-
Notifications
You must be signed in to change notification settings - Fork 86
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
BSIP74: Margin Call Fee Ratio #164
Comments
Thanks. IMO it makes little sense to open many issues here with individual ideas about how to optimize MPAs. Better discuss elsewhere and come up with one BSIP containing the results of the discussion. |
IMHO charging a fee would cause less margin calls being bought thus lead to a bigger premium in a downtrend. Due to the impossible trinity, we have to accept trade-offs. If we want to improve one side, the other side would get weaken. For optimizing MPAs, perhaps we can wait for @biophil's project: https://bitsharestalk.org/index.php?topic=28340.0 and/or work along with his team. Adding more parameters so the community can tune is generally good IMHO, more tools means greater flexibility. However, more tools also means greater complexity, we should have guidelines about how to tune the parameters, when should tune what and etc. |
Charge the fees was independent,MSSR still is 1.02, it wouldn't cause less margin calls being bought, the margin calls still will be eaten like now with MSSR=1.02. The research didn't have a exact name,maybe one month, maybe one year, and the result need to test and will change too much. Some project of MPAs have charged this fee for a long time, didn't find any problems, i think we can learn somethings from it. |
If to charge extra fee from collateral, that means the debt position would lose more collateral (than current rule) when it's partially filled, that means it's harder for the debt position to return to a healthy CR, that means stronger sell pressure, that means it's closer to black swan. |
We didn't charge extra fee from collateral, we charge the fee from the transaction of margin call. Someone borrow 100 bitcny when bts price is 1 bitcny, CR is 1.75, he collateralized 175 bts; Yes, as the current rule, when he set the target CR 1.75, he will need to sell a little more bts for the target CR. |
You can't buy more than the full debt. If you want to charge a fee you must take it from the collateral. (Suppose there was only one shorter - how would you buy more than the existing supply?) In your example, CR = 1.35 means price is 1.75/1.35 (about 1.3) BTS/CNY. So the call buys 100 CNY for 130 BTS. You can take another 10% (13 BTS) as fee and return 32 BTS (=175-130-13) to the shorter. The effect is the same as with a higher MSSR - in case of a partial fill, the remaining collateral ratio will improve less, because more collateral is spent. (Even if you take the fee from the CNY side the effect will be the same because less debt is paid back.) |
The penalty fees will not play the role like MSSR. Take the fee from the CNY and take the fee from the BTS which will be sold is different: The shorter set the target CR 1.75,
So, charge the fee from the collateral will have the smaller sell pressure than charge the fee from the CNY. |
@shulthz you kept talking about fully filled debt positions. I and @pmconrad concerned more about partially filled positions. It's different. I agree that there is little risk if we charge some when closing a position. However, the positions that are too big thus hard to close will be impacted as well. For example, if you have a position with 100 BTS collateral, it's easy to close if sell all, market price won't change much; if you have a position with 10,000,000 BTS, even you want to sell all collateral, likely there is no so big buy support in the market. |
I have thought about the target CR, the amount of bts which was sold is less than the MSSR=1.10, didn't take too much sell pressure. |
To be clear, I'm not against adding more features/tools. Just discussing to try to find out the pros and cons and guidelines about using the features/tools.
Every little bit counts. The last straw that broke the camel's back was not heavy at all.
IMHO sell pressure (actually the demand to close debt positions in a downtrend) is mainly due to MCR but not MSSR, although higher MSSR may lead to bigger premium in a downtrend which gives people an illusion "sell pressure is high" so may probably cause people to dump BTS in other markets. Reducing MCR can significantly reduce sell pressure but may lead to lower overall CR thus easier/closer to black swan. This is off-topic though.
It doesn't make sense to argue this way. Please focus on the topic. Although people can have fiat or other cryptocurrencies or a house to sell for bitCNY, or even loan elsewhere or get money by whatever means, it is off-topic. The situation is, there is only BTS in collateral, the collateral will be listed for sale automatically if CR <= MCR. This IS sell pressure. If the the owner of the debt position does nothing, charging extra fees doesn't help for mitigating sell pressure. Here I repeat again, please focus on topic. |
Focus on topic:
Charge fee(less than 15%) will give less sell pressure than MSSR=1.10 if we charge bts. Sell pressure was influenced by many factors, not only MCR and MSSR, if charge fee take the sell pressure not more than MSSR=1.10, I don't think it's a problem. Off-topic: When we change MSSR=1.02, the main influence is the price shock, but it also reduce the sell presure. "If the owner of the debt position does nothing, charging extra fees doesn't help for mitigating sell pressure." |
it lead to nothing but less borrowing and high premium and serious shortage of smartoin. |
“high premium”? where it come from? |
you can read here to get a full understanding: https://bitsharestalk.org/index.php?topic=28367.0 |
This didn't have any conflict with what you write, it's a supplementary rule, and the parameter can be adjusted by the community, and you will find how useful of the parameter and fees finally. (Edited by @abitmore: reformatted. By the way deleted 2 duplicate comments.) |
Personally I quite like this, and it's similar to part of an idea I've had and bounced around in TG. I'll try and get it fleshed out and add it to the conversation. |
10% fee for failing to increase your collateral is cannibalistic. Who in this earth is going to collateralize BTS to create BitUSD, after reading this conversation and watching the last two years BTS /USD chart? |
If you don't get it, I don't have time to try to convince you, sorry. I think you should check the MAKERDAI carefully and then think about what you say. |
Lol, you made me feel like a Larimer. Thanks for the compliment! |
Every serious business has an insurance fund.
A extra 0.1% margin call fee for bitCNY and 0.2% margin call fee for bitUSD should also work and has no real effect on the premium. No discount for users with a LTM, 100% to the insurance fund and reserve pool. I had a similar idea with a daily BTS buyback from the insurance fund: |
Hey guys, I wrote the following analysis regarding the issue. GS is only the manifestation of some more profound structural flaws in the Bitshares protocol. It's a pretty long read but, nonetheless worth the time.
|
@Inmortak I disagree with your analysis. |
I appreciate you took the time to read. I just saw this, otherwise I would have answered you before.
I am all in for the workers too. I am claiming that they must be paid in stable assets instead of bts. That's it. Precisely because, as you say, the process of paying for workers is independent of market conditions, is that we have a systemic negative feedback on immediate price. Of course, workers have a positive feedback on infrastructure for the whole ecosystem. But I am talking about the immediate effect on price here. If the workers receive thousands of bts daily that they have to short on the market, the immediate effect is a decrease in Bts price, which at its turn forces more bts to be paid next time to the workers in order to honor the USD value of the worker contract. Then, more bts dump and so on, etc.
A position with a CR>1.1 is not a healthy position. Healthy positions are those with a CR> MCR, which, at the moment is 1.6 for BitUSD. In fact, what I propose is a milder form of the parcial GS initiative in which all positions with a CR<MCR are taken over. Notice that this take over would only happen whenever the pool_cr < 1 and would not necessarily affect all positions as the take over would be on a one by one basis from worst position to better positions until Pool_cr > 1. Finally, I also propose that any taken-over position can be bought back by its former holder under the condition that its collateral is brought back to MCR. So, no punishment at all is mandatory.
Global settlement freezes the BitUSD market whenever it happens, which is whenever at least one debtor fails to keep his cr>1. What I am proposing is way different. I am proposing that under the super extreme case in which the pool_cr < 1 and all positions with cr<MCR have been taken over, the force-settement is suspended until the pool_cr > 1. Which, at its turn, is almost guarantee if the pool is always replenished. Nonetheless, your are absolutely right here in something: My proposal doesn't solve the problem completely. Problem that, by the way, is not totally solvable. Under an ultra-bear market for example, if bts price goes to 0, it's impossible to use it for collateral of even 1 BitCent.
You are right here too. And, unfortunately, I cannot provide better feedback here. I don't have the numbers involved nor I feel I have the required prerogative. In summary, I don't fully know the finances of the network. I am claiming, however, that a protection fund is required and its existence implies its replenishing. |
maybe charge 0.5% fees first. |
this BSIP will just provide the possibility to charge fees from margin call trading. how high a fee will be charged depend on what a consensus the community can reach. it's possible to charge 0% fee, it's also possible to charge a 13% fee, but I think it's not bad to begin with a value like 0.5%. |
This scenario is in scope of #193 (BSIP71: prevent global settlement). |
@ryanRfox please assign a BSIP number. Thanks. |
Assigned BSIP74 @bitcrab please make a PR with your draft text. |
Debt positions pay FP*MSSR exactly as they did before. Please explain why you think this increases system risk. |
The margin call fee does reduce the
When fees are not collected, this is not a problem and can be ignored. |
Yes. This is actually a good thing because it reduces the premium. Originally, the theory of our SmartCoins meant the premium to be an incentive for holders to sell their SmartCoins into margin calls. In reality this didn't work. Instead, the premium was perceived very negative by market participants because they expected the asset to trade closely around the fair value. |
You mixed the wrong sides, because there is no free lunch! |
As I explained, that was the original theory. We have seen that it didn't work even with MSSR=10%.
With "premium" I mean the difference between market price and feed price. For the market, Bob's price is relevant not Alice's. |
Because MSSR=10% did not work, doesn't mean MSSR can be 0%. Margin call fee should not be too high: https://bitsharestalk.org/index.php?topic=29138.msg333633
I hope you are right! |
Have all comments been adressed here, do we consider this ripe for voting? |
Assume
?
|
Well. I think the intent is clear enough. |
The "Stability Fee" makes minting new debt asset more expensive. |
No, the terms of the smart contract (MCR + MSSR) remain unchanged for the debt creator. With this BSIP it becomes possible to increase MSSR without affecting the peg, but
Yes.
Yes. It is up to the asset owner (issuer) to use the fee income appropriately. |
BitShares has no "Stability Fee", calling it "Stability Fee", like on MAKER doesn't make much sense. |
Exactly, it has no "Stability Fee" yet. I find "Margin Call Fee Ratio" MCFR not carrying the meaning of it. If we desire more BitShares like wording we could have Stability Fee Ratio SFR. MC is unnecessary, as its implied. MSSR and MCR also do not carry prefixed MC but refer to margin positions and calls. |
I have a thought, maybe the fee can share a part of it (e.g. like 20%) to the margin call buyer which can be decided by the asset owner. |
There will have two suggestions: 1.The fee can share a part of it (e.g. like 20%) to the margin call buyer which can be decided by the asset owner. 2.The assets owner have the choice to decide which kind fee they want to charge, the Smartcoin or Backing asset. |
When the margin call was closed by the market,the system can charge the shorter penalty fees from the transaction of margin call.
The margin call penalty fees will place in a insurance fund.
The penalty fees can be 0.1% ~ ?%(Adjusted by the community).
This will encourage the shorter to close the margin in time.
This wll suppress the aggressive shorter partly to reduce the global systematic risk.
The insurance fund will play a role in some place.
MSSR still will be MSSR.
When the shoter was be margin called, he will sell the bts to get the BITASSET to close the position, he sold 1 bts and got 0.4 bitcny, so the system charge 10% from the 0.4 bitcny as the penalty fees.
The text was updated successfully, but these errors were encountered: