Jake Brukhman of CoinFund and Tushar Jain of Multicoin Capital discuss a new trend among crypto funds: generalized mining, also called mining 2.0, in which investors participate in the networks in order to seed activity on them. For instance, an investor might offer disk space on a file storage network or provide capital on a decentralized lending network. They explain why it makes more sense to do this in the early days of a network than when it’s matured, whether this will lead away from a peer-to-peer vision for crypto toward more professionalization and how this affects the basic premise of the fat protocols thesis. We also discuss how this impacts how crypto funds hire, do their accounting and reporting and structure their LP agreements.
Thank you to our sponsors!
Abra: Click this special link for a free $25 in Bitcoin! https://www.abra.com/unchained
Altlending: https://altlending.com
WeTrust: WeTrust: Donate in crypto and have your donation matched by WeTrust through Giving Tuesday, November 27! http://wetrust.io/unchained
Episode links:
Jake Brukhman: https://twitter.com/jbrukh?lang=en
CoinFund: https://coinfund.io
Tushar Jain: https://twitter.com/TusharJain_
Multicoin Capital: https://multicoin.capital
Jake’s post on crypto borrowing: https://blog.coinfund.io/crypto-borrowing-and-staking-networks-e7d2d64a81a4
Tushar’s post on generalized mining: https://multicoin.capital/2018/10/23/the-evolving-role-of-crypto-investors/
Jake on LivePeer as a case study: https://blog.coinfund.io/livepeer-cryptoeconomics-as-a-case-study-of-active-participation-in-decentralized-networks-19a932415e0e
Unchained episode with MakerDAO: http://unchainedpodcast.co/why-its-so-hard-to-keep-stablecoins-stable
Unconfirmed episode with Rune Christensen of MakerDAO: http://unconfirmed.libsyn.com/rune-christensen-of-makerdao-on-its-15-million-from-andreessen-horowitz-ep039
Initial witness offerings, by Jake’s partner Alexsandr Bulkin: https://blog.coinfund.io/iwos-with-adapt-a-creative-technological-solution-to-a-regulatory-problem-513b0bc811ff
Jake’s post on fat protocols not being an investment thesis: https://blog.coinfund.io/fat-protocols-are-not-an-investment-thesis-17c8837c2734
Unchained interview with Joel Monegro, the author of the fat protocols thesis: http://unchainedpodcast.co/placeholders-joel-monegro-on-the-fat-protocols-thesis-today-ep65
Videos that Jake mentions at the end of the show:
Generalized Mining, An Introduction & Primer by Jake Brukhman, CEO of CoinFund: https://youtu.be/ceex9CN2YZU
Panel #1: Supply Side Services | Generalized Mining and The Third-Party Economy: https://youtu.be/Cr6H2FcidjY
Panel # 2: New Role of Crypto Investors | Generalized Mining and The Third-Party Economy: https://youtu.be/zakQc07GRXA
Panel #3: Staking Economic Design | Generalized Mining and The Third-Party Economy: https://youtu.be/ydViUpTZens
Transcript Below
Laura Shin:
Hi, everyone. Welcome to Unchained, your no-hype resource for all things crypto. I’m your host, Laura Shin. If you’ve been enjoying Unchained, pop onto iTunes to give us a top rating or review that helps other listeners find the show.
Here’s a pause for the ads.
The topic for today’s show is a new show that some people are calling generalized mining, others are calling it mining 2.0. Here to discuss are Jake Brukhman, founder, CEO, and managing director of CoinFund and Tushar Jain, managing partner of Multicoin Capital. Welcome, Jake and Tushar.
Jake Brukhman:
Hi, Laura. Thank you. Thank you for having me.
Tushar Jain:
Hi, Laura. Hi, Jake. Really excited to talk about this with you guys today. This is going to be a fun conversation. Thanks for having me on, Laura, big fan of the podcast.
Laura Shin:
Jake, since you haven’t been on the show before let’s start with you. What’s your background and what does CoinFund do?
Jake Brukhman:
Sure. My background is in mathematics and computer science. I spent about five years in the hedge fund world in New York on the financial technology side, I was then a technical product manager and engineer at Amazon for about two years working on ad tech products, and then CTO of a fintech startup called Triton Research where we did a lot of very interesting financial analysis on private technology companies. I learned about Bitcoin in 2011 and have been sort of in the space ever since, and when I saw the Ethereum white paper that really made a lot of sense to me, and CoinFund was formed around the concept of building portfolios to diversify digital assets. So CoinFund launched in July of 2014 as one of the first cryptoasset focused funds in the US.
Laura Shin:
And Tushar, although you were previously on the show with your partner, Kyle Samani, why don’t you briefly tell us how you got into crypto and what Multicoin does.
Tushar Jain:
Absolutely. So Multicoin is a thesis-driven cryptofund. I got into crypto actually through a similar story to Jake, where Ethereum was really what drew me in to the crypto world. I had seen Bitcoin before and Bitcoin was interesting to me, but the scope of Bitcoin was just so limited and Ethereum to me was a much more ambitious vision, and I saw new ways to coordinate human economic activity that weren’t possible before, and I knew that I had to be in this space once I fully understood the implications of what this technology could enable. So back in 2017, early 2017 my partner Kyle and I decided to launch Multicoin Capital and we went live last year.
Laura Shin:
A slew of crypto hedge funds launched in 2017 but I know several of them are now just falling themselves cryptofunds since they don’t really act like traditional hedge funds. Jake, I notice you’ve been using that phrase, cryptofund, as well. Why? What’s the job of being a partner at a cryptofund really like and how does that differ from managing a traditional hedge fund?
Jake Brukhman:
Yeah, thanks, Laura. I mean, I think that running a cryptofund is actually quite a bit different than running a hedge fund or running a VC fund. I think cryptofunds, you know, tend to be sort of structured and the vocation that one participates in when working at a cryptofund has to be tailored to the asset class, and you have a really interesting asset class here. You have an asset class of digital assets first and foremost, their liquidity profiles are quite different than what you would see in venture capital, private equity kind of markets, and more than anything else the technology around digital assets and things like Ethereum and Bitcoin and the centralized networks in general is pretty advanced and requires some probably pretty deep knowledge.
So, the approach that we have taken at CoinFund is that we’re a pretty multidisciplinary team. We bring together computer science, math, as well as financial experience and experience on Wall Street and investing to try to make sense of the space. And then when the fund manages assets I think it’s a bit of a hybrid between, you know, typically what people do in venture capital where they evaluate early stage teams and kind of make long term bets on the products that they’re building, and then, you know, also managing a portfolio of highly liquid assets that trade on 24 hour markets. So, I think cryptofunds kind of combine those two disciplines and really require a lot of different skills to make sense of the space.
Laura Shin:
And Tushar, how would you compare and contrast them?
Tushar Jain:
Yeah, I think Jake is completely right here. Cryptofund has caught on as a term because I think this is a new type of asset management firm. We are combining venture capital type diligence and venture capital type effort and work, you know, we’re working with the teams to help make them successful which is more akin to a VC firm to a traditional hedge fund. However, we do operate in public markets and we need to be able to leverage liquidity in order to earn the best returns possible, so we are also somewhat like a hedge fund in that regard. And then there are other aspects of what we do which are something brand new, which are very crypto-native which I think is what we’re going to talk about today, around how we can help the portfolio teams in this very blockchain-native or crypto-native way with these different tactics, and I think all of those combined together is what makes us a cryptofund and not just a VC fund or not just a hedge fund.
Laura Shin:
Yeah. That is the perfect segue to talking about generalized mining. Why don’t you guys describe for me all the ways in which cryptofunds can support decentralized projects and then explain what generalized mining is and how that fits in this larger universe?
Jake Brukhman:
You know, just stepping back for a second and just thinking about this concept of what does generalized mining mean, let me just say the following. So, I think over time what we’re seeing is more and more decentralized networks being built in the world, right? So, around 2015 the decentralized networks that we were very familiar with were cryptocurrency networks like Bitcoin, maybe asset issuance networks like Counterparty or NXT or BitShares, and then Ethereum came along and created a decentralized network which enabled Turing complete smart contracts and kind of subsumed the previous two.
The world that we live in in 2018 is a little bit more advanced, right? So, we see a proliferation of decentralized networks and it feels like they’re poised to proliferate further, and the applications of those networks are actually quite diverse, right? So today we not only see smart contract platforms and cryptocurrencies, we also see decentralized networks for storage, for computation, we see social media networks implemented as decentralized networks such as in the case of Steemit, we see different decentralized networks for various kinds of computational services like Livepeer which does video transcoding, we see financial services like lending protocols and exchange protocols.
The picture that begins to emerge is that in this world of very decentralized networks that have very specific applications and very specific domains, it turns out they’re not companies, they don’t typically have a management team. These are public networks and yet they need a variety of supply side kind of services to function, right? So, in Bitcoin those services are provided by miners and the service they’re providing is transaction validation, right? Same in Ethereum. When you go over to a social media network like Steemit you have people creating content and then you have other people or even bots curating that content. When you’re in a storage network you have people providing storage and that can be from their laptop, it could be from a data center. When you’re in some kind of decentralized finance network like a lending network then you need parties to come into that network and provide capital, and so on and so forth.
So, the number of roles in decentralized networks seems to be growing over time. We used to have miners and users, but then in proof of stake we have validators and users, and now a new kind of role called a delegator which could be a user or could be a speculator or investor or just someone interested in securing the network. You know, when you go to a network like Livepeer out of the gate you have five roles and that includes kind of miners, it includes transcoders, it might include software nodes that are orchestrators for the hardware nodes to support the scalability of that network and so on and so forth.
And so when we get to this concept of generalized mining the way that I think about it is there is a world of opportunities there in decentralized networks that would allow third parties to come in and provide these kinds of vital, you know, supply side services in exchange for compensation that’s allocated by the network, and those are financial opportunities, they’re opportunities to make a return, I think they’re slightly different return opportunities than if you were just investing in a blockchain startup or some kind of decentralized network token because they have different economics, and this space of generalized mining opportunities is also very diverse and different kind of players in the space fit better into different areas of that set of opportunities.
Tushar Jain:
Yeah, and to just dive a little bit deeper there, the way that I like to explain this generalized mining role is that we are investors in these crypto networks, these are early stage projects, and by being participants in these networks either on the supply or the demand side, and occasionally acting in ways that may not seem economically rational in a vacuum but are economically rational if you consider the entirety of the portfolio we are able to increase the chances that our portfolio company’s network effects actually catch on, right?
So, Jake talked a bit about Livepeer, and I think Livepeer is a fantastic example, right? With Livepeer as a decentralized video transcoding network it has a classic chicken or egg problem, and if we as investors can bring the chicken then that can help solve the chicken or egg problem, and that’s what I think at least for Multicoin is the intention behind generalized mining or this evolving role of crypto investors, is to use not only our capital but our technical knowhow and our other resources in order to accelerate the network effects of these projects that we’re invested in.
Jake Brukhman:
So, I agree with Tushar and in the context of this world it’s really interesting to think about what is the role of specifically funds kind of within the set of these opportunities. So, one thing that becomes really apparent is that a lot of staking networks today are coming out onto the market and they create an opportunity to help secure the network in exchange for sort of a return stream, right? And as a fund if you’re holding digital assets and there’s an opportunity to kind of make a return that’s there it’s a very compelling opportunity for funds and it also comes with some competitive pressure. So, if some fund adopts that strategy as a competitive way of making returns that might create some pressure on other funds to do the same.
There’s probably a lot of other reasons why a fund might want to participate here. If you’re a good cryptofund you should be invested, as Tushar alludes to, in the long term success of your portfolio companies, and if there’s a way that you as an investor can add value to the network that your portfolio company is building then that seems to be a compelling reason to engage that network, and I’ll give you a quick example. If you look at layer two networks, layer two payment networks kind of work on hubs today, right, and the capitalization of a hub also determines the throughput of that network. So, the more hubs you have the more capital is locked up in those hubs providing throughput kind of the more successful your network is poised to be because you’re helping to build the network effects of those payments, right? And if you as a fund who is invested in one of these projects can add that value that becomes incredibly valuable and measurable for the portfolio company.
Laura Shin:
Yeah. I think something that interests me is that earlier you were also mentioning that some of the other ways you guys can invest in these projects are through equity or investing in the tokens directly. So how do you compare this kind of service with those types of support?
Tushar Jain:
So, Laura, that’s a really interesting question. It’s something we think a lot about both from the investor perspective in terms of where should we allocate capital, and also from the perspective of how can we best support our investments. We believe that most of the value that’s created using this technology will be captured in tokens, not in the equity of corporations. There is a significant opportunity for these companies to emerge that build services on top of this technology. However, given how the value capture works and where the customer relationship really is, and we’re talking a little bit about aggregation theory here, we expect that almost all of the value ends up being captured in the protocol itself or the token of the protocol itself.
So this actually presents another opportunity for good crypto investors because in order to help the chances of the protocol being successful it can be a good decision to actually invest in a company that’s building on that protocol even if all things considered the protocol token itself presents a better risk-adjusted return than the equity of that company, and it would in many cases do so because if the company is successful it just drives most of the value back right into the protocol, but once again, we have a chicken or egg type problem, the protocol can only be successful if many companies build on top of it, so it can be worth it to fund those companies kind of like a VC would save capital for a follow-on investment in future rounds I think it makes sense for crypto investors to reserve capital in the same way to invest in companies or other projects within that ecosystem that help support their initial investment.
Laura Shin:
Yeah. I just realized that actually what this is is it’s investing in tokens and then putting work into the network, and it sort of tracks I guess traditional VC where, well, depending on which firm you’re talking about, they would invest and then do a lot of work to help that startup, but here it’s more measurable and it’s more technical. Is that a good way to think of it?
Tushar Jain:
Yeah. I think so.
Jake Brukhman:
I would say so.
Laura Shin:
Yeah, and actually, just a quick question I wanted to know, how did you come up with this term, generalized mining? I think some others call it mining 2.0. Are there any other terms?
Jake Brukhman:
Yeah. There is a bunch of people kind of discussing this idea. I think in the forefront of most people’s minds today are staking networks like Cosmos and Tezos, and so a lot of companies tend to think of this vocation as oh, we’re going to stake tokens and we’re going to make a return. To me the opportunity set is a little bit more broad than that, right?
So, you have some opportunities that are very hardware intensive, right? This is like, classical mining, this is providing storage, this is providing computation or building out data centers for various computational services. These opportunities are very hardware intensive and they need those economies of scale and they need that large capital.
I think in the middle of the opportunity space is a lot of these staking opportunities, right? This is where a new network comes up, the team provides some software that third parties can run such as node software, and then that node comprises the network, right, in exchange and compensating those third parties with digital assets from the network.
And I think on the other side of the spectrum you have things that are, you know, a little bit more interesting in the sense that they are a little bit more proprietary. So, these are things like market making algorithms, trading algorithms. These are things like maybe creating a machine learning algorithm to curate content on a social media network like Steemit and then performing better than others because your algorithm happens to be more advanced or something like that. Or if you think about the kind of services that are required by MakerDAO when they liquidate CDPs, right? They need third parties to sort of make an arbitrage decision about whether it would make sense to bid for a CDP and get it liquidated and then make a little bit of a return on that.
Laura Shin:
And so, what you’re telling me is it sounds like generalized mining is not limited to only proof of stake systems. You can have this with a proof of work system or any type of consensus algorithm and it includes not only hardware plays but also software. Is that a good summary?
Jake Brukhman:
Yeah. I definitely agree with that. You know, people have called this also supply-side services or network service provision, and generally I define this space as kind of any supply side service provided by a third party to a decentralized network in exchange for kind of compensation, usually but not always in digital assets, allocated by the network to that third party. So, this is most of the time a sort of on-protocol activity, it’s available globally because these networks are globally available and public, you know, and they present a variety of different return opportunities.
Laura Shin:
So then…because one question I was going to ask you is whether the goal was to accumulate more tokens or earn a return or simply grow the value of each token, but it sounds like it’s kind of all three.
Jake Brukhman:
Well, I think, you know, if you take a look at…let’s take staking as an example, right? If you take a look at most staking opportunities today this is an opportunity to take some tokens, to usually lock them up in some way in a decentralized network, there may be risks associated with that, there’s technology risk, there’s obsolescence risk, and of course there’s also sometimes risks on the protocol itself in terms of slashing, but in return what you get is a rate of return on your tokens in a token denominated way, right? So, you might make five percent a year in tokens, you might make about 20 percent a year. I think most of the opportunities there tend to fall into that range. Then you ask yourself as an investor, well, am I relying on the kind of token denominated return to say whether this was a good opportunity or am I really looking at the fiat-denominated return of my tokens? Right?
I think today still the fiat-denominated volatility of tokens is much, much higher than the five to 20 percent token denominated return of these opportunities, and so when someone goes into a staking opportunity, I think today what they’re saying is that we’re still betting on the success and growth of this network, this is still an investment decision for me, but by participating in this staking opportunity I’m actually increasing and maximizing my exposure to this investment opportunity that I have decided upon?
Laura Shin:
So then with the returns from generalized mining how do you think they would compare to either traditional VC or just passively investing in tokens?
Jake Brukhman:
Yeah. Well, again, so the staking opportunities, you know, they have a certain profile where they have a certain rate of return and maybe a certain set of risks. If you go over into kind of other opportunities, and one that comes to mind is you know, the Livepeer MerkleMine. The Livepeer MerkleMine was essentially a decentralized airdrop where software miners were incentivized to redeem an airdrop LPT, which is Livepeer Tokens, to 2.6 million Ethereum addresses, and when participating in that kind of opportunity what you’re doing as a miner is that you’re helping the Livepeer network in a very early stage of its life cycle. You’re basically helping the distribution of the token and in doing so you’re earning LPT so you’re taking a position in a token that really isn’t quite out on the market today, there’s very few venues where it trades today and there’s very little liquidity, and that’s fine because this is an early stage of the network, but the return opportunity there is similar to what you would get if you were an early stage investor in Livepeer, right? Like, in the sense that if the token goes 10x then you have acquired ownership of the token, you know, in a similar way to those early investors, but instead of participating in the round you have used technology to gain network ownership.
Laura Shin:
Interesting. Yeah, Tushar?
Tushar Jain:
So, I think these examples really help illustrate some of the generalized mining opportunities, but I think this can be expanded beyond that to not only include these opportunities where there are in-protocol rewards. So, something like the LivePeer MerkleMine, that was economically rational in a vacuum. Like, it was reasonable for someone to go and participate in that process regardless of what their opinion of Livepeer was or if they were trying to help support the network.
I really think of another example to illustrate where investors can use this generalized mining tactic as part of a broader strategy of supporting their investments, and the easy one is thinking of something like Filecoin where Filecoin, once again, has a chicken or egg problem upon launch, and there are some in-protocol ways to reward suppliers of storage on the network even if that storage isn’t being used, but it can be very economically rational for an investor who owns a large amount of Filecoin to go and bring a lot of storage to the network in the early days even when there isn’t enough demand for it to be utilized, even if provisioning that storage network at that point in time costs the investor money because if they can then help increase the chances that a new customer will come on to Livepeer and really kickstart those network effects, I think it’s worth it even if it’s not economically rational.
And so, I do think it’s worth differentiating between those two categories of generalized mining, one where there’s an in-protocol incentive and if you’re not doing it you’re leaving money on the table, if you’re not staking your Tezos you’re literally leaving money on the table that is yours by right. However, something like this Filecoin example isn’t something where you’re leaving money on the table, it’s where you’re going above and beyond to help your investment.
Jake Brukhman:
Yeah, and Tushar published a great post on this, you know, using that kind of loss leading example, and I think the other example you guys had, Tushar, was like, Maker CDPs, right? So, we could talk a little bit about that. Maybe to introduce the concept Maker CDPs…or I should say Dai doesn’t have any particular reason why third parties issue it, right? So if there is a lot of demand for Dai’s stable coin for whatever reason there’s kind of like, no immediate economic reason for people to lock up collateral at risk in a CDP in order to issue Dai, but in the same way that, as Tushar points out, a fund can kind of loss lead and participate in the network at an initial loss sort of looking at the long term view, you know, in that same way funds can go in and for example create more Dai supply to support that demand. Is that a correct assessment of that, Tushar?
Tushar Jain:
Absolutely. I actually think the Maker example is absolutely perfect because of exactly the dynamics that you’ve described, and an investment in the Maker system has a huge incentive to go and actually create more Dai, so they would have to lock up ether, they have to pay the interest rate to create Dai, so they’re going out of pocket to create this stable coin that they don’t actually need in that scenario, but if they then release that stable coin onto the market now there’s more Dai available and that makes it more likely that exchanges will offer trading pairs against Dai.
I mean, right now Dai has something like 60 million in market cap and Tether has something like two billion. It’s going to be really difficult for any exchanges to support meaningful trading pairs against Dai if there’s only 60 million dollars’ worth of Dai out there. There’s just not enough. So for investors in Maker who want to increase the probability that Dai becomes the collateralized, decentralized stable coin of choice it makes sense to go lock up 100 million dollars in ether or 200 million dollars in ether and go create more Dai, especially as multi-collateral Dai comes out and you can lock up different types of assets to create more Dai.
Laura Shin:
We’re going to discuss when it makes sense for funds to offer generalized mining services and more, but first I’d like to take a quick break for our fabulous sponsors.
Here is a pause for the ads.
I’m speaking with Jake Brukhman of CoinFund and Tushar Jain of Multicoin Capital. So, the example that you guys have been giving for the advantages that generalized mining provides tend to be around when projects are in their early stages. So, is that the only time that it makes sense, you know, when you want to get a portfolio project off the ground, or would you continue to offer generalized mining services even for mature projects?
Tushar Jain:
So, I think this is mostly helpful for early stage projects. The thing with mining is that it’s supposed to be a relatively permissionless activity if not completely permissionless in that anyone can do it, and that’s true of actual mining, like, you know, Bitcoin versus these generalized mining tactics that we’re talking about whether that’s supplying on Filecoin or creating Dai, etcetera. Anyone should be able to do them, and so what that means is over time as competitors enter, you know, because there is literally no barrier to entering, there should be no additional margins on that for the economically rational generalized mining, the ones where you are actually earning a profit.
The investments into protocols that are followed by supporting the protocol, that can exist longer into the protocol’s life cycle, but if at the medium stage or late stage maturity of the protocol it’s still reliant on investors’ good will to support the protocol and make sure that, you know, there is enough supply or demand on the network then the protocol has failed and the economics didn’t work out, and so it doesn’t make sense in that case either. So, I think in both the case where this generalized mining is driven by profit within that specific tactic or if generalized mining is driven by profit across the portfolio it really is only reasonable to do in the early days and maybe the middle days at the longest.
Jake Brukhman:
I agree with Tushar. I think most of the opportunities that we see today in generalized mining are kind of opportunities in early stages of networks, and that’s partially because most decentralized networks are in very early stages today. You know, I think as time goes on, we’ll see a lot more networks. I think we’ll see sort of less volatility kind of in the fiat-denominated sense of these tokens. I think we’re going to see a lot of opportunities like within decentralized finance to actually create portfolios and positions in these things that are less volatile.
I’ll give you a quick example, right? If you’re a staking company and you want to maybe build an initial stake of tokens to get into a network, like Livepeer might be an example, they have 15 validators, you need to cross a certain threshold in order to become an active validator on the network today, well, what you might do is you might go to a decentralized lending protocol that lends out LPT and you might borrow that, you know, borrow LPT from that protocol, use that to make a token-denominated return, and then return the borrowed LPT. So, what you’ve just done is you have created a kind of a way to neutralize your exposure to LPT and just sort of get that token-denominated return.
There are other strategies, for example periodic liquidation of those returns that further lowers the fiat-denominated volatility, and that might be an uncorrelated kind of source of returns for your fund.
You know, in the long term I really see people maybe building out different kinds of portfolios of these opportunities, and I think they also have different risk profiles, right? If you’re doing hardware mining, for example, like Bitcoin, you’re putting out a lot of capital and it’s like, relatively safe, you know, within the space of opportunities, and you’re making kind of a smaller return. And then on the other side of the spectrum you might be not putting in so much capital, you might be putting in some software into a network, and you’re betting on the success of that network, but the return could look more of a VC style return or something like that. And the combination of those things into a portfolio I think kind of creates a differentiator for funds, especially as you look at a lot of funds going long term, long on many of these opportunities or just signing equity deals or just signing SAFTs, you know. So, it’s an interesting sort of direction to explore in terms of making crypto portfolios more diversified and safe.
Laura Shin:
One other thing I wanted to ask about was Jake, your cofounder Aleksandr Bulkin proposed something called an initial witness offering, and when I read about it it seemed to me that it kind of worked in conjunction with generalized mining. Would you agree, and if so, describe how it works and how that connects with generalized mining?
Jake Brukhman:
I’m not an expert on the witness offering. There is a post by Alex there which you guys can read, but I think the motivation there is essentially trying to understand how a decentralized network can be sort of brought up and put into production using decentralized tools and in a way that tends to be, I guess like, more compatible with the regulatory environment that we have, and I think part of it is we’ve seen echoes of this throughout blockchain. So, you’ve seen fully mined assets for example like Bitcoin or Steem. Not a lot of people know that, actually. Steem never had an ICO, tokens have always been mined directly from the network, and the witness offering, you know, kind of takes a similar approach to bringing up the life cycle of these networks and getting them out.
And the way that I think it touches generalized mining is like a network that takes on the strategy creates, again, opportunities for third parties to come in to mine or to validate transactions and thus help to issue the asset.
Laura Shin:
Yeah, so one thing that I’m curious about is does this kind of take this vision of peer to peer crypto networks and mean that it’s just going to be more professionalized, where it’ll be more challenging for everyday people to contribute to these networks?
Jake Brukhman:
I mean, I think it’s actually quite the opposite, right? Like, the way the technology is heading you see better and better tools that enable kind of normal people to interact with networks. Like you see web 3 enabled browsers, that’s one thing, you see blockchain sort of networks that can use different kinds of technologies like zk-starks, for example, you know, to make the blockchain so small it can fit on your phone, and that actually democratizes the ability of normal people to just have a node that’s running with them at all times and creates a much more decentralized network than what you see in Bitcoin where kind of a handful of miners control the entire network. And also, hardware wallets, I think still in early stages here, but kind of looking forward a couple of years you can imagine waving your iPhone in front of some kind of decentralized application and being able to make a payment, so it becomes even easier that it is today with hardware wallets to interact with them. So overall, I think the trend is actually to put blockchains in the hands of more people, I would say.
Laura Shin:
Tushar, what do you think? And I wanted to ask, actually just to add one thing for context, I guess another reason I ask this is because your former research analyst Myles Snider left Multicoin to start an EOS block producer called Aurora EOS. So yeah, I just wanted to get a sense of like, how much of this is more professionalization versus like, actually furthering the decentralized model.
Tushar Jain:
I think it’s both. I think it’s more professional and I think I might disagree here slightly with Jake, that I don’t think that everyday people should be running a node. It just like, doesn’t add much value, and everyone having their own full node sounds more decentralized but it’s not because they’re just trusting the developers of the software, right? The centralization point just moves along that value stack somewhere else.
So, I think that having more professionalization actually leads to more decentralization because you have professionals who are qualified to be engaged in the network who can actually add real value to the network because they are working on the network all day, every day. I think the key for decentralization here is that those professionals are independent and are not all working for the same company or the same handful of companies, that they are distributed around the world and are independently making their own decisions, etcetera.
But I think have professionals and incentivizing professionals to add value to your network is the right path. Trying to win over the average layperson into adding value to your decentralized video transcoding network or your layer two Plasma chain solution is unlikely to me. I just think you’re going to have a lot of difficulty convincing them, educating them, and motivating them to do so. However, those opportunities can be extremely attractive for small professional organizations where you can start up something with one, two, maybe three people and be a generalized miner on a variety of these various networks and earn a decent return doing so, but that’s not necessarily amateur or regular consumer adoption, that’s really small professional team adoption.
Laura Shin:
And we did talk about how early this means that anybody who is participating in generalized mining will have more of an influence on the network simply because it is small. So over time how would you phase that out to make the network more decentralized?
Tushar Jain:
So, I think that’s a really interesting question, how do you phase out the influence of the early team and the early investors who own a large amount of the protocol? I think there are going to be a number of experiments done on this and we’re going to see the results. I can speculate, I have some ideas on how this might play out, but I don’t think this is proven yet. We haven’t seen many examples of projects go from, you know, the centralized launch to a decentralized governance or ownership of the project, which by the way, as an aside, I think is a good thing. I think in the early days you want to be centralized for agility, otherwise you will be outcompeted by a protocol that is a little more centralized that does then further decentralized.
I think some of the tactics that can be used to further decentralize a network are perhaps agreements by early investors or founders to actually sell some amount of their stake at some point. It’s not just a lockup but also something like the opposite of a lockup, saying if you’re an early investor in the project not only will you hold for at least the first one or two years, those are kind of normal lockups that we see, but also since you bought five percent of the total network, you know, you are a larger early stage investor, you also promise that after five years you will start liquidating your stake according to some schedule, and that way you can help decrease the ownership centralization of the network as it gets to the maturity stages.
You can also have commitments from the team ahead of time saying, like, we are going to work on this in this very centralized, agile manner for the first x months or x years, but after that we are going to transition governance over to this foundation that will hire other developers or fund other developers and other small teams in order to further the protocol.
So, through those types of mechanisms I think you will see the outsized influence of early insiders diminish as projects get more mature because it’s just necessary. It’ll be a good long-term selfish decision by those investors or by those founders to include these types of clauses that tie their hands because this then gives users of the network the confidence that, you know, there is a clear path towards decentralization.
Laura Shin:
Something else I was wondering about was for so long everybody was following the so-called fat protocols thesis where the theory was that value in this decentralized world would accrue at the protocol layer as opposed to the app layer, and then recently…I feel like…I mean, because everything is theoretical obviously as time goes on the thinking around these things shifts, and one good example of that is that USV, which had initially published the fat protocols thesis, they also recently published a new blog post about how it’s not only infrastructure that needs to be built at first in that the development of these things goes between infrastructure and apps because the fat protocols thesis led everybody to be like, I only want to invest in infrastructure. So, I’m just curious, how does now this theory about generalized mining, how does that fit in with the evolution of thinking about investing in decentralized networks?
Jake Brukhman:
Yeah. So, I actually wrote a post, Laura, called Fat Protocols Are Not an Investment Thesis where I tried to challenge a little bit this idea that you can use a blanket heuristic to see how value accrues in decentralized networks and tokens, right? And I think by and large it was an important post to sort of get people thinking along the lines of well, how does value actually accrue and where does it accrue and what is the mechanism by which it accrues, but I think because the design space of digital assets and the networks around them is so vast ultimately you need to examine opportunities case by case, and you know, in my post I kind of go through a couple of examples where, you know, here value may not accrue in the protocol token, or not completely be reflected in the value of the protocol token, and there’s a lot of counterexamples that you can draw sort of on an individualized basis to demonstrate that it’s not really, you know, quite so simple.
So kind of based on those ideas, right, you might think, yes, if I have a protocol that gets big and I own the token and the token accrues the value of that protocol that might be a good investment, but then you also have to consider the fact that this is not the times of TCP/IP where you essentially had TCP/IP and one other competitor. This is like, hundreds of competing protocols who are vying for the same or overlapping markets, right, and so the investment profile for protocol investing is actually kind of similar to sort of startup investing.
There’s certainly very interesting opportunities in middleware, right? When I say middleware, I mean kind of digital assets that are poised to be network agnostic, and where we’ve seen the most of this so far I would say is probably stable coins. You see a number of stable coins launching on Ethereum but also simultaneously launching their protocols on other networks like EOS, for example.
Finally, I think a number of VCs and investors in the space kind of over the last year have also come around to the idea that we wrote about of investing in DAPS, you know? A lot of people had said, well, you know, we’ve got to build the lower layers of the stack first and invest in DAPS later, but it’s also interesting to consider companies like Kick or YouNow who are kind of traditional companies with amazing teams with a lot of experience shipping software and hundreds of millions of users, right, and taking on some kind of blockchain strategy with a token and trying to accrue value in that token with an audience of people that they’ve already kind of converted into their products. I think that’s an interesting opportunity that a lot of investors have come around to.
And finally, there’s also opportunities outside of the decentralization stack that I think are really important to the space, and these are fundraising platforms and banking services and secondary market platforms that help get investment capital into the decentralization stack, that help to, you know, enable teams to operate and create great products, whereas it’s generally been pretty hard for teams who are fundraising in cryptocurrency, for example, to get bank accounts and things like that.
Generalized mining I think forces the person or the group that’s evaluating that opportunity to understand how value accrual works and it’s sort of motivating them to be a lot more specific than just kind of this fat protocol heuristic, and I think that’s a good thing, and in many ways I think it will actually support the idea that value accrues oftentimes in these sort of lower level assets but not always. So that’s sort of my take on it.
Laura Shin:
And I know you also have some interesting views on how this trend fits in with the shift toward scaling via layer two and state channels. Can you talk about that?
Jake Brukhman:
Sure. I mean, I think it’s going to be a while before we see, you know, a lot of layer two technologies out of production and actually supporting real world apps. There’s just a few examples of that today, one example is FunFair, another example is Bankchain that actually have layer two products in production, things like noncustodial hubs and payment channels. We’re sort of still waiting for generalized state channel frameworks and things like that, but I do think we’ll have some examples of that in the coming six to 12 months.
But as it relates to value accrual, you know, it’s interesting to think about, for example, if you compare Lightning Network above Bitcoin and maybe a generalized state channel network above Ethereum, right? So, in Bitcoin you’re going to be locking up some collateral on Layer 1 so that you can transact in layer two, and sort of the amount that you can transact is directly proportional to the amount that you lock up, and you’re actually taking units out of circulation. It’s interesting, how does that dynamic impact kind of the Bitcoin layer underneath? Well, it seems like you lower the circulating supply so that’s good for price, but at the same time you’re collapsing a lot of transactions that maybe would have been on chain into just two, right, into the transaction that opens the payment channel and the transaction that closes it. So, does that result in a decrease in transactions in that network or does the growth achieved by having a layer two payment system above Bitcoin actually outpace that? It’s an interesting question.
Now when you go onto generalized state channel networks on Ethereum then you realize that, you know, because you’re actually instantiating smart contracts in the state channels off the main chain you’re probably implementing some kind of business model using those contracts, right? So if you’re like, playing chess games for money and the provider is taking a fee or something like that, it turns out you’re taking away all of those transactions from the Layer 1 and you’re potentially, you know, not sharing the same amount of value that accrues in these off-chain applications with the base layer, and the question really becomes on a case by case basis, how does this impact, let’s say the price of ether underneath, if that makes sense.
Laura Shin:
And so generalized mining sort of solves a certain problem in the sense that you can still accrue value from the scaling of the network even if that value isn’t necessarily accruing at Layer 1?
Jake Brukhman:
Well, I think in the case of layer two specifically, right, if you sort of look at the landscape of what’s happening let’s say in state channel networks today, you have in the order of 15 or 20 projects that are working on this stuff, maybe like half or most of them are open source so you’re not really getting a lot of equity opportunities to invest in those teams, there are some. Out of those 15 or 16 projects only three of them have tokens, so if you’re taking a really sort of VC lens when you’re approaching investment you’re not going to find a lot of equity there, and you’re probably not going to find a lot of tokens in those networks. So, if a layer two network takes off and that network is really valuable but that network doesn’t have a token the question is, how did you get the exposure to that growth?
And sort of I think the best answer to that in today’s environment is, well, it makes sense to run, you know, some kind of infrastructure on this network, right? So, if you’re running a hub directly on the network then you can capture that growth by providing network effect, providing the ability of many different people to transact through your hub, and so of course you’re taking a transaction fee on that, and you’re making money. But if you’re sort of equity focused or token focused I think you’ll find it a lot harder to get that direct exposure to that growth, and plus you’ll have to contend with the fact that there are actually many competing payment networks in that layer.
Now a lot of founders who are building these things are actually saying, you know what? In order to monetize this for myself I’m going to build some applications on top and those applications are going to have token investment opportunities, or they’re going to have equity investment opportunities, and that’s great. I think it’s just going to come down a little bit down the line from now, and when you’re evaluating let’s say an equity opportunity in, I guess you would call that Layer 3 or something like that, right, you not only have to evaluate the application that’s being built on its own merits, but you also have to ask yourself, am I making the right underlying network investment decision? Because if I invest in this application on this network but then some other network becomes the dominant network, did I make the best investment there?
So it becomes a lot trickier to reason about tokens and equity in layer two rather than just saying hey, maybe I’ll just run five different kinds of hubs, watchers, or dispute resolvers on layer two and then try to get exposure that way and try to sort of diversify and make bets that maybe a few of these networks will become really big later.
Tushar Jain:
Jake, I think you said something really interesting earlier where you were talking about how some of these layer two solutions can go across chains. So I think that also presents another interesting opportunity for investors who are looking to invest in things other than the base layer protocol, is I think this lowers the risk, where if your product is really compelling and you built it on Ethereum for now but as it turns out one of these next gen, like smart contract platforms ends up actually fulfilling the vision and becoming the dominant smart contract platform, well, if you can transfer over the state and the code base of whatever the product is, whatever the DAP or smart contract is onto this other platform, well, then you’re not really exposed to the risk as either a founder or an investor of the underlying platform.
So, I think that’s another opportunity where the risk return calculation needs to be reexamined from first principles rather than necessarily, you know, looking at this the same way that it has been looked at in the past.
Jake Brukhman:
Yes. Exactly. Thank you for bringing that up, and you know, one way to just characterize that is sort of products that have interoperability with their base layers, right? So, these might be interoperability platforms like Polkadot or in some sense like what Cosmos is building. It might be what Chris Burniske likes to call middleware protocols, right, so these might be like stable coin protocols that live kind of in the middle of the stack.
The point here is precisely that, right, these kinds of interoperable mechanisms live higher in the stack than the base layers, right, and that’s another reason why you can imagine a bet on sort of the middle of the stack being a little bit safer than bets on fat protocols all the way at the bottom of the stack, if that makes sense, because if you have an interoperability platform that has 100 networks interoperating on it but then one of those networks fails that’s not going to be, you know, as bad as if you were all in on that one network that failed, right?
Tushar Jain:
Absolutely.
Jake Brukhman:
Create diversification.
Tushar Jain:
I think the diversification is really interesting here, but then if you contrast that with the fat protocols thesis, which I’ll stand up and say, I think I’m one of the few people who is willing to still defend the fat protocols thesis. I think it’s actually correct. I think that the vast majority of the value that is created through this blockchain ecosystem is going to be captured in some base layer protocols. I don’t know exactly how many, but my guess is it’s going to be a power law type distribution, and the largest one will be a smart contract platform just for security model reasons, right? If you have something built on top of this platform that’s more valuable than the platform, well, now it can justify an attacker to go and attack the base layer platform in order to get value on that layer two solution or whatever was built on stop of the platform.
So I think the base layer protocols actually present a more concentrated, riskier bet that has a potentially much higher routine than these middleware type solutions which can skate across protocols if necessary, they can interoperate across protocols if necessary, but they don’t have the same level of potential to, you know, have the huge mega outcomes that a base layer protocol does.
Laura Shin:
Yeah, that actually goes to that question I asked early on when I was asking you guys to compare this against things like just investing in the tokens directly, because just thinking, I agree with you that I do think it seems like the ROI here might be less, however the way in which it’s not less is that because it will help the network grow then perhaps the value of the token would grow or would ensure that the investment that you made in the token would succeed, something like that, or help ensure.
We’re actually running out of time, so I just want to ask you a few last quick questions. When VCs begin investing in tokens a lot of them had to change their LP agreements. Do you guys need to do that at all in order to make this work?
Tushar Jain:
Not for us. We are a crypto-native fund, so we launched with our LPA being amenable to us owning tokens.
Laura Shin:
And Jake, did you guys?
Jake Brukhman:
Yeah. So again, when we were kind of bringing up CoinFund we made sure that our agreement was sort of in balance with the kind of investments that we would make, and so in general I think we’re very, very flexible in how we can invest. We can make equity investments, we can sign SAFTs, or SAFE notes, or convertible notes as well as invest in tokens. As of this year we have some additions to our corporate structure to allow us to engage networks directly as well.
Laura Shin:
And so, with both of you now getting into generalized mining as a part of your strategy, how does that affect the way that you hire or other sorts of administrative or technical or strategic decisions you guys make as a fund?
Jake Brukhman:
From our point of view, you know, we have a full-time principle engineer named Jason on our staff who is kind of in charge of looking across these sorts of opportunities. Both me and my cofounder, Alex Bulkin, we are technologists and engineers, and so we participate in that process. Generally I think this creates kind of an onus on funds to be a little bit more technologically enabled, and so I think more technical hires are probably key to this, and then in addition to just participating in the opportunities there’s also a bunch of accounting considerations and tax considerations and working through sort of the reporting of what these activities that the fund has been engaged in are, and that creates, you know, additional sort of motivations for hires.
Laura Shin:
I would imagine actually maybe from a tax perspective it’s more advantageous because at least in the case where you’re sort of earning the tokens then…or not earning, but…no. Maybe it only applies if you are mining them and they’re being minted at that moment, because I was just thinking about how if you purchase them and then you later sell them then there’s two tax events, but at least in the case of mining I think where they’re being minted at that moment there’s only one tax event which happens when you sell, so can you just talk a little bit more about the differences in taxes here?
Tushar Jain:
Yeah. I think generalized mining actually introduces more complexities on taxes because of UBTI issues, which some funds and LPs have some restrictions around.
Laura Shin:
And what is UBTI?
Tushar Jain:
UBTI is unrelated taxable business income, so if you are a tax-exempt investor whether that’s some sort of foundation or trust or some other type of setup, and I am not a tax attorney, so to all the listeners, please consult your tax attorney
Laura Shin:
None of us are.
Tushar Jain:
But there are some restrictions around earning effectively operational income from an investment, and for tax-exempt investors that don’t pay capital gains taxes earning UBTI can actually pierce that and cause them to have to pay taxes. This is more relevant to the layperson through investing in their IRA where you have some restriction on the amount of UBTI you’re allowed to have in your IRA before you have to file tons more tax paperwork.
So, I think generalized mining adds some more complexity there, but at the end of the day I think it’s worth it, and going back to my earlier point, I think generalized mining is best executed by small professional organizations and not necessarily by laypeople. That means that the resources to actually account for all of this in compliance with all the relevant regulatory and tax law is actually much more plausible than someone who is a hobbyist going and doing this on their own, where it may not be really economically rational for them to do so once they take the accounting and tax prep work into account.
Laura Shin:
Interesting. Well, I guess we’ll see how all this plays out. Thank you, guys, both so much for coming on the show to discuss. We could keep going, there’s just so much material here, but we’re running out of time. So where can people learn more about you and your companies?
Jake Brukhman:
For us we have a website, coinfund.io, you can check out that and it has a description of our team and sort of broader investment thesis. On the generalized mining stuff, we actually just ran a very successful meetup over at DevCon in Prague where Multicoin Capital also participated and we’ll be releasing the video of that shortly and hopefully that gives some more material to this discussion.
Laura Shin:
Great. Tushar?
Tushar Jain:
And you can find out more about Multicoin Capital on our website as well, it’s multicoin.capital, and we have a good number of posts on there, and actually there’s one that Jake referred to earlier that I released a couple of weeks ago called The Evolving Role of Crypto Investors which specifically covers some of the topics that we’ve discussed today on this podcast, and I think if you’re still interested and want to dive deeper we will be releasing videos from the Multicoin summit where we had a panel called The Evolving Role of Crypto investors where Jake’s partner, Alex, also participated. So, there’s a lot of interesting content there to dig deeper and learn more about this topic.
Laura Shin:
Great. I will link to all this in the show notes. Thanks, you guys, for coming on Unchained.
Jake Brukhman:
Thank you, Laura, for having us.
Tushar Jain:
Thank you, Laura.
Laura Shin:
Thanks so much for joining us today. To learn more about Jake and Tushar check out the show notes inside your podcast player.
New episodes of Unchained come out every Tuesday. If you haven’t already, rate, review, and subscribe on Apple podcasts. If you like this episode, share it with your friends on Facebook, Twitter, or LinkedIn, and if you’re not yet subscribed to my other podcast, Unconfirmed, I highly recommend you check it out and subscribe now. Unchained is produced by me, Laura Shin, with help from Raelene Gullapalli, Fractal Recording, Jennie Josephson, and Daniel Nuss. Thanks for listening.
