Chicago’s population is about 4.5 times larger than Milwaukee’s, but, surprisingly, the arrest totals in the two cities have been slowly converging for many years. Here are the numbers reported to the FBI since 1980 (omitting a handful of years in which one city or the other did not report arrests):
To some extent, the convergence may result from underlying crime trends in the two cities.
In Chicago, as indicated in the figure below, arrest totals have indeed tended to move roughly in tandem with crime totals (measuring crime by reference to the “index crimes” reported annually to the FBI):
Since 1985, viewed in these five-year increments, arrests have gone up when crime has gone up, and arrests have fallen when crime has fallen.
The Milwaukee story does not appear quite so straightforward:
Arrests spiked dramatically in Milwaukee in the 1980s and 1990s, but crime moved much more incrementally and indecisively. Indeed, crime hit a peak in 1990 and has been in long-term decline ever since (a brief crime wave in 2005-2007 notwithstanding), but arrests continued to climb for several more years after 1990. (A graph with more complete year-by-year figures was included in the first post in this series.)
The Milwaukee Police Department’s great arrest spree in the 1980s and 1990s thus stands as something of a mystery. While Chicago’s arrest numbers began to recede as soon as crime dropped, Milwaukee’s continued to grow.
Arrests did finally drop between 1997 and 2005, and rather steeply at that. (Numbers are not available from 1998-2000, which makes it impossible to pinpoint the exact year in which the tide turned.) Ever since, Milwaukee’s arrest numbers have been relatively flat, even as Chicago has witnessed a second period of sharp decline.
Milwaukee and Chicago also differ in the offenses for which they arrest most heavily. Here was Milwaukee’s 2011 distribution among several important offense categories:
[Technical notes: “Violent crime” here refers to non-negligent homicide, robbery, and assault (aggravated and simple). I’ve not counted rape because of longstanding issues with the way the Chicago Police Department reports rape numbers. “Property crime” refers to the four property index crimes. “Drug crime” includes both trafficking and possession. “Vice” covers prostitution and gambling.]
Now, here is Chicago’s distribution:
Note that Chicago makes a much larger share of its arrests for drug crime than Milwaukee, and devotes a correspondingly much smaller share of its arrests to disorderly conduct. Another notable difference lies in the area of vice: Milwaukee’s vice arrests are almost negligible, but Chicago’s vice number is actually considerably higher than its numbers for vandalism, weapons, or DUI.
To put the drug comparison in more concrete terms, Chicago made about seven and a half times more drug arrests than Milwaukee in 2011, even though Chicago’s overall population is only four and half times larger.
It is hard to know what exactly to make of the relatively high number of Chicago drug arrests, since FBI data on the Windy City do not indicate anything about drug type or distinguish between trafficking and possession arrests. (Milwaukee’s numbers, unlike Chicago’s, do make the latter distinction; possession arrests here are about three times greater than trafficking.)
It is possible, though, to track how the drug proportion of total arrests has changed over time:
Two aspects of this figure particularly strike me. First, it is clear that Chicago’s “lead” in this area is not simply a function of different priorities by the current leaders of the two departments, but has instead been maintained in a remarkably consistent year-in-year-out fashion for almost three decades.
Second, notwithstanding the consistency of Chicago’s lead, the proportion of drug arrests in each city has changed a great deal over time. In Chicago, for instance, drug enforcement played an almost negligible role in the total arrest picture in the early 1980s, but has increased dramatically in importance since then, rising in an especially steep and sustained fashion across the 1990s. As we’ve seen, this was a time when “real” crime (that is, index crime) was falling. Yet, it was also a time when the Chicago Police Department went from doing drug arrests in the mid-30,000’s each year to the mid- and high 50,000’s.
At least two interpretations of these trends are possible. One is that the increase in drug enforcement caused the drop in real crime. Drug abuse and addiction, as well as the operation of often-violent inner-city drug markets, fueled a great deal of other crime. As a result, locking up more drug users and dealers had significant collateral benefits with respect to assaults, robberies, burglaries, and so forth.
The other interpretation would reverse the direction of causation: without as much real crime to go around, the CPD had to shift its resources in a new direction that would generate plenty of offsetting arrests and thus help to justify its budget.
It is possible, of course, that both interpretations have a measure of truth to them.
Milwaukee’s drug arrests have also been quite variable, although the direction of change in the 1980’s was the opposite of Chicago’s. Drug arrests started the 1980’s at about ten percent of total arrests, but then quickly dropped to two or three percent per year. Since the end of the 1980’s, Milwaukee’s drug arrest percentage has risen at about the same rate as Chicago’s, albeit from a much lower baseline number. As a result, we have now returned to where we were in 1980.
I have one final topic to explore in this post. In the pie charts above, it is interesting to see that Chicago’s proportion of disorderly conduct arrests is lower than Milwaukee’s to an extent that corresponds closely with Chicago’s higher proportion of drug arrests. This suggests an intriguing hypothesis: that, from a policing standpoint, drug arrests and disorderly conduct arrests are more-or-less interchangeable; both simply represent different techniques by which the police assert their authority on the street and take troublemakers out of circulation for a time. On this view, drug arrests are often pretextual; that is, the police are not really trying to stamp out drug use per se — if they were, we would see a lot more drug enforcement in affluent communities, where drug use is no less prevalent than in the inner-city — but are instead trying to accomplish different types of law-enforcement objectives, and it just so happens that many of the individuals who threaten public safety and the orderliness of public spaces are also drug-involved, providing a convenient justification for stops and arrests.
Take a look at the Chicago numbers over time:
There does seem to be some support for my hypothesis here: since 1985, the drug and disorderly conduct arrest numbers have been a mirror image of one another — as one goes up, the other goes down.
By contrast, no such pattern is apparent with arrests for the violent index crimes. As a simple matter of political survival, there’s not a lot of police discretion when it comes to serious violent crime; cases must be investigated, and, if there is evidence against someone, an arrest must be made. However, drug and disorderly conduct arrests tend to be much more discretionary (at the level of departmental policy and/or at the level of individual officer or unit choice). For that reason, it is not surprising to see much more volatility with the numbers in these areas.
In any event, and by contrast, the Milwaukee numbers do not seem consistent with the drug-d/c trade-off hypothesis:
In Milwaukee, it seems that disorderly conduct has always been a big slice of the arrest “pie,” and it has remained high without much apparent regard to changes in the proportion of drug arrests. It’s not possible for me to draw any definitive conclusions at this point, but these patterns may be evidence that drug enforcement has not been used as an order-maintenance technique in Milwaukee to the same extent as it has been in Chicago. It is not clear why this might be the case, let alone whether this is a good thing.
Finally, it is interesting to observe that the proportion of arrests for serious violent crime has moved within the same relatively narrow band as it has in Chicago — another indication that arrests for violent crime are not readily traded off against other types of arrests.
In the next post in this series, I will continue the Milwaukee-Chicago comparison by examining relative racial disparities in the two cities.