Invest Like the Best with Patrick O'Shaughnessy - Eli Dourado - Frontiers for Productivity - [Invest Like the Best, EP. 225]

Episode Date: May 11, 2021

My guest today is Eli Dourado, economist and senior research fellow at the Center for Growth and Opportunity at Utah State University. In our discussion, we touch on the ongoing stagnation in labor productivity, the system constraints, and some of the innovative technologies that could reverse this trend. While Eli identifies as an economist, his wealth of knowledge on biotech innovation, alternative energy, and the space opportunity are sure to leave you craving more. I hope you enjoy my conversation with Eli.    For the full show notes, transcript, and links to mentioned content, check out the episode page here. ------ Invest Like the Best is a property of Colossus, Inc. For more episodes of Invest Like the Best, visit joincolossus.com/episodes.    Stay up to date on all our podcasts by signing up to Colossus Weekly, our quick dive every Sunday highlighting the top business and investing concepts from our podcasts and the best of what we read that week. Sign up here.   Follow us on Twitter: @patrick_oshag | @JoinColossus   Show Notes [00:03:28] - [First question] - Why he’s interested the great stagnation [00:05:41] - Is a fair analog for total factor productivity technology? [00:07:29] - Leading theories for the stagnation in TFP [00:09:05] - Analysis of why growth is good, and stagnation is bad [00:10:19] - Rate limiters that are a key part of calculating stagnation [00:11:43] - Signs that we may be returning to a higher degree of TFP returning [00:12:24] - Exciting developments like MRNA in biotech that may lead to an explosion of innovation [00:16:16] - Functions of a protein and their role in advances in biotech [00:17:55] - What CRISPR is and what it unlocks for the future of humanity [00:19:31] - The pace of progression when rolling our clinical trials of cellular engineering [00:21:36] - How biotech may play a role in TFP as a proxy for progress and growth [00:22:51] - Interesting observations about potential innovation in the energy sector [00:24:10] - What currently requires energy that could be optimized if they had a lower energy cost [00:25:34] - Sources that could provide cheaper and more efficient energy [00:27:13] - Sage Geosystems and the future of the geothermal space [00:27:43] - The importance of batteries in the modern era [00:29:02] - Why energy should be a more pivotal focus in our future [00:29:41] - What’s interesting in the world of transportation writ large [00:31:02] - Boom’s story, supersonic air travel, and why concord shut down [00:34:55] - Mach 5 and March 12 supersonic air travel [00:36:19] - Second-order effects of reducing the time cost of air travel [00:38:19] - Liftoff, SpaceX, and the future of the space sector [00:43:07] - Other key players in space people could study; Blue Origin and Relativity Space [00:45:15] - What will we do in space once we can travel there cheaply [00:46:39] - What he’s most curious about in IT that could drive productive societal growth [00:50:07] - Ethereum and how a decentralized blockchain could change the world  [00:51:19] - The kindest thing anyone has ever done for him

This episode of Invest Like the Best is sponsored by Canalyst. Canalyst is the leading destination for public company data and analysis. I'd heard of Canalyst over the past few years and became more interested after meeting the founder and CEO last year to pick his brain about SaaS businesses. Founded by a former byside analyst who encountered friction in sourcing, building, and updating models, Canalist is now used by over 300 institutions, including the largest money managers in North America and by a number of guests on the show. With detailed company-specific models on virtually every investable public equity, Canalist clients are able to react more quickly. If you've been scrambling to keep up with the deluge of IPOs and SPACs these days,

Canalist has models on Coinbase, Roblox, Qualtricks, and everything in between. Their pre-IPO models are built as soon as the S-1 hits and include all segments, KPIs, and non-gap figures. If you're a professional equity investor and haven't talked to Cannibalist recently, you should give them a shout. Learn more and try Canales for yourself. at canalyst.com slash Patrick. That's C-A-N-A-L-Y-S-T-T-com slash Patrick. If you're curious to hear more about Canales, stay tuned at the end of the episode where I talk to Canalists's new chief product

officer, Jeremy Payne. This episode is brought to you by MIT Investment Management Company, also known as Matimco, the Endowment Office of MIT. Matimco seeks to find people who are focused on achieving exceptional long-term investment returns, partner with these firms early, and stick around for the very long term. Matimco doesn't care how small, new, or uninstitutional your firm is. If you have the potential to generate amazing results that supports MIT's pursuit of world-class education, cutting-edge research, and groundbreaking innovation. Despite their willingness to invest early, they do not ask for general partner economics,

and they commit their initial capital for 10 years. Matimco is also searching for an exceptional new teammate to join their internal investment team. Visit matimco.org, M-I-T-I-M-C-O-R to learn more. Click join to learn more about the global investor role at Mitimco's team, or click emerging managers to learn more about their emerging manager activities. Hello and welcome, everyone. I'm Patrick O'Shaughnessy, and this is Invest Like the Best. This show is an open-ended exploration of markets, ideas, stories, and strategies that will

help you better invest both your time and your money. Invest Like the Best is part of the Colossus family of podcasts, and you can access all our podcasts, including edited transcripts, show notes, and other resources to keep learning at join colossus.com. Patrick O'Shaughnessy is the CEO of O'Shaughnessy Asset Management. All opinions expressed by Patrick and podcast guests are solely their own opinions and do not reflect the opinion of O'Shaunacy Asset Management. This podcast is for informational purposes only and should not be relied upon as a basis

for investment decisions. Clients of O'Shaunice asset management may maintain positions in the securities discussed in this podcast. This is a conversation about the most exciting technology trends. in the world. My guest today is Eli Dorado, an economist and senior research fellow at the Center for Growth and Opportunity at Utah State University. We discussed the strange economic stagnation of the past few decades and the innovative technologies that could reverse this trend. Eli's wealth of knowledge on biotech innovation, alternative energy, and the space opportunity are sure to

leave you craving more. I hope you enjoy this great conversation with Eli Dorado. So, Eli, I think the right framing for our discussion, given the wide variety of your background and your work, is what we'll call the great stagnation. I'd love you to describe this problem, maybe introduce concepts like GDP or total factor productivity, ways of measuring what we'll call a stagnation. And then most of our conversation will be, how do we get out of it, assuming that we are indeed in it? But I think that's a great opening frame. Why are you interested in this problem? And what does it mean? A good way to start is to just think about economic output. and what are the inputs that go into it? So economists think in general about output as a function of

capital and labor and everything else that we can't measure. Everything else that we can't measure is actually the interesting part. If we apply more labor, if it means that we're working harder, we're all working longer hours and we get more output. That's like not a big yay. That's like a little yay. Maybe we got more output, but like we're all working harder. And that is a cost. But if you can get more output for the same amount of labor and the same amount of capital going in, that's really good. That sort of residual term is called total factor productivity, and it sort of represents the quality of ideas and institutions and the recipes that society has for combining labor and capital into output. Total factor productivity to me is like the key metric of like how

civilization is doing. And the history here is that basically from 1920 to 1970, maybe 1973, three, depending on who you listen to. TFP grew at about 2% per year. Then something changed. And it started growing at about less than 1% a year for the next few decades. And then around 1995, there was like a brief spasm where it went up about 2% a year again for a decade. And then since 2005, it kind of fell off a cliff. And now it's 0.3% per year. So you'd round it closer to zero than you would even to 1%. So historically, we've been at 2%. Now we're at 0%. That's to me is like the most fundamental metric of growth and we're basically not doing it. I think this has all kinds of implications for society. I mean, basically it creates in a way a zero-sum society

for not actually growing, which has all kinds of implications on the culture, on politics, on the way we think about the world if it's zero-sum rather than positive some. Is a fair analog for total factor productivity technology? Like, is that the right way of thinking about or is that too simple? I think that's most of it. That's certainly how, most economists have interpreted it in the past, but I would add in the element of it's also quality of our institutions. In the U.S., you can take a given quantity of labor and capital and produce a certain amount of output. In Venezuela, you can take the same labor and capital, and you cannot produce the same amount of output, but you produce less.

And why is that? It's because the state expropriates. It's because you can't get your rights enforced. You can't get the support you need from society to just do your thing in the most efficient way, that has to be an element as well if we're thinking rigorously about what TFP is. Is another way of thinking about this, Peter Thiel's famous, we wanted flying cars, but we got 140 characters. Like, is that another vector that's interesting here? Yeah, I think that that is very much related to the TFP slowdown.

If you want to break down the way that productivity has grown over the last, say, 50 years, it's grown a lot in IT, 140 characters, but also TV manufacturing and, you know, smartphones and the internet, those are all been bright spots in consumer goods, consumer electronics in general, have all been a bright spot in the economy over the last 50 years. If you look at healthcare, housing, transportation, that kind of stuff has not done as well. Healthcare costs are now something like close to 18% of GDP. So pretty massive housing is around that level 16 to 18% of GDP and higher, of course, in certain cities. Those are huge drags on productivity.

And in transportation, you know, we had in the 1960s, we developed supersonic airliners. And today we're flying around at the same speeds as we did in the 1950s. What are the leading theories for this stagnation in TFP? I think that there's basically two strains of thought. And so one of them is, I would call this like the Robert Gordon theory in the late 18, We developed indoor plumbing and we developed combustion engines. We developed pharmaceuticals, a couple other breakthroughs that he identifies. We developed those sort of in the late 1800s. It took time. It took decades for them to really start to affect growth. But that led to like this burst from 1920 to

1970 of fast growth in those areas. Then that sort of petered out. We did everything we could with internal combustion engines. That growth ended. And therefore, we're kind of of on a slow trajectory right now. It's possible another great invention will come along at some point, but when it does, it will take many decades for it to get to the point where it's affecting the macro picture a while, and then we'll write it for another few decades, and then it will peter out as well. That's just the way the nature of technology is. So I think the other school of thought is what my PhD advisor, Tyler Cowen, put forward in his book, The Complacent Class. which is basically we've kind of given up. We're much more concerned about comfort, about

making sure our neighborhoods don't change too fast, historical preservation of things. We're not really trying. We're engaged maybe a little bit too much in status fights in our culture rather than in actual like material improvement. Why is growth good and why is stagnation bad? That sounds obvious, but why is it the case that those things are good and bad? I think you can look at it a couple different ways. One is the levels view and one is the change view, the first derivative of the levels view. So the levels view is output matters because that's what determines our standard of living. There's a lot that isn't captured in GDP, time with family, and the ability to pursue artistic pursuits,

whatever you might think of that's not included in GDP, but it all correlates with GDP. So output matters because it reflects living standards. It reflects our ability to choose our own life. make a good life for ourselves. We're all wealthier. That's in general better. That's like one way of looking at it. Another way of looking at it is to say, well, society is just more stable when it's growing. You're not having to fight over a fixed pie as much when the pie is growing. So it's good. The ability since for the last 200 years or more to have a growing pie in some parts of the world It has been super important for peace and not being expropriated or killed.

Violence is coming down. The level of growth matters for that reason as well. One of the great examples that I think tells this story is the speed with which the Empire State building was built back in the day and how laughable it is that we might be able to do something like that in the future despite more resources, more technology, more know-how, etc. So with that gap, what explains some of that? Like, what are the rate limiters, maybe the exogenous rate limiters, whether that's regulation or other things that are also a key part of this calculus before we get into some of the really fun topics about how this might change over the next decade?

I think it's a lot of really boring regulatory stuff. That's what explains it to me. There was this law called NEPA that was passed in 1969, basically said that before the government, a federal agency takes any action, which would include approval of permitting or stuff that affects the private sector. It basically has to either document the environmental impacts or it has to prove that there are no environmental impacts. This has basically turned out to be about a five-year process to document the environmental impacts and about a two-year process to prove that there aren't environmental impacts. Very costly. You basically have a federal government, unless there is an exclusion for that kind of activity, they basically can't make the decision to like approve your new tunnel.

If you want to do a hyperloop network or something like that that involves their approval, you're going to have to wait years to get that approval. And what are the signs today that some of this might be changing or cracking or we might be returning to a higher degree of total factor productivity? I don't think it's a foregone conclusion. I think there's a lot of promising technology, but I do personally side in the sort of the complacency camp. It's not actually the lack of technological opportunity that's holding us back. It's we're not trying hard enough. We care too much about other things, other values besides just growing a material sense as fast as possible. We overweight those other concerns and underweight the material concerns. I think we might get there. But I think there's a lot of

pent-up innovation that is ready to burst forth if we decide to do that. Obviously a great place to begin the conversation now, kind of the uplifting or exciting part of this conversation in some of these areas that might see rapid change. And I think biotech and health is the right place to start because when you talk about pent-up innovation, everyone's amazed at the innovation that happened around the vaccines and some of the other things related to COVID, necessity, breeds, invention type scenario, sort of squeeze the innovation out of us. What is exciting about the seeds of MRNA technologies, protein folding, like what in this area might lead to a, the opposite of a stagnation, an explosion of innovation in the coming

decade? It's interesting, the crisis aspect of this. crises are by definition a time of low complacency where you're like actually willing to pay the cost to get things done. That in itself lends credence to the complacency theory of stagnation. So that's interesting. But MRI has so much potential. What MRI does is it basically allows us to produce arbitrary proteins inside the human body, inside the human cell. It's a lot like computer programming.

You can program the protein that you want to be built and insert it into the cell and the ribosome picks it up. And it just says, oh, this is the next thing I'm supposed to build. I'm going to build it. Combining that with the immune system is what enables us to make mRNA vaccines. You basically have a human cell creating a viral spike protein. And then your immune system uses that as sort of target practice so that when it sees the real spike protein in the actual virus, it knows how to attack it. This can be used for tons of other diseases. What's most interesting is the potential for treating cancer with it. So you could basically take a healthy cell. If you have

cancer, you could take one of your healthy cells, sequence it. You can take one of your cancer cells, sequence it. Look at the difference of those genomes. Find protein targets that are promising for your immune system to be able to pick up on and fight against and then give it all that target practice. you give your immune system the ability to find the cancer cells and take them out. That's hard work, but it's like definitely an implication of MRNA technology. The protein folding is also a really exciting thing that happened last year. This lab at part of Google, part of alphabet called DeepMind basically has over the last couple years cracked, effectively cracked the protein folding problem, which is that these proteins

are defined linearly as just linear sequences of amino acids. So it's just one of 20 amino acids followed by another, followed by another, followed by another. And yet what proteins end up doing is they end up folding after they're sort of extruded by the ribosome fold in upon themselves based on the laws of physics. And it's very, very hard to predict how they're going to fold in upon themselves. But then that three-dimensional structure is what gives them their function. function inside the body. So being able to predict that is really important. And it raises the

prospect ultimately in the near term, I would say it will enable us to find drugs that target certain proteins. It gives us the ability to reason about what kinds of molecules might affect what protein targets and let us activate or dampen the particular targets. In the long run, you know, what it enables is protein design. We're going to be able to design proteins that Nature didn't give us. So evolution selects for getting us to reproductive age and enabling us to successfully reproduce. It doesn't care about us anymore. So it gives us those proteins, but it doesn't give us the proteins to like sort of clean up our bodies and live a lot longer and stop from wearing out. Being able to do that protein design ultimately is going to produce

significant benefits in the long run. Why are proteins central here? Like what is the function of a protein. Why does that seem to be the core unit of interest here? So proteins are actually the bulk of what living is about. A lot of times I think we think of proteins in terms of like the macronutrient. Right. Steak. Carbohydrates, fats and protein. It's like pretty static picture of proteins. It gives structure to things and so on. But proteins are like the dynamic element of what's going on in the cell. You know, anything that's doing anything inside your body, like chances are it's a protein really is like the most fundamental aspect. of life. You can think of even your genes. Your genes are basically encoding for proteins. They're

basically instructions for how to make proteins. So actually, the molecules that are doing the stuff that makes you alive are proteins. They're not static. They move around. They have functions. They're like mini machines. There's proteins that pump things. There's proteins that operate like conveyor belts. There's proteins that have motor functions and stuff. It really is nanoscale machinery that is what keeps you alive. And that actually raises, to me, like the really far version of this is we're able to design proteins that help us design other nanotools, that then we can, through a few iterations of nanotools, designing other nanotools, designing other nano tools, you get to what Eric Drexler called, atomically precise manufacturing. You can imagine

we designed these tools, nanoscale tools that enable us to make physical stuff with, a minimum of energy input and extreme precision. That's the singularity stuff. We can't see beyond that. So it's not to leave out any of the buzz categories here. I remember reading about CRISPR in 2017 or 18, that great book on it, and thinking, wow, this is exciting. I don't really know what it means, but sure seems interesting. Where does CRISPR fit in the sort of like protein folding MRNA as a core technology or a method or an output? Like describe CRISPR a little bit and why it's important. CRISPR is a method for editing DNA. So DNA codes for proteins. So if you ever want to have your body produce a protein that it doesn't attack, that it doesn't go after as sort of an immune, it has an immune response against, do you probably want it stitched into your DNA at some point? Or if your body is producing a protein that is bad, that is harmful, causing you some sort of congeneral disease, you probably want that taken out or replaced with a different allele of the same protein. So CRISPR, they, basically enables very effective computer-like editing of DNA of living cells.

And it's invented in 2012. It has absolutely revolutionized the lab. So biology labs can use it to test various hypotheses on tons of organisms. It hasn't really yet reached the level of going through all the clinical trials and being able to be used in human diseases. But it probably will be soon. I think it's sort of going to happen first for things like congenital diseases, like the really bad stuff that breaks your heart when kid is born with sort of a genetic defect that causes

suffering or something like that. So I think it'll be used there first, but gradually expanding its scope to more and more stuff. What are the impediments here? So one of the things that pops to mind kind of like your regulatory discussion earlier is this is all stuff happening to humans. and it's very easy to iterate on code and a machine, probably pretty hard to iterate on code in a human, given there are negative potential consequences and we want to protect people. What's that structure like? Will this necessarily have to go slow because we're talking about programming human bodies versus just machines?

Yeah, I think CRISPR will probably be pretty slow progression in terms of like the rollouts of clinical treatments for humans because you have to go through clinical trials. You do have to try it on somebody first, and that somebody first is someone who has the really bad diseases. That's why they're happening first is because the bar to giving an experimental treatment for someone who has a really horrible condition is a lot lower than I want superpowers. And we're going to edit my genome to give me them. Yeah, clinical trials are very expensive. It takes a long time. There's other therapies as well.

They're similar. So I was following this guy at the University of Washington who, in monkeys, cured red, green, colorblindness. And so I was really interested in this because I'm somewhat red-green colorblind. And something like 8% of males in America have that condition. He can do it in monkeys. Basically, it's not CRISPR. It's a different payload delivery mechanism. But the eye is a confined space anyway. So whatever you inject into the eye isn't going to get out into the rest of the body. He figured out like, we can deliver this payload, change the gene. And the monkey can now see all the colors. And to me, like that's something that's probably be incredible

incredibly safe in humans. We actually use the same platform to do other sorts of treatments in the eye, but it just hasn't gone to clinical trial yet. It's not that it has taken too long. It's like they haven't done it. It's just sort of sitting there on the shelf. And I think it's because of the expense, the urgency, et cetera. If you follow that through and think about, well, okay, Christopher gives us even more power. How long is it going to take for us to actually reap the full benefit of this? It's going to be quite a long time. Trying to tie each of our categories here back to our opening frame of total factor productivity.

I'm an optimist by nature. If all of this goes right, why is this an interesting category for something like total factor productivity, which we're going to use as like a proxy for good progress and growth? I think it's just such a big chunk of the economic equation. Overall health is something like I said, 18% of GDP. And so if you improve everywhere else in productivity, but not that, then that only, is only going to grow as a percentage of GDP. What you actually want is to shrink that down. You want to get so productive that it's like a 10th. So if you got down to like 1.8% of GDP instead of 18% of

GDP, then that's like a massive boost to real economic living standards. That's the right way to think about it is like how do we take this big chunk of GDP and make it so productive that it barely registers anymore? It's interesting to me, just anecdotally, that a lot of the investors that I respect the most have developed a keen interest in this area, mostly because they feel uneducated, like they don't really know what's going on, but also feel like there's this enormous potential. How exciting all these technologies and COVID awful for most reasons, but maybe a kick in the pants to get things moving on the biotech side. Let's move from something that we want to preserve, our health, to something that we want to create, which is energy. The history of a lot of this

productivity increase, I love like Vaclav Smil's work on this topic where through discoveries like oil in the 1800s, drastically improving what we can do just through raw energy output, energy key input into a lot of stuff that we care about. What is interesting to you in terms of potential innovation, given where we've come from in the space of energy? Over 200 years ago, going forward, energy use per capita in the U.S. grew something like 2% per year. Every year, we would use more and more energy. And then something changed around 1977, 1978, and we started using less energy per capita. So, it was a lot of, was like a very consistent trend and then we started using less focused on energy efficiency

and energy efficiency is good. You can do more with the same amount of energy or so on. But I think that there's like a definite mindset shift that you want to, to my mind, like we've done the trying to get more efficient for several decades now. Instead of trying to do more with less, let's try to shift into more with more mode, try to do things that actually use more energy and not worry about it because it's gotten so cheap. That's like a mindset shift that it will have big ramifications throughout society, but to get there, you need much cheaper energy than we have today. What are the things that are more that require the most energy?

And then we'll talk about sources of that energy. Part of the answer is, I don't know, but there's a bunch of things. So one is taking CO2 out of the atmosphere that requires energy. So if you want to basically do negative emissions technology, there's like one that's really interesting that doesn't require a lot of energy, but a lot of the others do. I think materials is really interesting. Iceland has abundant energy, clean energy, geothermal, hydroelectric. They are the per capita leader in aluminum output.

That low energy costs gives them the ability to do that. What other materials? Instead of using building houses with two by fours, what if we could build with something stronger, not flammable, easier to assemble? So I think about materials and industrial processes as being interesting. Desalination. I worry about Yemen.

They're running out of freshwater. water. They're right next to an ocean. Wouldn't it be great if we had enough energy that we could just cheaply take the ocean water, take the salt out, give them fresh water so that they don't dive thirst. I'm really interested in vertical farming, new forms of agriculture. That's an energy equation, right? That's how much energy does it take to power LED lights to power the pumps that move the water around, heat the space so that it's good temperature. Basically, if energy is free, we can grow anything anywhere, which is really cool from a global agriculture perspective. What are the key sources here that are interesting in terms of what might provide both more

and cheaper per unit energy going forward? The last 10 years have been huge for solar and wind, massive declines. The one I really like going forward in the next 10 years is geothermal, particularly deep geothermal, so not tiny geothermal that we have today that is only used, where there's volcanoes and hot springs and geysers and stuff like that. Those places around the world, it's very easy to get geothermal. That's where we've done it in the past. But basically, if you dig deeply enough anywhere in the world,

there is enough heat to produce energy. And in fact, massively more than enough to provide all of the energy that humans will ever use. The stat that I've seen is between the surface and a depth of 10 kilometers, there is 50,000 times more heat energy than all are oil and gas reserves. You can't run out of it. It's replenished every year. So the Earth's soil has uranium in it. It decays. It produces heat replenished at a rate of twice humanity's current primary energy use. You can't run out of it. The advances that we've made in fracking, in shale gas, in tight oil, those advances subsurface engineering exploration. Those make it now viable to do deep geothermal. If we can do deep geothermal, then we can get energy that's actually better in terms of a profile. It's better than what you get in wind and solar because wind and solar is highly intermittent. And to some extent, unpredictable. If you can get baseload energy, which is what geothermal provides, it's 24-7, same output. That's a very nice compliment to renewables.

How near term of a possibility is this? Like, is there any actual producing energy deep geothermal project that's live today? Like, what's the timeline do you think? There's one being drilled right now in Texas, a company called Sage Geosystems that's doing their first demo well, and they're going to basically do it. And once that's producing, they say they're off to the races, they're going to do more. There's a bunch of concepts for how you would do this. They're all interesting. So it will be just within the geothermal space to see what concept wins. Do you think that, I mean, obviously if that were to happen, it seems like energy storage via batteries, maybe other means, becomes less important, but energy storage is still interesting topic, especially for some of the

less reliable, renewable energy sources. Any thoughts on batteries, battery technology, whether or not it's important. Do we overindex on this? Batteries are super important. We want to electrify all the things, so including cars and other big energy types of devices. So the question is kind of like, are batteries worth more in the edge of the network or in the core of the network? Are they worth more in cars or on the grid? And I think they're worth more in cars. We still have production bottlenecks. So I see all the time these people saying, like, we've gotten battery production costs down to $100 or something like that. Well, you have a supply bottleneck still. You can't do it everywhere. And even assuming very, very rapid growth in the supply side, I worry that it's not going to

keep up with the demand because I think everyone's going to be switching to electric vehicles and we're going to be electrifying more and more. So you don't want it, I think, primarily as grid storage. You don't want to use your sort of like scarce battery resources, grid storage. You want to use it on the edge of the network, on cars. And to some extent, you can connect the cars with the grid. They can be grid resources. There's that move, which is, I think, a good one. But you don't want it to be just pure grid storage.

So we've got this fascinating story of energy being an enabler, of all this other stuff, material science, lots of things that we want to do with energy. What else do you think about here in terms of why this is good and should be a focus? It just touches on basically every segment of the economy, everything we do. uses some energy. I think it's not properly accounted for in a lot of economic data because it's just so fundamental and affects everything, including the transportation side, which also affects everything. So it's like these chains of inputs that all work together to produce output and massively increasing the productivity of one of them is going to touch everything else.

What is most interesting about transportation? And I love how all of these are fundamentally the zone of atoms, maybe controlled by bits or made more efficiency by bit, but sort of all physical stuff. What is interesting in the category of transportation we finally get to talk about flying cars? Yeah, so flying cars are interesting. So I spent two years working on supersonic's. I was at Boom for two and a half years leading their policy team. I really like going faster. Again, there's different concepts for Supersonic now. So there's sort of the boom, Mach 2 model. There's the Hermius Mach 5 model. There's Venus Aerospace wants to do Mach 12. there's a low boom model for trying to do stuff over land. Again, it's like very open question to

see what model wins. Elon trying to do the point-to-point rocket anywhere on the planet in under an hour. Who wins? I don't know. But being able to go faster is super important because trade between any two parts of the globe is a function of the transportation costs, including the time costs of executives that are going back and forth. If you want to have a deal between a factory in the U.S. and a customer in Germany, how fast you can get between those two points for the executives to see each other and have a high bandwidth conversation and get to know each other and trust each other, that really matters for that deal. So reducing the cost on those people to get back and forth is important for being able to expand those deals and make economic activity happen.

It's interesting that Zoom may be one of the most compelling transportation-like technologies in that sense, that you can have all this picked up time by not traveling to Asia or something on a 747. What is the story with boom? So fascinated by that company. You mentioned Mach 12. I'm curious what the hell that is. Obviously, there's going to be reasons why people and stuff wants to move faster from point to point in the earth. What is the boom story? Why has there been so much stagnation? Why did the Concord shut down? What's interesting about supersonic air travel? Maybe it's starting with the Concord. It's a good way to do it. So there were 14 Concords ever saw service. It was designed by the French and British governments. They got

together and said, we're going to do this treaty. This part of the treaty, we're going to have our engineers work together to make this thing, the Concord. The problem with Concord is that it never got to scale. So with 14 airplanes, the maintenance costs on it were really high. So that means that the ticket prices had to be very high, which meant that they couldn't fill all the seats. So Concord exited service in 2003. And one of the calculations I did while I was in the industry was premium transatlantic travel has actually. more than doubled in the time since Concord left service. With double the demand for seats, you could actually get to a point where I think you could fill all the seats at a lower price.

The way the airline economics works is if you can't fill some of the seats, then you've got to charge higher price for all the other ones to add up to at least the maintenance costs. I think that with higher demand for travel today, or pre-pendemic anyway, Concord would have been a lot more economically viable today than it would have been when it actually flew, which means you could have had more of them and you could have had a bigger fleet and gotten the maintenance costs down. To me, like, that's a big part of the story is just that we live in a time where there is more demand for high-speed travel, or at least high-dollar travel between these two places. And you think about how Concord was designed.

It was designed with basically on paper, with slide rules. they didn't use computers, used wind tunnel tests, basically to test the aerodynamics. Test takes like six months to do. They were very handicapped relative to what we can do today. Secondly, materials. So Concord used aluminum. The reason Concord is Mach 2 is because if you go any faster, the aluminum starts to melt. Melt.

So even at Machu, it gets kind of soggy. So Concord would actually grow over a foot in length on every flight, grow and shrink. That itself was a design challenge. That doesn't sound good. You'd grow a foot. The outside grows, the inside doesn't, and it has to stay airtight. Think about that design challenge for a minute. Today, we can use carbon fiber.

Carbon fiber is like lighter, stronger, but more thermally resistant. It doesn't expand when it's exposed to those temperatures, much more stable. And it can be shaped better. So with aluminum, you can't have as many sweeping curves and so on. For aerodynamic purposes, you want to be able to do that. So that's like computers, materials. I think the third is just engines. So engines kind of have their own mini Moore's Law.

I think it's like much slower than Moore's Law, right? It's like something like half a percent more efficient per year. And so basically if you use a modern turbofan engine, you can get significantly more efficiency than Concord was able to get using the state-of-the-art engines that they had at the time. Those were great engines for 1969, but not great for modern day from an efficiency perspective. So if you put all those things together, computational fluid dynamics, carbon fiber, engines, and then the market demand point. Combine all of those together. Like,

it should be possible or viable supersonic aircraft to exist. There's an existence proof in Concord. So it's like no new fundamental science. It's just can you put together the pieces? And it turns out it's a very hard business. We'll see who succeeds. Save it more about Mach 5 and Mach 12. That sounds kind of insane. I don't know as much about the Mach 12 one. I know that they're saying we're going to go Mach 12. In Mach 5, people have been talking about it for a long time. Mach 5 is kind of like the limit where it stops making sense. It doesn't make a big difference to go Mach 6 if you're flying as a traditional airplane versus Mach 5 because you still have to spend some time as an airplane in a subzonic regime where you're circling the airport or flying at

low altitude and you can't go that fast. So you don't actually get a big speed up from going Mach 6 from Mach 5. From Mach 5 to Mach 4, like that's still a pretty big jump, relatively speaking. So that's like kind of what people have looked at before in terms of the next generation would be Mach 4 or Mach 5. And basically you can do it with new engine technologies that are export controlled. So you can't tell foreigners about them. So I probably shouldn't talk about them on this podcast. And also materials that are even more heat resistant than carbon fiber. And you can imagine an airplane that's probably autonomously piloted.

It flies a lot higher than Concord did in terms of altitude because you want the lower resistance of the atmosphere. There's a company called Hermius in Atlanta that's their design point that they're pursuing. And what would that mean in terms of time like to get somewhere? I think it's basically like from the U.S. 90 minutes to Europe. Crazy. Like four hours to Asia, something like that. Incredible. God, I hope we see that. That would be pretty cool. What are the second order effects of this? Like usually when you reduce friction, in this case reducing a time friction, You have weird, nonlinear, sometimes unpredictable things happen. It enables stuff that you can't even imagine.

How do you think about those second order effects of moving stuff faster? My speculation is basically the same as everyone else. I don't know the specifics of what people will do. You'll be able to go places for the weekend. Wouldn't it be cool, be like, oh, let's go to Japan for lunch. Like sushi in Japan. So there's that kind of stuff. Trains is an economist.

So I think about this through one of the most successful economic models. in sort of the history of that discipline, which is the gravity model of trade. So this is the idea that trade between two cities or two places is a lot like gravity. It's a function of how big is city one and how big is city two, the size, the mass of the two economic centers. And what is the distance between them, which is what is the cost of going between them, counting both money and time and risk and whatever else you want to count as a cost of moving between them? That's like the distance term in gravity. So the exponents might be a little different in sort of like the economic equation. It's not like an inverse square law like it is in physics. It is probably like inverse linear,

something like that. So being able to cut the cost between two cities of travel, if you can cut the cost by 50%, you should be able to get twice the amount of trade between those two cities. So twice the amount of economic activity happening between those two cities, twice the amount of just stuck. It doesn't even have to be like literal economic activity. And so people have done studies like when Southwest adds a route between two cities, do you get more collaborations between the professors that are in those two cities? If you could do that throughout the whole world, basically make it anywhere you go, you can get there and half the time. That's going to boost collaboration and cooperation across space a lot more than what we have today. I saved one of my

favorite topics for later in the conversation, which is space. It seems like it's obviously having a massive moment. SpaceX has been, you see these rockets like explode and then another one and get loaded right onto the launch plank and it's almost feels like software iteration or something. It's incredible. I highly recommend this book, Lyftoff that just came out about the early days of SpaceX. Yeah, I read it. It's good. What is interesting here? So space, cool, you know, we're relanding rockets. The reusable costs are coming down. What is going on that's so interesting in space, kind of like the Concord, you know, we did it a long time ago and then sort of went dormant and now we're back. Talk us through your interest in space.

what the variables are that matter, what it might unlock. You're right, the costs have come down. But the next order of costs coming down, it's so much more massive than any of the previous innovations, right? And that's Starship. As you said, like, SpaceX has taken this, like, amazing development approach to Starship. They're iterating very fast. It's a hardware-rich program.

They're willing to expend hardware to learn something. And they're just getting better every time. but ultimately, like, this is the vehicle that is being designed to take us to Mars. And so in order for that, to work for that, has to be dirt cheap. So they're talking about pumping them out at a high rate, the vehicle, upper stage, at like $5 million a pop. So it's made of stainless steel instead of, like, traditional aerospace materials. It's designed to be cheap to mass produce.

And sort of the fuels that it runs on are cheaper than traditional rocket fuel. There are also fuels that we could manufacture on Mars. It's designed not just to be sort of reusable in the booster stage the way the Falcon 9 is, but it's reusable in the first stage and the upper stage. So the whole thing is reusable. You're never like ditching part of the vehicle, discarding it, you know, and making a new one. So put all that together. Elon's talking about potential, like at a high rate, like you could launch.

And the other thing is this big. So 150 tons is what they're talking about. So 150 tons to lower orbit. And so if you can get 150 tons to lower the orbit, they're talking about at a high rate, a launch cost of something like $1.5 million, it might not be realistic. That's what they say.

And so that works out to $10 a kilogram. It's two orders of magnitude lower than Falcon 9, which was itself lower than everything that came before it. If they actually get to that cost, it's like sending stuff to orbit is like the same as sending it to Europe. It's that cheap. And then they're also developing star. ship to do is to refuel on orbit. Traditionally, the way you get mass beyond low Earth orbit

to Mars or the moon or wherever else you want to go is you just put less mass on a big rocket. You add more fuel to the rocket to take it the rest of the way. You trade payload for fuel. With Starship, you don't have to do that because you send it up to orbit with the full 150 tons, then you refuel it on orbit and then you send it on its way to wherever it needs to go. And so instead of sending like four tons to Mars at a time, we can send 150 tons to Mars at a time. I think it's hard to overstate the magnitude of the ambition and sort of like what that does for unlocking access to orbit. And then if you think back to like what I was talking about earlier about the gravity model of trade, think about like orbit as a place where we trade where sort of having economic interactions. What does a 200x reduction in travel costs do for trade?

well, it should 200x the amount of activity. We should put up probably like 200 times more valuable stuff into orbit and have more activity. So right now we basically do communication satellites and Earth sensing satellites, and that's like most of what we do in orbit. This opens up the opportunity for all kinds of stuff, for space manufacturing, for a sustainable human presence on other bodies, right, on the moon and on Mars, for private initiatives in space, private space stations. eventually for stuff like resource extraction, like asteroid mining and stuff like that,

probably still more than 10 years away. But that's all coming. And it's basically, you know, if you can get that launch cost point down low enough, you really start to make that basically inevitable. The other piece of it is that the launch cost plays a role in what kinds of things you can send into space. From a hardware perspective, if your launch costs $50 million a pop, you are going to spend millions of dollars at least on your spacecraft, on your satellite, to make sure that it's reliable, that it doesn't crap out as soon as the sun shines on it or,

you know, something happens. If you can get a ride share on a starship and it costs you $150,000, that's the point at which you can start using commodity hardware. You can start sending iPhones into space, using them as sensors and much more consumer-grade equipment into space, which, you know, enables much faster development and much more powerful satellites. Unbelievably cool. First of all, it's like a sci-fi fan or something. It's sort of surreal that some of this stuff might happen in our lifetime. Are there other companies, SpaceX is like such a dominant brand and force in this story? Are there other key players that you think people could or

should study if they're really interested in this? SpaceX has always been compared to Blue Origin. Blue Origin has been moving very slowly through their development cycle. So they have a notional idea of this New Glen rocket that will be not quite as big as starship, but bigger than Falcon 9 and maybe bigger than Falcon Heavy even. Big rocket, but they're not really doing that. They're basically focused right now or have been focused on their suborbital rock. So a rocket that can't make it to orbit that they can use for space tourism. And so space tourism, they are ready to launch that now with their last test. Won't be long before you can buy a ticket on a blue origin rocket and go above the Carmen line at least and become technically an astronaut.

How much are they going to charge for that? I don't know. Would you pay a quarter million dollars, just barely become an astronaut? I don't know for like a couple minutes. Maybe some people will. So that's happening. The other thing that's happening with them is that Bezos is quitting Amazon to focus on other things.

And I think that the other things, to considerable extent, it's got to be blue origin. It'll be interesting to see what kind of attention he brings to the company and if that improves their rate of execution. The other company that I think is really interesting right now is Relativity Space, a launch company that 3D prints rockets. So that is a really interesting manufacturing technology for aerospace in particular because it's so sensitive to weight and strength. And if they can simplify, reduce the part count and make it stronger and make it lighter and make the construction time faster, that's potentially really interesting technology. Does that alone catch them up to Starship? I doubt it.

But it is really interesting. and Bezos was seen touring their factory not long ago. I am personally wondering, like, does he come in and just buy them and merge them with Blue Origin? And then all of a sudden you have a very large company with a very interesting production technology. Seems obviously like competition would be great. You want as many people trying as possible for that cost to come down.

Exactly. I mean, the other aspect of it is like, what are people going to do in space? There's the earth sensing element. Of course, Starlink is super interesting from a communication side, having a low Earth orbit internet constellation that basically provides internet access anywhere on the planet, no matter how remote or rural, that's really interesting. Varda is doing in-orbit manufacturing. There's some products that you want to build that basically require putting together components in a way where the structure of the component itself might collapse on itself

in gravity. So gravity itself makes it unstable at some point in the construction of the good. And so being able to do it in microgravity, being able to like 3D print something in microgravity means that it won't fall apart while you're making it. There are some goods, especially in a starship world, it's probably worth producing in orbit and then bringing back to Earth. You could think of fiber optic cables or silicon wafers or certain pharmaceuticals or people have talked about 3D printing organs. Gravity wouldn't let you do it, but you could do it in space. That's a cool thing. There's a company called Axiom that's designing a space station that humans could go up into, right? The international space station is starting to show its age. Like, isn't it time for a

private space station that is in orbit that scientists could go up to and do experiments? That's all happening. The last category for us to discuss, well, there's two more, but they're both sort of information technologies. We started at the beginning, or you did, by saying this is the one area that we've seen a huge amount of progress, information technology. Still, obviously, there's more progress to come. Is there any interesting vector here that you're most curious about in terms of what might drive productive growth for society in IT specifically? I think there's a couple things, maybe three things. We kind of touched on this earlier when we talked about protein folding. We talked about deep mind alphabet lab solved protein folding. Well, they did it using machine

learning. And to me, that's a big breakthrough in machine learning because it's the first time that machine learning has been applied to a problem with superhuman results where the problem actually really mattered, where it wasn't like a toy. They solved chess. They can beat every human in chess, superhuman performance in chess. GP3 is like close to human performance in writing stuff, but it's not like a really big problem. It's still like most of the applications are toys right now.

Machine learning is like turning a corner, I think, in terms of being able to actually like beat humans at stuff that really matters. That's one aspect I think is really interesting. Another IT related thing that I think is important and interesting is the custom silicon trend. So Apple is ditching Intel processors, right? And they're going to, on their computers, house designed chips, basically the way they have been doing on their iPhone and iPad. So it's all a system on a chip. It's basically a continuation of the trend of trying to get computer components more densely packed, more transistors on a chip. But instead of just transistors on a chip, you're talking about all of the components of a computer as densely packed as possible on one piece of silicon.

The reviews of the M1 Max, sort of speak for themselves. But you see this trend also in machine learning computers, right? So people are doing chips that are basically entire systems, massive, massive systems on a very large ship, like full wafer size systems on a chip that can radically speed up machine learning. I think Tesla is doing this for their full self-driving computer. as well, it's all custom silicon. So that trend, I think, is only going to continue, right? Basically, we've got to be able to put all the components of a computer more densely packed right on the same piece of silicon.

That's going to enable us to like sort of the way it feels to compute. It's going to allow us to exceed Moore's Law for a little while. Things will get faster than Moore's Law. I think it's really interesting. The other IT thing that I talked about a little bit was crypto. I am interested in following it since 2011. I think I was the first economist in the world to write about cryptocurrency. To me, like, the key question is, like, what real world use will it get?

I am more interested at this point in Ethereum than Bitcoin. Bitcoin is like people are saying it's digital gold. They've changed the narrative. What's going to be a revolutionary payment system? Now they're like, oh, no, it's digital gold. I'm still interested in Ethereum because of all the smart contracts functionality, all the tokens, the identity, the new applications that can be built. Vitalik had a really good talk recently, Vitalik Buterin, on the future and the far future of Ethereum.

And it's just mind-blowing to think about all the different technologies that they are pouring into that chain. I think it's going to be like air. It's going to be everywhere. Your phone is going to be able to basically sync up to the blockchain in a few seconds and get cryptographic proof that it got to the latest block of the chain and just be able to go interact with this resource. What's most interesting to you in terms of what it might enable? Ethereum is fascinating, by the way, especially right now. And the question, of course, is like if it's the flexible cryptocurrency that you can build stuff on top of, what is most interesting to you in terms of what that might enable that

wouldn't be possible in the absence of a decentralized blockchain like Ethereum. One thing that I've really been interested in that already exists, the fees are too high to really use it, is prediction markets on top of blockchains, right? So I'm really interested in prediction markets for finding the probability of some event happening, using people who bet on the market to sort of like infer from the market price of that token a probability of an event. So basically markets for information. And I would love to see markets for basically every question you could think of. And then combining those markets in a combinatorial sense, you could think about what conditional on X happening, what is the probability of why? So you could even

think about conditional on the American Jobs Act plan passing Congress. What is GDP in 2025? Thinking about that, being able to say, well, if we do this, what do markets think is the likely outcome? To me, like, that's really interesting. That would change the world. This has been awesome. We've covered so much ground, so many interesting things that might change in the future. I love your framing about total factor productivity that we talked about at the beginning. And obviously, all of these things would be exciting, things that could bend that curve up into the right in ways that I think would be good for the world. I appreciate your time. I ask everyone the same closing question. What is the kindest thing that anyone's ever done for you?

Oh, man. That's really hard because I'm thinking of like so many, so many different people who have come along the way. I think that there's tons of value in coming along younger people and just encouraging them. If they're on the right track, just sort of like giving them a slight amount of encouragement. So I'm thinking about like when I was like blogging early in my career, people would go out of their way to, like, promote my work or something like that. Like, that is, to me, set up. I think of, like, so much of my career has come from writing online. Got me my first tech policy job, continues to pay benefits today.

People who have promoted the earliest, roughest iteration of my work and just encourage me to keep going. I'm really grateful for that. I love it. This has been so much fun. Thank you so much for your time and everything you taught us. I'm sure I'll keep reading. This has been great.

Thank you. Thanks for having me on. This episode was brought to you by Canalyst. In this four-part mini-series, I sit down with Canalyst's new chief product officer, Jeremy Payne, and talk about his background in fundamental data, the role Canales plays in the investment process for its clients, and how Canales products help investors better model and understand companies and their key drivers. In this week's episode, Jeremy and I discuss Canales' coverage of 10,000 businesses and how Canalist has rethought the process of updating your models for a business.

One of the things that we've talked about with Demir before is this notion of all the duplicated work that happens in this industry for bysiders. And obviously, Candlest is as a product reducing a lot of that valuable but not high value add work that investors are all doing some version of and they kind of come to similar conclusions. As you think about the product today, so let's say right now, Candelist is providing incredibly clean business specific financial statement and KPI data that. that helps eliminate the analysts need to do that crap, the same as everybody else. What's next there? Like, what are the other things that are in the by side analyst or PM's workflow that you think are important but not high value add in terms of differentiation between analysts that analysts might tackle in the future?

I think there's an incredible value ad that is provided by doing that work in a reliable and trusted way. The idea that there is an equity research shop that covers, today we cover close to 4,000 companies. Pretty soon we'll cover 10,000 companies. I don't think any of the big investment houses covers more than maybe 1,500 or 2,000 companies, and pretty soon we'll be covering 10. We cover them with a consistency that they would be jealous of if they could see them. And I think we wrap them up with some clever technology that makes them useful to,

an investment analyst in their process and adds additional value add. So in addition to doing a whole bunch of things that are repeated in the process of getting all of those facts together, historically, when one of the great challenges with getting a model from somebody else is what happens when the company reports that next period when the future becomes the past. And that analyst driven next quarter becomes a company reported actual. And it's time to, to put all of those new numbers into it, the cell side never had an answer for that. I mean, you could like Kleenex, you could throw the old one away and download a new one, but if you put anything in the other one, what do you do with it?

And that's one of the other answers that Canalist has come up with. Technology for updating your model at the end of the quarter, seamlessly and easily at the press of a button from right inside of Excel, gives you a variance report says, hey, here were all the assumptions that you had for Netflix, revenue and EPS for next quarter. Here's what they actually reported. Here's how close you were and set you up to do next quarter's estimates. If you enjoy this episode, check out join colossus.com.

There you'll find every episode of this podcast complete with transcripts, show notes, and resources to keep learning. You can also sign up for our newsletter, Colossus Weekly, where we condense episodes to the big ideas, quotations, and more, as well as share the best content we find on the internet every week.