No, we can’t address affordability by building lots more apartments

One of the (many) furphies that gets aired (frequently) in discussions around housing affordability is that we can build ourselves out of the problem by building a lot more high-density housing units rather than typical detached suburban houses.

In fact, the opposite is likely to happen should we attempt this course of action. Here's why.

First, the reality of construction economics is such as that every square metre of space in a high-rise apartment building is more expensive – roughly double or more – than the same square metre of space built in a detached single level home.

The traditional "sticks and bricks on slab" construction of detached suburban housing has been incredibly efficient over a very long period of time. As a building form, it lacks complexity (despite recent moves to introduce it via new standards). It's even been said that in post-World War II Sydney, as many as a third of the new suburban homes were by owner-builders.

Despite recent and rapid cost escalations, the build cost per square metre of a low-set detached suburban house is still around $1700 for the modest spec home. Not including the land, you can still find project home builders who will give you the keys to a brand new three bedroom, two bathroom house from around $220,000 to $300,000 – and that's for roughly 200m2 under roof.

It's a different story for multi storey apartments. These are more complex structures, with a good deal more going into them – more than just concrete and reinforced-steel slabs (which are themselves more expensive than timber frames or bricks for a low set house). There are also lifts, fire systems, standby generators, deep foundations and excavations for basement car parks… it's quite a list. Building them also takes a lot longer than a standard house and involves at least one tower crane, sometimes more.

All of which means each square metre of floorspace in that high rise unit tower is going to cost around $4,000/m2 for a structure of basic design. This applies to the whole floor of course – corridors and common areas included. The cost per metre for the actual living space, developers and QS's tell me, is closer to $5,000 per square metre. So a unit of 120 square metres will cost around $600,000 just to build. Not including the land or other development costs. This is around half the size but twice the cost of our detached new stick-and-brick example above – which includes builder margin.

Market prices need to factor in land and other development costs. The new low set detached house and land package will set you back from around $500,000 to $700,000, depending on location. (This is the discount end of the market of course). New home units however are well above this. Try finding a new three-bedroom apartment for $700,000 in Brisbane. Many are priced at $1m or more, even those well away from the urban core. (Spend a little while searching on realestate.com.au and see for yourself!).

What's happening to the theory that we can solve affordability by building more apartments? It doesn't stack up. You can't "solve" affordability by building more of the most expensive building types.

It gets worse. Developers will know that, given base build costs plus land, infrastructure levies, and other costs (including allowing for a margin), they will have a minimum retail price they can deliver apartments for. The build cost is virtually the same in middle and outer suburbs as the inner city (land being the only differentiator). But in middle and outer suburban markets, the price points that people will pay are below the cost at which the product can be delivered. So it's not worth doing. Instead, they will turn their attention to locations which can support higher purchase prices – and already well-off inner-city locations tick that box. There is less risk and a deeper market in building units that sell for over $16,000 per square metre (nearly $2m each or more) in any number of inner city locations, than trying to build lower cost units in suburban locations.

Which means the market supply of higher density housing will concentrate in and around the inner city where you will increasingly see new projects with price tags four times the Brisbane unit median. This is hardly doing anything for affordability.

A similar thing happened in Vancouver, Canada. There, as Patrick Condon (James Taylor chair in Landscape and Livable Environments at the University of British Columbia's School of Architecture and Landscape Architecture and the founding chair of the UBC Urban Design program) points out, the city tripled its housing via high density infill: "this was an unreservedly good thing – in all but one respect. This giant surge of new housing supply did not lead to more affordable housing as we all hoped. Somehow, confoundingly, the reverse happened."