Does new housing improve affordability? Economist Stu Donovan breaks down an important but controversial question.
Wellington City councillor Iona Pannett recently suggested new apartments in her neighbourhood of Mt Victoria wouldn’t improve affordability, because they were priced at $1 million and had “gentrified” the area. In another instance, Heritage New Zealand’s central region director Jamie Jacobs told Re: News replacing heritage buildings with new housing wouldn’t help first home buyers because the underlying land was expensive, so presumably any new housing would be, too.
Comments like this crop up regularly in discussions about housing, such as when people complain that new developments will just result in “luxury” apartments. Unfortunately, the views expressed by the likes of Pannett and Jacobs fundamentally misunderstand the important contribution that new housing makes to affordability for three reasons.
The first reason is that the price of new housing needs to be compared to other housing in the same area. In many neighbourhoods in Auckland and Wellington, for example, even new housing with a price tag in excess of a million dollars will be more affordable than the average dwelling in that same area. According to OneRoof data, for example, the median dwelling price in Mt Victoria is $1.2 million. By this measure, million dollar apartments are relatively affordable.
The second reason is that a simple comparison of prices doesn’t tell you much about quality. And new housing tends to be of a higher quality than existing housing simply because it is, well, new. Research from the US estimates that the capital value of houses depreciates at 2% per year even when allowing for maintenance. Accounting for this capital depreciation is especially important in Wellington because persistently low rates of building have left the city with a relatively aged housing stock. If we assume an average $1.2 million house in Mt Victoria was built in 1980 and has capital improvements worth $400,000 today, for example, then its new price today (adjusting for depreciation) might be closer to $1.8 million.
The third and most crucial reason is that new housing improves affordability far beyond the actual houses themselves. How? Well, new housing creates a series of vacancies whose effects ripple out over the city. People who can afford it move into the new houses, freeing up their houses for others. The vacancies that result from new housing help to put downward pressure on prices for everyone, even the people that choose to stay put where they are.
To put in simpler terms, imagine a hypothetical city in which there are 100 households with varying income levels. Then a developer comes along and builds ten new “luxury” homes in the city, which are purchased and occupied by the ten wealthiest households. This is, I think, the sort of “nightmare scenario” that looms large in the minds of people like Cr Pannett when they conclude that new housing doesn’t help with affordability.
The problem with this conclusion, however, is it ignores that the ten wealthy households have likely vacated their old homes, which then become available for other households. When these vacancies are taken up by other households, then there are another ten empty homes available for others to occupy. And so on, and so forth.
The process by which new housing creates vacancies that ripple out across the wider city is sometimes called “filtering” or “moving chains”. And the moving chain process is not merely a hypothetical theory proposed by a defunct Brisbane-based economist. Over the last decade, moving chains have been the focus of a large number of urban economic studies.
A recent study from Sweden, for example, considers how the effects of new homes varies with income. Although new homes tend to be occupied by wealthy people, like in our hypothetical city above, the authors find that the people who occupy the recently vacated homes are less wealthy than average. They also find that, in locations with higher housing construction rates, every income group enjoys better access to newer housing and more space. Another recent study from the US found new builds improve housing affordability for middle- and low-income households, even in the short run: “The effects are diffuse and appear to benefit diverse areas of a metropolitan area.”
A third study from Helsinki, Finland finds new housing “triggers moving chains that quickly reach middle- and low-income neighborhoods and individuals.” To illustrate these effects, the authors show that every 100 new houses in central locations results in 60 vacancies in zip codes in the bottom-half of the income distribution, with 29 of these in zip codes that are in the bottom-quintile of the income distribution. The authors conclude that new housing “is likely to improve affordability outside the sub-markets where new construction occurs and to benefit low-income people,” even when the new housing is priced at prevailing market rates.
Of course, this is not to suggest that all new housing will have the same effect. Evidence suggests that new housing makes a bigger contribution to affordability in cities with low vacancy rates and when the new housing is centrally-located. One possible explanation for the latter finding is that centrally-located dwellings are a viable alternative for more households, possibly because they’re accessible to jobs, education, and friends or family. This explanation seems to support policies that would allow more new housing in central locations.
Although the contribution of new housing to affordability is not immediately obvious, I think it’s intuitive when you account for location, quality, and these wider ripple effects. To use an analogy that may be even more intuitive: We wouldn’t measure the contribution of new cars to affordability based on their price. Indeed, new cars are expensive because they’re new!
And, if we were to turn off the tap on the supply of new cars, then older cars would quickly become more expensive. This is exactly what happened as we emerged from the pandemic, when global computer chip shortages pushed up the price of new cars and quickly fed through into higher prices for second hand cars. Similar market dynamics apply to housing.
The situation that Wellington now finds itself in, where even very old and low quality houses are relatively expensive, is a major problem and the direct result of a decades-long failure to build sufficient new housing. To start to turn this around, Wellington now needs to build a lot of new houses – even houses that might appear, at first glance, to be relatively expensive.
Stuart Donovan is a senior fellow for Motu Economic and Public Policy Research in Wellington. Stuart currently resides in Brisbane with his family and three worm farms. Stuart does not own property in Wellington, but he would consider doing so if prices were to fall.