What the San Francisco DA Recall Really Tells Us

What the San Francisco DA Recall Really Tells Us

And what it doesn’t.

John Pfaff — Read time: 11 minutes

This story was produced in partnership with the Garrison Project, an independent, nonpartisan organization addressing the crisis of mass incarceration and policing.

The June 7 election to recall San Francisco DA Chesa Boudin was clearly going to be interpreted as a national referendum on the idea of prosecutorial reform. Unsurprisingly, then, the coverage following the successful effort to unseat Boudin, a progressive prosecutor elected in 2019, has pushed the idea that the recall highlights a fundamental weakness with efforts to reform prosecutors’ offices across the country.

While this reads far too much into one fairly idiosyncratic election, we should not dismiss the recall as just some sort of idiosyncratic election. If we view it less as an overarching metaphor for reform and more as a single but salient data point, the Boudin recall provides a useful way to talk about what we know about the politics of and potential for progressive prosecution, and what we do not.

Sign Up for the Surge

The most important political nonsense of the week, delivered to your inbox every Saturday.

To start, though, it is worth highlighting the idiosyncratic aspects of this election. Boudin clearly struggled as a politician, including at one point saying that a person had committed murder during what appeared to be a “temper tantrum.” And unlike normal elections, recalls do not pit two candidates against each other, and thus may reflect people’s views of the person more than their policies (and evidence suggests San Franciscans supported Boudin’s policies more than Boudin himself). Not to mention that it is risky to draw big conclusions from low-turnout elections, something even those pushing a bigger narrative concede. And San Francisco voters were wary of Boudin from the start: By the end of the city’s ranked choice voting process in 2019, he barely won, edging out the much more moderate Suzy Loftus 50.8 percent to 49.2 percent.

At the same time, there are some national lessons we can learn from his loss.

For example, one aspect of reformist prosecution that has received little to no rigorous study is the significance of whether the candidate is an “insider” or an “outsider.” Boudin, like Larry Krasner in Philadelphia, ran for DA as an outsider to—and often former antagonist of—the office. Other reformers, like Parisa Dehghani-Tafti in Arlington, Virginia, Buta Biberaj in Loudoun, Virginia, and Satana Deberry in Durham, North Carolina, came from similar defense-oriented backgrounds.*

A significant number of progressive prosecutors have extensive careers on the inside: Brooklyn DA Eric Gonzalez has been a prosecutor in that office since the day he graduated law school. Cook County State’s Attorney Kim Foxx served over a decade as an assistant prosecutor, while former Suffolk County (Boston) DA Rachael Rollins and Steve Descano in Fairfax, Virginia, were both former federal prosecutors. George Gascón in Los Angeles was not only Boudin’s predecessor in San Francisco, but also served as a deputy chief of the LAPD and the chief of two other police departments.

It’s worth asking if a reformer’s status as an insider or outsider not only affects what they can accomplish but also the political pushback they may face. Do insiders inherit more loyalty, or are they more inclined toward incremental change? Are outsiders more politically vulnerable because it is easier to paint them as “caring too much about the criminals,” or because they tend to make certain kinds of problematic rhetorical moves that insiders instinctively do not? Do insiders receive more favorable coverage because of their backgrounds, or because of the media connections they make over their earlier career? (Of course, the potential recall of Gascón indicates that insiders can be vulnerable too, which further highlights how little we understand the politics here.)

We do not have answers to these questions yet. But rather than trying to determine what the recall means, it is likely far more useful—if difficult—to think about how it fits in with all the other examples of progressive prosecution we have so we can better understand the complicated politics of punishment.

Another way in which the recall could be a useful data point is with San Francisco itself, and its demographics. One major reason Boudin’s recall is seen as a national referendum on progressive prosecution is that San Francisco is seen as one of the most progressive cities in the U.S. So if a progressive can’t win there, where can a progressive succeed?

The underlying assumption of that frame is likely an overstatement. Voting for Democrats—which San Francisco does do—does not make a place progressive, and carceral Democratic mayors are the norm in cities including Chicago, D.C., New York, and San Francisco itself. The post-tech San Francisco is much less the countercultural bastion it was decades ago.

But even putting that critically important point aside, San Francisco looks quite different from most cities with progressive prosecutors. The push for reformist prosecutors—and criminal legal reform more broadly—is inextricably intertwined with the politics of race in the U.S. because the burdens of both crime and punitiveness fall on communities of color, Black communities in particular. It should not be surprising, then, that the counties with progressive prosecutors tend to have larger Black populations: Philadelphia (44 percent), Brooklyn (34 percent), Baltimore City (62 percent), Cook County (24 percent, and 30 percent in Chicago itself, where Foxx consistently performs better), St. Louis city (46 percent), St. Louis County (25 percent), Dallas County (24 percent), and Durham County (37 percent). San Francisco, at roughly 6 percent Black, is an outlier here.

These demographic differences could help explain why Boudin’s campaign was more vulnerable to fearmongering than Krasner’s—and, in doing so, perhaps also highlight a way in which the term “progressive prosecutor” may obfuscate as much as it clarifies.

A recent Pew Research Center study explains that the “progressive left” refers to a group that is distinctly non-Hispanic white (68 percent) and highly educated (nearly 50 percent have a four-year college degree). By referring to reform-minded prosecutors as “progressive,” we are effectively aligning them with this “progressive left”—that the electoral fate of reform-minded prosecutors turns on the support of these disproportionately white, well-educated voters.

And while it is true that the progressive left tends to favor reformers, we should also be concerned that when it comes to criminal legal reform, their support may be fickle, and that we should be concerned that white progressives likely often act like the “white moderates” Martin Luther King Jr. criticized in his “Letter From Birmingham Jail,” who are “more devoted to ‘order’ than to justice.”

It is increasingly clear that a core source of support for reform prosecutors comes from poorer, Blacker communities that bear the brunt of both violence and punitiveness. Preliminary results I have been working on—inspired by these maps comparing where Krasner won to shootings in Philadelphia—suggest that in elections across a wide number of cities, reform prosecutors consistently win the voting precincts where violence is most concentrated. Notably, these communities are less vulnerable to fearmongering because violence for them is far less an abstraction and far more a lived experience.

That is not to say that progressives cannot win in less Black counties. The same night that Boudin was recalled, progressive Kimberly Graham won the Democratic primary for county attorney (and thus likely the general election) in Polk County, Iowa, home to Des Moines and 77 percent non-Hispanic white and only 7 percent Black. But perhaps the Boudin recall—contextualized against events like Krasner’s and Foxx’s easy-ish reelection efforts, and alongside Graham’s win—can help us begin to understand the underlying factors that determine the sorts of political vulnerabilities candidates face.

When thinking about the Boudin recall, though, we must also consider how much we simply cannot know at this point.

For one, the push for reformist prosecutors is new. The term “progressive prosecutor” has many different definitions, so it is impossible to say exactly who the “first” reform DA was, but movement didn’t really start until 2015 or 2016, with the elections of people like James Stewart in Caddo Parish, Louisiana; Scott Colom in rural Mississippi; Kim Foxx in Cook County; Kim Ogg in Harris County, Texas; and Krasner in 2017 in Philadelphia. The movement is only about six or seven years old—or perhaps nine if we start with Marilyn Mosby’s election in Baltimore City in 2013.

Contested prosecutor races were almost unheard-of until the push for progressive prosecution. In a paper published in 2009, right before the reform movement really took off, Wake Forest’s Ronald Wright looked at data from 10 states between 1996 and 2006 and found that incumbent candidates ran unopposed in 85 percent of all primaries and of all general elections.*

When we put the Boudin recall in its broader but quite recent electoral context, it becomes even clearer how hard it currently is to determine any sort of pattern. Yes, the recall was a setback. And it wasn’t the only setback for reformers: Progressive candidates fell short in primaries in smaller, whiter suburbs like Marion County, Ohio, and Washington County, Oregon, as well as in Mecklenburg County, North Carolina, the home of Charlotte.* And in San Joaquin County, California, Tori Salazar may have become the first incumbent progressive to run for reelection and lose, albeit in a race with stunningly low (about 5 percent) turnout.

But there were victories for reformers over the past months as well. In Contra Costa, California, incumbent reformer Diana Becton held on against a more traditional challenger. Across from San Francisco, in Alameda County, reformer Pamela Price came in first and will face a runoff in November (although it’s worth noting there that the two more traditional candidates collectively received just over 51 percent of the vote). In Durham, North Carolina, incumbent progressive Satana Deberry breezed to victory with 80 percent of the vote in her primary.

We’ve seen these sorts of contradictory results before. In 2018, the defeat of three progressive candidates in California—including the heading-for-a-runoff Price—was taken as a sign of a broader shift away from reform, only to have three progressive candidates win strongly contested races in northern Virginia just one year later.

Moreover, not all races are equally important, given how concentrated people, and thus crime, and thus prosecutions, are. A Bureau of Justice Statistics study from 2007—sadly, the last year with data on the issue—found that only 10 percent of all prosecutor offices processed about two-thirds of the nation’s felonies. Or, put differently, the four biggest counties with progressive prosecutors—Los Angeles, Cook, Brooklyn (Kings), and Dallas—have about as many people as the 684 counties that make up the 12 smallest states with elected district attorneys. That means the overall impact of progressive prosecution will often turn on a very small number of elections, making prediction even harder.

There’s a broader point here about criminal legal reform worth noting. Perhaps the most important recent book on the politics of reform is Breaking the Pendulum, by Philip Goodman, Joshua Page, and Michelle Phelps. They argue that what looks like a somewhat predictable oscillation from punitiveness to progressiveness and back only looks predictable in hindsight. They demonstrate that there are never stable periods of punitiveness or progressiveness; at all times, the “out” group is trying to push through the changes it desires. And, more often than not, the forces that push the out group in and the in group out are broader social or cultural upheavals perhaps only loosely linked to crime.

This explains, for example, why the most recent successful reform efforts took off in the late 2000s—not because of crime, but as a result of political realignments brought on by the financial crisis. Importantly, the shocks that actually matter can be quite difficult to predict. Consider that we are in the midst of a period of upheaval, with the pendulum swinging wildly following the murder of George Floyd and the resulting protests, and then back again amid the stress of the pandemic and spike in homicides. What Goodman, Page, and Phelps show is that, while it may be hard to impossible to predict the sort of shocks that will matter, the ones that matter matter because activists and advocates are ready and able to exploit them.

If the Boudin recall teaches us anything, it’s not that voters are too scared to be interested in reforming the way we prosecute crime and other quality of life issues. Instead of attempting to put a fine point on what it all means, the recall outcome should help us clarify what we need to study. There’s no need to try to force a narrative onto an unprecedented, highly dynamic moment.

Correction, June 10, 2022: This piece originally misspelled Loudoun, Wake Forest, and Mecklenburg.