How To Address Concerns About Data on Deaths in Custody

One year ago, George Floyd was murdered by Minneapolis police as the world watched, spurring a blistering call for police accountability in the United States. Floyd is one among the countless Black Americans and other people of color killed by law enforcement: Breonna Taylor, Michael Brown, Freddie Gray, Amadou Diallo, Botham Jean, Alton Sterling, Daniel Prude, Philando Castile, Eric Garner, Tamir Rice, Atatiana Jefferson, Stephon Clark, and too many others. In the year since Floyd’s death, the list has grown longer still with the deaths of Tony McDade, Rayshard Brooks, Daunte Wright, Ma’Khia Bryant, Adam Toledo, Andrew Brown, and, again, too many others.¹

Roughly 1,000 people have been killed by law enforcement since May 2020, according to estimates compiled by the Mapping Police Violence project.² Yet this figure, while staggering, is almost certainly an underestimate of the total number of civilian deaths in the custody of the criminal justice system at large, since the database relies on media reports to track police violence and does not take into account deaths that occur in correctional facilities. Even still, these independent estimates are among the most complete accountings of fatalities within the system, the full scope of which cannot be determined due to a lack of official data.³

Data on deaths in custody are crucial for holding law enforcement and correctional facilities across the country accountable. The absence of accurate and complete information on the number of people who die in custody and the nature of such deaths stifles policymakers’ ability to examine the underlying causes, let alone determine what can be done to lower the incidence.

Federal efforts to collect these numbers have been in the works for more than two decades, since Congress unanimously passed the Death in Custody Reporting Act of 2000.⁴ Yet there are still no reliable, official data on the matter, even after Congress reupped the legislation by unanimously passing the Death in Custody Reporting Act (DCRA) of 2013.⁵ And while the U.S. Department of Justice (DOJ) began its DCRA data collection in 2020, outstanding funding and compliance issues could compromise the quality of the impending data.⁶ To take a meaningful first step toward accountability in the U.S. criminal justice system, Congress and state legislatures should use every policy tool at their disposal to resolve these concerns.

Background of the DCRA

Enacted to address a lack of reliable information about deaths in the custody of law enforcement and correctional institutions, the Death in Custody Reporting Act of 2000—which expired in 2006—required states to collect and report to the DOJ quarterly data on any death of a person who is in the process of arrest, is en route to be incarcerated, or is incarcerated at a municipal or county jail, state prison, or other local or state correctional facility.⁷ However, the law lacked teeth and funding from the start, resulting in underreporting so severe that the DOJ Bureau of Justice Statistics (BJS)—the agency responsible for gathering the data at the time—suspended its program for capturing arrest-related deaths in 2014.⁸

In response, Congress enacted the updated DCRA later that year, this time leveraging an enforcement mechanism to collect similar information from state law enforcement agencies.⁹ Unlike the original Death in Custody Reporting Act, the 2013 legislation authorized the AG to penalize noncompliant states by withholding up to 10 percent of their allotted funds under the Edward Byrne Memorial Justice Assistance Grant Program (JAG), the leading source of federal dollars for state and local criminal justice agencies.¹⁰ The law also required the AG to study means by which such data could be utilized to reduce the number of deaths in custody and to report the findings to Congress by no later than December 18, 2016.

However, no such report has been submitted to date, prompting Reps. Jerrold Nadler (D-NY) and Karen Bass (D-CA) to write a letter in January 2020 asking the DOJ inspector general to review the department’s continued failure to comply with the DCRA.¹¹ The letter reads, “In the absence [of] action by the Department [of Justice], the United States continues to face a persistent crisis of deaths in custody, the true scope of which remains unknown.” Later that year, in July, a group of 17 senators further underscored the importance of implementing the DCRA in a letter to then-U.S. AG William Barr.¹² The letter stressed that while the DCRA could not solve the underlying problem of police violence on its own, the data collected would be a “critical step” in the direction of accountability and would “inform the substantive and structural reforms that we must make in the weeks and months ahead.”

While a plan had been in place to not only collect but also verify such data in December 2016, it was never implemented under the Trump administration; 2020 marks the first calendar year in which the DOJ started collecting DCRA data from state law enforcement agencies.¹³

Unintended consequences of the DCRA’s JAG penalty

Even with the reporting finally underway, the incoming data could face potential issues stemming from the JAG penalty in the DCRA. A 2018 report from the DOJ Office of Inspector General (OIG) identified two unintended consequences of the provision that could negatively affect the quality of the data: 1) the transfer of data collection authority from one DOJ agency to another and 2) the creation of disincentives for state and local agencies to report their data.

First, because the provision renders the administration of the DCRA a “policymaking activity” rather than a purely statistical one, it disqualifies BJS from serving as the agent that gathers DCRA data from states.¹⁴ Although BJS specializes in data collection and has even piloted a promising approach for implementing the DCRA, it is required by U.S. Office of Management and Budget guidance to be an independent statistical agency and thus cannot participate in “policymaking activities.”

The DOJ Bureau of Justice Assistance (BJA)—the grant-making office that administers JAG funds—now takes on the role. However, due to budget limitations, BJA has not implemented BJS’ proposal of using a web-based, open-source application to search for reports on potential deaths in custody and surveying local law enforcement agencies to validate the leads, despite a pilot study showing the methodology’s success in enabling BJS to identify nearly three times more arrest-related deaths than before.¹⁵ Instead, BJA is relying on state law enforcement agencies to collect DCRA data from local agencies, which may result in incomplete reporting given that state agencies are generally not in the position to compel local agencies to report crime data without the requisite state laws. What is more, a similar approach taken by the BJS program that was suspended in 2014—before BJS developed the search-and-validate methodology—had captured only about 50 percent of law enforcement homicides.¹⁶

Second, because the penalty only authorizes the AG to withhold JAG funds from an entire state, if enforced, it could end up punishing compliant state and local law enforcement agencies for noncompliance by other agencies in the state. In some states, such as Alaska and Missouri, many state and local law enforcement agencies are not financially incentivized to comply with the DCRA because they do not receive JAG funds.¹⁷ The ones that do receive JAG funds, meanwhile, are financially incentivized to comply with the DCRA in order to keep that funding but could still be punished due to noncompliance by agencies that do not receive those funds. This creates a perverse incentive structure by which neither the agencies that have a financial incentive to comply nor those that do not have such an incentive are actually motivated to report their DCRA data.

Assuring the reliability of DCRA data

These issues raise significant concerns for the quality of the forthcoming DCRA data, as neither BJA nor state and local law enforcement agencies have a structure in place that would ensure complete and accurate reporting. Findings based on such flawed data—which have yet to be submitted to Congress—would not help policymakers understand the causes of deaths in custody or reduce their occurrence, the primary purpose of the DCRA.¹⁸

To address the unintended consequences of the JAG penalty, lawmakers at both the federal and state level must take action.

Actions for Congress

Congress should appropriate the necessary funding for BJA to implement a methodology to search for and validate leads on deaths in custody. In interviews with the OIG, BJA’s deputy director cited a lack of financial resources as an explanation for why BJA did not plan on using a web-based, open-source methodology similar to the one BJS championed.¹⁹ According to the deputy director, BJA could not leverage BJS’ search application because it was designed to meet requirements different than those of the DCRA and belonged as intellectual property to the contractor that had developed it. Moreover, even if BJA were to acquire a separate tool suited for meeting DCRA requirements, it would likely need to hire additional staff to validate the leads or contract with a third party to do so; BJA’s deputy director also acknowledged that her staff might not have the statistical expertise to reconcile the results with the reporting from states.

Congress should therefore appropriate funds to cover both the startup costs of acquiring a DCRA-specific search tool and the human-resource costs of validating the leads. This would give BJA the financial resources to put in place a methodology to capture deaths in custody that might otherwise go unreported and to verify the accuracy of state reporting. As a benchmark for Congress to consider, BJS spent more than $4 million of its own budget managing data on deaths in custody—including the cost of developing its search application—between fiscal years 2014 and 2017.²⁰

Actions for states

For their part, state legislatures should look to compel all state and local law enforcement agencies to report DCRA data—for example, by passing statewide legislation that would mandate DCRA reporting from all agencies, including those that do not receive JAG funds.

Some states already have laws that require all agencies to report data similar to those required by the DCRA; such cases can serve as models for other states to follow. California, for example, mandates any state or local law enforcement agency or correctional facility to report to the California AG all facts of a death in custody within 10 days of it occurring.²¹ Texas has a comparable requirement, while also making an investigator’s failure to report a Class B misdemeanor and fining agencies up to $1,000 a day for noncompliance.²² Maryland, meanwhile, requires officer-involved deaths to be reported to the Maryland Governor’s Office of Crime Control and Prevention on an annual basis.²³ Similarly, Tennessee requires law enforcement-related deaths to be reported to the Tennessee Bureau of Investigation on an annual basis.²⁴

These examples demonstrate how state legislatures across the country can and should take an active role in incentivizing DCRA compliance by all agencies. Doing so would prevent compliant agencies from being punished by the JAG penalty due to noncompliance by other agencies in the state, improve the quality of DCRA data, and bring about accountability in the criminal justice system. In Texas, for example, quality of data on deaths in custody and police shootings has markedly improved since the reporting requirements and penalties for noncompliance were enacted, leading to research, oversight, and even reform such as the adoption of evidence-based practices at community corrections.²⁵

Together, these policies would shore up the data gaps produced by the JAG penalty in the DCRA. The provision is an imperfect solution for DCRA enforcement by itself; however, when combined with the recommended congressional action and state legislation, it can create meaningful accountability for law enforcement and the criminal justice system at large.

Conclusion

For decades, the lack of reliable DCRA data has subverted the intent of the law—to help policymakers understand the causes of deaths in custody and curtail the incidence of these tragedies. Without such data, however, policymakers have not been able to assess the role that law enforcement and correctional facilities play in the deaths in their custody. While the DOJ started collecting DCRA data last year, there remain funding and compliance problems that could undermine the quality of the forthcoming data. Congress and state legislatures must therefore be proactive in pulling the available policy levers to tackle these issues.

One year after the murder of George Floyd, lawmakers at both levels of government can take a crucial first step toward police accountability. Floyd was not alone. Over the past year, the world has borne witness to too many other killings by law enforcement, not to mention the long history of police violence committed against communities of color. Still, the full extent to which the crisis of deaths in custody pervades the nation remains a mystery today. Congress and state lawmakers must rectify this problem immediately by working to promote accountability for the criminal justice system and keep communities safe.

Kenny Lo is a research associate for Criminal Justice Reform at the Center for American Progress.