SERVING THE QUANTITATIVE FINANCE COMMUNITY

 
User avatar
Traden4Alpha
Posts: 23951
Joined: September 20th, 2002, 8:30 pm

Re: End of bitcoin?

December 17th, 2017, 11:52 pm

Paul wrote:
How about a takeover, bitcoin buys out a 'better' cryptocurrency or vice versa. Bitcoin brand + better platform. You all see this as technology, I see it as a business. (Mainly because the technology is gibberish to me!)
It won't happen because it's entirely against the financial interests of the miners who maintain the blockchain and profit from its congestion and transaction fees. Bitcoin is a business and it's the miners' business.
 
User avatar
outrun
Posts: 4573
Joined: April 29th, 2016, 1:40 pm

Re: End of bitcoin?

December 18th, 2017, 9:11 am

Traden4Alpha wrote:
outrun wrote:
There are two different aspects:

* lack of liquidity during a crisis (no one willing to catch the knife)
* lack of market access.

I think both aren't much affected by the lack of transaction speed on the blockchain? E.g. suppose you want to quickly sell your bitcoin, ..

..1 ) but there is no-one there to buy it? That has nothing to do with exchanges or transaction speed. It's because potential buyers are having lunch, playing the tetris, aren't really interested no more in buying a bitcoin.

I'd think that the relative amount of interest in eating lunch versus buying bitcoins would be a strong function of how easily one can use bitcoins or sell bitcoins.  That, in turn, depends on the frictions in the transaction system.  Would you put your money in a bank that had ATM fees of €5 on some days but maybe €20 other days or that maybe you have to wait a hour for your money or maybe you have to wait 3 days to find out that your transaction never went through?  Transaction friction directly affects liquidity and liquidity (relative to holding times) directly affects price.

outrun wrote:
.. 2) You want to dump it at $45.000, just 5% of the value of the previous day,.. someone else really want's to buy it for $45.000 here at the  W forum, .. and so you agree to exchange the money and promise to move the bitcoin to his wallet when there is a transaction slot in the blockchain. You can write a simple contract.

I don't think thats how it works, transaction slots don't come up to you like a taxi and then you put in whatever order you want.  Instead, you submit your specific pre-defined irrevocable order with a pre-defined counterparty into an irrevocable queue.

that sound like a solid defined moment & fact of transfer of ownership of the coin leg of a trade agreement.
If the price of bitcoin falls while the order is pending, there's nothing the buyer can do except breach their side of the contract (or hope the blockchain order never gets confirmed).

The price of bitcoin doesn't exist as a global property, and coins can be transfered like any other type of transactions people do.

1. you can trade bitcoins without a price, I can donate bitcoint, no need to exchange them for dollars.
2. there are various exchanges where you can swap bitcoins for other currencies, different exchanges have different prices -not because there is no high freq arbitrage trading going on- but because each exchange has it's own credit risk: they can get looted and go bankrupt. That credit risk ends up in the price because before you can trade you will have to transfer funds to that exchange. This is very much like banks.
3. when you trade at an exchange there is no "price", when you want to sell you can look at bids and different exchanges will have different bids. This is very much the ame as you see with ECN exchanges.
4. you can trade with a paper contract and an invoice! You pay the invoice and 2 days later the money you've send to my bank account will arrive.
But lets say you are still willing to dump bitcoin for 5% of the previous day's price.  Why wouldn't the potential buyer think that they can wait an hour and get the order for 4% of the previous day's price?

When you trade with someone and agree on a price, you agree to some sort of contract that seals the deal. You either trade at 5% or not, .. you don't have an option to wait for another hour and change the price of your trade. E.g. if I would send you an invoice for 10 Euro based on some service I delivered as promissed then that's what you owe me. If you wanted to hedge that you have had to do it the minute you agreed to the deal were I would do a service and you would pay for it, that's the timestamp of our trade. The fact that you will pay later, and that the banks will hold the transaction in a twilightzone for two days doesn't change out agreement.
And to the extent that your order is public (which it needs to be if you actually want to find a decent buyer), then why don't all the other would be buyers of bitcoin revise their bids to 4.99% of the previous day's price?

But worse: if the buyer thinks the price is falling 16% per hour (i.e., a 95% drop in price per 14 hours) and they know that bitcoin's slow transaction system commits them to owning those bitcoins for at least 1 hour and possibly as long as several days, then why wouldn't all buyers wait on the sidelines for the price to stabilize?

[/quote]
You don't need settlement in the bitcoin blockchain to agree on a price or exchange ownership of coins vs dollars, and an agreement to exchange bitcoins for dollars is binding, you can't just pull out. Once you agree to a a trade at an exchange, the exchange will make sure that payment are done and coins exchanged. Fund and coins will imediately be transfered (un)locking the ability to immdiately trade once more at the exchange. The exchange keeps track of ownership, and that will materialize in the blockchain when you take the coins out of your exchange account.

However, nothing is stopping you to get drunk with a friend X and write on a matchbox "we agreed that t4a wil buy  from X 1`bitcoin for $19.000, friend X will transfer is within a week, t4a will move the money to an escrow account, bla bla bla more legal clauses". Now suppose you wake up the next day and the price of the bitcoin is $30.000. Your friend "can't rember this agreement you two make last night". Now what?
 
User avatar
outrun
Posts: 4573
Joined: April 29th, 2016, 1:40 pm

Re: End of bitcoin?

December 18th, 2017, 9:33 am

 ..and what about this scenario that shows that the liquidity problem is not unique to bitcoin.

I sign a contract where I saw "I'm going to give T4A one stock of IBM this very moment". You don't like that stock and you want to sell it in the next hour because you feel there is a crash comming. Would you be able to do so? I don't think so, even though you would legally own one as I promised. The reason is that in order to echange your stock for USD you would need to (indirectly via your bank or broker) be a member of an exchange.

This is interesting: *could* I even on paper promise you an IBM stock?? How is ownership transferred? In order to give it to you I would either have to have one, or buy one. But who is keeping track of ownership?
 
User avatar
Traden4Alpha
Posts: 23951
Joined: September 20th, 2002, 8:30 pm

Re: End of bitcoin?

December 18th, 2017, 2:24 pm

A contract is only as good as the probability of execution which then depends on the probability that both counterparties are both willing and able to execute that contract. Counterparties may become unable to execute their side of the contract through various physical, financial, or regulatory failure mechanisms. Counterparties may also become unwilling to execute their half of the contract if the net costs to the counterparty for completion of the deal exceed the net costs incurred by breaching the contract. Under some conditions (and with some counterparties), one counterparty or the other might willingly breach the contract. The chance of a breach would seem to be a function of the social connections or the anonymity of the parties, the prevailing legal systems of the counterparties' countries, the criminality of the counterparties, and the volatility of the system by which the favorability of the deal might change drastically during the contract execution process. During a bitcoin crash, I'd wager that the chance of the counterparty being both willing and able to execute a contract will drop.

The exchanges certainly do reduce the chance of a breach of contract. By taking custody of both the bitcoins and the cash, they become the virtual counterparties to BOTH sides of the transaction but with no exposure to the volatility of the value of either the bitcoins or cash. The exchanges essentially eliminate the counterparties' freedom to breach the contract which is a very attractive property that induces most people to seek on-exchange transactions over off-exchange ones (and the willingness of a counterparty to seek off-exchange transactions is a signal that they are more likely to breach the contract!) Yet the exchanges can still fail (the credit risk you mentioned), bitcoin holders still face the 4 transaction/second limits + confirmation delays + blockchain transaction costs in accessing the exchange, and exchanges are still limited by the balance of participants willing to buy bitcoins versus sell bitcoins (unless the exchange "borrows" participants' bitcoins and cash to make the market).

As you note, the blockchain does not record the price or terms of other side of the deal. Only one side of a bitcoin deal is secured by the properties of the blockchain. An off-exchange sender of a bitcoin really is in a precarious situation. They are being asked to drop their bitcoin into a cryptographic hole from which it can never be retrieved without the cooperation of the counterparty. Moreover exactly when it goes into that hole is not very clear.

The timestamp of a bitcoin transfer on the blockchain is indeterminate. It's certainly NOT the timestamp of the order submission because until the order is confirmed by several miners, it has not happened as far as the blockchain is concerned. And although the sender cannot revoke a submitted order, they can change the order to redirect the bitcoin to a different wallet (including one they control). (Note: they will incur a transaction fee for this and those coins will be in limbo until the replacement order is confirmed.) Similarly, the off-exchange recipient cannot revoke the order although they apparently can pay extra to expedite it.

P.S. If you promised me IBM stock, then I could unwind that transaction within seconds by shorting it. The US markets are extremely liquid with transaction capacities in the many tens of thousands of transactions per second and have an extremely tiny percentage of time they are constrained due to volatility. But if someone I don't necessarily trust promises me IBM stock, then I can't perfectly unwind the transaction until it is complete because maybe the stock will appear and maybe it won't.

P.P.S. If the resources of one's contractual counterparty are, in large part, defined by pending contracts with still other counterparties, then the chance of contract failure becomes chained across all the pending contracts. That induces a high correlation in the chance of failure and the distribution of outcomes becomes deeply bimodal (nearly all contracts succeed XOR nearly all contracts fail). That effect does not seem to play a role in the bitcoin universe (yet!) but it was at the heart of the 2008 financial crisis.
 
User avatar
Alan
Topic Author
Posts: 9536
Joined: December 19th, 2001, 4:01 am
Location: California
Contact:

Re: End of bitcoin?

December 20th, 2017, 5:03 am

 
User avatar
Alan
Topic Author
Posts: 9536
Joined: December 19th, 2001, 4:01 am
Location: California
Contact:

Re: End of bitcoin?

December 22nd, 2017, 2:35 pm

Feels like crash action at GDAX even by bitcoin standards. Also, the absurd premium on GBTC was wiped out overnight. 
 
User avatar
outrun
Posts: 4573
Joined: April 29th, 2016, 1:40 pm

Re: End of bitcoin?

December 22nd, 2017, 3:05 pm

There goes thr 4 day profit, we're back to last Monday levels!

I'm collecting orderbook data and the movement and waves you see are exactly the same like any other tradable asset. It's very interesting​ from a behavioural point of view. It has very little economic links like eg stocks have, use more pure in a sense.
 
User avatar
Traden4Alpha
Posts: 23951
Joined: September 20th, 2002, 8:30 pm

Re: End of bitcoin?

December 22nd, 2017, 4:21 pm

Indeed! The volatility is almost 100% endogenous dynamics.

What's fascinating is that Satoshi's algorithm creates a condition of artificial scarcity with strictly bounded supply. Unlike gold, for example, the increased price of bitcoins cannot somehow spur increased investment in mining to create increased supplies of bitcoins to match the demand and stabilize the price*. That seems to imply the price can only go higher. Yet although the total global supply of bitcoins is strictly bounded by Satoshi's algorithm, the marginal supply to the markets is not. Satoshi's algorithm strictly prohibits production of millions of new bitcoins to satisfy high demand but there's nothing to stop millions of existing bitcoins from flooding the market in a crash.

I look at the system in terms of elasticity of supply and demand. For each 1% move in price, what is the change in supply (% of bitcoin owners who now want to sell) versus the change in demand (% of cash owners who now want to buy bitcoins). Given that there's only about $150 billion in bitcoins but there's literally trillions of dollars in potential cash in the world, the % of buyers does not need to be very large to totally imbalance the market with buy orders and drive the price higher.

Elasticity is, in turn, a function of context and perceptions. As long as bitcoin's price is perceived as increasing, the market will be extremely imbalanced because very few bitcoin owners will want to sell their "excellent" investment and a great many cash owners will want to buy bitcoin. Yet what happens if perceptions slip the other way and bitcoin is not perceived as "a sure thing"? Supply will explode and demand will implode. It really is a perfectly cyclical system.

-----
*Note: Although increased bitcoin prices cannot spur increased bitcoin production, it can spur increased introduction of new cryptocurrencies.
 
User avatar
outrun
Posts: 4573
Joined: April 29th, 2016, 1:40 pm

Re: End of bitcoin?

December 22nd, 2017, 4:38 pm

I wonder what happens (soon) when the price drops below the marginal cost of mining (electricty). A mining rig can be seen as a real option (you can shut it down and reduce costs) and the value of options depends a lot on volatility. At the same time, the hashrate is ever increasing because mining rigs keep getting produced. The discount curve is strong with the bitcoin!
 
User avatar
Traden4Alpha
Posts: 23951
Joined: September 20th, 2002, 8:30 pm

Re: End of bitcoin?

December 22nd, 2017, 6:14 pm

That may have already happened.  I saw an article that alluded to Satoshi's time-decreasing function on the production rate of new bitcoins and that it has already made them too rare relative to the cost of mining.  Miners now depend, in part, on transaction fees.  At least that is the dynamic that must occur as the algorithm-defined rate of coin production drops.

The effects of a crash will be quite interesting.  Although the value of the coins will plummet, the transaction fee will surge as owners of bitcoin desperately try to dump their bitcoins on greater fools and bottom feeders.  During the crash, the miners will earn huge revenues.  But afterwards I'd think that transaction fees will drop, too.

What I don't understand about the bitcoin system is how transaction fees distribute across the set of miners.  It can't be winner-take-all with the first miner getting everything because there must be some incentives to generate multiple confirmations.  The sharing of the transaction fee surely influences the equilibrium size of the pool of miners.  But how?
 
User avatar
outrun
Posts: 4573
Joined: April 29th, 2016, 1:40 pm

Re: End of bitcoin?

December 22nd, 2017, 7:10 pm

Hmm, if transaction cost skyrocket then nobody will trade.. or will it always be less than eg 1%? Still a lot, but with this volatility ...
 
User avatar
Traden4Alpha
Posts: 23951
Joined: September 20th, 2002, 8:30 pm

Re: End of bitcoin?

December 22nd, 2017, 7:46 pm

Bitcoin transaction fees are not a percentage of the value. Instead, they are a fixed number tied to the size of the transaction data object which is invariant with the transaction size.

A person trying to quickly sell 100 bitcoins might readily bid $10,000 for the transaction fee (i.e., < 1% at current bitcoin prices). In a crash, all the largest transactions will go through first because those are the transactions that justify very high bids to get to the head of the mempool queue.
 
User avatar
outrun
Posts: 4573
Joined: April 29th, 2016, 1:40 pm

Re: End of bitcoin?

December 22nd, 2017, 7:57 pm

So you get paid in bitcoin (because you can send bitcoins between wallets without any exchange)?

And there is queue of pending transactions somewhere with transaction fee bids?
 
User avatar
Traden4Alpha
Posts: 23951
Joined: September 20th, 2002, 8:30 pm

Re: End of bitcoin?

December 22nd, 2017, 8:10 pm

Yes, if you want to send bitcoins between wallets, you submit an order to the system which distributes it to all the miners. Each miner has a mempool of pending transactions (with associated bids denominated in bitcoins). I'm not sure if these mempools are visible, though. If the transaction is not getting confirmation fast enough, the coin's sender can up their transaction bid or the coin's recipient can also expedite the transaction. It doesn't look like either party can cancel a transaction although the sender can submit a new transaction involving the same specific coin but sent to a different wallet address and with a higher transaction fee. The second order will probably go through first which will invalidate the earlier stalled order.

It does seem that there's some visibility on to the prevailing transaction fee bids that are clearing the market but whether one check one's position in the current queue is not clear.
 
User avatar
Paul
Posts: 8589
Joined: July 20th, 2001, 3:28 pm

Re: End of bitcoin?

December 22nd, 2017, 8:54 pm

The Winkelvoss twins are no longer billionaires?
ABOUT WILMOTT

PW by JB

Wilmott.com has been "Serving the Quantitative Finance Community" since 2001. Continued...


JOBS BOARD

JOBS BOARD

Looking for a quant job, risk, algo trading,...? Browse jobs here...