Forecasting failure

The International Monetary Fund last week released their latest World Economic Outlook and associated economic forecasts. The value of these forecasts is not so much that they are likely to be correct – economic forecasting is, after all, largely either a mug’s game or a delusion – as that (a) their forecasts for all countries are done in a fairly common way, and (b) their previous forecasts are well-documented and readily accessible.

I took a look at a few of the IMF’s New Zealand projections. This chart, which I showed on Twitter last week, compares the IMF’s October 2019 (ie last pre-Covid) and April 2021 projections for New Zealand’s real GDP per capita for the period 2019-2026.

IMF projections for NZ

I thought three things were interesting about these IMF projections:

  • over seven years, the IMF thinks that real per capita GDP will have increased by a little over 4 per cent (a bit under 0.6 per cent per annum),
  • even by 2024 (three years away) the IMF still thinks that New Zealand’s real GDP per capita will be almost 2 per cent lower than they thought it would only 18 months ago,
  • the total gap between the two lines over the 2020-2024 period (covered by both sets of forecasts) is almost 15 per cent of a year’s GDP – simply gone, and with no sign in these projections that that loss will ever be recovered (even if, at some distant date, perhaps the two lines eventually converged).

This isn’t a comment about Covid or New Zealand’s Covid policy – all sorts of things may be going on in the thinking of the Fund’s forecasters – just a report of how much poorer the IMF now thinks New Zealand will be over the coming years than it thought only 18 months ago.

But perhaps, charitably/hopefully, you are thinking “well, that doesn’t look too good, but surely the rest of the world – Covid mess and all – must be worse, or at least as bad?”.

So here is the same chart for Australia.

imf aus

Which is probably the picture most countries would hope to have seen – some (inevitable) income losses last year, but a quick rebound and perhaps even a move to a future slightly higher growth trajectory. Note the other contrast to New Zealand: over 7 years the IMF thinks Australia will manage per capita growth of almost 8 per cent. Some part of the difference – although not the bulk of it – will reflect the fact that the IMF expects Australia’s unemployment rate by 2026 to be below 2019 levels, while in New Zealand the unemployment rate is expected to linger a bit higher than the pre-recession trough (itself higher than the previous pre-recession trough).

I didn’t do this comparison for a large group of countries, although for what it is worth the US chart looks more like Australia’s and that for Canada (not a stellar productivity performer in recent decades either) looks more like New Zealand’s.

I did have a look at the IMF’s forecasts for growth in real per capita GDP across all the advanced economies in the OECD (and the few non-OECD advanced countries). Over the period 2019 to 2026 we were well into the bottom quartile of OECD countries, at least on the Fund’s view of the outlook.

But it was this chart I found most sobering. Many of the countries we often like to compare ourselves with – from days past – are now much richer and more productive than New Zealand is. They are nearer the productivity frontier, and future growth can’t “just” be catching up. But in this chart I’ve shown the IMF’s forecasts for real per capita GDP for the countries with a 2019 level of labour productivity (OECD real GDP per hour worked, in PPP terms) roughly 10 per cent either side of New Zealand (the IMF does not publish productivity forecasts).

IMF projections for NZ and productivity peers

Just woeful.

Of course, the IMF could be quite wrong, not just about the absolute outlook for the world economy (that is almost certain) but also about the relative performance of New Zealand. But when independent observers take a look at your country, its performance, and its suite of policies and think you will over the next few years do so much worse than a group of your peers (whether, in our case, considerably richer and more productive Australia – to which it is easy for young New Zealanders to move – or to a group of advanced countries with similar levels of productivity to New Zealand, all looking to catch up) it really should be a wake-up call for our political and official “leaders”.

But in New Zealand it seems that all those responsible would prefer that no one noticed, prefer to ignore the utter failure projected (on top of past decades of relative decline), to burble on about wellbeing, and to deliver to our kids some deeply diminished legacy and opportunities. We once were world leaders, but now this.

(Out of interest, I wondered what the IMF was assuming about immigration/population: of all those advanced economies – not just the ones in the chart – we are forecast to have the third fastest rate of population growth over 2019 to 2026.)

18 thoughts on “Forecasting failure

  1. I suspect a driver of the change in the IMFs view of NZ is their horror that we re elected a bunch of incompetent communistic clowns so NZ s majority population is unable to see the results of their last term of abject failure and now is letting the same bunch have another go – like shooting oneself in both feet and not understanding why both feet hurt.

    Liked by 4 people

    • You may be right but it reminds me of a comment made by an American acquaitance about a year ago when he quietly and somewhat shamefaced admitted to having voted for Trump in 2016: “but you have to consider what the alternative was”.


      • Now that Biden is in charge, most Americans probably are now likely thinking that they should have stuck with Trump.


  2. The Productivity Commission’s approach under new management to the low productivity growth problem seems to be to redefine the concept of productivity to make the term devoid of meaningful content: “I see productivity as the way we utilise, maintain, and apply our taonga to deliver wellbeing.”

    Trying to decipher the coy ambiguity, my guess is that “we” means “government” and “our taonga” includes the resources you thought you owned.

    The notion that sustainable real wage growth depends on labour-productivity growth risks being lost in translation. I hope the Productivity Commission can clarify what it really stands for.

    Liked by 2 people

      • Ganesh Nana was embarassing
        Bernard Hickey needs to be much more discerning as to his invitees


      • The productivity malaise can really only be solved by keeping our factories. Every highly automated factory that we shut down is a loss of productivity. We have reached peak cows with 10 million cows. Our rural sector has very simply added an extra cow to increase productivity because only human working hours are counted in the productivity equation. 10 million cows produce an output GDP of $18 billion is considered more productive than 1.5 million Aucklanders that produce an output GDP of $85 billion. My 12 year Old will tell you our economists have rocks for brains.


  3. Bryce, the best thing the Government can do with the Productivity Commission is to close it down, along with many other parts of Government, and return those funds to the tax payer.

    Our governments over the past two decades seem to have forgotten that they’re the trustees of OUR money.

    Liked by 1 person

  4. The IMF’s chart of forecasts for real per capita GDP caught my eye. On Saturday I visted the Auckland ‘Home and Garden show’ and returned home with only two food items: a jar of salmon and a jar of pickled herrings ($4 each). At home i discovered both are produce of Latvia but with English labelling; the assistant who sold them to me said in strongly accented English that she was from ‘a Baltic country’. Initiative. How many Kiwis are in Latvia selling them our produce?

    Liked by 2 people

    • There are 1 million Kiwis deployed overseas selling NZ produce. But if we are selling tourism, then expect productivity declines. If we are selling milk or meat, unfortunately we have reached peak cow with already 10 million cows.


  5. The really surprising aspect of this failure is the lack of media commentary (present company excepted).
    Recent Govt initiative to raise annual leave to five weeks, extend maternity/paternity leave, raise minimum wages, increase benefits, add another holiday, and add a swathe of restrictions and burdens on employers pass with widespread media acclaim.
    I can’t think of one Govt initiative likely to improve our woeful productivity stats.

    Liked by 3 people

    • Richard, you are mixing up productivity with profitability. The additional holidays do not reduce productivity as productivity measures output against working hours. The highest productivity gains have already been made when we reached the age of computerisation and robot’s in our highly automated factories. More holidays reduces profitability not productivity.

      Our productivity malaise is the outcome of shutting down our factories and not from being allocated more holidays.


      • When the government saved Tiwai Point Smelter from being shut down, they have at least maintained productivity. If that smelter had shut down, that would have been a loss of productivity. When NZ refineries shut down, NZ will suffer a loss in productivity. The 2,000 highly skilled and highly productive workers will redeploy as cleaners and retail assistants in the tourism sector and aged care with the resultant loss in productivity.


      • Productivity can be compared to hours worked or population or simply as a total – as in GDP, a nations total productivity with no reference to population or hours.
        If we had forty weeks annual holiday would total productivity – houses built or goods delivered say be the same? Is that what you’re seriously suggesting?


      • You will have to ask Michael. He uses working hours in most of his numeric and statistical presentations to calculate economic productivity. But I can see why. You can’t very well measure our growing numbers of aged, sick and dementia ridden people and have a government policy to increase productivity expecting the sick and dying working a few extra hours from their sick bed.


      • Horses for courses. Productivity is about output per unit of input applied, but it isn’t the only variable that matters. GDP per capita is a better measure of current average living standards, and explicitly takes account of how much work is or isn’t done.


      • David George, I believe Grant Robertson did come into government initiating the Future of Work with automation and AI robots fast replacing humans and its implication. 40 week holidays with 3D printed AI robots and increased productivity is really not that far in the future.


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 )

Connecting to %s