What thoughtful rich people call the problem of poverty…

I’ve recently completed two chapters for a forthcoming Oxford University Press Handbook on Economic Inequality. They’re rather long, but anyone who’s interested in a survey of the literature on top incomes, or health and inequality (coauthored with Christopher Jencks and Tim Smeeding), may be interested in them.

This entry was posted in Inequality. Bookmark the permalink.

7 Responses to What thoughtful rich people call the problem of poverty…

  1. Tim Worstall says:

    “Nonetheless, when income inequality
    changes in society as a whole, it is also likely to change in the same direction within
    reference groups composed of co-workers, relatives, or neighbors.”

    Really? I thought that the current complaint was that the top 1% were racing away, leaving the general populace stagnating or advancing only very slowly. As that 1% is most unlikely to be part of the peer group against which people compare themselves, I’d say that might not be all that supportable.

  2. Patrick says:

    Isn’t it very easy to imagine that income inequality is changing in the same manner within the top 1%?

    That is a very big group, after all: 200,000 Australians, nearly 3,000,000 Americans – the only thing special about the top 1% is the catchiness of the phrase!

  3. Andrew Leigh says:

    Tim, same direction – but I agree not necessarily the same magnitude. My top incomes paper shows some results on the correlation between top shares and the gini. Similarly, in the Luxembourg Income Study, the correlation between the 90/50 ratio and the 50/10 ratio is 0.9, indicating that when the 90th percentile is further above the median, the 10th percentile also tends to be further below the median.

    Patrick, we have some discussion of what’s happening within the top 1% in our Australia paper (the one where you didn’t like the comparison groups, if memory serves!).

  4. Patrick says:

    I think I took issue with groups like judges and public servants compared to CEOs, but not using professional firm partners, bankers, or other more apparently high earning groups, from memory, yes.

  5. Andrew Leigh says:

    Patrick, if we had good long-run series on bank CEO wages, I’d agree.

  6. Kevin Cox says:


    One measure of wellbeing used by the Victorian McCaughey Centre is how much cash a person could raise in an emergency – such as buying some medicine. This turns out to be a good indicator of how disadvantaged a person really is.

    I was astounded to find how little money people could raise in an emergency. From memory people were considered to be disadvantaged if they could not raise $100 from their own resources or from friends and family for an emergency.

    Perhaps there is a similar measure that defines the rich.

    Now I know it does not help with trends and looking into statistical entrails etc. but it might tell you more about how rich people behave and the influence they have on society. Perhaps a question might be along the lines of “how much money did you give over the past year to causes, or political groups”

  7. Thinking in old ways says:

    I found the top income paper a very nice summary – my only comment would be that I would have liked to have seen some earlier discussion of an issue you only pick up in your discussion of future directions, that of whether top incomes are permanent or transitory.

    This is particularly important where top incomes contain some components of capital gains. To a certain extent an increase in the share of top incomes which occurs simultaneously with an increase in the volatility of income with different people moving in and out of the top income group, is quite different to an increase in the share of top income where the same group remain at the top. (I have a suspicion, for example, that many of the cockies who got into the top 10% when wool was a pound a pound have not found their way back in since.)

    I haven’t read all the literature you cite on this – but had a quick look at the figures produced by Weber on the IRS Income Tax Return Panel it would look as if there is a lot of mobility especially in the top 1% of tax earners and that a possibly explanation of changes in the top share in terms of changes in the volatility of income should not be ruled out.

    I also wonder what the impact of this is on your Swiss estimates? Are the relative shares of top incomes in Switzerland understated (relative to other countries) because of the effective averaging of income over two years.

    The Health paper and economic inequality paper gave me some problems.

    In part this is because of the complex range of issues you are trying to cover in a very short review article. (Or in your terms too many theories for the available data points.)

    As I see it you talk about a couple of hypothesis on the possible link between income inequality and health – including: absolute income hypothesis (where income inequality enters into the debate because redistribution can be welfare maximising), relative income (place in pecking order/spacing of rungs on ladder – which is just one part of the social health gradient), inequality => crime => victimisation; public choice – either median voter will spend more of rich people’s money, or various inequality => fragmentation – either heterogeneity in public choice, or in fall of trust including in state.

    Along the way I think some bits fall to the side. Because you focus on income inequality and health –many of the aspects of the social gradient of health which are not purely income related (and definitely not causally related) seem to be passed over, or subsumed in a relative income framework (where as you point out reference groups are often more immediate – I earn more/less than my brother-in-law). While in some ways this approach might be seen as reasonable in that you are only interested in the income inequality health linkage, it reduces the overall value of some of the discussion. For example if you at least run through some of the other possible gradient issues (education, willingness to ask questions of professionals, control of work and life , etc) it would give readers a better idea of why a static income health gradient may not necessarily translate to a changing gradient with changes in the income distribution independent of these.

    Put another way taking the Whitehall study would an extra 10 pounds in the salaries of the public servants at the bottom turn them from overweight, beer swilling, smoking, oppressed and frustrated workers.

    In part I see taking this approach also builds a context for the ‘social epidemiologists’ to understand why income distribution may not be the critical issue. In the same way I think Mackenbach summary ‘Income inequality and population health’ in the January 2002 British Medical Journal (vol. 324, No.5) while now getting a bit dated is another useful reference.

    Anyway thanks for the posting.

Comments are closed.