We're Bringing 60,000 Super-Users to DeFi Overnight: Guy Benartzi

Hello defiers! Like I promised yesterday, today I’m publishing the full interview I did with Bancor’s Guy Benartzi, who co-founded the so-called liquidity network two years ago with his sister Galia. Bancor’s ICO in the 2017 go-go days became infamous for ...

Hello defiers! Like I promised yesterday, today I’m publishing the full interview I did with Bancor’s Guy Benartzi, who co-founded the so-called liquidity network two years ago with his sister Galia. Bancor’s ICO in the 2017 go-go days became infamous for raising $153 in three hours and slowing the Ethereum network to a crawl. It’s since been criticized for being too centralized as it kept an “emergency switch,” and for having a token when other platforms just use ETH.

In a series of changes it’s rolling out in the coming months, Guy says Bancor wants to prove that a platform token can be useful, it wants to gradually transfer control from management to the community, and it wants to entice more people to become liquidity providers by giving them a taste of market-maker fees, for free. He’s also proving that it’s possible to have a blockbuster ICO and also deliver a product.


Image source: Bancor

The interview has been edited for length and clarity and I bolded by favorite quotes.

Guy Benartzi: We are taking the reserve of the Bancor liquidity network, which exists today in the Ethereum blockchain, and as part of our increased decentralization of the network and its governance, we are essentially airdropping a ETHBNT relay, a token relay, on all of the BNT token holders. Every single person that holds BNT will also receive a second token, which is essentially a decentralized token pair.

You're going to have about 60,000 people, because over 60,000 people hold BNT, it's one of the most broadly distributed crypto tokens in the world, and they will become passive market makers. People no longer need to be professional market makers. It's enough that they have capital and they stake it inside these liquidity pools and they become like a pro rata owner of a token pair.

Today in all of DeFi you probably have a few thousand users. And in terms of how many people actually provide decentralized liquidity to these liquidity pools, my guess is maybe a few hundred people in the entire world. And so overnight it goes from a few hundred people to a few tens of thousands of people. And we hope that through this action many, many more people will get exposed to kind of this emerging world of decentralized finance or open finance.

Camila Russo: Can you give a sense of the range of fees token holders can expect to earn? What's the like magnitude of earnings expected or yield?

GB: The truth is, it depends on a variety of factors, including how much transactions will be on that relay on the Bancor network, how many people will choose to keep their token relay, because that would be the amount of people that it will be divided up pro-rata between. So there's obviously factors we don't know. One of the things we're going to do before the airdrop is to release a set of very simple, easy-to-use ROI calculators that you can look at it and you can see what it does and see how it's performing and see what it's making you, look at what the other ones are doing, and decide if you want to move some of it or all of it or none of it.

We're going to make it easy for people to visualize cause we kind of feel a bit of a responsibility bringing in, you know, tens of thousands of new essentially DeFi super users or customers. We want to try to make it simple for them. The whole idea of being an automated liquidity provider is that it's simple. You look at a graph, you see if it's making you money, if not, you just click a button and sell it. It's really something that anybody do.

In over $2 billion of conversions on the Bancor protocol we've learned something very interesting, which is that many token liquidity pools and token relays, they suffer from what's called impermanent loss. It means that if you put capital inside, you might actually withdraw less capital despite the fees because of the way that a token relay works. Now our goal in the kind of next generation of liquidity pools is to increase the range in which being a liquidity provider is profitable. That will encourage more usage and provide better rates because the liquidity pools will be deeper. You'll be able to convert bigger amounts with less slippage.

One of the reasons it's so complicated to even calculate if you're going to be making or losing money in a liquidity pool is that almost all liquidity pools have two tokens that are both speculative tokens. And that's crazy. That would be like having a pair on a stock market that's like Apple/Google.

And so I think that you're going to see all liquidity pools in 2020 essentially shift to a model where you have a synthetic stablecoin on one side and you have a token that's speculative on the other side. And in our case, what we did is we recently announced USDB, Bancor’s stable coin. And it works similar to Maker's Dai, except it's collateralized with BNT instead of with ether. You can use it inside of a liquidity pool in Bancor and that makes it really easy to calculate if the relay's going to be profitable or not. But you do it in a way that still keeps you connected to the entire Bancor network because behind that USDB is actually BNT that allows you to hop between blockchains. I think you're going to see everybody basically doing this.

CR: Can you explain a but more what happens to the reserve of BNT after the airdrop?

GB: The way that BNT works today is it has a single reserve of ETH that's equal to 10% of the market cap. So you can think of it like the reserve is almost like a kind of order book, ready to be liquid at any moment. And so because of that, you can very easily buy BNT, which is very essential because BNT is the network token built so that it gives you the shortest amount of hops between tokens in different blockchains. So it was very essential that BNT be liquid. And so it was essentially the first token that you could buy and sell directly from its smart contract. When you buy BNT it issues, new BNT and when you sell BNT to the smart contract it destroys BNT. It has a fluctuating variable supply.

And in this upgrade, what changes is that the BNT itself becomes a token that no longer has a fluctuating or variable supply. It has a fixed supply and instead a relay token is created into which the entire Ethereum reserve that exists today, so 10% of the market cap, will go into one side of the token relay and on the other side of the token relay, will be matched by the foundation with an equal amount of BNT. It'll have the exact same Ethereum reserve as right now, but we’re moving it from a liquidity pool that has a single reserve, which is Ethereum today, to a reserve that has two reserves, both Ethereum and BNT. So the end result is that it is no longer biased towards any particular blockchain.

CR: So you're saying that the purpose of this is to have BNT become blockchain agnostic and not depend so much on ETH, but I don’t get how this new mechanism helps that.

GB: Because by doing this, what we do is we airdrop the entire token relay on all the BNT holders and they can choose with their free will to do whatever they want. They could leave it in their wallets, and it’s like they own a pro rata piece of an exchange pair and collect the fees. They could decide they want to move it into a different token relays for some other ER 20 token. Or maybe they want to move it to EOSBNT, or maybe they don't even want to keep it, maybe they just want to cash it out.

CR: So now token holders are free to provide liquidity on EOS too. Can you tell me more about how the fixed supply and inflation will work?

GB: Once this airdrop is over and BNT is a fixed supply token, its initial inflation will be 0%, but the community will be able to choose to change this and to add inflation. And so let's say the community decides they want to have 2% inflation. And let's say that the market cap is $50 million just to make the math easy, then there's $1 million of new BNT. that's going to be issued in inflation and airdropped to wherever the BNT holders want it to go. They not only choose the inflation, but they choose the recipients of the inflation. As an example of who the possible recipients are, the first and obvious one is the token relays themselves. So token holders can decide which token relays are worthy of funding. This encourages more people to be liquidity providers. And it really illustrates what is unique about an automated liquidity network protocol that has its own network token because sometimes people would ask as they started to fork Bancor, why do you need the token?

Why not be like Uniswap, which just uses Ether? The answer is that there are many different ways to do something. It might be that there's not one that's wrong and one that's right. But there are different advantages and disadvantages to different ways of solving things. One of the advantages of having a network specific token to a liquidity layer on top of blockchain is one, that you're blockchain agnostic. So, you know, you don't have to put all your eggs in one basket. And I really honestly don't believe that we will have one single blockchain to rule them all. I think we'll have many different blockchains for different use cases.

The second one is that you can create the kinds of game mechanics or I should say the kinds of financial incentives that encourage the network to grow in the right ways via decentralized governance. And you also create longterm sustainability for the project because over the longterm the inflation will fund not just which token relays should be topped off, but which new developments should happen on the protocol. It gives a longterm sustainable, viable, let's almost call it a business model to a protocol, to a standard, that allows it to continuously evolve, um, without requiring, donations or, never ending money, which nobody has.

CR: If token holders decided to put inflation towards a relay, how will that impact token holders’ fees? Does increased liquidity mean they get higher fees?

GB: The relays' liquidity will increase because more capital will essentially be airdropped into it. But from the perspective of the relay holders, the relay got bigger, but no new relay tokens were issued. So the same relay tokens are out there, but they represent a bigger liquidity pool. So everybody relative amount has gotten bigger.

And there are other two things that they’ll be able to vote on. The first will be the liquidity pools, but the second is the oracles. Oracles are really important because they enable BNT to move between blockchains. Oracles essentially certify BNT was locked on one blockchain and then issued on another one. And then when it goes back, then it was destroyed on one and released on the other one. So that they ensure that the same amount of BNT always exists.

Then lastly, right now and probably for many years to come, we are giving out grants and we're developing the protocol ourselves and working with many core developers to build on it. But token holders will also be able to choose which developers they want to give inflation to. Maybe there's a cool project or a cool idea that's out there that the community really thinks should happen. They can choose effectively to fund new development through the inflation.

CR: All these voting mechanisms are basically one token, one vote. Will yours be like that?

GB: It's going to be similar. We haven't announced it yet. We're right now is simply in the phase where we're letting people know that this is even going to happen. After people have had a chance to move their ETHBNT into a noncustodial wallet or make sure that their tokens are held in an exchange, which is participating with us in this airdrop and making sure people get their allocation of the airdrop if their token is held in a centralized exchange, once that's behind us, then we'll start rolling out the first votes and the first votes will be on the simplest and most kind of straightforward subject that has the least ability to be manipulated, which is simply choosing the relays they want to top off. We want to get people into that phase where they start getting used to the fact that they have governance, that they have a voice, that they have influence, they can make decisions for BNT and to kind of see which percentage of the BNT community participates in the votes. Over the coming year, that's our plan. Then we’ll incrementally roll out to more complex, more involved votes where maybe it requires a little bit of due diligence, but that's a later phase.

CR: What Oracle's are you using right now?

GB: Right now we have a few different entities, companies, that are running oracles on the Ethereum blockchain and on the EOS blockchain, and it requires two out of three oracles to be validating a transaction in order for it to cross the bridge on Bancor X. Once we open this up and allow public voting on oracles, what we expect and what we hope to see is that there are going to be a significant increase in the amount of oracles, which will further decentralize the Bancor X bridge.

CR: And this bridge is the connection between blockchains, between Ethereum and EOS?

GB: Yes. It's the only cross-blockchain bridge, that's smart-contract enabled and in production. It’s about half a year old and it's been in production with millions of dollars moving between blockchains, between EOS and Ethereum. We have some big announcements that we plan to make with some very known players in DeFi that are going to be joining and taking nodes on the bridge. But we haven't formally made those announcements.

CR: What's the next block team that you think will be bridged?

GB: It's always a tricky subject and you know, we don't want to say things necessarily with certainty, before we're ready announce it. But what we do tell people is that we are completely focused not on hype and not on price and not on anything else, only where there is usage only where there are developers and there's two blockchains right now, smart-contracts enabled blockchains, that have usage from our perspective. One is Ethereum, which is the overwhelming dominant blockchain, has the most developers no doubt, and has kind of this Renaissance happening in DeFi, which is super exciting. DeFi seems to, on one hand, not be affected much by Ethereum's current bottleneck and speed. On the other hand, it really, really benefits from Ethereum's pretty good finality of consensus where people are pretty confident that transactions on Ethereum aren't going to get rolled back and there's not any kind of cartel that controls Ethereum. So for decentralized finance projects, Ethereum is the place to be.

Blockchains like EOS, which may be suffer on the confidence people have in the finality of the consensus due to block producer cartels, on the other hand is a whole different generation of blockchains in terms of power efficiency. You could write real applications that start to become indistinguishable from the apps you and I use on the internet where we don't even really know that there's a blockchain behind the scenes. So I think that's what guided us to EOS. And if you use that same kind of train of thought, I would imagine that the blockchains that are likely, even though we haven't made this announcement formally, but the blockchains that are likely to see a tremendous amount of usage are probably things like Telegram, which already has over 200 million users of their instant messenger, including the entire crypto community. And so we've been keeping a very close track on that and you know, hopefully we'll have something to announce there in the future.

And then of course, Libra. We're in regular conversations with Facebook. Again, nothing we can announce or want to talk about at this moment, and Libra of course is not out yet. There's only just kind of stuff you can play around with. And the jury's still out if they will get through some of their regulatory hurdles, they obviously have a very big spotlight on them, but assuming Libra does come out and I'm confident it will, I can't imagine that you won't see Bancor there on day one.

CR: Back to the airdrop, is it for token holders based anywhere? Because a few months ago Bancor had to exclude some users in the U.S. and other places. What's the status with that and how does it affect the airdrop?

GB: We are today in a reality in which there still is a degree of regulatory uncertainty, especially in the United State. And due to that, the Bancor foundation felt it was prudent, not close off U.S. users, but disable essentially conversions on the Bancor network if you're using the official front-end Dapp developed by the foundation and available at bancor.network. Of course, the smart contracts are decentralized and they are not under our control. There's no way for us to do anything even if we wanted to. And so in that sense we didn't block anyone or anything on the smart contract level.

We have since both worked with a number of different partners who have gotten grants from us as well as we're now seeing many different developers create liquidity portals where you can go and do whatever you want. Maybe they're in different jurisdictions or maybe they have different risk assessments or calculations on what they want to allow or don't want to allow from their front end.

In terms of the airdrop everyone will receive the airdrop, including U.S. users. But they won't be able to go to Bancor network and interact with their token relay. They won't be able to convert it. They won't be able to move it. But if they want they’re welcome to go to one of the other dozens of wallets or liquidity portals that are out there that have integrated Bancor protocol and use those and try those.

CR: That brings me to another question that I see you often get criticized for, which is that Bancor had a “back door” to allow management to take over, so it's not as decentralized as it claims to be. How much control does Bancor management or the foundation have over the protocol and users’ funds?

GB: So first of all, in terms of user funds there's absolutely no control of user funds, there is no user funds we hold anywhere. Even though we do provide our own noncustodial wallet, which people can use called Bancor Wallet. It's non-custodial, just like MyEtherWallet or MetaMask or anything like that. And you can use it or use any other wallet when you're using Bancor. So Bancor wallet is just another noncustodial wallet and Bancor by no means holds any user funds anywhere including in its wallet.

In terms of the smart contracts, when we first launched we had the same thing that I believe actually the majority of token projects have including, projects such as Kyber and other well known ones. But sometimes competitors like to pick on Bancor. Success does breed its own kind of haters. When we first created Bancor, we were very conscious of the huge Dao hack that had happened and had the potential to destroy all of Ethereum. We wanted to protect against a very, very early and nascent system that we didn't know how it was going to function. And we were proposing to do something kind of crazy; create decentralized liquidity pools, in which we'll put millions and tens and hundreds of millions of dollars on-chain. Nobody had done that before.

So we created an emergency switch that allowed us to pause BNT from being transferable and that switch was –and by the way, like I said, many of the most mainstream high profile token projects, like Kyber, which has Vitalik as an advisor, has or at least had the exact same type of functionality– we used it only one time when we had a hack a little bit over a year ago and I'm thankful that we used it because we were able to save the Bancor network by implementing that and being able to recover all of the BNT that was stolen by the hacker. Since then, later Bancor updates removed that ability once we had made certain security upgrades and felt that the protocol was in a much more mature place.

The place that Bancor is going to go to is actually very similar to the way the Maker contract works. It's not the way it works today. When there's an upgrade, like the one that's coming up, we do have the keys to the GitHub where we can update the contract, and we're upgrading the Bancor smart contract shortly. This ability will be part of the rollout of governance to the community, in the sense that just like Maker, in the near future when we want to make upgrades rather than us making the upgrade ourselves, we will propose the upgrade and the community will need to choose to accept the upgrade. Exactly like having the governance of Maker's token.

CR: That's interesting. So you removed the ability to freeze BNT now that the system is more mature and now you’re moving into a system of greater governance?

GB: But it's a little bit more nuanced. So there was a capability to be able to freeze, which has been removed. There's still technically if you, if you talk to some of let's say the people who have a more extremist position on decentralization, they will say the very fact that you can still upgrade the contract means that you can modify the contract, but you would have to believe that the Bancor foundation is suicidal in terms of wanting to destroy their own project. Everything we would do would be completely transparent and everybody would see it instantaneously.

I think in general you have you have a bit of a battle of the minds in the Ethereum community especially and in blockchain in general on who owns the definition of decentralization, what is decentralization?

Is there one “religion” of the version of decentralization that is the right version and everything else is not decentralized? Because from my perspective, decentralization means you can be forked. You know, I can't fork Facebook. If I don't like Facebook, I can't take the data. If you don't like what Bancor does, all the data is on-chain. Everything is transparent. You can fork it. For me, that meets my definition of decentralization. There are others, they will tell you Ethereum is not decentralized. There are people that will tell you Bitcoin is not decentralized, the majority of the miners are Chinese and they could be technically controlled and technically, technically, technically. Or the equipment that is running Ethereum nodes is on Amazon, which is a cloud service, and technically Jeff Bezos could control a 51% attack on Ethereum. So that's not decentralized.

So I don't think that there is one definition of the word decentralization, but it certainly is true that as time passes and systems mature, we are increasingly decentralizing both the network itself, the liquidity pools themselves, and now increasingly the governance of the Bancor network.

Sign up to get the best and only daily newsletter focusing on decentralized finance news, complete with analysis, exclusive interviews, scoops, and a weekly recap. All content is free for now. Those who become paid subscribers now, before the paywall goes up, get a big discount :)

Subscribe now

About the author: I’m Camila Russo, a financial journalist writing a book on Ethereum with Harper Collins. I was previously at Bloomberg News in New York, Madrid and Buenos Aires covering markets. I’ve extensively covered crypto and finance, and now I’m diving into DeFi, the intersection of the two.