This is a post from Peter Nunns. See Part 1 here.
In my previous post I gave an overview of the postgraduate research project I’ve been working on over the past year. It looks at the causes and economic consequences of rising regional housing prices in New Zealand.
Nobody reading this will be surprised to hear that house prices and rents have been rising much faster than wages over the last generation, and that Auckland has experienced some of the most rapid increases.
But I think that most people would have trouble explaining exactly why prices have increased so much. At the macroeconomic level, there are many potential causes: the growth of the financial sector and the increasing availability of cheap credit, tax policies that incentivise home-buying, a fast-growing population, and a variety of barriers to building more homes. A 2011 research paper from three OECD researchers summed this up nicely.
However, macroeconomic factors don’t necessarily explain why prices have risen more in some places relative to others. For instance, if banks are lending more freely, it should push up prices in Dunedin and Auckland to a similar degree. That hasn’t been the case, so we need to take a look at what’s going on at a microeconomic level.
If a house that was bought in, say, 2005 sells for a higher price in 2015, there are a couple of things that might have caused the increase.
First, the owner may have renovated the house or built an addition onto it. Investing in the property to make it nicer will increase its value.
Second, demand to live in that location might have increased. This could reflect changing preferences for different places, improvements to the quality of schools, shops, and public spaces in the area, or changes to transport accessibility, eg due to a new rapid transit route.
Third, the city may not have enough housing to keep up with demand. Scarcity can drive up prices even for crummy houses in bad locations, as people grow increasingly desperate for somewhere to live.
All three factors are at work in Auckland, and elsewhere in New Zealand. So I used a method developed by economists Ed Glaeser and Joseph Gyourko in a 2003 paper to identify how much of an impact each of these factors has had over the last generation.
The following chart shows the rise in the inflation-adjusted price for an average standalone house in the former Auckland City Council area, over the last generation. (I’ve focused on standalone houses as Glaeser and Gyourko’s statistical methods are easiest to apply to house sales, but the same basic dynamics apply to apartments as well.) In 1990-1992, the average house sold for around $330,000. By 2014-2016, prices had risen to over $1,360,000 – a fourfold increase.
The orange line in the following chart shows how much of this increase can be explained by building improvements and renovations. To calculate this, I used data from three-yearly ratings valuations, which assess the value of buildings on each site. This shows that, if section prices had held constant but homeowners had continued to invest in renovations, then average prices would have only risen to $520,000.
Next, I looked at the role of increasing demand for residential land in different places. To do so, I ran 1800 regressions – one for each council area over each of the 25 time periods in the analysis – and used the results to estimate the ‘hedonic’ value of land. In effect, I compared what buyers are actually willing to pay for a slightly smaller or larger residential site.
This analysis indicates that the underlying demand for residential land has increased in Auckland, reflecting growth in incomes, increasing preferences to live in cities, and improving urban amenities. If we also take this into account, it suggests that Auckland City house prices should have risen to around $780,000 over this period, as shown by the grey line on the following chart.
However, $780,000 is a lot less than $1,360,000, which means that increasing scarcity of housing in Auckland has played a large role in rising prices. The final chart suggests that our inability to build enough new housing in the right places has driven more than half of the house price increases that Auckland City has experienced over the last generation.
Finally, it’s important to note that many places have not experienced the same trends as Auckland. For instance, here’s the same chart for Whangarei District, where average house prices rose from around $160,000 to $400,000. Here, the role of improvements to buildings is proportionately larger, and the impact of increasing scarcity of housing is considerably smaller.
In fact, if you look around, Auckland seems like a major outlier: it has experienced larger scarcity-driven price increases than anywhere else. The following table shows how much of an impact a lack of housing has had on prices in different regions. (I’ve divided NZ into 22 areas that roughly correspond to commuting zones around major cities, plus rural areas.)
Prices in Auckland would be higher than elsewhere in the country even if housing was more abundant – people value living here! – but the gap would be much smaller than it is today.
This naturally leads on to the question of what else would have happened as a result. Would more people live in Auckland? Would fewer live in Whangarei? What impacts would that have on economic productivity?
Next week: Spatial equilibrium!