By Bill McBride, guest columnist and author of the Calculated Risk blog.
The U.S. Census Bureau and the Department of Housing and Urban Development reported this morning that housing starts declined to 708 thousand in May, on a seasonally adjusted annual rate (SAAR) basis, down from 744 thousand in April (April was revised up from 717 thousand). Single family housing starts increased in May to a 516 thousand annual rate, up from 500 thousand in April. The decline in total starts was related to the volatile multi-family sector, and overall this was a fairly strong report with an increase in single family starts, a sharp increase in building permits, and upward revisions to prior months.
For excerpts from the report and graphs, please see Housing Starts in May.
Two key points:
- Total housing starts are up significantly from the record low of 554 thousand starts in 2009. This nascent housing recovery is starting to make a positive contribution to both GDP and employment growth.
- Housing starts and completions have been at record low levels over the last four years. This low level of starts and completions has significantly reduced the number of excess vacant housing units in the US.
The Census Bureau started tracking housing starts in 1959, and the four years with the fewest starts were the last four; 2008 through 2011. Starts this year will probably be at the highest level in three years, but still very low historically. Total starts are on pace to be up about 30% this year from the bottom in 2009, and single family starts are on pace to be up about 15% from the record low 431 thousand single family starts in 2011.
Several Factors in Increase
This increase in starts is being driven by several factors. First, the demand for rental housing has increased sharply due to favorable demographics (a large cohort is moving into the 25 to 35 year old age group), because people who have lost their homes to foreclosure (or short sales) are turning to renting, and some people have probably stopped "doubling up" with friends and family. This has pushed down the rental vacancy rate, and pushed up rents. The builders have responded by starting more multi-family housing units. (Also see: Some thoughts on Apartments and Rents.)
For single family starts, the excess vacant housing supply is starting to be absorbed in some areas. This is more difficult to quantify because the best data on housing units is available once a decade from the decennial Census, and is quickly out of date. Also there is no timely measure of household formation in the US. However using the 2010 Census data, and extrapolating forward — and factoring in record low completions, and some increase in household formations — would suggest the excess supply has declined sharply over the last two years.
Now, instead of being a drag on GDP, residential investment — a category that includes new single and multi-family construction along with home improvement and brokers' commissions — will add to GDP this year for the first time since 2005. This is the end of six consecutive years of drag on GDP from housing, the most consecutive negative years since the Bureau of Economic Analysis (BEA) started tracking residential investment in 1930 during the Depression.
The increase in housing starts will also lead to an increase in construction employment this year too. Last year, construction employment increased by 69,000 jobs, the first increase since 2006. Construction employment will probably increase further in 2012.
And the housing recovery is just getting started. Even with the severe recession, and many households doubling up because of hardship, the US still added 11.2 million households between 2000 and 2010 according the decennial Census. And the US added 13.5 million households between 1990 and 2000. Looking forward, the US will probably add around 12 million household this decade, and if there was no excess supply, total housing starts would be 1.2 million per year, plus demolitions and 2nd home purchases. So housing starts will probably come close to doubling from the current level over the next several years.
However, in the near term, the recovery in housing will probably remain sluggish because of the high number of distressed sales, and the fairly weak employment growth. Still the change from housing being a drag on the economy to adding to economic and employment growth is significant.
Bill McBride is the author of the Calculated Risk blog that he started in January 2005. He is best known for writing about the housing bubble and predicting the collapse in house prices. Previously Mr. McBride was a senior executive and on the board of directors of a small public company based in Orange County, California. Mr. McBride holds an MBA from the University of California, Irvine.