Topic: *old* BitShare Economic Theory 10 BTC bounty to prove me wrong... paid. - page 11. (Read 10080 times)

I have a much improved white paper published on google-doc available for viewing:

https://docs.google.com/document/d/1FqDgA7J_O2hufetbNY51xue7yE57Euk2_nTa5IOhnJA/edit?usp=sharing

Granted it is still in draft, unformatted form but it takes a lot of effort to explain complex economic relationships in understandable and 'provable' terms.  The reality is it gets simpler the more I explain it.  


Proposal:

In this paper I present a new crypto-currency with aim of supporting the creation of many sub-currencies that closely track the value of any other item in the market without the need to any issuers.  An analogy can be made to a distributed peer-to-peer bank that accepts and pays interest on deposits in any currency.  This bank operates without counterparty-risk or IOUs.

This currency will have the following Properties:

• The native currency will be called a share and is mined into existence on the same schedule as Bitcoin.
• Shares will pay dividends from half of the mining fees and rewards
• Users may issue new sub-currencies by ‘shorting the sub currency’ and backing the short position with dividend payments from a defined number of their Shares.
• Owners of sub-currencies will receive the dividends from the all Shares used to create them proportional to their balances.
• Short positions can only be redeemed by the issuer.
• Users will be able to trade among all shares and currencies and via a built-in exchange.
• All shares and sub-currencies will have the same transaction properties as bitcoins.  

Economic Effects of this new Currency:
      1.  Shares derive their value from the same sources as Bitcoins.
       2.  Because shares pay dividends the shorts effectively pay interest to the longs.  This interest payment is what regulates the supply and demand for deposits and withdraws and ultimately maintains price parity.  

How does crypto-USD get created?
   Someone must go short crypto-USD and back their short position with interest bearing BitShares.

Does it cost them anything to maintain a short position?
  Yes, the act of going short causes the interest rate of crypto-USD to go up relative to the interest rate of  Bitshares  AND the individual who is short crypto-USD forgoes interest payments on their Bitshares.  

Why does the interest rate of crypto-USD go up relative to the interest rate of Bitshares when someone shorts?

    Because the individual short-selling crypto-USD is taking an exchange rate risk, he is unlikely to do so at the current market price.   After all, he will have to buy back crypto-USD in order to recover his bitshares (which he expects to go up in value).   So, he only goes short at a strike-price below the current exchange rate with enough margin to cover his risks.   If USD goes up then it will be more costly to cover his short position and he will lose money.  The risk of USD going up is equal to the risk of BitShares going down.

     Because the exchange occurred at below market rates, the dividend payments which are in Bitshares are necessarily valued at ‘above market rates’.  In other words, the percent gain in interest rates is proportional to the margin built into the short position.

What impact does the change in relative interest rates have on the market?

    Higher interest rates always attract more depositors which will drive up the price of crypto-USD.  This works against the short-seller who needs the price of crypto-USD to go down relative to Bitshares or else he will make a loss.  This is part of the reason the individual short crypto-USD must maintain margin.

    Higher interest rates also trigger other Bitshare holders to consider selling their Bitshares for crypto-USD.  By swapping their savings they may yield a higher return.  This also serves to drive up the price of crypto-USD (and discourage the short-seller and necessitates caution when issuing new crypto-USD).

If the change in interest rates brings in more depositors AND causes existing bitshare holders to switch to crypto-USD for savings, what market forces bring the price of crypto-USD back down?

    Holders of crypto-USD may see the potential for equity gains in bitshares outweigh the difference in interest rate.   In this case they would take the same position as the short-seller in believing bitshares will appreciate.

    Holders of crypto-USD may wish to withdraw their money into paper-USD.  


What happens if there is a run on the bank and everyone wants to redeem their crypto-USD?

If Bitshares lose all value, then crypto-USD no longer pays interest of any value whatsoever and the short-seller has nothing to gain.  Everyone involved would lose everything they deposited with the bank.  This scenario is unlikely so long as bitshares continue to provide a valuable service as a medium of exchange and the underlying cryptography, internet, and power systems remain available *somewhere* in the world.  Thus any depositor who puts money into crypto-USD is implicitly trusting that the underlying Backing of crypto-USD (bitshare dividends) will not go to 0.  

Assuming bitshare do to not go to 0, then crypto-USD will have value to all short-sellers.     crypto-USD will also have value to anyone who expects the bank-run to pass and for depositors to return at some point in the future.   In this case, speculators will start buying up crypto-USD at below face value prices and short-sellers will start covering their positions taking crypto-USD out of circulation.  These factors will conspire to provide support to the crypto-USD price.

In a panic the rush to withdraw will push the paper-USD to crypto-USD price down which is economically identical to increasing the interest rate paid to new depositors in crypto-USD.  The increased interest rate will cause cash to flood into the system to meet the demand for withdrawals.  

The depressed paper-USD to crypto-USD price will also discourage many people from trying to withdraw during the run.  They decided to ride out the storm rather than take a loss in a panic.

So the question becomes what would trigger a run on the bank?  Because there are no IOUs involved and all crypto-USD are backed by real value that does not have counterparty-risk (bitshares) then it is unclear what could cause such a rush short of something threatening the very infrastructure (internet/power) that the system is built on.  

What happens if someone dies before they close out a short position?

Then there will be a permanent bias in interest rate in favor of owning crypto-USD over bitshares.

What happens if Bitshares fall in value by 99%?

Assuming the 99% fall in value is not part of a complete crash to 0 which I have already covered, then I will assume that the Bitshares found support at 1% of their high.

Then the value of interest payments paid to holders of crypto-USD will fall by 99% while the interest rate paid to holders of Bitshares will remain at 10%.   As a result those who are seeking a 10% return will sell their crypto-USD and buy bitshares.  This will depress the price of crypto-USD and help support the price of Bitshares.   Short-sellers (crypto-USD issuers) would be taking it in the shorts so they would be actively redeeming as much crypto-USD as possible as soon as possible to cut their losses.  

Some short sellers may not be able to redeem their positions.  In which case the effect would be the same as if the short seller had died and there would be a permanent interest rate bias in favor of owning crypto-USD vs owning bitshares.   Once Bitshares reach a new equilibrium after the market adjustment, then this interest-rate bias will encourage others to redeem the crypto-USD for paper-USD.

How will the block-chain algorithm determine what exchange rates to use when issuing Crypto-USD?

The blockchain will not have to decide, market participants will estimate the risks of going short to acquire crypto-USD.  Therefore the price will be decided by the user in an intentional act to issue Crypto-USD.    Likewise, these same participants will decide at what price it makes sense to cover their position.

How do you prevent abusing short positions to manipulate the market?

Taking out a short position requires capital of equal value and incurs opportunity cost in foregone interest payments.  These forgone interest payments are what back the value of the short position.  Therefore there are no naked shorts and the shorts have no more power in the market than the longs.  

What prevents people from creating a million different currency types?

First of all, creating a currency type means taking out a short position in that currency which means it incurs a transaction cost.  Second, others must understand and have consensus about the meaning / relationship of that currency type to Bitshares for it to have any value.   Third, the short position must accurately reflect market prices or the issuer will take losses.  If the price too low, the the market will push the price up causing them to lose money on the short position.  If they price it too high, then they will only ever be able to redeem it with themselves and thus would incur 2 transaction fees for wasting everyone’s time.

I'm happy to help; personally I think multi-currency is a key feature in the next generation blockchain (and sub-chains, described in several postings recently, but much earlier here: https://bitcointalksearch.org/topic/m.1019392) .  PM me a google doc link or something if you want to co-author.

I am working on just such a paper and explanation.  All of your questions are exactly what I need to do.  Thanks. 

bytemaster, I have read you reply but do not have time right now for a point by point reply.  But let me ask a few questions and offer one piece of writing advice that I hope will clarify things.

Advice:  you tend to write anecdotal examples and stories and "prime" your and other people's mind to a sub-conscious assumption by using terms like crypto-USD.  In other words, you assume crypto-USD and try to argue backwards why it works.

Instead what you want to do is something like a true white paper:

Proposal:  A crypto currency X with these properties:
1
2
3

Effects:  Combining these properties creates the following interesting effects:
1
2
3:  [Ultimately I think you want to say] Payment of dividends creates a negative feedback loop that drives the price of X in THC (the "native" currency) towards parity with the price of the USD in THC.
This is clear because [now prove it].

Extreme situations:  [Prove the effects work during major market upset]

Result:  crypto currency X will track the value of USD


Questions:

1. How does information about the price of USD in THC enter the system?

2. When I was talking about anyone creating their own currency... you said at one point that "we" will start with just the major currencies.  Who is "we" here?

3. How does crypto-USD enter the system?  I get that someone buys it for USD, but I mean technically how is it actually created?

4. How does your dividend system cause crypto-USD to approach USD value?


And then an observation:

By not letting other people "mint" their own instruments, you tremendously limit the crypto-infrastructure.  Doing so would let:
A company issue its stock or bonds
An individual finance his own mortgage
(and a million variations)

Today, all current crypto-currencies (bitcoin, litecoin, etc) are really the same; they may have minor variations but hold the same fundamental role.  The above allows crypto-currencies to hold roles that interact directly with real life entities, rather then through the indirection of a floating exchange.

Whether an individual instrument is legally issued and enforced is not a problem to be solved by the crypto-infrastructure, just like the same is not solved by the paper these instruments are written on today.  In fact, a crypto-infrastructure really is like digital legal paper, allowing verifiable signatures and near-instant transfer.  But ultimately courts will decide about the legality of issuance and enforcement of a particular instrument.

 

I am trying to follow what you are saying, but think what I have proposed effectively accomplishes the same thing.  Could you explain what problem your solution attempts to address?

The economics section already addresses the hoarding myth promoted by mainstream economics and throughly debunked by austrian economics.  I will not debate this particular issue on this thread.

Could you elaborate? 

No I am trying to achieve parity between crypto-USD and paper-USD by allowing BTC  to float against crypto-USD.

I am trying to allow one block chain to handle crypto-USD, crypto-GOLD, crypto-EUR, and to allow parity to be maintained without counter-party-risk or IOUs.

Thanks for your summary of the field, but from what I can tell these paragraphs are your most direct challenge to the viability of a block-chain based exchange.   So lets see if I can address this very valid and insightful comment.

Not all trades must occur on the blockchain provided the blockchain facilitates the level of trading required to handle currency issuance and redemption.  Currency only has to be issued or redeemed at the edge cases because anyone willing to hold a balance in crypto-USD because it pays dividends would be a cheaper source of liquidity than issuing.  Crypto-USD can be traded on a centralized exchange but is ultimately backed by trades / arbitrage on the block chain.  So you can think of trading crypto-USD on the block chain like having a direct connection to the stock exchange.  It can be expensive... most people just use a broker which can be faster, cheaper, and lump large trades together.   All that really matters for this system to work is that crypto-USD can be created,  spent like BTC, and exchanged between currencies easily enough to maintain near parity.   HFT can AND SHOULD occur off chain.

After all blockchain real-estate is a finite resource and HFT would bid up the transaction fees forcing only bulk trades to occur on-chain and everything else could be resolved off-chain.

Such a thing already exists today, it is called Bitinstant.  They charge a fee for 'instant funds' in Mt. Gox and the reason they can charge that fee is price difference.

Actually, you can set a BitShare price for crypto-USD without having to trade with anyone but yourself.  It would lock in the 'exchange rate' and if you were the ONE AND ONLY issuer and held all crypto-USD then your dividend payment would be EQUAL to holding just BitShares and thus there is NO RISK in placing bids without having $USD in hand because the value you receive is proportional to the DIVIDENDS received which is the same both immediately before and immediately after your bid is accepted.   Now if the price changes while you hold-crypto-UDS then you could lose money due to exchange rate changes between crypto-usd and bitshares, but that is what markets do.

Ver interesting. There are some important elements, especially the dividends idea.

But there is also a lot of confusion here. Honestly, I would recommend to studying how a paper fiat money system and an exchange works first. Its not at all trivial. Did you know that every bank transaction gets wired over the central bank? Every bank holds an account at the central bank. The way banks exchange money is that they use the central banks system. And the central bank is backed by the government.

In any case: think about it - how can you convert 1$ into 0.01 BTC? You have to transfer the 1$ to a place or a person who has BTC and wants $. The only way to transfer $ is through an entity that has an account at the Federal reserve, aka a bank. Or the entity has USD cash, and you can accept cash. Those are the only ways to transfer fiat money without an IOU scheme. Every holder of USD has a counter-party risk against his bank and against the central bank. Which is why Bitcoin is so brilliant in the first place. But the legal frames make a total bootstrap very unlikely IMO. In some ways it implies that nation-states can't tax people. And they don't like that. Compare the AAPL discussion recently and the broken international tax regime.

Compare that with Ripple where you have all kinds of layered couterparty risks. In your personal IOU network and against the network overall. Which it makes it worse than fiat in a way.

First of you need to be much clear of how such an exchange would allow throughput of 100'000 of orders per day. Then you will see the problem. Take the MtGox volume and calculuate the necessary throughput time. Take into account the speed of light (no kidding). The maximum time for execution would be rather in the 50 ms range or lower (in fact the speed of light makes auto-trading across the globe impossible, but auto trading is the base requirement to process most than 10M$/day). Certainly a blockchain makes such an exchange simply impossible. Just do the calculations of execution time that are needed, say based on 100x current volume at MtGox. Imagine BTC would rule the world, then you need up to 10^12 more volume, which is roughly what the current markets handle. (Global money in circulation is very roughy <10 Trillion$ per day, compared to ca. 10 million $ per day at MtGox). Of course nobody would trade significant volume at 1% fee. Bitcoin exchange fees should in theory approach 0.01%.

An exchange works like this. You have a central datastructure, which is the limit order book. Because it is central, by definition, a distrubted block choin simply does not make any sense. An exchange has to be fast, and so a good exchange has to have a synchronous architecture. In fact that is its primary feature, namely liquidity attracts liquidity. If one single trade takes 10 minutes, you simply can't operate.

If you study this you see where Bitcoin differs and how it improves it. But there is a significant problem (not vulnerability) in Bitcoin in terms of potential adaption. And the problem is that Bitcoins has to interface with fiat banks. Which makes the BTC exchanges act as banks. So if I want to use BTC I have to receive money in BTC, be it a wage, or capital income, etc. Otherwise I have to go through the exchange, and the governments of the world have the ability to shutdown the exchanges, as they already have.

However, what you could do is to have gateways which are liquidity providers. These could meet at an efficient exchange. The gateways would handle fiat money and the exchange handles orders.

Say you are in a country, where you don't run the risk of getting jailed for ML, like you would establishing such a system in most developed countries (have you thought about that?). Then you would do this in a TOR like fashion. Doing this I would consider very, very risky though. Even it would work the first customers are going to be people who want to ML. I would suggest you look at B24 and how this has been going. SH now faces serious charges and in part rightly so, because his exchange clearly enabled ML. Which shows some of the big problems with Bitcoin, and why my enthusiasm is now greatly reduced after seeing this. Essentially BTC in part leads in part to what is considered criminal in most jurisdictions. Which at some point there will be massive backlash. B24 and Bitfloor are closed now. MtGox potentially could be closed at any time.

I just now got it... you are trying to achieve parity between usd/btc with the new blockchain, correct?

this encourages hoarding all coins

I believe the amount of coins you are holding should give you more coins from the new block chain (example: 50,000 dev coin gives you more coins from the new block chain than 1 btc)

to critique your idea

What if the alternate block chain verified the transactions of a peer 2 peer decentralized crypto-currency exchange

each transaction would cost the regular fee of 0.2% of whatever currency is being exchanged and those holding said currencies would recieve interest from that

but what if each transaction also cost x of the new coins created by the new block chain

the new block chain i bleieve would act as a fail safe against alot of problems crypto-currencies currently have

also you could integrate some kind of decentralized peer to peer social network with a rating sytem, like a bitcointalk.onion... kinda

it would make lending and borrowing decentralized and earning more coins for everyone holding

Thanks for the Tip... all tips will be funneled into this project (or the bounty).  I look forward to answering any questions you may have resulting from your research!

Hmm, you are starting to make me change my mind. Right now I'm at the point I was at two months ago with Bitcoin. I was intrigued, and thought they might be on to something, and it took me a couple of hours of reading up on the algorithms to convince myself the concept was sound. I'm going to put in those hours again, though I can't promise you I'll reach the same conclusion. And even if I should remain unconvinced, you have certainly opened my eyes to new possibilities. If it works, it's a clever innovation based not so much on new technology, but on a good understanding of (Austrian!) economics. Sent you a small tip for the entertainment value of this alone.

BTW, this reminds me of the unfinished Market.h and Market.cpp that Satoshi accidentally committed and subsequently removed. You are not secretly Satoshi, are you? If so, your cover is blown.

I certainly want to structure things clearer, and the goal of this bounty was to bring out errors in my beliefs so I appreciate your response.


This is not a trust-based issuer situation.  Crypto-USD has a fluctuating market price relative to paper USD that trades in a range similar to the transaction fees paid to move money into or out of Mt. Gox, your ATM, Dwolla, etc.   Because the price fluctuates it is backed by supply and demand and nothing less.  The supply is provided by individuals who see a price difference where crypto-USD is worth MORE THAN paper-USD.  This price difference must be sufficient to motivate them to give up the DIVIDENDS they are receiving on their BitShares (thecoin).  BitShares + their dividends have value for the same reasons bitcoins have value.  Thus new crypto-USD is created increasing the supply and driving the price back toward parity.

Crypto-USD is redeemed when the price of crypto-USD is LESS THAN paper-USD.  In this case anyone who issued Crypto-USD can buy crypto-USD with paper-USD at a price below face value and then use crypto-USD to redeem their BitShares held as collateral and start receiving BitShare DIVIDENDS again.  Thus there is profit to be made by redeeming shares and no need to 'trust' anything but the markets profit seeking initiative.    

As a result there is never any IOU in this system.  As all 'crypto-USD to BitShare' exchanges occur on the block chain and all issuance of crypto-USD occurs via a Bid transaction there will be a definite and changing ratio between crypto-USD and BitShares but a relative stable ratio between crypto-USD and paper-USD.


First of all, all crypto-USD is backed by dividend payments paid in BitShares.   The source of the dividend payments on USDs comes from the BitShares used by the system to issue the crypto-USD in the first place.  The source of BitShare dividends comes from half of the mining fees + rewards.  Thus if BitShares have value like Bitcoin then what is backing the crypto-USD is the future DIVIDEND earnings on the Bitshares held as collateral.   The ratio of dividends between crypto-USD and BitShare balances is proportional to the exchange rate.



I think that with the explanations above this statement is clearly 'wrong' as crypto-USD is not an IOU *issued* by anyone.   It is not a promise any more than a bitcoin is a promise.  

Ok, now that we've got that out of the way, lets structure the system.



Yes I was thinking of something similar for defining a new coin-type.  The problem is the network requires an exchange rate to be established before THC can be used to back the coin-type.   So I figured that a new coin-type would be allowed once a certain THC value in open bids for that coin were published on the network.  These bids would be backed by THC (until they were canceled).  With a high enough value threshold there should be enough 'volume' to get a market started and thus the network would allow the creation of the new coin-type in response to the highest available bid and work its way down.  Publishing all of these 'bids' and keeping them open would require transaction fees, time, and locking up of real value and therefore could not be spammed.

As a result there is no need for a new coin-type to be associated with any particular private key.  To accumulate sufficient bids those placing the bids would have to know what it was they were bidding on and what it represented.   For version 1.0 I would probably pre-define the set of national currencies + popular crypto-currencies + gold and silver.   Only after the concept was well understood would I enable the feature to create arbitrary coin types (such as company stocks, oil, etc).

Coins are not issued nor are they IOUs.  They are their own free-floating currencies which are issued in response to a bid on the market and thus are issued in proportion to the value relative to THC.  Of couse this relative value is constantly changing.  The block chain knows the sum total of all THC to OtherCoin issuance and thus knows how much of the THC dividends should be paid to OtherCoin balances and in what proportion.   Thus the DIVIDEND rate as a PERCENT OF VALUE is the same across all currencies.

Therefore if THC were BTC and there was a bid to buy 125 Crypto-USD for 1 BTC then that bid could be satisfied by issuing 125 Crypto-USD backed by the DIVIDENDS paid on 1 BTC.    Later BTC goes up in value and
there is another bid to buy 150 Crypto-USD for 1 BTC.  In this case 150 Crypto-USD would be issued backed by just 1 BTC.  Over time the all Crypto-USD currently issued will yield DIVIDENDS of value at about the average recent exchange rate.  Any differences in the value of the DIVIDENDS in various crypto-currencies would be taken out via arbitrage as people trade between Crypto-USD and THC.

Now the purpose of issuing a Crypto-USD is because the buyer who is paying with paper-USD wants to avoid exchange rate risk.  So he is issued a currency that pays DIVIDENDS of value proportional to the exchange rate.  Market forces will work to keep this as close as possible.   He knows he can then redeem the crypto-USD from the 'THC believers' at the same value regardless of exchange rate fluctuations because what he is really redeeming is a THC dividend bearing bond.




First of all a bid can be broadcast and matched before even getting into the blockchain.  The only bids that would be in the block chain are 'outstanding bids'.   Second, transactions can be pruned.  Third, eventually they do get in the block chain as part of a transaction.  Fourth, a bid with no transaction fee (or low fee) will be broadcast but not included in the chain until a matching transaction was made.    Fifth, the block chain needs to know about bids / asks to establish the exchange rate for issuing currencies.   That said, nothing stops out-of-chain exchanges from being setup like Mt. Gox for those who want to do HFT.  Mt. Gox would be able to operate using this system without having to take deposits and therefore could even operate behind a TOR hidden service.   Thus, let 3rd party dedicated services handle the high-frequency stuff and let the block-chain handle 'trustless' exchanges that might be slightly slower but still not that bad for most peoples purposes.  

All of that said, another way to create a bid is as an INPUT to any standard output that allows it to be spent as part of a valid exchange provided the bid is signed by someone allowed to spend the output.  Therefore the bid could be broadcast without actually creating a new output type.  Spam on the bids could be controlled by proof-of-work applied to the bid.   Bids can now be broadcast/cached and only people accepting the bids would include them as part of a transaction.



Once again, thank you for your response, you clearly took time to think about things and respond thoughtfully.