The Mt Gox Master Plan

[Tasty Champa]

especially with the 1.3% Mt. Gox was putting on each trade during the entire operating weekend.

[Torminalis]

You know what the main problem with this Bitcoin forum is?

It is the number of paranoid fantasists that are attracted to it. I can't wait until more normal, rational folk get involved.

[Klestin]

I couldn't care less whether an Mt.Gox associate, or Mt.Gox itself, had an account on the site.  I DO care if this was the compromised account. The have a clear conflict of interest in the decision to roll back all the associated trades when it's their money being lost.

For future reference, at what point will the theft of BTC, and subsequent sale, result in a rollback?  Is it a size threshold?  A market impact threshold?  Or simply a question of who owns the account?

[andes]

So by your judgement, insider trading benefits the market generally! Ha ha...

Yes, it benefits the market generally. It may harm specific groups of people, particularly when it's a violation of a confidentiality agreement, but it benefits the market generally.

For example, when a CEO trades against his own stock, this hurts his stockholders, and he has a legal obligation to protect their interests. But it benefits the market generally by appropriately reducing the value of the stock as quickly as possible.

If you believe otherwise, feel free to present an argument.

You previously wrote "Hidden information is very bad for a market.". And I agree. So consider this: a quasi monopolical exchange that has access to information that the rest of the market has not, is in a position to use this information to gain profits from the use of this hidden information. This is in effect a continous transfer of wealth from the rest of the market to the controller of the exchange. This wealth transfer is not due to their useful work as an exchange (0.65% fee) but due to hidden information they only have access to. This hidden information does not benefit the rest of the market. Opennes of information would benefit the market as a whole, beacuse we would not have this transfer of wealth due to monopolization of information. This of course is only one of the drawbacks of this private closed monopolistic exchange.

You have to take into account that the natural tendency of a global market like this would be to have only one big exchange, as everybody wants to trade where the biggest market is. So this is a feedback loop that kills the small exchanges and feeds the biggest. This will in the end give too much power to the exchange and would endanger the original goal of bitcoin of being a descentralized and free currency.

As long as bitcoin is mainly a speculative commodity that depends on exchanges, this threat is very real. Once bitcoin (hopefully) becomes mainly a currency/money that you can buy most things with, exchanges would loose power as people would not need to exchange bitcoins for dollars or other currencies. But I dont see that the world will widely accept bitcoins for most goods and services soon, this cultural change will take years or decades. In the meantime, lets not allow exchanges accumulate too much power over bitcoin.

For now, exchanges are almost the only interface between bitcon and the real economy, this is too much responsability and power to be in the hands of one private closed quasi monopolistic exchange. We need to find a better solution than honours the vision of bitcoin as free descentralized currency.

[andes]

(created by mistake)

[JoelKatz]

You previously wrote "Hidden information is very bad for a market.". And I agree. So consider this: a quasi monopolical exchange that has access to information that the rest of the market has not, is in a position to use this information to gain profits from the use of this hidden information. This is in effect a continous transfer of wealth from the rest of the market to the controller of the exchange. This wealth transfer is not due to their useful work as an exchange (0.65% fee) but due to hidden information they only have access to. This hidden information does not benefit the rest of the market. Opennes of information would benefit the market as a whole, beacuse we would not have this transfer of wealth due to monopolization of information. This of course is only one of the drawbacks of this private closed monopolistic exchange.

This pretty much can't happen. If this did happen, I could open a competing exchange and charge a 0.1% fee. Their monopoly would evaporate rapidly, and I could take over that continuous transfer of wealth.

You have to take into account that the natural tendency of a global market like this would be to have only one big exchange, as everybody wants to trade where the biggest market is. So this is a feedback loop that kills the small exchanges and feeds the biggest. This will in the end give too much power to the exchange and would endanger the original goal of bitcoin of being a descentralized and free currency.

If they can get a better rate on a smaller exchange, why would they prefer the bigger one?

[andes]

If they can get a better rate on a smaller exchange, why would they prefer the bigger one?

Why would a smaller Exchange be able to offer much smaller rates that a bigger one? There is a concept called economies of scale. Moreover if the bigger exchange makes profits from insider info and volume, there is even less chance a smaller one will be able to compete in rates.

And all else being equal, I would allways prefer being in a big exchange, because there are much more oportunities for supply to meet demand. Price wise and volume wise.

[andes]

You previously wrote "Hidden information is very bad for a market.". And I agree. So consider this: a quasi monopolical exchange that has access to information that the rest of the market has not, is in a position to use this information to gain profits from the use of this hidden information. This is in effect a continous transfer of wealth from the rest of the market to the controller of the exchange. This wealth transfer is not due to their useful work as an exchange (0.65% fee) but due to hidden information they only have access to. This hidden information does not benefit the rest of the market. Opennes of information would benefit the market as a whole, beacuse we would not have this transfer of wealth due to monopolization of information. This of course is only one of the drawbacks of this private closed monopolistic exchange.

This pretty much can't happen. If this did happen, I could open a competing exchange and charge a 0.1% fee. Their monopoly would evaporate rapidly, and I could take over that continuous transfer of wealth.

"This pretty much can't happen.". What do you mean? The quasi monopolistic situation? You must be kidding.

[JoelKatz]

"This pretty much can't happen.". What do you mean? The quasi monopolistic situation? You must be kidding.

No, I mean the quasi-monopolistic situation where the exchange is profiting significantly from its own inside information. You can have a monopolistic situation if, for example, the exchange cuts their rates so low, operates so smoothly and reliably, and makes so little hidden money that there's no point in bothering to compete with them. But if they were making huge profits, the incentive to undercut them would likewise be huge. (Assuming no significant barriers to entry.)

[andes]

"This pretty much can't happen.". What do you mean? The quasi monopolistic situation? You must be kidding.

No, I mean the quasi-monopolistic situation where the exchange is profiting significantly from its own inside information. You can have a monopolistic situation if, for example, the exchange cuts their rates so low, operates so smoothly and reliably, and makes so little hidden money that there's no point in bothering to compete with them. But if they were making huge profits, the incentive to undercut them would likewise be huge. (Assuming no significant barriers to entry.)

The barriers of entry are indeed huge, beacuse everybody wants to be in the biggest market. You cannot suddenly bring everybody to the other market. Also when potential competitors make a cost benefit analisys they would probably not be able to take into account the profits for using hidden information. This is by definition hidden data.

Are you from the Chicago School of economics?

[JoelKatz]

Why would a smaller Exchange be able to offer much smaller rates that a bigger one?

It wouldn't. That's why the hypothetical of the large exchange raking in dough on insider information is implausible. If the hypothetical were true, as I argued, a smaller exchange would be able to offer much lower rates.

There is a concept called economies of scale. Moreover if the bigger exchange makes profits from insider info and volume, there is even less chance a smaller one will be able to compete in rates.

No, not at all. The higher the profit margin, regardless of the source, the easier for people to enter the field.

If McDonald's was barely making money, you'd have a hard time competing with them. You'd have to be almost as efficient as they are just to break even. But if McDonald's was raking in the dough hand over fist, you could be profitable even if you couldn't match their economies of scale.

And all else being equal, I would allways prefer being in a big exchange, because there are much more oportunities for supply to meet demand. Price wise and volume wise.

Right, but in the hypothetical, all else is not equal. In the hypothetical, the big exchange is drastically overcharging and raking in money hand over fist while the small exchange only needs to make a reasonable profit (though would prefer to make more if it could, of course).

[JoelKatz]

The barriers of entry are indeed huge, beacuse everybody wants to be in the biggest market. You cannot suddenly bring everybody to the other market.

I don't follow why people would want to be in the biggest market. If they're trading BTC for USD, they want to be in whatever market pays them the most USD for their BTC (or failing that, the lowest transaction cost), assuming it can accommodate their volume. You may not get the biggest trades at first.

Also when potential competitors make a cost benefit analisys they would probably not be able to take into account the profits for using hidden information. This is by definition hidden data.

You are saying the risk of these hidden profits is enough that we should design exchanges and policies based on it, otherwise this whole discussion is mental masturbation. Then you are saying we won't be able to know that they exist with enough certainty to risk money on it. You can't have it both ways.

Either the claim that there could be massive hidden profit can be supported by facts or it cannot. If it cannot, then why are we discussing it? Let's worry about aliens taking over bitcoins -- that also can't be supported by any facts.

[danglybits]

I prototyped an exchange before abandoning the idea due to legal concerns.   My plan to 'compete' with Mt. Gox was to 'auto-magically' arbitrage between our two sites anytime the price difference was greater than Mt. Gox's fee.  I would charge a much lower fee to cause people to switch to my site, but make money on any price differences between the two sites.  This would give the new, 'smaller' site a larger 'volume potential' than Mt. Gox because it would represent the sum total of all of my customers plus all of Mt. Gox's. 

In a free market this kind of arbitrage is possible by any new entrant and would quickly undermine any even marginally profitable exchange.  Ultimately, all exchanges would end up performing this kind of brokering for their customers.

A standardized inter-exchange bid transfer system would help, but is not necessary.

[bitrebel]

This is not a question of IF. It is obvious they can profit form information only they have. I am not saying they are using this information for personal gains right now, but hey, who could avoid the temptation in the long term?

We would hope they wouldn't, because it benefits everyone for them to do so. By acting on secret information, they make the market price reflect that secret information, effectively publicizing it.

It works the same way as insider trading. Say you know a company's stock is likely to dip. You sell that stock. This causes the price to dip, communicating to the market that the price is too high. This cushions the drop so it is more spread out rather than sudden, benefiting the market generally.

Your saying insider trading benefits the market?
Now I know im at Disneyland.

[DonnyCMU]

So... the 'Master Plan' is to... buy low... and sell high... -_-"   and accumulate your way to 500k coins

Well I wish I could master this plan too!

(the title is so misleading)

With that money you could afford to buy bitcoins daily at the best prices it reached each day. It's not a cheating thing to do, really, but isn't it a conflict of interest to do that in his own exchange?

In a very short time, he could have easily acquired the 500,000 bitcoins, but would it have been completely legal?

Son, what you're saying completely make sense.... If Mt.Gox had a god damn TIME MACHINE!

Buy at the lowest, best price each day?? and sell them off? and buy again at another lowest price another day???
Are you serious? How do you know when are the price 'lowest'? The only advantage Mt.Gox have is black pool. That could be useful in speculating support and resistant, but certainly not enough to determine if 'now' is the 'lowest' point of the day. I think you have a gross misunderstanding about trading and equity market.

Mt.Gox being in Japan doesn't mean it can hire DORAEMON, you know.

[bitrebel]

So... the 'Master Plan' is to... buy low... and sell high... -_-"   and accumulate your way to 500k coins

Well I wish I could master this plan too!

(the title is so misleading)

With that money you could afford to buy bitcoins daily at the best prices it reached each day. It's not a cheating thing to do, really, but isn't it a conflict of interest to do that in his own exchange?

In a very short time, he could have easily acquired the 500,000 bitcoins, but would it have been completely legal?

Son, what you're saying completely make sense.... If Mt.Gox had a god damn TIME MACHINE!

Buy at the lowest, best price each day?? and sell them off? and buy again at another lowest price another day???
Are you serious? How do you know when are the price 'lowest'? The only advantage Mt.Gox have is black pool. That could be useful in speculating support and resistant, but certainly not enough to determine if 'now' is the 'lowest' point of the day. I think you have a gross misunderstanding about trading and equity market.

Mt.Gox being in Japan doesn't mean it can hire DORAEMON, you know.

I think you misunderstand where I was going with this.
I did not mean literally each day, at the lowest price. I just meant that with the profits, each day, buying some when they dipped down, and selling when they spiked, with any trading program, you could build up a nice supply of bitcoins in short time.

[Gandlaf]

I think you misunderstand where I was going with this.
I did not mean literally each day, at the lowest price. I just meant that with the profits, each day, buying some when they dipped down, and selling when they spiked, with any trading program, you could build up a nice supply of bitcoins in short time.

I´m sorry, but this whole thread is a bit pointless.
Either you assume that MtGox adheres to it´s policy of not being a counterparty(i.e. not participating in the market) or you assume that they are not honoring their word.

If you believe they decided to cheat/manipulate the market, following your strategy of buying low and selling high, would be a rather inefficient way of going about it. It would be far more easy and lucrative to simply run the whole exchange on a fractional reserve basis, especially given that they have full and unaudited control over entries in their database and a strong influence over how and when outgoing payments are executed.

So either MtGox is being true to it´s word(no buying for their own account), or they are cheating in which case following your ´Master Plan´ would be about the most inefficient way of going about it.

 

[bitrebel]

I think you misunderstand where I was going with this.
I did not mean literally each day, at the lowest price. I just meant that with the profits, each day, buying some when they dipped down, and selling when they spiked, with any trading program, you could build up a nice supply of bitcoins in short time.

I´m sorry, but this whole thread is a bit pointless.
Either you assume that MtGox adheres to it´s policy of not being a counterparty(i.e. not participating in the market) or you assume that they are not honoring their word.

If you believe they decided to cheat/manipulate the market, following your strategy of buying low and selling high, would be a rather inefficient way of going about it. It would be far more easy and lucrative to simply run the whole exchange on a fractional reserve basis, especially given that they have full and unaudited control over entries in their database and a strong influence over how and when outgoing payments are executed.

So either MtGox is being true to it´s word(no buying for their own account), or they are cheating in which case following your ´Master Plan´ would be about the most inefficient way of going about it.

Thanks Gandlaf.

 

[Gandlaf]

I think you misunderstand where I was going with this.
I did not mean literally each day, at the lowest price. I just meant that with the profits, each day, buying some when they dipped down, and selling when they spiked, with any trading program, you could build up a nice supply of bitcoins in short time.

I´m sorry, but this whole thread is a bit pointless.
Either you assume that MtGox adheres to it´s policy of not being a counterparty(i.e. not participating in the market) or you assume that they are not honoring their word.

If you believe they decided to cheat/manipulate the market, following your strategy of buying low and selling high, would be a rather inefficient way of going about it. It would be far more easy and lucrative to simply run the whole exchange on a fractional reserve basis, especially given that they have full and unaudited control over entries in their database and a strong influence over how and when outgoing payments are executed.

So either MtGox is being true to it´s word(no buying for their own account), or they are cheating in which case following your ´Master Plan´ would be about the most inefficient way of going about it.

Thanks Gandlaf.

 

Don´t really see why you would thank me, I basically just called you(and this thread of yours) a troll, whilst trying to be polite about it.

[bitrebel]

I think you misunderstand where I was going with this.
I did not mean literally each day, at the lowest price. I just meant that with the profits, each day, buying some when they dipped down, and selling when they spiked, with any trading program, you could build up a nice supply of bitcoins in short time.

I´m sorry, but this whole thread is a bit pointless.
Either you assume that MtGox adheres to it´s policy of not being a counterparty(i.e. not participating in the market) or you assume that they are not honoring their word.

If you believe they decided to cheat/manipulate the market, following your strategy of buying low and selling high, would be a rather inefficient way of going about it. It would be far more easy and lucrative to simply run the whole exchange on a fractional reserve basis, especially given that they have full and unaudited control over entries in their database and a strong influence over how and when outgoing payments are executed.

So either MtGox is being true to it´s word(no buying for their own account), or they are cheating in which case following your ´Master Plan´ would be about the most inefficient way of going about it.

Thanks Gandlaf.

 

Don´t really see why you would thank me, I basically just called you(and this thread of yours) a troll, whilst trying to be polite about it.

I guess i'm thanking you for being polite, because i'm not a troll. I just needed someone to point out a few things. I'm trying to figure out why certain things are happening, that I don't understand.