Mexico's government argues that killing and arresting the leaders of drug cartels does not lead to increased violence. But is the "myth-busting" by President Calderon's security team just another type of spin?

In a new public relations campaign, the Mexican government is out to question the top 10 "myths" of the violent war against the country's drug cartels. According to government security spokesman Alejandro Poire, these are frequently repeated "arguments, opinions, phrases" which are "false, and lack basis in reality."

Posts on the presidency's official website have already addressed some of these "myths": that there is no strategy to the government's security policy besides the use of force, and that the army is taking over the duties of the police, leading to increased human rights violations.

Now, for its third "myth," the government addresses the idea that violence has risen in areas where Calderon deployed the military. The latest post, accompanied, like the others, with an animated video, (see below, and illustration), appears to be a response to criticisms of Calderon's so-called "kingpin strategy;" which focuses on taking out the heads of drug trafficking organizations (DTOs).

The debate has been driven in part by two articles published this year in Nexos magazine: one by Jose Merino, a professor at the Autonomous Technical Institute of Mexico (ITAM), and another by sociologist and crime analyst Fernando Escalante. After examining and compiling national census data, both found that homicide rates grew disproportionately in states where the government sent in the army.

The government's counterargument seems to be this: in states where the Mexican security forces were deployed and were able to kill or arrest top-level operatives of the drug cartels, this, for the most part, had no impact on already rising murder rates. In some cases, the government argues, the increase in murder rates actually slowed down after a drug capo was taken out.

The usual critique of the kingpin strategy is that eliminating the heads of DTOs creates a power void, leading to the fracturing of the organization and thus revving up homicide rates as rival groups battle to inherit the goods. The theory is used to explain why violence is now spiking in Mexican states like Tamaulipas and Nuevo Leon, where the arrest of Gulf Cartel leader Osiel Cardenas led to an eventual split between the group and its armed wing, the Zetas. In Chihuahua, Mexico's most violent state, a similar phenomenon occurred with the split between the Sinaloa Cartel and the Beltran Leyva Organization.

The Calderon team has a different interpretation: taking out drug kingpins does not lead to increased homicide rates in states that are already among Mexico's most violent. By this theory, the army are sent in to states where violence is already increasing; their presence is a product, not a cause, of rising murder rates.

The presidency's website lists ten cases -- including the deaths of "Tony Tormenta," "El Chayo," Arturo Beltran Leyva and "Nacho Coronel," and the arrests of "El Teo," "La Barbie," "El Amarillo," "El Ingeniero" and Eduardo Arellano Felix -- in which the security forces appeared to successfully "decapitate" a DTO. Only three of those cases accelerated the increase in homicides in the DTO's area of influence, the government argues.

But the analysis has some problems. The government only compares homicide rates six months before and after the removal of these kingpins. This is an extremely narrow window of time to make such comparisons. As InSight has pointed out, often the effects of removing kingpins can be delayed for years, as was the case with the Gulf Cartel and the Zetas: the two factions did not formally split until almost three years after Cardenas' extradition to the U.S.

The government also chooses to use selective data. Mexico has three sources of statistics on homicide rates: the National Institute of Statistics and Geography (INEGI), which, like Colombia's Medicina Legal, uses death certificates for its figures; and the National Public Security System (SNSP), based on police reports. Finally, there is the federal government's Associates of Organized Crime (ACO). This appears to be the presidency's source for data in the "kingpin" blog post.

To be precise, the blog cites the ACO's data on homicides related to "criminal rivalry," its new term for data on killings related to organized crime.

The government's use of such definitions explains why the data in the blog post is so markedly different from the murder trends identified by INEGI. The government's analysis maintains that in states like Morelos, the rate of killings linked to organized crime actually slowed after the death in 2009 of Alberto Pineda Villa, alias "El Borrado," a leader of the Beltran Leyva Organization. However, as INEGI's data points out, total murders in Morelos jumped from 317 in 2009 to 559 in 2010.

The Calderon administration may be right to question whether there is a direct, causal relationship in the deployment of the military to certain areas and the subsequent increase in violence. Explaining the worsening security in states like Chihuahua, Nuevo Leon is complex, and it would be wrong to say that the deployment of the military to these areas is solely responsible for soaring violence. But, as some analysts have showed, there does appear to be a correlation between the reliance on the military and police to deter DTO activity, and the escalating bloodshed.

By attempting to cite "successful" examples of the kingpin strategy, Calderon's team has not built a very convincing counter-argument. Instead, it appears that the government has picked a definition of "murders related to organized crime" that best suits them, then picked a time frame that supports their hypothesis. It does make for a good example of how best to "spin" a damning set of data.