Episode Summary
Executive Summary: The episode surveys U.S. policing and criminal justice through a data-driven interview with economist Jennifer Doleac. It highlights mass incarceration, racial disparities, police misconduct, and reform options, while stressing that evidence is often weak and policy should be tested rigorously. Promising approaches include better training, accountability, recruiting, community trust, and non-police crime prevention like CBT, lead abatement, lighting, and DNA databases.
Main Topics: Mass incarceration and U.S. criminal justice scale: The host lays out the size and composition of incarceration in the U.S., emphasizing how unusually high imprisonment is compared with other countries and how much jail churn, pretrial detention, and racial disparity shape the system. Policing problems: force, bias, and discretion: Doleac argues the biggest policing problems are unnecessary escalation and racial bias, especially where officers have wide discretion. She discusses how difficult it is to measure misconduct and how it concentrates in low-level decisions. What evidence says about reforms: The conversation reviews procedural justice training, body-worn cameras, police diversity, union accountability, and implicit bias training, stressing that some popular reforms have mixed or null evidence while others show promise. Crime prevention outside the criminal justice system: A major theme is that prevention may be more effective than punishment, with examples such as CBT, lead abatement, street lighting, economic support, and DNA databases showing large crime reductions. Police unions, accountability, and culture change: The interview argues that unions and weak discipline make it hard to remove bad officers, and that changing incentives, hiring, and accountability structures may be necessary for lasting reform. Methods, evidence quality, and policy evaluation: Doleac explains why natural experiments, robustness checks, and transparent causal reasoning matter, and cautions against overselling weak evidence or pretending advocacy is settled science. Diversity and inclusion in economics and policing research: The discussion broadens to diversity in economics, with Doleac arguing that more women and people of color improve both the moral legitimacy and the intellectual quality of the field.
Key Arguments: U.S. incarceration is extraordinarily high, but most incarcerated people are there for violent or serious offenses, so decriminalizing drugs alone would not solve mass incarceration. Police misconduct is difficult to measure comprehensively, but available evidence suggests racial bias appears more strongly in discretionary, low-level decisions than in serious felony enforcement. Hiring more police generally reduces crime, mainly through deterrence, but this does not mean current policing practices are socially optimal because they also impose major harms. Better accountability is needed because current systems make it difficult to fire officers with repeated complaints or misconduct. Procedural justice training can reduce complaints and use of force, showing that relatively small interventions may matter if they change officer behavior and public trust. Body-worn cameras have not consistently reduced force or complaints across randomized trials, so they are not a silver bullet. Police diversity can improve trust and reporting in communities, as seen in studies of court-mandated hiring of Black and female officers. Police unions likely impede reform by protecting bad officers and weakening accountability, though the exact effect varies by jurisdiction. Evidence-based reform should be transparent about uncertainty; advocates should avoid claiming strong causal proof where only correlational comparisons exist. The most promising way to reduce crime may be upstream prevention: CBT, lead abatement, better lighting, youth jobs, and economic support. DNA databases can substantially reduce recidivism by increasing perceived certainty of detection. If reforms reduce police-community trust, they may reduce cooperation and clearance rates, creating indirect crime harms that are often overlooked.
Data Points: U.S. incarceration rate: about 1 in 110 adults - Current U.S. incarceration level cited in the introductory statistics section Total incarcerated population: 2.3 million - Combined count of prisons and jails in the United States State prisons population: 1.3 million - Breakdown of the 2.3 million incarcerated people Local jails population: 630,000 - Breakdown of the 2.3 million incarcerated people Federal prisons and jails population: 225,000 - Breakdown of the 2.3 million incarcerated people Jail check-ins per year: 10.6 million - Annual jail churn far exceeds prison admissions Share of jail population unconvicted: around three-quarters - Many jail detainees are awaiting trial or may never be convicted Black share of incarcerated population: 40% - Racial disparity in incarceration versus population share Black share of U.S. residents: 13% - Baseline population share for comparison Misdemeanor charges per year: 13 million - Includes low-level offenses such as jaywalking Low-level offenses share of jail population: nearly 25% - People held for offenses with penalties under a year Share incarcerated for violent crime: 47% - Among convicted incarcerated people Share incarcerated for drug crime: 19% - Among convicted incarcerated people Share incarcerated for property crime: 17% - Among convicted incarcerated people Share incarcerated for public order offenses: 17% - Among convicted incarcerated people Private prisons share: less than 9% - Used to argue private prisons are not the main driver of mass incarceration Serious violent crimes reported in 2017: about 2 million - Used to estimate arrest likelihood for serious violent offenses Arrests for serious violent crimes in 2017: about 450,000 - Compared against reported violent crime volume Estimated chance of arrest for serious violent crime: at most 22% - Rough calculation from reported crimes versus arrests Clearance rate for violent crimes: 46% - Police-reported clearance by arrest rate for the U.S. overall Clearance rate for property crimes: 18% - Police-reported clearance by arrest rate for the U.S. overall Average time served for drug possession: 15 months - From state-prison release data in 2016 Median time served for drug possession: 10 months - State-prison release data in 2016 Average time served for robbery: 4.7 years - State-prison release data in 2016 Median time served for robbery: 3.2 years - State-prison release data in 2016 Average time served for murder: 15 years - State-prison release data in 2016 Police killings per year: around 1,000+ - Approximate annual U.S. police killings referenced in discussion Police per-capita effect on crime: 0.1% to 1% crime reduction for each 1% increase in police - Elasticity estimate attributed to the literature, especially Chelfin and McCrary DNA database effect in Denmark: 43% reduction in reoffending - Study of more than 38,000 males added to DNA database DNA database effect in U.S.: 17% lower likelihood of a new conviction within five years for serious violent offenders - Estimated effect from U.S. DNA database expansion DNA database effect for serious property offenders: 6% reduction - U.S. DNA database estimate Lighting study effect: 36% reduction in nighttime outdoor crimes - Referenced floodlight/public-housing lighting experiment Body-worn cameras evidence: over a dozen randomized controlled trials - Used to assess camera effects on police behavior Chicago procedural justice training: 8,000+ officers - One-day training rollout studied via random assignment of training timing Procedural justice training effect: reduced complaints and use of force - Staggered rollout study in Chicago found meaningful improvements Police department count in U.S.: 18,000 police departments - Illustrates extreme decentralization and variation across jurisdictions
Pivotal Quotes: "the problems that are certainly at the forefront of everyone's minds right now are around the unnecessary escalation of incidents between community members and law enforcement" — Jennifer Doleac: Describing the main policing problem she sees "we know that racial bias is a problem throughout a society in the United States and lots of other countries too. And so this is a problem that is everywhere" — Jennifer Doleac: Explaining why bias in policing should not be surprising "I think the challenge here, right, is how you define criminal" — Jennifer Doleac: Discussing police violence, legal discretion, and accountability
Implications: The episode argues for pragmatic reform: test interventions carefully, prioritize accountability and trust, and invest more in upstream crime prevention. Listeners should expect progress to come from state/local experimentation rather than one sweeping federal fix.