A summary of some of Peter Tulip's papers

Below is a non-technical summary and discussion of some of Peter Tulip's key papers.

[This note may contain errors and inaccuracies. It was mostly written just for myself, in preparation for my live podcast conversation with Peter Tulip.]

Peter Tulip is an Australian economist who has held senior research roles at the Reserve Bank of Australia and the US Federal Reserve. He's currently Chief Economist at the Centre for Independent Studies.

His research focuses on housing and monetary policy. He's one of Australia's leading experts on the economics of housing.

Two papers on the effects of planning restrictions

(1) 'The Effect of Zoning on Housing Prices', Ross Kendall and Peter Tulip (2018).

In this paper, Kendall and Tulip attempt to quantify the effect of 'zoning'* on housing prices in Australia’s four largest cities: Sydney, Melbourne, Brisbane and Perth.

According to their estimates, zoning restrictions contribute significantly to the average price of detached houses in those cities. Here are zoning's contributions:

• Sydney: 42%
• Melbourne: 41%
• Brisbane: 29%
• Perth: 35%

It's important to note that this does not necessarily imply that, if we removed zoning restrictions, prices would fall, e.g., 42% in Sydney. The paper presents only a partial equilibrium model and doesn't estimate how supply and demand would respond if zoning was removed. But it is indicative.

How do Kendall and Tulip arrive at their estimates?

They follow the approach pioneered by Ed Glaeser and Joe Gyourko (and used by many researchers around the world). I'll sketch this approach below.

First, assume that if a housing market was healthy—that is, if supply was allowed to respond readily to demand—then over the long run, house prices should more or less equal the sum of housing's two inputs: land and buildings.

(This assumption arises from noticing that if the price of a good or service is above the cost of producing it, that will attract more producers into the market due to the potential for profit, thereby increasing supply and pushing the price down. Conversely, if the price is below production cost, that will cause producers to exit the market, reducing supply and pushing the price up.)

Now, the empirical work begins. Kendall and Tulip estimate land value by looking at what homebuyers will pay for a slightly larger block. Imagine two otherwise identical dwellings—in the same location, with the same number of bedrooms, bathrooms, etc—except one has slightly more land. Kendall and Tulip use sales price data to see how much homebuyers pay for that extra yardage. They call this the physical value of land (because it doesn’t come with the right to build an extra dwelling on it).

Next, they work out the costs of building houses or apartments on that land, to calculate construction costs.

Finally, they add together land value and construction costs. If the sum is less than observed house prices, then the difference is assumed to be due to zoning restrictions.

A critical point to note here is that the zoning effect is treated as a 'residual'—that is, whatever chunk of the average price is left over after accounting for the cost of the structure and the value of the land, is assumed to be induced by zoning.

This means it's very important how researchers like Kendall and Tulip calculate those costs. If they underestimate them, the zoning tax will appear larger than it really is; if they overestimate them, it'll appear smaller than it really is.

Kendall and Tulip address some of these concerns on pages 14-15.

(2) 'The Apartment Shortage', Keaton Jenner and Peter Tulip (2020).

This paper is probably the most important of Peter's housing research papers.

Its primary contribution is to quantify the effect of zoning restrictions on apartment prices (in contrast to the above paper, which examines their impact on detached housing prices).

Why is this paper arguably more important than the one above? Because apartments—not single-family houses—are central to policy debates about boosting supply. As Jenner and Tulip put it:

"Whereas estimates of the zoning effect for houses answer the question: why is housing so expensive?; estimates for apartments are more relevant to the question: what do we do about it?" (p1)

On Jenner and Tulip's analysis, it's clear that Sydney has the worst apartment shortage of any major city in Australia.

The textbook definition of 'shortage' used here is: 'excess demand', or the gap between price and marginal social cost. (As with the above paper, it is argued that this gap must be due to planning restrictions.)

So what did Jenner and Tulip find? How much were zoning restrictions contributing to the average prices of apartments? Here are their estimates (p6):

• Sydney: 41%
• Melbourne: 16%
• Brisbane: 2%

The paper highlights another interesting result for policy debates: high rises are a much less costly way of increasing supply than medium density (p25). This finding runs counter to the YIMBY focus on the 'missing middle' in the middle ring of our major cities, though obviously it's just one consideration.

Some other papers on the housing market

(3) ‘Is Housing Overvalued?’, Ryan Fox and Peter Tulip (2014).

In this paper, Ryan Fox and Peter Tulip present a test for determining whether housing is overvalued and apply it to the Australian market. This is obviously relevant to debates about bubbles.

Here is the test:

"We assess houses as 'overvalued' if home buyers pay too much, in the sense that they would be better off renting than buying." (p1)

More precisely, the test compares rentals yields with the 'user cost' of housing.

The user cost attempts to capture how much it actually costs (on an annual basis) to own a home.

The annual cost of owning a home can be written as:

Cost (in dollars) = P (r + c + s + d – π).

Where:

• P represents the price of the property
• r = the real interest rate
• c = other running costs (such as repairs, rates and insurance)
• s = buying and selling costs (stamp duty, agent commissions, etc)
• d = the physical depreciation rate
• π = the expected real appreciation rate of the property on a constant-quality basis

(To express the annual cost as a percentage of the purchase price, divide the left-hand side by P.)

Fox and Tulip argue that the user-cost-versus-rental-yield measure of overvaluation is superior to other popular measures like the price-to-income ratio or the price-to-rent ratio. They argue that it's not obvious that individuals actually consider price-to-income or price-to-rent ratios when deciding whether to buy. People ultimately need to live somewhere, whether that's a rental or their own home. And the relevant cost of owning a home isn't the sticker price that gets decided at an auction; it's the 'user cost'.

Applying their test to past prices, do they find any periods of clear overvaluation, i.e. bubbles?

Yes, they detect strong overvaluation in the boom period from 2002 to 2003.

As a reminder: in the two years to December 2003, real houses prices grew at an astonishing annual rate of about 13%.

According to Fox and Tulip's model, in 2003, home buyers were acting as though they expected real appreciation of almost 4% per year—which was noticeably above the historical average (p18). The 2002-2003 boom is therefore largely attributable to expectations of further capital appreciation.**

Fox and Tulip do not, however, detect a bubble at the time of the paper's publication in 2014. Nor would their test—to my understanding—detect a bubble today.

In Tulip's thinking, if the expected capital gain at which homebuyers would break even is close to its long run historical average, then there is no bubble. House prices are extremely high, but this is a rational valuation of a scarce asset. Whether or not you think the historical average is too high or too low hinges on whether you think zoning restrictions will get tighter or looser—and that requires forecasting political decisions.

(4) 'A Model of the Australian Housing Market', Trent Saunders and Peter Tulip (2019).

This paper presents an empirical model of the Australian housing market that combines a number of key relationships (see page 1 for those relationships).

Trent Saunders and Peter Tulip find two demand-side changes have been especially important in explaining the run-up in Australian house prices:

1. Much of the growth in prices to 2019 (the year the paper was published) was driven by falling interest rates.
2. The immigration boom of the mid 2000s also helps explain the growth in the late 2000s.

Some people may wonder: If demand has been driving changes in house prices, how do we reconcile this with Peter's papers on the effects of planning restrictions?

Econ101 answers this question: prices result from the intersection of supply and demand. (As Alfred Marshall put it, asking whether supply or demand determines a price is like asking which blade of a pair of scissors does the cutting.)

Saunders and Tulip essentially argue that it's useful to conceptualise the Australian housing market as having a near-vertical supply curve with fluctuating demand. If supply is inelastic, variations in demand (due to immigration or interest rates, for example) will cause changes in price:

"In terms of a conventional supply-demand diagram, zoning restrictions make the supply curve steep, while lower interest rates (or other factors) shift the demand curve out. It is the interaction of rising demand with inelastic supply that explains higher prices." (p23)

There's a distinction between what's been driving prices historically and what would lower prices counterfactually. In other words, how you arrived at a higher price level doesn't dictate that you attempt to bring prices down the same way. The cause and the cure can be different.

For Tulip, the key policy question is: Why doesn’t supply respond? (His answer to that is, of course, that planning restrictions artificially constrain housing supply.)

Herein lies the central problem in debates over housing policy. So often these debates feel like people are talking past each other. That's rooted in the fact that prices are a function of the interaction of supply AND demand. Thus people who dislike immigration, for example, will never miss an opportunity to tell you about how it's driving up prices—and in a narrow sense they're right.

On the other side, economists like Peter Tulip will argue that increases in demand only push up prices insofar as supply is inelastic, and that the welfare-maximising policy solution is not to reduce demand (which arguably also reduces living standards), but to allow supply to be more elastic.

Also of note: this paper originates Tulip and Saunders's famous rule of thumb:

"[E]very 1 per cent increase in the number of dwellings (when driven by an increase in supply) lowers the cost of housing by 2½ per cent." (p28)

(Note that this effect is small, meaning that we need BIG increases in supply to solve the problem.)

Some summaries by Peter himself

Since leaving the RBA in 2020, Peter has increasingly occupied a public education role. Below are some highlights from a couple of his summaries/overviews of the housing debate. I found these papers to be incredibly clear and helpful overall. To the final paper, I also add some of my own commentary.

(5) 'Misunderstandings about Planning Restrictions', Peter Tulip (2021).

Page 11 contains a neat response to people who object that there's no housing shortage because, for example, housing completions have exceeded household formation over certain periods. Peter notes that the relevant quantity is the stock of housing, not its flows:

"A more fundamental problem is that arguments like these apply to changes in the housing shortage when the problem is the level. Rapid growth in supply, relative to changes in population or the number of households or construction in other countries, implies that conditions are improving and the housing shortage is decreasing – it does not imply that the stock of housing is adequate. That should be judged by whether price is close to marginal social cost. Essentially, in response to arguments that the stock of housing is inadequate, these objections amount to “but the flow is fine”. Perhaps, but beside the point."

This is just one paragraph that stood out to me on my most recent reading of this paper. Others may find much else of interest in the paper.

(6) 'Housing Affordability and Supply Restrictions', Peter Tulip (2024).

Page 6 contains a neat rebuttal to the concern that building 'luxury housing' doesn't reduce prices across the board:

"Newly-constructed housing is often more expensive than old housing, giving rise to worries that ‘luxury’ construction will primarily benefit the wealthy. However, this concern is misplaced, because it does not consider the more important indirect ‘filtering’ or ‘cascade’ effects of increased supply on existing properties.

When wealthy residents occupy new housing, they vacate other housing, which falls in price. Those vacancies are filled by those on moderate incomes, who vacate other housing. That increases supply and lowers prices for those on lower incomes. And so on...

A shortage of housing is like a game of musical chairs. Regardless of who is playing or the quality of the chairs, if there are not enough then the weakest will miss out. If supply is abundant, those at the bottom benefit most."

The central problem of NIMBYism is that the (perceived) costs of new development are concentrated locally whereas their benefits are dispersed societally. Peter explains on page 8:

"If one considered that opposition to new development was widespread, society would face a trade-off. Elected representatives would need to weigh the desire of wealthy neighbours for easy parking and aversion to shadows against potential residents’ need for shelter.

Current institutional arrangements make this trade-off by giving those wealthy neighbours a say, via their local councils, while the views of potential residents from outside the area are ignored. State and federal politicians representing wider electorates would place more weight on the latter group and make different trade-offs. Broader social welfare is advanced by taking the decision to restrict housing away from local representatives."

This inherent asymmetry gives rise to a free-rider problem:

"[H]ousing affordability can be seen as a public good, subject to a free-rider problem. There is little that one council, acting alone can achieve. But if all councils allow more building, housing costs will fall substantially. So, it is rational for individual councils to contribute if, and only if, other councils are also required to contribute." (p13)

Obviously resolving this free-rider problem requires coordination at state and federal levels.

I knew that Peter wanted to make federal grants and expenditure on infrastructure conditional on extra housing.

Until recently, I wasn't clear on the precise policy reforms he wants to see at the national level. Page 12 helped me understand the key one:

"The federal government has allocated $3.5 billion in incentive payments for more housing (Albanese, 2023). This encourages more building, helps alleviate infrastructure bottlenecks and amplifies the message that more housing is needed. Reform of the Grants Commission, to direct more funds to those states that build more housing, would reinforce this. This could be achieved by revising the Grants Commission formulae to treat housing construction as a disability, in the same manner as transport spending. This would be fair, given that more housing requires more capital outlays on public infrastructure (above that paid for by infrastructure charges on developers)."

On page 14, Peter summarises what needs to happen to solve the housing crisis and finishes with a call for a "change in social values":

"Planning restrictions mean housing prices and rents are too high. The solution is for state and local governments to stop saying ‘no’ and start saying ‘yes’. Setting and enforcing high targets for local councils would help achieve this. Making federal grants and expenditure on infrastructure conditional on extra housing would also help.

More fundamentally, we need a change in social values. As a society, we need to be more accepting of higher density. We need to put more weight on the interests of renters and future home buyers and less weight on the interests of nearby residents."

Now I offer a critique: this massively understates the social changes that will be needed.***

Moving from a regime of high house prices to a new equilibrium of much lower prices (which I understand is what Peter wants—not merely slower growth, but falling prices) will exact enormous social and economic pain. The mere prospect of that pain will pose tremendous cultural and political challenges.

On this pain, in December 2024, I noted on X/Twitter that to be a more effective movement YIMBYs should:

"Recognise the narrowness of the path we must walk. In countries like Australia and Canada, households are massively leveraged, and most of that is mortgage debt. If asset prices fall too quickly against fixed nominal debt, balance sheets are wrecked and suddenly you're the US circa 2008."

But macroeconomic risks are the least of it. The first-order consequences of falling prices will be immense. The expectation that home equity will increase over one's lifetime is deeply entrenched in Australian society, and many people have planned their lives around it. Consider:

• Australia's home-ownership rate sits at around 65%. There are about 6.2 million households—therefore probably at least 10 million Australians—who own their own homes.
• The total value of the residential real estate market is about $11 trillion—or nearly triple the value of the entire pool of superannuation funds. Most Australians view their home as a nest eggs—it's core to their wealth-building and retirement plans. Tax concessions support this—primary residences are not subject to capital gain taxes nor are they included in the pension assets test.
• About 15% of Australian taxpayers—or 2.2 million people—own at least one investment property, and about half of those taxpayers are negatively geared, meaning they are very likely to be investing only for the capital appreciation.
• In the 2020-21 tax year, Australian real estate was the largest source of realised capital gains, accounting for at least 40 percent of net gains. This has been true for many years. In Australia, property is the most reliable way to get rich.

Reaching Peter's new equilibrium (which, for the record, is the equilibrium we should all want) will not only require enormous sacrifices—arguably it requires political courage and a deep cultural adjustment.

As Saul Eslake has pointed out for a long time, the home-owning constituency dwarfs the first-home-buying constituency.

The last two decades of Australian housing policy have been bookended by a resolute commitment to high house prices by both sides of politics.

In 2003, Liberal prime minister John Howard famously remarked on ABC radio that:

“I haven’t had anybody shake their fist at me and say: ‘Howard, I’m angry with you for letting the value of my house increase’.”

In 2024, Labor housing minister Clare O'Neill (in)famously remarked on triple j radio that:

"We're not trying to bring down house prices... We want house prices to grow sustainably."

Regarding the cultural adjustment, in January 2024, I wrote a correspondence to Alan Kohler's Quarterly Essay. I wrote:

"[I]f solving our housing crisis requires abandoning the idea that property is a vehicle for building wealth, perhaps it also means embracing the notion that creating valuable ideas or companies is the most noble thing a citizen can do.

This would require a complete inversion of our national outlook. Under the tyranny of tall poppy syndrome, it’s as if property is the most excusable way to get rich in Australia: if you found a start-up, you’re long on yourself; but if you invest in property, you’re just long on Australia.

But we may not have a choice. For if pouring ever-larger piles of credit into unproductive assets is a sure-fire way of doing less with more, innovation has always meant doing more with less."

Obviously I was short on concrete solutions here. But I still think these are the right questions to ask.

Peter sweeps all of this under the rug.

It's as if, in his heart of hearts, he doesn't really believe substantial change is likely to happen, or to happen very soon, and so it's not even worth discussing what that transition would actually look like.****

I'd like to propose an idea: To start the conversation on how we get to the new equilibrium, we need something like the Accord but for housing.

*Traditionally, 'zoning' refers to regulations dictating how land can be used; zoning restrictions are therefore a subset of planning restrictions. But in this paper, as in much of the public conversation, 'zoning' is used conveniently to refer a broader class of regulations (closer to that which is called 'planning restrictions')—for example, minimum lot sizes, maximum building heights, and planning approval processes.

**This is consistent with other measures that detected a bubble during that period. It also accords with the concerns held by the RBA at the time. The bank raised rates twice in 2002, and twice again in 2003. It was the first and only OECD central bank to raise rates during this period. Former Governor Ian Macfarlane explained on my podcast in 2020: "I don't know whether we did it, but certainly house prices stopped rising... Certainly that's what we were trying to achieve, and it was a gentle landing, they didn't collapse...".

***I'm grateful to David Orsmond for encouraging me to think about this again recently.

****Tellingly, Peter's response to my December 2024 tweet was: "I agree that falls in nominal house prices are undesirable and unpopular. We are a long way from that. But when the YIMBY movement is 10 times as successful as it has been, it will need to slow down."