Locked Out
Felon Disenfranchisement and American Democracy
By Jeff Manza and Christopher Uggen
Oxford University Press
New York, 2006
This sustained examination of American policies depriving felons of the right to vote, and the consequences of those policies, ends with these words from a South African Constitutional Court decision allowing felons the vote:
The universality of the franchise is important not only for nationhood and democracy. The vote of each and every citizen is a badge of dignity. Quite literally, it says that everybody counts. In a country of great disparities of wealth and power, exalted or disgraced, we all belong to the same democratic South African nation; that our destinies are intertwined in a single interactive polity. Right may not be limited without justification and legislation dealing with the franchise must be interpreted in favor of enfranchisement rather than disenfranchisement.
"No clearer statement of the argument of this book," the book's authors conclude, "can be found." Locked Out is a sustained case for democracy, for dignity, for citizen participation, and for equality. Jeff Manza, a sociologist at Northwestern University, and Christopher Uggen, a sociologist at the University of Minnesota who blogs on crime issues at Public Criminology, build this case through detailed consideration of the origins of felon disenfranchisement laws, who they prevent from voting, and the impacts of that disenfranchisement.
Despite its important historical role in the spread of representative democracy, by contrast with other democracies, the United States has exceedingly restrictive laws on felon voting. Many countries do not prevent voting even by current inmates, while very few have any restrictions on voting by those felons who have served their sentences and been released; 13 US states impose a lifetime ban on voting except for those felons who successfully petition to have their rights restored (which can be a difficult and time-consuming process, and one many may not know is available).
Uggen and Manza estimate that on Election Day in 2004, 5.3 million people were disenfranchised due to felony convictions. Still more were functionally disenfranchised because they were imprisoned without felony convictions (minor offenses and pretrial detention) or because they did not realize that they had regained their eligibility to vote. The vast majority of the 5.3 million were not in prison on Election Day, but were on probation, parole, or had completely fulfilled their sentences. Only two states - Vermont and Maine - have no restrictions on felon voting. In six states, more than 5% of the voting age population is ineligible, with a high of nearly 10% in Florida.
The picture regarding African Americans is especially troubling, and is a foundational lens through which felon disenfranchisement must be viewed; the authors examine the origins of disenfranchisement laws and conclude that:
When African Americans make up a larger proportion of a state's prison population, that state is significantly more likely to adopt or extend felon disenfranchisement. To be sure, many states adopted an initial felon disenfranchisement measure after property and other restrictions on lower-class white men were lifted (and before blacks could vote). Individual state histories may be more complex than our race thesis implies. But the overall evidence we have presented supports a strong conclusion about the political significance of race in driving the adoption of felon disenfranchisement laws.
And to the extent that depriving African Americans of the right to vote has historically been an unacknowledged (and sometimes acknowledged) reason for depriving felons of the right to vote, it has worked. In 14 states, more than 10% of black people are disenfranchised, and in 5 states, the number is more than 20%.
If the simple notion that in a democracy, millions of adults should not be denied the right to vote is not enough to make you oppose these laws; if you do not believe that, having served their time, people deserve to be functioning citizens of this country; if the racial disparity introduced by, and probably an intended goal of, the laws is not enough - how about a partisan argument?
While Democratic candidate Al Gore won a plurality of the popular vote, defeating the Republican candidate George W. Bush by over 500,000 votes, he narrowly lost in the Electoral College. Had disenfranchised felons been permitted to vote, we estimate that Gore's national margin of victory in the popular vote would have surpassed 1 million votes. Regardless of the popular vote outcome, however, the outcome in Florida determined the electoral college winner. If disenfranchised felons in that state had been permitted to vote, Al Gore would certainly have carried the state, and thus the election. There are more disenfranchised felons in Florida than in any other state (approximately 827,000 in 2000). Had they participated at our estimated rate of Florida turnout (27.2 percent) and national Democratic preference (68.9 percent), Gore would have carried the state by 80,000 votes. But even if we make drastically more conservative assumptions, Gore would still win. For example, if we halved the estimated turnout rate and consider only the impact of allowing ex-felons to vote, Gore's margin of victory still would have exceeded 30,000 votes, more than enough to overwhelm Bush's narrow victory margin (and to reverse the outcome in the Electoral College). The outcome of the 2000 presidential race thus hinged on the narrower question of ex-felon disenfranchisement rather than on the broader question of voting restrictions on felons currently under supervision.
It would be deeply cynical, and not a little immoral, to want to allow people to vote simply because you believe they'd vote for your party. And of course, Republicans portray this as a prime motive any time a Democrat seeks to lift even a small part of the restrictions:
A New York Post editorial, for example, declared that a proposal by Senate Democrats to enfranchise ex-felons "appears to be angling, not for the votes of centrists, but for the votes of the most dedicated left-wing constituency in America: Criminals."
That may be their thinking, and the flip side of that claim may be an important reason Republicans push back hard every time an attempt is made to expand voting rights even slightly. But, as Manza and Uggen show in so many ways, as much of a tragedy as it has been for the US to have President Bush instead of President Gore, it is an equal tragedy that this came through depriving millions of people of a fundamental right of citizenship in a democratic country. And it is difficult to think that the blatant abuses of existing law, including the illegal disenfranchisement of so many, in Florida in 2000 are unrelated to the fact that Florida legally disenfranchises the most people of any state. It is a democracy compromised both within the law and outside of it; tragically, it is only an extreme and not an anomaly in this country.
Some notes on reading this book:
Locked Out is a serious and methodologically sophisticated work of academic sociology that has its challenging moments. It is also an important work of public sociology, a considered intervention in a political debate. In either category, it is well worth reading, but here are a few suggestions for avoiding rough spots and finding the points of most interest.
Like many social science books, the first chapter sets up several key concepts and surveys their history; Manza and Uggen dispense with this in relatively short order, but if you find the first chapter rough going, skim it for key concepts and move on.
Readers who are not well-versed in and/or interested in quantitative methodology - the details behind the statistics - should be prepared to skip some short sections in which statistical models are described. Aside from a few such sections, though, presentation of statistics is clear and non-technical. On the other hand, readers with an interest in statistics will no doubt enjoy the extensive appendices.
Chapter five attempts to answer the question of whether voting may reduce the likelihood of recidivism. It's an interesting question, but not one really answerable with existing data. This chapter is therefore 23 pages leading to the conclusion that "the foregoing results are suggestive, but insufficient to establish a causal link between voting and desistance from crime." One is glad they asked the question, but the general-interest reader can probably skip the process of finding that a definitive answer is not possible at present.