What’s driving house prices?
Well we know that there’s some artificial scarcity driven by the rationing of land for housing is an important contributor. For instance Canberra has lots of land, but very high house prices (and pretty cruddy little blocks on the outer edges) and its hard to believe strangled land release policies are doing this. I’ve seen allegations the truth of which I don’t know that this scarcity is partly to appease the Greens.
But Sydney? Well, yes there’s a land release problem there too. But I’ve always assumed it must be a small part of the problem. And the PC report on first home ownership highlighted land release but seemed to suggest there were no miracles available from that. Not so according to a highly informative but politically partisan website (deomographia) that Access Economics got a lot of publicity for quoting today.
How’s this for a comparison:
Lets compare Houston, USA and Sydney, Australia. Expressed in local currencies, the median household incomes for both cities were, during 2004 (in local currencies) $50,400 and $57,100, with median house prices of $138,350 and $505,000 respectively. So it would take 8.8 years of median household income for a Sydneyite to purchase a median priced home, but just 2.7 years of median household income for an Houstonian.
Expressed another way, if the Sydneyite on the median household income, was to get the same deal as her counterpart in Houston, she should only be required to pay 2.7 times her household income of $57,100 – that is $154,170 ($57,100 x 2.7), some $350,830 ($505,000 – $154,170) less than she is currently forced to pay.
I don’t know much about this, but it seems like a big deal in influencing our lives. And there seem to be real paradoxes. Demographia’s opposition to “Smart Growth” policies in favour of urban consolidation would in all likelihood increase urban sprawl. Whether that’s good or bad, that’s what we have in a place like Brisbane. But Brisbane and L are some of the sprawliest cities in the world (on a different scale obviously) yet they have some of the relatively highest house prices. Brisbane is said to have just passed Melbourne in its median house prices (which given Brisbane’s substantially lower wages is quite something). Yet its’ surrounded by sprawl.
Can any commenters offer wise counsel on these matters?
Confused.
Port Melbourne.
Houston is not a good comparison, San Fran, NY or DC would be better analogies with Sydney. They are world cities that are the epicentres of the American Western, Eastern and Mid-Atlantic economies.
Using Houston is like comparing New York to Forbes, or San Francisco to Lithgow.
The DC area is very unaffordable, and its sprawl is right through Virginia, Maryland and even into Pennsylvania and West Virginia now. New York has sprawled through New Jersey and into Pennsylvania too. In Nth Virginia a starting single family home is $600,000. Townhomes are in the $400,000 at the outer edge of the Dulles Greenway.
There is no rhyme or reason to it, other than access to cheap money in the form of low interest rates. There is constant development in Nth Virginia, new areas are going up everyday. Yet the house prices of the new areas are inflating at the same rates as established property. In Nth Virginia it has not been unusual to see 100% inflation in property prices over three years.
I recall seeing a graph the other day about how money has been made cheaper through interest only loans, IIRC it was something like 40% of people using them for home loans. The old fixed interest rate loan is obviously too expensive for many people.
Surely most of the interest only loans are for investment properties which are also negatively geared. The increased borrowing capacity of an interest only loan compared with principal plus interest is surely only in the order of two to five per cent before you factor in tax deductibility.
A major factor in inflated houuse prices would be the rise of the two-income family in our lifetime, plus the “leg-up” from parents who assist with the first home purchase. Two indirect indicators: (1) the secretary of a GPS school reported that a third of the cheques for fees were written by grandparents. Many parents and grandparents are helping out in the short term rather than leaving it all in the estate. (2) the late Colin Simkin, when he was Prof Emeritus of Economics at the Uni of Sydney, did some calculations on house prices compared with income data and estimated that there was a gap of $100,000 (circa 1990) between the median house price and what people could have been expected to afford on the basis of the income data (even allowing for two income households).
“There is no rhyme or reason to it”. Cameron there must be some systematic forces at work affecting supply and demand in different cities. The Demographia people argue that most of the unaffordable US cities are unaffordable because of artificial land scarcity. And if you think about it, there should be no reason that land is so expensive in any of our cities ALL THE WAY TO THE FRINGE where land use competes with market gardening and even extensive agricultural use of the land.
Rafe, The US doesnt have negative gearing. You can claim the interest from your mortgage on your taxes for your primary residence, not sure with investment properties. I will try and find that graph I mentioned. This has become an issue that has popped into the mainstream media in the last few months.
On an anecdotal level, I have friends who have interest only loans for their primary residence. IIRC, one of the loan products is that after five years it kicks in as a variable interest rate. It is the only way a lot of people can afford a house/townhome/condo.
Nicholas, By no rhyme and reason, I probably meant human irrationalism. A sad fact is, I bought a house here three years ago – I could not afford to live in this neighbourhood now. It is a typical middle class neighbourhood too; retirees, firemen, school councillers, software, landscape business owner etc etc.
The urban sprawl of Nth Virginia is huge, endless and constant. It also extends across the entire north of the state into West Virginia. This didnt exist twenty years ago. People have been flooding into Nth Virginia, homes cant keep up. I am part of the problem of course. I moved to Nth Virginia four years ago.
Properties have been getting smaller and smaller too. I noticed a new development up Route 7 now has houses that are joined by garages, so they arent really townhomes, but are sorta homes. Maybe it is because people are demanding single family homes rather than living in high density areas.
Anecdotal again, but in the US, housing prices didnt pop until money got cheap. Interest rates have stayed low, and there has been new mortgage products which has made money even cheaper. Rather than an under-supply of houses, it might be an over-supply of cheap money.
You’re undoubtedly right Cameron that cheap money has driven much of the boom. But in Australia we no longer have particularly cheap money and it never was nearly as cheap as in the US in any event. But the interest is comparing different areas of the country. LA and Miami have very high prices, but other places don’t. I don’t think its clear why. Why should Houston have much more affordable prices than Brisbane. It’s got cheaper money, more income per person and a higher population.
Houston is a college town, bit like Wagga Wagga hahaha. It is not the epicentre of an economy or a global city. It used to have lots of oil money, especially in the 80s, but not so much now.
Maybe globalisation is the issue. I recall reading something a while ago how power and wealth with globalisation had collapsed into a few small global cities with dominant economies, and drained the countryside, kind of in the same way industrialisation in Britain did
in the 1760s.
The big global cities have higher salaries (more inequity too?), more oppurtunities, more high-tech, and larger economies. DC, NY and SanFran fit that. They are the epicentres for the Mid-Atlantic, Northern and Western economies. Though Boston is a global city too. The southern US global economy is probably dominated by Miami.
Australia has two economies; the eastern service based one, and the western primary based one. Brisbane is part of the eastern economy, along with Sydney and Melbourne. They are global cities in the same way DC, NY etc are. So it would be expected that those three cities would mimic the housing trends of DC, NY etc.
Speculating ….
Doh, thinking of Austin, not Houston. It is the college town. Houston was big on oil money though in the 80s.
In the last 30 years some interesting things have happened in Australia.
First, the interest rate on home loans has been deregulated.
Second, the average size of houses has doubled [OK, at the same time, more of the housing stock is units but apartments are getting bigger as well].
Thirdly, there has been a big increase in two income families [and more important you are allowed to count more of the second income for borrowing purposes] and also people are having children later and fewer of them (someone could put numbers on all those things).
Fourth, more people are getting substantial inheritances (the fruits of the hard work of others in the fifties and sixties).
Fifth, most people have hugely increased aspirations for the quality of their first home and they are less willing to start off in a fibro shack or an old unit.
People also forget that one of the drivers of the prices is the fact that people have been willing and able to pay them.
Can someone who works in the industry tell us how much more you can borrow with an interest only loan, I can’t believe that is the diffence that really matters, you could just as well buy a slightly cheaper place.
The high level of taxes might have something to do with housing costs in Sydney.
Lets compare a can of coke to a median Sydney house to give an insight as to what is going on.
A can of coke is cheap to produce and it sells for about a $1 with a small amount of tax incorporated in the price. It is easily affordable and available nearly everywhere.
If we put $18,000 stamp duty on it (approx.duty on a median Sydney house), it then becomes very expensive and trade in cans of coke would be severely curtailed as less people could afford to buy or sell them. Less cans of coke would be produced so production costs would also rise. Those that have to borrow to buy the can of coke would also be hit with stamp duty on their loan. Newly released cans in Sydney will also have the recently introduced $100,000 state and local government contribution levies added to their price. (see reason 7 http://www.reinsw.com.au/dir111/rei/Library.nsf )
When a Sydney shop keeper buys a slab of 24 cans and subdivides it for individual sale things get really interesting. He buys 24 cans at 80 cents each hoping to sell them at a dollar (20% margin on sales). On purchase the shopkeeper pays stamp duty 24 x $14,400 (80% of $18,000) = $345,600. He then pays the Sydney contribution levies 24 x $100,000 = 2,400,000. He then pays GST calculated under the margin scheme thus; Profit on 24 cans $4.80 + $345,600 + $2,400,000 = $2,745,604.80 divided by 11 = $ 249,600.44 GST payable on sale of the slab of 24 cans. This figure is then divided by 24 to give the amount of GST payable on an individual can; $ 249,600.44 divided by 24 = $ 10,400.
The retail cost of can is arrived at 80 cents + 20 cents + $ 14,400 stamp duty + $ 100,000 site levies + $10,400 GST = $124,801.
Taxed the same as a new median priced Sydney house the retail price of a can of coke the on the shelf would be $124,801. The purchaser (assuming not a first home buyer and no mortgage) then goes to the Office of State Revenue and pays $18,000 stamp duty (approx. duty on median Sydney house) at the time of purchase. This brings Total tax on the new Sydney can of coke to $142,800; being 2.5 years of median household income for a Sydneyite.
Supply to the Sydney housing market is very heavily restricted because of the large tax burden in proportion to median household income on transactions . This restriction must inflate prices. Further inflating the market is the tax on newly developed greenfield housing that is introduced to the market with total taxes of $142,000.
The Houston house-price situation is indeed an anomaly; under any measure, prices there are ridiculously cheap. My hunch is that the Enron factor, and fashion, are part of the explanation. That is, major corporate collapses (speaking from experience: there were several in my home town of Melbourne c. 1990) both directly deplete the pool of funds available for housing and, more nebulously, “spook”
econowit, in the US lots and lots of public goods (schools, police, etc) are funded by local property taxes. In most US cities taxes on property would be substantially *more* than anywhere in Australia.
Which of course is part of the point. A tax on land is likely to *lower*, not raise, the price of land because the UCV of land is a pure rent. You don’t have to be a full-blooded Henry Georgist to believe that a land tax is actually in principle a very efficient tax, and not bad on the equity stakes either. Arguably, though, it ought to be an ongoing assessment rather than a transaction tax.
I actually think one of the hidden things in this argument is the cost and ease of transport. Where this is poor (eg Sydney) it drives up the price of land close to where people work. I wonder what Houston’s transport is like?
There are a couple of factors that have not been mentioned, or only alluded to.
One is that scarcity of land is not always artificial. A city founded in the a location where it can expand in all directions should have cheaper land than one like Sydney, which has a solid wall not far from the centre in one direction. This means all sprawl has to go one way, and even if infinite land is made available in one direction it won’t entirely cut prices – why would some one want to live two hours from the city if they work there? Prices should be cheap in the outer reaches of Western Sydney, but that wouldn’t be expected to cut prices closer in if the outer reaches are just too far.
Another factor is simply expectations. Most Australians have a belief that house prices always rise (or occassionaly go sideways, but never fall). If you believe this, a house is a smart investment at almost any price, and people will take out ridiculous mortgages to pay for it.
However, if you don’t believe it (perhaps becuase your city once went through a period of house price deflation) then you consider prices much more carefully, and are willing to pay far lower prices.
I don’t know much about Houston, but I suspect that prices fell at some point, and consequently people there don’t believe in the myth of ever rising prices, and they’re probably paying sensible prices as a result.
In Melbourne I meet people all the time who state baldly to me “house prices never fall”. They tend to be pretty shaken when I tell them that many other countries have experienced falls of more than 20% in house prices.
Rafe says:
“Can someone who works in the industry tell us how much more you can borrow with an interest only loan, I can’t believe that is the diffence that really matters, you could just as well buy a slightly cheaper place.”
The difference can be significant.
With an interest only loan, the amount you can borrow for a given payment is:
PV = pmt / i
Take payments of $1,000/month, and interest of 0.5%/month.
PV = $1,000 / 0.005 = $200,000.
With a traditional loan, the amount you can borrow is:
PV = pmt (1/i – 1/(i(1+i)^n))
Take n = 240 months (20 years):
PV = $1,000 (1/.005-1/(.005(1.005)^240)) = $139,600.
For this case, the interest only loan lets you borrow an extra 43% over the traditional loan.
That’s in theory. In practice at least in Australia, loans are typically 30 years long which would increse the loan that could be serviced a great deal. Further, serviceability is assessed as if the loan were a principal and interest loan in almost all circumstances. Perhaps this will change in future.
Can’t get you a loan can I?
Dr Peach – http://www.peachhomeloans.com.au
“…loans are typically 30 years long which would increse the loan that could be serviced a great deal.”
I provided the formula so that you could actually play with it, not just make statements like “increase… a great deal” :)
With the same interest rate, and n=360, the PV increases to $166,800, so the interest only loan gets you an extra 20%.
It’s also interest rate sensitive. I used a nice round number, but a value of 0.62%/month would be closer to current rates. That would give you $124,700 P&I over 20 years, or $161,300 interest only, an increase of 29%. Over 30 years, you’d get $143,900 P&I, so the interest only option only gets you as extra 12%.
Nick,
Sorry, I should’ve clicked the link. I didn’t mean to try to teach you your own business. :)
One unmentioned factor is the increased cost of building materials. I have been involved with costing a building project at my church, and as we price materials our Architect regales us with how the cost of electrical outlets (the small bit of plastic on the wall) has doubled in the past two years, the cost of steel has increased by however many multiples in the last five years, etc. etc. All in all, he admitted to us that if we had been costing the same project three years ago, the cost of materials would have been half what it is today.
One reason why Houston is a bad example is that it is famous as one of the least planned cities in the world. My recollection is that Houston boasts zero city council zoning controls, as a result, development is all over the place and property is frequently converted from residential to commercial to industrial and back again. Another thing worth mentioning is the effect of undocumented migrant workers, who from the stories of friends living in Houston make up a large proportion of the population. They earn less than official average earnings, but aren’t included in earnings surveys, so would distort those results. They also are prevalent in the construction industry driving down the cost of construction, particularly in the less skilled, low-rise residential sector. They also demand a lower standard of housing than the rest of the population, and this housing is included in the statistics, thereby radically distorting the income / house price ratio on both sides (income and house prices).
For what it is worth, here’s an extract from the Midwood Report – a respected real estate report but one relating only to Qld.
“Brisbane construction costs have risen by 25% since 2001 . . . [but] have peaked and are expected to subside over the next 2 yearss or so, but not the extent of the 25% increase we have just had, as in the long term they tend to rise at slightly less than inflation.”
One cause of increased prices may be increased demand due to the increased concentration of industry and commerce. Someone mentioned decreased urban transport as something that keeps prices up. Loss of regional job opportunities and the collapse of agricultural employment through mechanization/industrialization give people no option than to stay where the work is, in the cities. Closeness to places of employment is the ultimate driving force behind cities in a way, even though it is self sustaining, this in turn does depend upon the resources available to a city.
The relationship between work and home, is IMHO, economically inelastic. Sydney workers, I suspect , spend the most time travelling to and from work , with many spending four hours a day. Spending any more time than this is probably unsustainable, thus driving demand up and prices because the supply can only be increased by increasing density (which is more expensive).
I suspect that travel time plays a part in Brisbane’s price rise as well and that it conversely has kept down Perth’s prices because of the extension of high speed rail/freeway network in the last 30 + years.
Derrida wrote, “in the US lots and lots of public goods (schools, police, etc) are funded by local property taxes.”
I can attest to that. I get taxed at four levels in the US; the federal, the state, the county and the town. The state, county and town take property taxes from me.
The town I am in takes 21c per $100 in Real Estate taxes, and takes $1 per $100 for the worth of my car. The county takes $1.04 per $100 in real estate taxes and $4.20 per $100 in personal property taxes.
The northern US states are worse than the south or mid-west. it is not unusual for a house in NJ to pay $8000 USD in property taxes, for what is an 800 sq ft home. The counties pay for most of the schooling costs in the US, so if the county you are in has a bump in school age students, property taxes rise quickly.
The county I am in has been raising property taxes at 10%-15% each year, just due to house appreciation values. Their budget that they have been spending has been rising at the same rate, despite them not having new liabilities to meet. Thankfully the town is more level headed, and caring of the fiscal situation of its citizens, and has frozen their property tax rate.
So whats the deal for Canberra? Transport to work isnt a problem (I doubt anyone spends more than 45 minutes travelling, if that). Perhaps Canberra just has the higher income/low interest rates and lots of money sloshing around. Rents are actually amongst the highest in Australia.
I was living in HK during their property crash. After 1997 prices were skyrocketing, people were buying apartments (the main form of housing) purely as speculative investments, on-selling within weeks and often before settlement. It was not uncommon for there to be 10+ buyers/sellers between the actual owner and the actual buyer who took possession. The govt was caught out as they only taxed the final contract, not sub-contracts – thus the in between sales had no cost (other than real estate agent fees – I heard of one agent who received 8 commissions on the same property in 2 months).
Anyway, by the time I got there in 2000 prices were down 50%+ (more for low grade commercial). Litigation over conveyancing was keeping the lawyers very happy; but many home owners faced significant negative equity (even the conservative ones got caught out – fearing prices would continue to rise they took the plunge and bought in). Bankruptcies were pretty common.
Of course, HK is not much use for Nick’s question since land supply is intentionally restricted. The govt actually owns all the land and just grants 99 year leases, restricting land means the govt gets more money from selling the land.
Nonetheless, while possibly somewhat extreme, it does show that investment can really drive prices up. A few years ago in Australia you could pretty much have a self funding loan for an investment property (rent+tax benefits such as depreciation being more than the cost of interest). Your return was running at 5-7% p.a, pretty good for ‘no risk’ but lower than shares and (for a while) interest rates.
However, when capital gains tax changes came into effect, your 7% return suddenly became net 14% (tax halving), still for no risk. Interest rates continued to fall, so the cost of the investment was even lower (indeed, if you bought right you could positive gear and make capital gains) and your gain was approaching 20%+. Thus it could be argued that housing prices simply increased to ensure that the investment return remained comparable to other investment alternatives.
In addition the increase in size and quality of housing needs to be factored in; plus the ‘sunk’ cost of those improvements has increased because the cost of labour is so much higher. In other words, a 3 bd house on a block of land has increased by more than just the value of the land because the replacement value of the house has also increased.
Those are good points ctd, I tend to downplay the human factors, seeing things from a biological perspective.
One could argue that the increase in the physical cost of housing is also driven at least partly by increased availability of funds. Much recent housing is notoriously unsustainable in terms of energy efficient design, which I guess is because people are able to discount the future in such a fiscally liquid environment.
I’m not sure that I would agree that the cost per cubic meter has increased in the last 50/20 years. Things like concrete pads have cut labour costs, but there are so many if’s and but’s and questions about what to count. It’d make a good Ph D. if it hasn’t already been done.
As the instigator and researcher of the Demographia International Housing Affordability Survey 2005, i have enjoyed reading the contributors comments – and trust my further comments are of assistance.
Its important to realise that over a reasonable time in real terms, construction costs have not increased very much at all. So the only other key component is land.
section prices within the many dysfunctional urban markets of Australia and New zealand have gone through the stratosphere over the past thirty years or so.
May I suggest commentators read the Australian Housing Industry Association Affordability Report 2003 and the Housing Cost Report 2004 by DTZ NZ for the NZ Centre for Housing Studies. Both are available on the web.
These clearly spell out and quantify the section / block cost problems.
For a good read on the differences between Sydney and Houston – may I suggest you access a recent speech by Bob Day, National President of the Australian Housing Industry Association – titled “A tale of two cities”. It can be accessed at http://www.nationbuild.com then go to the Housing Affordability Section.
This issue is really about getting down to the “nuts and bolts” of housing costs so that we can all hopefully better understand the problems.
New house purchasers land costs should not for example be any more than around 20% of total development costs. They have been cranked up to between 40 – 70% in our Australian urban property markets.
The 2006 expanded Issue of the Demographia International Housing Affordability Survey comes out late Jan / early Feb next year. Hopefully we will be able to better deal with the points raised above and elsewhere then.
With best regards
Hugh Pavletich
Christchurch
New Zealand