Aftermath of Carpenter: An Empirical Study of Fourth Amendment Law, 2018-2021

A forthcoming Harvard Law Review article reviewed 857 cases that cited Carpenter v. United States, the landmark Supreme Court Fourth Amendment case, from its publication in June 2018 to March 2021. The purpose of this study was to evaluate the landscape of post-Carpenter Fourth Amendment law. The full text of the article can be found here. The Carpenter decision held that parties may retain a Fourth Amendment privacy interest in data disclosed to third parties. Specifically, the holding pertained to the collection of cell phone GPS data. This was a dramatic departure from the default rule, the third-party doctrine, under which parties lost their constitutional privacy interest in their data if they voluntarily disclosed it to a third party. The Carpenter Court abstractly weighed several factors as to what constitutes a digital data search; they granted lower courts a liberal license in their own application. The factors included the following: the revealing nature of the data, the amount of data collected, the inescapable and automatic nature of the data disclosure, the number of people affected, and the low cost of data collection. This study examined how lower courts have applied the Carpenter factors and any statistical correlation with outcome. "Outcome" refers to whether the court found there to be a search or not. Only three of the factors discussed in Carpenter had an impact on outcome: the revealing nature of the data had the strongest impact on outcome; the automatic nature of data collection had the second strongest impact on outcome; and finally, the amount of data collected had a still significant impact. The low cost of the search, the number of people affected, and the inescapability of the data collection had no statistically significant effect on whether the court held it to be a search. Surprisingly, the "good faith exception" to the exclusionary rule played a much larger outcome in courts' findings than any of the Carpenter Court's factors. Courts resolved the admissibility of evidence on the good faith exception before reaching the search issue in 36.1% of the cases that applied Carpenter substantively. Even in 2020 and 2021, courts admitted evidence for searches in 30% of cases based on the exclusionary rule. The effect is that likely hundreds of criminal defendants have been convicted based on searches that may have been unconstitutional had Carpenter been applied, even years after the Carpenter holding. There is concern that the good faith exception may incentivize law enforcement to aggressively collect as much data as possible before courts impose a warrant requirement. In an ever-changing digital data landscape, there will be a perpetual lag between the courts and technological development. As evidenced here, the good faith exception ensures that convictions achieved with that data will be upheld, even years after the courts match pace with technology. The article calls the courts to reexamine the good faith exception, given its powerful impact on rulings. Venue unexpectedly correlated with outcome. Federal courts that applied Carpenter found a search in only 21.9% of the rulings. State courts that applied Carpenter found a search in 60.6% of the rulings. Most likely, state courts had less experience applying the third-party doctrine than federal courts. Therefore, federal courts were more apt to narrowly apply Carpenter, partially due to habit and partially to status quo bias. A key takeaway is that state courts are statistically more likely to find a search when substantively applying Carpenter. There is no clear roadmap for how to apply the Carpenter factors, but the revealing nature of data collected has had the biggest impact on outcome thus far. Finally, the author anticipates that the Supreme Court will continue to expand Fourth Amendment protections, but the impact will be delayed by courts' generous application of the good faith exception to searches.