The US economy got crushed in the second quarter, with the worst decline in real GDP for any quarter since the Great Depression. However, the long road to recovery has started and, for now, we're penciling in real GDP growth at a 20% annual rate for the third quarter. Of all the parts of the US economy that have weathered the COVID-19 storm, none has been as resilient as the housing market.
Homebuilders started homes at a nearly 1.6 million annual rate in December, January, and February, before the Coronavirus and government-mandated shutdowns wreaked havoc. Those were the best three months since 2006 and showed that residential construction had finally fully recovered from the housing implosion that was a center point of the last recession.
Then, during the shutdowns, homebuilding plummeted: housing starts bottomed at a 934,000 annual pace in April, before gaining in May, June and July, hitting an almost 1.5 million pace last month.
We have been saying for the past several years that the fundamentals of the housing market suggest an underlying norm of 1.5 million housing starts per year. This is based on a combination of population growth (more people mean more housing) and scrappage (homes don't last forever, either because of voluntary knockdowns, fires, floods, hurricanes. tornadoes,...etc.).
However, in the ten years ending in February (March 2010 through February 2020) builders had only started 1.011 million units per year. Part of this made sense: home builders started too many homes during the housing bubble and the only way to work off that excess inventory was to build fewer homes than normal. But, in our view, the inventory correction went too far. In the 20 years through February (March 2000 through February 2020), housing starts only averaged 1.265 million. Too low.
All of this suggests to us that home builders still need to make up for lost time, until the long-term average is closer to 1.5 million per year, which could mean reaching, and then averaging, a pace of something like 1.8 million starts for the next several years.
But it's not only home building that's recovered so quickly; home sales have revived, as well. Existing homes were sold at a 5.76 million annual place in February, the fastest pace since the housing bubble burst. Then sales plummeted in March, April, and May, bottoming at an annualized pace of 3.91 million, the slowest since 2010. Since May, however, sales have soared, hitting a 5.86 million annualized pace in July, even beating where we were in February.
Part of the recent gain was likely pent-up home purchases: people who wanted to buy earlier in the year but got temporarily thrown off track by the Coronavirus, massive economic contraction, as well as general uncertainty. But including the drop and the rebound, the average pace of sales in the last five months (March through July) is still slow, suggesting some further gains ahead. Ditto for new home sales, although neither existing nor new home sales will grow every month.
In terms of prices, we expect national average home prices to continue to grow, but with a wide dispersion. Dense cities hit hard by COVID-19, or which have seen social unrest (or both!), especially with the newfound ability to work remotely, are going to be relative losers; other metro areas are going to experience faster gains.
Yes, a Biden win in November could end up expanding the state and local tax deduction, helping some beleaguered cities. But that election outcome is not assured. The enlarged standard deduction would still mean fewer people itemize, and the Biden campaign wants to limit the "value" of itemized deductions to 28% (instead of a proposed top tax rate of 39.6%).
Bottom line: housing is going to be a significant tailwind for the US economy overall, but not everywhere.
Brian S. Wesbury - Chief Economist
Robert Stein, CFA – Deputy Chief Economist
Click here for PDF version