Tuesday, 26 July 2011

Reflections on São Paulo H1 crime figures 2011 vs 2010



Following a scandal over the sociologist Túlio Kahn who is accused of supressing his own crime figures whilst clandestinely selling the data to private companies, Geraldo Alckmin, the governor of São Paulo, anounced that forthwith crime figures would be published monthly and broken down by district as opposed to every 3 months at city level only. That was back in March 2011.

Now that the full first semester is in we have the usual articles comparing 2011 to 2010 and yesterday a good few were celebrating a 29% fall in homicides in São Paulo capital. Now, don't get me wrong, a drop in murder rate is usually a good thing but there is some detail in the stats that doesn't make as nice reading as the 29% fall headline would seem to imply:

  • First up, violent crime may have dropped in the last few years for various reasons and some may have dropped versus 2010 which is all good but they have dropped from an already very high level - however big the city is if you compare the below figures with the same 180day period in most western countries your per capita incidence metric will be through the roof. 

    • Although murder fell by 29%, attempted murder (which is similar in terms of intent but perhaps lacking in execution (excuse the pun, poor taste I know)) rose by 18% which suggests that the murder drop is not as dramatic and part of it is perhaps coincidental. Either that or the assasins are getting clumsier. 

      • Murder is not the only crime on the list. Drug trafficking is up, as is car theft and GBH. As you can imagine these were not exactly low in 2010 but they all increased. Car ownership did increase too so we would have to look at whether the rise is proportional (i.e. the risk to car owners is the same) but in absolute terms in went up 

        • Another interesting one to notice is that the incident of hostage/kidnapping for ransom is actually statistically very low. If you were to judge by the conversations of the wealthier residents of São Paulo they would have you believe this was extremely likely to happen, everyone has a story they have heard about kidnapping, but if you look at the stats there were only 13 cases in H1 2011. Unlucky number for some but contrary to the media hype around it quite a low number for a city of 20m. So maybe a nice topic to sell papers with but not necessarily the most worrying criminal activity in SP.

          Full table below (sorry, no translation this time)

          Ocorrências na capital paulista 1º semestre 2010 1º semestre 2011 Diferença
          Nº de vítimas de latrocínio 32 47 47%
          Latrocínio 32 46 44%
          Tentativa de homicídio 480 567 18%
          Furto (outros) 83.320 98.112 18%
          Tráfico de entorpecentes 2.802 3.263 16%
          Lesão corporal dolosa 17.038 19.765 16%
          Roubo de veículos 17.414 18.796 8%
          Roubo a banco 76 78 3%
          Furto de veículos 21.106 21.076 0%
          Estupro 1.210 1.186 -2%
          Lesão corporal culposa por acidente de trânsito 13.422 13.002 -3%
          Roubo de carga 2.186 2.108 -4%
          Roubo (inclui roubo de carga e banco) 56.676 54.026 -5%
          Homicídio culposo (inclui mortes em acidentes) 460 428 -7%
          Homicídio culposo por acidente de trânsito 426 378 -11%
          Nº de vítimas em homicídio doloso 596 501 -16%
          Homicídio doloso 568 470 -17%
          Extorsão mediante sequestro 16 13 -19%
          Lesão corporal culposa (outras) 998 799 -20%

          2 comments:

          IsaacRiquelme said...

          i wonder how do they make the estimation about kidnaping (among others)i guess we should be sceptical about this numbers, and undertand the wholeprocess of making a report in SP, Anyway very interesting to see how they do classify the crimes which ones grow and wich doesnt. Today I read a story in the red bar... Nice to read from this place.

          Pierre Larose said...

          Well there's obviously a gap between reported crime and actual crime but presumably the gap is smaller for more serious crimes: murders and kidnappings for example. You're unlikely not to report a murder or a kidnapping right?

          As for political manipulation that's a whole other issue but the fact they are now published publicly and broken down in detail makes you trust the figures a little more.

          Post a Comment

          Related Posts Plugin for WordPress, Blogger...