NATSEM at the University of Canberra has released a report [pdf] on trends in spatial socio-economic inequality from 2001 to 2006 [via Peter Martin]. As Martin suggests, it’s a useful corrective to claims lacking nuance that the Howard years saw a rising tide of prosperity lift all boats equally. There were definite winners in the household income stakes, as this graph demonstrates:
I wonder how much of the extra growth for the top 10% can be attributed to tax policy, particularly the cuts in the top marginal rate of income tax?
Robert - that’s gross income in the graph above isn’t it? So tax changes are not taken into account. I would have thought an after-tax income comparison would have been a much better comparison - one which takes into account both the big tax cuts as well as the huge amounts of middle class welfare that were introduced by Howard.
Also because of the progressive tax system, someone in top decile getting a before-tax increase of say 35% will end up with (handwave) a 20-25% increase in net income, whilst someone on a low income who gets a 25% increase in gross income will get (again big handwave) an 18-25% increase in net income. And in tax cuts over the 5 years and things like baby bonuses, LITO changes etc and who knows what it really looks like?
Chris and Rob, yopu each seem to be assuming that the graph refers to deciles of household income, when in fact it refers to ’spatial’ deciles. That is, it ranks Australia’s neighbourhoods in deciles from poorest to most affluent, then shows that the average gross income in our poorest neighbourhood rose 30 percent while average household income in the richest neighbourhood rose 36 per cent.
Snapshot figures of spatial inequalities is fraught because, basically, people move around, they come and go, and there may always be some ’sorting out’ as peoples’ incomes change, whereby lower income people and higher income people end up in different neighbourhoods - this is what real estate markets tend to do, or government policies that prioritise ministry of housing allocation to low income people, especially welfare recipients, where there are still large concentrated stocks public housing in use.
This dynamism means that to say the average income of households in a certain neighbourhood rose by X per cent doesn’t tell us much about how actual incomes of particular households rose if the neighbourhood is now composed of different households from when the last snapshot was taken. There’s no evidence that the people in the poorest (or richest) decile of neighbourhoods in 2006 are the same people who were there in 2001.
No surprises here if you’re on a pension.
Its hard to say how much low income people move about. It would depend if the rent went up and they had to look for a cheaper place, and if theyt were trying to buy a place (there are some) it would depend how thry were affected by interest rate moves. Especially when the rates were going up. And once the baby bonus is used up, thast’s it. so far as its impact is concerned.
There would be differences depending on where you lived too. The relatively low rent one pays in some regional areas, eg Armidale would not apply to the metropolitan areas, or, for that matter, to all rural areas. And prople on fixed incomers or low wages would be worse off because of rising energy, petrol, food proces, even allowing for the CPI being indexed.
Well, that’s my penny’s worth on it.
So mobility would have to be another variable? And not just spatial, also ‘class’ mobility. So it’s really a whole set of variables.
Something must have been going right - the lowest increase for any group was 28.4%.
Also interesting that the graph scale has been cropped on the y-axis to exaggerate the visual impact of the increase for the top 20%. The average looks to be about a 30% increase - the top 20% are up about 35%. So the difference is less that 1% growth each year. Not much really.
I wonder whether the richer ’spatial’ groups had higher increases because of asset inflation (housing and equities)? I wonder whether the next few years will show them below trend as the asset bubbles deflate?
“The relatively low rent one pays in some regional areas, eg Armidale would not apply to the metropolitan areas, or, for that matter, to all rural areas.”
I’ve wondered about this, but haven’t looked closely at the figures. I remember that friends of mine living in Bathurst were paying very high rent for studio apartments compared even to the inner city (in Sydney) a couple of years back. I wonder if more supply would lead to relatively lower rents in (outer) metropolitan areas, when compared to regional? It probably depends on the size of the regional centre.
Kim
Contrary to your claim, this graph seems to confirm Howard’s rejection of the cliche that under his government the poor went backward and that old chestnut ‘the rich are getting richer and the poor getting poorer’. In fact, this graph is quite astonishing in the picture of egalitarianism it paints. I wonder if similar graphs exist for the Keating years?
Paul, it’s not just a question of how much a static group of ‘low income people’ move around, but that people’s income itself moves around (you lose a job, you get a job; you divorce, you repartner; you’ve got dependent kids at home, the kids move out or start earning a decent income of their own etc). As I suggested, some aspects of real estate markets and/or government policy work to translate this income mobility into spatial mobility: this would be one reason why the composition of spatial deciles would change between census dates.
I would be very interested to know how much income mobility there really is. The classic accounts suggest virtually none, even intergenerationally, but there is probably more now than a hundred, or even thirty, years ago.
I’ve seen stats on both sides of this argument, and this lot also has me reaching for How to Lie with Statistics. And that’s just the attempt to show whether people are worse or better off! There’s more to this social justice than this.
How easy is it for people to move up (or indeed down) the SE scale? My feeling is that is isn’t very easy for people to move up off the bottom, and that the Rodent wasn’t terribly interested helping, but I have Blind Freddy next to me rather than any real stats.
“I wonder whether the richer ’spatial’ groups had higher increases because of asset inflation (housing and equities)?”
Andrew, for Census purposes, gross household income means exactly that: income (wages, business income, rent, dividends, interest, government benefits, super, child support, workers comp)
“I would be very interested to know how much income mobility there really is. The classic accounts suggest virtually none, even intergenerationally”
Klaus, I don’t think the classic accounts (are you thinking of Goldthorpe etc or earlier?) really tried to measure income mobility. They tried to measure class mobility or occupational mobility, along the lines of whether the kids had the same job as dad. The low level of mobility in this sense may well persist (I’m not up with the latest sociological data), but could in any case be quite consistent with a high volatility of income measured over relatively short periods whereby between one census date and another people move into and out of income deciles or above or below some notional poverty line etc.
Yes you’re right Kim, 30% is far worse than 33% and 36%.
Yes, I was thinking of Goldthorpe and the like. It’s been a while. But I think you’re right to distinguish between class mobility and income volatility - especially given the extent of casualisation in some sectors.
Its also worth looking at the graph which takes into account housing costs where the distribution looks even flatter - very flat except for the top two deciles. As the report mentions the areas with the lowest income had the highest proportion of home ownership due to retirees.
Have to be a bit careful about equating income with wealth or even quality of life.
Our libertarian friends keep telling us the lefty obsession with fairness and social justice is nothing more than envy.
Have a think about the wage and salary expectations of those in the higher income deciles, as well. Successful professionals, managers and business owners tend to expect that their remuneration will follow a significantly upward trend from year to year to reward their performance and ambition, much more than the yearly increment or CPI adjustment for someone doing night stacking in a supermarket. Anyway, given the inflation figures for those years, I think the graph markedly does show a rising tide lifting all boats.
There’s been a lot done internationally on intergenerational income mobility. Low cross-sectional inequality correlates with high mobility, and vice versa. Thus the US - the “land of opportunity” - actually has about the lowest such mobility in the OECD. However we are a bit of an exception - despite the fact that our cross-sectional inequality is higher than the OECD average (though there’s a good case that part of that is due to measurement differences) we have above average intergenerational mobility.
Oh, and Anthony is dead right - this chart is not really showing what the rest of you think it does. This link explains why.
“I wonder how much of the extra growth for the top 10% can be attributed to tax policy, particularly the cuts in the top marginal rate of income tax?”
Not much, since this didn’t happen until 2006, which is when the data stopped.
“I think the graph markedly does show a rising tide lifting all boats.”
Not really on topic, but that analogy shits me to tears. Tides lift all boats because they’re all at the same level in the first place!!!
It does a shit job on its own terms, and invites the rejoinder “a storm at sea will mostly sink the poorest boats”.
Shorter NATSEM:
Most neighbourhoods did pretty bl**dy well, with the really rich ones doing slightly better than the rest.
My commentary: If every period of 5 years was like that none of us would have cause for complaint.
“Most neighbourhoods did pretty bl**dy well”
Andrew, you can say bloody.
And these neighbourhoods who did so well out of John Howard voted him out.
Ungrateful c-words
Same water level, but some boats are taller than others
Was waiting for someone to play ecological fallacy bingo, DD (although it’s a while since I’ve heard it called that).
I thought it was clear from the post that the data was on spatial inequality. The reason why it’s worth focusing on spatial inequality is that it’s much more pronounced in Australia now than it was, and there are strong correlations between all sorts of outcomes that might contribute to social mobility and indeed current social and health outcomes and spatial advantage or disadvantage. Bob Gregory, I think, was the first to point out how much more important this had become in Australia as compared to the 70s.
Nor am I sure why the ecological fallacy comes into play here when the graphs are read in the context of the report’s discussion.
Kim - I think the graphs are misleading if they are to be interpreted as the report and you conclude that they are indicative of changes in living standards. Net income which was not considered is much more relevant. As I mentioned in an earlier comment, just because the higher deciles may have had higher percentage increase gross income, doesn’t mean they ended up with higher percentage increase in net income.
I wonder why NATSEM chose gross income, Chris? However, I suspect given the reach of income transfers into the higher deciles in this period, there would have been a relationship.
Kim, Bob Gregory and Boyd Hunter’s work on spatial disadvantage which you avert to - and generally I have the greatest respect for a lot of Bob and Boyd’s work - was, I suspect, even more fraught than NATSEM’s more recent work on spatial inequality. And they used more loaded language about ‘ghettoes’ which (wittingly or unwittingly) just reiterated underclass or ‘culture of poverty’ theories about, you know, too many poor people all living together reinforcing each others’ bad habits. Peter Whiteford did a good critique of their analysis, whereby he mainly focussed on the problems of crude measurements of spatial inequality which I have tried to raise here. I’m not saying there is no story to tell about growing inequality in Australia, just that the graph produced about spatial deciles might be a particularly opaque and problematic way of doing so, given the issues of spatial mobility etc etc I referred to earlier.
Point taken, Anthony, but the methodology and caveats are noted in this NATSEM paper. I think there is value in looking at spatial inequality for at least two reasons:
(1) Many services are delivered geographically (ie health, education) and access to them may significantly change the life chances and social amenity of those on low incomes - that is, those who are favoured in terms of geographical access to services effectively get something like a social wage compared to those on the same incomes who are not so favoured.
(2) This of course is a value judgement, but it seems to me that Gregory’s point about social cohesion and the spatial distribution of inequality has some force.
Klaus, as I understand it in the historical view, from the 1850s through to the First World War Australia had enormous social and income mobility, greater by far than the mobility due to mining prosperity than we have now, at least within a relatively small ethnic group (native-born or UK-born Anglicans, Protestants and Anglo-Catholics). The great gain of Australia’s last thirty years in my view has been the extension of social and income mobility to immigrants from outside the UK and Indigenous Australians, and that, alas, is a quality left unmeasured in measurements of average income growth.
There is certainly more income mobility now than there was thirty years ago, but then thirty years ago the “Australian settlement” was still in force, settling a staid fairness at the expense of income mobility. Those things were then, of course, seen as irreconcilable aims.
Sorry, Liam, I’m not clear what sort of concept of “income mobility” you’re working with there - do you mean increasing income over the lifecourse?
No, I suppose not. I really mean class mobility, and should get my terms straight.
How would you define that? Marxist or Weberian? Subjective or objective?
Serious question though. My recollection is that the last big project seeking to analyse inequality by class - the Western et al thing from UQ - used an “objective” measure of class - basically according to occupational stratification - ie “managers”, “professionals”, etc. - using adapted ABS categories and found there wasn’t much mobility between occupational grade.
Why, objective, of course, Kim. I’m not that far gone down the dark path of Luvvieconomics.
I’m not a social scientist. At all, but that never stopped nobody from blog commentin’:
If the study by Western you’re referring to is from 1991, and I’m reading the same one you are, it tends to bear out my huge unsubstantiated generalisation: that the Australian settlement of white trade union bargaining tended to settle Australians’ mobility up-and-down classes within their society, in favour of a sense of fairness, at least until its dark bloody end at the hands of Paul Keating as Treasurer and Prime Minister.
Since the mid-80s we’ve been seeing a return to certain aspects of the late Australian nineteenth century: a general growth in income across the board thanks to resource wealth, an easing of any restrictions on absolute smacking fortunes being made by clever well-connected people, and (at least while immigration to Australia remains restricted) a shortage of skilled labour, at the expense of income equality.
Let’s see if it results in an outpouring of art equal to that of the 1890s through 1910s: I bags the role of that militarist ratbag CJ Dennis.
I’m not meaning to be pedantic, Liam, but my recollection from the Western study was that it was very rare for individuals to move between less skilled and more skilled occupations during their adult lives, and the blue-collar white collar divide was very rarely crossed. Are you talking about the relative distribution of incomes maybe?
Yes, your first statement is exactly what I mean, Kim, at least in the twentieth century.
In the twentieth century, Australians’ income and professional mobility was rare and difficult. In the nineteenth century, in the odd colonial space Australia was, there was great opportunity for a small number of people—immigrants from the right kind of place—to “improve” themselves, in terms of skill and income. The whole story of the Labour movement in this era is that of skilled but uneducated men (and almost only men) entering the professional/bureaucratic class.
I meant to take up Klaus K on this, about the nineteenth century.
Integenerationally, you mean, Liam?
I hadn’t though of that but yes—intergenerationally, as the second and third generations of the Australian immigrants have tended to do better and better for themselves in the very late twentieth and early twenty-first century.
Though I hasten to add that my claims here are based on nothing but the basest of ludicrous generalisation and a-while-ago reading of admittedly dated Australian colonial history. How do economic historians admit ignorance again?
[flick, flick]
Of course as yet I’m not willing to state this conclusively, without a more comprehensive survey following a literature study.
That’s cool, Liam. I just thought there was a pretty manifest contradiction between the two poles of your argument - viz. that there wasn’t much social mobility under the “Australian settlement” but that was compensated for in various ways, and that there was a move from the Anglo-Celtic (forgiveness begged) male working class up the ladder over time. It starts to make sense if you distinguish between social mobility over the lifecourse of one individual and intergenerational mobility.
Yes, good points about C19th Liam, and you are right about the importance of the Australian settlement in C20th. Of course, part of what underwrites that colonial-era mobility of the C19th is dispossession, but the parallels in terms of a resource boom and the other things you mentioned make for an interesting comparison with the present.
Is it just me or is the second example in that wikipedia description of “ecological fallacy” actually a description of “confounder”, with the unobserved variable being proximity to the factory…? Obviously the first example is an ecological fallacy (people aren’t the same as their neighbourhood). But the second one seems to confound the discussion of fallacy with the discussion of confounding - a study which correctly adjusted for the position of the factories would still be subject to the ecological fallacy.