:format(webp)/cdn.vox-cdn.com/uploads/chorus_image/image/62047535/GettyImages-458406740.0.jpg)
Florida voters will soon decide whether to return the right to vote to almost 1.5 million ex-felons.
The proposed amendment, known as Amendment 4, has led some established election analysts, such as Nate Cohn of the New York Times, to speculate about the partisan consequences of such reform. Had ex-felons been allowed to participate in 2016, the story goes, their votes would have wiped out President Trump’s 113,000-vote victory in Florida. The implication is that a successful amendment would dramatically benefit Democrats throughout the state.
There are many reasons to support restoring the right to vote to those citizens who have completed their sentence. But Democrats should not do so because of perceived partisan gain. Neither should Republicans fear a “blue wave” by supporting the initiative. As political scientists who study election administration and have written extensively on felon disenfranchisement, our analysis of Florida’s previous attempt at a similar, but smaller, reform suggests that a blue wave is unlikely.
Florida’s disenfranchisement policy, which permanently strips the right to vote from anyone who has been convicted of a felony, is one of the harshest voting laws in the country. It first came under scrutiny when more than 600,000 ex-felons were barred from participating in the state’s contested 2000 election. The fact that the race was ultimately decided by fewer than 600 votes sparked a debate about the extent to which George W. Bush — and other Republican candidates — electorally benefit from this democratic exclusion.
But the state did restore the vote to a small group of ex-felons from 2007 to 2011 under then-governor (and then-Republican) Charlie Crist. (Florida vests the governor with the exclusive power to restore the vote through an executive clemency process.) Crist restored the right to vote to about 150,000 ex-felons convicted of less serious offenses. Although Republican Gov. Rick Scott rescinded the policy in 2011, these 150,000 ex-felons remain eligible to vote.
By combining information contained in public records (for more information on our process, see our previous work), we can assess how often these 150,000 individuals vote and which party they register with. We believe this offers the best evidence about how a massive restoration of ex-felons’ voting rights would have reshaped Florida’s 2016 election — and what this would mean for the partisan balance of the state. Our analysis suggests that Amendment 4 is less about swinging the state than about who has the right to participate in our democracy.
Ex-felons vote at low rates. And when they do, there is no strong partisan lean.
One thing that limits the electoral impact of restoring ex-felons’ voting rights is that they turn out at particularly low rates. To demonstrate this, we first look to the population of ex-felons who were restored the right to vote under Crist and calculate what share voted in 2016. We find that just 16 percent of black and 12 percent of nonblack ex-felons voted. (We defined nonblack as white, Hispanic, Asian, and other due to data limitations.)
:format(webp):no_upscale()/cdn.vox-cdn.com/uploads/chorus_asset/file/13373025/Artboard_2_80__1_.jpg)
The potential electoral impact of Florida’s ballot initiative is further limited by the fact that the ex-felons who do vote are not politically uniform. While black voters within this population overwhelmingly register with the Democratic Party (87 percent), nonblack voters within this population were more likely to register as Republicans (40 percent) than as a Democrats (34 percent). The fact that 26 percent of the remaining nonblack voters affiliate with neither party suggests that their votes may not reliably be cast for either party.
:format(webp):no_upscale()/cdn.vox-cdn.com/uploads/chorus_asset/file/13374303/Artboard_2_copy_80.jpg)
According to the Sentencing Project, there were 1,487,847 ex-felons in Florida who were unable to vote during the 2016 election, of which about a quarter, or 418,224, were black. Even though black ex-felons are overwhelmingly Democratic, the racial breakdown reveals that there are many more nonblack ex-felons, whose political preferences are much less certain.
Although people do not always vote consistent with their party of registration, we can approximate the electoral impact by assuming that they would. Specifically, we multiply the estimated disenfranchised population first by the turnout rate and then by the party registration rate for both black and nonblack individuals. Had all ex-felons been eligible to vote in Florida in 2016, we estimate that this would have generated about 102,000 additional votes for Democrats and about 54,000 additional votes for Republicans, with about an additional 40,000 votes that could be cast on behalf of either party.
The question of restoring ex-felon voting rights is about democracy, not swinging the election
The partisan framing of criminal disenfranchisement is not without basis. The partisan ramifications, though, depend on the racial demographics of the ex-felon population, which varies widely state by state. Florida’s Democrats would stand to gain some votes by extending the franchise, but the net gain of 48,000 votes is only about one-quarter of 1 percent of the more than 15 million people of voting age in Florida. Such evidence cannot support the stronger claim that Trump would have lost the state in 2016 but for criminal disenfranchisement.
Our analysis is far from perfect, but these limitations shouldn’t affect our conclusion, at least in the near term. For one, even if all ex-felons who register with neither major party vote for Democrats, the net Democratic advantage would still be less than Trump’s margin of victory. For another, even if the nonpartisan Florida Rights Restoration Coalition, the main advocacy group behind the amendment, turns its efforts to a unique registration and turnout drive, ex-felon turnout — among blacks, or among blacks and nonblacks — would need to more than double to make up Trump’s margin of victory.
Our past work suggests it would be difficult at best to bring up turnout from the teens to more than 30 percent. We’ve published estimates of ex-felon turnout in the 2008 and 2012 presidential elections in Iowa, Maine, New Mexico, New York, North Carolina, and Rhode Island and found it ranging from about 8 percent to almost 14 percent.
Advocacy groups could have boosted ex-felon turnout in Virginia after a recent mass restoration of voting rights in the state. But this is difficult to evaluate because ex-felons were restored voting rights on a rolling basis, with the governor prioritizing those who had registered before and thus were more likely to vote. Still, the limited data available on the reported participation of the broader ex-felon population in the state’s 2017 election is only slightly above what we observed in Florida.
Amendment 4 is one of the most important questions any American will vote on in the midterm. If successful, it would be a watershed moment for our democracy, restoring the vote to almost half of all disenfranchised ex-felons in the country. We recognize, though, that many people do not just think about voting rights in these normative terms but, perhaps increasingly so, in partisan ones.
For these people, in this era of “voting wars,” our analysis suggests the normative victory wouldn’t come at much of a partisan cost. Voters should consider the amendment a question of who gets to participate in our democracy instead of a question of the partisan realignment of the state — after all, this is unlikely to happen.
Marc Meredith (@mieuque) is an associate professor of political science at the University of Pennsylvania. Michael Morse (@MichaelLMorse) is a JD candidate at Yale Law School and a PhD candidate in Harvard’s department of government.
The Big Idea is Vox’s home for smart discussion of the most important issues and ideas in politics, science, and culture — typically by outside contributors. If you have an idea for a piece, pitch us at thebigidea@vox.com.
