Projecting, Estimating and Whispering about Venezuela’s Homicide Rate

David Smilde

Venezuela’s Attorney General is getting some deserved applause for actually providing the AN with homicide statistics for 2015, something that has not happened in years. She said there had been 17,788 murders in 2015 putting the homicide rate at 58 per 100,000. That is bad as it is, however, her numbers are being criticized as low, in part because she did not include “deaths under investigation” and those killed in the context of “resistance to arrest.”

In January, the Observatorio Venezolano de Violencia (OVV) put out some projections for 2015, saying Venezuela had suffered 28,000 murders, putting the murder rate at 91 per 100,000.

However, I do not think the OVV’s numbers should be included in the same category as estimates based on 2015 data. Their projection is  based on 2013 and 2014 data and assumes continuity. But the whole idea of collecting data is to allow it to mark breaks in continuity and judge if things are getting better, worse or staying the same as before. Simply projecting from past data precludes that possibility. It is a justifiable exercise given the dearth of official numbers. But it should be remembered that it does not represent new data that can be used to discuss the evolution of crime in Venezuela. Rather it repeats what we know from previous years.

The Mexican Citizen Council for Public Security and Criminal Justice (Consejo Ciudadano para la Seguridad Pública y la Justicia Penal- CCSPJP) recently named Caracas the most violent city in the region. However this was based on one article in which a Venezuelan journalist cited unnamed CICPC sources. The latter suggested there had been 22,748 murders in 2015 for a rate of 75 per 100,000. But this also does not include “deaths under investigation.” It also likely does not include deaths in the context of resistance to arrest.

The last year we have figures that breakdown homicides, deaths under investigation, and resistance to arrest was in 2013. The figures for 2013 and previous years show that these latter two categories add approximately 50% to the total (see page 439, of PROVEA’s 2013 annual report). If 50% were added to the total of of the unofficial data from the CICPC that would actually give us a number of 34,122, pointing to an astronomical murder rate over 110 per 100,000 and surpassing OVV’s estimate.

If we do the same operation with the Attorney General’s numbers, adding another 50% to include resistance to authority and deaths under investigation. We would get 26,682 for a murder rate of 88 per 100,000 persons. This is ironically close to OVV’s projections, as Dorothy Kronick suggested in the Insight Crime article (I was actually off in that article when I suggested the CCSPJP figure was probably closest. At the moment of the interview I didn’t realize the figure didn’t include deaths under investigation and resistance to arrest.)

It should be noted, however, that simply including these two categories is going to overestimate the murder rate by an unknown percentage. Certainly some “resistance to authority” deaths are justifiable and should not be considered homicide, and certainly not all “death under investigations” are actually homicides.

In the coming weeks there will be some scholarly estimates of the murder rate, including the first release of data of the 2015 Victims Survey led by Luis Gerardo Gabaldón. When that happens I will revisit this issue.