1/ Those of us who have studied the relationship between the size of a city's police force and crime including @emilyweisburst, @mellosteve2 and @ProfEmilyOwens among others are finding our work in the spotlight of late. Here is a tweet which I hope will be useful in sorting

2/ out what we know as well as what we don't know about the effect of police on crime. Views are my own but I am trying my best to summarize what we know absent a political agenda.1) Over the last few decades, when U.S. cities have increased the size of their police force,

3/ crime has declined. The effects are what I'd call modest: A 10% increase in police force size has led to something like a 3-10% decline in crime, depending on the estimate. This includes serious crimes like murder and robbery. 2) But there's more.... when police

4/ deployments change for idiosyncratic reasons (e.g., because of terror alerts), crime declines in the areas receiving additional police presence. And when police are pulled off their regular beats due to non-crime emergencies like traffic accidents, crime rises where police

5/ presence declines.3) There have also been a number of randomized experiments in which high-crime areas in a city ("crime hot spots") have been randomly assigned to receive additional police presence. These studies generally find at least modest reductions in crime due to

6/ police presence which are yet larger when the police actively engage in problem-solving. See: https://www.tandfonline.com/doi/full/10.1080/07418825.2012.673632?casa_token=2XZhCJeVRfAAAAAA%3A542Vb0HyU9clY6Iif3HNBuVdrOtl-kVpbFUIl4Uxwn5V7as_L2HgxULJQLV4YLfNz95pbJE3oWU4) Do

7/ people feel safer/happier when there are more police officers in their community? The evidence here is mixed. Overall, there is more evidence that police are effective in reducing crime than in improving perceptions of community safety and well-being.5) Given that public

8/ safety is sensitive to police force size, does that mean we shouldn't consider "de-funding" police departments? Not necessarily. There is evidence that a number of potentially scalable social service strategies (e.g., summer jobs, CBT) and disorder reduction strategies

9/ (e.g., restoring vacant lots, improving street lighting) can be effective in reducing crime too. Likewise, even well-managed policing will inevitably lead to collateral harms in the unequal world that all of including police have inherited. In the absence of good

10/ management, serious inequities can and do accrue.6) So what does the research tell us about de-funding police departments? My take is that we should proceed with caution. The demography of crime victimization suggests that the same people who might benefit from less

11/ policing also bear the considerable risks that de-funding police departments will entail. Armed with lots of promising research perhaps we can re-allocate away from police towards other crime control strategies, achieving lower crime with less enforcement. But history

12/ suggests that we haven't yet figured out how to do that. And if we fail, it is disproportionately the most vulnerable who will bear the costs.7) Ultimately there is no research that tells us what happens when a city fundamentally changes its policy regime from one that

13/ uses the stick to one that uses the carrot. This has simply never happened in the U.S. Evidence in favor of social service strategies implicitly takes as given business-as-usual policing. Likewise, evidence in favor of policing takes as given business-as-usual social

14/ welfare spending. In thinking about a fundamental shift in how public safety is provided, research can take us only so far.