Last week the Financial Times used the latest Office for National Statistics (ONS) housing price figures to assess the impact of the stamp duty raise on the housing market. Since the beginning of the month, purchasers of buy to let property and second homes indeed face a 3% stamp duty surcharge. Some experts believed that prospective second home buyers would rush to complete their transaction before the deadline and would thus generate an artificial and temporary price increase.
The FT article took the opposite view and concluded that “UK house price growth weakened slightly in the year to February, to 7.6 per cent […] meaning that a boost reported by mortgage lenders and estate agents ahead of stamp duty changes for buy-to-let investors is yet to show up in official data”. This statement surprised me and I decided to dig further into the data. My take is the following: although the statement may be true at a global level, by performing the analysis for each Government Office Region separately, one could conclude that the impact of the stamp duty reform was actually much more significant.
To perform this exercise, I used two publicly available sources of data:
- ‘Number of residents with a second address in a region, who are usually resident outside of that region’ and ‘Number of usual residents in a region with a second address outside of that region’ as per the 2011 census;
- ‘Mix-adjusted average house prices by region’ published monthly by the ONS.
First, we need to understand that all UK regions are not equal. The more second houses a given area hosts, the more impact the stamp duty reform should have had. The most recent information on the topic comes from the 2011 Census which gives us the number of residents elsewhere with a second address in a given region – e.g. in 2011 184,616 people had a second address in the South East.
Separately, the ONS data gives us the year-on-year house price evolution in each area from July 2015 to February 2016.
Nonetheless we are not interested in the price growth rate but by the increase of the growth rate as the stamp duty implementation deadline approaches. I have therefore compared the average housing price increase (in percentage points) in January and February 2016 with the one witnessed in July and August 2015 in each region – I have taken a two-month average to smooth out any shock. The analysis shows wide discrepancies between regions – for instance the growth rate increased by 5.0 percentage points in London over the period but actually decreased by 2.7 pts in Yorks & Humber.
Finally, I look for a correlation between the number of second addresses and the increase in housing price growth rate.
The correlation between the two variables is clear and we could therefore conclude that the stamp duty change has ‘warmed up’ the housing market in the regions with a high number of second addresses. The stamp duty effect was indeed pushing buyers to complete the transaction as fast as possible, even if it meant paying a higher price (up to 3% more actually). The surge in gross mortgage lending witnessed in March by the Council of Mortgage Lenders supports this conclusion. On a side note, this is another piece of evidence illustrating the fact that the UK housing market remains a ‘seller’s market’ – in a ‘buyer’s market’ you would have conversely seen a drop in prices as sellers try to get rid of their property before having to pay the surcharge.
It would be interesting to keep an eye on the market prices in the future. Given that purchasers have by definition a limited purchasing power, was the bump just bringing forward future increases (in that case the market should cool off for at least a few months) or will future sellers be able to maintain enough competitive tension to use the ‘overheated’ prices as the new normal?