Key Points From This Episode:
• Eli starts by explaining what total factor productivity (TFP) is and why it matters.
• Why he is focused on TFP as a means for future economic growth.
• Five inventions that led to past TFP growth, from the internal combustion engine to electricity.
• Learn about the Great Stagnation and why Eli is concerned about TFP growth since 2005.
• How reliable total factor productivity statistics are and how they are calculated.
• Why there appears to be a disparity between dissemination of products and the rate at which innovation is impacting TFP growth.
• What neo-medievalism is and how charter cities and online communities fit into this concept.
• How technology drives political change and the interplay between technology and geography.
• Eli shares his predictions for a technology like crypto currency in the next 10 years.
• Discover the technological developments Eli is most excited for, from biotech to supersonics.
• How we should think about leveraging policy reforms to impact the development and deployment of new technologies.
• What percentage of TFP decline can be attributed to regulation versus other cultural factors.
• How a higher percentage of linoleic acid in our diets is literally making American softer.
• Find out what NEPA is and how it prevents the government from making swift decisions.
• Eli reflects on the social change necessary to embrace necessary new infrastructure.
• How environmental organizations use NEPA to prevent beneficial projects from going forward.
• The growing eco modernist movement advocating for growth and environmental protection.
• Eli shares his idea for a remotely piloted airship that could move more cargo, more quickly.
• What Eli believes we can expect the impact of AI to be over the next decade.
• Learn why it would be a positive step to climate change to bring back wooly mammoths.
• What key industries Eli would target in a charter city to allow for greater technological innovation than the US; housing,
• The potential benefits of medical tourism for charter cities to recruit wealthy residents.
Mark: Hello, and welcome to the Charter Cities Podcast. I’m your host, Mark Lutter, the Founder and Executive Director of the Charter Cities Institute. On the Charter Cities Podcast, we illuminate the various aspects of building a charter city, from governance to urban planning, politics to finance. We hope listeners to The Charter Cities Podcast will come away with a deep understanding of charter cities, as well as the steps necessary to build them.
You can subscribe and learn more about charter cities at chartercitiesinstitute.org, follow us on social media, @cci.city on Twitter and Charter Cities Institute on Facebook. Thank you for listening.
My guest today is Eli Dourado. He’s an economist and regulatory hacker living in Washington, DC, and a senior research fellow at the Center for Growth and Opportunity at Utah State University.
Mark: Welcome to the show, Eli.
Eli: Thanks for having me, Mark.
Mark: What is total factor productivity and why does it matter?
Eli: Total factor productivity is all the growth that you get that doesn’t result from just applying more inputs. So, you could always grow your GDP by having more labor and more capital applied or everybody works longer hours or everybody saves more, and so on. That’s output that there’s some cost to you, right? You have to work more hours and that’s the cost.
Total factor productivity is the change in output that you get for free, that just hopefully increases year over year with either new technology or better institutions.
Mark: Yeah. I mean, it’s not free, is it?
Eli: It’s not easy. Let’s put it that way. It’s not easy to get but if you improve your TFP, you get more output forever, in principle.
Mark: Yeah. No, it’s just what it is. Why does it matter?
Eli: If GDP matters, and I think that’s the starting point is does GDP matter? I think GDP is pretty well correlated with all kinds of values that we care about, in terms of material welfare, in terms of living standards, in terms of how much leisure time we can have, or lots of other values that we care about, even including non-economic values. We have more time and resources for them when GDP is higher.
Then, TFP is the best way to get GDP because it doesn’t take up additional labor hours or other resources to get them, to get higher GDP. So, TFP is sort of the purest form of economic growth is just getting more output with the same resources year after year.
Mark: Why don’t – alright, because you’ve been kind of beating this drum for a while, why do you think other people are not as focused on TFP as you are?
Eli: That’s a great question. I don’t know. I’ve been thinking about economic growth since I was nine years old. That’s just unusual, right? For people to be that obsessed with it. It’s the statistics come out once a year, or once a quarter or something that. So, it’s not that relevant in the news. It’s very highly, highly abstract. It’s not a source of policy debate. It moves very slowly, relative to the things, the levers that government has.
Mark: Yeah. I mean, I get all of that, but you’d still think there’s lots of – I don’t know, weird, esoteric debates that occupy certain niches of the internet, where you might have a few dozen people arguing or a few dozen researchers. You’re – there is kind of the progress studies as a whole, and even within that, you’re already the only one who’s beating the TFP drum a little bit more heavily.
Eli: Yeah. So, within progress studies, there are people who focus on science productivity, and is the science getting better? There are people who think about historical, what led to progress in the past in terms of historical circumstances and so on. My view is that it’s underrated to focus specifically on the last step in the innovation process, the one that actually gets you products at scale that can affect productivity, and look at that in a forward-looking direction rather than just through a historical lens.
I don’t know why more sort of progress-oriented folks aren’t taking that approach, but I’m hoping to sway some of them. I think it’s good that we have these other approaches as well. So, I’m not trying to get everyone has to be on my train. But I would like to get a little bit more momentum behind it.
Mark: What single biggest invention or institution in the past led to the greatest TFP increase?
Eli: So, Robert Gordon. I think he’s wrong about a lot of things, but he is probably right about the great inventions in terms of what happened in the last century. So, he kind of says, argues that there’s five great inventions, things like internal combustion engine, the idea of pharmaceuticals and chemicals, the idea of sanitation and indoor plumbing and things like that, and a couple others that escapes me right now. Electricity is one of them, and then there’s one other that I forget.
I think Gordon’s historical work is pretty good, as far as it goes. I think it’s the interpretation that maybe is messed up. I would say he’s probably nailed it, that that was a huge deal from 1920 to 1970. Those inventions that were all invented in the late 1800s took a few decades to get started in terms of commercialization and privatization to get to, and then sort of scaled up over that 1920 to 1970 period. Then, that was a period that we had two percent TFP growth per year. That’s pretty awesome, and we should try to do that again.
Since then, we’ve had, I think he would say, and I would agree with it as far as it goes, that we’ve had the one great invention; the internet or IT. There hasn’t been something comparable. You could maybe say, in the last 10 years, fracking was pretty good, but –
Mark: But I think it might be helpful for some of our listeners to just go through, , reconstruct that argument broadly. You are kind of doing it now, I think, just responding, somewhat circumstantially to the question. But what is the progress studies argument broadly with TFP put at the forefront in terms of how that’s changed since the mid to late 19th century and today?
Eli: For most of human history, the TFP didn’t budge, right? So, if you go back, hundreds of thousands of years, it didn’t budge. It started growing in the Industrial Revolution, maybe a little bit before, and then these inventions in the late 1800s took a few decades, then middle of the 1900s, two percent per year. Then, since 1970, in 1970, it fell by over half, so less than 1 percent a year. There was a little resurgence in the mid 1990s. It lasted about 10 years in which it was growing at two percent again. Then, since 2005, it basically flatlined or, I mean, it’s 0.3 percent a year since 2005. Especially if you use utilization-adjusted TFP, which is take account of the business cycle, which is important.
If you look at it in that time span, it’s concerning, and Tyler Cowen has raised the point, since the 1970s, that’s the Great Stagnation, the period since 1973, he would say. But I think actually the period since 2005 is actually a lot more concerning in the sense of the amount that it dropped – the TFP growth dropped, was pretty substantial. We went down to 0.3 percent a year since then. That is really low. I mean, that’s pre-20th century levels of TFP growth. That’s kind of scary. We need to put it at the center and figure out how to do it. I think part of that is fundamental science, but the bigger share is actually getting products to market so that they can be at scale, and so that they can affect, actually, productivity.
I’m guilty of this as much as anybody but there’s this thing called science porn, where you read some exciting new journal study, and you tweet about it all excitedly. I do that all the time. But every new discovery in a journal, it doesn’t affect TFP until it becomes a product that can be disseminated at scale and actually affect how people are doing stuff. That’s the step of the process that I’m most concerned about, is barriers to taking these ideas. Even Gordon is a little guilty of this because he talks about great inventions. Well, it’s not the inventions that did the stuff. It’s the dissemination of the inventions. That’s what’s important.
Mark: How reliable are TFP statistics? How are they calculated? You say there’s this substantial drop in 2005, and my impression of some economic statistics, it’s like, “Alright, there’s something there.” TFP, to me at least, there’s no easy, natural intuition where it’s unemployment rate. You call up 5,000 people and ask, are you employed or not? Obviously, there’s challenges with respondents with cell phones, blah, blah. But there’s some kind of baseline understanding going on with TFP. It’s like, alright. I presume there’s, I don’t know, four statistics that are inputs to this, but that noise ends up getting pretty substantial at some point.
Eli: Sure. It’s totally possible. In terms of how it’s actually inferred, it’s basically a regression residual, right? So, if you take, log GDP and regress it on log labor hours, and in log capital. Some fancier models use a measure of human capital as well. Then, everything that is not explained by sort of the effect on output that’s not explained by the real inputs, right? Capital and labor is attributed to this sort of more abstract level of technology and institutions, TFP.
On one hand, that’s kind of a weird way to do economic statistics. You’re not measuring it directly, you’re inferring it based on everything else. On the other hand, we do have pretty good labor statistics. It’s not like there’s lots of capital that is missing in terms of our statistics. I would say one other thing is that, I would definitely be sympathetic to the idea that you could miss on a TFP estimate for one year or one quarter or something like that but, for the trends to be so striking and so consistent, that, to me, it doesn’t seem that’s a measurement error.
Mark: You mentioned previously, the commercialization and distribution of these inventions is good. How do you think about that? Typically, when I think about what might be called broadly industrial cycles, you basically have 60 to 80-year cycles, from kind of first deployment of a technology to that what might be described as an 80 percent utilization rate. Is that the right framework for thinking about it? And when does that out of deployment actually start showing up in TFP statistics? Because if we think about the internet, that started kind of in the early ‘90s. We’re 30 years into this cycle and it’s not only not showing up, it’s actually like we’re seeing a drop. Maybe it is showing up, it’s just that the drop in other areas is bigger. But that strikes me as, I don’t know, somewhat out of whack with what my intuition would be.
Eli: So, in terms of actual dissemination of products and stuff, things are getting faster over time. It took a long time to electrify the country, and it’s taking less time for everyone to adopt with smartphones. So, in terms of dissemination of products, it’s getting faster. You would think the rate at which innovations are starting to affect TFP should be going faster, but doesn’t seem to be the case. A lot of this IT stuff doesn’t seem to be massively increasing TFP on an appreciable rate or a rate that we can say, “Oh, there’s a big bump from it.”
Mark: Alright, well, let’s get on to other non-TFP related questions. What’s neomedievalism?
Eli: Neomedievalism is this idea, popular in the late 1990s, or not popular, but first discussed in the late 1990s, that power is getting less monopolized or power is not being tiered as much the way it used to be. If you think about, in the medieval era, people had loyalty to their local Duke, they had loyalty to the Catholic Church, they had loyalty to their professional guild, and all these institutions had jurisdiction or real political influence over people’s lives. But they weren’t territorial in the same way or, if they were, the territories were overlapping and the way that power operated was not just monopolized by a king who had Dukes beneath him with exclusive territories, and they had counts beneath them with exclusive territories, or something like that. It was just much more squishy and messy in the post Westphalian era. We’ve gotten to the modern world where states exist and they have territorial sovereignty.
Basically, all our allegiances are tiered up to one institution. So, the idea of neomedievalism is that we’re going back to that messier era, where our political allegiances and actual exercise of power could potentially be more messy, not tiering up to just a single entity, and more interesting, ultimately, right? I expect it will be interesting to watch if this trend, indeed does continue to play out.
Mark: What do you include in that, I guess, trend? Do things like sanctuary cities, potentially CHAZ, charter cities, do all of these fall into these buckets?
Eli: It could fall into that bucket. The sanctuary city is still fully encompassed by some political entity and subordinate to it, they’re just sort of resisting a little bit. I think it could be that or it could be also things like online communities that vie for our allegiances a little more. So, it could be that sort of the socialists are all, they’re more loyal to global socialism than they are to the government that they happen to live under. Or it could be that other people have online communities that they feel very strong affinity to and that they’re more loyal to, in some ways at least, than the formal political institution that governs them.
I think it is interesting for charter cities, because it means you’re not trying – you might not be trying to replace the state at the height of its high modernist power, where it has to be everything for everybody. It can just be focused on providing the sort of the necessities and allow people to have their other political communities that can influence their allegiances as well.
Mark: How do you think that affects the development and distribution of technology? I mean, sometimes I think people, particularly libertarians, I think, underrate the modern state in the sense that having all of these different local municipalities oftentimes can be very bad, right? In Germany, you would pay taxes every mile down the Rhine, because there’d be a different institution that ends up severely restricting trade. Then, two is just having a large internal market is good. That allows you to do very capital-intensive investments with long-term horizons, knowing that there is the possibility to recoup those investments at some point in the future when, if you have much smaller markets, that can be much more difficult.
Eli: It’s a great question. I don’t know if there’s a clear answer. I wouldn’t say that the nation state is going away necessarily. So, there could still be a United States with a multi-trillion-dollar budget. I just think that there are going to be subnational, supranational, transnational institutions also, that vie for political power or that actually exercise political power to some extent. I think it’s also the case that technology is – primarily, it’s technology that drives the political change. There may be a feedback loop where this change in the structure of power also affects the kinds of technologies we get, but I think a lot about the dialectic between technology and geography, because so much of political institutions comes out of geography, of big territories, of transportation networks, of the availability of boards. That’s the kind of stuff that I think, gradually, technology could conquer and make location irrelevant.
I have two thinkers that live rent free in my head. One is Peter Zeihan, who is like, geography is everything and describes all of political power. Then I have who is like, good geography doesn’t matter anymore. We’re transcending it almost completely. I think, Peter, probably more close to right in the short term, but could be right in the long term. That interchange between geography and technology and the way that they affect power is, I think, an interesting thing to watch.
Mark: Are there any kind of specific testable predictions that you put on that in the next, let’s say, 10 years?
Eli: An interesting thing to watch will be the crypto world, right? So, I think, one way this could play out would be that a chain like Ethereum, that reaches scale and is able to do – is programmable and can do smart contracts, something like that could become a global financial backbone. That is something that could make nation state based financial networks irrelevant. So, you still have nation states, in terms of police power, and all that stuff. But your financial system becomes essentially globalized, because the rails are truly global. So, that’s an example of an institution where it doesn’t – the locus of power shifts away from just a sort of monopoly by an actor in a single territory.
Mark: I think you mentioned this kind of briefly before, but what technological developments are you most excited about over the next, say, 20 to 30 years?
Eli: Let’s see. I think biotech is really hot right now in terms of science, in terms of the scientific breakthroughs. So, I think one thing I’m interested in watching is, are those being translated into into therapies and treatments and devices that can monitor us? I’m watching the longevity space pretty closely. I’m really interested in the way that we do medicine in the future, in terms of – so, if you go to the doctor today for your annual checkup, they actually don’t do that much to you. They listen to your lungs and ask you a few questions and so on. It’s actually not that high quality of a method of surveilling your health. So, medical devices, in terms of, you can imagine diagnostic grade, medical devices that you have with you all the time. Think of that the Apple Watch, as it will exist 20 years from now. That could replace a lot of primary medicine and I think that’s really exciting.
Energy and transportation are kind of two areas that I’ve focused on a lot. I’ve spent a few years working on supersonics and I think, over the next 20 years, I think we’ll definitely see a resurgence in high speed, flight and energy. I think everybody wants to do something about climate and I think that’s great, but I think a good side benefit of that is that people are rethinking energy. It’s possible we’re going to come out of this climate thing with new modes of clean energy that actually get us more energy, and much higher energy output than we have today. We’re going to do a lot of interesting things with that, new materials, new processes, and so on, that could have a pretty big effect.
Mark: How should we think about the policy lessons of supersonic? Because you and Sam Hammond wrote that paper, what, four years ago? And now we’ve seen it with Boom and a handful of other supersonics companies that are getting a lot of attention and are kind of taking off. I don’t actually know if there are any – I think there were meaningful policy changes that resulted in part from your paper, but how should we think about levering policy reforms to impact the development and deployment of new technologies?
Eli: Yeah, so I think there has been policy progress on supersonics lately. Yeah, that paper came out in 2016, and then I started Boom in 2017, and by early 2018, we had a good change from FAA on sort of the way they were approaching a takeoff and landing noise standard, which was one of the key issues that we identified. There has been a lot less progress on the area that Sam and I said was most important to work on, which is the overland ban. So, supersonics, it’s still – you cannot – without special permission for a test flight or something that, you cannot fly supersonic overland in a civil aircraft. And we’re like, “Well, let’s just replace the band with a noise standard. It’s not that hard.”
Mark: Would that be FAA? Or would that be a legislative decision?
Eli: That’d be FAA. FAA could do it. The challenge is they’re required to do something that’s – they’re required by statute to do something that is appropriate for the kind of aircraft, right? So, what they would need to know is what is a feasible noise level? Based on a relatively mature design, what is a feasible noise level that an aircraft could reach on cruise in terms of sonic boom? And then the other thing that they need to know is that they would need to do an environmental review, if they wanted to lift the ban, they would need to do a study. They don’t have, what they think is enough data to do that, in terms of the effect on the human response, and so on.
I mean, it’s been studied to death, right? There is tons of research, back from the 1960s and ‘70s on this, but they’re like, “Oh, people respond differently today than they did in the ‘60s and ‘70s, to sonic boom.” So, in terms of that, there is a NASA program to basically build a low boom supersonic jet, and fly it, and see how people respond. That will probably happen in the next few years. But to my mind, you just need to have a company that is going to build a low boom supersonic jet, and they’re going to say, “We’re going to build this no matter what,” and then FAA is not going to block that. They would take the data from that aircraft and use it to sort of inform what the standard should be. But you need pretty ballsy investors to go for that, and I think that’s the main obstacle so far.
We’ll see. There’s a company called Exosonic that is building a jet. So, we’ll see if they get far enough and mature enough to force the hand of FAA. In terms of how to do this generally, I think a lot about – my approach is, learn about the technology, learn about what it would be – I put myself in the shoes of, what would it be like to have a company that is trying to do this? What would the barriers be that I experience? And try to work on the policy changes that are actually blocking the entrepreneurs that are doing this thing. That’s the kind of the mode of policy work that I’d to see more of.
Mark: What percentage of the slowdown in TFP do you think is for regulation, versus what might be described as a general cultural malaise, versus just the fact that there aren’t as many low hanging fruit as there used to be?
Eli: Yeah, it’s a great question. Definitely, some of it has got to be cultural and institutional, right? Especially, this is obvious if you look at a bunch of other countries. Italy, and Spain, I think is actually – TFP is slightly going negative lately. That’s not attributable to people forgetting technologies or anything like that. It’s got to be something cultural and institutional. And then Brazil, I think it peaked in 1980, the TFP. So, today, you need 50 percent more real resources to get anything done to get the same amount done in Brazil, that you did in 1980.
I think some of the decline has got to be cultural and institutional. I don’t really attribute much to the lack of low hanging fruit. I’ve wrote a blog post six months ago, arguing about, here’s all the things that we could do in the next 10 years. It’s a pretty good list. So, I think there’s plenty of possibilities in terms of technological progress. I do think a big obstacle is regulation and barriers to commercializing technologies, especially at scale. I would say a big chunk of its institutional and the rest is regulatory.
Mark: So, have Americans gotten softer over the last 100 years?
Eli: It’s a great question. In the obscure corners of the internet that I like to read in, I’ve come across a theory on and it’s basically due to vegetable oils, which basically were zero or a very low percentage of American diets 100 years ago and are a high percentage today. And in particular, vegetable oils, like soybean oil in particular has 50 percent one particular fatty acid called linoleic acid, which is an Omega six fatty acid. It’s just highly unusual, sort of in the evolutionary environment that we all evolved in to get that much linoleic acid or that much omega six. You’re not going to be able to do that.
Linoleic acid is something that the body does not produce itself. You can only get it from what you eat. You do need a tiny amount of it, 1 percent of your calories or something like that would be plenty. People are getting things 10 percent or more of their calories from linoleic acid and that is just highly unusual, and it bioaccumulates. So, it’s one of the one of the few things you can eat that, if you eat a lot of it, you will become more of that thing, to a considerable extent.
There’s one guy, Brad Marshall, who blogs at Fire In a Bottle. He has a really interesting background. He’s sort of a trained – has a degree in molecular biology or genetics from Cornell, and worked in that field for a while, but he’s also a pig farmer, and a chef, and a butcher. He came to it through noticing the fat quality on the pigs that he was raising and butchering would vary based on the feed. So, that’s when he realized, “Oh, this is bioaccumulating, it’s kind of weird.” Basically, I think the fat that we consume has gotten much, much more unsaturated, and that has metabolic effects. When you burn it in your mitochondria, it behaves differently than it would if you’re burning a saturated fat or monounsaturated fat, and then that can trigger, because of the byproducts of that burning, it can trigger different reactions in your body than it would if you were burning saturated fat.
That’s the smoke, and then the all these other things that come downstream of that are probably the fire. So, basically, it’s just changing our metabolisms by eating different stuff.
Mark: So, it just puts us in a 1/10 of a hibernation state all the time because our metabolism is slow when we get fat?
Eli: It is linked to torpor, hat some animals do to hibernate, right? Hibernation is not sleep. You’re actually in a torpid state. Your body temperature goes down and your metabolism is in a conserving mode. The way that animals get into that state is they eat a lot of linoleic acid. So, they’ll eat acorns and stuff like that. They’re high in linoleic acid. That is what triggers it in basically all mammals that hibernate is something like that. This a highly conserved genetic function. I think it predates the bird mammal split. It’s really a highly conserved evolutionary mechanism. So, it makes sense. There are lots of mechanisms. So, it’s not proven. I think we need more research to prove that this is what’s going on. But that’s my current thinking on why it might be that we have an obesity epidemic.
Mark: Have there been any kind of studies where people restrict their intake of seed and tuber oils for a year and what the outcome is?
Eli: No, there haven’t been. So, that’s something we should do is – there’s been studies on mice and so on, and there’s short-term studies on people, but you wouldn’t expect the short-term studies to make much of a difference because the half-life of this stuff in your body is on the scale of years.
Mark: What is NEPA?
Eli: NEPA is the National Environmental Policy Act. It was passed in 1969, took effect January 1970, and it basically says that, before the federal government can make any decision, or take any action that could or that would affect the human environment, they have to produce a detailed statement of what those effects are. That in principle is, before we decide we’re going to do this, you got to say what the effects are. In the initial days after NEPA was passed, the government would do, take some action, they would put forward a 10-page statement saying, “Okay, this is what – we think this is going to have these effects. We’re going to do this anyway,” but it’s a 10-page paper, and that was fine.
Then, through litigation, and implementing regulations, and so on, that sort of paperwork burden grew and grew and grew over the last several years. Now, if the federal government is going to take any action that could affect the human environment, it’s a hundreds of pages report that they got to do, sometimes with thousands of pages of appendices, and they’ve got to do public hearings, and they’ve got to have public comments on it, and so on.
Once the decision is made and finalized, then they can be sued, and they can go go to court, and the decision can be stayed by a judge while that’s pending. So, it gives people cause of action to sue through the Administrative Procedure Act, that this report didn’t cover all the issues, let’s say. The remedy there is go back and do more research, and go back and improve the report even further. It delayed a lot of projects, and it enables people to sue and stop things that they don’t like.
The problem is that it gets even worse than that, because it also implicates every other decision that doesn’t affect the human environment. So, let’s say I’m an agency, I’m going to take an action that doesn’t significantly affect the human environment. Well, I have to prove that. So, to prove that this 500-page-long, five-year process thing, to prove that I don’t have to go through that, I have to do another report, which is a little bit shorter sometimes because you don’t have to consider all the alternative actions you could take. But it’s still very often takes two years or more to do that, and you also have to have public hearings and all that stuff around that. So, the longer report is called an environmental impact statement, the shorter report is called an environmental assessment.
The government does a high number of environmental assessments, tens of thousands a year of these two-year process types reports. So basically, what it means is the government cannot make decisions quickly. I think NEPA has basically lobotomized the federal government in terms of its ability to actually decide or approve an action. And this affects the private sector because it affects all permitting action. So, if I want to build a tunnel that takes people between DC and Baltimore, as Elon Musk is trying to do, you have to go through an environmental assessment process that takes years. If you want to build a new nuclear power plant, guess what? This is one of the biggest obstacles there. If you want to do geothermal well on federal land, you might have to go through this process multiple times.
Basically, it means it’s very hard to build in this country and this is a big part of why. A lot of times, we think about it being difficult to build in terms of zoning rules for housing, but at the infrastructural level, it’s NEPA that’s holding everything back. Even supersonic flight that we were talking about earlier, this is the reason that FAA has to do an environmental review on what it would be to allow sonic booms, it’s NEPA. So, it just touches almost everything in the economy and it makes the government slow and stupid.
Mark: Let’s then, I guess, chat a little bit about social change. How do we then – , but then also part of it is people truly believe this kind of stuff, and think that it is good. And maybe it’s just because it has the word environmental in the name, but how do we actually go and move from a people who are kind of serious about building in the next 20 to 30 years? How do we ensure that we actually are from a kind of regulatory as well as a cultural perspective, able to embrace these positive, constructive elements that allow us to build things?
Eli: Yeah, so I think with NEPA specifically, nothing is going to change until we get people who actually care about the environment to realize you’re never going to get everybody. There are some people you’re not going to get, and I’ll talk about that in a second. You need to get people who care about the environment, genuinely care about the environment, to see NEPA as an obstacle to their goals, right? And to see , “Okay, this is a big procedural thing.” The government can decide to go through with something, even if it does harm the environment. So, in that sense, NEPA doesn’t actually protect the environment directly, not in the same way that the Clean Air Act or the Clean Water Act sets standards for what you put in the air and the water. NEPA doesn’t do that.
We’re going to need a lot of new infrastructure, green infrastructure, high speed rail, all this stuff to improve the local environment in terms of air pollution and water cleanliness. You need new infrastructure to do that. Then also to fight climate change, you need to build a whole lot of things and we’re not going to be able to do that effectively if NEPA doesn’t change or get replaced. The big problem that you have is that NEPA gives this legal cause of action to environmental lawyers. There’s a whole class of existing environmental groups.
Mark: So, first kill the lawyers?
Eli: Basically. So, this is their bread and butter, right? So, there are a lot of big, nonprofit environmental organizations, this is what they do, is they use NEPA as a way to get a cause of action to stop things, bad things from happening, that they don’t want to happen. That’s their business model. If they didn’t have NEPA, they wouldn’t be able to raise the money, I would be super happy if environmental groups would lobby for stricter environmental standards, or something like that. But that’s actually not their main thing that they do. They sue to stop projects from going forward. That’s their business model, and you’re never going to get them to say, “Yeah, we should get rid of NEPA, because it’s holding back progress,” because it eviscerates their business model, right?
You need to get genuinely left to center environmental people who realize this is not a good system. To me, it’s winning over those people, that’s how we would ever be able to get a lasting change in the way we do environmental review in this country. So, Trump put forward a regulatory change in the NEPA implementing regulations. He got it through in 2020, and now Biden’s probably going to undo it all, at least some of the key aspects. It is a really unhelpful dynamic.
In the past, both Republican and Democratic administrations have tried to rein in this permitting problem, but because Trump made it sort of a Make America Great Again type program, I think that Biden now feels like he’s got to do something about it.
Mark: Yeah. I think the way you’re describing it, I’m just not as optimistic. I agree, alright, let’s get the greens on the side of common sense. But then you see the kind of dominant energy is not just “ban nuclear power” but “shut down existing nuclear power.”
Eli: Shut down existing. Yeah.
Mark: So, it’s literally, I mean, Germany shuts down existing plants and and they increase their coal production. They’re shutting down, what was it? Diablo canyon in California that produces about 40 percent of the non-carbon polluting energy. Let’s convince this group but kind of, by all appearances, they celebrate Greta Thunburg, who literally wants to end industrial civilization.
Eli: Yeah, I totally agree with you. It’s interesting, there is a small but hopefully growing eco modernist movement, right? People who say the FDA environment is critical, but we also need to grow, and we also need to grow so that we can preserve the environment and grow so that we can protect the environment, and also growth is good. I think it’s the sort of the future of environmentalism is going to be eco modernism versus eco primitivism, the people who think we should all be living in straw huts in developing countries and reduce our standard of living to that level. Probably, a lot of us are going to starve and so on. They would think that it would be good.
Mark: But they’re not the ones who are going to starve.
Eli: Well, yeah, exactly. But they would, right? If they actually got their way, a lot of them probably would. I think that’s implicitly the next 20 years of environmental politics is eco modernism versus eco primitivism. If that precedes, especially if it can become more of a high-profile fight, in terms for the soul of the movement, I think you’re going to see a lot of sensible moderate left people being , “Oh, yeah, eco modernism. That’s what we want. We want growth, and we want environmental protection as well.”
Mark: I really hope you’re right. But just looking at it, this fight has been going on for the past 30 years, too, right? You have people like , who at least, I don’t know how true his narrative is. But the way he tells it, he was a lefty student who was doing statistics and heard somebody come on the radio and basically say that most of the environmental statistics are wrong. He was like, “No, they’re right.” He went online, looked it up, and then he found out that this dude was right, and kind of converted his beliefs because of that. So, he tells the kind of experience of conversion through exposure to data, but he’s just pegged as a righty now.
Eli: Yeah, he hasn’t changed his fundamental politics at all. He’s still very left of center.
Mark: My kind of worry is that, basically, as soon as there’s been a handful of people who are able to go right kind of ego modernists without being tagged as righties. Brand is one who has enough lefty credentials. But by and large, I think most people if they start engaging in these kinds of public discussions, they’re just going to be pegged as a righty, and then they’re going to be ignored.
Eli: It’s interesting, also, the left wants infrastructure. It’s like, “Okay, you want infrastructure, if you want to build, we’re going to do it.” And I think also what’s really interesting is in 2009, the Obama stimulus, the American – I forget what it’s called something Recovery Act ARRA, they had to do tens of thousands, hundreds of thousands of NEPA reviews to implement all that, to spend the money. None of it got spent during the recession, or a small amount of it got spent during the recession. It was all getting spent in actually in terms of jobs and stuff in 2011, once the the formal recession was over. If you care about things like being able to respond quickly to an economic crisis, you need to reform some of this.
I think in general, if you think about the intelligence of an organization, as the ability to make quick and high-quality decisions, NEPA lobotomizes the federal government. It destroys the ability of an organization that the left is supposed to cherish, the state. It destroys the ability of agencies to make quick decisions. That’s not good, if you are a generally a pro government person. You should want the state to have high state capacity. You can’t have that. You can’t have that and NEPA. You can have one or the other.
Mark: I fully agree with your arguments. I’m just kind of thinking about it from a cultural slash political perspective of will they persuasive? Where, I mean, we’re seeing this out and kind of COVID, there are some corners of the internet that are basically saying, look, “There’s a massive failure of public health institutions.” And then you drive for Bethesda, and people have , Fauci yard signs. You’re starting to see the penetration in some circles. I kind of joke that the most important electrical trend today is the GMU to the former Vox pipeline, where you have implying and have been GMU build to certain extent, and are kind of realizing this, and they’ve got enough lefty credentials that they’re not getting exorcized or kicked out of the “movement”. But right, it takes it does – I don’t know, I’m figuring out how to thread that needle and get that consensus that allows reforms, I think, is really critical.
Eli: Yeah, totally. I mean, I agree with you. It’s going to be a hard slog to improve NEPA or to reform it in a way that lets a lot of progress happen. But it’ll be a fun fight either way.
Mark: So, you want to start an airship company. Tell me about that.
Eli: Yeah. So, I don’t know if it’s me starting it. But I think that somebody should –
Mark: You want to convince somebody else to start if, so have a business model –
Eli: I would be happy to even be involved, right?
Mark: Assume the future co-founder is listening to this podcast. What’s the pitch?
Eli: Yeah, so the pitch is that airships get more efficient the bigger they get. So, there’s a square cube law, the lift of an airship is proportional to its volume, and the drag of the airship is proportional to its cross-sectional area. So, the bigger the airship gets, the better the L over D ratio gets. Then, there’s other scaling factors as well, so if you are talking about a rigid body or shape, which I think you should, the weight of it doesn’t scale as fast as the volume either.
What this amounts to, is what we should use – we could use airships for cargo to take 500 tons of containers at a time, thousands of miles, and you could do it a lot more cheaply in terms of per ton. You could do a lot more cheap – or per ton mile, you can do a lot more cheaply than a 747 freighter. What it does is, so you can think about the existing cargo market today –
Mark: I want to compare to rail because that seems to be the kind of primary competition.
Eli: I think the way you would want to do this is transoceanic.
Mark: So, what would be it competing on? Speed, then?
Eli: Right now, the sort of transoceanic market, you either, your choice is pay a really high rate per ton mile to take it on a plane or a very absurdly low rate to put it on a ship. And, you know, if you put it on a ship it’s, it takes weeks or months to get to your final destination, because the ship stops at every port along the way. If you did point to point on an airship, you could do it in – so, basically it’s a five to 10x reduction in cost over airplane freight, and a five to 10x speed up over ocean freight. It’s a pretty nice middle market.
I think it’d be ideal for things sort of middle value to weight ratio, things like machinery, automobiles, parts, and equipment for factories, and so on. That’s probably the market you want to put it. If it’s higher value to weight ratio, you probably want to put it on a plane. If it’s lower value to weight ratio, you’d want to put it on a ship. But basically, if you target that kind of market, and let’s say you get five percent of today’s ocean container volume, and you put it on there, that’s a fleet of 4,000 airships and couple hundred billion dollars in revenue a year, maybe 100 billion dollars in profit a year, operating profit.
It’s a big market if people do it. Now the problem is, you can’t start this off with a sub scale airship. If you started off with – I’m going to do a 500-ton payload airship, I’m going to do a 10-ton payload airship, your economics get way worse, because of that scaling factor that I was talking about earlier. Because it’s more efficient the bigger it is, you want to start with a really big one. There is no path to sort of gradually scaling up, you need to start big. So, that’s the challenge with it. I think it’d be a really good business.
Today, more than ever, the time is right, because we have the autonomy systems to do this sort of either remotely piloted or with some sort of AI on board that pilots the thing. That saves you from crew costs, from having to put bunks on board for rotating crews, labs, heads, et cetera. You don’t need galleys, you don’t need any of that. You don’t need food on board. You don’t need to support the crew on board at all, and you can just do it completely autonomously. So, I think now’s the time to do it. It’s a good moment.
Mark: Yeah, I guess kind of thing that I think about when we chat about that is is Africa, because Africa has very poor internal infrastructure. And then two, is sometimes the climate is so poor that the maintenance costs are substantially higher than in more temperate climates, as well as you have a much more limited amount of work time availability, given that it’s very difficult to be productive on large infrastructure projects, when you’re getting several inches of rain a day, which might happen for several months a year. Because of that, it’s possible to imagine airships, not even competing with, because the rail infrastructure in Africa is quite poor, but kind of leapfrogging and allowing for countries to have more access to greater markets without the kind of rail or even road infrastructure that would otherwise be perhaps more costly.
Eli: Yeah, I think that’s right. So, Africa is interesting, because they actually have very few deep-water ports. I think it’s eight or nine for the whole continent, in terms of truly able to receive, the biggest cargo ships. Then also, as you said, the rail infrastructure is really bad, because they actually haven’t standardized the rail gauge, so they actually have to change real gauges several times. Every time you do, you have to unload the cargo from one train, and they don’t actually unload it from one train to and put it on another, they unload it, and take it off the ground – and set it on the ground, and it sits there for a couple weeks, and you’ll have to pay a bribe to get it back on to the other train. So, it’s an opportunity for some sort of tax to be extracted.
A decade ago, I was in Rwanda for a while. I heard, basically, to get a container to the port at Dar Salam from anywhere in the world is 30 bucks, to get the 700 miles from there to Kigali is – a couple orders of magnitude higher, right? A thousand bucks or something like that, to get a container there, to be able to just go straight, skip over that, and just go straight inland. There are many landlocked countries and just be able to serve them directly, I think it’d be a huge advantage.
Mark: I mean, it’s sort of, there’s been the drones to deliver medicine and goods for remote villages and regions, this is kind of that but on a much, much grander scale.
Eli: Yeah, 500 tons of time, right?
Mark: Yeah, writings, and I think I saw that much about AI. So, in the last year or so we’ve had several substantive advances, or maybe GPT-3 was two years ago, I don’t remember. We’re at GPT-3 as well, as we’ve seen, I think was DeepMind that did AlphaFold. What can we expect the impact of AI to be over the next decade?
How does that kind of work with all of the other, I guess, I don’t know technological innovation emergence, because they’re really optimistic. AI people are like, this is bigger than electricity and everything else. I don’t have that kind of a baseline for it. I’m like, that might be true, but I don’t really know.
Eli: The way I had approached this before DeepMind, before AlphaFold, was there has been a lot of excitement about AI in the past. Going back decades, there has been AI, but then there has always been this disillusionment, this AI winter. So, I was like, okay, there’s going to be an AI winter. This is still hype and stuff. Part of what I would have said a year ago to justify this, I would have said, look, if you think about GPT-3, it’s basically a toy. It’s not superhuman performance. If you think about something like AlphaGo or AlphaZero, the chess engine that DeepMind did, it is superhuman performance, but it’s basically a toy. Everything so far that before protein folding was either sub-human performance still, or working on a problem that didn’t matter. So, chess, okay, chess. It’s cool, but it’s not going to increase TFP.
AlphaFold is the first example of we are actually able to do something that we could not otherwise do with AlphaFold and it matters. So, to me, that’s a shift that is worth noting. I think the question is just how quickly can we come up with with more of these? I think that’s how it will play out. It will be relatively discrete domains, where we kind of have AlphaFold breakthroughs. I’m not expecting general intelligence that is able to replace skilled labor or anything, or skilled general-purpose intelligence, the ability to creatively problem solve. I don’t think it’s going to take all the jobs. It will take jobs in particular narrow domains.
Mark: Why should we bring back mammoths?
Eli: So, it would actually be, this is another eco modernism thing, right? It would be great for the climate.
Mark: Your first answer should be they’re awesome.
Eli: They are awesome. They are awesome. I think another thing is a justice argument, right? We killed them. They were on the planet for thousands of years and, you know, maybe hundreds of thousands of years and or millions of years. When humans showed up and we hunted them to extinction pretty fast. Basically, sort of if you think about these ecosystems Siberia, the way it used to be was it was grassland, and it would snow of course, but the fauna would trample the snow, basically, make it not as insulating on the ground, and it would uproot a lot of the trees that started to grow. Because actually, elephants and mammoths are a kind of elephant, but elephants really enjoy pulling up trees. Apparently, they do it for fun. The mammoths basically kept the ecosystem grassland, they kept it grassland. That enabled the cold air that was circulating above the ground, to penetrate, the coldness of it, to penetrate the permafrost and keep it frozen.
Now, with trees, which are darker and absorb more sunlight, with thicker layers of snow, which are insulating in the winters, basically the permafrost is not so perma anymore. It is melting, because the temperature is getting warmer. Basically by hunting the madness to extinction, we’ve created this situation where the the permafrost is melting and there is potential for it to release a lot of greenhouse gases, especially methane, which is more potent than CO2. As it melts, it’s going to make climate change, potentially accelerate. So, the argument is bring back large herding animals or grazing animals, herds of ideally mammoths, but maybe other large grazing animals as well that trample things and eat grasses and are going to be foraging for grasses and disrupting the snow cover.
Mark: We should bring back bison?
Eli: I think, absolutely. Bring back bison and mammoths and you would have these sort of grass –
Mark: with airships?
Eli: Absolutely. Sky dark with airships, but also bring back mammoths. I think it’s a good future.
Mark: Okay, so let’s finish up on charter cities. CCI mostly focused on charter cities in emerging markets, figuring out how to do catch up growth. But there’s also these ideas about using charter cities to kind of accelerate technological innovation, allow for a better use of regulatory environment than might be in the US or in Europe, or wherever. If you are building a charter city to accelerate technological innovation, what would kind of be the key industries that would you target? And how would you create regulatory regimes that would allow for them to be more innovative than the US?
Eli: I think in terms of infrastructure, particularly with the NEPA discussion that we’ve had, what have you made it super easy to build things, right? Housing also, get housing super cheap, make it dense, super cheap. If you cut out all the things that we do to make infrastructure expensive, all the way, we have American provisions, we have union labor rules, we have all these redundant employees doing a bunch of things. We have all this environmental review, which means you have to go through years of review before you can even put a shovel in the ground and think about the ROI that that does for the capital markets, the way it affects – reduces the ROI, because it takes longer.
I think it’d be great to just have a city that is optimized for building, where the rules are designed upfront with “How can we make it easy to build really high-quality infrastructure that is going to improve the lives of everybody and all the residents?”
Mark: What about biotech? That’s kind of where the innovation – I’m probably most excited about biotech and AI as well, but in terms of biotech, that’s one of the areas that typically has some of the strongest arguments about and slowing down innovation. It’s really challenging because you need the kind of agglomeration combined with the regulatory environment. In reality, you could go to a lot of emerging markets, and many emerging markets are not particularly strong biotech regulatory states. But going there, .
How do you actually create this agglomeration with a possibility of kind of pushing the frontier? The way I would think about starting, is doing direct to try, which is kind of unobjectionable and will allow for some innovation on the market and start to get moderation and kind of branch from there. You could also do, for example, medical tourism, and then once you have the base medical tourism set up, even if it’s not innovative at all, then you have at least some degree of infrastructure established that you can start allowing for experimentation with new therapies, new drugs, new medical devices.
The way I kind of think about it is you have to target Series B plus companies, because they’re the only companies that are risk loving enough to do it and have enough capital to open a second lab or something in a new location. Where big pharma is just way too risk averse, that they’re not going to bother with it.
Eli: Yeah, so I think medical tourism is a really good application, particularly if we’re talking about stuff like longevity. I think the market for sort of off label or not fully approved longevity treatments, as those get developed in the next couple years. I think that being the market where somebody can go and spend a few days in your charter city and receive some of these treatments, that they couldn’t get in the US. I think that’s a great application. And then they stay for a few days, and maybe they love it, and then maybe they move, and they come. So, it’s actually a potentially good recruitment.
You want to recruit wealthy residents to be sort of tentpoles of the of the economy there. That’s potentially very useful. In terms of competing, if you’re trying to compete globally on drug approvals, it’s really hard, because your market just isn’t big enough.
Mark: It has to be very high value. It’s like the airship. It’s high value, one shot, it can’t be anything that’s chronic, or has to be people fly in, take it, fly back. Maybe once a year, once every five years, but it can’t be any more frequent than that.
Eli: Right. Exactly. So, I think it’s really tough. But I think, just starting with medical tourism and especially longevity tourism could be really good.
Mark: All right. Well, great. Thanks for coming on the podcast.
Eli: Thanks for having me, Mark. It has been super fun.
Mark: Thank you for listening to The Charter Cities Podcast. For more information about this episode and our guest, to subscribe to the show, or to connect with the Charter Cities Institute, please visit chartercitiesinstitute.org. Follow us on social media, @cci.city on Twitter, and Charter Cities Institute on Facebook. I’m your host, Mark Lutter, and thank you for listening to The Charter Cities Podcast.