IANAE but in June last year there was a spate of discussion over productivity and real wages having a one to one correlation. Nicholas Gruen wrote in a comment to his article Economic Nonsense:
But in the long run, you expect to see income trending towards productivity. People or nations with high productivity put themselves in a pretty good position to make more money.
and Andrew Leigh wrote:
Absolute levels of productivity, not levels of productivity relative to our trading partners, are what matter.
So we can take from those quotes that productivity is an important indicator of current and future wealth. Peter Martin had an article the other day, titled: What if Paul Keating was right?
Productivity is the engine that drives our future prosperity. The Treasurers budget papers reveal that it has stopped growing. But you wont hear that admission from the mouth of the Treasurer himself.
Peter produces charts from the ABS to back his argument up. This is the chart that Peter mentions from the Budget Papers:
IANAE, so one for the economists that haunt this modern digital republic of letters; what is your opinion?
- pdf[↩]
Though the government’s policy for some years has been to increase output through both higher workforce participation and higher productivity. At least in the short term, the former will work against the latter – not because existing workers become less productive (or even because their productivity stops growing) but because new workers or people returning to the labour force are less productive than people who have been working continuously, and this shows in the aggregate statistics.
Andrew, That reading of new people coming into the workforce, from what, about 1% of the workforce (assuming a drop of 5.3% to 4.3) means that productivity is super-sensitive then. If a small number of low skill workers coming into the workforce under workchoices (assuming all AWAs are for new worker and not existing ones being renegotiated) can flattern productivity growth then that means productivity as an economic reading is pretty tenuous. Wouldn’t the productivity from the highly skilled service, financial and manufacturing sectors swamp retail and other low skilled segments?
the labour turnover in the economy is much higher than merely the changes in the unemployment numbers. Women returning to work into part-time jobs. Retirees replaced by the new cohort of school and university leavers; in and out-migration; etc. As a casual rule of thumb, I’d guess as much as 5% of the workforce changes every year which leaves a lot of room for changes in composition and average productivity.
If you look at the skill component of new jobs, you do see that in the last 3 years or so, we’ve finally had a bit of growth in the full-time jobs for low-skilled, so yes, compositional changes would have to be the front runner in the explanation for the productivity slowdown. And dont forget: services make up about 70% of the economy and we have no real way of measuring productivity in the service sector so I believe we just count wages.
Paul – Turnover is much higher than that. In the last labour mobility survey more than 10% of workers changed employers, and that’s not counting the many more who entered and exited the labour force and who changed jobs but kept the same employer.
Cam – I’m not saying there definitely isn’t a problem, but I think a lot of caution is required. Another issue is that GDP figures are often revised so productivity estimates based on recent GDP statistics should be treated with added caution.
We should keep in the mind that business pursues profit, not productivity. They will take on workers that make them money regardless of what that appears to do to their average output per hour per worker.
Paul, If turnover is consistent from year to year it shouldn’t impact productivity though right. Because the churn is the same from year to year.
Andrew, That graph from the budget papers looks spurty too, like productivity gains come in rushes. Though the graph that Peter Martin had from the ABS looked less jumpy.
Cam,
These issues were run with the political parties in opposite drag in the 1980s. The Accord was very good at generating employment and did so by holding wages down (reducing pressure on businesses and workers to innovate to be able to afford wage rises.)
Keating pointed all this out, but when the Accord atrophied and wages and productivity went up he changed all the numbers he talked about. Pretty much any set of numbers is ‘beautiful’ when you pick the ones you like. Do you recall what happened at the very depth of the recession? Keating started talking about the beautiful inflation numbers. How inflation was the central building block of a decent modern economy – etc etc etc. There was no evidence for that either. Of course low inflation is preferable to hyperinflation and if you get inflation where we have you should bust a gut to keep it there. But there’s precious little evidence that inflation where the recession got it is much better than where it was in the eight years before – around 5-7% from memory.
But there is a difference. The Howard Government has been a strange schizophrenic mix of fecklessness and crazy brave on micro-economic reform. Feckless on most of it and crazy brave on beating up on their class eneemies (WorkChoices) and on Howard’s strange passion for a GST. The Hawke Govt was thoughtfully and courageously reforming in scores of places in the economy.
Cam, I can’t figure out why Peter is placing so much weight on what happens to productivity in the current financial year. If you average out the forecast in the graph up to 2046, you get about 1.7% per year. That’s lower than the 1990s, but about the same as for the previous four decades taken together. That doesn’t mean the arguemnt about Workplaces isn’t correct, but it doesn’t stand or fall on the Treasury’s forecast for 2007-8, for heaven’s sake.
The really important point in that budget paper is that while labour utilisation has an upper limit, productivity does not. That’s why Countries like Norway can have higher GDP per capita than the US and still work a lot less hours on average.
Cam,
“Paul, If turnover is consistent from year to year it shouldnt impact productivity though right”
depends. I wasnt talking about people changing jobs, but about people entering or leaving the workforce (which is a lot lower, hence my lower guestimate than Andrew’s number which includes job changers. I’m guessing job changers wont add that much to productivity). I can think of good reasons for each component in this workforce turnover to be associated with lower skills. The additional part-time women are likely to be the less productive ones compared to those already in the labour force. Having made it harder to get into DSP, perchance some of the less productive older workers stay on longer in their job rather than go into DSP. The increase in temporary working visas is likely to be associated with the less productive migrants; etc. I will say though that a productivity slowdown is often something you expect to see at the end of a boom, not in the middle of one. Its a sign.
Peter Lindert argues with reference to Europe that lower workforce participation might have little impact on productivity as some potential workers might be extremely unproductive even if you pushed them into the workforce. Also if worker resistance is a block to managers adopting changes that increase productivity, then workplace reform should led to a permanent increase in rates of productivity growth as managers will be able to make changes more quickly. The evidence seems to be against this however. Curious that the press gallery loves AWAs, presumably on the grounds that they increase productivity, but then judge every other item of government policy by the sole criteria of short-term imagined political effectiveness.
Geoff, Lindert’s point is exactly the one that Andrew and Paul made. It’s backed up by a recent OECD cross-national study that found that high participation by marginal groups was correlated with low average productivity.
Andrew is echoing Treasury’s view that this is a short-term effect that should disappear as the new workers acquire human capital. Personally I reckon there’s neither theoretic nor empiric support for that, though admittedly the argument is a tad arcane (it involves relative rates of human capital accumulation between different groups). Anyway this is not in itself a killer argument against getting more marginal workers into jobs – there are other reasons than aggregate GDP why you might want to do that.
cam, economic theory (backed up by experience) suggests very strongly that in the long run wages should follow productivity. But that’s not necessarily true for the short run. There are two real problems with drawing any sensible policy conclusions from short-term productivity movements:
– it’s a ratio of two rather badly measured quantities, so it’s “noisy”. Its measured value jumps around a lot.
– it’s affected by many different things, some of them not well understood and many with quite non-linear effects. That means that its technically difficult to identify what the effects of any individual policy on short-term movements are or will be.