My 4 month experience as a liquidity provider

[mhps]

if BKS holders decide to give NBT liquidity operations reduced or no trade fees.

That is a way for Nu to generate income. We all agree that Nu should find more ways. many potential ones … setting higher spread, asking exchange to give a cut of trading fee, generating more on-chain tx fee by increasing fee rate or increasing transaction, lending, trading by payment processors …

[Dhume]

Cutting down operational costs would also help, using our own funds instead of borrowing large sums of money every month should help reducing our costs.

[mhps]

Cutting down operational costs

I think making it profitable to sell insurance is an effective way

Incentivise pool cost reduction

Currently Nu pays 6% - 9% per month interest to LPCs to provide liquidity, and pays 1% for pool managers. There doesn't seem to be more incentive to reduce interest cost except for shareholders (which all pool managers are) altrruism. I think if we give incentive to people who can reduce liquidity interest cost, Nu will provide liquidity more efficiently. Let's look at the interest in more details, taking NBT/BTC pair - The interest is for compensating exchange default risk, which is con…

[dysconnect]

I think we can all agree that BTC/NBT is very costly not just to Nu but also to the liquidity providers. So what is your opinion of moving away from BTC/NBT on other exchanges once B&C exchange starts picking up some volume?

No I can’t agree it’s costly to liquidity providers. You might have a point when NuPool is buggy, but not for NuLagoon. Did you read my post again? And reconsidered the issue with accounting in BTC? For the 6 months where BTC when down and up and down and up in the 220 - 270 range, NuLagoon’s Pool A NAV has increased from 1 to about 1.2. Am I making it simple enough for you?

[Dhume]

[quote=“dysconnect, post:24, topic:2964, full:true”]No I can’t agree it’s costly to liquidity providers. You might have a point when NuPool is buggy, but not for NuLagoon. Did you read my post again? And reconsidered the issue with accounting in BTC? For the 6 months where BTC when down and up and down and up in the 220 - 270 range, NuLagoon’s Pool A NAV has increased from 1 to about 1.2. Am I making it simple enough for you?
[/quote]

I’m not sure why you seem to have a personal vendetta going on against me here, I did not make this thread to blame or complain against anyone or anything. What I’ve merely stated is my experience as a liquidity provider and why I feel pegging BTC/NBT is more costly then we perceive it to be.

You might want to consider that people who put in BTC for liquidity provision party do so because they are in BTC anyway (for various reasons) and see liquidity provision as a way to earn some additional interest (which comes at a
considerable risk) over their BTC. When the net result is traders are losing BTC it means people will be less willing to put in BTC.

We are supposed to be paying interest to cover their costs, you feel that a net result of losing BTC is justified if its NBT value went up I disagree. Let alone the risk of exchange default which could lead to a complete loss of funds for the liquidity provider.

Now we can debate this all we want and we are probably not going to see eye to eye, so lets focus the discussion on the other reason why I started this thread. Namely a discussion about a possible future business model for NU and how we are going to deal with liquidity provision and defending the peg.

[mhps]

For the 6 months where BTC when down and up and down and up in the 220 - 270 range, NuLagoon’s Pool A NAV has increased from 1 to about 1.2

Why 6 months? The last 6 months is when BTC bottomed out and climbed from 220 to 400. Why not imagine what could have happened in the last 1 or 2 years? LPs would have had 50 - 70% losses.

[Nagalim]

people who put in BTC for liquidity provision party do so because they are in BTC anyway

right, we need to attract more custodians that hold NBT rather than BTC. If those people don’t exist, how do we create them? Perhaps non-zero park rates?

[Dhume]

[quote=“Nagalim, post:27, topic:2964, full:true”]right, we need to attract more custodians that hold NBT rather than BTC. If those people don’t exist, how do we create them? Perhaps non-zero park rates?
[/quote]

One solution that I’m proposing (not for right now but in the foreseeable future) is to consider to start using our own funds rather than borrowing funds to defend the peg. If B&C leads to increased demand and starts seeing some decent volume we should have a huge stream of revenue coming in. Part of which we could use to reduce or remove our reliance on liquidity provisions and reduce our operational costs since we don’t have to pay interest to ourselves.

[Nagalim]

our own funds

So what happens when exchanges get hacked? Isn’t that exactly what we were doing 6 months ago when we decided to move to a trustless model? BTW, I assume by ‘our own funds’ you mean T4, cause we don’t have any other funds.

[Dhume]

[quote=“Nagalim, post:29, topic:2964, full:true”]So what happens when exchanges get hacked? Isn’t that exactly what we were doing 6 months ago when we decided to move to a trustless model? BTW, I assume by ‘our own funds’ you mean T4, cause we don’t have any other funds.
[/quote]

I’m talking about doing this on B&C which if everything works as intended can’t be hacked. Well T4 is relatively small now but say demand go’s up 300k suddenly we have 300k revenue of which a part we could use to defend the peg.

[Nagalim]

I agree that we should increase T4 as the NBT marketcap increases. However, I disagree fundamentally that Nu should be using T4 constantly. If anything, I think the threshold by which we engage T4 should be decreased to 35% instead of 40%.

pay interest to ourselves

Who is going to run the bot that controls the funds? That person will likely want to be paid, and will constitute a centralized point of counterparty risk. The main argument I’m making here is that it shouldn’t matter whether we use T4 funds or T1-3 funds, either way the result is the same that the network will experience a certain loss from volatility. Using the decentralized T1-3 scheme instead of T4 for normal every day operations, however, greatly increases our flexibility by having many people involved instead of few. Given two choices equivalent in all other ways, I will choose the more decentralized one every time.

[Dhume]

I agree that we should increase T4 as the NBT marketcap increases. However, I disagree fundamentally that Nu should be using T4 constantly. If anything, I think the threshold by which we engage T4 should be decreased to 35% instead of 40%.[/quote]

Well we could have the tiers remain the same just that tier 1-3 would be filled with our own funds rather than borrowed money.

[quote]Who is going to run the bot that controls the funds? That person will likely want to be paid, and will constitute a centralized point of counterparty risk. The main argument I’m making here is that it shouldn’t matter whether we use T4 funds or T1-3 funds, either way the result is the same that the network will experience a certain loss from volatility. Using the decentralized T1-3 scheme instead of T4 for normal every day operations, however, greatly increases our flexibility by having many people involved instead of few. Given two choices equivalent in all other ways, I will choose the more decentralized one every time.

I envision a lot of different liquidity providers (or should I say custodians) holding small percentages of the funds, in order to protect us from theft they’d have to put in some collateral. Running a bot is not that difficult once its set up. I can’t imagine not finding a decent amount of shareholders willing to put in some collateral and run a bot for a fee much smaller then we pay to our LP’s currently. We’d still have decentralization, running it basically the same way only with reduced costs.

A nice bonus would be a more secure peg since shareholders running a bot have no reason to ever pull funds. This would make our peg a lot more secure.

[Nagalim]

collateral.

If you figure out the collateral question, you figure out how Nu can make a business loaning NBT. This is not a trivial concern. Banks enforce this with repossessers and the court system, Nu has no such mechanisms. Bitshares is based off this very concept, if we could duplicate or improve upon it in Nu it would mean a lot of profitability.

[mhps]

I can’t imagine not finding a decent amount of shareholders willing to put in some collateral and run a bot for a fee much smaller then we pay to our LP’s currently.

What’s your reason? You stated that current fee (7%) for currrent exchanges is not enough. Let’s say we need x% extra to break even. Then if you agree B&C only cuts ~4%/mo cost, Nu would still have to pay more than (3+x)%/mo (36+12x% pa) for custodians on B&C at least. Collaterals incurrs opportunity cost. Placing orders on B&C have cost… It won’t be much cheaper however you cut it.

[dysconnect]

We are supposed to be paying interest to cover their costs, you feel that a net result of losing BTC is justified if its NBT value went up I disagree. Let alone the risk of exchange default which could lead to a complete loss of funds for the liquidity provider.

It’s not exactly personal, but I tend to get touchy when the frivolous spending on liquidity interests is said to be “not enough”. Especially with an argument that I can’t accept. Even so when you said “I think we can all agree” despite seeing my “vendetta”.

I don’t buy your point; people have varying investment interests. Those who are “in” BTC but don’t hold enough would be bitter when BTC rises, but they can laugh at others who went all in while it tanks.

It’s about risk aversion, and those with the right investment appetites would find it justified. There exist plenty of people who would account in NBT and be pleased with the rise in NBT value despite the exposure to BTC, and they can come in if people who prefer to speculate on BTC don’t want to.

To put it in an other way, the interests are there to compensate people so they agree to sometimes hold the volatile, unstable BTC instead of hedging with some other things. Not the other way round. There may be ways to fine tune the system so it can be customized for different appetites, there’s a point on exchange default (though I don’t think our interests are too cheap for that), but I categorically reject any argument based on accounting in BTC.

Finally I don’t believe people can predict BTC price consistently. It’s the sole motivation of creating a stablecoin. The “volatility” cost mentioned by @Nagalim is basically that liquidity providers are sometimes forced to buy high and sell low. Sometimes it’s because the price updates lag behind, sometimes it’s because traders won a bet on BTC price. This can be combated with higher spread, though still leaves behind an issue: some liquidity providers would end up on the bad side of trades more often than others, so if we’re to make everyone happy we need to raise interests to compensate for bad luck, rather than the average case, which is costly.

using our own funds

It’s all about decentralization, and something like NuPool already removes one layer of counterparty risk. In other words, “using our own funds” isn’t feasible until B&C comes out. Even with B&C I think this form of liquidity provision is still going to be a preferred in many cases.

And NuLagoon is more centralized but it’s the only way to get something like Pool C/D at the moment.

[dysconnect]

I was talking about March to September. Or if you look at 2015-04-09 to 2015-10-08, the BTC prices at the end points are about the same, but swung back and forth in between, but the NAV went from 1.055 to 1.283. This means, the volatility risks did not dominate the interests.

[dysconnect]

I’ll add here that despite our disagreements on this matter I still welcome you in FLOT and encourage you to propose a motion to designate yourself as a signer, like this. My apologies if my criticism was too personal.

[Dhume]

If you figure out the collateral question, you figure out how Nu can make a business loaning NBT. This is not a trivial concern. Banks enforce this with repossessers and the court system, Nu has no such mechanisms. Bitshares is based off this very concept, if we could duplicate or improve upon it in Nu it would mean a lot of profitability.

Put the collateral in a multisig group, akin to how we want to handle the reserve funds. Person x applies for Liquidity Custodian position of xxx amount of funds, puts in xxx amount of Nushares roughly equal or a factor x greater than the grant amount and receives either nubits or bitcoins corresponding with the size of the grant.

Person x puts the funds into his B&C exchange account designated for the liquidity provision post his account details that enable us to check what kind of orders are being placed and if he’s not stealing funds from the account (since if I understand correctly everything should be on the blockchain so I’m assuming account info is as well).
Once a month he reports operations (we can check his account info to verify) and applies for additional funds in case some of the funds have been lost due to operations. In the same proposal he could also request Blockcredits for the account so blockshareholders could if they so choose grant these custiodians Blockcredits in order to have reduced or 0 trade fees.

What’s your reason? You stated that current fee (7%) for currrent exchanges is not enough. Let’s say we need x% extra to break even. Then if you agree B&C only cuts ~4%/mo cost, Nu would still have to pay more than (3+x)%/mo (36+12x% pa) for custodians on B&C at least. Collaterals incurrs opportunity cost. Placing orders on B&C have cost… It won’t be much cheaper however you cut it.

Lets say we use NSR as collateral then a fair amount I feel would be say a 0.5/1% to cover running the bot and not having access to the funds combined with the amount of NSR the person would receive minting in that same time period for which the grant is valid. This combined with a parametric order book (which from what I understand is possible on B&C) should offset the remaining costs significantly.

I’ll respond later, time to sleep now after look at BTC charts all day.

[Nagalim]

Put the collateral in a multisig group, akin to how we want to handle the reserve funds. Person x applies for Liquidity Custodian position of xxx amount of funds, puts in xxx amount of Nushares roughly equal or a factor x greater than the grant amount and receives either nubits or bitcoins corresponding with the size of the grant.

So in bitshares, this is called a ‘short’, because you’re essentially selling the nushares for nbt. I can give my NSR to Nu and get rewarded with other types of currency, which I can then walk away with. The only thing stopping me is the factor ‘x’, which is precisely what BitAssets has been dealing with in version 2.0. You are looking to create a virtual exchange, which would mean we’d need to basically build BitAssets inside Nu.

[cryptog]

At the end of the day, what matters is whether or not Nu makes some profits.
If higher rewards for liquidity providers are needed to generate liquidity that then causes more traders to use nubits, which then in turn induces new nubits sales, then they should be increased, as long as nu is in the black. Right now, we have more buy side than sell side despite the btc rally. So it s encouraging.
Of course, the lower the costs, the higher the profits.
So b&c should help in this reagard.
It seems that getting away from nbt/btc will not happen tomorrow.