In an intermittent series on cryptoeconomics, Olaf Carlson-Wee and Ryan Zurrer of crypto hedge fund Polychain Capital describe what cryptoeconomics is, what goals it typically helps networks accomplish and what behaviors token systems might someday incentivize. We discuss when cryptoeconomic models don’t make sense, how the type of consensus algorithms a blockchain chooses can affect behavior in that system and which consensus mechanisms excite them now. We also dive into whether or not it’s desirable for a cryptoeconomic system to depend on a small number of knowledgeable participants, how to manage on-chain governance so networks don’t vote themselves into a “black hole,” and what disciplines are helpful in designing smart cryptoeconomic systems.

Polychain Capital http://polychain.capital/

Thank you to our sponsors:

Preciate: Recognize someone with a shout out: https://preciate.org/recognize/

Bitwise Asset Management: https://www.bitwiseinvestments.com/unchained

Keepkey: https://www.keepkey.com/

Transcript:

Laura Shin:
Hi everyone and welcome to Unchained. The podcasts where we hear from innovators, pioneers and thought leaders in the world of blockchain and cryptocurrency. I’m your host, Laura Shin. If you’ve been enjoying Unchained, pop into Itunes to give us a top rating or review. That helps other listeners find the show. And, be sure to follow me on Twitter @Laurashin. Unchained is sponsored by Preciate. Founded by Ed Stevens, Preciate is building the most valuable relationships on earth. In each episode of Unchained, Preciate sponsors the recognition of an individual or group in crypto for an achievement. Who in crypto will be recognized today? Stay tuned to find out.

Bitwise: 00:00:39
This episode of Unchained is brought to you by Bitwise Asset Management. Last year, Bitwise created the world’s first cryptocurrency index fund. The Bitwise HOLD 10, which holds the top 10 crypto currencies and rebalances monthly. The fund has several hundred LPs and is currently accepting accredited investors. To learn more and invest in the Bitwise cryptocurrency index fund, visit www.bitwiseinvestments.com/unchained.

KeepKey: 00:01:04
Today’s episode is brought to you by KeepKey, the easy, safe, and simple way to protect your bitcoin, ether, litecoin and many other digital assets. There’s no time like the present to protect yourself from hackers, malware, and viruses. Rest easy knowing that your digital assets are protected. Visit keepkey.com to order your secure hardware wallet today.

Laura Shin: 00:01:25
Today’s topic is cryptoeconomics, which is easily one of my favorite things to think about in this space. Here to discuss this topic are Olaf Carlson-Wee, and Ryan Zurrer of Polychain Capital. Welcome Olaf and Ryan.

Olaf Carlson-Wee: 00:01:36
Hey Laura, thanks for having us.

Ryan Zurrer: 00:01:38
Thanks for having us on and to talk about my favorite topic in the world, cryptoeconomics. Super excited for this jam.

Laura Shin: 00:01:46
Thanks Ryan. I’m glad that you’re excited. Let’s start with the basics. What is cryptoeconomics?

Ryan Zurrer: 00:01:54
So for me, cryptoeconomics is the field that studies using tokenized representations of digital scarcity to incentivize a distributed network of actors, and these actors usually contribute some valuable resource to a network. And, self organize in a specific way and then are remunerated for either the contribution of these resources or pay for a specific resource on a network.

Olaf Carlson-Wee: 00:02:28
And the simplest and first version of a cryptoeconomic model is really the block reward in Bitcoin, leading to a very secure network for the users of Bitcoin.

Ryan Zurrer: 00:02:37
And, that’s a really a key point because cryptoeconomics typically either drives security on a decentralized network or accelerates network effects. And, what we look for in cryptoeconomics is typically one of those two things or a combination of them.

Laura Shin: 00:02:54
Yeah. Actually… well that was going to be my next question is what behaviors you are trying to incentivize or what problems you’re trying to solve? But, are those really the only two or are there other ones?

Ryan Zurrer: 00:03:04
So there’s a number of different resources that could be exchanged on an open source network. So, it could be, trust. It could also be say computation or storage or basically anything else that we can bring a distributed group of network actors together to exchange and in some sort of value. And, one of the things that we like to talk about here at the Polychain office quite a bit is the resources on these decentralized networks that will be exchanged. There are a number of different ones that we haven’t even thought of yet and we’re really excited about what could come up next. So it’s definitely not just security and network effects among users. It could be a range of different things.

Olaf Carlson-Wee: 00:03:49
Yeah, I think to date, securing blockchains has been by far the number one usecase of cryptoeconomic models. I think, looking into the future though, you could imagine for example, the contribution of intelligence leading to sort of block reward rather than a kind of proof of work or proof of stake style, block reward. And, instead of securing a blockchain, you could grow a pool of data say, or a pool of people that are curating content and you could gather a group of people that are curating content in an intelligent manner. So I think that going forward we’ll actually see many more experimental types of cryptoeconomic models that accomplish many different goals outside of just security of networks.

Ryan Zurrer: 00:04:40
And then, really we’re also seeing the emergence of sort of crypto commodities, right? And again, this could be computation like, which is represented by gas on the Ethereum network. It could be storage, for example, with IPFS, Filecoin, and any sort of important online resource you could essentially drive down into a cryptoeconomic model and achieve greater levels of market efficiency for trading those resources. This could be bandwidth, it could be energy one day, it could be, essentially anything. Effectively where today we have a network and then on top of that network, either an intermediary or maybe a very inefficient market. That will be disrupted and will be made into a protocol that operates on a perfectly competitive and highly efficient market.

Laura Shin: 00:05:38
Yeah, that’s something I wanted to ask about. What’s an example of where a cryptoeconomics model make sense or where one doesn’t make sense? And to my mind, I’ve been thinking a lot about how tokens or cryptoeconomic models only makes sense in networks that are actually displacing middlemen. So you can either have kind of like a centralized company, like let’s say a Facebook. Or, you can have a decentralized or distributed social network that’s powered by tokens. But, that’s why, and I sort of hesitate to name names, but there are some even really huge coins in the marketplace that aim to somehow maintain middlemen and yet still have coins. And, those don’t make sense to me. Am I just missing something?

Olaf Carlson-Wee: 00:06:25
I think in general, peer to peer tokenized models are incompatible with centralized revenue extraction. I just think it’s too easy for an open source developer anywhere in the world to really fork the network and remove the revenue extraction mechanism. You could, for example, imagine an alternative Bitcoin where a company or the founder of Bitcoin, really tried to charge 1% transaction fee for every transaction on the Bitcoin network. But, this would be trivially easy for anyone to sort of route around and not abide by that rule. So I do think that in general, combining a for profit company that has revenue extraction and a peer to peer token model often doesn’t make sense.

Ryan Zurrer: 00:07:13
The only reason why some of these, what we call rent seeking tokens have not been forked out of their networks is because there just isn’t enough sort of resources or developers building in this space right now, but I think in relatively short order you will see some of these, not very sensible token models that are effectively rent seeking on top of their networks, be forked out and that network be offered free of that rent seeking.

Laura Shin: 00:07:43
Yeah. So we’ll see. Hopefully as this space develops, we’ll see some of these tokens that I think don’t make sense, get shaken out and hopefully the people who’ve bought them won’t lose too much money. So let’s talk about actually these basic problems then. Let’s start with security. How does cryptoeconomic solve for security? And by that I mean, why don’t you describe how proof of work, works and some of the other consensus algorithms.

Olaf Carlson-Wee: 00:08:09
I think in a proof of work system, what is happening is you’re incentivizing rational, for-profit actors who are really not behaving altruistically in any manner. Who are just trying to receive block awards for hashing. In the case of proof of work. Those actors though who are performing those hashes are actually effectively adding security to the blockchain so that it becomes harder and harder, the larger those block rewards get. As the value of the underlying coin increases, it becomes harder and harder to attack that network. For an adversary to kind of attack that network and alter that ledger. So just by virtue of pure rational selfishness, these miners are actually, in a sense, providing a massive service to the users of a proof of work network. Now that is being paid for in a sense by the users through inflation, right? So it’s not to say that it’s free, but the users of Bitcoin effectively pay the inflation rate to these block validators. And, in return, can be sure that their transactions aren’t going to be reversed at a later time.

Laura Shin: 00:09:23
And you and I have actually talked about this before, but I just want to point out how you keep saying that they’re not doing this altruistically. And, we’ve spoken before about how we have had decentralized or peer to peer networks before Bitcoin, but those were ones where people operated them altruistically just out of the goodness of their heart where they put files on, Napster or on the Tor network or whatever, but they weren’t being compensated. And so this really was a breakthrough where you don’t need for people to just decide to make this charitable action. And instead, they can be compensated just the way that you would if you were doing a job.

Ryan Zurrer: 00:10:11
Well, I think the really interesting thing here is that when you add in incentivization, and this is sort of the “Aha” moment of cryptoeconomics. When you can add in incentivization, you not only drive network effects on these decentralized networks such that the amount of people willing to participate goes up quite expressively. You accelerate these network effects like rocket fuel. And, this is really one of the great things about cryptoeconomics is that using these digital incentives, we can create a certain mass very quickly that are willing to contribute, say computation or some kind of resource to a network and you don’t just rely on altruism. But then, you can also program in further incentives such that again, security is delivered so that people won’t say, try to include transactions that are a double spend or some kind of transaction that shouldn’t be there. You force people to sort of act but act honestly, because of their own sort of economic self interest in doing so.

Olaf Carlson-Wee: 00:11:22
I would also say that it depends on the level of adversary you’re dealing with. So for example, the Tor network which you mentioned Laura is a very, breakthrough and innovative anonymity network. However, I would argue that it is insecure against nation state actors who are trying to decode kind of who’s using that network and actually connect endpoints despite the fact that packets are being encrypted and routed. So to me, if you were to build a sort of tokenized Tor, so to speak, which wouldn’t be that dissimilar from say the Bitcoin network where users could effectively pay to use the Tor network in the form of inflation that would go towards the nodes in the Tor network. I would argue that if design successfully and like for example, the project, the Orchid protocol is working on this. That you’d actually get a much, much more secure, version of the Tor network. And, I don’t mean like a little bit more secure. I mean like a hundred or a thousand times more secure because people are now doing this in a kind of professional, industrial scale and for profit. And this would replace all of the current nodes in the Tor network, which are a more kind of academic or, experimental and they aren’t really being done at scale and it doesn’t provide sufficient defense against a really sophisticated adversary.

Ryan Zurrer: 00:12:54
Not only greater security but all to greater performance than what we see today on Tor.

Laura Shin: 00:12:59
Interesting. Just out of curiosity, did you guys invest in Orchid?

Olaf Carlson-Wee: 00:12:59
Yes, we did.

Laura Shin: 00:13:08
OK. Because, I have heard quite a lot of criticism also about that project. I don’t want to get off into that tangent right now. Maybe I should explore that in a future episode, but I have not explored it in any depth. But I definitely know that there are some questions being raised about the viability of that. But, I do agree just like on an abstract level that of course when you have direct economic incentives that people are going to be much more motivated. And there’s just going to be a lot more activity and I think this goes to an earlier point that Ryan was making about the speed with which these networks can take off. I mean we’ve seen it time and again. If you look at the DAO. If you look at the way ICOs took off. If you look at CryptoKitties. What’s so crazy to me about this whole space and Olaf you made this point in your speech at Consensus last year, is that, I do think that because the incentives are much more direct than you would find in most other systems that we partake in the world, that, that is part of the reason why we’re seeing this level of speed for a lot of these projects. So to kind of keep going in this vein around security and talking about proof of work. So I’m curious to know what you guys think are the cryptoeconomic differences between proof of work, proof of stake, delegated proof of stake and any other consensus algorithms you find interesting? And as you describe them, if you could sort of define some of the new terms for the listener, that’d be great.

Olaf Carlson-Wee: 00:14:36
Yeah. So one of the major breakthrough consensus mechanisms that we’re very excited about is the Dfinity consensus mechanism. Which is called Threshold Relay. And so, Threshold Relay uses something called a BLS signature, which is a cryptographic signature that was developed at Stanford. And the Thresholds Relay in simple terms, allows randomness to be natively generated within the block creation of the protocol. And this randomness allows you to randomly select the next block validator set from a massive potential validator pool. So you have a pool of call it 500 potential validators and each block is validated by a small subset of those. And it also selects what the next subset of validators will be. In short, this model allows for very, very efficient, block times. So in the Dfinity system, blocks take about one second to be created. And because this is not probabilistic, the way proof of work validation is probabilistic in nature. You have finality in just two blocks. So the equivalent, kind of finality in Bitcoin, we would take an hour. In Ethereum you’re looking at 17 second blocks and maybe about 30 blocks for that level of finality. Whereas in Dfinity, this is just two seconds. So this is the type of improvement in the underlying consensus mechanism that’s actually been enabled by cryptography that was literally invented since the creation of Bitcoin.

Laura Shin: 00:16:20
Actually, I want to ask about that. So, I understand something like proof of stake or proof of work where the probability that you’ll get the next block reward is based on how much you’ve put in, whether it’s hash power or coins, but in this case where it’s random, is there some way to, kind of improve your chances of being the one to add the next block?

Olaf Carlson-Wee: 00:16:46
That’s the brilliance of the design is there is in fact no way to kind of game that system. Again, this is all experimental. I don’t wanna make any guarantees about anything that hasn’t been live and tested for a few years. But at this stage, it appears that this kind of randomness is kind of true randomness and cannot be gamed by that validator set.

Laura Shin: 00:17:11
But then does that remove the incentive in some fashion? Do you know what I mean?

Olaf Carlson-Wee: 00:17:19
Well, so as a validator though, your profit is still quite predictable. It just, based on probabilities. Over a long time horizon, it becomes very stable. Obviously it’s random on a block by block basis, but over an entire day, each of let’s say 500 validators should receive a pretty equivalent number of blocks.

Ryan Zurrer: 00:17:39
And then there’s sort of a second part to this with validation towers such that the subset that proposes the transactions in a specific block are randomly selected. But then the validation through kind of the validation tower height, which is based on who owns what, looks somewhat like the payout there would be somewhat more akin to who owns what is similar in proof of stake. So just to review them, proof of stake, effectively your level of confidence and say the transactions proposed by a given miner, is proportional to the amount that the miner has at stake if they’re proven to be wrong and thus are slashed. So basically, it’s just an economic Nash equilibrium where you’re not going to try to say, stick in faulty transactions or double spends in an economic system that you are highly committed to and thus your stake is sort of the level of confidence, that the network has in what a miner is proposing. That needs to be then paired with a sort of a challenge period where others could challenge in case somebody with a high amount of stake were to still propose blocks or transactions that are double spend or, we’re not supposed to be on a blockchain.

Laura Shin: 00:19:09
We skipped over like delegated proof of stake. Is that another one that you could discuss a little bit in what you think that’s useful for?

Olaf Carlson-Wee: 00:19:19
Yeah, so I think delegated proof of stake is also significantly faster than Bitcoin’s proof of work. However, I think that dedicated proof of stake runs a risk that’s pretty well understood at this point. Having a somewhat small pool of validators and risks a sense of potential centralization and thus censorship or control by Endogenous parties to the network. So to me, delegated proof of stake, similar to Threshold Relay is a continued area of research, is very promising, but I think that until these things are in the wild for many years, it’s hard to know the full implications of these decisions.

Ryan Zurrer: 00:20:04
Yeah, and that’s kind of similar to say proof of authority. So with delegated proof of stake, you’ve got a group, all the token holders in a network that will sort of delegate mining authority to a subset of miners and those are probably very professional miners that are well structured to do that. However, that creates a certain amount of centralization similar to proof of authority where miners are chosen based on some selection criteria. Maybe they’re like notaries or maybe they have some kind of celebrity status in that network, and so they’re chosen as authorities and thus trustworthy in that network. But then, the problem with that is that it’s sort of orthogonal to the very culture of our space where we’re giving power back to some authority and it’s not decentralized and in the hands of each node on the network. And so we tend to prefer a permissionless system, an open network rather than say like some kind of proof of authority network that’s somewhat similar to what we already have today in the world, right?

Laura Shin: 00:21:25
Yeah. I think actually some of these systems lead and this was maybe the point Olaf was making to is that they lead a little bit toward oligarchy…

Ryan Zurrer: 00:21:25 Which is what we kinda right?

Olaf Carlson-Wee: 00:21:38
The thing is though, that it’s very tricky to predict how this will play out on a mass scale over many years. So for example, Bitcoin’s proof of work, which I think on paper is a very decentralized system. In reality, as we’ve seen play out in the market, I think that just a handful of mining pools dominate validation in the Bitcoin network. And, I think that kind of centralization that came with economies of scale of mining hardware and power was very hard to foresee. So to me, we can predict as best we can, but this is one of these interesting things about this area is that when it plays out in the wild, in an adversarial environment. And, in the real market, it can often be a bit different than it was thought to be on paper.

Laura Shin: 00:22:30
This will definitely be a continued area of research. So we’re going to discuss governance and other issues. But first I’d like to take a quick break to tell you about our fabulous sponsors. Founded by Ed Stevens, Preciate is building the most valuable relationships on earth and each episode of Unchained, Preciate sponsors the recognition of an individual or group in crypto for an achievement. Today, Preciate is recognizing Bill Shihara. founder and CEO of Bittrex. A leading US based cryptoexchange. Bill recently gave a keynote speech in our nation’s capital to a group of CEOs and leaders from around the world. He was candid and thoughtful and spent hours after his keynote helping anyone who asked him. Bill is a leader among leaders, and he showed that by being so generous with his time and insights. Thanks Bill. Listeners, if you know someone in crypto who should be recognized on a future episode of Unchained, take action and go to preciate.org/recognize.

KeepKey: 00:23:26
Cryptocurrency is vibrant and exciting, but it’s not without its share of bad actors. Exchanges and personal accounts can get hacked. Computers can be infected with malware. Left unprotected, your digital wealth is up for grabs. Don’t let yourself be a victim. KeepKey is the safest and simplest way to protect your bitcoin, ether, litecoin and other tokenized assets. This hardware wallet is a separate device that you control. Brought to you by the pioneering team at ShapeShift. KeepKey works with the wallet software on your computer to manage your private keys and transactions. Your device is pin protected, which renders it useless even if it falls into the wrong hands. Its large display, let’s you carefully review and approve every transaction, and if your KeepKey is ever lost or stolen, you can safely recover your device without compromising its private keys. The bottom line, you’ll sleep easier knowing that your digital wealth is safe and secure. Visit Keepkey.com to order yours today. It works on PC, Mac, Linux and Android.

Bitwise: 00:24:26
Bitwise Asset Management is the creator of the world’s first cryptocurrency index fund. The Bitwise HOLD 10. The fund holds the top 10 cryptocurrencies by five year diluted market cap, rebalances monthly and takes care of secure storage and taxes. It’s an easy, secure way for long-term investors to get diversified exposure. Bitwise is backed by Khosla Ventures, General Catalyst, Blockchain Capital, Naval Ravikant, and several others. They are a trusted partner to individual investors, wealth managers, family offices, and large institutions who are navigating the crypto space. The fund has several hundred LPs and is currently accepting accredited investors. To learn more about the Bitwise, cryptocurrency index fund or download research, visit www.bitwiseinvestments.com/unchained.

Laura Shin: 00:25:19
I’m speaking with Olaf Carlson-Wee and Ryan Zurrer of Polychain Capital. We were talking before about block rewards and I know that was something you wanted to explore, but I didn’t know if we’d gotten to everything that you wanted to say about them.

Ryan Zurrer: 00:25:30
We’re very optimistic of the block reward. We had this moment in 2017 where projects were selling some kind of a majority of their tokens all at once in a crowdfunding moment. And then with Juan Benet’s Filecoin, we saw what I call the return to the block reward where Juan prioritized fully 70% of his network over time to incentivize that network over time. And we think that that was a very prudent move. We’d like to see that in other teams that the reward, the token is doled out over time to incentivize actors over the long run to contribute a specific resource or a relevant activity on that network. And what we’re also seeing is the emergence of block rewards that can go to different actors in different ways. So where a network needs a combination of say software developers developing applications and miners confirming transactions and other entities doing other things and contributing to other resources, the block reward… The interesting way to approach the block reward is to have governance that allows it to be doled out to these different constituents in these different value add groups in accordance with what is important or what is valuable for that social network, or that network of nodes.

Ryan Zurrer: 00:27:01
And so we’re spending a lot of our time today jamming with projects on how to design their block reward such that they bring together all of the different groups that they need to incentivize as they’ve gotten a higher abstraction of software developers building applications that they can also incentivize low level protocol layer development. While you know, having security and transaction confirmation and different things. And the block reward doesn’t need to be just isolated to one group. You really should think very carefully about all of the different constituents in your network.

Olaf Carlson-Wee: 00:27:42
So you can imagine a thought experiment where instead of 100 percent of block rewards going to validators or block validators in that network. You can imagine say 10% or 20% of that block reward going to a Treasury. This could be a multisignature treasury, which could be managed, sort of like a DAO or decentralized, autonomous organization where the token holders can actually vote on how that treasury is spent. So, there’s no reason you couldn’t use the block reward to subsidize other types of activities outside of just block validation which adds security to the network. One thing that would be very interesting to think about is suppose in Bitcoin, if 50% of the block reward were paid to core developer contributions as determined by the community of users. Now there’s some pretty difficult governance problems around voting and how that allocation is decided based on the community of coin holders. But the hash rate of bitcoin would be, mathematically about one half of what it is today, which I don’t think would materially alter the security of the network in a meaningful way. But then you would have literally tens of billions of dollars in a sort of centralized treasury controlled by the decentralized network of users that could actually be allocated to core protocol developers to anything that could increase the size of the network, increase the reliability of the network, increase the visibility of the network. So this idea of a kind of a decentralized DAO treasury that could actually be used by the community to pay for things outside of just block validation is a very interesting concept to me.

Laura Shin: 00:29:32
Yeah, I love that. And, just so I understand then those funds would be held in a smart contract and then there would be some function by which people would vote on how to dispense the funds?

Olaf Carlson-Wee: 00:29:45
Yes, exactly. It would likely be a system of smart contracts. One would probably hold funds another would be a voting contract, which could uniquely control the actual contract that held the funds. So it’s not a simple architecture, right? And the security of such an architecture would obviously be paramount, but it is the type of thing that I think we will start to see emerge. So that funds can effectively be used by the community in ways that… we’ve never seen that type of capital coordination built directly into a protocol. And I think it would be a very, very interesting experiment to see play out.

Laura Shin: 00:30:25
Are there any particular tokens that you think are really interesting in this way?

Olaf Carlson-Wee: 00:30:30
So anything that is really focused on on-chain governance is capable of creating such a system as this, Systems that are working on on-chain governance like Tezos and Dfinity, I think are some of the most promising contenders there.

Ryan Zurrer: 00:30:57
Also, Polkadot. There’s really interesting work coming out from that team on on-chain governance. But on-chain governance is very difficult. And we are still sort of researching and developing around security and secure models for on-chain governance.

Olaf Carlson-Wee: 00:31:11
The problem with… I’m glad you brought up security Ryan, because there’s not just a sort of network security of on-chain governance. There’s kind of like ways you can vote yourself into a black hole, right? You vote for no more voting or you vote for giving all the tokens… like for example, you could imagine a network, we’re 51% of the people or rather of the tokens voted to steal the tokens from the remaining 49% and grant it to themselves.

Laura Shin: 00:31:11
[Laughing] Yeah that would not be good governance.

Olaf Carlson-Wee: 00:31:47
Yeah, and one would hope that the market will react in such a way that there would actually be a net loss of value for that 51% that voted that way. But, these are the types of problems that are not really network problems. They’re more like game theory problems for how these voting and governance systems could play out. So I do think we’ll see some really, really interesting experiments over the next few years. I think some of which will succeed wildly and some of which will blow up in a large way.

Laura Shin: 00:32:16
How do you try to prevent something like that? Because actually before we got into this topic, one of the questions that I had for you was that, if I think about kind of the pros and cons of on-chain governance versus more like an Ethereum or Bitcoin model, my take on it is that obviously we’ve seen the pitfalls of off-chain governance. However, with on-chain governance, that can be somewhat more easily gamed since people sort of know the thresholds they need to meet in order to effect some sort of change that they might want. Whereas with something like Ethereum or Bitcoin where human judgment plays a bigger role and so governance is less predictable. Attackers can’t just say to themselves, “Oh, here’s what I need to do. This is the minimum necessary to push this decision to my advantage.” So how do you prevent that kind of thing from happening?

Ryan Zurrer: 00:33:08
I would say with respect to on-chain governance, yes, it’s difficult, but that doesn’t mean that we shouldn’t move towards that. So it is a more difficult goal to achieve, but it’s a very noble and I think it will be a very valuable goal once we get there. Now with that in mind, it’s important for projects to grandfather in governance over time and there are some checks and balances that you can put in place to help do that. For example, this is what we did with Maker where the core team held sort of veto power on any decisions with respect to Maker and its development. But they fostered this community of token holders that were very participatory. So that once a week, on Sunday mornings there’s a governance call that all the major token holders participate in. There’s kind of a culture of expectation that to be a major a token holder in the network that you have to participate and have to be vocal and have to contribute. And then, this voting became sort of a natural process and then the whole community was also able to vet out over time that there is a threshold or quorum of reasonably competent participants in that network, such that, there’s confidence there that when Maker migrates to fully on-chain governance that there are a group of actors that understand the decisions that they’re making in that there’s kind of a voting block that can prevent negative things from going through. But again, that needed to be developed over time through the culture of that network, which I don’t think we talk a lot about in our space. But really, it’s an important element there. And I think you’ll probably see it in some of these other, emerging governance platforms that there are some large token holders that are known entities that are, benevolent actors that will contribute to the network in a positive way. And there’s an expectation of that. And, they work together and coordinate such that attacking the network is then more difficult. Now that does unfortunately mean that sort of like the whales have a certain amount of power in that network, but again, I think the benefits of on-chain governance and the speed that it will offer to a network in evolution of itself, of the network, will dramatically outweigh the drawback.

Laura Shin: 00:35:43
My thing about that is then it doesn’t sound… it sounds quite centralized. Is it just that these people… that their votes are weighted more heavily… I’m just trying to figure out because that just doesn’t sound as decentralized as I typically think of a decentralized network being.

Ryan Zurrer: 00:36:10
Sure. But we just went through a few minutes ago how Bitcoin and also Ethereum have kind of consolidated around some powerful nodes that do a lot of the mining there. It has been this migration towards some level of centralization. Decentralized consensus mechanisms do not necessarily mean radically decentralized consensus mechanisms. There sort of a happy medium towards efficiency and evolution. And then on the other side, sort of massive decentralization. I think the more that you, that you have participation and the more that you incentivize nodes to be positive contributors to a network, the more that other people will start to onboard into that network. And then over time, again, in this grandfathering process, it will decentralize and you’ll have say closer to democracy than plutocracy. But at the same time, because some of the early tests in governance have been total blow ups, mostly because they haven’t been able to get to quorum for even very obvious decisions. I do think that having this grandfathering process and governance make sense at the juncture that we’re in and it’s better than just throwing our hands up and saying, “Well, governance is hard, so let’s not do it. Let’s just trust the dictator.”

Laura Shin: 00:37:41
And when you say that there have been these blowups, what examples are you thinking of?

Ryan Zurrer: 00:37:52
Well, the DAO is obviously the big one, right? Like to not have voted in favor of the moratorium. That was such an obvious decision, but we couldn’t get to the threshold to push that decision through because not enough token holders had voted in that moment, even though at the time it was a frankly obvious decision and that was because we didn’t have the right tools to make it easy for token holders to vote. There were obviously a certain threshold that were speculative asset holders and not active participants of that network. And that network in that moment would have benefited from say, a minimum quorum of well capitalized benevolent actors that could push a reasonably non-controversial measure like the moratorium at that time through.

Laura Shin: 00:38:51
Yeah, I definitely see what you’re saying and this actually leads us to another question I was going to ask you, which was for on-chain governance, whether or not we really should be trusting people who are busy, with their work and their families and everything to make a decision on probably stuff that they haven’t done research in. And, if I look at even just governance is like here in the US and I see how… I live in California where we’re supposed to vote on all these referendums all the time. Every time I’m like, “Oh my God, I have no idea about these issues and I am supposed to make this decision?” So I definitely see what you’re talking about, but I also do think that… it’s like a tight rope, you know, you don’t want to go too far one way or the other.

Ryan Zurrer: 00:39:42
Well this is the beauty of cryptoeconomics though, right? Because when you add incentivization in, then people get very interested very quickly. And, I would argue that if you were paid a reasonable amount for your votes, and you were also tracked and measured based on your contribution to your voting than, in California, over the long-term, you would probably, or maybe not you, but a threshold number of people will be very active and given the metrics of the wisdom of the crowd and the strong academic research in favor of that. I would argue that the crowd would arrive at good decisions probably more often than the benevolent dictator situation, or another political model. Adding in that economic incentivization matters, a lot.

Laura Shin: 00:40:42
And why wasn’t the economic incentive there with the DAO, because wouldn’t they have had that incentive in order to preserve the value of their tokens? So there was an incentive there and yet people still didn’t do it.

Ryan Zurrer: 00:40:57
Well, not necessarily, right? You need a couple of different things. You need the access and the information, right? And so there wasn’t the access and for some people there wasn’t the information. And then you also need the incentive to do the action, right? There is this sort of Nash equilibrium that went the wrong way where people are like, “Oh, well other people who can run command line, will just do the voting and the quorum will go through.” Almost like Brexit, right? And then like it happened, and then you’re like, ” Whoa, wait a second. That’s actually not what I wanted.” Kind of thing. But like people were sort of lazy because they didn’t receive direct benefit for that action. So if you receive direct benefit for taking action, then that’s where, again, a cryptoeconomic model can adequately incentivize very strong governance.

Olaf Carlson-Wee: 00:41:50
And what we are getting at is a larger issue around voting, really being something that when you think about it from a game theoretical perspective in say the US elections, it’s really quite rational to not vote at all because the probability that your individual vote will swing even to say a local election is extremely low. And so it leads to a sort of tragedy of the commons. Just our normal democratic system. And it’s why in part, you see such low voter turnout even in, in very important elections. So to me, one of the things that’s so fascinating about this is that we move from a designing democratic and voting systems using governments and nation states with extremely high stakes, to using open source software systems and blockchains, which I think everyone can agree iterates at about a 1000x speed to the development of nation state systems, right? So now you have open source developers all around the world actually designing the tip of the spear on what governance voting and incentive systems look like. And so to me, the fact that we can move this to an offline world that’s often a bit lower stakes and much, much more experimental and open to experimentation is something that will have long lasting effects.

Olaf Carlson-Wee: 00:43:10
And I think we’re gonna see this in other ways with blockchains, right? So, we’re seeing this early, new financial instruments that are on-chain. So things using the 0x protocol, which is sort of like a trade execution protocol in a smart contract. On top of that, we’re seeing a Dharma, which is a peer to peer loan system. DYDX, which is a peer to peer options and derivatives system. Over time, these are projects which are mimicking financial instruments from the offline world. Once these are all created in that blockchain smart contract environment, we’ll actually start to see the tip of the spear of novel financial instruments occurring in an open source software development environment on blockchains. Instead of that occurring, in the sort of traditional financialization that’s kind of a complex intertwining of Wall Street and the government and various regulatory bodies. We’re going to see it evolve in a global open source software environment. So to me, the fact that we’re pushing a lot of these very hard questions that actually are unsolved, to this open source software development speed and iterative ability, is one of the things that’s so fascinating to me about this whole ecosystem.

Laura Shin: 00:44:32
So I actually want to go back to what we were discussing earlier where you were talking about how you could govern your way into a black hole. How do you design a system to prevent that kind of thing from happening? As you mentioned, there’s just so many ways that these could blow up and everything’s an experiment, but obviously I’m sure as these teams are going about trying to design these projects, they’re trying their best to account for all the different failure scenarios. So how do you guys think about that when you’re trying to help these teams develop their systems?

Olaf Carlson-Wee: 00:45:07
So as Ryan said, I think it’s important for things to start gradually. You don’t want to codify something that you’re not sure if it actually works and kind of vote yourself into a black hole. But there are all sorts of things you can do. Ryan mentioned outreaching quorum. If you don’t reach quorum, you can reduce the threshold which is required to get to vote programmatically. So it’s the type of thing where if you can see the problems before they happen, you can continually iterate and codify around them. The other thing is that voters can actually change the voting system, right? This, again is kind of tricky and an almost paradoxical because you have the people participating in that voting system, voting to change the rules and thus their rights within that voting system, which can again create like mixed incentives for certain participants and everything. The main thing for me is that there’d be a number of experiments run, and that we encourage experimentation in this area and that over a long enough time horizon, we will see what works. And, continue to build on those core things that work and continue to experiment and iterate. I don’t think anyone today, even those of us that spend a lot of our day thinking about this, can really determine what is the exact best mechanism right now. I think this will be an iterative process that takes a long time to determine, but I also think that when it is figured out, I would say that ICOs to date have been the most efficient form of capital coordination we’ve seen since the development of capitalism itself. $30 million in 30 seconds is kind of unprecedented, right? And I think that, if we can move to these systems, on-chain voting of governance in an effective manner, we’ll actually see even, more efficient and large scale, form of capital coordination beyond anything we’ve ever seen in the world.

Laura Shin: 00:47:09
I also want to ask you about a couple of the governance mechanisms that get talked about a lot like Futarchy and Liquid Democracy. Can you just define those and also describe what you think of them?

Olaf Carlson-Wee: 00:47:20
Yeah, what a lot of these come down to is on a very high level, putting the vote to the market. So, in a where you have token based voting, those largest token holders, it’s literally kind of one token equals one vote, which is very different system than we have in say the US offline democracy. It’s one person, one vote. And, the blockchains though you don’t really have this sense of personhood yet. So to me, I am very excited about experiments like Futarchy were effectively prediction markets, can determine outcomes and so people have to sort of put their money where their mouth is so to speak. And I think that you get much more honest assessments of positions. A much more honest reveal of my preferences when you asked me to actually bet on, on the outcomes that I want. So I very much am looking forward to larger scale experiments of these systems. And I think that there really is no clear answer though today about what is going to work the best.

Laura Shin: 00:48:31
Are there any projects that are trying Futarchy?

Ryan Zurrer: 00:48:37
I think very wisely, there are not any projects that I can think of off the top of my head that are going to implement Futarchy in their first iteration of governance. However, it’s often a footnote that you’ll see in many white papers that governance will eventually move towards a Futarchy model. That specifically, having essentially a prediction market be the governance mechanism is something that needs a thorough amount of research and experimentation because it can create essentially a self fulfilling prophecy where if you like rally enough people around a specific concept and they’re sort of like all betting towards that concept, then then you can kind of in some certain decisions and in some instances can bring it into being. And that could go the wrong way. Because we don’t have a lot of experimentation around prediction markets to date, but we’re on the cusp of seeing a number of different prediction markets be live and at scale. I think, in very short order we will see a lot of projects move their governance towards a Futarchy model. I know a lot of people are excited about it.

Laura Shin: 00:49:57
I also wanted to discuss ICOs which Olaf had mentioned. Obviously, there have been a number of maybe less than savory outcomes we’ve seen in the ICO space and maybe incentives aren’t always aligned in such a way as people might like to see to incentivize the developers to actually build a network once they’ve raised all this money. So what kinds of governance do you think you can bake in or cryptoeconomics can you bake in to the way ICOs are being done now in order to improve the incentives there?

Ryan Zurrer: 00:50:57
One of the things I mentioned earlier, is just favoring a block reward that incentives the network over time. That is one. So having an ICO thats only a minority of the tokens and not a very significant component of the tokens. And then raising only enough to get the network live. So that the distributed network will have enough tokens leftover to incentivize resource allocation by individuals over the long term. That sort of one easy one.

Olaf Carlson-Wee: 00:51:05
Yeah. I think there’s also just some basic stuff that exists in a pre blockchain world for a reason. Like for example, vesting, which is just, you know, that the creators of these blockchains, should dedicate to a lock up so that they’re in it for the long-term. I think that makes a lot of sense. Though, I think a lot of the kind of wild west we see in the ICO markets is a result of the fact that in block chains, you really give financial sovereignty to everyone. And I think that not everyone is totally used to that kind of power. I think that when you give financial sovereignty to everyone, there’s like a flip side to that coin, which is that we see a lot of fundraising and contributions and investments that as an outsider don’t make a lot of sense. Yeah. I think that there’s a flip side to everything, when you give everyone free speech, say, people say unsavory things.

Laura Shin: 00:52:08
One other thing I was wondering about, because you guys haven’t raised this, but obviously you are investors and I know the way that you invest is a little bit more of a venture style where you really work with the teams to try to ensure their success. But, I’ve seen a lot of different analyses that look at how you can design a cryptoeconomic system to keep the value of the token from plummeting toward zero as the velocity or the usage and therefore the turnover of the tokens rises. So is that something that you guys think about and if so, what do you think are some good designs to keep the value of the token rising along with usage of the network?

Olaf Carlson-Wee: 00:52:44
I think what you’re referring to is the MV=PQ kind of monetary equation? I think that can maybe usefully inform a way to think about token values. I think we’re dealing with a totally new asset class and it is incomplete, to use that sort of valuation model. That again, is I think it’s struggling to take an old system of thinking and attach it to the current things and innovations we’re seeing in cryptocurrencies and I don’t think it’s the full picture. So every kind of valuation model that I’ve seen, I really applaud those that are working towards coming up with something that makes sense. But I also think you need to think about this from a kind of first principles perspective without being too caught looking in the rear view mirror at equations that have worked for historical movements, historical assets, and apply them too aggressively to this new area which is frankly just a completely new asset class.

Ryan Zurrer: 00:53:51
So for example, some people are of the opinion that lowering the velocity of money on these networks does increases the value of the token. And so, they are coming in with longer staking lockups and, things like that. But in reality, in the nature of our space and these networks, you could think of it totally counter-intuitively as well and say increasing the value of tokens means that you’re increasing transaction throughput, which is that actually network effects, which means that the network itself is increasing in usefulness. And so again, you don’t just exemplifying last point and further to that, I’m not sure that we have arrived at a simple and true in all senses equation for valuing tokens or valuing these networks. And again, we support experimentation and continue to be open minded about how to approach these networks.

Laura Shin: 00:55:01
Well, when you’re helping teams design these systems, do you guys think at all about how to inflate the money supply or whether to cap it or, or things like that? And, and if so, how do you make those decisions?

Ryan Zurrer: 00:55:18
It really depends on… it’s a case by case basis. So when we bring in projects and jam on Cryptoeconomics, it’s important to note and to remember that a good cryptoeconomic model requires computer scientists and economists and behavioral psychologists and math and gametheory experts and it’s really this amazing field because it’s a combination of so many different fields. And so, depending on the resourcing question that’s being traded on the network, depending on the nature of that network, depending on whether say velocity is a desirable thing on that network or whether you’re trying to say maximize psychological ownership of the token or that networking. So locking it up makes more sense, we’ll help design a system that sort of makes sense for that specific network. I mean, the capping or uncapping will usually be the result of a number of different variables that are very particular to what you’re trying to achieve. Whether it’s trust or whether it’s exchange of value or whether it’s exchange of a resource. Again, how you approach each one of those would be different. And the cryptoeconomic model there would would also be different and needs input from all these disparate fields

Laura Shin: 00:56:47
We’re running out of time. But, one other thing I wanted to ask you about and Olaf knows that this is something I have been curious about. I feel like every time we’ve seen sort of a really big decision that needed to be made and people had entrenched views on either side. There’s been a fork. So I just wonder for all the talk of on-chain governance, how do you prevent, they’re just being a ton of forks all the time. I feel like, if you end up with some system where there’s a lot of voting going on about a lot of different changes and there’s a few every so often where people just feel super passionate about it and they just decide to fork. Then we’re just going to end up with all these split coins. But Olaf didn’t agree with that. And I’m curious to just hear your opinion on how you disincentivize that and keep everything on one chain.

Olaf Carlson-Wee: 00:57:36
I think that in a democratic or voting system, people tend to accept that if the system is designed correctly, they will accept the outcomes even if it was not their preference. So for example, if the president of the United States that is elected is not the candidate I voted for, it doesn’t mean I leave the United States. It often just means I accept that in this case, I was in the minority of voters and there really was a preference, that wasn’t mine that actually represented the most people. So to me, if you’re buying a token on a blockchain, that has on-chain voting and governance, you should also be buying into the fact that when you are the minority voter, that you accept the outcome of that vote as legitimate, even if it was not your preference. Its not to say that this will stop forks, right? But I do think there’s more of an understanding that you will not have the outcome that you always want, but you still accept the legitimacy of the overarching system.

Laura Shin: 00:58:40
Yeah. We’ll see, I think the place where I disagree with you slightly and maybe it’s just that we’re on different places on a spectrum is that I feel like with an open source system you can so much more easily leave. Whereas obviously if you live in the US, you kind of have to upend your life to protest the outcome of an election. So it’s quite a different thing. But I guess we’ll find out what ends up happening. Is there anything else that I didn’t ask you guys about cryptoeconomics that you want to mention?

Ryan Zurrer: 00:59:12
One thing that I’m observing and we’re starting to jam on here at the office and do some research on is really the power of psychological effect in cryptoeconomic systems. And this has come up fairly recently with respect to Maker and the Dai. So as ether went down quite precipitously over the last few months, there were some instances where one would’ve expected the Dai to become unpegged and adjust downward, at least moderately from the one US dollar value that it is. And in fact it has maintain stability remarkably well, even though it’s still today, just the single collateral which is ether. And, we’re starting to kind of think around just the psychological effect there that people have decided that Dai is a dollar, and even though the metrics in certain moments would have indicated that it should be adjusted down slightly because, there is a certain black swan risk. Now, again, just a caveat, the Black Swan risk will be reduced when it’s a basket of underlying collateral. Unfortunately today, it’s still not a basket of underlying collateral, but even in this incipient moment, we have stability and we only have stability because of the psychological effect that the market has deemed it to be $1. And I find that incredibly fascinating. And again, it calls into attention the fact that strong cryptoeconomic models are a combination of economics, behavioral economics, behavioral psychology, Computer Science, security, gametheory, math and so on. This for me is what makes this field the most interesting on the planet, is this combination of these otherwise disparate fields in a new way that is so fascinating and has such an amazing effect on large groups of people who otherwise have no interest or or no commonality amongst themselves globally. And, it’s happening at such an accelerated rate. I find myself often saying that we are innovating in the field of economics more in the next 24 months than have happened in the last 24 years. And that for me is just incredible.

Laura Shin: 01:01:46
I agree. Olaf, did you want to add anything?

Olaf Carlson-Wee: 01:01:48
No, just thanks for having us on the podcast. Laura,

Ryan Zurrer: 01:01:53
Congratulations on all the success with this new podcast. It was expected. And, we’re just so excited about all your success here.

Laura Shin: 01:02:03
Thank you. This has been such a great conversation and I would like to have you back because I’m sure in just a few months time there will be much more to discuss on this same topic. But before we go, where can people get in touch with you or learn about Polychain?

Olaf Carlson-Wee: 01:02:19
Think if you go to our website, you can get about as much information as we have out there, which is polychain.capital.

Laura Shin: 01:02:28
Right, It’s like three sentences or something.

Olaf Carlson-Wee: 01:02:31
We are hiring for a variety of roles. So if you’re interested and listening, feel free to reach out.

Ryan Zurrer: 01:02:37
Yeah, there is one link to a jobs board and you can also reach out to [email protected]. We are hiring both for internal functions as well as a number of functions around our portfolio. So if you are active in any of the fields that I just mentioned, and excited about what’s happening in our space, and in cryptoeconomics, please feel free to reach out. Unanimously across our portfolio projects are looking for talented people to come and help them. And again, in a variety of fields, not just computer scientists.

Laura Shin: 01:03:16
Great. And as I’ve said before, not on the podcast, but on the web, this whole field is full of people who were doing something else before. So don’t necessarily think that you don’t have experience because basically none of us have experienced. Thanks guys for coming on the show. Thanks so much for joining us today. To learn more about Olaf and Ryan, check out the show notes inside your podcast episode. New episodes of Unchained come out every Tuesday. If you haven’t already, rate, review and subscribe on Apple podcasts. If you liked this episode, share it with your friends on Facebook, Twitter, or LinkedIn. Unchained is produced by me Laura Shin with help from Elaine Zelby and Fractal Recording. Thanks for listening.