In real estate, just as in shares and other investments very high rates of gain are not sustained, the gains in prices made during bull markets are by no means typical of average long term growth, and after a while every bull market fizzles or falls, bringing back long term gains more toward average returns. This is known as regression to the mean.

Recently I have been to a number of real estate seminars put on by agents wanting to drum up business, just curious I suppose about what goes on in these things. Real estate agents are of course only putting on these seminars to generate business so you can understand that they may wish to put real estate in the best possible light, while trying to discourage you from putting your money into alternative investments.

What I keep hearing at these seminars is that real estate goes up at a rate between seven and eleven percent per annum. They said that property almost never goes down in value and that all you need to do is buy some property to gain vast, virtually risk free profits. By the same token they are always quite keen to try to put shares down, dwelling on the crash of 1987 and how much money everyone lost, and the number of years it took to break even.

A common rule of thumb that applies to real estate is that people are reluctant to spend more than one third of their income on mortgage payments or rent. Banks tend to enforce this rule by rejecting borrowers that want to borrow more than this, but it is nevertheless true that few people would willingly buy a property that will cost them more than one third of their total income.

Think about what factors drive the cost of a mortgage. There may be a few other factors but really the ones that completely dominate the equation are wage growth and interest rates. If wage growth speeds up people can afford nicer houses, or at least more expensive ones, if interest rates fall the same thing occurs.

What drives wage growth? Inflation mainly, though there are fluctuations in so-called real wage growth, where people's earnings do not necessarily move in lock-step with the cost of living. Real estate is not one giant national market the way the stock exchange is, prices in individual areas tend to be influenced to a great extent by local factors, which is why various suburbs seem to boom while others seem to bust at any given moment.

There are of course obvious local factors such as new roads, employment facilities, shopping centres and schools that may affect local prices but in some ways it is not these that affect prices but a factor that follows on from all of these.

That extra factor is the earnings growth of people in the area. This is one of the most important factors and is a result of all of those above. When an area becomes more desirable because of some of the changes mentioned in the previous paragraph it will attract wealthier new buyers. This is particularly so in inner city areas that are undergoing rejuvenation, when all of the old cottages that used to be owned by the working class suddenly become fashionable, prompting the café latte set to move in.

So areas that are attractive to wealthy people will appreciate at a higher rate than the market average because of this status change, the median wage in that area moves within a few years from that of a factory worker to a city office worker, in some cases this gain can be huge.

The other factor that influences property prices is of course interest rates. When rates are lowered all of a sudden new possibilities are available to everyone, and they can now afford more expensive houses. Of course low interest rates aren't available to just one demographic, they are available to everyone, so they tend to produce a universal rise in property values. Falling interest rates are enough to stimulate strong growth in values in every sector.

By the same token if interest rates go up this will slow the property market, possibly even sending it into decline.

Now just consider how these two factors influence the real estate market as a whole. Interest rates won't rise or fall for ever, after a while all trends will be reversed. The movements of interest rates will stimulate various boom and bust cycles but they will all tend to even themselves out in the end. From a very long term perspective you might as well ignore interest rates and merely watch the other big factor, wage growth, when forecasting likely long term returns.

Ignoring local changes to an area that may lead to an upgrade in the average wage, it is the average growth in wages for the entire population that will push up property prices on average. These real estate agents were saying that median house prices have gone up by eleven percent per annum over the past twenty years, is this a great rate of return?

Over the same period of time, people's wages have risen by an average of just below 6%. In December 1981 average weekly ordinary time earnings (AWOTE) was $276.90, in December 2000 it was $802.60. I can't quote the December 1980 figure because this figure was not published before 1981, nevertheless this figure shows over the last 19 years that AWOTE has increased by 5.76%pa.

On the other hand average weekly earnings has been tracked for a much longer time and from December 1980 to December 2000 it has increased from $230.50 to $644.40, an increase of 5.27%pa. (source, the Reserve Bank's web site, labour costs data spreadsheet).

So just under 6% of this price increase can be accounted for purely by looking at wage growth, this leaves just over 5% in median house price growth unaccounted for.

What were interest rates in 1981 like? Looking this up again on the Reserve Bank's web site I see that in December 1980 the average variable home loan rate was 11.00%, in December 2000 the average lending rate was 8.05%, that is a 27% drop in the cost of bank financing over this time.

Another figure you can find at the Reserve Bank's web site is one of their measures of inflation, there is an index that tracks the cost of housing. In December 1980 this index was 43.3, in December 2000 the index was 107.7, which says that housing has grown in expense by just 4.66%pa. Is it just a coincidence that housing affordability has lagged behind wage growth by approximately the same amount as lending costs have fallen?

According to one well known organisation that promotes property almost as hard as they denounce shares, in Sydney in 1980 the median house price was $71,000 and in 2000 it was $315,000, which I figure as a compounded increase of 7.73%pa. Somehow they got their maths a bit wrong and they say this increase was 9.1%pa, but I'll forgive them for that, after reading what they say about shares and superannuation my expectations have been lowered somewhat.

I must admit that I'm a little confused by how prices could get so high, working out the mortgage payments on my financial calculator I figure that for a 30 year loan at 11% the payments should be $676.15 per month, which seems a little harsh compared to the average weekly earnings of that time, which were around $250 depending on which statistic you want to believe. This means Sydneysiders were paying two thirds of their income on mortgage repayments!

Graphs comparing real estate price increases with inflation do indeed show that property did very badly for the next few years, perhaps a result of the exorbitant costs. On the other hand when you look at median house prices in 2000 at $315,000, assuming an 8.05% interest rate and a 30 year loan, monthly mortgage payments would be around $2,322, or $536 a week, which is now almost 80% of average weekly ordinary time earnings!

Over the very long term these cycles even themselves out. The difference between inflation and property price increases is almost negligible.

In other capital cities, median house prices are a little lower, in Perth in 2000 they were $156,700 and in Melbourne around $241,000. These mortgages are a little easier to pay, in Perth the monthly mortgage payment would be $1,160 and in Melbourne $1,780 or about 40% and 64% respectively. These are still expensive prices by the historical standard of only paying 1/3 of your wages out to the mortgage.

What is going on? Why are people willing to pay such prices? For starters Sydney does tend to have a lot of higher income earners. This is a "median" house price, which means they make a big list of house prices and then take the one from the middle, as opposed to a mean list where they add up all prices and divide by the number of houses sold. If it was an average house price there would be a fair few really expensive waterfront mansions pushing up the price but with a median price quote this is not the case. Instead we can only assume that residents of Sydney have a higher weekly wage than the national average (this is true) and that people in Sydney find real estate far too expensive, which is also true.

Charts stretching back over the last 40 years show an interesting trend in residential real estate prices. Although it is common for real estate exponents to talk of a 7 year cycle or a cycle going over some other number of years, there have been only two significant bull markets in property prices in all that time. Graphing real returns (after inflation), the periods between 1969 and 1974, and 1985 and 1988 were the only periods of significant upward movement in prices with respect to inflation, the late 90s showed some gain against inflation, but nothing spectacular.

This is in terms of real returns, which means returns indexed against inflation. As inflation does tend to trend upward this masks the cycle from the view of less sophisticated investors. If inflation runs at 10%pa but property gives a negative real return of -3%pa, that still gives a 7%pa increase in the nominal price, which is very impressive to someone that doesn't know that an investment is a failure unless it gains value at a rate higher than inflation depreciates what you can buy with that money.

You don't necessarily need to have growth exceeding inflation to make money out of real estate. Although when you try to figure out your return on investment you won't find property too impressive, but if you borrowed all of that money anyway this might not concern you. Even fairly modest capital gains can give relatively impressive profits when gearing is involved.