On the profitability of liquidity provision

[cryptog]

The success of bitcoin (so far) is due to a large degree to the fact that mining has been an economically profitable activity: miners secure the network and in exchange they get rewarded in bitcoins (that have a significant FIAT value because there is still a substantial demand for it), though it is not sustainable on the long run (this is what Sunny King foresaw, which gave birth to peercoin and then later on Nu…) but this is another story.

Likewise, I feel tat the success of nubits will be due to the fact that liquidity providing will be a profitable activity: liquidity providers secure the peg and in exchange they get rewards in nubits.

The issue is that liquidity providing is at this stage not predictably profitable.
Even if pools offer interest rates, the volatility of bitcoin is not predictable, making profits uncertain. This is another reason why, I believe, we have seen so few lpcs so far…

So the question I have: how are we going to make liquidity provision a predictably profitable business?
If it is not predictably profitable, the peg will not be kept and Nu will fail.

Some possible solutions:

• include into pools some tools to hedge but hedging looks complicated and costs.
• support only nbt/fiat but most exchanges do not have these pairs.
• increase the spread but it would make trading nubits expensive.

So are we stuck?

Well, decentralized exchanges could help us increase the spread while still keeping it low.
Decentralized exchanges would be able to offer very low trade fees compared to centralized exchanges.
That means, liquidity providers would be able to set a very low spread in order to compensate for that low trade fee.
So even if liquidity providers increase that spread, it would still be cheap compared to when they provide the same liquidity in a centralized exchange.
That increased spread could bring them the sufficient revenues to make liquidity provision a predictably profitable business.

What are your thoughts on that?

[Nagalim]

The trading fee is only .2%, you aren’t going to make a huge spread out of that. Hedging is our best bet because it is something that can improve with time as more and more people customize the system.
Even better, we can provide different compensation based on what tools are used. This will already be the case when we introduce those tools because of the competition model we currently use. For example, let’s say I introduce a Boolean value that, when true, causes orders to be pulled during periods of high volatility. Anyone who has this value nil will gain extra compensation during those times if the pool is competitive because there will be less competitors. Giving liquidity providers more options like this allows them to tune their bot to be more profitable than others and allows for a game to be played.
In addition to putting up an order or taking one down based on market conditions, we can modify the spreads (within the 0.5% allowed) to follow different patterns. As a simple example, if you believe in BTC and want to be sure you have BTC most of the time, you can place your NBT buy orders on the far end of the spread and your NBT sell orders near the market price.
The hedging tools I mentioned are are fairly easy to implement, but they can be expanded upon. It is theoretically up the the liquidity provider to do this because it’s open source, but as you say shareholders should help liquidity providers be more profitable. I’d be willing to try my hand at making some hedging tools.

[masterOfDisaster]

support only nbt/fiat but most exchanges do not have these pairs

Let me give an example with the NBT/BTC pair - I’m going to use BTC and crypto exchangeable as well as USD and fiat.

Exchanges that don’t have NBT/fiat pairs, but NBT/BTC pairs, often have BTC/fiat pairs as well.
One way to fight the volatility issues the NBT/BTC pair poses, is to shift BTC to fiat and create a virtual NBT/fiat pair by chaining the two pairs NBT/BTC -> BTC/fiat.

Drawback I can think of:

• additional fees
• slower response to market
• additional complexity
• you can’t withdraw fiat as easily as crypto

Benefits:

• lower volatility risk for NBT/crypto pairs

I wouldn’t want to mirror each and every transaction that is made on a NBT/BTC pair to the NBT/fiat pair.
Instead I’m thinking of using the tiered model for providing liquidity and move a part of the funds to Tier 2, but stored in fiat and not in BTC.

The 0.4% conversion fee (for those exchanges that have 0.2% per transaction) might be less than the loss by a dropping BTC price.
That fee could very well be recovered by increasing the spread for NBT/BTC pairs.

I remember that NBT is pegged to USD and Nu should think twice before playing with the spread of NBT/USD pairs.
But I don’t see why customers shouldn’t participate in the costs for converting NBT to crypto pairs like NBT/BTC.

Those who don’t want to withdraw the NBT from the exchange might prefer BTC/USD anyway.
Those who trade NBT/BTC back and forth to make money from fluctuating BTC prices drain money from Nu.
Those who want to withdraw NBT (because they can’t do that with USD) might be willing to pay the price for the increased spread.

So the basic question is: to what level does Nu want to buy trade volume on exchanges (which it does by offering that service at no cost but associated with a financial risk) by supporting NBT/BTC?
The answer to that question can be used to determine

• how much to increase the spread for NBT/BTC and
• how much of the funds to move from Tier 1 (BTC) to Tier 2 (USD) on the exchange.

I’m aware that trade volume on exchanges creates attention.
From a marketing perspective I second that.
From an economical perspective I need to search for alternatives or improvements.

[cryptog]

you aren’t going to make a huge spread out of that

Nothing would prevent the LPC from increasing his or her spread, in my opinion.

let’s say I introduce a Boolean value that, when true, causes orders to be pulled during periods of high volatility

So it would be done automatically based on a volatility data feed?

is to shift BTC to fiat and create a virtual NBT/fiat pair by chaining the two pairs NBT/BTC -> BTC/fiat.

So is it some sort of hedging:

• when you are long NBT, you buy BTC
• when you are short NBT, you buy USD
  ?

[masterOfDisaster]

So is it some sort of hedging:

Briefly, yes.
But I wouldn’t dare to do this as a means of speculation.
I’d be more interested in storing value in Tier 2 in a way that is not prone to fluctuating crypto currency exchange rates.