Possibility of negative token interest rate for loan tokens #947
Comments
|
@uzyn: Thanks for opening an issue, it is currently awaiting triage. The triage/accepted label can be added by foundation members by writing /triage accepted in a comment. DetailsI am a bot created to help the DeFiCh developers manage community feedback and contributions. You can check out my manifest file to understand my behavior and what I can do. If you want to use this for your project, you can check out the DeFiCh/oss-governance-bot repository. |
|
Has any well experienced finance or economist looked over this to doublecheck? |
|
What is preventing dusd to spike and liquidate most dusd vaults? |
|
@uzyn thanks for letting this happen so quick ! great work ! |
|
Sounds good! The dUSD price really prevents me of getting into stock-LM and "pushes" me into minting dUSD by myself. Even though I said "i'll not take a loan, i'll simply buy dUSD/dTOKEN on the DEX and do LM". |
I would say we would need to wait for the next voting round |
EDIT: misunderstood something - this comment is irrelevant!as I understand this, its only planned for DUSD, right? to be honest, I don't know exactly if I understand the whole thing - but this looks a little odd to me right now |
|
Having the dUSD at such a high premium makes it impossible for me to buy dUSD to pay back my loan without a huge loss, which poses a great risk for me, and at the same time having my funds blocked prevents me from participating in the Stock Liquidity Mining. So yes, we need this, and I hope we can have an extraordinary voting round asap and DON'T need to wait until the end of the month! |
|
sorry if I offended someone - didn't want this! |
|
This proposal should be prevented BY ANY MEANS. The price of DUSD reflects its value, which is significantly higher then the value of a dollar, as you can get extremely high interests. The price of DUSD will drop automatically if the interest rates drop over time. By printing unbacked DUSD you try to regulate a self-regulating free market. If this proposal is applied I hope there will be a chainsplit into DeFiChain and "FEDChain". Why do you want to manipulate the price of a token that has a significant higher value than its nominal value to its nominal value? DUSD is not a classical stablecoin. People who do not understand the difference between price and value simply should not go into DUSD. This proposal is VERY DANGEROUS!!! and destroys the otherwise free market. What will come next? Giving Stock holders free (unbacked) stocks when their price is higher than their nominal value? THIS MIGHT DESTROY THE WHOLE DEFICHAIN PROJECT!!! |
dUSD should reflect 1 USD |
|
Why not just add a dUsd/dUsdt and dUsd/dUsdc Pool Pairs? Should be enough to handle the stability of dUsd - without a fork. |
Great Idea! This would in fact help to accelerate the devaluation of DUSD without manipulating the free market. And it will help people to understand that DUSD is not a classical stablecoin (which most people obviously do not understand). |
Why? The price should reflect the value, which is obviously higher than 1 USD. The manipulation of the price will lead to an inflation (and thereby indirectly reducing the interest rate). Have a look at the GBTC discount. You can get "BTC" for 15% less of its USD-price. However, people do not buy it. Why? Because the market offers better fully backed BTC products. Let the market decide, don't copy the FED. |
|
The problem is that there is no free market. Because there is no secondary market for DUSD, it cannot balance itself out. The only way the rebalancing can take place is if everyone goes back through the same DUSD gate and corrects the price (but this can only be done by losing money). But if this is really the case, there must be a mistake in my thinking somewhere, because I cannot imagine that this was not foreseen by the people at DEFIChain. |
|
I think the negative funding rate is a great idea. It will allow price of DUSD to be slowly returned to $1. If USDT-DUSD liquidity is provided at this current state, there is a chance that DUSD still trades at a premium to USDT because the liquidity pool is isolated within the DeFiChain ecosystem. This means that unless a pool with sufficiently deep liquidity of USDT-DUSD is provided, the idea of a pool is unlikely to work as of the current status. A negative funding rate is ideal because it minimizes risk. By charging interest to borrow DUSD when borrowing is high and refunding negative interest when borrowing is too low, the system is less likely to experience price shocks, because there is still a lack of active arbitrating mechanism. Instead, arbitragers are now forced to play a longer term game in order to profit reliably from their positions. That said, the exact mathematical parameters are beyond my abilities, but what I can foresee happening is that if the magnitude of interest rate is too small, you might not have any visible impact in the short run, which could disincentivize the arbitrating mechanism. Too high and you risk over-correction, resulting in a stablecoin that keeps fluctuating back and forth the dollar peg. As for the other dTokens, I believe once DUSD is fixed, they should be able to be arbitraged effectively to oracle prices over time since the premium paid for these stocks currently are the indirect consequence of a demand issue for the stocks, which is also the supply side issue of the DUSD. Once the supply of DUSD is restored via incentives for people to mint DUSD, the premium for these stocks should go down as well. However, this argument assumes that all dTokens are currently traded at similar levels of premium which will wane equally when DUSD price peg is fixed. That assumption may not hold, which is why additional measures can be considered again once DUSD peg has been fixed. TLDR, imho, it would be optimal to introduce the funding rate mechanism, then add USDT-DUSD liquidity pools, then address dToken prices, if they remain far from the peg. |
|
Great to see some healthy discussions on going! Prioritising the design on this slightly higher - please keep the feedback coming :) |
dUSD should reflect 1 USD! 100% Agree! dUSD should refelct 1 USD. If that's not the case there will be a problem of onbording new users to the eco system. If 1 dUSD is more than 0.05% off to the USD we can kiss the idea of a decentralised stock exchange good bye. At the end of the day thats the utility of DefiChain ... not Staking Rewards (Negativ) interest rates are a super efficent way to do this - given that not a human actor is setting them but an algorithm. |
|
I think variable interest rates for minting DUSD resolves the actual issue. Additional trading pairs e.g. dUSDC - dUSD or dUSDT - dUSD would not resolve the high arbitrage issue. I do not see how additional pool-pairs for USDT and USDC increase the incentive to mint DUSD. There are a lot of people minting DUSD swap it to DFI and then swap it to USDC or USDT. Additional pool pairs would only shorten this way and may remove DFI from the "trading" chain. When there is not enough DUSD in the market (DUSD > 1 USD) additional DUSD is "created" by reducing the loan of a "DUSD"-Vault and therefore you have to payback less loan. When DUSD hits < 1 USD or near to 1 USD you have to pay interest on your DUSD-Loan and reducing the amount of DUSD in the market. I agree with @otto1593, at least no one would buy dTokens with a premium of 15 to 20 percentage only to hold a dToken like dTSLA - and why should another exchange list your dToken with such a high premium? EDIT: By now 56,377,451.62634288 (creation height: 1,367,556) are mintet (https://defiscan.live/tokens/15). Actually you have to pay 1 to 2 percent interest on 56 mio DUSD 560 k - 1,120 k interest per year. If the DUSD minters have to pay interest by themselves you would penalize minting DUSD and therefore no one would mint DUSD anymore and the price of DUSD would sky rocket. |
|
We also should talk about the deviation of the Stock DEX price vs. the Oracle Price. For me it's a feature that deviation is possible. Especially during times where there is no price feed available. But if the DEX separates too much from the oracle price, I also would support negative interest rates to get them back to the oracle price. At the moment we only have one side of the medal positiv interest rates ... |
|
@uzyn how you are going to calculate the token interest rate? The pink paper or your description does not specify that, does it? Maybe: Oracle's USD/DFI / DUSD/DFI on DEX - 1 Or even better use a new Oracle for DUSD/USD which contains multiple DUSD/USD calculations? |
|
@uzyn @chriger @otto1593 @Rno123 @Gainto : I assume there is some reason why people buy dStocks at a premium of around 50% (33% (USD=>DUSD) x 12.8% (DUSD=>dStock) = 50%) over the real world price, namely: they expect an APR that is higher than the "loss" (it's unlikely but it might be that they can sell back their dStocks with that 50% premium again, probably slightly less due to the decrease in LM rewards over time) they make. They know that this dStock has a higher value than its real world counterpart due to the fact that they can make profits by liquidity mining. Now, when you devaluate DUSD by 33% (to 1 USD), what do you assume will happen to the price and premium of that dStock w.r.t. DUSD? Will it go down? Or stay the same? Or go up? Why? I personally assume that the price w.r.t. DUSD will go up, as the value of that dStock stays the same (you still can get your LM rewards). So every (artificial) decrease in value of DUSD will cause an (nominal) increase in dStock prices. Now you will have to explain investors why they have to pay a 50% premium over the real world stock price. Until some smart guy comes up with the idea #1324 of also manipulating the dStock price by printing more (unbacked) dStocks. What are your thoughts on that? In my opinion, the idea technically is fascinating as I love these mechanics of decentralization technology. But it won't solve the problem that an asset that has a special quality (making profits by putting it into a liquidity pool) is not traded for its "nominal value" as it is a combination of the representation of the underlying asset PLUS its ability to create profit. I think it would be better to communicate transparently that the value of DUSD is higher because of it's special quality. Isn't it much cooler to have that special asset with a special price? Not trying to imitate the boring real world dollar? Just to explain investors then why they have to pay a 50% premium for dStocks? The premium of DUSD over USD is not a bug, it's a feature. Just sell it as such to new investors. If you adjust the DUSD premium you will just shift the problem to dStocks. |
|
Yes but the LM-Pools should provide liquidity to those who want to buy a dToken. Liquidity mining is not the main feature for the dTokens. The main goal is to give people access to tokens whose price reflects to the real world asset. By now you have high rewards in the dToken LM pools whose are decreasing over time until there is a point you will get commission for providing liquidity to a pool and no more rewards or a little fraction of the rewards today. When you only want to buy a dToken and only want to hold it, you wont buy it with a premium. No one who wants to buy and hold only dStocks (without LM) cares about the system and logic behind it, they only see: "TSLA is 20 percent more expensive, i wont buy it". |
|
@chriger so you do not think that the dStocks prices w.r.t. DUSD will go up if you inflate DUSD? Or that investors will think: "Ah, finally the DUSD price is at 1 USD. NOW I am willing to pay 50% premium on the dStocks." |
I understand you payed a premium and didn't mint your tokens to do LM. I hope for you the LM will compensate you for any loss you might have, if the overpriced asset you bought goes back to its fair value. LM is a mechanism, it's not the value DeFichain wants to provide. We want to open the investment opportunity of the stock and commodity market to people who are unbanked. Part of that is that, you also have to be able to exit your trade and store in a "stable" currency. There is no point in having DUSD if the price jumps around like crazy. That's a bug not a feature and it needs to be corrected to fulfill the value preposition of defichain. The reale thing we should talk about is, if NEGATIVE interest rates on DUSD are enough, or if we need them also for the commodity and stock tokens! |
|
We need to solve the problem soon, because it's unbalanced and too complex for the most users. The shortest way to solve is to change the ration DFI to stable coins in the vault. Maybe only 10% DFI with 90% USDT / USDC can fix the problem. It's then attractive to mint DUSD and buy DFI and swap it to DUSDC / DUSDT. Then we can discuss maybe other options. |
|
The DFI % in vault has to be kept high for generate demand in DFI. We do not want degenerate DFI to a governance token. |
MX24C is right, DFI demand needs to be kept high. What about allowing Usage of Liquidity-Pool-Tokens as Collateral #951 ? Anyone sees any problem with that? Or might it incentivize people to mint more? |
|
@Kassius84 there is a proposal for that already: Usage of Liquidity-Pool-Tokens as Collateral #951. Nobody seems to be interested. I think it's very nice as you can get LM-rewards and provide dTokens. |
Hey @defiprop, thanks for the hint. I like the idea of using the LP tokens as collateral. It's necessary to have detailed look at the relationship between LM rewards per token and price difference dUSD/USD again in more detail. But I can imagine that the possibility LM rewards for a security in the Vault and the price difference between dUSD and USD, creates many incentives to mint dUSD. Kind regards 👍 |
|
I like the idea of @BalthasarBecker that we can use DFI (Oracle Price) to payback the loan. It's simple, it's understandable and it doesn't hurt the ecosystem. It doesn't mean I don't like this proposal. I'd like to see both proposals to be released. The discussion about it occurs on Twitter |
|
I like the idea to pay back loan in DFI (oracle price). This way you can mint stock tockens asset backed for oracle price and sell back minted stock tokens for oracle price. That is genious, because this is simple to understand and explain. It does allow to attract a lot of investors that are used from CEX that they can rely on prices on the stock market. This looks really to contribute to roadto50. |
|
+1 for the DFI (oracle price) payback option discussed at https://twitter.com/julianhosp/status/1470632959048880130?s=20 Easy to use, easy to arbitrage and you have an easy way out of dToken loans. |
+1 I like this idea as well Make sure to vote in the Twitter poll: |
I think I got it. It's a good idea. If you payback the loan back with DFI. You mint the dToken and pay it istandly back with dfi which get burned. You can use 100% of your DFI. In my understanding it's basically the same principle as getting a 1:1 vault 100% collateralized with DFI and then mint dToken. If collateral ratio falls below 100% collateral gets burned. Same mechanism at the end imo. Let the devoloper decide what is easiest to implement. |
I love this idea!! Wont this be the easiest way to draw dUSD or other stock tokens back to market prices? |
|
Burning the DFI would be probably extremly deflationary, BUT then we have dToken in circulation, where there is no collateral existing respectively you will change the DFI amount to dToken but permanently. When you change the loanscheme to 100% (maybe 105%) the arbitrage possibility will reduce the difference of the pool to 5%, but you can go backwards to pa the loan. |
I think with 105% collateralized vault the pool difference will level around 97,5% (e.g. DUSD is 2,5% more expensive) . Assuming the same buying demand. |
Then we have to collect the DFI (we do not have to burn) in a special wallet to not decrease the col. ratio to much. |
But what should be done with these DFI? |
|
Just a quick question here. If we were to go with the "light version" that is we can only pay back DUSD using DFI but you cannot payback dTokens using DFI. Would that not increase the premium in the dToken pools? That is the dTokens get even more expensive as the oracle price? |
|
True. A lower dusd price will enhance demand for stocks. |
I agree, this will happen probably. Please comment if anyone think it will not happen and why. |
I had this concern too some days ago:
But I discussed that topic with two of my friends, Jerome and Christine (I am not at liberty to give any details about their jobs... Let's just say they work in the financial sector) and they assured me that this effect will only be "transitory". |
|
There seems to have a preferred solution around paying back loan by burning equivalent DFI based on oracle price. This way will be very efficient bringing DEX dUSD price to the oracle price but I'm not buying the idea of having uncollateralized dUSD loating around . They would be backed by utility but surely uncollateralized. Also having them as collateral that early seems a bit odd to me. Some creative and (seriously) smart solution are under review but I think there is a simpler way. a dDFI token might do the trick. It will bring DFI value to the "dTOKEN" layer without the need to burn it.
So if dUSD is more expensive on the DEX, We can borrow dUSD sell them on the DEX to get dDFI, payback borrowed dUSD with dDFI (oracle price) and pocket the dDFI remaining. We can even swap them back to dUSD for a bit less (bu still) profit. On the other end, If dUSD DEX price is less than oracle price, people will mint dDFI, sell them on the DEX payback borrowed dDFI using dUSD and pocket the remaining dUSD. It's as simple and straight forward as using native DFI but all key player stays collateralized and no need of dUSD as collateral. The potentiel issue I see is the "Fake Vault Lock" effect of dDFI: Creating a vault a and mint dDFI right after to get dDFI which looks like unlocked DFI. The thing is we will have interest rate on dDFI as we mint them from a vault. So minting them will require to pay interest back. In fact I belive that in long run, we should have dTOKEN/dDFI pairs instead of dTOKEN/dUSD. Other swaps made possible by compisite swap. |
|
Given the high utility of DUSD that might constantly drive up its price above $1, will there not be a constant inflow of unbacked DUSD through arbitraging - if DUSD loans can be payed back with DFI? If so, is there not a danger that this will slowly but continuously drain utility from DFI (and hence its value)? For example, if we take any asset-DUSD LP token, currently the DUSD half gives utility to DFI, which was needed to mint the DUSD, but the more unbacked DUSD is going around, the less its use implies the utility of DFI (as more and more emerged through arbitraging instead of minting). So, worry: a steady increase of unbacked DUSD would mean a steady decrease in DFI's utility? Or am I missing something? |
|
Personally I don't see any reason to make a claim that any dUSD is unbacked when the loan that minted it was paid with DFI. It's still backed, because you retired it's loan with equal value of coin in DFI. Since dUSD is a fungible token, not an NFT, it's always equal to the value and backing of the entire pool. It's not like we're printing them out of thin air, they're minted from the value of the collateral and even when paid off with other coin they still are "backed" by same value. It's simply a trade off that allows the value to be slightly arbitraged back to what it was minted to be. |
|
@MartinAdriaan How do you want to reduce the price of DUSD/dTokens without reducing its value? This is the idea of the whole thing. Sure, then you can mint (print) DUSD/dTokens, sell them, take a part of the rewards and pay back your loan, increase your collateral (with the remaining DFI), mint (print out of thin air) even more and so further. You will have a high inflation. But after one hour this whole thing will be over (that's probably why my friends said the inflation will be "transitory"). The minters will make a lot of money whereas the dToken holders will be expropriated by the amount of the premium they paid. That's how socialism works. I think it's a great idea as long as you are on the minters (printers) side. Long live socialism :D I will go ALL-IN on that one! |
|
If the interest is burned too (like @uzyn mentioned in the tweet below), isn't it inevitable that we will end up in a endless loop, where we always need new loans just to cover up the interest of the previous ones? This does not sound healthy. |
|
If we want reduce premium on dUSD and stock tokens we should reduce the cost of minting them. What about offering loan schemes of 125%, maybe 110% to reduce capital cost for minting - combined with direct liquidation of vaults when falling below (no auction, because anyway not interesting to bid on it)? |
Jap I als believe it's the cleanest way. Maybe we just let them go to auction, because there is still the chance that the capital from collateral gains its value back. Could it be a good idea to indroduce then for all liquidated vaults - which carry unbaked tokens - a backup pool. It would work like that as more tokens (vaults) are unbacked that pool would incentive to put collateral inside e.g. DFI. It's a little bit like stacking just for the pool to back the tokens. Are all tokens backed the incentive goes down. It would be a easy and risk adverse trade. You could gain little interest with not to much risk. And would have stake alike returns without the need to run a masternote. If the collateral prices rise again (mostly DFI) some could buy back the loan... Etc.. I also know this Gremium is wrong placed if you want to reach out to the community you probably have to use Twitter or reddit. Feel free to share the thoughts there if you like it... |
|
@wmbst @goscha80 @LumpiesRevenge @blttgrst So, there is an "emergency" update to make the "arbitrage"-deal become a success. Congratulations! Well, at least mainly the Cake users who paid the premium (and can't create vaults) are the screwed ones. And, surprise, surprise, now the premiums are shifting to dStocks/dETFs... Who told you so?
However, it was fun to discuss the topic... Nevertheless the fact that there is an "emergency" update makes me kind of suspicious. Maybe the next "emergency" is that the master-node holders don't get enough money? How could we solve this? Correct: Another "emergency" update (New "emergency" update to triple master-node rewards) tripling the master-node rewards. And who will vote for it? Correct, the master-node holders. |
|
No one is screwed. LoL. They can simply do what I did and reverse their positions and gain back what premium they paid. I did that because I didn't have certainty around when or how fast all this would get resolved. So rather than mining the new "riskier" D stocks, I went back to my other position in BTC/DFI mining. |
|
Closing this in favor of DeFiCh/dfips#99 |
and others
What would you like to be added:
Loan tokens should be allowed to have a negative interest rate.
Effective token interest rate (unchanged) = vault interest rate - token interest rate.
When effective token interest rate is negative, interest owed is reduced, even to the point of it being negative. There could be case when amount owed is less than principal.
Destruction and burn behavior
Currently at Fort Canning
Loan token's principal is destroyed upon payback, excess, i.e. interest, are swapped into DFI and burned.
Proposal to be added by the next upgrade
To prevent the baseline of having all loan tokens be backed by cryptocurrencies from continuing to shift forward without correcting itself, Untethered Loan Token counter is to be introduced.
When entire payback happened and it is less than principal (due to negative interest rate), insufficient loan tokens are destroyed, The excess loan tokens are therefore being added to a global Untethered Loan Token counter.
When there are excess, total loan token to be destroyed should be principal (of vault) + minumum of (global untethered loan token counter, excess amount). This should therefore reduce unthetered loan token counter, with the aim of getting it back to 0 when interest rate is positive.
Only if there are any excess loan tokens from the above will it be further swapped into DFI and burned.
Why is this needed:
AMM DEX is lacking of order book. This limits arbitrage possibilities of dToken.
When dToken interest rate is negative, it allows DEX of dToken to be arbitraged without being dependent on AMM DEX correcting itself.
The text was updated successfully, but these errors were encountered: