Protecting our social security system from churn and burn

Today’s Crikey has a brief piece by me in reaction to a piece by Christian Kerr on Wednesday. I’ve written about this a couple of times on Troppo before.

Here’s my reprise for Crikey.
What makes Australia’s social security system great

The memes are out in force again, I see. Given the breadth of issues journalists cover and the speed at which they must work, one can’t blame them for swallowing the odd line set for the unwary.

On Wednesday, Christian Kerr fell for what he himself called an “old chestnut” — “why the hell are we paying public servants to take money away from people in tax and hand it back to them in welfare?”

I’ll tell you why because, like all other countries our tax system raises revenue according to individual earnings, but family payments are designed to assist according to household rather than individual needs.

That difference underscores an under-appreciated gem of Australian public policy.

As in other countries, our social security system is messy. But when it comes to what matters getting the best social results for the least cost well, the clich© “world’s best practice” doesn’t begin to describe it.

Our system provides European-style protection for the least well off and for low and middle income children at near American-style revenue cost. It’s nearly three times as targeted as the next best system New Zealand’s. Even poverty traps have been tackled here better than most other countries.
progressivity of the tax system.jpg

* Ratio of dollars to poor households per dollar to rich households. Source (ppt): Whiteford, OECD, 2005.

Of course some “churn” reflects increasing politically targeted middle class welfare and to that extent Kerr’s comments hit home.

But Australia gets the stellar results it does precisely because of the way the tax system raises money and the way the payments system then recirculates it according to household need.

The real question is the administrative efficiency of both systems acting together not the magnitude of “churn”. Indeed, if we called family payments “tax credits” as some other countries with less efficient systems do measured tax churn would plummet.

Of course nothing real would have changed. But the “tax churning” meme could be laid to rest.

Then we could move on and get churned up about something more worthwhile.

This entry was posted in Economics and public policy, Politics - national. Bookmark the permalink.

36 Responses to Protecting our social security system from churn and burn

  1. derrida derider says:

    The real question is the administrative efficiency of both systems acting together not the magnitude of “churn”

  2. Peter Saunders says:

    Nick keeps using Peter Whiteford’s table to support his argument that we do not have a churning problem in Australia. But as I have pointed out several times before, the Table is comparing tax-welfare flows at extreme points in the income distribution. Look at the pattern across the distribution overall and you get a very different story.
    It is true that Australia targets cash on poorer households more successfully than other countries do. This is because of our tight means testing and the fact that all benefits are paid out of general taxation rather than from a national insurance fund. Whether or not this is a good thing is of course debatable, for we certainly pay a high price in the high EMTRs which are unavoidable when you withdraw means tested payments from people as soon as they start to earn a bit more of their own money. This creates huge work disincentives from which our system inherently suffers. Nevertheless, it is true that we devote a higher proportion of our total welfare spending to the lowest 30% of earners than virtually any other OECD country (something the welfare lobby persistently fails to recognise, incidentally).
    It is also true that our system churns back fewer tax dollars to the richest 10-20% of the population than other countries do, but this is almost entirely due to our different age pension arrangements. Because we do not have a European-style insurance system, the highest earners in Australia do not get a state age pension when they retire, but in most other countries the pension is paid on the basis of eligibility through past contributions rather than on the basis of ‘need’. Because the richest Aussies do not get a government age pension, their share of total welfare spending is lower than in other countries.
    The key point, however, is that other welfare payments are spread thickly across the middle classes, and this has got worse as family payments have been made more generous. Nick fails to mention that, below the top 20% of income earners, churning is every bit as great in our system as elsewhere. As Peter Whiteford states in the paper which Nick keeps referring us to: “Australia has roughly the same share of middle class welfare as most other OECD countries” (p.11). He estimates that the middle 60% of Australian earners get 56% of all welfare transfers. This means there is a huge problem of churning in our tax-welfare system, and Ann Harding’s modelling at NATSEM bears this out.
    This is not just a problem due to the inefficiency and expense. Much more important is the erosion of independence. Millions of people who could otherwise look after themselves are having money taken from them in tax so that the government can give it back (and demand their gratitude) in payments and services they would not have needed if they had not been taxed so heavily in the first place. This is why the churning debate will not go away, no matter how much Nick (or Peter Costello for that matter) wish it would. “Stellar results”? Hardly.

  3. James Farrell says:

    What’s with this ‘targeting’ business? Are you some kind of patriarchal paternalist? Get with the program.

  4. vee says:

    I read this in Crikey before here and I don’t have the brain capacity of those that posted before me.

    Doesn’t this sort of churn suggest that those on middle income are not as good as money managers as those on low incomes and high incomes?

    Doesn’t that in turn suggest that those on high incomes care about money and want to for arguments sake re-invest their money in X, and those on low incomes need all the money they can get just to meet an adequate standard of living for say their family, whereas those on middle incomes are a bit blaise?

  5. Peter Whiteford says:

    I feel that I should point out that while my SPRC paper – in an attempt to be “balanced” – does note that “in this specific way Australia has roughly the same amount of middle class welfare as other OECD countries”, this specific way means as a share of total spending. However, because of targeting Australia spends less than the average, so the paper also shows that when you calculate churning across the whole income distribution and – not just the extremes – Australia still has the lowest level of churning of any OECD country, when you express churning as a % of household disposable income.

    In any case, in the paper I also argue that this measure of churning is fairly primitive. It is an overestimate because it is calculated from grouped data rather than from unit record data. More importantly, silly policies can reduce measured churning. For example, if all disabled or old people living with relatives set up separate households then churning would go down because it would reduce the overlap between a household receiving benefits and paying taxes. Would this be a good policy to support? Unfortunately poverty would go up unless the taxpayers voluntarily transferred more money to their relatives (and in case they would lose the economies of scale arising from sharing households).

    As Nick and Derrida Derider point out – The real question is the administrative efficiency of both systems acting together not the magnitude of “churn”

  6. James Farrell says:

    In case anyone wonders or cares, my comment was directed at Nicholas, not Peter (whose comment I hadn’t seen).

  7. Christine says:

    Interesting stuff.

    Peter S: Are there many retirees in the top 10-20% of the income distribution? If not, are these figures calculated on a lifetime basis? I would have thought the answers to both these questions would be ‘no’, in which case I don’t quite see how lack of government old age pensions in Australia relative to other countries could be responsible for a relatively low rate of spending on the top 10-20%.

    Much as I dislike baby bonuses, they would also contribute to what seems a lot like churning, but is actually pulling money away from singles to families, but it might look a lot like churning if the group is defined on the basis of income. (Which is just another way of making Peter W’s point that how the group is identified will matter.)

  8. Thanks for your comments Peter (S). Thanks for your comments too Peter W, but these comments are addressed to Peter S :)

    You commented on this when I posted on it a while ago and I was hoping you’d expand. It seems to me that you’re using churning to suggest that there’s massive inefficiency in the system and that we can end it all by just giving the money back. Well I broadly endorse the way the money is being given back through the family payments system now. Of course we could quibble, and I guess we could both think of some middle class welfare we’d like to do away with.

    So I’ll agree with you that it could well be better ‘optics’ to call all of the cash circulated as family benefits ‘tax credits’. Wherever you do it that gets rid of the churning. One is left with the real question which is ‘where are your priorities in terms of targeting tax relief?’ Right now the relief is targeted to middle and lower income families. How would the relief you want to give differ?

    On the last thread I made the following comment but didn’t get a response. I hope you might be able to expand a little more here.

    Thanks for your response. I’ve read your papers and my own bias is the same as yours – to encourage self-provision where possible. But . . . Whiteford’s comments on churning runs for seven pages including over a page of specific responses to your suggestions. He’s fairly sceptical of your proposals without rejecting them out of hand.

    (For instance just trying to think it through intuitively, it seems to me that using the tax free threshold is a very blunt instrument because it is so indiscriminate. This means that it gives away large amounts of revenue and so undermines our capacity to cross subsidse those who are in poverty more than temporarily and families with children. (I’ve take it from comments you’ve made in various publications that you’re broadly in support of these goals, even if you want to scale them back to the extent you can bring about self-provision.)

    Can you tell me if there’s a detailed response to Whiteford’s comments.

  9. Cam Riley asked me to post the original piece by Christian Kerr. Here it is.

    “More than 900,000 working age Australians will keep less than half of their next dollar of private income, a report shows, and more sole parents and women face higher effective marginal tax rates than they did a decade ago,”

  10. Peter Whiteford says:

    I have redone the calculations of churning in OECD countries and excluded the top 20% of the income distribution in each country to deal with Peter S’s point above. When you do this, Australia moves from having the lowest level of churning in the OECD to the third lowest – after the US and France (but France pays for most of its welfare system with employer social security contributions, which are not included in household income surveys, so this is misleading). It also means that the US spends a lot more of its lower level of social spending on the rich than does Australia!

    These calculations are static and do not measure lifetime churning, and they also do not include indirect taxes and government non-cash benefits. If one is concerned about this broader measure of churning then it would be much higher than if you simply look at direct taxes and cash benefits. However, Australia would still have a relatively very low level of chuning as our government non-cash benefits are around the average, but our level of indirect taxes is lower than most other countries.

  11. Peter Saunders says:

    Thanks to Nick and Peter W for thoughtful contributions. But I fear there is a basic philosophical gulf between us.
    Nick talks of “giving the money back” to taxpayers, but that is not what I have been arguing for. I don’t want it taken from them in the first place. Similarly, he thinks raising the tax free threshold means “giving away revenue”, but I think it means not collecting revenue.
    Nick thinks the difference between giving people their money back and never taking it away from them in the first place is just a semantic quibble (which is why he thinks relabelling FTB payments as tax credits will answer my concerns) but it is not. There is a huge difference between depending on the government to make payments to you, and being independent of government by retaining your own income earned through your own efforts. And this, for me, is what the churning debate is all about. Nick keeps insisting the concern is about efficiency, but in a paper I published in July last year ( I identified efficiency as only one of six reasons for trying to reduce churning (the others being strengthening work incentives, improving sustainability over time, personal empowerment, enhancing social solidarity and depoliticising civil society by stopping politicians vote-buying with voters’ own money). I am currently trying to write a short monograph on all this which will hopefully make these points in more detail.
    Turning to Peter W, it seems to me the appropriate basis for measuring the extent of tax-welfare churning is as a proportion of total government spending, for the debate is precisely about how much tax governments should take and how much benefits they should hand out. It therefore remains true to say that Australian government churning generates as much ‘middle class welfare’ (relative to total government budgets) as other western governments do.
    More importantly, Peter accepts that his analysis excludes lifetime churning (tax taken today but handed back later rather than immediately), it excludes the value of services in kind (notably education and health, the first of which is known to favour the middle classes more than the poor) and it excludes indirect taxes (thereby massively underestimating the amount of tax paid by those on lower incomes who receive most of the targeted spending). These are three huge omissions. Analysis has been done (by Ann Harding and others) that includes all these things, and it reveals extensive tax-welfare churning in the Australian system (I reviewed some of this literature in my first churning paper published in April last year –
    Peter rightly says that including all these things would still show Australian churning as lower than in most of Europe, but so what? The task as I see it is to refashion the twentieth century welfare state to take account of the growth of affluence over the last 50 years. The European welfare states are the last place to look for ideas on how to do this.
    In his SPRC paper, Peter rightly says that there are major practical difficulties in significantly reducing churning. I agree with this – but it doesn’t mean we shouldn’t try. I believe, for example, that there is a lot of scope in developing personal accounts to replace mass dependency on government provision, but the models here are found in places like Chile and Singapore, not in Europe.
    My argument has always been that large numbers of Australians could afford to insure themselves, or save, to cover many of the risks that 2 or 3 generations earlier they could not have afforded to cover, but the high taxes they have to pay to fund the state system of provision precludes many of them from doing this. Superannuation is the obvious example of the viability of enhanced self-reliance – most of us can now save and invest for our retirement in a way that would not have been possible 50 or 100 years ago. What is true of retirement is true too of other key areas like health and income insurance.
    A move in this direction may or may not prove more efficient (I think it would but others may disagree). But it would certainly be more liberating.

  12. Peter,

    I look forward to your responding to my points at some stage. You paint my views as in some way fundamentally different to yours. Well I tried to engage you by conceding your point. I could concede it ‘for arguments’ sake, but I also concede it more genuinely. I agree with you that if possible it’s better if money isn’t taxed from people and then given back to them.

    But you’re not arguing to do away with some of the distributional concerns of the payments system. As I understand it you want to favour people with children (by giving them larger tax free thresholds). I’d rather do it with tax credits. The difference as far as I can see it is that both policies tax families with children less but using tax credits would introduce ways of targeting that tax relief. They would target the tax relief to poor and middle income families and means test away the high income families.

    So at least in terms of distribution, it seems your proposal is similar to the status quo with the advantage of it being done by not raising revenue rather than recycling it (which is a good idea by me). But your proposals make the tax relief less targetted and available to rich and poor alike. Which of course means if we did it on a revenue neutral basis, the size of the benefits will have to be reduced.

    Am I getting warm?

  13. backroom girl says:

    If I could just add a few cents worth to this argument (sorry, discussion). The issue about how to treat Family Tax Benefit is key. Conceptually, Family Tax Benefit is part of the tax system, not the social security system (that’s why it’s called family tax benefit, not family benefit). It’s just that in Australia, unlike some other countries, we recognise that such a benefit is more useful to many people as a fortnightly payment than only at the end of the year, so we give people the option to claim it fortnightly as if it was a social security payment But anyone who wants to reduce their level of tax/social security churning can always wait until the end of the year and claim their FTB then. The fact that so few people do this may illustrate just how important FTB recipients think churning is.

    Unfortunately, because FTB is means-tested it leads to high EMTRs (though I would point out that these are very much lower than they used to be, precisely because this government has made FTB so much more generous). But the alternative to this is to have very generous universal family assistance which would have to be financed by, you guessed it, high tax rates. Or you just have much higher tax thresholds, as Peter S seems to prefer, but that leaves you with the problem of how to assist families who don’t have a high enough income to benefit from tax breaks. As soon as you introduce some form of refundable tax credit to deal with this issue, you are right back where you started.

    By the way, the high EMTRs that NATSEM seems to be so worried about are at the low end of what you would expect from a means-tested system – in fact most of the people experiencing EMTRs greater than 50% have EMTRs between 50 and 60%. But if you defined high EMTRs as more than 60% you wouldn’t have much of a story, would you?

  14. Peter Saunders says:

    Backroom girl, I don’t see why it should make any difference to my argument whether you are given your benefit fortnightly or annually, the point is you are still being made to rely on a payment by government when there would be no need for you to be in this demeaning position if the money hadn’t been taken away from you in the first place. (I have, incidentally, made the same point in relation to our absurd income tax system which deliberately over-taxes you so you can be grateful when the ATO gives you a little bit back again at the end of the tax year).
    Nor do I think an EMTR of 50+% is a minor concern. I agree with you that high EMTRs are inherent to means tested benefits systems, but isn’t this an argument for looking for an alternative system that does not rely on means tested benefits?
    This brings me on to Nick’s point about redistribution and support for families with children. My view (set out most recently in my essay with Barry Maley in my edited CIS volume, ‘Taxploitation’, chapter 7) is that the TFT should ideally be set as equivalent to the welfare floor for any given family type – thus, around $13K for singles, $21K for couples opting for joint taxation, $21K for single parent with one child, $27K for couple with 1 child, $30K for couple with 2 children, etc. This would guarantee that nobody is taxed before they have earned a subsistence income. It would also mean that nobody earning above these thresholds would need be given any further state support. Middle class welfare thus all but vanishes.
    The problem, of course, concerns people earning less than their TFT. Given the minimum wage, this will only affect families with children, so the single and joint couple thresholds are fine. The problem concerns low income families with children (e.g. a couple with 2 kids earning less than $30K).
    Assuming we want to avoid the present system of gradual means test tapers creating extensive welfare dependency across a wide span of incomes, there are only 2 possible solutions that I can think of. One is to have a 100% withdrawal rate below the TFT, but this would create a major disincentive effect over a tight band of income, which I consider unattractive. The other is to turn the child’s component of the TFT into a non-means-tested tax credit. Parents could then claim, say, a $4K tax reduction for each child, and those earning less than $4K per child above their threshold could claim the shortfall as a cash payment.
    A system like this would keep the great bulk of families out of the government payments system, in contrast to our present system which sucks more and more into it. It would also overcome the work disincentives associated with means testing, for many low income families would pay no tax, and no low or middle income families would face an EMTR above 30%.
    I estimate this change would ‘cost’ $12 billion in revenue foregone, but some of this would certainly be recouped in efficiency gains and higher employment rates.
    I hope this answers Nick’s question (all of this is spelled out in my essay with Barry Maley).

  15. backroom girl says:

    Peter, just a few points in response

    I’m not sure why a means-tested tax credit is ‘demeaning’ and an increase in your tax threshold (supplemented by a means-tested refundable tax credit) is not. While I’m sure that most Australians would rather pay less tax than they do at present, I’m not aware of any evidence that FTB recipients in Australia regard receiving FTB as demeaning. I also don’t believe that most Australians regard themselves as ‘welfare dependent’ if they receive FTB. (In many households, of course, the person receiving the FTB is different to the person who would be paying less tax, which might affect some people’s views.)

    As to whether it is necessary to have EMTRs of less than 50%, I am not aware of any realistic proposal that would allow income support for people without work to be paid at existing levels and not have EMTRs of more than 50%. Any degree of means-testing will create EMTRs this high over at least some of the income distribution and I believe that a universal non-means tested basic income that was enough to actually live on would require every dollar of private income to be taxed at more than 50% just to pay for it in the first place. It really isn’t possible to reduce everyone’s EMTR to 30 per cent without leaving a hell of a lot of people worse off than they are at present.

    Assuming that your estimated $12 billion in revenue foregone (presumably this is after subtracting off the money that is currently paid out as FTB?) is correct, that is a fairly large hole in government revenue. Do you have any proposals for how that hole would be filled – or would government just have to cut expenditure in other areas by that much?

  16. Peter, who are the winners and who are the losers in your proposal – compared with the current system?

  17. spog says:

    Backroom Girl and others have it quite wrong. Most of what passes for FTB is welfare money, rebadged as a tax measure. Just because people pass it off as being part of the tax system doesn’t really make it so. Having the butcher deliver bread wouldn’t make it sausages. And I’m sure BG knows this; he/she is probably just being agreeable.

    Peter S. has the makings of the right idea by having tax free thresholds that sit on a par with the payments to the various household types. However, the claim that this would prevent people being in both the tax system and the welfare system just cannot be so (unless there is some cunning trickery at work). It implies a withdrawal rate on the income support side of 100%, and that income support is not assessable income for income tax purposes. It also overlooks one of the great divides between tax and the welfare system – one is assessed on yearly income; the other fortnightly (more or less).

    People should wrap their brains around the idea that the welfare system needs to interact properly with the Pay As You Go Tax system, not the tax system in the broad.

  18. spog says:

    The other point I wanted to make is to take exception to the claim in Nick’s original article that the churn is due to family payments. Some undoubtedly is (most of it, probably, because we don’t have proper tax deductions for children any more). However singles and couples without kids are churning away as well, without FTB in sight.

    Consider the changes the Gov’t introduced to Newstart Allowance, etc, income tests from July this year. They knocked between 10 and 20% off the withdrawal rates as an incentive meausure (pulling down EMTRs). This means that the cutout point for a single Newstart payment is now a little under $800 a fortnight. But the last $130 of this is pure churn – the person pays more in tax than they get in Newstart. For childless couples it’s about the last $220 a fortnight.

    People in these income ranges show up in the welfare stats, but they are net taxpayers. If the tax and welfare interaction had been designed better, they wouldn’t be on payment, and could probably still be getting the same in-hand dollar amounts. I think this is poor design. It also means people need to take the welfare dependency numbers with a very large grain of salt.

  19. Spog,

    I am not defending any or all of the payments system. I’ve been critical of the idea that getting rid of churn should be a high priority in this area. If the matters you have drawn attention are inappropriate, they should be dealt with but not primarily because churn is an issue – it is that the net effect of the various policies are not working well together.

    As I’ve said, I’m actually in agreement with Peter that appearances are of some significance here – that it’s better for people to get things as targeted tax relief rather than get payments from the government. But that’s because you care about appearances – because you think appearances have significance in terms of values and culture.

    But that can be separated from the argument about who we help (with grants or tax reduction). I think Peter’s support for reduced churning mixes up the two issues.

    But beyond arguing the broad merits targeting assistance according to household income (adjusted for how many people live in the house) I’m not supporting any particular aspect of the current payments system – at least not in this discussion. Further I agree with Peter’s opposition to ‘middle class welfare’ so I don’t agree with lots of those cheques that go out in multiples of $100 the oddest being the $800 for apprentices. On the other hand quite a lot of middle class welfare is the result of reducing EMTRs – you can’t argue for benefits and against middle class welfare and high EMTRs at the same time – they mutually exclude each other.

  20. Peter Whiteford says:

    Peter Saunders: A number of comments

    To a significant degree you seem to me to be concentrating on appearances rather than reality.

    Like backroom girl I don’t understand why current family payments are demeaning and why they reduce people’s independence, but if they are paid as a tax reduction they are OK. Is there any evidence that recipients of family payments would think of them differently if they were paid in a different way?

    As I read your proposals families would receive non income-tested payments below their new tax thresholds, so for these families your system is the same as it is now, so in your terms you still have a system that is demeaning .

    To me the problem is that if you wish to go further and encourage private accounts for education and health, the effective tax thresholds you are talking about must be even higher. If for arguments sake the cost of a public secondary school place is about $4,000 a year and you have a marginal tax rate of 30%, then you need to increase the tax threshold for a one-child family by an extra $12,000 a year. Health costs increase this.

    Let’s say that we introduced extensive personal accounts along the line of Singapore’s – wouldn’t most people perceive these as no different from taxes in the sense that the Government is telling you what to do with your money?

    I would note in passing that I have severe doubts that Australian social values are anything like those in Singapore – a point you have made, I think, about people who advocate a Swedish social welfare system. Singapore is economically liberal when it comes to macro-economic policy, but the government is authoritarian and social values are conservative (and presumably derived from confucian social and family values – not exactly prevalent in australia). The Gini coefficient in Singapore is about 0.49 compared to roughly 0.30 in Australia – and they have one of the lowest fertility rates in the world – in my view an indication of a society with significant internal strains (mainly relating to the position of women). Also while they have high rates of home ownership most of the housing was built by the government and sold off at subsidised prices.

    In contrast, Australia has one of the most redistributive social security systems in the world and a strong commitment to egalitarianism and questioning of authority – about diametrically opposite to Singapore.

    Your proposal also does not deal with the churning that is actually heaviest on low income people – that is the level of indirect taxes on their consumption. why does reducing income taxes have higher priority than reducing indirect taxes?


  21. Peter Saunders says:

    Backroom girl (post 16) and Peter W (post 21) both query why government top-up payments should be considered demeaning. The point is that learning to rely on government to give you money is destructive of the spirit of self-reliance. If many recipients do not find this demeaning, it is only because we have all become so habituated to dependency on government hand-outs, and because so many people receive them that it now seems ‘normal.’
    As illustration: At the time of the last federal budget I participated in the ‘Life Matters’ programme next morning where we responded to listeners’ calls. Overwhelmingly, callers went on about what they were being given, or not being given, in FTB, Child Care Benefits, etc., expressing indignation that some people were getting more than them, or that their needs were not being properly recognised by the government. It was a thoroughly depressing experience, for the subtext was that we are all needy victims requiring and deserving government assistance.
    One of my core objections to churning is the way it transfers power to politicians while making great swathes of the working population dependent on their arbitrary beneficience. The results are seen in the scramble for handouts on the one hand (“what is the government doing for me?”), and the brazen political auction of taxpayers’ money for votes on the other (“vote for us and we’ll give you even more of your own or other people’s cash”).
    None of this happens if people are simply allowed to keep their own money.
    Incidentally, Peter W asks whether “recipients of family payments would think of them differently if they were paid in a different way?” But like Nick’s comments about “giving the money back” (see my post 11), this again fails to grasp what I am saying. To reiterate: I am not saying the money should be paid “in a different way,” nor that it should be “given back.” I am saying it should not be taken away from people in the first place.
    Peter also doubts whether people would regard compulsory payments into their own personal accounts as any different from paying tax to the government. Answer: yes they would! Look at super. If people saw no difference between the money they (through their employers) pay into their own super accounts and the money the government takes off them in tax, they would not make additional voluntary contributions into super. When did you last hear of anyone making additional voluntary contributions to the ATO?
    Just as receiving a government payment is very different from keeping your own money, so too paying tax to the government is very different from paying into your own savings account. These things may appear to be economic equivalents, but sociologically, psychologically and politically they are complete opposites.
    Finally, thanks to Spog (18) for insisting that FTB payments are welfare payments no matter how much Mr Costello may insist they are tax transfers (although Costello actually tries to have this both ways – see And yes, Spog, it is true that my proposals do not eliminate tax-welfare churning altogether, but they do eliminate middle class welfare (for nobody earning above the threshold needs or gets any kind of government payment).
    I think this discussion has been very useful, but I’m now bowing out as I’ve been away from CIS for the last 3 months and lots of other things are pressing for attention. Thanks to all contributors.

  22. backroom girl says:

    Peter S

    I know you have now withdrawn from this conversation, but I would like to assure you that I do actually have quite a bit of sympathy with your preference that people be more self-reliant and for government policies to encourage and assist this.

    It’s just that I think you underestimate the practical difficulty of delivering what is currently paid through FTB through the tax system, not least because it is employers that have to administer the tax system and they are not interested in anything that makes their lives more complicated. I think there is a very good reason why most countries in the end have opted for a system of income transfers to assist families with children (even in cases such as ours where they pretend that they are part of the tax system). Income transfers are designed to be targeted (or not) on a family basis, while most tax systems are based on the individual.

    You can only really make tax-based family assistance simple by making it universal, so that you don’t have to take account of the income of both parents at once. I agree that this would significantly reduce churning, but it would also have to be either much more expensive than what we have now, or much meaner in terms of maximum assistance, or both. The system we have, imperfect as it might be, is in the end a compromise between trying to give people without their own income enough to live on and keeping the overall costs down, while trying to ensure that there is still sufficient financial incentive for people to become more self-reliant. Nicholas is quite right when he says that it’s not really possible to be against both high EMTRs and middle-class welfare – you have to have one or the other unless you want to abandon any pretense of providing an adequate income at the bottom of the scale.

    While I would be one of the first people to say that I think the current government has gone further than necessary to throw more money at families, the political reality is that now that they have, changing the system in any significant way would be extremely difficult. But good luck with your campaign.

  23. Peter thanks for the time you’ve given us, but aren’t you being a bit of a tease?

    I don’t know about others, but I’ve got this feeling that we’re just inches away from nailing down our differences. Just another 2 minutes of your time for you to set out who wins and who loses from your proposals?

  24. Bring Back EP at LP says:

    Can i just say thanks to Nick for first of all using that large Gruen intellect to write on this topic and for the comments made after that. Special thanks in particular to the two peters but in essence all contributions have been highly educative for boofheads such as I.

    I do agree with nick. the evil Peter S is not very far away from nick.

  25. Peter Saunders says:

    Never let it be said I was a tease! Actually, I published some benefit/loss estimates over 2 years ago.
    We don’t have the resources properly to model these proposals at CIS, but when Barry Maley and I first started toying with the idea of combining TFT rises with non-means tested child tax credits (and scrapping other benefits), we estimated patterns of gains and losses as follows:
    Small gains for most single, childless workers (non-workers on benefits unaffected);
    Lower net incomes for families relying entirely on welfare, and for those with one part-time minimum wage earner (welfare claimants with no other source of income are the biggest losers – a single parent with 2 children who has no other income is 13% worse off);
    Slightly lower net incomes for families with one full-time minimum wage earner, but substantial increase in net incomes for families earning between the minimum and the average wage. Above that people make the same absolute gains (but lower relative gains).
    These patterns are not surprising – part of the intention is to strengthen work rewards and incentives, and that clearly is achieved, for nearly all households with at least one full-time worker benefit, and their EMTRs are slashed. But as Backroom Girl predicted, we lower EMTRs and end middle class welfare at the expense of lower levels of welfare support for families with no other earnings.
    If these reduced incomes for welfare dependent households are deemed unacceptable (or politically impractical), Barry and I thought one possibility would be to enhance the value of the Child Tax Credit for children under five (where there should be no expectation of the primary carer getting a job), although this would add another $2.5 billion to the total cost. (incidentally, answering an earlier query from Backroom Girl, my view is that we should require government to find savings to cover these changes and should not be bound by the revenue neutrality condition, which simply institutionalises current high levels of public expenditure).
    All these calculations (which are very rough, and which rest on an earlier and slightly different version of our proposals) can be found on pages 16-20 of Maley and Saunders, “Tax Reform to Make Work Pay” (March 2004) – download from These proposals are not cast in stone – we can play around with them to generate somewhat different outcomes for specific groups – but this is basically the benefit/loss pattern, and it’s not one that I believe is totally unreasonable, though I appreciate others will disagree.

  26. spog says:


    In response to your 20, I think churning has more significance than logic (or perhaps just arithmetic) would suggest. One of the points Peter S. is constantly making is that it might just be better to not take the money in the first place, rather than recycle it back, even if the end benefits (financially) are the same on paper.

    Horizontal equity measures are a case in point. Look at the noise level around Family Tax Benefit Part B. Then cock an ear toward the dependent spouse tax offset – isn’t the silence on this deafening in comparison? What’s the difference between them for households with a tax liability? Functionally they are for the same thing, but one is despised and taunted and the other is ignored. The difference? One is a cash payment and the other is reduction in tax collection.

    Similarly, the point I made in my post 19 was that welfare dependency, as measured by numbers on payments, is artificially high because the Gov’t went down the direct payment route when it could have reduced the tax take. This has knock-on impacts that are additional to the churn involved. For example, these people are picking up concession cards that are really conceptually for people on low incomes, not higher income people who are just on benefits because the Gov’t couldn’t be bothered doing the harder work of fixing tax-transfer interactions. They are lumped in with the poor quite inappropriately.

    Reductions in churning would require a return to sanity on the issue of horizontal equity, something I don’t see coming when people endlessly talk about a need to target. In fact that illustrates my point. As soon as we go down the cash payment route, targetting comes up. Horizontal equity goals are lost if we switch to cash payments. They survive better in the place they should live – reduced tax collection.

  27. Leo says:

    If I read Mr Saunders post above correctly – under his proposal, people on welfare are worse off, most people on the full-time minimum wage or less are worse off, most people above that are better off.

    Is that correct?

  28. I’m in a bit of a Robin Goodfellow mood – I’d just like to remind everyone that the Titanic had targetted lifeboats.

  29. Surely it is not impossible to combine the family benefits with individual payments – the UK has been doing it for years and, at least from the point of view of a simple taxpayer (me) it worked well.
    For example, my wife and I were both working and were eligible for certain tax reductions based on our family situation. When starting with an employer I gave them a form from my previous employer showing how much I had earned in income and paid in tax up to the point where I left.
    I then filled in a form indicating which tax adjustments I was entitled to (as we were both working it took a quick discussion with my wife to work out which ones she was already claiming) and then at each pay date through the year my correct tax liability up to that point was worked out. Any outstanding tax was then paid to the Inland Revenue after being deducted from my pay and, unless there was anything odd in my tax arrangements, most of the time a tax return (and a tax refund / payment) was not needed. Simple.
    Where there was a change to the rules an email came around from HR advising us of it and asking us to update their records if we were affected. I would then check with my wife if she was already claiming it.
    It meant that most of the burden of the year end forms was shifted to the point where you joined an employer and any changes were dealt with on the fly. Although many of them were accounted for by the government as payments, they were actually never taken out of our pay in the first place. A much better and easier system with less work to do at the year end for everyone. It also gets rid of the payments back and forth we get every year.

  30. spog says:

    That’s actually a good point Andrew. The Australian system of Family Tax Benefits can work the same way – people can take it as a reduction in tax deducted rather than as cash payments. Hardly anyone does, which is probably an indication of something (possibly ignorance, but it might also reflect some preferences within couple households for who gets the money in their hand).

    The Australian system requires you to guesstimate your income for the financial year, something at which people are spectacularly bad. There is a rumour that the estimates people submit are recorded by Treasury and then compared to their actual incomes. Anyone who has correctly estimated their income and expenditure in advance is then hired by Treasury in an effort to improve that Department’s own forward estimates processing.

  31. Peter Whiteford says:

    As Spog points out it is possible to take the family tax benefit as a tax refund at the end of the year, but you have to be happy with giving the Tax Office an interest-free loan – and you also can’t do this for any rent assistance you are entitled to. You can take FTB in tax reductions throughout the year, but then you have to be happy telling your employer what your husband or wife’s income is, and also for them to calculate what your payments will be, so some people may not like this for privacy reasons (and you also can’t do this for rent assistance).

    It is inherently difficult to deliver large income-tested payments through the tax system, however. I’m not sure from what he says whether Andrew Reynolds was receiving any of the Child Tax credits that now exist in the UK, but they have exactly the same overpayments problem that Australia had in the first few years of the the Family Tax Benefit. See for some of the gruesome deails.

    Like Australia they are solving the problem by throwing a lot of money at it.

    It is worth noting that for a couple family on benefits or combining benefits with low pay (up to about 40% of average wages), Australia now has the most generous family payments in the OECD (and presumably therefore the world), while for a family with somewhat higher pay (around 50% of the average), the UK is the most generous, followed by Australia.

  32. Peter W,
    No, we were not there for that pleasure (?). I doubt that we would have been eligible. To me, though, the system where the calculation is remade each pay is the best one – ensure that, provided you are a normal PAYE taxpayer and you have no abnormal circumstances, you do not need to submit a return. This cuts down on so much work, saving everyone a lot of time and effort – except for the employer, but even there the load is not too heavy as the computer programs take care of most of it.

  33. derrida derider says:

    Of course the FTC faced considerable opposition from employers in the UK for exactly the reason Andrew stated. But it also caused some fears that employers would use it to lower some wages (as they know what the person’s household net income is) and therefore appropriate some of it for themselves (don’t forget that Australia’s original Child Endowment was introduced in wartime as an anti-inflationary measure explicitly to do this).

    Wherever you have a degressive effective tax scale (ie EMTRs higher on lower incomes than for slightly higher incomes) and collect tax/pay benefits during he course of the year will inherently overpay some people on a full-year basis. As I’ve pointed out elsewhere a practical EITC in Australia would have to face this issue, f’rexample.

    Backroom girl has made some references to the “wallet vs purse” issue (ie who within the couple actually gets the money). Historically this was a big factor in design of these things – the sociologists provided strong empiric evidence that using tax and child benefits to transfer money from the wallet to the purse boosted the amount spent on the kids. That is, intra-household churning was deliberate. I dunno if this is still the case though – the “male breadwinner/female housewife” model ain’t what it used to be.

  34. derrida derider says:

    Oh, and BTW Peter S – give Ann Harding a call if you want to model these things. STINMOD is really easy to use. Or you could use MITTS in Melbourne which also does the behavoural changes in labour supply, but it has a much steeper learning curve (ring Kostas Mavromaras if you’re interested).

    Both bodies price-discriminate according to customers’ elasticity of demand (as revenue-maximisers should), so a plea of poverty probably won’t fall on deaf ears.

  35. Lance Kelly says:

    Australia is on the downward spiral. As I am sitting here at my desk I am wondering about my return to Australia after 7 years sabbatical in another country. Leaving as a single and coming back as a family I find my mind churning over the issues that lie prevalent in Australia every time I read some news about Australia. The burgeoning welfare state mentality that society leans on, the lack of efficient business entreprenualship the policies of free education and free medical care at the cost of low to mid income earners.
    All these issues have their drawbacks and their advantages but high on my mind is the very severe lack of Australian Identity and patriotism. Yes, it is for many Australian a politically incorrect word, but when you visit and live in other countries you realise that without it you are just not a country at all and very lax in identity. Most countries are proud of who they are and what they are because on the one hhand the government encourages it and on the other the national days of holidays bring about festivity and togetherness. As an Australian growing up in Australia I found this sort if activity lacking in attendance or indeed non-existant.
    Why? I think it is simply our lack of thought on producing positive minded people in Australia that are proud of who they are and what the are. Our Army is very small compared to the size and population of our country. Any time Army Service is touted people run for cover, but it is the training ground for the appropriate future leaders of a country, without it you end up with social upheaval and welfare state mentality. Business is surrounded by political interference in just about every area in Australia. It’s no wonder many small businesses end up on the heap, with the sort of interference that runs rampart.
    When you grow up inside Australia you dont know any different and its impossible to objective about what you dont understand nor care too. With the rise of globalism and world economics alomg with free trade, Australia cannot depend on the way oversized public service to bring it into the modern globalism age of the 21st Century. One only has to look from outside in to Australia to actually see what is wrong with it. The idea of a fair go has been a misnomer for many a year and we need to get on with the business of making trade and manafacturing and business in a truly global sense without the infringment of beauracracy.
    Reduce public service and increase business should be th motto on everyones lips these days not what money can I get from the government or how can we make that business pay for our welfare.

Leave a Reply to derrida derider Cancel reply

Your email address will not be published. Required fields are marked *

Notify me of followup comments via e-mail. You can also subscribe without commenting.