Centralized Policing Is the Wrong Solution

Elinor Ostrom and the case against centralization

Let's take it for granted that you live in a world where cops can stop you (or worse) for a wide range of petty behaviors, where the culture of law enforcement has been infected with the idea that the police are domestic soldiers, and where federal programs are providing even the most local agencies with arsenals. Which would you rather live under, a small-town police force or a big-city police force?

The conventional wisdom right now is that you're better off with the bigger department. After the Michael Brown shooting in Ferguson, Missouri, and the government's heavy-handed response to the protests that followed, a wave of stories have suggested that the root problem with policing in the St. Louis region is the area's patchwork pattern of local governments. There are 90 municipalities in St. Louis County, plus the City of St. Louis next door; there is a jumble of school districts, fire districts, and other local authorities whose jurisdictions do not precisely match the municipal boundaries; there are hundreds of street associations that serve as purely private "governments" owned and run by their residents. BusinessWeek called this fragmented landscape "a key factor in the tense race relations that contributed to the riots and, perhaps, the shooting itself." The New York Times published an op-ed that claimed consolidating local governments would "empower the black citizens of Ferguson," carefully eliding the fact that such a change would transform Ferguson's blacks from an electoral majority to a minority. "For Anything to Change, Missouri Should Consolidate St. Louis," The Atlantic echoed. The most extreme version of the idea appeared in The Washington Post in late November, when a Los Angeles cop published a piece arguing that the country's local police forces should be replaced by statewide agencies.

Technocrats are constantly calling for consolidated regional governments, so it's no surprise to see them taking an opportunity to do it again. But some libertarians have echoed parts of their critique, not least because these local authorities really do behave atrociously sometimes. In September, my former Reason colleague Radley Balko filed a devastating dispatch from St. Louis County in The Washington Post, showing in close detail how the region's towns squeeze the poor with petty fines and fees. ("I was actually let go as a municipal judge from a town because I wasn't generating enough revenue," one former jurist recalled.) Balko cited the region's splintered system as a core part of the problem: "there are just too many towns, too many municipal governments, too many municipal employees, and not enough revenue to support them." That financial shortfall, he wrote, creates an incentive to squeeze people still more.

That incentive is certainly there, and some towns are shameless about pursuing it. Last week, after months of negative coverage of precisely this practice, we learned that Ferguson plans to deal with a budget gap by issuing more tickets. But it isn't self-evident that centralization is the solution. Indeed, it could make things worse.

Consider a series of studies conducted by the economist and political scientist Elinor Ostrom and her colleagues in the 1970s. These came at a time when academic and political opinion on local power was being tugged in two different directions. On one hand, there was a drive toward merging municipal governments and tightening their professional standards, moving away from the sorts of part-time work and volunteerism that many small communities rely on. On the other hand, a vocal group of dissidents—some on the right, while others came out of the New Left and black power movements—pushed hard for decentralization and community control.

Ostrom, who would later win a Nobel Prize in economics, decided to put the arguments to an empirical test. The results helped convince her that highly centralized government was inferior to what she called "polycentric" systems, in which political units of varying size can cooperate but act independently, without a clear hierarchy. She felt this idea applied to the police as much as anyone else: While economies of scale might make it sensible for, say, a single crime lab to serve a large region, she felt it would be better for that lab to contract with many relatively small departments than to be an arm of one big one.

In the first study, published in 1973, Ostrom and the political scientist Gordon Whitaker took advantage of a natural experiment in Indiana. When Indianapolis and Marion County consolidated into a single government, dubbed Unigov, some towns in the county maintained a separate identity with their own services. Ostrom and Whitaker compared three of those towns with three Indianapolis neighborhoods of a similar size and demographics, going beyond the standard measures of police performance to survey residents on a range of questions about their experiences with law enforcement. They found that the smaller, locally controlled forces outperformed their big-city counterpart on most measures and did about the same on the others.

A host of similar studies around the country followed, some conducted by Ostrom and some not. All of them reached similar conclusions. Two of those follow-ups are particularly relevant to the issues raised this year in Ferguson.

In one paper, Ostrom and Whitaker looked specifically at policing in predominantly black communities. The literature that already existed showed a great deal of black dissatisfaction with the police, part of it related to brutality and racial harassment and part to insufficient control of real crimes. With that in mind, the scholars went to Cook County, Illinois, where they compared three neighborhoods in the City of Chicago to a pair of nearby black villages with very small, largely part-time police departments. The residents here turned out to be less satisfied with their small-town forces than the residents in Indiana had been. But the Chicago police scored even more poorly (and were also more costly). So the same basic pattern replicated itself: Citizens still preferred the locally controlled cops to the larger department, albeit with less enthusiasm.

In the other follow-up, Ostrom examined St. Louis itself. This time the results were a little different: There were some reasons to prefer a middle-sized police force (with 11 to 76 officers on the payroll) to a tiny one (employing 10 cops or less). But it was the big departments, again, that were most likely to have severe problems.

Obviously, a lot of time has passed since these studies were conducted; much has surely changed since the '70s. (It's worth noting that one of the black townships that Ostrom and Whitaker studied in Cook County—the village then known as East Chicago Heights and now as Ford Heights—recently replaced its police force with county-provided patrols.) But all three papers provide strong evidence against the idea that solving our police woes requires some sort of great urban merger. When people argue for centralized systems, they are comparing the warts-and-all world of a cash-strapped suburb to a professionalized ideal that large city governments rarely meet in practice.

At any rate, you hardly need a big city-county consolidation to get rid of a small unit that isn't performing well. There was a time when St. Louis County had even more municipalities, reaching a peak of 98 of them in the '50s. Since then, several towns have dissolved themselves without forcing the entire region to combine.

Albert O. Hirschman famously identified two major ways people can influence an institution, voice and exit. Consolidation weakens both tools: The larger electorate makes it harder for a single voter's voice to be heard, and the larger territory makes it harder to escape bad governance by exiting to a different burgh down the road. The problem in Missouri—well, one of the problems—is that small towns have found their own way to evade both voice and exit, through a method we might call piracy. With speed traps, high court fees, and similar mechanisms, they put a lot of the weight of funding themselves on the backs of people merely passing through town.

Fortunately, there are ways to reel in such practices. Last year the state of Missouri passed a law reducing the percentage of a city's budget that can be drawn from traffic revenue, dropping it from 35 percent to 30 percent. There is now a push to bring it down further, to 20 percent. Other potential state-level reforms include limiting the use of speed cameras, repealing nonviolent offenses that have become opportunities to levy fines, and requiring municipal courts to adopt protections already found in the state's circuit courts. The St. Louis Post-Dispatch, for example, has editorialized that the state Supreme Court should:

• Order courts to cease issuing warrants on the charge of failure to appear unless defendants will receive the same protections they would receive in circuit court, including the appointment of public defenders.

• Order all municipal courts to operate openly.

• Order that in all cases, defendants are brought before a judge within 72 hours, or 24 hours in cases where an arrest is made without a warrant.

If, having lost these sources of money, some local governments then feel the need to merge with a neighbor, so be it. If they cope instead by reducing expenses or by drawing on more honorably acquired funds, that will do too. The point is to change the incentives they face.

That won't end the problems that have fueled abuses across the country, in police departments both large and small. But it would be a start.