## Monday, April 11, 2011

### No evidence of supply-side constraints in approvals data

Possibly the central lesson of my previous post was that planning controls and development approvals by local councils are not a factor that limits the quantity of new homes constructed. Council behaviour in these areas could limit housing supply if councils began a system of quotas for approvals. But they don’t. They provide limitations on the location of new supply in their planning instruments, and they approve the quantity of homes demanded by the development industry - which is a reflection of the number of new home sales. Sales volumes of new homes and land determine the rate of supply of new dwellings.

In my last post I provided no evidence for my assertion apart from logically examining the process of development in a hypothetical scenario.

So is there evidence that councils are limiting the supply of new dwelling through their planning controls and approvals processes?

No.

Let’s look at my home town of Brisbane. The following table shows the stock of approved house lots in Brisbane and surrounding local council areas that are yet to be developed (All data from here - Table 1. Excludes building units and retirement homes).

In Brisbane, where broadacre land is arguable more physically constrained, the stock of approved housing lots has remained relatively constant. And as you would expect, in the fringe areas, the stock of new house lots has grown far more rapidly than sales of lots or construction of housing. This indicates that councils approve far more housing lots than the market can absorb.

Yet it is this development approval that many claim is a hold-up to development.

In Brisbane, this reserve stock equates to about two years supply, while in surrounding areas there is between 3 and 10 years supply (Logan City and Somerset respectively) already approved. Of course, there are many thousands more lots that could be approved under existing planning schemes should demand arise.

Remember, this is just the stock of new land developments. If data were more readily available for unit developments there will no doubt be a similar story (in Brisbane attached homes are about 50% of new stock).

The clear message that comes from actual data on planning approvals is that they are not a constraint to supply. This might be one reason why these figures are never mentioned by ‘supply-siders’, even in the most detailed documents outlining supply side concerns in the housing market such as the 2003 Prime Ministers Taskforce on Home Ownership Report.

1. perhaps we're on the same page if I suggest that the constraints are really that the councils actually want the developers to pay money towards infrastructure capacity building (which the new development of course necessitates by brining in some hundreds more people)?

No matter which way I look at it I see that increases in population bring increased costs and long term increases in maintenance (wonder when we'll have to cost out replacing the backbone of reticulated water in Brissy?)

2. I don't even agree that developer contributions to infrastructure influence the rate of supply of new homes.

If the charge is known in advance, then it is another cost that is factored into the amount offered for the site.

If by chance the charges are increase between buying a site and lodging a development application, the developer has two choices - continue with reduced profits, or sell the site.

If he sells the sit he makes a loss because the next buyer will be factoring in the extra costs and pay less of the site. So he continues.

Of course, if this is an apartment project and he needs financing for construction perhaps these costs decrease his ability to finance the next stage. He can then bring in another equity party or sell up.

But if he sells for any reason the next buyer will develop the site

3. perhaps my point was badly phrased, I was suggesting it was leading to increased prices (not a constraint in itself)

4. Increased costs cannot be passed on. If the costs of development rise, this risk is fully borne by the developer. He cannot increase prices to compensate or people will buy elsewhere. They are a price taker at the extreme.

So my answer is no with one disclaimer - that sufficient development sites remain with residual values above the value for the next best use. For example, if the costs of development rise so much that residential developers will pay $1,000/sqm for an urban apartment site, but the value to the existing industrial users is$1,500/sqm, the site won't be able to be developed. If the site was sold the buyer will be another industrial user.

The same reasoning applies to subdivisions in rural areas - if the agricultural value of $150/sqm but residential value less costs is$100/sqm for the site, it won't be developed.

5. interesting, so are you saying that if developers paid nothing to councils for the developers fees (which are ostensibly to be put back into infrastructure) that the house prices would be exactly what they are now?

I guess this means that the developers would then simply be taking that slice and keeping it?

6. Exactly. For two reasons.

1. Because developers would be able to bid up the price of the site a bit further. This reduction in costs would accrue current land owners of developable sites.

2. There is never an incentive to sell below the market price. So if developer was holding land (land banking) and government decided to reduce charges (as they have done this week in Queensland), it would be a windfall for those developers and they would continue to supply homes at market prices at a rate they can achieve sales at that price.

7. ok, so if no developers were paying any developer fees would not the market find a lower level?

8. Alan W. Evans points out in his authoritative work, "Economics, Real Estate, and the Supply of Land", that the effect of developer fees varies according to how strictly the supply of land is controlled. If "planning gain" (the amount that land inflates in prices as the result of semi-monopoly advantages to the owner of the site) exceeds the development fee, then the development fee becomes merely the local government share of planning gain. Evans discusses the process, with historical examples, of the way the participants in the process fight for share in "planning gain"; the local government, bureaucracies, central government, developers, speculators, incumbent property owners, consultants, lawyers. The effort expended on this fighting is "dead weight loss" to the economy, along with most of the "planning gain".

In the event that a government with low restrictions on development introduces developer fees, the fees will be built in to the sale prices of the resultant homes, and the equilibrium of supply and demand will be at a slightly lower quantity. But this is much preferable to rationing of supply that has much more severe consequences.

There IS a social justice issue involved, in that everyone who already owns their home has not paid for such fees, while those who DO pay these fees (and those who buy their first home in the slightly price inflated total real estate market that results) are still going to pay taxes at the same rate as those who did not pay the fees.

Unfortunately, the younger generation today seems to be far too stupid to catch on to the great rip-off that is being perpetrated on them, especially in the grossly unjust costs of a first home due to the supply restrictions.

If a government has ALWAYS paid for infrastructure with developer fees rather than general taxation, there is no inequitable impact.

If you haven't caught up with Evans' whole authoritative theoretical discussion of real estate economics, you should do so, and join the ranks of the enlightened who are calling for reform of the great local government land supply racket.

9. This comment has been removed by the author.

10. This comment has been removed by the author.

11. I have now read Evan's book, but I wouldn't call it an authoritative work. For example, in Ch17 p222-223 (and again on p233) he confuses land value and the value of improvements to draw his conclusion that Henry George's logic leads to ultimately suggest that all land should be nationalised (although technically it is, by definition, owned by the State, with a bundle rights granted to owners who generally pass on a selection of those rights to renters).

Evans argues that if a land value tax was 100% then landlords would have no incentive to improve the value of their property. In fact the opposite is true, since the value of the land is derived not from what he does to the land but its natural bounty and location. Landlords have far more incentive to IMPROVE their land and reap the returns from those improvements than to leave the land idle and reap the rewards of increases in the value of land due to improvements external to their land of which they have played no part (eg, new roads and railways, economic growth in general etc).

Further, I don't argue that controlling the rate of land supply in the residential market doesn't impact prices. I argue that current planning schemes actually do not restrict supply of land, and in fact local councils approve far more residential development than the market can absorb. Therefore the rate of sales of new land and housing is the limiting factor.

Remember, planning schemes control the location of development to minimise externalities from different land uses (amongst other social and political objectives). They don't (although they could) control or restrict the quantity of land available to be converted to residential use. Thus is they don't affect the rate of land supply for housing, they can't affect the price level.

I would also add that Evans doesn't do a very good job refuting the quote on p17. Specifically his analysis in later chapters appears to support the Grigson qoute - "House prices determine land prices, not the reverse, because the builder's estimate of the selling price of the building will largely determine his bid for the piece of land"

To put my argument simply:
Supply and demand of housing determine rents. From these rents value is derived for housing. From this value, the value of land for housing is determined by subtracting the costs of development.

I argue there is no supply restriction because rents are not rising. Values of housing, and therefor land, are rising at a far greater pace than rents. While some of this can be explained by interest rates, much of this is pure price speculation, and should not wrongly be attributed to planning controls.

The graph at the link below shows this deviation of prices from rents. You might argue that the recent increase in rents above CPI might be used to argue a shortage, but this has happened twice previously on that graph only to even out. Recent data also suggest a high CPI and low rent increases so it is reasonable to expect then to converge again.

http://www.whocrashedtheeconomy.com/blog/?p=1503

12. Wow, you are trawling desperately through Evans' book to find things to criticise. Good on you for reading it, though. I hope you learnt a lot from all the bits you can't disagree with.

His explanation of classical economics theory on land supply is THE major contribution in there. I have no opinion one way or the other on whether Henry George was right or not. I hope you especially noted Evans' discussion of "planning gain".

The reason why rents can remain low, while property prices can be in a bubble situation, is simply that investors chasing capital gains, buy up properties faster than there is rental demand for. This also makes the "bust" more severe when it comes. It also makes it possible to have "enough" houses, AND bubble prices, AND empty houses, AND young people who cannot afford to buy them staying at home with their parents. If you've got parents that you can board with, why would you bother to rent a home?

This only applies to "first phase" bubbles after land supply regulations create distortions. The second phase and later phases (after the first one has been through its "bust") are usually accompanied by actual shortages of homes and rents rising almost as fast as house prices - eg in Britain, where they have now had 4 world record setting house price bubbles, on a "cycle" of about 15 years, since they enacted the 1947 Town and Country Planning Act. And they are now "short" literally millions of homes.

California is tracking them, about 1 "cycle" behind, although their "cycles" are shorter.

Australia doesn't HAVE to join them.

13. Maybe I wasn't clear. Evans' arguments rest on the assumption that supply is restricted. What he doesn't explain is how a modern planning scheme actually restricts the quantity supplied. Most planning schemes (in Australia at least) cannot control the rate of supply of land for housing, merely the location.

Hence, most of his arguments about supply restrictions are not applicable to a modern planning scheme.

Yes supply restrictions CAN be the major determinant, but this is not the case presently.

I did agree with much of what he said in the early part of the book, yet nowhere did he refute the p17 quote that land prices arise from rents, and he often used the term rent in a confused manner.

I still find it difficult to agree that supply restrictions, is whatever form, contribute to current elevated prices in Australia. If rents were abnormally high I would agree. But they are not. Hence the elevation of prices above the capitalised rent is mere specualtion (which you obviously agree with).

From what I can tell rents in Britain also have tracked CPI quite closely, at least in the past 15 years. http://timetric.com/index/uk_cpi_actual_rents_for_housing/

14. Testing......just seeing if this posting "deletes" the previous one - a major problem I have been having on this blog.

15. It did.....! So here it is again:

Have you got the same Evans that I have? Alan W. Evans (University of Reading, UK), in his book "Economics, Real Estate and the Supply of Land" (2004).

The Evans I know, goes to pains to clarify the meaning of the term "rent" in different settings; in fact he criticises the widespread misunderstanding of the difference between economic rent and commercial rent. I guess that having explained it, probably better than most, he then expects us to follow him logically through the rest of his book.

Indeed he does agree that land prices arise from rents, but he comprehensively explains why it is wrong to use this theory to claim that disallowing competition between all potential users of land, will NOT result in the inflation of the price of the artificially limited portion of the REAL total land supply. Australia is not Singapore, or even Britain.

Agricultural land is a vast "reservoir" of low cost land that SHOULD ensure a near-flat, and "flattening", low price curve at urban fringes everywhere. It is simply not the case that the single-digit-percentage excursions of urban land into this vast reservoir, ever pushes this low, flat price curve up by more than low single digits. In fact, the terms of trade for agriculture have been trending downhill so consistently for decades, that this should more than negate any urban encroachment effect.

It WOULD be possible to have a "planning" scheme that did NOT limit the supply of land for urban use; I would regard most of the heartland USA cities that have avoided having any price bubble, as falling into this category. There is virtually NO city that does not have a "planning" department of some kind; they merely differ in the extent of how "market driven" they are.

But you know and I know, that planners in Australia are driven by a stated policy intention to CONSTRAIN urban growth to within a certain delineated border for any given time period. This fits Evans' description of how land owners in locations priveleged to be included in the plan, can hold out for a share of "planning gain", and developers have to bid against each other for land banks or be forced out of business.

The French are masters at dodging international trade agreements by using bureaucratic bottlenecks to de facto ration the quantity of imports of certain goods. "But, monsieur, we do not have quotas; we wouldn't breach the terms of our agreement by having quotas, quelle horreur, non". But it MIGHT take you 6 months to get the customs entry processed.......

On the role of speculation; here we can agree. Some academics have developed models that explain the relationships at work by making the speculation endogenous to the inelastic supply. We have discussed this. Let me re-frame the argument. A market driven planning system simply IS now a well proven tool by which speculation is prevented from starting at all. The following is intuitive conclusions from the evidence I have studied; the INITIAL "planning gain" from supply being limited to some amount below 30 years, probably drives median multiples up from around 3, to around 4.5. It is the speculation that takes it to the levels beyond that.

16. Testing, testing.....

17. I have seen no evidence of any policy tools on the DEMAND side that successfully prevent these effects. They have all been tried, in Korea, Japan, various States of the USA.....(capital gains taxes, etc). The only areas that have avoided price bubbles, have done so by a very small variety of supply-side policies. Ironically, I believe that Australian State governments DO have powers over ownership of developable land, that mean that they COULD keep the prices low, as government does in the Netherlands, but they perversely exercise their powers to extract monopoly rents.

Evans wrote 2 books at the same time, the other one, "Economics and Land Use Planning", discusses a lot of what I am saying here, that the first book does not. I have still never seen any other books as good; Evans seems to me to deserve to go into the econ history books as the next crucial developer and refiner of land econ theory following on from Alonso and Muth.

I have insufficient data for Britain, but I would be surprised if their earlier bubbles (1950's onwards) did not include a temporary disparity between rents and sale prices of houses, which disparity lessened as the actual shortage of homes began to really be embedded by perverse supply regulations. I am not at all surprised that in a mature housing-shortage market, rents can be seen to be tracking the CPI; I AM sure that they are far too high, along with house prices, (compared with historical or international or free-market alternatives) due to the easily quantifiable under-supply of houses in Britain.

Bear in mind too, that the CPI includes rent as an input; AND that everything is too expensive in Britain as the result of high commercial land rents and lost urban economy efficiencies.

In Britain, developers have to cope with the cumulative effects of the need to hold land banks, and to compete in Dutch auctions for land banks, and to go through lengthy permission processes, and to attempt to follow volatile pricing cycles, and to attempt to bring one's "supply" to market BEFORE the market "peaks" each time; all this has led to more and more developers simply exiting the industry, and employment in the building sector steadily dropping - all this while the shortage of houses gets all the more starkly quantifiable.

It is interesting that Japan did not enter into this "repeat cycle" phenomenon, but had an extremely LONG unwinding after their famous bubble - which also had to do with supply issues - as did Korea's famous bubble; and China's bubble today. Boy, oh boy - watch that space. Will Aussie's bubble blow by itself first, or follow the Chinese one?