This report analyzes the London real estate market behavior and applies the bubble methodology to capture the presence of bubbles in the last 6 years. Real estate bubbles are characterized by an explosive behavior of house prices, called as “boom”, followed by a fast decline of it, called “bust”.
In this report, a dataset composed by two variables will be used. The first one is the House Prices, and the second one is called Price-to-rent Ratio. The price-to-rent ratio is a well-established economic principle used for real estate valuation. It is typically calculated as the ratio of home prices to annualized rent in a given location. At its most basic level, the price-to-rent ratio is a benchmark for understanding whether is it better to rent or buy a property. The combination of these two variables results in a data set with a total of 71 observations for each variable.
The Augmented Dickey Fuller(or ADF) test tests the alternative hypothesis of stationarity or trend-stationarity behavior for a time series. Its statistic is a negative number. The more negative it is, the stronger the null hypothesis rejection. In the real estate market, a sequence of this test is applied to determinate where there is a greater probability of explosive behavior. However, there is more than one sequence type of this test. The two most used types of sequences are the backwards ADF sequence and the backwards SADF. The backward ADF sequence uses a fixed window and applies the test to the entire time series. The backwards SADF sequence also applies the ADF test for all the time series but it also varies the end point. Once one of this sequences is calculated, we need to determinate where is the bubbles. To do it, first a confidence level is fixed, usually 90%, 95% or 99%. In this report we are gonna use the most common one, the 95% confidence level. Then, a 95% critical value sequence is calculated and the periods of time with backwards ADF sequence values bigger than those is considered a bubble period.
In this session is presented the graph of the backwards ADF sequence with the 95% critical value sequence and the real time series (the price to rent ratio variable).
The graph below shows the backwards SADF sequence, the 95% critical value sequence and the price to rent ratio variable values.
This report also applies the bubbles methodology for the London house prices. The graph below shows the backwards ADF sequence for the house prices.
And finally, here there is the backwards SADF sequence for the London house prices.
Every time a value of the sequence exceeds the 95% critical line, we have a indicative of a bubble. The charts above are only two examples of sequence to detect bubbles. We can see that the Backwards SADF sequence has more sensibility than the Backwards ADF sequence at some points. Although that, analyzing the price to rent ratio behavior we can understand that the peaks around 2013 and 2014 are possible bubble points, captured by the backwards SADF sequence. As for the house prices, there is no greater peak, but this series presents a continuous growth with similar peaks all over it. However the Backwards ADF sequence shows some explosive behaviors due to the greatest values about 2015 and 2016. Taking a general looking we can affirm that the London real estate market indicated a bubble before the end of 2011. Since then the house prices has increased slowly and the price-to-rent ratio, after a big decline, tends to grow around 2014.