Chase Lochmiller (Crusoe Energy Systems) on turning stranded energy into Bitcoin (EP.131)

Yeah, thanks for having me. Happy to be here.

We historically haven't been super public about a lot of our operations, but we're starting to come out of some degree of secrecy and talk more in a public facing manner.

I did my undergraduate work at MIT focused on math and physics and then I got into the quantitative trading space. I worked for a couple of different big electronic and algorithmic trading firms. One called Getco which was one of the largest market makers in the world, then acquired a company called Knight Capital that had sort of a big trading snafu and we ended up going public and that entity has since merged with Virtu. So that's all kind of part of a large public electronic trading company. Now I ended up leaving Getco at some point and going to work at another electronic trading company called Jump Trading and Jump actually has a pretty big footprint in the crypto space now. And I think a lot of people in crypto are aware of Jump Trading and what they're doing on the market making and trading side of various crypto exchanges. But I ended up getting super interested in the digital asset and cryptocurrency space. And I ended up joining Polychain Capital as an early partner in 2017, sort of at the beginning of the height of all the ICO madness that sort of took over in 2017. But I ended up leaving in 2018 to sort of pursue a personal dream of mine, which was to climb the highest mountain in the world, Mount Everest. So I spent about two months kind of thinking about my life personally and what I wanted to do next. One of the big things that I was really really excited about was this notion of being able to deliver computation cheaper. Just being able to bring down the cost of compute I think has been one of the large driving forces for innovation across society and has been one of the big enabling factors for large scale artificial intelligence to really make an impact on people's lives. And then it's also brought a tremendous amount of security to proof of work blockchains like the Bitcoin blockchain. So, I ended up after summiting Mount Everest, I came back and I was doing some more thinking and I was in Colorado and I ended up meeting up with a very close high school friend of mine who's now my co-founder, Cully Cavness. And Cully, we went to high school together, but we ended up going very, very different career paths. He went to Amherst and he studied geology there. He was a Watson fellow after graduating and sort of traveled around the world studying the various energy resources around the globe ranging from coal fired power plants in China to stranded geothermal assets all over the world and he spent a good deal of time in the geothermal industry before sort of making his way back to oil and gas. His family had been in that industry and was kind of somewhat of his calling. So he ended up working in various angles of the oil and gas industry ranging from the banking side. Cully had most recently been working for an upstream oil and gas company and he was dealing firsthand with this flaring issue. And just to give you a quick introduction to what flaring is and why it happens. So when oil companies drill oil wells, they're underwritten on the economics of the oil production expected from that well. Now, as a byproduct of the oil production, they also produce natural gas and water. And so, these three different things come out of the ground. But there are certain cases where they actually don't have the infrastructure to get the natural gas to a downstream market where it can actually be sold. So, this is a case where either they're either not connected to a natural gas pipeline or they're connected to a pipeline and that pipeline's fully at capacity. Now the water they can take and they can transport it via truck to a water treatment facility. The oil they can transport to an oil refinery with a truck as well. But the gas because it's a gaseous state, being able to transport it in lieu of a pipeline is actually a huge challenge. So the most economic thing to do with the gas for these oil companies is actually just to burn it on site. They end up burning it. And it seems like an insane crazy thing that this exists if you've never spent time in an oil field. Driving around an oil field and seeing this flaring taking place is really a sight to behold because you just you're driving down the road and you'll see these giant fireballs right next to the side of the road. So to put it in perspective on how much 14.5 billion cubic feet a day is, this is enough to power the entire continent of Africa. So it's a really tremendous waste with a big emissions footprint and zero beneficial use for humanity. So when Cully kind of came to me with this problem, the real insight that we had about it was it's a waste not because oil companies somehow fundamentally hate the environment and they're trying to pollute more, but it's truly like the most economic thing for them to do. And the reason it's that way is it's fundamentally a transportation problem. They can't transport the gas from the place that it's being produced into a marketplace where it could actually be consumed. It's the same problem that faces dairy farmers with biogas and the cows are belching and farting and creating methane emissions and they would love to be able to capture all of that and transport it to a downstream market where it can actually be consumed and sold, but that problem is difficult. So anyway, what we ended up coming up with was this idea of basically taking a very energy-intensive demand and taking it directly to this source. And that demand was in the form of computing. And we started initially with basically working with an oil and gas company in the Powder River Basin in Wyoming. And that was our initial pilot project. And we started in 2018. And we basically built this mobile modular data center that we took directly to the oil well and kind of had it outfitted for rugged oil field conditions and ran it through the winter there and experienced multiple different bomb cyclones. And I think the day we were in negative 40 degrees in sort of the eastern plains of Wyoming which were very cold and windy. But being able to sort of operate through those harsh conditions really was a testament to I think that we sort of had something there that could work by absorbing this very very large and wasted energy resource with computation and specifically with Bitcoin mining.

Crypto for me, it always resonated with me just having some sort of algorithmic monetary supply. Like it just made a lot of sense to me that way. I think when I was much younger, I was a lot more like hardcore libertarian. I would say I've kind of eased off on being as hardcore in my older age and sort of seeing a role for government in a lot of different aspects. But when I was younger, I think I first found out about Bitcoin in 2010. There was a developer I worked with that told me about it, and I was like, 'Oh, wow. This makes a ton of sense, and this is really cool. This is like a really, it's cool to have sort of a native internet money where the monetary supply is governed algorithmically versus by the Federal Reserve.' So, that was my initial interest. I stayed in sort of the algorithmic trading space for a long period of time, just kind of passively observing and sort of watching the crypto space. But I think what really got me more interested in it was as it started gaining more traction and greater wider adoption, I think it brought greater credibility to the space. It could be used for more things than just Silk Road and some of the early popular applications of Bitcoin. And I think at that point it felt like it was something that I wanted to spend a lot more time on. That was kind of my personal journey into the cryptocurrency landscape. Having something that's tradable 24 hours a day, I think makes a ton of sense. Always resonated with me. Having traded in traditional markets, I always kind of thought it was bizarre that we would just like all of a sudden like the markets are closed. I mean, at the end of the day, like the portfolio manager, if the strategy is running or it's not like it takes a ton of effort for most of these automated trading strategies to continue to run overnight or whatnot. So, it was always kind of like they have after-market trading, but why can't I buy shares of Apple at like midnight if I wanted to? That doesn't make any sense that I wouldn't be able to do that.

From the operator's perspective, there's really like three different interested parties here. You have the landowner, which is actually typically like a farmer or rancher, someone like that. Then you have the operator, which is the oil company. And typically what's the arrangement that was set up is that the oil company has a mineral rights lease to the subsurface usage of the minerals that are below the land and then we are sort of the service provider here. The landowners typically paid a royalty on anything that sort of gets extracted from and sold from their land. But the United States is actually a very interesting and unique place in the world because it's the only country in the world that has private subsurface ownership. So every other country in the world, everything that's underground is sort of owned by the government. And this is actually one of the main reasons you've seen the most innovation within the energy sector occurring domestically here in the United States because those mineral rights are privately owned and operated which has driven a tremendous amount of distributed and capitalism-fueled innovation. Like the whole concept of horizontal drilling and hydraulic fracturing like that could have never occurred anywhere else in the world because nowhere else in the world they would have had the right monetary incentives to be that innovative. So anyway, back to your original question of what the conversation looks like. Flaring is an issue that's governed on a state-by-state basis. There are EPA guidelines as far as emissions go, but typically the main restricting factor for oil and gas companies is managed on a state-by-state basis. So in Colorado, you have the Colorado Oil and Gas Conservation Commission. In North Dakota, you have the North Dakota Industrial Commission and the NDIC. In Texas, you have the Texas Railroad Commission. So each one of these commissions has a different regulatory approach to how they manage flaring. And depending on where you're at, flaring is actually a nuisance to these oil and gas companies. They're producing this gas and the gas is really a nuisance that they're not expecting necessarily to make a ton of money in the perspective of what they're the main profit driver for them is the oil and it can often become an inhibitor if there's a regulatory requirement that basically forces them to flare under a certain amount of gas. That was early on probably one of our main value propositions was being able to help oil and gas companies operate more efficiently in a more environmentally friendly capacity that helps them achieve regulatory compliance within local jurisdictions. So from that standpoint I think it was a very clear value added proposition because it basically helped them produce oil and they're very familiar with other types of technologies. What we were approaching them with was a much more economic solution than some of the alternatives. So to give you an example, one alternative that people can do to basically manage this flare gas is to convert it to LNG, liquid natural gas at the location. Now the problem is the cost of doing that conversion at a small scale. That's basically a pod level of gas production is just not economic. You're going to end up spending way more money to do that conversion process and the transportation of the LNG to a market than what you're actually able to realize from a revenue perspective. So, it's basically a way of losing a lot of money in a compliant fashion. What we brought to the table was a much more economic solution for operators to operate in a compliant fashion. So, people were very very excited about it.

We don't shy away from the fact that we're mining Bitcoin or anything like that. It's a very clear and open conversation. We tell them kind of what's happening and how the mining process really is a critical infrastructural process to the Bitcoin network existing and how we're validating and securing the transactions that are occurring on this decentralized network and people get it. I think particularly in the energy industry, I think people are very receptive to the way that that system works.

Our relationship with the energy companies is it's not typically like a joint venture partnership where we're helping them mine Bitcoin. Typically, there's a transaction where we're actually taking the gas from them, purchasing the gas from them and the transfer of ownership happens at that point and then we sort of deal with all of the Bitcoin trading and monetization piece of it.

Exactly. I mean, we like them to know and understand what we're doing just so that they're very comfortable with it because we have to get all of our surface use. We have to get all of our equipment approved and everything that we do is ultra compliant with all of the standards associated with both the electrical standards as well as kind of the oil and gas specific standards around where we can place equipment on the location.

I think Bitcoin mining industry has kind of like undergone like the classic S-curve industrial explosion dynamic where like you said early on you could mine Bitcoin on your laptop and then as things got more advanced people were starting to mine it on GPUs and then people built the first ASICs and then the speed at which ASIC innovation was taking place was incredibly rapid and at one point you would place an order for one ASIC and by the time you took delivery of it it was no longer economic and there were all these kind of crazy games that were unfolding. I think that was really kind of this explosion vertical part of the S-curve. But I think since we're sort of reaching a plateau where I think a lot of the innovation taking place specifically in the ASIC space has flatlined to a certain extent. I mean, I think it's just growing a lot less quickly to where you can predict these things a little bit better and have a little bit more insight as far as what's expected from an efficiency standpoint of future generations of ASICs. And a lot of this is just based on the global silicon market and they're sort of just Bitcoin's right at the cutting edge of that curve. But because it impacts so many other industries, I think it's just become a lot more predictable. So there's definitely an art to just staying plugged into what's happening within the community, having conversations with various vendors and other partners in the space. But it is something we think about a lot. But I think fundamentally we're trying to stay ahead of the curve by having a low operating costs. I would say globally competitive operating costs. We feel like if we can kind of be in that top tier of operating costs that we can maintain a margin even if there's sort of unexpected technological advancements.

I think just access to machines, just having the right supply chain in place because our effective power price is so cheap that we don't have to have access to the latest and greatest, most efficient units to maintain profitability. There's always a trade-off between being at the very front of that and then sort of being able to make money from the things that aren't economic for a lot of different people. I think we kind of participate across that whole spectrum, but a lot of it's just about having the supply chain in place and really being able to access machines and kind of know what they're worth.

I think fundamentally like at a high level Crusoe's mission is to unlock the value in stranded energy resources with the power of computation. And we think about that more broadly beyond flare gas as well as beyond just Bitcoin mining. So we have some kind of early pilots in other energy-intensive computational applications that are for things like training large scale artificial intelligence algorithms, things like drug discovery through massive large scale protein folding simulations and things like graphical rendering to make the next Pixar movie look ultra sharp and make sure the kids think it's super realistic or whatever. But I think all of these markets are markets that really really stand to benefit from this type of cost reduction that you're seeing unfold in the computation space. There are many different types of stranded energy resources. There's a handful that we're looking at as a next market to tap into beyond just flare gas mitigation. I can't really discuss it in too much detail today, but you should be on the lookout for something from us here, I would say, in the next 6 to 12 months. One thing I would say that Crusoe tries to focus on fundamentally is like you said there has been a lot of things published around Bitcoin being a wasteful use of energy. And I think from our perspective, what we try to do is we try to flip that and try to find wasted energy resources and places where widescale emissions are either occurring or where there's renewable resources that could be made more economic by basically adding on this base load of Bitcoin mining. And we try to really really just focus on the emissions reduction problem. If we can be helpful in sort of the instead of kind of adding to the problem, if we can only focus on ways that we can help reduce things that are stranded or otherwise wasted, I think that's really one of our core focuses here. So from that standpoint, I think the world actually stands to benefit quite a bit from the Bitcoin mining industry. Bitcoin mining has a lot of unique elements to it because it's fundamentally a decentralized architecture. Contrasting it to other types of compute applications, Bitcoin mining is an interruptible application. I mean, you don't want to be off more than you need to be, but it's not like the network suddenly collapses just because one Bitcoin miner turns off. The network continues to persist and transactions continue to happen and network continues to be secured so long as that's not a broad scale everybody turning off at the same time type of thing. But that interruptibility actually provides a very very unique feature just from the standpoint. I mean, I know it was super helpful for us with mining with flare gas in the oil field because that fuel supply, it's not necessarily a consistent fuel supply. So, wells have to go down for maintenance periodically. They have issues that come up. There are things that happen in the physical field where lines can freeze or things can happen that tend to cause shutdowns. So having that ability and that flexibility inherently to interrupt the process is actually a super valuable and underappreciated aspect to the Bitcoin mining process.

Yeah, absolutely. Broadly speaking, I think Bitcoin mining is probably the or proof of work blockchains more broadly are probably the most commoditized compute applications that exist today. But I think fundamentally you're going to continue to see that trend unfold in other sort of broader digital commodity spaces. So I mean what I mean by digital commodities are sort of compute as a compute basically flops as a digital commodity, storage time as a digital commodity and network transfer is sort of a digital commodity. And I think fundamentally those commodities should have prices in the same way that pork bellies have a price that gets set on the CME futures market or orange juice futures kind of get set in the same way or corn futures or whatever. I mean these are all sort of commodity assets that are physical commodity assets as opposed to digital commodities. But I

I think Bitcoin and proof-of-work blockchains are really paving the way for a future for broad-scale digital commodity adoption, and then in turn basically commoditize pricing, which will I think be a net benefit to any user of basically the internet.

They will be, I would say, kind of in the future. I think most of them are early stage enough where it probably doesn't make sense from a decentralized perspective, but I don't think they've been able to attract kind of enough buyers and sellers to make it meaningful. And I think a big part of this really inherently is because when you look at the large cloud providers, AWS and Azure and then GCP, really what end cloud compute consumers are the biggest value-add that they get from that really is the platform. So I think to the extent that open-source platforms gain more adoption and gain more usage and utility to end enterprise cloud services compute consumers, I think you're going to see greater utility and demand for commoditized back-end digital commodity consumption. So I think you're already starting to see a lot of these trends with Google kind of open-sourcing Kubernetes and all of the development that you're kind of seeing in that space. I think could pave a path forward for this type of future.