How the hell do we measure a police department's effectiveness?
If we reward increasing arrest rates (because presumably we're arresting more and more of the criminals) the straightforward solution is to arrest more people, and that would seem to lead to more and more draconian law enforcement to reach quota.
If we reward declining arrest rates (because presumably there are fewer and fewer crimes being committed), then the straightforward solution is to stop making arrests and thus allow real crime to proliferate.
This just seems like a no-win solution for honest departments and civil rights.
A good question, I would not base it on arrests though, that is like measuring developers by bug reports.
Start with the purpose of a police department, law enforcement and in carrying out that duty a mitigation of unlawful activity. So in terms of achieving their goals one would start with comparing unlawful activity in their jurisdiction vs a similar jurisdiction with a different agency. Compare per capita costs (hence the tax burden) and compare customer satisfaction (the citizens).
In a squishy way, do the citizens feel safe? Do they engage in activities that would not be possible if they felt danger? And How much do we spend compared to other cities with similar economic demographics ?
So we reward arrests that accurately convict the perpetrator responsible in as short as time as possible. Always finding your crook, but taking on average 14 - 18 months to do so, not good. Solving 90% of your caseload in under a week is good; Having more than a 1% false positive rate is bad. Citizens trusting/partnering with the police is good, people who can't trust their police is bad.
That's one of the reasons I have pushed for the 'civil treason' sort of crime for violating the public trust. In order to trust the integrity of a police officer, I need to know that the risk they would face of violating that integrity is so high they don't do it. Life in prison for a police officer that commits perjury for example. Not regular citizens, but someone entrusted with enforcing the law. Similarly for judges that take bribes. In order for democracy to work you need a justice system that is harder on itself than it is on criminals.
I would be more comfortable with the surveillance request if I knew that if the officers were lying about the need for it they could be sentenced to life in prison. That makes them less likely to lie.
"This just seems like a no-win solution for honest departments and civil rights."
My thesis is that the root of the problem is that its a "no lose" solution. There isn't any penalty for being a bad cop, you might get a reprimand for misstating that surveillance was the law but that's it. If the penalty, the 'lose' part, was there, then the conversation changes.
The number of crimes committed in an area is often used. If the number goes down, the police are doing a good job of discouraging crime.
This results in in things like absurdly dysfunctional web forms for reporting crimes. People just give up on reporting their car being broken into rather than waste an hour figuring out why their report keeps getting rejected.
Thus there is less reported crime, and the police pat themselves on the back.
I don't think it's that hard. There are a set of crimes, like burglary, mugging, and armed robbery that are at the core of the quality of life in a city. Clearance rates for those crimes, plus a survey that measures reporting rates, should give a pretty good measure of a police force.
Arrest rates, unless they are in these categories, tell us very little.
His point was that nuclear weapons were seductive because they were cheap. And even though they might cost $100M and require $25M/year in maintenance, that was a pittance compared to training, equipping, and deploying a couple of armored divisions which could similarly level a city. So having a "cheap" nuke meant you didn't need a couple of divisions of marines.
Cops have the same problem. There are only so many of them, and yet there is a lot of crime. So the combined forces of budget cuts and crime fighting pull them toward solutions of "efficiency" (which is code for more crime fighting for less money). At a logical (but immoral) extreme, if you have cameras recording everything on every street in a city 24/7 with cameras that can see in all conditions, and a crime occurs, you don't need cops you don't need witnesses, you just rewind the tape to when it happened, see who did it, and go arrest them. A police force of 150 officers reduced to one fat guy [1] sitting in a room full of monitors and recording devices. That is why these guys practically orgasm over things like the DARPA Argus-IS system [2].
Watching re-runs of Starsky and Hutch and Cagney & Lacey vs modern cop shows like Castle or NCIS gives you a good feel for the difference in efficiency that tech has brought into law enforcement.
So the weird thing is that people demand these sorts of capabilities and tools for good reasons (crime mitigation, catching bad guys after the fact) and ignore the potential for abuse (which is more often covered up). When you see editorials that the biggest drug gang in LA used to be based in the Rampart police station you realize that no amount of "good intention" protects you from a few now truly well equipped "bad apples." I've proposed to my representatives a law that would make "violation of the public trust" a capital crime. It is the civil equivalent of treason. Not a lot of traction on that effort though. That is what it will take though to protect yourself from folks who would abuse these toys to do bad things. I can't see trying to deny them the efficiencies as a winning proposition you come out looking guilty.
[1] Its a trope I see all the time, the guy watching the monitors is morbidly obese (always male) and generally portrayed as a loser. Of course their job is redundant so in the "real" world nobody would watch the monitors.
[2] http://blogs.discovermagazine.com/80beats/2013/01/29/watch-t...