Association for Public Policy Analysis & Management

In The New Jim Crow: Mass Incarceration in the Age of Colorblindness, lawyer and civil rights activist Michelle Alexander develops a compelling argument that mass incarceration of young Black men constitutes an extension of “racial control under changing disguise.” Racial control is itself an extension of slavery and, later, the Jim Crow laws of the first half of the last century—a desire on the part of the white power structure to maintain a racial caste system and suppress minority political power, especially in the realm of voting. In the United States, not only are incarcerated persons denied the right to vote, but many disenfranchisement laws also apply to parolees or former felons. Moreover, many of the formerly imprisoned face legal discrimination in housing, education, employment, and public benefits.

As intriguing as this research is, no one has yet made a direct empirical connection between the enforcement of voting rights, via the Civil Rights Acts and Voting Rights Acts of the 1960s, and the response of local law enforcement—the first line for creating a permanent underclass of current and former prisoners. In this work, I examine local arrest data from 1960 to 1980, and estimate how arrest behavior changed in jurisdictions that fell under Section 5 of the Voting Rights Act of 1965. Section 5 requires that jurisdictions receive federal approval before implementing changes to election laws. Because enforcement of Section 5 rolled out in certain U.S. counties between 1965 and 1973, I am able to use the roll-out as a source of identification in the effect of the law on local law enforcement. In a difference-in-differences analysis of the law, I find that the rates of Part I arrests for Black persons increased in jurisdictions within counties where Section 5 was applied compared with other jurisdictions. Because jurisdictions fall under Section 5 based on a formula that includes a 50% cutoff for voter registration, I also use a regression discontinuity approach, allowing me to compare covered jurisdictions to those that just miss the cutoff. Finally, using more recent data, I examine whether arrests increase in the months before an election day. The findings of this work have implications for the development of policies regarding civil protections for current and former felons.