If only we could persuade poor people to adopt the values and behaviours of their rich neighbours we could end poverty in a generation. Or at least that’s the impression you’d get from reading the never ending stream of books and articles about the culture of poverty, the underclass and the socially excluded. In an essay for the Boston Review Stephen Steinberg challenges the idea that poor Americans are trapped in a cycle of disadvantage that only they can end.
In 1965, Daniel Patrick Moynihan argued that black families and communities had become dysfunctional. Conservatives seized on Moynihan’s work to argue that American society offered all Americans ample opportunity to get ahead and that the only thing holding poor Afro-Americans back was their own dysfunctional way of life.
According to Steinberg, Moynihan’s fatal error was to invert cause and effect. Steinberg writes:
… even Moynihan’s harshest critics did not deny the manifest troubles in black families. Nor did they deny that the culture of poor people is often markedly at variance with the cultural norms and practices in more privileged sectors of society. How could it be otherwise? The key point of contention was whether, under conditions of prolonged poverty, those cultural adaptations “assume a life of their own” and are passed down from parents to children through normal processes of cultural transmission. In other words, the imbroglio over the Moynihan report was never about whether culture matters, but about whether culture is or ever could be an independent and self-sustaining factor in the production and reproduction of poverty.
The argument about cause and effect has a long history. For example, in 1970 Lee Rainwater argued:
The special ways of adapting by the poor suggest only that effective poverty strategies have to change their income situation before requiring changes in their behavior and attitudes. The major reason for the failure of most anti-poverty programs so far is that they require the poor the change their behavior before they have gained the resources that would change their situation (quoted in The Social Inclusion Agenda).
Over time, poverty and programs designed to alleviate it became increasingly identified with black Americans. The problems of a small subset of the most disadvantaged welfare recipients generated stereotypes which then applied to all welfare recipients. Those who were non-poor and non-black came to see poor Americans — particularly those who relied on welfare — as a distinct and alien population that had somehow developed an immunity to opportunity.
Recently the culture of poverty thesis has made it into the New York Times where Patricia Cohen argues that it’s back on the research agenda. According to Steinberg: "the ballyhooed ‘restoration’ of culture to poverty discourse can only be one thing: an evasion of the persistent racial and economic inequalities that are a blot on American democracy.”
“The problems of a small subset of the most disadvantaged welfare recipients generated stereotypes which then applied to all welfare recipients”
Not in Australia — I thought most welfare recipients were middle class, but most people don’t even think of the type of subsidies they get as welfare.
“… most people don’t even think of the type of subsidies they get as welfare.”
I agree. And it’s true in the US as well. David Super of the University of Maryland writes:
Conrad – Indeed, middle class welfare is called ‘family tax benefit’ to try to distinguish it from what poor people receive.
But critical as I am of FTB, I don’t think it is going to seriously damage middle-class norms. These are pretty resilient (in the 1970s social critics feared the demise of the work ethic; within a generation social critics were complaining that people were working too hard). For most families, FTB supplements earned income.
Rubbish, Conrad and Andrew – Australia actually has less middle-class welfare than any other developed country, and by a considerable margin. Read Peter Whiteford’s stuff. “Most payments go to the middle class” is actually a very harmful myth that has distorted much Australian policy debate – I actually blame Peter Saunders of the CIS for promulgating it, which is infuriating because he knew better but had a hidden agenda.
It’s because of our mania for means testing. Means testing FTB, for example, immediately reclassified it from a “horizontal equity” instrument (ie compensating for the fact that if you’re supporting kids then you have less capacity to pay net tax than someone on the same income without kids) into a “welfare” measure. There’s actually a good argument that we’d be much less prone to creating a “small subset of the most disadvantaged” if we had a more universal welfare state that aims more at buffering life risks than directly addressing poverty. If you pay only the poor, it only pays to be poor.
The bit about a “poverty culture” being endogenous is obvious, as critics pointed out in Moynihan’s time. The perpetuation of it is also easily explained by Schelling models of discrimination – basically if you’re in a group (not necessarily a racial group) that is discriminated against, then “poverty perpetuating” behaviours become perfectly rational (there’s no financial payoff to education, for example, if no-one gives you a well-paid job. And life is so risky that there’s no point in suffering current hardhip for future possible benefits by saving). Of course, the more of those poverty-perpetuating behaviours in the group, the more the group will be (rationally) discriminated against.
DD – Yes, I know that Australia is the means-testing capital of the world. But that doesn’t stop FTB being paid to the vast majority of families with kids, most of whom are not poor by any sensible definition of the term.
I don’t think Saunders ever claimed that most payments go to the middle class – only that far more went to them than was desirable.
Andrew, DD’s point is that FTB was not designed as a welfare policy and so it shouldn’t be evaluated in those terms. Indeed, as far as welfare goes, a middle class person that experiences a bout of unemployment (or some other event that affects their earnings potential) gets screwed compared to most other OECD countries because the replacement rate for most benefits is so low (and not dependent on previous earnings). In doing so, Australia’s system functions less as a social insurance system that buffers everyone against unforeseeable shocks and more as a system of extreme poverty avoidance. Of course, the tradeoff is that the tax burden on the middle class in Australia is low compared to many other countries (thus, one could argue that they are being asked to self-insure out of their post-tax income). The irony in all of this is that the anti-welfare rhetoric in Australia is STRONGER than most other countries, despite the system being so stingy. Concomitantly, we also perpetually battle the myth of the overtaxed middle class.
LO – I realise the language was supposed to steer us away from the idea of welfare.
FTB (or at least its scale and reach) have really been quite a departure from earlier Australian policy trends, which as you note has otherwise been characterised by high levels of middle class self-reliance. Australia also has unusually high levels of private health and education, though it is partially subsidised in most cases.
I don’t have any comparative internationaal measures of tax and welfare rhetoric, but intuitively it seems plausible that a middle class that gets little for its taxes would complain about it more than a middle class which gets a lot.
“Andrew, DD’s point is that FTB was not designed as a welfare policy and so it shouldn’t be evaluated in those terms”
I’m not really sure what the difference is between something designed as welfare and something not designed as welfare, since the end effect seems the same. Getting away from the FTBs, maybe the medicare rebate wasn’t designed as welfare, and, more controversially, maybe negative gearing wasn’t designed as welfare for the rich, but basically they enrich one group at the expense of another, neither of whom generally deserve it based on their income. I’d be happy to see those two (and others) entirely chopped and the proceeds spent on increasing the tax-free threshold and unemplyoment benefits for people that really do have no money.
“Australia actually has less middle-class welfare than any other developed country”
Whilst I’m happy to admit that Australia is a fairly low tax country (as LO points out), surely that can’t be true unless Singapore, Taiwan and Hong Kong arn’t considered developed countries.
“The irony in all of this is that the anti-welfare rhetoric in Australia is STRONGER than most other countries, despite the system being so stingy.”
Perhaps Australians just like victimizing people more than other places.
oops, that should be the private health insurance subsidy, not medicare rebate.
I don’t think its victimising, conrad, that ‘irony’ seems to be the same as our debt: we have the least of almost anywhere and yet we have massive anti-debt public sentiment.
Of course, in both the case of middle-class welfare and debt, the ‘and yet’ should perhaps be ‘because’.
Conrad – I am not sure that you need your correction. For a middle class person the Medicare rebate and private health insurance rebate are conceptually the same – both pay part of the cost of private health care.
Andrew, when you say that the middle class doesn’t get a lot for its taxes, what exactly do you mean? I’d interpret it a little differently – the middle class gets exactly what it pays for – it isn’t particularly highly taxed and that is matched by a relatively limited (or perhaps better put – a highly targeted) state. Just take a look at the revenue and spending shares of GDP (at all levels of government) in Australia – we are well below the OECD average. Conrad is of course correct that there are some Asian countries that also have fairly limited states, but that doesn’t change the general point that Australia is in the bottom quartile of the distribution for developed countries.
Conrad, I don’t think that anybody claimed that all government programs are well targeted at the poor. But if you look at it comparatively, I think you’ll find that the largesse directed at Australia’s middle to upper classes is small compared with most other countries. You mention negative gearing for example. But in the US, all mortgage payments are tax deductible – a policy that affects a much larger share of the population than Australia’s negative gearing.
I reckon you’re wrong about Singapore, Conrad – look at their massive Provident Fund FFS. Like compulsory super in Australia, just because the money doesn’t get passed through Consolidated Revenue doesn’t mean it’s not taxing and welfare – failure to pay your Fund contributions gets the same coercion as failure to pay your tax.
Of course we have strong anti-welfare rhetoric in Australia – that’s precisely because we tightly target our payments to groups that the average punter assumes he or she will never be part of. It’s simple political economy really.
More broadly, “middle class welfare” may be an issue if your sole concern is with income inequality or income poverty (though even then then there’s difficult tradeoffs with work and saving incentives). But that is only one function of real life welfare states – the other, critical, one is buffering against uninsurable risks. Large welfare states are a substitute for missing insurance markets – missing because adverse selection problems make them unviable. Try reading “The Welfare State As a Piggy Bank” by Nick Barr.
In some areas – notably health – you can also create large welfare (in the economic sense) gains by creating a single provider to offset large private rents due to severe information asymmetries – that’s why semi-socialised European health systems get far better “bang for buck” than the privatised US one.
LO — I don’t disagree with you that Aus is a low tax country and may be directing money better than others (I have lived in France!), but that still doesn’t mean there are many programs that give significant amounts of money to groups that don’t need it that attract very little attention compared to programs that give money to those that do. Now, part of this might be because it’s very hard to target just groups that need it (I wouldn’t dispute that for things like schools, for example), but if you can think of things that could be chopped tomorrow, then part of it clearly isn’t — it’s a political problem — Perhaps what it really shows is that it is very easy for society’s dominant groups to engineer things in their favour.
Patrick: “I don’t think its victimising, conrad”
I do — and I don’t think that the debt analogy is great because debt applies to a single government entity and hence the belief (even if irrational) is that it affects basically everyone’s future, but denigrating the poor breaks individuals into groups.
For example, let’s say you’re a member of everyone’s favourite old group to put the boot into, an unemployed young single mother (you can be black too if you want, just to make it worse). My bet is that you’d love to have a job playing $100K, but due to circumstances, you don’t. You are going to get sneered at by a lot of people. Now let’s say you’re earning 100K and saving an amount of money via a negatively geared share portfolio that the unemployed single mother gets in benefits. Now people will tell you what a good idea that is. So we have two groups, both which receive an identical amount of government freebies, yet they are treated entirely differently. That surely qualifies as victimization in my books.
LO – A 40% marginal tax rate feels like a lot, even if someone in Scandanavia pays far more.
It’s terribly out of date, but the most recent (2003-04) ABS estimates had the top 20% of households paying about $650 a week more in tax than they received in benefits, and the fourth quartile paying about $220 more in tax than they received in benefits.
The third quartile was about even, and the bottom three quintiles significant net beneficiaries.
Andrew, the ABS FIS is right as far as it goes, but it includes cash benefits and personal income tax only. So it’s actually leaving out most tax and spending. What, for example, is the distributional impact of the GST? Of company tax? Of tobacco taxes? Of Medicare? Of lauranorder? And so on.
Which illustrates an important point – focusing too much on whether individual polices are regressive or progressive will mislead you about the overall effect of the system.
“focusing too much on whether individual polices are regressive or progressive will mislead you about the overall effect of the system.”
Alternatively, there may be particularly expensive individual policies that should be chopped by themselves. For example, if I’m reading chart 12 correctly, then unemployment benefits cost the government about 5 billion in 2007. Alternatively, digging around on the web, negative gearing on housing cost the Australian government about $3 billion, and I assume that 99.99% of those getting the benefit didn’t need it. These are both dwarfed by payments to families and children, and so surely some of those payments could at least be better targeted.
DD – They include indirect taxes including GST, and sin taxes, and on the spending side education, health and housing benefits.
There are of course interesting questions about tax incidence, but for company tax at least the ABS figures are likely to understate taxes incurred by high income earners relative to low income earners.
Conrad – I think it is misleading to construe negative gearing as simply a benefit. The principle of taxing net rather than gross income is a sound one. The issue is more that we have an industry built on the assumption of capital gains rather than getting a return on investment from rental.
Are you seriously pretending that we don’t transfer wealth away from the person earning 100k and toward the welfare recipients? How then do you figure that the person who is losing wealth is actually getting “government freebies”, just on the basis that they could potentially have been losing even more?
In the particular case of negatively geared share investment, if you take take out a large loan to buy shares you are also taking a large risk… if you happen to lose on this, no way will government be offering freebies to pick up the pieces, you get to keep your loss. If you make a profit they will take CGT out of that profit. Besides, you are paying interest on that loan which becomes taxable income for the next guy, so government doesn’t even lose any tax base (it merely shifts sideways). Finally, if the share investment wasn’t at least a little bit attractive, then the money would not get borrowed, the shares would not get bought, and the person earning 100k would just shrug and take 6 months off work to relax because trying to save the money only vanishes in tax anyhow, why bother?
I can imagine someone standing in court after cutting off another man’s leg:
“Oh no your honour, I actually gave him a leg!”
“But the evidence shows you chopped his leg right off.”
“Ahh, but I might have chopped BOTH of them off.”
“Are you seriously pretending that we don’t transfer wealth away from the person earning 100k and toward the welfare recipients?”
Tel, I’m not sure that I’d really want to argue for a system where people get more money back from the government if they have put in more. This is of course what happens in some places of Europe, and all it does is increase inequality (since it’s a rich get richer scheme), and so to try and fix this, you get ever higher rates of tax and redistribution into more and more things. I think a better way of looking at it is that the government shouldn’t be giving rich people money back unless it has some greater benefit. Obviously greater benefit is arguable, but depending on your political bent, that could include things like education and socialized health-care. Alternatively, I don’t see it including subsidizing investments schemes, giving out free money to people wanting to buy houses that don’t need it etc. .
This is an interesting discussion (although perhaps not entirely relevant to Don’s original post).
On the middle class welfare issue, I see the basic cause of the disagreement as being different definitions of what constitutes “welfare”. (It’s also worth noting that in the USA “welfare” is used virtually universally to refer to programmes for the extremely poor, so it’s a much narrower definition of welfare than we commonly use in Australia.)
Derrida Derider quotes my work that when you look at cash transfers only – age pensions, unemployment benefits, disability pensions etc plus family benefits – that Australia targets more to the poor by an extremely wide margin than any other OECD country and that we have less middle class welfare than any other country. More than 40% of what we spend on cash benefits goes to the poorest 20% of households, compared to about 25% in the USA for example, and 16% in France to take Conrad’s example. (It is also worth noting that even though the poorest 20% of the population in France get “only” 16% of the cash benefit spending, this is still redistributive, since they pay a much lower share of taxes and have a much lower share of private income, so 16% of benefits increases their share of disposable income.)
Now it is perfectly possible to target more than we do and to have even less middle class welfare than we currently have, but basically this will require higher effective marginal tax rates on lower income groups since we would have to withdraw benefits at a faster rate than we do currently.
However does welfare include public education or public health and does it include tax expenditures? Now some people here and elsewhere might think that it should include all social spending, and it is certainly true that if you include health and education that the distributional profile of spending becomes less pro-poor and more directed to the middle class. But then why not include roads and public transport and telecommunications and all government spending and not just social spending? I think that there are strong arguments for assessing the overall impact of spending and taxing, but whether it should be described as “welfare” is a moot point.
In case of tax expenditures I take Tel’s point. From some perspectives, it may seem strange that taking less tax from taxpayers should be described as “welfare”. But tax expenditures refer to situations where the tax concessions depend on people making specific expenditures, so governments are rewarding specific behaviours, not providing overall lower taxes for everyone. And in rewarding specific behaviours they are in some cases doing this in ways that benefit higher income groups (negative gearing) and in other ways that benefit lower income groups (non-taxation of imputed income from owner-occupied housing – at least at a point in time).
The case for treating tax expenditures as equivalent to cash spending is fairly clear in the case of family payments. In 1975 tax deductions/ rebates for children were fully replaced by family allowances. Did the shift from tax rebates to cash transfers somehow amount to a large increase in middle class welfare? If you say that spending is welfare but tax reductions are not, then you would say that middle class welfare increased. But in fact the effect of the change was to increase support for low income families and over the next few years to reduce the support for higher income families, since Family Allowances were not indexed and tax rebates were supposed to be indexed. And as DD points out a few years down the track we income-tested family allowances anyway.
Overall, I think that the problem is that we are talking at cross-purposes. But I’m not sure that there is one indisputable way of defining middle class welfare. As DD points out there are trade-offs with work and savings incentives. My view is that we should look at alternative definitions of middle class welfare and see what differences they make; that it is valuable to look at the overall profile of public spending and taxing; and that rather than getting hung up on processes we should try to judge how effective different approaches are in achieving desired outcomes.
Conrad,
I’m not arguing for “fixing” anything, merely trying to call a spade a spade. Someone who contributes in the nett balance, is a taxpayer (i.e. some of their productive output goes into government coffers) someone who consumes in the nett balance is a welfare recipient (i.e. they are supported by someone else’s productive output). All this business about people getting some small percentage of their money back is smoke and mirrors — just look at the overall nett amount that flows in or out. On that basis please show me some of this “middle class welfare”.
The middle classes have always been the prime targets of taxation, and always will be for the simple reason that they are wealthy enough to be worth taking from, but not powerful enough to defend themselves.
http://www.budget.gov.au/2010-11/content/overview/html/overview_33.htm
There’s a convenient overview calculated to show the threshold for each category of person, where you would switch from being a nett welfare recipient to a nett productive contributer. The biggest behavioral factor that influences this is getting married and having children, for the obvious reason that government is interested in producing more population. You will note that all of the thresholds are well below 100k.
By the way, there’s also a simple breakdown of where the money comes from and goes to:
http://www.budget.gov.au/2010-11/content/overview/html/overview_37.htm
If you add together welfare, health and social work (the things that most directly benefit the poor), that’s half the federal spending right there. Add together income tax, fringe benefits tax and superannuation tax (all of which are essentially income tax by any other name), and you get approx half of the federal government revenue. Thus, (speaking of nett real figures only) we can say that ALL income tax goes toward helping the poor.
I fully agree it would be informative to put a cash equivalent value on services payed out of the public purse, where they directly benefit the individual (subtracting away the user-pays components now that train tickets generally cost more than the equivalent fuel if you happen to have both a car and available parking). It is however, quite difficult to do this because not all people actually use those services, and areas such as transport, communication, education, etc are so heavily regulated that it has become essentially impossible to measure a private equivalent. We could make some sort of statistical guess, but things like education deliver diverse benefits, the government itself benefits from a better educated population, so does business, as well as those individuals who improve their position because of their education.
It’s quite reasonable to use the argument of “for the general public good” when arguing for government spending on roads, education, defense and similar things. However, it’s a bit contradictory to then start counting up individual benefits from these things as cash-equivalent.