Old housing can depreciate and still become less affordable

Nygaard and colleagues test a familiar theory of housing affordability: as new homes enter at the top of the market, older homes depreciate relative to them and gradually become available at lower price points. Their evidence from Melbourne and Sydney shows why that process is not automatic.

For filtering to materially improve affordability, new housing supply has to create enough slack relative to new household formation and other demand, and new housing has to be an effective substitute for existing housing. In Melbourne, the researchers find very low neighborhood-level supply responsiveness. In Sydney, older rental properties did move down relative to prevailing market rents—but that depreciation was not large enough to offset the broader increase in rents. Older housing became relatively cheaper while remaining increasingly difficult for low-income households to afford.

The authors caution that the process varies by submarket and can be interrupted by redevelopment, conversion, neighborhood effects, and weak substitutability between different kinds and locations of housing. The evidence is Australian, so it should not be read as a direct estimate of U.S. housing markets.

Original source

Filtering as a source of low-income housing in Australia: conceptualisation and testing

Christian Nygaard, Ryan van den Nouwelant, Stephen Glackin, Chris Martin, and Alistair Sisson · Australian Housing and Urban Research Institute (AHURI) · September 15, 2022

Evidence basis

Melbourne housing-stock, sales, and neighborhood-income dynamics, including change from 2006–2016, plus Sydney private-rental property histories using rental-bond, sales, and development records. The authors use conceptual, statistical, and econometric analysis to test whether aging housing actually filters toward lower-income households and how supply responsiveness, location, and substitutability affect that process.

Read at Australian Housing and Urban Research Institute (AHURI) ↗

The read

In a 2018 essay, Why older homes become more unaffordable if new homes cost more to build, I argued that the economics of producing the next home affect the affordability of the homes we already have. My formulation then was too categorical in places, but the underlying mechanism was the important part: when new production cannot keep pace with demand, scarcity does not stop at the newest building. It travels through the existing stock.

Nygaard et al. sharpen that argument in a useful way. Filtering is not a clock. It is a market condition. An old apartment can become cheaper relative to a new apartment and still become less affordable in absolute terms because the entire market is appreciating faster than the old unit is depreciating. Their Sydney evidence shows exactly that distinction.

Age creates the possibility of relative depreciation. It does not create affordability by itself. For that depreciation to translate into meaningfully cheaper housing, the market needs enough supply response to create slack relative to demand—and enough substitutability for that competition to travel across the housing stock.

That is a cleaner version of what I was reaching for in 2018: the affordability of an old home depends, in part, on whether we can build the next one.