by Stephen Lacey
February 12, 2018

Stephen Lacey: This is The Interchange, conversations on the global energy transformation from Greentech Media. I'm Stephen Lacey with Shayle Kann and in today's episode we've got a new segment that will hopefully make us a little bit smarter about blockchain and how it will impact the energy industry, we're calling it Consensus. Okay Shayle, we've done a couple episodes on energy blockchain applications. Why are we picking this back up?

Shayle Kann: Well because first of all, there are a lot of blockchain applications in the energy sector that are emerging and evolving, and every time that we talk about one of them on this podcast, another one springs up and it's different and it's worth a conversation because we need to understand it and we need to figure out whether we believe that it's useful application or just kind of vaporware. And also because just the blockchain universe is changing so fast right now and with all the hype and all the new companies springing up in it. So I think every time something new comes up as you said, we find ourselves just as confused as we had been about the previous application before we learned about it. So there are just going to be a bunch of opportunities over the next few years to take a step back and trying to understand the new thing in block chain and energy.

Stephen Lacey: We are going to bring a blockchain related topic that we don't understand, a term, a business model, an application and present it to GTM's resident blockchain enthusiast, Scott Clavenna to see if he can help us out. Scott is the Chairman of the Power & Renewables business at GTM and Wood Mackenzie, and as regular listeners know, he's studying this space closely. In fact, Scott is also putting together a block chain and energy forum in New York City on March 8th. Scott, how's it going?

Scott Clavenna: Good, good. Thanks for having me on. This sounds like fun.

Stephen Lacey: Are you to see if you can help us newbies out?

Scott Clavenna: Yes. I guess. I feel very sympathetic to the notion that blockchain ... studying blockchain is literally like you open one door that you think will solve or will answer one of your ... the last key question you have about it and you realize there's a thousand other doors behind it with ... that exposes the depth of your ignorance. So I'm getting there, I'm walking through the doors, but every time I figure something out, I realize I need to figure 50 other things out.

Shayle Kann: Can I ask you a question about that? We've talked about blockchain before; once in a while you've made an analogy to the internet, your lead is the internet, which you also were involved in at the time as I could tell, communications expert. Did you feel similarly about the internet in its early days in terms of like how hard it was to understand?

Scott Clavenna: Yeah. I had a conversation about this last week and I think I did not. I did to some extent. There were times where looking at optical networks I realized I had to understand like how telecommunications transmission networks worked to understand the value of one particular component or some breakthrough. So I had to do some more reading and often times I get asked just about photonics, like the underlying technology or physics of how optical networks work. So I'd have to do a little reading, but I never got ... I never found just one answer led to so many other questions. I mean here, I think because blockchain, it applies to so many different industries and so many different applications, you're not solving just one thing with it. That even as we talk about blockchain, you're talking about cryptocurrency.

So you have to suddenly ... You realize you have to step back and say ... answer some really simple questions that are not easy to answer. Like what is money, what is a store of value versus a currency, like what's the difference between gold and cash. You know? And these are things you all take for granted but if you really want to answer questions about blockchain and cryptocurrencies, you have to feel like you understand all of that. What's an international trade? How do you settle transactions across borders? Like I don't know. Then you realize like blockchain has a great application for that, but then the only way to talk about it is to back up and say, "Okay. I need to figure out like cross border settlements and counterparty risk." And you know, it's just ... It's endless.

Shayle Kann: Totally, and it's also like a very multidisciplinary problem and that you just described a bunch of questions that are sort of answered by like macroeconomics and international trade as disciplines. But then there is another set of questions in order to understand blockchain that requires an understanding of like how computers work and how trades are settled in financial markets and ... So you ... trust and security, cyber security, like all of those things. If you want a full understanding of blockchain, you have to like get all those disciplines, which I feel like very few people do.

Stephen Lacey: Okay. So we do have a couple concepts that we want to explore today and we'll see where that takes us and if we start unpacking some of those much simpler concepts that I feel like maybe some of us should know, but it's laying bare the limits of our expertise, or at least mine. The first is this emerging energy trading firm, WePower and I want to understand how tokenized energy trading will work and how WePower differs from other companies out there. And Shayle, you've got big questions about the energy use of bitcoin.

Shayle Kann: Yeah. My set of questions isn't about a specific application of blockchain in the energy sector, but rather the energy use of cryptocurrency mining, which has gained a lot of attention and for which there are a whole bunch of questions within that.

Stephen Lacey: Let's start with WePower and this concept of using the blockchain to verify energy trading. This is probably the first mover application for a lot of firms in this space and it's going to be important for retail electricity sales and wholesale market trading, and it seems like a lot of companies are pursuing the wholesale side. So that's where WePower is at. Again, it's this blockchain based platform for renewable specific trading. In a sense as I understand it, it's a way to raise capital for a project. So if developers want to raise money for a solar project, they issue tokens on the platform for future electricity delivery and buyers of those tokens are basically investing in a project with the promise that electricity will get delivered at a later date, a forward contract of sorts, and I want to understand whether this mirrors existing forward contracts. This is an example where I sort of understand each element of the company's business model and some of the corollaries out there in the market, but I'm having trouble with aggregating them all together. So maybe to the best of your ability Scott, you can quickly summarize what WePower does.

Scott Clavenna: Sure. Sure. Yeah, it's a new ... I think it actually is fairly unique. I think WePower by focusing right now exclusively on the project development side of renewable energy is actually kind of unique. It's not one of these peer to peer, more retail focused energy trading mechanisms. It does tokenize generated electricity from renewable energy assets or it could be a battery, just anything where electricity is delivered from an asset. It tokenizes that and allows people in a market then ... It creates a market for those tokens that can be traded in various ways among market participants or to actually trade for use of that energy. But it is fairly unique in that its focus is to use blockchain as a platform to create a market that then creates a financing platform for renewable energy projects.

Stephen Lacey: I got to stop you right here. I'm sorry to interrupt you but we already hit a concept that probably needs some explaining and that is tokenizing. So like ... I think that we're just automatically assuming that people understand what that term means. And before we get into the transaction itself, like tokenization as I understand it is like the assignment of random values to a packet of information, which is like basically-

Scott Clavenna: No. No.

Stephen Lacey: No?

Scott Clavenna: Not exactly.

Stephen Lacey: Okay. So here we go.

Scott Clavenna: There's a --

Stephen Lacey: Let's embark on this journey together.

Scott Clavenna: There is tokenization in the IT world that's somewhere around there. But tokenization in this world of blockchain and crypto is ... it's converting rights to an asset into a digital token on a blockchain. So it's an assignment of rights to anything, that can be a physical asset or any kind of virtual asset. It could be copyright, it could be ... It's just ... An asset then gets a broad definition here, it can be fungible, it can be physical, all this but it's the assignment of rights to that. And so the value of it is set by a market or by the issuer but the key thing is that tokenization is assigning rights.

Shayle Kann: So as example then, when I say buy stock right now, like if I go onto my Fidelity account and I buy one share of a company, that is not currently tokenized because there is no sort of digital, unassailable unique asset assigned to it. But if there were, this is what cryptocurrency allows. Right? Is that in theory, I could buy stock that was tokenized and if I did, the difference would be my Fidelity account might look the same either way. But what would actually happen in that case is that like a unique code would get assigned to me that everybody within a blockchain recognizes as mine and then I would be trading that token.

Scott Clavenna: Yes.

Shayle Kann: Right. So it's that ... The distinction is we still ... We would buy and sell stuff online. Right? But the stuff that we buy and sell online, if it's not using blockchain just does not have this kind of unique, verified identifier attached to it.

Scott Clavenna: Yep. No, that's ... I think the promise, the beauty of all the tokenization is that once you have tokenized something and it is in this digital format. Not only have you now digitized sort of ownership of everything in an atomized kind of fashion, you could sell fractional ... sort of shares as it were, fractional ownership of literally anything, of an activity, of a copyright, of a physical asset, literally anything. But you also have done it on a blockchain that's global and standardized, like you have a platform that crosses borders, that crosses any kind of regulatory regime, different currency regimes, anything. You have just like a digital token that can be traded more or less instantaneously, anywhere that the blockchain or that that market has been created.

Shayle Kann: Right. Okay. So let's now apply this to a renewable energy project. So I build or plan to build a solar project in wherever ... in Latvia and that project is physical, but I've decided to tokenize the generation from it. Now I guess my first question in terms of what WePower is doing is are they assigning rights to like equity shares of the project or are they assigning ... Is one token equivalent to X percent, 'cause that's the ... You know there was a wave of sort of solar crowd funding a while back and I think that's what ... it wasn't using cryptocurrency at that point but what WePower's doing sounds roughly similar to that, like assigning rights to a diverse group of individuals. But they were largely assigning like rights to a share of the ownership of the project and this is different. Right?

Scott Clavenna: This is different. Yeah. As far as I understand it and I guess I should say I haven't spent any time talking to the founders of WePower. I've read a bunch about it, spent some time with the white paper, talked to some other experts about solar project finance and how this compares. So with my understanding that's not what you're when you're buying ... when you're taking part in the financing of one of their renewable energy projects. So they would still raise capital in the traditional means, there would be equity investors and lenders that would provide say 90 percent of the capital for a project. Then what they would do is that remaining 10 percent or 20 percent, so it's only a minority share of the financing required, they sell through tokenized energy to be produced from that platform in the future. So that's what you're buying. You're not really an equity stakeholder, you're more or less being ... You're buying a fairly liquid, short term PPA.

Shayle Kann: Right. Except that you ... So then the tokens are measured in kilowatt hours or megawatt hours. They're measures of-

Scott Clavenna: Yes. One ... Each token equals one-

Shayle Kann: Right.

Scott Clavenna: It's one to one.

Shayle Kann: So then what happens at the actual project is the project gets built and starts operating and sells power into whatever market it's in. Now, I the token holder don't actually have to be a consumer in that market, so I can't utilize those kilowatt hours or megawatt hours myself. So what I actually get from that is what? My ... For every kilowatt hour generated, the project generates revenue and I get assigned some portion of that revenue? Or how does it actually ... Like what do I get?

Scott Clavenna: The way they described it in the white paper and the way they're out talking about it and they ... I guess I should preface this with a bit of a disclosure. So WePower is in the midst of a initial coin offering, they're going to sell a ton of these tokens. Not these kilowatt hour tokens, which are going to sell WPR tokens, which just give you priority access to the WePower network. And with that priority access, then you get to participate in these auctions of the tokens that we're talking about here associated with individual projects. So the key thing being there is an ICO coming up. Just one disclosure, I'm not involved in WePower. I'm not involved in that ICO and nothing we're talking about here today is meant to be an ICO preview or assigning any value or opinion on it. We're just going to try and describe this as best we can, so you have an understanding of it.

Okay. So when these projects then generate power and thereby tokenizing the electricity, you have these tokens and so WePower has given you three options of what to do with those tokens. If you are in that market, you actually can use those tokens to literally consume the energy that is produced. So you can be a buyer of that energy with your ... you trade your tokens for that energy, and a key point that I overlooked is that the whole value of buying these tokens in these auctions to ... this fractional share of the energy, is that it's already been negotiated at a discount from the prevailing market rate for power in whatever market that's in. So it's 10, 20 percent below market rate. So you are basically through providing funding up front, you're entering into a kind of a liquid PPA that is already set at below market rates.

Shayle Kann: Okay. So I'm now ... Let's just take this example for a minute before we move on to the other ways you can use the tokens. I'm ... So I know that WePower has a partnership in Estonia with a big utility there.

Scott Clavenna: That's a test bid. The key thing is Spain, where they're actually going to do the first project.

Shayle Kann: Okay. So Spain. So Sam is Spanish --

Stephen Lacey: This is a Gibraltar-based company.

Shayle Kann: Okay. So say I'm a Gibraltar-based consumer of electricity, and I buy some of these tokens ... So I buy into the ICO, which then allows me to participate early in a token auction because they have a project in my region that is coming up or getting developed. So I do that, I buy a bunch of tokens. I now hold tokens that represent kilowatt hours in my market, generation in my market and the price of those is ... somehow is 20 percent less than my retail price electricity otherwise would be. Can I actually assign those tokens directly onto my bill? Meaning have they associated a deal with the utility such that I ... it actually is incorporated into my bill as savings as if I had solar on my roof or is this sort of a separate out of network transaction that just like settles to 10 or 20 percent less?

Scott Clavenna: That is a good question that I'm not a 100 percent sure if it goes all the way down to you if you the market participant are in a home. I think that's what they're going to test out in Estonia where there's a 100 percent smart meter coverage, then they could extend that all the to the home because the smart meter would be that digital meter of record of consumption. It may not be the case in Spain, it may be more wholesale to retail transactions. So it's not necessarily to an individual homeowner, but someone operating in a retail space could buy lower priced wholesale power.

Shayle Kann: Okay. All right. So one way or another, that use case seems pretty straight forward to me. I use it to offset my bill and I save some money on it as if I had installed my own generator.

Scott Clavenna: If you're a retail provider, you could be buying energy from the wholesale market with these tokens. Say the wholesale market's at four cents, you could be buying it at three cents. Right? You can exchange your tokens for the equivalent of it as though they were three cent kilowatt hours.

Shayle Kann: Yeah, that version looks a little bit more like an offsite, corporate renewable PPA-

Scott Clavenna: Yes. It's exactly like that.

Shayle Kann: In the U.S., yeah.

Scott Clavenna: Right.

Shayle Kann: Like whether or not it's on your bill. Okay. All right, that's use number one. What's use number two?

Scott Clavenna: Two, you could actually sell that even before it's produced. You could just find another buyer in the market. So if you're sitting here in the U.S. and you're sitting on a million tokens, you could find a retailer in Spain and say, "Hey, I'll sell you these tokens at a little bit of a ... I'll sell them to you for three and a half cents." So you take half of the profits, they still want it at a discount, or maybe even at the current market rate and they're going to hold them long because they think power's going to continue to go up. You just have a ... You become a market trader.

Shayle Kann: Wait, that actually raises ... You said that prices are going to go up, that raises a question for me. If I own a token that is equivalent to say one kilowatt hour of generation. At any point in time, can I call that from any kilowatt hour produced when ... for the next 25 years of the project's operation or is it like assigned a time?

Scott Clavenna: Apparently ... It's assigned a time and it's a little ... I will say it, it's a little thin in here. It says, "Typically four to six months from when the project comes online."

Shayle Kann: But that doesn't make any ... I mean so the project is generating kilowatt hours for the next 25 years.

Scott Clavenna: Right.

Shayle Kann: They ... So I guess then they're assuming they're going to sell enough tokens to max out the number of kilowatt hours that they're willing to sell for the next four to six months and then they'll have another token sale six months later when the project generates a bunch more power?

Scott Clavenna: No, no see you've reached to one of those doors that when you get to it, it doesn't open. So it's possible that yeah, these are just rights to that much generation and you could hold it and sell it whenever you see the greatest spread between what your price and what the market price is, or you're assigned a particular time. That's not super clear in the white paper.

Shayle Kann: Yeah, and the other thing that scares me about this use case, the idea that you buy it for resale ... I mean selling it to a retailer, sure that part makes sense but this also seems like the use case that is like grown to speculation, which I think has been the case in a lot of these initial coin offerings, is just like I buy it because I think somebody else is willing to pay more. And I'm not sure why or who but there's going to be somebody out there and like that's the worst case scenario for all these auctions 'cause it's not like reflective of the intrinsic value of the asset.

Scott Clavenna: Yeah, yeah. Well ... Okay. So the third one is just a variation on that. The third option is you automatically sell the energy to the wholesale energy market once the energy is produced and you immediately settle. So that's just saying ... That's sort of you're giving the developer the benefit of the financing up front in exchange for the moment that power is generated, you're just going to take your profit. So it's just you set and forget it, and you're just an investor with a ... hopefully a relatively well known spread between what you paid and what the market will be.

Shayle Kann: Except wholesale prices are pretty volatile generally. I mean I don't know about Gibraltar specifically, but I'm thinking about the U.S. and like wholesale market. Wholesale are ... go all over the place and so if I don't control the time of the time stamp of when my kilowatt hour is generated that can have a huge impact on its value in the market.

Scott Clavenna: Yes. There's an interesting sentence in the white paper that goes, "This leaves an ambiguity of the final settlement price as it does at the market price in that specific moment."

Shayle Kann: Right. So like is my token ... I mean I'd love to be able to say like the clever version of that would be I want to buy the token that reflects January 22, 2019, at six PM. You know? And there's going to be ... like I can calculate ... I could forecast wholesale prices and figure out when I think it's going to have the biggest spread. But if I don't know, that just seems like a huge element of risk.

Stephen Lacey: Okay. So what is the application here? Is this a way to facilitate cross border energy transactions? I mean it sounds like a more distributed way to execute wholesale electricity trades but they're ... If in theory someone from a completely different country can ... an American investor can purchase the electrons from a Gibraltar based solar project, is this a way to facilitate those kinds of cross border energy transactions?

Scott Clavenna: Well it's ... I think it's more about the availability of capital for renewable energy projects, like increasing the availability. And what I find interesting is that if you build most of these projects in places where the IRR's are pretty high ... pretty reliably high, in the mediterranean, middle east, Africa but you're allowing investment from all over the world through this token platform, it does seem like ... In an ideal world, it does seem like a very efficient way to get a lot of capital to these projects from a lot of sources, in areas where the likelihood that the projects are in a good environment for high returns, that feels like a positive thing in general. Good terms for the investors 'cause they're in higher IRR regimes and good terms for the developers because they have access to not all the capital, but 10, 20 percent of the capital ... is coming from this outside source through a fairly well standardized, smart contract based, blockchain based market system.

Shayle Kann: Do you know when in the stage of a project's development cycle they plan to issue these auctions of the individual tokens? Is it like when the project's about to begin operation? Is it in the early development stage? Is it raising capital for construction finance? Do they say even?

Scott Clavenna: They put it back to when a plant is first planned, we will estimate its energy production pattern, then the green energy developer will be able to bundle up the desired amount of future generated power and set a price for it. So it seems fairly on in the project development.

Shayle Kann: Right. So just stepping back like as it stands today, when you develop a renewable energy project, they're sort of decreasing risk as time goes on and the project gets further developed that ... And the way that it works out today is you get like very risky capital early on that is for origination, which doesn't cost as much money but ... So you get riskier capital with lower dollars and then you get a further stage, you have to sign a PPA, you have to get an interconnection agreement, you have to secure your equipment and all that kinda stuff. And then you need to finance the construction, which is the big expensive and is less risky than origination, but more risky than what ends up being the long term owner who owns the asset once it's begun operation. At that point a lot of the risk is gone except for the sort of market risk and the risk that project fails or degrades faster. And this feels to me like you're getting ... you're selling electrons before the project has been developed, which puts you in the early very risky category. So what happens if you buy the tokens for project that fails, which happens ... tons of projects fail.

Scott Clavenna: So then it comes down to ... That's a really good question because the way it's described ... It describes it how they come about assigning the value to these tokens and actually creating the predicted share of the total investment that will be tokenized, but it doesn't say actually when they hold the auction. So the key moment in time is when you log into WePower, you see there's a project in Spain that they're holding an auction for next week, that's when you would want to know where ... That's when you need to do some diligence. You should hopefully do a lot of diligence, but it does feel like that moment is the auction held once they have all the ... the rest of the debt and equity lined up and the off take of everything, or are they holding that auction very early on saying, "This is a speculative project. We're looking for investors early and then we're going to add all the rest of the pieces ..." That's not clear from here either.

Stephen Lacey: We have another major rabbit hole to jump down here. Before we climb up this one and jump down that other one, I want to ask an even more basic question. Why? Why should a platform like this exist? What do you think the benefits are, Scott?

Scott Clavenna: If there are a lot of individual investors, a lot of people that might want to put 5, 10, 15, however much ... however many thousands into renewable energy projects 'cause they see it's a good return for them, and then they feel good about supporting renewable energy around the world. It does feel like in the grand scheme of things if that's what happens, is it unlocks a vast global appetite for investing in renewable energy, then that answers the why. If there's hundreds of millions of dollars kind of out there in aggregate, enough people willing to invest in renewable energy, then this is a great way to get it to those projects in theory.

Stephen Lacey: The Interchange is brought to you by Fiveworx. Times there a changing for utilities. In this digital age, the world expects more and in the utility space, that means beyond meter data. Not only are you being asked to better engage and service your customers, but to anticipate their changing expectations and preferences. So what does it mean to truly know your customers and can you leverage your data and the rising standards for customer engagement to benefit your business? With Fiveworx, absolutely you can. Fiveworx personalizes communication and drives customer action at scale through behavioral science, psychographic personas and machine learning. Enabling you to deliver the right message to the right person through the right channel at the right time. Go to fiveworx.com/theinterchange for more. That's Fiveworx with an X, fiveworx.com/theinterchange or follow the link on the podcast page to begin engaging the customer of one.

Shayle Kann: Okay. So onto our second complicated, confusing issue of the day, and so as I mentioned at the beginning, this one is not about an application of blockchain in the energy sector. It is about blockchain's impact on energy. So there's been a lot of press about this recently because the energy consumption of cryptocurrency mining and in particular bitcoin mining, has been growing basically exponentially for the past year as the price of bitcoin has been growing exponentially. And it's gotten to the point where currently by most estimates, bitcoin mining in aggregate represents as much energy consumption as a small country. It's about as much as Bulgaria that as far as I've seen recently and with their predictions, Morgan Stanley predicted that it's going to be as much as Argentina by the end of this year. And then there've been a bunch of really alarming forecasts that look forward and suggest that bitcoin mining is going to become ... in some scenarios could like dwarf electricity consumption from everything else in the world by a decade from now. But even the less alarming ones than that suggest that it's going to create more new load than electric vehicles, it's going to represent a medium size ... even a large size country. And then that creates a bunch of additional questions about where that electricity consumption would come from. Is it going to be clean, is it going to be new, is it going to be dirty and so on.

But ... And so I've been diving into this some and I want to talk about a few bits of it. The first one, just starting at the most basic level is why does bitcoin mining use so much energy and what is the relationship between mining and energy consumption? So Scott do you want to kinda kick us off there and then I'll jump in?

Scott Clavenna: Sure. Sure. That is not a simple question to answer, but I think fundamentally it requires a lot of energy because it requires a lot of computing power, because there is no trusted third party. There's no trusted intermediary in this very critical financial system that is cryptocurrency. And so if you've eliminated your trust in banks and brokers and anyone in charge of managing your balances and ledgers, then you ... Fundamentally, the way they came up with bitcoin was to say you're going to put your trust in math. It's distributed, everyone agrees that math is math, everyone trusts math. And so the idea is the way in which every transaction is validated is to run very computationally, intensive hashing mechanisms on that transaction, that then is validated by everyone else on that network as well. So everyone's participating in running this ... the cryptographic algorithms that have to run to make sure if I send you one bitcoin, my balance goes down by one, yours goes up by one and I can't mimic that onto anyone else's account, and it's a perfect sort of balance sheet bitcoin.

And to do that it does ... There's no authority other than this ... the cryptographic math that's going on and that math is hard and it also because as the value of bitcoin goes up, more and more ... Not the number of transactions, it's not ... Energy use isn't really tied to transaction volume, it's more tied to the value of bitcoin itself. So as the value goes up, it incentivizes ... it creates an economic incentive for more miners and these are people that would own computing processing resources to participate in bitcoin. It incentivizes them to devote those resources to mining because they're rewarded with bitcoins for doing that mining and as the value of bitcoin goes up, it allows them to invest in more computers and it kind of builds on itself. So whatever the market cap of bitcoin is, there's associated value of compute resources that miners invest in. So it has gone up quite a lot.

Shayle Kann: Can I attempt another version of that explanation? That is I'm going to try to make it as simple as possible. I was just trying to do this 'cause I wanted to write up an explanation of it for my newsletter and this is the simplest thing I've been able to come up with and tell me if and where I'm wrong. Okay. So the way that new bitcoins come into existence is through mining. The way that mining works is that a bunch of miners, which are computers, solve difficult math problems. Difficult math problems require a lot of computation, a lot of computation requires a lot of energy. The harder the math problem, the more energy you end up using to solve it. If you're a miner your goal is to be awarded the next block in the blockchain and the way that bitcoin works is that there is one block awarded every ten minutes. And so if you're the lucky miner that gets awarded a block, you get a fixed number of bitcoin in exchange. That currently sits at 12 and half bitcoin per block, so if I'm a miner and I happen to get awarded a block, I get 12 and a half bitcoin in exchange that currently ... Bitcoin is currently sitting over $10,000 per, so it's pretty lucrative.

But the bitcoin algorithm or I guess the bitcoin system is designed to ensure that a new block is created exactly every 10 minutes. So if a bunch of new miners pop up onto the network and start solving math problems, it doesn't mean that you get a bunch more blocks in the same amount of time. It means that the math problems basically get harder, so they adjust such that they end up awarding one every 10 minutes. So what ended up happening recently is that the price of bitcoin went way up. So the revenue you could generate from getting one block was increasing and as a result, we got a bunch of new miners, which meant then that the math problems adjusted to become more difficult, which meant that they had to use more computational resources to mine a block, which meant that they needed more energy. And so this is the sort of scary spiral that people have talked about, which is if the value of bitcoin keeps going up there is sort of nothing to stop this from continuing to infinity, which means that the energy consumption could grow by orders of magnitude. Did I get anything wrong there?

Scott Clavenna: No, that's pretty ... That's right. Yep. It works in reverse too, so if bitcoin price goes down, then there's less of incentive in ... people would ... there would be less energy consumed. But yeah, it's tied to that 10 minute figure and it's tied to the market cap or the value of bitcoin. Those are the two key figures.

Shayle Kann: Okay. So then two quick additional threads there. The first one is a ... You mentioned this and this is a really key point, which is that I presume then that all these projections about bitcoin energy consumption going through the roof, rely upon the assumption that bitcoin prices will continue to increase. Is that right?

Scott Clavenna: Yes. That is the key, key thing. There's often confusion around tying it to the number of transactions, but it's really not. It's tied very tightly to the value of bitcoin.

Shayle Kann: Right. Okay. So if you believe that bitcoin is currently in a bubble and the price is going to crash, then probably you shouldn't be that worried about energy consumption. The other thing that I guess makes ... should make you maybe slightly less concerned as I understand it, is that the other ... sort of bit of how the bitcoin system works, is that the number of bitcoin you get for mining a block falls over time. So it falls ... It basically gets cut in half every four years. So initially in the early days of bitcoin, when you mined a block you would get 50 bitcoin. Already we're down to 12 and a half bitcoin per block and that's going to get cut in half again every four years to infinity, which is how the system is designed to ensure that at the end of the day, only a certain number of bitcoin will be generated. And so that means that even if prices increase a little bit every year, as time goes on you're going to get less bitcoin for every block that you mine, which again is going to create a disincentive for you to add new miners and new resources to mining, and thus less energy consumption.

Scott Clavenna: Right. And I guess the one thing in addition to that is that we've seen in computing for decades, is that computers themselves get more energy efficient and faster every year. So there's not a one to one relationship between the algorithm and the energy.

Shayle Kann: Right. Okay. So like I can get ... So here's where I ... This is as far as I can extend. So I can get to the point where I now understand why bitcoin uses a lot of energy and I can understand the dynamics that would cause it to keep using more and more energy, which is in short bitcoin prices keep increasing, computing efficiency doesn't increase by a ton. And those two things alone probably do the trick and that things that would cause bitcoin energy consumption not to be a big deal, which is prices stabilizing or falling and/or computing resources or computing efficiency increasing. Now there's kinda separate conversation going on about changing the way that bitcoin mining works in order to reduce its energy consumption impacts, and this has to do with the current mechanism for mining, which is called proof of work versus this idea, which is called proof of stake. And as hard as I try, I have a really hard time grasping those two notions and how they are different from each other. Can you do it?

Stephen Lacey: I was in the car the other day and I tried to ... My wife Sandy asked me to repeat back to her what I thought both meant and she just like shook her head. She was like, "Nah, you don't really get it, do you?"

Shayle Kann: That's the worst thing about cryptocurrency, people. I'll just say every time we do a ... like a blockchain episode on this show, like somebody runs into me on the street and explains to me something that I got wrong on the episode. But they're usually like crypto-nerds and they always have this very disdainful attitude toward like my ignorance about the minutia about how crypto-mining works.

Stephen Lacey: Yeah.

Shayle Kann: Crypto mining works.

Stephen Lacey: Indeed. I mean I've seen Twitter wars over whether you call it a blockchain or the blockchain. There are lots of semantic battles over how you talk about this stuff. But anyway, this is really ... I mean it's complicated stuff. Scott, what is your ability to describe both of these concepts?

Scott Clavenna: Well. Okay. So proof of stake is much less well defined. I mean I ... One thing I think that's probably important is that people think that ... You could get the impression at least from articles that there's this proof of stake concept and it's very energy efficient and if only bitcoin would just adopt it, they would solve it. The problem with proof of stake is that it's not even well defined yet. It's not proven yet to be as secure as proof of work is, of running that ... those very computationally intensive validations on the bitcoin currency transactions. So the issue is it's still not well defined. Even people within the Ethereum community, anyone who's moving to proof of stake, that's probably where it would happen first, is in the Ethereum blockchain, which currently uses a kind of proof of work as its validation ... They could move to proof of stake but even there, there's lot of debate about whether it's mature enough, whether it actually is secure or what have you. But proof of stake, basically adds another ... it removes the computationally intensive part of bitcoin mining and it still has a lot of cryptography around it. But the stake ... the proof of stake is that it basically makes anyone who's a validator on the network, more or less put a very large deposit of currency down.

So they put a stake and so the idea is built into the ... built into that ... The proof of stake process then is mechanisms that you can validate transactions because you have a sizable stake of currency at stake as a validator, and if you do anything wrong or if you do anything that looks suspicious, you forfeit your stake. So there's a penalty associated with it ... a personal penalty associated with it, which is not the case in proof of work. Proof of work, you don't even have to own any bitcoin. You just have to devote computational power to it. So you're just sort of risking your own investment in that, where proof of stake, you're actually part of the sort of currency holders and a fairly large part, and so you stake it on that. And then there are some other layers around that, so that you can't just be big whale and then just corrupt the whole thing, because there's validation that is ... there is consensus still. It's not that you are a single validator, but you sort of drive it and ... Then I start to run out of the clarity because there is a lot of different proposals around it and Ethereum can't even agree on that yet.

Shayle Kann: The reason that proof of ... No, the current world is proof of work and bitcoin, and the people who want to move it to proof of stake because of the energy consumption benefits of that, want to do so because proof of stake does not create an incentive for new mines to just pop and use really high powered computers to get bitcoin. You could no longer do that in this ... in the proof of stake world. Instead, if you're already a holder of bitcoin, you could put up your bitcoin like ... Almost like put it up in escrow in exchange for mining. Is that right?

Scott Clavenna: Yep.

Shayle Kann: Okay. So is this not just a way for the rich to get richer, which seems like totally anathema to the all notion of bitcoin? In other words, is it not like saying in order for you to mine for gold, you have already have a lot of gold and put a bunch of that gold in a vault that you will lose if you create fake gold or something like that?

Scott Clavenna: The key distinction is the word mine. I think in the ... in where you would use proof of stake, is that all the currency's already been minted. So you're just getting transaction fees and you're sort of ... you're granted the rights of a validator because you have a big stake at risk. But that doesn't actually earn you any reward of ... You're not mining. You're not minting any ... You're not getting a reward of actual cryptocurrency by doing it. You're getting some transaction fees, but any ... others can do that as well. You're just a key validator on the network.

Shayle Kann: Ah. So it basically eliminates mining in other words. It's just a different way ... I mean the point of mining as it stands right now is to be able to devote computational resources to validating transactions. Like that's the whole point of having computers doing all this work in the first place. It's not just to mint bitcoin, that's like the way that they incentivize you to mine. But what the actual hard math that you're doing, like the point of that is to validate transactions, which is the whole thing with bitcoin.

Stephen Lacey: Well let's get back to the very premise of this conversation, and that is that bitcoin mining is going to be a global catastrophe because of exponential energy use if you believe that the price will continue to go up. Scott, I don't think that you're quite sold on that, are you?

Scott Clavenna: Not really. I think this has happened before, even in the internet. There was a lot of talk of ... with Facebook and Google, anyone who ran large data centers that were going to be very computationally intensive. Data centers that housed all the information on the internet, like Facebook and Google, anyone else, that they were having to move their data centers to Iceland or low cost electricity areas, and that is true. But technology tends to always get one step ahead of these issues, because there's such an economic incentive that ... Personally, I just think it's not going to be that hard to solve. In fact, almost anything in cryptocurrency or blockchain, when there's a ... For me, I have a bias toward reacting to any of these issues around scalability, around energy use, security to some extent, that if the problems are digital, I have a tendency to believe they are eminently solvable.

Shayle Kann: I guess I'm a little bit more concerned about it than it sounds like you are. I don't know that I'm necessarily am the fully alarmist, like this is going to double the pace of global warming side. But you used a good example of the data centers, which is real, data centers do use a ton of electricity. It happens to be that a lot the big companies that require a lot of data centers, they have commitments to renewable energy and so that I think has alleviated some people's concerns about the climate impact of data center energy usage. But were that not the case, it would be an issue and you can already see this with bitcoin mining, which is like these miners are ... many of them were in China. Now China has sort of clamped down on them and so they're moving to other places and that's displacing generation, that otherwise would have been produced for some other purpose and like ... The energy impacts can still be pretty big and thus can still have an impact ... a meaningful impact on climate change, even without it just growing exponentially to infinity.

Scott Clavenna: I agree with that. I think the thing is you could have ... I don't have a good link to it, but I know there were articles back in early 2000's that the internet was going to swamp all electricity in the world if it kept growing like it did. And so is bitcoin ... Is cryptocurrency going to be an electricity intensive part of the economy? That seems almost guaranteed, like anything else that moving to a digital infrastructure, then yes, I would say I'm there. But is it massive? I think people will innovate around that. I mean one thing we didn't talk about is like the Lightning Network, a bunch of ... just the ways in which you use transactions or the ways in which you use cryptocurrency. There's a number of proposals already and ideas floated out there where you only use the blockchain as like your final record of transactions, but you conduct all the rest of your transactions on sort of private channels outside of it. And then dump them back onto the blockchain just to like only use that intensive cryptographic hashing rarely, instead of for every single transaction. There's a bunch of proposals already floating out there this year.

Shayle Kann: You literally just opened a whole new door, to a whole new world I've never heard about and don't understand at all. It's a perfect way to close it up.

Stephen Lacey: Close it. Close that quick.

Scott Clavenna: All right. I'll give you two things. Look at Lightning Network and Raiden. R-a-i-d-e-n. Those two are worth looking at.

Shayle Kann: They sound cool. I'll tell you that.

Stephen Lacey: Okay. We'll try to address those in a future episode. As you can see, these doors continually open and we step through them and enter a whole new world, and I'm looking forward to exploring this more and more. I think there's going to be a lot of questions from audience out there, people in the energy business, investors, developers and crypto-currency nerds who are really interested in these blockchain energy applications.

Shayle Kann: Yeah. You know if we got anything wrong here, as I'm sure we did, tweet at Stephen, it's @Stphn_Lacey.

Stephen Lacey: If you noticed, I tried to hang back a little bit so that I wouldn't keep saying the wrong thing, get all the heat. Also, actually if anyone knows of a good study or is working on any numbers comparing existing fiat currency with cryptocurrencies or bitcoin mining itself, I'd love to see something. So you can send that to [email protected].

Shayle Kann: You're saying from an energy perspective, like how much-

Stephen Lacey: Yes. Sorry, from an energy consumption perspective.

Shayle Kann: Which is an interesting question, like how much energy do we actually consume in the literal printing of paper currency and the shipping thereof.

Stephen Lacey: Yeah. You got to cut down trees.

Scott Clavenna: Oh, you could go farther. Look at the all the people that work in the Visa headquarters.

Stephen Lacey: Right, right.

Scott Clavenna: The MasterCard ... Like literally, we could just ... This is disintermediating all of them. So the energy intensity of transactions is quite high when you factor all those people in.

Stephen Lacey: And another door opens. We're going to close it out there folks. Thanks to Scott and Shayle for a great conversation. Hopefully we all reach consensus and we're going to continue to try to reach consensus in coming episodes. In the mean time, again reach out to us at [email protected] if you want to provide commentary or suggest other concepts that you want us to tackle on this show. I'm Stephen Lacey with Shayle Kann and Scott Clavenna. This is The Interchange, weekly conversations on the global energy transformation. Thanks for joining us. We'll catch you next time.