Pod Insights Summation with Auren Hoffman Construction Physics' Brian Potter - why America excels at fracking but struggles with construction About this episode Brian Potter is a structural engineer and author of Construction Physics, a weekly Substack about the economics, technology, and productivity of building and infrastructure. He's also a Senior Fellow at the Institute for Progress. In this episode of World of DaaS, Brian and Auren discuss: Why construction productivity has flatlined for decadesThe failure of modular construction and automation barriersThe unprecedented innovation at Bell Labs and modern research comparisonsWhy America do...
Transcript Speaker names marked ? are inferred from speaking time and the show’s known hosts — they are not voice-verified.
0:02 Auren Hoffman? Hello fellow data nerds. Welcome to World of DAS. I'm your host, Orin Hoffman, CEO of Incubate and GP of Flex Capital. Discover more episodes, get weekly data as a service news, original content, articles on data and more at worldofdas.com. That's worldofdas, D-A-A-S, dot com. Hello data nerds. My guest today is Brian Potter. Brian is a structural engineer and author of Construction Physics, which is a weekly blog about economics, technology and the productivity of building and infrastructure. He's also a senior fellow at the Institute of Progress, where he's exploring why construction productivity has stagnated while the rest of the economy has accelerated. Brian, welcome to World of DAS.
0:47 Speaker 01 Thanks for having me.
0:48 Auren Hoffman? Super excited. Now, I'm a huge fan of construction physics. I read it every week. I send it out to my newsletter, my five links and everything like that. I really love it. But I've been meaning to ask you this one question, which is, I live in Northern Virginia, often drive on the GW Parkway. For the last three and a half years, there's been this two mile stretch, which has just been under construction forever. It didn't seem like they make any progress at all. I feel like if I was in China, this would have been done in like two or three weeks. Why does it take so long just to slightly move some structural things on a highway?
1:27 Speaker 01 So my sense is for roadways specifically, I think there's a lot of what you would call collective action or coordination problems, where it's basically, there's a very large reluctance to disrupt existing operations and people's lives, basically. People don't want to close down a main road to just rebuild it, even though that would be much faster. And so they have to structure things in such a way, so it's as minimally disruptive as possible to the existing users of the road.
1:58 Auren Hoffman? It was two lanes each way, so they made it one lane each way. And it's been one lane each day for, it'll probably be another five years before it's done.
2:05 Speaker 01 You keep the lanes open, you try to maybe minimize the super disruptive stuff. So it's only on nights or only on weekends or only on certain days or whatever. And then all that difficulty just ends up adding quite a bit of time. I think that's a big chunk of it. And you see that in various other guises as well. So for like New York subway construction, they famously want to excavate all their stations instead of just coming in down from the top, which is a construction method called cut and cover, which is often much faster than just having to tunnel in from below. You just cut open the road, build up your station, build the road back up. It's a little, it's quicker to do things that way.
2:45 Auren Hoffman? And then don't do it in New York because you'd have to close down Second Avenue for a month or something?
2:50 Speaker 01 Yeah, you'd have to disrupt all the existing operations. People already hate these big disruptive construction projects. So you do this more expensive, less disruptive way. And there's not difficulty of getting to people affected by something to go along with what you're trying to do is often quite difficult. And so you see that show up in like other ways, like horizontal infrastructure construction generally, like you're trying to build a transition line and you're trying to get the right of way from all the existing landowners or whatever. That can be quite difficult. That's a major blow.
3:24 Auren Hoffman? But wouldn't you get the right of way first before you started anything?
3:27 Speaker 01 Well, you do, but acquiring all that right of way takes a very long time. So that's one of the reasons why it takes 10 years to build a transition line.
3:34 Auren Hoffman? And in places like China, they could just force everyone to do it.
3:38 Speaker 01 Yeah, the government says, we're going through. The US used to be more willing to just do this disruptive shit, basically, and upend certain people's lives, build some big piece of infrastructure, use eminent domain and stuff like that. And people used to be perhaps more tolerant of it, or perhaps they just had less of a civacy in whether it went through or not. But for better or for worse, we are often will sort of bend over backwards to not disrupt existing operations for things, much to the detriment of actually getting stuff built quickly and efficiently.
4:13 Auren Hoffman? There is some sort of narrative that regulation is a massive percentage of cost of construction. Where do you fall on that?
4:21 Speaker 01 It's a complicated question. I think it's definitely true that regulation has made stuff more difficult to do in construction. The question is, is to what extent that differentiates construction from other industries? Because regulation of everything has gotten a lot more stringent over time, manufacturing, agriculture, blah, blah. And those industries have seen substantial productivity increases. And so the question is, is construction more regulated than other industries? Is construction differentially affected? Is regulation like uniquely burdensome to construction in some way that it isn't in other industries? And I think there's some evidence that that is the case, that construction is, regulation is like more stringent, affects construction more than
5:06 Speaker 01 other industries. One way that that shows up is that the death rate of construction workers on the job has fallen a lot more than the death rate of agriculture or manufacturing. So construction has gotten a lot safer over time than other industries have. And if you look at OSHA fines, fines workplace safety violations, they're way, way, way higher in construction than any other industry. By far, the most industry that gets fined is construction. And so there's these pieces of evidence that suggests that construction is a little bit more affected by regulation than maybe other industries are.
5:42 Auren Hoffman? Okay, interesting. You worked at Cetera. I love that idea of being able to like modularize things in some sort of way. Why did that not work?
5:52 Speaker 01 So that's a pretty standard playbook for trying to, you know, someone operating like a construction startup or a new construction operation, someone thinking, oh, Construction is so inefficient. Everything is built onsite by hand. It's very, very labor intensive. Each building is unique in most ways. I can make building more efficient by doing it in a factory. I can mass produce buildings, just like Ford did for the Model T, and I will use that to sort of make the whole industry more efficient. And lots and lots of other people have had this idea over the years and tried to execute on it. And they mostly have gone bankrupt trying to do it.
6:31 Auren Hoffman? Why is that the case? Is it just because moving it is really hard? Or the final assembly onsite is still so big?
6:38 Speaker 01 Yeah, it's a complex bundle of those things that you mentioned and other things. But at a high level of abstraction, it is very hard to cut labor and efficiency out of the process and make it more efficient. It's hard, even if you move it into a factory, it's very hard to dramatically rework your process in a way that you can make stuff really without labor, without adding cost back in somewhere else in the process. So you move your process into the factory. You're producing some sort of module that you're going to make your building out of. Even if it's done in the factory, there's still not all that many tasks that can be really, really automated the way that a car assembly line is very, very highly automated.
7:24 Auren Hoffman? Why bathrooms or bathrooms or walls or walls?
7:27 Speaker 01 Yeah, so again, it's a complex bundle of things. But one reason is that buildings are different enough and you can't transport your buildings far away. That like the amount of volume that a single factory can produce in terms of like number of buildings is like not all that high.
7:45 Auren Hoffman? Even the insides of buildings are different, like the countertops and stuff.
7:49 Speaker 01 Yeah, a lot of that stuff is factory produced. Windows are like factory produced these days, stuff like that. So you do see a creeping level of modularization, but it's pretty like small components. The stuff that like can be is repetitively Used in very, very large numbers and doesn't change much from building to building. That stuff does tend to be like factory produced. So windows, drywall sheets, doors.
8:11 Auren Hoffman? Doors, windows, those types of things. Yeah.
8:14 Speaker 01 Fans, stuff like that.
8:16 Auren Hoffman? I assume that even the wood like is pre-cut to the right two by fours or something like that, or?
8:21 Speaker 01 It is not to as much as you would think. Typically on site, they will get like a big bundle of lumber and then just cut it to what they need. Basically because the workers can like do that like reasonably quickly.
8:36 Auren Hoffman? It might be like a tiny fraction of an inch difference or something or whatever.
8:40 Speaker 01 This touches on another big thing as to why it's so difficult to factoryize some of construction is that the amount of equipment and machinery that you need to like duplicate what two guys can do with a hammer and a saw basically is a lot. All this material and these components are big and bulky and they need a lot of pretty substantial equipment to like manipulate them. And so it takes hundreds of thousands of dollars worth of equipment to even do relatively simple jobs that like two guys could do really, really quickly.
9:11 Auren Hoffman? Is it because things like homes are still so different? Whenever I go to someone's home, they may have one of four cars in it. So the cars are all the same. They all have the same phone in the home. They all have roughly the same TV. But then each home is a complete snowflake. If the homes were more similar, if there's only seven models or something, would it be a lot easier?
9:33 Speaker 01 To some extent, a lot of that is dictated by site constraints.
9:37 Auren Hoffman? Like you're on a hill or?
9:38 Speaker 01 Yeah, there's a hill where there's trees, there's water, stuff like that, or just the size of the plot is different. Where the utilities come in is different. So there's only so much that you can make it uniform just by the very nature of like you're building this thing in like a specific place with requirements. To go back to what I was talking about earlier, there's only so far that you can ship this stuff because it's so big and so bulky. A car factory will produce hundreds of thousands of cars a year, millions of cars a year. But it's pretty hard to get more than a couple thousand. Home units out of a single housing factory, even stuff that is like really, really repetitive, like a manufactured home, like a mobile home. Those factories tend to be pretty small scale. They're not all built in one factory in Nevada or something.
10:23 Speaker 01 These big manufacturers, they have 20 facilities scattered across the country, each one making a relatively small number of these homes because there's only so far that you can ship this stuff economically because it's so big and so bulky. So these things have like a catchment area of a few hundred miles and that's your market. So you have a few hundred miles that you can ship it. Within that, you have a bunch of different demands for different kinds of housing, different customers that can afford different things, different plots, stuff like that. And so like the uniformity really gets like whittled down.
10:51 Auren Hoffman? So if we want to drive down the cost of construction, what are the big drivers that we could do as a society to do that?
10:58 Speaker 01 I'm pretty optimistic, medium term, about automation and robotics, just because-
11:06 Auren Hoffman? On-site automation?
11:08 Speaker 01 Yeah, onsite or like some combination of something like a pop-up factory or something like that, that you set up near the site and then your shipping distance is greatly reduced. There's a lot of popularity for that idea, too. And the reason is this basically is uncreative is like, well, I can't think of any amazing ideas of what would optimize construction. I spend a lot of time racking my brain, but lots of other people have lost large amounts of money trying to come up with a really clever idea. Nothing really seems to work. But even if nothing else works, if you replace your relatively extensive labor with a robot that costs $5,000 or something that can work 24 hours a day, that would effectively solve your problem.
11:47 Auren Hoffman? World of DAS isn't just a podcast. World of DAS membership is a private invite-only community for founders and executives of important data and AI businesses. It's a place for off-the-record, high-trust conversations focused on the biggest questions data leaders are facing. The community is about peer learning, curated intros, and access to private events. Everyone in the group is building or leading something meaningful in data. If that sounds like you, apply to join at worldofdas.com. That's world of D A A S dot com.
12:29 Auren Hoffman? Now you have people using forklifts. You have all these different tools that these people are using, which do make them, I assume, much more productive than they were before.
12:36 Speaker 01 You would think that. And that's, I think, one of the big mysteries around the construction industry is that why that didn't have a bigger effect. The mass introduction of power tools, basically.
12:46 Auren Hoffman? The number of people that you need to make a house and the number of man-hours seems the same as it was many years ago.
12:54 Speaker 01 It is. It's very, very similar.
12:57 Auren Hoffman? Which is crazy. That doesn't make any sense to me.
12:59 Speaker 01 Yeah, that's very strange. And what's also very strange related to that is between, say, 1950 and today, the technology of putting up a house hasn't changed that much. There's some stuff that's different. Drywall was pretty uncommon in 1950. It's ubiquitous today. Stuff like wood trusses became adopted. Certain materials became more common. But the basic process wasn't all that different. The basic materials you were working with didn't change too terribly much. But between, say, 1900 or 1890 and 1950, everything changed about it. But even over that earlier period between 1890 and 1950, the data we have suggests that it didn't really get that much more efficient building.
13:43 Speaker 01 We don't have great data, but we have cost time series and stuff like that. And it doesn't seem like it really got that much cheaper over that period, even while the technology of building totally changed. Again, there's still a lot of things, reasons that this is like not amazingly well-understood as to what specifically makes it so difficult to improve it.
14:03 Auren Hoffman? What countries are doing this well? Are there countries that we could like look to at all?
14:08 Speaker 01 There's a couple of ways of thinking about it. From what I've seen, there's not really any countries. I've only seen a limited number of data sets. So some other country out there might surprise me. But there's not that many countries that are avoiding the problem of construction costs constantly going up, construction productivity relatively flat, as far as I can tell. If you look at cost per square foot or whatever productivity indexes, they show the same trends as the US.
14:36 Auren Hoffman? And some of the countries, their labor costs are going up much faster than US. I presume their costs are going up with labor or something.
14:43 Speaker 01 Yeah, or just they're under similar sorts of constraints. What you do see is that other countries operate a lot closer to the efficient frontier. The most inexpensively you can put up a house. So an apartment building in a major European city or something will be much cheaper to build than a major US city.
15:02 Auren Hoffman? Why is that?
15:03 Speaker 01 Again, it's a combination of things. High labor costs, a lot of these coastal US cities have very strict and burdensome regulation. New York has very high construction costs and they have a lot of like really, really onerous regulations for building. It's very hard to use a tower crane in New York for some reason. So you have to use these smaller, less efficient cranes, takes longer to do. You have to use union labor for a lot of stuff and that drives up costs.
15:30 Auren Hoffman? Is it also you have to guard against the wind that may only happen once every 400 years or that type of thing?
15:38 Speaker 01 That's less of a thing. There's some of that. The US in general has, I'm not sure about winds, possibly about wind, but like high seismic stuff. There's some places, earthquakes, you have earthquake design is more stringent in a lot of places in the US, especially on the West Coast than it is in most places in Europe. I would not say that's an enormous driver of costs. It's a factor, but I would say it's much more regulatory labor costs, general coordination issues. Stuff that's like very highly regulated. European countries tend to be better and more efficient.
16:12 Auren Hoffman? Every time you read anything on this, it all, you always say something like France can make their subways at one tenth of a cost as New Yorkers. You always see these weird stats like that and it just seems like you're like, well, France, really? Like, why is France like?
16:27 Speaker 01 Spain, very, very good at building subways very, very cheaply, which again, you don't think of Spain as. The incredibly efficient country of Spain, that's not the first thing that jumped. Italy too, another one, I think they're pretty good at building subways. And that's just this complex bundle of issues. The government agencies in charge of. Managing these metro construction projects, they a lot of time don't have in-house capacity to like do the design and manage them very well. And so they end up outsourcing a lot of that. And so you end up with these principal agent problems where they're not necessarily doing a good design. And the agencies aren't be able to oversee them very well and there are conflicts at work. You see stuff like that across the board.
17:06 Auren Hoffman? Is the hard part about building subways literally boring the tunnel that they go? Or is it all the other stuff afterwards, like the tracks and the trains and getting people there and all that other types of things?
17:17 Speaker 01 I'm not an expert at transit construction. There's so many people who are like so enthusiastic about like transit construction. That I've in some ways stayed away from that topic, but my sense is that a lot of these are used pretty similar technology to actually drill them. Use like a big tunnel boring machine. Basically it's more like doing it with an efficient amount of labor. And doing the coordinations, they're doing the right jobs in the right place at the right time and stuff like that.
17:43 Auren Hoffman? Okay, interesting. You also spent a lot of time studying shipbuilding. Why is that so important?
17:48 Speaker 01 Well, it's important now because if you think their major global competitor is China. And you think there's a reasonable chance we're going to fight a war with China in the South China Sea over Taiwan or something like that. Then that's going to be a naval conflict. You're going to be able to build a large number of ships and then just more generally.
18:09 Auren Hoffman? Why would ships be important in like a drone world?
18:12 Speaker 01 Well, because you need your ships close to deploy the drones, deploy larger aircraft, fire missiles and stuff like that. It's not like these drones can all just come. You have to get your stuff into the field.
18:22 Auren Hoffman? I'm saying they're already there. It's more like the US would need some ships.
18:26 Speaker 01 China is building enormous numbers of ships. China has a bigger navy than the US now.
18:30 Auren Hoffman? Why would that be important in a conflict like in the Ukraine conflict today? Russia has a massive navy and Ukraine doesn't, but it doesn't help them because their navy can't get anywhere close to Ukraine because of the drones.
18:43 Speaker 01 Yeah, that's a good question. There's a lot of thoughts. I'm not an expert on this. If you want an expert, you should talk to my friend Austin Vernon about drone combat. It relates to sort of modern warfare, but I think in general, it is not clear how drones are going to interact with modern ships. And you're still basically going to need modern ships, transport weapons, transport troops, transport larger aircraft.
19:02 Auren Hoffman? Yep. Or there'd be like counter drone things.
19:05 Speaker 01 Yeah, counter drone thing, stuff like that. Drone carriers, I think, is a possible area where you might see ships and stuff like that. But anyway, you're going to need a lot of ships for a conflict in the South China Sea or basically anywhere because the US is basically protected by oceans from most countries that could be a potentially major adversary. And so for any major conflict, you basically need ships to move your material around. So in a Gulf War, basically, for instance, we needed a huge number of ships to move the material over into the Middle East.
19:38 Auren Hoffman? And I assume most of those ships today are Chinese-made ships.
19:42 Speaker 01 There's actually, I think, a big issue with whether we have enough sea lift capacity, is what it's called, to sort of move large amounts of military material to a foreign country if we needed to. There's a lot of that capacity, I think, has been reduced quite a bit.
19:54 Auren Hoffman? Initially, in the 50s, Japan was dominating. And then maybe starting in the 90s, China started dominating. How did these shifts, was it from a top-down country perspective where it's like, this is a priority and we're going to make this happen? Or how did these shifts happen?
20:09 Speaker 01 It's pretty much that. It started as a top-down priority. We need to have a strong shipbuilding industry for whatever reason, because we're a maritime country and we get all our goods via ocean trade. Because it's a key component is to building up your industrial economy more generally because it sort of builds manufacturing capability. It's a big user of steel, which we're also trying to build up and stuff like that. It's something that we can sell to foreign countries and earn foreign exchange. Basically, it was a top-down thing, is the way these countries decided that was important. And they just invested huge amounts of money building up the infrastructure to do it. They all had relatively low-cost labor. They ended up getting often low-cost steel, too. Didn't necessarily start out with low-cost steel, but they built up their steel industries over time and were able to reduce their steel prices.
20:54 Speaker 01 They were able to reduce the cost of their inputs so much that they were able to build these ships like very, very, very competitively.
21:00 Auren Hoffman? How does it actually work? So ships are massive. Is it some point where you're like literally building it on the water? You can't put them in a building that would fit. Or are you building in this modular way and then you stitch them together on the water? I assume at some point, the final assembly of the ship is on the water.
21:17 Speaker 01 So it's usually done in what's called a dry dock, which is basically a big recess with the concrete bowl, basically with flat on the bottom. And they basically build the ship in big chunks. It'll be one small chunk of the hull or whatever.
21:31 Auren Hoffman? It's outdoor. I assume it has to be outdoor.
21:34 Speaker 01 Most of them are outdoor, but sometimes they will build these things indoor. There's a few facilities that have just a big, giant, huge building with a dry dock in it. They can flood. I think some cruise ships are built that way. But yeah, there's some facilities that are indoor, but most of them are, yeah, it's like a big outdoor truck. And so they assemble the whole ship. It's sitting on the bottom of this dry dock, basically. And when the ship is done, they open the gates and they flood it. Ship floats and then they sail it off.
22:00 Auren Hoffman? Oh, that's so cool. I imagine just the infrastructure of that is massive. It's part of the reason the US just can't do it.
22:07 Speaker 01 It's not so much that it's physically difficult or like so complex to do. That's one of these reasons that you keep seeing new countries come along and get a big share of the shipbuilding industry is because it's relatively simple to do compared to something more complex, you know, like semiconductors or aircraft or something like that. It's much simpler to build a ship than things like that. But it does take a lot of space. And a lot of these existing shipyards in the US, they like don't have space to expand. And so all these countries that built up their shipbuilding industries over time, they were able to build this big, huge, new shipbuilding facilities. And willing to invest the money to do that in like these spaces where there was huge amounts of room to expand and build all the room to like move parts and pieces around and store the big modules and stuff like that.
22:52 Speaker 01 In the US, like shipbuilding shipyards, they don't really have room to expand in the same way. There's not necessarily the money has not been allocated to find places to build huge, giant shipyards. Perhaps someday there will be.
23:06 Auren Hoffman? I went once to the Boeing aircraft facility in Seattle.
23:10 Speaker 01 Yeah, the Everett factory.
23:12 Auren Hoffman? Yeah, it was amazing. First of all, it's so big. Just the scale of it was so cool just to see. But why are Airbus and Boeing, why do they persist as this duopoly? Where it just seems like the last place. The US has lost things like shipbuilding and so many other types of things.
23:29 Speaker 01 My sense is that the aircraft industry is so small in terms of number of airplanes sold a year. For a large commercial aircraft, Boeing 737 or whatever, they only sell maybe a thousand, fifteen hundred of those a year. There's not that just much room to like carve up the market between all that many participants.
23:51 Auren Hoffman? And I assume like safety is really important there and efficiency and stuff.
23:56 Speaker 01 Yeah, it's just so expensive to build and develop a new model of aircraft, partly because of safety, partly because you have to test it so much. It's so expensive. You need to earn that back over a certain number of sales. And if the market is carved up too much, there's not enough basically market share to go around. And so what you saw in like the seventies and the eighties and nineties or whatever, is that you had more competitors. And then they just got like slowly whittled down because it was just so expensive to like build these aircraft. There just wasn't a market share.
24:27 Auren Hoffman? Donald Douglas of the world.
24:29 Speaker 01 Yeah, so like Lockheed was a competitor in like commercial aircraft for a while. And they developed this airline, this aircraft called the TriStar. It cost them like billions of dollars to develop and they only sold like a tiny number.
24:40 Auren Hoffman? I never even knew Lockheed made commercial.
24:43 Speaker 01 Yeah, they lost huge amounts of money on it.
24:46 Auren Hoffman? I don't think I've ever flew on a Lockheed commercial aircraft.
24:49 Speaker 01 Yeah, it would be pretty uncommon because it was built in the seventies. I think there's maybe one still flying. I feel like it's used for like some special purpose or something like that.
24:58 Auren Hoffman? Are there examples of recent infrastructure projects in the US that are going really well? Like, I don't know, maybe in some place in Texas and something we can learn from them?
25:07 Speaker 01 So the US is pretty good at building a lot of natural gas power plants, stuff like that. Those, if you look at the time lapse of those going up, those get built really, really quickly. But they avoid a lot of these problems that we were talking about earlier with so many infrastructure projects. It's not like a tradition line or a subway where you're incredibly disruptive to a huge number of people that you're trying to appease.
25:29 Auren Hoffman? Right, it's just in the middle of nowhere.
25:31 Speaker 01 You haven't get a bunch of land easements or whatever. It's a social that they dodge a lot of those permitting difficulties and stuff like that. Industrial facilities in general are pretty okay at building. A big car factory or something will go up reasonably quickly. So stuff that avoids a lot of these coordination challenges.
25:47 Auren Hoffman? So basically, if you could build it out of population center, whether it's a power plant, or a factory, or a warehouse, we can do it. It's when you had a close population centers.
25:58 Speaker 01 Yeah. You don't need approval of a bunch of little individual landowners. If you don't need federal government approval or whatever, avoid a lot of like your complex permitting problems. It's quite a bit more straightforward. We're a little bit better at it.
26:11 Auren Hoffman? It's interesting because 20 years ago, there was all this talk in the US about peak oil, not just the US but worldwide. And right as the fracking movement was happening, but it's happening under everyone's noses. Do you have a sense of why that was so successful? In some ways, it is a manufacturing success for the US. Why was that so uniquely successful and so uniquely American?
26:35 Speaker 01 Who is this company in Texas, Mitchell Energy, I think of it. And they just experimented, they had a sense that they could extract large amounts of gas in these formations that had traditionally been thought that you couldn't get it out of. And they just kept fiddling around with drilling and fracturing parameters for long enough and kept experimenting with it until they basically cracked it, essentially, until they figured out, okay, here's what you need to do. Drill these formations in a way that we can extract large amounts of gas and energy from it. And once they started figuring it out, more people poured into the sector, they sort of more development. They kept figuring out better and better ways to do it. In general, the oil and gas industry in the US has been pretty innovative.
27:17 Speaker 01 It's very, very competitive. And there's lots of constant technology changes to sort of find ways to do this drilling more efficiently and extract more oil and gas from these formations. For the past 100, 150 years, or whatever, there's just been constant, constant innovation pushing the technology forward to try to make it better over time. And the fracking is incredible and huge amounts of energy.
27:40 Auren Hoffman? Why has that been so uniquely American? It seems like in most other countries, if we can build something quick, they can do it. They can sometimes even do it way better than us. This is like an industrial, very industrial thing where it seems like America does it the best.
27:54 Speaker 01 I have a vague sense, so I'm going to put something that's a placeholder that you or listener can go look into lately. I think the way that the US treats mineral rights is a little bit different than other places. I have a sense that in a lot of places, the government basically owns mineral rights.
28:11 Auren Hoffman? Yeah, most countries, the government owns whatever's below the ground.
28:15 Speaker 01 And in the US, I don't think that's true.
28:17 Auren Hoffman? Yeah, if you own an acre, you own what's underneath the acre.
28:20 Speaker 01 You own it and then you can also sell it, basically. And I think that has been a key factor into why there's so much more incentive to find ways to like develop ways of getting minerals out of the ground, basically oil and gas than maybe in other places. I think that's part of it, but you'd have to ask somebody with a deeper knowledge of the oil and gas industry to get a more detailed answer.
28:41 Auren Hoffman? And one of the things I loved was you wrote this great essay on Bell Labs, which basically invented so much stuff, such an innovative place. Why do you think they were so uniquely innovative?
28:51 Speaker 01 Yeah, so it's this perfect Goldilocks combination of arrangement of parts and pieces, basically. So we had this huge monopoly that could funnel huge amounts of money into this operation. And so they didn't really have to worry about competition, but also the monopoly was so big that they were always on the verge of being too big for the federal government to accept, and they were always worried.
29:14 Auren Hoffman? Yeah, you have to waste money.
29:16 Speaker 01 Always worried that the government was going to like come in and break them up and just so appease them. They're like, okay, you need to be constantly improving your services and constantly driving down costs. So they had like a very, very long time horizon because they were a monopoly. They didn't really have to worry about competition, but they were also very, very strongly incentivized to constantly be improving their technology and developing new capabilities that they could hold out. US government says, look, we made this other thing. Please don't break us up. And then they just hired extremely, extremely, extremely talented people and managed them in a very good way where they gave them a lot of freedom to pursue whatever they thought would be useful, but they also subtly directed them into solving problems that were like
29:58 Speaker 01 related to business that they were in, which is seen improving communications.
30:02 Auren Hoffman? Because like a Claude Shannon could have gone into like a university or something, but somehow they were more effective in a Bell Labs.
30:09 Speaker 01 Yeah. So Bell Labs, they had huge amounts of funding. The equipment was much better. They didn't have to worry about teaching or finding grants or stuff like that. So it was a very cushy job to like do research in Bell Labs. And of course, for most of the time, since it was also very, very prestigious, it was as prestigious, more prestigious than like a major professorship at a major university or something like that, which is not generally true industrial research type places.
30:33 Auren Hoffman? There's some people make analogy today to like what Google does for research or what some other companies do for research. Google, they published a transformer paper, which really has led to the AI revolution. Is there some analogy to today or that was a unique moment in time, which will never be replicated?
30:52 Speaker 01 Google is probably the closest. There's stuff like transformer, a lot of their AI work, they were funding self-driving cars for like years and years and years and years and years, which is only now starting to like bear fruit.
31:05 Auren Hoffman? Only a company with a massive time horizon compared to that. Yeah.
31:09 Speaker 01 Yeah. Lots of other stuff that escapes me at the moment, but even then they're a little more focused on the bottom line. I think they tend to like spin their efforts out as separate companies a lot sooner than maybe would be done, would be turned into a commercial operation at something like Bell Labs. Google is perhaps the closest, but even Google is maybe not quite at the same level of truly free from immediate concerns about like competition. We only need to worry about making this useful technology in the long run.
31:37 Auren Hoffman? You think of Bell Labs, the 40s and 50s, and then at some point you think of Xerox Park or in the 70s or something, somehow the innovations start to move to that type of place.
31:48 Speaker 01 I would say that after the transistor, which was a huge deal after Bell Labs sort of invented the transistor, Bell Labs was incredibly, incredibly prestigious and everybody was like, holy shit, if these industrial labs can produce these hugely important valuable products. And so there was a lot of companies that were like, oh, I need to get an operation like that in my own big company. And so companies like Ford, companies like Xerox, IBM, to some extent, Exxon, they all set up these industrial research labs that were basically modeled on Bell Labs idea, which was hire a bunch of super smart researchers, give them a bunch of freedom to do whatever they think is valuable. And they did a lot of groundbreaking
32:31 Speaker 01 research. IBM Research has won a couple Nobel Prizes. Xerox, of course, invented a whole bunch of modern computing technology. Exxon, a lot of the initial research for like the lithium ion battery was done there. But none of these really managed to turn their research into huge money-making products. They did a lot of prestigious research and stuff that ended up being important in general, but not especially valuable to the company itself. Xerox, very famously, never managed to really capitalize on its part developments. I don't think Exxon really ever did anything with its lithium battery. So they just had a hard time turning this research into products that showed up on the bottom line. And so
33:14 Speaker 01 they all ended up moving away from this model where they would just give these researchers free rein, just do whatever they wanted.
33:21 Auren Hoffman? There's been a lot of talk about rare earth minerals in the US and how do we develop them or how do we get more, whether it's lithium or copper, whatever it might be that's out there. Is that something you think that the US is going to have to do more to just be more self-reliant? And if so, like, how are we going to get there?
33:40 Speaker 01 That's a good question. This is on my list of things to research. Rare earths, I have a big list of topics. So like you should learn about more about rare earth mining. You keep hearing about it in New York Times or whatever. I don't know a huge amount about it. I get the sense that the US has a very large number of deposits of these things. It's just a question of is it possible to extract them in an economically profitable way, basically?
34:04 Auren Hoffman? It kind of rhymes with fracking in a way, right?
34:06 Speaker 01 A little bit. I get the sense that the US has not been especially dominant in mining type developments. My vague sense is that that was an early casualty of not wanting like these incredibly environmentally destructive and disruptive industries. And so it's gotten very hard to like open a new mine in the US.
34:26 Auren Hoffman? And also the process of rare earths is very.
34:28 Speaker 01 Yeah, a lot of it's like pretty dirty, pretty gnarly. There's definitely more interest and more self sufficiency and stuff like this for defensive purposes or for whatever reason. I suspect we'll get back, do a little bit more of it. I don't really know the trajectory. I don't have a good sense of what the trajectory of it will be. I don't know. I haven't researched this area all that much. I don't have too, too terribly much to say about it.
34:50 Auren Hoffman? So I was an engineer. I went to college in the 90s. And if you went to college when I went to college and you were an engineer and you studied electrical engineering, computer science, maybe chemical engineering, this would be like a very lucrative field. If you studied things that are a little bit more traditional aerospace engineering, nuclear engineering, mechanical engineering, et cetera, you probably didn't do nearly as well on a relative basis over the last 30 years. And my guess is that a lot of people decided not to go into those fields because of that. How is that affecting today's market?
35:27 Speaker 01 I saw something just the other day. I think if you look at the top median earnings by career or something, it's still all engineering careers that are right at the top of the list. Electrical, aerospace, industrial engineering, all that stuff is pretty high up. I don't have a sense of why we're producing fewer engineers than maybe we'd like. I think for a lot of period in the 20th century, it was very prestigious to be like an engineer. Like that was something that people aspired to be. I think the prestige attached to it is not nearly the same as it was.
35:59 Auren Hoffman? And what do you think the reason is? I have a theory.
36:02 Speaker 01 I wonder if it's downstream of perceived less technological progress.
36:09 Speaker 01 In general, outside of a few things like semiconductors and stuff like that. So in like first half of the 20th century or whatever, there's all these incredible technology developments. Cars and airplanes and nuclear bombs and stuff, it's all getting better and more capable and stuff all the time. And so an engineer is that technology progress personified to some extent. And so if you're an engineer who works on that stuff and you're in some vague or indirect way responsible for this technological progress, that's quite prestigious or whatever. But if the last 50 years have been, a lot of this technology seems like it's maybe not advancing quite as fast. In many ways, it seems very similar than they did in the 70s and 80s and stuff like that. Maybe it's not quite so prestigious to become an engineer as maybe it was when you have progress looks a lot different.
36:53 Speaker 01 And maybe that's why China is able to produce so many engineers, or at least part of it, is because they've really seen huge amounts of development of progress overall and just development of living standards and the results of building like huge amounts of infrastructure and how that's been able to improve people's lives.
37:08 Auren Hoffman? It seems like if you're an engineer, you want to actually build something. That is the reason you're an engineer, you want to actually do something. So if you could build the SR71 in two years or whatever it took back then, you're like, oh, great. Like this is amazing. Like that must be a really fun project to work on. If you'd work on that same project today and it takes 20 years, that's your whole career to do one thing. And so you might as well just go into finance and make some money now, whereas before they got to build cool stuff.
37:36 Speaker 01 I've heard stories that traditional automakers like Ford or GM or whatever, that you would have people who's engineers who they only work on door handles or whatever.
37:43 Auren Hoffman? Right. Like terrible. Imagine 30 years in and you could show the three door handles you made or something. I don't know. It just doesn't seem that exciting.
37:53 Speaker 01 Anyway, what's your theory?
37:54 Auren Hoffman? I think that's exactly it. You want to see what you did in the wild. My wife's grandfather was a food engineer. He worked for General Mills and he got to make super cool stuff like Tang and Cool Whip and
38:08 Speaker 01 Ah, that's cool.
38:09 Auren Hoffman? All these other types of stuff. Every year he got to come up with something cool. He didn't make a lot of money. He worked for General Mills. He was just an engineer. But less than every year, he got to see a new invention that his lab did. And then he could see the neighbors buying that invention. So that was cool.
38:25 Speaker 01 Cool.
38:26 Auren Hoffman? We do have a shortage of engineers in the US, right?
38:28 Speaker 01 I don't know.
38:29 Auren Hoffman? You don't know? Okay.
38:30 Speaker 01 Yeah, I don't know. I know there's constantly concerns about labor supply. I'm more familiar with the construction industry. There's constantly concerns about having enough construction workers and stuff like that.
38:39 Auren Hoffman? Why aren't more people going into construction? Because it does seem like a fairly lucrative. If you're an electrician or something like that, it seems like a very, very hard skill to learn. But once you do learn it and once you're relatively competent, you have an extremely good guaranteed income. It seems like so few people are going into these types of fields in the US.
38:59 Speaker 01 Yeah, I think part of it is, what we talked about earlier is the status of these things has been lowered relative to other things. There's so much emphasis on like going to college and getting a degree and doing some sort of knowledge work, basically. And the relative status of construction work has fallen. I think it's not quite as lucrative as a lot of people say. You see these threads on Reddit, it's like, oh, you should go and be a plumber and then earn $150,000 a year. Very few plumbers make that much money.
39:26 Auren Hoffman? Yeah, maybe if you live in Manhattan and you have to earn like five grand a month on your one bedroom.
39:31 Speaker 01 Yeah, I'm sure you can, but I don't think it's the modal outcome. A lot of these jobs are really quite taxing to work in. Construction is very physically demanding, not necessarily after you're like super strong or whatever, but you're on your feet all day. You're bending back and forth in awkward positions. It's very taxing on the body to work a lot of these jobs. So it's not amazingly surprising that people are like less interested in that and more interested in jobs where they can work in a nice air conditioning office and boot numbers around and Excel spreadsheet.
40:03 Auren Hoffman? Yeah, that's a good point.
40:05 Speaker 01 Yeah, I've spent time doing construction work quite badly, I will say, when I did do it. And it sucks.
40:14 Auren Hoffman? Yeah, pouring concrete.
40:15 Speaker 01 Part of that is a personality thing where I'm not particularly well suited to it. But yeah, it was quite taxing. You're out in the sun all day doing manual labor. An office job is comparatively much more attractive.
40:26 Auren Hoffman? Yeah, it is funny. These people, they'll do the office job, but then they'll go every morning to Barry's Boot Camp or something.
40:32 Speaker 01 Yeah.
40:35 Auren Hoffman? What are some topics that you're researching beside the rare minerals that you haven't yet felt confident writing about yet?
40:41 Speaker 01 So there's usually a quick turnaround between when I'm interested in a topic and when I start researching it and when I write about it. I have a list of things that I want to learn more about, but usually, once I start learning about it and diving into the research, the essay starts to shoot up pretty quick. Water infrastructure is something I think I'm gonna start
41:01 Auren Hoffman? Oh, yeah.
41:02 Speaker 01 devoting more time to.
41:02 Auren Hoffman? That's interesting.
41:04 Speaker 01 I'm a civil engineer, so you sit by degree. So you would think I would understand it better.
41:08 Auren Hoffman? So you're talking about things like treatment plants, the movement of water, all that stuff.
41:13 Speaker 01 Treatment plants, how water infrastructure works at a high level and how it gets built, what gets built and what do we need to do different types of water tasks and how much water is used and where it's going. I just don't have a good sense of any of that, really.
41:27 Auren Hoffman? I 100% want to read that when you write that. That's super interesting.
41:30 Speaker 01 That's a pretty important part of infrastructure that I haven't really touched on at all.
41:34 Auren Hoffman? I don't know that much about it, but intuitively, it seems like something US does pretty well. Pretty much anywhere I go in the US, I can drink out of the tap, where that's not true in a lot of countries. I love the tap water in the US. Unfiltered tap water, it just tastes good. It's relatively good for you, I think. It's clean. It's ubiquitous, pretty much anywhere you go. That is a weird thing the US does well.
41:58 Speaker 01 Yeah. Again, this is the thing I don't know is how good is the water in the US compared to other countries? That's actually the interesting angle that I hadn't considered until just now. But how good is our water? You see stories like Flint.
42:09 Auren Hoffman? Right. There are some terrible stories, too.
42:11 Speaker 01 But yeah, that would be actually interesting to compare. How good is the water in another country in the US compared to other countries? How does it vary across the geography of the US? That'd actually be interesting to look at. So I'm gonna add that to the list of topics to write about.
42:22 Auren Hoffman? Interesting. All right. We have two more questions we ask all of our guests. One is, what is a conspiracy theory that you believe?
42:28 Speaker 01 Oh, gosh. I don't know. For a while, it really seemed like A lot of this UFO stuff, it was like, oh, there really are actually UFOs. And everybody in the Pentagon just knows this and we just didn't really talk about it. So that was becoming quite compelling to me for quite a bit of time. And then it was, wait, are there actually UFOs? Why isn't anybody talking about this? And then just recently I saw some article that was, oh, this was just disinformation deliberately introduced by the Pentagon to like,
42:58 Auren Hoffman? Yeah, it's like a psy op.
43:00 Speaker 01 Confuse some other people in the Pentagon, basically. So I don't know what to believe. It still seems perhaps somewhat plausible to me and I would love it if it were true.
43:09 Auren Hoffman? Is there any conspiracy theory in the construction industry that people have? There's some monopoly that's trying to move prices or anything or?
43:17 Speaker 01 Oh, I've seen discussion as to like, there's conspiracy among large home builders to like, restrict the availability of land, basically effectively drive up prices essentially. I haven't spent a ton of time looking into it, but historically in the US, home building was really, really, really fragmented. And even the largest home builders only had a pretty small chunk of the market. That's not nearly as true today and the largest home builders have a pretty big chunk of the market, actually.
43:46 Auren Hoffman? And probably even locally, they have a bigger chunk. Where I live in Northern Virginia, there may only be a few.
43:51 Speaker 01 Yeah, companies like Toll Brothers and stuff like that, that would be, and I like these big national companies are a lot bigger than they used to be. I haven't looked into it deeply, but it doesn't strike me as crazy just based on what I know about the changes in market share over time.
44:07 Auren Hoffman? There's also been some conspiracy theories that the train companies have been against the trucks or against the pipes. These other different conspiracies that are out there.
44:16 Speaker 01 Yeah, I think a lot of that is just the economics of something is bad and so it doesn't work out. And people don't accept that conclusion or whatever.
44:23 Auren Hoffman? Our last question we ask all of our guests, what conventional wisdom or advice do you think is generally bad advice?
44:29 Speaker 01 So my thinking about advice is that all advice is directional. So it takes you in a direction from a certain point, but you should do more of this, you should do less of this. You should spend more time exercising, spend more time with your family, should spend less time working, blah, blah, blah. But because it's directional, how good it is depends on where you are starting from. There's people that are like, you work too much and spend less time with your fam, you know, you should spend less time working, more time with your family. There's people that do not work nearly enough, but well-served by buckling down and focusing more on your job and grinding it out. So I think in general, people place too much emphasis on broad advice
45:11 Speaker 01 that's not tailored to their necessarily specific set of circumstances. And in general, advice is good to the extent that it takes into account where you're starting from and what would move you in a most profitable direction. And I think most advice is not like that kind.
45:26 Auren Hoffman? I highly encourage everyone to subscribe to Construction Physics. I also follow you at underscore Brian Potter on X. By the way, why the underscore? Was Brian Potter taken or something?
45:37 Speaker 01 Yeah, Brian Potter was taken. I could have done Brian Potter 2727 or something like that. But the underscore seems good. I saw a few other people do it. I was like, oh, this seems like a good way to get something that's close to your actual name.
45:50 Auren Hoffman? Yeah, okay. You have to somehow buy it from the other guy.
45:52 Speaker 01 Yeah, I know. Maybe somebody will sell it to me.
45:56 Auren Hoffman? Well, this has been awesome. Thank you, Brian Potter, for joining us at World of DAS. And I definitely encourage our listeners to both subscribe and follow you on X. This has been really great.
46:04 Speaker 01 Yeah, thanks for having me.
46:08 Auren Hoffman? Thanks for listening. If you haven't already done so, please subscribe to World of DAS wherever you consume your podcasts, YouTube, Apple, Spotify, and more. And please help us get discovered by leaving a review. And check out worldofdas.com. That's worldofdas.com. And of course, connect with me on Twitter, at Auren. That's A U R E N. Would absolutely love to hear from you.
Previous Airtable CEO Howie Liu - Excel vs Access, Founder Mode, and the Future of SaaS Next The LM Brief: Conditional Voting Rights in VC Term Sheets Page updated November 1, 2026 · transcript and summary are machine-generated