1. The Fryer paper does not provide evidence about a causal link between number of interactions and rate of crimes. It merely notes that this is one possible mechanism for the observed relationship between PDs that are investigated because of high profile brutality or killings and subsequent increases in crime. Even if we accept a causal link between number of interactions and rates of crime, please note that the actual number of officers did not change. The officers simply stopped interacting as much despite still being there.
Are you suggesting that interactions would be expected to stay flat if there were less cops in the streets? How long would their shifts be and how large of a geographical area would they need to cover? Is this feasible? And how would those interactions be anyway, do you think they would become more or less violent as a result?
And also, your example about Toronto is not causally interpretable either.
Pants-of-dog wrote:2. Feel free to quote the other research. I find it interesting that you refuse to compare Toronto because the cities may differ too much, but cities in Argentina seem to be comparable.
The paper doesn't compare cities in Argentina with American cities. It compares car thefts in blocks that didn't get an increased police presence with blocks that did.
Di Tella (2004) wrote:Abstract
An important challenge in the crime literature is to isolate causal effects of police on crime. Following a terrorist attack on the main Jewish center in Buenos Aires, Argentina, in July 1994, all Jewish institutions received police protection. Thus, this hideous event induced a geographical allocation of police forces that can be presumed exogenous in a crime regression. Using data on the location of car thefts before and after the attack, we find a large deterrent effect of observable police on crime. The effect is local, with no appreciable impact outside the narrow area in which the police are deployed.
Pants-of-dog wrote:3. Why would police inefficiency (in terms of deployment) be a reason to keep police around and pay them as much or more? Does rewarding inefficiency usually work? This inefficiency in deployment seems to be another reason to defund the police.
It doesn't indeed, but what makes you believe that firing current officers and defunding PDs would fix this problem? Do you want to know what would? Some good old police union busting combined with an investment on consulting on the matter. Once that happens, yes, a reduction in the number of officers would be feasible.
That's also what NYC did for that matter: They hired consultants who provided advice on where to best deploy their police forces, and began to decrease the number of police out in the streets as a response to a decrease in the crime rates and not the other way around. From your own source (emphasis added):
USA Today wrote:...
James McCabe, a retired New York Police Department official who travels the country as a police staffing consultant, says there is little clear connection between staffing numbers and crime. “New York City made the conscious decision to reduce the number of cops,” he noted in an interview. “And crime continued to go down. It’s not what you have, it’s what you are doing with them.”
The NYPD is one of a few departments that routinely recalculate how many officers they need to staff a 24-hour cycle. Called the Patrol Allocation Plan, the statistical model studies 911 calls and calculates such variables as how the time of day and type of crime affect an officer’s response time.
NYPD’s model is similar to what McCabe, and consultants like him, preach as the gold-standard in police staffing strategies. These “workload allocation models” are time-consuming and require statistical skills that most police departments lack. Instead, cities sign pricey contracts with Weiss, McCabe and their competitors to perform data-driven analyses of how officers can best use their time.
But the suggested reforms don’t always stick, because they entail a lot of bureaucracy and require wholesale support from City Hall and local police unions. The recommendations typically involve assigning longer work shifts, moving officers out of jobs that don’t require guns—duties such as crime scene investigations and administration—and hiring more civilians.
If anything it would seem that police budgets may as well need to go up, at least at the beginning. Then they may go down as conditions allow.
The article also warns us to avoid comparing staffing and crime rates between different cities, which probably makes sense:
USA Today wrote:...
“It’s helping communities figure out how they can get the most out of their police department,” said Jeremy M. Wilson, a Michigan State University criminal justice professor and police staffing expert.
Wilson discourages police departments from comparing staffing levels to cities of similar size, and instead suggests basing police deployment primarily on the numbers of 911 calls and allowing time for cops to get out of patrol cars to talk to people. “It’s important for each community to understand what the community wants and can afford,” Wilson said. “Some communities want a community-oriented style, some want a law-and-order style, or service model. That has implications for deployment, costs, number of officers.”
Memphis has a murder rate worse than Chicago’s and a police force that has shrunk by nearly a fifth since 2011, to a head count of 2,020. The city is on its fourth round of outside police staffing consultants in eight years.
The most significant change, prompted by an outside expert, was made on previous mayor A.C. Wharton's watch. In 2013, the department redrew its precinct maps in response to a 2012 analysis from a Washington D.C think tank, the Police Executive Research Forum.
PERF’s analysis, obtained by The Marshall Project, pointed out that the downtown tourist district had up to seven times as many cops on patrol per square mile as the city’s more violent areas. Officers in high-crime North Memphis spent half their work day chasing 911 calls, leaving less time to learn about the neighborhoods they protect. Their counterparts assigned to the tourist zones spent only a fifth of their time answering the police radio.
The Memphis Police Department carved out new precinct boundaries, but rejected other advice. For example, the consultants found that officers were spending a lot of time chasing reports about burglar alarms and vicious dogs. City officials debated and rejected the idea of freeing up cops by outsourcing low priority complaints to other city agencies, or to private security firms.
“That was going to be a pretty steep hill for us to get over with the citizens, to say we aren’t providing that service anymore,” said former Memphis Police Director Toney Armstrong. “It is hard to stop doing something that you have traditionally done.”
Chicago and Memphis are yet another reason why raw comparisons between cities are probably not a good idea, even if they may share some things in common. For instance as a resident I'm also willing to bet that Chicago is actually more heavily policing the downtown area and upscale neighborhoods than poor inner-city ones. Up to an extent, it's not that surprising since those areas also happen to be where most of the city's economic activity and thus tax revenues are generated, on top of being safer for cops themselves to be deployed at (people don't like to work in dangerous places, and this includes those whose work consists in doing just that - particularly when they are senior officers). I guess a similar thing happens in Memphis.
Yet none of this actually supports defunding the police, at least not for now.