Attorney General Jeff Sessions speaks at the Association of State Criminal Investigative Agencies event in Scottsdale, Ariz., last week. (AP Photo/Ross D. Franklin)

In two speeches now, one on May 8 to the Gatlinburg Law Enforcement Training Conference, and then one Monday night to the National Association of Police Organizations (an association of organizations!), Attorney General Jeff Sessions has blamed the ACLU for the surge in Chicago homicides in 2016.

The source for this claim is a recent paper by Paul Cassell and Richard Fowles, two professors at the University of Utah. Cassell, it’s worth pointing out, is a former federal judge and defender of law-and-order policy. He recently represented (successfully) the Utah police officer who was facing criminal charges for shooting and killing Danielle Willard. (He also blogs at the Volokh Conspiracy, formerly hosted here at The Washington Post, and now hosted at Reason magazine.)

In their paper, Cassell and Fowles claim that murders spiked in Chicago because of an agreement between the city and the ACLU that requires police in the city to document each street stop they conduct. Note that the agreement does not prohibit the stops. It only asks that they be documented to assess whether police are violating the rights of Chicago residents. The aim is to determine if these stops are being done because of reasonable suspicion or if they’re motivated by unconstitutional racial or ethnic bias. The number of stops plummeted by 75 percent the following year — 2016, the same year that saw a 60 percent spike in homicides. Cassell and Fowles link the two statistics to come up with the “ACLU effect” that Sessions is touting:

“We conclude that, because of fewer stop and frisks in 2016, a conservative estimate is that approximately 236 additional homicides and 1115 additional shootings occurred that year. A reasonable estimate of the social costs associated with these additional homicides and shootings is about $1,500,000,000. And these costs are concentrated in Chicago’s African-American and Hispanic communities.”

The study has been criticized in a number of places, including by Salvador Rizzo here at The Washington Post, and by John Rappaport at Slate. The main criticism is that street stops in other cities have also fallen — in some places pretty dramatically — without a corresponding rise in homicides such as the one we’ve seen in Chicago. For example, Street stops in New York City fell from 685,000 in 2011 to just under 23,000 in 2015. Despite dire warnings from supporters of stop and frisk that the city was about to endure a bloodbath, the city’s homicide rate hit record lows.

In their paper, Cassell and Fowles dismiss what happened in New York as an “anomaly.” They suggest that the two cities may have seen different results because New York has more police per capita than Chicago.

Another problem in equating New York’s circumstances with Chicago’s is that the level of police power is different. New York has high levels of law enforcement. For example, if we look to the FBI’s Uniform Crime Reports for 2016, New York had 51,399 police employees compared to 13,135 for Chicago, which even when adjusted for population size, means that New York’s overall police force was, on a per capita basis, about 25% larger than Chicago’s. But this comparison fails to capture the true workload differences between the two cities. Using the same crime (homicide) as used by the Crime Lab study, New York had about 153 law enforcement employees for every homicide committed in the city, while Chicago had only about 17 employees for every homicide committed — about a 800% difference.

But this number is inflated by the NYPD’s substantial bureaucratic bloat — NYPD employs about 5,000 people who aren’t uniformed police officers. It seems unlikely that civilian employees working in, say, the NYPD human resources department are going to do much to deter crime. If we look at the number of actual police officers — Chicago actually has slightly more cops per capita than New York.

But New York is far from an anomaly. As Rizzo points out, cities such as Newark, Seattle and Philadelphia also dramatically decreased the number of stop and frisks and didn’t see a Chicago-like spike in homicides. Philadelphia, for example, entered into a consent decree in 2011 that reduced the number of stop and frisks. Two years later, homicides in the city hit a 45-year low. (The city did see a five-year high in homicides in 2017, but also saw an increase in gun arrests the same year. Meanwhile, all other violent crime declined.) Rappaport also notes that homicides did spike in other cities in 2016 at rates similar to Chicago, but those cities did not significantly alter their policies on stop and frisk. All of this suggests that the spike was driven by other factors.

The ACLU itself also responded to the Cassell-Fowles study. The group noted that the agreement with the city of Chicago was similar to a state law that took effect in 2016 requiring all police agencies in the state to create a record of street stops. None of the state’s other largest cities saw a significant increase in homicides.

Critics of the study also point out that Illinois has been reducing funding for a number of anti-violence initiatives over the past several years, some of which have been found to produce real results. The critics also note a possible phenomenon we’ve discussed at length here at The Watch: As minority communities lose trust in and respect for the police, they stop cooperating with law enforcement. Crime goes up. There’s good anecdotal evidence for this. Crime has spiked in places such as Baltimore, the St. Louis area, Cleveland and Chicago. Perhaps not coincidentally, the Justice Department (before Sessions took over) issued blistering reports for each of these cities showing widespread police abuse, little unaccountability for bad cops, rampant racial profiling, and constitutional violations so common that they’re customary. More recently, a study in Milwaukee found that after high-profile incidents of possible police abuse, 911 calls from black neighborhoods declined. The same researchers found a similar pattern after other high-profile police shootings in other parts of the country.

There’s ample reason for Chicago residents to distrust their police department. The city, of course, saw the release of video of the Laquan McDonald video. The city is still sorting out the repercussions from the revelation that for two decades some officers tortured people without consequence. The recent Justice Department report on the Chicago Police Department was jaw-dropping. A Chicago Reporter investigation found that between 2011 and 2016, the city paid out over $280 million to settle police brutality lawsuits. In 2015, the Guardian reported that city police took suspects to “black sites” where they were detained and beaten.

Over at the Volokh Conspiracy, Cassell has written an unsatisfying response to some of these criticisms. In response to the argument that the state’s decrease (or in some cases, elimination) of funding for anti-violence programs may have contributed to the homicide spike, Cassell writes, “For example, Rappaport suggests that a reduction in funding for a social program known as “Ceasefire” might be the cause. An analysis by the Chicago Crime Lab found that this explanation doesn’t fit, because the cessation in funding was well before the spike took place.”

It’s interesting that Cassell favorably cites the Chicago Crime Lab, because the same report is also dismissive of the theory that fewer police stops caused the murder spike.

As for CeaseFire, the report’s language isn’t outright dismissive; it only suggests that the timing isn’t conducive to studying cause and effect. And the funding wasn’t cut off “well before” the spike began; it was less than a year. Moreover, as Rappaport notes, a recent Northwestern study did find that CeaseFire (now called Cure Violence) “reduced shootings by 41 to 73 percent in seven Chicago communities.” The group itself also has some convincing data showing that after funding was cut, homicides rose dramatically in the communities where the group had previously been working.

Moreover, the Chicago Crime Lab report does address the funding cuts to community and anti-violence programs more generally. Far from concluding that the cuts had no impact, it instead explains that there isn’t enough data available to say one way or the other:

. . . the recent deterioration in the financial condition of the state of Illinois, and the resulting budget impasse, seems to have caused disruptions for many community organizations that rely on state funding. The cumulative effect of these service disruptions and their impact on gun violence in 2016, however, is difficult to gauge without more detailed data on their exact magnitude and timing.

As for the argument that public scandals and high-profile shootings such as that of Laquan McDonald may have caused Chicago’s black residents to be less trustful of police, Cassell writes:

. . . in any event, we collected monthly 9-1-1 call data from Chicago during the relevant time period. Directly contradicting Rappaport’s speculative hypothesis, the call data show a longterm downward trend—no sharp break around the time of the McDonald video release.

Of course, the argument that Chicago’s black residents have a fractured relationship with the city’s police doesn’t start and stop with the McDonald video. It’s about the torture scandal. It’s about the abuses documented in the Justice Department report. It’s about the hundreds of millions of dollars paid out for police brutality settlements. It is, in fact, precisely the sort of culmination of scandals, reports and day-to-day interactions that would result in the “longterm downward trend” Cassell describes. And according to Five Thirty Eight, the spike in both fatal and nonfatal shootings began just after the release of the McDonald video in November 2015, not after the implementation of the new policy in January 2016.

But the strongest evidence that Chicago residents — and black residents in particular — are losing trust in their police department is polling data. For example, here’s a Chicago Tribune editorial about a poll released in February 2016, the same year of the homicide spike that’s the subject of the Cassell-Fowles study.

Most Chicagoans — 70 percent of all respondents — said they don’t believe police officers treat all citizens fairly. Nearly as many — 64 percent — said cover-ups and a code of silence in the Chicago Police Department are widespread problems. … Only 20 percent of respondents said they believe city cops treat all citizens fairly.

Distrust is rampant citywide, but perceptions vary by race. Among whites, 53 percent said they don’t think the police treat everyone fairly, while one-third said they do. But in the African-American community, faith in the police to do their job well is practically nonexistent. Do Chicago police treat all citizens fairly? Among blacks, 85 percent said no, while just 6 percent said yes. Among Hispanics: 69 percent said no, while 23 percent said yes.

A separate 2016 Northwestern study found that among Chicago’s black residents who had been stopped and frisked, trust in police dropped by 45 percent.

For comparison, national polling done at about the same time put overall trust in police at 57 percent, and support among blacks at 30 percent.

Cassell also responds to the criticism that his study focuses only on one city, and that their conclusions about Chicago are contradicted by data from other cities.

[Rappaport] also notes that in 2016, other cities had homicide spikes, suggesting that homicides can increase for reasons apart from changes in police tactics. But our research tried to explain what was going on in one important city—Chicago—where our research tools and data could explore what was happening. Every city has its own story to tell, and we have tried to tell one chapter of Chicago’s.

But of course the intent of the paper is to prescribe stop and frisk as a national policy. That’s certainly how Sessions is using their study. And it’s how the authors themselves characterize its conclusions.

. . . these findings shed important light on the on-going national debate about stop and frisk policies. The fact that America’s “Second City” suffered so badly from a decline in stop and frisks suggests that the arguably contrary experience in New York City may be an anomaly. The costs of crime—and particularly gun crimes—are too significant to avoid considering every possible measure for reducing the toll. The evidence gathered here suggests that stop and frisk policies may be truly lifesaving measures that have to be considered as part of any effective law enforcement response to gun violence.

In other words, they’re suggesting that Chicago is the norm, New York is the anomaly, and that stop and frisk should be considered “part of any effective law enforcement response to gun violence.” (Emphasis added.)

In closing, Cassell points out that he isn’t trying to “scapegoat anyone.” But of course he is. In specifically naming the ACLU, and putting the organization in the title of his paper, he’s giving fodder to politicians such as Sessions to blame Chicago’s murders on the people and groups who work to protect the constitutional rights of Chicago’s residents. It’s red meat for law-and-order voters.

Even if Cassell and Fowles are correct (and again, I don’t think they are), the communities where stop and frisk is most heavily enforced are essentially Fourth Amendment free. According to the survey done in conjunction with that 2016 Northwestern study, nearly 7 in 10 young black men in Chicago had been stopped and frisked in the previous year. And according to that study, many had been subjected to it more than once: “Blacks and Hispanics were also repeatedly targeted and stopped multiple times in the course of a year, especially while they were out walking.” In one neighborhood, blacks made up 9 percent of the population, but comprised 60 percent of police stops.  Citywide, 7 in 10 streets stops were black suspects, and blacks were far more likely to be frisked after a stop.

(These disparities have continued after implementation of the new policy, though the number of overall stops has dropped dramatically.) Overall, just 6 percent of stops produced drugs or weapons. A common response here is that blacks are more likely to commit crimes; therefore, they’re more likely to be targeted by police. The problem is, that assumption isn’t backed up by the data. A Chicago Sun-Times study found that just 2 percent of all stops produced a gun, and another 4 percent produced guns or other contraband. That means that 94 percent of the people stopped were innocent. Importantly, those figures were consistent across races and ethnic groups.

It’s worth noting again that the ACLU agreement with the city did not prohibit street stops or stop and frisks. It merely required that the city’s police officers document those stops so their constitutionality could be assessed at a later date. Cassell and other defenders of stop and frisk argue that the burden of filling out a form dissuaded the city’s cops from making such stops. That’s one possible explanation, though it’s always interesting when defenders of law enforcement argue that police officers are so sensitive that they’ll simply stop doing their jobs at the slightest inconvenience or risk of oversight. Another possible explanation is that city police knew that most of the stops that they had been making were unconstitutional — they were based on little more than the race or ethnicity of the suspect. They knew that documenting such stops would get them into trouble, so they stopped making the unconstitutional stops. In fact, before the new policy, the ACLU of Illinois estimated that at least half of such stops in Chicago were unconstitutional.

It’s of course important to point out when law-and-order types (or anyone else) have mistaken correlation with causation, as Cassell and Fowles seem to have done. And the good news is that there’s plenty of evidence to support the notion that constitutional policing is pretty effective at keeping the country safe. But in the end, the Fourth Amendment and other constitutional protections aren’t (or at least shouldn’t be) subject to a cost-benefit analysis. If we were to repeal the Bill of Rights entirely and give police free rein, crime would probably drop pretty dramatically. After all, violent crime is nearly nonexistent in most totalitarian states. We value our rights more. Sometimes, valuing our rights means sacrificing public safety. Fortunately, when it comes to stop and frisk, that just doesn’t appear to be true, no matter how many times Jeff Sessions says it.