In the 80s and 90s, Australian governments were hugely successful in reshaping the economy: they introduced new financial paradigms, reformed industrial relations, lowered trade barriers, and did it all for the better of the country—but they stopped short of reforming the policies that govern our physical world.

In order to build the Australia of tomorrow, we must get serious about reforming the way we regulate and utilise land. This is the greatest and most important policy challenge of our era, and it’s what Boundless is about.

This is the book Howard and I have wanted to write for a long time, and we’re very excited to be doing it. Currently, we are both buried deep in Australian history—beginning all the way back in 1788—laying out the history of land use, and building a strong understanding of how we got to where we are today.

I rebooted this O’Mail substack like one would a fibremaxxing diet—with great ambitions of regularity. For the time being, though, I’ll be directing all the energy I have for writing toward the book. If you hear from me here, it won’t be much. Between running both YIMBY Melbourne and Inflection Points, and now writing Boundless, I am not exactly oozing spare time from my pores. (This very bodily paragraph brought to you, most likely, by cortisol).

That all said—the best way to know about the book is to subscribe to this mailing list here. (I am increasingly negligent of Twitter, and LinkedIn’s algorithm is unknowable.)

I’m looking forward to sharing more when we can.

As I said in my tweet: this shook me to my core—and so I decided to investigate: why does Australia have auctions, and who are they good for?

Subscribe now

Auctions were first required by law—and have remained popular since

Land auctions were first mandated by the British Colonial Secretary in 1831

In the early years of the Australian colony, most land was distributed by grant. Free land was used to entice settlers, awarded to military officers, and even given to ex-convicts who had served out their term.

When a single male ex-convict was freed, he would often be granted 30 acres. If he had a wife, 50 acres. If he had two kids—a whopping 70 acres. If you’re like me and no good at thinking in acres, let me help you out: 70 acres is 283,000 square metres. The mythical quarter-acre block is about 1,000. The median lot size in inner-city Melbourne is around 500 or less.

This immense generosity was reigned in over time, and on January 3, 1831, land grants came to an end all together.

Previous Prime Minister of the United Kingdom Viscount Goderich was serving as Colonial Secretary at the time. And so it was within his purview to issue a despatch to the colony, expressing dismay at the way land was being allocated and offering a correction.

The Viscount’s feeling was that the Crown’s wanton allocation of large land tracts was extremely inefficient. “The Regulations now in force,” he wrote to the Governor of New South Wales, “have not had the intended effect of preventing large tracts of land from being appropriated by Persons unable to improve and cultivate them.” Basically: the land they gave away for free was not being well-used.

Goderich was charmingly dry about the whole state of affairs. He wrote: “This result does not surprise me”. Me neither.

The letter in full is an interesting record of Australian cities’ path-dependency. Within its few pages the Viscount expressed alarm at the way settlers had spread themselves out across the land, due to the large acreages the Crown had granted them:

These two apparently inconsistent evils of a high price and of a want of demand lead me to believe that cultivation has been too widely extended, and that it would have been more for the interests of the Colony, if the Settlers, instead of spreading themselves over so great an extent of territory, had rather applied themselves to the more effectual improvement and cultivation of a narrower surface.

With concert and mutual assistance, the result of the same labour would probably have been a greater amount of produce; and the cost of transporting it to market would have been a less heavy item in the total cost of production. A different course however has been pursued, chiefly, as it appears, owing to the extreme facility of acquiring land.1

The land granting process may in no small part underpin the remarkable lack of density in Australian cities today: from before our nation began, households were incentivised to sprawl further and further away from each other.

But we are getting sidetracked. We now know the Viscount’s diagnosis of the problem: giving random people large amounts of land for free creates suboptimal outcomes. So what is his solution?

I am sure you must have found the impossibility of giving satisfaction to all the applicants for land and of reconciling contending interests, and that you will gladly be relieved from the irksome and ungracious task of endeavouring to do so.

[...]

The measure, to which I allude, is that of declaring that, in future, no land whatever shall be disposed of otherwise than by sale, a minimum price (say five shillings an acre) being fixed, but this price not to be accepted, until upon proper notice it shall appear that no one is prepared to offer more, the highest bidder being in all cases entitled to the preference, ten per Cent on the whole of the purchase money to he paid down at the time of sale, and the remainder at an early period after the sale and previous to possession being granted. This last Regulation I conceive to be of great importance, and it ought uniformly to be adhered to.

And so the colony began to sell all land by public auction.

Auctions were quickly popular, and have remained so

People started hanging out at auctions almost immediately, and as the majority of auctions moved from the Crown to the private market, the spectacle remained. Per Dr Richard Gillespie for Museums Victoria:

The new suburban estates spread along the expanding rail and tram lines; prospective buyers of land were lured with the offer of free railway passes and lunches of chicken and champagne prior to the auction.

These sorts of incentives still exist as part of the property sales pipeline, though mostly for boutique off-the-plan apartments, which are not typically auctioned.

Nonetheless, Australians remain happy to rock up to Saturday morning auctions. Bad coffee, it seems, is a fine substitute for champagne.

In 2021, more than one in eight properties were sold at auction. In Melbourne, Canberra, and Sydney, this number was more than one in four.

Interestingly, very few properties go to auction in Perth or Hobart. To the point where Domain—back when they used to publish a monthly auction report—excluded those two capitals.2

Why is that? The answer, once again, is regulation.

Who are auctions good for?

For sellers, auctions offer certainty

The crucial thing to know about house sales by auction is that they are binding. There is no cooling-off period, meaning that neither party can back out of the sale once the deal is signed.

At an auction, the sale is unconditional from the moment the hammer falls: the highest bidder is legally bound to purchase, and they must do so on the spot.

In Australian states outside of WA and Tasmania, the benefits of auctions for sellers are obvious: the lack of cooldown period offers greater certainty.3 But in states where auction and private treaty contract requirements are identical, the additional costs of an auction—property setup, fees, auctioneer, weekend hassle—offer no extra benefit to the seller.

But what about buyers?

For buyers, auctions offer tradeoffs

The obvious benefit of a live auction is price transparency: each buyer knows exactly what the other buyers are offering. In closed negotiations, information is asymmetric: the seller knows all the offers, and each buyer knows only what they hear from the seller. This leaves competing buyers in private treaties vulnerable to overbidding.

On the other hand, the rush of a bidding war can also lead to overbidding, and result in a phenomenon known as the winner’s curse, in which the victor of an auction overpays and as a result experiences suboptimal long-term outcomes. This phenomenon is not limited to property—it’s also been observed in places like stock market initial public offerings.4

But the most recent Australian study I can find—a 2012 paper by Frino, Peat, and Wright—found that when properly controlling for relevant factors, there was “no significant difference between prices of properties sold at auction to those sold by private treaty”.5 So it kind of feels like a wash.

Auctions are fine, actually

The majority of objections to Australian home auctions are aesthetic: the spectacle, perhaps, feels like a gross extension of our property-obsessed culture. And maybe it is.

But when you look at both the history and the outcomes of auctions, the whole thing looks a lot more like a cultural curiosity than a substantive factor in the successes and excesses of our inflated housing market. But it’s an interesting curiosity nonetheless.

I’m not enamoured with either take. The obsession with new cities, I think, comes mainly from the fact that thinking about new cities is a lot of fun. In practice, of course, creating new cities is very costly, and rarely ever works. This is because cities are labour markets, and labour markets are difficult to create from scratch. The most recent example of a new Australian city is Canberra, a place that only exists because the government forces the bulk of the public sector to work there.

This gets to the heart of the new city problem: individuals and firms need a good reason to move. For any new city to succeed, it must offer some combination of resources or amenities that existing cities do not. Otherwise, no one will go there.

It is difficult to imagine how new cities might differentiate themselves in the Australian context. Local governments have limited ability to meaningfully vary policy,2 and it’s not like a new city can offer meaningful tax and transfer incentives to entice would-be residents. A new Australian city would just have to be really, really valuable on its own merits—but it almost certainly wouldn’t be.

All of this is okay, of course, because I come bearing good news: we do not need to create new cities.

This article will do two things: first, it will weigh in on the recent new-city discourse.

Second, it will lay out the fact that Australia actually has a large number of cities. We just don’t use them very well, and could very easily make that change.

Setting the “new city” record straight

Australia is not particularly centralised

One of the important things to remember is that Australia does actually have a lot of cities. City Density creator Jonathan Nolan laid this out, demonstrating that Australia’s centralisation in just two cities is not actually unique.

One mildly interesting fact is that because Sydney is only 1.03× the size of Melbourne, Australia’s top-two cities are most similar in size when compared to any other top-two city pairs internationally.

But while we are ranked first on this metric, we are by no means an outlier—meaning that this is more of a party fact to tell your uncle this Christmas, rather than a north star to inform your next policy submission.

Specialised cities are not necessarily better

Manning’s article focuses on Marshallian externalities—the agglomeration benefits of specialisation, often driven by a few dominant firms. I think this is a fine story to tell, but it is incomplete, and understates the reasons we might think that Australia’s generalist cities are actually internationally successful, and well-placed to continue that success.

To get a more full understanding of how cities drive growth, we can look to de Groot, Poot, & Smit’s 2009 paper Agglomeration Externalities, Innovation and Regional Growth: Theoretical Perspectives and Meta-Analysis. Crucial to understanding this meta-analysis is that there are three kinds of agglomeration externalities:

• Marshall-Arrow-Romer externalities: the role of concentration and specialisation

• Porter externalities: the role of competition intensity

• Jacobs externalities: the role of diversity and cross-sectoral spillovers

Focusing only on Marshallian externalities, then, is incomplete: there are other ways we might foster healthy, growing cities. Analysing 322 articles, de Groot and friends looked at exactly this, and they found some pretty stark results: almost 4 in 10 studies found that specialisation has a negative impact on growth outcomes—whereas only 1 in 10 found a negative impact from diversity.

The overall evidence of the meta-analysis based on a simple counting of conclusive effect sizes reveals that relatively many primary studies conclude in favour of significantly positive effects of diversity and competition on growth. No clear-cut evidence was found for the effects of specialisation.

Economically diverse cities are the most successful cities.

Diversity is more often the key to a city’s success than specialisation.

This is a big deal. Manning may be right: there are certainly some cases where specialisation can yield enormous upside. But it’s also clear that specialisation carries large risks, and I’m not sure they’re risks worth us taking as a nation. If it were advantageous for industries or firms to centralise in single Australian cities, they would have likely already done so.

As the de Groot meta-analysis demonstrates, a lack of specialisation does not indicate that our cities are worse off than the counterfactual—if anything, it may explain their relative strength. The Australian economy has gone a long time without downturn—and our cities have each done the same, in no small part because diversity creates resilience to shocks.

I think where Manning’s article is most correct is when it discusses the failures of Australian industrial policy. What it misses is that policy attempting to create specialisation among our cities may just be yet another expensive exercise in picking losers.

Australia has many under-utilised cities

Let’s say you’re unconvinced by the above, and still think we need new cities. You may, for instance, be under the impression that Australia has relatively few sizeable cities compared to peer nations. But this is also not really the case.

The relative size of Australian cities stacks up globally

In the chart above, you can see that Sydney and Melbourne are very close to each other in size—but that the following drop-off in Australian city size is not all that precipitous.

It is only in 5th place where we see Adelaide drop below the US’s equivalently ranked Dallas, and at 6th where the Gold Coast drops below Ottawa in Canada. We stay above England until 15th, where Cairns as a proportion of Melbourne drops just slightly below Bradford as a proportion of London.

We aren’t the United States, but crucially—neither is anyone else. That said, how might we get closer?

Small Australian cities are subject to stringent land use regulations

Australia’s 18th-largest city is Ballarat. Its origins of growth can be found in the gold rush of the 1850s, and between 2014 and 2024, it remained the 12th fastest-growing Australian city. But the city is subject to significant land use restrictions that force most housing development outside the city, and restrict the development of commercial capacity.

YIMBY Melbourne compiled Ballarat’s zoning data as part of our submission to the Inquiry into the supply of homes in regional Victoria. Our focus was on residential zoning, but because our mixed use density category is a combination of commercial and mixed-use zones, you get a clear idea of just how prohibitive the zoning in Ballarat is for new entrants.

While Melbourne and Ballarat allocate similar amounts of land to industrial use, Melbourne enables mixed-use and commercial development on more than 30% of its land. Ballarat allows the same on just 4%.

Even if Ballarat wanted to specialise, it would likely struggle to do so, because the rules that govern its growth are prohibitive. The scarcity of commercial land will mean reduced availability and higher costs—not exactly the conditions for fostering innovation outside the big cities that one might hope to find after so many hours of inland travel.

We should enable all our cities to grow

Of course, Ballarat is just one city—but this story generalises across our nation’s regions.

The point I’m making here is one that the YIMBY movements of our nation’s capitals do not often have time to emphasise: that the restrictions our large cities face are faced in equal measure by our smaller cities. That in fact the case of Ballarat demonstrates that these restrictions may well be more extreme in our smaller cities. The NIMBYism of residents is almost certainly as fierce or more.

I mentioned earlier that I am wary about policymaking that tries to pick winners. By divvying up land arbitrarily, legacy planners have created an entire school of policy based around the idea that picking winners is a) desirable, and b) reliable. But it is neither of these things.

If you want new cities, or better cities, or cities that specialise—the policy prescription is the same: loosen land use restrictions in every city across our nation, and let people and firms vote with their feet. We should want to create the best possible set of Australian cities—but that requires letting people build them in the first place.

Footnote for skeptics: chart with additional comparison nations

And an obligatory link to Zipf’s Law.