I hadn’t had a look for a while at the OECD labour productivity (real GDP per hour worked) data, but the release of the latest OECD Economic Outlook the other day prompted me to spend some time in the (less user-friendly than it was) OECD database.
It takes a while for all the data to come together, and it is only annual, so the most recent near-complete data set is for 2023. On the OECD’s estimates – using national data, but converted at (estimated) PPP exchange rates – New Zealand stood 29th out of the 38 OECD countries (remembering that the OECD now has four Latin American “diversity hires” – all much poorer and less productive than the rest of the “club”).
Treasury highlighted a few years ago that the absolute level of reported New Zealand labour productivity may be understated (because of technical issues around how hours worked numbers are calculated/used, relative to the approach used for a number of other countries). The difference isn’t small but, as they noted, what is involved is largely a level-shift, and doesn’t affect materially comparisons of growth rates (of productivity) over time. Nor, of course, does it make any difference whatever to actual wages or living standards.
It is now quite well-recognised that productivity growth has been quite poor in many countries over the last decade or so (and the contrast between the US and some of bigger European economies has been remarked on often). Productivity growth in the typical high productivity OECD country has not been great – for example, for the eight years from 2015 to 2023, the median total growth in labour productivity for the top 10 countries [excluding Ireland and Luxembourg, for international tax reasons] was 4.9 per cent.
But much as we might like to catch up again with these high productivity countries, perhaps the most relevant comparators for us are the countries either side of us on the league tables, well behind the global productivity frontiers. In 2015, the start of our period, these were the countries with real GDP per hour worked already within 20 per cent either side of New Zealand’s. There are big gaps both above Slovenia and below Latvia (and, thus, even if the hours issue is fixed up, and there are no changes for any of these other countries, we wouldn’t even come close to the next country above Slovenia),
How has our labour productivity growth compared against these countries over 2015 to 2023?
The median productivity growth rate for this group of middle-to-lower (levels) countries was 17 per cent. New Zealand, by contrast, managed 3.5 per cent growth, not even managing to keep up with the median growth rate of the group of highest productivity countries (see above).
It really is a woeful record. And in case you are wondering if perhaps 2024 might have made all the difference, on national data (GDP per HLFS hours worked) average productivity in 2024 was about half a per cent higher than the average for 2023, so no, not really. Just possibly SNZ data revisions might lift our past productivity growth a bit, but (a) these 2023 estimates should already include last year’s SNZ updates, and b) even at the most hopeful, it is doubtful any revision would lift our past productivity growth even to Japanese rates. It seems pretty likely that we were in fact better only than Greece among this mid to lower productivity group of countries.
People tend to push back and say “yes, but so many of those other countries are in Europe”, and sure, about two-thirds are. But it isn’t as if being next to the US has done much for Canada’s recent productivity growth (productivity growth over this period was a touch worse than New Zealand’s, and the level of Canadian productivity is far below that of the United States), and quite a few of these countries border either Russia or Ukraine (Estonia appears to have taken quite a hit), and Israel has been fighting a war (there is 2024 data for them, no higher than the 2022 numbers). And to the extent geography matters, and it almost certainly does, it is a binding constraint we have to live with, not an excuse for perpetual underperformance (we were, after all, even in my lifetime – just – still in upper tier of advanced countries). It is a reflection of a series of poor policy choices, and the evident growing indifference of our politicians (and bureaucrats?).
And worse, there is no sense of urgency, about outcomes that shape the lives and options of this generation and the next. The glib “oh, they can always go to Australia” – itself not a stellar performer – is no decent basis on which to build a country.
croaking cassandra 
Hi Michael
When calculating productivity are all working age New Zealanders taken into account, or just those who are in employment? I believe we have something like 12% of working age Kiwi’s fully dependent upon a benefit. Calculated or not, this must have a dampening effect on our national productivty.
I note this is up from around 3% in the early 1970’s.
LikeLike
These measures use hours actually worked (other measures, less commonly quoted, are per person employed). The point you make – not uncommon in other countries either – drags down GDP per capita, but shouldn’t materially affect GDP per hour worked.
LikeLike
Thank you for the clarification.
I appreciate that economies are complex, that there is a voluntary sector that contributes as well, and there is a mix of those over 65 who are either fully retired or work full time or part time. That said, is there another population based index that takes all the variables into account including those fully dependent upon the state for one reason or another?
I’m guessing those who read your articles are interested both in where we sit compared to other nations, and also the trajectory we are on from an economic perspective. For example the productivity index would appear not to change if we had 20% of our working age population on welfare, but life for the taxpayer would likely change a great deal.
LikeLike
I’m pretty sure there is no single indicator that will give you what you are after, but do note that labour force participation rates here are fairly high by OECD standards (my impression is that our welfare dependency numbers while up a lot are not unusually high by OECD stds, much as we might still deplore them). Note too that govt spending and revenue as a % of GDP are a touch under the OECD average and our net debt as % of GDP is below average. For all our other problems, economywide productivity is really the most glaring failure.
LikeLike
Can you expand on any Government policies over the last 20yrs which you think have had major impacts on our growth? Thanks
LikeLike
If you mean a list of new measures that have helped materially growth in productivity, it would be a short list (I can’t really think of any). But on the other hand the persistence of high policy-driven rates of immigration, high rates of taxes on business, tough planning restrictions, FDI restrictions have acted as drags, probably accentuated by rising energy prices.
There are small things working both ways. Govts like to talk up trade agreements for example but actually trade as a share of GDP has been trending down this century.
LikeLike
Michael
Do OECD do productivity per hour by industry category? There may be answers in there as to our relative problem… majoring in the wrong industries perhaps. I thought I had recently seen the ranking by country of bureaucrats per head of population and we were well outside the norm. Again that is partly because we are small but likely also simply ineffective. All that drags on total productivity….
Paul Chrystall
LikeLike
People have played around with such data (eg Productivity Commission and MBIE/Tsy) for NZ. An obvious contrast of course is with Australia where labour productivity in mining is huge (v capital intensive industry) altho of course Aus is far from being a productivity superstar.
On the bureaucrats I don’t think those numbers (I think I saw a similar chart) are correct (prob classification issues). Total govt spending as a % of GDP here is little different than the OECD average/median.
LikeLike
Here are Treasury’s forecast labour productivity growth rates for Budget 2015 as in fsm BEFU15). Nope, sorry, the system will not let me paste the table into this comment after all.
So, in words, labour productivity growth was forecast to increase by 1.4, 1.7, 1.7 and 1.2 percent in the years ended June 2016 to 2019 respectively, before hitting and sustaining the sweet spot of 1.5% p.a. in the year to June 2020 and beyond.
This is not to sneer facilely at Treasury. The forecast errors are sobering for the macroeconomic forecasters more generally.
LikeLike
I put your observation in a comparative chart on Twitter
https://x.com/MHReddell/status/1930470290750071270
LikeLike
Thanks Michael, and yes, I did mean the epidemic lockdowns, not the GFC. Apologies
LikeLike
P.S. Of course, a GFC would not be forecast in a most likely scenario back in 2015
LikeLike
Pandemic I’m guessing you meant? (on the other hand, the default assumption prob would have been another recession at some point in the coming 10 years, based on past form)
LikeLike
Hi Michael,
What impact do labour market regulations and laws have on productivity per hour worked? New Zealand in my observation tends to have labour market settings that – all else being equal – tend to encourage high participation.
Compared to settings in some European countries in particular which make employers reticent to hire staff for fear of the consequences of mishires and having to carry more staff than otherwise have lower participation.
Would this mean that the average European worker is more productive as the less productive workers are not participating in the labour market?
To add to this New Zealand labour laws make it quite difficult to cost effectively move non-performing/productive employees on.
It seems to me that we therefore have a situation where our least productive workers are encouraged into work but employers are generally discouraged from either training them (a cost rather than investment) or holding them to account on productivity.
If they do so, the power of the system is very much in favour of the employee.
Contrast this to the American (and some Asian market) system which is very much designed to ensure productive contribution from employees less they suffer the consequences of job loss. A brutal system perhaps but it appears to deliver results
Combine the New Zealand settings with high immigration and some extremely high marginal tax rates for workers due to various transfers it feels like a recipe for the results in your post.
LikeLike
Those composition effects (least productive potential workers are less likely to be actually working in some of those other countries) are likely to be real but also aren’t likely to explain much of the productivity gaps. A couple of the highest productivity countries (Sweden and Switzerlannd) have higher labour force participation rates than NZ does (and Denmark isn’t far behind). When the productivity gaps are 60% or more it is the overall typical erconomywide experiences that matter more than whether some potentiallyl lower productivity people (prob at most 5% of the potential workforce) are in or out of the labour market.
LikeLike
I would argue that a older employee is more productive than a more youthful employee. If you value experience then you would find that a older employee with the benefit of experience completes his day to day task significantly faster than the younger less experienced employee. A younger employee may appear to be working harder and putting in more time and effort but the lack of experience means that the younger employee spends more time chasing his own tail than actually getting anything done.
LikeLike