FDIC for BitUSD

[bytemaster]

I think we can control the risk significantly by requiring the number of shorts to balance the number of potential calls before new shorts can be executed.   To the extent that we can avoid a lop-sided order book we could be better protected. 

Ie:.... if building up to the collapse the bid side of the orderbook got very thin we have a higher risk of collapse than if there was plenty of bids.

The challenge we face is that people can cancel their orders.  Unless we prevent people from canceling orders if canceling the order would put the market with insufficient depth. 

So many potential rules and they all have nasty consequences. 

[bytemaster]

Everyone who is worried about "dilution" beyond 2 billion to bail out BitUSD holders have a very simple way to hedge that doesn't involve them selling their full BTSX position.  Simply keep a percentage of your holdings in BitUSD.   If BTSX is diluted due to the volatility your losses from BTSX side due to dilution are more than made up for with your gains on the BitUSD side.   (Assuming initial price parity) Someone who has 50 XTS and 25 USD will end up with 100 XTS (or more) in the event of a black swan so their "stake" grows.   Furthermore if dilution does occur they are likely to receive enough of the dilution via their USD holdings to cover their losses on the XTS holdings.   So for all of you worry-worts who see the evil of being diluted beyond 2 Billion, then just buy some BitUSD as a hedge.

All of that said lets consider very carefully this flash crash event because the numbers being thrown around about potential dilution are FUD.

Step 1)  Ask what percent of the BTSX market cap does BitUSD represent.   If it is 1% then your risk of dilution is 1% and if it is 33% then 33% is the maximum dilution one need fear.  33% is the maximum amount that shorts would be willing to go anyway.

Step 2) Ask what percent of the margin positions have more than 2x collateral because they already had some unrealized profits in them or because those holding the position wanted to be pro-active about avoiding an event where they run out of margin and pay a 5% fee.  To the extent that this is the case the network has even more of a buffer before having to worry about dilution.

Step 3) Ask what percent of the margin positions would be closed out normally during the first part of the crash?  I would guess that the majority of margin holders would be covering as quickly as possible (or adding collateral to avoid getting called).   This means you can generally look at the order book and see how many bids are available that are above 50% the current price.    You can safely assume that the USD up for sale (or short) will be used to close out a large number of the margin positions. 

After considering steps 1, 2 & 3 of the short, then crash and cover process you will hopefully conclude that if you start out with 10% of market cap represented by BitUSD obligations, that by the time all collateral and sufficient bids are consumed that you will be looking at a very small percentage of the original USD that needs some money printing.


So the "worst case" is that all bids cancel at once and the new "high bid" is 25% of the old high bid (75% fall instantly).   All margin positions get called at once and 66% of the USD is purchased back by the collateral.  The remaining 33% of the USD is purchased back by issuing new shares.   So if you started with 30% of the cap represented by USD then you are looking at a 10% dilution in a "worst case" 75% crash.    In practice crashes will be slower, many of the positions will be covered without any dilution and less than 30% of the cap will be represented by USD.    So I think what we are looking at here is a whole lot of fear about a "bad event" that you can "hedge against" and is "unlikely" to happen where the potential losses from "dilution" are less than the average daily volatility of Bitcoin.

[luckybit]

There should be no more than 2 billion BTSX at any time for any reason otherwise the protocol is going to be perceived as broken.

It should be impossible to go beyond the cap. That should be a hard limit. If new BTSX get created then just increase the burn rate to destroy BTSX to make up for that. There is no reason to discuss going beyond 2 Billion BTSX unless you're trying to make people panic.

If there are more BTSX than 2 billion then everyone is going to know something is seriously wrong with the algorithm. If it never can surpass 2 billion then people have the same sort of certainty that they have with Bitcoin. Let the burnrate increase as the BTSX approaches the cap and if it hits the cap then resolve it using other mechanisms.

I think people will tolerate small corrections, adjustments, and similar volatility if and only if the trend is going in the right direction. We tolerate Bitcoin volatility because the price keeps going up in the long term otherwise who would invest?

I agree with this. I think you will create unnecessary panic which will back fire in all of us. Although from an economic perspective it may be necessary i am not sure if it should be implemented as an option now.
In the end of the day if we come to this a new bitshare can be created which honors the previous one with the twist of increasing the supply or not?
Anyway, I don't have much experience in all these and how cryptotraders respond to these strategies so my opinion should not be considered much.

The risk of a black swan event which would require us to breach the cap of 2 billion is extremely low. Why contemplate an event which isn't likely to ever happen?

Events do happen but if it does happen in a way where we would have to breach the cap then I think we are better off restarting the market manually than doing that. Sure it might seem less autonomous if human beings have to intervene to restart the market but that is better by far than breaching the 2 billion cap in my opinion because then there is no algorithmic certainty to the protocol.

It would be like Bitcoin going above 21 million temporarily in the case of a black swan event. Most people would probably prefer a hard fork rather than to see that happen. In fact most users would probably think Bitcoin itself broke if that happened.

I think 2 billion is where we started, it's the algorithm, and improvements should be designed to preserve the cap as it was originally written.

A manual restart would be freezing the market until someone among us is willing to take a loss to restart it. One person taking a loss is better than possibly millions of people taking the loss and then getting out in a panic. The moment dilution occurs beyond 2 billion cap everyone is going to panic and there is no guarantee Bitshares would recover.

As much as you might tell people to be calm and rational if people who got in at the first day when it was 2 billion are in a panic and these are the earliest adopters with the majority of the shares then there will be a problem. This is why I said originally to never breach that cap so that the earlier you got in the safer you are.

[mf-tzo]

There should be no more than 2 billion BTSX at any time for any reason otherwise the protocol is going to be perceived as broken.

It should be impossible to go beyond the cap. That should be a hard limit. If new BTSX get created then just increase the burn rate to destroy BTSX to make up for that. There is no reason to discuss going beyond 2 Billion BTSX unless you're trying to make people panic.

If there are more BTSX than 2 billion then everyone is going to know something is seriously wrong with the algorithm. If it never can surpass 2 billion then people have the same sort of certainty that they have with Bitcoin. Let the burnrate increase as the BTSX approaches the cap and if it hits the cap then resolve it using other mechanisms.

I think people will tolerate small corrections, adjustments, and similar volatility if and only if the trend is going in the right direction. We tolerate Bitcoin volatility because the price keeps going up in the long term otherwise who would invest?

I agree with this. I think you will create unnecessary panic which will back fire in all of us. Although from an economic perspective it may be necessary i am not sure if it should be implemented as an option now.
In the end of the day if we come to this a new bitshare can be created which honors the previous one with the twist of increasing the supply or not?
Anyway, I don't have much experience in all these and how cryptotraders respond to these strategies so my opinion should not be considered much.

[Riverhead]

option 3 is not reasonable, why bts owner need to take losses for the one who short bitu?
bts owner have already lose much for the price down.

the one who short bitu have already pay all backup bts. then can do nothing.
the one who owner bitu is the  winner for  the bitu price up, they should take the remaining losses.

I think of it like a premium for market stability insurance. The market, and thus those holding BTSX and assets, would be better off taking a short term hit for long term stability. Ya, it sucks but that's how all insurance works. The person that the market is bailing out has already lost their collateral so they're not getting off free either. If option 3) comes into play we're all in danger anyway so I'm OK with paying a bit for some security.

[luckybit]

What if the blockchain not only is a market maker but controls fee destruction such that within a year time it always caps to 2B BTSX supply? In order to prevent dilution.

Blockchain cannot control total fees paid.

It can! based on estimates adjust a destroyed surcharge included in the fees every 1000 blocks or so, and only do that if total supply has expanded beyond cap or 2B. You don't change anything about delegate fees/etc. just an "anti-inflation" fee that only kicks in when conditions require it.

This will keep balance and faith in the system as it is self-correcting and continues to function as intended for both investors (BTSX holders) and hedgers (BitUSD and the like holders)

There should be no more than 2 billion BTSX at any time for any reason otherwise the protocol is going to be perceived as broken.

It should be impossible to go beyond the cap. That should be a hard limit. If new BTSX get created then just increase the burn rate to destroy BTSX to make up for that. There is no reason to discuss going beyond 2 Billion BTSX unless you're trying to make people panic.

If there are more BTSX than 2 billion then everyone is going to know something is seriously wrong with the algorithm. If it never can surpass 2 billion then people have the same sort of certainty that they have with Bitcoin. Let the burnrate increase as the BTSX approaches the cap and if it hits the cap then resolve it using other mechanisms.

I think people will tolerate small corrections, adjustments, and similar volatility if and only if the trend is going in the right direction. We tolerate Bitcoin volatility because the price keeps going up in the long term otherwise who would invest?

[alt]

I support option 2.

option 3 is not reasonable, why bts owner need to take losses for the one who short bitu?
bts owner have already lose much for the price down.

the one who short bitu have already pay all backup bts. then can do nothing.
the one who owner bitu is the  winner for  the bitu price up, they should take the remaining losses.

some bitu(x style) will destroy as trade fee, some bitu short position will destroy at the margin call,
bitu don't need to as  same as the short position, that make sense.

[Overthetop]

Maybe I missed something because of 6 pages discussion.

I am afraid all of the 3 options are not good enough,we should find out another better ways .

My consideration is below：

The case happens by the trigger of inadequate collateral.

This kind of cases maybe always happen suddenly and sharply as some critical issues were exposed.

Think about the value of BTSX drops 70% suddenly because of the explosive panic .

If we take #3 mechanism, the BTSX would be diluted more than 50% by the program automatically.

The problem is , because of the severe dilution ,the confidence of the market would crupt, shareholders may dump BTSX frighteningly... ,that is the vicious circle.

I think the footstone of  BTS system is the confidence of market, and we should do our best to avoid any potential and even tiny hurt to the coinfidence.

3# mechnism maybe is the better choice than others, but ,I think we should try to find out the better.

Some guys point out we can set up reserves for this ,  maybe it is a good idea.




 

 

[bytemaster]

Could you incentivize bitUSD sellers in the event of a BTSX sell off? Like in fees. 1/10 the fees to to sell bitUSD and buying BTSX.

Who pays the fees in the two party transactions anyway?

Both parties pay to submit their order and they get what they ask for.  Any overlap is kept by the network.

[Ggozzo]

Could you incentivize bitUSD sellers in the event of a BTSX sell off? Like in fees. 1/10 the fees to to sell bitUSD and buying BTSX.

Who pays the fees in the two party transactions anyway?

[bytemaster]

What if the blockchain not only is a market maker but controls fee destruction such that within a year time it always caps to 2B BTSX supply? In order to prevent dilution.

Blockchain cannot control total fees paid.

It can! based on estimates adjust a destroyed surcharge included in the fees every 1000 blocks or so, and only do that if total supply has expanded beyond cap or 2B. You don't change anything about delegate fees/etc. just an "anti-inflation" fee that only kicks in when conditions require it.

This will keep balance and faith in the system as it is self-correcting and continues to function as intended for both investors (BTSX holders) and hedgers (BitUSD and the like holders)

Shareholders are in control and delegates can voluntarily implement that.

[bitmeat]

What if the blockchain not only is a market maker but controls fee destruction such that within a year time it always caps to 2B BTSX supply? In order to prevent dilution.

Blockchain cannot control total fees paid.

It can! based on estimates adjust a destroyed surcharge included in the fees every 1000 blocks or so, and only do that if total supply has expanded beyond cap or 2B. You don't change anything about delegate fees/etc. just an "anti-inflation" fee that only kicks in when conditions require it.

This will keep balance and faith in the system as it is self-correcting and continues to function as intended for both investors (BTSX holders) and hedgers (BitUSD and the like holders)

[Empirical1]

Won't a small BTSX crash 25%+ cause an already extremely negative market to think a dilution event could be triggered soon? Thereby creating the flash crash you'd like to avoid as everybody rushes for the exits before a potential dilution event is triggered.

I don't think any market can offer participants 100% certainity that individual assets will be be redeemed at full value. Participants will already be evaluating the peg on risk of delegate compromise, technical bug & others things. The black swan trading event is just another risk that doesn't seem worth risking BTSX total issuance for. (Even though Bitcoin has inflation, a large part of it's value is derived from the certainity of Bitcoin issuance imo.)

You said participants can buy dilution event insurance/hedge, so I would let them buy it/offer it to them at purchase myself.

Small crash will NOT create potential for dilution because all positions will have been covered and thus all "risk" consumed.

Thanks well I'm always happy to go with BM choice anyway just trying to see a different POV.

I realise a small crash will not create the potential for dilution. However a >50% flash crash could trigger up to a ~25% dilution was my understanding.

So it would be like watching Bitcoin fall rapidly from $600 to $420 and knowing that if it hits somewhere in the  $300 range a ~25% dilution event could start kicking in that takes it to $240. (& probably more panic selling after that.) 

So while at $600 there's only a 33% flash crash risk. At $400 in a flash crash there's a 33% crash risk + huge dilution event risk (relative to that price). So there's an intelligent rush for the exits after a rapid drop from $600-$420 which actually helps create the 50-65%+ drop which otherwise would be incredibly unlikely to happen.

So it seemed that we'd now included incentives for market participants to turn a small flash crash into a big one was my concern.

We could limit the maximum XTS held in collateral to limit the exposure of the network.    After all it is all about relative market cap... if there is $1 BitUSD with a $100 Million XTS market cap then the potential dilution is very small (less than 1%).   I don't think many people would fear that.  It is only when $BitUSD as a percentage of the market cap approaches something like 33% that you risk such a huge dilution event.  I think you would see a slow steady fall when USD demand is that high.

I might have misunderstood the last sentence, but yeah I don't think BitUSD approaching 33% of CAP will create a crash. I think a BTSX flash crash will be instigated by an unrelated event but at that point, the presence of 25-33% BitUSD & or other assets to CAP will become the factor that could turn an unrelated ~25% flash crash into a bigger one by the manner described above.

Don't know if it's a big issue though. Just putting it out there.

[bytemaster]

Won't a small BTSX crash 25%+ cause an already extremely negative market to think a dilution event could be triggered soon? Thereby creating the flash crash you'd like to avoid as everybody rushes for the exits before a potential dilution event is triggered.

I don't think any market can offer participants 100% certainity that individual assets will be be redeemed at full value. Participants will already be evaluating the peg on risk of delegate compromise, technical bug & others things. The black swan trading event is just another risk that doesn't seem worth risking BTSX total issuance for. (Even though Bitcoin has inflation, a large part of it's value is derived from the certainity of Bitcoin issuance imo.)

You said participants can buy dilution event insurance/hedge, so I would let them buy it/offer it to them at purchase myself.

Small crash will NOT create potential for dilution because all positions will have been covered and thus all "risk" consumed.

Thanks well I'm always happy to go with BM choice anyway just trying to see a different POV.

I realise a small crash will not create the potential for dilution. However a >50% flash crash could trigger up to a ~25% dilution was my understanding.

So it would be like watching Bitcoin fall rapidly from $600 to $420 and knowing that if it hits somewhere in the  $300 range a ~25% dilution event could start kicking in that takes it to $240. (& probably more panic selling after that.) 

So while at $600 there's only a 33% flash crash risk. At $400 in a flash crash there's a 33% crash risk + huge dilution event risk (relative to that price). So there's an intelligent rush for the exits after a rapid drop from $600-$420 which actually helps create the 50-65%+ drop which otherwise would be incredibly unlikely to happen.

So it seemed that we'd now included incentives for market participants to turn a small flash crash into a big one was my concern.

We could limit the maximum XTS held in collateral to limit the exposure of the network.    After all it is all about relative market cap... if there is $1 BitUSD with a $100 Million XTS market cap then the potential dilution is very small (less than 1%).   I don't think many people would fear that.  It is only when $BitUSD as a percentage of the market cap approaches something like 33% that you risk such a huge dilution event.  I think you would see a slow steady fall when USD demand is that high.

[tonyk]

I think it is the long work days...
    coupled with a sleep deprivation....

[edit]
I do not want to think about the other possibility because it is 'replaced by the aliens'