4th April, 2021
It's just over 9 years since Sir Paul Callaghan lost his battle with cancer.
And, this week, 10 years since he spoke at a StrategyNZ workshop called “Mapping our Future”. Then he was still just Professor Paul Callaghan (he was Professor of Physical Sciences at VUW - not perhaps the most obvious qualification for the talk he presented), and his talk was “Sustainable Economic Growth For New Zealand: An Optimistic Myth Busting Approach”.
I know this wasn't the first time he gave a version of this talk, and it definitely wasn't the last - I wasn't at that workshop, but did have the opportunity to see it in person twice subsequently, and it evolved over time1. However, this was the one that was recorded and is available on YouTube, so in some ways it is the reference case.
Many people have enthusiastically adopted the memorable expressions he used in this talk (e.g. "be the place where talent wants to live"), and we still do a lot in his name - not least Callaghan Innovation.
But how many have taken the time to really understand what he was trying to communicate, or with any thought to how that message might have been updated, if he were still here.
A lot of what he talked about in 2011 has really informed me on the topics I'm trying to write about this year. So, rather than three different things this week, I'm going to focus on this one…
First things first, I encourage you to actually watch the talk.
It's just over 20 minutes long (13½ if you watch at 1.5x speed).
There are parts of it that have aged quite significantly, but most of it feels just as relevant today as it was 10 years ago.
Some of you may have seen this previously. Watch it again, with fresh eyes if you can.
And if not, sit back and enjoy for the first time...
While there are some big ideas in this talk, for me the biggest takeaway is actually Paul's style of presentation, which I assume reflected his style of thinking.
He presents a common belief, exposes the flaws and then (importantly) in each case proposes a better and different way of thinking about it. He doesn't only tell you the correct answer (from his perspective, at least) he explains why your assumption was wrong!
I love it.
For those of you who didn't follow the instructions above, let's catch you up on the full list of seven common myths that he spoke about in 2011, and re-consider each of them now with the benefit of hindsight...
1. New Zealand is an egalitarian society
This is something that we still like to believe, although I feel that we're now a little more honest about our current reality.
In 2010 the value for New Zealand was 0.32. In 2011 (the data he used in his slides) the value was 0.35, which ranked us 26th out of the 33 "rich" countries.
As for international comparisons, the most recent league table is from 2014. At that point in time we had dropped one further place to be 27th out of 33 countries.
It's difficult to imagine our ranking has improved significantly since then.
In other words, despite what we think, nearly all of the other rich countries that we like to compare ourselves with are more egalitarian than us.
2. New Zealand is 'clean and green'
The example he used was the impact on water quality from increased dairy farming.
I wonder what he might have said then if he had known that between 2011 and 2020 the output from dairy farming would increase a further 22% (from 17,339 million litres of milk to 21,145 million litres).
He also included this graph, to highlight the hypocrisy of expecting other developing countries to be more respectful of their environmental impact now, when our own history in this regard is poor, especially from 1800 onwards:
All of that still feels very topical today, although perhaps the focus may have shifted away from our marketing messages and more towards our actual response to climate change.
3. We don't need to be more prosperous
“We don't need to be rich like those Australians”
This was a very personal point for Paul. By 2011 he was already fighting cancer, and explained: prosperity is actually what allows us to afford the things we all want more of, like better cancer drugs, and better infrastructure.
In other words, this is why we can't have nice things!
He specifically focussed on the income gap with Australia. In New Zealand we love a per capita measure of success. But, unfortunately, this is not one of them!
I was born in 1976, which is the last time that New Zealanders earned more per capita than Australians. During the 35 years shown on this graph the gap steadily increased. By 2011 the difference was 35% or around NZ$45 billion (i.e. to match the prosperity of Australia, New Zealand would need to earn NZ$45 billion more per year).
So, what's happened since then?
In 2019 (the most recent year we have data for both), Australia earned US$53,068 per capita (~114% of OECD average) and New Zealand earned US$45,301 per capita (~97% of OECD average).
This means the gap has grown to ~NZ$47 billion.
We're further behind now than we were then.
Paul was also giving this talk in the wake of the Christchurch earthquakes, and he specifically highlighted the predicted cost of the rebuild as "severe". It's chilling to think that the government portion of that rebuild was $7 billion, compared to the $50 billion the government has allocated in the last year on the Covid response (of which $38 billion has already been spent) and the $42 billion earmarked for new infrastructure spending in the next four years.
There is a lot of work to do!
4. New Zealanders have a relaxed easy-going lifestyle
(aka, to be more prosperous we'd need to work harder)
This is perhaps the most confronting statement Paul makes in this talk:
“We are poor because we choose to be poor”
What did he mean?
Specifically he meant we choose to own and work on businesses that require long hours for little output per hour worked.
He showed a plot comparing how many hours we work vs productivity (output per hour):
Note: I’ve updated this based on OECD Data, but using the same subset of countries that Paul used in 2010 - this really should be updated to include all of the OECD countries now in the dataset.
In 2011 we worked longer than just about anybody, but earned less per hour than everybody except for Greece (at least amongst that OECD group of developed countries).
So, again, what has happened in the decade since then? Have we improved at all?
In fact, we now work even harder than we did then, and with only a marginal improvement in output per hour.
We are still poor because we choose to be poor.
5. More tourism would be good for the NZ economy
Next, Paul breaks it down by sector, to try and understand the component parts.
To do this he calculates the revenue per employee required to maintain then current levels of GDP, and comes up with $120,000 per job (or, more accurately per full-time equivalent employee)4.
He compares that average to the values for the two biggest sectors:
Tourism: $80k per FTE
Dairy: $350k per FTE
The two lessons he draws from this still feel very contemporary:
Firstly, it's a mistake to think that we could grow more prosperous on the back of more tourism. This is because each additional job in this sector, as it's currently configured, drags the overall average down (and if we target only high value tourism jobs, as is proposed, those are by definition likely to be significantly fewer jobs and so will have less impact on overall numbers).
Secondly, we need to be careful before we dump on dairy too hard. As he says "if it weren't for Fonterra we'd be desperately poor". I think this point in particular is often missed by those who want to encourage growth in the "tech" sector - he wasn’t suggesting we needed to replace jobs in tourism or agriculture with technology jobs, but that we need to create new businesses in addition to those existing sectors.
I need to do a lot more number crunching before I comment on how this has shifted over the decade since. When I tried to re-calculate Paul’s 2010 values using OECD data I got a revenue per FTE value of $110.5k (slightly lower than his $120k). My equivalent value for 2019 is $141.1k (an 11% increase in real terms)5. So, that’s a good benchmark for your own business - are you dragging the average up or down?
I'll write more about this soon. But for now, it is interesting to consider revenue per FTE for Xero, which in 2011 wasn't on the radar, but is now increasingly held up as a case study:
In FY20 Xero reported revenue of $235k per FTE. This value has increased every year, with FY21 results due soon, and has been improving the national average ever since 2015.
6. You can't manufacture in NZ
“You can manufacture in NZ, you just need to make products with high profit margins and high value per weight”
This is an interesting one.
In 2011 the most successful kiwi “tech” companies were nearly all manufacturers: F&P Healthcare, Rakon, Tait Electronics, Gallagher. Even Paul's own company, Magritek.
So thinking about the contribution that manufacturing, and specifically ELT manufacturing (meaning elaborately transformed) could make was logical.
However, as we'll see below, many of our recent successes have actually been software companies, with their weightless exports.
He also highlights our very low levels of R&D investment. Interestingly this has become a big focus for Callaghan Innovation in recent years.
“We have the sort of businesses that don't need much R&D”
Unfortunately the progress in the 10 years since once again doesn't make great reading:
2011: Just under 1% of GDP, compared to OECD average of 2.1%
2018: 1.4% compared to OECD average of 2.3%
We've closed the gap a little, but at our current rate it’s going to take a long time to catch up.
I'd like to understand the sector-by-sector numbers better, and see how much of this increase has been driven by sectors that previously under-invested in R&D getting better vs the growth of new sectors where levels of R&D are just higher. Does anybody have this data?
7. We are small so we need to specialise
“The words of truth are often paradoxical”
The instinct to pick winners is seemingly timeless. As Paul explained, trying to anticipate the specific types of businesses that are likely to be successful is much more difficult than it seems on the surface (part of the reason for this, as I tried to describe a couple of weeks ago, is because building a successful business is not just about an innovative idea, it's mostly about execution).
So, to use his 2011 examples, while it might seem like we should have competitive advantages in biotech or cleantech, actually the large amounts invested in those areas from the top down had little impact. Instead he predicted “we better be prepared to be good at some pretty weird stuff”.
So, where have our successes in the time since actually come from?
Before we can answer that we need to skip ahead to his prescription.
In the talk he highlighted the top 10 exporters from the 2010 TIN Report6. Together those 10 companies generated $3.9 billion of revenue per year. The top 3, F&P Appliances, Datacom and F&P Healthcare, accounted for nearly 60% of that.
His proposal was simple in theory: to generate the additional $45 billion needed to close the gap with Australia we needed 100 companies of that size (actually 115 additional companies, for a total of 125 companies, but let's not get distracted by detailed maths).
So, how how much progress have we made towards this target?
By 2020 the TIN Report listed 18 companies that were at least as large as the 10th largest from 20107:
Eight companies that are still in the top 10 (although two of these, NDA Group and Tait Communications, reported reduced revenues compared to 2010); plus
Five companies that were in the top 100 in 2010, but not the top 10, who have grown revenues sufficiently; plus
Five newbies, including Xero and PushPay as well as Livestock Improvement Corp, founded 1909, and Transaction Services Group, apparently founded ~2001 (it’s not clear why those two weren't on the 2010 list?)
So, a net increase of eight companies.
Together those 18 companies contributed additional revenue of $3.53 billion. That means in ten years we got less than 10% of the way towards Paul's target.
This trend is even more depressing, if you consider that the very first TIN Report in 2005 contained 13 companies with reported revenue greater than $100m. This increased to 17 companies by 2010 and to 25 companies by 2020. So even over that longer time period the growth of new companies in that bracket has been pretty anaemic.
There is a litany of problems with the TIN Report, in my opinion - both how the data is collected and reported, which companies are included or not, and who can access the data because of how it's packaged and sold. But, let's not dwell more on this today.
Reflecting on the prescription that Paul gave us all, I wonder if he actually got some important things wrong:
Firstly, when you look at the companies that have grown the most in that 10 year period many are software companies. When we think about manufacturing we're looking for exports that are higher volume by weight, and thinking about the expensive R&D that is typically required to develop those types of businesses. So, where do software companies fit in - with technology that is often actually low-tech, but with potential that has been proven multiple times now to generate a lot of weightless export revenue? Of course, this should be a “yes, and …” not a “yes, but …”.
Secondly, the results further down the list have been much more impressive. In 2010 the 100th biggest company listed in the TIN Report was Triodent with revenues of $12m. In 2020 the 100th biggest company was Modica Group with nearly twice that much revenue ($21.2m). In 2010 there were 51 companies listed with more than $30m revenue. By 2020 that number had increased to 79 companies (a 55% increase). So, maybe rather than focussing on a small number of very large companies, the "weird stuff" we are good at is actually going to produce a much larger number of smaller companies.
Which leads us to an even biggest mistake, in my opinion. Towards the end of the talk he says:
“It takes one genius entrepreneur to make a company like this”
Reflecting his own myth busting approach back at him... I think this is a common misconception.
We like to believe that this is all it takes, just as we like to believe that NZ is clean and green or that to be more prosperous we'd just need to work harder. And, it's true that there is often just a single name attached to each of the businesses that becomes big enough to get attention (Xero is Rod Drury, Rocket Lab is Peter Beck etc etc). But, as soon as you scratch the surface of any of these companies you find that their path to success was actually built by a large team of people, with a wide range of skills and specialities, all working together.
We don't only need 100 crazy founders. We need 10,000 people contributing their expertise to these businesses. Then 100,000. What are the constraints to that?
Again, I'll write more about this myth soon.
I only met Paul in person a couple of times, and only briefly. I was delighted to discover his interest in what I was doing at the time, and I enjoyed those conversations.
Go back to the very first words he says in this talk:
“We don't think very well about the future”
The thing about the future is it's happening constantly. The future he was trying to think about in 2010 is what we now call the present.
A couple of weeks I quoted Glenn Ashby from Team NZ: (the recurring pattern this week seems to be our efforts to catch up with Australians!):
“Throw the ball as far away as you can and then run to try and get to where it lands”
I think the biggest mistake we can make today, in thinking about Paul's legacy, is to just repeat his mantras without understanding that if he were still here now his thinking would have developed and evolved, based on the results that have been achieved and the lessons learned in the process.
So, the question should be: how do we build on the ideas he shared, and update them for our current reality?
In case it's not obvious, that's what I'm hoping to do with this 2021 project. I've thrown the ball. I'm interested to see if I can reach it. Thanks for reading the drafts each week and helping me to make them better.
What did I get right or wrong this week?
There are a lot of numbers and charts this week, hopefully most of them made some sense.
I'm grateful for the following organisations for the data sources which help to shine a light on the questions I’m asking:
Notably, the only data set I've relied on which isn't freely available is the TIN Report. The 2020 version of this report cost me $399+GST. For that I got a link + password to access a digital copy of the report, which I couldn't even download to use offline. I encourage those at NZTE and other government organisations who have funded this report extensively over the years to stop and think about what they've enabled in the process.
Top Three is a weekly collection of things I notice in 2021. I’m writing it for myself, and will include a lot of half-formed work-in-progress, but please feel free to follow along and share it if it’s interesting to you.
As Nicola Gaston pointed out in her interview with NBR this week, Paul tailored his message depending on the audience that he was speaking to, so perhaps not surprising that a lot of the intent has been lost over time.
It would be useful to run this analysis using wealth as the measure rather than incomes. I believe the OECD Gini Coefficient numbers are calculated excluding housing costs. Given the extent to which house prices have had attention in the news (and public policy announcements) in recent times, what impact does that have when they are included?
John Key would later say that 100% Pure was “like a fast food ad” and “should be taken with a pinch of salt”
This is a relatively simple calculation, using his numbers: GDP per capita = ~$40k, Population = 3.9 million, FTEs = 1.3 million, so each of those 1.3 million FTEs needs to earn ~$120k to get to a GDP of $40k across the whole population.
F&P Appliances ($1.1b)
F&P Healthcare ($503m)
Tait Electronics ($200m)
Douglas Pharmaceuticals ($145m)
Curiously Moffat itself was not one of those having dropped to $80m revenue - their 2010 result was just an estimate, so perhaps an over-estimate at the time?