Topic: MasterCoin: New Protocol Layer Starting From “The Exodus Address” - page 150. (Read 448489 times)

The cheapest below target is going to be at or below  Fund / Owed...  which will thus represent the same risk profile as owning MC directly without any of the gains.

Yes yes and yes! MasterCoins should be a better investment over the long haul, but some people want stability!

Wait, did we just agree? At least, I don't see anything in this post that I disagree with. Maybe I'm missing something . . . .

Ah. There's the spiral_mind I know and love!

The escrow fund buys the cheapest coins below target. If somebody has offered them for sale below the target price, the escrow fund can buy them at that price. If they didn't want to sell at that price, they shouldn't have offered them for sale. The escrow fund merely buys and sells what is out there.

The escrow fund is created the moment the first person buys a new pegged currency. 100% of their money goes into the escrow fund.

Price would move for its own sake, not because of your actions.  Price is driven by supply/demand for Gold IOU backed by 100+X% margin and the only thing your fund would be doing is maintaining the margin or decreasing it to some minimal level, say 150%.  

You still have the initial condition to address.   Everyone initially agrees to buy GoldCoin and 'self insure' everyone else against temporary losses due to volatility with a baked in assumption that MC will go up relative to Gold.    Everyone also recognizes that this insurance only works as long as everyone continues to hold a balance and not withdraw their funds and who ever holds on the longest when the price goes down takes the largest loss.  

In theory, as long as the fund never debased the escrow ratio and MC was intrinsically deflationary and growing in market demand then anyone who could 'hold on until the price recovery' would get their money back.   However, they could have held MC itself and had a better investment because they wouldn't be magnifying their losses by propping up (insuring someone else) and they would have an opportunity to profit from the gains in MC.  

Effectively this system transfers losses from short-term volatility to those who are holding for long-term gains in MC value.  Those who are LONG MC would be better off holding MC than holding their value in GC which has almost 0 upside potential (only the premium from increased margin) and more than 100% of the long-term downside potential.  

You must follow the money and always identify who is profiting and who is taking the loss and make sure that the profit == loss and thus a value transfer occurs.   Then ask yourself if the parties signed up for that particular transfer and what compensation did they receive for the risk they took?

If you sell for higher than market price there's no reason anyone would buy from you. By definition a "market price" implies that this is a well functioning market which requires there to be plenty of buyers and sellers (not just you). If others are selling the same asset as you but for 1% lower there's no reason anyone would ever buy from the escrow.

By the way where are you proposing this escrow even comes from? So far you've just proposed that money sent to the "exodus address" go directly to funding you, your wife, and your kids.

Well, the escrow fund hits the bid or ask as appropriate. If you do that enough times, the price moves. It's logic would be "buy the cheapest 1% of coins for sale below the target" or "sell new coins to the 1% highest bids above the target"

It seems we're finally getting somewhere after all these walls of text. I was starting to wonder if it would ever end, but I think I see the light at the end of the tunnel!

So, we've established that an over-funded escrow fund can provide downside protection. I argue that a break-even fund or even a modestly underfunded one can also provide protection. For instance, a 90% funded escrow fund can survive 80% of the GoldCoins being cashed out in a panic (and be healthier in the end!). Your instinct is correct that this fund is vulnerable, but only to a sufficiently large panic.

If the fund sells at the ask price of $1.2 then presumably it will not result in the price falling, but instead the MC Held / Value Owed ratio would be unchanged.

You are correct, the fund would never have an opportunity to act to push the price down because no one would bid it up until the margin was insane.  So your debasing algorithm could work assuming it never debased the fund below 150 to 200% margin.

Unfortunately, to get there everyone had to bet right initially.   People would probably be willing to pay above market rates proportional to their demand for margin and thus GoldCoin will always trade at a premium to Gold when the escrow fund is above 150% and the algorithm will not debase beyond 150%.

On the down side though, your algorithm cannot prop up the price because it will create a positive feedback loop devaluing GoldCoin to nothing by continually debasing it.


This means that your system only offers down-side protection *after* it has already profited to the upside.  The initial buyers of GoldCoin have no such protection and would be better off not owning GoldCoin because they are exposed to losses but not gains in MC.  

No, that would be the opposite of a premium - that would be a giveaway!

The escrow fund sells at $1.2 when the ask is $1.2. Consequently, the ask price is now less than $1.2 by some amount.

Is the fund really profiting by selling for $1.1 when the current ask price is $1.2?    Every other actor on the market believes a GC is worth $1.2 and yet the escrow fund is selling for $1.1....  when it does this value is neither created nor destroyed, only transferred.  It gets transferred from everyone who holds GoldCoin to the new owner.   In this case $0.1 is transferred and the average backing after the freshly printed GC is less than the average backing before and therefore even when selling for $1.1 the fund is de-capitalizing itself rather than capitalizing itself.   Profit or loss from the perspective of the fund must be measured in terms of   MC Held / Value Owed.   Show me once how your escrow agent is ever able to take an action that increases the MC Held / Value Owed?

When someone wants out they are willing to take a loss.   The normal loss would be at the price of MC Held / Value Owed, however, you are attempting to prop up the price and let them out for more than MC Held / Value Owed and as a result the fund makes a loss.  

You are right about the possibility of collapse when the escrow fund is unhealthy. I think we are all agreed on that, although we probably have different ideas of how unhealthy it would have to be before it reached the tipping point.

Your examples (both here and previously) seem to assume that an over-funded escrow fund would cause investors to over-pay for a currency and take a loss when the escrow fund debases them. I can't think of why someone would do this, knowing they would take a loss. Consequently, it seems clear that the escrow fund would remain 150% over-funded indefinitely.

You are only looking at the market from the perspective of the escrow fund, but not from other users.

Why should *I* buy above the target price?   If I do, the fund will push the price down and I have guaranteed losses.  Conclusion: no market participant with any sanity what-so-ever would ever bid above the target price and therefore your assumption that the escrow fund could raise money this way is wrong.  The only time the escrow fund could sell GoldCoin would be when the value of the escrow fund is over 100% and it wouldn't be selling at a profit.

The escrow fund socializes the losses and privatizes the profits.   If MC goes up the fund keeps the profits and prints new GoldCoin transferring the gain from the fund to those who buy the new GoldCoin.  After all they would be buying GC with 1.5 backing for the price of 1 until the backing fell back to 1:1.

If MC goes down then the fund is unable to make everyone whole and is left playing a confidence game / ponzi scheme of temporally propping up the price by stealing the backing from other GoldCoin holders.  As long as the current holders are ignorant to the fact that the only thing propping up the price is the backing of their own coins it will work.

How can you tell if the automated system is making or losing value?   If it makes decisions that no rational actor would make, it is losing money.  

Remove the 'magic' of your algorithm for a second and instead lets pretend I create a bank, the ByteMasterBank  BMB.

I tell people that I will issue a BMB IOU for $100 USD if you deposit 1 BTC into my bank and that I will always be willing to redeem that $100 USD IOU at face value, but you will have to accept $100 worth of BTC in place of actual USD when you redeem your IOU.   Assuming price stability I will always be able to meet my obligations.   If the price of BTC goes up then I can make a killing because my IOUs are denominated in USD.

So the price of USD goes up and I now have enough BTC in my fund to buy back all outstanding USD loans 2x over.   Great!  I can now print up additional USD loans and start buying things with them all while maintaining over 100% reserves!  

The next week the price of USD goes up and all of a sudden I am insolvent, I only have the ability to buy back 50% of the outstanding USD loans.   As long as no one catches on I can play the game of a fractional reserve bank and still honor withdraws (redemptions at face value).   How ever, once 50% of my depositors have withdrawn their money that game is over and everyone else loses everything because I am unable to pay.

Unlike traditional banks, I don't earn any interest on my loans and must cover all losses from depositor funds.  I also don't have the ability to hide the true state of my balance sheet nor do I have any ability to raise any additional capital on my own.   Everyone that continues to bank with me knows they are playing a game of musical chairs and they will only continue to trust it as long as everyone else does.   It will not last long, especially for a crypto-currency not backed by the full faith and credit of a big bank.

The problem you have is that you need the system to be workable on day one when MC is worth almost nothing and has no reputation.  The first 'panic' or 'market correction' it faces would entirely collapse all of MC because everyone with GoldCoin would lose and once that is revealed all other currencies (not backed by a public issuer violating bearer bonds laws) would also make a rush to safety.  

I just realized that this very intelligent question came from SPIRAL_MIND!

I think maybe he's just messing with me now. Is this really the same spiral_mind?

When a target currency is above target, that means that people want that currency so badly that they are willing to pay a small premium for it, which the escrow fund collects.

When a target currency is below target, that means that people want OUT of the currency so badly that they are willing to pay a small premium to get out, which the escrow fund also collects.

It would be irrational to give that premium to the escrow fund if there was another choice, but there is not.

A good counter-example is fractional reserve banking. Like MasterCoin, banks which don't have enough money on hand to cover all deposits are inherently unstable and might crash. On the other hand, like MasterCoin they also have the potential to keep going for many years without crashing.

Banks fundamentally violate this law, and take on risk as a result. MasterCoin makes similar trade-offs.

Right but people are inherently self interested in economics. They won't sell to the escrow unless they can make money from doing so. They won't buy from the escrow unless that also makes them money. So how does the escrow ever make any money unless people are willing to sell at a loss?

I certainly hope you aren't claiming to have changed the laws of economics and made people fundamentally not profit motivated?? Can you address this problem? How do you get people to sell at a loss (which is what is required for the escrow to gain money).

Have you factored in that your built in exchange will have an order of magnitude more trx volume than the Bitcoin block chain?   Looking at the size of one message is totally missing the point and assuming bitcoin-level trx volume when people are actively playing the market is also a very unsafe assumption.

A conservative guess is that there would be 10 market transactions for every traditional bitcoin transaction.  

The escrow fund buys coins below the target, and sells them above the target, making a profit. I haven't seen anybody post a convincing argument that this is not true (as long as the escrow fund is reasonably healthy).

I did address it the first time it came up. The footprint of MasterCoin is very small in the block chain, and consists mostly of transfers of value. However, it is true that MasterCoin transactions do contribute to block-chain bloat just like other bitcoin transactions do. The total cost per MasterCoin transaction is about 2 cents, which is obviously not going to interfere with adoption.