Can Homebuilding Ever Become Efficient and Affordable? Brian Potter's Book Has Disturbing Answers
- Craig Webb
- 3 days ago
- 6 min read
By Craig Webb, President, Webb Analytics
Brian Potter used to believe that factory-built homes were the key to solving America's home building and home affordability woes. Now he knows better, because he's taken a phenomenally deep dive into how to build anything better, faster, and cheaper.
What Potter learned results in a just-published book, The Origins of Efficiency. In it, he analyzes dozens of the world's greatest product achievements and explains why the efficiencies that made mass production of other goods possible have been so difficult to duplicate in homebuilding. He also concludes that factory-built homes as they currently exist merely nibble at the edges of improving quality and costs.
Achieving the efficiencies that match the lower costs and higher outputs at other industries will require massive changes in how we operate today. One vision he presents involves squads of robots, massive computer power, materials being manufactured at the job site, and perhaps the abandonment of 2x4 framing lumber.
That world is decades in the future. In the meantime, Potter suggests the most logical way to make housing more affordable is mainly through reg reform, with modular housing providing modest help.
Birth of the Blues
The Origins of Efficiency grew out of two experiences. The first was his years with Katerra, the multi-billion-dollar attempt to transform construction. Having been a structural engineer for 10 years, Potter was excited by Katerra's vision. But in a little over three years, the company crashed.
“I came to believe their thesis that if you just build these buildings in factories, it would be much cheaper. It was not really correct," Potter said in an interview. "And if you look at the history of the construction industry, you see lots of other businesses basically trying the same idea and not really succeeding at it. You can build a successful business prefabbing houses, but in general, outside of a few exceptions, no one has really been able to dramatically and broadly reduce the cost of construction.”
"I became obsessed with understanding why things had gone so wrong, and what it would take to actually make the construction industry more efficient," Potter wrote. That led him to create Construction Physics, the Substack column where more than 67,000 subscribers go for his weighty analyses into questions like why construction is scared of innovation and why bricklaying machines aren't the norm. Now comes the book.
What Makes Efficiency Possible
Potter's Origins of Efficiency argues that inventions are wonderful, but they don't help humanity until other people figure out "the step-by-step series of transformations" to make that invention widely available at an affordable price. Take penicillin: Discovered in 1928, the original way it was produced yielded such small amounts that it took a year to create enough penicillin to treat just one patient. But over time, scientists found better growth materials (involving, at one point, moldy cantaloupe) and engineers developed better production methods. By 1943, output reached 80 million units per month. A year later, production jumped to 18 billion units per month, while the cost of the drug had plummeted.
Agriculture, textiles, toys, books, you name it: The lifestyle we enjoy today comes thanks to people who made inventions available to all. And in general, Potter says these advances come from a few key factors. Among them are: Economies of scale, production process technology improvements, and cutting out unnecessary steps. All three work against housing.
By scale, Potter emphasizes big numbers. “Sweden often gets held up as a model the U.S. should emulate because roughly 80% of their houses are built in factories," he said in an interview last month. "But if you actually look at like their costs, they’re actually a little bit more expensive than U.S. houses which are built on site by hand using 'traditional methods.'" Why? Labor costs are one factor, but a bigger one is that Sweden struggles to start 50,000 homes in a year. That's not a big enough scale. One U.S. example where scale did work was Levittown, the enormous post-war housing project created in Long Island after World War II. But that combination of pent-up demand, limited options, enormous land, and scant regulations is nearly impossible to duplicate today.
Improving Production
As for the production process, consider these six qualities at right that Potter lists as vital elements. Together, they made it possible to increase the rate for producing the glass used in light bulbs from 150 per hour in Thomas Edison's day to 120,000 an hour by the early 1980s. But in home building, each of these factors fall short.
In fact, the construction process dominates Chapter 10 of the book, which Potter devotes to "Failures to Improve."
"What prevents a production process from getting cheaper over time?" Potter asks in the book. "There are, broadly, two reasons this might happen. First, the process improvements discussed in the previous chapters might be blocked. It might not be possible to find better production methods, use cheaper materials, or make the process more reliable. If process improvements can't happen, costs won't fall. Second, a process might accumulate additional, uncontrollable costs over time, offsetting any efficiency improvements and resulting in steady or increasing costs even as the process becomes more efficient."
These blocked paths can be because technology hasn't solved a problem yet. They could be market-related, such as when you can't achieve economies of scale because of low production volumes. Or the technical problem might be that factory-built houses are big, heavy things with low "dollar density" and thus relatively expensive to transport. A 20-foot shipping container can hold $10 million worth of iPhones but only $30,000 worth of a single-family home, Potter notes.
Policy Can Be a Drag
Then there are political blockades, such as NIMBYism, homeowner association covenants, land-use fees and licenses, building trades' opposition to prefab construction, mandates to install (e.g. smoke detectors) or not install (e.g. gas ovens) particular products, permitting requirements, and prescriptive building codes. Purely political motives also can figure in such as a requirement that multifamily builders in Chicago doing city-supported work must do expensive research to establish that their company never profited from slavery.
New technology also requires a shakeout period, to which Potter spends a chapter explaining. But in home building, "the introduction of a new process is quite risky," he writes. "In general, construction projects have a fat-tailed, right-skewed distribution of outcomes. ... The potential upside of a novel production method is relatively low, whereas the potential downside is extremely high."
An efficient production process takes away impediments, just as Toyota did with its famed process improvements. But housing seems to get encounter more speed bumps all the time. To Potter, the blockades our industry needs to work on first are regulatory.
How to Change
Most building product manufacturers have been able to implement many of the process improvements that other industries use, and they're already producing at massive scale. Thus, "there's not an amazingly obvious path to me to make materials more cheaply," Potter said in the interview. "It's also hard to use fewer materials in in a home. The most obvious gains have already been made."
So that leaves regulations and labor as the nuts to crack. Potter's vision of future home construction starts with automation: software that can process millions of variables to come up with a home design that meets the buyer's family's needs as well as society's and the building code's requirements.
Once the design is set, an "automated foundation installation truck" arrives at the jobsite and drills several large steel screws that will provide the home's foundation.
"Now a second truck arrives on-site, this one packed with a dozen humanoid robots and various pieces of construction equipment," Potter continues. He believes robots that could do almost anything a typical crew member does now ultimately could reduce labor's share of construction to 15% from the roughly 50% it is now. (In the auto industry, labor is about 10% of the total construction cost.)
As for materials, "the home’s structure will consist almost entirely of thin sheets of steel, bent into the proper shape by truck-mounted sheet-bending equipment: steel box sections for the heavy beams that span foundation posts, lighter C-shaped joists for the girders, steel roof trusses, and so on," Potter writes. "Each one is fabricated on the truck, moved into position by a pair of robots, and quickly welded in place with welding guns mounted to the robots’ arms. The walls, floors, and roof consist of steel triple-sandwich panels. These are made by taking a sheet of steel and placing a layer of expanding foam insulation on top of it, followed by another sheet of steel."
In other words, no wood is used on the building envelope.
Then comes a truck full of prefabricated windows and doors, which the robots again will install. Other robots might be called on to install HVAC, plumbing, and electrical.
"No home inspection is needed; pictures and other data were automatically recorded during the fabrication process, and as installation took place, a variety of sensors (visual, thermal, and so on) monitored progress to ensure that everything was connected and working properly," Potter says. Before leaving, the robots check each switch, cabinet door, faucet handle, toilet flush, and every other home feature, recording the result. The data is fed into the local jurisdiction’s automatic checking software, which instantly verifies that everything is working and signs off on the home. The entire process, from starting design to being handed the keys, has taken less than 24 hours."