Along the same lines of reasoning as yours, approximately 1/2 our murders are committed by black Americans, and the proportion of people who are black are well correlated with murders. So one could equally well say that the murder rate is high because of the presence of black people. Would you endorse this conclusion? If not, why not?

(Let me emphasize I'm not endorsing the conclusion I derive here - I simply plan to repeat whatever argument jaquesm has against the above conclusion back at him w.r.t. guns.)

IF this was true, we would see a correlation between gun ownership (or access) and murder rates. If anything, the correlation is slightly negative, both in the US and globally.

Where you DO see the correlation is inequality. In US states, using data from wikipedia, I find a 70% correlation between gini coefficients and homicide rates. The same is true globally.