Real estate promoters are very fond of pointing to enormous profits being made by all in certain fringe suburbs that are due for an upgrade. It isn't all that rare to see median house prices pick up by more than 30% in a year. A spectacular return! All this from something as safe as houses...

Lets take a look at a fictional suburb that we'll call Slumdale. For years Slumdale has been an awful place, with dead cars sitting on blocks in every second front yard, streets you wouldn't want to walk at night, graffiti, high burglary rates, bad facilities and so on.

Sound familiar? It ought to, there are plenty of suburbs like that where you can buy some rat infested fibrous cement slum on a big block of land for $60,000. Next to Slumdale is Toff's Peak, a really nice new suburb built on a hill with beautiful views of the distant ocean, if you can see past the eyesore beneath it.

In 1998 the state government announced that Slumdale would be a candidate for one of those urban renewal programs. Values jumped slightly on the news and amazed onlookers then were treated to an amazing spectacle, a rebuilding boom started as houses were demolished and replaced with nice houses, Slumdale was renamed "Emerald Gardens Estate" and the local park finally got working swings and a sprinkler system. Security patrols began and graffiti was halved. What really gets you is that median house prices rose over 50% in the next year or two. Makes you wish you could be a slum lord doesn't it?

That too should sound familiar, it happens all over the place and real estate agents gain great mileage from pointing out such successes as definitely achievable. One might wonder though if it is achievable, can you really time the real estate market by watching long term planning strategy at the local governments and get in on the booms before they happen? Assuming that you could anticipate such a move, there are certain factors that make the median house price rises a mirage.

Lets just say that at the start of 1998 every house in Slumdale was worth $60,000. Developers start to encroach on the borders of Slumdale building reasonably nice houses to extend the suburb of Toff's Peak. Over the next two years, 20% of houses in Slumdale, or Emerald Gardens Estate as it now is called are demolished and rebuilt.

So houses are bought for $60,000, demolished, rebuilt at an expense of another $60,000, $15,000 is added on as the developer's mark-up and then sold to investors for $135,000.

Another 20% of houses get renovated, at an expense of $30,000, and are then sold for $95,000.

The remaining 60% of houses are still fibro hovels, but at least some of the inhabitants start to take pride in their "new" suburb and paint their woodwork, put in a nice picket fence or do something about the garden. Between these improvements and the appreciation brought on by the boom the average value of each of these other houses is now about $75,000.

If each sold, what percentage profit would they make? The builders would make about 12.5% before taxes. The renovators would make 5.55%. The passive investors would make 25%, quite a good profit for just one year.

These are great returns for a single year, though you do have to take into account the fact that they are not really repeatable. Chasing such returns will be futile because it is likely that the boom will move on to another area next year as the urban renewal bandwagon rolls on. Following that prices will probably stagnate like real estate prices usually do, and move up at the same rate as the average weekly income of people that want to live in that area, which could actually be a good thing if yuppies decide to move in because they think Toff's Peak is overpriced.

But what about the indicators, the average house prices that real estate speculators who were sitting back watching will all go GA GA over?

The mean house price in 2000 is now $91,000. Which means the average house price has gone up by a staggering 51.7%. Even better, watch out for the median house price. These numbers reflect only houses sold in that area over the last 12 months, so they exclude the values of houses that are not bought and sold.

Lets just say that Emerald Gardens Estate has 1000 houses, then 200 will be new buildings, 200 will be renovations and 600 will be unimproved. Of the 200 houses that were rebuilt, 150 get sold when the building is complete. Of the 200 houses that were renovated, 100 were sold. Of the 800 houses that were unimproved, 100 were sold. The median house price will be $95,000. It is higher than the mean house price because it only reflects houses actually sold in 2000, which automatically biases the sample toward the redeveloped properties because it is these, not the fibro slums, that are being sold. So in this case the median house price gain would be 58.3%.

Just for interest's sake, the mean price of houses actually sold would be $106,428, a gain of over 77%.

This is why median and mean house prices are not a very good way to track improvements in property prices, they massively exaggerate price appreciation because they completely ignore the costs involved in preparing a property for resale. In this fictitious example everyone made money, in real life budgets go overboard and people over-capitalise. The median house price gain would be just as impressive, but everyone's actual profits would be diminished. Not only is it hard to anticipate real estate booms, but even when you do the returns are usually pretty lukewarm anyway. Then you take off 5% for the real estate agent's commission when you sell, a bunch of taxes and stamp duty, then factor in all the trouble you went to.....