Yes, and the operative statement used in the study by Parker and Vadheim is "short term". From a utilitarian perspective, of which Worstall is a blind advocate, it makes sense to buy from warlords who are committing lower levels of violence against civlians in case they move on and start committing violence elsewhere.
Of course, as it is with economics and sociology, a few caveats are added to the paper to make sure that what they're not saying should be taken as conclusive. For one, the figures "imply" and "suggest" the above, as "there is almost certainly measurement error in the ACLED conflict data". Secondly, the paper notes that, to put it in terms Worstall can understand, correlation does not equal causation. Confounding factors to the thesis include:
- increased reporting of violence enabled by Dodd-Frank (enabled by removing the financial base of static warlords and meaning they have to move into other territories where they have less control over the population).
- movement of workers into other regions where the government may have less control (qualitative data included in the report suggests that the safety of workers was compromised during the move; the paper says that Ndjingala is "nearby" Bisie, but the reality is that Ndjingala is a day's walk from Bisie, one that has to be completed on foot).
- commodity price increases that have led to the increased violence (several examples are cited, including from tin and coltan).
What you have to remember is that this paper is designed mostly to support a theory: that policy, not price shocks are responsible for the increased violence. This monocausal explanation is not tenable at any level of policy study. It is more likely that there are multiple explanations, all of which in some way contributed to the measured increase in violence.
Dodd-Frank has actually done what it was designed to do - it has put a stop to Western companies using exploited labour in the developing world by adding systems of accountability to a previously murky supply chain. The reality is that policies that are designed to deal with one problem often have other effects. Of course the warlords and gangs were going to find other ways to supplement their income - but the question remains as to whether or not that's actually happened as a direct result of the policy changes alone, or as a result of a combination of factors.
But regardless, there is no justification for continuing to buy from warlords who commit violence against civilians. Dodd-Frank was supposed to put a stop to this enabling behaviour. And it did.