Join Delphi Research today and immediately get access to our full Member Portal!
Join Delphi Research today and immediately get access to our full Member Portal!

Terra Autumn Series Ep. 1: Kickstarting Terra Autumn with Do Kwon

Oct 4, 2021 · 58 min media

By Jose Maria Macedo

Jose Maria Macedo sits down with Do Kwon, Co-Founder and CEO of Terraform Labs, the team behind the Terra blockchain, a fully decentralized and programmable money for the internet. The two discuss decentralized stablecoin mechanisms and pitfalls, Terra’s core ethos, regulatory risk for Terra, and much more!



Show Notes: 

(00:00:00) – Introduction.

(00:01:04) – Do’s background. 

(00:03:40) – The importance of stablecoins.

(00:06:26) – An overview of stablecoins.

(00:11:09) – Overcoming the challenges of an algorithmic stablecoin.

(00:16:01) – Making $UST the most useful form of money. 

(00:20:12) – Terra’s DeFi ecosystem. 

(00:24:40) – The metaverse on Terra. 

(00:27:21) – Terra’s scalability.

(00:30:22) – Terra’s main goal / Getting $UST on other chains.

(00:33:40) – Using $UST for payments / CHAI payments.

(00:39:45) – Regulatory risk for Terra.

(00:45:14) – Terraform Lab’s plan for their treasury.

(00:46:53) – Sustainability of the Terra ecosystem without incentives.

(00:50:08) – Managing the concentrated liquidity trade-off.

(00:51:34) – Do’s biggest concerns for Terra. 

(00:53:20) – Closing thoughts from Do.

Interview Transcript:

Jose (00:01):

All right. Hey, Do. Great to have you here, kicking off our Terra podcast series. As you know, we’re huge fans of what you’re building with Terra. We think it’s one of the most important and also underrated protocols in crypto. Over the next week, we’re going to be talking to builders from all the major protocols on Terra to give people the intro into what Terra is, why it matters.

Jose (00:19):

Also give them a glimpse as to all the innovation that’s happening on there. So before we start and just for full disclosure, Delphi Ventures holds a position in LUNA and Delphi Labs is actively incubating a bunch of different projects on Terra. So many of our listeners will have watched my real vision interview with you. So I’ll try not to repeat too much content, but there’s probably going to be some overlap in terms of LUNA’s design and stuff like that.

Jose (00:40):

But we’re also going to cover some new ground especially towards the end. So for today’s interview, I’d like to try and explain, first of all, why decentralized stablecoins matter, why UST is the only decentralized stable coin, but also how you think about building on Terra, regulatory challenges and some of the long term vision for it.

Jose (00:57):

So to start with, first of all, thanks very much for being here. It’s awesome to have you here and then to start with, could you give us a quick background on yourself, how you got into crypto, but also why you chose to start Terra? You had a great thread on it recently. I’d love to hit on that.

Do (01:16):

Sure. I’m an engineer by trade. So I studied computer science at Stanford and then after graduation, I was an engineer at Microsoft for a while. I had the great fortune of being placed in a really boring team. So they weren’t giving me too much work. So in my spare time, I worked on a lot of side projects. Then one of them that was sufficiently compelling enough to get me to leave the company and to start my own project was something I do with WiFi mesh networking technology.

Do (01:47):

So I developed software whereby user devices like phones and laptops could connect to each other. WiFi direct and Bluetooth, such that even if you don’t have direct internet access, you could piggyback off of a multi hop network of the peers around you to get internet access. So we package that as a B2B networking solution to large enterprises like amusement parks and shopping malls and others.

Do (02:13):

So while doing that you can imagine that a lot of standard search queries for mesh networks or distributed networks returns hits like bitcoin and Ethereum. Then that’s how I got sucked into crypto. So I spent a lot of time better parts of 2016 and 2017 looking into white papers, going to meetups and just found myself spending more and more time in the crypto industry.

Do (02:39):

So what was really interesting at that time was that it was always confusing to me for an industry that calls itself, the cryptocurrency industry, that there were very few people working on the problem of making currencies better. Most people were either working on very niche apps. Things like betting markets or, different types of decentralized exchanges or things like that. Or they were focused on building better blockchains in terms of making Ethereum faster.

Do (03:10):

But I felt like the most important problem to work on is to think about a situation where a cryptocurrency could be more useful than fiat currency or bitcoin. I thought about that as the Holy Grail of cryptocurrency. That was the founding motivation for Terra and it is still the mission that we operate by today.

Jose (03:34):

Your intuition was definitely right. Stablecoins have since I think arguably become crypto’s killer app. Why are stablecoins so important? Why would you choose, I guess, a stablecoin as the design for this hyper useful currency? Why have they gained so much traction in your opinion?

Do (03:55):

I think the most important feature of stablecoin is that it allows users and most users in the world to nominate either their net worth or their financial holdings in context of the fiat currency where they happen to be domiciled in. So for lots of people in the US and the Western world, it is the US dollar. In China it’s the Yuan. In Korea is the Korean Won and so on and so forth.

Do (04:23):

So in order to build, let’s say deep IX apps on top of blockchains, it’s natural that the base currency that constitutes the liquidity for all of these different applications, be denominated in that same currency. So if you look at how crypto operates today, most of the crypto applications either poke prices or execute trades with the stablecoin as the base Terra. That’s why most people believe the stablecoin so far has been the killer after crypto.

Jose (04:57):

Do you think that will always be the case? Like in a, let’s say like a hyper bitcolonization world from the maxis or in a world where US investors control of its monetary policy, do you think it will always be denominated in dollar, or are you interested in some of these experiments, like RY and these non-stablecoins or like different ways to denominate?

Do (05:21):

I think I’m more or plausible case is that fiat currencies tend to peg its… I can see a version of the world where the monetary policy paradigm changes such that the fiat currencies themselves peg to things other than the CPI. So for example, like the US dollar can at some later point in time, choose to peg itself to some combination of a basket that includes bitcoin for instance.

Do (05:44):

But I think fiat currencies will always remain in some form or fashion. So for example, when the US dollar was tethered to gold, it was still the US dollar. Just because the pegging mechanism changed does not mean that the US government collects taxes in gold or other types of currency. So in so far as most users in the world are paid wages in something that’s fiat denominated and have to pay taxes in something that’s fiat denominated, I think stable points will always have a place irrespective of how much adoption other cryptocurrencies gain.

Jose (06:17):

Makes sense. I agree with that. I think for the foreseeable future fiat-pegged stablecoins are going to be the dominant way people denominate and the medium of exchange. Could you give us a quick breakdown as to how current stablecoins work, what the different architectures are and the risks of each one?

Do (06:34):

Sure. So there’s a lot of newer experiments that are happening in stablecoins these days, but as a very broad categorization, there’s three categories of stablecoins. So number one is fiat backed stablecoins. So these are basically tokenized dollars. So the premises that some issuer let’s say, a tether incorporated, or circle keeps a dollar in the bank account for every stablecoin that they issue.

Do (07:02):

So the basic premise is, is that a user should have redemption rights to a dollar to which he can redeem that to his bank account for every stablecoin that he happens to be holding.

Do (07:13):

The second design is something that follows the paradigm of MakerDAO. So the idea is that for every single DAI stablecoin that is minted in the market, there is over collateralization in some sort of volatile crypto asset that’s backing it. So the idea is that at any given time you can redeem one stablecoin for something that’s a little bit more than a dollar. So that’s what allows the stablecoin to keep its [inaudible 00:07:40].

Do (07:40):

The last category of stablecoin are algorithmic stablecoins. This is the category that Terra operates in. The idea is that we use a set of on-chain incentives to allow the stablecoin to maintain its price peg against dollar. So in terms of the trade-offs between the different designs, for fiat backed stablecoins the first category, the problem is that there’s always risk in sort of the integrity and the security of the issuer or the regulators that seek to bring that issuer under their purview.

Do (08:12):

So for example to no one’s surprise, people are starting to figure out that, coins like USDC and USDT do not actually have a dollar in the bank account for every stablecoin that’s issued. So there’s large integrity risk on the part of the issuers as to whether they’re actually keeping dollars in the bank account. What happens if the bank accounts get seized or the bank fails? Then in that case, the entire stablecoin ecosystem collapses.

Do (08:39):

Another problem is that depending on where the issuer or the bank accounts might happen to be domicile, the integrity of those bank accounts are held hostage by the regulators that are overseeing those accounts. For example if you have a bank account in Taiwan as the issuer, then in some cases, all the different types of DeFi apps and other things that use these stablecoins are held hostage to the whims of the Taiwanese government.

Do (09:07):

The risks of the second model is a little bit more nuanced, but the problem here is that the model works pretty well in terms of making sure that there’s enough collateral to back each stablecoin issued. But the issue here is that the expansionary pressure for stablecoins do not actually keep in line with how much demand there is for stablecoins.

Do (09:30):

So for MakerDAO when it had a single collateral model, it had the problem that people only wanted to mint Dai when they wanted to take out a leverage long position on Ethereum. Not when there was an additional demand pressure to purchase Dai from the open market. This led to a supply and demand mismatch, which meant that there was always an over demand for Dai tokens from the market, which led to Dai consistently trading at a slight premium to the dollar for most of its history.

Do (10:00):

The risks of the third model is that, it relies on a set of on-chain incentives. So in terms of how these incentives work given that there’s no explicit collateral backing every stablecoin that’s issued, there’s a risk that these on-chain incentives could fail and the stablecoin could be deep end in the open market.

Jose (10:27):

Absolutely. I guess I’d add for the second model, for the MakerDAO model, because of the issues that you point out, they’ve also had to onboard different types of collateral, which themselves are sensible. So most of Dai is actually backed by USDC, which has the problems of the first model that you mentioned in terms of counterparty risk and regulatory risk, where circle can basically freeze the USDC holdings in MakerDAO vaults and MakerDAO insolvent.

Jose (10:54):

So you have a decentralized stablecoin that’s built on top of a centralized stablecoin, just ends up inheriting the problems of the centralized stablecoin. That was really great overview.

Jose (11:08):

In terms of I guess the trade-off between the second design and the third design, which are the only two decentralized ones, you mentioned the MakerDAO debt based design. It’s hard to expand, right? Because you need demand for debt to keep going up which is uncorrelated to demand for stable coin. So you end up with this mismatch.

Jose (11:30):

Then with the algorithmic design it relies on-chain incentive. So there’s always this fear of the death loop. Can you explain maybe the death loop a little bit and how different algorithmic stablecoins try to overcome this and maybe how LUNA has managed to really be the only one that’s really been able to avoid this fate?

Do (11:54):

So most algorithmic stablecoins follow the same design or a very similar design. So generally there’s a dual token model whereby there’s a stablecoin, and there’s a second token, which gives you redemption rights to that stablecoin. So for example, how Terra works is that at any given time, a user can trade in one Terra USD for a dollars worth of LUNA and visa versa. You can trade in $1 worth of LUNA for one Terra USD.

Do (12:27):

During a time when you want to redeem one Terra USD for one LUNA in the open market, you just need to burn one Terra. Then it gives you $1 worth of LUNA, which you can sell to acquire the actual US dollar, bitcoin or anything that you want. So the problem here could potentially manifest when there are too many people trying to redeem stablecoins at the same time, i.e a sharp contraction in the stablecoin economy.

Do (12:53):

There could be a lot of people that are rushing to the exits. So what happens here is that if a lot of people are swapping Terra USD to LUNA, the value of LUNA could be slipping as people are looking to trade out of it. This increases beta risk for people that are swapping out of Terra stablecoins. So this could lead to a situation that is commonly referred to as a debt spiral.

Do (13:15):

In the sense that as the stablecoin is being redeemed, the price peg of the stablecoin starts to seep because the cost of doing this redemption transaction increases, which leads to more people wanting to swap out of the stablecoin and so on and so forth until the economy collapses. This has in fact happened to many earlier algorithmic stablecoin efforts.

Do (13:39):

So the reason why Terra has been resilient to a lot of these contractions and in fact, in May, we had a 25% contraction in the overall Terra stablecoin economy, when US piece market cap was around $2 billion in the grand crypto crash in May, we saw about $400 million of Terra stablecoins cells and redemptions hitting the market, which was absolutely massive.

Do (14:07):

We handled our prices pretty well. So as swaps were happening we saw the price peg of Terra USD seep 6 and 7% for a period of few days, but then it recovered as redemptions started to smooth out in the open market. I think the reason that I would have to give as to why Terra is more resilient than other types algorithmic stablecoins is because there’s a vibrant economy that is being built on the Terra blockchain.

Do (14:39):

I think that’s the best defense that algorithmic stablecoins can have against that spirals.

Jose (14:44):

Yeah, absolutely. Because I guess most of the algorithm stablecoins that came before and I proudly lost money on a lot of them. They tried to fix this with incentives, right? So having some mechanism design where below a dollar, there’s some incentive for people to come in and buy bonds, which will be redeemed later, but it ultimately relies on faith by speculators and investors. Because the algo stable itself doesn’t have utility.

Jose (15:10):

Whereas I guess the approach you’ve taken is you do have some smart mechanisms like increasing the tax rate on transactions when UST is below a dollar and stuff like that. But you also have, I think the key, which is creating real demand for the stablecoin, which is also what we saw during the May crash, where there was a big contraction because there was a broader crypto crash.

Jose (15:31):

That’s how these death loops generally are initiated, but you did see for instance UST and Anchor go up, which is huge because it means that a lot of people are fleeing to safety, which they see as UST and they see as UST on Anchor. No one’s fleeing to like ESD during a crash or to any of the other algo stables. I thought that was pretty huge.

Jose (15:54):

That goes into the way you think about what you’re doing with Terra. Could you maybe talk a bit about, like, you’ve talked about your mission at TFL is helping make UST the most decentralized and useful form of money there is, I guess, can you talk a bit about how you’re doing this with CHAI, with TFi and Mirror and all of that and how you think about your role?

Do (16:16):

So to speak a little bit more about why algo stablecoins fail is that the launch sequence for most algo stablecoins look exactly the same. I think this is consistent across most DeFi launches, but this is especially problematic for algorithmic stablecoins. So generally the scheme boils down to look, if you buy and hold some of our algo stablecoins in let’s say a pull zero, we’re going to give you more algo stablecoins.

Do (16:45):

So these are recursive holding benefits that help bootstrap the TVL of these stablecoin projects to let’s say, hundreds of millions or billions of dollars very quickly. But the problem here is that there’s no reason to hold these algorithmic stablecoins when there’s a loss in faith in the yield or some bright, new shiny thing comes along, that offers better yield.

Do (17:05):

Then in that case, all the capital fleas, the stablecoin barms at once, and then this leads the entire stablecoin peg to come crashing down because there’s no consistent source of demand that that has formed at the base of these stablecoins. At Terra we operate things a little bit differently. So the way that we look at it is that there should be no recursive holding incentives.

Do (17:29):

The only reason why users want to hold Terra is because it’s a more useful form of money compared to fiat currency. That’s actually pretty simple to do because money is a product with very simple features. You can either spend it, you can use it to invest in things and you can hold it.

Do (17:47):

So basically you need to offer better savings experience. Whereby users can earn a higher yield by holding Terra stablecoins versus keeping money in a Wells Fargo savings account. Second, you need to be able to use these stablecoins to invest in any type of asset class in the world, be it equities or ETFs or different types of [inaudible 00:18:06].

Do (18:08):

Third, these stablecoins need to be easy to spend. So if you make money in the first two categories, you need to figure out a way where you can help to pivot users money at rest into money in motion. So this involves different types of payment apps and different types of integrations like that. So at TFL, we focus on making sure that these three features of Terra stablecoins as money is better than opportunity costs or other types of opportunities that people can get in theo currencies or other types of stablecoins.

Do (18:43):

So this is why for savings we built Anchor. So Anchor is a protocol whereby people can deposit Terra stablecoins to earn a high stable yield powered by block rewards that are coming in from multiple different proof of state blockchains. So right now Anchor offers 19.5% yield on stablecoins, which is obviously much, much higher than what you can get at the savings bank or other lending protocols that you find in DeFi.

Do (19:12):

For investing that’s why we started to work on efforts such as Mirror protocol. So this is a synthetics protocol whereby users can use Terra stablecoins to invest into any asset class that they want. So you can buy price exposure through mirrored synthetics for let’s say equities, different types of cryptocurrencies and ETFs.

Do (19:34):

Lastly, we built out a number of different payment apps, such that Terra stablecoins are easier to spend. These are reflected in CHAI and [inaudible 00:19:42] in Korea, MemePay and Mongolia and some other countries.

Jose (19:47):

Awesome. I guess you mentioned the three goals there, so savings, investment and spending. Let’s split them up, I guess the first two more concern Terra’s DeFi ecosystem, and then the third is separate and requires on ramps in each country and stuff like that. So maybe we can go through those in turn.

Jose (20:12):

So in terms of the layer one DeFi ecosystem, what’s the traction like, how are you thinking about it? Maybe just give people an overview of what’s going on there. Because I think most people don’t realize that Terra’s actually the third biggest layer one ecosystem by TVL. It’d be interesting to give people that background.

Do (20:35):

It kind of goes up every day, so it’s hard to quote exact numbers. But Anchor for instance, launched in February of this year. So today Anchor has $4 billion of TVL locked in the platform in either borrower or deposits. Month over month Anchor is scoring at maybe 60, 70% month over month. So pretty quickly.

Do (21:03):

Mirror protocol, for instance, for synthetics has about $2 billion in TVL locked up across various different types of synthetics, such as mirrored Apple, mirrored [inaudible 00:21:15], mirrored Robinhood and so on and so forth. So there’s a vibrant synthetic shall we say, equities or trading experience that’s happening there. There’s also a number of very interesting earlier projects.

Do (21:28):

So Loda for instance allows people to lock off funds and then invest in lossless lotteries through the yield that is being generated in the Terra DeFi ecosystem. There’s ApolloDAO, which was one of the winners of the previous Delphi hackathon. So they build a yield aggregator on top of Terra stablecoins. So in a relatively short period of time the Terra DeFi ecosystem really only started in 2021 with the launch of Mirror.

Do (21:57):

So in the first three quarters of this year, we were able to start from zero to collectively, let’s say $7, $8 billion in TVL and a vibrant ecosystem that encompasses synthetics, savings, decentralized exchanges, betting and all sorts of different things with a ton of developers joining the platform every day.

Jose (22:20):

Absolutely. There’s also some interesting stuff happening in the NFT space and stuff. I think there’s some interesting innovation our outside of finance that’s starting to seep into Terra. How do you see that and how does it fit into the three goals that you mentioned before?

Do (22:41):

I think NFTs represent a really interesting opportunity. I don’t think those opportunities are being tapped into today with color drops on Ethereum. But I think the real opportunity here is allowing digital property rights to transition from communism to capitalism. So what I mean by this is that human incentives in the real world operate via private property rights.

Do (23:02):

So it always relies on a conception of scarcity, but our digital experience doesn’t really abide by the conception of scarcity because digital goods are almost always fundamentally not scarce. So this means that in all the metaverses that we participate in, be it let’s say LinkedIn or Twitter, there’s a grand equality that applies to all sorts of property rights.

Do (23:26):

Which means that anyone can create the same property. So if you’re on Twitter, like a billionaire has the same digital property, as let’s say somebody that has $20 in the bank account, which is great. But this also means that it makes it difficult for people to try to cultivate property or economic identity on top of these metaverses. NFTs change this.

Do (23:51):

So for example, if we had a set of NFTs whose property rights are well respected across multiple different metaverses that we operate in, then in that case that creates a conception of scarcity in a digital context. It allows people to take the one version that only existed in the real world and move it into the metaverse.

Do (24:12):

You can imagine a situation where you have avatars and NFT properties that you migrate from, let’s say, an MMRPG game, and then you bring it into your SNS service. Then you might be able to take it into, let’s say your email and so on and so forth. I think that’s going to be the connective economic tissue that constitutes all metaversus that we operate in together.

Jose (24:37):

Very interesting. I have very similar thoughts on that. I guess, how do you think about that existing on Terra? Because there’s, I guess, limited block space on Terra. How do you think about what should exist there and what are you focusing on and how does it fit into the goals that you mentioned before having this metaverse available?

Jose (24:58):

Because it’s like most of the stuff that you get on Terra so far, the DeFi stuff, there’s a clear use case for why it makes UST as money more useful. Whereas NFTs and metaverse, you start to seep into more of a pure layer one, where you’re just becoming a decentralized database for these metaverses to run on. How do you think about those two goals?

Do (25:23):

Let me put it this way. So how everybody in the NFT space operates today is that they’re really focused on supplying more NFTs. Which means that everybody’s really busy creating new NFTs that have the same limited features, because there’s a lot of money to be made in that market for now. But I think a bigger opportunity that is being neglected is giving people an opportunity to use these NFTs.

Do (25:49):

I think that’s super interesting. So for example, if you had some experience whereby you can either… so it’s like let’s say there’s a shopping mall metaverse where you can showcase your crypto punks or pudgy penguins, or whatever NFTs that you purchased and then showcase them, you can import them into this metaverse and then interact with them in a way simply more than posting as your Twitter profile pick. Then that could get a lot of value locked up in this metaverse.

Do (26:25):

What you could do with that attention, where people are spending lots of time is that you can add a commerce layer to it. So for example, if there’s a metaverse whereby people are bringing in tons of crypto punks and pudgy penguins and ether rocks and things like that, then in that case an interesting experiment here could be in this metaverse can we add, let’s say a movie theater, let’s say in partnership with Netflix, then have people pay for that service with UST?

Do (26:53):

Or for example, if there was like a bartering marketplace within this metaverse, can we get UST to be the core currency for all those markets? Those are the types of opportunities that we’re interested in.

Jose (27:05):

That makes sense. So it’s facilitating that third goal, but rather than real world spending is like this digital spending. So you build a metaverse in order to make UST the base currency and decentralized currency of this metaverse. That makes sense. I guess it’s relevant to this. How do you think about scalability because for crypto to do all the things that it’s meant to do and also for Terra to facilitate, or like, for instance, the metaverse use case that you’re talking about, there’s clearly a need for greater scalability. Do you have a plan for that? Is it something that you think about actively or?

Do (27:44):

Sure. Terra is built upon Tendermint from the Cosmos SDK and the Cosmos vision is pretty simple. Every application should have its own blockchain and then all these app specific blockchain should be connected to each other via a bridge or a messaging protocol called IBC. So the entire premise of the Cosmos infrastructure is a world whereby different blockchains can interoperate with each other.

Do (28:16):

I actually think that the branding of IBC could be better because it’s actually a great scaling solution. So one way that you can think about this is that not all different types of smart contracts and applications need to interface with each other. Logically, a lot of these things are separate. So for example, a lot of different DeFi+S might have to interact with Terraswap or Uniswap because they depend on that base infrastructure for liquidity.

Do (28:42):

But it makes perfect sense that something that has to do with payroll has nothing to do with that DeFi cluster of applications. Which means that we could logically separate them into charts for most intense and purposes. So what you could do is let’s say you can have a logically separated chart that computes a lot of DeFi settings.

Do (29:06):

Then you could have a logically separated chart that has something to do with NFTs and user types of activity, a separate start chart for content. Then you can have the main Terra blockchain be the connective tissue that binds all those different blockchains together. Then all these different blockchains can connect to each other via IBC.

Jose (29:28):

Very nice. Sounds like what Cosmos was meant to be. I guess, what you see is building your own ecosystem of DAPS and of different experiences and then having Terra validate across them in a parallelized way or?

Do (29:49):

I would say this is pretty consistent with the Cosmos vision. I don’t think it’s necessary competitive with what Cosmos was initially trying to do, but I think the idea is that it’s just a change of perspective. So interoperability and scalability are not that different things because you logically separate state machines anyway. But I think it makes more sense to call it a scaling solution.

Jose (30:20):

That’s super interesting. One thing I wanted to dig into as well is you’ve been pushing pretty hard and successfully to get UST on other chains, right? You guys backed Wormhole, you’ve been doing incentives on [inaudible 00:30:34] which is portfolio company, by the way, and I think it now has more liquidity on salina than it does on Ethereum or at least it did a few weeks ago.

Jose (30:43):

How do you see that go, I guess getting UST on other chains, and do you think there’s a scenario where Terra can succeed, even if it fails as a layer one ecosystem, just because it’s so prevalent on other chains?

Do (31:01):

I mean, this is like a confession but Terra became successful as a layer one almost purely by accident. So our main goal was to make sure that we could create use cases around Terra USD. In the beginning, when the Terra blockchain first launched, it didn’t have a smart contract layer. But in order to build use cases around UST, we needed smart contracts. So we added a Cosmos module on top of it.

Do (31:28):

Then we built out the initial applications, Mirror and Anchor on top of Terra and that gave the inspiration for tons of developers to either build apps that are adjacent to Mirror or Anchor or to build something on top of it. So our recent success in terms of rising TVL and developer attention wasn’t designed from the beginning. It just happened as an accident as we were trying to build first use cases around USD.

Do (31:55):

But the way that we look at it is, most people look at Terra and they’re confused by what we’re trying to do. Because we had built and successfully spun out a payments company. It is a layer one. We have stablecoins that sprawl across multiple different ecosystems. I think it’s like four or five blockchains at this point. So it’s hard to pinpoint what exactly Terra’s trying to do, but at TFL there’s no confusion about what we want to do here.

Do (32:23):

Our main goal, and the only product that we look at is the Terra stablecoin. Our goal is to make that stablecoin the most decentralized and useful form of money there is. For everything else it’s just a means to an end. So if Terra succeeds as a layer one smart culture platform, then great why not. But from our perspective, we place no difference between UST being used on the Terra blockchain versus Ethereum or solana or avalanche.

Do (32:56):

We’re pretty agnostic to that insofar as lots of users are gaining benefit by using the Terra stablecoins. We are indifferent to be context in which these stable points are used.

Jose (33:08):

That’s very cool. So Terra’s product is UST and the smart contracting features of the layer one are just a way to bootstrap the utility of UST and make it more useful. But actually it’s possible that at some point if another layer one ecosystem ends up being the winner, whether it’s solano or Ethereum, whatever UST could still be the settlement currency of that ecosystem, as long as it wins as a decentralized stable, even if Terra as a layer one doesn’t win.

Do (33:35):


Jose (33:37):

That’s interesting. That’s super interesting. Then we’ve spent a bit of time on Terra’s finance ecosystem. I’d like to spend some time on the spending because for me, they’re like two parallel tracks. One of them is making Terra more useful as investment in savings and sorry, UST more useful for investment, for savings, for financial primitives, with all the stuff that’s happening there.

Jose (34:00):

Then the other one is making it more useful in the real world in terms of actual payments. There’s a lot happening there as well. I guess at some point the plan would be to link them so that you have like this stablecoin that’s hyper useful as part of the centralized financial system. Then you can also spend it everywhere.

Jose (34:17):

Maybe could you give some update on what’s happening on the spending side with CHAI, cash ,Alice, all of those?

Do (34:26):

Sure. The motivation for wanting to start payment companies was pretty simple. So I wanted to get to a state where I could simply do all my finances in the entire blockchain and then never have to actually touch a bank account. For me, unless I have to get some paperwork signed or something like that, I don’t go to the bank. So for example, I take on my staking rewards and then convert that to KRT, pop that up into CHAI and then make purchases for whatever I need across restaurants, general purpose e-commerce, food delivery all from the CHAI app or by using the CHAI card.

Do (35:03):

So we wanted to close the retail loop on what people can do with their money. So in terms of the things that we’ve been doing with payments is about two years ago, we created a payments app called CHAI which was powered by the Terra blockchain. Then we started building integrations with the pop e-commerce merchants in South Korea.

Do (35:26):

So today CHAI operates across about 50 of the largest e-commerce merchants in the country, as well as a few offline chains, such as like the largest convenience store or the largest movie theater, bookstore and so on and so forth. Then in terms of online e-commerce, it fills out the entire gamut in terms of the most popular internet shopping businesses in the country.

Do (35:50):

So you can use it to shop, let’s say baby diapers on Timon or order food from Yogiyo, or you can use it to buy online groceries from Market Curry. So pretty much full gamut as a Korean citizen, what you can do with KRT to buy lots of things.

Do (36:11):

Recently we’ve started to work with or observe the creation of different types of neobanks and wallets in different jurisdictions. So for example there’s Alice coming out of the US, which is a neobank where users can deposit UST to earn Anchor yield and then hook up those hook up that yield with a debit card that they can spend anywhere across the world.

Do (36:38):

There’s similar efforts such as [inaudible 00:36:41] in the Scandinavian regions. There’s Cake in Australia as well as tons of different wallets, payment providers and neobanks that are popping up across the world.

Jose (36:57):

That’s awesome. I didn’t even know about those. All of these will enable you to spend UST in their various jurisdictions. So they’re getting the appropriate licenses and payment processing licenses in order to be able to do that.

Do (37:11):


Jose (37:12):

Those are popping up independently, or are you like funding or encouraging those in some way?

Do (37:18):

To the best of my knowledge, there’s more than 10 different payments plays built on Terra that’s going on. Because that was a part of core to our DNA when we first got started. I did fund the Alice because the founder Icho was a intern at Terra TFL for a while. Then he left TFL to start this business, but a relatively small stakeholder.

Jose (37:47):

Very cool. When do you see the bridges happening between these payment apps and the TFi, Terra’s DeFi ecosystem? When do you see, for instance like an Anchor or Mirror integration into CHAI or into some of these payments apps where people can pay through them, but they can also save, they can also have access to investments and all of that? Do you see that coming anytime soon?

Do (38:14):

Yeah. So for CHAI Terra Station is already integrated. So if you deep link Terra Station once, then you sign a special transaction that allows CHAI to withdraw up to X amount in Terra stablecoins, then in that case you can shop on CHAI without having to pop a manually from Terra Station. So I guess a low hanging fruit that we can do is to integrate Anchor into Station.

Do (38:40):

I think we’re actually working on that now, in which case, you could be earning Anchor yield while at the same time shopping in CHAI.

Do (38:48):

In terms of a native deposit experience in the CHAI wallet itself, there’s a new license that’s coming out in September 25th in Korea, in which case payment providers would be able to offer some crypto facility insofar as they meet the proper requirements, which CHAI already does. So I think starting at Q4 this year, we will start to meet some of the qualifications for a native deposit experience. I think we’ll be able to do lots of interesting things.

Jose (39:18):

That’s super cool.

Do (39:19):

In terms of Alice they have an alpha out so I think they’re gearing up to a public launch in November. So that should be exciting. There’s a few other neobanks that are in extreme stuff, but I think those should be rolling [inaudible 00:39:35] as well.

Jose (39:37):

That’s super exciting. I mean, that’s super exciting. So to pivot over to some of the risks and especially the regulatory side, earlier this year you tweeted that crypto is at war. What did you mean by this and I guess, how do we win this war?

Do (40:01):

So not surprising to myself or other founders in DeFi, but recently there’s been a lot of chatter about stablecoins and crypto having to be more regulated, especially from the United States. So for example, Janet Yellen made some comments about how stablecoins have to be regulated by the US treasury and then the creation of new stablecoins has to be licensed by the approval of the Secretary of the Treasury.

Do (40:35):

Now, the interesting thing is the comments didn’t pertain to only US pegged stable points, but any fiat-pegged stablecoins. Which was interesting that the US Treasury was trying to claim jurisdiction over stablecoins that had nothing to do with the sovereign currency.

Do (40:52):

But I think one of the things that’s going to happen as a result of this within, let’s say the next 24 months is that I think there’s going to be a lot of regulatory tension on centralized stablecoins. Where the underlying bank deposits are under direct purview of regulations in each nation state.

Do (41:15):

I think what that means is that all the DeFi applications that are built on top of centralized stablecoins are now a censorship risk because it’s trivial to, let’s say Censor, Avi or Compound some other protocol, if the regulator could control the base layer money that is locked up in most of these protocols. So they don’t even have to touch the decentralized protocol that sits on top of it.

Do (41:41):

If they censor the base their money, then they can bring the entire thing crushing down. So I think this is why it’s really important for crypto to decentralize or die. So every protocol should have radical decentralization in their mind or be prepared to fail. I think a huge component of that is making sure that the DeFi stack is built as much as possible on decentralized stablecoins like UST. That’s how-

Jose (42:08):

A 100% agree. I think it’s inevitable and I think the low hanging fruit will definitely be the centralized stablecoins where there’s a bank account, there’s an entity, the entity actually has the ability to freeze accounts. So it’s very easy for a state to coerce them, whether it’s by freezing their bank account or by just telling them freeze this wallet and all of DeFi is at risk.

Jose (42:33):

Most of DeFi primitives, whether it’s Curve, Avi, heavily reliant on centralized stablecoins, which if they were frozen would lead to insolvency. I agree with that, but interestingly the language that was used as you mentioned was pretty broad to the extent that it could apply to something like UST or anything that’s pegged to the dollar Gensler asserted mandate over.

Jose (42:58):

How do you think about those risks for Terra? To what extent is Terra vulnerable to a regulatory attack and what are you doing to decentralize or die?

Do (43:11):

So an interesting component of how Terra works is that there’s no centralized minter. So anybody can mint Terra stablecoins by burning an equivalent value of LUNA. In order for regulators to claim that people that are minting Terra are broker dealers, or they need some sort of licensed capacity, in an open protocol like Terra, is that they would need to bring action against, let’s say [Hunsha 00:43:36] in China, or like Ken, who is scratching his butt in his mom’s basement while he’s minting Terra stablecoins in Ohio or something like that.

Do (43:45):

So if you have a large community of people that are just taking the right and interfacing with an open protocol, then it’s just not possible to regulate. Like TFL plays no part in people minting stablecoins and then transacting with it in whatever sense. We don’t even operate any nodes or validators on top of a protocol. I think from that we have a massive advantage over centralized stablecoins.

Do (44:15):

In terms of the steps that we are taking to decentralize, I think this is going to be a gradual process. But one of the things that I stress to people that work at TFL is that TFL is a transient aberration. So it’s sole purpose is to make sure that the Terra ecosystem can get to a state where it gets to be a self sustaining economy.

Do (44:35):

In the sense that even with that TLF’s help, there’s always a new influx of people that are building use cases and apps on top of Terra stablecoins, that Terra economy is in a constant increase. Like the stability mechanism has been stress tested to such a level where it doesn’t need people that are solving coordination problems around, let’s say network upgrades or R&D about how to improve the mechanism.

Do (45:00):

In which case TFL will finally dissolve. We will either find ways of seeding our assets to the community or to burn our remaining assets and we will just disappear.

Jose (45:12):

Interesting. One of the questions we actually got from one of our subscribers as well was what’s Terraform Labs plan for the treasury that it holds right now and LUNA and UST? Is there any plan to diversify away from those assets in an effort to I guess be a buyer of last resort for the peg? What’s your general thoughts on treasury use long term?

Do (45:41):

These days I’m starting to notice that $150 million rounds are small. So one of the things that I was thinking about is to do raises with the stability reserve that we are custody. Then use the proceeds from that to fund, let’s say stability pools against let’s say other stable assets such that, then people have more optionality in terms of redeeming out of their Terra stablecoins into other assets.

Do (46:12):

They can add or redeem out into LUNA, which is the base layer stability mechanism, but they can also have a stablecoin forex market whereby with deep liquidity whereby they can transition out of Terra stablecoins into let’s say well, knock on wood USDT or USDC.

Jose (46:33):

That’d be pretty cool. That’s super interesting. What’s the timeline? Is this just an idea or are you looking at doing this soon?

Do (46:44):

It’s in the idea phase right now, but find me a lead investor and we’ll make it happen.

Jose (46:52):

I think we know one. Then the other thing people asked is a lot of Terra’s success right now, but I think this applies to most layer one ecosystems other than eth is sustained by incentives, right? Whether it’s like the yields are sustained by Mirror incentives or Anchor’s borrowing is being subsidized NC tokens. What do you say to people that worry about the sustainability of that long term and what are the plans to have real demand come in to substitute that source of yield?

Do (47:27):

Sure. I think a different response is warranted for each of Anchor and Mirror and Pylon, because I think each protocol operates a little bit differently. But just to talk about Anchor and Mirror. So for Anchor, I think the ANC incentives was just a temporary incentive until Anchor can operate across multiple different POS staking derivatives.

Do (47:54):

So in the beginning when we first started, we just had LUNA as the accepted collateral type within Anchor. Now we have things like bonded Ethereum and we’re adding bonded [inaudible 00:48:04] as well. Then eventually we’ll add a bunch of different POS staking derivatives, such as bonded atom, bonded [inaudible 00:48:13] and so on and so forth.

Do (48:14):

So I think the ANC incentives, a, traits token holders from multiple different ecosystems to interact with Anchor. So I think that’s a huge plus. But I don’t think ANC incentives would need to consist in perpetuity in [inaudible 00:48:29] protocol to work. So at steady state, the way that I’m thinking about it is that the borrower is able to take out a loan in Anchor without having to pay explicit borrowing costs.

Do (48:40):

Then just the yield that is accruing to staking derivative collateral gets converted to the lender in the form of established interest rate. So in some sense, Anchor becomes a staking yield stabilization protocol. I think after that ANC incentives wouldn’t need to continue.

Do (48:59):

For Mirror, I think the MIR emissions play a heavier role than do ANC emissions in Anchor. But I think eventually the state that Mirror needs to get to is that mirrored assets need to operate over concentrated liquidity. So right now mirrored assets are on Uniswap B2 and Terraswap. So liquidity is being provided for all possible changes in asset values.

Do (49:26):

But that’s a waste of money because for example, like the price of Apple stock is not going to drop to zero overnight. It’s just not going to happen. So in that case, the concentrated liquidity model makes a lot of sense because you just need to make sure that liquidity is available in a very tight end around the current stock price, maybe 10% up 10% down, or maybe a little bit above that for more volatile assets.

Do (49:49):

Then that should increase as the yield that goes to liquidity providers. I think once we have that concentrated liquidity model and there’s more apps that leverage mirrored assets, there should be efficient returns from liquidity providers, even if there aren’t incentives.

Jose (50:07):

Makes a lot of sense. I know we’re working on this right now, but how do you think about the concentrated liquidity, the trade off between capital efficiency, but then also the composability, I guess? Because you’d lose the ability for LPs to just passively stake their LP tokens and earn a yield and you’d need to have either strategies. Some kind of active strategies that they’re honored as the canonical LP token that are made into CW20s.

Jose (50:33):

Or you need to have these professional LPs come in that are doing that work for users. How do you think about that?

Do (50:43):

As you know, as report is building in concentrated liquidity for the early version of the protocol… was that not [inaudible 00:50:53]?

Jose (50:53):

No, no, it’s fine. That’s totally fine.

Do (50:58):

So concentrate on liquidity rumored to be coming to as report and some of the things that we’re building on top of that is an auto rebalancing there. So the idea is that users can simply deposit their tokens into this post similar to how they will provide liquidity to places like Terra as well.

Do (51:15):

Then depending on how prices are moving periodically, rebalancing occurs to form liquidity around the current spot, such that liquidity providers aren’t missing out on that sweet composability they’ve come to expect and love [inaudible 00:51:32].

Jose (51:33):

Very cool. Well, this was awesome. I want to ask maybe one last question. If I came to you from the future and told you that LUNA had failed, what would think the most likely reason is? What’s the stuff that’s keeping you up at night, that you’re most worried about if anything? You don’t strike me as much of a worrier to be honest, but what are the things that most worry you?

Do (51:57):

Well, I sleep like a baby, so there isn’t too much of that. I don’t know, like a million lunatics get hit on the same day with the bus. I guess there’s that. So look, I mean, there’s couple of failure points that I spend a lot of time thinking about. So right now a lot of TVL in Terra is concentrated across a couple of protocols.

Do (52:22):

Our smart contract failure or some economic failure, one of them could have pretty severe consequences for the underlying stablecoin economy. Something that I would really like to do is to get to a state where no one protocol is holding too much assets in UST. So diversification is definitely important. It’s similar to how if Samsung failed, the Korean economy would collapse and that’s what would happen to the Terra economy if Anchor collapsed. Right?

Do (52:56):

I just found out today that Samsung pays one third of all taxes in the country. It’s like 30% of the GDP.

Jose (53:03):

That’s crazy.

Do (53:03):

It’s crazy. So diversification is important. I think we need to keep up developer mind share so that there’s lots of chains that want to integrate with UST, lots of DeFi apps and of course, lots of developers that want to build using UST.

Jose (53:20):

That’s awesome. Do you have any last things you want to mention or anything that’s coming soon to Terra that people should be looking out for? Otherwise, we can call it. This was super cool.

Do (53:32):

I mean, folks, if you’re building a crypto application, don’t worry about getting rug pulled by some DeFi developer. If you keep using centralized stablecoins, they’re eventually going to rug you. So might as well start using UST now.

Jose (53:47):

100% back that sentiment. Well, thanks very much for being here Do. I think our subscribers are going to really like this and it was a great kickoff to the Terra series.

Do (53:57):

Yeah, for sure. Thanks for having me.