The inequality hypothesis: thesis, antithesis, and a synthesis?

In the pages of this journal Min Hua Jen, Kelvyn Jones and, Ron Johnston recently presented a paper claiming to have evaluated Richard Wilkinson’s hypothesis that social inequality damages the health of populations (Jen et al., 2008). It was an interesting paper with findings not originally recognized in Wilkinson’s hypothesis, but it suffers from one major flaw: the findings are not a direct evaluation of Richard Wilkinson’s hypothesis. We show why below. It is important to point this out as the balance of much very new evidence now points towards inequality having damaging effects on society in all kinds of ways (Wilkinson and Pickett, 2009), not just in terms of health (but for a review of that see Ram, 2006) and it would be unfortunate if the value of these new findings were cast into doubt by studies that did not evaluate the actual hypothesis, but something else, which in turn turns out to be quite interesting. In hindsight it is easy to see how Jen et al. could have thought that they were evaluating the inequality hypothesis. Part of the argument of their paper is that quite complex methods of synthetic data creation and multilevel modeling are needed to evaluate the inequality hypothesis. This might be the case, although others find the existing evidence in support of Wilkinson’s hypothesis to be near overwhelming (Dorling et al., 2007 provide an argument for extending his findings to poorer countries). However, it is possible that it was dealing with that complexity of synthetic data and multiplicity of hypothesized levels which led to the very simple oversight we highlight below. We should note that our response is far from being a simple critique, as we find Jen et al.’s analysis and findings very interesting, just in a way that they didn’t. This is thus an attempt at a constructive critique and we hope it is seen in that way and