THE RECESSION set off by the covid-19 pandemic has strangely coincided with surging asset values, and residential property is no exception. Taking an equal-weighted average of 16 rich-world markets, real prices have risen by 5.4% during the most recent 12 months with available data, compared with 2.2% a year earlier. In America values in the 12 months to January jumped by 11%, the fastest rate since 2006.
Does the boom have room left to run? Bulls and bears both make strong cases. Buyers note that fiscal stimulus and lower consumer spending have left household balance-sheets unusually healthy; that remote work has encouraged people to prioritise their living environments; and that interest rates are near all-time lows. Sellers counter that those rates have been rising quickly; that mounting debt will force governments to tighten their belts; and that rents in big cities have fallen sharply.
To help sort through this uncertainty, we have fired up The Economist’s statistical forecast of house prices. Built in 2019, our model uses a “random forest”—an algorithm so named because it blends the output of thousands of “decision trees”, each containing a series of yes/no questions like “is inflation above 2%?” It is trained on a dataset of national price averages going back to 1990, and uses 25 different variables, such as price momentum, household incomes and credit growth. During the covid-19 pandemic, financial gyrations have caused its short-term forecasts to be unusually inaccurate. But once economies open up, the factors that have historically preceded rising or falling house prices should regain their predictive power.
Giving equal weight to each of the 16 countries for which the model has data, its central estimate is that the rally is likely to stall but not reverse. It expects appreciation to slow to 3.4% this year, and to 0.7% by 2023. Much of this deceleration comes from America and Britain. However, because prices are currently soaring in those markets, the model projects them to climb a bit further before reaching a plateau.
Elsewhere, the model does expect price declines. It is most dubious on Spain, which faces a mix of deflation, high unemployment and stagnant incomes. The most likely scenario is that prices there eke out paltry gains in 2021 and then turn south. Sweden and Canada also have unemployment rates above 8%, and the model predicts they will start facing losses in 2023.
Among big markets, the model is most bullish for the next two years on France and Germany, where solid fundamentals back up recent price increases. Yet by mid-2023, it expects annual appreciation of just 1.2% and 1.6% in those countries.
The longest-lived bright spot is Ireland, predicted to lead the table in 2023 with a gain of 2.5%. The model does not explicitly count Brexit-driven relocations to the EU’s lone English-speaking country. Instead, it is impressed by Ireland’s healthy numbers on jobs and incomes. Moreover, household credit in Ireland is far below the rich-world average, implying that buyers have room to increase borrowing. Irish homeowners may still be scarred by their losses a decade ago, but the data suggest that the Celtic Tiger looks poised to reawaken.■
Source: The Economist house-price forecast model. See economist.com/house-price-forecasts for details
This article appeared in the Graphic detail section of the print edition under the headline “Approaching the ceiling”