This time I am going to analyze a housing bubble that Australia is currently facing. I did it once to model a Real Estate (RE) demand in Israel and wanted to extend it to any other country, which RE market is going through the similar behavior. Well, apparently, it is going to be Australia.
Google Trends is exposing unique data of demand changes and as a result various micro and macro-economic derivatives. I tried to use this tool to model RE market dynamics in Israel and later in other countries and found interesting behaviors of different markets. In this work we will try to understand a bit better reasons for growth in RE prices, we will try to understand better the market, compare RE markets models in different countries, will try to get tools to predict the market based on different scenarios, we will see the RE price hysteresis and will try to explain it.
First part of the work is only exposure of data with no/minimal conclusions or analysis.
Databases (period 2004-2013):
Google Trends (Weekly. Based on keywords “property for sale” and “houses for sale”)
Real Estate Prices table (Quarterly. Australian Bureau of Statistics)
Home loan lending Rates (Monthly. F05 Indicator)
Central Bank Interest Rates (Monthly. Reserve Bank of Australia)
Australian Inflation rates (Reserve Bank of Australia)
Real Estate Demand data analysis
Let’s check the Google trends data for search keywords “property for sale” & “houses for sale” as well as “Mortgage” & “home loan” in Australia. Those indexes represent RE demand indexes and actual RE purchase phase. We will sum first two to create a “RE Demand Index” and last two to create a “Mortgage Demand Index”:
I. Demand for housing and mortgage over time. Split at some point.
It is important to mention that those numbers are not absolute, but normalized and relative to max within the examined period (i.e. k*Q instead of Q).
You can see that till some point in 2009, those two indexes followed each other and since then are separated. It is easier to see if we plot one versus another:
II. Comparison of apparently dependent demands and change of behavior at some point of time (red)
You can wee black dots that are till Jan 2009 and red dots since then. Very similar behavior you can see in Israel (same data types, just in Hebrew and for Israel):
III. Demands for Mortgage (left blue) and RE (left red) over time and one versus another (in Israel)
Why do we see two different behaviors of consumers? It seems like a panic or euphoria that is causing “abnormal” explosion of demand for RE (even though people not actually buying property, but only willing to do so very much).
If we plot “Price Index of Project Homes: Weighted Average of 8 Capital Cities” versus “RE demand” index (a.k.a. “Demand curve”) we will get the next one:
IV. ”Demand Curve” of housing in Australia
We can see that the after the Jan 2009 demand went up and prices has followed it in non-linear matter. For those that are familiar with Demand Curve, it should look familiar.
What had happened at that point in time?
To show you that you are not alone, here is what happened in Israel:
IV. ”Demand Curve” of housing in Israel
One of most important factor that impacts the demand for RE is an Interest rate (obviously as people need to pay mortgages according to it), so let’s plot a price versus Australian Central Bank Interest Rate:
V. Homes price – Interest Rate dependency and RE bubble triggering moment
You can see an interesting behavior when higher IR 2004 to 2008 caused higher prices! I thought that this is because of Real IR (Bank IR minus Inflation), so I have calculated a real IR based on Inflation rates (by Reserve Bank of Australia) and, indeed, this can partially explain the behavior (unless you argue with that):
VI. Homes price – Real Interest Rate dependency and RE bubble triggering moment
While the Real IR, which is a very tricky indicator, is:
VI. Real interest Rate over time = IR – Inflation. Going negative at 2009 and approaching zero at 2014
You can see that up until mid of 2006 the price was opposite to a Real IR (Congrats, Australians – it is not trivial that people follow up a real IR!). In 2006, even though the central Bank IR increased a bit, the Inflation rate is dramatically dropping during 1.5 years (until end of 2007) which caused significant growth of real Interest rate to its peak value, but prices did not drop and even got higher (V green). Yep – I am going to need an explanation here!
And then something interesting is happening.
During 2008 the inflation rates are going the peak of 5%, but what was a major trigger for explosion is a sequence of Central Bank IR reductions from 7% to 3% at the mid of 2008 within 0.5 year, crossing point of Inflation (and making the real IR to be negative for the very first time). This is the point where the Demand explodes (V and IV red dots).
In Israel, the Central Bank was frightened by the explosion of prices and tried to raise IR back, but prices DID NOT FAIL. This is a phenomenon of “Hysteresis” that I will try to describe the model in the next part, once I will write it – or you can read it in my second half of previous article in Hebrew )))
VII. RE Price Hysteresis: Price as function of IR (Israel). Also refer to chart V
Well, it says that even if Central Bank is raising an Interest Rate, once triggered, the price will not go down back on the same path it climbed and there is a need in different type of instrument to avoid the development of housing bubble. Why? because of type of demand curve and location of equilibrium. Stay tuned and do not forget to subscribe/like! 🙂