For some time now we’ve been seeing housing consents set new records set again and again. In total, 16,656 consents were issued in the 12 months to the end of December, up 10% on the same time the year before. We’re also seeing new records for the types of dwellings being consented with 44% (7,285) of those consents being for townhouses compared with 39% (6,535) for single houses.
There’s a lot to celebrate and the Unitary Plan has certainly helped kick off significant development around the region. Yet at the same time, the housing crisis continues unabated and the records we’re are yet to make up for crater that occured from about the mid-2000’s through to the mid-2010’s that saw annual consents drop to just over 3,100.
There seems to be a couple of reasons for that drop. The first major drop in around 2004/5 seems to stem from the changes to the building act in 2004 which came in response to the leaky homes crisis and would have taken both confidence and capacity out of the market – with resource being diverted to fix up the stock we had. The second drop in 2008 was the impact of the Global Financial Crisis.
But just how much have these record levels of consents over the last few years eaten into our housing shortfall? Have they at all? There are some estimates around but with this post I thought I’d make my own.
For this I started by looking at Stats NZ estimates for how our population has grown between 1996 and 2020. Our total population has grown from 1.116 million in 1996 to 1.717 million in 2020.
I then focused that on looking at just the adult population as children don’t live in houses alone and Auckland’s adult population is growing at more than twice the rate of growth of children and so I think it gives a better representation of what’s needed.
The number of adults in Auckland grew from about 779,000 in 1996 to 1.277 million in 2020. There will various reasons for this growth, such as children becoming adults and immigration but that’s not really the point of this post. You can also see that annual growth below and has similarities to the shape of the consents graph above with with a peak in the early 2000’s and trough during the GFC before strong growth again.
Next I looked at census data on how many dwellings there were in Auckland based on data from the census. This showed we grew from about 380,000 dwellings in 1996 to 538,000 in 2018. To that we can add the number of Code Compliance Certificates (CCCs) the council issued through to 2020 to get an estimate for how many dwellings Auckland had as of the most recent population estimates. That suggests we have about 561,000 dwellings in mid-2020.
When this is compared to the adult population it shows that at the lowest, in 2001, we had one dwelling for just over every 2 adults in Auckland. That figure grew with each subsequent census and in 2020 it stands at nearly 2.3 adults for every dwelling. I suspect a decent chunk of this adult children staying at home for longer due to unaffordable house prices.
From this we can estimate that if we had the same ratio of adults to dwellings as we did in 2001, then in 2020 we should have about 636,000 dwellings. That’s means we have a massive shortfall of about 75,000 dwellings. Furthermore, if you look at how many dwellings are needed to support the population growth shown in the graph above, that shortfall is still growing. For example, about 11,000 CCCs were issued in the year to mid-2020 but Auckland’s adult population grew by about 33,700 meaning we’d need 16,800 dwellings to keep up.
Whether 2001 is the right year to base this on is certainly something that could be up for debate. I haven’t done anything to look at that in more detail but house prices were certainly more affordable then.
Finally, while there is not as much (public) history on CCCs, like there is for consents, what we do have suggests most new dwellings are completed within two years of getting consent and the majority of the rest being under five years. Another way of looking at this is to say the lag for reaching milestones is just over two years. This suggests that the record consents we’re now seeing will take about 26 months to flow through to completed homes. Whether that will be enough to help eat away at the shortfall will depend a lot on what happens with population.
Whatever happens, it’s going to take a very long time to clear that backlog even if population growth slows and we keep building at current rates.
Update: a couple of other graphs I was working on but didn’t originally include as we only have CCC data from 2013 onwards.