A pragmatic idea for labour market reform

The case for labour market reform is obvious enough. If people have a marginal productivity of labour below the minimum wage, then they will not get a job. The only way to fix that is to (1) increase their productivity; or (2) lower the minimum wage. Increasing productivity is a slow long-run project, so the only short-term option to increase employment is to lower the effective minimum wage (including all the costs of employing somebody, not just the wage). 

Equally obvious is the political reality that prevents real reform. One of the reasons that the Liberals lost the last election was the unpopularity of Workchoices — the overly complex and bureaucratic attempt to marginally increase labour market flexibility. The consequence has been that labour market reform has stalled and gone slightly backwards. 

In this context, what can be done? I have one minor suggestion…

Every year the government wages watchdog considers the case for an increase in the minimum wage. There needs to be a subtle (but important) change of focus. Instead of aiming at a set minimum wage, the wages watchdog should aim for a set level of disposable income.

After all, the whole idea is to help low-income people by making sure they have more money. So the important point is how much money they have, not the level of the minimum wage.

The benefit of this is that after the wages watchdog picks their preferred increase in disposable income, then there can be a separate debate about how this should be achieved — through wage increases,  lower taxes or a handout (wage subsidy).

The ensuing debate will highlight a number of points — such as how 60%, 70% or even 80% of a wage increase can be lost to the tax/transfer system… how the minimum wage is not well targeted, with 25% of minimum wage earners coming from families in the top 30% of disposable incomes… how lower taxes on workers will not only increase their disposable income but also improve the efficiency of the economy… and how higher minimum wages will reduce employment while using the tax/transfer system will increase employment opportunities. 

The downside of using the tax/transfer system is that it will impact on the government budget (through lower taxes or higher welfare spending). This may encourage the government to still use wages policy sometimes… but I don’t think it would cost the budget too much in any one year and so especially in times when unemployment is a political issue we may see the government pursue tax/transfer instead of wages policy to address disposable income targets.


28 thoughts on “A pragmatic idea for labour market reform

  1. If there is unemployment because the minimum wage exceeds labour’s marginal product won’t it always be best to go for a wage subsidy/tax cut if you do want to increase incomes. That outperforms pure welfare because you get something back from it in output. It outperforms forcing a higher wage since that would worsen unemployment.

  2. Totally agree with you hc. This idea has been around a long time, with the “5 economists” outlining their plan in 1998 for a minimum wage freeze (in nominal terms) mixed with an earned-income tax credit.


    But politically it is a hard sell. That’s why changing the focus from “wage” to “disposable income” could be useful… to change the political dynamic and shift the focus on to what matters.

  3. Isn’t that simply propagating bad policy though? If we accept the market does arrive at the right market rate, why screw up our finances by giving people what appears to be an undeserved increase. It’s not going to come from a tax cut as minimum wage people generally pay almost no tax. Furthermore government rises. There’s really no free lunch.

  4. Many people on minimum wages do pay tax. The tax-free threshold is $14,000 and many people earn more than that.

    Nobody is talking about a free lunch or utopia. But if you have to choose between using the wage system or the tax/transfer system to increase low incomes, then which do you choose?

  5. “The downside of using the tax/transfer system is that it will impact on the government budget (through lower taxes or higher welfare spending).”

    Economics is really not my strong suit but I don’t understand how this would be a problem. Surely any loss of income by the Government would be more than offset by having the general population disposing of more income thereby paying more GST

  6. This is essentially a new way to sell an old story. I think it has potential, but it requires an agenda change. The CIS could put it out there but unless someone in the government of opposition picks it up, it won’t go anywhere.

  7. Cameron — you are right that some of the tax cuts or transfer payments will come back to the government through the GST and other taxes (customs, excises etc). However, (1) GST revenue goes to the states, not federal govt; and (2) the govt won’t get all the money back as some will stay in the private economy.

    In the medium-long term there should be a growth dividend, leading to more revenue… which might help the policy to pay for itself. But that won’t work in the short-run, so any change in the tax/transfer system will impact on the budget position of that year.

    Having said that — I think it would be quite affordable, and a very good excuse to cut income taxes for people on low incomes.

  8. Trading wage rises for tax cuts is essentially what the accord was all about. Not quite the same as what is proposed here but similar logic.

  9. “The case for labour market reform is obvious enough. If people have a marginal productivity of labour below the minimum wage, then they will not get a job. The only way to fix that is to (1) increase their productivity; or (2) lower the minimum wage… (including all the costs of employing somebody, not just the wage).”

    That’s wrong in general, though it’s an accurate observation about present circumstances and institutional arrangements. What counts isn’t whether the marginal productivity of labour is below the minimum wage (with on costs) but whether it’s below the marginal cost of labour – which only works out roughly the same in special cases, including as it happens today’s distorted Australian economy, so it’s accurate to that extent.

    For a fuller discussion see my second submission to the Henry Tax Review and its references, where I describe “a Pigovian virtual wage subsidy integrated with the tax system, to improve both employment and GDP without the problems usual with wage subsidies, Negative Income Tax, etc.” (although both paths are long run equivalent). The rest of this post is going down that second, more awkward path, complete with budget implications, costly lags and so on. The approach I outline has fewer political ramifications and shorter lags, with no added costs during those.

  10. Trading wage rises for tax cuts is essentially what the accord was all about.

    Really? Here’s me thinking it was a trade between real and nominal wage rates.

  11. PML — the only way that the minimum wage doesn’t represent the marginal cost of labour is if the minimum wage isn’t binding. So my statement holds in all cases where the minimum wage is effective… which is the only minimum wage that matters.

  12. No, JH, you haven’t considered the wider and more general range of things. Try working it through with some example numbers, with things as they now are and with the approach I suggested in my submission.

    Suppose JQ (let’s pick on him) can produce a value of $15,000 p.a. but the minimum wage with on costs is $20,000 p.a.; as things are now he isn’t worth hiring, as any employer would be $5,000 p.a. out of pocket by doing so. But suppose hiring JQ generated a tax break of $10,000 p.a. for an employer; then he would be worth hiring, as the employer would be $5,000 p.a. ahead by doing so. The thing is, with the tax break the marginal cost of labour isn’t the minimum wage, it’s the minimum wage minus the tax break. And, of course, there are more possibilities that could affect that marginal cost than just tax breaks – but we don’t need to look any further.

  13. Bah, that’s just semantics.

    If you include the $10,000 tax break in the effective minimum wage (and I clearly said I was talking about the costs of employing somebody, not just their wage) then the minimum cost to the employer in your example is $10,000… which is lower than the marginal productivity in your example and so the “minimum wage” (broadly defined) is not binding.

    As you say, tax breaks aren’t the only thing at play. There are many factors the impact the real cost of employing somebody, including compulsory holiday/sick leave, unfair dismissal legislation, workplace health & safety regulations, maternity leave provisions, special rights for unions etc etc.

  14. Broadening the term “minimum wage” to incorporate other factors seems like a more tortured use of semantics.

  15. JH, if, when you write “effective minimum wage” you are not using the ordinary meaning of “minimum wage” as a component, then your “Every year the government wages watchdog considers the case for an increase in the minimum wage” is a switch of meaning and you are not using terms consistently.

  16. I think it is important to recognise both the effects of minimum wages and non wage costs. I don’t think that is an issue of semantics.

  17. I think it was blindingly obvious that I was talking about the cost of labour to employers.

    The reason I think this is blindingly obvious is that I wrote “including all the costs of employing somebody, not just the wage”.

    If you feel the need to substitute any set of words you like for my short-hand use of “minimum wage” then please do so. It will make absolutely no change to the substantive point (and hence is simply semantics) but if it makes you happy, then please mentally substitute any words you like. Translate the entire post to German if it will help. After all — I want you to be happy.

    But perhaps comments about the substantive points (and not your preferred substitute semantics) would be more helpful.

  18. “Increasing productivity is a slow long-run project….”

    Its not a slow process. And as well as that these aren’t the only two measures you can take. Why not increase productivity? What you’ve done is made an excuse to ditch this option and so you’ve just dismissed it.

    Mass-sackings in the public sector is the first step to full employment. Since it enables more business-business spending which leads to higher productivity, a greater demand for labour, and as well you can afford a defacto wiping out of the company tax via double expensing of the first few hundred dollars every week for each separate employee.

  19. Of course we should strive to increase productivity. I never said otherwise. But (1) it takes time; and (2) it’s only effective if we also stop increases in the effective minimum wage.

  20. It doesn’t take a lot of time. But the idea of focusing on after tax income for the poor is a good idea. What we are really talking about here is mass-sackings in the public sector to increase the tax free threshold.

    There is no getting around the need for mass-sackings and closing down bureaucracies by the bakers dozen. There is no maths-magic that you can perform to evade this need.

  21. The resources we need to increase after-tax-income while stalling pre-tax income…. The resources we need to increase the productivity of labour.. Or for double-expensing of some labour cost. Or for getting rid of the company-tax direct…

    The resources we need to attack nearly all aspects of the unemployment problem has to come from mass-sackings in the public sector. Mass-sackings is at the core of any sensible solution.

  22. Graeme – you could completely get rid of personal income tax by simply freezing real per capita government spending for 10-15 years. Whether that is more or less politically viable than mass sackings I don’t know. However it is an alternative.

  23. Making the tax free threshold on payroll tax dependent on number of employees, rather than size of payroll, would be another way to tilt the field in favour of full employment. Although obviously killing off payroll tax entirely still ought to be a priority.

  24. GP, well, yes, that proposal does increase the (virtual) tax rate, just not the intermediate figures used in calculations. It does so by having a graph of amount due against payroll that intersects the amount due axis at a positive amount due, which means there is an effectively infinite slope (and so, rate) there, since firms reaching that zone would liquidate rather than pay it. That highlights an underlying problem: how to assess the tax for new firms coming in? And, as a corollary, does the approach for those firms give old firms an incentive to do a “phoenix” reorganisation, closing down and restarting as new?

    The approach I outlined only presents a psychological barrier of a “higher” rate even if you kept the carrying tax revenue neutral as at the implementation date, since it would only be an intermediate figure anyway (after that date it would only be budget neutral). But if you follow the link to my detailed proposal, you will find I canvass not raising the rate on the carrying tax but on other taxes – which might be done conveniently at state level as the states get GST revenue. That might be done through land taxes etc., or (as I prefer) indirectly by means testing state benefits and services. The latter would cause no hardship for people with secure employment and no change for those who hadn’t yet got that, and would tend to wind back state delivery of those benefits and services in favour of private arrangements, which for me is a strong point in favour.

Comments are closed.