Land price & house size jumps: How NZ’s 1989 tax experiment ignored the potential impacts on land values, house sizes and property prices as govt chased a short-term revenue kick but doomed future generations

By Alex Tarrant

Why are land values in New Zealand so high? Why are we building oversized houses? And why has this been happening for so long?

Interest.co.nz last week ran a presentation by economist Shamubeel Eaqub on why housing in New Zealand had become so expensive and change would be so hard. Eaqub railed against failures by successive governments as policy actions (or inaction) over the past 30 years as dooming those following the Baby Boomers into ‘Generation Rent’.

There was a timely follow up this week. It should be mandatory reading for every politician, Treasury boffin (and anybody wanting to enter one of these professions), anyone trying to buy a house, and every homeowner who can’t understand why the young are complaining about not being able to buy property.

Motu economist Andrew Coleman on Wednesday released a paper titled Housing, the ‘Great Income Tax Experiment’ and the intergenerational consequences of the lease.

It focussed on one policy action that might just have contributed most to dooming generations of younger New Zealanders to renting longer and requiring more time to save for a house deposit than their Boomer parents, and being faced with having to pay up to buy oversized houses built by their Boomer parents.

Before we go further, Coleman did point out that the Great Income Tax Experiment of 1989 has been just one of the many drivers of property price rises and house size increases since 1990. He also said that although the outcomes match the theory of what would be expected by the tax change, other changes in 1989 – such as the introduction of the Reserve Bank Act reducing inflation – meant that it cannot be seen in the data which change can be attributed to which factor.

But he warned that in 25 years’ time, when we might have fixed every other problem addressing New Zealand’s residential land supply woes, one change still required will stick out like a sore thumb. It needs to be dealt with, and there are ways to do this. Will our politicians listen and act?

New Zealand’s Great Income Tax Experiment of 1989 – the basic facts

  • New Zealand new house sizes grew faster than in the US and Australia after 1990, without a much bigger rise in incomes. From an average 110sqm in 1974 to about 200sqm today.
  • New Zealand had the largest increase in house prices out of the 24 OECD countries monitored by the Dallas Federal Reserve between 1990 and 2016.
  • From 1975-1990 New Zealand had the third lowest increase in house prices. Now this was partly because 1975 was a peak year for New Zealand. “But we weren’t always the fastest increases in house prices in the world.”
  • Since 1990 we are the biggest – in fact bigger than by about 25% than the second-highest OECD country.
  • Tax changes introduced in 1989 created one of the biggest tax system distortions in the OECD between how property and sanctioned savings schemes were taxed.

Tax system distortion

To start with, here are the slides – they’re a quicker way of going through it all. The below is based on Coleman’s Motu Public Policy Seminar in Wellington on Wednesday.

In 1989, the Lange Labour government, based on a review carried out by Don Brash, changed New Zealand’s tax settings. Until then, New Zealand was like most other OECD countries in that it applied an expenditure tax treatment on savings placed in sanctioned savings schemes (ie what KiwiSaver is today). 

The change was that we scrapped NZ’s ‘exempt exempt tax’ system (EET), where income was exempt from tax when earned then saved, exempt when it accumulated and earned interest and dividends, but was taxed at a person’s marginal rate when withdrawn and spent.

We replaced that with income tax treatment, so it would match other forms of savings, except owner-occupied housing. This meant there was a distortion created in our tax settings.

Here are some of the countries that do what New Zealand did up until 1989:

There’s the United States, Canada, Mexico, Japan, South Korea, Austria, Belgium, Finland, France, Germany, Greece, Ireland, Iceland, Norway, Netherlands, Portugal, Poland, Spain, Switzerland, Turkey, the United Kingdom. And Hungary has a very similar way. Furthermore, Denmark, Italy and Sweden exempt tax from when it is saved and put into a retirement savings account.

So we used to be part of the mainstream OECD approach to expenditure taxes. But the Lange government needed revenue, and the change brought forward revenue into the government’s coffers.

An experiment that went horribly wrong

In a new paper titled Housing, the ‘Great Income Tax Experiment’ and the intergenerational consequences of the lease, Economist Andrew Coleman claims that was an experiment that went horribly wrong.

“New Zealand has one of the most distortionary tax environments for housing in the OECD. Most other countries tax all other assets – if they’re in a sanctioned retirement scheme – on an expenditure basis, the same basis as housing. But New Zealand does not.”

If you wanted to close the wedge properly, you would tax everything on an income tax basis. But that’s not what happened. We taxed sanctioned retirement funds on an income tax basis, along with all other assets, but we didn’t change the taxation for owner-occupied housing.

The paper discusses the implications for the size of new houses and the price of land. And there is a fix. Not by raising the taxes on housing, which we’ve been unable to do politically, but by reducing the taxation of retirement savings.

“Really, I don’t do much in this paper other than apply standard tax theory.”

Tax distortions created bigger houses, fewer term deposits

If you had a neutral tax system, it wouldn’t make any difference to the size or quality of the house that you choose.

A neutral income tax system, would mean we taxed imputed rent at income tax rates – the implicit rent you pay yourself for living in your home. We would also have an accrual-based capital gains tax system, we’d make an allowance for depreciation (because houses wear out) and for interest payments, and we’d take into account local property taxes.

Incidentally, this is the sort of thing Gareth Morgan’s talking about right now.

So how does this affect house sizes? Suppose you had $10,000, Coleman says. You could either build a bigger bathroom or some extra closet space or a third garage, lend it to the bank (ie a term deposit), or invest it in shares.

“If you lend it to the bank, you get taxed on it. If you put an extra addition to your house, or you buy a bigger house, you’re not taxed on it.”

A neutral income tax system, this decade, would have meant you would require annual benefits from the extra bathroom of $550 a year (5.5%). If you could get more than that by putting the money in the bank, then you’d do that.

But we don’t have a neutral income tax system. Under the current system, because we tax lending (term deposits in the bank) but we don’t have a tax system which is neutral towards imputed rents, you would only need a $460 annual return from the bathroom to make it more worthwhile than putting the money in the bank.

“The current tax system provides an incentive to accept a return from housing that’s about 20% less than under a neutral tax system.”

“And this could lead to an incentive to build houses that are 20% better – 20% bigger – than you would under a neutral system. You don’t need so much return from having that extra closet, that extra bathroom, that extra garage, as you would otherwise, because of the tax system.”

Now, this could be fixed largely by a tax on imputed rents, if the inflation rate was the same as the depreciation rate, which is that’s the case in New Zealand. We don’t have one of those taxes.

Land prices – why are they so high?

Moving on to land prices. New Zealand’s tax system incentivises people to buy better located land when a neutral tax system would not, Coleman says. He introduces the key concept of ‘convenience yield’.

This is the annual benefit from being closer to desirable amenities or jobs. For example, you might be willing to pay $10,000 extra a year to live in Oriental Parade rather than Upper Hutt because the Parade is closer to things you like to do and to your job.

Convenience yield depends on the cost of transport to those amenities. So if there’s a great train line from Upper Hutt into town you might not be willing to pay as much extra to live in Oriental Bay.

How much more you would pay to buy that Oriental Bay section to provide $10,000 of benefits a year is inversely related to interest rates. So if interest rates were 5% you would pay $200,000 to get a $10,000 benefit.

This would be the captialised land value under a neutral tax system. “The problem is, we don’t have a neutral tax system – we don’t have a neutral income tax system.”

With a non-neutral tax system, the value will be capitalised to a different amount. And the amount that we capitalise depends on interest rates, the way we tax things and the expected appreciation of land prices.

“Even if you think that property is going to go up at 1% a year, you will pay approximately twice as much for conveniently located land under our current tax system as you would under a neutral tax system,” Coleman says.

“This premium is going to exist even if we build lots of houses up at Upper Hutt, or in South Auckland. Even if we get rid of current supply problems, our tax system is still going to give an incentive to bid up the price of conveniently located land.”

This could be partially fixed by a capital gains tax, but not by an imputed rent tax, which is inconvenient because it means that the solution is different for houses and land.

On rents, house prices and becoming a landlord

The distorted tax system creates incentives to become a landlord rather than lend money, Coleman says. “We have a distortionary income tax system – even though rent is taxed, capital gains are not.”

And this provides an incentive for landlords to reduce rent-to-price ratios. You can do this either by lowering rents, or bidding up the price of houses; Most of this seems to have occurred by bidding up the price of houses rather than lowering rents.

Fun fact: In 1989 there were 62,000 private sector landlords. There’s now 276,000, which suggests there has been some change in the way people are investing their money.

And that problem can be entirely fixed by an accrual-based capital gains tax on nominal capital gains.

Why should we care?

Future generations will be paying artificially high prices for land. They will be worse off than their parents.

“If you tax land at concessional rates relative to other assets, the price of land is likely to become artificially high. And if you have artificially high land prices, you induce an intergenerational transfer to existing land owners and away from all future generations of land owners,” Coleman says.

“They are paying artificially high prices for land, and that stops them from earning other income from that money or having a good time from that money, if they wanted to.”

What’s worse, the economics show “this is likely to reduce the local ownership of the capital stock and wealth.”

Translation: Ongoing current account deficits. Sound familiar?

Well, “that’s what happens if you have a tax system that causes artificially high land prices.”

What to do?

“I think this is hardly a scheme that you would willingly introduce if you could avoid it,” Coleman says.

The Great Income Tax Experiment has created incentives to create bigger houses and to bid up the price of land. And the biggest beneficiaries were the Baby Boomers (Coleman himself is one of them).

Reforms should allow for an intergenerational transfer, he said. There are three possible solutions:

One, do nothing “and keep a tax system that’s particularly distortionary on housing markets.”

The second is creating a political consensus to tax imputed rents and capital gains on an accrual basis. However, this is “something that seems to be an extremely difficult task. Even if we did introduce it, it’s not clear we could sustain it.”

The third option is to move back to an EET retirement tax system, like the rest of the world. And this is the option Coleman is now throwing his weight behind, due to the political impossibilities of option two.

“This last option isn’t actually easy. But it does have proven international durability – it’s done in 25 countries around the world. I also say, it’s got some intellectual respectability.”

He says it’s time New Zealand reconsidered the Great Income Tax Experiment of 1989, and thought again about whether that wedge between the ways we tax assets inside retirement savings accounts and outside retirement savings accounts was worth doing,

“Because it has inadvertent consequence for the price of land, which may be causing worse problems than the problems that we’re trying to solve.

“Even if we fix up the supply of land, which I would encourage, the tax system is still going to create artificially high land prices, and it’s still going to impose costs on younger people and future generations. We need to fix both problems.”


Footnote: Coleman better not hold his breath.

Interest.co.nz understands that, while the dichotomy between the way property and retirement savings are taxed is something on the government’s radar, Budget 2017 isn’t likely to deal with it (as things stand today).