How does a 76000% tax rate affect taxpayer behaviour?

Bruce Chapman and I have released a paper looking at the effect of the HECS repayment threshold. In 2003-04 (the most recent year we looked at), earning $1 over the threshold reduced your post-tax income by $760, making it – on one view – a 76000% tax rate. The abstract:

Do Very High Tax Rates Induce Bunching? Implications for the Design of Income-Contingent Loan Schemes
Bruce Chapman & Andrew Leigh
We test whether very high marginal tax rates affect taxpayer behaviour, using a unique policy. Under the Higher Education Contribution Scheme – an income-related university loans scheme in Australia – former students with a debt face a sharp discontinuity. At the first repayment threshold they are required to repay a percentage of their entire income, resulting in an effective marginal tax rate that could be regarded as being as high as 76,000 percent. We formally model the taxpayer decision, and then use a sample of taxpayer returns provided to us by the tax office to investigate whether taxpayers bunch below the repayment threshold. We find a statistically significant degree of bunching below the threshold, but the effect is economically small. On net, we estimate that both the deadweight cost and the budgetary loss are less than A$1 million per year, a small fraction of the amount annually repaid through the Higher Education Contribution Scheme. The result has an important implication for the design of income contingent loans for higher education, such as those being introduced in the UK for tuition in September 2006. This is that it is possible to design arrangements in which the first income threshold of repayment is apparently high, but which are still able to deliver relatively high revenue streams in the early stages of income contingent policy reform without important tax payment avoidance consequences. Our findings also reinforce earlier research suggesting only minimal bunching around kink points in taxation schedules.

Apologies for the gratuitous British references – we’re taking a shot at a UK journal first.

This entry was posted in Economics of Education, Tax. Bookmark the permalink.

62 Responses to How does a 76000% tax rate affect taxpayer behaviour?

  1. Sacha says:

    Sorry Andrew, I’ve only glanced at this. Just an anecdote: I find that I’m trying to reduce my taxable income partly to go onto lower HECS repayment tax rates – have to see what eventuates at tax time!

  2. Ben says:

    Another anecdote: My wife was promoted and received a fairly considerable gross pay increase but we were a little shocked when her take-home pay was less than pre-promotion because she went up a HECS repayment band.

  3. I presume you wouldn’t argue against a more rational phasing mechanism.

    One can pretty much keep what you’re after without having EMTRs of over 100%. I remember a Japanese CEO saying to me that when he found things like this in the tax schedules that affected his own company (I forget what scheme it was but there was a short band over which there was an EMTR of >100%), he thought it reflected badly on the govt – that it looked amateurish. There are probably good arguments to grin and bear high EMTRs to reduce the interval over which payments are phased down, but I would have thought it’s reasonable to try to avoid any EMTRs over 100%. Your thoughts?

  4. Tell me about it, Andrew.

    My current gross pay is $500 below the top cut off. If I get an increment, I’ll get less in hand. Oh the joy!

  5. Sach, the one advantage of course is that you pay the debt off quicker. You should probably factor that into your calculations. And also whether it might be more sensible to borrow the money commercially to kill it off in one go and get the discount for a lump sum payment (if the debt isn’t huge, and you can borrow at a low rate, this can make sense).

  6. Sacha says:

    Thanks Mark – that bears thinking about. Is the discount 15 or 25% ?

    If I was to stay on a good salary over the next few years I’d seriously consider it, at the moment I’d like an overseas postdoc which as you know doesn’t pay a huge amount.

    Think I might do some calculations!

  7. Sacha says:

    Along with the HECS marginal tax rates, there are the SFSS marginal tax rates, although I bet that these are far less significant than the HECS onces as SFSS was only around for a short time and the amount you could borrow wasn’t absolutely huge (double the austudy?).

  8. Andrew Norton says:

    I’ve been in the situation of having to consider whether or not to accept a pay rise, because of the effect of HECS on my disposable income. But Andrew L’s paper suggests that not all that many people hold their incomes down in this way. Part of the reason is that this is not an EMTR in the way normally discussed, since the money goes toward paying off a debt you owe – it is not money lost entirely as it is for welfare recipients moving to work. The only reason for staying below the threshold is a strong need for immediate cash flow (as was the case when I was making the decision, but I ended up being very modestly ahead in take home pay for an otherwise good pay increase).

    While it sometimes has perverse effects, I think the current system is the right one. For the alternatives to be revenue neutral, you would need much higher marginal rates as income rose, which would also have disincentive effects, and from a political point of view, much greater sticker shock. And as Andrew L’s paper shows, the tax revenue losses from people holding their income down are fairly small.

  9. derrida derider says:

    Gee, Andrew, Mark’s point is an obvious one. If you scrape in under the threshold you have merely *deferred* repayment, and you pay interest (albeit concessional) on that deferral. That’s a very, very different incentive to actually *avoiding* repayment. I’m not surprised you actually find little bunching.

    If you want to see bunching at a rate threshold, chase Centrelink for bunching near the pension free area and (even more marked) near the pension cutout. There’s a lot of retirees who manipulate their income to just get a little bit of pension so as to keep the Pensioner Concession Card.

  10. Andrew Leigh says:

    Wow – who knew so many bloggers were at the repayment threshold. Perhaps the biggest deadweight loss of HECS is in reduced blogging time….

    DD, you’ll find the discount point addressed at some length in the paper. We estimate that it brings the marginal tax rate at the repayment threshold down from 76000% to about 20000%

  11. Sinclair Davidson says:

    ‘at the moment I’d like an overseas postdoc which as you know doesn’t pay a huge amount’

    Do you pay HECS while overseas? Surely not.

  12. Tanya Price says:

    Andrew, these results are another confirmation that your co-author really did do a great job in designing HECS. I agree with Mark that the study suggests something of the psychology surrounding HECS repayment and would add to it something about financial illiteracy even amongst the apparently educated. Despite Chapman’s work, most folk think of HECS as an interest-bearing debt that must be paid back now and regardless of income.

    British students have been so recalcitrant on the introduction of fees that I have to wonder if a higher proportion wouldn’t minimize their incomes just to prove a point. Are there as few opportunities to minimize income in the UK?

    Sinclair points to one of the problems with HECS: no overseas collection. Death, low income, and living overseas – which avoidance mechanism to choose?! Surely, there will have to be some serious movement towards overseas (even just in the US, UK, and Canada) collection soon??

    “Therefore our data were extracted specifically for this project by the ATO. To our knowledge, this is the first time that the ATO has made taxpayer records available to economists.”
    This is great news – well done!

  13. Tanya – I have a paper advocating that Australia collect from HECS debtors overseas, as they are doing in Britain and New Zealand for their student debtors.

    The issue of charges will calm down in Britain, as it has here.

  14. Sacha says:

    My motivation is really to have a small tax bill when I put my return in! HECS is just a part of that.

    When I actually start thinking properly about these things I might get myself sorted – but I’m not thinking strategically at the moment – just paying off the credit card and organising to go overseas on holidays/conferences (work is paying for me to go to the conference – very nice!).

    I don’t see why people shouldn’t be paying off the HECS if they work overseas. One element about this, and I havn’t read your paper, Andrew N. and so I don’t know if you’ve talked about it, is that the cost of living and tax rates in other countries may be quite different to the cost of living and tax rates in Australia, and it might not leave people with much money to live on if they have to pay an additional amount in tax.

    It might not be reasonable to say convert their overseas income into Aus dollars and use that as a basis for taxing them. Eg, in Vietnam the cost of living and average income is pretty small in Aus dollars, the HECS payable on income turned into Aus dollars might be zero. In the UK, you might earn a lot in Aus dollars but the cost of living is very high, so what appears to be a lot of money is not actually a lot.

    Just some thoughts (sorry for the long post!).

  15. Sacha says:

    re: Tanya’s comment. My partner, when working for NSW National Parks, helped design and administer a cadet ranger course for indigeneous people. He told me that quite a few people in indigeneous communities (the older ones, I think he said) had the idea that HECS was a debt and that they had an aversion to debt (and thus for the kids to go to university). I assured him that it was really just a higher rate of income tax for a while, if you earned enough.

  16. spog says:

    Andrew Norton’s comments about peoples’ views on HECS repayments is, I think, valid for EMTR calculations in general and has, for a long time, cast doubt in my mind on their usefulness as a measure of work disincentive.

    EMTRs are a sum of the various tax-transfer and related components withdrawal/taper/incidence rates but its not clear to me that adding them this way can really provide a number that can be used without considerable care. If one imagines a new form of loan, repayments of which are income contingent, it would be possible to include the effect of that in an EMTR calculation, if you wished. Having used such a loan to buy my new BMW, I doubt that I’d think about the % contribution its repayment makes to my EMTR in quite the same way as I’d feel about income tax generally.

    Similarly, a hypothecated tax that went toward a cause that I had passionate feelings about (positive or negative) is also going to be viewed differently to general tax rates. I’d include welfare payment withdrawals in this too – some people think they are too high; others think reducing their call on the community at a dollar for dollar rate would be quite appropriate.

    I know many former students who don’t view HECS repayments in anything like the same light as tax rates, but some obviously do.

    Having said this, I still think that a design with the kind of sudden drop in income that hitting HECS thresholds (plural – it happens at each of the change points in the repayment schedule, not just the first, albeit with smaller impact) is poor/incompetent/ lazy. Even something like the medicare shade-in arrangement is preferable to this, especially given Andrew Leigh’s finding that most people appear unpurterbed by the current imposition. Presumably they wouldn’t be fazed by the higher phase in rate either.

  17. derrida derider says:

    On the EMTR measurement, it depends what you call “marginal”. I assume you’re going in margins of $1 a year. But why not $100 a year? Or $1 a week? Or 1 cent? Of course, if it’s a true margin in the mathematical sense the EMTR will be infinitely large at any “sudden death” threshold like this, even if only a single cent of income is lost. This is the thing with EMTRs – a single figure is pretty meaningless on its own. You have to know the range over which it applies. And in the case of a sudden death threshold, what the actual amount of loss is.

    But you probably covered all this. I can think of another possible critique to do with the mechanics of tax assessment (hint: there’s an endowment effect; if you’ve already paid the bloody bill in PAYE during the course of the year you’re less likely to to give a last minute charity donation to sneak in under the threshold than if you’re otherwise facing a bill), but I suppose I’d better read the bloody thing before I shoot my mouth off anymore.

  18. Sinclair Davidson says:

    ‘I don’t see why people shouldn’t be paying off the HECS if they work overseas.’

    Probably because it’s an income contingent loan adminisetered by the ATO. In practice it becomes an Australian-income contingent loan. Of course, if you plan to return there should be nothing stopping you from paying the HECS – but I don’t know enough about HECS to comment. If you’re going somewhere with a strong currency vis-a-vis Australia you could think about borrowing the amount, paying the HECS and then paying the loan in forex. (Many of my contemporaries had bank loans to finance their studies and went overseas after finishing uni and paid back their loans that way – having a high paying foreign job and gaining on the exchange rate).

  19. Sacha says:

    I think you’re right, Sinclair 🙂

    Andrew N, I just read through your paper. Is this the one that Jenny Macklin said was “extreme”?

    Wholly agree that the HECS debt (I’ll just call them all that as I don’t know all the nomenclature) shouldn’t be extinguished upon someone’s death. This doesn’t make much sense.

    Agree that people overseas should also pay the HECS debt, in principle. The thing is how to implement it given what I brought up before – different exchange rates/standards of living/local tax rates.

  20. Sinclair Davidson says:

    Sacha – you might also want to read the exchange between Andrew n. and myself on HECS after death (

  21. Andrew Leigh says:

    DD: yep, we covered that too 😉

  22. cba says:

    lots of interesting comments… so sorry, long post & a tad off the original topic (indulge me, andrew)

    from Jenny Macklin: “Students being pursued beyond the grave, real rates of interest applied to student debt, and loans for the wealthy are again on the education policy agenda, with the release of a new paper from Andrew Norton, a former adviser to the Howard Government.The latest extreme proposals from the Centre for Independent Studies (CIS) are a recipe for the Howard Government’s continuing AMERICANISATION of Australia’s education system.” (my emphasis)

    if Jenny did her research on the US system as opposed to spouting the usual party-line nonsense she’d note that:
    1. the US forgives the loans of the dead and disabled (also from time to time the loans of certain target professions, e.g. teaching)
    2. with consolidation of loans into fixed rates it’s actually been possible in the US system to lock in rates below the rate of inflation at times – although on average the rate is somewhere between risk free (t-bill) and commerical rates.
    3. loan limits on govt sponsored loans in the US are not based on ability to pay. (The smaller, but well-developed private market for student loans in ths US presumably does discriminate in this way)

    my 2 cents on Andrew Norton’s proposals
    *if you planned to offer loans by ability to pay, then beyond a certain threshold those loans should be offered at commerical rates (then you would no longer require discounts to encourage early repayment by a group whose case for subsidisation is shaky at best) – this add-on should be a pseudo-private market (i.e. loans should be originated in the private sector with some backing by the federal govt to promote liquidity).
    *collecting from deceased estates is probably only worth the trouble if (a) delinquency through death is signifciant (speculated to grow, but as yet unproven) (b) loans accrue near market interest rates (otherwise the PV of deceased estate collections at time of origination will be vanishingly small relative to the amount borrowed). as far i can tell, this is a second order issue except for the case of well-educated house husbands and wives dying without ever repaying, but collecting the HECS of dependants would better solve that.

    one last point on collection from overseas. The first question that needs to be asked is whether expat migration a significant drain on HECS to warrant the effort? i don’t have an answer

    even if it is, the fundamental character of an Income Contingent Loan ensure that collection is prohitively expensive to implement without using the ATO and their definition of income. obviously, ex-pats rarely have Australian income, hence they do not repay HECS. Without a total redo on the notion of tax residency and unwinding/rewriting every tax treaty Australia has this isn’t going to change. Asking other foreign nations to use their tax system to collect from ex pats isn’t going to happen either (at least not for free). Any other means of collection will be about as successful as chasing unsecured credit card debts across borders.

    An alternative might be to charge commercial rates on the outstanding debt as soon as someone disappears overseas until they return (imagine filling out the “Do you have a HECS debt?” question on your departure card). that might actually make crossing borders to chase the delinquent debt worthwhile after a few years…

  23. cba says:

    apologies for the grammatical bumps in the road (written hastily)

  24. Tanya Price says:

    Andrew N and Sinclair, thanks for the links. Andrew L, I hope you don’t mind the posts going slightly off-topic.

    Andrew N, I agree that HELP debts should definitely be collected from expats. The individual amounts are just too high to be ignored. The nature of HELP provides greater latitude to collect on

    For HECS debts, we need better modeling on current and potential outstanding HECS debts of current and future overseas Australians. I say future because the global labor market for professionals is growing rapidly – i.e. the number of disappearing HECS debtors is only going to increase

    I’d propose a ‘small-scale’ foreign collection program. If we focus on the UK, the US, and NZ, we will obtain most of our high-income earning, HECS-debt bearing expats. The UK and NZ have a definite incentive to do a reciprocal tax-system-based deal with us. That’s less the case with the US, however the US IRS administers foreign tax treaties and we do have the (admittedly still small) new E3 visa scheme. Set the thresholds according to Purchasing Power Parity plus a little extra for generosity’s sake (PPP would sort out the cost of living problems raised by Sacha).

    I would stay away from punitive schemes like imposing real rates of interest or restricting access to passports … they will only produce behaviours that are counterproductive to Australia’s interests. Folk will nick off and never return.

    Sinclair, can I gather from their strategic behaviour that your ‘contemporaries’ are of the dismal sort?

  25. Tanya Price says:

    Opps, forgot to finish this: “The nature of HELP provides greater latitude to collect on” … debt. HELP is clearly a loan and those that take out HELP should be made aware that they should repay regardless of where they live.

  26. cba says:

    i don’t understand in any sense (other than political expediency) how the zero real rate of return for any HECS debtor is in Australia’s best interest if paying for education is. This is especially true for offshore debtors who can delay payment indefinitely yet have an ability to pay.

  27. Tanya Price says:

    CBA, we could debate the zero-interest-rating of all HECS debts but what is the advantage in applying interest to only the debts of Australians that are overseas? Such a scheme would be morally problematic, administratively difficult, tactically counterproductive from a national human capital perspective, and, most likely, unprofitable.

    Consider some of the administrative issues … Many overseas HECS debtors earn below the current payment threshold. Should they pay interest simply because they are overseas? What if they are studying or are undertaking work experience that will be helpful for Australia? Should Australia punish them for that? Or should we ask people to predict their income and then apply interest accordingly? Could we trust those ‘predictions’? How/when/where would they actually pay in ways that works with debtors and the government? What do we do if they don’t pay? We don’t have access to their assets, so what’s the point? Should we threaten to whop them hard on their return? Why set up a department of big bad bureaucrats whose job it is to send pointless letters of demand? Meanwhile, where are the $$$?

    Gentle, low-cost sticks that work within the central – and well tested – logic of HECS would best suit everyone’s long-term interests.

  28. Andrew Norton says:

    If we do a reciprocal tax-based deal with the UK and NZ, one option may be for HECS debtors to repay at the same rate as debtors under their national loans schemes (or for UK and NZ citizens here, at Australian rates). This would be a relatively simple way of doing it, keeping in mind that employers of graduates have to do the initial calculations of repayments. Tax arrangements for each country are complex enough without requiring them to keep track of an aspect of another country’s system as well.

    I also agree with Tanya that this problem is likely to grow due to the globalisation of labour markets. I only included the most recent figures in my paper, but numbers in the departure from Australia categories I examined had increased significantly in the 1990s and early 2000s, consistent with what we believe is happening to labour markets.

  29. cba says:

    tanya – in both of your posts i think you make some good points, but i’m not giving in 🙂 …

    i agree that the list of exemptions would have to be pretty long (education, charity, government special interest etc – many of these actually retain their australian tax residency anyway, right?) and, yes, even this could be costly to administer.

    i don’t know about fairness… it’s not like there isn’t a precedent for such discrimination: the current discount scheme gives an advantage only to those with an ability to prepay. i guess you think well why should the expats who do the right thing pay for the blackguards that never repay? in my view anything less than a “fair” market rate (say t-bill + 1 or 2%, perhaps even capped as protection from bad monetary policy/fiscal ill-restraint) is still more than fair. the blackguards still have to contend with that, and it might actually be an estate worth chasing down the road.

    i think perhaps you underestimate the cost of an international income contingent loan collections scheme by comparison with my (admittedly hastily conceived) proposal, even one that is bilateral. but more importantly, I think you overstate the likelihood that a hecs/help debt, or interest charged on it, is the deciding factor in an ex-pats decision to return to oz, which presumably means you think it is a big part of the reason to leave in the first place? i don’t know about others, but it certainly wasn’t for me or any other ex-pat I know.

    i really think that HECS/HELP loans need to be thought of as loans rather than contingent income tax obligations, especially for ex-pats. so i’m going to have to disagree with your final para.

  30. Sinclair Davidson says:

    ‘Sinclair, can I gather from their strategic behaviour that your ‘contemporaries’ are of the dismal sort?’

    Actually, no. They were mostly medical students who had accumulated HUGE debts over their six-year degrees and internships.

  31. Sacha says:

    Cheers Sinclair, I’ll check the link out.

    This is an interesting thread to read! It’s good to see.

  32. Sacha says:

    “If we do a reciprocal tax-based deal with the UK and NZ, one option may be for HECS debtors to repay at the same rate as debtors under their national loans schemes (or for UK and NZ citizens here, at Australian rates). This would be a relatively simple way of doing it, keeping in mind that employers of graduates have to do the initial calculations of repayments.”

    Andrew, this is a good suggestion at first glance, as presumably the PPP issues have been dealt with when these tax rates were set. It’s possibly administratively easy to do, although govts would have to remit monies to each other.

    Some labour markets are effectively global at the moment, eg in the sciences – I’m hoping to take advantage of that!

  33. Tanya Price says:

    Andrew N, using the UK and NZ schedules is a great idea. It solves both the administrative cost burden and local cost of living issues.

    CBA, I think there is considerable variation in the incentive/disincentive structure of HECS debts as they relate to temporary/permanent migration. I don’t think that having a debt is the reason why most that leave do leave. As Graeme Hugo’s work shows, interest in seeing the world and taking advantage of work opportunities not available at home are the clear winners. Ditto for the return home: Family, lifestyle, etc are the winners there. However, I am concerned that selectively-applied, punitive measures add to the disincentives to returning home: the tax situation, IMHO, needs improving and the job situation is comparatively limited. We also need to think about the increasing size of HECS debts. I started university in ’89 and ended up with a debt of $10,000 or so from memory. However, more recent students have much larger debts – does that mean different ways of thinking about repayment?

    As I wrote, we could debate the costs and benefits of the zero-interest-rated HECS vs standard loans. I do agree with you that there are benefits to loans. One of my major concerns with using HECS rather than loans has to do with debt aversion … but it’s a complicated issue with ramifications beyond sorting out the no overseas collection problem.

    My late Friday night take on the topic: Overall, HECS is an excellent policy and may just need tightening up around the edges to fit changing conditions like increasing expat debtors. Let’s find high return and minimally-punitive fixes.

  34. spog says:

    Sure, but to return to the original point, can we do it in a way that does not involve reductions in income as one crosses a threshold. Regardless of whether people are not particularly responsive to the effect, some will be hurt by it. It’s inequitable and lazy policy setting.

  35. Sacha says:

    It is lazy policy, isn’t it?

  36. derrida derider says:

    spog’s hit on the other thing about high EMTRs. Economists tend to think about them as purely an *efficiency* issue. But where they’re high enough they raise *equity* issues too – in fact it’s the equity issues that dominate popular (as distinct from academic) discussion of incentives. Most punters think it unfair, rather than economically wasteful, to have someone work harder and end up worse off as a result.

  37. spog says:

    I hope I’ve done the ‘rithmetic right on this, but these numbers might be of interest.

    Using the 2005-06 HECS repayment arrangements, a person with an income of $35,870 is just under the HECS trigger point. Give this person an increase in income, HECS kicks in and they lose money. They won’t return to their previous net income until they get to around $38,100 (at that point, their effective average tax rate on that $2,230 is 100%).

    Continuing with this process up to the final HECS trigger point at $66,924 results in an effective average tax rate over the entire $31,054 range of about 50%. Interestingly (to me) this is about 20% higher than the tax rate otherwise applicable over this range (31.5%). I say interesting, because in my first post I mentioned the medicare shade-in arrangements as a possible alternative approach. Medicare is supposed to be a flat 1.5% rate, but lower income people don’t pay it until they reach a threshold, and then the rate cuts in at 20% until it catches up with what the 1.5% liability should have been.

    It would seem possible to run exactly the same approach – use a 20% rate from the existing HECS threshold – and by the time the top trigger income was reached, the amount recovered would be the same. Presentationally, you could still claim that HECS was calculated using the existing schedule, but that there were special arrangements to ensure no-one went backwards. This would give an EMTR from go to whoa of 51.5%, which is not much over the existing top marginal tax + medicare rate. But no-one would go backwards.

    The question is, would people react differently to this “concessional” scheme than to the sudden loss approach we now have?

  38. Andrew Norton says:

    Spog – Given the added complexity and compliance costs involved with your scheme I expect it will be very attractive to the Howard government!

    More seriously, I think Andrew L’s paper suggests that though in theory the HECS repayment scheme can have some perverse effects such as people holding their income down, in practice these are not that common – perhaps because they don’t really ‘lose’ their money, they just reduce cash flow now and increase it later (because they are paying off the debts more quickly). Therefore in the interests of simplicity we should keep the system as it is.

  39. Tanya Price says:

    dd:“Most punters think it unfair, rather than economically wasteful, to have someone work harder and end up worse off as a result.”
    Yes, but they are paying off a debt rather than paying an ongoing tax. Despite the popular commentary, I wonder if many debtors see this as similar to forced saving. While it does seem inequitable at the lower end of the schedule, these folk may be the most likely to accept the idea of rapid and inescapable debt repayment. Of course that doesn’t mean that it’s not inequitable in a relative sense, but the real inequality of HECS repayment is in the discounts for upfront / lump sum payment.

    Spog: ” “The question is, would people react differently to this “concessional” scheme than to the sudden loss approach we now have?”
    As Andrew’s paper suggests, probably not. Well, not in the sense of income minimization around the threshold. They would have extra cash that they might save and use for lump sum repayments though at such a low income I’m not sure of the likelihood of that occurring.

  40. spog says:

    Andrew: Which seems more complicated to you?

    The current scheme:

    HECS repayment (via tax system). Find your income in the table, then apply the % rate to the total

    From To Rate
    0 35879 0.0%
    35880 39987 4.0%
    39988 44147 4.5%
    44148 46487 5.0%
    46488 49971 5.5%
    49972 54131 6.0%
    54132 56991 6.5%
    56992 62763 7.0%
    62764 66923 7.5%
    66924 whatever 8.0%


    HECS is the lesser of
    8% of income; or
    20% of income above $35,879

    For an employer with a payroll system, this is done automagically, so neither approach is more complex. For manual calculators, I don’t think one is particularly harder than the other.

    For the ATO doing at end of year tax time, neither approach would tax its computer system.

    Complex it isn’t. And even if some people see it as more complex, the difference is so minor that I don’t think simplicity is a valid excuse for the silly approach we now have.

  41. spog says:


    That looked completely different in type-in box. Now the table is hard to read.

    Maybe I could argue that makes the current system even more complex!

  42. cba ;) says:

    spog: you’d probably need to put in a cap on the max payment under your scheme, or our friends on high incomes would cry foul [cap could be based on a standard loan repayment term of say 6 years]. it’s a good proposal otherwise

    tanya: “the real inequality of HECS repayment is in the discounts for upfront / lump sum payment” just to clarify – the discounts aren’t necessarily unfair, they’re in part a natural by product of the zero real interest rate. the present value (PV) of a $1,000 hecs loan is less that $1000 because the loans don’t accrue real interest unlike commercial loans. any reasonable budgetary treatment should account for time value of money, at the least. for the govt, it makes sense to offer some amount of discount if (discount rate)*1000

  43. Tanya Price says:

    CBA: Thank you :o) To clarify … from a collection perspective, it makes sense to offer a discount. However, the fact is that those with $$$ – particularly, parental $$$ – get a discount. Should the discount be removed? No … there are better ways to deal with this problem … well, that’s if one thinks it’s a problem.

    Spog: So what do you think the consequences of your proposal might be? How about ball-parking it?

  44. cba ;) says:

    odd – my post was chopped. (the end bit was agreeing with you that it could be unfair under those circumstances you identified, tanya, especially for borrowers with a large balance who would repay in quick time due to their high income)

  45. Andrew Leigh says:

    Spog, thanks for taking the time to do the calculations – along with Bruce Chapman and Andrew Norton, I think you’ve now brought about a 50% increase in the number of people seriously modelling alternatives to the current system.

    I do worry a bit about asking for 8% back from low-income earners, though. One of the neat features of the present system is that we don’t ask for a cent of the HECS back until you’re doing better than the average worker.

  46. Spog – Wouldn’t the EMTR be over 60% for graduates in the next tax bracket?

  47. spog says:

    Andrew N: I haven’t done the modelling specifically, but I had assumed that at or around $66,924 the 20% rate would end and the 8% take over. Given that the next bracket (42%) doesn’t start until $70,000, then it wouldn’t be 60+%. However, given that the HECS limits are changed each year, but the tax thresholds are not, there might come a day when an overlap raises its head. This will happen under my suggestion, or the current scheme though.

    Andrew L: I think you may have misread my suggestion. Nothing would be deducted until the current threshold. Then it would be 20% on each dollar over the threshold. This is less of a deduction than the current 4% on every dollar from $0. The 8% wouldn’t really kick in until the current 8% threshold (roughly – my 20% suggestion is only a guesstimate based on the contribution to the effective average tax rate over the HECS range as mentioned before). We already do this in medicare – instead of 1.5% of taxable income from $0, it’s really 0% up to $15,902, then 20%, then back to 1.5%. I’m just copying an existing approach really.

    CBA: Because the rate falls back to 8% at approximately the current threshold for that rate, the amount collected is never more than under the current scheme, which I believe is uncapped (except to the limit of your debt, obviously).

  48. cba says:

    spog: you’re right – i read it too quickly.

  49. matt cowgill says:

    “it makes sense to offer a discount. However, the fact is that those with $$$ – particularly, parental $$$ – get a discount.”

    That’s actually the main thing that frustrates me about HECS… I have no issue with repaying the money (and my HECS debt is over $30k and still rising), but I resent the fact that kids with wealthy parents end up paying less for their education.

  50. Christine says:

    Matt/Tanya: cba’s right: part of the idea is that those who ‘borrow’ the money from the government are subsidised because of the zero real interest rate on their borrowing, so the up front discount is really meant to offset that. (Also I think in the early days a cash flow argument, but that shouldn’t apply anymore.) I don’t know whether the zero interest subsidy is on average equivalent to the up front subsidy or not, which would be necessary to figure out whether it constituted a discount, but someone must have done some calculations on this. The system would be clearer if the fee were reduced, but there was some positive interest rate applied to the debt. Plus, I would say, some grants to students from low income families.

    You’ll probably never be able to completely remove any benefits from being a kid from a high income family (I’m still jealous of how those guys who went to a swank private high school that had language labs were so much better at Japanese than me going into first year Uni), but HECS is still pretty good relative to lots of other systems.

  51. The ‘discount’ isn’t really a discount. Because real interest rates are politically sensitive the system has a debt surcharge instead, ie the real price is the up-front price and if you don’t pay that you pay a surcharge in lieu of paying real interest on your loan.

    This also redistributes risk; ie those who have high earnings and repay quickly end up paying higher effective interest rates.

  52. cba says:

    AN: the problem with the discounted prepayment or discounted up-front price is that it can only be set to capture the foregone interest on a HECS loan of a particular term. borrowers with capacity to prepay who expect to repay their loan earlier (later) than that term have an incentive to prepay (not prepay). a more efficient discount would condition on balance outstanding and income (to determine the expected repayment time) but this would be a mess to implement and probably still easy to game.

    in any case, to the extent that the discount works, you’re basically saying it’s all price illusion: we either take a discount on an inflated HECS fee, or pay real interest on a lower fee.

    ultimately, HECS fees (and any interest) should be recouped through the labor market if education actually generates valuable skills rather than just providing a screening mechanism (which is one thing you have to be leary of when comparing OZ and US wages, especially starting salaries). i think the move to real interest rates is inevitable in the longer term once we embrace the value of higher education.

    the problem with the risk re-distribution argument is that the guy who doesn’t go to uni on average does worse in the labor market, but is also a tax payer and ultimately shoulders some of this risk shifting burden. it’s certainly not clear cut that risk re-distribution benefits everyone…

  53. Sacha says:

    I remembered this morning that a few years ago, when I was studying part-time and working part-time, that while working out my tax return, I realised that I’d have to pay a couple of hundred dollars in tax as I was a bit over one of the HECS thresholds.

    This would have represented a bit of a problem as I wasn’t earning very much, and I managed to find some legal deductions that put me under a HECS threshold, and the tax office then owed me a small amount.

  54. Tanya Price says:

    “… if education actually generates valuable skills rather than just providing a screening mechanism (which is one thing you have to be leary of when comparing OZ and US wages, especially starting salaries).”

    CBA: Would you mind explaining what you mean by this? Specifically, the connection to the different OZ and US wage levels.

  55. spog says:

    I think it’s interesting that HECS can be included in the PAYG tax calculations by employers, as well as just an extra that pops in at the end of year calculation by the ATO. This means that crossing a HECS threshold won’t just mean a larger end of year bill than was expected, but could produce a reduction in take home pay.

    Neither of these is attractive, I suppose, but the take home pay thing strikes me as marginally worse. I remember the interest generated by last year’s wage case, when it was revealed that for some housholds, the wage rise triggered a net fall in household income. That wasn’t HECS though, just the impact of the rococo intricacies of the tax-transfer system.

  56. ” guy who doesn’t go to uni on average does worse in the labor market, but is also a tax payer and ultimately shoulders some of this risk shifting burden.”

    The working class guy paying for someone else’s uni education is a good populist way of arguing against government subsidies, but is it not really true, since such people are net beneficiaries of the tax-welfare system, ie they don’t pay for what they receive, let alone for what other people receive.

    University subsidies are middle-class churning.

  57. spog says:

    Just to put some numbers under Andrew Norton’s point about net beneficiaries, here are the ROUGH income points (yearly) at which a few different household types become “net taxpayers”.

    Single person – $15,500
    Single income couple – $26,000
    Single parent, two children – $44,000
    Single income couple, two children – $44,000

    As you can see, there is a clear relationship between the number of words used to describe the household, and how far to the right the $ amount appears. Not sure if that is particularly relevant though.

  58. cba says:

    Andrew N/spog: I didn’t intend to make a “populist” argument. i could just as reasonably argue: “the guy who doesn’t go to uni [i.e. doesn’t have a HECS debt] is also a tax payer and ultimately shoulders some of this risk shifting burden” (i.e. irrespective of how life sans uni treats him). why? because investment in education is not a risk free enterprise – its payoff is risky both individually and to some extent in aggregate. the extent of the aggregate risk affects aggregate HECS repayment and some of that risk is borne by non hecs owing tax payers. now don’t get me wrong – i’m all for the diversificatoin of idiosyncratic risk that HECS provides, but my suspicion is there’s some hidden element of aggregate risk hidden away in the hecs system. on balance, it may be too small to matter relative to the benefits, but that’s an empirical issue.

    Tanya: I’ll try to expand (somewhat haphazardly as i’m a little pressed for time)
    (a) That australian university education is inferior to the US is not my claim (certaintly at undergrad level i suspect the opposite).
    (b) (incentives, productivity benefits, externalities etc aside) after accounting for differences in tax and tuition, wages should be roughly the same under a user pays system versus a public provision or combination system – this is what i meant by “ultimately, HECS fees (and any interest) should be recouped through the labor market” and my main point. in the US, the user pays closer to the full cost of provision than in Australia.
    (c) I think there is an important distinction between the productivity benefits of education on wages versus the signalling/screening value of education on wages – i would argue the latter is more likely to simply redistribute wages and is an expensive redistribution mechanism at that since taylored IQ tests could probably achieve the same result. (i think there is more to be said about this that ties in with (b), but it will have to wait)
    (d) do wages differentials in the US and OZ have anything to do with tuition? I’m not trying to say this is the only diff between US and OZ wages, just a factor that needs to be considered. I’m loathe to provide specific comparisons of US and OZ wages because there are too many factors to control for beside tuition. It just strikes me that with a typical $US50,000 tuition bill (thats probably a low estimate, tuition at private universities is now pushing $US30,000 a year) versus half or less in Australia something has to give somewhere.
    I’m happy to hear your thoughts if you have another take or interesting observation to make…

  59. Tanya Price says:

    CBA: Thank you for taking the time. Unfortunately, I don’t have any great observations right now. I need to give it some more thought … and ideally after the end of the semester (!). … I keep coming up with a series of ifs and buts that ultimately point to the difficulties in comparing Aus to the US. For all the problems with the Aus situation, it is much neater than the US situation. If we look across the entire US system, there is substantial variation in both sticker prices and actual prices alone. I used think that the quality of undergrad ed was equivalent but given the restrictions on inputs in Aus, I’m less certain these days. I’d say that the Go8 unis compare well to the US state flagships, but Aus suffers from the lack of high-quality liberal arts colleges. We don’t just need more resources but more institutional differentiation as well. It’s great to see institutions like Melbourne trying to pull away.

  60. Sacha says:

    Andrew, I read the little story in the HES yesterday – did the UK govt not consider the Australian system of repaying HECS?

  61. anxiety says:

    Missing: Tony Luzio Jr.
    Look here for anxiety to call with tips on the possible whereabouts of anxiety people profiled on Without A Trace , and shown on The Early Show . The latest: Tony Luzio Jr.

Comments are closed.