In the well known fable the slow and steady tortoise won its race against the quick but inconsistent hare.
We all know that some local property markets offer more consistent returns than others. But what relationship exists between inconsistency (or volatility) in residential real estate returns and the returns themselves? Is it better to invest in a ‘slow and steady’ location, where prices move incrementally upwards and don’t swing much, or is it better to risk the highs and lows of a more volatile market?
It’s obviously a complex question, but it’s much easier to approach with access to the correct datasets.
For example, with our proprietary £/sqft transaction price dataset for the UK, we can return the monthly median £/sqft value for all property types, by postcode district for the past 25 years. With some basic spreadsheet work, we can create a time series of the YoY historic returns, the 12 month rolling standard deviation in returns, and the correlation between the two:
Of course the eye immediately starts looking for patterns, like turning points in prices or returns. While we can see certain historic events in the national data, at a local level we could actually find no consistent strategy which related volatility to prices, and which worked regardless of location or timing.
However through this process we did discover some relationships which are interesting, and potentially actionable:
Firstly we decided whether a location was in a ‘High Volatility’ or ‘Low Volatility’ period based on the 12 month rolling standard deviation, relative to the previous two years rolling standard deviation (effectively 36 months). To put this another way: ‘if the past twelve months’ average volatility in this location is higher than the same for the previous twenty four months, then the location has relatively high volatility’.
Across our dataset there were 177,000 instances of ‘high volatility’ and 209,000 instances of ‘low volatility’. We looked at what the 12 month returns were across all locations and time periods, after a period of high or low volatility.
On average we found that in the 12 months following a period of low volatility, returns were 7.7%. However after a period of high volatility returns were 8.8%. The average area had 92 periods of low volatility and 78 periods of high volatility over the last 25 years.
Based on that result, investing during a period of relatively high volatility would appear to provide a slight outperformance relative to investing during periods of low volatility.
Below is a distribution of the frequency of returns in low and high volatility periods. As you can see the higher volatility periods (blue) have higher frequencies in the higher forward return periods.
However these high volatility periods are, as previously stated, less frequent than low volatility periods.
Finally we found that this insight could be optimised a little further. In areas which tended to have relatively low volatility, when they did have higher volatility periods, the 12m returns were higher still, achieving an average of 9.1%.
Today there are 28 postcode districts which fit that criteria, and seven of them are in a surprising area of London. To try this yourself, contact our sales team at email@example.com,and ask them about our Data as a Service offering.
© Treex 2020