A housing bust got this recession started, so it is important to know when housing construction will resume at a normal pace. However, important determinants of housing demand have moved in opposite directions, so a housing recovery could be as close as this summer, or as far as 2011. The release later today of the House Price Index from the Federal Housing Finance Agency will provide insight into the relative likelihood of these scenarios.
The black line in the chart below shows an index of housing inventory per person at the end of each year from 1990 to 2008 (with housing inventory measured as square footage, adjusted for quality). Housing almost always increases faster than population, but the housing boom of 2002-06 stands out compared to the other years in terms of the relative rate of housing construction.
Housing construction will resume when housing demand exceeds housing supply. But that raises the question: What will happen to housing demand?
One simple housing demand forecast is to assume that housing demand will return to its 1990s trend, shown as a green line in the chart. This approach suggests that housing demand would match supply — and therefore housing construction would resume more normal rates — by the end of this year.
However, several things have changed since the 1990s, and are not likely to return fully to their 1990s situation. There have been two recessions since 1999, which will leave 2009 family incomes about 6 percent below the 1990s trend. Presumably housing demand grows more slowly when incomes grow more slowly.
Based on the income growth alone, it looks as if housing demand would be 7 percent below supply (104 rather than 112) at the end of 2009 (see the red line in the figure), and that the excess supply of houses will be at least as much this year as last year. According to this approach, it could take until the end of 2011 for housing construction to resume, and for now housing prices would continue to fall sharply.
Although the banking and real estate industries have received deserved attention for their failures, they have also succeeded in some ways since the 1990s, and those successes will not be erased. Information technology has advanced, and can help lenders assess borrowers’ ability to pay, as well as help home buyers search for a home and for a mortgage. Thanks to this progress, housing demand might be able to stay above the trend of the 1990s.
The housing boom may have left some homes in the wrong places, such as Arizona, California, Florida, and Nevada. If so, then housing demand may soon catch up with supply outside of these areas, while housing long remains depressed in them. The national total of housing construction could therefore resume earlier.
The housing market itself shows a few signs that prices and construction are not falling as they were in 2008. After declining several months in a row in 2008, housing starts in February and March 2009 are above what they were to begin the year. Pending home sales increased in February. After falling 10 months in a row, the House Price Index rose 1.7 percent from December 2008 to January 2009 (not surprisingly, housing prices continued to fall in the Pacific region).
Thus, the release later today of the office’s measure of February 2009 prices is of significant interest. If it fell more than 2 percent from January, this suggests that the 2008 rate of decline continues and 2009 may not be a year when many new homes are built. Otherwise, we have more support for the conclusion that the 2008 housing crash is largely finished (for now) in most of the United States, and that the next step is for resumed housing construction.