Economic Conditions and Crime

1
Economic Conditions and Crime

• Naci Mocan
• Department of Economics
• Louisiana State University
• and
• National Bureau of Economic Research

2
Economic Conditions and Crime

• Unemployment and Crime
• A 1 percentage point increase in the
  unemployment rate (e.g. from 7 to 8) increases
  Property Crime by 2.2.
• A 2 increase in wages of low-skilled workers
  decreases crime by 1.
• The same is true for the real (inflation
  adjusted) minimum wage

3
Economic Conditions and Crime

• A typical business cycle in the U.S.
• Average contraction (recession) is 10 months, but
  average expansion (recovery) takes 52 months
• A 1 ppt. increase in the unemployment rate takes
  7 months BUT a 1 ppt. decline in the
  unemployment rate takes 17 months.

4
Asymmetry

• New research shows that Crime is impacted MORE
  from a RISE in unemployment than from a DECLINE
  in unemployment.
• (The impact of unemployment on crime is
  asymmetric).

5
Sharp Rise-Gradual Decline
6

• NEW RESEARCH shows that
• Economic uncertainty (about the future) has an
  impact on crime
• Social interactions and peer effects have an
  impact on crime

7
Deterrence

• We also know that deterrence has a robust impact
  on crime
• If the arrest rate of an index crime goes up by
  10, the corresponding crime rate goes down by
  2-6.
• If the number of uniformed police officers goes
  up by 10, some property crimes go down by 6-7.

8
Broken Windows

• Broken Windows Policing Works
• In NYC- Mayor Giuliani and Commissioner Brattons
  implementation of broken-windows policing
  Increase in misdemeanor arrests (disorderly
  conduct, public drunkenness, vandalism, etc.)
• A 10 increase in misdemeanor arrests reduce MV
  thefts by 2, robberies by 3, grand larcenies by
  0.5.

9

• Why crime rates differ between states or cities
• Economic conditions differ
• In 2005, Male UR with Educationlt12 yrs.
• Average of all states 13
• AL 5, GA7, PA 9, OH11, MN13 MS 16
  IN 18, SD24, ND29, MT34

10

• Louisiana in 2005
• Overall Unemployment Rate in
• East Carroll Parish 12.4
• West Carroll Parish 10.2
• East Feliciana Parish 6.4
• West Feliciana Parish 7.6

11

• Demographic composition differs
• ( young, poor, educated, single parent,
  gun ownership rates, drug use, etc.)
• Police resources and enforcement differ

12

• Why criminal activity differs between
  neighborhoods within the same city
• Same reasons as before
• Plus
• Social Multiplier (spillovers, peer effects)

13

• If my criminal activity influences my peers
  criminal proclivity, aggregate crime will exhibit
  differential spatial persistence depending on the
  intensity of local social interactions
• Social networks (stronger in segregated
  neighborhoods)

14
Do We Know Enough?

• We dont know everything
• We dont know a lot
• We know a lot to implement good policy
• (police force, arrests, unemployment, wages)
• More research with better data will provide
  additional insights

15

• Crime is an individual behavior
• Should we analyze data on individuals or
  aggregates (crime rates)?
• Both are important
• Need better (focused, integated) data.

16

• Real World Experiments can be designed with data
  collection
• criminal activity, risk aversion, peer
  interactions, contextual variables
• Merge them with labor market, police and school
  data
• Precedent exists (Add Health data)
• UR needs to be measured MORE finely.