How profitable are exchanges?

[ShakyhandsBTCer]

Exchanges with and without unique client codes could coexist, so that clients could choose.   That is something that the market (or should it be the meta-market?  ) seems capable of deciding.

this would defeat the purpose of having client codes as it was said previously that they would be installed to make sure no one is doing anything nefarious and a criminal could simply use an exchange that does not use these client codes

The intended purpose of the unique client code idea was not to prevent criminal use of bitcoins by clients,  but to prevent certain unethical behavior by the exchange, possibly with complicity of some privileged clients.  (Of course it woudl only work  if the uiqueness of the codes was certified by independent, reputable auditors.)

Unique client codes would  also immensely improve our knowledge of the market.  For instance we could tell how many coins have been bought for each given price and not sold yet; that would provide an estimate of long-range liquidity that may be much more rreliable than the order book.  We could tell whether the market is concentrating or distributing bitcoin ownership.  How many active traders there are in each exchange.  And much more.

This is not something that even our stock market in the US has. In general most people (IMO) trust stock market exchanges and have confidence in the market.

What kind of behavior by exchanges do you think this would stop/allow to be discovered? If you are talking about an exchange trading for it's own account, then there is really nothing wrong with this as long as they are using their own money. If you are talking about an exchange having enough reserves then the exchange could simply have their books audited by an outside auditor   

[1715020012]

1715020012

[1715020012]

1715020012

[1715020012]

1715020012

[1715020012]

1715020012

[1715020012]

1715020012

[1715020012]

1715020012

[JorgeStolfi]

The intended purpose of the unique client code idea was not to prevent criminal use of bitcoins by clients,  but to prevent certain unethical behavior by the exchange, possibly with complicity of some privileged clients.  (Of course it woudl only work  if the uiqueness of the codes was certified by independent, reputable auditors.)

This is not something that even our stock market in the US has. In general most people (IMO) trust stock market exchanges and have confidence in the market.

That is because stock exchanges are strictly regulated and audited by the SEC, and many unethical tricks are crimes with stiff sentences.   I was assuming that bitcoin exchanges would remain free from such regulations for a while still.  If they are to be regulated the same way as stock exchanges, then the UCCs may add little to public confidence (but would still be very useful for market analysis).

What kind of behavior by exchanges do you think this would stop/allow to be discovered? If you are talking about an exchange trading for it's own account, then there is really nothing wrong with this as long as they are using their own money.

That is actually very wrong.  The exchange knows the client orders before other clients, so it can scoop up good trades that otherwise would have been exploited by them.  (I believe this is what is called front-running, yes?).  Any money that the exchange makes this way comes from the pockets of their clients, but they would not notice it -- they would just have less luck than they would in an honest exchange, and as a whole they would lose more money than just the trading fees.

I suspect that all exchanges are doing this (and/or other dirty tricks). If something is not illegal, cannot be detected, and is extremely profitable, why would a businessman not do it?

A standard audit of the books and accounts would not detect this.  But with certfied UCCs in the logs, people will notice that one client (or a few "friends of the house") will have extra luck and will always seem to know juicy orders beforehand.

[ShakyhandsBTCer]

The intended purpose of the unique client code idea was not to prevent criminal use of bitcoins by clients,  but to prevent certain unethical behavior by the exchange, possibly with complicity of some privileged clients.  (Of course it woudl only work  if the uiqueness of the codes was certified by independent, reputable auditors.)

This is not something that even our stock market in the US has. In general most people (IMO) trust stock market exchanges and have confidence in the market.

That is because stock exchanges are strictly regulated and audited by the SEC, and many unethical tricks are crimes with stiff sentences.   I was assuming that bitcoin exchanges would remain free from such regulations for a while still.  If they are to be regulated the same way as stock exchanges, then the UCCs may add little to public confidence (but would still be very useful for market analysis).

Would you prefer that exchanges be regulated like this? Having unethical tricks be crimes would deter exchanges more then simply opening their trading logs like this. IMO people would still utilize an exchange if it was well known that it was doing unethical things that didn't put their money in question nor were illegal. It was no secret that gox was having money problems (the extent of them were not known), and that it had many security breaches but they were the largest exchange for a long time after Bitstamp and BTC-e started seriously competed with them, and when they were de-throned as the largest exchange they still had a lot of trading volume by any measure. IMO money and security problems is worse then an exchange having unethical practices. 

What kind of behavior by exchanges do you think this would stop/allow to be discovered? If you are talking about an exchange trading for it's own account, then there is really nothing wrong with this as long as they are using their own money.

That is actually very wrong.  The exchange knows the client orders before other clients, so it can scoop up good trades that otherwise would have been exploited by them.  (I believe this is what is called front-running, yes?).  Any money that the exchange makes this way comes from the pockets of their clients, but they would not notice it -- they would just have less luck than they would in an honest exchange, and as a whole they would lose more money than just the trading fees.

Yes an exchange does know their customer's orders before they do, and they also know when there will be large buy orders in the near future (when they receive fiat deposits people who make the deposits will generally buy bitcoin). For all customers the time that they have to front run their customers would be seconds at best. The only type of customer that this could protect would be the whales (other orders would simply not be large enough to move the market) and the possible person who could front run would go from being the exchange to other market participants.

I suspect that all exchanges are doing this (and/or other dirty tricks). If something is not illegal, cannot be detected, and is extremely profitable, why would a businessman not do it?

A businessman would not do it because it would be unethical. Even if there is no way for the public to audit certain things, there is also the risk that someone would leak certain information. Most employers tell their employees to act in a way so that they would not be ashamed if what they did appeared on the front page of the newspaper.

What makes you think that all exchanges are front-running their clients?

A standard audit of the books and accounts would not detect this.  But with certfied UCCs in the logs, people will notice that one client (or a few "friends of the house") will have extra luck and will always seem to know juicy orders beforehand.

What would prevent the "friends of the house" from creating multiple accounts and only using each account for one trade? I would think it would be the expected behavior for most accounts to only have one buy or one sell transaction then become dormant as people would cash out of their bitcoin or buy bitcoin as an investment in one lump sum. 

[JorgeStolfi]

IMO people would still utilize an exchange if it was well known that it was doing unethical things that didn't put their money in question nor were illegal.

That is why there are so many scams in bitcoinland...  Most bitcoiners aparently expect from exchanges and bitcoin ventures only the same level of business ethics that they expect from a drugs dealer... 

But an exchange that does unethical things will of course "put the clients' money in question", at the very least by spoiling their "luck" at trades.

An exchange that uses a dirty trick against some other client may next use a different dirty trick on you.  That is why one should not tolerate ANY unethical behavior from  them, against ANYONE.  If clients don't mind other clients being scammed, the exchange will have no reason not to scam them, one by one.

Yes an exchange does know their customer's orders before they do, and they also know when there will be large buy orders in the near future (when they receive fiat deposits people who make the deposits will generally buy bitcoin). For all customers the time that they have to front run their customers would be seconds at best.

That is plenty of time to take advantage of the clients, big or small, even without any fancy psychological heuristics.

Say for example that the current spread is 600$ -- 610$.  At 10:45:00 Alice places a buy order for 10 BTC at 605$.  At 10:45:03, before Alice's bid gets posted to the world, Bob places a sell order for 10 BTC at 600$.   Carl, the exchange owner, then inserts between the two orders his own sell order for 10 BTC at 605$, and his own buy order for 10 BTC at 601$.  Only then he posts the trades to the world, namely

  10:45:01 sell 10 BTC at 605$
  10:45:03 sell 10 BTC at 601$

The first trasaction being from Carl to Alice, the second one from Bob to Carl.  Both Alice and Bob will be happy with the liquidity and the result, but Bob will get only 6010$ for his coins.  Whereas in a honest exchange the only transaction would be

  10:45:03 sell 10 BTC at 605$

from Bob to Alice. In this case Bob would get 6050 for the coins.  The 40$ that Bob failed to earn in the first case are pocketed by Carl.

Certainly many other such tricks are possible, e.g. when a client places a single buy order to scoop up N sell orders at different prices, Carl can run ahead and buy the N-1 lowest sells, then place a single sell order just below the last one.  Or whatever.

A businessman would not do it because it would be unethical. Even if there is no way for the public to audit certain things, there is also the risk that someone would leak certain information. Most employers tell their employees to act in a way so that they would not be ashamed if what they did appeared on the front page of the newspaper.

That does not seem to bother the typical bitcoin scammer.  I don't know if Danny Brewster got front page coverage anywhere, even in Cyprus; and he obviously did not get much worried about it.  The dozens (hundreds by now?) of lesser bitcoin scammers got no coverage even in the "bitcoin media".

What would prevent the "friends of the house" from creating multiple accounts and only using each account for one trade? I would think it would be the expected behavior for most accounts to only have one buy or one sell transaction then become dormant as people would cash out of their bitcoin or buy bitcoin as an investment in one lump sum. 

As I said, one would still require a trusted audit to certify that clients are real and that each client got a single client code.  The dirty tricks that I imagine are happening would be thousands of small "thefts of good luck" per day, as illustrated above, that could not be masked by using a few friendly clients in place of Carl.

[Alex_green]

APSJEX.com offer: The fee for input/withdrawal of BTC – 0%; Trading Fee as low as – 0.2%; Minimal round lot – no minimal amount; Maximum withdrawable sum – unlimited; Bank fee for money transfers – 1.7%;

[BitchicksHusband]

I don't know about the other exchanges, but monthly revenue for BTC-E is somewhere in the range of BTC2,000. That means around $1.2 million per month and $15 million per year. Don't know the net income.

That is huge profit for an exchange. No wonder no major exchange want to offer equity.

Right now I don't think that will be a good idea. 99% of the world population don't know about Bitcoin. The exchange rate and adoption will increase logarithmically in the coming few years. Once that happens, they can list the exchanges in NASDAQ or some other stock market.

It does not matter if a currency is listed on a stockmarket.

Once citizens of emerging markets start to use bitcoin then it can reach it's potential.

Why would citizen of emerging market will choose bitcoin vs their own currency?

Let me answer that in one word: Argentina.

[InwardContour]

I don't know about the other exchanges, but monthly revenue for BTC-E is somewhere in the range of BTC2,000. That means around $1.2 million per month and $15 million per year. Don't know the net income.

That is huge profit for an exchange. No wonder no major exchange want to offer equity.

Right now I don't think that will be a good idea. 99% of the world population don't know about Bitcoin. The exchange rate and adoption will increase logarithmically in the coming few years. Once that happens, they can list the exchanges in NASDAQ or some other stock market.

It does not matter if a currency is listed on a stockmarket.

Once citizens of emerging markets start to use bitcoin then it can reach it's potential.

Why would citizen of emerging market will choose bitcoin vs their own currency?

Let me answer that in one word: Argentina.

Argentina actually has a lot of foreign currency reserves. They have much more then they need to pay back their holdout creditors, they simply do not want to pay them as they have a history of being a deadbeat nation

[JorgeStolfi]

Why would citizen of emerging market will choose bitcoin vs their own currency?

Let me answer that in one word: Argentina.

Please stop that nonsense.  Bitcoin is not a viable hedge against inflation, not even in Argentina.  Pushing it on Argentinians as such is a scam, pure and simple.  Have people forgotten Neo & Bee already? 

[InwardContour]

One thing that would improve transparency a lot would be to attach a scrambled client ID to each bid/ask submission, and the two IDs to each trade

This would get a lot of people to stay away from exchanges, at least people who plan on buying/selling large amounts. If it was clear that scrambled id "1234" was buying large amount of bitcoin then other traders could also buy expecting the large buyer to further drive up the price. The result would be that this large trader essentially pays a higher price then if several small traders bought the same amount at the same time

Yes, many traders would not like even that limited increment in transparency.

Most, if not all largish (and bigger) traders would not like this. It would allow for people to have more information about who has how much coin.

If for example every time that trader "x" bought 200BTC within two hours there was 200BTC(+-5BTC) moved on the blockchain and some of the time the input address(es) were known exchange addresses. An attacker could then trace those coins to his identity.

[Gargulan]

Why would citizen of emerging market will choose bitcoin vs their own currency?

Let me answer that in one word: Argentina.

Please stop that nonsense.  Bitcoin is not a viable hedge against inflation, not even in Argentina.  Pushing it on Argentinians as such is a scam, pure and simple.  Have people forgotten Neo & Bee already? 

Yes. When a country is in trouble, people worry about how to take care of their family. Security, food and medical kits probably worth more to them than bitcoin.

[Harley997]

Why would citizen of emerging market will choose bitcoin vs their own currency?

Let me answer that in one word: Argentina.

Please stop that nonsense.  Bitcoin is not a viable hedge against inflation, not even in Argentina.  Pushing it on Argentinians as such is a scam, pure and simple.  Have people forgotten Neo & Bee already? 

Bitcoin is by concept a deflationary currency. Deflation is the opposite of inflation. When inflation rises, assets that are deflationary will rise, but will still be able to purchase the same amount of goods/services

[JorgeStolfi]

Bitcoin is not a viable hedge against inflation, not even in Argentina.  Pushing it on Argentinians as such is a scam, pure and simple.  Have people forgotten Neo & Bee already? 

Bitcoin is by concept a deflationary currency. Deflation is the opposite of inflation. When inflation rises, assets that are deflationary will rise, but will still be able to purchase the same amount of goods/services

Concept is not reality.

How much did 1 BTC buy on November 28, how much did it buy on January 18, How much now, how much on May 10?  And how much will it buy next month, next year?

Now compare that with gold, stock, or even the Argentinian peso...

Until there are concrete reasons to believe that it value cannot fall as much again, bitcoin cannot be considered a hedge against inflation; and "selling" it as such is scamming. 

[izanagi narukami]

MCXNow, Crypsy, BTCT, BTC-E?

I'm conducting a survey to determine the actual profitability of the popular exchanges.

Does anyone have any idea of what the volume is for these exchanges? What these exchanges make weekly or monthly and how we could find out?

looking by their volume transaction, it must be more than 1000 BTC / month
I think we can calculate by their FEE policy and most of the exchanger using payment vote.
Of course that will make the exchanger gain high profit

ah... i wish i know how to make exchanger 
of course i will not SCAM people because it's a big SIN 

[silvercane]

Inactive accounts with tiny amounts in them, fees, tiny deposits etc make exchanges a very large amount of money. Though, there are a lot of legal complications so setting up an exchange definitely is not an easy task

[Harley997]

Bitcoin is not a viable hedge against inflation, not even in Argentina.  Pushing it on Argentinians as such is a scam, pure and simple.  Have people forgotten Neo & Bee already? 

Bitcoin is by concept a deflationary currency. Deflation is the opposite of inflation. When inflation rises, assets that are deflationary will rise, but will still be able to purchase the same amount of goods/services

Concept is not reality.

How much did 1 BTC buy on November 28, how much did it buy on January 18, How much now, how much on May 10?  And how much will it buy next month, next year?

Now compare that with gold, stock, or even the Argentinian peso...

Until there are concrete reasons to believe that it value cannot fall as much again, bitcoin cannot be considered a hedge against inflation; and "selling" it as such is scamming. 

You cannot use short, specific time periods to compare buying ability. When you do that you can easily pick and choose periods that make the data work in your favor.

[JorgeStolfi]

How much did 1 BTC buy on November 28, how much did it buy on January 18, How much now, how much on May 10?  And how much will it buy next month, next year?

You cannot use short, specific time periods to compare buying ability. When you do that you can easily pick and choose periods that make the data work in your favor.

The point is that there have been periods when the purchasing power of 1 BTC dropped by more than 50%.

There is no reason to believe that it cannot drop another 50% or more from its current price, at any time.  There is no reason to believe that it will not stay low forever, if and when it drops.  Bitcoin enthusiasts hope that it won't, but cannot say why it would not.

So how could one honestly advise people to exchange their pesos for bitcoins?

Keeping one's wealth as cash or simple bank accounts (in pesos or any other currency) is not a good idea in any case, inflation or no inflation.  However, to escape infation one should pick an investment that can reasonably be expected to retain most of its real value, like real estate (if it is not in a bubble), or bonds that are automatically corrected for inflation.  Bitcoins simply cannot serve that purpose: logic says that its purchase power is unpredictable, and experience confirms that it can devaluate much faster than the peso.
 

[Harley997]

How much did 1 BTC buy on November 28, how much did it buy on January 18, How much now, how much on May 10?  And how much will it buy next month, next year?

You cannot use short, specific time periods to compare buying ability. When you do that you can easily pick and choose periods that make the data work in your favor.

The point is that there have been periods when the purchasing power of 1 BTC dropped by more than 50%.

You could pick two arbitrary points in time and could make a negative point about any currency or any investment. You are not going to find any financial unit of measurement that will go up in a straight line. Over long periods of time the purchasing power of bitcoin has increased.

[JorgeStolfi]

The point is that there have been periods when the purchasing power of 1 BTC dropped by more than 50%.

You could pick two arbitrary points in time and could make a negative point about any currency or any investment. You are not going to find any financial unit of measurement that will go up in a straight line.

Hm, when was the last 6-month period where the purchase power of the dollar fell 50%?  

The Brazilian real did that in 1996 (days after the reelection of the president whose only campaign point was its stability). Since then it has fallen gradually by another 30% perhaps.

Over long periods of time the purchasing power of bitcoin has increased.

Indeed, anyone who bought bitcoin at any time before November 1st, 2013 has seen about 1000% increase per year on average.

BUT those who bought after that moment have had a much mixed returns, from very good gains (if they bought in May, for example) to very bad losses (if they bought in January, or after mid-November).

So bitcoin is a good hedge against inflation only for those lucky ones who have a time machine in their attic that allows them to buy their coins in 2013 to sell now.  For those who are forced to buy now to sell in the future, it is not an investment: it is a gamble in a lottery with unknown prize and unknown odds.

[SunBin]

The point is that there have been periods when the purchasing power of 1 BTC dropped by more than 50%.

You could pick two arbitrary points in time and could make a negative point about any currency or any investment. You are not going to find any financial unit of measurement that will go up in a straight line.

Hm, when was the last 6-month period where the purchase power of the dollar fell 50%?  

The Brazilian real did that in 1996 (days after the reelection of the president whose only campaign point was its stability). Since then it has fallen gradually by another 30% perhaps.

Over long periods of time the purchasing power of bitcoin has increased.

Indeed, anyone who bought bitcoin at any time before November 1st, 2013 has seen about 1000% increase per year on average.

BUT those who bought after that moment have had a much mixed returns, from very good gains (if they bought in May, for example) to very bad losses (if they bought in January, or after mid-November).

So bitcoin is a good hedge against inflation only for those lucky ones who have a time machine in their attic that allows them to buy their coins in 2013 to sell now.  For those who are forced to buy now to sell in the future, it is not an investment: it is a gamble in a lottery with unknown prize and unknown odds.

Very practical analysis.

And yes, buying bitcoin is a gamble and not an investment.

[InwardContour]

How much did 1 BTC buy on November 28, how much did it buy on January 18, How much now, how much on May 10?  And how much will it buy next month, next year?

You cannot use short, specific time periods to compare buying ability. When you do that you can easily pick and choose periods that make the data work in your favor.

The point is that there have been periods when the purchasing power of 1 BTC dropped by more than 50%.

You could pick two arbitrary points in time and could make a negative point about any currency or any investment. You are not going to find any financial unit of measurement that will go up in a straight line. Over long periods of time the purchasing power of bitcoin has increased.

It is generally best to look at long term results of any investments to get a good idea as to how it has performed. You should also be looking at several starting points at equal intervals, for example a 3, 6, 9 and 12 month rate, or a 1, 3, and 5 year rate of return.