You touched on this with the too small driver but one other factor is that with Australia's population being under 30m, there's only enough demand to enable specialisation through export markets. It's not a coincidence that the one capital that's specialised does so in our number one export industry.
Out of the potential specialisations you mentioned I'd say the Melbourne one is most plausible since biotechnology is inherently such a global market
Sharp piece, Manning. From Hobart, I can’t help noticing that Tasmania never even makes the frame in these debates, despite being the most obvious test case for what you’re talking about. Hobart has some bones of specialisation - Antarctic science, MONA-driven cultural tourism, renewable energy - but policy settings here seem designed to sand off the edges rather than sharpen them. Instead of doubling down on a natural strength, we get the same scattershot hydrogen-hub announcements and “creative industries” strategies as everyone else.
Do you think our allergy to letting a city own a distinctive identity is cultural as much as economic? There’s a national wariness of appearing “parochial” that seems to cut against the very idea of specialisation. Curious whether you see that as a structural obstacle or just a failure of political imagination.
When you looked to other countries did you consider Canada? One of the curious things that has emerged in the Trump tariff regime is the fact that Canadian provinces have been able to impose tariffs and other trade restrictions against each other. So the level of economic integration in Canada is low and the provinces compete against each other for investment.
NAFTA and it predecessor made it easier to trade with the US and you got a different pattern of development and specialisation as a consequence that is a higher degree of integration with the US. Prior to Trump cross-border activity was strong with the Canadian hydropower and natural gas sector serving the North East US markets and importing US logs to produce furniture etc.
Each of the Australian colonies had an administrative and port city (entrepot) facilitating extraction industries (and inward migration).
Federation and subsequent periods 'import replacement ' and then open market policies have always built on this settlement pattern. To some extent the federal model fuels inter-state competition. There is very little imagination in industrial development policy at the federal level or states (with the possible exception of SA).
Looking at Newcastle, Wollongong, Geelong and Port Kembla historically you may have seen more specialisation by occupation/industry. This might unsettle your argument a bit but could be worth a look.
Once you move down the urban hierarchy in the US sectoral specialization reduces considerably also. Deindustrialisation has a kind of sameness to it.
Lastly, if you look a little more closely at say the finance sector you may observe differences between Melbourne (superannuation/pension fund management is strong) versus Sydney (not so much).
Interesting! My impression is that the US specialisation developed naturally. Philadelphia had a stock exchange 2 years before NYC for example.
I would worry that politicians would pick bad specialisations. I can imagine in 2021 some premier or mayor proposing to become the NFT capital of Asia!!
Great piece Manning. I agree about Australia simply not having scale to warrant differentiation.
I think an interesting creative experiment could be looking at the restaurant cuisine aggregation within suburbs in Sydney / Melb / Brisbane. The similarity here is there are spaces in Australia where there is a natural aggregation of places with similar specialised characteristics (e.g. Indian restaurants in Harris Park).
Further, I love the practical examples included and the research undertaken! I agree with potentially the Feds taking a more intentional role in designing policy to incentivise aggregation or specialisation, but inevitably there will be issues with politicians and bureaucrats picking and choosing winners. E.g. labelling Surry Hills a early stage venture zone indirectly through R&D incentives etc would draw critique from everyone not from Surry.
It is worth looking at the Australian capital cities in terms of Jane Jacobs and "city regions." The one for Sydney starts at Wollongong and extends past Newcastle towards the central coast. SE Queensland runs from Noosa right through into NSW and past Tweed Heads.
Melbourne is a fan shape - Geelong swinging round Ballarat and then out east past the Dandenongs. Perth is north-south finishing somwhere near Busselton- for the moment.
When you only have 5 big cities in a continent-sized country, maybe you shouldn't be shocked that they're all generalists. Being outsized regional hubs does that to metropolitan areas.
The whole piece and not a single mention of Australia’s most specialised city? So what about Canberra….?
But if you look at others, Adelaide was Detroit and is now turning into a defense & submarine location, the Gold Coast is Orlando with all the theme parks, Newcastle is steel, etc.
Though more general, I do think there is something to your description that cities are less specialised, though it can be explained by various aspects of Australia’s economy overall.
- Australia as a country is hyperspecialised into mining and education. Both sectors don’t have high location externalities, mining is constrained by the geography of natural resources and for education the link to domestic population implies that there aren’t that much advantages to a concentration in a single place (at least from a broad “industry” perspective, not from a top leading research one). This implies that both, mining and education is broadly distributed around Australian capital cities without being concentrated into a single one. And with 27mio it’s larger but not that much then the NY Metro area (20-22mio), so it would fit into the overall picture as simply another specialised area in the US.
- Australia is much less of a closed integrated economy. Going back into history, the eastern US seaboard and the early Aus colonies were both directly connected and linked to the UK. With the westward expansion, the US turned into its own ‘closed’ large economy trading internally, while Australia changed from shipping stuff to the UK to ….. shipping stuff to China. Specialisation requires trade to make sense, so the much more external orientation (and connected specialisation) of Australia implies less scope for internal specialisation.
- Australia also has a strange ‘duopoly’ capital city with Melbourne and Sydney both at 5pmio. Standard urban economics would imply more of a 7/3 split, a counterfactual with likely higher specialisation (finance and media probably goes into wherever the 7mio are….). So explaining that similarity in size might explain a lot here as well.
And as a final note, a couple years ago an overseas economist visiting Adelaide half-jokingly said that Adelaide should specialise into becoming a retirement home for rich Asian (ok, he formulated that a little bit nicer sounding….). Clean, safe, relaxed, good wine & food, decent health care, affordable prices, cultural amenities, relatively ok transport connections, and so on…. Economically it has its logic, politically…… yeah not going to become official policy ;-)
Living in Perth I got half way into this thinking hang on mate, and then read the Perth paragraph. Side note: Perth has many synergies to SE Asia that are indirect to the mining and resources industry. For example, many of my colleagues in professional services in Singapore and Hong Kong keep houses in and around Margaret River which are available on short term rent during most of the year. They either are from Australia or visited Perth on business. The timezone means they can easily work down here with their family. Interesting piece, thank you.
How much should new cities play a role? It’s utterly thinkable that Australia doubles in size in the next 80 years. Where do we put these 27.5 million new Aussies? At least a few million into new cities, I’d hope. How do we begin with creating a Phoenix out of the ashes / dust? What specialization do we pick for new settlements? Who picks? I enjoy looking at the Gold Coast as the only case study of a sizable new city since WWII.
Good article, although I don't see Sydney giving up its film industry to Brisbane without a fight! Also those US interstate move stats are not really comparable to Australia. The US has 50 states, a lot of them tiny. Moving from DC to the Virginia suburbs, or Manhattan to Newark, count as interstate moves, even though people can still commute to the same job. I suspect there is still a lot more meaningful movement in the US but we'd have to dig into the data a bit more.
You touched on this with the too small driver but one other factor is that with Australia's population being under 30m, there's only enough demand to enable specialisation through export markets. It's not a coincidence that the one capital that's specialised does so in our number one export industry.
Out of the potential specialisations you mentioned I'd say the Melbourne one is most plausible since biotechnology is inherently such a global market
Sharp piece, Manning. From Hobart, I can’t help noticing that Tasmania never even makes the frame in these debates, despite being the most obvious test case for what you’re talking about. Hobart has some bones of specialisation - Antarctic science, MONA-driven cultural tourism, renewable energy - but policy settings here seem designed to sand off the edges rather than sharpen them. Instead of doubling down on a natural strength, we get the same scattershot hydrogen-hub announcements and “creative industries” strategies as everyone else.
Do you think our allergy to letting a city own a distinctive identity is cultural as much as economic? There’s a national wariness of appearing “parochial” that seems to cut against the very idea of specialisation. Curious whether you see that as a structural obstacle or just a failure of political imagination.
When you looked to other countries did you consider Canada? One of the curious things that has emerged in the Trump tariff regime is the fact that Canadian provinces have been able to impose tariffs and other trade restrictions against each other. So the level of economic integration in Canada is low and the provinces compete against each other for investment.
NAFTA and it predecessor made it easier to trade with the US and you got a different pattern of development and specialisation as a consequence that is a higher degree of integration with the US. Prior to Trump cross-border activity was strong with the Canadian hydropower and natural gas sector serving the North East US markets and importing US logs to produce furniture etc.
Each of the Australian colonies had an administrative and port city (entrepot) facilitating extraction industries (and inward migration).
Federation and subsequent periods 'import replacement ' and then open market policies have always built on this settlement pattern. To some extent the federal model fuels inter-state competition. There is very little imagination in industrial development policy at the federal level or states (with the possible exception of SA).
Looking at Newcastle, Wollongong, Geelong and Port Kembla historically you may have seen more specialisation by occupation/industry. This might unsettle your argument a bit but could be worth a look.
Once you move down the urban hierarchy in the US sectoral specialization reduces considerably also. Deindustrialisation has a kind of sameness to it.
Lastly, if you look a little more closely at say the finance sector you may observe differences between Melbourne (superannuation/pension fund management is strong) versus Sydney (not so much).
Thanks for an interesting article!
Interesting! My impression is that the US specialisation developed naturally. Philadelphia had a stock exchange 2 years before NYC for example.
I would worry that politicians would pick bad specialisations. I can imagine in 2021 some premier or mayor proposing to become the NFT capital of Asia!!
Great piece Manning. I agree about Australia simply not having scale to warrant differentiation.
I think an interesting creative experiment could be looking at the restaurant cuisine aggregation within suburbs in Sydney / Melb / Brisbane. The similarity here is there are spaces in Australia where there is a natural aggregation of places with similar specialised characteristics (e.g. Indian restaurants in Harris Park).
Further, I love the practical examples included and the research undertaken! I agree with potentially the Feds taking a more intentional role in designing policy to incentivise aggregation or specialisation, but inevitably there will be issues with politicians and bureaucrats picking and choosing winners. E.g. labelling Surry Hills a early stage venture zone indirectly through R&D incentives etc would draw critique from everyone not from Surry.
It is worth looking at the Australian capital cities in terms of Jane Jacobs and "city regions." The one for Sydney starts at Wollongong and extends past Newcastle towards the central coast. SE Queensland runs from Noosa right through into NSW and past Tweed Heads.
Melbourne is a fan shape - Geelong swinging round Ballarat and then out east past the Dandenongs. Perth is north-south finishing somwhere near Busselton- for the moment.
When you only have 5 big cities in a continent-sized country, maybe you shouldn't be shocked that they're all generalists. Being outsized regional hubs does that to metropolitan areas.
The whole piece and not a single mention of Australia’s most specialised city? So what about Canberra….?
But if you look at others, Adelaide was Detroit and is now turning into a defense & submarine location, the Gold Coast is Orlando with all the theme parks, Newcastle is steel, etc.
Though more general, I do think there is something to your description that cities are less specialised, though it can be explained by various aspects of Australia’s economy overall.
- Australia as a country is hyperspecialised into mining and education. Both sectors don’t have high location externalities, mining is constrained by the geography of natural resources and for education the link to domestic population implies that there aren’t that much advantages to a concentration in a single place (at least from a broad “industry” perspective, not from a top leading research one). This implies that both, mining and education is broadly distributed around Australian capital cities without being concentrated into a single one. And with 27mio it’s larger but not that much then the NY Metro area (20-22mio), so it would fit into the overall picture as simply another specialised area in the US.
- Australia is much less of a closed integrated economy. Going back into history, the eastern US seaboard and the early Aus colonies were both directly connected and linked to the UK. With the westward expansion, the US turned into its own ‘closed’ large economy trading internally, while Australia changed from shipping stuff to the UK to ….. shipping stuff to China. Specialisation requires trade to make sense, so the much more external orientation (and connected specialisation) of Australia implies less scope for internal specialisation.
- Australia also has a strange ‘duopoly’ capital city with Melbourne and Sydney both at 5pmio. Standard urban economics would imply more of a 7/3 split, a counterfactual with likely higher specialisation (finance and media probably goes into wherever the 7mio are….). So explaining that similarity in size might explain a lot here as well.
And as a final note, a couple years ago an overseas economist visiting Adelaide half-jokingly said that Adelaide should specialise into becoming a retirement home for rich Asian (ok, he formulated that a little bit nicer sounding….). Clean, safe, relaxed, good wine & food, decent health care, affordable prices, cultural amenities, relatively ok transport connections, and so on…. Economically it has its logic, politically…… yeah not going to become official policy ;-)
Living in Perth I got half way into this thinking hang on mate, and then read the Perth paragraph. Side note: Perth has many synergies to SE Asia that are indirect to the mining and resources industry. For example, many of my colleagues in professional services in Singapore and Hong Kong keep houses in and around Margaret River which are available on short term rent during most of the year. They either are from Australia or visited Perth on business. The timezone means they can easily work down here with their family. Interesting piece, thank you.
Really interesting article!
How much should new cities play a role? It’s utterly thinkable that Australia doubles in size in the next 80 years. Where do we put these 27.5 million new Aussies? At least a few million into new cities, I’d hope. How do we begin with creating a Phoenix out of the ashes / dust? What specialization do we pick for new settlements? Who picks? I enjoy looking at the Gold Coast as the only case study of a sizable new city since WWII.
Good article, although I don't see Sydney giving up its film industry to Brisbane without a fight! Also those US interstate move stats are not really comparable to Australia. The US has 50 states, a lot of them tiny. Moving from DC to the Virginia suburbs, or Manhattan to Newark, count as interstate moves, even though people can still commute to the same job. I suspect there is still a lot more meaningful movement in the US but we'd have to dig into the data a bit more.