There is considerable commentary about the high level of house prices. There is general concern that an increase in interest rates may have a big effect on house prices.
We would like to develop a simple relationship we can use to do some projections to forecast what could happen to the general level of house prices, depending on how interest rates move in the future.
__What Impacts House Prices?__
There are many factors including:

Interest rates

Household income levels

Housing supply and demand

Immigration levels

Employment levels

Health of the general economy

The availability of home loan credit

And many more factors.

A complete model would incorporate all of these and would be reasonably accurate in forecasting house prices.

To simplify our model, we will solely focus on interest rates. This assumes that all other factors do not change and continue to have the same impact that they have in the past. This is valid for the time being as Sydney (and Australian capital cities generally) continues to be a city with a growing economy.

__10-Year Bond Index__

We will construct a 10-year bond index based on available data, which starts from June 1969.

To construct this index

We start with a 10-year bond worth $100.

We then move forward 1 month and see what our 9-year 11-month bond is worth at the new 10-year bond yield.

We sell the bond for this price and reinvest this capital to buy a new 10-year bond.

This process then repeats every month to construct the index.

Some observations here are:

If the 10-year bond is the same as last month, our bond will increase in value by approximately 1 month's worth of interest coupon.

If the 10-year bond is lower than last month, there will be an increase in value.

If the 10-year bond is higher than last month, there be a decrease in value.

The first effect will be additive to either the second or third effect each month.

So, even though rates may move up and down over time, each month we effectively earn some of the interest coupon, and we expect that this index will have a positive return over time.

*[NOTE: The exception to this is where we have a sudden large increase in interest rates and this index is measured only over a short time period.]*

__The Model__

Graph 1 at the end of the article shows:

Sydney Median House Prices from June 1969 to now;

the Starting Value for the Sydney Median House Price at June 1969 and increased this value by the 10-year Bond Index we constructed; and

also included the 10-year bond rate (on the Right Axis) for reference.

Note how correlated the Sydney Median House Price Index is to the 10-year Bond Index Adjusted Starting Value over the longer term.

There are periods where the Sydney Median House Price Index, is higher or lower than our projection using the 10-year bond index. However, this doesn’t last long as the two series tend to get closer relatively quickly.

__Future Interest Rates Sensitivies__

Assuming our Interest Rate Index Adjustment Model is reasonable, we will now project Sydney House Prices, under these five interest rate scenarios:

10-year bond rate increases at 0.0% p.a. for next 10 years (i.e. stays the same)

10-year bond rate increases at 0.5% p.a. for next 10 years

10-year bond rate increases at 1.0% p.a. for next 10 years

10-year bond rate increases at 1.5% p.a. for next 10 years

10-year bond rate increases by 5.0% p.a. in year 1, then 0.5% p.a. thereafter for next 9 years

The results are shown in the Graph 2 at the end of the article covers the 10 years from October 2011 to October 2021 and adds the 10-year forecast to the end of that period.

Interesting observations of each case are:

Case a) house prices should continue to rise over the next 10 years

Case b) house prices will fall by about 5.1% by January 2026. By March 2030, house prices would come back to the same level as October 2021. So effectively we would have 8 years and 5 months of zero net growth

Case c) house prices will fall by about 16.2% by December 2026. By October 2031 (the end of the 10-year forecast), house prices would increase from that low, but still be 3.1% below the October 2021 level.

Case d) house prices will fall by about 24.7% by November 2026. By October 2031 (the end of the 10-year forecast), house prices would increase from that low, but still be 5.1% below the October 2021 level.

Case e) house prices will fall by about 30.0% by October 2022. By August 2029, house prices would come back to the same level as October 2021. So effectively we would have 6 years and 10 months of zero net growth.

__Conclusion__
Using the return of the 10-year Bond Index as being a reasonable model of Sydney House Prices seems to provide an ability to forecast the general level of future house prices.
We have presented five interest rate sensitivity cases to show the extent of price falls we can expect as 10-year bond rates change.
Assuming we have very low interest rates or very low increases (no more than 0.5% pa) then house prices should fall no more than 5.1% and recover by the end of 10 years.
Where interest rates fall more than 1.0% pa, or we have an interest rate spike early in the forecast, there is a high likelihood of a 15% to 30% fall in house prices before they bottom out.
*It should be noted we have assumed that all other market factors are unchanged, otherwise they could affect the validity of forecasts under the 10-year Bond Index adjustment we have used in forecasting.*

**Disclaimer**

*This document has been prepared by Boston Money Pty Ltd. Documents published on this website are for general information *__only__* and are not intended to provide you with personal financial advice as we do not take into account your personal objectives, financial situation or needs. To the extent permitted by law, no liability is accepted for any loss or damage as a result of any reliance on this information. Replication of this information requires prior approval and appropriate referencing to the entire document.*

## Comments