AI summary by TwinMind

567. Do the Police Have a Management Problem? | Freakonomics Radio

Freakonomics Radio Network

47m 39s

Background and Origin Story

  • 2007: Doctoral student in chemistry at University of Chicago shot and killed during robbery attempt
    • Led to soul-searching about university's impact on local community
    • Sparked creation of crime lab as R&D partner to government
  • 2015-2016: Chicago murder rate jumped 60% (300 additional homicides)
  • Federal investigation of Chicago PD after Laquan McDonald killing
  • Obama Administration DOJ civil rights division connected crime lab to Sean Malanowski (LAPD)

Strategic Decision Support Centers (SDSCs)

  • Malanowski's team identified technology deployment without proper training
  • Chicago PD had 14,000 officers but zero civilian crime analysts (LAPD had 100 for 10,000 officers)
  • Built decentralized real-time crime centers in high-crime neighborhoods
    • Combined CCTV cameras, gunfire detection, license plate recognition
    • Paired civilians with police officers for data analysis
  • First SDSC in Englewood (South Side Chicago)
    • February 2017: First month without homicides in neighborhood
    • Initial success but effects didn't last after year
  • Problem: Commanders lacked training to effectively use data and manage operations

Research Foundation

  • Ludwig, Capoen, and Newman analyzed data from 50 biggest US police forces
  • Studied management turnover effects on crime rates and police use of force
  • Key finding: Police managers matter enormously - better leadership generates huge improvements in public safety and reductions in force
  • Peter Principle problem: Good patrol officers promoted to management without training
    • Example: College-educated officer becomes good at "chasing down criminals"
    • Gets promoted to run district with 500 officers and 200,000 residents
    • No management training provided

Policing Leadership Academy (PLA) Program Structure

  • First cohort: May 2022, University of Chicago campus
  • 24 departments represented plus Manchester, England
  • Participating cities: LA, NYC, Cleveland, Miami, Chicago, Indianapolis, Arlington TX, Dallas, Waco TX, Aurora CO, Boston, DC, New Orleans, Detroit, Phoenix, Madison WI
  • Format: One week per month over 6 months
  • Curriculum director: Sandy Joe MacArthur (retired LAPD Assistant Chief)
  • Only 3 of 24 departments had leadership training before promotion

Curriculum Components

  • Data literacy: Converting information into intelligence
  • Trauma-informed policing: Understanding impact on officers and community
  • Cultural competency: Natural curiosity without assumptions
  • Strategic thinking and risk management
  • Violence reduction and community resilience
  • Social network analysis for violence prevention
  • Psychology of conflict and organizational morale
  • "Show me clear" tactic: Mental reset between calls

Participant Examples and Projects

  • Stephanie Dresser (Madison WI Captain): Promoted without formal training
  • Michael Walley (Indianapolis Deputy Chief): Learned multi-agency coordination for bar district violence
  • Kenneth Corey (30-year NYPD veteran): Emphasized officer care and social network analysis
  • Ciana Paul-Noel (Miami Lieutenant): Co-responder model with mental health professionals for high utilizers

Economic and Scale Arguments

  • Training leaders vs. frontline personnel provides huge leverage
  • Low-cost intervention compared to salary expenses
  • Gun violence concentrated in ~200 cities
  • ~20 key leaders per department = 4,000 people to train nationally
  • Each serious crime reduces city population by 1 person
  • Each murder reduces city population by 70 people
  • Gun violence creates headwinds for economic development

Measurement and Evaluation

  • Randomized admission process for research validity
  • Departments must share data for participation
  • Success metrics: Reduced gun violence, reduced police use of force, fewer low-value arrests
  • Challenges: Self-reported police data reliability varies
  • Body cameras + AI tools emerging for measuring police-citizen interactions
  • Results from first cohort pending

Historical Context

  • 18,000 federal, state, county, local law enforcement agencies in US
  • Decentralized system with limited coordination
  • Historical issues: Slave patrols, corruption in 1920s-30s, militaristic responses in 1960s
  • Police became "enforcers" rather than "peacekeepers"
  • LAPD example: Chief Parker's military-style corruption cleanup had mixed results

Implementation Challenges

  • 99% of government programs never evaluated with data
  • Government lacks feedback mechanisms unlike private sector
  • Complex social problems difficult to measure
  • Need for systematic rather than individual solutions
  • Second cohort convening January 2024

TwinMind

Get summaries like this for your own meetings

TwinMind takes notes in the background during meetings, lectures, and calls, then writes the summary for you.

Free plan to get started

No bot joins your calls

Phone, desktop & Chrome apps

Your next meeting, summarized like this

TwinMind listens in the background and turns meetings, lectures, and calls into searchable notes you can chat with. Start free.

Available on phone, desktop & Chrome

Your next meeting, summarized like this

TwinMind listens in the background and turns meetings, lectures, and calls into searchable notes you can chat with. Start free.

Available on phone, desktop & Chrome

Your next meeting, summarized like this

TwinMind listens in the background and turns meetings, lectures, and calls into searchable notes you can chat with. Start free.

Available on phone, desktop & Chrome