Step One of a Housing Bottom

Last week’s report on residential construction provided more evidence that housing may be beginning to bottom. The headline evidence, noted by most media and economic pundits, is the rebound in housing starts over the past two months:

The rebound is from an extremely low level, so it’s hard to get too excited about it. But it does suggest that the plummeting of the past few years may finally be over.

As I noted last month, however, a bottom in housing starts isn’t a bottom in housing. From a macroeconomic point of view, the key thing is the amount of construction activity, which depends on both housing starts and housing completions. Not surprisingly, house completions plummeted along with housing starts, albeit with a lag reflecting the time needed for construction:

Completions have bounced around the 540 thousand level in four of the past five months, suggesting that they too may be putting in a bottom.

It’s crucial to note, however, that the number of completions (538 thousand in June) has been higher than the number of starts (470 thousand in June). As a result, the number of homes under construction is still falling:

Fewer homes under construction means fewer construction jobs and fewer purchases at Home Depot. So housing is still a drag on the economy.

The apparent bottoming of housing starts and completions should thus be viewed as only the first step in the larger process of housing making a bottom. The next step will be for home construction to stop falling

June’s data suggest that day may be coming. The gap between housing starts (which increase the number of homes under construction) and completions (which reduce the number of homes under construction) was the lowest its been since early 2006:

From a macro perspective, housing will have bottomed once housing starts have caught up with housing completions. (Of course, there’s also a third issue — when will there be a bottom in housing prices? — that I will address another day.)