"All the Dizzying Ways People are Farming Yield is Proof of DeFi Composability's Success:" Ampleforth CTO
Hello Defiers! In this week’s interview, I spoke with Brandon Iles, the CTO of Ampleforth. Brandon is a former Google and Uber software engineer, and built his first crypto app together with Coinbase CEO Brian Armstrong. Today, he’s trying to rebuild the way money works.
For its huge ambitions, Ampleforth is a relatively little-known project. It wants to create an entirely new category of monetary assets — one that doesn’t have Bitcoin’s gold-like fixed supply, it doesn’t have Ether’s oil-like utility, it’s not fiat-backed like USDC, and it’s not collateralized like Dai. So what is it? It’s a token — AMPL— whose value comes from having a price which moves in a predetermined band, thanks to a smart contract which automatically contracts or expands its supply. The end result is a token that trades at close to $1, the 2019 US dollar, to be precise, but Ampleforth doesn’t like to be called a stablecoin —read on to find out why :)
We also talked about the project’s recent liquidity mining program called Geyser, where traders can earn AMPL by supplying liquidity on Uniswap, and how for Brandon, the dizzying activity around these schemes is proof of DeFi’s success. Brandon also discussed the project’s roadmap, from being used as a way to diversify collateral in lending platforms like Maker and Compound, to denominating debt obligations, to very far out into the future, replacing central bank issued currencies.
🎙Listen to the interview in this week’s podcast episode here:
You’re a free signup, which means you get only part of the transcript below.
💗 Consider supporting quality DeFi journalism by contributing to The Defiant’s Gitcoin Grant! You can make a big difference with not that much; 1 Dai is now being matched by 138 Dai. Thanks so* much to my contributors so far!
🙌 Together with Quantstamp, a leading blockchain security firm keeping your money legos safe, Kyber Network, the on-chain liquidity protocol for the tokenized world, and Keycard, the secure, contactless hardwallet & open source API.
Brandon Iles: My background is very much in technology as a pure science in school, worked in research labs after that. I happened on crypto almost by chance, so I first encountered the Bitcoin white paper through Google Search, completely out of context with no prompt or anything like that, it was in 2009. And so, I read it and back then, you know, I thought it was interesting. But with no context, I thought looking at the various aspects of it, like scalability and that sort of thing, it seemed like an interesting toy, more than anything else.
Camila Russo: In 2009, it must have been hard to see it as something else.
BI: Oh, for sure. Yeah, I mean, it was basically a PDF format, it looks sort of like an academic paper, but didn't have any university markings on there, had this funny name on top. And so, that’s pretty much all I had at the time. I sort of forgot about it from that point forward, you know, it’s around the time I was going to work at Google. Move on to California, started working there. And then a couple years later, late 2010 or 2011, a friend of mine from college sent me an email and said, hey, have you heard of this thing called Bitcoin? We should play around with it. And so, together, we built one of the first Android wallet applications to be released on the Android store. That got released in July 2011.
CR: And that was related with Bitcoin or no?
Brandon: Yeah, so that application was called Bitcoin Android. That was just a straight up Bitcoin wallet using the current Bitcoin J library. And so, the friend I build that with was Brian Armstrong, who has continued to go in crypto…
CR: With Coinbase.
BI: With Coinbase. Exactly. At that time, Google was a really exciting place, because that was very much the hotbed of machine learning and artificial intelligence. That was where deep learning was starting to explode. So, I was getting very deep into that area of technology, machine learning, worked in the machine intelligence group, in search ranking group. I got the opportunity to meet all kinds of incredibly smart people and do lots of interesting work.
I stayed in machine learning for a number of years, I went to Uber for a couple years. And then with the release of Ethereum was where I started to get excited with crypto again. So, I thought, okay, this is actually an interesting, a generalized platform we can work with and one that we could potentially solve some of the earlier problems that I didn't quite like so much about Bitcoin back in the day, the elasticity and fixed supply.
And so around that time, I started talking a lot with another friend of mine, Evan Kuo, the other founder of Ampleforth, and we were both trying to figure out what was going on in the space. Because this was when there was a lot going on in the blockchain world with Ethereum, and with the whole ICO craze that was coming and going. And we were just trying to figure it out, because none of it really seemed to make much sense to us.
Brandon Iles. Image source: Medium
CR: Was this 2017?
BI: This was, let me see, summer of 2017.
CR: And where did you meet with Evan?
BI: Him and I had already known each other. So, we met through mutual friends a couple years earlier and we found that we had a mutual interest in mathematics and solving different kinds of puzzles. And so, we often get together and just talk about different puzzles we encountered.
Basecoin
And together we try to collectively understand this great puzzle that was going on in the blockchain world. We happened on the Basecoin paper. And so, when I was reading that, that's when things really clicked, that okay, we can actually design new monetary assets here with goals that we create ourselves.
“We happened on the Basecoin paper. And so, when I was reading that, that's when things really clicked, that okay, we can actually design new monetary assets here with goals that we create ourselves.”
CR: Just to give them a little context or people who might not know about Basecoin, but it was a stable coin that was supposed to be algorithmically stabilized, right?
BI: Pretty much, yes. So, the intent there was to create a stablecoin. It had a three coin systems. There was the stable coin and there was bonds that you'd be issued to help the market bail out the system if it ever became underwater and there were shares. This was a big deal back in the day, because they raised a lot of money to make this happen from very smart organizations. Ultimately, they ran into a number of problems and decided to close the project and return the funds to investors.
It was an exciting project and people still talk about it today, because I think it's easy to understand. So reading that got both of us really excited about the possibilities in the space. We initially started with a project that we called Fragments. So, Fragments was fairly complicated, much more complicated than Ampleforth. And it also had multiple coins, had a bond market in there and we had a reserve. But over time, as we worked with the system more and more, we started to lose faith in a lot of the mechanisms that we were sort of piling on, probably it seems increasingly fragile and risky. It had a lot of dependencies on markets that we didn't like and that sort of thing. We iteratively started taking these pieces out one by one and eventually, we ended up with a system, Ampleforth, that is so simple and sort of straightforward. We decided this just deserves its own name and so we came up with the Ampleforth name instead of the other name.
CR: Okay, perfect. And by this time was the ICO 2017 bubble over? So, because I see like a lot of building happens in that kind of 2018 bear market.
BI: Exactly, yeah. So, by the time we officially started, so we first started exploring ideas in the summer of 2017. But it wasn't until the end of that year and then we officially started in January 2018. So, we were definitely in the territory at that point.
CR: Yeah. And was it just you and Evan or had you grown the team by this point?
BI: I mean, at that point, it was just me and Evan, and then Jessica was also helping with some of the business aspects of it. And so, we started with three of us, two of us being technical and then we hired a first engineer. Eventually, some other people I worked with at Google and Uber also came over to join us.
CR: Nice. And did you have fundraising from VCs or were you bootstrapping?
BI: We did. We went the traditional VC route. We raised a seed fund from some very high quality folks like True Ventures, Founders Collective, and Pantera. We did think an ICO, that was definitely an option to us. But given the choice between getting just funds or funds with the advice of very smart, respected people, I think we chose a second one, which is the traditional VC route, which thankfully we had access to just given location where we were. I think that was definitely the right move.
Monetary Asset
CR: Now, can you get into how Ampleforth works?
BI: Sure, yeah. Ampleforth is the name of the protocol and then the token is called AMPL, pronounced “ample.” AMPL is a monetary asset like Bitcoin, but it's meant to solve the problems that come along with a fixed supply asset. So, Bitcoin as you know has a maximum supply of 21 million. It’s somewhat analogous to gold. But the demand for holding currency fluctuates, it goes up and down naturally with trade cycles, with booms and busts, with technological advancements. And so, when you have a fixed supply asset like that, a change in demand gets expressed entirely in the change of the price of that currency. We've seen this play out historically with Bitcoin, the prices has been incredibly volatile, going as high as $20,000 to as low as $3,000 and then back up to somewhere in the middle. So, it's not that surprising that we've seen that behavior.
“Ampleforth is the name of the protocol and then the token is called AMPL, pronounced “ample.” AMPL is a monetary asset like Bitcoin, but it's meant to solve the problems that come along with a fixed supply asset.”
So, when the price of that currency changes, that also means that it gets manifested in the changes of prices and things denominated in that currency, so goods and services, but also, most importantly, contracts and debt. If I were to take a loan from my friend for one Bitcoin to be paid back a year from now, I don't know exactly how much in terms of raw purchasing power I'm actually going to have to pay back at that point. I maybe on the hook for potentially twice or more.
This can lead to defaults, which is not good, especially when you have various pieces stacked on top of each other, which we’ve absolutely seen today in DeFi. One of the greatest things about DeFi is the composability aspect and sort of the Lego blocks you can piece together and the stack seems to be getting precariously high in the last few weeks or so. A default at the bottom layer can cause cascading failures all up the stack and reverberate through many other areas, which is not a healthy thing to do.
Image source: ampleforth.org
Systemic Risk
So, there are some great properties of fixed supply, like Bitcoin has and that's one of the things that people like about it. It's non-dilutive. It can't be inflated away by someone else who's outside of your control. So far, it's nice. But it's not able to support a complex economy on top, financial instruments on top. There's too much exposure to risk of cascading failures in what you might call systemic risk.
“(Bitcoin) is not able to support a complex economy on top, financial instruments on top. There's too much exposure to risk of cascading failures in what you might call systemic risk.”
CR: What about like compared with Ether and which is, I mean, if we're talking about DeFi, I think maybe Ether is kind of a good comparison to make. So, Ether, it doesn't have a fixed supply, but it's still really volatile like Bitcoin. So, how would Ampleforth compare with ETH?
BI: So, ETH is a different sort of asset with a different sort of monetary policy. So, on Eth1, it has a sort of linear inflation schedule, so it's rules based, which is good. There's no person or group behind the curtain, pulling levers to make things go up or down, that sort of thing. So, it's predictable. But there's no feedback cycle between the needs of the economy and the issuance of the currency. And then there's also one other difference that I think is important between ETH and BTC. Whereas ETH has use value, whereas BTC does not.
CR: What do you mean that?
BI: So, ETH, you can use to pay for gas to run computations on the platform. BTC is used to be BTC and nothing else. BTC has no real use value, the same way that something like oil would. So, oil, you can use for industrial applications and running your car. When you're talking about purely monetary assets, a use value, in some ways is a liability, because an outside source of demand that can impact the price of the currency. And so, in one hand, it gives it some grounding of value; on the other hand, it's another source of volatility that maybe you want in a pure currency. So, the US dollar, for example, also has no use value, which is kind of a nice property.
“When you're talking about purely monetary assets, a use value (like paying for gas with ETH), in some ways is a liability, because an outside source of demand that can impact the price of the currency.”
Not Pegged, Not Collateralized
And so, Bitcoin, I would say is defined primarily by its monetary policy’s fixed supply. ETH, I think is defined primarily, and people might disagree with this, but I think ETH is defined primarily by its ability to be used to pay for computation. We could still change monetary policy at ETH, and it will be ETH. But if you change anything on that with BTC, it would no longer be BTC.
So, I think they are somewhat different classes of assets, but they do have some similarities. Ampleforth is really closer to BTC in the sense that there's no use value and so there's no outside source of volatility.
CR: Okay. And so how is it designed to be a stablecoin? I understand it's not supposed to be pegged to the US dollar, but does it have any sort of functionality that makes it not be as volatile as other cryptocurrencies?
BI: Yeah, well, it's more a question of how that volatility is expressed, I guess in Ampleforth. So, the AMPL token has what we call a long run price circuit. So, we say that it’s not pegged per se, because there's no underlying redeemability for the US dollar the way there is for Tether or TrueUSD or anything like that. AMPL is non-collateralized the same way that Bitcoin or Ethereum is non-collateralized.
“We say AMPL is not pegged per se, because there's no underlying redeemability for the US dollar the way there is for Tether or TrueUSD. AMPL is non-collateralized the same way that Bitcoin or Ethereum is non-collateralized.”
When the price of the AMPL token in the marketplace is higher than the price target, then it means we need to increase the overall supply of AMPL to help bring the price back down to the target. And then when the price of the AMPL token is below the target price, then it means we need to reduce the supply of AMPL token to bring the price back up.
So, those are the two core rules. If we're above the price, we expand; if we're below the price, we contract. Now, the way the protocol does this is rather unique; in the sense that when there's a supply adjustment, it happens universally to all parties in the system. It happens proportionally to what you own. So, what this means is that the number of tokens you own in your wallet can change day to day.
CR: The number or the value?
BI: Both, right. The protocol adjusts the supply tokens, but its actors in the marketplace incorporate the supply back into price. So, this is the feedback loop. In a perfect world, if we could control the price of something in the marketplace, that'd be great. But you know, history has shown when governors try to ignore market forces, bad things happen. So, all we can do is let the market play out, but make supply adjustments to help it reach back into the equilibrium point where one token has the price target.
“The protocol adjusts the supply tokens, but its actors in the marketplace incorporate the supply back into price.”
CR: Okay. And so, how is this supply manipulated? Is it a smart contract that automatically kind of detects when a level has been reached and issues more AMPL or buys more AMPL from the market? Like how is it technically done?
Built on Ethereum
BI: Exactly, yeah. So, right now, Ampleforth is put onto the Ethereum blockchain. We made a decision early in the project not to build our own blockchain. But instead of creating our own silo, we preferred from the very beginning to be in the ecosystem where the most number of people are making transactions, and storing value. So, instead of building our own blockchain, we decided to deploy onto another blockchain Ethereum. Potentially, we could be also on other blockchains in the future as a sort of cross chain asset.
But the rules, the protocol, as you said, are deployed and codified into a smart contract on chain. And so, the idea is that the system is rules-based completely, with no room for discretionary policy.
And so, there’re three main components to the system. There's the monetary policy that has the rules about how to react to the marketplace and adjust the supply. There's the token itself, which is an ERC20 token that allows you to make transactions. Then there are the oracles that take the market data and then also the CPI index, the Consumer Price Index, it feeds that into the monetary policy, so that it’s able to make these decisions.
CR: What oracle are you using?
BI: So, we have two oracles. The first one provides the 24-hour volume weighted average price of AMPL versus the US dollar. The second oracle provides the PCE, the Personal Consumption Expenditures. So, this is a measure of consumer price index that’s published by the Bureau of Economic Analysis. And what this does is it allows us to target the 2019 US dollar. So, AMPL doesn't even target the US dollar, it targets the US dollar at a particular point in time...
CR: Okay. So, that's why you mentioned that CPI was another component. Why did you decide to do that, target the US dollar only in 2019?
BI: We wanted the value of AMPL to remain constant throughout at all time, is the idea. So, the value of the dollar naturally depreciates through the tools available of the central banking system. As long as this inflation is moderate, then it's no problem at all. But it does change the value of the US dollar over the years. One Ample targets about $1.01, right? Five years from now, one Ample might be $1.07. So, the idea is that we have a consistent value and unit of account through time.
Not Fans of the Word Stablecoin
CR: That's an interesting difference with most stablecoins, which reflect the value of the dollar. But in crypto many people are in this space because they disagree with central banks inflating their currencies. So that's always been a little bit of a disconnect, I think between crypto and the huge popularity of stablecoins still tracking the US dollar which is what people are supposed to be getting away from in the first place.
BI: Yeah, so back around 2015 or so, stablecoin meant something very different than it seems to now. It sort of represented the holy grail of crypto, the cryptocurrency with a predictable purchasing power.
But then, over time, as the needs of traders became very important for the development ecosystem, these exchanges needed some representation of US dollar to trade against, so stable coins became to mean a representation of the dollar on the blockchain.We were not a huge fan of the word stablecoin. I think, it's actually kind of a misguided framing. I like to ask, is the US dollar a stablecoin? And people don't really know how to answer that precisely because it is what stablecoins target. But is it a stablecoin? It's not really a stablecoin. It gives it stability through denomination and usage through pricing, denominating wages and contracts and that sort of thing, which gives it some nominal rigidity. And so, it's those market dynamics that give the US dollar most of its stability. So, the circulating supply of US dollar can change, but it takes a while until it actually propagates through the prices.
“We were not a huge fan of the word stablecoin. I think, it's actually kind of a misguided framing (…) back around 2015 or so, stablecoin represented the holy grail of crypto, the cryptocurrency with a predictable purchasing power. But then as the needs of traders became very important for the development ecosystem stablecoins became to mean a representation of the dollar on the blockchain.”
[ … ]
Paid subscribers have access to the full transcript, including sections on:
Why some volatility is necessary
Ampleforth’s minimized governance model
How the Bitfinex IEO was a double-edged sword
The under-appreciated risk in DeFi
Why AMPL is a good collateral candidate
Ampleforth’s new liquidity mining program
The project’s roadmap
Why yield farming is both exciting and worrisome
Subscribe now so you don’t miss any of The Defiant content. Subscribers reading this post: Head to posts marked with the little lock to see the full content.
The Defiant is a daily newsletter focusing on decentralized finance, a new financial system that’s being built on top of open blockchains. The space is evolving at breakneck speed and revolutionizing tech and money. Sign up to learn more and keep up on the latest, most interesting developments. Subscribers get full access at $10/month or $100/year, while free signups get only part of the content.
Click here to pay with DAI.There’s a limited amount of OG Memberships at 70 Dai per annual subscription ($100/yr normal price).
About the founder: I’m Camila Russo, a financial journalist writing a book on Ethereum with Harper Collins. (Pre-order The Infinite Machine here). 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.