Brian Potter on housing construction productivity

One factor in the high cost of housing is that we’re not better at building houses; construction productivity is relatively flat (not only in the US, but in many countries). It’s often presumed that this is because housing construction has mostly failed to move inside the factory (this piece by Arpit Gupta and Steve Teles, which we’ve covered before, gives some reasons why that might be). But in a new blog post, Brian Potter is skeptical of this explanation.

Brian reminds us that factory built housing is actually a thing, in the manufactured home industry, and it’s telling that this sector has often not found it worth centralizing production in large automated factories. He also suggests there might be a low ceiling on how much automation can reduce the cost of housing, by looking at the so-called Idiot Index: the cost of a good divided by the cost of its raw materials. In housing, Brian argues this is roughly 2, compared to 1.8 in the highly automated car manufacturing sector. Maybe the reason construction productivity is flat here and around the world, is that it’s already near a local maximum? And yet, and yet - other work finds manufactured homes cost substantially less to build than comparable site-built homes.