An idea I frequently encounter from urbanist folks is that a barrier to the efficient mass production of housing is the amount of variety people demand in their housing.
Interesting overview, and here’s a historical tidbit to further bolster your point: the old Sears Roebuck home kits came in a limited number of styles/layouts but were quite popular in the early years of the last century. My tiny Midwestern hometown had several. Apparently they were delivered by rail. (Not sure how inclined you’d be to dig deeper into the history there, but they clearly filled a niche.) Decades later those I saw seemed to be generally holding up well and had been modified extensively.
I think the idea is true to a very limited degree: yes, people prefer unique and distinctive houses but aren't willing to pay for them a significant premium. If cookie cutter mass production would bring a large productivity boost, that style would be dominant across all market segments except luxury/high end - because it would trickle down as cheap housing.
The fact that we don't see this dominance and that cookie cutters don't sell at a lower price than a comparable one off house suggests that the productivity gains of repetition are very limited. Or, in other words, that current production methods cannot exploit repetitive patterns to lower costs, unlike say the electronics or auto industries.
You are missing the point, SB. Municipalities are the culprit here. They don't want "those people" living in their city limits. I can build affordable housing forever, if the cities weren't in the way. Aesthetics are form of structural discrimination.
There were significant gains from repetition that made houses cheap, but it all happened at least a hundred years ago (we went from six or more people to a room, to two or three), and there are no more gains to be had from that source alone.
I also remember reading articles a few years ago that said, if you subtract out semiconductors, the productivity growth of manufacturing since the 1960s is broadly similar to that of services like retail or transport. The gains from repetition were exploited early in auto manufacturing also, and productivity growth since has not been stellar.
Why are one-off houses now not much more expensive than cookie-cutter? I think this is a recent development, and it is due to CAD and a lot of other software that has brought down design and permitting costs, in part by having a library of standard modules and allowing combinations of them. Truly "blank slate" designs are still expensive, I believe.
> this is a recent development, and it is due to CAD and a lot of other software that has brought down design and permitting costs, in part by having a library of standard modules and allowing combinations
Even if true, that would still be a rephrasing of Brian's argument: he's debunking that housing variety is the problem, while you say that even one-off houses aren't that varied, they are just small variations of a archetypical "house" product just like the various color, upholstery and trim option are the same basic car model.
Still, I don't think that stands up to scrutiny. The soft costs of even extravagant buildings are a small part of the construction budget, while most of the money is claimed by tasks that involve large amounts of manual and/or skilled labor on site (ex. electrical, plumbing, concrete). There is very little you can do to mass produce foundations on site even if they are a standardized shape. You have some learning curve effects and maybe create some rigid reusable formwork, but the productivity gains are limited and still need to build each one with traditional methods, you will definitely not have an assembly line of foundations like you have for cars or iPhones.
I think of the old adage "faster, better, and cheaper, pick any two."
The industry seems to do "faster" pretty well. "Better" (performance, materials, layout, design, aesthetics) might be possible but it isn't going to be cheap.
I'd happily wait a few years if cheaper and better were realistic options.
I think this mainly depends on culture, and on whether or not platting still exists.
If all newly developed land is sold in lots measured in square miles, then of course the only takers are large developers that come in with a few cookie cutter designs.
If land is subdivided and then sold in some fraction of an acre then individuals can come in with an architect and build their snowflake designs.
The latter is quite common in Belgium. There is a photo blog called ugly Belgian houses that collects the more infamous specimens. This is some odd cultural trait over there.
However. If those custom designs were expensive they would be a lot less common than they are.
Around me in New Zealand, cookie cutter designs are built in exactly the same way as snowflake designs. Foundations, well, what are you going to do? Then, wood frame. It is a pile of timber and a few guys nailing it together just the same. Same plumber and electricity vans. Flooring? Another van. I am not sure how much it matters to the flooring guy if all the floor plans are identical.
Much of the cost savings come from making stupid value engineering choices, because the developer doesn't have to live in it, so whatever.
Maybe the savings come from mass producing parts. Mass produced timber, windows, floor boards, etc.
A decade or so back I did a project on technology diffusion in residential construction. Drywall, with its clear productivity benefits over lathing plus plaster, too 30 years. Some was that early materials weren't that good, but part was that the people who actually did the walls had to learn to work with the new materials. I strongly suspect that the "big" builders work with local crews rather than taking their own from project to project, so the fragmentation is even greater.
My particular interest was improved HVAC, heat pumps with inverters and variable refrigerant flows. Those can work with forced air systems, but work better better with room units. That approach is the norm in (say) Japan. Architects didn't know how to design for those, I worked with both the HVAC company and an architect, to no avail.
Now such systems wouldn't lower the cost of a finished standalone house, and realtors are uninterested in trying to learn how to get US home purchasers to pay more up front for lower utility costs.
Geography may be a factor, roofing trusses and engineered wood joists probably have different suppliers due to shipping costs.
The underlying story is yours, that fragmentation makes anything new diffuse slowly. My add is that the value chain is fragmented, and skill sets are no longer gained through formal apprenticeship programs so are even slower to evolve. It's not just the outward facing builder / general contractor.
Build 'em all the same. Let the purchasers differentiate their homes over time. landscape, trees, paint are key. Keep the cost down to get the first time homebuyer back in the market. Attainability is more important than conformity.
Interesting overview, and here’s a historical tidbit to further bolster your point: the old Sears Roebuck home kits came in a limited number of styles/layouts but were quite popular in the early years of the last century. My tiny Midwestern hometown had several. Apparently they were delivered by rail. (Not sure how inclined you’d be to dig deeper into the history there, but they clearly filled a niche.) Decades later those I saw seemed to be generally holding up well and had been modified extensively.
I think the idea is true to a very limited degree: yes, people prefer unique and distinctive houses but aren't willing to pay for them a significant premium. If cookie cutter mass production would bring a large productivity boost, that style would be dominant across all market segments except luxury/high end - because it would trickle down as cheap housing.
The fact that we don't see this dominance and that cookie cutters don't sell at a lower price than a comparable one off house suggests that the productivity gains of repetition are very limited. Or, in other words, that current production methods cannot exploit repetitive patterns to lower costs, unlike say the electronics or auto industries.
You are missing the point, SB. Municipalities are the culprit here. They don't want "those people" living in their city limits. I can build affordable housing forever, if the cities weren't in the way. Aesthetics are form of structural discrimination.
There were significant gains from repetition that made houses cheap, but it all happened at least a hundred years ago (we went from six or more people to a room, to two or three), and there are no more gains to be had from that source alone.
I also remember reading articles a few years ago that said, if you subtract out semiconductors, the productivity growth of manufacturing since the 1960s is broadly similar to that of services like retail or transport. The gains from repetition were exploited early in auto manufacturing also, and productivity growth since has not been stellar.
Why are one-off houses now not much more expensive than cookie-cutter? I think this is a recent development, and it is due to CAD and a lot of other software that has brought down design and permitting costs, in part by having a library of standard modules and allowing combinations of them. Truly "blank slate" designs are still expensive, I believe.
> this is a recent development, and it is due to CAD and a lot of other software that has brought down design and permitting costs, in part by having a library of standard modules and allowing combinations
Even if true, that would still be a rephrasing of Brian's argument: he's debunking that housing variety is the problem, while you say that even one-off houses aren't that varied, they are just small variations of a archetypical "house" product just like the various color, upholstery and trim option are the same basic car model.
Still, I don't think that stands up to scrutiny. The soft costs of even extravagant buildings are a small part of the construction budget, while most of the money is claimed by tasks that involve large amounts of manual and/or skilled labor on site (ex. electrical, plumbing, concrete). There is very little you can do to mass produce foundations on site even if they are a standardized shape. You have some learning curve effects and maybe create some rigid reusable formwork, but the productivity gains are limited and still need to build each one with traditional methods, you will definitely not have an assembly line of foundations like you have for cars or iPhones.
How much does standardising the design reduce costs?
I think of the old adage "faster, better, and cheaper, pick any two."
The industry seems to do "faster" pretty well. "Better" (performance, materials, layout, design, aesthetics) might be possible but it isn't going to be cheap.
I'd happily wait a few years if cheaper and better were realistic options.
I think this mainly depends on culture, and on whether or not platting still exists.
If all newly developed land is sold in lots measured in square miles, then of course the only takers are large developers that come in with a few cookie cutter designs.
If land is subdivided and then sold in some fraction of an acre then individuals can come in with an architect and build their snowflake designs.
The latter is quite common in Belgium. There is a photo blog called ugly Belgian houses that collects the more infamous specimens. This is some odd cultural trait over there.
However. If those custom designs were expensive they would be a lot less common than they are.
Around me in New Zealand, cookie cutter designs are built in exactly the same way as snowflake designs. Foundations, well, what are you going to do? Then, wood frame. It is a pile of timber and a few guys nailing it together just the same. Same plumber and electricity vans. Flooring? Another van. I am not sure how much it matters to the flooring guy if all the floor plans are identical.
Much of the cost savings come from making stupid value engineering choices, because the developer doesn't have to live in it, so whatever.
Maybe the savings come from mass producing parts. Mass produced timber, windows, floor boards, etc.
A decade or so back I did a project on technology diffusion in residential construction. Drywall, with its clear productivity benefits over lathing plus plaster, too 30 years. Some was that early materials weren't that good, but part was that the people who actually did the walls had to learn to work with the new materials. I strongly suspect that the "big" builders work with local crews rather than taking their own from project to project, so the fragmentation is even greater.
My particular interest was improved HVAC, heat pumps with inverters and variable refrigerant flows. Those can work with forced air systems, but work better better with room units. That approach is the norm in (say) Japan. Architects didn't know how to design for those, I worked with both the HVAC company and an architect, to no avail.
Now such systems wouldn't lower the cost of a finished standalone house, and realtors are uninterested in trying to learn how to get US home purchasers to pay more up front for lower utility costs.
Geography may be a factor, roofing trusses and engineered wood joists probably have different suppliers due to shipping costs.
The underlying story is yours, that fragmentation makes anything new diffuse slowly. My add is that the value chain is fragmented, and skill sets are no longer gained through formal apprenticeship programs so are even slower to evolve. It's not just the outward facing builder / general contractor.
Watching a few hours of HGTV will also prove that people are totally fine with having the same thing over and over.
Build 'em all the same. Let the purchasers differentiate their homes over time. landscape, trees, paint are key. Keep the cost down to get the first time homebuyer back in the market. Attainability is more important than conformity.