The Dim-Post

October 9, 2012

Still in stage one of Kubler Ross model

Filed under: economics — danylmc @ 1:33 pm

DPF has a chart up of HLFS employment data mocking the claims that there’s crisis in manufacturing, but absentmindedly forgot to zero the y axis or include data-points prior to 2009, or adjust for population growth which happens to generate a chart that supports his argument. Here’s one from the same dataset, without those totally unintentional errors.

It’d be interesting to throw the exchange rate there on a secondary axis. My memory is that it was pretty high for the last years of the Labour government, but we didn’t see a decline in manufacturing until the GFC kicked in.

Also, the HLFS doesn’t include the September quarter yet, and the couple hundred job-losses a week for the last several months is what’s freaked everyone (except Key and Joyce) out.


  1. When John Key called a ‘jobs summit’ to address the jobs crisis, unemployment was 5.1%. Nearly four years later, he says there is no need for a summit on manufacturing jobs because there’s no jobs crisis. And unemployment is at 6.8%

    Comment by Deano — October 9, 2012 @ 1:38 pm

  2. Those damned lying statistics! They do get in the way.
    A recent Cabinet discussion has declared war on Statistics. They will be banned for everyone – except Joyce and English.

    Comment by xianmac — October 9, 2012 @ 1:58 pm

  3. It’s a pity that politicians have ignored the facts because a lot of people’s livelihoods are riding on this. Once again, I feel the need to point out that unemployment needn’t have risen over the last four years at all, if the money borrowed had made its way into providing fulltime permanent jobs for people instead of tax cuts and an absurd roading scheme.

    The government is in its last throes and, as such, is ignoring all opportunities to show the nation that they’re prepared to do something. Can’t see them winning 2014. This is where Shearer comes in.

    Comment by Dan — October 9, 2012 @ 2:04 pm

  4. Danyl,

    You’ve chosen manufacturing jobs as a proportion of the population. Why not as a proportion of the working population? Why should it matter if there are babies born each year?

    Comment by Ross — October 9, 2012 @ 2:07 pm

  5. Just tells you how useless summits are deano. They are PR stunts pure and simple. How’s that knowledge wave we’re riding?

    Exchange rates – from 60s to 80s to 50s against USD 2006-2008, manufacturing jobs flat during that time until 30k jobs lost in 2008.

    Comment by insider — October 9, 2012 @ 2:07 pm

  6. @2 “A recent Cabinet discussion has declared war on Statistics”

    Like Canada then, where the federal government (right of centre Conservatives) are cutting back on anything to do with information/knowledge, including Statistics Canada and the long-form census. Can’t have information tainting political decisions! Can’t give our opponents ammunition with which to criticise us!

    Comment by MeToo — October 9, 2012 @ 2:36 pm

  7. I guess you “absentmindedly” forgot to data-points prior to 2006. Whether you look at share of employment of share of output, manufacturing has been in decline at least since the late 1980s, which is as far back as the official statistics go. It’s happened in every rich country.

    Comment by Miguel Sanchez — October 9, 2012 @ 2:37 pm

  8. insider – would be better to use TWI to take into account the relativity of exposure to USD changes…

    danyl – why not share your chart so someone can do your alternative axis…

    Comment by The Baron — October 9, 2012 @ 2:42 pm

  9. No, it’s just that the population estimates prior to 2006 are in a different dataset.

    Comment by danylmc — October 9, 2012 @ 2:42 pm

  10. Working-age population estimates for the civilian, non-institutionalised population (on which the HLFS is based) are available back to 1986, by age group and sex:
    Group: Household Labour Force Survey – HLF
    Table: Labour Force Status by Sex by Age Group (Qrtly-Mar/Jun/Sep/Dec)

    From there you can derive a working-age population aged 15-64 years by subtracting the 65+ age group, or use the total, which is the 15+ population.

    The number of employed people by industry can be found here (note that the data in this table does not specify whether the employment is full-time or part-time). The Infoshare series goes back to 2003.

    Group: Household Labour Force Survey – HLF
    Table: Persons Employed by Sex by Industry, ANZSIC06 (Qrtly-Mar/Jun/Sep/Dec).

    Comment by Laura — October 9, 2012 @ 3:57 pm

  11. I’m not sure which of “debt monetization for the ChCh rebuild will save Western manufacturing” or “there is nothing wrong with Western manufacturing” is the more laughable idea?

    Comment by garethw — October 9, 2012 @ 4:57 pm

  12. Manufacturing has been in decline for years as a longer historic series would show. It was big in the 1950s employing something like 30% of the workforce but has now dropped to just 12%.

    This comment gives a clue as to why lately; “In the 2000s, most manufactured goods are imported. This has reduced employment in manufacturing in New Zealand, which by 2006 was just 12% of the workforce.”


    Comment by JC — October 9, 2012 @ 6:09 pm

  13. Manufacturing is going the way of agriculture. Productivity increases outstrip increases in demand. Hence a drop in employment. Its been happening since about the 50’s.

    Comment by swan — October 9, 2012 @ 6:53 pm

  14. “until the GFC kicked in” and as everyone knows that is John Keys’ fault. Bastard.

    Comment by gn35 — October 9, 2012 @ 8:05 pm

  15. …and as everyone knows that is John Keys’ fault.

    Feel free to point out where someone’s suggesting it’s his fault. More to the point is that he, English, Joyce et al are currently drawing fat salaries as the guys meant to take the effects of the GFC on things like manufacturing seriously and do something about it. Drawing the salary while pretending there isn’t a problem comes under the heading of “Performance Issues.”

    Comment by Psycho Milt — October 9, 2012 @ 8:12 pm

  16. Drawing the salary while pretending there isn’t a problem comes under the heading of “Performance Issues.”

    We’re a bit stuck as the alternatives are offering more inflation and giving the elderly less support. Aren’t we spoilt for choice.

    Comment by NeilM — October 9, 2012 @ 8:28 pm

  17. I am shocked that DPF is so lax in his analysis.

    I cannot imagine why someone of erudition could be so slipshod.

    Is it possible he is promoting the views of a certain political party?

    To hell with the facts!

    Spin matters.

    Spin counts!

    Comment by peterlepaysan — October 9, 2012 @ 9:08 pm

  18. > We’re a bit stuck as the alternatives are offering more inflation and giving the elderly less support.

    In the current situation, I’d have thought the government would be happy with this scenario (though both assumptions are tenuous at best). I mean, those being made redundant are hardly likely to be rushing down to the pub to celebrate the latest inflation figures.

    Comment by Ross — October 9, 2012 @ 9:21 pm

  19. You’ve chosen manufacturing jobs as a proportion of the population. Why not as a proportion of the working population? Why should it matter if there are babies born each year?

    Because all that would really show is that the number of people in factories is changing *relative* to all the other jobs. So it might not be changing much, everyone else might just be going in/out of work, or it might scale exactly because people are going in/out at the same rate across all industries and therefore look like there was no change.

    To track the exact change in numbers in order to see what that industry alone is doing, ignoring everyone else, would show the true changes but not how much they meant (e.g. are they going down because the population just halved because of a sudden plague, or going up because the entire economy/job rates are going up?)

    Tracking it against overall population means that we can see proportions changing without the total data being skewed by a workforce shifting out of sight into unemployment, or never actually finding a job, as well as the simple shift of people into different jobs away from manufacturing. Obviously we can’t tell WHICH they’re doing, and another line showing total population of working age would help, but for a simple graph, it does what it says, as long as you pay attention to what conclusions it actually lets you draw.

    Comment by Flynn — October 9, 2012 @ 9:35 pm

  20. NZ is the unlucky country but we are right next to the Lucky Country.

    Australia is our largest market, our exchange rate has remained stable for decades, and yet no opposition party has come up with any plan to try and make more of that.

    Comment by NeilM — October 9, 2012 @ 9:49 pm

  21. Flynn

    I’m not sure what you’re trying to say. Let’s say the total population is 100, and there are 10 workers employed in manufacturing. A baby is born so the population increases to 101. There are still 10 workers in manufacturing and the working age population remains unchanged. But the figures for manufacturing vis a vis the population goes from 10% to 9.9%. That is quite misleading…

    Comment by Ross — October 10, 2012 @ 10:32 am

  22. But if you restrict it to the workforce only, it becomes even more misleading, because the workforce would be say, 75% of the total population. So one more person entering or leaving would create a *larger* than 1% change. There’s no point to a graph that has one person leave from manufacturing and 4 people leave from all the other jobs, out of five categories. It would show no change despite going down.

    Aside from capturing the actual working population *and* the non-working population, which may be where all those manufacturing jobs have come from or gone to, it also gives you an idea of how big a manufacturing base the country normally supports.

    Comment by Flynn — October 10, 2012 @ 4:15 pm

  23. We all know that unemployment has risen over the last four years, so there is no legitimate reason for the figures not to be restricted to the amount of people in the workforce. In fact, it makes the most sense.

    Comment by Dan — October 10, 2012 @ 5:58 pm

  24. > because the workforce would be say, 75% of the total population

    Well, the workforce is always changing. But in the example I gave, the workforce didn’t change nor did the numbers in manufacturing. But you are OK that the birth of a child could distort the figures…I am OK with a changing workforce distorting the figures because we are talking about employment…not births.

    Comment by Ross — October 10, 2012 @ 8:12 pm

RSS feed for comments on this post. TrackBack URI

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

Create a free website or blog at

%d bloggers like this: