Do federal stats undercount police killings?

Read the Full Transcript


    The nation's been focused on Staten Island, Ferguson, and Cleveland in the last few weeks as citizens and law enforcement assess how they have and how they should deal with one another.

    This as a new investigation by the The Wall Street Journal reveals that accounting for killings by police might be grossly underestimated.

    Wall Street Journal reporter Rob Barry joins us now. So, how did you do your reporting, and what did you find?


    Thanks for having me. What we did was we asked about 105 departments to give us the number of people who have been killed over a five or six year period.

    And we compared those numbers to what had reported to the FBI. And we found that there was a lot of stuff that wasn't in the FBI's information.


    You said that at least 550 police killings between 2007 and 2012 never made it onto the books?


    Yeah, and that's only among the top 105, 110 largest agencies in the country.

    So there's 18,000 jurisdictions. So you know, that's just a small estimate of the total.


    OK, so for example, some jurisdictions could call something a justifiable homicide versus an unjustifiable homicide? Discrepancies in definitions? What do you mean?


    Yes, it was a wide range of things. That was certainly one of the issues there, that what we're dealing with here are essentially crime reports.

    And agencies who are forced to report information about unfortunate events where officers take someone's life. They don't really want to include that in a crime report.

    It's not a crime in their eyes. It was a justifiable event. So there was some concern by some agencies about that issue.

    There's also a lot of other issues involved. You had technical issues. So you have departments which, at least they told us, they thought that these things were being reported, they thought they were participating in this program.

    But then when they went in and they looked into it, when we said to them, "hey, here's the numbers that you reported," maybe there would be one, but they told us 10.

    They'd look into it and they'd say, "Oh, well it turns out that we haven't been keying it in correctly." So that was another issue.

    And then I think the largest issue was that three of the biggest states in the country – New York, Florida, and Illinois – have almost nothing reported.

    And that's because of how this process works. What happens is that when an agency wants to send this information to the FBI, they pass it through a state agency first.

    And in all three of those states' cases, there are issues, varying issues, with the way that the states then turn around and pass it up to the FBI.

    Such that there's no information about justifiable homicides from any of those states.


    What you're also saying here is that not everybody has to report.


    It's voluntary, exactly. And when you're dealing with 18,000 law enforcement agencies across the country, a lot of them small, a lot of these agencies only have five, 10, officers at them, these events are very rare.

    Reporting them is just not built into their process in many cases.


    So we've got different definitions, we've got 18,000 jurisdictions, it's all voluntary, and I'm assuming some of these departments are pretty sensitive about this information in the first place.


    Yeah, of course. I mean, each of these things are inflammatory or potentially inflammatory events.

    So there were concerns – a lot of departments asked me when we went to them and said, "Will you provide us with the number of incidents you've had," quite a few departments answered first saying "Why? Yes we'll give it to you, but just tell us why you want this."

    There's a lot of concern about it being used for comparisons. And I mean, you know, when we see what's happening across the country right now, I think that you can understand that concern.


    And your reporting is not saying that all cops. You're just pointing out that there's this discrepancy in how we're reporting the information.

    So if there are these gaps in the data set, how do we make any policy based on maybe faulty numbers?


    That is a great point. And that's why we were looking at this in the first place.

    We wanted to get a benchmark. We wanted to know how often does this happen and who does it happen to?

    When we tried to do that using the available information and took it to experts, everyone said well, you can't really do that.

    And we said, OK. And that raised the question of why not?

    In terms of solutions as to what we'd do from here, it's a complicated problem. And we've discussed most of the reasons why.

    Some of – some of the people involved in this are working now towards coming up with incentives, financial incentives from the federal government or some sort of mandatory process that would require these things to be reported.

    But from what I've been able to tell at this point, there's nothing concrete on the way.


    All right, Rob Barry from The Wall Street Journal, thanks so much.


    Thank you for having me.

Listen to this Segment