[OFFICIAL]Bitfinex.com first Bitcoin P2P lending platform for leverage trading

[Sukrim]

How about as a first test measure "FRR = (average of all current swaps/85)*100"? This would mean that the 15% fee of Bitfinex is added statically on top of the FRR, as currently if you lend at FRR, you receive only 85% of the average rate in reality.

This would mean the FRR rises the more it gets used and no fancy programming would need to happen (unlike "FRR +/- X%" scenarios, that can also end up cancelling each other out).

[QwertyCore]

To put the FRR problem a slightly different way... it's using "the average rate of all fixed-rate swaps" to try and approximate the equilibrium point between supply and demand, but it's got its finger on the scales of that equilibrium because so much of the supply and demand it's trying to measure is provided/satisfied by FRR swaps.

Currently, so long as there's a sufficient wall of FRR offers, you can take out a million-dollar swap and leave the rate unaffected... whereas if you did the same thing to the fixed-rate offers it'd tear through to the astronomical rates (if not the entire book of offers) and yank the average abruptly upward. If it could be reformed in such a way that the flash-rate moved in response to offers being taken at the flash rate then... whilst it would still be a bit market-distorting to have so many offers congregated at one point it might at least stop choking the life out of the swap market quite so much, and make it a viable proposition to set a fixed-rate offer higher than the FRR.

I would like to put the same question out to you (and every other swap lender): Can you think of any way to force/compel/prod Bitfinex to address this issue?

From my vantage point, there has been a very good discussions of the FRR, it's effects and possible alternatives over the past few days. Bitfinex has been noticeably absent.  If they have a viewpoint, even if it is that the FRR will remain unchanged, it would be encouraging if that would say something.

[QwertyCore]

How about as a first test measure "FRR = (average of all current swaps/85)*100"? This would mean that the 15% fee of Bitfinex is added statically on top of the FRR, as currently if you lend at FRR, you receive only 85% of the average rate in reality.

This would mean the FRR rises the more it gets used and no fancy programming would need to happen (unlike "FRR +/- X%" scenarios, that can also end up cancelling each other out).

Are you proposing including the FRR swaps in the calculation and not just the fixed rate swaps?

I don't see how your calculation will raise the rate due to more FFR swaps being taken. I think the FRR would get a one time boost when the new formula is implemented and then will proceed to drop just as it does now.

[bubfranks]

Is a "flash return rate" useful for efficient price discovery? If not, get rid of it. The market would adjust.

Anybody else having problems withdrawing? http://www.reddit.com/r/Bitcoin/comments/2jtfqn/bitfinex_not_processing_withdrawals/

[daddyfatsax]

Is a "flash return rate" useful for efficient price discovery? If not, get rid of it. The market would adjust.

Anybody else having problems withdrawing? http://www.reddit.com/r/Bitcoin/comments/2jtfqn/bitfinex_not_processing_withdrawals/

I did a withdrawal about an hour ago and had no issue.

[noggin-scratcher]

How about as a first test measure "FRR = (average of all current swaps/85)*100"? This would mean that the 15% fee of Bitfinex is added statically on top of the FRR, as currently if you lend at FRR, you receive only 85% of the average rate in reality.

This would mean the FRR rises the more it gets used and no fancy programming would need to happen (unlike "FRR +/- X%" scenarios, that can also end up cancelling each other out).

Are you proposing including the FRR swaps in the calculation and not just the fixed rate swaps?

I don't see how your calculation will raise the rate due to more FFR swaps being taken. I think the FRR would get a one time boost when the new formula is implemented and then will proceed to drop just as it does now.

Taken as written, without including FRR swaps in the average (setting a variable rate to the average of a bucket of rates including itself just seems too self-referential to work), it would at least move the wall up to "slightly above average", allowing swaps in the slender middle-ground between "average" and "FRR" to be taken, and hence allowing the FRR to move upward without the wall first being eaten.

It's not quite what I had in mind but I'm not sure how my thought could be implemented  mathematically - I'd want the wall to move upwards as it gets consumed, regardless of what fixed-rate swaps are doing.

[mjr]

To put the FRR problem a slightly different way... it's using "the average rate of all fixed-rate swaps" to try and approximate the equilibrium point between supply and demand, but it's got its finger on the scales of that equilibrium because so much of the supply and demand it's trying to measure is provided/satisfied by FRR swaps.

Currently, so long as there's a sufficient wall of FRR offers, you can take out a million-dollar swap and leave the rate unaffected... whereas if you did the same thing to the fixed-rate offers it'd tear through to the astronomical rates (if not the entire book of offers) and yank the average abruptly upward. If it could be reformed in such a way that the flash-rate moved in response to offers being taken at the flash rate then... whilst it would still be a bit market-distorting to have so many offers congregated at one point it might at least stop choking the life out of the swap market quite so much, and make it a viable proposition to set a fixed-rate offer higher than the FRR.

I would like to put the same question out to you (and every other swap lender): Can you think of any way to force/compel/prod Bitfinex to address this issue?

From my vantage point, there has been a very good discussions of the FRR, it's effects and possible alternatives over the past few days. Bitfinex has been noticeably absent.  If they have a viewpoint, even if it is that the FRR will remain unchanged, it would be encouraging if that would say something.

I have mentioned my thoughts on this numerous times, and we are listening to your discussion as we plan on how to improve the FRR.

[mjr]

How about as a first test measure "FRR = (average of all current swaps/85)*100"? This would mean that the 15% fee of Bitfinex is added statically on top of the FRR, as currently if you lend at FRR, you receive only 85% of the average rate in reality.

This would mean the FRR rises the more it gets used and no fancy programming would need to happen (unlike "FRR +/- X%" scenarios, that can also end up cancelling each other out).

Are you proposing including the FRR swaps in the calculation and not just the fixed rate swaps?

I don't see how your calculation will raise the rate due to more FFR swaps being taken. I think the FRR would get a one time boost when the new formula is implemented and then will proceed to drop just as it does now.

This is one of the main issues. Our goal is not to raise the rates, or lower the rates. Whatever the rate is, we would like to see it adjust to market conditions. That is one of the main problems with this discussion, is that we are looking for ways to improve the FRR, not improve returns for people providing swaps. Most suggestions are looking for ways to increase returns for swaps providers, and while many of those suggestions could work, they are not actually answering the question that was asked, how do we improve the calculation of the FRR? In my opinion, as bitcoin goes more and more mainstream, I think the swap providers will be competing with more of the open market for returns, and I don't see how rates can stay at this level, when they start competing with the expected returns that most people are used to.

That being said, I think we are getting close to an announcement on how we will adjust it. Stay tuned.

[DoubleSwapper]

How about as a first test measure "FRR = (average of all current swaps/85)*100"? This would mean that the 15% fee of Bitfinex is added statically on top of the FRR, as currently if you lend at FRR, you receive only 85% of the average rate in reality.

This would mean the FRR rises the more it gets used and no fancy programming would need to happen (unlike "FRR +/- X%" scenarios, that can also end up cancelling each other out).

Are you proposing including the FRR swaps in the calculation and not just the fixed rate swaps?

I don't see how your calculation will raise the rate due to more FFR swaps being taken. I think the FRR would get a one time boost when the new formula is implemented and then will proceed to drop just as it does now.

This is one of the main issues. Our goal is not to raise the rates, or lower the rates. Whatever the rate is, we would like to see it adjust to market conditions. That is one of the main problems with this discussion, is that we are looking for ways to improve the FRR, not improve returns for people providing swaps. Most suggestions are looking for ways to increase returns for swaps providers, and while many of those suggestions could work, they are not actually answering the question that was asked, how do we improve the calculation of the FRR? In my opinion, as bitcoin goes more and more mainstream, I think the swap providers will be competing with more of the open market for returns, and I don't see how rates can stay at this level, when they start competing with the expected returns that most people are used to.

That being said, I think we are getting close to an announcement on how we will adjust it. Stay tuned.

I am not necessarily trying to improve the rates of swap providers here. It's more of a fact though that the FRR is artificially surpressing the swap rate because of the way it works. This can be observed very well and in fact today was one of the best days to observe it.
With the FRR gone because of large demand in the past days only then we had large (and not really healthy or rational) upward spikes to astronomical rates because there is no incentive to put in orders at the top to create a deep order book.
Now that some large sums have been returned and were immidiately stacked up on the FRR rate we have a huge fastly downward moving wall which won't be eaten any time soon. This would have never happened without the FRR and because of the way the rate is calculated and every lender naturally trying to get his money lend out we are ALWAYS moving downward until we get so low that the FRR dries up and gets eaten. That's not how it supposed to happen and I don't seee any way why it has to be like this.
We had a 3m FRR rate for quite some time and now have just build up another 1m in less than one day. We would almost never see those huge walls without the FRR (in fact I've only ever once seen a single 7 figure offer in the last year). It's foolish to believe that such walls don't have an influence on the market automatically (and the influence is logically downward).

[nrd525]

All this talk of a FRR wall causing the rates to decline always strikes me as very silly. Why does the wall exist?  It is because there is excess supply due to swap rates being 2-10 times higher than the market rate, and due to friction costs of moving money in and out of Bitfinex.

I think the FRR wall isn't suppressing interest rates - in fact it is preventing them from falling fast enough to balance supply and demand. This was especially noticeable with the long decline from 0.1% to 0.04%. The 4 million FRR wall was a product of declining demand (due to the bear market), excessive supply,and a FRR rate that was too high.

[mjr]

All this talk of a FRR wall causing the rates to decline always strikes me as very silly. Why does the wall exist?  It is because there is excess supply due to swap rates being 2-10 times higher than the market rate, and due to friction costs of moving money in and out of Bitfinex.

I think the FRR wall isn't suppressing interest rates - in fact it is preventing them from falling fast enough to balance supply and demand. This was especially noticeable with the long decline from 0.1% to 0.04%. The 4 million FRR wall was a product of declining demand (due to the bear market), excessive supply,and a FRR rate that was too high.

I think that both of these are right. The key issue with the FRR is not that a lot of money sits there. It is that it adjusts too sluggishly. So it does keep the rate lower when it is rising, and keeps it higher when falling. We are mainly focusing now on how to make it more agile, because we still think it is a useful tool, I also really appreciate someone mentioning that if there is a wall, that must logically mean that there is an excess of supply and rates should be lower. If there was demand at that level, there would not be a wall. I am somewhat curious about the price sensitivity of the margin traders and I wish I knew more about how many people would actually not place a trade because they don't like the swap rate.

This is an interesting discussion, and I am still listening to anyone else's thoughts.

-Josh

[noggin-scratcher]

It's undeniable that a wall means an excess of supply at that price point, but I think it's arguable that it's being supplied at that particular price point only because the FRR is the only option for automated swaps, and has bent the market around itself. It's not that lenders have come to a reasoned decision that the FRR is the price point they want to set their rates at, it's just the only way to pick a rate automatically that's minimally market-sensitive. Making it more sensitive, so it adjusts upwards more quickly after it gets eaten, would be a positive improvement... prevent those somewhat-ridiculous occasions where the rates offered are "FRR: 0.06% ... nothing until ... fixed rates: 0.6%"

I'm wondering now if there's any improvement to be had by drawing a parallel to TCP's rate-setting mechanism. For those who aren't familiar with networking protocols, TCP carries traffic across the internet and dynamically adjusts the rate it sends data at to try and saturate your connection while avoiding causing congestion. Each time it gets an acknowledgement of a successfully received packet it increases by a linear amount, but as soon as it detects failure it "backs off" by cutting its rate in half to rapidly alleviate the congestion it assumes caused the packet loss (it's more complex than that in reality, but that's the simple version).

By analogy, if we take "the wall" as a simple fact of life, and treat "the wall has been exhausted" as a strong signal that the FRR rate is too low... could try simply doubling the rate whenever the total offered at the FRR goes to zero. Would solve the problem of it not responding quickly enough to increased demand. Although, I'm not sure how to handle the "linear decrease" portion once it's uncoupled from the average of fixed-rate swaps, and anyone with substantial existing FRR swaps would have quite the incentive to trigger the doubling mechanism as often as possible. Would at very least need to cap how often it can happen to prevent it being gamed.

[Killerloop]

Hello, I have a problem with the API, "POST /balances" call... it only appears to return balances in BTC and USD, no LTC/DRK/TH1.

Is there some extra fuction or parameter I am missing?

[vernes]

Small noob question here... searched allover the web but couldn't find anything

I send 2 bitcoins to my Finex account My balance is showing that I have 2 bitcoins


BTC   2.00
BTC Swap total:  2.00
BTC Swappable balance:  0.00


However when I create a Swap (30days,0.003,1order of 1 bitcoin)  order it states I do not have sufficient balance.

What did I do wrong>

[noggin-scratcher]

BTC Swap total:  2.00
BTC Swappable balance:  0.00

This suggests to me that you already have an active swap using those 2 BTC.

Go to the swaps page, make sure you're on the "BTC" tab, then click to open the list of "Swaps currently provided" - is it in there?

[vernes]

Yes, it took a while to show, that's what got me confused.

but now everything is fine, thanks for the help

[QwertyCore]

All this talk of a FRR wall causing the rates to decline always strikes me as very silly. Why does the wall exist?  It is because there is excess supply due to swap rates being 2-10 times higher than the market rate, and due to friction costs of moving money in and out of Bitfinex.

I think the FRR wall isn't suppressing interest rates - in fact it is preventing them from falling fast enough to balance supply and demand. This was especially noticeable with the long decline from 0.1% to 0.04%. The 4 million FRR wall was a product of declining demand (due to the bear market), excessive supply,and a FRR rate that was too high.

I think that both of these are right. The key issue with the FRR is not that a lot of money sits there. It is that it adjusts too sluggishly. So it does keep the rate lower when it is rising, and keeps it higher when falling. We are mainly focusing now on how to make it more agile, because we still think it is a useful tool, I also really appreciate someone mentioning that if there is a wall, that must logically mean that there is an excess of supply and rates should be lower. If there was demand at that level, there would not be a wall. I am somewhat curious about the price sensitivity of the margin traders and I wish I knew more about how many people would actually not place a trade because they don't like the swap rate.

This is an interesting discussion, and I am still listening to anyone else's thoughts.

-Josh

The cost of swaps has a minor role in the demand for swaps.  The demand for swaps is driven almost solely by the price movement of bitcoins, it is not dependent on the swap cost.  If bitcoins are going down, no trader is going to show an interest in taking a swap even if its 0.0001%.  The swap market is not driven by supply and demand of swaps until the price or perceived price of bitcoins is rising; it subservient to the whims of the bitcoin market.

Once that demand has be created, by the bitcoin price, not by the swap price, the acceptable swap cost is dictated by how much or how fast a trader thinks the price of bitcoin will increase.

To put this as simply as possible, market for swaps is driven by the rising price of bitcoins, not the cost of swaps.  Can we agree on this?

[eneloop]

Noob question:
What happens if someone can't repay? There is no swap-taking-user-information available.

[Sukrim]

Noob question:
What happens if someone can't repay? There is no swap-taking-user-information available.

They get liquidated before that can happen. If that is not enough, Bitfinex claims to have some vague "insurance" that would cover the difference.

[eneloop]

Noob question:
What happens if someone can't repay? There is no swap-taking-user-information available.

They get liquidated before that can happen. If that is not enough, Bitfinex claims to have some vague "insurance" that would cover the difference.

Thanks!