to regulators, both decisions left open important questions about the First Amendment treatment of government-mandated warnings that are neither “purely factual and uncontroversial” disclosures nor overt government-sanctioned opinions, and about whether graphic cigarette warnings belong in this middle ground. R.J. Reynolds only addressed the constitutionality of the nine warnings before it, and left unanswered whether another graphic warning depicting the negative health consequences of smoking could be constitutional. But, in characterizing FDA’s graphic warnings as “a much different animal” than the mandated statements to which the Supreme Court has previously applied rational basis review, and in viewing them as “intended to evoke an emotional response, or, at most, shock the viewer into retaining the information in the text warning,” R.J. Reynolds strongly implied that no graphic cigarette warning could ever receive rational basis review protection. Not only did the court seem to demand Central Hudson intermediate scrutiny review for all government-mandated graphic warnings, it created an overly burdensome intermediate scrutiny test by misapplying the Administrative Procedure Act’s (APA) “substantial evidence” standard to its First Amendment analysis, and by failing to look beyond the Court’s abstract statements about Central Hudson to its application of the test.
Part I of this Note outlines the Supreme Court’s commercial speech jurisprudence. It explains the Court’s differential treatment of commercial speech restrictions and compelled commercial speech, and it outlines the open question of whether Zauderer, and its accompanying rational basis review protection, is limited to mandates aimed at correcting deception. Part II discusses the various interpretations of Zauderer advanced by circuit courts. It defends the broader interpretation of Zauderer adopted by the First Circuit, Second Circuit, and, most recently, the D.C. Circuit in American Meat Institute, and it criticizes the narrow interpretation articulated in R.J. Reynolds. Part II likewise outlines the doctrinal support and policy justifications for a broader interpretation of Zauderer, with particular focus on the importance of recognizing the government’s interest in promoting public health as worthy of rational basis review. Part III then looks at how FDA could issue revised graphic cigarette warnings that would pass constitutional muster. It examines the type of graphic cigarette warnings that could potentially merit review under the Zauderer standard, argues that R.J. Reynolds misapplied the Central Hudson intermediate scrutiny standard, suggests a better view of Central Hudson as applied to graphic cigarette warnings, and describes post-Reynolds scientific research supporting the effectiveness of graphic warnings.
New Regulations and Pending Cases
Do new regulations apply to pending cases? The question is simple, but the short answer is a lawyer’s favorite: “It depends.” It depends on the organic statute, it depends on the regulation, and, unfortunately, it may even depend on the federal court of appeals that happens to decide the case. This Essay looks at this issue by examining the effect of the Department of Labor’s 2013 amendments to regulations governing claims under the Black Lung Benefits Act. This Essay explains why the new regulations are applicable to pending cases, even if the Department of Labor already issued its final decision on a claim and a party already petitioned a court to review that decision.
The analytical route to this result varies by circuit. This Essay explains why the factor-based retroactivity test used by most circuits better addresses fundamental due process concerns and is more administrable than the D.C. Circuit’s approach, which turns on whether a circuit split predates the new regulations. This Essay both provides a clear answer about whether the 2013 amendments to the black lung regulations apply to pending cases and suggests how courts should handle retroactivity questions for regulations more broadly.
If Professions Are Just "Cartels By Another Name," What Should We Do About It?
The state action doctrine has been a significant impediment in the campaign against anticompetitive conduct by provider-dominated state licensing boards. In Cartels by Another Name: Should Licensed Occupations Face Antitrust Scrutiny?, Professors Edlin and Haw argue that state licensing boards operate as a “massive exception” to the Sherman Act’s ban on cartels, and that the Supreme Court should use a pending case (North Carolina State Board of Dental Examiners v. FTC) to “hold boards composed of competitors to the strictest version of its test for state action immunity, regardless of how the board’s members are appointed.” They also propose the application of a modified rule of reason when deciding similar cases on the merits.
Professors Hyman and Svorny suggest three modifications to Edlin and Haw’s proposal. These modifications aim to limit occupational licensing’s anticompetitive tendencies and licensing boards’ anticompetitive behavior. First, in reviewing the decisions of licensing boards, courts should presume that states were not actively supervising the boards, absent compelling evidence to the contrary. Second, defendant–licensing boards should be required to present persuasive evidence of actual harm that their proposed licensing restrictions or restraints will prevent and should be required to show that private market and non-regulatory forces (including brand names, private certification, credentialing, and liability) are insufficient to ensure that occupations maintain a requisite level of quality. Finally, Professors Hyman and Svorny argue that legislators should take steps to roll back existing licensing regimes.