Monday, November 16, 2009

Some empirical support for land economics

There has been a paper recently published by Andrew Leigh, Economics Professor at ANU, which empirically estimates the impact of stamp duties on the housing market. I found out about this paper through Chris Joye’s blog at Business Spectator, although I regularly read the Core Economics blog, where it was also linked with some added thoughts from Andrew. I mention this only because I have now been involved with discussions about the paper at these sites.

I need to quickly summarise Leigh’s findings before moving on to the important theoretical and political implications.

His main finding is that if stamp duties are raised, house prices will fall by more than in the increase in the tax. Did you get that? If you increase stamp duty, the total price of housing (price plus stamp duty) will fall. Sellers suffer, buyers benefit. It’s a classic land tax - there is no deadweight loss, as shown in the figure below.

How can such a thing occur? For any other product, assuming a competitive market, if you add costs to production, prices will have to go up (even if quantity sold goes down), or margins will go down (temporarily at least).

Land, however, has some characteristics that make it quite different from other goods
  1. There is a fixed supply (vertical supply curve), and
  2. It is costless to produce (the producer surplus starts at a price of zero)
Some would argue that land available to be developed is not in fixed supply, and that town planning regulations can change that supply. I agree. But these are regulations, they are not market players, and that does not make the supply of land price elastic (although I would suggest the supply curve for serviced residential lots above the intersection with demand is quite elastic as land parcels are brought to market). I think both sides would agree that from a theoretical standpoint, the supply curve is vertical below the intersection with the demand curve.

It is the second point that is far more important to understanding the land market. Land itself is costless to produce. That means that the level of demand determines the price of land at any point in time. Not supply, demand. So when you increase a tax on land the total land and tax price stays constant, but the underlying value of the land declines (as shown by the reduced producer surplus in the figure above).

I have been quite baffled by the success of Christopher Joye’s argument that the supply of housing is a major factor in determining prices. He maintains two contradictory positions. The first is that we have a land price boom, not a house price boom. The second is that we should elastify the supply of housing to avoid further unnecessary price increases. Hang on chap. We don’t have a problem supplying housing. Our problem is that we all decided to pay ridiculously high prices for land.

There are two more characteristics to the land market that make analysis difficult. There is competitive behaviour in the market for buying land, both development sites and serviced land parcels, but not a competitive market for the sale of land.

Once a serviced land parcel is developed, there is no price competitiveness exhibited when selling to the final consumer market. The problem with land is that if prices fall, they can gather momentum as people sell to avoid further falls. Also, if developers seek to undercut the market price, it reduces the value of all their other land holdings. There is no incentive to release below market prices.

I also believe Leigh’s findings shed some light on my argument that changes to town planning rules, including increases in height limits and allowable building area, does nothing to affect home prices. Any site with increased development potential will fetch a higher price, and the resulting dwellings will be released at the market price.

While Leigh’s paper is just one simple analysis of stamp duty rates and house prices, the theoretically sound finding should put to rest some of the illogical arguments of the supply side warriors, and the property development lobby in general.


  1. Sorry Cameron, I still disagree, but at least now I can see where you and Leigh are getting it wrong. Any drop in price is as a result of a secondary effect, caused by the added difficulty in the buyer obtaining finance. It is not a direct effect of the stamp duty. In a world where all buyers paid cash, the theory would not hold.

    All the high stamp duty does is impede transactions, and deny borrowers access to finance by decreasing the cash available to be used as a deposit, and you have interpreted that as a force to lower house prices.

    A better understanding of that would be gained if you examined the liquidity of housing in various states, in the sub $500K category. It would be better to concentrate on FTB's as that would allow you to compare the states with zero stamp duty and those who still have high tax rates.

    Another way of testing the theory would be to examine what would occur if rates were raised even higher. EG What would happen if the tax exceeded the value of the property. Can the price drop to below zero? Does the Leigh model fit all scenarios then?

    Cameron I look forward to your reply.

  2. Peter,

    Not sure what I can add at the moment. We are obviously not going to agree in the near future.

    If the tax exceeded the value of the property we would have some kind of communism - there would be no incentive to own property - the government would be forced to hold all land, in which case they would have no tax revenue. I guess they would raise revenue by renting their new property assets. In any case, that's not what the argument is about really.

    Leigh's doesn't actaully have a model as such, based on some theoretical foundation. He uses a couple of techniques to estimate the affect of stamp duties based on real data around Australia.

    Imagine the second hand car market. A new annual fee is placed on car ownership. What will happen to the prices of second hand cars? They will drop. It's simplified, but I'm not sure what else I can add to my position here.

  3. Cameron - thanks for your reply, and I don't want to be loading muskets at dusk with you - Leigh has done some research but drawn some conclusions that do not hold water in practical terms.

    I'm at grass roots level, so I see every day Victorians with their 10% in savings wanting to buy a home. The reality is they cannot because even if they are first time buyers, the FHOG does not cover the cost of the SD. As they can't finance that duty, they cannot complete. In NSW and QLD the transaction would proceed due to lower or Nil rates for first time buyers.

    Therefore in Victoria, prices will be effected by the high duty, but at no-ones benefit except higher revenue for the state government in the transactions that do complete, but the government gets Nil return from the transactions that cannot complete.

    I don't think that any state duty should be an exercise in "what is the most we can get from the public for this service" but even in that equation some reduction in duty would allow more transactions to complete, and still drive the same income whilst increasing the citizens ability to make choices in their life. Have you compared duties raised in Qld and NSW on a population basis to see if there is a real difference?

    I also question the use of income sourced from land transactions, if it is used for the benefit of all, then ALL should pay. I accept that some costs should be incurred for the service, but do you think that state governments should tax the life out of a new home owner just because it can. Is this adding to lifestyle quality, or robbing natural choices that people must make in relation to their place of domicile?

    Do you have any data on housing liquidity. If you divided house sales in each state by the adult population, would we see that the higher taxing states are inhibiting population movement for work and lifestyle? If so is that to the citizens benefit?

    Your lucky to be a Qlder like me, because when you buy you won't be slugged with a massive stamp duty bill as residents of some state have to bear, but is that fair and to the benefit of all Australians?

  4. I want to vote for = No Stamp duty, No land tax & 15% GST

  5. Cameron, I appreciate the time you spend frequenting my blog and commenting on the content. And I apologise that I am not normally able to respond to your questions. I do have a moment to spare now, so let me quickly make a few remarks.

    Generally speaking, you seem eager to criticise my now well-accepted arguments that government impediments to the production of new housing supply (such as zoning restrictions, long approval processes, anemic land release programs, and the multiplicity of taxes and charges levied at the input level) are critically responsible for driving up the long-term cost of housing in this country.

    I must say that I found your efforts to refute these claims surprising given that there is near-universal agreement amongst academic economists and government agencies that there is much empirical evidence to support my statements.

    Your confusion may relate to the fact that you seem to focus on the theoretical costs of "producing new land", which is not my concern given that I am talking about the cost of producing new "housing supply" (as opposed to new land).

    If you read my 2003 report to the Prime Minister, which was co-authored with arguably the world's leading housing economist, you would be familiar with these essential facts:

    "Here it is useful to distinguish between market-based valuations that recognize control rights, and intrinsic measures of worth that make no attempt to incorporate such. Ultimately, a property’s costs of production will be determined by three factors: the physical characteristics of the dwelling structure, the innate value of the turf on which it was built, and land use regulations that interfere with the market’s estimate of the latter. These distortions may take the form of specific rights that attach to the lot in question (i.e., zoning), or holistic supply-side strategies that dictate the release of greenfield and brownfield sites. As we shall see, the soaring cost of owner-occupied housing in this country has much more to do with government restrictions of this type than, say, a shortage of exploitable land."

  6. Watch out Cameron, looks like Chris is using "authority" as an argument.