Exploring the Decentralized Stablecoin Design Space

Transcript

Introduction - Yield vs Minimizing Losses

the way that I approach the stable coin question is like I'm personally not as interested in like getting 7% yield instead of 4% yield the one the thing that drives me is not getting minus 100% yield um and you know there have been plenty of stable coins that have given people that um so like basically the way that I think about this is uh if the customer is not even necessarily users but applications right and as an application you know you're not able to kind of quickly between Things based off of uh know changing opinions about uh like what makes um which providers trustworthy which providers is not trustworthy you want to be decentralized yourself and like what is a mechanism

that you can like set down like be sort of minimal as credibly neutral as possible and be something that like you can make a five fiveyear bet on poly Market with right uh so I think uh basically the way that I think about this is that there's three major types of stable coins right so there's centralized stable coins so we know of you know usdc USD bu USD and so on and then you have what I call like Dow Dow governed real world asset backed stable coins and so D is uh the most famous example here though I think there's uh definitely room to try to innovate and make more examples and then there is governance minimized stable

Types of Stablecoins - From Centralized to Fully Decentralized

coins that are like very fully decentralized and Ry is the best example there right so I think the second category is interesting right and uh it's interesting because there's a lot of uh value that you can get out of like actually Connecting People directly into diff real world assets and uh like real world assets are ultimately the productive economy the point of a financial system should ultimately be to support the uh productive economy there's a lot of different things there that uh have different properties in terms of in terms of stability different properties

RWA-Backed Stablecoins and Their Risks

in terms of generating Revenue different jurisdictional properties um so like what principles do I think it makes sense for a dow governed real world asset stable coin to follow right I think to me one big one is uh total excess in collateral should be greater than the amount held by any single issuer or in any single jurisdiction right so this is uh like pretty basic if you care build decentralization right you don't want the thing to break solvency even if like any one specific um issue or breaks right and I mean diet self right actually historically has not um really uh uh satisfied this well right when uh black Thursday happened

back in 2020 they basically went immediately from being eback to being like basically a front end for usdc right and uh I think to me it's like if you're willing to trust a one single actor in order to be solvent then like you might as well just have a centralized stable coin and uh to me the goal of this is to basically have a governance structure that gives you some kind of like long-term reliability that you have something that like actually accomplishes the core property of decentralization which is basically no single issuer going wrong or no single jurisdiction going wrong should be able to like actually break the core

guarantee and the core guarantee is one unit of the stable coin is worth like a dollar or like one of whatever you're tracking right so like that like basically know rule of thumb in terms of collateral and that's like a very simple rule of thumb if you want to like go fancy and mathematically optimal about this then obviously yeah you know you go into like the the entire theory of like how Insurance works and uh you know you you start thinking about like correlations and what and uncorrelated things and like what kind of collateral ratios are needed based on like level of decentralization how correlated the risk is and you know there's entire textbooks on this I'm sure that I that I haven't read but if you want to make something perfect you probably should but like you

know basic rule of thumb start here right then obviously standard requirements for a d right basically you wants to avoid governance attacks and uh like to me this is still a very unex explored part of the design space right because uh like most people it like it feels like we have this effect that I really hate in the crypto space which is like you know how there's this meme that like nobody got fired for buying IBM and it's like in the crypto space the meme is nobody got fired for foring compound uh but and like we've gotten this like meme in our heads that the only way to be decentralized is to have token voting and like token voting is an awful mechanism right token all voting gets

The Limitations of Token Voting

easily it's like basically one of two things easily dominated by very small highly centralized group or it's uh some you get something with a tragedy of the commons effect and like sometimes you'll need a small you only need a small percent of the supply to take over the thing and it's like even in theory you have this uh you you have this problem that okay I'm trying to figure out if that was a metaphor for something um like even in theory you have this problem that like if the the the market cap of your assets under management cannot exceed the market cap of your governance token because someone could just buy the governance token and then cause all the oracles to like go to zero and uh or whatever numbers uh allow them

to steal all the assets under management and then you're screwed right and uh like to me the only way to break past that barrier is to have some form of non-f financialized governance right so at the very least like you want some kind of delay mechanism right like you know if you have to have a governance token you have them vote on oracles and then you have a rule you can change at least like maximum one of like say 13 oracles every week and that way you know that even if a governance attack starts you still have like six weeks of on hisorical within which you have to like get your money out and switch to something else right so avoiding governance attacks avoiding um you know vote buying avoiding like voter non participation also at the same time like actually ensuring quality though you

could also make an argument that like it's a standard argument that the thing that democracy is worst at is like making optimal decisions the thing that democracy is best at is like making mid decisions but avoiding terrible ones and maybe that's like exactly what a stable coin requires right because it's like if I have a stable coin I want to freaking not lose my money if I'm okay losing all my money for the sake of like better returns I just going to hold e right uh so um right and then there's like interesting ideas about futarchy like voting Dynamics right which is basically that you know if you vote for a delegate who vote for who like make some decision that's Upstream of a collateral asset

failing then your tokens get slashed right and the nice thing is that that's something that you can like kind of see and kind of uh enforce in the uh in the protocol which is interesting um so let's get into the like fully decentralized stable coin world right so what I always found interesting about Ry is like it really tries to do the minimized governance thing right like the goal of Ry is basically that the governance does exactly one thing and that exactly one thing is managing a price Oracle right and uh Ry has this very simple model where there is a Target price and that's a basically a Target price of Ry so there's a Target

Governance-Minimized Stablecoins and Wind-down Mechanisms

price of Ry denominated in US Dollars and then the oracle tells you the USD eth price and then you can look up the Ry eth price on Unis swap so like the app itself can just go in and do that and then you just like multiply one by the other and you can see whether the actual price is lower or higher than the Target and what you also have is you have like a built-in interest rates right so the Target price adjusts by like some percent per percent per year and if the actual price goes too high then the uh interest rate starts going lower if the Target price goes to goes low then the uh interest rate starts going higher again right so that's like a pretty simple mechanism right it's how Ry works and the basic idea is basically

that like what you want is you want to find an equilibrium right you want to find the equilibrium interest rate at which the number of people willing to hold positive units of your current of your stable coin and equals the number of people willing to hold negative units of your stable coin right uh so and know rise been running for a couple of years and it seems like it actually does that so I wrote this post about two years ago on thought experiments for evaluating decentralized stable coins right and I had two thought experiments right so one is is a safe wind down possible Right is it possible for a system to safely wind down to zero users in such a way that no

one loses money and uh obviously Luna is like one big example of this not being true right so uh who you remembers Luna from the moon so you know let's uh basically you know first we Bas the way that Luna worked is uh it was this like system where it's like actually kind of similar to this the soraj Sher paper by Robert Sams back from 2014 uh where you you have two coins you have this small too coin in the stable coin and you can like constantly trade one for the other but the key property is that like the total a quantity of stable coins that gets issued is positive and those tokens get backed by the volatile coin right and if

the total issued quantity of positive uh of of stable coins ends up like actually exceeding the uh market cap of the volatile coin then like actually you're screwed right and the system has this potential to like basically flip and completely crash right and the problem is that Luna like basically did everything that they could to like exacerbate this risk right because they started like adding these Pony Dynamics they started adding this mechanism where you can like stake your tokens you can get 19% aprs and like if someone offers you 19% aprs there are like two exactly two possibilities one is that they're scamming you and the other is that that it's denominated in a fiat currency

that's scamming you uh so basic but you know the a lot of people were more optimistic and then they ended up being very wrong and people lost like $30 billion right and and because of this a lot of people dis ended up like being afraid of algorithmic stable coins as a concept but to me that's the wrong lesson right to me it's uh that you want to be afraid of a specific type of design of algorithmic stable coins and the rule that would have prevented this situation completely is that if you have a decentralized stable coin the the total quantity of stable coins issued has to be zero that means is for every positive one you issue you have to

issue a negative one right and this is actually something that D does it's something that Ry does um and basically you know you keep the uh you know you keep the amount issued and you keep the debt exactly the same and that way you guarantee that you can pass the windown test because during a wind down you basically just like match every positive with a negative and then everyone gets back the the value that they had and then it all comes out of collateral so that's experiment one thought experiment two which is like what happens if demand for holding the E stable coin exceeds demand for holding negative amounts of the stable coin even at a zero interest rate right and this was actually the

Exploring Fully Decentralized Stablecoin Models

thing that caused D to like pretty chaotically flip over from being ethb to being a usdc front end back in 2020 right which is basically that there was not enough demand to hold negative units of d that were back that were backed by eth and so basically demand for positive die was high but then demand for putting in collateral was low and if they had done nothing the price of D would have like started going up and D would have like entered a speculative bubble and so just like basically quickly scramble and like throw in huge amounts of usdc as collateral right and uh I think right now this is one of those things that's easy to forget because uh if uh right

now we're in this relatively High interest rate regime but macroeconomic conditions can change lots of things can happen maybe in 2027 we're going to be back in negative interest rate land like actually people love making you know like all kinds of prognostications that are like super confident but like really nobody has a freaking clue right so I think any protocol that exists now like it has to you have to keep in mind the possibility that you are going to have a sustained uh negative interest rate regime at least at four some portion of the time during which the protocol is operational and you basically have two possibilities right one is you accept the possibility of your stable coin

having a negative interest rate and so in that case you accept a floating Peg and like you can change it around on the UI level right the other thing you can do is you can have like an erc20 where the negative interest rate is kind of baked in so basically every time you do Ascend the negative interest rate gets like calculated first and then you lose that extra amounts but then it's like basically you know you uh you do end up having to like suffer one one uh sort of ux issue over the other and the other possibility is like just accept the risk that your stable coin is going to turn into a bubble or like turn into something else that's bad in this situation right and like personally out of this uh I prefer option one um so

that's um like another part of of the Yes stable question I think another thing that's really interesting is this question of what index do you follow right so basically yeah if we wants to accept the uh user experience and the accounting downsides of not just perfectly following USD um then there's an argument that like maybe you want to actually go further and we want to actually be less like skom morphic with these things right so skoric is this like term for ux Design Land that's like basically making like computer user experience like items things that are that like really try to look like physical things that they're modeled

Decentralized CPI Indices for Stability

after right and uh the ultim basically the equivalent here right is that ski morphism and stable coins is giving people US dollars because people are used to US Dollars now the argument here is well what if we don't be skew morphic and what if we take a first principles approach to a stable coin right if I hold a stable coin what do I like actually want like if I want stability what do I actually want and the answer basically is that I want to guarantee that if I have K times n units I can pay for n days of my expenses right like I think that's ultimately what uh stability really means from just like a very practical perspective right and so

the conclusion of this was like you make an index that actually tracks people's expenses um so decentralized CPI is you know a meme and it's like starting to become an actual thing I mean I've talked to like even people here working on decentralized cpis I've uh you know I've seen people elsewhere I mean fra is uh I mean they're just like basically have a pointer to the Federal Reserve CPI um so like a lot of different people working on uh different indices and like today I personally think that Federal Reserve numbers are reliable for inflation so sorry to like any Shadow stats fans in the room uh but uh you know there are tail risk situations where that like might end up actually

changing quickly and there's an argument that I think you wants to internationalize and you wants to like minimize dependency on the economic situation of any one country right and so if you do that then like maybe like these are arguments that like we actually do want some kind of crypto native decentralized CPI in some sense and like what would that mean right that could mean like you have a dow that manages oracles uh but then those instead of those oracles being dollars those oracles like actually are a basket of all kinds of different expenses that people have right uh so I think that's interesting I mean I uh I think there's

also another radical idea which is like you just uh create this like crazy multi-dimensional Market where each user chooses their own index out of uh like thousands of uh like basically literally everything that has any kind of price feed anywhere and then you have this multi-dimensional Market that just balances interest rates across all of these and you make UI wrappers that make it really easy for people to like make their own CPI out of whatever they want and then this like completely breaks F fungibility basically everyone has their own coin uh but then uh what you get out of and then the way that anyone actually pays anyone else is like you just basically shove every payments through like Unis swap or whatever and the

theory is that like this could actually work relatively fine with mature enough infrastructure probably aided by like AI or like something something something and like hackathon people figure it out right uh so so that's so I think it's another interesting topic right um so another topic is uh what's the collateral right so I think historically we've seen like basically a couple of opt really trying hard to be governance minimized and so Ry had eth as its collateral right eth credibly neutral asset um you know I don't know I guess eth used to be money eth is ticker now

What is the Collateral for Algorithmic Stablecoins

apparently uh so you know um but you know credibly neutral asset in the base of the ethereum ecosystem and it's great but just holding eth is uncompetitive if you miss out on all the on all the all yield and I think this ended up being a big reason for um like a rise underlying unpopularity right and like even though you know like I definitely don't believe in you know like risky yield maxing for stable coins I do think that you have to be reasonably yield competitive and the basic reason is like there's a lot of applications that break if you're not yield competitive right so like if you like from if you think about prediction

markets for example right so in prediction markets like about so like literally today right you can go on poly market and you can make make a bet you can bet that there will not be a civil war in the UK in the next uh 2 months and one day and if you make this bet then uh the price the price of that is 975 and so you can make 2 and a half% right so 2 and a half% in two months 15% yield minus you know risk of an actual Civil War which I think is like counts for like 1% and then minus the risk of Errors which is under 1% right so it's like you know like you can kind of make like 13% yield but then there's a lot of other questions that are like that where the yield of like take basic of like

actually betting against crazy people is like above zero but it's not high enough right and so basically what you get out of that is that you get like a bunch of PE a bunch of uh you know like third party candidates uh just uh that like actually stand no chance of winning just have like lingering percentage rates of like 1% and then a lot of people just get convinced that like oh this guy actually has a chance when like actually they don't right and so like to me it's like I think the over waiting small probabilities and under waiting probabilities that are close to one is like like basically everyone agrees that it's like the biggest systematic weakness of prediction markets and the problem of course is that you if arbitraging against it requires you to

like accept CRA like very crappy or like zero yield then people are not going to Arbitrage against it and so the crazy people are going to set the price right so the again you know you do want an asset that where like if the intended use case is poly market right you want a an asset where one the asset is the governance and like all of the different properties around it are robust enough that that people feel comfortable uh sticking money into it in 2024 and getting it back in 2029 but two you do want the yield to be competitive enough so that if you put the money in now in 2024 like as an arbitrage strategy then in 20 the basically the amount that you

lose between 2024 versus to 2029 versus an optimal portfolio is like relatively small so that Arbitrage strategies are actually possible across longtime Horizons right so this is you know basically the problem that Ry ran into and so now I mean they're doing high and there's other stable coins and they're getting into Stak eth the problem with Stak eth is of course like what is the governance min minimized eth staking solution right and uh like one approach is to just just like pick one right like everything into rocket pool or like everything into Lio or everything into something else and you know there's like centralization risks of this and know

you could make a plausible argument that like maybe one of these things should just create its own staking pool that's uh reasonably decentralized enough and that way like if you participate like you're not actually going to be increasing ethereum centralization on some but the problem like whatever you do once you get into the staked e game it's like either you are contributing at least somewhat to ethereum staking centralization or your thing is not governance minimal at all right and like that's one of the problems now there recently there's been a lot of interest in like I layer and avss and basically yeah use it like what that actually like works out to is you're basically saying

accept anyone's stake teeth as collateral the problem with accepting anyone's stake teeth as collateral is like there's this tail risk situation where what you're actually doing is you're like enabling attacks on ethereum that are cheaper because someone could go in like 51% attack and then if they 51% attack they're not going to 51% attack with your eth instead they are basically going to like put up their validators and then they're going to put their validators up into this thing and then they're going to 51% attack and then they're like going to get most of their money back because the like their the E that they have in that in that stake is not theirs it's like actually like lent out through these protocols right and so that's uh like basically

the risk that you're incurring if you just uh do this kind of let anyone participate approach right and then option three is um of course just like plain eth but then wait for the possibility that ethereum will will incorporate some kind of stake capping which will would realistically mean like much lower staking rates and then at the same time I know thanks to the verge you just you get much lower difficulty of staking and so instead of staking becoming this kind of like high difficulty uh relatively high yield thing it just becomes like a low difficulty low yield thing and then like some people who choose to participate can people who don't choose to participate including these protocols don't have to unlik if the penalty you

incur by holding plain eth versus Stak e is like 1% then like maybe for 5year poly Market questions that's kind of okay um so those are like some of the nuances around collateral so conclusion right so like Optimal open areas to explore and and this is uh especially if you're like really designing stable coins for the stability use case um so be one of them is governance uh mechanisms and I think a key question that's really important to think about is how do you achieve Security even above cryptoeconomic bounds right so how do we achieve Security even in a situation where your assets under management are greater than

Conclusion and Open Areas to Explore

the market cap of any protocol token that you have and potentially going even above that like just plain buying 50% of the tokens is one kind of attack another kind of attack is like vote bribing and so how do you try to achieve some degree of Security even under those kinds of situations so that's one question another question is like indices right so tracking the dollar at least for like decentralized stable coins is like one thing you could do and it's like kind of obvious but then if you're willing to incur the costs of like not being a perfect dollar then like actually there's a lot of other things that it make sense to track that are not dollars right and so this could mean CPI and potentially some kind of decentralized CPI it could

even mean like you have different indices right so you know like you one typee of that you could try to track is uh you know things like food and commodities another type of index that you could try to track is like rent prices in specific cities you can try to track a lot of different things right and then finally collateral types right and uh for collateral types uh like that kind of splits up into two sections right one section is like the very governance minimalistic approach in which case like basically the most governance minimized collateral with close to competitive yield and like is that eth is that like some specific version of staked eth is uh is that some kind of basket approach of different

types of eth if so like how is that being managed and then the other approach is this kind of maker Dow style approach where you have very active management and that puts money into things which could be onchain defi protocols and it could also be like offchain like real world you know government bonds corporate bonds whatever and uh then the question is like actually becomes you know like how do you choose these which ones of these do you exist like what kind of formula do you try to use how do you like balance between like yield and resilience to tail risks and uh how does how do you ensure that the governance of

like that whole thing actually really stays viable in the long term uh so yeah I think these are interesting questions and I you know I yeah look forward to more people trying to solve [Applause] them M um the toal M [Music] toal MH like 10 mil you know maybe of like Ry High and gry

and so and then total Mar of sa coins is like 170 Bill and so like that differentiation you know how do we get more people giving a about decentralized stable coins is it just about yield which I think like you know part of this presentation is like the yield is not competitive but like yeah I mean I think like the real one very just like realist answer is like yield being like at least greater than zero is is part of the answer right CU like it does seem like there's uh like legal barriers to usdc types offering greater than zero yield and like Ry Ry types actually do right so that's uh I think that that that's one part of the answer another

Q&A: Regulatory Challenges and Self-Regulation in DeFi

part of the answer is like uh it's like realistically if uh usdc like if the yield question did not exist right so if uh in the long run we're on average in a very low interest rate environment then like there is an argument uh that you know if usdc proves to be trustworthy over a hundred years what's the point of having anything other than usdc right and so the contrapositive of that is basically that like well any argument for things other than usdc like has to connect to like some like risk that people actually are concerned about that actually is connected to usdc and like I think political tail risks are definitely a valid argument right and like you know there's risks to usdc as a

whole and then there's also obviously know like risks in terms of like what holders will usdc continue being being friendly to under what conditions right so that and then with usdt it's like you know basically the same argument though I think for USD there's uh like it's definitely less transparent than than usdc right and so there's more uh arguments that like you know we actually want something where it's like very uh very clear like where every part of the collateral is coming from um so it is possible that the answer just is that like you you have projects that are

going to keep uh basically doing their thing with relatively small market caps uh until some kind of event happens that suddenly makes them interesting for people and then the other and then the other answer is like you know you don't need maximum yield but like actually creating a stable structure that gives people a couple percents that they would not have if they just held usdc and usdt is uh also like that like that's something that would attract users today um building off of James point that um decentralized stable coins only have like what like bips of a percentage of the um like total circulating supply

Q&A: Stablecoins as Yield Vehicle

of the centralized um of the more centralized stable coins do you think that it's better to like going off of what you were saying earlier um lean way more heavily into kind of like moving away from the skor fism so that like at the end of the day we all recognize that these are basically like power user instruments and when you lean into that and think about it as kind of like I don't know like a Vanguard retirement you know 2065 fund then we're kind of able to create a more like elegant solution ignor not ignoring but like accepting the fact that this this isn't necessarily something that people are supposed to be using for everyday payments but it's something of a as you were just saying like a you know set and forget type yield accumulation vehicle

yeah I think uh I mean I think that's definitely one uh one type of a good strategy that you can have I think uh one kind of nice complement to that on the payment side is that I think uh at least from a ux perspective we do need to be moving toward a payment world where it just becomes really easy to pay whatever currency you have and then it just automatically gets routed into whatever currency the uh recipient wants to receive and I think some of the uh discussion around like cross layer 2 stuff and uh like chain specific addresses and incorporating M 7683 into wallets and like that whole direction is already going to get us about halfway there right because it already like at

least sets the standard that uh a send is uh operation that could be executing complex logic involving like exchange and liquidity providers in the back end and then once you're at that point it's like relatively a smaller step until like extend it to say well actually if someone asks you for $100 and you don't have dollars then like the wallet should automatically offer to let you pay in like eth or Ry or you know CPI tokens or whatever else you happen to have right and if we have a world where uh like which uh like currency Network effects are less relevant from a payment

perspective then that gets us uh into a world that's like practically speaking more open for people to have different types of stable or approximately guess approximately stable assets and then like I I mean hopefully you know yes uh like we we will have more room for like more custom things like that to exist right and I think the there is just to think about is how globally speaking we already have a kind of ecosystem where we have lots of approximately stable assets and uh people can pay with whatever and receive whatever right because that's like how Fiat and the credit card system works already right um and uh I think if you compare to that

like the uh crypto world and like even with exchange spreads and all that is already very competitive mhm um yeah yeah thanks for the insightful talk and answers uh my question is uh regarding regulations of uh stable coins after the Tera Luna fell from the Moon uh and Regulators really activated in terms of um especially in Europe but I guess in other parts of of the world as well and uh my question is is it the time to really build decentralized table coins right now when when Regulators are so active in trying to capture this uh field thank

Q&A: Regulations and Decentralized Stablecoins

you I mean I my own impression is that like there's definitely like some level of uh like regulatory hawkishness on this but it's uh like it's definitely far from um you know like people trying to uh trying to ban the category right and I think there are things that the cryto space can do to try to like front run the situation and uh like prevent you know another Luna from happening and and making the that whole situation worse and I think the thing that we should do is like basically actually try to do the self-regulation thing and uh like basically like actually make the

information more accessible for stable coins of like what are the trust properties of a stable coin what are the assumptions that you're making what are things that might break on the governance side what are things that might break on the Oracle side um and the so like there was one s group know blue chip that's like starting to do that and I think it also probably makes sense to I me either you know like work with them or have another that's and like basically also try to cover sort of you know the more kind of defin native and like the more decentralized end of the spectrum and uh if we uh have like a pre uh like if we have our own pre-existing Frameworks where yeah know

there's clear sites that you can go to and you can see like you know these are the relatively safe stable coins and like these are the uh the the The Crazy Ones then you know if a uh if one of The Crazy Ones breaks then uh like our community will have a much stronger position uh to respond to that situation from so like that's one thing that I think like we could do and like actually yeah get started on more now to try to like make sure things don't get worse in that regard I think we have one more time for one more question I thanks uh might be a little uh dumb question but uh I find the argument that

Q&A: Yield vs Convenience for Adoption

uh decentralized stable coin adoption is low mhm uh because yield is low is sort of a very supply side focused argument um for me uh when I use these stable coins sometimes I just think of them as like a substitute for dollar but not chain uh and in that aspect there's no notion of yield like when I use that it's it's just something that I use and it's uh so I wonder what or how should I think about suppose something is very convenient to use and there's a lot of consumer demand for like the circulating ones right does that create uh pressure to to generate more adoption

on the supply side or are they completely sep like I I haven't thought about this I'm not sure what to think about it I see so I guess the question is like uh can you motivate people to hold these stable coins with convenience rather than with uh yield I I think it's uh I mean it's a valid point though I think you know you also wants to kind of game out like what this will look like I mean like 5 years from now or 10 years from now right and uh like the one outcome like I want to think about is the is the possibility that like the user paying with one thing and the recipient getting another thing is something that's going to be as frictional in uh the ethereum world as it was like as it is with credit cards

and you know and like with 10 times the lower spreads and if uh so like that's like one one thing that might change right and if in that case it's like convenience would not really be yeah like like the degree to which convenience is a property of an asset would decrease right because it's like every asset would have similar convenience properties the yeah I guess uh you know when I I mean realistically like when I think about why like I hold hold usdc it's like basically two reasons right one is just laziness like I receive it and then I know that I'm going to have future

expenses in it and then the uh and like if there was infrastructure to let me swap it into like dollars that I could trust that would have an expected yield greater than zero I probably would and then the other and then the other reason is just that like usdc is what poly Market accepts right no thank you so much