Very interesting article from the Residex Monthly Newsletter
The Shape of Things To Come: Part 1 (Sydney and Brisbane)
25/07/2005
At the end of every financial year, we do a major long-term forecast for the capital cities of the state. Over the next couple of issues of this newsletter, we will be presenting the forecast graphs of the major capital cities and seeing what the future holds for these different markets.
The common theme in all of the cities is that they have come off the boil in terms of property growth rates over the last 12-18 months. The hardest hit were Sydney and Melbourne. Brisbane had a milder slowdown but it was very sharp and quick compared with its southern cousins. These corrections are normal at the end of the housing cycle. However, these corrections, especially in Sydney and Melbourne, are lasting longer than normal.
While each city exhibits its own unique growth characteristics, there are broad similarities in terms of the timing of market corrections etc., but the amount of expected future growth will vary from city to city depending on its economic fortunes and government policies.
This month we will look specifically at the most likely outcome for Sydney and Brisbane.
The graphs below are produced by our computer prediction model. This is a stochastic model, meaning that we are modelling processes which are based on probability. In this case, our model is developed by looking at movements in the property market going back over 100 years. Some of the information we look at is what happens over short time periods (1 or 2 years), medium time periods (5-8 years) and longer time periods (9 years plus). Other information taken into account is duration of cycles, growth rates following periods of correction and population growth, etc.
From looking at these various movements, and other factors that influence the market, we can build up pictures of what kind of outcomes are likely to occur under certain circumstances.
The thing to remember is that there is never just one likely outcome for the future. In any given situation, there will always be a range of possibilities for what is likely to occur in the future. While there are no crystal balls that can tell you exactly what will occur in the future, the statistician can work out the range of possible future outcomes.
In the graphs below, you will see that there are three lines for the future portion of the graphs. These lines mark out the range of possible future outcomes. Some of them are quite low and some of them are quite high. This is because there is always a chance that the market could do exceptionally well or exceptionally badly.
However, the most likely outcome is the middle line on the graph. There's a 50% chance that the growth could be lower than that line, and a 50% chance that it could be higher.
Another factor that needs to be kept in mind is that even though there is a 50% chance of the growth being lower or higher than the middle "most likely" line, that doesn't mean that the potential for higher or lower growth is equal. So for instance, there might be a 5% chance of getting growth that's 10% higher than the middle line. This does not mean however, that there's an equal chance of getting 10% lower than the middle line. It might be that that there's an equal chance of getting only 4% growth on the lower side.
If this were the case, the gap between the middle line and the upper line would be bigger than the gap between the middle line and the bottom line.
For example, on the Sydney graph below, the gap between the "most likely" line and the upper dotted line is larger than the gap between the "most likely" line and the bottom dotted line. This means that if the future turns out to be lower, there's less potential for it to go very low. However, if the future turns out to be above the middle line, there's a great potential for this growth to be higher.
We can also see that after the quite large drop in growth over the last 18 months, our model is showing that the market has bottomed out and should start to move forward again.
Given this graph, we expect the growth to be in the order of 5% pa or better over the next two years for Sydney. Demand in the house and land market remains with land in short supply and a population in many cases "sitting on their hands" doing nothing and waiting for the Government to alter the tax regime. They are also avoiding debt due to uncertainty about the potential for interest rate and petrol price increases. The asset cost has reached a level where minor changes to these two items have a significant impact on the perceptions of people with respect to their borrowing capacity.
By contrast, Brisbane's outlook is a little bland. Unlike any other market, Brisbane has had a short, sharp decline in growth just over the last 12 months. This followed from an unprecedented high growth period. (This compares with, say, Sydney, whose growth took almost two years to completely slow down from a relatively smaller peak.)
The reason for this is simply because Brisbane has really had all the capital growth it should have expected over 10 years in a short two-year period. Brisbane needed to slow down because it could not possibly sustain the price levels given the land volumes available, and the loss of investment competitive advantage compared to Sydney and Melbourne particularly in the rental return stakes.
If these high price rises had occurred in another city, the outlook for the next 8 years may well have been one of negative or low growth. However, Queensland has been going through a period of high growth resulting from a resources boom and Brisbane has benefited from this. The boom, the competitive advantage in terms of house prices, and rent return have been driving house price growth and building activity, creating jobs and causing a boost to the Brisbane economy. Because of this resource boom, the risk of a significant correction is small and the outlook for the next year or two is small but positive growth. The graph above displays the results from the computer generated predictive model.
However, the reliance on the resources boom means that Brisbane is at risk from any adverse change in the resource-driven economy. The result is the exhibited higher potential of lower growth rates than the median position shown by the "most likely" black line. If the resources boom continues and things go well, we will see modest growth. If the resources boom halts, then the corrections will be noticeable. This is why the gap is larger between the middle line and the bottom line than between the middle line and the top line.
The models used to generate the graphs above form the basis not only of our general predictions for major cities but also in our prediction reports, where we identify the best likely performing areas over the next 5 and 8 years. These models have now become sufficiently advanced to allow us to develop predictions down to a suburb level and as a consequence for the first time all our prediction reports are now available for suburbs. You can purchase these here.
Next month we will release the quarterly Residex Report for each state and those reports will contain expanded versions of this information on our big picture city predictions for each major capital city. Additionally we will provide information on our predictions for interest rates, inflation and rental returns. To ensure you receive the Residex Report early, you can subscribe to an annual subscription here.
John Edwards
John Edwards is the CEO of Residex.
25/07/2005
At the end of every financial year, we do a major long-term forecast for the capital cities of the state. Over the next couple of issues of this newsletter, we will be presenting the forecast graphs of the major capital cities and seeing what the future holds for these different markets.
The common theme in all of the cities is that they have come off the boil in terms of property growth rates over the last 12-18 months. The hardest hit were Sydney and Melbourne. Brisbane had a milder slowdown but it was very sharp and quick compared with its southern cousins. These corrections are normal at the end of the housing cycle. However, these corrections, especially in Sydney and Melbourne, are lasting longer than normal.
While each city exhibits its own unique growth characteristics, there are broad similarities in terms of the timing of market corrections etc., but the amount of expected future growth will vary from city to city depending on its economic fortunes and government policies.
This month we will look specifically at the most likely outcome for Sydney and Brisbane.
The graphs below are produced by our computer prediction model. This is a stochastic model, meaning that we are modelling processes which are based on probability. In this case, our model is developed by looking at movements in the property market going back over 100 years. Some of the information we look at is what happens over short time periods (1 or 2 years), medium time periods (5-8 years) and longer time periods (9 years plus). Other information taken into account is duration of cycles, growth rates following periods of correction and population growth, etc.
From looking at these various movements, and other factors that influence the market, we can build up pictures of what kind of outcomes are likely to occur under certain circumstances.
The thing to remember is that there is never just one likely outcome for the future. In any given situation, there will always be a range of possibilities for what is likely to occur in the future. While there are no crystal balls that can tell you exactly what will occur in the future, the statistician can work out the range of possible future outcomes.
In the graphs below, you will see that there are three lines for the future portion of the graphs. These lines mark out the range of possible future outcomes. Some of them are quite low and some of them are quite high. This is because there is always a chance that the market could do exceptionally well or exceptionally badly.
However, the most likely outcome is the middle line on the graph. There's a 50% chance that the growth could be lower than that line, and a 50% chance that it could be higher.
Another factor that needs to be kept in mind is that even though there is a 50% chance of the growth being lower or higher than the middle "most likely" line, that doesn't mean that the potential for higher or lower growth is equal. So for instance, there might be a 5% chance of getting growth that's 10% higher than the middle line. This does not mean however, that there's an equal chance of getting 10% lower than the middle line. It might be that that there's an equal chance of getting only 4% growth on the lower side.
If this were the case, the gap between the middle line and the upper line would be bigger than the gap between the middle line and the bottom line.
For example, on the Sydney graph below, the gap between the "most likely" line and the upper dotted line is larger than the gap between the "most likely" line and the bottom dotted line. This means that if the future turns out to be lower, there's less potential for it to go very low. However, if the future turns out to be above the middle line, there's a great potential for this growth to be higher.
We can also see that after the quite large drop in growth over the last 18 months, our model is showing that the market has bottomed out and should start to move forward again.
Given this graph, we expect the growth to be in the order of 5% pa or better over the next two years for Sydney. Demand in the house and land market remains with land in short supply and a population in many cases "sitting on their hands" doing nothing and waiting for the Government to alter the tax regime. They are also avoiding debt due to uncertainty about the potential for interest rate and petrol price increases. The asset cost has reached a level where minor changes to these two items have a significant impact on the perceptions of people with respect to their borrowing capacity.
By contrast, Brisbane's outlook is a little bland. Unlike any other market, Brisbane has had a short, sharp decline in growth just over the last 12 months. This followed from an unprecedented high growth period. (This compares with, say, Sydney, whose growth took almost two years to completely slow down from a relatively smaller peak.)
The reason for this is simply because Brisbane has really had all the capital growth it should have expected over 10 years in a short two-year period. Brisbane needed to slow down because it could not possibly sustain the price levels given the land volumes available, and the loss of investment competitive advantage compared to Sydney and Melbourne particularly in the rental return stakes.
If these high price rises had occurred in another city, the outlook for the next 8 years may well have been one of negative or low growth. However, Queensland has been going through a period of high growth resulting from a resources boom and Brisbane has benefited from this. The boom, the competitive advantage in terms of house prices, and rent return have been driving house price growth and building activity, creating jobs and causing a boost to the Brisbane economy. Because of this resource boom, the risk of a significant correction is small and the outlook for the next year or two is small but positive growth. The graph above displays the results from the computer generated predictive model.
However, the reliance on the resources boom means that Brisbane is at risk from any adverse change in the resource-driven economy. The result is the exhibited higher potential of lower growth rates than the median position shown by the "most likely" black line. If the resources boom continues and things go well, we will see modest growth. If the resources boom halts, then the corrections will be noticeable. This is why the gap is larger between the middle line and the bottom line than between the middle line and the top line.
The models used to generate the graphs above form the basis not only of our general predictions for major cities but also in our prediction reports, where we identify the best likely performing areas over the next 5 and 8 years. These models have now become sufficiently advanced to allow us to develop predictions down to a suburb level and as a consequence for the first time all our prediction reports are now available for suburbs. You can purchase these here.
Next month we will release the quarterly Residex Report for each state and those reports will contain expanded versions of this information on our big picture city predictions for each major capital city. Additionally we will provide information on our predictions for interest rates, inflation and rental returns. To ensure you receive the Residex Report early, you can subscribe to an annual subscription here.
John Edwards
John Edwards is the CEO of Residex.
Comment