Please stop confusing supply and density in housing debates
People say they get it. But they don't.
It is nearly the end of the financial year in Australia. A paid subscription to Fresh Economic Thinking could be a tax-deductible expense for you. Please consider supporting to get access to this and all articles, the audiobook of The Great Housing Hijack, the charts page, and more.
A dominant line of thinking in policy-land is that allowing higher housing densities increases the rate at which new housing is produced.
Few people express it so precisely, but many believe it to be true. In doing so, they unwittingly conflate the two completely separate concepts of ‘density at a site’ and ‘rate of housing production across a market’ under the single label of ‘supply’.
Indeed the less aware they are of the distinction, the more cocksure they are in calling out idiots who don’t see the world as they do. After all, “it’s ECON101, stupid”.
But supply is not density.
I’m reluctant to use the word supply at all.
Supply, in its specific and technical economic meaning, is not a quantity of existing homes nor a rate of production of new homes. Supply is a concept of market exchange. Think of it as an offer curve at a market exchange—all the sellers line up with the price they would accept to trade. This is why, with the same stock of homes, you can have a market boom or a market crash depending on supply (willingness to sell in a period) and demand (willingness to buy in a period).
Saying that a price change is due to supply and demand is not an explanation but a tautology. The explanation needs to answer the question of why sellers and/or buyers changed their willingness to sell and/or buy.
As Brian Albrecht has explained, supply represents the demand from the current owners of the product being traded. While we often think that supply is independent of demand, it really isn’t. There is no such thing as an independent supply.
I’ll do my best to talk about the stock of dwellings, the growth in the stock or the rate of new housing production, and other specific terms. Though it is hard to escape the word supply and I might just use it at times to mean the rate of new housing production.
Now, back to the frustrating topic of confusing density and the rate of new housing production. You will find an abundance of examples of this in housing debates.
Here is an especially clear example from the New South Wales Productivity Commission (NSW PC). They wrote a report asking about how much lower rents would be if all new buildings were denser but with the effect of assuming that dwellings were built faster.
About 1,500 new apartment buildings were built in Sydney between 2017 and 2022. These buildings averaged seven storeys and contained ten dwellings per storey (NSW Productivity Commission, 2023a). If instead we had permitted modestly denser development—for example, if apartments had averaged ten storeys instead of seven—then an extra 45,000 homes could have been provided, all without using any extra land and with minimal effect on neighbourhood character.
The additional 45,000 units would represent a little over two per cent increase in Sydney’s private dwelling stock. Typical rules-of-thumb suggest this extra supply would have lowered apartment prices and rents by 5.5 per cent, all else being equal (Saunders & Tulip, 2019). In dollar terms, this is a saving of about $35 a week in rent on the median apartment – or $1,800 a year. For a median-income earner, this is equivalent to a 2.75 per cent increase in their real purchasing power, similar to a typical year’s wage rise.
Here’s a replication of this approach for Vancouver, and you can see in this plot the counterfactual “extra” homes marked for the five years 2019-2023 due to an assumption that more density is a faster rate of new housing production.
I want to make it very clear here that density and the rate of new housing production per period are different concepts, different dimensions, with different units of measurement.
Confusing them is one of the most common problems in the analysis of housing “supply” in both academia and popular commentary.
Does this diagram help?
I am a visual learner. Maybe the below diagram of density and the rate of production on a two-axis chart will help you too.
Keep reading with a 7-day free trial
Subscribe to Fresh Economic Thinking to keep reading this post and get 7 days of free access to the full post archives.