Do Police Make Us Safer?

W. David Ball

Americans regularly believe crime is rising nationally, even when actual crime rates are decreasing. We are wrong about that. But what if we are also wrong, fundamentally, about what police do? What evidence do we have that policing strategies are effective, or that they are the most cost-effective way of creating a safe society?

Our ideas about how crime can be best addressed are not necessarily based on sound evidence. Consider “broken windows policing,” the theory that signs of disorder give rise to crime, which was used to justify large-scale stop-and-frisk, zero-tolerance policing programs. In the 1982 article which laid out the broken windows theory, support came from a Philip Zimbardo experiment in 1969 that involved a car with its hood up in exactly two places: the South Bronx and Palo Alto. While the car in the South Bronx was stripped quickly, the Palo Alto car was untouched for a week—only when Zimbardo “smashed part of it with a sledgehammer” did the car get further damaged. This did not stop police departments across the country from rolling out massive stop-and-frisk programs, but it should not be surprising that there is "no support for the idea that broken windows enforcement activities… represent the optimal use of scarce government resources."

The benefits that police do provide do not extend to all citizens equally. In Los Angeles, whether a homicide results in an arrest--not a conviction, just an arrest--depends on what race the victim is. When an African-American or Latinx person is killed, an arrest is made less than half the time.

There are also costs to policing—not just in terms of budgetary resources, but in terms of mental health, feelings of safety, and, ironically, crime fighting. This goes beyond police killings of citizens, which, like so many of the costs of policing, are borne marginalized groups. When arrestees are detained pretrial (due to failure to pay bail), they cannot work, which means they are much more likely to lose their leases (especially since almost half of Americans do not have $400 for emergencies). For young people, “Contact with the justice system … promotes delinquency.” To repeat: treating kids like criminals is a self-fulfilling prophecy: they grow into the way police treat them. Jails are a major vector for the spread of Covid-19, and could lead to 100,000 more deaths than current estimates. Some of these harms—say, those from pretrial detention—are both certain and widespread. Time in jail will certainly result in those harms. Crime victimization is speculative and rarer—depending only on one’s ability to imagine “what if this person gets out and hurts someone.”

We need to understand that police are not the only or even the best way of responding to criminal problems. A large-scale study of Florida counties found “no evidence” that increases in prison population growth are accompanied by lower crime rates. The recidivism rate of those who have been in prison is “stubbornly high.” Prisons and jails don’t make people better. As radical as it may sound, there is no single model of policing, and there have been many different models of policing in this country. We did not have professional police at the founding of the country, and the first professional police force was not established until 1838. Before then, citizens were organized in the watch system. This is not to suggest that these alternatives were benign. The watch system gave policing power to those who had power in the first place—white men. In the American South, slave patrols served as, essentially, organized forces of lawlessness, terror, and torture. The reality of policing is complicated, and history shows us that it is not enough to have different policies and procedures: what really matters is who executes them and how, and how they are kept accountable.

William Blackstone argued in Commentaries on the Laws of England (1765-69) that “preventive justice is, upon every principle of reason, of humanity, and of sound policy, preferable in all respects to punishing justice.” Two hundred and fifty years later, we should reconsider his approach. How much better would we all be if we focused on what causes violence, rather than simply reacting to it? We know from programs like the Perry Preschool Project that investing in children pays for itself and has poverty-reducing effects that last for generations. Giving poor African-American children two years of quality preschool resulted in much lower rates of arrest and reoffending. We know that community nonprofits were key drivers in reducing murders, violent crimes, and property crimes. We know that childhood trauma results in all kinds of adverse health outcomes: depression, obesity, and substance use disorder. These traumas—physical abuse, but also parental alcoholism or mental illness—should be addressed directly. Nobody is healed in the criminal legal system, yet about half of those in prisons and jails have mental health problems, and 65 percent of prison inmates meet the DSM-IV medical criteria for alcohol or other drug abuse and addiction.