Elected judges hand down harsher sentences as elections near. The United States has 1.5 million people in its prisons. (Eric Risberg/AP)

California voters last week recalled a judge for the first time in more than 85 years. The politics of punishment are already pathological; the recall will make them worse.

The particulars of the case by now ring familiar: Santa Clara County Superior Court Judge Aaron Persky sentenced Stanford student Brock Turner, convicted of sexually assaulting a woman on campus, to six months in jail (he served three), three years of probation (with the risk of 14 years in prison for a violation) and lifetime registration as a sex offender, along with costs and fees. Persky followed the sentencing recommendation made by the probation department, but many people viewed the sentence as too lenient, reflecting both a lack of respect for victims of sexual abuse and an attitude favoring the well-off and privileged. In response, Stanford law professor Michele Dauber launched the campaign that led to Persky’s removal.

As an academic who studies criminal justice, I have opposed the recall effort since I first heard about it because of potential consequences that reach well past Persky’s now-former courtroom: The recall will make judges more punitive, thwart progress toward scaling back mass incarceration and — though Turner and Persky are both white — hurt minorities disproportionately.

A central reason the United States punishes its citizens more than any other country is that actors in our criminal justice system face more political pressure than they do elsewhere. Only this country allows judges to be elected, which 39 states choose to do. It’s a consistent theme: We are also the only country that elects its prosecutors. While a concern in the Andrew Jackson era about corrupt appointment processes drove the decision to elect judges, more recent concerns about the costs of a politicized judiciary have led to increasing calls to return to appointing them.

In criminal justice, the costs of politicization are unambiguous: They make judges more punitive. The empirical studies on judges and crime tell a consistent story. Judges sentence more aggressively as their election dates near and as their elections become more contested. Elections make judges nervous, and nervous judges are harsh judges.

This harshness is entirely logical. Judges are harsh because the costs of mistakes are asymmetric. There is little downside to harsh sanctions, because the error costs are invisible: How do you show that someone would not have reoffended had they left prison sooner? The costs of being overly lenient, however, are inescapable. That sort of failure produces an identifiable victim for political opponents to capitalize upon.

The recall turned on a slightly different asymmetry but one that equally pushes judges toward severity. An overly lenient sentence will be seen as insulting the victim, while an overly harsh one will be seen as unfair to the defendant. The former error, as the Persky recall demonstrates, is costlier (unless, perhaps, the defendant is politically powerful).

Defenders of the recall dismiss this concern by pointing out that recalls are rare. But the lesson here isn’t only about recalls. The Persky case makes clear to judges and their detractors alike that judges can lose their jobs — in a recall, in a primary, in a general election — if just one or two decisions anger someone with sufficient political capital to oppose them. The Persky recall campaign highlighted only five decisions out of thousands that the judge handed down. Persky was cleared of any wrongdoing by California’s Commission on Judicial Performance, and public defenders in Santa Clara were quick to argue that he was a fair judge. Even the prosecutor in Santa Clara opposed the recall. Dauber, however, is a politically well-connected professor at a nationally acclaimed law school with strong media ties. The success of her campaign tells judges, and the politically powerful who are unhappy with their decisions, that these campaigns can work even with little evidence, as long as there are one or two bad cases to point to.

The recall’s political costs are already apparent. Not only did Democratic legislators pass new mandatory minimum sentences for sex offenses in response to the recall to make sure they looked tough enough on crime, but public defenders in California also report that judges seem harsher now, out of fear of being targeted next.

Some defenders of the recall concede that it may make judges harsher, but only regarding sex crimes. The judges, they say, are smart enough to limit what they have learned to the facts of the recall. But this is overly optimistic. Judges have no idea what issue will trigger the next recall or primary challenge, only that such campaigns can work.

Take the infamous case of William “Willie” Horton. Horton was serving time for homicide in Massachusetts when he was allowed to go home under the state’s furlough program. He instead absconded to Maryland, where he committed a vicious assault and rape. His case was later used in a flagrantly racist television ad that attacked Massachusetts Gov. Michael Dukakis (D) during his 1988 campaign for president. But here’s the thing: The Horton ad didn’t just discourage states from adopting furlough programs or from allowing those convicted of homicide to participate. It encouraged elected officials to be punitive across the board — even if the ad’s reach was limited, and even though the furlough program had a 99 percent success rate. No one knows what the next “Willie Horton” will be. A single scary outlier anecdote always reaches far beyond its narrow set of facts.

One final cost of the recall might be the most consequential. It won’t take much of a change in judicial toughness to completely halt the nascent efforts to scale back mass incarceration in this country. After 40 years of unceasing growth, the U.S. prison population — the highest in the world at just over 1.5 million  — has, since 2010, started to slowly decline, by about 110,000. This is good, as the evidence is clear that the costs of incarceration dwarf its public-safety benefits and that there are other less expensive and less harmful ways to keep people safe.

But simply put, even a small change in how we treat just sex offenders could bring that decline to a halt. We admit about 30,000 people to prison for serious sex offenses each year, and they serve an average of about 5½ years. If the political fears stoked by the Persky recall lead judges to send, say, 5,000 more people to prison for such crimes and to make them serve an average of 6½ years instead of 5½, then the total number of people in prison for sex offenses will rise by about 62,000 — offsetting nearly two-thirds of the 110,000-person decline and effectively ending the overall drop.

And that’s only sex offenses. What if judges also become slightly more aggressive against aggravated assaults (65,000 admissions, with sentences averaging two years) and robberies (48,000 admissions, with average sentences of four years)? Our decline in incarcerations is already slow and tentative and halting, and it won’t take much pressure to end it.

The recall will make judges more aggressive, and in ways that will never be neatly confined to the issues in the Turner and Persky cases. More people will be sent to prison, and that increase won’t make us safer. And since a majority of people in prison are black or Hispanic, the impact of this toughness will fall disproportionately on minorities. For those hoping to see the United States become a less punitive place, the recall’s success is disappointing.