Total housing starts fell slightly in July because of weakness in multi-family. But starts of single-family homes increased to 490 thousand (at an annual rate), the fifth straight monthly increase and the highest level since last October.
As the chart shows, this rebound is off of extremely low levels, so we shouldn’t get too excited. But it does appear that single-family starts bottomed last January and February (at 357 thousand).
That’s the first step of a housing bottom.
As I’ve noted in previous posts, however, that isn’t enough to declare a bottom in housing activity. Housing activity depends on the number of houses under construction, which depends on both housing starts and housing completions. Completions have exceeded starts for more than three years. As a result, the number of houses under construction has fallen for 41 straight months.
For me, the big news in the July data is that this decline may be ending.
Single-family starts were essentially equal to completions in July (completions were 491 thousand, just a smidgen more than starts). In other words, the wide gap between completions and starts largely disappeared in July:
As a result, the number of houses under construction was essentially flat in July:
If construction activity does indeed level off, or even rebound slightly, in coming months, that will be the second step of the housing bottom. (The third step — a bottom in house prices — is a topic for another day.)