Sunday, June 11, 2006

Record low unemployment

Even with the qualification that the basis for measuring unemployment has changed over the years, the fall in the Australian unemployment rate to a 30-year low is a considerable achievement – 4.9% measured unemployed, while still high, is indeed ‘a beautiful number’. An extra 56,000 people won work in May and the overall workforce participation rate increased.

Several lessons can be learnt. (i) To reduce unemployment to this level we have not needed lavish government-funded job creation schemes of the type funded by Labor in the early 1990s. (ii) The decline in trade union power has not hurt workers – they have never been better-off in terms iof wages or employment levels. (iii) The new WorkChoice laws can be expected to further eliminate the constraints that inhibit the ability of employers and employees to arrange mutually beneficial wage-output deals, thereby further reducing unemployment.

For economists the myth of a natural rate of unemployment that would need to be maintained in the long-run to prevent excess demand and consequent inflation can be finally discarded. It is good to see that the RBA’s Governor Macfarlane rejects it - it is not based on solid economic theory or post-war economic history and it promotes unwarranted policy pessimism in addressing unemployment issues. For most of the 1950s and 1960s Australian unemployment was below 2% - it was a scandal for the Whitlam Government in 1974 to have it rise to 3%!

We still have unemployment doomsayers. Indeed it is amazing how the fall in unemployment is pounced on by the press as bad news - interest rates will have to rise , the economy risks overheating and so on. Others see the decline in unemployment as due to the commodities export boom – this is not supported by the facts - the biggest decline was in the relatively resource-poor state of NSW.

The significant residual problem is the 712,000 people on the Disability Support Pension and those on single parenting payments. While 1.8 million jobs have been created by the Coalition since 1996, non-dole social welfare receipients have increased by about 60,000. Again some are treating the high level of DSP as a type of giant fudge of the unemployment outcome – it isn’t – but these people need to be encouraged back into the workforce with both ‘sticks’ and ‘carrots’.

Tax reforms that reduce the effective marginal tax rates of up to 60 per cent that apply to people when they switch from receiving social security to joining the workforce. These high marginal tax rates provide disincentives to work and it should be a simple matter to redesign benefit schedules so that extra work is possible while higher levels of benefit are retained. This redesign comes comes to a negative income tax – a low enough very basic level of benefits is provided to ‘non-shirkers’ (nor surfies who elect a non-working lifestyle) with extra earned income always then always being subject to non-punitive loss of benefits and taxes.

Job subsidy schemes are likely to outperform work-for-the-dole schemes which should mainly be directed towards the hard core residual ‘work shirkers’. But the major way of reducing unemployment will be via retraining schemes which help our workforce to adapt to a changing economic structure and to technological unemployment.

The challenge is to reduce unemployment much further.

12 comments:

Anonymous said...

Harry, I believe that 5% unemployed is considered "full employment" in the US, basically, anyone who wants a job can have one. Would that figure also be considered full employment in Australia?

Anonymous said...

I'm glad you report the disability numbers (and underemployment would be good too), although I'm not sure why you think they *arn't* a big fudge. These numbers have increased huge amounts in the last decade or so. Its only logical to conclude

(1) Some amount is due to aging of the population and (2) some amount is due to what would otherwise be long term unemployed people moving onto pensions as it pays slightly more and the government doesn't bother you as much. Even if (1) accounts for a fair amount of people, I can't see why a fair number of them couldln't work if there were jobs to take them, hence they should be considered unemployed.

I can also add a third thing to note which never seemed to get a mention until recently (and now it is mentioned as a "good" thing) -- even if people have enough money of their own not to work (or whether due to other subsidies by the goverment), but could, this is still a loss to the country as a whole. In this case, most people simply ignore the governments indirect contribution to most people that someone has to pay for(healthcare, police, education subsidies etc.).

hc said...

civitas, 5% is not treated as full-employment in Australia - still many people desparate for work. Many long-term unemployed who need retraining.

conrad, 1.8 million jobs created and only 60,000 added net to DSP numbers so no fudge. But yes a problem. Aging of population a bit but I think (2) more important.

I wasn't quite clear on your point about public goods. Are you saying that the provision of public goods to those not workking but able to support themselves is an extra burden? Some will still be paying taxes and will have paid taxes in the past. There is some cost.

Anonymous said...

Harry,
I fail to see how falling unemployment discards the natural rate concept. This is a purely economic concept that basically says that using DEMAND side policies you cannot permanently 'peg' the level of unemployment at some low level (see e.g. Friedman 1968). The natural level can obviously vary over time (with developments in the labour market). Your mention of monetary policy contradicts your point since this natural rate idea is the building stone of inflation targeting currently practised all over the world.

Anonymous said...

An addition to my previous post: instead of saying 'cannot permanently peg the level of unemployment at some low level' I meant to say 'at some level that is LOWER than the level given by the real factors/fundamentals'

hc said...

Jan, It isn't the standard used by the RBA - read the comments of Macfarlane where he describes the idea as not used. What is the value of thinking about a particular unemployment rate which can only be exceeded if inflation occurs if it doesn't occur? Every time unemployment falls below the natural rate level without inflation occurring we say the natural rate has dropped.

Friedman 1969 talked about the natural rate as that rate ground out by the solution to a Walrasian general equilibrium - but how do general equilibrioum models 'grind out' equilibrium unemployment rates?

Anonymous said...

What is full employment in Australia then, Harry. It cannot be 0% as there will always be people between jobs. So what is considered full employment?

Anonymous said...

This may be from 1999, I cannot find anything more recent, but it pegs NAIRU at 6.8% for Australia.

http://www.oecd.org/dataoecd/44/50/2086120.pdf

hc said...

civitas, Bodman in 1999 here estimates frictional unemployment at 5.3% well above current actual levels. I don't believe these figures and I don't believe econometrics has the ability to come up with half-decent estimates off such things.

There is quite a bit of structural unemployment in the Aussie economy due to the commodities boom and the shift of peak economic growth to the western and north of the country.

A specific figure - I don't know but less than 4.9%.

Anonymous said...

Harry,
the concept is useful to think about money (non)neutrality, the short run vs the long run, the self-correcting mechanism etc. The fact that we do not have a good measure of it (especially ax ante) does not reduces the usefulness of the concept. Central banks use the concept of the output gap, i.e. the difference between actual and natural rate of output. Estimating this is (and hence the natural rate of output/unemployement) is crucial in determining the optimal policy response).... The fact that policymakers had an incorrect estimate of the natural rate is one of the plausible explanations for the 1970s inflation, see work by Orphanides e.g. http://www.federalreserve.gov/pubs/feds/1999/199938/199938abs.html

Anonymous said...

I'm just saying that people that don't work and don't collect direct benenfits still get supported by the government indirectly (via things like infrastructure etc.) -- but this group almost never gets a mention, and the attitude of people towards them is completely different.

For example, lets say I work overseas for 10 years, save lots of money, move back to Australia and retire. My cost to the Australian public has been huge (20 years of education, health, police...), and it is never going to be recouped (in fact the cost will continue), despite the fact I never took any direct benefits. In fact, my cost to the Australian public is probably a fair bit more than if I had worked 30 years and was on the dole 20 (You can just calculate it all through).

Thats an extreme example, but I presume a fair chunk of the population fits into at least a milder version of this category(early retirees and non-working parents being probably the biggest groups), and I also presume that this category is going to more skilled than the long term unemployed and the like that do get a fair bit of publicity. It would be good if some attention was given to them as the resources needed to get them working might provide a bigger bang for the dollar than some of the other groups (presumably excluding childcare).

Anonymous said...

Have you heard of the GST?

Actually I am in favour of cutting income taxes by about 20 billion and increasing the GST to 20% with a low-income-household rebate of 20% of $25,000 declining by 1% per thousand dollars from $35,000 - eg at $40,000, a rebate of 15% (20-5) of $30,000 ($25,000 + 5,000).

Basically low-income households would not pay any GST, and the rich would pay more tax, with lower income taxes making Australia a more attractive place to work (our prices are really low compared to Europe anyway).

But the mere presence of the GST probably answers all your worries even at 10%. Imagine if your putative person buys one Mercedes at $85,000 - that's actually about $13,000 in taxes, I believe, or the equivalent to earning about $60,000.