[Fund] Exchange Rate Differences: Arbitrage Opportunites

[azaniet]

I am sure everybody is aware of the arbitrage opportunities caused by differences in BTC exchange rates.

For example, on October 8th I bought BTC32.73 for $4052.39 (€3000) and sold BTC30.56 at Mt.Gox for €3,030.30. These transactions happened in a 10 minute time frame. Today, I received €3000 (99% * €3030.30) on my bank account. So, with this transaction I gained BTC2.17 (i.e. 7.1% return) in 15 days.

Would anybody be interested in investing in a new fund solely devoted to the above type of (risk-free) transactions? If so, please comment on how you would like this type of fund to be set up.

[1714655736]

1714655736

[1714655736]

1714655736

[1714655736]

1714655736

[1714655736]

1714655736

[1714655736]

1714655736

[1714655736]

1714655736

[SebastianJu]

You surely didnt try to get these $ to your bankaccount. Since you will have to wait weeks, sometimes months at mtgox. Thats why no one is selling bitcoins there. And thats why the bitcoin price there is not real.
Arbitrage is a risk with mtgox. At the moment you get the money the bitcoin already could be worth way more.

[Rannasha]

You surely didnt try to get these $ to your bankaccount. Since you will have to wait weeks, sometimes months at mtgox. Thats why no one is selling bitcoins there. And thats why the bitcoin price there is not real.
Arbitrage is a risk with mtgox. At the moment you get the money the bitcoin already could be worth way more.

Well, he posted that he received the money today, so 2 weeks after the transaction (euro-withdrawals are allegedly somewhat faster than dollar-withdrawals).

Also, his type of arbitrage is not exposed to fluctuations in the exchange rate, since he buys on Stamp and sells on Gox at the same time. It means you lock in both buy and sell prices, but you need to have both fiat and coin available to make it work (as well as wait for Gox to pay out the fiat before you start a second round)

[SebastianJu]

Looks like i overread that he already got the money. But as far as i read thats nothing he can trust about. Or did Mtgox change his behaviour? It would astonish me since the exchangeprice is still so high. If its really only 2 weeks normally now then i await that the prices are getting a bit closer to other exchanges now.

I dont think its needed to have bitcoins or money on both platforms since he bought bitcoins. They can be withdrawn pretty fast and uploaded fast on mtgox too. The shareprice wont change too much in that timeframe.

[azaniet]

Well, he posted that he received the money today, so 2 weeks after the transaction (euro-withdrawals are allegedly somewhat faster than dollar-withdrawals).

Also, his type of arbitrage is not exposed to fluctuations in the exchange rate, since he buys on Stamp and sells on Gox at the same time. It means you lock in both buy and sell prices, but you need to have both fiat and coin available to make it work (as well as wait for Gox to pay out the fiat before you start a second round)

I dont think its needed to have bitcoins or money on both platforms since he bought bitcoins. They can be withdrawn pretty fast and uploaded fast on mtgox too. The shareprice wont change too much in that timeframe.

I agree with Rannasha, since it would still imply more risk if you transfer (approx. 1 hour). A lot can happen in one hour especially on the BTC market. I think it would always be better to avoid this risk, especially, when I would hold responsibility for other people's funds.

Looks like i overread that he already got the money. But as far as i read thats nothing he can trust about. Or did Mtgox change his behaviour? It would astonish me since the exchangeprice is still so high. If its really only 2 weeks normally now then i await that the prices are getting a bit closer to other exchanges now.

The average withdrawal might take more time than the 15 days in this particular case. But, let's say it would take 1 month, then still a return of +7%/month is still worth pursuing in my opinion.

[Deprived]

Well, he posted that he received the money today, so 2 weeks after the transaction (euro-withdrawals are allegedly somewhat faster than dollar-withdrawals).

Also, his type of arbitrage is not exposed to fluctuations in the exchange rate, since he buys on Stamp and sells on Gox at the same time. It means you lock in both buy and sell prices, but you need to have both fiat and coin available to make it work (as well as wait for Gox to pay out the fiat before you start a second round)

I dont think its needed to have bitcoins or money on both platforms since he bought bitcoins. They can be withdrawn pretty fast and uploaded fast on mtgox too. The shareprice wont change too much in that timeframe.

I agree with Rannasha, since it would still imply more risk if you transfer (approx. 1 hour). A lot can happen in one hour especially on the BTC market. I think it would always be better to avoid this risk, especially, when I would hold responsibility for other people's funds.

Looks like i overread that he already got the money. But as far as i read thats nothing he can trust about. Or did Mtgox change his behaviour? It would astonish me since the exchangeprice is still so high. If its really only 2 weeks normally now then i await that the prices are getting a bit closer to other exchanges now.

The average withdrawal might take more time than the 15 days in this particular case. But, let's say it would take 1 month, then still a return of +7%/month is still worth pursuing in my opinion.

Problem is that +7% in a month turns into a loss (expressed in BTC) if BTC rises by more than 7% whilst your withdrawal is stuck in limbo.

Add to that two layers of counter-party risk (you and Mtgox) and it suddenly doesn't look very attractive.

Arbitrage opportunities where you're stuck on one (the worst) side of the deal for any length of time cease to actually be arbitrage - as the end result depends far more on how the exchange-rate moves during that period that on the relatively small difference in exchange-rates at the point at which you initiated the trade.

Just consider if you'd done such an arbitrage a week ago and were now waiting for euros to arrive.  Would you REALLY think you were better off than if you'd just held BTC throughout?

[azaniet]

Well, he posted that he received the money today, so 2 weeks after the transaction (euro-withdrawals are allegedly somewhat faster than dollar-withdrawals).

Also, his type of arbitrage is not exposed to fluctuations in the exchange rate, since he buys on Stamp and sells on Gox at the same time. It means you lock in both buy and sell prices, but you need to have both fiat and coin available to make it work (as well as wait for Gox to pay out the fiat before you start a second round)

I dont think its needed to have bitcoins or money on both platforms since he bought bitcoins. They can be withdrawn pretty fast and uploaded fast on mtgox too. The shareprice wont change too much in that timeframe.

I agree with Rannasha, since it would still imply more risk if you transfer (approx. 1 hour). A lot can happen in one hour especially on the BTC market. I think it would always be better to avoid this risk, especially, when I would hold responsibility for other people's funds.

Looks like i overread that he already got the money. But as far as i read thats nothing he can trust about. Or did Mtgox change his behaviour? It would astonish me since the exchangeprice is still so high. If its really only 2 weeks normally now then i await that the prices are getting a bit closer to other exchanges now.

The average withdrawal might take more time than the 15 days in this particular case. But, let's say it would take 1 month, then still a return of +7%/month is still worth pursuing in my opinion.

Problem is that +7% in a month turns into a loss (expressed in BTC) if BTC rises by more than 7% whilst your withdrawal is stuck in limbo.

Add to that two layers of counter-party risk (you and Mtgox) and it suddenly doesn't look very attractive.

Arbitrage opportunities where you're stuck on one (the worst) side of the deal for any length of time cease to actually be arbitrage - as the end result depends far more on how the exchange-rate moves during that period that on the relatively small difference in exchange-rates at the point at which you initiated the trade.

Just consider if you'd done such an arbitrage a week ago and were now waiting for euros to arrive.  Would you REALLY think you were better off than if you'd just held BTC throughout?

I apologize beforehand, because my first language is not English, but I think that you misunderstood my explanation. I will try to clarify with an example using current exchange rates:


Preparation:

• Let's say that I hold €10,000.00 at my bank account which I deposited to Bitstamp: $13,650.00 at current rates (no deposit fee).
• Additionally, I own BTC70 which is deposited to my Mt.Gox wallet.

Transactions:

• I buy $13,650.00 worth of BTC (0.4% fee) at Bitstamp for $199.30 ea: $13,650.00/($199.30/0.996)=BTC68.22.
• At the same time I sell €10,101.01 worth of BTC (0.6% fee) at Mt.Gox for €158.25 ea: (€10101.01/158.25)/0.994=BTC64.21.

Final picture:

• I now own 70-64.21+68.22 = BTC74.01 (5.7% return).
• And I withdraw the fiat money: €10,101.01 * 0.99 = €10,000.00 (1% fee) which I will receive x days later.

Thus, the amount of BTC held becomes larger with each transaction. The time it will take to receive the fiat determines how many opportunities (if there are any) may be exploited.

Additionally, the amount of EUR and BTC available determine to which extent a price difference can be exploited, of course.

[SebastianJu]

Like Deprived wrote... you actually lost money with the trade. The Europrice for a bitcoin was around 100€. Today its 145€. So the bitcoins now are 45% more worth. Way less then 7%.

Of course thats probably a problem that doesnt occur often but the longer it takes to repeat such a cycle the higher is the risk. I exchange bitcoins for fiat too but i do it at bitstamp because when i do it at mtgox i dont know when i get the fiat and if i get it bitcoins might be way more worth. You might say you invested USD and now have more USD. But the long time to get the USD is a risk that easily can make you regret why you didnt keep Bitcoins instead.

Arbitrage with MtGox was done pretty often some years? ago but as long as i know it was not popular then because Mtgox has these problems with paying out fiat.

Well, he posted that he received the money today, so 2 weeks after the transaction (euro-withdrawals are allegedly somewhat faster than dollar-withdrawals).

Also, his type of arbitrage is not exposed to fluctuations in the exchange rate, since he buys on Stamp and sells on Gox at the same time. It means you lock in both buy and sell prices, but you need to have both fiat and coin available to make it work (as well as wait for Gox to pay out the fiat before you start a second round)

I dont think its needed to have bitcoins or money on both platforms since he bought bitcoins. They can be withdrawn pretty fast and uploaded fast on mtgox too. The shareprice wont change too much in that timeframe.

I agree with Rannasha, since it would still imply more risk if you transfer (approx. 1 hour). A lot can happen in one hour especially on the BTC market. I think it would always be better to avoid this risk, especially, when I would hold responsibility for other people's funds.

Looks like i overread that he already got the money. But as far as i read thats nothing he can trust about. Or did Mtgox change his behaviour? It would astonish me since the exchangeprice is still so high. If its really only 2 weeks normally now then i await that the prices are getting a bit closer to other exchanges now.

The average withdrawal might take more time than the 15 days in this particular case. But, let's say it would take 1 month, then still a return of +7%/month is still worth pursuing in my opinion.

Problem is that +7% in a month turns into a loss (expressed in BTC) if BTC rises by more than 7% whilst your withdrawal is stuck in limbo.

Add to that two layers of counter-party risk (you and Mtgox) and it suddenly doesn't look very attractive.

Arbitrage opportunities where you're stuck on one (the worst) side of the deal for any length of time cease to actually be arbitrage - as the end result depends far more on how the exchange-rate moves during that period that on the relatively small difference in exchange-rates at the point at which you initiated the trade.

Just consider if you'd done such an arbitrage a week ago and were now waiting for euros to arrive.  Would you REALLY think you were better off than if you'd just held BTC throughout?

I apologize beforehand, because my first language is not English, but I think that you misunderstood my explanation. I will try to clarify with an example using current exchange rates:


Preparation:

• Let's say that I hold €10,000.00 at my bank account which I deposited to Bitstamp: $13,650.00 at current rates (no deposit fee).
• Additionally, I own BTC70 which is deposited to my Mt.Gox wallet.

Transactions:

• I buy $13,650.00 worth of BTC (0.4% fee) at Bitstamp for $199.30 ea: $13,650.00/($199.30/0.996)=BTC68.22.
• At the same time I sell €10,101.01 worth of BTC (0.6% fee) at Mt.Gox for €158.25 ea: (€10101.01/158.25)/0.994=BTC64.21.

Final picture:

• I now own 70-64.21+68.22 = BTC74.01 (5.7% return).
• And I withdraw the fiat money: €10,101.01 * 0.99 = €10,000.00 (1% fee) which I will receive x days later.

Thus, the amount of BTC held becomes larger with each transaction. The time it will take to receive the fiat determines how many opportunities (if there are any) may be exploited.

Additionally, the amount of EUR and BTC available determine to which extent a price difference can be exploited, of course.

[azaniet]

Like Deprived wrote... you actually lost money with the trade. The Europrice for a bitcoin was around 100€. Today its 145€. So the bitcoins now are 45% more worth. Way less then 7%.

Of course thats probably a problem that doesnt occur often but the longer it takes to repeat such a cycle the higher is the risk. I exchange bitcoins for fiat too but i do it at bitstamp because when i do it at mtgox i dont know when i get the fiat and if i get it bitcoins might be way more worth. You might say you invested USD and now have more USD. But the long time to get the USD is a risk that easily can make you regret why you didnt keep Bitcoins instead.

Arbitrage with MtGox was done pretty often some years? ago but as long as i know it was not popular then because Mtgox has these problems with paying out fiat.

Please explain how I lost money with the trade, because I did not. I got back the fiat and retained my original btc fund + BTC2.17.

I think that you see the trade this way (correct me if I'm wrong): I am buying BTC at bitstamp, selling it at mtgox and waiting for the fiat to return before buying BTC again, and thereby risking the increase in BTC while waiting on my fiat. This would be a high risk endeavor and would've caused me to lose money. This is not what my transaction entailed.

[SebastianJu]

Like Deprived wrote... you actually lost money with the trade. The Europrice for a bitcoin was around 100€. Today its 145€. So the bitcoins now are 45% more worth. Way less then 7%.

Of course thats probably a problem that doesnt occur often but the longer it takes to repeat such a cycle the higher is the risk. I exchange bitcoins for fiat too but i do it at bitstamp because when i do it at mtgox i dont know when i get the fiat and if i get it bitcoins might be way more worth. You might say you invested USD and now have more USD. But the long time to get the USD is a risk that easily can make you regret why you didnt keep Bitcoins instead.

Arbitrage with MtGox was done pretty often some years? ago but as long as i know it was not popular then because Mtgox has these problems with paying out fiat.

Please explain how I lost money with the trade, because I did not. I got back the fiat and retained my original btc fund + BTC2.17.

I think that you see the trade this way (correct me if I'm wrong): I am buying BTC at bitstamp, selling it at mtgox and waiting for the fiat to return before buying BTC again, and thereby risking the increase in BTC while waiting on my fiat. This would be a high risk endeavor and would've caused me to lose money. This is not what my transaction entailed.

I see it this way... Buying BTC... keeping... when you need Fiat then selling at bitstamp and having the full value in 2-3 workdays.

Would you have sold the 32.37BTC today at bitstamp (i know its unfair since it takes some time to withdraw there too) then you would need, at a Price of $201.5, $4135.5, which means you sell 20.52BTC, got the 3000€ for it and still would have 15BTC.
I mean time is something crucial for bitcoins.

[azaniet]

Like Deprived wrote... you actually lost money with the trade. The Europrice for a bitcoin was around 100€. Today its 145€. So the bitcoins now are 45% more worth. Way less then 7%.

Of course thats probably a problem that doesnt occur often but the longer it takes to repeat such a cycle the higher is the risk. I exchange bitcoins for fiat too but i do it at bitstamp because when i do it at mtgox i dont know when i get the fiat and if i get it bitcoins might be way more worth. You might say you invested USD and now have more USD. But the long time to get the USD is a risk that easily can make you regret why you didnt keep Bitcoins instead.

Arbitrage with MtGox was done pretty often some years? ago but as long as i know it was not popular then because Mtgox has these problems with paying out fiat.

Please explain how I lost money with the trade, because I did not. I got back the fiat and retained my original btc fund + BTC2.17.

I think that you see the trade this way (correct me if I'm wrong): I am buying BTC at bitstamp, selling it at mtgox and waiting for the fiat to return before buying BTC again, and thereby risking the increase in BTC while waiting on my fiat. This would be a high risk endeavor and would've caused me to lose money. This is not what my transaction entailed.

I see it this way... Buying BTC... keeping... when you need Fiat then selling at bitstamp and having the full value in 2-3 workdays.

Would you have sold the 32.37BTC today at bitstamp (i know its unfair since it takes some time to withdraw there too) then you would need, at a Price of $201.5, $4135.5, which means you sell 20.52BTC, got the 3000€ for it and still would have 15BTC.
I mean time is something crucial for bitcoins.

You're implying that on October 8th I should have only bought btc to for example sell it today? What if the price would have declined? Anyway, that would require the ability to predict the future btc price...

And I still have the BTC32.73! Meaning that my transaction has only increased the amount of BTC in my portfolio:

• Ocotober 8th: I had €3000 + BTC30.56.
• Today: I have €3000 + BTC32.73

Anyway. I understand that most people here have faith in the btc price increasing. This doesn't mean that we cannot make money of price differences as illustrated.

[SebastianJu]

You're implying that on October 8th I should have only bought btc to for example sell it today? What if the price would have declined? Anyway, that would require the ability to predict the future btc price...

And I still have the BTC32.73! Meaning that my transaction has only increased the amount of BTC in my portfolio:

• Ocotober 8th: I had €3000 + BTC30.56.
• Today: I have €3000 + BTC32.73

Anyway. I understand that most people here have faith in the btc price increasing. This doesn't mean that we cannot make money of price differences as illustrated.

I only say that i dont use mtgox to sell Bitcoins because i think its too risky that i will get less fiat than i should at the time i actually have the fiat. And i think so because i believe that bitcoins generally will have the tendency to rise in price for some time now.

I forgot you had the 3000€ on top. So if you had turned them to bitcoins 2 weeks ago you now could exchange them to 45% more fiat, not only 7%, so in fact you lost money by not keeping everything as bitcoins. Something that never could have happened when mtgox wouldnt be so slow. Thats why i keep my money in form of bitcoins. And i dont sell bitcoins at mtgox because i know that such things like in the last 2 weeks can happen anytime. On top i dont plan weeks or months in advance to when i need fiat on my bank. And i cant guess in how many weeks mtgox will give my fiat.

[pascal257]

Sebastian, Please keep in mind that the bitcoin exchange rate may also decline. You're assuming that BTC will always raise. In that case you would even increase the expected profit.

[Rannasha]

You're implying that on October 8th I should have only bought btc to for example sell it today? What if the price would have declined? Anyway, that would require the ability to predict the future btc price...

And I still have the BTC32.73! Meaning that my transaction has only increased the amount of BTC in my portfolio:

• Ocotober 8th: I had €3000 + BTC30.56.
• Today: I have €3000 + BTC32.73

Anyway. I understand that most people here have faith in the btc price increasing. This doesn't mean that we cannot make money of price differences as illustrated.

I forgot you had the 3000€ on top. So if you had turned them to bitcoins 2 weeks ago you now could exchange them to 45% more fiat, not only 7%, so in fact you lost money by not keeping everything as bitcoins. Something that never could have happened when mtgox wouldnt be so slow. Thats why i keep my money in form of bitcoins. And i dont sell bitcoins at mtgox because i know that such things like in the last 2 weeks can happen anytime. On top i dont plan weeks or months in advance to when i need fiat on my bank. And i cant guess in how many weeks mtgox will give my fiat.

Your proposal of simply converting the 3000€ to BTC is a different investment, relying on the increase in price of BTC. The arbitrage-scheme discussed here relies on the exchanges operating properly. It is perfectly reasonable for someone to keep part of his capital in fiat currencies and if you have the fiat available, the scheme discussed here can use that fiat to obtain BTC-denominated profits and this is a viable course of action if you assume the reward outperforms the risk of MtGox defaulting.

[SebastianJu]

Sebastian, Please keep in mind that the bitcoin exchange rate may also decline. You're assuming that BTC will always raise. In that case you would even increase the expected profit.

I know that. But i believe that crashes could be relatively good be forseen in the past. And i believe too that bitcoins have to rise in price when they are used more widely. And i believe thats the case. The more people want to use it the higher the price will go.

You're implying that on October 8th I should have only bought btc to for example sell it today? What if the price would have declined? Anyway, that would require the ability to predict the future btc price...

And I still have the BTC32.73! Meaning that my transaction has only increased the amount of BTC in my portfolio:

• Ocotober 8th: I had €3000 + BTC30.56.
• Today: I have €3000 + BTC32.73

Anyway. I understand that most people here have faith in the btc price increasing. This doesn't mean that we cannot make money of price differences as illustrated.

I forgot you had the 3000€ on top. So if you had turned them to bitcoins 2 weeks ago you now could exchange them to 45% more fiat, not only 7%, so in fact you lost money by not keeping everything as bitcoins. Something that never could have happened when mtgox wouldnt be so slow. Thats why i keep my money in form of bitcoins. And i dont sell bitcoins at mtgox because i know that such things like in the last 2 weeks can happen anytime. On top i dont plan weeks or months in advance to when i need fiat on my bank. And i cant guess in how many weeks mtgox will give my fiat.

Your proposal of simply converting the 3000€ to BTC is a different investment, relying on the increase in price of BTC. The arbitrage-scheme discussed here relies on the exchanges operating properly. It is perfectly reasonable for someone to keep part of his capital in fiat currencies and if you have the fiat available, the scheme discussed here can use that fiat to obtain BTC-denominated profits and this is a viable course of action if you assume the reward outperforms the risk of MtGox defaulting.

I see that. But i think mtgox isnt really a good partner for this. Search the forum. Some users had to wait months! to get their fiat. And even when its only 2 weeks. Thats something you can do 24 times in one year only. I admit that overall you might get a positive result though. I only dont like the unreliability of mtgox in this game.
But maybe thats only my personal view. If it works for others its fine...

[azaniet]

Sebastian, Please keep in mind that the bitcoin exchange rate may also decline. You're assuming that BTC will always raise. In that case you would even increase the expected profit.

I know that. But i believe that crashes could be relatively good be forseen in the past. And i believe too that bitcoins have to rise in price when they are used more widely. And i believe thats the case. The more people want to use it the higher the price will go.

You're implying that on October 8th I should have only bought btc to for example sell it today? What if the price would have declined? Anyway, that would require the ability to predict the future btc price...

And I still have the BTC32.73! Meaning that my transaction has only increased the amount of BTC in my portfolio:

• Ocotober 8th: I had €3000 + BTC30.56.
• Today: I have €3000 + BTC32.73

Anyway. I understand that most people here have faith in the btc price increasing. This doesn't mean that we cannot make money of price differences as illustrated.

I forgot you had the 3000€ on top. So if you had turned them to bitcoins 2 weeks ago you now could exchange them to 45% more fiat, not only 7%, so in fact you lost money by not keeping everything as bitcoins. Something that never could have happened when mtgox wouldnt be so slow. Thats why i keep my money in form of bitcoins. And i dont sell bitcoins at mtgox because i know that such things like in the last 2 weeks can happen anytime. On top i dont plan weeks or months in advance to when i need fiat on my bank. And i cant guess in how many weeks mtgox will give my fiat.

Your proposal of simply converting the 3000€ to BTC is a different investment, relying on the increase in price of BTC. The arbitrage-scheme discussed here relies on the exchanges operating properly. It is perfectly reasonable for someone to keep part of his capital in fiat currencies and if you have the fiat available, the scheme discussed here can use that fiat to obtain BTC-denominated profits and this is a viable course of action if you assume the reward outperforms the risk of MtGox defaulting.

I see that. But i think mtgox isnt really a good partner for this. Search the forum. Some users had to wait months! to get their fiat. And even when its only 2 weeks. Thats something you can do 24 times in one year only. I admit that overall you might get a positive result though. I only dont like the unreliability of mtgox in this game.
But maybe thats only my personal view. If it works for others its fine...

If you would be able to do this 24 times a year though (assuming you have enough EUR available) :

1.06^24 - 1 = 305% (6% is quite a low average still)

Or even 12 times a year would mean your get a 101% return.

But yes, the main (and only major) risk here is Mt.Gox defaulting.

[azaniet]

Related: https://bitcointalk.org/index.php?topic=313358.0

We're discussing a whole other fund with a different type of strategy here. Though, "bonds" might be a means to set up a fund as discussed here to obtain BTC-denominated profits.

[Deprived]

Well, he posted that he received the money today, so 2 weeks after the transaction (euro-withdrawals are allegedly somewhat faster than dollar-withdrawals).

Also, his type of arbitrage is not exposed to fluctuations in the exchange rate, since he buys on Stamp and sells on Gox at the same time. It means you lock in both buy and sell prices, but you need to have both fiat and coin available to make it work (as well as wait for Gox to pay out the fiat before you start a second round)

I dont think its needed to have bitcoins or money on both platforms since he bought bitcoins. They can be withdrawn pretty fast and uploaded fast on mtgox too. The shareprice wont change too much in that timeframe.

I agree with Rannasha, since it would still imply more risk if you transfer (approx. 1 hour). A lot can happen in one hour especially on the BTC market. I think it would always be better to avoid this risk, especially, when I would hold responsibility for other people's funds.

Looks like i overread that he already got the money. But as far as i read thats nothing he can trust about. Or did Mtgox change his behaviour? It would astonish me since the exchangeprice is still so high. If its really only 2 weeks normally now then i await that the prices are getting a bit closer to other exchanges now.

The average withdrawal might take more time than the 15 days in this particular case. But, let's say it would take 1 month, then still a return of +7%/month is still worth pursuing in my opinion.

Problem is that +7% in a month turns into a loss (expressed in BTC) if BTC rises by more than 7% whilst your withdrawal is stuck in limbo.

Add to that two layers of counter-party risk (you and Mtgox) and it suddenly doesn't look very attractive.

Arbitrage opportunities where you're stuck on one (the worst) side of the deal for any length of time cease to actually be arbitrage - as the end result depends far more on how the exchange-rate moves during that period that on the relatively small difference in exchange-rates at the point at which you initiated the trade.

Just consider if you'd done such an arbitrage a week ago and were now waiting for euros to arrive.  Would you REALLY think you were better off than if you'd just held BTC throughout?

I apologize beforehand, because my first language is not English, but I think that you misunderstood my explanation. I will try to clarify with an example using current exchange rates:


Preparation:

• Let's say that I hold €10,000.00 at my bank account which I deposited to Bitstamp: $13,650.00 at current rates (no deposit fee).
• Additionally, I own BTC70 which is deposited to my Mt.Gox wallet.

Transactions:

• I buy $13,650.00 worth of BTC (0.4% fee) at Bitstamp for $199.30 ea: $13,650.00/($199.30/0.996)=BTC68.22.
• At the same time I sell €10,101.01 worth of BTC (0.6% fee) at Mt.Gox for €158.25 ea: (€10101.01/158.25)/0.994=BTC64.21.

Final picture:

• I now own 70-64.21+68.22 = BTC74.01 (5.7% return).
• And I withdraw the fiat money: €10,101.01 * 0.99 = €10,000.00 (1% fee) which I will receive x days later.

Thus, the amount of BTC held becomes larger with each transaction. The time it will take to receive the fiat determines how many opportunities (if there are any) may be exploited.

Additionally, the amount of EUR and BTC available determine to which extent a price difference can be exploited, of course.

To make a BTC-denominated profit, the TOTAL value of your holdings (in BTC) has to rise.

Yes - you make a 'profit' on the trade.  But if you look at the TOTAL value of your capital then whether you actually made an overall profit or not depends far more on the how the exchange-rate moved than on the difference between the exchange-rates at different exchanges.

What you're proposing only makes sense for people who want some of their BTC to be converted to fiat and held as fiat.  i.e. those who believe BTC won't move much or will fall vs fiat over whatever period they intend to be invested for.  And because the BTC exchange-rate tends to move a lot, the end result they achieve will have more to do with how the rate moves than with the 'profit' from the arbitrage.

Simply put : for a BTC-denominated investment to make a profit it must end up with more BTC than it started with.  And you have to value ALL capital (even the part held in euros) in BTC to measure that.  I'm not saying it's necessarily a bad thing to do - just that it doesn't make sense to do it IF you want to hold BTC.  If investors intended to hold a mix of BTC and fiat anyway then it starts to make sense - as they DO end up making a profit compared to just passively holding the mix of currencies.

In fact it makes most sense as an investment for those who want to try to profit from BTC without committing to holding any long-term (as the BTC part is always liquid so can be exchanged out if the rate starts to move unfavourably).

The mistake you made is in portraying it as a means by which a profit in BTC could be made - when whether any such profit is made or not depends far more on how the exchange-rate moves (and hence the value of the part locked up in fiat) than on the difference between the rates on different exchanges.

[azaniet]

To make a BTC-denominated profit, the TOTAL value of your holdings (in BTC) has to rise.

Yes - you make a 'profit' on the trade.  But if you look at the TOTAL value of your capital then whether you actually made an overall profit or not depends far more on the how the exchange-rate moved than on the difference between the exchange-rates at different exchanges.

What you're proposing only makes sense for people who want some of their BTC to be converted to fiat and held as fiat.  i.e. those who believe BTC won't move much or will fall vs fiat over whatever period they intend to be invested for.  And because the BTC exchange-rate tends to move a lot, the end result they achieve will have more to do with how the rate moves than with the 'profit' from the arbitrage.

Simply put : for a BTC-denominated investment to make a profit it must end up with more BTC than it started with.  And you have to value ALL capital (even the part held in euros) in BTC to measure that.  I'm not saying it's necessarily a bad thing to do - just that it doesn't make sense to do it IF you want to hold BTC.  If investors intended to hold a mix of BTC and fiat anyway then it starts to make sense - as they DO end up making a profit compared to just passively holding the mix of currencies.

In fact it makes most sense as an investment for those who want to try to profit from BTC without committing to holding any long-term (as the BTC part is always liquid so can be exchanged out if the rate starts to move unfavourably).

The mistake you made is in portraying it as a means by which a profit in BTC could be made - when whether any such profit is made or not depends far more on how the exchange-rate moves (and hence the value of the part locked up in fiat) than on the difference between the rates on different exchanges.

Your analysis is true of course, i.e. you mean that:

On October 8th, I had €3,000 + BTC30.56  which was worth (in btc) 30.56 + 32.73 = BTC63.29 on that day.
And on October 23rd I had BTC32.73 + €3,000 which was worth (in btc) 32.73 + approx. 22 = BTC54.73 on that day.

Meaning that if I had bought €3,000 of btc and did not sell any on October 8th, i.e. going long in Eur/btc, I would have had more Euro's worth of BTC now, True! Though, €/BTC could also decline.

Nevertheless, it all depends on the way that a potential fund would be set-up. One way could be that the amount of fiat that I own will be used as a means of growing a BTC-denominated fund. This will mean that investors' funds in BTC would grow aside from the unchanging amount of fiat. This would not require the investors' btc to be converted to fiat and investors will be benefiting from an increased btc exchange-rate.

[Deprived]

To make a BTC-denominated profit, the TOTAL value of your holdings (in BTC) has to rise.

Yes - you make a 'profit' on the trade.  But if you look at the TOTAL value of your capital then whether you actually made an overall profit or not depends far more on the how the exchange-rate moved than on the difference between the exchange-rates at different exchanges.

What you're proposing only makes sense for people who want some of their BTC to be converted to fiat and held as fiat.  i.e. those who believe BTC won't move much or will fall vs fiat over whatever period they intend to be invested for.  And because the BTC exchange-rate tends to move a lot, the end result they achieve will have more to do with how the rate moves than with the 'profit' from the arbitrage.

Simply put : for a BTC-denominated investment to make a profit it must end up with more BTC than it started with.  And you have to value ALL capital (even the part held in euros) in BTC to measure that.  I'm not saying it's necessarily a bad thing to do - just that it doesn't make sense to do it IF you want to hold BTC.  If investors intended to hold a mix of BTC and fiat anyway then it starts to make sense - as they DO end up making a profit compared to just passively holding the mix of currencies.

In fact it makes most sense as an investment for those who want to try to profit from BTC without committing to holding any long-term (as the BTC part is always liquid so can be exchanged out if the rate starts to move unfavourably).

The mistake you made is in portraying it as a means by which a profit in BTC could be made - when whether any such profit is made or not depends far more on how the exchange-rate moves (and hence the value of the part locked up in fiat) than on the difference between the rates on different exchanges.

Your analysis is true of course, i.e. you mean that:

On October 8th, I had €3,000 + BTC30.56  which was worth (in btc) 30.56 + 32.73 = BTC63.29 on that day.
And on October 23rd I had BTC32.73 + €3,000 which was worth (in btc) 32.73 + approx. 22 = BTC54.73 on that day.

Meaning that if I had bought €3,000 of btc and did not sell any on October 8th, i.e. going long in Eur/btc, I would have had more Euro's worth of BTC now, True! Though, €/BTC could also decline.

Nevertheless, it all depends on the way that a potential fund would be set-up. One way could be that the amount of fiat that I own will be used as a means of growing a BTC-denominated fund. This will mean that investors' funds in BTC would grow aside from the unchanging amount of fiat. This would not require the investors' btc to be converted to fiat and investors will be benefiting from an increased btc exchange-rate.

Your last paragraph is true.  But there's a problem with it - specifically this:

If you raise BTC from investors then you need MORE fiat to actually use those invested BTC withut exposing investors to fiat-risk.  If you had those extra fiat then you wouldn't need to be raising funds from investors in the first place.

Put simply: if you have enough fiat to be able to shield investors from fiat-exposure (i.e. keep their investment always denominated in BTC) then you don't need their investment at all.

My view on investments like this is that if they want to be denominated primarily in one currency then they need fiat-denominated investment not BTC-denominated  - because you can minimise BTC exposure easily (due to it being liquid) but can't do the same with fiat exposure (due to a lack of ability to move it rapidly between exchanges).  Don't interpret my comments as me being opposed to fiat-exposure - I'm not, and maintain significant fiat-exposure with my BTC myself intentionally.  What I'm opposed to is attempts to sell 'BTC' investments where the investment can make a 'profit' but still end up worse than just holding BTC (which, incidentally, includes nearly all mining 'investments').

Something like this works nicely if (and ONLY if) investors WANT significant fiat exposure and are happy to accept that they'll do worse than holding BTC if BTC rises vs fiat.  Specifically they need to understand, accept and even WANT that if BTC rises fairly sharply vs fiat then they'll get back less BTC than they put in.  In return they do, of course, do a LOT better than just holding BTC if BTC falls even slightly vs fiat.

For those saying the main risk is MtGox - just no.  Historically the main risk in ANY security is the issuer.  Even when GLBSE closed, the exchange itself only stole a very small percentage of invested funds (under 10% of deposited cash and next to nothing from value tied up in securities) - which is nothing compared to what issuers steal.  That's not a comment about this particular (potential) issuer - I'm just pointing out that issuer counter-party risk can never be treated as something negligible.