SPECIAL IDEAS REPORT

« How Far Will You Go? | Main | Deep Impact »

22 July 2009 3:00 PM

The Death and Life of American Cities

Ryan Avent:

One of the persistent themes of urban change is the presence of positive and negative feedback loops. Crime provides an excellent example. An increase in crime stretches police resources, decreasing the odds that any individual crime is solved and thereby increasing the return to crime -- and generating more of it. Which further stretches police resources.

This process touches off other feedback loops. Rising crime reduces property values which reduces property tax revenue. This limits city resources and further strains the police force and the public services which might otherwise keep at risk residents from turning to crime. Declines in public safety and service quality encourage mobile residents to leave, and since richer residents are more mobile that has a strong negative impact on the revenue base, further complicating matters.

Out-migration also reduces property prices which then attracts people who need cheap housing, which will tend to be economically distressed individuals and households. These families then demand more city services while contributing less to public coffers, and so on.

Luckily, he says, there are positive feedback loops too.


TrackBack

TrackBack URL for this entry:
http://ideas.theatlantic.com/mt-42/mt-tb.cgi/12180

Comments (1)

John Harrold

I believe the examples cited above are "positive feedback loops". The term positive in that statement doesn't refer to the social impact the effector (crime in this case) has on the system. The term positive there refers to whether or not the effector ultimately amplifies itself. If an increase in crime leads to a series of events that creates more crime that is a "positive feedback loop". If an increase in crime leads to a series of events that regulates crime and forces it back to "normal" levels, then that's an example of a "negative feedback loop". Or at least that's how these terms are normally used in biology, process control, etc.

Comments on this entry have been closed.