In a 1982 issue of The Atlantic, a renowned theory was published. It was the broken window theory. The theory basically states that if a building has a broken window, vandals will likely break the other windows and/or break into the building. The article theorizes that small, petty crimes lead to a culture of disorder and crime; therefore, small, often victimless crimes should be prosecuted with “zero tolerance” policies to reduce petty crimes and reduce major crimes. I wonder if there were any studies that support or reject this theory.
Many proponents of this theory point to New York City during the 90’s. William J. Bratton was appointed police commissioner, and he developed a “zero tolerance for graffiti and turnstile-jumping.” During this time, violent crime fell by 51% and homicides fell by 72%. his seems to back the hypothesis that a zero tolerance for petty crimes decreases major felonies, but correlation does not equal causation. I was immediately susceptible to confounding variables. It could also be reverse causation. Could violent crime cause more smaller, petty crimes?
A study conducted by the University of Chicago Law Review proposed the decline of the violent crime rate was largely due to the decline in the crack epidemic in America. Evaluating the different precincts in the city supported this hypothesis. Their study found that “those precincts that received the most intensive broken windows policing are the ones with the largest increases and levels of crime during the city’s crack epidemic. Consistent with findings elsewhere from city-level data, 99 jurisdictions with the greatest increases in crime during this period tend to experience the largest subsequent declines as well.” By looking at the individual precincts of law enforcement in the city, this study is able to show how confounding variables, not the zero tolerance policy, was the cause for a decrease in crime.
Testing this theory is very difficult. In a perfect world, I would like to see experiments set up. Random cities would be chosen, and each city would be randomly instructed to adopt or not adopt zero tolerance policies for small crimes. If enough cities and towns were involved in the study, and allocation of enforcement type was random, reasonable evidence should arise to support or reject the alternative hypothesis. This experiment should also be conducted at the same time to attempt the elimination of third variables. A third variable that may be time sensitive could be seen during the 90’s, when crime went down in virtually every city, not just NYC, due to the decline in crack.
The broken window theory is a popular theory for discussion, but there is little evidence to support or reject both the null and alternative hypothesis. Even the creator of the theory, James Q. Wilson states, “I still to this day do not know if improving order will or will not reduce crime. People have not understood that this was a speculation.”
Before reading your article, I had never heard of the broken window theory, but I now have a better understanding of it. It makes sense that if there is a building with broken windows that looks rundown, people will assume it is abandoned and make that a hangout spot, even though they may most likely be trespassing. Although I do not think enforcing a zero-tolerance policy will have a drastic effect on the broken window theory, there is still room for me to be proven wrong since there is no answer to the theory yet. The experiment that you proposed seems as if it could work really well, especially if the cities are randomized enough that the demographics of the area also vary. I like how you also questioned the possibility of reverse causation and third variables. Good job!
Your blog focused more so on crime rate but when intrigued me was the broken window theory. Towards the end you said there wasn’t much research done on the theory which is disappointing because it seems to be a great sociological opportunity to expose and decrease human crime rate but if I were to conduct an experiment I would pick towns of each social class, taking third variable into consideration. (current crime rate, what kind of crimes are most common vs least common, economic income, etc.) In each area a new building would be built and after construction scientists would break one window. Scientists would then observe the building and see what or who inhabits it after that point, what crime is associated with it, and what third variables influenced the kind of behavior within or surrounding the building with the broken window.