Australia’s productivity growth stalled by mass immigration

The Conversation’s editor, Peter Martin, has penned an article bemoaning the collapse in Australia’s productivity growth since 2005:

Australian labour productivity growth

Australia’s labour productivity has collapsed since 2005.

According to Peter Martin:

  • The high point for Australia’s productivity growth was 2005 (as shown above), “driven by an ever-building surge in output per working hour (“productivity”)”.
  • But from 2005 onwards, Australia’s productivity growth collapsed. “In the 15 years since, Australia’s output per working hour (productivity) has grown by just 17 per cent”.
  • Martin notes that “things we do these days are harder to automate” mainly because 80% of Australians now work in services jobs.
  • He then concludes that “we might be coming up against hard limits in the amount we can squeeze out of each hour of paid work”.
  • This decline in productivity growth will make it harder to sustain an aging population.

As we know, Peter Martin is a rabid supporter of the ‘Big Australia’ mass immigration policy (see here, here, here, and here). So it is no surprise that Martin conveniently omitted to mention that the ramp-up in immigration from 2005 has driven much of Australia’s productivity decline:

Australia's net overseas migration

Australia’s immigration intake accelerated after 2005, coinciding with the collapse in labour productivity.

Australia’s net overseas migration (NOM) jumped from an average of 89,000 between 1991 and 2004 to an average of 215,000 between 2005 and 2020 – an annual average increase in immigration of 140%.

This surge in immigration would have adversely impacted Australia’s productivity in three ways.

First, the massive rise in population growth created chronic infrastructure bottlenecks across the capital cities, driving up congestion costs. It also encouraged growth in low productivity people-servicing industries.

The infrastructure investment required to keep pace with population growth was also much more expensive than in the past, due to diseconomies of scale (e.g. tunnelling and land buy-backs).

Second, the migrants that Australia imported are typically less productive, as evidenced by them earning lower than median wages and suffering higher unemployment and underemployment than the Australian-born population (see here).

Third, the downward pressure on wages caused by mass immigration necessarily disincentivises employers from investing in labour-saving technologies and automation to lift productivity (see here and here). After all, why invest in these productivity enhancements when you can simply bring in cheap migrant workers to do the task?

In summary, if you want to know why Australia has suffered simultaneously from low wage growth and low productivity growth, look no further than the nation’s mass immigration policy supported by the likes of Peter Martin.

Restoring the mass immigration ‘Big Australia’ policy post pandemic would repeat the same mistakes and consign the nation to a low wage, low productivity future.

Unconventional Economist


    • Returning residents and citizens fleeing the virus for relatively safe harbour in Australia.
      Plenty of PRs apparently returned to India to live well on Aussie earnings from family or other sources, now crying out to return to OZ.

  1. Just for the record, Leith: Peter Martin of “Crawford School” is already out there, policing any miscreant passers-by, who dare to disagree with his self-serving exclusion of the bleeding obvious p-for-population factor.

    It’s why we totally respect The Conversation’s integrity so much, isn’t it. Academic rigour, journalistic flair. The same article also replays in today’s Canberra Times. Let’s see if they permit any polite rebuttals.

  2. I see that a contributor called Phillip Brown has linked a MB article on immigration in the comments to the Martin piece. Martin, predictably, pretty much ignores it.

  3. Second, the migrants that Australia imported are typically less productive, as evidenced by them earning lower than median wages and suffering higher unemployment and underemployment than the Australian-born population
    I think you’re confusing the cause and effect here: shovelling in more immigrants may well be the cause of lower productivity, but their lower wages/higher unemployment are NOT the cause of it.

    • Even StevenMEMBER

      I think you need to re-read Leith’s comments, Mr Tezza. He is saying the lower wages are evidence of lower productivity.

      Granted, it’s an assertion, and i can come up with other explanations, but it is indicative.

  4. What happened to productivity in other countries? Have countries without high immigration also experienced declining productivity? The article mentions USA, but it also has high immigration. What about Japan, Denmark, Netherlands?

  5. I’m not sure why Australia’s management class gets away with this scott free.
    In 2005, investment % of nominal GDP was just under 30%. It has now trended down to about 23% – reducing capital per worker – and barely maintaining the earning power of the capital stock once you factor in depreciation.
    The quality of capex and acquisition decisions by any measure has also deteriorated.
    Yet executive compensation has skyrocketed.
    When executives are effectively raiding profits, not only do you lose the incentive of productivty growth, you also 1. encourage poor management & 2. reduce the capacity for business investment.
    It becomes “great management” to lift the price of stamps by 1/3 and then walk away with $11m. What does that do for productivity?

  6. Productivity WTF is it?
    it is easy to understand Productivity in a Manufacturing context
    it is possible to make sensible conclusions about Productivity in an Agriculture context (if suitable allowances are made for soil quality degradation and other environmental impacts (water, salinity, pesticides etc) of increased Ag output)
    Productivity as a measure still makes perfect sense Mining context if you just compensate for the lag between spend and output (and don’t take too much notice of the decreasing quality of the remaining unmined assets)

    But productivity wrt services like what on earth are you measuring?
    When a teacher takes the time to explain something to last laggard in the class, is that Productive or counter Productive?
    When an Aged care worker is expected to look after 30 end stage dementia patients on her own, is this outcome Productive? It is definitely not beneficial in that the service quality definitely declines, but I guess it’s “Productive”
    It’s simple, really simple, Productivity as a metric makes no sense in a Service based society.
    In Australia I’d say Productivity is a term that has out lived its usefulness, trouble is we’re addicted to the metric and we’ve embedded it into all our activities, heck we even want our “Productivity Commission” to weigh in on things like Road and Transport system upgrades…like why?

    • simple way for service businesses to boost productivity is to halve their management headcount.

      • Why would you think that less managers would boost productivity in a Service economy?
        Labour is the Input and Labour is the output, not to mention that even the necessity of the job “needing to be done” is something which is presumably defined by management. See what I mean? Productivity drops if we simply define less as needing to be done and Productivity increases if we simply define “more as needing to be done” …think KPI’s in the service sense it’s a BS in and BS out system. Most of the defined KPI tasks never actually needed to be done so without these requirements our measured labour productivity would be trending towards our broader actual labour Productivity (How much aggregate external output our economy/labour pool produces for the total hours worked)

        • if you halve management headcount for the same frontline staff and same output, that would boost productivity.

        • fyi labour isn’t the output.
          In a classroom student classes are the output, aged care – bed days etc.

    • This stuff must be very difficult to measure. An example… I work in software. Specifically mapping (well, geo-spatial data, but it’s basically mapping). Back in the day, maps were laboriously produced by hand. They were physical things, and you could treat them as manufactured goods. Some number of people worked some amount to produce some number of maps that were bought for some amount of money.

      Now, we have software engineers that produce software that produces digital representations of reality. The marginal distribution cost of this data is almost zero. Therefore, it counts as a service. However, it has completely displaced a “manufactured” good. We see this all over the place. For example, custom mechanical units have been replaced with commodity computers with commodity sensors driving commodity actuators… all the money is in the software that drives it.

      • Good points LD, as I said to begin with wft is Productivity anyway.
        What’s the Productivity of a game developer? Is it higher if he sells a DVD and lower if he sells a data stream
        the whole measurement metric makes no sense anymore. The sooner we dump this Productive concept the better.
        If you ask me which matric should replace Productivity I’d answer Connectivity. (especially within the context of Economic Complexity ala Harvard/MIT)

        I’d say this measure better reflects the lived experience of most Australians over the last two decades.

    • Productivity = no of folk though Sydney & Melbourne airport straight to Real Estate office & bank (if necessary) to sign up loan, mortgage, sale and or lease etc. nothing more NOTHING LESS

    • Good comment.
      I measure productivity by an increase in human wellbeing.
      Others measure productivity by how many more dollars go to privileged elites.

  7. SnappedUpSavvyMEMBER

    little peter martin lives in canberra far from any crush loaded cities and has a fence around his house

  8. Ian Verrender blamed our educational system for the skills shortage. But I think it’s unfair to pin all the blame on the educational system. The dysfunctional job hiring culture is also a major culprit for the atrophy of skills in Australia. Basically, employers want to hire workers that are trained and experienced at someone else’s expense. The problem is, since most employers have that attitude, then who’s the one doing the training and giving workers the chance to grow to gain experience?

    These article explained the situation brilliantly:

    The macro-economic consequences are as follows:

    Give that Medium author some love and clap for writing these 2 articles.

    • Do you really believe this?
      Do you believe that I would have trouble getting a job today if I graduated from a good university with skills tailored towards where we know that industry is headed?
      Lets get specific. If I graduated from say UoN (Unversity of Newcastle) not a top school but also not a bad school, with a degree in say Mechanical engineering, (for your info UoN is very highly regarded in the field of Control systems)
      Do you think I would have any trouble getting a job in a modern Robotics company?
      I can’t imagine that I’d have any trouble landing a job (it might not be a job in Australia) but I still believe I’d have a good relevant engineering job within say 3 months of graduation.
      I don’t get people who study something which has zero real job prospects and only discover this fact after 4 or 5 years.

          • drsmithyMEMBER

            You are talking about Engineering graduates, almost certainly going into specific graduate programmes which are, by and large, still competently run by actual Engineers (or at the very least, with their direct involvement).

            Others with similarly well-defined jobs and competently run industry systems to absorb graduates and traing them (eg: medicine) would be in the same boat.

            However. the articles are talking about workers who don’t fall into that sort of category, navigating the nightmare of increasingly nonsensical and fine-grained job descriptions with associated (and often automated) keyword-based applicant filtering managed by corporate HR departments (or, even worse, “HR partners” and contracting agencies who don’t even have a stake in the employer’s business) with basically zero understanding of the job they are trying to fill and the kind of applicant who can fill it.

            Two entirely different worlds.

          • Thanks for the response
            I guess I’m just glad I don’t know anything about that job market
            But that said nobody forced all these people to make these study choices, many of these tertiary education choices have much higher atar entry requirements then Engineering (and most have better overall career prospects at least in Australia)

          • drsmithyMEMBER

            You are still talking about graduates.

            The articles are talking about established career workers and/or the unemployed, who are job seeking.