University of Otago's Andrew Coleman looks at the intergenerational effects of the tax system on New Zealand’s housing markets

The following article is the seventh chapter from Progressive thinking, ten perspectives on housing, a Public Service Association (PSA) publication. Interest.co.nz is publishing all 10 chapters from different authors on various aspects of housing.

By Andrew Coleman*

The tax system plays a crucial role in New Zealand’s housing markets. At the simplest level, the Goods and Services Tax is applied to new land development and new house construction, raising the price of new housing by 15%.

But the effects of the tax system are more complicated than this.

Since 1986 several tax changes have caused an intergenerational rift in New Zealand society by increasing the prices young people pay to purchase houses.

Some of these tax changes appear justifiable on efficiency grounds, but even these have made it more expensive for young people to purchase or rent property.

In conjunction with other tax changes that have artificially raised property prices, a generation of older property owners have become rich at the expense of current and future generations of New Zealanders.

The scale of the problem

The scale of the problem is seen by observing how average property prices have increased by over 220% in inflation-adjusted terms since 1989; the highest rate of increase in the developed world.¹

The average size of new houses has also increased more quickly than in Australia or the United States, the only two countries that publish this data.

The average size of a new dwelling in 2013 was 198 m2, up from 125 m2 in 1989, and nearly twice as large as the average new house in Europe.

The tax changes that have affected housing can be divided into those that affect the cost of supplying housing and those that affect the demand for housing.

Unfortunately, unravelling the effect of taxes on house prices and rents is challenging.

The effects depend on the extent that the supply of new housing is responsive to prices.

If the supply of housing is very responsive to prices, taxes that affect supply prices (such as GST) become fully reflected in prices, while taxes that affect demand (such as the relative size of taxes on housing income and other assets) do not. Conversely, if the supply of housing is not really responsive to prices, supply taxes like GST have little effect on prices but demand taxes have large effects.

The analysis is further complicated because the supply of land – particularly land in good locations – is less responsive to price than the supply of new houses.

It is quite possible that a particular tax can simultaneously lead to higher land prices but not much new land, and larger houses but not much of an increase in building costs.

Since 1989, the ways that the tax system affects the demand for housing has been the biggest problem.

The fundamental difficulty is that the returns from other classes of assets such as interest income are more heavily taxed than the returns from housing.

Because interest is more heavily taxed than the returns from owneroccupied housing – which are essentially the rent people get from their own home – people have an incentive to live in larger houses than otherwise, and pay more for well-located properties.

In the absence of this tax distortion, many people would choose to live in smaller houses and land prices in major cities would be a lot lower.

It is not unreasonable to suspect the premium people pay for well-located properties is twice as high as they would pay under a non-distortionary tax system.

Incentives

But this is not all. The tax system provides incentives for landlords to pay a much higher price/rent multiple for the houses they lease, largely because the absence of a capital gains tax.

Because the houseprice/ rent multiple could increase either because house prices increase or because rents decline (or some combination of both), the tax system could make buying more expensive or it could make renting more affordable. Most of the evidence suggests house prices have increased rather than rents have fallen; either way, the result is a tax-induced decline in home ownership rates.

When the tax system causes artificially high house prices, costs are imposed on current and future generations of young people, who have to borrow more and pay higher mortgage costs.

Why 1989? New Zealanders have never paid tax on the capital gains associated with house price increases, and the way housing is taxed was not fundamentally changed in 1989.

This is true. But the distortionary effects of taxation depend on the way houses are taxed relative to other asset classes, and in 1989 the government changed the way some other capital income is taxed.

Until 1989, money placed into retirement saving schemes was tax deductible, and the earnings from this money were not taxed as they accumulated.

Under this tax scheme – which is used in most developed countries including the United Kingdom, the United States, France, Germany, and Japan – the money placed in these savings schemes is taxed in a similar way to housing.

It reduces the incentive for owner-occupiers and landlords to overinvest in housing.

While the distortions in the current tax system could be eliminated by introducing a capital gains tax on housing and all other assets, and by taxing the rent you implicitly pay yourself when you own your home, most countries have found this too difficult to do.

As they have discovered, it is far simpler to change the way other savings are taxed.

On the supply side, in addition to GST, the Local Government Act (2002) has also affected the cost of supplying housing by changing taxes.

Instead of levying property taxes (rates) to fund the costs of developing new sections, local governments have progressively imposed development charges.

This change has improved efficiency by moving the costs of a larger city to the new people populating it, but it has also increased the price of housing right across cities.

People who bought before 2002 shifted the cost of new development to others, increasing the value of their houses, even though their development costs had been paid by other ratepayers.

For a long time, economists have pointed out that if you tax the income from housing less than other assets, you tend to increase land prices.

At the macroeconomic level, they have noted that this tends to increase national debt levels, and lower national income.²

The first owners of land benefit from these schemes, but everyone else loses.

Perhaps this is a reason why other countries have been concerned to tax housing on a similar basis to other assets.

It is unfortunate New Zealand does not do so, even if the tax changes implemented since the late 1980s have proved very advantageous to middle-aged and older generations.

¹Data on house prices is from the International House Price Database provided by the Federal Reserve Bank of Dallas. See Mack, A., and Martínez-García, E. (2011) “A Cross- Country Quarterly Database of Real House Prices: A Methodological Note.” Globalization and Monetary Policy Institute Working Paper no. 99 (Federal Reserve Bank of Dallas, December).

²Feldstein, M. (1977) “The surprising incidence of a tax on pure rent: a new answer to an old question.” Journal of Political Economy 85(2) 349 – 360.


*Dr Andrew Coleman has half-time positions at the University of Otago, where he teaches economics, and at the Productivity Commission. From 2008 – 2016 he researched the interactions between taxes, housing markets, and retirement schemes at Motu Economic and Public Policy Research and the New Zealand Treasury, and in 2011 he was a member of the Saving Working Group.

Note: The views expressed in Progressive thinking, ten perspectives on housing belong to the authors and do not necessarily represent the view of PSA members or the organisation.

The foreword is here.
The first chapter is here.
The second chapter is here.

The third chapter is here.
The fourth chapter is here.
The fifth chapter is here.

…and the sixth chapter is here.