By Ryan Greenaway-McGrevy*
One of the stated goals of the Auckland Unitary Plan (AUP) is to restore housing affordability in the city. But it will not be easy to tell if it is working – at least not at first.
Let me explain.
The plan is designed to increase the supply of residential dwellings by, among other things, relaxing restrictions on urban density. If successful, that increase in supply will help put downward pressure on prices.
But relaxed urban density restrictions will also increase the value of land. That will put upward pressure on the price of dwellings that sit on land that is able to be profitably redeveloped. This is because the value of the option to redevelop the property is embedded in the sales price. In other words, the right to pull down the old bungalow and replace it with a block of units is going to attract a premium.
Disentangling these offsetting effects when measuring overall trends in property prices will be key to evaluating the efficacy of the AUP. But the widely-reported price indices published by Quotable Value and the Real Estate Institute of New Zealand are an average across all residential sales, which obscures any bifurcation in the market. As such, these indices may not be that useful when evaluating whether the plan is restoring affordability to the Auckland housing market.
What we need are price indices that distinguish between properties that can be profitably re-developed on the one hand, and those that cannot on the other.
This redevelopment premium depends, in part, on where a property is located, since density is only permitted to increase in certain targeted areas of the city. But it also depends on the existing extent of site development – which I will refer to as ‘site intensity’. While it makes sense to pull down a three-bedroom bungalow to build an apartment block, it makes less sense to pull down a block of apartments to build another block of apartments. In economic terms, the opportunity cost of the forgone rent from the demolished bungalow is much less than the opportunity cost of the forgone rent from the demolished apartments.
One simple way we can evaluate whether the AUP is working is to first identify sales transactions that are located in areas zoned for additional density, and then distinguish between low-intensity and high-intensity properties before constructing price indices for each separate group.
If intensification is working, we should expect to see an increase in the price of low-intensity housing that has been upzoned to support additional density. Think about a three-bedroom bungalow sitting on a quarter acre that that falls into the ‘Terrace Housing and Apartments’ residential zone, which is one of the highest density residential zones under the AUP. This low-intensity housing should have appreciated significantly in value, reflecting the increase in the value of the option to redevelop the site.
Meanwhile, existing high-intensity housing that has been upzoned should not increase in value, and, over time, should decline in value if intensification brings additional construction of high-intensity housing to the market. Think about an apartment block on Mt Eden Road that has just been upzoned under the AUP. The apartment block already is the kind of housing that the plan is designed to encourage, and it makes little sense to redevelop the site. The redevelopment premium embedded in this kind of property should therefore be rather small. And if the zoning changes encourage construction of more high-intensity housing, the apartments within the block should also decline in value (at least relative to price trends in the rest of the housing market).
But in order to observe these price trends, we would have to construct a different price index for each type of property – high-intensity and low-intensity – for each residential planning zone in the city. This is exactly what Kade Sorensen and I do in our Centre for Applied Research in Economics discussion paper.
We first define a ‘low-intensity’ property as having an intensity ratio below the median intensity ratio of all transactions in the quarter of sale, while a ‘high-intensity’ property has a ratio above the median. The intensity ratio, in turn, is the ratio of the assessed value of improvements to the total assessed value of the property. This ratio is our measure of the existing extent of site development. A small state house sitting on a quarter acre typically has a low intensity ratio, while at the other end of the scale, an apartment typically has a high intensity ratio.
Having sorted transactions according to their intensities and residential zone, we construct price indices for each different group. The price indices are based on repeat sales – this is the same methodology used in the US for the Case-Shiller price indices. Repeat sales circumvent many – but not all problems – associated with measuring house price inflation.
The figure below exhibits prototypes of these price indices for Auckland from quarter one 2011 to quarter four 2015 for the ‘Terrace Housing and Apartments’ residential zone. This zone has the least restrictive land use regulations among the residential planning zones in the AUP. For example, it permits seven to eight storeys to be built. We begin in 2011 as this is immediately after amalgamation into the super city. Our data ends in 2015, but we soon hope to update our results to 2016.
The blue line is a price index based on sales of high-intensity properties, while the red line is a price index based on sales of low-intensity properties. The price indices are normalized to one for Q1 2011.
Evidently there is a marked divergence in prices across the two groups. High-intensity properties appreciated by 77% between Q1 2011 and Q4 2015, while low-intensity properties appreciated by about 95%.
This result is what we would expect if upzoning inflated the redevelopment premium. Within the Terrace Housing and Apartment Zone, low-intensity properties (e.g., the bungalow sitting on a quarter acre) have appreciated by significantly more than high-intensity properties (e.g., apartments and flats).
If the AUP is successful, we should expect this divergence to continue. But, more importantly, the price index on high-intensity properties in this zone will decline if construction brings additional supply of terraced housing and apartments to the market.
Next, let’s have a look at price trends for properties located in the ‘Single House’ residential zone under the AUP. This is one of the lowest density residential zones in the plan.
High-intensity properties in this zone appreciated by 74% over the five-year period, which is qualitatively indistinguishable from the 76% increase in low-intensity properties. These properties have generally not been upzoned for additional density – and thus we would not expect to see a big difference between the price indices.
In the full paper we also examine price trends in the ‘Mixed Use Urban’ and ‘Mixed Use Suburban’ zones. We observe patterns that are consistent with upzoning inflating the price of redevelopable properties. Within the Mixed Use Urban zone – which permits up to three storeys – low-intensity properties appreciated by 14.8% more than high-intensity properties over the 2011 to 2015 period. Meanwhile, within the Mixed Use Urban zone – which only permits up to two storeys – low-intensity properties appreciated by only 2% more, on average, compared to high-intensity properties.
Unfortunately our sample ends in 2015 due to data constraints. But we hope to be able to update these and the other indices in our discussion paper soon – and on an ongoing basis – in order to see whether intensification is bringing affordable housing options to Auckland.
*Ryan Greenaway-McGrevy is a Senior Lecturer in Economics and the Director of the Centre for Applied Research in Economics at the University of Auckland. Kade Sorensen is a PhD candidate at the University of Auckland and recipient of the Kelliher Charitable Trust PhD Scholarship in Economics.