Home > Economics - General > One percenters — Crooked Timber

One percenters — Crooked Timber

September 6th, 2010

In my post on EU-US convergence, I found that the US was similar to the leading eurozone countries in both productivity (output per hour worked) and employment-population ratio, so that the difference in income per person is mostly explained (in some cases more than explained) by differences in hours worked per employment person. I didn’t take the distribution of income into account, since the data sources I was using there did not provide anything useful. But commenter Detlef found a blog post by Maximilian Hagemes at the World Bank site which links to a useful paper by Piketty and Alvaredo on cross-country comparisons of income concentration. For most eurozone countries, they show that the top 1 per cent of households gets about 8 per cent of total income (the presentation is graphical, but in any case, there is no point in going for spurious precision with numbers like this). For the US, the most recent data gives an 18 per cent share. So, the share of national income going to the remaining 99 per cent is about 10 per cent smaller in the US than in the eurozone.

There are a couple of ways of looking at this.

Update I was too subtle for my own good in the original version, putting up the idea that the share going to the rich could be viewed as a kind of tax, and leaving it to readers to draw the conclusion that (99 per cent of) Europeans get better returns from the money they pay to governments than do their American counterparts from a system where a large share of total income goes to the top 1 per cent. In the update, I’ve spelt this out.

The data on the top 1 per cent share I used was out of date. The figure for 2007 was 23.5 per cent. At this level, the share of US income that flows as disposable income to the bottom 99 per cent is probably below that of most European countries (the exact outcome depends on how much tax is paid by the top 1 per cent). The recession probably took a bit more from the very rich, but they are already bouncing back, at least on Wall Street.

The simplest is to adjust the comparisons given in my previous post by excluding the income of the top 1 per cent. There we saw that income per hour worked in the US is about 15 per cent higher than for the EU as a whole. But, given the much larger share going to the top 1 per cent, most of this difference disappears for the remaining 99 per cent, taken as a group. Within that group, Americans in the the top decile or top quintile (but not the top 1 per cent) do much better in relative terms than Europeans, while those in the remaining 80 or 90 per cent do relatively worse, as Piketty and Alvaredo show. So, in terms of income per hour worked, most Americans are probably a bit worse off than their counterparts in the EU as a whole.

But of course, the EU is a very disparate place, as is the US. France, Germany, Belgium and the Netherlands are all above 90 per cent of the US level in terms of total income per hour worked, and therefore ahead of the US as a whole once the top 1 per cent is excluded. We can then say that (excluding the 1 per centers) income per hour worked in the US is significantly below that in the leading eurozone countries. The gap is made up by longer hours of work and (to a smaller extent) by higher female participation in market employment[1]

There’s another way of thinking about this that I find a bit more interesting. In some sense, the top one-percenters are competing with the state for the job of running the economy. If they aren’t burdened down with high taxes and income redistribution, so the free-market case goes, they will do a much better job in promoting innovation, better resource allocation and so on, to the benefit of all. And through philanthropy, the top 1 per cent can help to provide public goods, health services and so on that would, in a social-democratic system be the responsibility of the state.

So, the income share accruing to the top 1 per cent may be seen as an alternative to paying more of national income in tax revenue for governments. It makes sense then, to combine the two, to see how much of total market income is being turned over to those who run the economy, fund public services and so on. Conversely, post-tax market income, excluding the proportion going to the top 1 per cent, is what is left over for private expenditure.

As with the earlier post, the answer is that, on this combined measure, the EU and US are fairly similar. Wikipedia cites Heritage estimates that the US tax revenue share of GDP is 28 per cent compared to 40 per cent for the Netherlands, 41 per cent for Germany and 46 per cent for France. Simply adding in the proportion going to the very rich (a very crude way of doing the adjustment, but this is a blog post after all) brings the US share up to 46 per cent, while the other three come out at 48, 49 and 54 per cent respectively

Within the inevitable margin of error, we can reaffirm the conclusion from the earlier post that there is no significant difference between the US and the eurozone leaders on output per hour worked or on employment population ratios. The big differences between the two are
(a) Employed Americans work longer hours (offset by the fact that Europeans do more household work)
(b) In both the EU and US, ordinary income earners receive about half of total market income as private disposable income. In the US, however, a much larger proportion of the other half goes to those in the top 1 per cent, while in the EU it is mostly tax revenue. It seems clear to me that that the Europeans get a better return for their money

fn1. Conversely, other parts of the EU are behind. And while it’s difficult to get really accurate state-level or regional figures for the US, it’s obvious that you can go the other way, and point to US states that are better off or worse off than the EU as a whole or the leading eurozone countries.

Posted via email from John’s posterous

Categories: Economics - General Tags:
  1. Andy McLennan
    September 6th, 2010 at 10:25 | #1

    In comparing US and European tax revenue share, it is important to
    bear in mind that in Europe a greater share of health care and higher
    education is funded by the government, perhaps among other things
    like day care in some countries. (Admittedly total government expenditure
    on health care is greater in the US than in all but a couple European countries,
    but that is a different story.) My general impression is that if you
    look at taxes + health care + higher ed you find quite similar numbers
    across US and Europe.

  2. Peter Whiteford
    September 6th, 2010 at 17:47 | #2


    I approach these issues from the perspective of studies of household income distribution, so that the preferred measure of individual economic wellbeing is household disposable income per equivalent adult. Adjusting for household size is quite important since households at the top of the income distribution in virtually all OECD countries are larger than households at the bottom – being families with children at the top and single aged people at the bottom. Also because the USA has higher fertility than most EU countries and a younger age structure, this will tend to produce different outcomes from the studies by Piketty and colleagues which I understand to be usually derived from tax data and refer to individuals.

    The OECD data in Growing Unequal (2008) show that in 2005 the share of equivalent household disposable income of the richest decile ranges between 20% in Denmark (the least unequal OECD country) and 29.2% in the USA (the most unequal rich OECD country). In France, Germany (and Australia) the richest 10% receive about 23% of equivalent household income.

    So these are significant differences between the USA and EU countries, but nowhere near as large as you get from the unadjusted data for personal incomes that Piketty and colleagues quote.

    As a result when you compare the PPP adjusted equivalent household incomes at different points in the income distribution, you find that in a significant number of original EU countries the bottom 30 to 40% have higher real incomes than the corresponding households in the USA, but at the median, US households are better-off than median EU households except in Luxembourg and the Netherlands (and also Switzerland, and are about the same as in Norway, but these are not EU countries).

    So, overall you find that the top half of US households have higher “real” incomes than corresponding EU households. Overall, therefore, my reading of this related literature is that it is not just the top 1% or 10% who do better in the USA, but closer to the top 50%.

    Also government services (health, education etc) are not counted as part of cash income, and in my view would tend to favour the EU over the USA. On the other hand, private health cover is also very large in the USA and treating it as a fringe benefit would tend to favour higher income Americans – but then higher income Europeans get much more generous public pensions.

    This does not account for hours worked of course. I think that the appropriate way to do this would be to add imputed income from home production to the household income concept, which as you suggest would tend to favour Europe.

    But on the other hand (!), imputed income from owner-occupied housing would probably tend to favour the USA.

    As far as I am aware no one has done a study that incorporates all these possible extensions of the household income concept and looked at the effects on distribution of outcomes.

    On balance, however, I would tend to think that a fairly sizeable group of US households – significantly more than the top 10% but probably less than the top 50% have higher real household incomes than corresponding EU households even after all these adjustments.

    But it still sounds like a good research agenda!

    Also of course, the GFC appears to have had a much greater negative effect on US households in terms of unemployment and lower housing and pension wealth than in most European countries, disregarding Ireland and Spain.

Comments are closed.