BitShares Forum

Main => General Discussion => Topic started by: bitcrab on May 10, 2016, 06:55:57 am

Title: [BSIP#16]Optimization to Force Settlement Parameters of BitCNY
Post by: bitcrab on May 10, 2016, 06:55:57 am
A BSIP is created based on the discussion in last several weeks regarding modifying bitCNY parameters: https://github.com/bitshares/bsips/blob/master/bsip-0016.md (https://github.com/bitshares/bsips/blob/master/bsip-0016.md)

Here is the discussion thread: https://bitsharestalk.org/index.php/topic,22355.0.html (https://bitsharestalk.org/index.php/topic,22355.0.html)
There is also a discussion thread in Chinese:https://bitsharestalk.org/index.php/topic,22401.0.html (https://bitsharestalk.org/index.php/topic,22401.0.html)

Please input if you have any thoughts, if there's no persuading objections, committee will create the committee proposal in next Monday.
Title: Re: [BSIP#16]Optimization to Force Settlement Parameters of BitCNY
Post by: ebit on May 10, 2016, 06:58:17 am
 +5% good
Title: Re: [BSIP#16]Optimization to Force Settlement Parameters of BitCNY
Post by: xeroc on May 10, 2016, 07:00:43 am
... committee will create the committee proposal in next Monday.
As stated in the BSIP .. the proposal will have an expiration of 30 days to give shareholders enough time to cast a vote and shorters enough time to settled/sell
Title: Re: [BSIP#16]Optimization to Force Settlement Parameters of BitCNY
Post by: cube on May 10, 2016, 07:06:00 am
... committee will create the committee proposal in next Monday.
As stated in the BSIP .. the proposal will have an expiration of 30 days to give shareholders enough time to cast a vote and shorters enough time to settled/sell

The proposed change will affect holders of bitCNY the most.  Should you have any concerns, please voice them before the committee members make their final decision.
Title: Re: [BSIP#16]Optimization to Force Settlement Parameters of BitCNY
Post by: Chronos on May 10, 2016, 04:20:47 pm
@bitcrab there's a typo in your opening post that makes the first link invalid due to extra ']' characters.
Title: Re: [BSIP#16]Optimization to Force Settlement Parameters of BitCNY
Post by: pc on May 10, 2016, 07:33:52 pm
Here's the correct link: https://github.com/bitshares/bsips/blob/master/bsip-0016.md

This proposal is full of lies and unfounded assumptions.

Forced settlement happens at par value, therefore no "compensation" for the shorter is necessary.

Shorters can mitigate their risk by selling only for a premium above the feed. Robbing *current holders* of CNY, who may be completely unaware of this proposal, is not necessary.

The proposal does not create a new "fee". Fees in BitShares are handled separately. Calling this robbery a "fee" is a plain lie.

Forced settlement is not a "service", it is a guarantee that defines the price floor for BitAssets. Changing this guarantee in mid-game without the consent of current CNY holders is robbery. Committee members voting for such a proposal may be legally liable for the financial damages this causes.

Changing the price floor will obviously affect the trading price of CNY, therefore the claim that only forced settlement is affected is another lie.

Changing an existing guarantee in mid-game will be disastrous for the credibility of *all* BitAssets, and thus harm our entire ecosystem.

Finally, the proposal was created by Bitcrab, who is known to be in a short position and would directly profit from the proposal. He also has a history of acting in his own interests wrt forced settlement. Edit: I take back that statement, after abit's analysis

Edit: looked up some quotes
The settlement will always happen AT the feed

http://docs.bitshares.eu/_downloads/bitshares-general.pdf section 4.1.1:
Quote
The  core  product  of  BitShares  is  a  class  of  assets  referred to  as  Market-Pegged  Assets  (MPA),  BitAssets,  or  SmartCoins and  represent  a  crypto-token  that  has at  least the  value  of  the underlying asset. For instance, a bitUSD can always be sold for $1,  either  to  a  merchant  at  face-value,  or  to  the  network  (by means of settlement of a contract) in return for BitShares’ core currency (BTS) worth $1.

http://docs.bitshares.eu/bitshares/user/mpa.html
Quote
Everyone that holds your (MPA) coin in BitShares can convert the coin into the backing asset at a fair price. This procedure is called “settlement” and ensures that your MPA is always worth at least the fair price.
Title: Re: [BSIP#16]Optimization to Force Settlement Parameters of BitCNY
Post by: Chronos on May 10, 2016, 07:53:08 pm
@pc you make very good points.

From the proposal:

Quote
As a reference, TCNY has been set a 2% force settlement offset, the consequence is that almost no holder wanted to request force settlement even when the price for the collateral was very low.

Is Bitcrab involved in TCNY in any way? Perhaps this experimentation should be on that asset instead.
Title: Re: [BSIP#16]Optimization to Force Settlement Parameters of BitCNY
Post by: pc on May 10, 2016, 08:19:32 pm
Is Bitcrab involved in TCNY in any way? Perhaps this experimentation should be on that asset instead.

AFAIK the T in TCNY is for Transwiser. See the signature below bitcrab's posts. The issuer of TCNY is "transwiser-mpaadmin".
Title: Re: [BSIP#16]Optimization to Force Settlement Parameters of BitCNY
Post by: Moon on May 11, 2016, 06:47:01 am
... committee will create the committee proposal in next Monday.
As stated in the BSIP .. the proposal will have an expiration of 30 days to give shareholders enough time to cast a vote and shorters enough time to settled/sell


hey,xeroc

We all know mindphlux was a thief,

link:
https://bitsharestalk.org/index.php/topic,21212.msg275947.html#msg275947


so, Why do we support him .Almost 1 year,
today,he have 215,272,616 BTS vote
how do you think  the price fall? maybe perhaps you can see from here,

we have been around a thief

This is our committee ?
This is our shame !



please revoke your vote
Title: Re: [BSIP#16]Optimization to Force Settlement Parameters of BitCNY
Post by: xeroc on May 11, 2016, 07:00:37 am
It is a little far fetched to call the proposal "full of lies". It may present different interpretation of things though (e.g. fee vs. robbery).

Anyway, it seems to be a bad timing for starting such a discussion just when many of the community are busy helping out growing STEEM.
I certainly don't want to rush anything of that importance

I wonder if there are other businesses that would like to support this move besides transwiser
Title: Re: [BSIP#16]Optimization to Force Settlement Parameters of BitCNY
Post by: abit on May 11, 2016, 07:51:02 am
I wonder if there are other businesses that would like to support this move besides transwiser
BitCash
Title: Re: [BSIP#16]Optimization to Force Settlement Parameters of BitCNY
Post by: bitcrab on May 11, 2016, 09:44:57 am
let me repeat:

In common financial logic the collateral should not be force settled without any compensation if the collateral price does not fall below the margin call price.

In common financial logic the collateral should not be force settled without any compensation if the collateral price does not fall below the margin call price.

In common financial logic the collateral should not be force settled without any compensation if the collateral price does not fall below the margin call price.


and please think globally, the point is not that transwiser can be better as a shorter, but this change can make the bitCNY relevant business grow and benefit the whole economy.

we need to make bitCNY supply grow, to make it a useful trading basis, not just a toy,  that is why most of China community members support change.

hlet's also review the history, in v0.9, the bitCNY is there without force settlement feature, at that time bitCNY already works well, when 2.0 times come, the force settlement comes and had scared many shorters, after argument the maximum settlement volume was reduced from 20% to 2%,change always happen, it's not "mid-game" change, it's just continuous improving.

I don't think the input from @pc is persuading enough, and actually I have answered most of the questions in previous discussions, we need not to rush, but need not to fall in endless argument neither.



Title: Re: [BSIP#16]Optimization to Force Settlement Parameters of BitCNY
Post by: Chronos on May 11, 2016, 01:37:04 pm
In common financial logic the collateral should not be force settled without any compensation if the collateral price does not fall below the margin call price.
@bitcrab What do you think about a collateral offset of 0.1% instead of 0% or 1%? If the feed is accurate, that still results in guaranteed profit for the shorter.
Title: Re: [BSIP#16]Optimization to Force Settlement Parameters of BitCNY
Post by: xeroc on May 11, 2016, 01:46:08 pm
@bitcrab What do you think about a collateral offset of 0.1% instead of 0% or 1%? If the feed is accurate, that still results in guaranteed profit for the shorter.
no matter what you set it, there is no such thing as "guaranteed profit" .. you only make this x% profit if there is a settlement and YOUR call position is settled.
To be the one that gets settled, you need to have lowest collateral which leaves you position at risk of a margin call!!
Title: Re: [BSIP#16]Optimization to Force Settlement Parameters of BitCNY
Post by: pc on May 11, 2016, 06:52:33 pm

In common financial logic the collateral should not be force settled without any compensation if the collateral price does not fall below the margin call price.



Reference, please.

The term "compensation" implies that there is some damage caused by the settlement, which is often the case in "real-world" finance, but not in BitShares.

If the shorter sells BitAssets with a 1% premium he achieves the same effect for himself as with a 1% settlement penalty, only with the premium existing holders are not hurt.
Title: Re: [BSIP#16]Optimization to Force Settlement Parameters of BitCNY
Post by: pc on May 11, 2016, 06:54:24 pm
It is a little far fetched to call the proposal "full of lies". It may present different interpretation of things though (e.g. fee vs. robbery).

Calling it a fee IS a lie.

Yes, I'm using strong wording, because I have a strong opinion on people who make guarantees only to break them later. As well as on people who invent nice terms to cover up the awful truth.
Title: Re: [BSIP#16]Optimization to Force Settlement Parameters of BitCNY
Post by: bitcrab on May 12, 2016, 03:10:11 am

In common financial logic the collateral should not be force settled without any compensation if the collateral price does not fall below the margin call price.



Reference, please.

The term "compensation" implies that there is some damage caused by the settlement, which is often the case in "real-world" finance, but not in BitShares.

If the shorter sells BitAssets with a 1% premium he achieves the same effect for himself as with a 1% settlement penalty, only with the premium existing holders are not hurt.

here the finance principle are same for human society, no different.

existing holders have enough time to execute settlement with 0 offset before the change applies, they can also use the transwiser bitCNY withdraw service. they have ways to avoid the so called "1% loss" after the change.

Title: Re: [BSIP#16]Optimization to Force Settlement Parameters of BitCNY
Post by: bitcrab on May 12, 2016, 03:24:52 am
It is a little far fetched to call the proposal "full of lies". It may present different interpretation of things though (e.g. fee vs. robbery).

Calling it a fee IS a lie.

Yes, I'm using strong wording, because I have a strong opinion on people who make guarantees only to break them later. As well as on people who invent nice terms to cover up the awful truth.

the force settlement feature is new in BTS2.0, in your words we can say it also break the rule in 1.0, however this feature is good, but only when there are reasonable parameters, at first the community is not so aware of this. but now it's time to review and refine the parameters.

we always need change, be flexible, not so stubborn.

ultimately, let the vote speak.
Title: Re: [BSIP#16]Optimization to Force Settlement Parameters of BitCNY
Post by: xeroc on May 12, 2016, 07:33:12 am
The term "compensation" implies that there is some damage caused by the settlement, which is often the case in "real-world" finance, but not in BitShares.

If the shorter sells BitAssets with a 1% premium he achieves the same effect for himself as with a 1% settlement penalty, only with the premium existing holders are not hurt.
Very good point ... and part of the problem as this results in the premium we can see on the bitassets/BTS markets.

I think the question we need to answer to ourselves is this: Is a bitUSD worth $1 or not? Then we need to think about WHO is asking the question! A merchant who wants to exit bitUSD would say, YES because he can settled at the feed price (currently without penalty). A user however may say that he cannot enter bitUSD because it as a premium (at least in terms of BTS).
However, the latter argument is only true when people want to *enter* bitUSD, if they instead get paid in bitUSD already, then it's all a wash and we can say 1bitUSD=$1.

Thinking about it I actually prefer it that way, because it supports MERCHANTS and those that want to use it can accept a small premium because bitUSD is WAY WAY more flexible than real cash.
We need to support MERCHANTS and thus it (IMHO) makes sense to keep the 1:1 settlement
Title: Re: [BSIP#16]Optimization to Force Settlement Parameters of BitCNY
Post by: bitcrab on May 12, 2016, 09:01:46 am
The term "compensation" implies that there is some damage caused by the settlement, which is often the case in "real-world" finance, but not in BitShares.

If the shorter sells BitAssets with a 1% premium he achieves the same effect for himself as with a 1% settlement penalty, only with the premium existing holders are not hurt.
Very good point ... and part of the problem as this results in the premium we can see on the bitassets/BTS markets.

I think the question we need to answer to ourselves is this: Is a bitUSD worth $1 or not? Then we need to think about WHO is asking the question! A merchant who wants to exit bitUSD would say, YES because he can settled at the feed price (currently without penalty). A user however may say that he cannot enter bitUSD because it as a premium (at least in terms of BTS).
However, the latter argument is only true when people want to *enter* bitUSD, if they instead get paid in bitUSD already, then it's all a wash and we can say 1bitUSD=$1.

Thinking about it I actually prefer it that way, because it supports MERCHANTS and those that want to use it can accept a small premium because bitUSD is WAY WAY more flexible than real cash.
We need to support MERCHANTS and thus it (IMHO) makes sense to keep the 1:1 settlement

I don't know how many MERCHANTS are there accept bitUSD, seems a lot? anyway let's focus on bitCNY here.
AFAIK there are no MERCHANTS in China that accept bitCNY, I don't expect there will be some in the future. even btc38 accept bitCNY mainly because transwiser play the role of gateway.
you may ask then why we need the supply of bitCNY to grow up?
the answer is: DEX

let me tell one plan of transwiser: to introduce ANTSHARE to DEX for trading.
antshares is a famous blockchina project in China(see more in www.antshares.com (http://www.antshares.com)), which focus on company register and share management, last year it get 2100 BTC in crowdfunding, now it is still under development, maybe will be launched in a couple of months.

ANTSHARE is already in  trading as pre-sales in DEX, after the launch of antshares,   more than 10M CNY worth ANTSHARE will be issued and traded in DEX, so we have a problem, to make which one the main trading pair?
maybe transwiser also need to issue TRANS.BTC and TRANS.CNY for this, but why not use bitCNY here in DEX?

now seems in DEX OPEN.BTC is most wildly used in trading, I know there's some difficult to overcome to make smartcoin wildly used in DEX trading,  however, if we cannot make smartcoin wildly used in DEX, how do you think we can make MERCHANTS accept it?

that's why I focus this problem mostly: how to encourage(or at least do not discourage) shorter to supply bitCNY, with less than 1M supply bitCNY is just a toy.
don't care too much on the MERCHANTS problem in China, in China MERCHANTS do not care whether it is 100% or 99% force settlement, they only care if there is a reliable gateway service.




Title: Re: [BSIP#16]Optimization to Force Settlement Parameters of BitCNY
Post by: ElMato on May 12, 2016, 04:10:26 pm
Coming late to the discussion but i have been following this thread with much attention.

Here is what i think in a very disorganized way.

Its a delicate thing to change the offset parameter once its been in a certain value for a time (not so long time .. 5 months?)

Saying robbery its a bold argument since you can still trade bitCNY for BTS (or any other asset) at a fair price, the difference is that you now will get less (CNY value in) BTS if force settle.

@bitcrab is one of the major players pushing, developing and creating bitCNY (and the DEX, and BitShares as a whole) even after the problems they suffered in the 1.0 -> 2.0 upgrade they still commit to the project and is something that we can't disregard easily.

Also they are doing 1:1 conversion (or similar) from bitCNY <-> CNY (alipay). So the value of bitCNY won't be affected in a significant way even if we move the offset 1%.

Are "normal" users of bitCNY doing force-settlement periodically? is a common usage pattern or they are using the gateway basically?

I also recall BM saying that the cheapest way to buy BTS without moving the price is doing a FS on a smartcoin ... so adding a little percent to the offset i think is fair.

Bottom line: I do support the change in the settlement offset from 0% to 1%, but we have votes .. so lets see what the community thinks.
Title: Re: [BSIP#16]Optimization to Force Settlement Parameters of BitCNY
Post by: Chronos on May 12, 2016, 09:16:51 pm
The purpose of a forced settlement offset is not to protect any party specifically. The purpose is to maintain the stable and accurate price of the smartcoin, which in turn benefits all users of the system.

So, the best metric is this: how has the trading price of BitCNY compared to the price feed, over the last months? Can we overlay the price feed to the historic price and take a look ourselves? I don't see the price feed in a graph here: http://cryptofresh.com/a/CNY
Title: Re: [BSIP#16]Optimization to Force Settlement Parameters of BitCNY
Post by: Empirical1.2 on May 12, 2016, 09:49:07 pm
 +5% Very positive change that should improve BitCNY

1. What about lowering the request settlement to 99%?
It should lower the premium and it's fair that there should be a cost associated with accessing liquidity at the expense of shorters.

I think lowering forced settlement to 98/99% would be beneficial.
Title: Re: [BSIP#16]Optimization to Force Settlement Parameters of BitCNY
Post by: abit on May 15, 2016, 12:21:19 am
The purpose of a forced settlement offset is not to protect any party specifically. The purpose is to maintain the stable and accurate price of the smartcoin, which in turn benefits all users of the system.

So, the best metric is this: how has the trading price of BitCNY compared to the price feed, over the last months? Can we overlay the price feed to the historic price and take a look ourselves? I don't see the price feed in a graph here: http://cryptofresh.com/a/CNY
Are you looking for this http://cryptofresh.com/a/BTS_CNY?
Title: Re: [BSIP#16]Optimization to Force Settlement Parameters of BitCNY
Post by: pc on May 15, 2016, 10:02:35 am
I suggest we all take a step back and take 15 minutes to (re-)read https://bitshares.org/technology/price-stable-cryptocurrencies/ .

What I take from that

My conclusions

Furthermore, every piece of documentation I have seen tells BitAsset buyers that they can always exchange their BitAssets for an equivalent amount of BTS. The word "guarantee" is used in several places in that context. Do we really want to unilaterally declare that guarantee void, thereby destroying all credibility of the very concept of BitAssets?

From all of this follows that
Title: Re: [BSIP#16]Optimization to Force Settlement Parameters of BitCNY
Post by: Empirical1.2 on May 15, 2016, 05:19:28 pm
I suggest we all take a step back and take 15 minutes to (re-)read https://bitshares.org/technology/price-stable-cryptocurrencies/ .

What I take from that
  • The settlement price guarantees price stability to holders of BitAssets. This guarantee defines the peg.
  • Shorters are supposed to sell their shorted BitAssets at a premium to cover their risk.
  • BitAssets are expected to trade at a premium, due to the inherent risk of shorters. The premium is expected to be higher in a BTS bear market.
  • The premium is supposed to encourage merchants to accept BitAssets.
  • The premium does not play a significant role to traders or other users of BitAssets.

My conclusions
  • BitAssets are working right now exactly in the way they are supposed to be working.
  • The premium we are seeing in BitAsset markets is due to the BTS bear market. The settlement price is not the cause of this. (Coincidentally, BitCrab's attacks on forced settlement always seem to happen in a bear market.)
  • When (if) the BTS price is rising again, the premium will automatically reduce to something close to zero. (IIRC, two months ago there was actually a negative premium on the markets, where some people bought below the feed and used forced settlement to make a profit.)
  • The premium is good to have, for shorters, merchants and customers. It is the natural market mechanism to keep everyone's interests balanced.
  • We want bitCNY to be pegged to CNY. By changing the settlement offset to 1% we are moving the "guarantee", which in turn means that bitCNY will be pegged to .99 CNY. This is an undesired side effect.

Furthermore, every piece of documentation I have seen tells BitAsset buyers that they can always exchange their BitAssets for an equivalent amount of BTS. The word "guarantee" is used in several places in that context. Do we really want to unilaterally declare that guarantee void, thereby destroying all credibility of the very concept of BitAssets?

From all of this follows that
  • the proposed change is not necessary - it is the function of the premium to encourage shorters, not the function of the settlement offset
  • the proposed change does not have the desired effect - the reason for the premium and the risks of the shorters is the BTS bear market, not the settlement offset
  • the proposed change is extremely harmful to our ecosystem, due to its side effects - it redefines 1 bitCNY to be worth 0.99 CNY and destroys our credibility, because we will be breaking our own guarantees.

@pc you make some good points and are clearly passionate about your position.

Your first argument is that adjusting BitCNY parameters would represent the breaking of an implied guarantee. I & I think most would disagree, SmartCoins are still in the nascent stages of design/adoption/parameter tweaking and are subject to change if agreed by shareholders. (They have already undergone major and controversial changes such as the removal of yield and the introduction of forced settlement itself.)

Your second argument is that SmartCoins are functioning well and the change is not desirable anyway.

While I believe you are correct that part of the premium at any time is related to the BTS market, bear/bull the 100% forced settlement feature also plays a role.

1.  When you guarantee liquidity at the expense of shorts and at 1-1, this adds a significant burden to shorts that they will price in via an additional premium. By lowering the forced settlement number as well as the amount that can be forced settled you reduce this burden and therefore not only bring the price close to the peg (which is favourable) but also reduce the risk premium shorts need to charge & thus create a tighter peg overall.

1. What about lowering the request settlement to 99%?
It should lower the premium and it's fair that there should be a cost associated with accessing liquidity at the expense of shorters.
Hopefully market makers will step in the majority of the time but the ability to get $0.99 on the dollar in 24 hours regardless of market conditions would still be very appealing to longs and merchants imo.


2. The closer forced settlement is to 1-1 the higher the potential for forced settlement manipulation which has already been an issue in the past and creates an additional risk premium too.

Does the BitShares price feed reference last traded prices at weighted exchanges or does it average last traded prices over a certain amount of time?

What I mean is there is that there is usually a 1-2% spread between buy and sell prices for BTS on BTC38 and Polo.

If you are force settling and you know it will reference last traded prices at a specific time then it would be in your interest to make tiny sell trades so that the last traded price is in your favour by 1-2%. It your force settle is going through at a thinly traded time it may also be in your interest & very cheap to move the price by another 1-2%.
That would assume that you know when witnesses publish a price and how many witnesses publish it ..
This could be fixed by tracking trades over time or running a weighted average over time but that involves alot of work to upgrade the feed script into a time-sensitive feed script.
Fortunately, this time-sensivity is also achieved by different witnesses running their scripts at different time instants.

We may consider replacing the last price with the "(highest bid + lowest ask)/ 2" that would make it a little more "fair"

Cool, yeah I'm just wondering if the forced settlement at 100% can currently be exploited.

I think your other suggestion of introducing a 1-2% forced settlement fee could also help address that if it was a potential issue...

- Settlement can be discouraged by asking for a percentage fee (1-2%) (this will move the peg AROUND parity. Flat fee for settlement can also be increase which has some negative effects on pred. markets

3. Merchants - I believe merchants are most interested in the size of the potential market of users as well as the speed, simplicity and cost of fiat conversion via something similar to BitPay.

Without a tightly pegged, liquid SmartCoin there will be no users and no attraction for merchants to serve that market. (So it's not chicken and egg, users come first)  Second is middlemen who can convert from SmartCoins to fiat. Transwiser provides this service for BitCNY and they believe they can do it better with a lower forced settlement so it should actually improve merchant adoption from that perspective too.

3. SmartCoins are not working well imo. They have gained low/no traction with the current parameters since being introduced 7 months ago. They have a tiny 2-4% of existing market share (never mind potential market share.)  and no/low merchant adoption and clearly need some improvements.

In addition to lowering forced settlement, I am personally also in favour of other measures that reduce the burden on shorts so that their risk premium is reduced thus creating a tighter peg overall.

I am also in favour of subsidizing liquidity & I am also in favour of subsidizing SmartCoin yield. https://bitsharestalk.org/index.php/topic,21597.msg286581.html#msg286581

While clearly not optimally designed yet, SmartCoins have the potential to be absolutely huge and when you get the formula right & I believe they will really take off.
Title: Re: [BSIP#16]Optimization to Force Settlement Parameters of BitCNY
Post by: pc on May 15, 2016, 06:08:41 pm
1.  When you guarantee liquidity at the expense of shorts and at 1-1, this adds a significant burden to shorts that they will price in via an additional premium. By lowering the forced settlement number as well as the amount that can be forced settled you reduce this burden and therefore not only bring the price close to the peg (which is favourable) but also reduce the risk premium shorts need to charge & thus create a tighter peg overall.

Your wording implies that forced settlement is in some form damaging to shorters. That is wrong. Settlement happens at the fair price and does not change the NAV of either involved party. The shorter can take the remaining collateral and the proceeds from selling the originally shorted smartcoins and immediate extend his short position, should he want to stay short. The only "damage" here is the required fee, which is negligible (less than 1 US-cent I believe).


Without a tightly pegged, liquid SmartCoin there will be no users and no attraction for merchants to serve that market.

Maybe. My point is, the settlement offset does not make the peg more tight. The premium will remain, because the reason for the premium is BTS price volatility.


3. SmartCoins are not working well imo. They have gained low/no traction with the current parameters since being introduced 7 months ago. They have a tiny 2-4% of existing market share (never mind potential market share.)  and no/low merchant adoption and clearly need some improvements.

With "working well" I meant the mechanics that are supposed to keep the price stable. You are right that the market acceptance is lousy, and finding ways to improve that would be desirable. The proposal would have the opposite effect though, because unilaterally declaring a guarantee void will destroy all trust in the system, and nobody is going to touch our "smart" coins every again.


In addition to lowering forced settlement, I am personally also in favour of other measures that reduce the burden on shorts so that their risk premium is reduced thus creating a tighter peg overall.

The "burden" on shorts is an inherent factor in *every* short position. The reward for that position is the leverage effect, which works for the shorter in a BTS bull market. Actually, in BitShares the shorts are relatively well protected. In real-world shorts you can literally lose everything, not just your collateral.

Increasing the settlement offset to 1% would reduce the risk premium by 1%, right? How far above the feed are smartcoins trading currently? Would 1% really make a significant difference wrt the tightness of the peg?
Title: Re: [BSIP#16]Optimization to Force Settlement Parameters of BitCNY
Post by: Empirical1.2 on May 15, 2016, 07:22:20 pm
1.  When you guarantee liquidity at the expense of shorts and at 1-1, this adds a significant burden to shorts that they will price in via an additional premium. By lowering the forced settlement number as well as the amount that can be forced settled you reduce this burden and therefore not only bring the price close to the peg (which is favourable) but also reduce the risk premium shorts need to charge & thus create a tighter peg overall.

Your wording implies that forced settlement is in some form damaging to shorters. That is wrong. Settlement happens at the fair price and does not change the NAV of either involved party. The shorter can take the remaining collateral and the proceeds from selling the originally shorted smartcoins and immediate extend his short position, should he want to stay short. The only "damage" here is the required fee, which is negligible (less than 1 US-cent I believe).

In addition to lowering forced settlement, I am personally also in favour of other measures that reduce the burden on shorts so that their risk premium is reduced thus creating a tighter peg overall.

The "burden" on shorts is an inherent factor in *every* short position. The reward for that position is the leverage effect, which works for the shorter in a BTS bull market. Actually, in BitShares the shorts are relatively well protected. In real-world shorts you can literally lose everything, not just your collateral.

Increasing the settlement offset to 1% would reduce the risk premium by 1%, right? How far above the feed are smartcoins trading currently? Would 1% really make a significant difference wrt the tightness of the peg?

Forced settlement forces the short to make a trade when they don't want to. (I would hate if my shorts on Poloniex could be closed out at any time by longs forced settling.) 

Lets say the trade was just between you and me.

I short BitBTC because I believe this is a good time to short. Yay, turns out I was right and the general trend is clearly moving in my favour due to news/other!  If you wanted to me to cover you'd have to sell your BitBTC at a discount but instead you force settle me at 1-1. 

- It's statistically more likely you'll force settle when the market is moving in my favour and against you.
- It puts me in a position where I have to actively manage my short position more because I may be force settled unexpectedly.
- I'll  lose any potential gains for the period I'm out of the market
- I have to repay fees again
- I possibly have to pay a higher premium to re-enter my position if the market is clearly trending against the longs. (Which might be why they chose to force settle.) 

By reducing forced settlement or even charging a reasonable forced settlement fee, traders who are long can't used forced settlement as a trading tool/advantage and will have to find a willing buyer as opposed to a forced buyer. This levelling of the playing field removes/reduces the negative factors the short would have to price in thereby hopefully lowering the spread by far more than the 1% offset.

Being guaranteed 98/99% is still very attractive for genuine longs. especially if the premium/spread to enter the position has lowered as a result. 

Also how I 'think' potential manipulation could work...

I go long X BitBTC/SmartCoin. I request force settlement for 3 am the following morning.  The spread on BTS on exchanges is usually 0.4-1.5%. In 24 hours at the thinly traded force settlement time I make tiny BTS sells just so that the traded price referenced is in my favour by 0.5-1.5% at the shorts expense. (Annualized that's a significant edge that would have to be priced in) Depending on depth and size of settlement it may be in my interest to push the BTS price at settlement time in my favour more. (Reducing forced settlement or charging a hefty fee reduces the potential for for forced settlement trading/manipulation too.)
Title: Re: [BSIP#16]Optimization to Force Settlement Parameters of BitCNY
Post by: bitsharesbrazil on May 25, 2016, 04:51:07 am
God this not easy.....
Im not tech or economics.....but we have to make smartcoins happen.....trully dex
Title: Re: [BSIP#16]Optimization to Force Settlement Parameters of BitCNY
Post by: cylonmaker2053 on May 25, 2016, 02:27:34 pm
As one of the community's most active traders and from a financial market theory perspective, i support @pc perspective on this. I also share @Empirical1.2 frustrations in having many short positions force settled on me; however, I've been on the settlement initiation side of trades many times and it is a very valuable feature.

@pc is correct that this is mainly an issue in bear markets. From personal experience, it has SUCKED pouring more and more collateral into my shorts just to keep them open, but that's a reasonable price to pay for the leverage on the upside when market conditions reverse, and for the fact that I've sold all of my short assets at premiums to begin with.

#1 priority is to guarantee parity with the underlying asset. Forced settlement seems the best way to do this. We cannot compromise on this priority or we lose one of our biggest value propositions in the smartcoin market.

#2 there are times when asset holders, mainly Chinese CNY from my experience, value immediate liquidation and so put their assets up for sale at discounts to the feed price; this is where market makers like myself and @JonnyBitcoin come in, buy at a discount, and settle for some risky profit (risky in that the feed can fluctuation in the next 24 hours).

#3 we all know the rules of the game going into any trade, these are built into our expectations ex ante, and so the prices we choose to initiate ought to compensate for the risk of forced settlement.

What could be improved is the old idea we've punted in this community about compensating short positions with some sort of yield on collateral. This would further incentivize short initiations, especially in bear markets. Once we're back to a bull market, it'll be irrelevant.
Title: Re: [BSIP#16]Optimization to Force Settlement Parameters of BitCNY
Post by: bitsharesbrazil on May 25, 2016, 03:39:20 pm
I like your view ylon. Thanks for sharing
Title: Re: [BSIP#16]Optimization to Force Settlement Parameters of BitCNY
Post by: xeroc on May 25, 2016, 06:08:37 pm
that's a reasonable price to pay for the leverage on the upside when market conditions reverse, and for the fact that I've sold all of my short assets at premiums to begin with.
Very good point!!
Title: Re: [BSIP#16]Optimization to Force Settlement Parameters of BitCNY
Post by: xeroc on May 27, 2016, 08:48:39 am
I was hoping for a more neutral approach for getting shareholders approval so I published a quick article on Steem:
https://steemit.com/bitshares/@xeroc/vote-please-cast-your-vote-for-the-bitcny-settlement-offset-worker
Title: Re: [BSIP#16]Optimization to Force Settlement Parameters of BitCNY
Post by: cylonmaker2053 on May 27, 2016, 03:40:51 pm
I was hoping for a more neutral approach for getting shareholders approval so I published a quick article on Steem:
https://steemit.com/bitshares/@xeroc/vote-please-cast-your-vote-for-the-bitcny-settlement-offset-worker

how do you plan on voting, @xeroc?

you guys have my take below. on this specific proposal, i'm for keeping the status quo, even as a major short position holder, which would be on the benefit side of the change.

Short initiators already have two big incentives: leverage on the upside and selling their shorts at premium to begin with. Yes, it sucks in a bear market to be caught on the short side, i just don't see that as reason for changing the rules of the game.

if there really is compelling need to further incentivize short positions, then some yield from fees might be a better approach.

final point, i do not like the idea of changing contract rules for one smartcoin and not all of them. that can set a bad precedent for plenty of discontinuities, complications keeping users away, and potentially non-value-added arb opportunities akin to regulatory capture just bc the rules of different, not all users may understand differences, and so will get swindled.
Title: Re: [BSIP#16]Optimization to Force Settlement Parameters of BitCNY
Post by: xeroc on May 28, 2016, 08:01:14 am
I am voting for the "no changes" worker:
http://cryptofresh.com/workers
Title: Re: [BSIP#16]Optimization to Force Settlement Parameters of BitCNY
Post by: cylonmaker2053 on May 28, 2016, 03:42:55 pm
I am voting for the "no changes" worker:
http://cryptofresh.com/workers

thanks for the link. this is a cool way to see what's being worked on, but can you explain the different columns, especially the "supporting stake" column? And why is the bitCNY forced settlement split into two workers, one YES and one NO instead of one worker that gets voting on for yes or no? sorry for the newb governance questions...

(https://cdn.pbrd.co/images/1glkPkdM.png)

Title: Re: [BSIP#16]Optimization to Force Settlement Parameters of BitCNY
Post by: Chris4210 on May 28, 2016, 08:49:28 pm
As one of the community's most active traders and from a financial market theory perspective, i support @pc perspective on this. I also share @Empirical1.2 frustrations in having many short positions force settled on me; however, I've been on the settlement initiation side of trades many times and it is a very valuable feature.

@pc is correct that this is mainly an issue in bear markets. From personal experience, it has SUCKED pouring more and more collateral into my shorts just to keep them open, but that's a reasonable price to pay for the leverage on the upside when market conditions reverse, and for the fact that I've sold all of my short assets at premiums to begin with.

#1 priority is to guarantee parity with the underlying asset. Forced settlement seems the best way to do this. We cannot compromise on this priority or we lose one of our biggest value propositions in the smartcoin market.

#2 there are times when asset holders, mainly Chinese CNY from my experience, value immediate liquidation and so put their assets up for sale at discounts to the feed price; this is where market makers like myself and @JonnyBitcoin come in, buy at a discount, and settle for some risky profit (risky in that the feed can fluctuation in the next 24 hours).

#3 we all know the rules of the game going into any trade, these are built into our expectations ex ante, and so the prices we choose to initiate ought to compensate for the risk of forced settlement.

What could be improved is the old idea we've punted in this community about compensating short positions with some sort of yield on collateral. This would further incentivize short initiations, especially in bear markets. Once we're back to a bull market, it'll be irrelevant.


I was reading all comments and opinions about this proposal.

After considering the case I decided to vote against the BSIP 16 and to not support a change of the CNY settlement parameters.



PC made his point clear enough that we should not break the trust in the guarantee of getting at least 1 CNY back. This could have another negative impact on the BitShares ecosystem and further harm our reputation in the crypto community. I think the current settlement parameters are fine and the spread will get tighter when volume picks up. Please don't forget we are at the beginning of this payment platform and we see many anomalies due to low volumes.

I also think that market makers like cylonmaker2053 and JonnyBitcoin know the trading risks and can evaluate the associated risks and potential profits. Shorting and providing liquidity is not for everyone and should be handled by experts who know how to create a derivate. I am looking forward to adding bankers to the BitShares platform so that they can add their market expertise and help us with the market creation. Creating a derivate is banking 101. Just replace BTS with USD and every banker knows what you are talking about.

cheers Chris4210
Title: Re: [BSIP#16]Optimization to Force Settlement Parameters of BitCNY
Post by: abit on May 28, 2016, 09:03:25 pm
I am voting for the "no changes" worker:
http://cryptofresh.com/workers

thanks for the link. this is a cool way to see what's being worked on, but can you explain the different columns, especially the "supporting stake" column? And why is the bitCNY forced settlement split into two workers, one YES and one NO instead of one worker that gets voting on for yes or no? sorry for the newb governance questions...
The "supporting stake" column means how much stake has chosen to vote for the specified worker. Just ignore the colors and other workers, focus on these 2 workers.

Because there is no longer a way to vote "no" on a worker. So here comes 2 workers,
* if you agree to change, vote for the "yes" worker;
* if you don't agree to change, vote for the "no" worker;
* if you want to stay neutral, or you're not aware of this voting, don't vote.

Quote
(https://cdn.pbrd.co/images/1glkPkdM.png)
Title: Re: [BSIP#16]Optimization to Force Settlement Parameters of BitCNY
Post by: abit on May 28, 2016, 10:00:57 pm
To @pc:

From the blockchain, we can see that @bitcrab is now only have a short position of 60K bitCNY (see https://cryptofresh.com/u/btareserve which is his main shorter account), which is about 10% of total bitCNY supply. His collateral level is the 3rd high among all shorters, most likely won't match any forced settlements any time soon. There are a few other shorters who have bigger position than him ( https://cryptofresh.com/u/aboy, https://cryptofresh.com/u/code). In the mean while bitcrab is holding more than 60K bitCNY (see https://cryptofresh.com/u/bitcrab, https://cryptofresh.com/u/btareserve and https://cryptofresh.com/u/transwiser-wallet). When I'm writing this post, there is around 37k bitCNY in btc38's account (https://cryptofresh.com/u/btc38-cny-kuos-72722), which will be sold to bitcrab when btc38 needs fiat. So it seems bitcrab is currently a bitCNY holder rather than a shorter.
Title: Re: [BSIP#16]Optimization to Force Settlement Parameters of BitCNY
Post by: pc on May 29, 2016, 07:51:51 am
Thanks for the analysis! I take back my statements implying that bitcrab is acting out of self-interest. (Edited my earlier post.)

My other points remain.
Title: Re: [BSIP#16]Optimization to Force Settlement Parameters of BitCNY
Post by: bitcrab on May 29, 2016, 11:40:28 am
As one of the community's most active traders and from a financial market theory perspective, i support @pc perspective on this. I also share @Empirical1.2 frustrations in having many short positions force settled on me; however, I've been on the settlement initiation side of trades many times and it is a very valuable feature.

@pc is correct that this is mainly an issue in bear markets. From personal experience, it has SUCKED pouring more and more collateral into my shorts just to keep them open, but that's a reasonable price to pay for the leverage on the upside when market conditions reverse, and for the fact that I've sold all of my short assets at premiums to begin with.

#1 priority is to guarantee parity with the underlying asset. Forced settlement seems the best way to do this. We cannot compromise on this priority or we lose one of our biggest value propositions in the smartcoin market.

#2 there are times when asset holders, mainly Chinese CNY from my experience, value immediate liquidation and so put their assets up for sale at discounts to the feed price; this is where market makers like myself and @JonnyBitcoin come in, buy at a discount, and settle for some risky profit (risky in that the feed can fluctuation in the next 24 hours).

#3 we all know the rules of the game going into any trade, these are built into our expectations ex ante, and so the prices we choose to initiate ought to compensate for the risk of forced settlement.

What could be improved is the old idea we've punted in this community about compensating short positions with some sort of yield on collateral. This would further incentivize short initiations, especially in bear markets. Once we're back to a bull market, it'll be irrelevant.


I was reading all comments and opinions about this proposal.

After considering the case I decided to vote against the BSIP 16 and to not support a change of the CNY settlement parameters.

PC made his point clear enough that we should not break the trust in the guarantee of getting at least 1 CNY back. This could have another negative impact on the BitShares ecosystem and further harm our reputation in the crypto community. I think the current settlement parameters are fine and the spread will get tighter when volume picks up. Please don't forget we are at the beginning of this payment platform and we see many anomalies due to low volumes.

I also think that market makers like cylonmaker2053 and JonnyBitcoin know the trading risks and can evaluate the associated risks and potential profits. Shorting and providing liquidity is not for everyone and should be handled by experts who know how to create a derivate. I am looking forward to adding bankers to the BitShares platform so that they can add their market expertise and help us with the market creation. Creating a derivate is banking 101. Just replace BTS with USD and every banker knows what you are talking about.

cheers Chris4210

I haven't expected that it's so difficult to push such a change.

China community is the main bitCNY ecosystem,  here users can deposit/withdraw bitCNY with fiat,  most of the users support this change,  but seems their opinion is less important than some outside speculators'?

Bitshares is a decentralized platform, anyone can short if he like, this is simple trading, no need to suppose only seldom experts can/should do that.

if shorters are forced to sell his collateral without any compensation when the collateral price is above the margin call price, there is no reputation on this system. you guys always mention "guarantee" to holders, but who has thought in the place of shorters?

we cannot expect a financial system where everyone like to hold but seldom like to short can grow big enough. 

if we do not change, shorters will always worry of being force settled and we can not expect enough supply of bitCNY.

20+ days left for the committee proposal, let's see what will happen.
Title: Re: [BSIP#16]Optimization to Force Settlement Parameters of BitCNY
Post by: cylonmaker2053 on May 29, 2016, 04:18:19 pm
The "supporting stake" column means how much stake has chosen to vote for the specified worker. Just ignore the colors and other workers, focus on these 2 workers.

Because there is no longer a way to vote "no" on a worker. So here comes 2 workers,
* if you agree to change, vote for the "yes" worker;
* if you don't agree to change, vote for the "no" worker;
* if you want to stay neutral, or you're not aware of this voting, don't vote.

So the way to read the contest here is that a 266M BTS stake is voting YES and 144M voting NO?
Title: Re: [BSIP#16]Optimization to Force Settlement Parameters of BitCNY
Post by: cylonmaker2053 on May 29, 2016, 04:26:33 pm
I'm not seeing the big issue with forced settlement. You aren't losing anything you haven' already lost when forced to settle, the system is just closing out your position earlier than you might have liked, but you're welcome to re-open another position with the returned collateral. This simple mechanism keeps the entire system stable by cutting imbalanced positions early.

The idea of modifying system rules market by market instead of having a coherent system doesn't at all seem worth it in this case.
Title: Re: [BSIP#16]Optimization to Force Settlement Parameters of BitCNY
Post by: Empirical1.2 on May 29, 2016, 04:55:38 pm
I'm not seeing the big issue with forced settlement. You aren't losing anything you haven' already lost when forced to settle, the system is just closing out your position earlier than you might have liked, but you're welcome to re-open another position with the returned collateral. This simple mechanism keeps the entire system stable by cutting imbalanced positions early.

The idea of modifying system rules market by market instead of having a coherent system doesn't at all seem worth it in this case.

You're forcing shorts to play the role of market makers for longs providing liquidity on demand at 1-1.

A trader doesn't have to charge you a significant premium but a market maker who is guaranteeing you liquidity at 1-1 whenever you want to sell will charge you a very big premium for that privilege. 

So a large part of the huge premium you're seeing on the SmartCoin markets, are market maker premiums charged by shorts for the risk and opportunity cost of providing liquidity to longs.

Imagine if your BitUSD long position (or any other crypto position you hold) could be converted at any time to BTS/BTC, would you be willing to pay close to 1-1 for it?

Rather let the free market bring buyers and sellers together and don't force either side to play market maker in a tight range around the peg. The purpose of forced settlement is really to give market participants confidence SmartCoins are backed by actual collateral that can be redeemed for most of their value in an emergency. So even forced settlement at 0.95 would be very attractive.

Then you could let market participants buy/sell/short without being forced to play market maker and if you did want some market making or guaranteed liquidity near 1-1 because the value to BTS of SmartCoin adoption is high, then you could subsidize it and pay market makers so that the cost isn't built into a wider peg that makes smartcoins unappealing.   

Title: Re: [BSIP#16]Optimization to Force Settlement Parameters of BitCNY
Post by: cylonmaker2053 on May 29, 2016, 06:14:12 pm
The purpose of forced settlement is really to give market participants confidence SmartCoins are backed by actual collateral that can be redeemed for most of their value in an emergency. So even forced settlement at 0.95 would be very attractive.

Absolutely, the purpose of settlement is to give market participants confidence that our assets are backed by collateral. The 100% settlement at the peg means our assets are 100% collateralized, which i certainly prefer. Any discount from that, like 0.95, just means we'd be advertising smartcoins with 95% collateral banking. I don't see that as beneficial.

All market participants go into trades knowing the rules of the game ahead of time. i like 100% collateralilzation over any arbitrary fractional value, but if the system changed (which it changes too often IMO), then i'd just rationally discount the value on the long side and price my bids at 95% or whatever the collateral value. All we'd do is skew the bid-ask window in the other direction.

Like i've repeated many times on this forum, i do see value in providing incentive to traders willing to short assets into existence. Right now that compensation is implicit with high spreads  and shorts get leveraged upside. That's been enough for me to trade frequently on both sides of the market. However, if there is a big desire in this community to further compensate short traders, then i suggest directly subsidizing shorts with some yield on BTS locked in collateral; penalizing buy-side traders who want to settle assets at peg is not the best way to provide this subsidy.
Title: Re: [BSIP#16]Optimization to Force Settlement Parameters of BitCNY
Post by: Empirical1.2 on May 29, 2016, 07:47:30 pm
The purpose of forced settlement is really to give market participants confidence SmartCoins are backed by actual collateral that can be redeemed for most of their value in an emergency. So even forced settlement at 0.95 would be very attractive.

Absolutely, the purpose of settlement is to give market participants confidence that our assets are backed by collateral. The 100% settlement at the peg means our assets are 100% collateralized, which i certainly prefer. Any discount from that, like 0.95, just means we'd be advertising smartcoins with 95% collateral banking. I don't see that as beneficial.

All market participants go into trades knowing the rules of the game ahead of time. i like 100% collateralilzation over any arbitrary fractional value, but if the system changed (which it changes too often IMO), then i'd just rationally discount the value on the long side and price my bids at 95% or whatever the collateral value. All we'd do is skew the bid-ask window in the other direction.

Like i've repeated many times on this forum, i do see value in providing incentive to traders willing to short assets into existence. Right now that compensation is implicit with high spreads  and shorts get leveraged upside. That's been enough for me to trade frequently on both sides of the market. However, if there is a big desire in this community to further compensate short traders, then i suggest directly subsidizing shorts with some yield on BTS locked in collateral; penalizing buy-side traders who want to settle assets at peg is not the best way to provide this subsidy.

The reason longs don't need guaranteed liquidity from shorts at $1.00 to buy at $1.01 is the same reason shorts don't need guaranteed liquidity from longs at $1.10 to short at $1.09.

Everyday people trade billions in CMX gold contracts even though they are hardly backed at all. If the shorts were forced to guarantee liquidity at the gold price for the longs, the increased cost burden would make the spread too wide and effectively kill the market which is what we've done to SmartCoins.

Having forced settlement at say 0.96 or say 1-1 with a 4% fee would be similar to a service like bullionvault.

Traders agree they are buying and selling vaulted gold using the current gold price as a reference but to actually redeem it from the underlying its collateralised with in the event you couldn't find a buyer/other would cost you about 4-5% but it still trades at 1-1 with the actual gold price.

The point is if you know there is actual collateral backing the token even if you'd have to pay 4-5% to get it, you can comfortably buy and sell it round 1-1 and make a market. If you place a burden on either buyer/seller/shorter that they have to pay 1-1 whenever someone else wants to settle, the premium they will price for being a forced market maker/liquidity provider at 1-1 will effectively kill the market.
Title: Re: [BSIP#16]Optimization to Force Settlement Parameters of BitCNY
Post by: bitsharesbrazil on May 29, 2016, 08:03:21 pm
As I am new here I glad that you guys discuss this in a open way so I can learn n improve my knowledge n judgment. Thanks again
Title: Re: [BSIP#16]Optimization to Force Settlement Parameters of BitCNY
Post by: cylonmaker2053 on May 29, 2016, 08:13:34 pm
@Empirical1.2 i agree that 100% collateral isn't required, i just prefer it and think it's an excellent long term selling point. It's premature to say we've killed the smartcoin markets; i'd argue they haven't yet been discovered. We have an excellent trading technology as is and whatever rules we set will be internalized by traders ex ante. Whether we set 100% collateral assets or 95% or whatever, those values just get capitalized into expectations.

We have a good enough system the way it is, we just need to get more traders into it. Traders are just one group, btw. It'd be great to get actual end users for our smartcoins, like people using PoS apps for bitUSD, bitCNY, bitEUR, etc. like @kenCode is doing. 100% collateralized assets are a big selling point for that application.

I'm all for subsidizing collateral used to maintain shorts, but just prefer doing it directly via some yield. I'm also fine without subsidies bc of the indirect benefits of shorting mentioned earlier.

Also keep in mind, this proposal is to change one of the markets on the DEX, bitCNY. Doing so would unnecessarily complicate the DEX with markets trading with arbitrarily different rules.
Title: Re: [BSIP#16]Optimization to Force Settlement Parameters of BitCNY
Post by: Empirical1.2 on May 29, 2016, 08:31:49 pm
We have a good enough system the way it is, we just need to get more traders into it. Traders are just one group, btw. It'd be great to get actual end users for our smartcoins, like people using PoS apps for bitUSD, bitCNY, bitEUR, etc. like @kenCode is doing. 100% collateralized assets are a big selling point for that application.

I'm all for subsidizing collateral used to maintain shorts, but just prefer doing it directly via some yield. I'm also fine without subsidies bc of the indirect benefits of shorting mentioned earlier.

Also keep in mind, this proposal is to change one of the markets on the DEX, bitCNY. Doing so would unnecessarily complicate the DEX with markets trading with arbitrarily different rules.

Competitors like NBT have no guaranteed redeem-ability at all (& I think they're struggling in this BTC surge) I think merivercap's BitCash has zero as well. I disagree with both those strategies but I do believe 95% is more than sufficient for the market to trade at 1-1 and will significantly reduce the spread without the need for subsidies & given that we haven't even subsidized SmartCoin liquidity yet I'm not sure how realistic it is.

Given the stagnation of the 2.0 SmartCoin markets which have been operating with the current rules for over 7 months, I think trying a rule change at this stage is warranted.

I think it's good that it's just one of the markets as we'll be able to see if it's a positive change before applying it elsewhere. Personally I'd prefer to make the change bigger so the differences would be more easily observable & actionable.


Title: Re: [BSIP#16]Optimization to Force Settlement Parameters of BitCNY
Post by: cylonmaker2053 on May 29, 2016, 08:35:15 pm
Also, as a practical matter from trading...

I've invoked forced settlement many times in both bitCNY and bitUSD ...actually, today has been a weird bitUSD trading experience in that people have been dumping at about 6%-7% below peg all afternoon and i've been the highest bidder soaking up the supply. Since I can't find anyone else to buy anywhere near peg, i've been invoking forced settlement. If there were some penalty for doing this, or if assets were only 95% collateralized, i'd be doing the same stuff but at 15%+ below feed. What's the point of that?

The way i've seen forced settlement is that it acts as a natural supply reduction mechanism when we have imbalances and not enough buyers. It brings supply down to what the market is actually willing to support.
Title: Re: [BSIP#16]Optimization to Force Settlement Parameters of BitCNY
Post by: cylonmaker2053 on May 29, 2016, 08:39:40 pm
We have a good enough system the way it is, we just need to get more traders into it. Traders are just one group, btw. It'd be great to get actual end users for our smartcoins, like people using PoS apps for bitUSD, bitCNY, bitEUR, etc. like @kenCode is doing. 100% collateralized assets are a big selling point for that application.

I'm all for subsidizing collateral used to maintain shorts, but just prefer doing it directly via some yield. I'm also fine without subsidies bc of the indirect benefits of shorting mentioned earlier.

Also keep in mind, this proposal is to change one of the markets on the DEX, bitCNY. Doing so would unnecessarily complicate the DEX with markets trading with arbitrarily different rules.

Competitors like NBT have no guaranteed redeem-ability at all (& I think they're struggling in this BTC surge) I think merivercap's BitCash has zero as well. I disagree with both those strategies but I do believe 95% is more than sufficient for the market to trade at 1-1 and will significantly reduce the spread without the need for subsidies & given that we haven't even subsidized SmartCoin liquidity yet I'm not sure how realistic it is.

Given the stagnation of the 2.0 SmartCoin markets which have been operating with the current rules for over 7 months, I think trying a rule change at this stage is warranted.

I think it's good that it's just one of the markets as we'll be able to see if it's a positive change before applying it elsewhere. Personally I'd prefer to make the change bigger so the differences would be more easily observable & actionable.

i like that idea of testing changes in one market before applying to all, so i agree with that point. However, unless we have more control other than pure mob rule for forcing changes in a particular market, then we'll likely end up with fragmented rule sets across markets.

yes, we've been operating with these rules for 7 months, but i still argue that things haven't taken off bc of the rules, but bc we haven't done a good job attracting traders to the DEX.
Title: Re: [BSIP#16]Optimization to Force Settlement Parameters of BitCNY
Post by: Empirical1.2 on May 29, 2016, 09:30:29 pm
Also, as a practical matter from trading...

I've invoked forced settlement many times in both bitCNY and bitUSD ...actually, today has been a weird bitUSD trading experience in that people have been dumping at about 6%-7% below peg all afternoon and i've been the highest bidder soaking up the supply. Since I can't find anyone else to buy anywhere near peg, i've been invoking forced settlement. If there were some penalty for doing this, or if assets were only 95% collateralized, i'd be doing the same stuff but at 15%+ below feed. What's the point of that?

The way i've seen forced settlement is that it acts as a natural supply reduction mechanism when we have imbalances and not enough buyers. It brings supply down to what the market is actually willing to support.

I'd speculate the reasons BitUSD holders are willing to offer a 6/7% discount today rather than waiting for 1-1 in 24 hours is because they think BTC is surging so fast that they can make up the difference. If forced settlement was 1-1 with a 4% fee they still would be unlikely to sell much lower than a 7% loss because they're unlikely to make up a 7% loss. So in a way 1-1 might be of limited value other than to traders like yourself who are able to make good daily gains in these scenarios.

While from the shorts perspective a lot of them are going to wake up tomorrow having been unexpectedly force settled and then have to compete with each other if they want to rapidly re-short but then find there are not a lot of willing BitUSD longs because they all want to be in BTC atm so this will disrupt the market, discourage shorts and ultimately raise premiums. A 4% fee would have forced longs to wait 24 hours because it wouldn't make sense to sell at a 9-10% loss today and shorts would have received some compensation for having provided liquidity.

If liquidity close to 1-1 was something we wanted to provide to SmartCoin customers then liquidity subsidies would be the way to do it rather than disrupting the market, negatively impacting shorts & raising the premium the way the current system does imo.   
Title: Re: [BSIP#16]Optimization to Force Settlement Parameters of BitCNY
Post by: cylonmaker2053 on May 29, 2016, 10:52:48 pm
Also, as a practical matter from trading...

I've invoked forced settlement many times in both bitCNY and bitUSD ...actually, today has been a weird bitUSD trading experience in that people have been dumping at about 6%-7% below peg all afternoon and i've been the highest bidder soaking up the supply. Since I can't find anyone else to buy anywhere near peg, i've been invoking forced settlement. If there were some penalty for doing this, or if assets were only 95% collateralized, i'd be doing the same stuff but at 15%+ below feed. What's the point of that?

The way i've seen forced settlement is that it acts as a natural supply reduction mechanism when we have imbalances and not enough buyers. It brings supply down to what the market is actually willing to support.

I'd speculate the reasons BitUSD holders are willing to offer a 6/7% discount today rather than waiting for 1-1 in 24 hours is because they think BTC is surging so fast that they can make up the difference. If forced settlement was 1-1 with a 4% fee they still would be unlikely to sell much lower than a 7% loss because they're unlikely to make up a 7% loss. So in a way 1-1 might be of limited value other than to traders like yourself who are able to make good daily gains in these scenarios.

While from the shorts perspective a lot of them are going to wake up tomorrow having been unexpectedly force settled and then have to compete with each other if they want to rapidly re-short but then find there are not a lot of willing BitUSD longs because they all want to be in BTC atm so this will disrupt the market, discourage shorts and ultimately raise premiums. A 4% fee would have forced longs to wait 24 hours because it wouldn't make sense to sell at a 9-10% loss today and shorts would have received some compensation for having provided liquidity.

If liquidity close to 1-1 was something we wanted to provide to SmartCoin customers then liquidity subsidies would be the way to do it rather than disrupting the market, negatively impacting shorts & raising the premium the way the current system does imo.

i guess my main point is that i don't see changing collateralization or charging fees to force settlement as materially changing premature forced settlement; traders will rationally price all of this in from the start. They'll offer less for smartcoins to account for either a discount to par collateralization or to capitalize the fee ex ante. Then they'll still force settle and induce unplanned early settlement on short positions.

this process of capitalizing fees ex ante comes at the cost of losing 100% asset backing for our smartcoins, which then hurts the PoS marketing message of being able to get at least the full pegged value at any time. It's a much easier story to sell that 1 bitUSD can be settled at any point for 1 USD worth of BTS. It's unnecessarily damaging to sell merchants on getting 0.95 on the dollar or whatever.

Shorting is for professionals, or those at least willing to keep an eye on positions, up margin when needed if they want to keep positions from being called. If being force settled is something a prospective shorter is concerned with, they should set collateral sufficiently high to make it a non-issue. This will still be an issue with charging fees to force settle, or eroding collateral requirements.
Title: Re: [BSIP#16]Optimization to Force Settlement Parameters of BitCNY
Post by: Empirical1.2 on May 29, 2016, 11:11:54 pm
this process of capitalizing fees ex ante comes at the cost of losing 100% asset backing for our smartcoins, which then hurts the PoS marketing message of being able to get at least the full pegged value at any time. It's a much easier story to sell that 1 bitUSD can be settled at any point for 1 USD worth of BTS. It's unnecessarily damaging to sell merchants on getting 0.95 on the dollar or whatever.

This is one of the main arguments against, but liquidity subsidies would solve this imo and probably make BitUSD even more appealing than 1-1 forced settlement.

Anyway we obviously disagree on a few issues so it will be interesting if the BitCNY change does go through whether we're able to gleam anything one way or the other.
Title: Re: [BSIP#16]Optimization to Force Settlement Parameters of BitCNY
Post by: cylonmaker2053 on May 30, 2016, 02:30:29 pm
this process of capitalizing fees ex ante comes at the cost of losing 100% asset backing for our smartcoins, which then hurts the PoS marketing message of being able to get at least the full pegged value at any time. It's a much easier story to sell that 1 bitUSD can be settled at any point for 1 USD worth of BTS. It's unnecessarily damaging to sell merchants on getting 0.95 on the dollar or whatever.

This is one of the main arguments against, but liquidity subsidies would solve this imo and probably make BitUSD even more appealing than 1-1 forced settlement.

Anyway we obviously disagree on a few issues so it will be interesting if the BitCNY change does go through whether we're able to gleam anything one way or the other.

yeah i guess we'll see how it goes. i don't see any reason to think anything would change wrt forced settlement. spreads will just be skewed left and buyers will further discount CNY and continue force settlement for whatever fee or discount to par that gets build into the system. that's at least how i'll adjust my own behavior in that market. the end result isn't limiting forced settlement, just breaking the 100% collateralization, which kills PoS for the Chinese market. my guess is that the few active traders in that market don't care about PoS, nor do they understand the financial economics for how the change will be capitalized ex ante and market behavior unchanged.