I live in a part of south London called Balham. It’s about 6 miles south of the central area and part of a postal district named SW12 (the 12th SouthWest district in London). It was famous a long time go for being lampooned in a radio comedy as “Balham, Gateway to the South” but things have changed. I’ve lived here over 20 years and property prices have shot up and I wanted to analyse this.
The UK Land Registry provide details of every one of the 18 million property transactions since 1995 in one rather large CSV file which can be found here. This contains the price paid, the date of transfer, the postcode and the house number. An example of a postcode is SW12 0EN. The UK postcode narrows down to a set of about 20 properties, usually on the same street – together with the house number it uniquely identifies the property. There are over 600 postcodes in SW12 and over 1 million in the UK.
My friends at the Information Lab have kindly made publicly available a dataset that geocodes every postcode in the UK. The details are in this blog post. Geocoding ties a location, in this case a postcode, to a latitude/longitude so Tableau can map it. So now I can plot all the house purchases on a map.
This compares the property sales in prices in 1995 and 2013. The size of each circle represents the purchase price. Hover over a particular spot to see the sales details.
This motion chart shows the 14,000 property transactions since 1995, year by year. Running it through the years gives an impression of the activity, including the dip in 2008 due to the credit crisis.
I’m only interested in my neck of the woods but the Land Registry data obviously covers the whole country.
Obviously you can do a lot more with this data – for example filter by price paid only to look at properties within a certain price band. Or you can look at those properties that have been bought & sold several times in the last 18 years and use these to calculate average annual price increases.