Essays on compensation, incentives, and representation




Lee, Andrew Ji-Hoon

Journal Title

Journal ISSN

Volume Title



This dissertation consists of three essays on compensation, incentives, and representation in the United States criminal justice system. The first two chapters study flat fee compensation structures for publicly financed defense lawyers, their implications for case outcomes of indigent defendants and lawyer effort, and their role in distorting incentives away from effective legal representation in the spirit of the Sixth Amendment. The third chapter examines whether pre-arrest diversion programs impact later-life outcomes, such as recidivism, among first-time misdemeanor offenders.

In Chapter 1, I empirically examine whether compensation structures for court-appointed attorneys impact case outcomes of indigent defendants. I study this question in the context of a natural experiment in North Carolina, in which six counties were arbitrarily chosen to change compensation for assigned counsel from a schedule of statewide hourly rates to a flat fee per case disposed. Linking detailed administrative on attorney pay records with court records for the universe of indigent criminal cases in the state, I find that moving from hourly to flat fee pay resulted in a higher incidence of adverse case outcomes: defendants were 11% more likely to be convicted and 37% more likely to be incarcerated, a result driven by an increase in guilty pleas.

In Chapter 2, I investigate two potential mechanisms through which compensation structures for court-appointed attorneys may impact case outcomes of indigent defendants. Examining the intensive margin response on effort, I find evidence that lawyers reduce their effort under flat fee pay: lawyers spend 11% fewer hours on each case, dispose of cases 25% sooner, and are 36% more likely to dispose of a case on the same day as their first meeting with the defendant. Second, I examine selection of lawyers on the extensive margin, but do not find evidence that moving to flat fee pay significantly impacted lawyer composition in the treated counties.

In Chapter 3, I turn my attention to the long-term impacts of criminal records, which can create barriers to employment, housing, and higher education throughout an individual's life. I focus on the pre-arrest misdemeanor diversion program (MDP) in Durham County, NC, which has been offered to first-time offenders from 2014 to the present day. Key features of the program include rehabilitative treatment (such as mental health counseling) in lieu of arrest and prosecution, a clean criminal record for participants upon successful completion, and law enforcement officer discretion in diverting first time offenders as an alternative to arrest. I propose two empirical strategies to study the impact of pre-arrest diversion on later-life outcomes, particularly future criminal behavior, of first-time offenders. First, I propose a regression discontinuity approach that exploits cutoffs at (1) the start of the program in March 2014 for 16-17 year old offenders and (2) an expansion of the program to 18-21 year old offenders in October 2015. Second, I propose an instrumental variables design that leverages as-if random assignment of offenders to law enforcement officers, who vary in their probability of diverting first-time offenders in lieu of arrest.



LCSH Subject Headings