Solution to High Volatility potentially breaking the BitUSD Peg

[theoretical]

What about decreasing the margin to 2x and using fees to recapitalize losses from blown shorts gradually over time?  That way, XTS holders who don't touch BitUSD don't risk a sudden dilution of their capital from an adverse event.  Rather, dividends are reduced or eliminated.

You could also "buffer" the fees in a reserve balance, which would simplify bookkeeping enormously.  E.g., whenever a blown short happens, you just add the negative BitUSD balance and positive collateral XTS balance to the reserve balance, and zero out the position attached to the account like a normal margin call.  Then the total negative BitUSD in the reserve balance is how many BitUSD you want to purchase on the market to get rid of the excess uncollateralized BitUSD, and the positive XTS balance is how much capital you have to make that happen.  The actual open market operations to be performed with the reserve balance would be specified by some fixed algorithm.

The reserve BTS balance should decay exponentially over time (I'd suggest a half-life of ~120 days).  This means that reserve balance will eventually be paid as dividends to XTS holders (by destroying XTS) if no blown shorts occur.  Also, the fraction of the reserve available for open market purchases in a single block, and the market depth to which it can go, should probably be limited.

In an event with lots of blown shorts, the reserve funds might be inadequate to make good on them right away, but BitUSD holders will know that eventually the extra money will be destroyed.

Of course there are lots of flourishes you could add.  For example, the reserve BTS balance shouldn't decay when the reserve is in danger of insolvency.  (I.e., XTS assets of reserve balance divided by XTS liabilities of reserve balance gets lower than some threshold K, e.g. K ~ 2, as determined by valuing BitAsset balances at the current market price).  Or maybe we could temporarily increase fees when the reserve balance is hurting.

[bytemaster]

Would making the minimum trading unit bigger also decrease the volatility?

I.e Lets say 1 Bitcoin was trading at $1000

If the minimum trading unit was $0.10 then manipulating the price a few $ might start a small short squeeze.
If the minimum trading unit was $1 then manipulating the price tens of dollars might have the same effect.
" " " was $10 then they would have to move it a few hundred $.

Obv. $10 would be extreme, but as the blocks will be processing slower than real time trading, Bitshares hopefully won't be practical for HFT bots or people trying to make 0.X% gains so you should be able to make the trading unit bigger than most other platforms which should also suit the slightly longer-term trading mindset of the users and reduce volatility?

Minimum trading unit will be driven by transaction fees and I don't see small purchases having any effect at all on short squeezes because anyone that close to the edge risks a 5% penalty for having a margin call executed on them.

[Empirical1]

Would making the minimum trading unit bigger also decrease the volatility?

I.e Lets say 1 Bitcoin was trading at $1000

If the minimum trading unit was $0.10 then manipulating the price a few $ might start a small short squeeze.
If the minimum trading unit was $1 then manipulating the price tens of dollars might have the same effect.
" " " was $10 then they would have to move it a few hundred $.

Obv. $10 would be extreme, but as the blocks will be processing slower than real time trading, Bitshares hopefully won't be practical for HFT bots or people trying to make 0.X% gains so you should be able to make the trading unit bigger than most other platforms which should also suit the slightly longer-term trading mindset of the users and reduce volatility?

[bytemaster]

Yes, you have just quoted two phases of our design and we would like to experiment with interest as an alternative iteration.

[theoretical]

there is no interest charged for holding a short position

I swear there's at least one official I3 video which stated that dividends generated by BitUSD held as collateral would be paid to BitUSD long positions.  And it was stated that this would be equivalent to charging interest on short positions.

The top post of the top sticky in this forum states:

short positions will pay a 5% borrowing cost.

This sticky makes it sound more like a one-time fee than an ongoing interest charge.

Also, If there is a 5% fee / interest on short positions, doesn't greatly increasing the margin requirement (from 2x to 10x) also greatly increase the amount the price has to move to overcome the fee enough to break even (let alone make a profit)?

If there is no 5% fee / interest on short positions, doesn't that reduce pressure on short holders to take profits?  If there's less pressure on short holders to take profits, doesn't that hurt market liquidity and make the price less stable?

[bytemaster]

How is my description not accurate?

My main objection is this:  What Bad Things would happen if you increased the margin requirement to 1000000x from 2x?

How do you know that 10x margin requirement isn't enough to make those Bad Things happen?

If nothing else, with 10x margin requirement, I'm thinking many investors will be scared away by the small upside and large downside.  The remaining capital will be stretched quite thinly, and most of it will be inefficiently deployed insuring extreme events.

In other words, I think 10x margin is so favorable to BitUSD holders that there won't be enough BitUSD short sellers to make the system work at any price.

Because there is no interest charged for holding a short position, taking a short position is a way for long-time-holders (like most of us in BTC) to earn a return betting on the growth of the system.   Remember, the 10x margin increase does not actually change the risk profile for the short position except in extreme events, it just changes the amount of leverage they can apply.   The extreme events are less likely to occur with a 10x margin than a 2x margin.

This will simply make BitUSD more scarce and the demand for it will drive the value of BTS up.

You do make a good question regarding why not use 1000000x vs 2x vs 1.0000001x.   My answer to you is that increasing the margin requirement gives non-linear reduction in risk.  The benefits of going from 1.000001x to 1.1x are greater than the benefits of going from 1.1 to 1.2.... Thus the higher the margin the less benefit you get per increase.  There is a bell curve an the margin requirement determines how many standard deviations away from normal market conditions a price change of a given magnitude is likely to occur.   Going from 1.000001 to 1.000002 almost doubles the area under the risk curve, but only handles an insignificant fraction of the risk.   Going from 2x to 10x goes from covering 90% of the risk to 99.99% of the risk.  Going from 10x to 10000000x only covers you for events that are unlikely in our lifetime. 

The numbers used are merely an example of the mathematical relationship between increasing margin and reducing risk. 

[dv]

How is my description not accurate?

My main objection is this:  What Bad Things would happen if you increased the margin requirement to 1000000x from 2x?

How do you know that 10x margin requirement isn't enough to make those Bad Things happen?

If nothing else, with 10x margin requirement, I'm thinking many investors will be scared away by the small upside and large downside.  The remaining capital will be stretched quite thinly, and most of it will be inefficiently deployed insuring extreme events.

In other words, I think 10x margin is so favorable to BitUSD holders that there won't be enough BitUSD short sellers to make the system work at any price.

Agreed. Especially when combining with the enforcement of 1% minimal market depth, perhaps it's very hard to make the system start working.

[theoretical]

How is my description not accurate?

My main objection is this:  What Bad Things would happen if you increased the margin requirement to 1000000x from 2x?

How do you know that 10x margin requirement isn't enough to make those Bad Things happen?

If nothing else, with 10x margin requirement, I'm thinking many investors will be scared away by the small upside and large downside.  The remaining capital will be stretched quite thinly, and most of it will be inefficiently deployed insuring extreme events.

In other words, I think 10x margin is so favorable to BitUSD holders that there won't be enough BitUSD short sellers to make the system work at any price.

[bytemaster]

Your description is not quite accurate, but we have ultimately addressed this issue by simply increasing the collateral requirements to 10x from 2x.  The probability of a blowout when down by more than 5x because it is non-linear.

[theoretical]

Let me see if I understand the problem you're trying to solve in this thread.

Let's say Alice shorts 1000 BitUSD at a price of 0.1 BTS / BitUSD.  Bob has a pre-existing bid order which is matched to Alice's short sale.  Alice and Bob are each required to have 100 BTS in their account for this match to occur.  (For simplicity, I'm neglecting fees.)

The 200 BTS (100 each from Alice and Bob) is set aside into a "pie" that Alice initially "owns" 50% of.  Alice's equity -- her piece of the pie -- as a function of the BTS / BitUSD price is given by f(x) = 200 - 1000*x.  Initially her equity as given by this formula is f(0.1) = 100 BTS, exactly equal to the capital she contributed; it decreases as the price increases until Alice is "wiped out" (zero equity) when the price is 0.2 BTS.

My understanding is that Alice's position will be liquidated if the price rises above some liquidation threshold L.  (I think L is the average of the initial price and the wipeout price, 0.15 BTS, but I'm not sure about this.)  This means that, if the market price rises above L, a "liquidation" bid order will be placed on Alice's account automatically by the system.  If this order is matched, then the resulting purchase price is paid out of the 200 BTS "pie" set aside earlier, the negative dollars are wiped from Alice's account, and she receives the remaining BTS from the pie; the liquidation order is priced such that this remainder is exactly equal to her equity (again, I'm ignoring real-world complications like the 5% liquidation fee, or what happens if the liquidation order is only partially filled).

However, if the price rises above 0.2 BTS / BitUSD, then the liquidation bid order cannot be matched.

The fact that some BitUSD "escaped" being destroyed means that the total market capitalization (as determined by number of BitUSD in circulation times the market price) can potentially exceed its total collateral value (the sum of all the pies of all short accounts, minus the sum of all the short-holders' equity).

Is this a fair description of the problem?

[Liberty]

You can also glean ideas by searching for post-crash proposals to avoid flash crashes in markets.

Can you give a few starting points?

Here is a detailed analysis of the May 5, 2010 "flash crash":

http://www.sec.gov/news/studies/2010/marketevents-report.pdf

YouTube is a source for many congressional hearings. The video series "Federal Regulation of Financial Markets" is close. I don't have links for the testimonies that I'm familiar with. The ones you are interested in are the videos of exchange executives speaking about the systems already in place and in development to avoid problems (and thereby encouraging legislation that makes their solution a costly compliance requirement for competitors). To see the truly useful recommendations it helps to start with an understanding that everything seen there (especially from the politicians running the show) is posturing for increased theft from someone.

A Google search for "Interval Price Limits" and "Reasonability Limits" will get you to information about one solution that has spread to other exchanges. The basic idea is to impose artificial trade limits on price volatility. I'm not going to risk giving examples of how I think trade limits could be implemented by III. I doubt any technical solution will be immune to dedicated manipulators with resources, but there are ways to make it more difficult. Volatility diminishes as the number of participants increases. An incentive that creates high market participation is a better approach than price limits, but markets still evolve to have whale participants that can use their influence to direct price movements for their gain. I like how derivatives provide a reward for finding market consensus (and punish those that stray), but first you need markets with higher participation than the derivatives themselves can influence.

[toast]

You can also glean ideas by searching for post-crash proposals to avoid flash crashes in markets.

Can you give a few starting points?

[Liberty]

"Solution to High Volatility potentially breaking the BitUSD Peg"

Traditional derivatives exchanges use Interval Price Limits and Reasonability Limits. Anchor price can be set from that last (or recent) block prices. The limits also help traders avoid price error. You can also glean ideas by searching for post-crash proposals to avoid flash crashes in markets.

[bytemaster]

You don't seem to understand the mechanics or economics behind how BitShares X works.  You cannot resort to price fixing and the collective capitalization of BTS holders has nothing to do with begging BitAPPL to APPL.     You just need enough people willing to speculate on BitAPPL but the amount they each speculate is irrelevant.

As long as there are currently people who are essentially receiving early shares in these assets at extremely low prices, there is almost no way they don't sell far below real world market value. Essentially, for all of these assets, as long as there are large amounts of holders before the public gets full access to it, their values will NOT be pegged to real world values for a very long time, because no matter what, these early holders will be tempted to sell out early when they see such a large percentage on their gains.

No one gets early shares in these assets below the peg.

[zavtra]

You don't seem to understand the mechanics or economics behind how BitShares X works.  You cannot resort to price fixing and the collective capitalization of BTS holders has nothing to do with begging BitAPPL to APPL.     You just need enough people willing to speculate on BitAPPL but the amount they each speculate is irrelevant.

As long as there are currently people who are essentially receiving early shares in these assets at extremely low prices, there is almost no way they don't sell far below real world market value. Essentially, for all of these assets, as long as there are large amounts of holders before the public gets full access to it, their values will NOT be pegged to real world values for a very long time, because no matter what, these early holders will be tempted to sell out early when they see such a large percentage on their gains.

[bytemaster]

Ok so pretend this thread is titled "Solution to High Volatility potentially breaking the BitAPPL Peg". You didn't solve the problem.

The value for APPL will come from the value APPL produces. If BitShare Holders simply see no value in APPL, there can not possibly be a peg.

We of course cannot flood the market with a BitAsset for every company right away, because Bitshare holders simply do not have the collective market capitalization to match real world prices.

I suppose what you can do is create Precious Metal BitAssets, and then tie the BitUSD DIRECTLY to those BitAssets in terms of real world prices.

For instance, an ounce of silver costs 20 dollars. Therefore, you may purchase 1 BitSilverOunce for exactly 20 BitUSD. In addition, BitSilverOunce may not be purchased via any other method, thus tying the value of BitUSD to a virtual precious commodity (which is a lot more to say than for the actual USD).

We need to build this system from the ground up, and not worry so much about what the current "real world" pricing will be for the time being, because if you truly believe in this system, what you are dealing with right here right now WILL BE the 'real world'.

You don't seem to understand the mechanics or economics behind how BitShares X works.  You cannot resort to price fixing and the collective capitalization of BTS holders has nothing to do with begging BitAPPL to APPL.     You just need enough people willing to speculate on BitAPPL but the amount they each speculate is irrelevant.

[zavtra]

Ok so pretend this thread is titled "Solution to High Volatility potentially breaking the BitAPPL Peg". You didn't solve the problem.

The value for APPL will come from the value APPL produces. If BitShare Holders simply see no value in APPL, there can not possibly be a peg.

We of course cannot flood the market with a BitAsset for every company right away, because Bitshare holders simply do not have the collective market capitalization to match real world prices.

I suppose what you can do is create Precious Metal BitAssets, and then tie the BitUSD DIRECTLY to those BitAssets in terms of real world prices.

For instance, an ounce of silver costs 20 dollars. Therefore, you may purchase 1 BitSilverOunce for exactly 20 BitUSD. In addition, BitSilverOunce may not be purchased via any other method, thus tying the value of BitUSD to a virtual precious commodity (which is a lot more to say than for the actual USD).

We need to build this system from the ground up, and not worry so much about what the current "real world" pricing will be for the time being, because if you truly believe in this system, what you are dealing with right here right now WILL BE the 'real world'.

[toast]

The solution is simple, friends!

In reality, fiat currencies will not survive long-term. There is no viable reason to invest in BitUSD, because it is ultimately worthless.

The total number of BitUSD shares should be directly equivalent to the number of USD in circulation (say a few trillion, give or take 10 trillion, plus the total number of mass-produced counterfeits in circulation).

Perfect, now we know that in relation to BTS, USD is actually going to be worth very little. As more nations enter currency crisis, such solutions such as BTS will be seen as viable alternatives, and yet when people finally come into BTS, do you think the first thing they are going to buy will be BitUSD?

NO! That'd be like buying BitZimbabwe or BitGermanMark.

People will more likely be looking to invest directly in real companies such as BitAPPL, as well as DAC's such as Ticketcoin/Biticket.

Think real hard about it. BitUSD in fact should be the last thing we are even concerned about.

Ok so pretend this thread is titled "Solution to High Volatility potentially breaking the BitAPPL Peg". You didn't solve the problem.

[bytemaster]

1 BitUSD will eventually be worth 0 when the dollar dies, but 1 BTS will eventually be worth millions.   Future value and present value are two entirely different things.

[zavtra]

The solution is simple, friends!

In reality, fiat currencies will not survive long-term. There is no viable reason to invest in BitUSD, because it is ultimately worthless.

The total number of BitUSD shares should be directly equivalent to the number of USD in circulation (say a few trillion, give or take 10 trillion, plus the total number of mass-produced counterfeits in circulation).

Perfect, now we know that in relation to BTS, USD is actually going to be worth very little. As more nations enter currency crisis, such solutions such as BTS will be seen as viable alternatives, and yet when people finally come into BTS, do you think the first thing they are going to buy will be BitUSD?

NO! That'd be like buying BitZimbabwe or BitGermanMark.

People will more likely be looking to invest directly in real companies such as BitAPPL, as well as DAC's such as Ticketcoin/Biticket.

Think real hard about it. BitUSD in fact should be the last thing we are even concerned about.

[bytemaster]

The key to a successful crypto network is that as long as the rules of the transaction ledger are followed there is nothing to prosecute and all is fair.  If the rules allow manipulation by anonymous parties at a profit then the rules need to change.   Resorting to criminal prosecution of someone doing something 'by the rules' is a shortcut and we should strive to find a better solution.

I am working on this issue very much right now.  New thread will be dedicated to making the initial chain bullet proof.

[bitbadger]

I solved a major issue in the design of BitShares X as it relates to how we handle the case where the collateral is insufficient to back the short position and thus the result is unbacked BitUSD in circulation.    Considering the value of BitShares X is directly related to the degree to which the holders of BitShares X are willing to guarantee the purchasing power of BitUSD it seems that a decentralized Bank should take the same approach as their centralized counterparts...   Sell new shares in the bank to raise capital to cover the losses.   In effect all BTS holders would provide 'insurance' against the 50% discontinuity event that would blow out a short position and leave unbacked BitUSD.  The reason why BTS longs would insure the BTS shorts is because the entire value proposition of BitShares X is derived from the promise of BitUSD remaining pegged.   If the BTS holders can increase the value of BTS by providing this insurance against a very rare event, it should in turn increase confidence and thus increase the value of BTS.

This change would shift the losses that BitUSD holders would currently pay to the BTS holders and thus collectively BTS holders are backing all BitUSD and BitUSD holders have something with much lower risks in these rare events.

It is curious to see a proclaimed student of Austrian Economics turn to central bank intervention of markets. A liquidity problem seems like the real risk, and that is a market participation problem. Is it still correct that people get to own BitUSD from ownership of PTS (and now also AGS) at time of fork, and purchasing power increases 5% per year even though they never participate in the markets? If that is true then it seems natural that market participation would be a problem. People would hoard the stronger form of a currency (BitUSD) and trade the weaker one (USD). I presume I don't understand your plan because I don't see a market incentive to sell BitUSD as the market discovers it to have greater value than real USD.

Many replies to your post indicate a general understanding that your central bank intended to create new BTS as necessary to fund the taking long positions by the bank. It was interesting to see that some even liked the idea of devaluing BTS (which imbalances all other products) to attempt to balance one product. The way I read your post however is that you intend to create a central bank that people can invest existing BTS into. I much prefer that idea, but wonder how that too would work. Investment in this bank subjects capital to more risk; do you have a reward in mind for this risk, and where does this reward come from? Do you intend for the actions of this bank to be automated and not otherwise subject to moral hazard? How do you imagine events would play out if Gresham's Law causes BitUSD to become systemically more valuable than USD in a way that the bank can not compensate for? Do you imagine a bank for each product, or one bank for a group of products?

You may find this a useful perspective on central banking: http://www.cato.org/multimedia/events/31st-annual-monetary-conference-panel-1-100-years-fed-what-have-we-learned.

I do not think you understand what BitUSD is.  BitUSD is pegged to the dollar.  BitUSD are created by using BTS as collateral.  Going short BitUSD is essentially going long BTS.  Going long BitUSD is essentially going short BTS.  You are betting on how many USD will buy one BTS; or conversely, how many BTS will buy one dollar.  Same thing.  BitUSD should never differ much from the actual price of the actual USD.  If people want "good money" per your Gresham's Law reference, then they will have to go somewhere else, because that's NOT what BitUSD is designed to do.

bytemaster: I have been thinking about things and the need to make this absolutely bulletproof.  If it catches on as we hope it will, it will be subject to attacks.  People will be trying to break it constantly.  Attempts at market manipulation will be a daily occurrence.... and unlike Wall Street, access to the network is unlimited and anonymous.  So we have no recourse, no ability to find people's names and prosecute them if they intentionally foul things up.  Individual nodes will be hacked or DDOS attacked to reduce liquidity.  It does not take much DDOS power to derail a home PC with a consumer level ISP connection.

So... how is the Peer-to-Peer aspect of BTS implemented?  Are IP addresses printed in the transactions?  Is there a TOR-like system of IP obfuscation, or are all IP's in the clear?  What happens to the orders/positions generated by a user/node, if that node is knocked off the network?

[Liberty]

I solved a major issue in the design of BitShares X as it relates to how we handle the case where the collateral is insufficient to back the short position and thus the result is unbacked BitUSD in circulation.    Considering the value of BitShares X is directly related to the degree to which the holders of BitShares X are willing to guarantee the purchasing power of BitUSD it seems that a decentralized Bank should take the same approach as their centralized counterparts...   Sell new shares in the bank to raise capital to cover the losses.   In effect all BTS holders would provide 'insurance' against the 50% discontinuity event that would blow out a short position and leave unbacked BitUSD.  The reason why BTS longs would insure the BTS shorts is because the entire value proposition of BitShares X is derived from the promise of BitUSD remaining pegged.   If the BTS holders can increase the value of BTS by providing this insurance against a very rare event, it should in turn increase confidence and thus increase the value of BTS.

This change would shift the losses that BitUSD holders would currently pay to the BTS holders and thus collectively BTS holders are backing all BitUSD and BitUSD holders have something with much lower risks in these rare events.

It is curious to see a proclaimed student of Austrian Economics turn to central bank intervention of markets. A liquidity problem seems like the real risk, and that is a market participation problem. Is it still correct that people get to own BitUSD from ownership of PTS (and now also AGS) at time of fork, and purchasing power increases 5% per year even though they never participate in the markets? If that is true then it seems natural that market participation would be a problem. People would hoard the stronger form of a currency (BitUSD) and trade the weaker one (USD). I presume I don't understand your plan because I don't see a market incentive to sell BitUSD as the market discovers it to have greater value than real USD.

Many replies to your post indicate a general understanding that your central bank intended to create new BTS as necessary to fund the taking long positions by the bank. It was interesting to see that some even liked the idea of devaluing BTS (which imbalances all other products) to attempt to balance one product. The way I read your post however is that you intend to create a central bank that people can invest existing BTS into. I much prefer that idea, but wonder how that too would work. Investment in this bank subjects capital to more risk; do you have a reward in mind for this risk, and where does this reward come from? Do you intend for the actions of this bank to be automated and not otherwise subject to moral hazard? How do you imagine events would play out if Gresham's Law causes BitUSD to become systemically more valuable than USD in a way that the bank can not compensate for? Do you imagine a bank for each product, or one bank for a group of products?

You may find this a useful perspective on central banking: http://www.cato.org/multimedia/events/31st-annual-monetary-conference-panel-1-100-years-fed-what-have-we-learned.

[bytemaster]

Here is my thinking...

This attack is only made possible by the insurance system proposed in this thread which can create 'unlimited buying power'.   However, like all things there is diminishing marginal utility as buying power goes from 0 to infinity as the risk of such a market move triggering that kind of event goes to 0.   

So if we consider the likelihood of blowing out a 100% margin position to be very low... then if we provide network-wide insurance up of say 200% margin position then the probability of BTS falling from    1:1 to .25:1 is is more than 2x as unlikely as .5:1.   This puts an upper limit on the potential debasement by market manipulation or attempted short squeezes with thin markets.

Through normal operation BitUSD is taken out of circulation by market fees.   This will correct any small unbacked issuance over time.   Likewise, BTS is taken out of circulation from fees and will correct any new issuance required when a short is blown out.   

This is kind of like a limit on the FDIC insurance payout of $100K per account... in this case our limit is 4x the margin.

[jifehuang]

I have concern over short squeeze that it may be exploited by an attacker to manipulate the market. Suppose the attacker is holding plenty of xts and several bitusd. He can create an order asking 1 bitusd for 1M xts. Of course no one buys his bitusd right now.

Then he starts to buy in bitusd with his xts. By rapidly doing it, bitusd rises and margin calls are triggered and short squeeze more margin calls.  Eventually no bitusd is left for sale except for his, and margin calls are forced to buy his bitusd even though the asking price is ridiculously high.

Whether this will be successfully depending on market depth. If there are lots of bitusd listed for sell at one moment, the price for the attack will be high. However one cannot guarantee market depth at all times. It can be low sometimes, or if the attack holds lots of xts, he can still initiate the attack. Once the attack is initiated, most people may be watching instead of getting ready to catch the squeeze. More over, since margin calls are automatic, manual orders may not be added fast enough to catch the squeeze if they are not listed before the attack.

Remember that a few days ago, btc was down to 102 on btc-e? Isn't it an example of short squeeze? On traditional exchanges, most people are not playing margin account or stop buy/sell, so the implementation of short squeeze may not be easy. However on btsx margin accounts are everywhere, and short squeeze should be much easier to happen.

I have 2 ideas to solve this issue. One is to make the margin call an uncancellable limit order instead of market order. This should prevent chain effect from happening.
Second is put an up limit for bitasset every day or every certain number of blocks. This allows plenty of time for people to join to catch the squeeze.

This is called a short squeeze and if it happened there would be many people looking to enter new shorts (fully collateralized shorts) to make a profit and keep the price in line.   The challenge will be if the short squeeze happens all in one block.   I suspect there would be a large number of orders ready to catch any squeeze and put a stop to the rise.

[santaclause102]

...always thought about that: The BitAsset speculator would have to take into account not only the future price developement of his Asset but also the price developement of BTS. That would not be good for someone that is looking for a simple way to hedge something or an expert in Gold/Oil/whateverAssett who is not also an expert on the future developement of BTS price..

not a problem?

Not a problem.   

ok
can you or anyone elaborate on this?
... going short bitusd = going long bts. That's a bet in two ways with two variables that matter to the outcome of the bet... where is the fault in how I think about it?
 

[bitbadger]

The talk of robots has me thinking.  I am sure that many people will write their own robots to essentially act as market makers.  (Surely the robots mentioned previously would not be "official" in any way?  Thus creating a single point of failure...)  These privately-acting robots will always have buy orders of BitUSD at 0.99 and sell orders of BitUSD at 1.01, along with a good stock of BitUSD and BTS to adequately cover these positions.

Now there will be a good number of these critical market-makers.  Maybe even hundreds.... but a few hundreds of IP's, most of them presumably with non-redundant, non-hardened network connections, are not difficult to DDOS attack.  So what happens when critical market participants (who provide liquidity, have tons of orders in the system, and constantly generate new orders based on market conditions) suddenly drop off the network?  Is there a fail-safe somehow?  Because it seems like DDOS'ing the biggest market participants would be the first step in an attack on the market to try to break the buck.  Coupled with the fact that blocks are generated on a predictable schedule, I think there is a vulnerability there.  The blockchain itself would not be affected, but the transactions generated during the attack could unfairly distort the market. 

I suppose that slower block generation would act as somewhat of a fail-safe because of the network hashpower being knocked off, but I don't know enough about the BTS implementation of Proof-of-Stake to know if this will be meaningful or not.  I also know that the network is "Peer-to-peer" like all cryptos, but this term has many meanings.  Could a person theoretically discover every node on the network?  Or is it more like Tor, where the network is multi-layered and obscured?

[bytemaster]

The main difference of the special account and bm's solution is that the BitUSD which is used to cover the uncovered short position is not bought from the market, but took out from the former transaction fees.

Yes, you got me.
when the margin call execute, the current sell orders in the market get an absolutely power to make deal, these order get much benefit, it's unfair to others.
Bought from the market just lead  to unstable state:
the margin call execute -> buy a lot of bitusd from market -> price up -> more margin call  -> buy more -> price up -> more margin call .......
It's unstable feedback.

This is called a short squeeze and if it happened there would be many people looking to enter new shorts (fully collateralized shorts) to make a profit and keep the price in line.   The challenge will be if the short squeeze happens all in one block.   I suspect there would be a large number of orders ready to catch any squeeze and put a stop to the rise.   

[Markus]

The other solution is equivalent to that if the collateral is not enough, some BitUSD will be created directly to make a total cover. There's no more market transactions to do such a cover. That will lead to a less influence on the market.

That won't work. If the collateral is not enough there will be a surplus of BitUSD that has to be removed - not additional BitUSD to be created.

[toast]

The market will already take the locked up fees in the robot account into account as if the total XTS supply was lower. Destroying/creating and locking up / reallocating are the same thing when it comes to shares of a whole. Are you just saying one is less likely to cause problems because of exclusively psychological factors?

No. There's a little difference even considering the equivalence of destroying/creating and locking up/reallocating. Let's just use destroying and creating instead of the other two to make it more clear.

When margin call occurs, all the collateral bts is used to buy enough BitUSD to cover the short position. The current solution is that if the collateral is not enough, some bts will be created to buy BitUSD to make a total cover. 

The other solution is equivalent to that if the collateral is not enough, some BitUSD will be created directly to make a total cover. There's no more market transactions to do such a cover. That will lead to a less influence on the market.

Ah, I see - I assumed it was the second one (alt's suggestion?).

[ul]

The market will already take the locked up fees in the robot account into account as if the total XTS supply was lower. Destroying/creating and locking up / reallocating are the same thing when it comes to shares of a whole. Are you just saying one is less likely to cause problems because of exclusively psychological factors?

No. There's a little difference even considering the equivalence of destroying/creating and locking up/reallocating. Let's just use destroying and creating instead of the other two to make it more clear.

When margin call occurs, all the collateral bts is used to buy enough BitUSD to cover the short position. The current solution is that if the collateral is not enough, some bts will be created to buy BitUSD to make a total cover. 

The other solution is equivalent to that if the collateral is not enough, some BitUSD will be created directly to make a total cover. There's no more market transactions to do such a cover. That will lead to a less influence on the market.

[clout]

Why not charge a capital gains fee of 10-20% for short positions. Hold it in the blockchain as collateral for that particular asset. If bts appreciates in price relative to the asset by a certain percent, lets say 50% then redistribute the bts from the fee as dividend for the stakeholders in network.

[Markus]

To prevent margin call flash crash:
How about triggering a margin call not based on the current price but based on the trailing average price of the last x blocks?

[alt]

The main difference of the special account and bm's solution is that the BitUSD which is used to cover the uncovered short position is not bought from the market, but took out from the former transaction fees.

Yes, you got me.
when the margin call execute, the current sell orders in the market get an absolutely power to make deal, these order get much benefit, it's unfair to others.
Bought from the market just lead  to unstable state:
the margin call execute -> buy a lot of bitusd from market -> price up -> more margin call  -> buy more -> price up -> more margin call .......
It's unstable feedback.

[jifehuang]

I guess the main concern here is how to avoid chain effect on margin calls when someone pulls bitusd double price. Should there be a daily up/down limit?

[bytemaster]

Why we must cover all the bitusd on margin call? In the case when the system is out of bitusd sell order we won't even be able to cover it no matter how many more xts created. Could it be simply to sell the collateral, and try to cover as much bitusd as possible. If not sufficient, take the rest as a global bad debt?
One way to deal with bad debt is to use transaction fees or interest obtained later to cover it before paying back to stakeholders. Considering the bad debt is a rare event, it could be get controlled within a reasonable level. Just my 2 cents.

This is effectively the solution.   

[jifehuang]

Why we must cover all the bitusd on margin call? In the case when the system is out of bitusd sell order we won't even be able to cover it no matter how many more xts created. Could it be simply to sell the collateral, and try to cover as much bitusd as possible. If not sufficient, take the rest as a global bad debt?
One way to deal with bad debt is to use transaction fees or interest obtained later to cover it before paying back to stakeholders. Considering the bad debt is a rare event, it could be get controlled within a reasonable level. Just my 2 cents.

[toast]

Also, Alt's idea about the market robot is also very interesting. The main difference of the special account and bm's solution is that the BitUSD which is used to cover the uncovered short position is not bought from the market, but took out from the former transaction fees. I'd like to prefer Alt's idea once taken into account the solution's influence on the market.

The market will already take the locked up fees in the robot account into account as if the total XTS supply was lower. Destroying/creating and locking up / reallocating are the same thing when it comes to shares of a whole. Are you just saying one is less likely to cause problems because of exclusively psychological factors?

Am I going crazy here?

Imagine this:
current price is 1000 usd/xfs
some sell orders for bitusd in the market, maybe 10,000,000.
first I offer an sell order: sell 1000,000 bitusd with price 1usd/xfs
then I buy all 10,000,000 bitusd  with 10,000 xfs.
now, the margin call execute.
all the backup xfs will buy my bitusd with price 1usd/xfs. I can get 1000,000 xfs.
If I change the price to 0.0001usd/xfs, I can destroy the BTS system.

Would this not be just as much of a problem if the market robot ran out of funds? I thought you guys had a solution to this trade pattern earlier

[ul]

If we have a company account, the margin call is very easy:
transfer all short position and backup to the company account.

these short position will be cover with the bitasset in the account.

And I think It's very cool when you give the BTS  such a robot account.

margin call executed because price of bitusd rise up.
In this situation, we still use backup xfs to buy bitusd with any price, even create new xfs.
I think it's not ok for price stable.
Maybe there huge danger to the whole network.

I think transfer the short position and backup xfs to the robot account is more simple.

Yes. If the price suffers from a sharp fluctuation, we can expect the emergence of a large amount of uncovered short positions. According to the lastest strategy by bm, enough bts will be created by the system to buy adequate BitUSD to cover the short position (regardless whatever high price of the bitUSD) . Then the price of BitUSD will be pushed even higher and that will help to further exacerbate the market fluctuation. It's a positive feedback progress.

So I think maybe we can just ignore the issue and just leave the unbacked BitUSD in circulation. The total amount of BitUSD and BTS(collateral) need not have some fixed relationship. All these changes will ultimately be reflected by the effective market.

Also, Alt's idea about the market robot is also very interesting. The main difference of the special account and bm's solution is that the BitUSD which is used to cover the uncovered short position is not bought from the market, but took out from the former transaction fees. I'd like to prefer Alt's idea once taken into account the solution's influence on the market.

[willj]

To clarify: This would be the case when BTS would decrease in price more than twice compared to the point in time when the short positions was taken (if a short position requires 2x Asset value in BTS)?

so i guess this describes the margin call case. How can the system run out of bitshares like mentioned in the OP?

If the starting collateral is 100 BTS backing 50 USD... and the price of 50 USD rises to more than 100 BTS so fast that the market couldn't cover the position in time then there would not be enough BTS in the short position to buy back 50 USD... so the network would be forced to create new BTS to close the short position.   

As I put money in BitShareX, so I keep thinking it in my head.
Please think about this idea.

BitShareX system must very powerful, and make sure BTS has the real value.
BTS's value is real and stable, the marketing price only show how much people like it. But not effect BTS's real value.

My idea is:
suppose BTS born by 4000 BTC(at 1000 USD), and 1,500,000 PTS(at 40 USD)
INIT value:1 BTS_value == (4000 * 1000 + 1,500,000 * 40)/ 4000,000 == 16

when 4,000,000,000 USD want join BitShareX for shares,
BTS born 4,000,000,000 bitUSD,
1 BTS_value == 1016 + fees

BitShareX only decrease value when value decrease.
e.g. people get money back from BitShareX(and do mandatory liquidation base on BTS_value).

if you want 5% double shares, buy bitUSD;
if you want own the growing of the BitShareX system, buy BTS with price higher than the BTS_value and compete with more likers;

if you want sell 1000 bitUSD value BTS at 1 bitUSD, it is up to you. but this not effect the BTS_value.

[alt]

Imagine this:
current price is 1000 usd/xfs
some sell orders for bitusd in the market, maybe 10,000,000.
first I offer an sell order: sell 1000,000 bitusd with price 1usd/xfs
then I buy all 10,000,000 bitusd  with 10,000 xfs.
now, the margin call execute.
all the backup xfs will buy my bitusd with price 1usd/xfs. I can get 1000,000 xfs.
If I change the price to 0.0001usd/xfs, I can destroy the BTS system.

[alt]

If we have a company account, the margin call is very easy:
transfer all short position and backup to the company account.

these short position will be cover with the bitasset in the account.

And I think It's very cool when you give the BTS  such a robot account.

margin call executed because price of bitusd rise up.
In this situation, we still use backup xfs to buy bitusd with any price, even create new xfs.
I think it's not ok for price stable.
Maybe there huge danger to the whole network.

I think transfer the short position and backup xfs to the robot account is more simple.

[bytemaster]

That's what we're already doing (the "rules" are not exactly the same, like your margin suggestion is 10% etc), except we're creating/destroying BTS instead of having an account that does trades. As long as your design only has one "network market bot", these designs are equivalent to the user when you view your funds as % BTS on market

but  still have something I want to issue:
1. bitusd never be create
2. can we limit total amount of bts not more then 400,000,000?

In this particular case, BTS will only grow in the event BitUSD needs to be honored.  The supply is limited and shrinking except in rare events (that may never happen).

Dan, do you think adjusting the transaction fee formula in a way that it would get rid of BTS more quickly if ever the amount of outstanding BTS was more than 4 million is an option? I suggested that in the second post in this thread.
Or do you think this scenario is too rare to be considered?

I think it is very rare and that any increase in transaction fee would harm the network by reducing transaction volume and thus devaluing BTS.

There is one additional fact that should be considered... from time to time the network destroys BitUSD in market fees.  There is no need to issue new BTS so long as total BitUSD in circulation is less than total short positions.  Thus the network has a built in buffer from fees that prevent this kind of inflation from having much impact on the long term supply.    This is all just handling very rare corner cases that when they do happen will probably result in insignificant inflation that is unlikely to be more than prior deflation.  Any very rapid declines probably mean something else bad happened.   

[Markus]

That's what we're already doing (the "rules" are not exactly the same, like your margin suggestion is 10% etc), except we're creating/destroying BTS instead of having an account that does trades. As long as your design only has one "network market bot", these designs are equivalent to the user when you view your funds as % BTS on market

but  still have something I want to issue:
1. bitusd never be create
2. can we limit total amount of bts not more then 400,000,000?

In this particular case, BTS will only grow in the event BitUSD needs to be honored.  The supply is limited and shrinking except in rare events (that may never happen).

Dan, do you think adjusting the transaction fee formula in a way that it would get rid of BTS more quickly if ever the amount of outstanding BTS was more than 4 million is an option? I suggested that in the second post in this thread.
Or do you think this scenario is too rare to be considered?

[bytemaster]

There is an automated stop loss that kicks in long before collateral runs out in 99% of the cases.   

This is only for the rare extreme events where the stop loss cannot kick in fast enough.

[NewMine]

If the starting collateral is 100 BTS backing 50 USD... and the price of 50 USD rises to more than 100 BTS so fast that the market couldn't cover the position in time then there would not be enough BTS in the short position to buy back 50 USD... so the network would be forced to create new BTS to close the short position. 

Let me get this straight:

1 Bitshares =1 bitUSD.
2 Bitshares are required to short 1 bitUSD (2:1ratio no mater the price).
Maximum risk of the short trader is 2 Bitshares.
maximum risk to the network is limitless, in theoretical extremes.
Maximum risk of long bitUSD trader is 1 bitUSD.
Maximum profit of short Trader is 1 BitUSD.
Maximum profit of long Trader is limitless.
Maximum profit to network does not exist.


Isn't this just shifting the risk from the person who shorts USD to every BTS holder through dilution? I think an automated stop loss would do the trick. One that automatically buys back the short position when it nears the collateral limits.  Although you could then have whales attempting to hit stop losses to gain momentum especially if there is an open ledger with these types of transactions visible.

Am I seeing this right?

[toast]

The more I think about it the more I like this idea

Before I could never get a feel for what the "equilibrium equations" for BTS market forces are, now things fit more nicely. BTS holders have a few more incentives in the right places

[bytemaster]

That's what we're already doing (the "rules" are not exactly the same, like your margin suggestion is 10% etc), except we're creating/destroying BTS instead of having an account that does trades. As long as your design only has one "network market bot", these designs are equivalent to the user when you view your funds as % BTS on market

but  still have something I want to issue:
1. bitusd never be create
2. can we limit total amount of bts not more then 400,000,000?

In this particular case, BTS will only grow in the event BitUSD needs to be honored.  The supply is limited and shrinking except in rare events (that may never happen). 

[alt]

That's what we're already doing (the "rules" are not exactly the same, like your margin suggestion is 10% etc), except we're creating/destroying BTS instead of having an account that does trades. As long as your design only has one "network market bot", these designs are equivalent to the user when you view your funds as % BTS on market

but  still have something I want to issue:
1. bitusd never be create
2. can we limit total amount of bts not more then 400,000,000?

[alt]

If we have a company account, the margin call is very easy:
transfer all short position and backup to the company account.

these short position will be cover with the bitasset in the account.

And I think It's very cool when you give the BTS  such a robot account.

when the margin call execute, we have right to sell all the backup to public to cover the short position.
but now, the buy order(for xts) match from the margin call  have an opportunity to get more benefits from backup,
this result in backup  not enough, it's unfair.

And I think this rule  maybe attract people to rob the backup xts from control price.

[toast]

That's what we're already doing (the "rules" are not exactly the same, like your margin suggestion is 10% etc), except we're creating/destroying BTS instead of having an account that does trades. As long as your design only has one "network market bot", these designs are equivalent to the user when you view your funds as % BTS on market

[alt]

Do you not see how having such a robot account lock up backup funds is *exactly equivalent* to the network printing/destorying shares on demand? If they're out of circulation then supply-side things like the same in both cases. Or am I missing part of what this network account does?

we can design  rules for this robot account.
For example, if we want to keep price stable, we can give this rule:
when  the price rise up to 10%, offer a sell order to the market. otherwise, offer a buy order.

Destroy bitusd will make the amount of  bitusd keep down from amount of short position.this can't revert.
one day, maybe we have 100,000 bitusd short position, but only 10,000 bitusd. I don't know if it's a problem.

with the robot account. the BTS network will keep be a real zero-sum.

[bytemaster]

To clarify: This would be the case when BTS would decrease in price more than twice compared to the point in time when the short positions was taken (if a short position requires 2x Asset value in BTS)?

so i guess this describes the margin call case. How can the system run out of bitshares like mentioned in the OP?

If the starting collateral is 100 BTS backing 50 USD... and the price of 50 USD rises to more than 100 BTS so fast that the market couldn't cover the position in time then there would not be enough BTS in the short position to buy back 50 USD... so the network would be forced to create new BTS to close the short position.   

...always thought about that: The BitAsset speculator would have to take into account not only the future price developement of his Asset but also the price developement of BTS. That would not be good for someone that is looking for a simple way to hedge something or an expert in Gold/Oil/whateverAssett who is not also an expert on the future developement of BTS price..

not a problem?

Not a problem.   

[santaclause102]

To clarify: This would be the case when BTS would decrease in price more than twice compared to the point in time when the short positions was taken (if a short position requires 2x Asset value in BTS)?

so i guess this describes the margin call case. How can the system run out of bitshares like mentioned in the OP?

...always thought about that: The BitAsset speculator would have to take into account not only the future price developement of his Asset but also the price developement of BTS. That would not be good for someone that is looking for a simple way to hedge something or an expert in Gold/Oil/whateverAssett who is not also an expert on the future developement of BTS price..

not a problem?

[bytemaster]

So is the new solution basically "the exact moment all collateral is gone, fill everyone's buy-BTS-for-BitAsset order filled for free"? Or is it at the price the shorts blow out at?

This would only affect ONE position at a time and not be a universal thing.  If one position goes insolvent it is unlikely all will at the same time.  Further more, anytime margin is called successfully there is a 5% fee, thus these fees will probably make up for the times where margin runs out entirely.

[toast]

This is quite an interesting concept, you have the potential for a huge reallocation of BTS towards people who were holding assets at the time of a catastrophe... give the next set of shareholders a chance try and see if they can keep the bank solvent

[toast]

Maybe it should buy 10% of unbacked assets per block or something to give real life BTS holders time to react and buy back assets at a loss

[toast]

So is the new solution basically "the exact moment all collateral is gone, fill everyone's buy-BTS-for-BitAsset order filled for free"? Or is it at the price the shorts blow out at?

[toast]

en.....
I still have a little doult about it.
When I short the bitusd, I promiss with double price for backup.
I think double price is the most I should be responsible.
the solution you point  out take more from the shareholder, It's unfair.

I still prefer to force  offer an buy order: sell  the backup bts to buy full bitusd to cover the short position.
I have give all my backup, it's enough for me.
for this buy order, If it can match the sell order in the marketing , I don't care, it's the choice of the marketing.

If we have a company account, the margin call is very easy:
transfer all short position and backup to the company account.

these short position will be cover with the bitasset in the account.

And I think It's very cool when you give the BTS  such a robot account.

Do you not see how having such a robot account lock up backup funds is *exactly equivalent* to the network printing/destorying shares on demand? If they're out of circulation then supply-side things like the same in both cases. Or am I missing part of what this network account does?

[alt]

If we have a company account, the margin call is very easy:
transfer all short position and backup to the company account.

these short position will be cover with the bitasset in the account.

And I think It's very cool when you give the BTS  such a robot account.

[alt]

en.....
I still have a little doult about it.
When I short the bitusd, I promiss with double price for backup.
I think double price is the most I should be responsible.
the solution you point  out take more from the shareholder, It's unfair.

I still prefer to force  offer an buy order: sell  the backup bts to buy full bitusd to cover the short position.
I have give all my backup, it's enough for me.
for this buy order, If it can match the sell order in the marketing , I don't care, it's the choice of the marketing.

[santaclause102]

I think what he's saying is that in the case of more than 2x drop of BTS/BitUSD and collateral runs out then the network will print more BTS to buy back BitUSD... this is paid for by BTS holders (remember "create/destroy" and "redistribute" are completely equivalent)

To clarify: This would be the case when BTS would rise in price more than twice compared to the point in time when the short positions was taken (if a short position requires 2x Asset value in BTS)?

I thought it was the opposite?

You are right. I confused it. It should read: This would be the case when BTS would decrease in price more than twice compared to the point in time when the short positions was taken (if a short position requires 2x Asset value in BTS)?

Something else: Going long BitUSD (=buying BitUSD) doenst need a collateral, correct?

[bytemaster]

Toast got it right. 


Sent from my iPhone using Tapatalk

[toast]

I think what he's saying is that in the case of more than 2x drop of BTS/BitUSD and collateral runs out then the network will print more BTS to buy back BitUSD... this is paid for by BTS holders (remember "create/destroy" and "redistribute" are completely equivalent)

To clarify: This would be the case when BTS would rise in price more than twice compared to the point in time when the short positions was taken (if a short position requires 2x Asset value in BTS)?

I thought it was the opposite?

[alt]

Dan, waiting for your solution 

[bytemaster]

Actually you all are missing the idea entirely.   But I cannot answer from phone.   


Sent from my iPhone using Tapatalk

[santaclause102]

the case where the collateral is insufficient to back the short position and thus the result is unbacked BitUSD in circulation.

To clarify: This would be the case when BTS would rise in price more than twice compared to the point in time when the short positions was taken (if a short position requires 2x Asset value in BTS)?

...always thought about that: The BitAsset speculator would have to take into account not only the future price developement of his Asset but also the price developement of BTS. That would not be good for someone that is looking for a simple way to hedge something or an expert in Gold/Oil/whateverAssett who is not also an expert on the future developement of BTS price....

[willj]

 

not get it.

I think if UserA can create as many bitUSD as wanted, than the going short must be success.

[alt]

Actual, we can give the company more effective rules to keep marketing stable, like:
when price rice 10%, sell some asset. 20% -> sell more ....
otherwise, buy some ...

[alt]

short posision will be covered by the bit assets in the Company account.
what I mean destroy is for now, the new solusion will not destroy it , but transfert the fee to the Company account.

all the bit assets which are destroy should transfer to this account.

If they are destroyed they can't be transferred because they don't exist anymore.

and all short possision which can not cover should transfer to this account.

A short position without collateral has a negative value. Nobody would want to have it - least buy it.

[Markus]

all the bit assets which are destroy should transfer to this account.

If they are destroyed they can't be transferred because they don't exist anymore.

and all short possision which can not cover should transfer to this account.

A short position without collateral has a negative value. Nobody would want to have it - least buy it.

[alt]

what's your solution?

My new solution is add a special account to the network(actually it's not an account, transaction about it is programed in the bts source code)

all the translation fee which are destroy should transfer to this account.
all the bit assets which are destroy should transfer to this account.
and all short possision which can not cover should transfer to this account.

we can give the company more effective rules to keep marketing stable, like:
when price rice 10%, sell some asset. 20% -> sell more ....
otherwise, buy some ...

[Markus]

(Only just saw you posted this in the status thread as well.)

Do you mean buying back the surplus BitUSD with newly minted BTS? You'd get my vote on that - even though it might theoretically tear down the 4M BTS limit. Maybe increase transaction fees in the unlikely event that there are more than 4M BTS around?

[bytemaster]

I solved a major issue in the design of BitShares X as it relates to how we handle the case where the collateral is insufficient to back the short position and thus the result is unbacked BitUSD in circulation.    Considering the value of BitShares X is directly related to the degree to which the holders of BitShares X are willing to guarantee the purchasing power of BitUSD it seems that a decentralized Bank should take the same approach as their centralized counterparts...   Sell new shares in the bank to raise capital to cover the losses.   In effect all BTS holders would provide 'insurance' against the 50% discontinuity event that would blow out a short position and leave unbacked BitUSD.  The reason why BTS longs would insure the BTS shorts is because the entire value proposition of BitShares X is derived from the promise of BitUSD remaining pegged.   If the BTS holders can increase the value of BTS by providing this insurance against a very rare event, it should in turn increase confidence and thus increase the value of BTS.

This change would shift the losses that BitUSD holders would currently pay to the BTS holders and thus collectively BTS holders are backing all BitUSD and BitUSD holders have something with much lower risks in these rare events.