Plain English with Derek Thompson - Why Is It So Expensive to Build Stuff in America?

Episode Date: September 12, 2023

We're in the midst of a great affordability crisis. It's not just the inflation crisis. It's a greater cost crisis of the last few decades. Everything that matters most in life—health care, housing, education—is getting more and more expensive. Why? One way to investigate this question is to look at the cost and speed of building physical things in America. We build urban transit more slowly than we used to, we build highways more slowly than we used to, we build energy infrastructure more slowly than we used to, we build skyscrapers more slowly than we used to, and we build housing more slowly than we used to. Brian Potter, the author of the newsletter 'Construction Physics,' explains the forces behind the great slowdown, why it matters, and how to turn things around. If you have questions, observations, or ideas for future episodes, email us at [e-mail].  Host: Derek Thompson Guest: Brian Potter Producer: Devon Manze Learn more about your ad choices. Visit podcastchoices.com/adchoices

Hey, it's Bill Simmons. Did you know I've had my podcast for 15 years? Do you know that it is the most downloaded sports podcast of all time? Did you know I have guests from the sports world, from the culture world, people who work for the ringer, people outside the ringer, celebrities, experts, you name it. It's on my podcast three times a week. Late Sunday night, late Tuesday night, late Thursday night, late Thursday night, the Bill Simmons podcast. Check it out on Spotify. Today's episode is about a question I've been thinking about a lot over the last few months,

not only because it's a question very close to the heart of the book I'm working on, but also because it's a question I've been thinking about and writing about for probably more than a decade. And that question is, why is everything that matters most getting so damn expensive? Since the year 2000, the inflation-adjusted price of television sets has declined by 90% TV's, flat screen TVs are way cheaper than they used to be, especially when you're looking at the exact same television, because TVs get better year by year. The price of electronics, like smartphones, has declined. Toys are much cheaper, simple furniture, much cheaper.

But the price of healthcare has gone up faster than just about anything else. Education has gotten more expensive relative to wages. The same is true for childcare. The same is true for housing, especially in America's richest, most productive cities. Now look, TV, toys, they're great. I love TV. I love toys. But they're not as important to human life, to human flourishing as hospitals, schools, or houses. So it really is, like everything that matters most is getting more expensive. I think that this affordability crisis in America is at the heart of a lot of dissatisfaction in this country. It just sucks, I think, to feel like elites are

telling you that the economy is good and progress is great and yay science and yay technology, but you can barely afford rent or your mortgage, and a hospital bill could send you in a bankruptcy and your $40,000 in debt from college. It's not a position from which it's safe to root for progress or the status quo. It feels more like a state of disorder, like something in the system has gone terribly wrong. And that last bit, that's not a conspiracy theory. Something in the system really has gone terribly wrong. If you live in a city, you'll be interested to know that we now build urban transit, like subways, much slower than we used to. If you live in the suburbs, you'll be interested to know that we build highways much slower and much more expensively than we used to.

Highway construction costs tripled between the early 1960s and 1980s, and then they just kept rising. If you like using electricity, you'll be interested to know that we build, energy infrastructure, like transmission lines, much slower than we used to. We build skyscrapers slower than we used to. We build houses slower than we used to. On a population-adjusted basis, the 12 worst years on record for new home construction in American history were all between the years 2008 and 2019. Something has gone terribly wrong. And it is weird that in the same period that everything in the digital space, everything that's made contact with software and computer chips, has gotten faster and more efficient in the last two decades, while everything

in the physical meat space has gotten slower and more expensive. Brian Potter is a writer and a kind of internet historian of building things in America. He's the author of the newsletter, The Fantastic Newspetter Construction Physics. And today, Brian and I talk about the golden age of building stuff quickly in America. why we are now in the sludge age of building stuff today, why it's so hard to innovate in the construction industry, and why this question of building things, and building things quickly, matters so much.

I'm Derek Thompson. This is plain English. Brian Potter, welcome to the show. Thanks, Derek. It's great to be here. You have written that the U.S. seems to build things much more slowly than we used to. So, for example, the average time to construct a nuclear power plant in the U.S. rose from about four years in the late 1960s to about 14 years for nuclear power plants completed this century.

Well, you look at apartment buildings. The average time required to build a 10-unit apartment building went from about eight months in 1971 to 15 months today. And we're going to spend the next 45 to 60 minutes talking about why this has happened and why it's so important. But let's say we didn't have 60 minutes. Let's say you had 60 seconds. What would be your 60-second answer to the question,

why does the U.S. build things slower than we used to? So at a very high level, I would say it's basically a case of we've steadily made it more and more difficult to build things in the U.S. with rules and regulations and that sort of thing, and have not had commensurate technology and productivity increases that have kind of been able to offset that. So the regulations and the difficulties and stuff like that just kind of adds more and more and more burdens over time. And there's just, there has not been an offset technological improvement to sort of counter it. I'm going to take that answer, and I very much appreciate that it was so much less than 60 seconds,

as a kind of table of contents for our discussion. We're going to talk first about rules and regulations, and then we're going to talk about why, as you put it, I think so clearly, innovation has not increased as fast as rules and regulations have increased. So let's start our analysis with a very specific example. Let's talk about New York City skyscrapers. The Empire State Building was built in about a year in 1930. The Chrysler Building was built in 20 months in 1928. you take the largest skyscraper completed in New York City this century, the One World Trade Center. That took eight years to construct, about six to eight times longer than skyscrapers built 90 years ago. So to get us started, Brian, when did New York get so slow at building skyscrapers? So it's really kind of been, if you look at the data, and this is data based on a large database of just skyscrapers constructed in the U.S. over the past century.

New York, of course, has a lot of skyscrapers. So there's a big, juicy data set to draw from. And it looks, it's really kind of a steady increase. The Empire or a steady decrease in a time it takes to a construction speed. The Empire Stage Building is actually a sort of a huge outlier. If you look at terms of like square foot, you know, look at a square foot per year, because it's such an enormous building that was built just so, so, so quickly.

So that really kind of skews the, um, the construction speed upward by quite a bit. But outside from that, you see kind of a pretty high construction speed up until like the 1960s. And then starting in around the 60s and the 70s, then you start to see a pretty big decline. And it's slowly, it's kind of been tapering down since then,

but right. in the kind of the 60s 70s, is when you see kind of a big jump in New York skyscrop construction specifically. Before we get to the Great Slowdown, I just want to hold on the Empire State Building, which is so famous in this space for being essentially the best example of a building

we built really, really, really fast. How did we do it? How did we build the Empire State Building in just over a year? So the building was basically designed from the ground up to be really, really fast. to assemble.

It's funny if you read like sort of descriptions from the architect at the time, they're like, we didn't, you know, and they're talking about the design of it, it's like, we didn't basically think about the design of it at all. We basically, the design is a reflection of the fact that we sort of, all the elements were arranged to sort of make it build, to be able to build as fast as possible. And the sort of design, the actual, what it looks like is essentially just a reflection of that.

So they basically, you know, they use. like off-the-shelf parts and sort of very standard components wherever possible. There was not really any sort of custom materials or, you know, custom shapes or connector, stuff like that. It was mostly sort of off-the-shelf. And they arranged those components in such a way that they could be just put in really, really, really quickly and repetitively. They used like the exact, you know, they had like a really small number of windows that

They used to use the exact same window design over and over and over again. They had these sort of exterior panels that were arranged to sort of drop in really, really quickly. They had done, they did stuff like they had this sort of, you know, mine cart situation they set up on every floor. And because this was, of course, the days before forklifts and stuff like that. So they would like raise the materials up. And then this mine cart situation would bring it down to sort of where it was needed. So basically the whole building was essentially designed from top to bottom to be really fast to assemble. And they just executed really, really well is, of course, the other part of that.

And now we're in the Great Slowdown, where skyscraper construction is getting slower in New York and other cities. How dramatic has this declined been? How much slower are we building skyscrapers now than we used to? And please, use the most fair apples-to-apples comparison here, maybe something like square foot per floor. So on average, it's maybe like a little bit less than half the speed that it was in like the 60s and 50s and stuff like that. It's a pretty, in terms of like square feet per year on average, it's a pretty, it's a very substantial difference. And let's say that timing is suggestive. And this is a leading question because it's my theory that this timing is very suggestive.

I think a lot of things happened to the 1960s and 1970s on the regulatory and rule side that might have directly decreased the speed with which. we built stuff like houses, nuclear power plants, and New York City skyscrapers. Walk us through what 1960s, 1970s changes you think are most important in explaining the decrease in building speed. So in the 60s and 70s, you really see this huge sea change overall in like what the government was going to be responsible for and how they should intervene in society to, you know, try to make it safer and to try to make it better and to try to improve it. So you just see just a huge increase in regulatory burdens, you know, burdens, is perhaps biased in the answer a little bit,

but a huge increase in just regulations across the board. Environmental laws are like a huge one, right? And of course, those have had huge benefits, right? Like environmental controls on coal power plants have made the air much, much cleaner, have reduced the health costs of those plants. But they did have a huge impact on like the cost and the time it takes to build plants like this. You know, the National Environmental Policy Act is the environmental law that basically makes us do environmental impact statements, environmental reviews for any big federal project. That also came in the 60s and 70s. You know, Clean Water Act, all these, you know, acts to prevent pollution of like streams and rivers all come.

You know, the environmental movement is really ascendant. in this time period. And then you also have stuff like OSHA, the Occupational Safety and Health Act, I believe, comes in the late 60s, which dramatically increases federal regulation of, like, you know, work safety, which, again, has had really huge benefits. It made construction much, much safer.

But again, these things have, you know, have basically made things just more a little bit, more difficult to build things. Then the mid-early 1970s is when, of course, you have the energy crisis, and that's also when you see building codes start to get much more strict. That's when you start to see a lot more state adoption of building codes and then start to include energy efficiency provisions and them and stuff like that. And kind of, you know, once you have these rules, they just kind of wrap it up or ratchet up

in stringency over time. Yes, that's about the, that's a high-level look at it. Yeah, I think it's a very complete answer. I mean, we could talk for hours, and there are many hundreds and hundreds of page books that are about this moment in history, which is incredibly successful on the first-order basis of cleaning the air and cleaning the water,

getting pollutants out of the biosphere in America, reducing lead exposure, making it much, much-savisor, for to work. I mean, you had this statistic from your newsletter on the state of iron work in the 1910s that found that in 1912, the International Association of Bridge and Structural Iron Workers had a death rate of 1,000 per 100,000 workers. That's twice the death rate of U.S. soldiers in Afghanistan in 2010. So work in this golden age of building speed was incomprehensibly dangerous compared to what construction work or manufacturing work is today.

And so there's a big thorny story to tell here, but one executive summary is that we pass all of these laws, environmental laws and work safety laws, the 1960s and 1970s, that succeed on the first order basis of cleaning up the environment and making it safer to work, but have these second order effects of dramatically slowing down the rate of construction and making it harder to build many of the things that are most important to build in the 21st century, not just skyscrapers, but also apartment buildings. And while we might not touch on it in depth, energy mega projects and transmission lines as well. I want to move through this regulatory story because I think it's really important. But one thing that I think your work does in a really original way is point out

that even though the regulation story matters a lot, the innovation story might be just as important. So let's tiptoe into innovation in this way. If you take a field like agriculture in the U.S., the U.S. saw productivity advances in agriculture that have absolutely changed the planet. Like in the 1800s, more than half of American workers worked on or around farms,

and today just 2% of the economy is devoted to agriculture, and yet we have way, way more food and much, much better food. That is a productivity success story in agriculture. Construction, by contrast, is one of the only industries that has, by some accounts, seen its productivity decline. So let me put it to you this way. Why did agriculture mechanize in a way that advanced its productivity while construction did not?

Yeah, it's a very, it's a very, you know, big thorny question to answer one. The very high level answer to that question is there are specific things that you can do to make a process more efficient, to produce something more efficiently, using less time, less energy, less resources. And all of those things are very difficult to do in construction. I mean, you talked about one of those is mechanization is, you know, take what used to be like labor intensive, take a lot of manual labor to do, find a way to make a machine to do, to do it. it. It's very hard to mechanize construction tasks. People have been, you know, people have been trying for, you know, again, like decades. And it's just, they've had very relatively little success for it. The biggest impact is probably things like power tools, but those have a relatively, you know, the impact of those is kind of relatively narrow. It doesn't improve, you know, apparently a productivity, all that much. You still, it still requires a worker to hold

the power tool, of course. It doesn't remove the labor from the job site. I wrote a sort of long essay looking at the history of brick robots, bricklaying robots. Because setting bricks, it seems like the perfect task to get to mechanize, right? To automate the solution.

Because brick lane is so incredibly repetitive, right? You're putting an identical brick down over and over and over and over again, thousands and thousands of bricks. Even on like a small house, we'll have thousands and thousands of bricks on them if it's made with brick. You know, it satisfies like the three requirements of things that are, seem like good candidates for automation, which is dirty, dull, or dangerous, right? It's all three of those things.

Bricks are really, really heavy. It's very stressful for the workers to have to lift these things all over again, over and over again. And people have been trying to sort of build a mechanized brick lane system for, like, decades. And it's, you know, they have, there's systems that can do it. There's, there's robots on the market that you can get that will set a brick for you, but they basically don't really work as well or as cheaply as just a person setting a brick. So when you have, you know, bricks or like concrete, masonry units, stuff like that, that really have not been able to sort of automate that

yet. It's still like sort of a manual task, depending, just despite the, you know, decades and decades and millions of dollars have spent trying to automate it. Make me smarter about why. Help me understand why it is that robots can build cars, but they can't build houses. Obviously, I know the difference between a car and a house. I know that one is bigger than the other. I know there's all sorts of differences between a car and a house.

But, like, explain to me at, like, the most tactile level possible. Like, why can't robots build houses? So when you're trying to mechanize or automate some tasks, there are essentially kind of two parts of it. You have the actual physical thing you're trying to do. Like, I'm going to physically move this brick from point A to point B. And you also have like an information processing component where you sort of sense what is going on in your environment. And you take that feedback and use it to sort of modify your actions as necessary. So, you know, the brick is maybe a little bit to the left or a little bit.

bit to the right and sort of I adjust my grip to sort of, you know, put it where it needs to be and put the brick where it needs to be. If the wall sort of is bowing in a little or the wall comes up to a corner, I can change the actions that I need to do based on sort of the feedback and how things are going. At a very high level, you know, simplification, automation is very good at like, you know, automating that physical part of it. And it has historically been, not nearly as good or competent as that automating the information processing part of it. Making the actual decisions responding based on feedback to the environment is something that automation has really only now is starting to get a lot better at with things like self-driving cars,

which can, of course, respond flexibly to their environment. So historically, what you see with automation is that you figure out the example, movement or sequence of actions that you need to take and you set it up so you can just perform those exact motions over and over and over and over and over and over again without, you know, without really having to sort of modify them very much. So if you look like agriculture mechanization, it's like these simple machines that sort of do some very, very repetitive action. And in that action, they're able to sort of filter out the sort of plant, you know, the fruit of the plant from the sort of rest of,

of it because there's like physical differences between the plant that you want to harvest and the other part that you plant away and you can make a repetitive machine that sort of can act on those differences but you know a you know a combine that's harvesting wheat or corn or something like that does not it does not work in the same way that a person that a manual farm labor harvesting the corn would work right it doesn't like reach out and pick each individual ear of corn it sort of just does this repetitive motion and it filters that it filters the corn out from everything else. The plants that you and the sort of agriculture that can't do that,

where you do need to sort of have an information processing component where you need to pick and choose exactly where you need to move based on where the plant is and stuff like that, those have been much, much harder to develop. So like corn harvesting got mechanized in the 1930s or the 1940s. But strawberry harvesting, which strawberries are much softer, you can't, it's so much harder to get like a machine to sort of pick them, that's still, you know, still kind of an open problem. Those are still harvested by hand. And so in construction,

construction is just has a hard time with that, you know, because of that information processing component, it has a hard time mechanizing sort of the actions. Part of that is because each building tends to be unique and different. So you can't have just, just set up your operation to run, you know, hundreds of thousands of times, sort of like that. You have to like, you know, you're going to do it to build the building and then you're going to move on to another building that might be different. Part of it is that it takes place, you know, it's outside in an uncontrolled environment. So you have to, you know, wind and rain and dirt and people moving all over the place.

And it just, the environment itself is, is varying in a way that maybe is not in sort of a, you know, the ideal automation environment. And so there's, you know, stuff like that is, is kind of a big reason why. That's a great overview. Let me ask one quick follow-up question. So robotics, it seems to me, based on your answer, has what I'm very, very stupidly going to call a corn to strawberry spectrum, from easy to mechanize and automate to harder to mechanize and automate.

Are there parts of the house that are more like corn than like strawberries? are there parts of a typical home or apartment building that are incredibly important, a huge part of the construction costs, and slightly easier to make without so many people? Yeah, and I mean, that's definitely true. And the stuff that you, that is highly automated is basically the materials itself,

you know, drywall manufacturing or lumber manufacturing, It's interesting if you look at like sawmills, right, where they take like a big giant log and chop it up into dimensional lumber and then that lumber gets like dried out and then it gets sent to Home Depot or whatever. That process is very highly automated, even just like the examination of the individual lumber to see where the defects are and see where the grade, what grade it can be. That is all automated. It's all done with like computer vision that can look at it and see where the defects are. and what grade needs to be. So, like, you know, if you go out a step back from the actual construction process into the production of, like, materials,

that's all, you know, that is basically just like any other aspect of, you know, production, where it's just gotten much more productive over time. And if you look at, you know, the costs of build materials, those do tend to get cheaper over time, or at least they rise in costs lower than inflation, which is, you know, effectively the same thing.

There's a recurring theme of your newsletter, which is that you have this failure of pre-fabrication, a failure of building homes and factories rather than building them on site, which would make it easier to modularize, which would make it easier to incrementally experiment and therefore innovate and therefore maybe gain some few efficiencies

that could bring down the cost of building a house. I don't think most people know just how popular manufactured homes used to be, that is mobile homes or sometimes called trailer homes. At their peak in the 1970s, the mobile home industry accounted for 20% of new housing units, 20%, now I think it's less than 6% according to your reporting. What happened to mobile homes, or aka manufactured homes in America? Yeah, so mobile homes really blew up.

Like you say, in the late 60s, they were incredibly popular for kind of a very, you know, for a sort of narrow window of time. And then, yeah, the industry basically collapsed in the early 1970s and didn't really recover. It kind of depends on how you look at it. And it kind of collapsed for similar reasons that you see factory-built housing of more conventional type collapse, which is basically there was a huge downturn in the housing market. and when a downturn like that happens, a normal builder, they can weather it by just laying off workers and just riding it out.

They cut staff and they bring their expenses down. And then when things turn around, they can sort of hire back up. A lot of them are just, they are very, very capital light, very asset light. It's very easy for them to sort of, comparatively easy for them to scale up and down their operation, especially because so much of the work of a typical. construction site is I'm subcontracted. With a factory-built housing producer, you don't really have that option. You have to pay the rent on your factory, regardless of how many housing houses you're actually selling. And so a huge housing downturn, these are these factory-built housing producers

are faced with like relatively high fixed costs that they keep having to pay. And that kind of sets them out of business. You actually see that in 2008 when the housing market took a huge down turn. Factory-built housing producers basically went out of business left and right, and you see a huge decrease in factory, you know, the rate of like modular housing. And you see that in like the early 70s with like mobile homes. You know, they just, a lot of these guys go out of business. And it's very, It was very hard for them to bounce back when sort of the housing market turned around. And then you kind of have a couple other things that have sort of made that basic mechanic even worse. A big one was just, you know, again, similar to 2008.

The mobile home industry, which caters to like a lower income, you know, clientele, they had sort of been relaxing their lending standards to try to get more customers, for their homes. And then that really came back to bite them when the sort of economy turned, they ended up repossessing a lot of these mobile homes. And then so they were in this situation where there was a big, actually a glut of supply on the market because there was all these extra mobile homes that had been repossessed. And also the housing market itself had declined enormously. And then also at the same time, because they'd gotten bit, they made their lending standards much more stringent.

So that, those effects combined just had a very brutal effect on the mobile home industry. And you actually see the same thing in the very early 2000s, a few years before the sort of broader housing crisis where the exact same thing basically happened. You had mobile home manufacturers extending credit to people who probably couldn't actually afford to buy a mobile home and then sort of the market turned and they had all these repossessions, which gave this big glut of supply at the exact time when it wasn't needed. And at the exact time when the lending standards got more stringent, and again, it just had an extremely brutal effect on the market.

So you've seen that same mechanic repeat actually multiple times in the industry. Yeah, I think it's important to bring in economics here, as you've just done for the mobile home or manufactured home industry, because we didn't mention that at the top. We talked about regulation and we talked about innovation. And I do think that those are two absolutely critical parts of the story. But the worst decade for home construction per capita in the last 60 years in the U.S. was the 2010s.

And the 2010s were not a period where, oh, suddenly regulation got so much worse or, oh, suddenly innovation fell off a cliff. No, we had a housing crash. we had a housing crash, and as a result, a lot of builders pulled back their construction of single family and apartment buildings. And as a result, because of the macroeconomics of the Great Recession, we had essentially a decade of under construction of houses, under construction, especially relative to the fact that this millennial generation, the largest generation in U.S. history, was about to enter their 30s. were going to need homes to move into, as they were moving out of, you know, bunking up with five roommates or living with their parents,

they didn't move into a house, and we hadn't built the houses for them. And so as a result, what happens? Well, just look at vacancy rates. They're near all-time lows. Look at housing appreciation. They're near all-time annual highs. So it's really good, I think,

to bring in the economic side of this alongside the regulatory story, which is crucial, and the innovation story, which is also crucial. I want to go back to the innovation, story, I want to talk about what an innovation moonshot in housing construction would look like. Before we get to robots and ways that we can essentially think about building a house more like

the same way we build a car, let's talk about material science. Are there ways in which we could have moonshots for the materials that we use to build a house that might make it slightly more efficient to build good, sturdy, long-lasting homes? Yeah, I think there's a lot of opportunity there, actually. I mean, you know, very, very roughly the cost to physically construct like a single-family home in the U.S., irrespective of like land costs or development costs or stuff like that, is about half labor and half materials. So if you can't address, like, the material aspect of it, that really, like, caps how much

you can sort of lower the cost, improve the process of building. One big thing that I think, you know, I'm sort of optimistic on is the sort of improving lumber. And it doesn't necessarily seem like it would be an obvious thing that you could improve because, you know, wood comes from trees. You know, we've built, building stuff out of wood for thousands of years, right? What left is there to possibly improve?

if you look at most crops that we produce, right? Like almost all of them have been like selectively bred for, you know, thousands of years to have more desirable properties that for people, for what, you know, people want. To return to our example of corn, if you look at like what corn is now compared to the plant that it originally started out as, it's like completely different. It's this huge, enormous cob with hundreds of kernels. If you look at like the original plant, this plant called Teoscent, it had like maybe 10 or 12 kernels on it in this tiny little cob

that actually shattered when it was ripe. And the kernels would like scatter all over the ground. And basically that was turned into modern corn over thousands and thousands of years of just selective breeding of getting, you know, getting plants that had a bet, you know, more fruit or more, you know, didn't pop, didn't spread their seeds all over. And they just people bred those and harvested the most desirable, traits for thousands of years and very gradually turned it's turned it in to a plant that was much

more desirable from a human point of view. And of course all, you know, almost all crops are like that. You know, if you look like a modern chicken, it's like, you know, three times a size of a chicken, even just a few decades ago, just from like selective breeding and modern as animal husband techniques. And of course, now we have genetic modification and stuff like that. So there's really huge opportunities from like improving, quote unquote, you know, plants to sort of have more desirable traits for like human use. And we've, we do do that with trees, but we've really only just started doing it. So like corn, like I said, has been improved through selective breeding for like thousands, thousands of years. We're only in a few generations, like two or three generations of doing that for trees.

It's really a post-World War II phenomenon. And of course, corn, which, like, you know, is harvested every year and you have a new crop every year. You know, trees take 15, 30 years to grow. So it's like a much slower process of sort of improving these traits. But, you know, you can imagine a tree that, like, grew much, much faster and then grew much much straighter and didn't have defects. One big limiting factor on how, you know, strong wood is is just the defects that, exist in the wood. And there's this huge category of materials that are called engineered wood that is basically,

essentially what they do is they just chop up wood and they glue it back together. And so you don't have these defect concentrations that limit the strength of your thing. There's a whole category of materials that just do that. But you can imagine that if you bred trees to like get that level of strength without having to go through that process or trees that were like, you know, had introduced to be rock resistant or moisture resistant or stuff like that. So I just, I don't know, I can imagine that just trees be, you know, a very intensive tree breeding program or tree improvement program could really, there would be a huge amount of upside in that.

Convierte your passion in a business with Shopify and bathe records of ventas with the form of pay with a better conversion of the world. Has heard of bien? The mejor conversion of the world. The incredible system of of Pago by the website facilita the website on your

website, and in the world. That is music for your ears. No, you'll

do you do this your business will a super-exit with Shopify. Empieza your period of a

per a month on Shopify. coms bar records. I love the idea that we've spent

you know 5,000, 7,000 years, you know, breeding kale and cauliflower and Brussels

sprouts and corn to, you know, be perfect for our taste and to be, you know, to be easy to eat when you, you know, cook it with fire or whatever it was that our ancestors were using to, uh, to eat these vegetables with. But we haven't spent those same five to seven thousand years breeding trees, super trees that are perfect for home construction. What is the bottleneck? What's the innovative bottleneck here, as you understand it, to breeding, as I suppose I'll just keep calling them super trees that are perfect for home construction. I mean, yeah, that's a good question. I don't have a super clear answer to that.

I think just basically the time and expense that it takes to do this, I suspect it would be very hard from like an investment point of view to have some sort of, you know, any sort of reasonable return on that. As far as I understand it, and I'm of course not an expert, this sort of improvement is mostly done in universities. And it's had really, it's had really, so I guess, you know, to sort of rephrase a little bit, people are doing this and it has had big effects. Like a modern tree plantation will produce much more than a tree plantation from like 60, 70 years ago. I just think, you know, it's just a higher level. I think that process just could continue. And we can imagine just trees getting better and better and better and better. It's already being done, but there's just so much room to go, I think. I wonder if another one here is drywall. I mean, are people trying to figure out, like, better, faster ways to produce some kind of substance that's basically, you know, not super trees, but super drywall?

Is that another vein on this sort of innovation? Yeah. So drywall is, yeah, is another big, is another big pain point. Actually, if you act like, ask like factory-built housing, factory owners, what their one sort of what are their main, you know, factory difficulty is where one of the main operational difficulties is drywall is like very, very high on the list because it's a very, it's so far been like a labor intensive, very time consuming, very sort of finicky process to get like the very smooth inside surface that basically people want. It just takes a lot of time and effort and manual labor to do.

That's actually a case where there is a, you know, there is a robot that can now, set drywall, but still the whole process is still, you know, quite, quite slow and labor intensive. And people have been, you know, people have been looking for a very long time for it to find a way to sort of put, you know, an interior finish surface that doesn't take, you know, days and days and tons of comparatively high amount of labor to produce. It's just very hard to find a sort of material that has the sort of nice combination of drywall properties, which is inexpensive and produces the nice seam-free surface that people like, and you can take a lot of different finishes, you can paint in different colors,

you can put wallpaper on it, and is fire-resistant and is cheap. It's very hard to find a combination of all those things. And so there's different materials that you can use for the interior finish of a room, but they tend to sort of be lacking in one or more of those, one or more of those things. And that sort of has really hindered an adoption. Most people still, you know, in the U.S., especially, you know, still use drywall for the interior. And on the part of the process that is the actual labor, the putting up of the house, not building the super trees and developing the super drywall, but actually finding a way

to build houses in less time. I'm sort of reminded of the fact that we had some people talking about fusion technology as being sort of the dream of limitless clean energy. Is there like a fusion dream but for housing construction innovation that someone's working on or maybe lots of different companies are working on that's like this is what we ultimately want to try to make? So I mean, we talked a little bit about factory built housing and mobile homes. the big one that has come up over and over again is, you know, we should build houses in factories

the same way we build everything else. You know, people look at the process for making, you know, a car or a computer, you know, literally anything else, right? And it's just, you know, cranked off some assembly line, like very high speeds and gives us, like, very affordable consumer products. And people look at the, you know, housing process, which is built on site by hand, not that different from the way houses were built a hundred years ago. And people, you know, looks archaic and like, well, we just need to take this process that we figured out for everything else and apply it to home construction, building construction, and we will make everything much more efficient. And that's sort of been something that people have been tried, trying over and

over again for, again, you know, decades and decades. I myself worked at a construction startup that raised several billion dollars trying to basically take that approach. You know, build buildings and factories. We build everything else and we'll build them so much cheaper and be the low cost supplier in the market and have, you know, be able to produce in huge volumes and we will become the Henry Ford of housing. It just did, you know, it did not work for them. They spent $2 billion of, you know, investor money trying to do that unsuccessfully.

And it hasn't really worked for the for the other people that have also tried, which is not of course to say that you can't build a building in a factory. You of course can, and many people build successful businesses doing exactly that. And especially in Europe, factory-built housing is much more popular than it is in the U.S., Japan as well. But what you can't do is no one has really figured out how to do is become the Henry Ford of housing, where you've come up with this factory-built method that is just so much more efficient and so much cheaper. than currently existing methods, that it just wipes up, it wipes out the previous way of doing things, and it's unimaginable going back.

Quick question. Quick follow-up. Is the problem, as you see it, for factory-built housing in America, on the supply side, because there are constraints that innovation can't overcome? or is it also partly on the demand side that Americans, perhaps unlike Europeans, although in this question I'm literally just guessing, we have certain expectations that our houses be original, that we not live in cookie-cutter mansions,

that there's a kind of American desire for uniqueness that makes it a little bit harder to get the kind of, of efficiencies that you get from modularizing and repeating the same construction process over and over again. So I guess how much of the issues that your startup faced, how much of it was a supply problem versus what you perceived or experienced as also an American demand problem? Sure. So I will separate out, you know, the startup that I worked at from the problem more generally. A startup that I worked at had, you know, various challenges that were specific to them and operational issues and stuff like that.

But to talk about the sort of fraud or problem, yeah, there's definitely like surprise problems in the sense that it is hard to do that and dramatically, you know, build, building housing and a factory and dramatically reduce your cost while doing so. The very high mechanic is you don't get as many, the very high level mechanic is that you don't get as many efficiencies as you think that you might because you just don't have the production volume to produce like hundreds of thousands of units and a lot of the efficiencies that you do get tend to be eaten up by high transportation costs. So that's like a high level look at the at the supply problem. The demand problem I think is more a function of the fact

that in the U.S. especially it's just you know we you know with our sort of federal system we just have, and the fact that, you know, building regulations are often done at, like, a state or local or, like, municipal level. We just have, like, thousands and thousands of permitting jurisdictions in the U.S., each one, which might have, like, its own requirements, each one, which might, you know, want to review your specific building and see if it fits in with, with the other buildings around it and stuff like that. And even in the absence of that, every sort of site is a little bit different. The workforce is always a little bit different.

So there's all this like variability that makes it very hard to crank out like a uniform product. I actually don't really think that it is on like the consumer side. I actually think empirically people seem pretty okay with living in very, very uniform houses. If you look at actually most new housing developments, it's really just like six or seven floor plans copied over and over again. Maybe they changed the paint color.

Maybe they flipped the orientation back and forth. But it really is quite repetitive. It's really pretty much the same unit over and over again. If you look like apartment buildings, it's really just a few different floor plans. Again, copied over and over and over again. Empirically, people don't seem to value uniqueness really that much. And also, I mean, I think if you look at other things, this is also the case. Everyone always makes this observation that if you look at cars, every single car manufacturer has basically the same, you know, or mass market car manufacturer has like the exact same model offerings. It's almost everything else. It's just, you know, you have an SUV, you have a Ford or a sedan, you maybe a slightly smaller compact. And everyone has a nearly identical looking version of basically the same five or six types of car, right? And it's really, and of course you have different sort of trim options that you can do to customize your specific taste. And of course, that also.

exists in housing. You can change the paint color and sort of change the finishes in your bathroom or your kitchen and stuff like that. So I think the level of, I think the level of, you know, customization that people demand is not incompatible with like sort of a factory built housing system. It's just there's all these other constraints that make it quite hard. Last question. If I told you that 20 years from now, by the year 2043, this multi-decade trend of construction productivity going down, down, down, had suddenly reversed itself. That actually, in the golden years between 2023 and 2043, construction productivity had suddenly taken off. That's your news headline from 2043. would you be more likely to believe that this was a regulatory success that the Yimbys, the Yes in My Backyard movement,

the abundance movement, all of these pro-build, pro-growth movements thrillingly succeeded and got federal state and local governments to pull back regulations and made it easier to build and that unleashed construction productivity that had been pinched in the previous half century?

Would you be more likely to believe that story or would you be more likely to believe the story that the fusion breakthrough in construction had finally come, we'd figured out some way to eke out extraordinary productivity gains and cost declines in the mere construction of homes, even though nothing changed in the regulation side?

Which side of this problem do you think is more likely to see its breakthrough, regulation or innovation? Ooh, that's a good question. I mean, I think it's going to be very hard to solve it without both, but if you're forcing me to choose, I think probably the thing that would be most likely to solve that

is just if automation gets really good. I kind of talked before how automation has sort of this physical aspect of it and this informational processing aspect of it, and we're slowly getting better at the informational processing aspect of it. If you could have like a really general purpose automation, you know, like a humanoid robot or that was basically as capable as a manual labor and wasn't outrageously expensive or something to that equivalent.

That would really change the game in construction. That would enable a lot of, I think, productivity improvements that we haven't really seen. It's unclear to me how close we are to something like that. Some people think we're quite close. I myself am perhaps a little bit more skeptical. But I think that would be like a huge deal. Yeah, I personally am not.

not holding my breath for C3PO, the bricklayer. If we get it, fantastic. I don't think that particular breakthrough is imminent. But, yeah, as always, the cop-out answer is probably the correct answer, which is that we probably need both. Brian Potter, thank you so much for doing this. I really appreciate it. Yeah, it's been great to be here. Thanks for having me.

Plain English was hosted and reported by me, Derek Thompson, and produced by Devin Manzi. We'll see you back here every Tuesday for a brand-new episode. Have a great. Thank you.