Possibility of negative token interest rate for loan tokens

[uzyn]

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.

[defiprop]

How about introducing a secondary market for dUSDC/T and dStableTokens (I'll call it the "oracle market")? Smart people, who understand that there is a difference between price and value can stay in the original "overpriced" market (DUSD/dTokens) whereas other people with less understanding of economics (or no interest id DFI rewards) could trade in the "oracle market" (where liquidity mining rewards are reduced to fees). We only need to find a way to keep the prices in the "oracle market" pegged to oracle prices. Then everyone can decide whether to pay a premium (and get additional DFI mining rewards) or don't pay a premium and get rewards that come from trading fees. Thereby we do not have to change the (running) system and smart people as well as other people are happy. Once the additional DFI rewards on the original DEX are going to 0 the premiums will fall to 0.X% and the markets can be merged (e.g. by exchanging the assets of one market by the ones of the other market).

[derstrecker]

If the dUSD will be "regulated", people immediately will ask themselves: What else is maybe regulated at the DeFiChain? It very much can appear like traditional finance. Of course the code shows openly, nevertheless if one thing is regulated, who stops from regulating more? We shouldn't even start this in my opinion. To settle the price correctly we need arbitrage. To do so, we need a circle for the trades. Best would be dUSD-dUSDT and dUSD-dUSDC Pools, with small APR. To keep the demand for DFI still high, I suggest to be able to create a pooltoken for the stablecoin-pools, a third party is needed: DFI. So you need dUSD, dUSDC/dUSDT a n d DFI. Maybe 20% have to be in DFI, to create a pool token, but this calculation should be done by s.o. else.

[defiprop]

Why not inflate all dTokens at one given date (e.g. Block 1.500.000)? All dToken balances are updated (according to their deviation w.r.t. the oracle prices) and we are done. All will be happy and investors can come in to trade! Simple, efficient and fair.

[Rno123]

Why not inflate all dTokens at one given date (e.g. Block 1.500.000)? All dToken balances are updated (according to their deviation w.r.t. the oracle prices) and we are done. All will be happy and investors can come in to trade! Simple, efficient and fair.

@defiprop simple yes. Efficient, no. Fair, absolutely not. Unless you're updating dToken balances to vault holders and not token holders. Otherwise, you're essentially punishing vault owners and reward token holders. Your solution assumes vault holders will hold their tokens, but they're more likely to have sold half and provided liquidity with them. This solution will solve the problem in the instant it's implemented, but a significant number of vault owners will close their vaults as a result of this act of "bad faith" and that will remove dTokens from the circulating supply, and this premium will return almost instantaneously again. Not efficient and most certainly unfair, imo.

[gsi2help]

What do you think about an easy option for all vault holders who mint tokens to arbitrage the pools... by closing the loan via the oracle price of a different token than the minted one.

For example: Create a vault, mint dTSLA, swap them to DUSD (via DEX, Pool dTSLA-DUSD) so the pool will go nearer to the oracle-Price; then you can close the loan with your DUSD (changed from the DEX) at the oracle-price of dTSLA without using the DEX.

So the arbitrage can be done very fast to all pools.

The only need is to close the loan with another dtoken as you minted, because of the minted one is in the pool to stabilize the price & to grow the demand.

What do you think about this?

[defiprop]

@Rno123 Well, the vault holders of course would also get the airdrop on the dTokens they minted. Let's assume I put DFI into a vault. I can print my DUSD. The airdrop comes, I get 33% DUSD on top (having 133%). I pay back 75% of my DUSD (as I now have 33% more), get my collateral back and still have the airdropped DUSD (for free, like any other DUSD holder). So why do you consider this unfair?

[Rno123]

@defiprop because the minters are capital inefficient and are still paying interests. Naturally each vault has a different context but what this does essentially rewards ppl for buying DUSD from the open market with DFI, since DUSD is the one getting airdropped. So ppl will sell DFI for DUSD, and get more DUSD airdropped to them. Liquidity providers and vault holders (who would have either sold their DUSD for other assets or provide liquidity with DUSD) would be the ones who are at the losing end.

At least, that's what I expect would happen. Unless you do a retroactive snapshot, which might also cause another round of debate about when is a fair time to take the snapshot, because the snapshot ratio may be very different from present day ratio. And ppl got in at different times.

[Rno123]

What do you think about an easy option for all vault holders who mint tokens to arbitrage the pools... by closing the loan via the oracle price of a different token than the minted one.

For example: Create a vault, mint dTSLA, swap them to DUSD (via DEX, Pool dTSLA-DUSD) so the pool will go nearer to the oracle-Price; then you can close the loan with your DUSD (changed from the DEX) at the oracle-price of dTSLA without using the DEX.

So the arbitrage can be done very fast to all pools.

The only need is to close the loan with another dtoken as you minted, because of the minted one is in the pool to stabilize the price & to grow the demand.

What do you think about this?

I think this is a decent idea but I'm not sure if it can be implemented. You would need to enable repayment of loan across different dTokens. This might cause dTokens to be unpegged from a corresponding underlying vault. So for eg, if u repay dTSLA with DUSD, the sum of all minted dTSLA in vaults can be less than dTSLA in circulation.

[Rno123]

I think the idea of negative interest rates is experimental but it is still code-governed. Moreover, it is going to release DUSD into the market much slower. It will also incentivized people to mint dTokens and DUSD in a more sustainable manner.

I know there's alot of concern surrounding the intervention. But any other method is also direct intervention anyways (airdrop, new liq pools, etc). DeFiChain has always been a robust protocol and given the fact that it's Turing incomplete, I think the idea of negative interest rate is worth a shot.

That said, I think all other ideas have their merits too and I think it's more important to align the interests of the community than to decide on the method first.

From this chat, I gather that there are various goals:

1. Return the peg of DUSD and dTokens
2. Minimize price shocks in doing so
3. Not break anything
4. Keep DFI price afloat
5. Avoid centralized interventions

Are there more goals to include or is this more or less it?

[gsi2help]

It must be possible to grow to bigger poolsizes at the prices near by the oracle-price without dumping the DFI-Price in the DFI-DUSD pool, but how can this work - I think minting and waiting for the right price is not the solution, because when money comes in the system it is now only possible by DFI (from cake-users and users which do not want to have a loan).

Am 10.12.2021 um 19:12 schrieb Rno123 @.***>:

 I think the idea of negative interest rates is experimental but it is still code-governed. Moreover, it is going to release DUSD into the market much slower. It will also incentivized people to mint dTokens and DUSD in a more sustainable manner.

I know there's alot of concern surrounding the intervention. But any other method is also direct intervention anyways (airdrop, new liq pools, etc). DeFiChain has always been a robust protocol and given the fact that it's Turing incomplete, I think the idea of negative interest rate is worth a shot.

That said, I think all other ideas have their merits too and I think it's more important to align the interests of the community than to decide on the method first.

From this chat, I gather that there are various goals:

Return the peg of DUSD and dTokens Minimize price shocks in doing so Not break anything Keep DFI price afloat Avoid centralized interventions Are there more goals to include or is this more or less it?

— You are receiving this because you commented. Reply to this email directly, view it on GitHub, or unsubscribe.

[blttgrst]

I made now some suggestions. I really don´t understund, why this game-theory nobody haven´t seen. Now it is a more complex financial structure build, but all impacts should have been thought through. Why can´t cake have a monstrous vault for their users, so that they the customers can mint coins as well. I know that it´s not my decision, but this has the way to be. As well the risk of being liquidated don´t has to be, when not your complete vault has to be liquidated, than maybe percentage-wise, so that there isn´t a complete loss. We do see now a big sell of DFI through the pool and, when the APR is more and more dropping, the big run out of DUSD is coming. Maybe it is stable this way, but I don´t see the road to 50 currently.

[goscha80]

As well the risk of being liquidated don´t has to be, when not your complete vault has to be liquidated, than maybe percentage-wise, so that there isn´t a complete loss.

There is nearly no loss on vaults with 150% collateral on a liquidation. The arbitrage-traders uses this as exit from the dStock -System now. Just take a TSLA Loan backed with DFI sell it against dUSD -> sell the dUSD against DFI. Voila you have your DFI back (+-1%).

[derstrecker]

As well the risk of being liquidated don´t has to be, when not your complete vault has to be liquidated, than maybe percentage-wise, so that there isn´t a complete loss.

There is nearly no loss on vaults with 150% collateral on a liquidation. The arbitrage-traders uses this as exit from the dStock -System now. Just take a TSLA Loan backed with DFI sell it against dUSD -> sell the dUSD against DFI. Voila you have your DFI back (+-1%).

Still you have to pay back the dTSLA ... and you don't have 100% of your DFI back. That's not arbitrage as I understand it. Update: Ok, got it now :). Because of the different prices, this is possible. But why should it to this routine? I leave DFI behind to receive almost the same amount of DFI?

[goscha80]

Of course not. But this is the only way now. It's the reason why the prices of the Stocks are limited now (they would/could be much higher). The guys dont care about a liquidated/open vault. This is just a snapshot what now is going on. We just have to give them the right tools. A Exit appotunity to the oracle prices.

[derstrecker]

Now I understand! Thanks. They can't loose, only win. Thanks for the insight. Trading is somewhat easy for me, but such tactics are not - yet :).

[goscha80]

Now I understand! Thanks. They can't loose, only win. Thanks for the insight. Trading is somewhat easy for me, but such tactics are not - yet :).

Only if the prices are in the right correlation. And they are multiple time a day. :)

[derstrecker]

But this has value only if you believe the stock goes down. You are just not in a hurry or have problems, when you are wrong. Is there also a similar tactic for going long on a token?

[goscha80]

But this has value only if you believe the stock goes down. You are just not in a hurry or have problems, when you are wrong. Is there also a similar tactic for going long on a token?

The apportunities are much bigger then i could imagine. As long dfi rises faster and higher than Stocks. If not you have nothing to lose.

[goscha80]

Here is an example:

https://defiscan.live/vaults/f2d852fb9cb52daddafb57a3502491720eb452b13737bdd4138c516f29e9a40b and the owner Address https://defiscan.live/address/dQDqxersjvtxy1wZxr1R4VrYM6TfDobKAp

The owner has enough money to prevent the liquidation.

[blttgrst]

I know this, but that´s part of the problem as well. It is only locking the arbitrage in the vault, and when the prices are shifting, you can get the DFI. BUT: It doesn´t help to get the DEXes to the right prices at all.

You have to close the system!

[defiprop]

I think the problem is the following: due to the high incentive for liquidity mining (due to the additional DFI you get besides the fees) the dTokens are very attractive and valuable. Then uzyn came up with his idea of manipulating the prices artificially. Julian made the video in which he called short-selling of DUSD "arbitrage" (which it is definitively not, as DUSD is not a stable coin) and a "no-brainer" (which it only would have been if the proposal was accepted yet, but it was not and still is not). In fact, short-selling DUSD is simply a speculation on whether this proposal is accepted or not. It is definitively no arbitrage.

Some people thought "Great, a risk-free trade, I'll do it" without realizing that this is a speculation on the proposal being accepted. Now it turns out that it might not get accepted, as the realization would be similar to what the central banks are doing. And we all know that printing money out of thin air causes problems (otherwise we wouldn't be in crypto, right?). So now the people, who did not think about but just followed the idea Julian spread are getting nervous. And trying to push this idea.

Luckily, Julian did not put too much money into that, so maybe he won't play out his influence to make it happen (I'm sure he could if he wanted to). So in the end all know that bringing this proposal to life would be a simple manipulation of prices that would make the "no-brainer" reality. This is a severe political problem: Should some people (with influence) decide how to manipulate the market (to take advantage of people which analyzed the rules correctly) or should only the market work on this?

Anyway, I made huge gains with DeFiChain already, so when the decision makers decide to steal 30% I would survive this easily. However, I do not think it is the right way to change the rules in such a manner. Either all tokens (in vaults, liquidity pools or just as tokens) should get inflated proportionally to their deviation to oracle prices (of course stock tokens need be inflated by the DUSD-USD difference PLUS the nominal difference between DUSD and the oracle prices) or nothing should be done.

The argumentation of "we want to bring dToken prices closer to real world prices" to justify giving free money exclusively to the minters and thereby devaluate the assets of the dToken holders is clearly unethical.

[goscha80]

I know this, but that´s part of the problem as well. It is only locking the arbitrage in the vault, and when the prices are shifting, you can get the DFI. BUT: It doesn´t help to get the DEXes to the right prices at all.

You have to close the system!

This is what my entire proposal is about. It would correct the prices immediatly and balance the worth of DFI and the dStock in the dUSD-Ecosystem. without of big changes to the whole system. By the way would it solve the Problem with the not covered interests (There are not enough dStocks to payback all loans and ineterests). I try to summarize: A additional 1:1 vault model with 100-105% collateral and very high interest rate (maybee 100-200%). must be installed. The details have to be discussed.

1. Arbitrage traders would then have the opportunity to take out a short-term loan at almost 1:1 at oracle prices. They would use this to generate profit as fast as possible and to level the price on the DEX.
2. The short-term loans would be liquidated with high probability. An additional liquidation process would be helpful for this kind of vaults. I can imagine to burn a part or all of the collateral the rest could go to the Airdrop/Cfip etc. Wallet.
3. Probability of using the vault-model for other than arbitrage purposes is very low.
4. All stock pools could be leveled independently but also across them

So we would have a fully functional arbitrage apportunity on the DEX. Or I'm wrong?

[LumpiesRevenge]

I think goschaa80´s strategy is the way to go since it is very elegant. It seems to solve the biggest problems and doesn´t require new and complex mechanisms.

Concerning the 1:1 Arbitrageur Vaults I guess you´re meaning high interest rate instead of APY?

@Liquidation mechanism for Arbitrageurs: The new vaults should feature a "liquidate immediately"-Button. That would be a great service to the abritageures because they could instantly realize their profits. What has to be done to the Collateral of these vaults? Auction, Burn, keeping in a treasury, adding it to the Liquidity Miner APR Pool ... that should be thought through thoroughly. Maybe we should elaborate on that?

Concerning the existing vault system I would strongly suggest the implementation of just only one Collateralization scheme (150% or even 125%). With ONE dynamic" interest rate for every dToken which is calculated by the difference betwenn DEX and Oracle price. . Multiple collateralization schemes are absolutely superfluous because: 1) 200%, 250% and higher schemes reduce the amount of minted dTokens 2) The deciding factor ist the LTV. If you´re betting on a maximal price drop of your collateral of 50%, you have to provide triple loan amount as collateral in any scheme. Interest rate for classic vault should never fall below Zero.

[goscha80]

Concerning the 1:1 Arbitrageur Vaults I guess you´re meaning high interest rate instead of APY?

Yes, of course.

[goscha80]

@Liquidation mechanism for Arbitrageurs: The new vaults should feature a "liquidate immediately"-Button. That would be a great service to the abritageures because they could instantly realize their profits. What has to be done to the Collateral of these vaults? Auction, Burn, keeping in a treasury, adding it to the Liquidity Miner APR Pool ... that should be thought through thoroughly. Maybe we should elaborate on that?

"liquidate immediately"-Button is a great idea. To prevent a missaunderstanding the vault could be renamed to oneway-Swap or simmilar. But it has to follow the same rules. Auctions is not the way to go, in my opinion, as long the auctions stay how they are designed. (its more complex, but shortly: They would provide a third way to get in the dStocks-System without a collateralisation-ratio and bypass the LM-Pool). The collaterals in the liquidated 1:1 vault are the to much worth of DFI in the dStock system. They have to flow out. In my opinion this worth belongs the community (For the first it could be accumulated). At this moment they go completly to the arbitrage-traders.

[LumpiesRevenge]

@Liquidation mechanism for Arbitrageurs: The new vaults should feature a "liquidate immediately"-Button. That would be a great service to the abritageures because they could instantly realize their profits. What has to be done to the Collateral of these vaults? Auction, Burn, keeping in a treasury, adding it to the Liquidity Miner APR Pool ... that should be thought through thoroughly. Maybe we should elaborate on that?

"liquidate immediately"-Button is a great idea. To prevent a missaunderstanding the vault could be renamed to oneway-Swap or simmilar. But it has to follow the same rules. Auctions is not the way to go, in my opinion, as long the auctions stay how they are designed. (its more complex, but shortly: They would provide a third way to get in the dStocks-System without a collateralisation-ratio and bypass the LM-Pool). The collaterals in the liquidated 1:1 vault are the to much worth of DFI in the dStock system. They have to flow out. In my opinion this worth belongs the community (For the first it could be accumulated). At this moment they go completly to the arbitrage-traders.

Hmmm, I´ve been thinking a while about the Collateral of the "liquidated one-way swap vaults: Since the minted dTokens remain in the market, the Collateral MUST be kept in some kind of treasury to secure the underlying dToken. Otherwise we would create new atrocities. The new Arbitrage-Vaults should definitely allow stable coins only as Collateral (USDC and USDT) to eliminate/reduce volatility risk. Otherwise the volatility of the collateral could cause problems under certain circumstances. If the dToken price falls below the oracle price, the protocol could remove them from the market via the existing DEXes: USDC/USDT -> DFI -> DUSD -> DSTOCK and then burn the dToken. The liquidity of the accumulated treasury stable coins would flow back into the market. But new questions emerge:

Why should the dToken price ever fall below the oracle price? There is constant demand pressure executed by Liquidity Miners and Stock Traders. So if this scenario never comes into reality: What to do with the Collateral (stable coins) from the liquidated Arbitrage Vaults?

What about growth stocks? Imagine this scenario: the treasury is backing a lot of dBABA stocks in the market who are collateralized by the today´s low low prices. In the meantime BABA has made a +200% bull run. Now those dBABA have become heavily under-collateralized :( A solution could be to use the dynamic classic vault interest (raise to deter Minters from producing more of the under-priced dToken) to replenish the "treasury".

What if demand for dToken constantly plummets, then all stable coins from the treasury have flown back into the market and the dumping continues? For this scenario an additional mechanism to the Arbitrage vault system might be helpful: a possibility to sell dTokens from the DEX in the VAULT for the oracle price there? But who pays the price difference then? So it would be better if the treasury collateral is used for this purpose? Under which circumstances would this mechanism fail? Everything should be water proof.

If the new vaults offer mechanisms to arbitrage in both directions ...

A) dTOKEN DEX price > dTOKEN Oracle price: mint dTOKEN -> sell it on DEX -> dUSD -> DFI -> USDC/USDT = Exit, now liquidate the vault B) dTOKEN DEX price < dTOKEN Oracle price: ??? I dont know how but it would be very elegant to enable Arbitrageurs to do the work and profit from it

... it would be fit to name it "Arbitrage Market" :)

[goscha80]

Since the minted dTokens remain in the market, the Collateral MUST be kept by in some kind of treasury to secure the underlying dToken.

This is exactly the point. Try to imagine the whole System. When the worth of DFI rises, there is to much of them in the System. For example: At the creation time 1DFI had a worth of 3dUSD. Sometimes later 1DFI->6dUSD but the other 3dUSD was never minted. So you have disbalanced worth between the assets in the system. If you put 0.5DFI out of the system its balanced again (and dUSD is backed anyway).

Why should the dToken price ever fall below the oracle price?

This could/will happen. So we have to take this in account. (see below)

What about growth stocks? For example: the treasury has a lot of dBABA stocks in the market which are collateralized by the current low prices. If BABA makes a +200% bull run over the next years/months those dStocks are under collateralized? A solution could be to use the dynamic classic vault interest to adapt the collateral.

This is already done. Your collateral ratio in the vault sinks.

What if demand for dToken constantly plummets, then all stable coins from the treasury have flown back into the market and the dumping continues? For this scenario an additional mechanism to the Arbitrage vault system might be helpful: a possibility to sell dTokens from the DEX for the oracle price there? But who pays the price difference then? So it would be better if the treasury collateral is used for this purpose? Under which circumstances would this mechanism fail? Everything should be water proof.

With the 1:1 Vault-model The arbitrage-traders will bring new DFI's into the system by demand. if the price sinks under the oracle price they will take a loan over a normal Vault(150%) according to oracle (higher price) and pay them back by buying cheaper dStock over the LM-Pools. (the missing DFI-worth is back in the system.) And the price stabilized.

[wmbst]

What should be part of this discussion is, that having individual interest rates per d-token, to have an incentive for the token creators to move the price of the tokens closer to the oracle price has been part of the pink paper from the very beginning. This is normal method for crypto backed tokens. The only new part is, that this interest rate could be negative to increase the speed how fast the token price will move towards the oracle price. This is anyway a long-term thing. It will not move over night but over weeks and months. So for dUSD I think having a negative interest rate is necessary, as it is a bottleneck and price needs to be arbitraged faster than on stock tokens. Also It is absolutely necessary in the long term to get dUSD close to 1 USD to be considered a stablecoin from exchanges and players outside the still small defichain world.

It is not printing money, it is not complicated like a lot of suggestions made here and it is a long term solution, not a short term price manipulation. I would love to see an algorithm which automatically adjusts the interest rate for all d-Tokens depending how far off the oracle price they are. And definitely for dUSD also the option to go negative to increase the speed a little bit.

[defiprop]

Why not inflate all dTokens at one given date (e.g. Block 1.500.000)? All dToken balances are updated (according to their deviation w.r.t. the oracle prices) and we are done. All will be happy and investors can come in to trade! Simple, efficient and fair.

@defiprop simple yes. Efficient, no. Fair, absolutely not. Unless you're updating dToken balances to vault holders and not token holders. Otherwise, you're essentially punishing vault owners and reward token holders. Your solution assumes vault holders will hold their tokens, but they're more likely to have sold half and provided liquidity with them. This solution will solve the problem in the instant it's implemented, but a significant number of vault owners will close their vaults as a result of this act of "bad faith" and that will remove dTokens from the circulating supply, and this premium will return almost instantaneously again. Not efficient and most certainly unfair, imo.

So, what will happen if this proposal is accepted? Don't know what other people would do, but I can tell you what my strategy would be:

1. I'd liquidate all liquidity containing dTokens (about 1 week before the change is applied)
2. I'd sell all dTokens
3. As I expect about 90% of the dToken holders will do the same (as they do not want to get ripped off) therefore:
4. Premiums for dTokens would drop significantly (i.e. vault holders would nearly get nothing of the out of thin air printed dTokens when the propsal kicks in, as the premium w.r.t. oracles would be very low. Or might it even be that the DUSD drops below 1 USD due to the loss of trust into the system. Would vault holders be expropriated then? Would short-selling of dTokens still be a "risk fee trade", a "no-brainer"?)
5. I'd wait until the change is applied
6. I'd buy back the dTokens at near to market prices
7. I expect about 90% of the former dToken holders will do the same (as they want the LM rewards) ...

So, right now I am more open to ripping off dToken holders by implementing this proposal as suggested as there is a counter strategy.

[LumpiesRevenge]

What about growth stocks? For example: the treasury has a lot of dBABA stocks in the market which are collateralized by the current low prices. If BABA makes a +200% bull run over the next years/months those dStocks are under collateralized? A solution could be to use the dynamic classic vault interest to adapt the collateral.

This is already done. Your collateral ratio in the vault sinks.

I´m not talking about the dBABA token who were minted in the classic vaults. I´m talking about those dBABA tokens, that have been minted in the Arbitrage Vault = backed by 100,5% USDC/USDT = currently "owned" by some sort of "Protocol Treasury" When the Oracle Price of BABA does a 2x (which is very likely in the future), these dBABA tokens ("owned" by the treasury) are then merely collateralized by 50%. Don´t You see a problem there? ;)

What if demand for dToken constantly plummets, then all stable coins from the treasury have flown back into the market and the dumping continues? For this scenario an additional mechanism to the Arbitrage vault system might be helpful: a possibility to sell dTokens from the DEX for the oracle price there? But who pays the price difference then? So it would be better if the treasury collateral is used for this purpose? Under which circumstances would this mechanism fail? Everything should be water proof.

With the 1:1 Vault-model The arbitrage-traders will bring new DFI's into the system by demand. if the price sinks under the oracle price they will take a loan over a normal Vault(150%) according to oracle (higher price) and pay them back by buying cheaper dStock over the LM-Pools. (the missing DFI-worth is back in the system.) And the price stabilized.

Great! Now I see. Good to know that the arbitrating could easily be done this way with the existing pools ^^

[goscha80]

When the Oracle Price of BABA does a 2x (which is very likely in the future), these dBABA tokens ("owned" by the treasury) are then merely collateralized by 50%. Don´t You see a problem there? ;)

Not really. Therefore is the high interest rate. The collateral ratio in this 1:1 vault would sink and it gets liquidated. As i said col. ratio (100%-105%) with high interest rate (100%-200%). details must be discussed.

[wmbst]

Not really. Therefore is the high interest rate. The collateral ratio in this 1:1 vault would sink and it gets liquidated. As i said col. ratio (100%-105%) with high interest rate (100%-200%). details must be discussed.

How will it get liquidated? You cannot auction it as nobody would bid on a vault wich is below 100%.

[LumpiesRevenge]

So, right now I am more open to ripping off dToken holders by implementing this proposal as suggested as there is a counter strategy.

I don´t think that we have dToken holders at all apart from liquidity miners. So I wouldn´t be much concerned about them. With Your current idea, the Liquidity Miners would pay the profit realized by arbitrating. But keep in mind: in order to fulfill the dToken value proposition the protocol/algorithm has to produce a dUSD/USDC,USDT ratio very cloe to 1:1. Since this will affect the DFI price very positively, the Liquidity Miners will still profit greatly if "road to 50" is reached :) This should be communicated publicly via our current "ambassadors"/channels to be fair. So the Liquidity Providers can decide themselves. If some of them quit providing liquidity and realize their Premium by buying back DFI more expensive, it has at least a positive effect on the Oracle vs DEX price ratio under current circumstances. If they have trust in the protocol and stay, they will likely make good profit in the short term.

[goscha80]

Not really. Therefore is the high interest rate. The collateral ratio in this 1:1 vault would sink and it gets liquidated. As i said col. ratio (100%-105%) with high interest rate (100%-200%). details must be discussed.

How will it get liquidated? You cannot auction it as nobody would bid on a vault wich is below 100%.

The short-term loans would be liquidated with high probability. An additional liquidation process would be helpful for this kind of vaults. I can imagine to burn a part or all of the collateral the rest could go to the Airdrop/Cfip etc. Wallet.

please read the whole strategy.

[LumpiesRevenge]

When the Oracle Price of BABA does a 2x (which is very likely in the future), these dBABA tokens ("owned" by the treasury) are then merely collateralized by 50%. Don´t You see a problem there? ;)

Not really. Therefore is the high interest rate. The collateral ratio in this 1:1 vault would sink and it gets liquidated. As i said col. ratio (100%-105%) with high interest rate (100%-200%). details must be discussed.

When the Oracle Price of BABA does a 2x (which is very likely in the future), these dBABA tokens ("owned" by the treasury) are then merely collateralized by 50%. Don´t You see a problem there? ;)

Not really. Therefore is the high interest rate. The collateral ratio in this 1:1 vault would sink and it gets liquidated. As i said col. ratio (100%-105%) with high interest rate (100%-200%). details must be discussed.

gosha80, you´re getting me wrong. The Arbtrageurs have already realized their profit in my scenario. I´ll make it very clear this time:

Imagine we have the current situation (all of the dToken DEX prices are higher than the Oracle prices) WITH an implemented ARBITRAGE VAULT which works with this parameters:

• Loan Scheme 99,5% of the Collateral
• Only USDC or USDT are allowed as Collateral
• Any dToken can be minted
• Liquidation on demand is a function

How the Arbitrageur realizes his profit:

• Provides Collateral in USDC
• Takes a 99,5% Loan in dUSD
• sells dUSD in the dUSD-DFI pool
• sells DFI in the DFI-USDC pool
• presses the Instant Liquidation Button in his vault
• the Collateral (100,5% USDC) is taken over by the "Protocol Treasury"

-> the Arbitrageur makes 32% profit instantly -> his minted dUSD remain in the market and shift the Oracle-DEX-Ratio closer to 1 -> these dUSD are now held by the treasury and collateralized by the stable coins from the "liquidated" arbitrage vault.

Then my dBABA scenario kicks in ...

1) today (dBABA DEX prixe = 110% Oracle price) PLUS dUSD-DFI has successfully been arbitraged to 1:1 ratio (as it should happen in the future: dUSD-DFI will be repaired first due to the higher margin, then the Arbitrgeurs will tackle the dStocks) Arbitrageur provides USDC and takes a 99,5% Loan to mint dBABA, sells dBABA for dUSD, sells dUSD for DFI, sells DFI for USDC then liquidates his Vault -> Arbitrageur realizes 10% proft -> dBABA is collateralized by the protocol with 100,5% USDC

2) THEN BABA goes 2x in the real world -> the oracle price goes 2x -> the dBABA (who remained in the market and are currently held by the protocol) are under-collateralized by 50% :-(

Would it be any good to liquidate the DeFi Chain? ;-)

[wmbst]

I read it, but you just write liquidated, without explaining how this could happen. So in this case you suggest, somebody takes a 100% Loan for dUSD. It gets liquidated when it falls below 100% a minute later. The collateral just gets burned. And the owner still has the dUSD wich now is backed by nothing just thin air. That part doesn't make sense for me.

[Rno123]

Why not inflate all dTokens at one given date (e.g. Block 1.500.000)? All dToken balances are updated (according to their deviation w.r.t. the oracle prices) and we are done. All will be happy and investors can come in to trade! Simple, efficient and fair.

@defiprop simple yes. Efficient, no. Fair, absolutely not. Unless you're updating dToken balances to vault holders and not token holders. Otherwise, you're essentially punishing vault owners and reward token holders. Your solution assumes vault holders will hold their tokens, but they're more likely to have sold half and provided liquidity with them. This solution will solve the problem in the instant it's implemented, but a significant number of vault owners will close their vaults as a result of this act of "bad faith" and that will remove dTokens from the circulating supply, and this premium will return almost instantaneously again. Not efficient and most certainly unfair, imo.

So, what will happen if this proposal is accepted? Don't know what other people would do, but I can tell you what my strategy would be:

1. I'd liquidate all liquidity containing dTokens (about 1 week before the change is applied)
2. I'd sell all dTokens
3. As I expect about 90% of the dToken holders will do the same (as they do not want to get ripped off) therefore:
4. Premiums for dTokens would drop significantly (i.e. vault holders would nearly get nothing of the out of thin air printed dTokens when the propsal kicks in, as the premium w.r.t. oracles would be very low. Or might it even be that the DUSD drops below 1 USD due to the loss of trust into the system. Would vault holders be expropriated then? Would short-selling of dTokens still be a "risk fee trade", a "no-brainer"?)
5. I'd wait until the change is applied
6. I'd buy back the dTokens at near to market prices
7. I expect about 90% of the former dToken holders will do the same (as they want the LM rewards) ...

So, right now I am more open to ripping off dToken holders by implementing this proposal as suggested as there is a counter strategy.

So your solution doesn't solve the root problem. You just want to buy back stocks at market cost. Premium will return after airdrop. I don't see how this is viable.

[goscha80]

I read it, but you just write liquidated, without explaining how this could happen. So in this case you suggest, somebody takes a 100% Loan for dUSD. It gets liquidated when it falls below 100% a minute later. The collateral just gets burned. And the owner still has the dUSD wich now is backed by nothing just thin air. That part doesn't make sense for me.

Try to imagine the whole System. When the worth of DFI rises, there is to much of them in the System. For example: At the creation time 1DFI had a worth of 3dUSD. Sometimes later 1DFI->6dUSD but the other 3dUSD was never minted. So you have disbalanced worth between the assets in the system. If you put 0.5DFI out of the system its balanced again (and dUSD is backed anyway).

[wmbst]

Try to imagine the whole System. When the worth of DFI rises, there is to much of them in the System. For example: At the creation time 1DFI had a worth of 3dUSD. Sometimes later 1DFI->6dUSD but the other 3dUSD was never minted. So you have disbalanced worth between the assets in the system. If you put 0.5DFI out of the system its balanced again (and dUSD is backed anyway).

So your system needs DFI to go up at any time. As soon, as DFI goes down it breaks and dUSD is not backed anymore and there is no chance to get it backed again. I definitely don't like such a system.

[defiprop]

So, right now I am more open to ripping off dToken holders by implementing this proposal as suggested as there is a counter strategy.

I don´t think that we have dToken holders at all apart from liquidity miners. So I wouldn´t be much concerned about them. With Your current idea, the Liquidity Miners would pay the profit realized by arbitrating. But keep in mind: in order to fulfill the dToken value proposition the protocol/algorithm has to produce a dUSD/USDC,USDT ratio very cloe to 1:1. Since this will affect the DFI price very positively, the Liquidity Miners will still profit greatly if "road to 50" is reached :) This should be communicated publicly via our current "ambassadors"/channels to be fair. So the Liquidity Providers can decide themselves. If some of them quit providing liquidity and realize their Premium by buying back DFI more expensive, it has at least a positive effect on the Oracle vs DEX price ratio under current circumstances. If they have trust in the protocol and stay, they will likely make good profit in the short term.

Well, not sure if you got this proposal right: the plan is to print dTokens out of thin air exclusively for vault holders. This means that dToken holders (and thus also liquidity miners in the dSpace) would face the same problem savers (in the real world) face these days: the out-of-thin-air USD printed by the FED create an inflation (currently more than 6%, transitory, isn't it? Oh, wait a minute, the FED turned around by 180 degree, it is not transitory any more...), eating up their savings (if you could eat 100 days one year ago, now you can eat only 94 days, 6 days starving). So this is what is planned by the proposal. With the current premiums DUSD holders would be expropriated by around 30%, dToken holders around 40%. But anyway, leaving the dSystem temporarily will solve their problem, so I give the plan to rip dToken holders off (as the FED currently does with savers) a GO! Maybe afterwards the people who followed Julians plan to "arbitrage dUSD" (as he calls it, in fact it was/is a speculation on this proposal being accepted) will be happy and the whole thing can go on in peace. I am so sorry for those pseudo "arbitrage-specialists" (as the missed out on the LM rewards cuz they did not think) so I am fine with their plan to rip off the others (well, at least not me). Afterwards, I will continue with business as usual. Probably I will switch to some DFI-Crypto pool for those two weeks until the whole thing is settled after the proposal is implemented. Meanwhile (until this proposal is discussed, approved and implemented and the aforementioned "arbitrage" specialists don't get nothing for their so smart trade) I will enjoy my ~1% rewards per day. Everything easy.

[goscha80]

Try to imagine the whole System. When the worth of DFI rises, there is to much of them in the System. For example: At the creation time 1DFI had a worth of 3dUSD. Sometimes later 1DFI->6dUSD but the other 3dUSD was never minted. So you have disbalanced worth between the assets in the system. If you put 0.5DFI out of the system its balanced again (and dUSD is backed anyway).

So your system needs DFI to go up at any time. As soon, as DFI goes down it breaks and dUSD is not backed anymore and there is no chance to get it backed again. I definitely don't like such a system.

If DFI goes down the collateral ratio in each vault sinks. So the dToken are backed at any time.

[wmbst]

One suggestion. This topic is about: "Possibility of negative token interest rate for loan tokens"

But discussions are about a lot of totally different proposals. Could you please make a new topic for your proposal, so interested people can discuss this proposal specifically?

At the moment this is quite a mess and not really possible to follow the different discussions.

[LumpiesRevenge]

So, right now I am more open to ripping off dToken holders by implementing this proposal as suggested as there is a counter strategy.

I don´t think that we have dToken holders at all apart from liquidity miners. So I wouldn´t be much concerned about them. With Your current idea, the Liquidity Miners would pay the profit realized by arbitrating. But keep in mind: in order to fulfill the dToken value proposition the protocol/algorithm has to produce a dUSD/USDC,USDT ratio very cloe to 1:1. Since this will affect the DFI price very positively, the Liquidity Miners will still profit greatly if "road to 50" is reached :) This should be communicated publicly via our current "ambassadors"/channels to be fair. So the Liquidity Providers can decide themselves. If some of them quit providing liquidity and realize their Premium by buying back DFI more expensive, it has at least a positive effect on the Oracle vs DEX price ratio under current circumstances. If they have trust in the protocol and stay, they will likely make good profit in the short term.

Well, not sure if you got this proposal right: the plan is to print dTokens out of thin air exclusively for vault holders. This means that dToken holders (and thus also liquidity miners in the dSpace) would face the same problem savers (in the real world) face these days: the out-of-thin-air USD printed by the FED create an inflation (currently more than 6%, transitory, isn't it? Oh, wait a minute, the FED turned around by 180 degree, it is not transitory any more...), eating up their savings (if you could eat 100 days one year ago, now you can eat only 94 days, 6 days starving). So this is what is planned by the proposal. With the current premiums DUSD holders would be expropriated by around 30%, dToken holders around 40%. But anyway, leaving the dSystem temporarily will solve their problem, so I give the plan to rip dToken holders off (as the FED currently does with savers) a GO! Maybe afterwards the people who followed Julians plan to "arbitrage dUSD" (as he calls it, in fact it was/is a speculation on this proposal being accepted) will be happy and the whole thing can go on in peace. I am so sorry for those pseudo "arbitrage-specialists" (as the missed out on the LM rewards cuz they did not think) so I am fine with their plan to rip off the others (well, at least not me). Afterwards, I will continue with business as usual. Probably I will switch to some DFI-Crypto pool for those two weeks until the whole thing is settled after the proposal is implemented. Meanwhile (until this proposal is discussed, approved and implemented and the aforementioned "arbitrage" specialists don't get nothing for their so smart trade) I will enjoy my ~1% rewards per day. Everything easy.

I´m all on your side. I´m against Uzyn´s proposal. I currently favor your idea heavily: an extra vault for arbitrageurs, which enables them to do proper arbitrage = instant profit != waiting for a price constellation that might never happen

Maybe you should open a new thread where your idea is thoroughly expressed and adjusted by positive feedback loops? I would love to join the conversation there. If we find sth. great and elegant, you could make a great community proposal ^^

[defiprop]

@LumpiesRevenge @wmbst great idea! I will write a new proposal!

Edit: Done! @ALL: feel free to discuss proposal #978: Bring dTokens Prices closer to Real-World Prices in a fair Manner

[goscha80]

Sorry I misunderstood it, thought this is about solving the problem.

[wmbst]

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 if i mint DUSD i have to pay the LOAN + interest rate. I only can pay the additional interest rate if anyone else mints more DUSD and sell it to me so i can pay the interest. On the long run i would say DUSD would get more expensive.

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.

I totally agree. And this is already part of the pink paper and should be implemented as soon as possible. It does not need an dfip to be implemented as it is an absolutely necessary feature and as already said, it is part of the pink paper. If I remember correctly Daniel already mentioned it in several news shows about the Loans.

The only thing to discuss is, whether we should implement negative interest rates, to speed things up for the dUSD pool. The pink paper doesn't forbid this as it doesn't say interest rates have to be positive.

[defiprop]

@wmbst @goscha80 @LumpiesRevenge @blttgrst

Well, folks as I learnt the whole discussion is irrelevant: the decision maker (in this case prasannavl already has moved the proposal from Great World (in the far future) to Fort Canning Museum (the next update planned). No voting, critical voices ignored, no expert consulting... the decision maker has taken the decision for YOU!

Welcome to the wonderful world of democracy in a "decentralized" project!

(Keep your discussion going! HAHAHA!)

Edit: Uhhh, I was wrong about that. Thought the proposal was put into Fort Canning Museum and thus would be implemented. But there will be a voting on that. Sry, I am a noob. Just making money without fully understanding the system...

[Dcshel]

Lot's of great ideas and discussion here. I'm no financial wizard, yet I don't think the system needs a huge fancy interest variation for dUSD to fix the problem. I'm only into Defichain a few months and a new masternode operator, so I'm still learning. But; It seems to me all we really need is an outside world way to arbitrage the value of the dUSD tokens back to one dollar. Could we not simply allow for any vaults to be paid back by existing dUSDC or dUSDT stable coins for their dollar Oracle value ? Or at minimum dUSD loans? If dUSD is 20% more expensive, I'd just use real stable coin to retire a loan. Just doing this would provide us an easy way to allow arbitrage and pressure the dUSD value back to actual dollar value. Doing this and burning those other stable coins on retired loans with interest would seem to me to create an inflation of supply of those overpriced token assets lowering demand pressure and leveling out their price. Am I missing something?

[Dcshel]

@wmbst @goscha80 @LumpiesRevenge @blttgrst

Well, folks as I learnt the whole discussion is irrelevant: the decision maker (in this case prasannavl already has moved the proposal from Great World (in the far future) to Fort Canning Museum (the next update planned). No voting, critical voices ignored, no expert consulting... the decision maker has taken the decision for YOU!

Welcome to the wonderful world of democracy in a "decentralized" project!

(Keep your discussion going! HAHAHA!)

What does that mean? That the decision was made without a vote by Masternode operators to implement a negative interest or that this problem has been expedited to be voted on really soon?

It needs to be expedited IMHO It is a pretty urgent need, and something should be done if we want mass adoption into this protocol. I cannot see a lot of people joining as is. I didn't want to toe up tons of unworking capital in a loan, or risk being too close to liquidation, which I tried for a couple days. After taking a 5,000 dollar hit trading for dUSD and dstocks, I calculated I could gain it all back in mining rewards in 30 day IF the Apr stayed like 450 to 500%, but already that has significantly dropped, so I unwound all my stock token mining and reverse traded back to DFI and mining only on the BTC/DFI pool because I got all my losses back and don't have to risk it, plus the higher rates on the BTC pool are nice without risk of vault liquidations or loss of value in doing direct swaps. Many people may likely do likewise. By the time I figured my much lower available capital use in a vault loan even the much higher stock LPs didn't make me any more revenue. Of course the vault option give one more upside (or downside) exposure to DFI token value.

[arakis2016]

I am speaking from the perspective of a DefiChain customer. At the moment, it is simply not worth entering the LM token pools. The price for the dUSD token is too high. Mining via the vaults will not compensate for this in the long term. The situation at the moment is unsatisfactory for investors. I would very much welcome the approach with negative interest rates on dUSD. The bottom line is that everyone would win.