Has unemployment peaked?

ANZ has an interesting note out today asking the suddenly all too real question:


•Employment rose by 44k in March, well above market expectations of +6.5k. The unemployment rate was steady at 5.2% (mkt: 5.3%), and has been broadly stable since July 2011.

•In a trend sense, the multi-speed nature of Australia’s economy is alive and well in the data. WA’s labour market has tightened considerably relative to the rest of Australia. Nonetheless, labour market conditions in NSW have also improved in recent months.

•The better-than-expected data saw the AUD jump around ½ a cent against the USD shortly after the release and cash rate futures sold off modestly.

•RBA watch: we still expect the RBA to ease by 25bps on 1 May, pending a benign Q1 CPI inflation outcome on 24 April. However, today’s labour force data, coupled with recent modest improvement in other domestic activity indicators, make selling a May rate cut more difficult for the RBA (one wonders if they now wish they had cut in April). Importantly, current evidence casts significant doubt on calls for a follow-up rate cut soon after May – recent job ads data may be telling us that the unemployment rate has already peaked (Figure 1).

The broader definition Roy Morgan unemployment series continues to show significantly less work available to those who want it and the NAB Business Survey also has ongoing weak employment intentions so I’m not fully convinced we’ve seen a peak. There is also the simple fact that the declining participation rate has disguised a decent rise towards 5.8% or so already.

But we are seeing all sorts of peculiar data outcomes for the simple reason that the changes underway in the Australian economy are unprecedented: a slow melt in credit issuance, as well as a terms of trade driven adjustment in the tradeables sectors. These other traditionally reliable data series could be getting it wrong. And as you can see, so far, the ANZ job series has tracked the unemployment rate very nicely.

I can’t discount a peak in the unemployment rate but I can’t yet believe it either. Check out the state by state break down of the job ads:

That is most of the economy going sideways. To me it is very unlikely that these adjustments can transpire smoothly. Labour is not especially fungible. Just because someone loses a job in Victoria, doesn’t mean they’ll be retrainable for one in the Pilbara, or willing to move there. Having said that, Australians are a redoubtable bunch.

At the very least, the ANZ job series must take on greater importance for economic watchers. I have been bearish on employment since mid last year but am now definitely bringing in my unemployment forecasts for the immediate period ahead.

Taking a bigger picture view, it is perhaps not surprising that unemployment has been capped with the government not yet returning to surplus and private credit creation, whilst decelerating, not negative. The second half remains an open question, if the government pursues its surplus and the net effect of a trade deficit and weak private credit does not take up the slack.

Labour Force Mar 2012

David Llewellyn-Smith
Latest posts by David Llewellyn-Smith (see all)


  1. The ABS trend unemployment rate actually looks to have been very slowly improving over the past six months – but trend unemployment has now stopped improving and has levelled out.

    Perhaps the improvement has bottomed? I don’t expect Swan’s slash and burn to do much good for employment.

  2. Given these figures it’s hard for economists to justify rate cuts or complain about swannie’s surplus. In fact he should go for a greater surplus as clearly any unemployment is soaked up!

  3. Nice to see a reappraisal upon receipt of changed data. Certainly agree that recent economic data has resulted in ‘peculiar’ results and fortunately, little to date, has been as bad as predicted.

    If unemployment has peaked, then we truly are ‘different’, indeed, the Lucky Country!

  4. Not sure if this is the right thread, but here is more “peculiar” data, given the strength of the employment figures…

    “JUST 12 commercial property sales resulted in more than $85 million worth of property transactions last month.”

    “The biggest losers were the commercial and industrial sectors. There were just four commercial sales last month, totalling $5 million, while in March last year there were 13 sales amounting to almost $17 million”.

    The number of industrial sales halved from four to two over the period, with the value of industrial sales falling from $7.8 million to just $1.1 million. In February, there were seven industrial sales adding up to more than $10 million.”

    “Leasing activity was also well down in March, with just 28 lease deals completed.

    Retail lease deals were almost half that of the previous month while industrial leases fell from 16 deals in February, over 17,434sq m, to just seven deals across 13,622sq m.

    More than 50 lease deals were completed in March last year, with 21 industrial deals and 17 retail deals.”


    One or two things seem a bit inconsistent with the article’s downbeat tone (total value of sales), just thought I’d throw it in there anyway.

    • I read the article and don’t know what geographical area that was referring to, but the message is rubbish.
      For sales > $5 million, there were 3 retail shopping centres totalling $394 million, 5 office towers totalling $314 million and 8 industrial facilities totalling $245 million.
      The commercial sector in Australia is ticking along quite nicely with very good buyer depth and many not having trouble accessing funding.

      • You’re saying that the author of the article is attempting to mislead us and that growth is actually strong? Fair enough, I won’t argue with you since I’m not 100% certain how to interpret the figures given and have no frame of reference to make comparisons. If you could share your knowledge with me it would be appreciated.

        Regarding the employment figures, the ever-cautious Bill Mitchell reminds us not to get too excited over volitile monthly figures. http://bilbo.economicoutlook.net/blog/?p=18974#more-18974

      • Thanks georgie. Residential property certainly does dwarf commercial.

        But is pessimism spilling over from the poorly-performing residential sector really the only reason or is comercial property actually a bit on the lacklustre side of late? I guess ABS lending finance data might give a clue but I don’t think the March figures are available yet.

  5. Bloody hell. What is it about these employment surveys that has the media (including MB it seems) running around like headless chooks every time the ABS releases one? Its a survey people, with a huge margin of error, and huge revisions (which no-one ever reports)

    Wasn’t it just yesterday a rate cut in May was a dead certainty? Has anything really changed? The employment trend still looks pretty bloody weak to me. Next month, employment could come in at -10,000 and all of this month’s “gain” gets quietly revised away.

    As Colebatch says:

    The problem is that for the past four months, the seasonally adjusted figures have followed a zig-zag pattern: up one month, down the next. They fell sharply in December, bounced back in January, fell again in February, and now rose again in March. Anyone who focuses on the jump in March will be misled.

    That is why the Bureau of Statistics tells us to focus on its trend figures, which smooth out the zigs and zags. Even they vary with the latest readout, but over the last six months they estimate that just over 25,000 new jobs have been created, all full-time, with unemployment flat at 5.2 per cent.

    But those figures vary greatly from state to state. All of the net job growth and more was in WA, where 31,000 jobs have been added since September, almost all of them full-time. NSW has added just 6000 jobs, Queensland 4000, while Victoria has lost 23,500 jobs in the last six months, with unemployment rising to 5.5 per cent, from 4.8 per cent a year ago.

    Settle down people.

    • Yes when I corrected my error in the COTD post, I was reminded again that the UE rate can be off +/- 0.2% because of the 95% confidence level of the survey.

      So UE could be 5% or 5.4% – we just DONT KNOW!

      But it made Adam Carr’s day.

      • But it made Adam Carr’s day.

        I don’t read Mad Adam much these days but I stumbled on one of his pieces the other day. He is completely detached from reality.

      • Adam Carr went nuts today on twitter today. expect a big dose of “i told you i would get a volatile stat right one day” on bus spec tomorrow.

    • I agree. But where would commentators, pundits and blogs like this be without all the noise that goes along with the release of any monthly statistical data.

      Out of curiosity Lorax, if today’s data have read unemployment at 5.3 or 5.4 of even 5.5 would you still have been tempted to comment as above. 🙂


      • Out of curiosity Lorax, if today’s data have read unemployment at 5.3 or 5.4 of even 5.5 would you still have been tempted to comment as above.

        I think I did last month. Dunno if it was here or elsewhere, but employment surveys are volatile with a capital V, and to be treated with huge buckets of salt.

        • Lorax, if the movement is statistically insignificant, is it any more informative to look at the trend? Or is that insignificant too?

        • The problem is that the RBA uses the results of volatile and suspect surveys to drive the direction of interest rates.

  6. Has unemployment peaked? No way. although stronger than expected it just continues the trend. one good month where everyone gets excited then a bad month when reality sets in. the unemployment rate never goes down though, only up or unchanged. these are still crap numbers on a medium term veiw.

  7. I heard on the radio this morning during my commute, a robust discussion about the “legality” of “unpaid internships” in this country. The luvvies hate this kind of thing, must be stamped out ! But why ever would it be occurring if we had full employment?