Single-Family Housing Starts Fell In January, Despite What The Headlines May Have Told You

Annualized Single-Family Housing Starts dropped 1 percent in January to 413,000 units nationwide, it’s lowest reading almost 2 years.

A “Housing Start” is defined as a home on which construction has started. 

Now, if you had only seen the Housing Starts story in the headlines today, you wouldn’t have known that single-family starts fell at all. It’s because of how the story is being reported.

Most commonly, newspaper headlines are reading something similar to “Housing Starts Jump 14.6%” with the lead paragraph making mention that “housing starts are at their highest levels in 4 years”.

It’s a true statement, but it’s misleading, too.

This is because, despite the Census Bureau reporting Housing Starts by property type — single-family, multi-family, and apartments — the media often lumps them into a single data set.

It’s a categorization that helps investors in homebuilder stocks, but it does little for everyday Walnut Creek home buyers. The huge majority of buyers aren’t buying multi-units or whole apartment buildings — they’re buying 1-unit homes.

Here’s how January’s Housing Starts broke down by type:

• Single-Family Homes : Down 4,000 units, or -1%
• 2-4 Unit Homes : Negligible change
• Apartment Buildings : Up 46,000 units, or +80%

Clearly, the surge in Housing Starts can be attributed to the rapid rise in the 5-unit-or-more sector. Single-Family Starts were weak, by comparison.

Even with all of this noted, however, we can’t even be certain that the January Housing Starts data is accurate anyway. A footnote in the government’s report shows that, although single-family starts are said to have decreased 1 percent, the data’s margin of error is ±8.6%.

This means that the true Single-Family Housing Starts reading may be anywhere from -9.6% to +7.6%. The data is throw-away. Housing Starts may have actually increased in January, but we won’t know until revisions are offered later this year.