A journalist pointed me to this recent report from the New York City Department of Investigation, which begins:
Between 2010 and 2015, the New York City Police Department (NYPD) issued 1,839,414 “quality-of-life” summonses for offenses such as public urination, disorderly conduct, drinking alcohol in public, and possession of small amounts of marijuana. . . . NYPD has claimed for two decades that quality-of-life enforcement is also a key tool in the reduction of felony crime . . . NYPD has claimed for two decades that quality-of-life enforcement is also a key tool in the reduction of felony crime.
Here’s what they find:
OIG-NYPD’s analysis has found no empirical evidence demonstrating a clear and direct link between an increase in summons and misdemeanor arrest activity and a related drop in felony crime. Between 2010 and 2015, quality-of-life enforcement ratesand in particular, quality-of-life summons rateshave dramatically declined, but there has been no commensurate increase in felony crime. While the stagnant or declining felony crime rates observed in this six-year time frame may be attributable to NYPD’s other disorder reduction strategies or other factors, OIG-NYPD finds no evidence to suggest that crime control can be directly attributed to quality-of-life summonses and misdemeanor arrests.
I took a quick look, and they make a reasonable case that there’s no evidence from 2010-2015 that so-called quality-of-life policing has any effect in reducing serious crime, hence there’s a good case for doing less of this sort of harassment of citizens on the street, given that, as they say in the report: “Issuing summonses and making misdemeanor arrests are not cost free. The cost is paid in police time, in an increase in the number of people brought into the criminal justice system and, at times, in a fraying of the relationship between the police and the communities they serve.”
But I don’t know how relevant this is to claims about the effectiveness of quality-of-life or “broken windows” policing in the past. The argument was that quality-of-life policing was necessary in the 1970s/80s/90s because the law was not widely respected. Now that behaviors, attitudes, and expectations have changed, perhaps an intense level of quality-of-life policing no longer has the effect it had earlier. So it’s possible that the ramping-up of those sorts of police actions was a good idea in the 1990s, and that the ramping-down is a good idea now.
P.S. I don’t know who wrote the report in question. The only names I see are Mark Peters, Commissioner, and Philip Eure, Inspector General for the NYPD. I find it difficult to interact with a document with no listed author. At the end of the report, it says, “Please contact us at: Office of the Inspector General for the New York City Police Department” and then gives an address and some phone numbers and emails. That’s fine, and I understand that it’s the whole Office of the Inspector General for the NYPD that takes responsibility for the report, but there’s still an author, right?
P.P.S. Here’s a news article by Nick Pinto with some background.
P.P.P.S. Lots of informed discussion here from Peter Moskos.
I’m unsure why anyone expects a direct link between this enforcement and crime decreasing, given that the posited causal chain is broken windows -> lower respect for law -> more crime. Given past success, the question is the counterfactual; would failing to enforce these measures reduce respect for law? And would this lead to an increase in crime levels? These are by no means obvious in either direction, but it isn’t even addressed in the analysis, from my quick reading.
(Also, NYPD presumably uses editorial authorship as a way to endorse the report and insulate the authors, who are not public figures, and should not be held personally responsible for the contents – and you may dislike this, but it’s standard in many agencies that all authorship is ex-officio, in the name of the Comissioner and IG, as the report lists. I know other NYC government agencies do the same.)
“broken windows -> lower respect for law -> more crime”
Is not exactly what the original idea was. Respect for the law would be something at the individual level, while broken windows is really about places and what individuals do in places. Broken windows are not illegal but they change perceptions.
The original idea was:
Broken windows -> perception that no one cares about a place/that there will be no consequences -> negative behavior by some and fear for others.
So for example if you come into a place with trash all over, weeds in the vacant lot, unleaded up graffiti, etc you are going to think it’s okay for you to throw trash, write graffiti, smoke pot in public, and further that no one will call the police if you break into a store, etc. They also included “disorder” as part of the idea of broken windows, that where there was disorder people (public drinking and drug use, aggressive begging, etc) there would also be a perception of dangerousness and this would create of vicious cycle of crime ->fear ->more crime
From the article
“The citizen who fears the ill-smelling drunk, the rowdy teenager, or the importuning beggar is not merely expressing his distaste for unseemly behavior; he is also giving voice to a bit of folk wisdom that happens to be a correct generalization—namely, that serious street crime flourishes in areas in which disorderly behavior goes unchecked. The unchecked panhandler is, in effect, the first broken window. Muggers and robbers, whether opportunistic or professional, believe they reduce their chances of being caught or even identified if they operate on streets where potential victims are already intimidated by prevailing conditions. If the neighborhood cannot keep a bothersome panhandler from annoying passersby, the thief may reason, it is even less likely to call the police to identify a potential mugger or to interfere if the mugging actually takes place.”
So broken windows policing would have originally been to focus on addressing those perceptual issues and the reality they may have been indicia of. Literally, fix the broken windows but also discourage public drug use, urination etc, but fixing broken windows would actually help with that. And of course the perception was also a reflection of reality, in many places like that no one did care and it was unlikely someone would call the police, and if called it was unlikely they would come. On the upper east side if there was an overflowing trash can it would be dealt with, not so in the South Bronx.
Later, the model changed to one focused on enforcement of low level rules breaking aka quality of life crimes.It’s really disorder reduction policing which is sort of broken but also policing that is based on a few different ideas, which is one reason it became useless. One idea was arrest people for small crimes and they will be incapacitated, another really takes the whole idea in reverse and says make relationships between community and police more like one of an occupying army and that will reduce crime, while if you read the original Kelling and Wilson article the point is that if you engage community and police to work together
Also, whether or not there are “past effects” is very much subject to question. Although the place based model
http://cebcp.org/evidence-based-policing/what-works-in-policing/research-evidence-review/broken-windows-policing/
http://www.theatlantic.com/magazine/archive/1982/03/broken-windows/304465/
But let’s include a bit more specific about the context here. Where socio-economic status is high enough (Upper East Side) that window replacement is not a financial burden, people are too busy working to spend time hanging out on the streets, and law enforcement has little to do but identify people in the neighborhood who like they don’t belong and make sure they don’t stay too long.
The South Bronx differs from this substantially. I agree that these issues should be addressed, that windows should be repaired, and that trash should be picked up wherever it is found. However, I disagree with the notion that excessively harsh punishments, fines that are unreasonable given the SES in the area, and prison time for these types of offenses will do anything but push the problem down the road and increase retaliatory actions on those who harassed by said police.
Edited:
But let’s include more specifics about the context here. Where socio-economic status is high enough (Upper East Side) that window replacement is not a financial burden, people are too busy working to spend time hanging out on the streets, and law enforcement has little to do but identify people in the neighborhood who look like they don’t belong and make sure they don’t stay too long.
The South Bronx differs from this substantially. I agree that these issues should be addressed, that windows should be repaired, and that trash should be picked up wherever it is found. However, I disagree with the notion that excessively harsh punishments, fines that are unreasonable given the SES in the area, and prison time for these types of offenses will do anything but push the problem down the road and increase the likelihood of retaliatory actions by those harassed by said police.
The point is that “broken windows theory” is not at all about harsh punishments, fines etc, it is about police and others working with the community to make sure that these things are taken care of. It What things evolved to in NYC and what happens in places like Ferguson where police are basically used to finance city government has nothing to do with that model. If you read the research what you actually find is that disorder reduction broken windows models may work best when specifically targeted at individual block segments that have large amounts of crime or sometimes even at specific buildings, lots or corners.
My point is that the goal of reducing crime and civil disorder is good. However, pretending that a faulty causal mechanism is actually true, does not get us any closer.
Pretending that civil rights are not compromised in the process does not mean it is constitutional to do so. While I applaud the goal, I disagree that harsh punishments are the only plausible strategy, or that it is even a good strategy.
What if we decided we wanted to punish people on the upper east side who did not water their plants enough? What if we decided that fines were not effective and instead we would use jail as the solution?
“…would failing to enforce these measures reduce respect for law?”
Good question. Many years ago I hypothesized that creating laws or rules (I was a teacher at the time) which you were unable or unwilling to enforce would lead to greater contempt for and less conformance overall. Seems that this analysis fails to support the hypothesis. Counter-intuitive to me, and I’d like to see some independent verifications, but good to know if valid.
It seems to me that there is often needless criminalization of what are effectively serious social faux-pas. Of course, if someone breaks my window, I expect them to pay for it :)
I wonder whether the NYPD ever does randomized experiments to address issues like this. If not, why not? Is it fear that if it ever got out that they were ‘experimenting with our safety’ there would be irrational public outrage?
Yes. Think of how much backlash Facebook got for doing tests on altering people’s moods by manipulating news feeds. Now consider if they had been randomly sending people to jail instead just showing them different Facebook posts. I’m not sure that public outrage would necessarily be “irrational”.
How would outrage be rational? Because the police should keep nonrandomly following a potentially bad strategy? If you believe policing is too aggressive, how is it worse to (temporarily) have a 50% chance of living in a neighborhood that gets policed aggressively than a 100% chance? And if you believe policing is appropriately aggressive, then sure the experiment might upset you, but not as much as a new commissioner coming in and completely eliminating aggressive policing, which is totally possible given the genuine disagreement on the issue. Whatever your belief, your treatment assignment in the experiment would be completely possible in the absence of the experiment. And afterward, we’d get to know that the best strategy was being implemented.
While I understand your point (ie risking temporary loss to become informed in order to reduce long term loss), this is basically medical ethics on a much more extreme level.
It’s not easy, nor should it be easy, to get approval for a clinical trial. Very importantly, enrollment must be voluntary.
You call performing an experiment ‘risking temporary loss’. What do you call not performing an experiment? ‘Risking long term loss and guaranteeing long term uncertainty’?
long term loss is just the integral of temporary loss rate. “Risking temporary loss” is also “risking temporary gain”. The real question should be the magnitude of the risks and returns.
Agreed
As I stated, this was your point and I understand the reasoning.
The counter point is that you do not need to look too far into human history to know that unregulated experimentation on human subjects very frequently leads into highly unethical behavior. Even if the stated goal is “gaining knowledge for the improvement of all mankind”. This is why you can’t go around doing experiments without heavy review.
And in the case in point, it seems like a quite reasonable conclusion that person A may be upset that they are being punished for doing the same thing that person B did, with the only justification being that
rbinom(n = 1, size = 1, p = 0.5) == 1
As for enrollment in medical trials being required to be voluntary, I’m not sure that’s true. How about trials of new protocols where entire hospitals are randomized to either adopt the new protocol or not. It’s not the case that every potential future patient at these hospitals must agree in advance to the hospital participating in a trial like this. Voluntary enrollment is great and should be required when it’s possible, not when it’s not.
You don’t have to try very hard to find plenty of rationally, legitimately upset people due to unequal treatment under the law. If you tell them that, even temporarily and for a putative good reason, you’re going to be arresting person A but not person B for the same crime, people are going to be upset because that isn’t how most people think justice should work. I’m all for randomized experiments but this might be a situation where you hope for a fortuitous natural experiment, like different precincts or maybe different cities implementing the policy at different times.
Unequal treatment under the law occurs naturally from precinct to precinct or city to city anyway. The only difference between that type of unequal treatment and the type in a randomized experiment is that at least we can learn from an experiment.
And do you actually believe assignment in the random experiment would actually be random? Who would control that? You already have unequal treatment under the law in the same precinct if you’re not white. And you have unequal treatment between different precinct due to many “non-random” factors (sozioeconomical status, ethnic compositions, amount of racist prejudices etc.).
> If you believe policing is too aggressive, how is it worse to (temporarily) have a 50% chance of living in a neighborhood that gets policed aggressively than a 100% chance?
Because fairness matters.
Overly aggressive policing results in false positives. Lax policing results in false negatives. Two examples of the cost of false negatives:
Case I: Imagine a law where compliance poses an inconvenience for me, e.g., a speed limit. Suppose I choose to be a good citizen, comply with the law, and accept the associated inconvenience to me. I do this because the community is generally better off when members choose to comply with community-established rules. If I make this choice then I’m going to be pretty unhappy if I see someone else in violation and the police not enforcing the law. I may choose not to comply in the future out of contempt for current failure to enforce.
Case II: Violation of the law by one member of the community causes harm to another. Justice is not served if the law is not enforced.
Cf. Michelle Meyer on the “A/B illusion”
http://papers.ssrn.com/sol3/papers.cfm?abstract_id=2605132
Dean:
I clicked through to the link, and the paper looked interesting, but I had a tl;dr attitude. It’s 60 pages long! I guess it’s a style thing. Papers for law journals are long long long, following the template of documents written for courts. I’ve done occasional legal consulting, and everything is dragged out. Depositions can be hundreds of pages long, it can take dozens of pages to describe a nearly trivial statistical procedure, etc. The issue for legal documents, I think, is that the trial happens only once, so you have to cover all possible contingencies. Now, there’s no reason that a law review article has to be 60 pages long, but if you’re operating under the norms and expectations set by legal documents, it makes sense that we’ll see such things.
Did you read the abstract? I think it makes the points thoroughly enough for this comment section.
How about a NYT version?
http://www.nytimes.com/2015/06/21/opinion/sunday/please-corporations-experiment-on-us.html?_r=0
Not just NYPD there have been many randomized experiments on policing.
Here’s a fairly recent meta analysis of disorder policing experiments and quasi experiments http://petermoskos.com/files/BW/Braga_2015_policing_disorder_reduces_crime.pdf
“This Report looks solely at the question of whether quality-of-life enforcement has any
measurable relationship to felony crime.”
Why would these researchers NOT first be interested in whether NYC quality-of-life enforcement specifically reduced the minor crimes it was directly aimed at ?
Why bother worrying about felonies if the treatment failed on misdemeanors ?
The Broken Window theory was devised by social scientists James Q. Wilson and George L. Kelling in early 1980’s. Its validity has never been established — with much contradictory evidence and controversy both within the social sciences and the government.
I thought the idea behind the Broken Window theory was that crime was more likely in places with lower quality-of-life. As far as I can tell, the results of this study show that a decrease in quality-of-life summons did not cause an increase of felonies. I couldn’t see anything that describes why there was a decrease in quality-of-life enforcement. If the number of quality-of-life violations decreased (and thus the number of summons decreased), the decrease in felony rates at the same time would seem to validate the the quality-of-life enforcement had a direct effect of decreasing quality-of-life violations and that this had an indirect effect in causing a decrase in felonies. It is only counterevidence against the theory if the decrease in enforcement occurred independently of a decrease in violations, unless the theory is that the enforcement measures were supposed to directly decrease felonies, instead of indirectly decreasing felonies by improving quality-of-life.
My first thought was similar: an increase in quality of life could result in a decrease in quality-of-life summonses, so the latter is not a good measure for the question being studied.
The decrease in the use of this model was a deliberate policy change. Was this done using randomization? No. But it was definitely done only because of a change in departmental policy, just as the original mass implementation was done as a change in departmental policy.
“But I don’t know how relevant this is to claims about the effectiveness of quality-of-life or “broken windows” policing in the past. The argument was that quality-of-life policing was necessary in the 1970s/80s/90s because the law was not widely respected. Now that behaviors, attitudes, and expectations have changed, perhaps an intense level of quality-of-life policing no longer has the effect it had earlier.”
I find this to be an interesting and important argument. One explanation of disappearing effects (http://www.newyorker.com/magazine/2010/12/13/the-truth-wears-off) could be that of a dynamic, or moreover, adaptive, population. When first introduced, treatment X may have a large effect of condition Y in the population. But as the population accepts treatment X, and it’s alternatives, increasing the level of X has a much smaller (and potentially inverted) effect than when it was initially introduced and studied.
I had this thought in regards to medical treatments, but I believe it’s a similar effect to what you’re suggesting; as the population adjusted to a specific style of enforcement, the effect of increasing that style of enforcement changes dramatically. It also further complicates replication studies; just because an effect on the population is negative now does not mean it had a negative effect at the time of the initial study. This is especially plausible if the results of the study had a large impact on the population of interest.
To me this time-series, state-space type model is VASTLY under-utilized in social sciences. The reality of decades of observation is that things could well work differently at different times even while a single dynamic explanation holds. Throwing a ball in the air, we initially find that it goes up quickly. if we look at a later time we find it nearly stationary, and then again later, it’s going down again… If we studied this simple problem in the way social scientists study things like police enforcement and crime rates… we’d never realize the single unifying process. Now, if we can’t figure out a simple 2-vector state space process using those methods, how are we supposed to figure out population criminal justice dynamics?
Have you looked at this literature? I don’t think there is “the way social scientists study things” (emphasis on the “the”). I think you would find that it is more complex than you are indicating. There are randomized trials but there are also time series studies as well as panel and non panel surveys such as NCVS not to mention ethnographic studies on the ground.
My experience is limited admittedly. Most of what I’ve seen is out of economics, and most of it is of the form “At time X policy change Y happened, and “the effect” was Z” where the effect is a coefficient in an ultra-simplistic curve fit.
My impression from that kind of stuff is that Economists have yet to realize that not everything is in equilibrium at all times, and that they don’t care about modeling causality only its derivatives with respect to certain variables at certain times. When you have a differential equation like dy/dx = f(t/tscale, q) and tscale is small compared to your observational time, and the solution has a horizontal asymptote / equilibrium, then you can typically replace this with a functional relationship between y and q. At every time that you’ll look, y is close to the value it would have at infinite time. But, lots of stuff has long time scales, of decades or centuries… and it’s totally inappropriate there.
My impression is that this heavily influences economics in large part because it lets them use their favorite tools, unbiased estimates of linear regression coefficients using least squares models or whatever.
I think it’s good that the economics profession is looking for the “natural experiment” type scenario to elucidate things, but I don’t think the typical modeling strategy *THAT I’VE SEEN* is appropriate (and yes, you’re right, I’m not an expert just a bystander who has occasionally read a few studies on topics of interest).
As an illustration, suppose you’re interested in suicide prevention. Typical stuff I’ve seen is you look at some political event that changed some kind of policy, and you look at suicide in 3 years before and 3 years after. You model suicide rate as two separate lines, one for before, one for after.
My preference would be to model a much longer timeseries with the rate changing in time according to some function that combines information about local economic conditions (GDP/capita, taxation, median real household incomes, labor force participation rates, etc), local politics, geo-politics and trade (trade imbalances, currency exchange rates, outsourcing of jobs), social effects (marriage and divorce rates, immigration, migration), race, gender, time-lagged feedbacks (suicide rate now as function of suicide rate when people were children for example), social services (suicide prevention efforts, health-care availability).
So you’d wind up with some plausible integro-differential equation involving many important aspects of people’s “life experiences” and then you’d split up the population by age, and solve this integro-differential equation from initial conditions at birth for birth cohorts for the last 100 years…
In the end you’d have posterior distributions over parameters that define the rate of change of suicide rate for birth cohort relative to the experiences that they experienced through time.
Is that the kind of stuff you’ve seen done?
I don’t know that I’ve seen exactly that but I have seen people working with time series data that incorporates a lot of this.
Most of this work is done by people who are specifically focused on criminal justice or public policy rather than by economists.
I don’t know much about suicide work but I do know that in looking at why crime (serious crime, all crime, homicides) changes over time people hae looked at many local conditions and a lot at the issue of age/period/cohort. For example the much studied huge increase in crime during the 1960s is partly a cohort flow issue (crime is mainly 16-24 year old males) where some explanation is the baby boom, but also some explanation was the changing economy,changes in education (start of the growth of college attendance), continued decline of agriculture, and then the period effect of Vietnam and civil rights. Likewise the increase and decline in crime in NYC has been attributed (and most decent models attempt to incorporate) growth/decline in size of youth population, decline of industry, changing employment, emergence of crack, emergence of mass incarceration, change in police practices including but not limited to stop and frisk and disorder policing. Of course both temporal and spatial autocorrelation play a role too and many models incorporate them though temporal more than spatial. The other thing is that using data at the city level is really not always the best choice. Sometimes census tract is going to make more sense.
Good, I’m glad if I’m wrong about the simplicity of the models. I do agree though that studying at the city level is less good. It would be interesting to see some spatial stuff in areas like LA county, where it’s all run together.
Since it doesn’t seem to be pointed out anywhere here, I’d just like to note that “quality-of-life” is in the eye of the beholder, and for those getting these police summons, the quality of their life is decidedly worse. There are horror stories about poor, usually minority individuals, ending up on Riker Island for years once they are ensnared in the judicial system, but that 1.8 million summons needs to be translated into the effect on people. A few examples–Baltimore typically issues 100,000 of these type of summons year (out of 650,000 population), while in Ferguson, Mo there are about 16,000 outstanding warrants of some type on a population of 21,000 or so. That last statistic came up in the Supreme Court dissent by Sotomayor over whether an illegal search of an automobile by the police could nonetheless be made legal if an outstanding warrant existed on the individual being stopped (the court said yes). Sotomayor’s point was this essentially criminalized the entire population of a city such as Ferguson and that police functions, which depend in the final analysis on the consent of the policed, should not do so. The fourth amendment was adopted because of the behavior of the British troops in the colonies and the reasons for doing so are still applicable today–people don’t like being at the mercy of often capricious behavior of the authorities.
This may not seem to have anything to do with statistics but statisticians have been complicit in social policies that we’d all rather not have happened (such as eugenics or much IQ testing). It’s important to keep in mind the real-world consequences of this mathematical abstraction we call probability theory and statistical inference.
+1
Ditto
Civilization has its discontents and does not have to accommodate every lifestyle. Policing places like Ferguson saves lives, violations of the law have to be punished and unfortunately police does not have too many instruments.
The telescreen will be installed in your apartment on Tuesday, since you are expressing concern about lack of instruments available to the police. Also, you might peruse
https://newrepublic.com/article/134712/wounds-incarceration-never-heal.
Dwight Schrute: “Better a thousand innocent men are locked up than one guilty man roam free.”
Civil rights are foundational to the U.S. system of government. The rights to be secure in one’s person, to freedom of movement and to due process (among others) enshrine in both principle and law that we choose to err on the side of freedom and liberty and thus on the side of false negatives rather than false positives. This means that laws and policing policies that directly contradict these values are not only unconstitutional, they are quite literally Un-American.
Laws that are written with the intent of circumventing these rights through punishments that exceed the harm done, that unfairly create a financial burden on the offender despite the lack of severity of the crime, that unfairly target groups of individuals are also Un-American. From my memory of reading the constitution, I do not recall a right to an ever increasing property value for which my neighbors are legally responsible.
A story about what happens when you stop policing petty crime:
“Like all young thugs I’ve encountered, he “ran hot” which meant he went from zero to full-rage in about a millisecond. After ~20 seconds of screaming every four letter word and negative gentrification invective in the book at me (he and I are the same race but that seemed to elude him), he started towards me, pulled up his shirt to show me his gun and told me he was going to shoot me (I’ve actually cleaned that language up quite a bit, I will let your imagination run with what he really said). I grew up in rural VT, hunted in my childhood, spent time in the Corps and am extremely comfortable around firearms, but I had never had one pulled on me before, or the threat of it. I simply reacted out of fear, grabbed him by the neck with one hand, put my hand on his gun with the other so he couldn’t draw it and slammed his head into the retaining wall in front of my house while trying to disarm him.
As a side note, having grown up in a rural VT community as a minority, I was always one of the few non-white people around, yet in all my travels personally and professionally, the District of Columbia is by far the most racist place I’ve ever been, and the overwhelming bulk of the racism is generated and perpetuated by fellow members of the black community. It is incredibly disappointing.
Anyway, the police came, ran him through the system, discovered he had a couple outstanding warrants (one for assault and another for weapons). He was 24 and had been in the system a number of times already. Police told me this would be the last straw, he would go away for some real time on this.
Less than 72 hours later he was throwing glass bottles at my front door.”
http://www.popville.com/2016/06/well-this-is-depressing/
Gulliver:
From your story: “discovered he had a couple outstanding warrants (one for assault and another for weapons).” I think everyone agrees that assault and weapons crimes should be prosecuted. “Quality-of-life policing” typically refers to offenses such as public urination, disorderly conduct, drinking alcohol in public, and possession of small amounts of marijuana.
I don’t want someone puking in front of my sidewalk either, but that’s nothing like assault or weapons possession.
One can typically hammer on another human being with relatively minimal consequences. Our society is (justifiably, in my opinion), cracking down on sexual assault, but if Brock Turner had just slugged his victim, broken her jaw and a few ribs, then given his privileged social status, he would have gotten the same 6 months if that. Echoing Andrew’s point, physical crimes against other people are why locking people away from other people is very reasonable and should be applied across the board for all of these offenses (those men who tend to assault women and other men also tend to sexually assault). The quality of life offenses can typically be handled without use of imprisonment, and in Ferguson, a number of the offenses were for things like not having curtains, illegally parking (for repair shops), etc, leading to the suspicion that the government was imposing them for pursuing revenue.
Andrew:
According to the author of the blog post what happened was a consequence of the attempts to “ease the burden”: it all starts with crackheads pissing on your lawn and ends with police and judicial system tolerating much more serious crime. Once normal people are gone, the neighborhood turns into a jungle.
numeric
“quality of life offenses can typically be handled without use of imprisonment”. Let’s say, perps are unemployed, don’t have any money and cannot pay fines. What instruments do you propose police should use as a deterrent?
Gulliver:
My impression is that the discussion was about “quality-of-life policing” which involved “summonses for offenses such as public urination, disorderly conduct, drinking alcohol in public, and possession of small amounts of marijuana.” I agree with you 100% that it’s a problem if people are going around assaulting people and the justice system is ignoring it.
The key intermediate step is the claim that an increase in quality-of-life policing for minor crimes, back in the 1990s, led to a safer, more law-abiding environment which has resulted in a drop in serious crimes. As I wrote in my post, I don’t know that the report discussed above, which uses data from the 2010s, says much about this claim, one way or another.
So I’m not saying I disagree with you. I’m just saying that you can’t really take an example of the police ignoring assault and weapons possession, and use it as an argument about quality-of-life policing. Whether or not there is quality-of-life policing, I think we all agree that the cops shouldn’t let people get away with assault.
Fine. Call it what it is. Locking people up for being poor and unemployed. That is the solution we have chosen as a society. There are other ways to deal with it. Granted, an individual police officer only has what is at their disposal in any given situation, but as a society we have other options.
Curious
“an individual police officer only has what is at their disposal in any given situation, but as a society we have other options.”
What options do you have in mind and why do you think that society has not tried and continues trying them? You can not simply appeal to the fact that there is more that can be done, since that is always true.
I agree that people who are poor and unemployed (both in America and abroad) need to be given enough money to satisfy all their desires. That way they will not have any inclination to damage your property, take your things or sexually assault your wife. If they continue to commit crimes, we conclude that they were not given enough money and we simply need to give them more. Solutions that rely on punishment, even if they are cost-effective, are clearly immoral.
Gulliver:
Please cut the sarcasm/trolling. Saying that maybe we shouldn’t throw people in jail for pissing on the street is nothing like saying that people need to be given enough money to satisfy all their desires so that they don’t rape people. On our blog we have thoughtful comments from all sorts of political perspectives; the point is to advance the discussion, not to troll.
work-release, community service, etc. None of these require any monetary contribution on the part of the “perp” (just like on CSI!).
numeric
“quality of life offenses can typically be handled without use of imprisonment”. Let’s say, perps are unemployed, don’t have any money and cannot pay fines. What instruments do you propose police should use as a deterrent?
since when is “assault with a deadly weapon” a “petty crime”? Assault only requires an attempt, so “pulled up his shirt to show me his gun and told me he was going to shoot me” was as far as his attempt went, because the assailant responded rapidly, but I think it still qualifies.
whoops, “victim responded rapidly” not assailant
This has nothing to do with arresting people for littering. In fact, it is an argument for focused attention rather than on putting hundreds or thousands of people with no serious history, no weapon, and no violent threat, into the system.
You are correct. It has to do with arresting people for not cleaning up after others who have littered. It has to do with circumventing the 4th Amendment by finding trivial matters that allow one to put someone behind bars who the police believe, “should have been there all along”.
I don’t see any thing like that in the blog post about the person explaining why he left the District. In fact I don’t think broken windows is particularly relevant for the story of the author’s experience with that one guy. I do think it is relevant where he talks about how no one would help his neighborhood.
The context of the post is ‘quality of life summonses’. The fact that they are ineffectual was the focus of the post. My focus is on the clear implications of this approach.
I’m not sure there’s anything of merit in the report. It’s all correlation, no causation. The closest it comes is in saying that enforcement of “quality of life” infractions has gone down for a few years but the felony rate hasn’t gone up. That tells us nothing about the counterfactual.
Perhaps some people thought that there is one and only one thing that affects the crime rate — whether police are also arresting people for minor quality-of-life offenses — and that the two trend lines will always perfectly match each other at all times. OK, that straw man has been refuted. But now what?
The two trend lines will be perfectly inverse to each other, that is.
In the Braga et al meta analysis http://petermoskos.com/files/BW/Braga_2015_policing_disorder_reduces_crime.pdf they talk about the cottage industry of people who look for explanations of NYC data and the fact the it is one of the places that have not done experimental research.