House prices have picked up around the UK since the pandemic began (Chart 1). One notable trend is that they have increased less in London than elsewhere. This blog dives into the data to see what's been going on in London.
The first thing to note is that while London stands out when we look at an index of all properties that have sold, it stands out a bit less when we break it down by property type (Chart 2). It's clear that prices of flats have risen less than houses since the pandemic. And London has a much higher proportion of flats than most regions, which goes some way to explaining why house price growth has been slower than in other regions.
The preference for houses is likely to be driven by a preference for more space, as people are spending more time at home. We can see this when we break price growth down by both internal floor area and garden size. Chart 3 splits all properties in London into five buckets (aka quintiles), according to floor area. So the smallest 20% of properties are in the first quintile, the next 20% in the second quintile, etc. It then shows price indices for each quintile, with 2019 set to equal 100. It shows that larger properties have seen higher price growth.
Chart 4 takes the same approach for all properties that have gardens, but adds an extra line for properties that have no garden. Again, the larger the garden, the higher the price growth.
Chart 5 shows estimates of commuting times to the City, using public transport. And Chart 6 breaks down price growth using these commuting times. Again, it supports a post-pandemic narrative of people caring less about short commuting times, as they spend less time in the office. Of course, this is pretty crude, as people work all over London, but the results don't change much if we swap 'the City' for other bits of central London.
There has been a long-term trend towards the higher-priced areas of London being ever more concentrated in the centre. To see this, we can look at the price per square foot of property across London and how that has changed over time. Charts 7 and 8 show price per square foot in 1995 and 2021 respectively.
It takes a keen eye to pick out the patterns, so Charts 9-14 help spell this out a little. Like Charts 7 and 8, they plot price per square foot by region of London. But they only plot it for regions with a price per square foot higher than the median level for London (i.e. the more expensive half of regions in London). This shows pricing pressure drift towards central London through the long house price booms of the 1990s and 2000s, with the trend continuing after the global financial crisis of 2007-9 right up until Brexit in 2019, since which it has reversed slightly. For fun, each chart also shows the total area taken up by all these houses and their gardens, which has declined markedly as pricing pressure drifted towards central London.
There were likely a number of causes of the push towards central London. Foreign investment and inward immigration (both from other parts of the UK and from the rest of the world) likely both led to preferences shifting towards more central property. Those drivers abated with Brexit, at least in the short-term. Where they go in the long-term is unclear. But it seems likely that they could combine with a post-pandemic re-assessment of lifestyles that sees people continue to put more value on space and less on reducing commuting times. This could see prices per square foot converge further. Given the huge disparity in starting points (Chart 8), there is plenty of scope for the trend to run for a long time, if it does materialise.
The areas used in the analysis are Middle Super Output Areas (MSOA) from the 2011 Census. They are constructed by the ONS to be fairly socially homogenous. There are nearly 1,000 in London, containing on average 3,700 residential properties.
Quintiles of properties by garden size and floor area are calculated using all properties, not just those properties sold in a particular time period. This helps protect the analysis from oddities in the changing mix of properties sold through time and is made possible by our detailed database of property data.
Travel times are calculated at Lower Super Output Area (LSOA) level, but are shown in Chart 5 at MSOA level for consistency with the other charts.
Please feel free to get in touch for more details.