Author Topic: Issuer Pools - Simpler Form of BitAsset With Near Perfect Pegging?(another idea)  (Read 1865 times)

0 Members and 1 Guest are viewing this topic.

Offline monsterer

I haven't analysed your proposal in detail, but to me replacing the complexity of a thousand different order types all operating in one market with separate market types seems elegant. The current system is so overcomplicated I'm surprised anyone can make sense of it. Even the developers of the wallet can't since the orderbook is crossed right now(!)
My opinions do not represent those of metaexchange unless explicitly stated.
https://metaexchange.info | Bitcoin<->Altcoin exchange | Instant | Safe | Low spreads

Offline starspirit

  • Hero Member
  • *****
  • Posts: 948
  • Financial markets pro over 20 years
    • View Profile
  • BitShares: starspirit
OK, I'm now enhancing this OP idea, to add mechanisms that better regulate supply and demand, thus avoiding rapid swings in the units on issue. I want to stress that this is still just an idea to consider, and there are other variations each with potential pros/flaws/cons.

The regulating mechanisms to add are as follows:

i) When bitCurrency demand is growing too rapidly for the growth of the Issuer Pool to accommodate it, the price for new bitAssets at the Issuance Window is allowed to rise to a premium. The surplus is a profit to the Issuer Pool that incentivizes more investors to join.

ii) As alluded to in the OP, when bitCurrency demand is waning relative to supply, or there are more BTS bulls than current bitAsset demand can accommodate, ownership incentive can be provided through bitCurrency lending and borrowing markets that sit as an adjunct to the Issuance Pool. bitCurrency owners would be incentivised to lend bitCurrency rather than redeem them, and Issuer Pool investors would be incentivised to move money from the Issuer Pool (where their leverage will be dropping) to lending markets where they can get more leverage. This would help bring the supply and demand for bitCurrency units back into balance.

On this approach, there is no need for interest to be employed within the Issuer Pool itself.

BitShares could then run the following in combination:

A) A free market (limit and market orders only) in bitCurrency, plus
B) An Issuance Window, backed by an Issuer Pool, plus
C) Lending/borrowing markets for the bitCurrency at a range of maturities

The benefits of this approach, on top of those identified in the OP,  would now include the ability of both longs and shorts to position across a range of maturities, rather than being constrained to 30 day shorts for example.

This is not necessarily the ultimate model I am proposing, and I'm still working on other variations. But this hopefully provides some food for thought on the possible benefit of separating various functionalities. I am gradually coming to a view that bitAssets in their current form are mixing many of these elements of free market, cash pools and 30-day borrowing markets together, and perhaps making for a more complicated development process and user experience than might be necessary.
« Last Edit: April 09, 2015, 02:02:17 am by starspirit »

Offline starspirit

  • Hero Member
  • *****
  • Posts: 948
  • Financial markets pro over 20 years
    • View Profile
  • BitShares: starspirit
I still see the superfluidity of creation/destruction and the associated peg tightness as attractive features of this concept. So I may think more about the interest mechanism, maybe do a second take on it at some point.

Offline xeroc

  • Board Moderator
  • Hero Member
  • *****
  • Posts: 12922
  • ChainSquad GmbH
    • View Profile
    • ChainSquad GmbH
  • BitShares: xeroc
  • GitHub: xeroc
I think the devs discussed something similar in github a few weeks ago and they came to the same conclusion ..

Offline starspirit

  • Hero Member
  • *****
  • Posts: 948
  • Financial markets pro over 20 years
    • View Profile
  • BitShares: starspirit
On second thoughts, kill this idea for now.

In hindsight, I think it would be a mistake to take out the interest rate as a natural market regulator. This is more than a nice-to-have feature for bitassets - fundamentally it means that there is a regulating mechanism to adjust supply and demand without the bitCurrency supply having to expand or contract significantly if either side changes view. In the absence of such a mechanism, its likely that bitCurrency supply  and demand would be much more unstable.

An issuer pool is still probably a UIA idea to form a short pool for those not wishing to manage short positions themselves.

Offline xeroc

  • Board Moderator
  • Hero Member
  • *****
  • Posts: 12922
  • ChainSquad GmbH
    • View Profile
    • ChainSquad GmbH
  • BitShares: xeroc
  • GitHub: xeroc
Also we would lose one of the big seeling points: the yield

Offline monsterer

If users can buy an unlimited quantity of bitassets at the feed price, doesn't that create arbitrage opportunities for them on other exchanges which could lead to under collateralisation?
My opinions do not represent those of metaexchange unless explicitly stated.
https://metaexchange.info | Bitcoin<->Altcoin exchange | Instant | Safe | Low spreads

Offline starspirit

  • Hero Member
  • *****
  • Posts: 948
  • Financial markets pro over 20 years
    • View Profile
  • BitShares: starspirit
[Edit May 12: The ideas in this thread have now been superseded by whitepaper here: https://bitsharestalk.org/index.php/topic,15880.0.html]

This is not a proposal, just a personal idea, let me know what you think.

Basic Structure

There would be a standard free market in the bitCurrency (e.g. bitUSD, bitCNY, bitEUR...) against BTS, with only bids/asks, and no price feeds or shorts,

PLUS

There would be an Issuance Window, where users are free to create and destroy units of the bitCurrency instantly at the median feed price (MFP) +/- a small spread.

An Issuer Pool would automatically make the market at the Issuance Window, by accepting BTS to create new bitCurrency, or distributing BTS to destroy bitCurrency, instantly upon demand and in unlimited quantity for destruction, and nearly unlimited (see below) for creation. The Issuer Pool would be capitalised by a combined pool of investors seeking leverage to the BTS price, who have paid BTS into the pool by buying units in the pool (Issuer Units). The Issuer Units would trade in their own free market against BTS.

The leverage to BTS in the Issuer Pool varies with both bitCurrency creation/destruction (i.e. demand) and movements in the BTS price. The Issuer Pool is required to remain within specified leverage limits. If leverage begins to get too high, the Issuer Pool would issue more Issuer Units into the free market, tending to create a discount to NAV that would attract new investors and raise capital. If leverage begins to get unattractively low, the Issuer Pool would buy back Issuer Units in the free market, tending to create a premium to NAV that encourages capital release to increase leverage again. These changes in unit supply should be prescription-based and transparent to the market with notice, so that the market has time to prepare to take advantage of it.

The only limit on unfettered activity at the Issuer window would be that new creation would be stopped and queued if the Issuer Pool failed to raise more capital when beyond its leverage limit, despite significant discounts to NAV on the Issuer Units (e.g. if the Pools had grown too large relative to total BTS supply that is available to support them). In this scenario bit-Currency premiums in free markets would be unresolved, just as they would in the current system.

Key Benefits

- BitAssets would retain their key advantages over competitors, being it is transparently and overly collateralised, and supply is highly responsive to changes in demand in both directions
- The peg would be as close to perfect as possible, because market-makers could provide massive buy and sell walls around MFP in the knowledge they have access to the Issuance Window
- It removes the complexity associated with individual shorts, including expiries, calls, rolls and interest (shorts are now replaced by Issuer Pool investors)
- It allows free market trading in bitCurrencies:BTS in a simple form that people are traditionally familiar with, without needing to understand complicated market rules
- It allows other functionalities to be built in the BitShares environment that complement these markets, such as lending, depositing, margin trading etc (see below)

Impacts on bitCurrency users

They can have confidence operating in highly liquid strongly pegged markets, whether they ever personally use the Issuance Window or not.
They no longer receive interest (bitCurrencies are more like cash) but they can receive interest in lending or deposit markets (below), just like when they lend or deposit fiat cash. This is less confusing than the current market which combines properties of both cash and deposits.

Impacts on Issuer Pool Investors (replacing shorts)

Unlike shorts in the current system, Issuer Pool investors share the costs associated with managing a leveraged investment. Costs exist in both systems, but in different forms.

When supply is too high for demand in the current system, shorts would need to raise their interest rates to compete to retain their shorts beyond expiry. In the Issuer Pool, unit-holders get diluted by those accepting the buyback of Issuer Units at a premium, if they do not also accept it themselves and take capital out.

When BTS falls in price and collateral becomes low for some shorts in the current system, they can choose to add collateral, or get called and effectively lose BTS at a low price (which is a cost). If the Issuer Pool runs low on collateral, it sells new units at a discount, which dilutes any investors that choose not to participate by contributing more collateral.

If demand is running hot and outstripping supply, there are no costs for shorts in the current system in maintaining their position. However investors in the Issuer Pool will be diluted by new Unit issuance if they choose not to participate.

In the current system, shorts receive a leveraged return on their BTS, less interest, and less the costs associated with inopportune calls or expiries. In the Issuer Pool, investors receive a leveraged return on their BTS, plus a return from market-making at the Issuance Window, less dilution from new unit issuance or buybacks that they do not participate in pro-rata.

In the current system, shorts can attempt to earn extra timing return by shorting at premiums and covering at discounts. Issuer pool investors can attempt to earn extra timing return by buying more Issuer Pool units at discounts and selling at premiums.

Its not clear to me that there would be a big difference one way or the other, on average, from a purely economic perspective. However it is much easier for the average investor to participate in the Issuer Pool than it is to manage shorts, which may expand the available supply.

Impacts on market-makers

Market-makers would now make a market anywhere in the bit currency, and utilise the Issuance Window for hedging or offsetting. This makes the task easier and less risky, so spreads should be very tight. Gateways would likewise be able to offer very tight spreads.

Complementary functionalities

A direct lending market could be established in the bitCurrency, allowing bitCurrency lenders to earn interest at risk. The borrowers could use or sell the bitCurrency for any personal purpose, opening up the set of all available borrowers beyond just BTS bulls, in the same vein as BTCjam.
A banking/deposit market could be established in the bitCurrency, where an intermediary accepts interest-bearing deposits, and lends to a managed pool of borrowers, taking some equity risk.
A margin lending market could be established in the bitCurrency, providing margin for other on-blockchain investments such as BTS. This could operate similarly to Bitfinex.
Other possibilities are using the bitCurrency for margin accounts to underpin derivative and CFD markets.

Essentially there is no need for the properties of any of these markets to be built into the primary bitAsset market, because they can all be provided separately, such as via UIAs. In particular, it is not necessary to offer interest in a cash-creation market if lending and deposit markets can provide this.

There's more I could flesh out, depending on initial interest.
« Last Edit: May 12, 2015, 12:32:07 am by starspirit »