Author Topic: A New Proposal of Interest for BitUSD Holders  (Read 9075 times)

0 Members and 1 Guest are viewing this topic.

Offline liondani

  • Hero Member
  • *****
  • Posts: 3737
  • Inch by inch, play by play
    • View Profile
    • My detailed info
  • BitShares: liondani
  • GitHub: liondani
liondani, a quick look at my post history would show I'm on the side of a free ..................

Sorry my posted answer was for bytemaster's thread  :P

Offline jsidhu

  • Hero Member
  • *****
  • Posts: 1335
    • View Profile
In todays market there is something called a SWAP which pays an interest rate to the bank (or whoever you borrow from) which can be negative based on the currency you borrow vs the one you lend. ie holding long GBP/JPY pays positive interest because you buy GBP at a higher interest rate and sell JPY at a lower interest rate, making the difference. Could something be applied analogous to lending mechanisms today in the world of bitshares?

 +5%

This swap could be derived from the market inefficiencies.

What if you adjust the swap based on how far from the median price the order executed?

Lets' say someone shorts 20% below median price of BitUSD. Then they just entered a contract, in which the shorter pays 20% interest to the holder of the BitUSD. Of course trouble with this is that you can't have different classes of BitUSD floating around (i.e. earning different interest). Perhaps interest is only paid if the buyer is still holding.

Either way, some sort of incentive needs to be introduced, it is evident that the peg is not working even with the short restriction.

I think that is the idea of the interest rate I have been talking about.. so essentially the interest rate is working as a SWAP in this case... and something that BM et all don't agree that is necessary seemingly. This difference from the median would essentially be the premium one side pays from bitUSD of 1... to keep a peg, but you're example is to have a premium based on the median price derived from the price feed? hmm not sure why not just use 1 and negate the use of price feeds?

Anyways what I was thinking about was the use of multiple rates based on the asset somehow, and borrowing one against the other one comes to a known conclusion of how much they will earn or have to pay over time to hold the trade. THe way SWAP works is that there is a daily turnover and you pay or collect interest at a set GMT time... so we can essentially transfer premiums or swaps or interest or whatever each day based on the interest rate (per day) .. this would be different then the long term interest somehow, to motivate people to hold long term?

It might mean that we would have multiple prediction markets for each asset, or asset class that would set the willing interest rates of the participants... but personally for peg reasons I thought to just tie it into the peg of 1 for now... but a more thoughtout mechanism may use a prediction market somehow to arrive to a conclusion what the premiums should be.
« Last Edit: September 18, 2014, 07:36:51 pm by jsidhu »
Hired by blockchain | Developer
delegate: dev.sidhujag

Offline bitmeat

  • Hero Member
  • *****
  • Posts: 1116
    • View Profile
In todays market there is something called a SWAP which pays an interest rate to the bank (or whoever you borrow from) which can be negative based on the currency you borrow vs the one you lend. ie holding long GBP/JPY pays positive interest because you buy GBP at a higher interest rate and sell JPY at a lower interest rate, making the difference. Could something be applied analogous to lending mechanisms today in the world of bitshares?

 +5%

This swap could be derived from the market inefficiencies.

What if you adjust the swap based on how far from the median price the order executed?

Lets' say someone shorts 20% below median price of BitUSD. Then they just entered a contract, in which the shorter pays 20% interest to the holder of the BitUSD. Of course trouble with this is that you can't have different classes of BitUSD floating around (i.e. earning different interest). Perhaps interest is only paid if the buyer is still holding.

Either way, some sort of incentive needs to be introduced, it is evident that the peg is not working even with the short restriction.

Offline jsidhu

  • Hero Member
  • *****
  • Posts: 1335
    • View Profile
In todays market there is something called a SWAP which pays an interest rate to the bank (or whoever you borrow from) which can be negative based on the currency you borrow vs the one you lend. ie holding long GBP/JPY pays positive interest because you buy GBP at a higher interest rate and sell JPY at a lower interest rate, making the difference. Could something be applied analogous to lending mechanisms today in the world of bitshares?
Hired by blockchain | Developer
delegate: dev.sidhujag

Offline starspirit

  • Hero Member
  • *****
  • Posts: 948
  • Financial markets pro over 20 years
    • View Profile
  • BitShares: starspirit
liondani, a quick look at my post history would show I'm on the side of a free market incentives. The OP for this thread proposed a complicated rule system that I was opposed to, and along the way I thought of a what I think is a better alternative given a rule-based approach.

If there are no rules or other mechanisms in place, and price is free to float anywhere, I cannot see any reason why the consensus should necessarily settle on a price around the peg. It could settle anywhere and constantly shift. It would then behave like any independent crypto. At the moment, being within 10% seems like an excellent result IMO.

In the end, I think there needs to be some sort of mechanism to ensure this consensus settles and becomes stable around the peg. I do not think any of the rules introduced to date achieve this, because they assume ex-ante that there is some sort of magic attraction around the peg price (a lot of which seems to rely on the principle of keep saying it till everyone believes it), and a little bump here or there will remove the current gap. Instead I think there needs to be a mechanism in place that financially affects the results to all parties of trading around that peg price specifically, not just any price the market consensus might instead settle upon. That I suspect is where the self-perpetuation idea mentioned in the white-paper might break down - it is a "holding yourself up by the bootstraps" concept that only works if the market has already permanently settled on the peg price as being the one and only right price.

To give the peg price such financial tangibility, some possibilities are: (i) a floating (interest) incentive paid from shorts to longs (or visa versa), (ii) a floating (interest) incentive paid from BitUSD borrowers to BitUSD lenders in a separate lending market where BitUSD could be deposited for lending, (iii) market exchangers who are willing to make BitUSD 1:1 fungible with USD, subject to an incentive. There may be others you might think of. But even in these cases, we need to be careful assuming things about what the market consensus (if there is even such a thing) might be.

I think such a mechanism would see confidence in the peg rise considerably, especially in outside circles, and then we see potentially massive interest and growth.

Offline liondani

  • Hero Member
  • *****
  • Posts: 3737
  • Inch by inch, play by play
    • View Profile
    • My detailed info
  • BitShares: liondani
  • GitHub: liondani
It is getting to complicated !
Make it please simple!
Cut the Gordian knot!

In theory maybe it's ok, or maybe you will find even "better" solutions than the initial proposal but I think
it will not work! I feel allready I must go to a University to study bitsharesx before starting with trading!

Let the market free! Remove even the initial restrictions for shorting? What can go so wrong ? What are we afraid? Its normal on early days to have a peg even at -40% at some point because the volatility  is huge now for BTSX, but that will change after some weeks/months.... Its different to have a 60 million market cap and different with 6 billion (and we get there).
What do you think will the average investor/speculator/supporter do when 1 bitUSD is 0.6 USD ? (Personally I will only watch because I would have gave all my money for bitUSD when it was at 0.75 USD).And what will I do when 1 bitUSD goes to 1.1 USD? Guess!.... I mean it doesn't need a rocket scientist to predict what will happen! The early days the peg will deviate from parity and every day it passes and the market matures it will come closer and closer... Am I missing something??? We must give more time to the market!
And yes it is more importand to have a stable platform right now with lots of new futures, tweaks, design improvements,bug free and more secure, and of course begin the marketing aggressively. Without happy users (market) the peg will never be established even if we have 10 Agents on our team.

PS After all it's a bitUSD, think about it! The most stable cryptocurrency that gives you interrest....
PS2 Try to accomplish your goals only with a free market and .... interest/dividends
PS3 I really think it will make a big difference when you let the users make orders (bid or asks) depended of the ratio...    [(bitAsset price) / (median price of realAsset)]
      For example :  buy x bitUSD at ratio:85% nobody would be afraid to make orders !!!!!!!!!!!!  for me that's the cut of the GORDIAN KNOT!!!!!!!!! That's the way to help the peg!!!!
      Then we see on the charts BIG BIG WALLS.... ;)


EDIT: I POSTED THIS ANSWER ON A WRONG THREAD !!!   SORRY
« Last Edit: September 18, 2014, 09:50:51 pm by liondani »

Offline starspirit

  • Hero Member
  • *****
  • Posts: 948
  • Financial markets pro over 20 years
    • View Profile
  • BitShares: starspirit
I just had a subsequent idea that levers off your principle of a floating R, but better satisfies my concerns that R always acts to sustain rather than distort the peg. What if you had a simpler set of rules, that if BitUSD is below the peg price, R could be steadily increased once every hour say, and if above, R could be steadily reduced every hour. The rate of increase or reduction at the end of each hour could further depend on how far the price is from the peg price, the further it is requiring more aggressive change. The increments would step down again as the gap tightened.

It should also not be required to have any limits on R, as if they are hit, it will only mean the peg will break again. We should be prepared for R to get very high if that is what is required to balance the supply and demand, given the view is that the peg is more important.
This would still require price feed.

The question whether R limits should be lifted is open to discussion. They are there for preventing short-term market manipulation.
Yes a price feed would be required, which ideally we would do without. But without some sort of reference to a feed price, any rules for adjusting R could just as likely be distorting the peg as aiding it. The only alternative schemes to this would involve having a way for the market to agree on the level of R that equates supply and demand. This is being explored in some other threads.

I don't think an unrestricted R could lead to market manipulation. Market manipulation can occur where few trades set a price for the entire market. In this case, R could not be outside where the market wants it because to drive it to that level would mean either buying large amounts way above the feed price or selling way below it, and the rest of the market will be glad to take the other side of the trade. The manipulator would make losses. Bit I could be missing something. Can you provide a scenario would a manipulator might be able to screw the rest of the market?
 

Offline ripplexiaoshan

  • Board Moderator
  • Hero Member
  • *****
  • Posts: 2300
    • View Profile
  • BitShares: jademont
I strongly agree that we need more manipulation on bitUSD/bitCNY issuance based on the current situation. Now that the business market requirement of bitUSD/bitCNY is still very low, BTA issuance should be further limited by adding more cost to "short", besides price feeds. Price feeds can only guarantee a rough pegging, but a dynamic interest rate will guarantee an accurate pegging.  However, the rule that we need to follow to change the interest is still worth discussing.   
BTS committee member:jademont

Offline gulu

  • Full Member
  • ***
  • Posts: 121
    • View Profile
I just had a subsequent idea that levers off your principle of a floating R, but better satisfies my concerns that R always acts to sustain rather than distort the peg. What if you had a simpler set of rules, that if BitUSD is below the peg price, R could be steadily increased once every hour say, and if above, R could be steadily reduced every hour. The rate of increase or reduction at the end of each hour could further depend on how far the price is from the peg price, the further it is requiring more aggressive change. The increments would step down again as the gap tightened.

It should also not be required to have any limits on R, as if they are hit, it will only mean the peg will break again. We should be prepared for R to get very high if that is what is required to balance the supply and demand, given the view is that the peg is more important.
This would still require price feed.

The question whether R limits should be lifted is open to discussion. They are there for preventing short-term market manipulation.
BTC: 1FRsgcQELHKFY2VMFosBnyBaNxZgS4wj7x
PTS: Pf9yDYUknXShepu6YjwSEpMvm1ewNQLNgE
BTSX: gulu

Offline PTS中国

  • Sr. Member
  • ****
  • Posts: 416
    • View Profile
  • BitShares: ptschina
--------

PTS中国

Offline 当年很厉害

  • Sr. Member
  • ****
  • Posts: 310
    • View Profile
BTS粉里有SB!

Offline starspirit

  • Hero Member
  • *****
  • Posts: 948
  • Financial markets pro over 20 years
    • View Profile
  • BitShares: starspirit
I just had a subsequent idea that levers off your principle of a floating R, but better satisfies my concerns that R always acts to sustain rather than distort the peg. What if you had a simpler set of rules, that if BitUSD is below the peg price, R could be steadily increased once every hour say, and if above, R could be steadily reduced every hour. The rate of increase or reduction at the end of each hour could further depend on how far the price is from the peg price, the further it is requiring more aggressive change. The increments would step down again as the gap tightened.

It should also not be required to have any limits on R, as if they are hit, it will only mean the peg will break again. We should be prepared for R to get very high if that is what is required to balance the supply and demand, given the view is that the peg is more important.

Offline starspirit

  • Hero Member
  • *****
  • Posts: 948
  • Financial markets pro over 20 years
    • View Profile
  • BitShares: starspirit
Great to keep thinking about this problem, but I do not think this propel will work. There is no reason why any fixed rule (e.g. rule 5) should lead to the BitUSD being properly pegged. Buy and sell orders for BitUSD will be driven by many factors such as views on Bitshares, not just the discount/premium to real USD. So situations are possible such as where BitUSD is already at a discount to USD for some reason, people suddenly become more bearish on BTSX (buys increase on BitUSD) but the action of rule 5 would be to decrease interest paid on BitUSD and diminish the incentive further to own BitUSD over real USD.

The only way I can think of to get such a scheme to work is to actually fix the peg (i.e. fix BitUSD to real USD price) and float the interest rate. That is, people make bids and offers quoted as an interest rate rather than the price. The matching process defines the variable market interest rate at any time. With such a scheme, you would have to be ready for interest rates to vary dramatically and be extremely high at times of peak BTSX demand (e.g. see Bitfinex margin lending book as an analogy).
The rule5 can be modified to be some percentage of existing BitUSD. Not a big issue. When R moves from an equilibrium, it is the result of the interest rate acting as a counter-balancing force against initial market shift.

Bidding on interest rate would make BitUSDs non-fungible, because different BitUSDs would carry different interest rate.

Re rule 5, I'm definitely trying to understand the logic, but still not convinced that trying to temper the "driving" orders through changes in R always acts to reduce the peg, because it does not refer to the actual peg level in any way - at times it could distort the gap further. Suppose the value of BTSX rises, then sellers of BitUSD should drive price down to keep the proper peg. Won't rule 5 be counteracting that? Also you refer to an Rmin and Rmax limit, but the market may be in equilibrium outside of that, meaning that even at either limit the peg may not be holding (and liquidity is consequently thin). Now if the "driving" order were subsequently in the direction of reducing the distortion, wouldn't rule 5 again counteract that? There are many other examples I think showing that rule 5 is laden with assumptions about what is driving movement to or from the peg. In principle I think R needs to be set by the market, not a rule.

(Re bidding on rates, I accept your point on non-fungibility.)

Offline Empirical1.1

  • Hero Member
  • *****
  • Posts: 886
    • View Profile
I like your proposal.

Edit: But why don't we bring money into our system the way treasuries bring money into theirs?

This is my understanding of how BitUSD would be brought into use for a treasury in the real world..

Bonds are sold at auction by the treasury (the auction decides the interest rate)

In our case the shorts would be the treasury but unlike an overly indebted country we know the shorts are good for the interest because of the collateral requirements.

Kind of like this video up to 1:20 https://www.youtube.com/watch?v=p3_Q1SiRN-A

Why don't we bring BitAssets into existence similar to the way bonds are done (Excl. central banks and fractional reserve banking.) It seems to work so why re-invent the wheel. I don't understand enough though to figure out exactly how to cross apply it to BitAssets.
« Last Edit: September 02, 2014, 08:41:55 pm by oasis »

Offline gulu

  • Full Member
  • ***
  • Posts: 121
    • View Profile
Expected end results:
1) R reaches equilibrium. This is the market-driven interest rate, which represents the wills of both sides of the market. If one side disagrees, it would move R by buying or selling BitUSDs to the direction that it prefers.

This makes a lot of sense.
Assuming R has reached equilibrium - please explain how is this interest rate going to be applied?
I mean: how will bitUSD shorts be paying the interest R to bitUSD longs? What will be the mechanism?

Shorts will be paying the interest from their collateral. An average interest rate can be calculated every 8640 blocks, which is one day in time. For instance, when 1 BitUSD is covered and the long is entitled to 0.05 BitUSD in interest, he would receive 1.05 BitUSD worth of BTSX based on current market price (or moving average of last 1000 blocks).
BTC: 1FRsgcQELHKFY2VMFosBnyBaNxZgS4wj7x
PTS: Pf9yDYUknXShepu6YjwSEpMvm1ewNQLNgE
BTSX: gulu