Author Topic: Non-fungible BitUSD (i.e. just a CFD)  (Read 3059 times)

0 Members and 1 Guest are viewing this topic.

Offline monsterer

Not sure what you mean by 'true' CFD design. 

* There is no 'difference' payment - longs are essentially a spot trade and shorts end up in a different currency
The 'difference' payment is using BTS.  The BTS holder that is short USD gives another BTS holder the 'difference' between BTS vs USD prices at a certain time period.  The BTS holder that is long BitUSD is effectively hedged to the dollar.  If you package the 1) BTS collateral along with the 2) CFD agreement for the long USD party you create an asset BitUSD that is equivalent to USD.

True CFD is just that, Contract For Difference. That's pretty much it - your profit is based on the difference between the price when you opened the contract and the price when you close it. You start out in the same currency you end up in, hence the difference part.

Quote
What are 'True' CFD's?  If the agreement beforehand is that settlement for the difference will be at the market at any time why would that be un-'True'? The elegance of the design is precisely because you are making the CFD fungible and are able to create BitUSD as stated above.   

I don't think it is that 'elegant' because it forces the huge compromise of requiring the long to persist after the contract ends, which is why the short has to cover in the bitAssets and not BTS, to keep the hedge in place.

bitAssets are compromised CFDs IMO - the compromise is the persistence of the long position.
My opinions do not represent those of metaexchange unless explicitly stated.
https://metaexchange.info | Bitcoin<->Altcoin exchange | Instant | Safe | Low spreads

Offline merivercap

  • Hero Member
  • *****
  • Posts: 661
    • View Profile
    • BitCash
We can discuss other markets in the future, but I think this CFD design is good and elegant so I just want to get this to be the best it can be first before adding other types of assets.

Isn't it important to recognise that none of the bitAsset proposals has been a true CFD design, though?

* There is no 'difference' payment - longs are essentially a spot trade and shorts end up in a different currency
* True CFD's do not persist after the contract ends
* CFDs do not create other asset types
* CFSs are 100% equal on both sides (same margin requirements, interest payments)

I think starspirt is right when he says there is no one solution to fit all of our requirement perfectly. The only trouble is, I don't think you'll be able to implement all the different asset classes as plain privatised bitAssets because the shorting mechanism will be the same across all of them.

Not sure what you mean by 'true' CFD design. 

* There is no 'difference' payment - longs are essentially a spot trade and shorts end up in a different currency
The 'difference' payment is using BTS.  The BTS holder that is short USD gives another BTS holder the 'difference' between BTS vs USD prices at a certain time period.  The BTS holder that is long BitUSD is effectively hedged to the dollar.  If you package the 1) BTS collateral along with the 2) CFD agreement for the long USD party you create an asset BitUSD that is equivalent to USD.

* True CFD's do not persist after the contract ends
* CFDs do not create other asset types
What are 'True' CFD's?  If the agreement beforehand is that settlement for the difference will be at the market at any time why would that be un-'True'? The elegance of the design is precisely because you are making the CFD fungible and are able to create BitUSD as stated above.   

* CFSs are 100% equal on both sides (same margin requirements, interest payments)
I agree balance is important.  Margin requirements aren't needed for the long bitUSD because you're packaging the BTS collateral with a CFD the moves against it.  The BitAsset is a hedged position.  However all other parameters should be balanced.

Dividends are burned under Bitshares, so stakeholders are not actually paid BTS dividends.  Hence this BitUSD CFD is structured like a Total Return Swap.  In a Total Return Swap no one receives actual interest, coupon payments, or dividends so there is no need to account for that in the agreement.  Everything is embedded in the price.  If you tried to create a BitAsset for an asset where one party received actual interest/dividends you would have to account for the pre/post dividend/interest date on the price feeds.  (It probably will be adjusted for in market prices anyways).  For bonds you can design a CFD as a zero-coupon bond.  Hence any instrument that has coupons/dividends can be accounted for in prices.  If you wanted to peg to actual bonds or stocks you can just think of these BitAssets as pegged to non-dividend paying versions of the stocks or zero-coupon versions of the bonds. 

Anyways there doesn't need to be a one-size fits all solution, but this CFD solution or variations of this design are very good and can solve many problems in many asset classes.  I mentioned before you can create BitAssets by blockchain fiat and have a way of redeeming its value based on the value of BTS collateral, but you still need a BTS vs BitAsset market and so you'll probably eventually end up with something similar to the current design.  I'm open to different designs, but I think we're really close to getting a good BitAsset system with just a few tweaks. 
BitCash - http://www.bitcash.org 
Beta: bitCash Wallet / p2p Gateway: (https://m.bitcash.org)
Beta: bitCash Trade (https://trade.bitcash.org)

Offline Permie

  • Hero Member
  • *****
  • Posts: 606
  • BitShares is the mycelium of the financial-earth
    • View Profile
  • BitShares: krimduss
This is effectively what a CFD manager does when they offer a spread around fair value. To be able to do this, the CFD manager needs to absorb the difference between longs and shorts, and hedge their book of exposures in the external market. We could create something like this in bitShares. We would need a pool of users to act as a decentralised CFD manager and offer a spread around an external fair value, and be willing to hedge the long and short positions in the external market, using futures, ETFs or other available instruments. The pool would take income in the form of a market spread as we all as possibly a fee charged on the collateral of both longs and shorts. I would be interested to explore this concept further.

In the end I think bitShares could have many different classes of markets (currency, bonds, CFDs, ETFs, futures, options, shares, credit,...), each serving a different role for users, and each privatised. There is no need to think of all these things as competing ideas. They are complementary. There is simply no perfect solution that will meet all needs.

We have all been so obsessed by trying to find the perfect solution to currency, that we are missing the fact we need customised solutions for different needs. We are banging our heads against each other in vain. Everyone's needs can be met, but not through the same vehicle.
+5%

I've been thinking about a fund (maybe UIA) that employs traders or a bot-programmer to arbitrage bitAssets and make a low risk steady return. With the low liquidity we have right now I'm sure there is profit to be made on the spread.
If the trader(s) employed by this fund are human and fallible, then the fund could purchase insurance.

Another UIA could be issued that pays a dividend for every month the hedge (?) fund makes profit and doesn't fail. A selection of Delegates could be employed to act as arbiters in case of a claim.

If BitShares can provide products that earn a better return than other savings vehicles then new users will WANT to trade on our platform and we won't need marketing.
JonnyBitcoin votes for liquidity and simplicity. Make him your proxy?
BTSDEX.COM

Offline monsterer

We can discuss other markets in the future, but I think this CFD design is good and elegant so I just want to get this to be the best it can be first before adding other types of assets.

Isn't it important to recognise that none of the bitAsset proposals has been a true CFD design, though?

* There is no 'difference' payment - longs are essentially a spot trade and shorts end up in a different currency
* True CFD's do not persist after the contract ends
* CFDs do not create other asset types
* CFSs are 100% equal on both sides (same margin requirements, interest payments)

I think starspirt is right when he says there is no one solution to fit all of our requirement perfectly. The only trouble is, I don't think you'll be able to implement all the different asset classes as plain privatised bitAssets because the shorting mechanism will be the same across all of them.
My opinions do not represent those of metaexchange unless explicitly stated.
https://metaexchange.info | Bitcoin<->Altcoin exchange | Instant | Safe | Low spreads

Offline merivercap

  • Hero Member
  • *****
  • Posts: 661
    • View Profile
    • BitCash
I think it will work fine if settlement is agreed to be only on margin calls.  Theoretically it's the same as having a daily settlement and then having everyone repurchase.  I don't mind a daily settle, but it's cleaner to just wait for margin calls if we have a lot of long term BitUSD holders and shorters.  I'm not sure there isn't a historical precedent with market-settled contracts, but in any case I would prefer to evaluate a system with first principles and praxeology. 

It's necessary to have a large stable float of BitUSD to make an impact in the real world.  My biggest concern is that not enough traders will want to short for the exact reason stated by the original poster.   I want to help create BitUSD ecosystems in the real world and without enough supply of BitUSD it will be difficult.  It seems evident to me people don't want to short even after the run up in prices. 

I agree with your i) but not ii).  A system can use a CFD manager, but you can also just have natural margin call settlement.  I think it's unbalanced to have only one side be able to force settle.  It's better to have either both sides have the option or neither side.  The force settlement has an 'option value' and I think that is a large factor in the premium.   Again, I want to go long BitUSD, use BitUSD and have it circulate.  I don't think it's necessary for me to have this extra force-settlement option, especially because I'd be more afraid there would be less BitUSD supply. 

We can discuss other markets in the future, but I think this CFD design is good and elegant so I just want to get this to be the best it can be first before adding other types of assets. 
BitCash - http://www.bitcash.org 
Beta: bitCash Wallet / p2p Gateway: (https://m.bitcash.org)
Beta: bitCash Trade (https://trade.bitcash.org)

Offline starspirit

  • Hero Member
  • *****
  • Posts: 948
  • Financial markets pro over 20 years
    • View Profile
  • BitShares: starspirit
Yes 'having to cover' is a disadvantage, and something I wanted to remove from the system.  That's why I am advocating no user-generated forced-settlements and only automatically forcing under-collateralized positions.  That would make the market symmetric. 

merivercap, I understand your desire, most traders just want to trade without being forced out of anything unless they fail to maintain collateral. Unfortunately somebody must offer settlement in some form, either continuously or at an expiry, at a price near or at fair value, in order for any derivative to offer exposure matched to the underlying asset (a pegged asset is simply a form of derivative). All traditional derivatives have settlement, and there is no historical precedent for anything otherwise that I am aware of. I am not sure you accept this, and am happy to discuss my opinion further, perhaps outside this thread to not distract this discussion too far.

But assuming this is true, I think the desire you are expressing is:
i) for longs and shorts to simply be able to hold positions for as long as they like (i.e. no expiries), and
ii) be able to settle that against an intermediary rather than parties forcing the hand of opposing parties.

This is effectively what a CFD manager does when they offer a spread around fair value. To be able to do this, the CFD manager needs to absorb the difference between longs and shorts, and hedge their book of exposures in the external market. We could create something like this in bitShares. We would need a pool of users to act as a decentralised CFD manager and offer a spread around an external fair value, and be willing to hedge the long and short positions in the external market, using futures, ETFs or other available instruments. The pool would take income in the form of a market spread as we all as possibly a fee charged on the collateral of both longs and shorts. I would be interested to explore this concept further.

In the end I think bitShares could have many different classes of markets (currency, bonds, CFDs, ETFs, futures, options, shares, credit,...), each serving a different role for users, and each privatised. There is no need to think of all these things as competing ideas. They are complementary. There is simply no perfect solution that will meet all needs.

I've opened the possibility of an equivalent to ETFs in another thread, ideal for use as an investment vehicle. CFDs as above are ideal for leveraged trading, at a cost. Futures and options are ideal for managing exposures over a defined period.

We have all been so obsessed by trying to find the perfect solution to currency, that we are missing the fact we need customised solutions for different needs. We are banging our heads against each other in vain. Everyone's needs can be met, but not through the same vehicle.

As the OP infers without saying so, speculative traders should not be the party offering fungible non-expiring assets. This should be offered by parties willing to earn an income from making and supporting those markets. Speculators can have many other avenues to ply their trade as we develop them.

Offline merivercap

  • Hero Member
  • *****
  • Posts: 661
    • View Profile
    • BitCash
I think this is a great idea - you could even add leverage to it. It's one that I've considered suggesting myself.

Yes.  If you lower collateral requirements and shorten settlement durations you can effectively increase leverage.  Although I would not recommend high leverage for the main BitUSD, I think for a Privatized BitAsset this may work well.
BitCash - http://www.bitcash.org 
Beta: bitCash Wallet / p2p Gateway: (https://m.bitcash.org)
Beta: bitCash Trade (https://trade.bitcash.org)

Offline merivercap

  • Hero Member
  • *****
  • Posts: 661
    • View Profile
    • BitCash
bitAssets 2.0 will lower the collateral to 100% to each side...check!

- for the daily cover process, i personal think it is not needed, but i would love to cover with the collateral

I should have read that massive thread, but how does only 100% from shorters make a blackswan unlikely? There were times when the BitUSD collateral went down to 183% (down from 300 obviously).

And having to cover means that shorters are still guaranteed to be at a disadvantage, no? Short term CFDs would still be desirable.

Yeah you should have read the massive thread.

Black swans are based on time of settlement.  The original design has a 30-day hold so a 66% drop in 30-days would result in a black swan reset.  If you don't have a 30-day settlement period and allow settlement on any day a Black Swan will only occur when the market drops 50% in one day.  That is extremely unlikely and hence 100% collateral is sufficient.  Having higher collateral requirements than that is not necessary, but if we want to be conservative you can increase that requirement as well. 

Yes 'having to cover' is a disadvantage, and something I wanted to remove from the system.  That's why I am advocating no user-generated forced-settlements and only automatically forcing under-collateralized positions.  That would make the market symmetric. 
BitCash - http://www.bitcash.org 
Beta: bitCash Wallet / p2p Gateway: (https://m.bitcash.org)
Beta: bitCash Trade (https://trade.bitcash.org)

Offline merivercap

  • Hero Member
  • *****
  • Posts: 661
    • View Profile
    • BitCash
I havent been keeping up with all the latest revisions (too much day-job) so sorry if this idea has been discussed before, but...

At the moment no one is shorting BitUSD, IMO because shorting requires too much commitment.

Shorters have to:
-Put down twice as much collateral as the longs
-Are not guaranteed that they can easily/quickly get any BitUSD to cover
-Their collateral cant even be used to cover, they have to buy more BitUSD to cover (is that still the case?)

The result is that shorting is just waaay too stressful (I have never shorted BitUSD, it looks suicidal).

So how about reusing the price feeds to allow very short term CFD bets that are totally symmetrical to both sides? Both sides put in 100% worth of collateral, and the CFD is closed after 24 hours. The CFD is not tradeable so no covering is required, and you dont have to constantly monitor your position because you know that it will be fairly settled after 24 hours anyway.

You could possibly mix the markets together for extra liquidity: some BitUSD longs might not care about fungibility or not, so they could either get matched by a 200% BitUSD shorter for a long term BitUSD, or just a 100% shorter for a short term CFD.

Good idea? I personally would love to take short term bets on BTS/BTC - there's just no way in hell I am willing to go through the stress of putting down twice as much collateral and then desperately looking around to find BitBTC to cover afterwards.

This is similar to what I've been advocating and the current design already proposes 100% collateral vs 200%.   I'm open to a daily settle, but I think it may be just as effective to settle any undercollateralized positions (ie. you don't have to settle any shorts unless you are below 100% collateral.)  I am open to a full daily settlement of the entire BitUSD float, but I thought it may take up too much resources on the blockchain and ultimately may not be necessary.   I also don't like relying on the price feed so heavily either.

Not sure what you mean by matching or mixing markets.  The market should be fungible.  If you do a daily settle, you just need a way for BitUSD holders to repurchase each day so they can continue to hold BitUSD long term.
BitCash - http://www.bitcash.org 
Beta: bitCash Wallet / p2p Gateway: (https://m.bitcash.org)
Beta: bitCash Trade (https://trade.bitcash.org)

Offline speedy

  • Hero Member
  • *****
  • Posts: 1160
    • View Profile
  • BitShares: speedy
bitAssets 2.0 will lower the collateral to 100% to each side...check!

- for the daily cover process, i personal think it is not needed, but i would love to cover with the collateral

I should have read that massive thread, but how does only 100% from shorters make a blackswan unlikely? There were times when the BitUSD collateral went down to 183% (down from 300 obviously).

And having to cover means that shorters are still guaranteed to be at a disadvantage, no? Short term CFDs would still be desirable.

Offline Shentist

  • Board Moderator
  • Hero Member
  • *****
  • Posts: 1601
    • View Profile
    • metaexchange
  • BitShares: shentist
I havent been keeping up with all the latest revisions (too much day-job) so sorry if this idea has been discussed before, but...

At the moment no one is shorting BitUSD, IMO because shorting requires too much commitment.

Shorters have to:
-Put down twice as much collateral as the longs
-Are not guaranteed that they can easily/quickly get any BitUSD to cover
-Their collateral cant even be used to cover, they have to buy more BitUSD to cover (is that still the case?)

The result is that shorting is just waaay too stressful (I have never shorted BitUSD).

So how about reusing the price feeds to allow very short term CFD bets that are totally symmetrical to both sides? Both sides put in 100% worth of collateral, and the CFD is closed after 24 hours. The CFD is not tradeable so no covering is required, and you dont have to constantly monitor your position because you know that it will be fairly settled after 24 hours anyway.

You could possibly mix the markets together for extra liquidity: some BitUSD longs might not care about fungibility or not, so they could either get matched by a 200% BitUSD shorter for a long term BitUSD, or just a 100% shorter for a short term CFD.

Good idea? I personally would love to take short term bets on BTS/BTC - there's just no way in hell I am willing to go through the stress of putting down twice as much collateral and then desperately looking around to find BitBTC to cover afterwards.

bitAssets 2.0 will lower the collateral to 100% to each side...check!

- for the daily cover process, i personal think it is not needed, but i would love to cover with the collateral

Offline monsterer

I think this is a great idea - you could even add leverage to it. It's one that I've considered suggesting myself.
My opinions do not represent those of metaexchange unless explicitly stated.
https://metaexchange.info | Bitcoin<->Altcoin exchange | Instant | Safe | Low spreads

Offline speedy

  • Hero Member
  • *****
  • Posts: 1160
    • View Profile
  • BitShares: speedy
I havent been keeping up with all the latest revisions (too much day-job) so sorry if this idea has been discussed before, but...

At the moment no one is shorting BitUSD, IMO because shorting requires too much commitment.

Shorters have to:
-Put down twice as much collateral as the longs
-Are not guaranteed that they can easily/quickly get any BitUSD to cover
-Their collateral cant even be used to cover, they have to buy more BitUSD to cover (is that still the case?)

The result is that shorting is just waaay too stressful (I have never shorted BitUSD).

So how about reusing the price feeds to allow very short term CFD bets that are totally symmetrical to both sides? Both sides put in 100% worth of collateral, and the CFD is closed after 24 hours. The CFD is not tradeable so no covering is required, and you dont have to constantly monitor your position because you know that it will be fairly settled after 24 hours anyway.

You could possibly mix the markets together for extra liquidity: some BitUSD longs might not care about fungibility or not, so they could either get matched by a 200% BitUSD shorter for a long term BitUSD, or just a 100% shorter for a short term CFD.

Good idea? I personally would love to take short term bets on BTS/BTC - there's just no way in hell I am willing to go through the stress of putting down twice as much collateral and then desperately looking around to find BitBTC to cover afterwards.
« Last Edit: June 04, 2015, 09:21:29 pm by speedy »