What policies shouldn’t we implement?

Economists spend large amounts of time evaluating existing policies or pushing for some particular new economic policy. Equally important, but less frequently done, is to say what should NOT be done and why it shouldnt be done. Of course these ‘warnings’ have to relate to a policy that is simmering below the surface and is just waiting for some intellectually challenged politician or selfish interest group to advocate it. I invite everyone to add to my following top-5 of ‘no-no’ economic policies which to my mind are in danger of becoming advocated:

1. Make mortgage interest payments tax-deductible. My country of origin – The Netherlands – implemented this policy, with the ensuing predictable rise in housing prizes and the associated large tranfers of the poor and the young (who rent) to the rich and the old (who own and have higher marginal taxes for their deductibles). Its one of the recognised mistakes of other countries and Australia is well-warned not to copy this mistake because once this mistake is made, it would take decades of political fighting to get rid of it.

2. Implement negative income taxes (also known as basic incomes) in order to osensibly solve the incentive problems with targeted welfare programs. The reason not to do this is because it is monstrously expensive: if you dont want to seriously reduce welfare levels, youre going to have to give every adult over 18 at least $10,000 AUD per year. Also, youre going to have to give everyone with dependent children a blanket subsidy (most of the disincentives come in via taper rates on the benefits for the kids), in the order of $6,000 per dependent child. Thats $32,000 AUD for your average 2-adult, 2-kids family. For the whole of Australia (some 15 million adults and 6 million kids) that would be a bill around $186 billion AUD per year. Thats 20% of GDP, whereas the current bill is about 7% of GDP. This creates two types of work disincentives. The first is that it takes away the stigma of not working since the whole population gets welfare. The second is that this additional welfare will have to be paid by additional taxes on those that work of course. Hence the plan would involve massive work disincentives for those currently working as well as opening up the attractive option to welfare for everyone.

3. Adopt a carbon trading scheme that gives all the initial emission rights to the current industry. This would effectively mean a huge transfer from the community (who implicitly owns all emission rights) to polluters, rewarding them for previous emissions.

4. Expand public-private partnership schemes whereby the government guarantees extraordinary returns to a private contractor (for instance when building a road or a bridge) in return for some up-front private funds, and where the contractor is insured against the risk of few future customers. It should go without saying that such schemes a) have huge long-run costs to the community in the form of the promised extraordinary returns, b) do away with the principle that private market parties make decisions on risk, c) coopt governments into using money from the public purse to subsidise companies just in order to make the short-run budget look good.

5. Exempt pensions from income taxes both at the moment of saving and at the moment the pension gets taken up. Such a policy would be a monumental mistake because it is firstly highly inequitable: the rich have more disposable income to save than the poor and face higher marginal taxes (thus having more to gain from tax exemptions). The more important problem with such a scheme is that it opens up incredible arbitrage opportunities that undermine the income tax system as a whole. Any bank could then offer any earner a loan with the future pension payouts as collateral – effectively bypassing income tax altogether, which again will mainly favour the rich. Such a policy would also quickly lead to a large group of stakeholders to make it politically difficult to undo it. And of course the wealth effect of such a policy will give strong incentives towards early retirement for the coming generation of workers, thereby leading to lower participation rates. Oops, Australia has just implemented this policy, making Peter Costello at a stroke an economic incompetent. I fear that Oz economists will be spending decades trying to undo ‘Peter’s Prank’.

This entry was posted in Economics and public policy. Bookmark the permalink.

30 Responses to What policies shouldn’t we implement?

  1. Jc says:

    “….. mortgage interest payments tax-deductible.”

    I would argue that you shouldn’t pay tax on interest either then, Paul. The logic of your argument is intellectually indefensible if you think otherwise. If you pay tax on interest you ought to also deduct interest.

    I don’t think interest deductibility has any bearing on house prices.

    The US has interest deductibility and like for like cap gains relief on homes. I am pretty certain that Oz has experienced higher increases in house prices over the past 30 years.

  2. Jc says:

    ummm should read ……shouldn’t pay tax on interest earned……….

  3. Yobbo says:

    “Also, youre going to have to give everyone with dependent children a blanket subsidy (most of the disincentives come in via taper rates on the benefits for the kids), in the order of $6,000 per dependent child. Thats $32,000 AUD for your average 2-adult, 2-kids family. For the whole of Australia (some 15 million adults and 6 million kids) that would be a bill around $186 billion AUD per year”


    Only if all 21 million have no income. A fairly unlikely sceneria.

  4. Yobbo says:


  5. conrad says:

    “I dont think interest deductibility has any bearing on house prices.”

    Are you joking JC, or have you forgotten that there are supply problems in all of the major cities in Australia where people live ?

    • Patrick says:

      Yes, in none of which is personal mortgage interest deductible as it is in the US (on up to about $1m of loan capital).

      I’m a lot less convinced on this one than Paul is.

  6. Guise says:

    I’d add a corollary to #3: compensate heavy users of water for reductions in their allocations. Match this to an allocations trading system and it’s effectively another way to privatise a public resource.

  7. Mark Hill says:

    1. How come Keating and Hawke made a similar backflip on CGT?

    It should not be implemented, although it will make housing more affordable. All taxes should be cut across the baord.

    There is a laundry list of reasons why housing is expensive, all of which lead to some form of generational inequity.

    Simply cutting or abolishing taxes and regulations on the transfer and development of property as well as land release will help, along with tighter inflation targeting. Environmental regulations like BASIX wouldn’t be worth a pinch of crap if market prices were charged for utilities. (As they should be).

    2. You can’t be serious, opposing a NIT or BI and VAT like this. Costed analysis sees a total reduction in the total tax bill and Govenrment spending, and results in a balanced budget.

  8. Jc says:

    “Are you joking JC, or have you forgotten that there are supply problems in all of the major cities in Australia where people live ?”


    you may then want to explain why US houses prices haven’t performed as “well” as ours over a 30 year period. The US has interest deductibility up to $1,000,000 of interest.

  9. paul frijters says:


    you say
    “I dont think interest deductibility has any bearing on house prices.”

    well, I do believe there’s a relation because tax relief on housing costs changes the trade-offs between the consumption of housing versus other expenses. Call me an old-fashioned economist but I think incentives matter.

    Yobbo, the 186 billion price tag is my estimated cost of the basic income scheme. Of course you might hope to off-set this cost with (increased) income tax receipts, but that doesnt change the cost of giving everyone a basic income. Note that in the present welfare system there are large groups without access to welfare (for instance dependents of rich partners) who would be given access under a BI.

    Mark, I’d like to see those costings you speak of because I’d be ex ante highly sceptical of them. You need an enormous positive labour supply response to make a BI cost-effective whereas I’d in fact expect a negative labour supply response from this scheme since the welfare stigma is taken away. This long-run effect is notoriously hard to estimate and is thus usually ignored in simulations even though the history of welfare reform suggests long-run effects dominate short-run effects.

  10. Jc says:


    You may want to explian that anomaly

  11. Jc says:

    Paul you say:

    “I do believe theres a relation because tax relief on housing costs changes the trade-offs between the consumption of housing versus other expenses. Call me an old-fashioned economist but I think incentives matter.”

    Housing is consumption. Compare like with like. People have to either buy a home or rent. There is no tax adavantage to do either.

    Compare that to the US. There is a clear tax advantage in the US, yet their hosue prices haven’t risen as much as ours.

  12. conrad says:


    the US has an oversupply of houses at the moment (I believe). They also have many cities without such ridiculous planning restrictions as Australia has, and they have more decent cities you can live in too, so if you don’t like living in an expensive city, you can simply move to another.

    If you compare the big cities that have population pressures like New York, I think you’ll find they have had greater price growth than Australian ones. You can check the figures yourself. Try comparing:

    http://www.census.gov/hhes/www/housing/census/historic/values.html for the US


    http://www.blackwell-synergy.com/doi/pdf/10.1111/j.1467-8462.2005.00373.x?cookieSet=1 for Australia.

    This doesn’t even include the last bit of US price growth before the current problems (markets like Sydney have been flat for a few years now).

    It seems to me that in these big cities, many people are willing to borrow as much as they possibly can (as highlighted by the current situation), and thats probably true of Australia and the US. Any measures that allow you to borrow more are therefore likely to increase prices (perhaps that won’t be true for the next few years, depending on how the recent problems pan out). The problem is likely to be worse in Australia than the US due to supply differences.

  13. derrida derider says:

    Well, I’d agree with all Paul’s points except the BI (though note that the carbon trading scheme is only a problem if you give the rights away. It’s because of the likelihood that horse-trading politicans would indeed give them away to favoured players that I prefer a carbon tax to tradeable quotas – exempting a mate from tax is more transparent than fiddling quota levels.

    As I’ve written elsewhere, what matters is not the paper flows through the goverment’s coffers of a BI but the net behavioural effects. Under a true BI/FT proposal the size of those paper flows is extremely sensitive to details of implementation. Do you do it as a demogrant accompanied by a flat tax with no tax-free threshold or do you do it as a phased out GMI accompanied by a phased in tax credit? The first of these implies massive churning but in fact is adminstratively far the easier (a flat tax with no threshold is easy to administer – not to mention hard to evade – and a demogrant carries virtually no administrative burden). But because the after-tax position of individuals is the same in both arrangements they would likely have very similar behavioural effects.

    On those behavioural effects, most simulations in static models of labour supply show that replacing real real life tax and welfare systems with a BI that approximates to the level of welfare payments gives a small (typically 3-5%) drop in aggregate labour supply. It’s true, though, that these models perforce leave a lot out – eg Paul’s bit about long-run social norms, the fact that the biggest reductions in labour supply are amongst the least productive people, etc. Against this there are the well known Mirrlees-Diamond results in optimal tax theory that suggest that a BI/FT gives the biggest redistributive bang for your efficiency loss buck.

  14. Brendan Halfweeg says:


    Where did you get your figures and basis of negative income tax? I think it reasonably assured that a NI’s aim is to reduce the overall welfare budget. Your figures don’t really make a lot of sense and I question the disincentive aspect.

    Any dependant who claims the full NI allowance, say $9,000 under the LDP 30/30 tax policy, would be simply splitting their income with their household. The houshold as a whole will still see a marginal tax rate of 30%. So if you are a dependent of your wife, who has an income of $30,000, the household’s total tax bill is $0. From an egalitarian perspective, income splitting is a good thing, I’d have thought. But your objection is to the rich income splitting through trusts and the like, but 30/30 would reduce the incentive for the wealthy to minimise their taxation by bringing income tax and company tax into alignment.

  15. Niall says:

    regardless of what economic policy is floated with regard to the housing cost/benefit problem, someone is going to lose out. Frankly, I don’t see any broader economic benefit in granting mortgagors a tax deduction on the residence they chose to borrow for. What portion of the debt do you allow deductibility on? The original purchase amount reducing over time? Do you only allow it on fully amortised loans? Do you penalise those who choose to conduct interest only loans, relying on the eventual sale & capital gain for their strategy. How heavily do you tax the capital gain?

    It’s a ludicrous postulation. There’s nothing wrong with deductibility on investment interest where the debt incurred can definitively be proven to support investment housing. Extract of equity for personal use is not acceptable. This can be accommodated within the present taxation admin structure. CGT on investment property ought to be completely abolished as an incentive to take that profit and turn it back into more investment housing. Certainly, this scenario doesn’t solve the rich-getting-richer paradigm, but then it’s not meant to. It’s mean to help resolve the housing shortage.

  16. Damien Eldridge says:

    DD, the relevant comparison is not between exempting someone from a tax and allocating quota. After all, regardless of how the permits are allocated, they still need to be surrendered for emissions. If the permits have a positive price, then there will still be an opportunity cost attached to emissions.

    Instead, the relevant comparison is between the use of the emissions tax revenue and the use of the implicit revenue from the tradable permits. (That is, the revenue that would have been raised if the permits were auctioned.) In effect, a gradfathered allocation amopunts to giving money to the recipients. Since this is unlikely to be the best use of such money, it is a bad idea. It would be better to either reduce other taxes or fund the provision of additional public goods.

    However, you’re underlying point is correct. I think that one of the advantages of an emissions tax over a tradable perit scheme is that it makes the use of the revenue more transparent. This should increase the costs of rent seeking and thereby hopefully reduce the amount of rent seeking that takes place.

  17. Chris Lloyd says:

    Jut a minor quibble Paul. You say

    Exempting pensions from income taxes both at the moment of saving and at the moment the pension is paid ..is..highly inequitable

    becaue the rich are on higher marignal tax rates. That argument pretty much kills any tax deduction, let alone reductions in marginal tax rates.

    “Progressive” tax scales are already inequitable – they are biased against the rich. Calling tax deductions or reductions in marginal rates “inequitable” is, to my mind, an abuse of language.

  18. Yobbo says:

    “Yobbo, the 186 billion price tag is my estimated cost of the basic income scheme.”

    No it isn’t Paul.

    186 billion is

    $10,000 x 15 million


    $6000 x 6 million

    = $186 billion.

    Therefore, your “estimated price tag” is actually the price tag of the NIT if every single Australian was unemployed and had zero income.

    This is completely idiotic and the entire section regarding the NIT should be deleted.

  19. Patrick says:

    Well, France made personal (domestic) mortgage interest tax deductible last month. We can watch the results there for a real-time illustration.

  20. Paul Frijters says:


    the basic figures are easy. Lets take the figure of 10,000: a single unemployed person in Oz got about 12000 dollarsd in 2002 (2002 OECD figures. This includes 9600 in New Start, 600 in income tax credits, and 2300 in rent assistance). This is still an underestimation of the total amount of transfers, which also includes medical benefits and other entitlements. Hence the 10,000 per adult is a gross underestimation. At present I’d estimate the true welfare transfer for a single unemployed adult to be closer to 14000. Similarly, the 6000 derives from looking at what a single unemployed mother gets over and above a single person without kids. Those are the relevant groups to look at because they tell you what you’d need to give everyone if you dont want to short-change some existing groups and nevertheless want to do away with the incentive effects.

    Yobbo, I am basically calculating how much one would initially have to hand out in terms of giving everyone a basic income. The 186 billion is an underestimate of how much it would cost. If you want to make that amount look lower by off-setting it by tax receipts, then that’s your business but you are then no longer talking about the cost of giving everyone a basic income but about some more abstract concept. And dont fool yourself: introducing a BI also entails lowering the marginal tax rates a long way up the income distribution (otherwise you’re back where you started), and hence that 200 billion dollars wont be recouped in a hurry, making a comparison with the current system invalid.

  21. Paul Frijters says:

    what is equitable or inequitable depends on the point of view you take. I take the point of view of the existing prior situation and simply look at who is likely to gain more than others, which allows me to say quite unequivocally that mortgage tax relief is going to be highly inequitable because the rich and the old are going to gain whilst the young and the poor are going to lose. Why I cant call that inequitable is beyond me. Now, of course, if you take the existing prior situation with its progressive income taxes as inequitable itself, then you might call the change in favour of the rich equitable. Its your prerogative to abuse language in that way, but its an unusual way of looking at whether a change is equitable.

  22. crocodile says:

    Policies that should not be implemented !

    Programs such as the likes of private health companies, child care centres and LPG converters etc etc derive a substantial share of their income via taxpayer funded subsidies and handouts.

  23. Yobbo says:

    Paul: I don’t think you understand how a negative income tax works.

  24. Paul, your figures don’t align with what a NI is aiming to do. A negative income tax only guarantees a minimum income for unemployed people and a top up of income to lowly paid people. Once you start earning more than the threshold, NI doesn’t apply and it simply works as a tax free threshold, albeit considerably higher than the $6,000 that we get now. You aren’t paying tax AND receiving welfare like you do now.

  25. derrida derider says:

    Actually Brendan you might be paying tax and getting welfare at the same time, depending how you set your scheme up. As I’ve argued above you might want to do a massive churn because it is in fact simpler and more administratively efficient than alternatives that achieves the same distributional and allocative efficiency goals. Put another way, depending what you do adminstratively Paul’s $200B figure could be right, but the paper flow of $200b does not have the adverse behavioural effects that an additional flow of $200b would under present arrangements.

    Which just goes to show that there’s an awful lot of complete hogwash written and spoken in Australia about churning, and by people who have the expertise to know better. But that’s a story for other threads.

  26. conrad says:

    Paul, I don’t get your analysis either — I get a different number from everyone.

    To make the calculation easy, I’m going to delete all the part-time people that don’t earn 30,000. Instead, I’m going to say that about 40% of Australians don’t work and are over 18, and these people all currently earn nothing. Everyone working gets to earn over 30,000 so they are not included. I note here that I’m including people married to rich people and so on who don’t get welfare now, which is why I have pulled 40% out of a hat. No doubt this number is going to rise a lot as Australia’s population ages.

    This means 8 million people get $14,000 = 112 Billion ($10,000 + $4,000 administration). To me this number says its a hopeless scheme which costs more than now, but not as bad as your figures.

  27. NicM says:

    Re Chris Lloyd (#17)

    Progressive tax scales are perfectly equitable. We all get the same tax free threshold and all pay the same amount of tax at each income level. It just that some of us have more income than others.

  28. Brendan Halfweeg says:

    Derida, are you talking about giving everyone a basic income (BI), setting that BI as the tax free threshold (TFT) and then having a flat tax above the TFT. I suppose it work out more efficient for employers, since all they would need do is deduct tax at the flat tax rate on ever dollar earnt, rather than try to balance the NI. Mathematically it works outs the same.

    Personally, I’d like to see the witholding tax eliminated, it simply transfers responsibility for administration of your income tax to your employer, increasing the cost of employing people. Having to write a big cheque to the state at the end of each financial year would sharpen the sense of how much tax you are paying and make voters more financially aware of the impact of their decisions in the pollbooth. Witholding tax lessens the hurt of paying tax and disguises the state’s light fingered ways with your hard earned. Abolishing it would also end the churn.

  29. Pingback: Club Troppo » Tuesday’s Missing Link on Tuesday (I think)

Leave a 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.