Topic: [OFFICIAL]Bitfinex.com first Bitcoin P2P lending platform for leverage trading - page 104. (Read 723903 times)

So, I'll keep this short, since the rates are too low for me to find interest in discussing margin lending anymore, but you did specifically mention MarginBot, so I'll respond.

If everyone who was currently using FRR switched to a bot, I can pretty much guarantee rates would be more responsive, and in all likelihood, higher, for a couple of reasons.

1) Minimum rates - you can't set this with FRR.  You claim people are happy to take what they can, and to some extent that's true, but part of this is because they CAN'T choose the minimum they'll take with FRR...  This option alone would have a HUGE effect on rates, even if people literally installed the bot, added their API Key and walked away, the rates would be floating closer to .065% (default minimum).  One way or the other, this would at least make people spend just a few seconds thinking about what they're willing to take.
2) Spread Lending - right now all the money going to FRR is one big ass lump.  Spread it out, and the "wall effect" would go away, or at least be reduced.  Each upward step would have a smaller wall to climb, small runs would still move us up, not sideways.
3) Reactive lending - loan targets would be constantly moving, looking for new targets every few minutes, instead of every hour.  Rates would adapt quicker.

I could go into detail on lots of other, far more complex reasons as well, but honestly, I'm not particularly interested in this discussion.  I never was, but somehow got sucked in.  FRR is broken.  You talk about wanting a free market, but then manipulate the hell out of it with the FRR.  Honestly, have you ever seen a financial market anywhere else in the world with an auto-averaging system like this?

Like I said, I wish I could. Unfortunately, it is not that simple, you have a couple million in offers at that rate, just what...cancel their offers? The implementation of that, affecting potentially thousands of accounts, is kind of tricky, as are all large changes on a platform like bitfinex. We have been pretty stable over the years, haven't been hacked, and have had very little downtime. We don't like to take risks, we play it safe when it comes to an ecosystem that a lot of people rely on. But, if the FRR were magically removed, just so we could prove a point, currently, it would change nothing unless your order was going long for more than ~$300k. The reason I wish we could see this, is because I wonder what rate the almost 3 million at the FRR would pick...I am guessing it would prob be around 0.025%. If THAT were true...then in order to make rates go up, it would take an order of around 3 million dollars to move the rates (assuming they all simultaneously put in their offers at the same time, if they put them in over time, you could see the rates go much lower).

People who are content to take whatever they can get, will probably take whatever they can get...another alternative is that it is too much work and that 3 million vanishes from the book. Either way, it doesn't change the fact that some guy is perfectly content offering a swap for .025%

I am not sure if that is currently on the list, but it definitely will be. There have been quite a few things on the frontend that we want to revisit, but its all hands on deck for the backend upgrade, so I definitely think those suggestions make sense, especially as we see more institutional and business clients. It would make sense to be able to split up privileges for different functions.

I am not aware of the VIP program at okcoin...what do they offer? Currently, if you are a large trader (25k coins per rolling 30 days) you pay no fees to place an order on our book. Also, through the affiliate program, if you are introducing people to Bitfinex, you are rewarded with a portion of their fees. I believe you have to request access to that. If you can let me know exactly what the VIP program is, I will definitely look into it more. 

Not sure about Bitstamp, when I read that story, I didn't even think about that, because it has been so long since we really did much over there. I will get back to you, but I don't believe that it has any effect on us.

So, lets talk about the FRR stuff (again).

So, if the FRR is broken, and with an increase in demand, the rates responded and now the FRR for BTC is twice that of the FRR for USD. That is exactly what I would have expected to happen, with or without FRR. Oh, a lot of people want to short, the rates for shorting went up. I think that the FRR probably acts as a counterbalance against large swings, dampening effect either way, but with autolending bots all over the place, I really think it is completely backwards to want to get rid of it. You are literally asking for the people who are not managing their funds, content to sit on the sideline, to be forced to jump into the market. In other words, those people are NOT competing, and it is preferable for them to start competing. I don't see how that would possibly help people who want rates to go up. All of a sudden, they have to pick their own rates, and don't really care that much, as evidenced by the fact they use the FRR...so they should download any marginbot, set it to "as long as i get something", and that would get rid of the downward pressure? I understand you don't like FRR, but how do you account for ALL the offers IN FRONT of the FRR...those people are CHOOSING a lower rate. If you pick whatever rate you like, I don't think that most people are going to say, oh, well if his rate isn't the FRR then I don't want to beat it...anyway, coming from the perspective of someone who doesn't care at all what the rate is, I don't see how that is better. I think the fact that the rates have risen, basically shows, definitively, that with enough demand to overtake the supply, the FRR won't stop rates from rising. The rates can rise, if demand is high enough, as demand for longs sagged, they just weren't enough to raise the USD rate...the USD rate would have fallen either way.  I'm not sure why rates dropped (have to look into that), but in the short term, yes, rates are erratic, over the longer term, they make more sense.

Short answer: Demand for BTC grew, rates rose. Demand for USD dropped, rates dropped. That is what I would expect.

The other thing you mentioned. Which is separate from the FRR. I think that is a GREAT idea. Very useful and constructive. I think separating out the books makes sense (somewhat), and in theory, would be great. I am not sure if it is that easy to do, however. Either way, it is definitely worth talking about, and looking into. I don't really see a problem with it, but I have just started thinking about it this morning. These are the kind of ideas I would really like to hear more of. I have wondered if there was a time premium placed on longer length swaps, and what exactly it was. This would be very helpful in getting more data, and probably help price discovery (how much of the rate is specifically because I am willing to give up use of funds for 30 days vs 2). The thing though, is that if a person seeking a swap is going to simply take the best rate possible from any of the books (per the example), then aren't we really just talking about a different view of the same book that exists now? I can choose which offer I want to take, and I specify the time. So, if people automatically take whatever is lowest on any of the separated books, all the individuals are still competing with each other. I am just starting to think about this, so I could be wrong, lets talk more about this though.

Please, since there would be no difference in your opinion as well: just remove it and prove me wrong.
If it's the same situation, there's no harm trying, right?

Can you tell us if API security is coming soon? Like a key for swap deposit operations only, another for exchange only, one for margin trade only. Also the ability to "lock" your account type [ deposit/exchange/margin ] will be awesome.

Thank you for your time,
S.

PS:. Please VIP program for large accounts like okcoin

If this is actually happening...I think the people who should plan for this are the people who are offering swaps.

If you believe:
A) There is a sudden exodus of funds about to happen due to rates decreasing.
B) There will be a shortage of funds, and margin traders will need to pay much higher rates to be able to open a position

Then the liquidity provider can plan for this by simply putting out an offer at a much higher rate. Once the lower rate offers exit en mass (why they currently have offers that aren't filled, and are content to leave those offers on the books...its not my theory, I don't know) you are now at the top of the book. Congrats, you are now receiving the returns that you think are appropriate.

The reason this doesn't work, and why it won't work, is because there is no shortage. There is more supply than is currently demanded. We DID see the opposite however. The BTC swap market was much thinner, but lo and behold, when demand for short positions started increasing, somehow, BTC appeared on the books...

If you believe, as I do, that markets are usually the most efficient method of allocating resources, then you see that there is NO problem with supply or demand. The rate will move just enough to incentivize just enough dollars or BTC, and you can see this in the BTC swap market. The rates were, if i remember correctly, 10% of the return that you could get on the USD side. They are now around double. People wanted more BTC short positions, people offered BTC as they saw they could get better rates, as more people jump to that side, the better rates will dry up, until people stop offering.

So now, Bitfinex, tell us, that Bitstamp is down and possibly hacked: Do you have meaningful funds left on Bitstamp that could cause liquidity problems for your operation?




+++++
Barabbas is making some clever and useful proposals for changing the swap market and making it more efficient. I've already given up on BFX changing anything fundamental though and now it's merely a question for me whether keeping the funds on the platform is worth it or not.

Funny thing is that now that BTC swaps have risen in interest rate they seem to suffer from the exact same FRR related problems USD swaps have been plagued with. I'm wondering whether BTC swaps have only been so low the entire time because of the FRR related mechanics as even mjr seemed to be wondering about the normally super low rate of BTC swaps.

Hi all,


Ok, First, I am going to (again) talk about the FRR...if you read my other points, basically the same thing I usually say (tl;dr markets set rates, sorry if you don't like what is currently available, thats how markets work). If that bores you, please feel free to skip to the bottom where I will discuss our new plans for the upcoming year.

I still don't understand the debate about this. If the rate is low, you express that opinion by either leaving an offer at a higher rate, or simply not offering at all. The rate wouldn't be, and in fact COULD NOT be this low, if people weren't clicking buttons and entering a lower rate than you would prefer. People are CHOOSING to offer funds at these rates, so the fact that these rates EXIST means that they aren't too low.

Again, whenever we discuss this, people take the opinion that we should have a rate in mind. We shouldn't. If it is too low, then you are free to not take it. What you are seeing is the greater supply chasing the limited demand. This is classic, texbook even, supply and demand. You can blame the FRR all you want, but honestly, I have always felt that the rates people had seen in the past were really great...in other words, you were taking advantage of a market inefficiency. As people discovered it, it goes away. I currently have a LOT of people who are specifically interested in offering out millions, they find the rates very acceptable, and are curious how they can proceed as a business.

If you, today, got rid of the FRR, the rate would not change at all...because the FRR is not the rate that you are currently competing with.

Total Number of Swap Contracts    Number of Swap Contracts    Rate           Time    Offer Count
427.79                                  427.79                          0.0253%   2      1

It was whichever random user decided that for HIM, 0.0253% was a good enough return for his $427.79. Without the FRR, this would be EXACTLY the same situation you currently face. I find it funny though, we are in a HUGE bear market, and people are wondering why the demand for going long (and with it the subsequent rate for a USD swap) are dropping, while, for the first time in a LONG time, the demand for BTC is up...its almost as if there is some sort of correlation between the demand for a certain position, and the rates that people are willing to give...

In a previous post, I specifically said that I thought it was very strange that the BTC rate was so low. I thought it was an inefficiency, in that if I had to choose to short or long over the next 30 days, I would specifically look for shorting opportunities, because they cost a lot less. I honestly wish we could get rid of the FRR, just to prove that it wouldn't make any difference at all...is it an ideal tool for calculating a "going rate" for swaps, NO. Could it be better, yes! But you will see that the rates are just reacting to the market. There is a finite, but unknown, demand for going long, whoever is willing to offer a swap for the lowest rate gets to be first in line to fill that demand. For a lot of people, something > nothing.

One last point, I find it funny that people think that for us this is some sort of financial move or something, but then point out that we could make more by raising rates. That is mathematically correct. Our rates are 15% of the returns generated, if we increase your returns, we would get 15% of a larger number. This is math, and we are aware of this. We have no interest in raising rates, in order to pad our cut. We would rather that supply and demand lead to whatever rate it leads to. In other words, a market.

Now that I have responded (again) to the FRR discussion. I want to wish you all a Happy New Year.

We have a live, working version of our new backend, and are currently testing it. It looks fast...REALLY fast. This is going to open up a lot of doors for us, in this new year.

We are going to be able to offer a FIX gateway, for institutional players who have been patiently waiting for one.

We will also have an official websockets API.

We are currently looking into other pairs that we may offer, including other fiat currencies.

We will finally be able to divert some resources into the growing list of suggestions and requests, specifically better reports and more user customization in the interface. 

We are moving cautiously, we definitely don't take risks when it comes to our users funds and we appreciate the trust you place in us. I am SO excited about this, it is going to be a real game changer (sorry if I was grumpy in the first part, I am just really bored and annoyed by FRR talk, and the new backend is WAY more exciting and interesting).

I am looking into perhaps setting up a simulation/testnet environment, where perhaps some people could test it out, etc. No promises yet, but it is an idea I am throwing around. I know that a lot of people here have been with us since the beginning, and I really want to hopefully reward your patience, and show you guys what we have been working on. I know that things aren't perfect, I know that there are a lot of people who don't have things just the way they would prefer. I am sorry about that. We are working really hard, doing our best, to make the best exchange for bitcoin in the world. 2015 is going to be a great year, I hope. If you have more questions, feel free...

By the way...what are your opinions on the current prices? I am kind of shocked to see sub $300 bitcoins again. Any theories?

Wow, we've gone full swing to traders wanting outright manipulation and market fixing to protect their rates. The solution is quite simple, first, remove FRR. With the number of autolending bots, there's no need for it.

Second, put lend durations on their own books. Have, say, four lend books total, 2 day, 7 day, 14 day, and 30 day. This more accurately reflects how true fixed interest markets work. It would also prevent bid/ask from crossing. Additionally, because the lends are callable (the borrower can end the loan and return the funds at any time) borrowers can automatically take the best rate from any book they want. All the time duration risk lies on the lender and can be reflected in yield curve.

Did the algorithm change already, somehow?

Yesterday, while there was HUGE demand for BTC-swaps in the selloff to ~$260, the BTC-FRR sharply DROPPED from about 0.11% to 0.07%.

The only way this could possibly explain that imo is:
- FRR calculation model changed or
- someone accidently lent out a big amount of BTC for 0.01 instead of 0.1, or some similar mistake.

Didn't know upper limits weren't enforced! Then I change position on this. No limits at all.

The upper limit is also not "enforced", it just requires manual interaction. Requiring manual interaction for loans outside of a certain range of percentages could be a viable option. Currently there's only an upper limit - if you want to borrow funds above 1%/day, you need to go to the respective tab and manually accept the offer(s), then you can trade with these funds. Usually this is not necessary, as opening a position automatically grabs the necessary amount of the cheapest funds available for lending.

Lower limits concern lenders more, so e.g. Autolend could stop working if it would lend out funds below 0.1%/day. I don't see how this would work out in practice to limit traders to a certain automated offer-taking (imagine some funds being available at 0.05% and some at 0.11% - would traders now no longer be able to automatically fund their positions, would they just get the 0.11% despite a cheaper option being available or something else?). This also would offer some DoS/annoyance attacks, such as offering small amounts of USD at low rates to force borrowers to manually process their funds.

Regarding limits, I am either on the side of having no limits at all, or having both upper and lower. If the lending platform proves to offer stable rates without upward or downward spikes, as a result of manipulation or the inability to balance itself during times with high volume, then obviously there is no need for limits.

Which funds get lent out when bids and ask meet, if they are at the same price? Obviously, the moment there is an offer and a demand placed at the same rate, it gets executed.

Okcoin offer 20x leverage, so there it can happen quite easily that a position cant be fully liquidated when the price changes fast (a 5% move can force-liquidate, and 5% isnt much in bitcoin-world). An insurance helps then.
BFX offers 5x leverage. So the price would have jump by 20% to force-liquidate. No need for an insurance. Plus every insurance costs money.


I dont see how there can be a lower limit. Apart from disturbing a free market even more than the FRR-wall, I see practical problems:
What if there is a hypothetical 3M swap demand and 15M swap offers at a lower limit: which funds get lent out then? 20% of all available offers? In practice, you only have 1/5th of the interest rate of the lower limit then. This doesnt solve any problem.

Hello bitfinex,

I have some suggestions to make regarding the lending service and FRR.

There is a limit on how high the interest rate is allowed to be, but not how low. Of course there are reasons to put a cap on how high one can set the interest rate. People may get into a position where the only available option is 10000% interest per year, either by someone manipulating it, or by the demand increasing too suddenly. This would obviously make the margin trading platform look bad as profits may get diminished by interest. But, if the platform recognizes this, then why does it not recognize that interest rates may go too low, either by manipulation, or suddenly decreased demand, or - an unfortunate truth - the FRR pushing the rate lower than the actual decrease in demand.

The last part regarding FRR pushing the interest rate is true and is evidently something many are complaining about. There are several reasons why there should be a lower limit, since we recognize limits are necessary (there is an upper limit). I will make a list below

• Lending is risky. If someone gets margin called but there is not enough liquidity on the orderbook, the lender loses out.
• Added risk: lending is done on the bitfinex platform. Considering MtGox lost 600k BTC and tens of millions of USD, one is wary about leaving their funds on an online exchange
• Lending has to compete with other investment vehicles, and risk has to be considered here.
• The profitability of lending vs borrowing has to be balanced. Too high interest rates cuts traders profits, but too low rates make it almost free to borrow (a $10,000 loan at 0.03% costs a measly $3 per day)

I think 0.1% per day is a reasonable lower limit.

The second suggestion is my vote on how the FRR algorithm should work. Here is what is exactly happening now. If there are an equal amount of USD being lent and borrowed each day, say $20m, and $3m waiting to be lent, some people will by default try to get under the $3m FRR wall to get things going, and the FRR algorithm will see this as a signal to lower the rate - despite the fact that the demand or supply never changed. Once the rate drops a little, the FRR adjusts existing loans, making those who want to lend their USD quickly go even lower. Then this cycle repeats. There are charts showing that the demand and supply hasn't changed yet the rate drops consistently.

My suggestion is that the FRR algorithm should not be based on the average rates of previously lent money. Instead it should be based on the demand and supply of USD for margin trading. I mean this type of logic:
Let X be supply (available USD), Y be demand (new swaps per day and/or sum of swaps), Z be the lending rate

• If X increases while Y increases less, remains unchanged, or decreases, then the rate decreases proportionally
• If Y increases while X increases less, remains unchanged, or decreases, then the rate increases proportionally
• If both remain unchanged, the rate remains unchanged

Surely both lenders and borrowers would agree on a system which says "I'll loan to you at a rate depending on how much demand there is for my money". Having the ability to adjust the variables for this would be a plus, as some people would like their money lent as fast as possible while others want the highest rate.

Newbie questions here:

1. I see that Okcoin is offer a 10% insurance option, is there something similar offered by Bitfinex?
2. My understanding is that If I'm lending bitcoins the offers  are being matched by people who want to borrow bitcoins for them to open short BTC positions. The system is designed to close positions if traders are loosing and don't have enough money in their own account to cover the loses. However If the BTC price suddenly goes up is there a risk the system can not react fast enough and trader defaults on the borrowed amount?


Basically I want to understand what are all the risk involved for lending bitcoins (beside the obvious one of keeping them at a 3rd party).

 

hi

how about amount I can deposit USD and exchange toBTC?
1000000USD OK ?
1000k USD a month?
too big amount need more  vertification?
my account aleady vertified

/thread

That's why we can't have good conversations. When you use language like that, you demonstrate that you have no grasp on what currency actually is.

1) Put a USD Swaps demand offer for 30 days at the rate you want.
2) Buy all the Bitcoins
3) Enjoy the knowledge that you won't be paying more than 0.278% (like 10%/year) the next 30 days
4)
5) Take loss (or profit if this thing actually moves up)

See, the secret that is used so successfully by the trader elite is to reserve the USD at low rates for long periods before buying all the bitcoins. You may actually end up paying interest for USD you don't use for some hours or a day but that's life.

As for those who say "these rates are too low": People are, for some reason, willing to lend out at these rates. If you are a lender and you find these rates to low then go find something else to do with your digital american toilet-paper.

And there is generally no reason to worry about "Bitfinex running out of USD". If there is little USD available then rates go up and millions of fresh USD magically appears. It's supply and demand, it's that simple.