For a more down to earth example consider professional sports. Back in the day professional sports players earned high yet “reasonable” salaries. As new technologies increased the revenues of the sports the highly talented players came to demand a larger share of the pie. Today top tier players in any major sport earn astronomical salaries while lower tier athletes struggle or make nothing at all. This is supremely unequal, yet nobody is clamoring for super-stars to take “fair” salaries. They have the talent, they should earn whatever that talent can get them.

That the best argument this piece can come up with to illustrate the damage of high income inequality is “stress” shows that this problem may be a tad overrated. While it is sensible policy to ensure that wealth does not beget wealth w/o regard to talent (ie the estate tax), in a world where single individuals can create outsized value is greater inequality really a problem?

What people need is equal opportunity to rise according to their talents and in America I don’t really see any move by those with wealth to shut out deserving individuals so that their otherwise undeserving offspring can maintain power like some feudal lord.

Specifically in hte proposed mechanism by which inequality produces these bad health outcomes: status.

For Japan has a low income inequality and yet is, compared to most other societies being mentioned, hyper-sensitive to status.

So from the income part, we’d expect Japan to be on the “good” end of all of the hypotheses….which it pretty much is (although Wilkinson unaccountably drops Japan when comparing inequality and working hours). But if we take the suggested mechansim through which it all works, Japan should be at the “wrong” end. Which it isn’t.

Which leads to the suggestion that there’s something screwy about the mechanism at least.

If you would have read the book, you would also have been updated on the extended discussion of the differences and importance of relative poverty vs absolute poverty.

This distinction is important because there is no policy option that reduces inequality while keeping total wealth fixed. There is generally a trade-off between equality and economic efficiency.

To Nymnchen I would respond that the quality of statistical reasoning employed in the medical profession is sometimes very low indeed (see for example the literature on publication bias). Indeed, medical professionals know a great deal about medicine, but not necessarily about statistics.

In order to have an opinion of the quality of science, I advice you to first read and understand it. I understand from this blog that a number of people havent come that far in the discussion.

(When it comes to Hong kong there are numerous indeptht discussions of how this citys population is made up by the upper crusts of both the chinese and the brittish societies. Since it has been concluded that it is our socioeconomic staus as children that will influence our health and behaviour for the rest of our lives http://goo.gl/gT3r , it is of no surprise that the inequalities of Hong Kong have little negative impact on its population. Similar conclusions can be made for Singapor)

For instance, Hong Kong & Singapore have some of the highest levels of income inequality but are nearly the best in the world for having low infant mortality rates and low crime.

A theory of everything that explains nothing

In the edition of The Spirit Level I have, the discussion about the causes for the correlation, the social stress, is quite elaborate. Firstly, the authors conclude that “stress” is a poor description of the phenomenon, since stress is a widely used term, but decide to stick with it in lack of better terminology. Second, that this kind of social stress has direct and measurable physiological effects has long been known in social medicine (lately, but not exclusively reported in for example “If your shoes are raggedy you get talked about”: Symbolic and material dimensions of adolescent social status and health. Sweet E. Soc Sci Med. 2010 Mar 16. [Epub ahead of print]). There are obvious medical explanations for a variety of physiological and psychological problems seen in the wake of inequality. So, social hierarchies, and especially a low standing in a social hierarchy, cause well known effects on medical conditions (observed in both human and different animal populations in different experiments and studies) and ultimately health. We have an effect as show by Pickett and Wilkinson (and not only them, they refer to 170 different studies in my copy of the book). As a medical scientist, cause and effect, the causal correlation, is beyond question. What we can debate is the importance of this effect.
As I pointed out, that the poorest 20% in a society will suffer a bad health from inequality is a prediction made by medical research long time ago, and not in any way controversial. The much more interesting conclusion reached by Pickett and Wilkinson is the one where they claim that social stress is problematic for the rest of the society, also for the richest 20%. There are great implications of this conclusion and I hope we will see more research on this subject.

I am a medical scientist and the more I read how economists discuss The Spirit Level, the more I doubt you are even qualified to understand basic medical reasoning. As I have learned, conclusions in economics can be disregarded and accepted as a matter of taste. This is not the case in medical research. You can not, with maintained credibility, ignore conclusions based on large amounts of published data without offering an alternative explanation that fit the entire set of data better, or atleast as well. (then you can ofcourse claim that the data is invalid for some reason, but since it is the same data as all other makroeconomists use, I havent seen any attempts to this). Posts like the above puts the ability of economists to analyze medical and especially epidemiological research in to question. Not surprising, maybe, since extensive training is required in order to become an epidemiologist. However, I think economists in general would benefit from being more humble in the evaluation of the kind of research presented by Pickett and Wilkinson, since after all, epidemiology, as opposed to economics, have proven a powerful instrument to improve wellbeing in society. I think that in order to offer scientific criticism of the conclusion in The Spirit Level, it is instrumental that the original peer-reviewed papers referred to within can be read and understood. (in addition to the book, I may add). So, but failing to offer an alternative, testable hypothesis to the explanation offered in The Spirit Level, you are demoting your role in the discussion to mere producers of opinions.

On another note, I think it is very welcome if the discussion about The Spirit Level increase the criteria of what can be regarded as causal evidence. I look forward to seeing future (and past) scientific conclusions in the field of economics being analyzed according to these new criteria for causality imposed on the conclusion of The Spirit Level.

This is a very interesting post. I suspect there are a range of things going on and your second chart is a better representation of reality than the first chart from Wilkinson and Pickett.

One problem with the first chart is that Japan is not a very equal society and indeed never was. There was an article in the Review of Income and Wealth from the late 1980s which showed that the view that Japan was a really equal place was based on surveys that excluded single person households but also excluded the self employed including farmers i.e. getting on for 40% of the population. Moreover, income inequality in Japan has risen from whatever it really was, in part due to the ageing of the population and also because of the casualisation of the younger workforce. So we should move Japan from the left hand side towards the right hand side of the chart. It has high life expectancy, but it is not a low inequality society – and my understanding is that the same is true for South Korea and Taiwan.

But also why is life expectancy in Denmark so much lower than in Sweden despite the similar levels of inequality ? Well my understanding is that people in Denmark smoke a lot more than people in Sweden.

Dalton Conley, in the linked American Prospect article, says “A few years ago, there seemed to have emerged a cottage industry on the deleterious effects of economic inequality on health”. He proceeds to argue that links between inequality and health are always found to be mediated by other factors. He says: “Other studies have shown that U.S. states with more inequality are less likely to be healthy. But what really makes the difference between states are policy differences — like minimum wages, welfare rules, ease of divorce, and speed limits — which suggests that, like countries, whatever states do to produce equality also produces population health; equality itself does not produce better health”.

This looks like a distinction without a difference. If those policy differences both produce more equality and better health, then go ahead and implement them. You will have better health and more equality, and, if Mr Conley is right (in the later part of his article), you’ll also have more honest politics.

Frankly, if studies linking inequality and poor health are a “cottage industry”, articles like Mr Conley’s look like a “palace industry”, aimed at maintaining the status quo with quibbles. The approving millionaire reader can then go, after finishing reading what to him/her is a feel-good piece, to the exclusive dating service advertised on the same page of American Prospect, where an outfit called Elite Meeting – Romance For the Successful and Attractive promises access to its “membership of successful and attractive singles”. What a life!