Higher housing prices due to COVID-19 will have a lasting impact on inflation and complicate the lives of central bankers for some time.
The COVID pandemic has caused a sharp increase in the use of telework, a significant part of which is permanent. Many statistical institutes have shown that this coincided with the departure of a certain number of people from large cities to smaller towns, to telecommute there. The impact on real estate prices is far from trivial, and might even have consequences for monetary policy.
The example of France
The COVID crisis has profoundly changed the way of working. It is now well documented that with teleworking, a number of people have left the big cities to live in smaller towns and that this has propelled real estate prices out of the big cities. If we take the case of France, INSEE figures indicate that the average price of apartments in Paris barely changed between the start of 2020 and the start of 2022, while it increased by more than 8% in Ile de France excluding Paris and 15% in the provinces. This marked a change in dynamic from previous years.
In the United States, more than half of the rise in house prices
We feel with the example of France that telework has affected property prices in a differentiated way depending on the area, but one of the questions that can be asked is that of the impact on property prices at the national. Two researchers from the San Francisco Fed carried out this work in the case of the United States. They worked on 895 geographical areas of at least 10,000 inhabitants, which covers a very large part of the United States. They have recovered for each zone the evolution of the share of employees practicing permanent teleworking: at the national level, this proportion has increased from around 5% before the pandemic to more than 16% in 2020. They have also recovered for each zone the net population migration, ie the balance of the population that moved into the zone and the population that left the zone. Finally, they also recovered for each zone the evolution of real estate prices calculated by the company Zillow and for a slightly more restricted sample the evolution of rents calculated by the company Apartment List.
A first part of the study shows that the share of the use of telework before the pandemic is strongly correlated with the increase in the use of telework during the pandemic: in other words, it is in the areas that were most suitable for telework before the pandemic that teleworking has increased the most, presumably due to specific characteristics, such as having a pleasant environment or different amenities.
Next, the authors show that the areas with the highest shares of telecommuting experienced the largest house price increases. By performing longitudinal regressions, they arrive at the result that each additional percentage point of teleworking observed in 2020 is associated with an additional increase in property prices of 1.5% over the period from December 2019 to November 2021. The results are very relatives for rents. An interesting point here is that the authors find no link between pre-pandemic price dynamics and the share of telecommuting among employees, suggesting that there was indeed an exogenous shock related to telecommuting during the pandemic.
However, as the authors point out, it is difficult to extrapolate what these results represent at the national level because the upward pressure on prices created by households arriving in an area to telecommute there may be associated with downward pressure on prices in the area they just left. The authors therefore sought to neutralize the effects linked to migration in order to isolate an effect “purely” linked to the search for real estate in areas suitable for teleworking. They conclude that net migration accounts for regarding a third of the price effect discussed earlier and that following controlling for the effect of net migration, an additional percentage point of teleworking observed in 2020 is associated with a increase in real estate prices of almost 1% over the 2020/2021 period. For the authors, the sharp increase in the use of teleworking would thus explain 15 of the 24% increase in average property prices observed in the United States between December 2019 and November 2021: more than half of the increase in property prices in this period might thus be explained by the increase in the use of teleworking.
The authors therefore believe that the rise in house prices in 2020 and 2021 reflects more of a fundamental development than a speculative bubble and that the Fed’s monetary policy would have been of only secondary importance. They also hypothesize that the future evolution of real estate prices will depend a lot on the evolution of teleworking. If telework declines, one might imagine a cancellation of the observed phenomenon… but if it persists and strengthens, the effect on property prices and therefore on inflation would be even stronger.
A challenge for the evolution of inflation over the coming quarters…
As we have just seen, the increase in the use of telecommuting seems to have strongly contributed to the recent rise in real estate prices in the United States. But it is important to stress that the magnitude of the price increase is extraordinary. Considering the long series compiled by Robert Shiller (which go back to 1890), we see that the only time when prices rose as quickly as in recent quarters was at the end of the Second World War. The real value of real estate is almost 15% higher than the peak of 2005/2006. The configuration is very different from that of the cycle of the 2000s, since the latter was characterized by a phase of overinvestment, whereas the 2010s were marked by historic underinvestment.
This is a subject that central banks must follow very closely. In the United States, rents (or equivalent rents for landlords) are a very important component of consumer price indices (CPI or PCE). The sharp rise in housing prices has been accompanied by a sharp rise in rents, which generally takes around a year to be taken into account in the price indices. The non-energy housing component of the CPI has steadily accelerated in recent months and in June reached its highest level since 1982. A recently published NBER working paper co-authored by Larry Summers predicts that “even if the rest of the CPI basket returns to the 2% target, housing will push the core CPI to almost 4% in December 2022”. According to this study, underlying inflation should remain quite high in 2023 due to real estate.
The long-term effects of the digitalization shock that took place during the COVID crisis are still poorly understood. Nevertheless, it seems that the sharp increase in the use of teleworking has had a very significant impact on the real estate market in various countries, and in particular in the United States. The resulting rise in property prices will have a lasting impact on inflation and complicate the lives of central bankers for some time. This is a clear example that societal changes can have significant macro-economic consequences.
