Building a Benchmark Rate and Interest Rate Derivatives in DeFi - TOKEN2049 London 2022

Well, we have a small and friendly audience here, so we'll make it a little more personal. Who among you has heard of Libor? Okay you has heard of Libor? Okay , keep your hand up if you actively participated in the LIBOR manipulation. Okay, maybe it was a bad joke to begin with, but it's a bit of history about bringing Libor back home.

Taking LIBOR and transforming what this benchmark rate is for DeFi, leveraging the best of the blockchain market to make it more transparent, open, and auditable. A little about me: I've been building in the crypto space for the past 11 years. Very happy to introduce you to Ipore, which truly represents the base layer of the DeFi credit market. So, Ipore is actually a protocol with two different parts. One, a reference rate and two, a set of interest rate derivatives.

If you 're here, you might know a little about DeFi credit markets. So we won't talk about innovation. Let's talk perhaps more about what's missing. We really lack this timeframe structure. We have these very innovative but very short-term debt markets, right?

And this is very interesting, but it's not really useful. Let's say, if you want to take out a mortgage loan from a DeFi bank. OK. Next, there is a lack of transparency in the rates. And that's partly due to the fragmentation between different markets.

But where does that lead? You have different rate behaviors in different markets and there is a real lack of tools to manage risk and enable arbitrage between these fragmented rates. If we think about how to build this base layer for DeFi credit markets, we need to look at two things from traditional finance. One is a benchmark interest rate and two, a set of interest rate derivatives that support these credit markets. Okay, if we look at traditional benchmark rates, let's talk about LIBOR and SOFR.

You have these banks or financial institutions that report their own rates. And if we look at the LIBOR scandal, we know that there is a problem of trust and self-reporting. But in the blockchain world, we have a much more transparent dynamic for this, which we will see in a second. But what is the publication time, for example in LIBOR, 11:00 a.m.?

m. BST? Is this the magic moment on which all credit markets should be based ? Or, if we look at something like SOFR, what does "from one day to the next" mean? Someone's "from one day to the next" is always someone else's daytime schedule .

And if we look at volume weighting, we can use BSBY, which weights all transactions on the Bloomberg terminal by volume. We can take these dynamics from traditional markets and apply DeFi mechanisms to make it more transparent and real-time. For example, in the DeFi sense, when moving from LIBOR to IOR, we replace banks and financial institutions with protocols, and these protocols are transparent. They are based on smart contracts. So, instead of asking someone to report their rates, we simply access smart contracts to understand what the market rate is.

Regarding the publication, again, someone's "from one day to the next" is always someone else's daytime schedule. Instead of having a geocentric publication time , we get what is the IOR, or interprotocol rate per block; It is published and calculated block by block. It's the closest thing to real-time that we can achieve in DeFi. And finally, we can clearly see the transactions that occur through smart contracts on the Ethereum blockchain. For example, Ethereum blockchain.

For example, how is the IOR index weighted? Well, once again, take a sample of these smart contract protocols. In its initial construction, it takes data from Aave and Compound, and technically it is not an offered rate. It's more of a mid-market rate, due to how the rate dynamics are structured. So it will be interesting to see how this index rate evolves in the future.

How are the rate behaviors of different protocols incorporated? OK. This index is also a public good. If we look at the lack of transparency or the "black boxes", we see the discontinuation of LIBOR. There are these long-term instruments that need to be liquidated, right?

But there really isn't much visibility about them. From a blockchain perspective, the IOR rate is published block by block, meaning it is a public good. This public good can be used by other protocols through a smart contract call to another. This same structure of debt, derivatives, and agreements can be maintained. Thus, these informal agreements, or perhaps a mortgage or auto loan structured from the IOR index, can be seen in the same way that loans for different stable currencies are structured.

And this is truly its value. We call the IPO the heartbeat of DeFi because it keeps the pulse of the DeFi credit markets block by block. And as you have this very open data framework and these instruments linked block after block, the real long-term value of the IPO instrument is how many of these other instruments are linked to it. What is the value? linked to it.

What is the value? What is the volume and velocity of these external instruments that clearly make these smart contract calls and basically function as a public good at the base of DeFi credit markets ? So, in reality, the IOR DAO may be one of the most interesting places where decentralized governance will be available to, say, prevent something like LIBOR manipulation. For example, how are new protocols added to the index? Well, that's a great question, and you're not asking me that .

It will actually be an IPO DAO governance decision. Okay, so how do you remove a protocol from the index? We know that DeFi protocols are innovative. They move fast and sometimes get hacked. So, in the event of a major hack, how would you remove a fee from the protocol?

How would you actually make this protocol multi-chain and capture fees from different money market protocols on another chain, for example, outside of Ethereum, on Avalanche or Algorand? I mean, this is a very interesting thing and a great use case for decentralized governance. for decentralized governance. How is the index rebalanced? So , how do you actually create a new index?

Currently there are IPO rates for USDT, USDC and DAI, but perhaps a new stablecoin will emerge with entirely different rate dynamics. OK. And finally, the last part deals with risk management in the instruments themselves. So, risk management: how do you change the structures and constants in AMM functions? How do you change the risk dynamics to determine how much risk the protocol can underwrite ?

Again, these will be decentralized governance decisions. OK. We believe that there is currently an active interest rate swap on Ethereum, which we consider the first instrument that truly references the IOR. It is a 28-day peer-to-pool cancelable swap peer-to-pool cancelable swap . Now, if you come from traditional finance, it might seem like a ridiculously short instrument and the optionality is very strange the optionality is very strange , but it's something that really matches the behavior of the DeFi credit market.

So, if we look at who the parties are here, instead of, say, a traditional swap, which is usually bank-to-bank, with a very fixed term structure and where the price is set by an order book or broker. Well, we really use the best we have in DeFi. The structure of Ipore swaps and the way to start a market where it has never existed is through a peer-to-pool construction. So this passive liquidity fund acts as a passive market maker and offers these swaps that can be taken by traders, where the traders are the counterparty. Instead of having this rigid structure, it's more like an RFQ or a request for quotation.

As soon as the trader takes possession of the instrument, the countdown to the 28- day expiration begins. So, again, this is about designing an instrument that fits the temperament, time frame structure, and behavior of today's DeFi users. And finally, the price is determined by an automated market maker. And really, this is the secret ingredient that sets Ipore apart. When creating an automated market maker for interest rates, which behave very differently from assets, we cannot use the simple formula xy = k that you find on Uniswap simple formula xy = k that you find on Uniswap .

Because? Because you're not setting the price based on volume or demand. You're really pricing it largely based on volatility. That's why we use, we're turning to the quantitative toolkit. We use something called Hull-White, which is the traditional quantitative model for the behavior of interest rates behavior of interest rates .

And then we borrowed other models that really match this DeFi behavior. For example, we use jump diffusion, which models these volatility spikes and then the reversion to the mean. And you can think of it as a kind of upward premium. And that's very interesting today, because the Ipore rate jumped from 2% to 40% and now, hopefully, it's reverting to the mean. It was close to 17% before he went on stage.

But this is the model we use to project these particular jumps. And then we use Longstaff-Schwartz, which is the assessment of optionality. If you perform all these calculations off-chain, you can do whatever you want, right? It's not reinventing the wheel. The real beauty of designing this instrument on-chain is making these complex mechanisms a good enough approximation to manage the risk of liquidity providers, while also being cheap enough to implement on-chain.

So taking an IOR swap right now isn't much more expensive than taking a Uniswap swap. And that is truly the engineering beauty behind the IPO protocol. Very good. So how do they test these models? Well, we only have about two and a half years of actual rate behavior in decentralized credit markets.

So, once again, we turn to the quantitative toolkit. We use something called the Hawkes process. What Hawkes' process does is analyze a set of data and extrapolate it over a long period of time. So we take this limited dataset that we have. We recreated it in several different situations and then ran the AMM models on these sets to understand how they work sets to understand how they work .

And all of this is done within an internal testing framework called ITF, which we use to perform retrospective testing of the models and see how they would behave on the chain. And this is really the interesting process of marrying the quantitative with what is in the chain to create these instruments. Good. OK. The key takeaway is, and probably will be, terrible English.

Please excuse me. I am American. But it's really about taking the best- known fundamentals of traditional finance. You combine them with all the capabilities you have in DeFi. You offer something that is genuinely useful and designed for the temperament of the people you are targeting, but you also provide the experience they are accustomed to in the traditional market accustomed to in the traditional market .

So when institutions come to manage their rate risk from traditional finance to DeFi, they get the same experience they 're used to. So, what is the future of the IPO protocol? We see this IOR protocol as the base layer of DeFi credit markets. If we visualize it as a software stack, this data layer is really the foundation. It is the transparent benchmark rate that everyone can link to.

We see this layer of derivatives as this stabilizing factor. And because DeFi is like building blocks of money, you can actually put them together in a couple of different contract calls to create structured products. And these are the DeFi Legos that will go into these structured products, different types of funds that we will be building in the future. And what about 5 years from now? Perhaps you ask a DeFi bank for a mortgage loan and that loan is priced according to the IPO index, and there are these interest rate derivatives that back the loan, and perhaps this whole process is completely invisible to the end user, and that's really the goal we have at IPO.

Okay, very quick about the team. Our team comes from a deep history. For example, our specialists have spent 20 years in traditional markets as fixed income analysts . our software developers. They come from enterprise software development and moved to blockchain.

They have built banking, payment, and insurance systems. Therefore, building these robust and scalable systems in DeFi is extremely important. Again, I've been building in the industry for 11 years. We have a very deep talent. And we have strong support from Arrington, GMI, CMT Digital and GSR.

So if you're interested in learning more about the IPO protocol, visit ipore.io. We have perhaps one of the best documentations in the industry best documentations in the industry . You can review the quantitative models behind the AMM and behind all the test suites. Come and talk to us.

We have a stand in there . Thank you so much. And uh