Loading Events
  • This event has passed.

Modibo SIDIBÉ (Duke University) – "Police Officer Assignment Mechanisms and Neighborhood Crime"

May 11, 2021 @ 3:00 pm - 4:15 pm
Microeconometrics Seminar: Every Tuesday
Time: 03:00 pm – 4:15 pm
Date: 11th of May 2021
by visio
Modibo SIDIBÉ (Duke University) – “Police Officer Assignment Mechanisms and Neighborhood Crime 

Abstract: In most major American cities, police officers are assigned to neighborhoods through a bidding process that grants priority based on seniority. Such seniority-based mechanisms typically give rise to an equilibrium allocation in which experienced officers are more likely to be assigned to neighborhoods with lower (violent) crime rates, raising important efficiency and equity concerns. In this paper, we estimate a model of the bidding process used to assign police officers to neighborhoods in Chicago and use it to study the economic implications of the current and alternative assignment mechanisms. We estimate the model using detailed police personnel records from 2004-2015, which provide a continuous record of the assignment of officers to police districts, including the winning “bidder” whenever a vacancy is posted in each district. The estimated model reveals clear officer preference for police districts with lower rates of violent crime. To identify the causal impact of officer seniority on neighborhood crime, we use a simulated IV strategy that uses the estimated model of the bid process to isolate a component of the evolution of the composition of each police district that is a function of only the initial conditions in each district and aggregate changes to the composition of the police force and/or crime rates in the city as a whole over the study period. The estimates imply a significant negative impact of officer seniority on neighborhood crime, but at a magnitude that is only about 25 percent of what OLS estimates would imply. A comparison of the current assignment to counterfactual simulations in which district-specific financial bonuses are used to induce more senior officers to remain in districts with higher (violent) crime rates implies that equalizing officer seniority across neighborhoods would reduce the overall violent crime rate in Chicago by 5 percent and substantially reduce the inequality in crime rates across neighborhoods.

Organizers:

Benoît SCHMUTZ (Pôle d’économie du CREST)
Anthony STRITTMATTER (Pôle d’économie du CREST)
Sponsors:
CREST

Microeconometrics Seminar: Every Tuesday
Time: 03:00 pm – 4:15 pm
Date: 11th of May 2021
by visio
Modibo SIDIBÉ (Duke University) – “Police Officer Assignment Mechanisms and Neighborhood Crime 

Abstract: In most major American cities, police officers are assigned to neighborhoods through a bidding process that grants priority based on seniority. Such seniority-based mechanisms typically give rise to an equilibrium allocation in which experienced officers are more likely to be assigned to neighborhoods with lower (violent) crime rates, raising important efficiency and equity concerns. In this paper, we estimate a model of the bidding process used to assign police officers to neighborhoods in Chicago and use it to study the economic implications of the current and alternative assignment mechanisms. We estimate the model using detailed police personnel records from 2004-2015, which provide a continuous record of the assignment of officers to police districts, including the winning “bidder” whenever a vacancy is posted in each district. The estimated model reveals clear officer preference for police districts with lower rates of violent crime. To identify the causal impact of officer seniority on neighborhood crime, we use a simulated IV strategy that uses the estimated model of the bid process to isolate a component of the evolution of the composition of each police district that is a function of only the initial conditions in each district and aggregate changes to the composition of the police force and/or crime rates in the city as a whole over the study period. The estimates imply a significant negative impact of officer seniority on neighborhood crime, but at a magnitude that is only about 25 percent of what OLS estimates would imply. A comparison of the current assignment to counterfactual simulations in which district-specific financial bonuses are used to induce more senior officers to remain in districts with higher (violent) crime rates implies that equalizing officer seniority across neighborhoods would reduce the overall violent crime rate in Chicago by 5 percent and substantially reduce the inequality in crime rates across neighborhoods.

Organizers:

Benoît SCHMUTZ (Pôle d’économie du CREST)
Anthony STRITTMATTER (Pôle d’économie du CREST)
Sponsors:
CREST