Speech to a housing conference in Wellington highlights the high social and economic costs of restricting the availability of residential land in major New Zealand cities
Ladies and Gentlemen,
We all know that over the last few years the price of houses has risen enormously in almost all developed countries.
We all know that in New Zealand house prices in our main urban areas are now among the most unaffordable in the world, relative to household incomes.
And we’ve all heard the multitude of reasons produced to explain this situation: greatly reduced interest rates (so that borrowers can afford to borrow more than previously); low levels of unemployment (so that people feel more comfortable taking on large amounts of debt); financial deregulation (so that banks have been driven to compete for borrowers more aggressively than before); tax laws which favour investment in residential property; and, in some countries, high levels of net inwards migration.
And there’s no doubt that all of those factors have, to varying degrees, added to the demand for houses.
But an increase in demand for houses will only cause an increase in the price of houses if there’s no commensurate increase in the supply of houses.
Of course, in the short term it’s not easy to increase the supply of houses. So if, for example, there’s a sudden surge in net inwards migration, it wouldn’t be in the least surprising to see the price of houses increase until developers and builders are able to respond to the increased demand by sub-dividing more land and building more houses.
And it’s also true that a downwards adjustment in the level of interest rates will tend to lead to an increase in the price of all assets.
Similarly, more aggressive lending behaviour by banks brought about by financial deregulation will lead to an increase in the price of many assets.
But none of these factors seems adequate to explain what’s happened in many developed countries in recent years.
We could reasonably expect that, after a year or two, developers and builders would respond to an increase in demand for housing arising from a surge in net inwards migration by increasing the supply, with a tendency for house prices to return to their previous level.
We could expect to see a step increase in the level of house prices as a result of lower interest rates and more aggressive lending behaviour by banks, not steady increases, year after year, to the point where houses have become severely unaffordable in a great many urban areas.
Something else has been going on to produce these very large increases, extending over a decade and more in some countries.
I believe the annual Demographia surveys point to that “something else”. What all those surveys strongly indicate – including the most recent published just two months ago and covering 227 urban areas across six English-speaking countries – is that houses are affordable where local governments apply a light-handed approach to the zoning of land for residential development, and they are unaffordable where local governments take a highly restrictive approach to the zoning of land for residential development.
Those who, for whatever reason, disagree with this conclusion have come up with all kinds of nonsense in an attempt to discredit the Demographia surveys.
To me, US experience provides the most conclusive evidence in support of the view that it is overwhelmingly the degree to which local governments have restricted the availability of residential land which explains the rapid escalation of house prices in recent years. That country provided more than half the urban markets covered in the latest Demographia survey – 129 out of 227. In many of those urban areas, 46 to be exact, housing is eminently affordable, with the median house price no more than three times the median household income in that city; in some cases, this despite very large population in the urban area surveyed and rapid population growth (cities such as Atlanta, Dallas-Fort Worth, and Houston, cities which are among the fastest growing major urban areas in the developed world). At the other end of the affordability range, with the median house costing more (and in many cases much more) than six times the median household income in that city, were 18 urban areas (cities such as Los Angeles, San Francisco, New York and Honolulu).
And yet all of these cities – those with affordable housing and those with extremely unaffordable housing – operate within the same country: their interest rates are the same; their tax laws are closely similar; the extent of financial deregulation is essentially identical; and people are free to move from one area to another.
What the authors of the Demographia survey found was that the main explanation for the difference in the affordability of housing between these different urban areas within the same country was the severity of the restrictions on the availability of residential land. It’s these restrictions which have caused the price of residential land to escalate from around a quarter of the price of house and section together as it once was to the point where in many highly restrictive markets the price of land is half the price of house and section combined, and in some cases more than half. It’s not the cost of constructing a house which has escalated enormously in recent years – though construction costs have gone up, no doubt in large part as a result of increasingly bureaucratic consent processes which must be endured before construction can even begin – but the cost of the land that the houses sit on which has escalated.1
And if confirmation of this was needed, it was surely provided by Arthur Grimes and Yun Liang in their study of land prices in Auckland, published in the middle of last year. This quite remarkable study looked at the influence on the price of land of the Metropolitan Urban Limit imposed by the Auckland Regional Council. As most of you will know, this Limit requires new residential subdivisions to be undertaken within an area defined by the ARC. What they found, after controlling for all sorts of other influences on the price of land, is that the price of land just inside the Metropolitan Urban Limit was between eight and 13 times the price of land immediately outside the Limit.2 And of course, the price of land just outside the Metropolitan Urban Limit is itself pushed up by speculation that, eventually, the Auckland Regional Council will be obliged to extend the Limit beyond its current boundaries, taking in land which is at present just outside the Limit.
We should be getting “starter sections” on the fringes of our major cities for between $30,000 and $40,000, not the $200,000 to $300,000 nonsense which is all too common.
The same study by Grimes and Liang cited a number of academic studies conducted in the US which show a very consistent pattern: limiting the availability of land for residential development inevitably results in higher housing prices.
So that’s clearly the first implication of restricting the availability of residential land: doing that greatly reduces the affordability of housing, with all the consequential effects in terms of a decline in the proportion of families who can own their own homes, with that effect almost certainly having a disproportionately large impact on low income New Zealanders, many of them Maori or Pacific Islanders.
But I’ve come to believe that restricting the availability of land for residential development has consequences which go far beyond its effect on the number of people who can own their own homes. The rapid increase in the price of houses has almost certainly had far-reaching economic consequences as well.
It’s always dangerous to speculate on how things might have been different “if only”. If only the Archduke Ferdinand had not been assassinated in Sarajevo in June 1914! If only Hitler had been assassinated in July 1944! If only the Auckland Regional Council, or the Canterbury Regional Council, or the Wellington Regional Council, had been less restrictive in their approach to the zoning of residential land.
But let me speculate a little.
We know that as house prices have increased strongly over the last 20 years, and especially over the last decade, New Zealanders have felt very much wealthier and have changed from being modestly frugal to being, on average, net dis-savers – indeed, probably the champion dis-savers in the world!
We know that this has perpetuated a situation where, as a country, we are heavily dependent on the savings of others, with that dependence having become markedly greater in the last few years.
In September last year, the New Zealand banking system had made loans totaling $273 billion to New Zealand-based borrowers, but had deposits totaling only $175 billion from New Zealand-based depositors, leaving a funding gap to be borrowed from overseas of $98 billion.3 So much for the argument that New Zealand interest rates are in some sense “too high”!
Or put it another way: between 1990 and 2007, deposits from New Zealand households grew at an average rate of 7% per annum, while home loans made by the banking system grew at 14% per annum over the same period. The difference was, to a large extent, financed by borrowing the savings of foreigners.
We have supported our living standards by borrowing from foreigners to the point where we are, relative to the size of our economy, more heavily indebted to foreign residents than any other country in the world. Once upon a time, the New Zealand government borrowed overseas to fund its spending. Now, the New Zealand government has no foreign debt expressed in foreign currency, and very little foreign debt even denominated in New Zealand dollars. The overwhelming bulk of the obligations owed by New Zealanders to foreign residents is owed by the private sector, and a great deal of that is owed by ordinary New Zealanders, through the banking system, to fund a bubble in house prices and a level of consumer spending which we have not earned. This has been facilitated by the restriction on the availability of residential land.
The rapid escalation in house prices fuelled by restrictions on the availability of residential land has also had an impact on the conduct of monetary policy. As the Reserve Bank noted in its submission to the Finance and Expenditure Committee of Parliament last year, rising house prices don’t in themselves cause inflation. But the Bank noted that once house prices start rising, whether because of increased net immigration or for some other reason, restrictions on the availability of residential land create a momentum for further increases in prices, which require tighter monetary policy to contain. So all borrowers end up paying a price in the form of higher interest rates, and all exporters and their staff pay a price to the extent that the exchange rate is pushed to levels well above what the fundamentals justify.
Pushing up the price of housing through restricting the supply of residential land almost certainly has another effect on economic growth. Not only has this encouraged New Zealanders to spend more than they earn, financing the gap by borrowing from foreigners, it has almost certainly encouraged more of the total available savings to be channeled into housing, with less going into more directly productive assets. While home loans from the banking system grew at 14% annually between 1990 and 2007, as I’ve noted, other loans to New Zealand borrowers – mainly New Zealand businesses and farms – grew by just 6% annually.
In the United States, the increase in house prices to the point where in many major urban markets they have become severely unaffordable encouraged some grossly imprudent lending practices, with banks and other lenders willing to lend on the flimsiest of security and poor loan-servicing capacity because of a naïve belief that they would be saved from loss by a continuing escalation of house prices above the rate at which incomes are growing. The unwinding of this situation is having a serious impact on the economic growth of the United States, and potentially of the whole world.
It is often argued that the restrictions which local governments have placed on the development of land for housing have at least reduced the total area of New Zealand devoted to urban development (as if this were some kind of desirable outcome in itself). But assuming that 80% of our population is in some sense “urbanized”, and assuming an average urban population density of 1900 per square kilometre (which is the density in Christchurch – both Auckland and Wellington figures are higher), only 0.7% of our total land area is urbanized at the present time. If New Zealand’s population grows by 50,000 a year, and we housed all these additional people in fringe housing around our urban areas at the population density of Christchurch (requiring roughly 26 square kilometres, or one hundredth of 1 per cent of our total land area, each year), we couldn’t urbanize a further half per cent of our total land area over the next 50 years if we tried!
And does the refusal of local governments to zone more land for residential purposes on the outskirts of our major cities actually reduce the land devoted to housing anyway? Probably not – it may just increase the amount of commuting which people have to do. Instead of being able to live on the outskirts of our larger cities, they may be forced into satellite towns some miles beyond the artificial boundaries of our cities – in Rolleston and West Melton in Canterbury, in Pukekohe and Warkworth in Auckland.
But let’s suppose the tight restrictions which local and regional governments have imposed have actually managed to reduce the land devoted to housing in New Zealand. Let’s suppose that without tight restrictions the area of New Zealand devoted to urban development would already be 1% of our land area. Yes, it is possible we may have needed to invest a little more in transport and other infrastructure.
But I haven’t the slightest doubt that that infrastructure could’ve been financed without difficulty if we’d charged appropriately for those services. And the consequences of that alternative policy would’ve been vastly less damaging – for the ability of young New Zealanders to buy their first home, for New Zealand’s dependence on the savings of foreign residents, for the interest rates paid by all borrowers, for exporters, and for our standard of living – than what we’ve experienced.
The policy of tightly restricting the availability of residential land, a policy followed by most of the main regional and urban local governments in New Zealand, has been an unmitigated disaster from almost every standpoint.
1 Hugh Pavletich has drawn my attention to a newspaper advertisement for starter homes in Woolston, a low income suburb of Christchurch, advertised at a price of between $399,000 and $445,000. Assuming that the houses are in the order of 200 square metres in area, and construction costs are around $800 per square metre, this implies a cost for each 450 square metre section of between $239,000 and $285,000, with the price of the land being over 60% of the combined price of house and land.
2 Spatial Determinants of Land Prices: Does the Metropolitan Urban Limit Have an Effect?, Motu Economic and Public Policy Research, August 2007.
3 Reserve Bank of New Zealand data.