NuPawn - An Nubit colleteral backed lending business

For a while now myself and other have contemplated about a Nubit lending scheme, as an entity that can create its own currency at will we have similar functionality akin to a bank, money lending is their prime business and it could be ours. However the sharp distinction between us and a real world bank in terms of money lending is that banks have legal instruments (law, police, etc) to insure that lenders are enforced by law to pay back their loans. We as a DAC in the crypto world do not possess such means.

So the question then becomes how do we issue loans but at the same time protect ourselves against counterparty risk? The answer is collateral, if we hold something of equal or greater value then the funds loaned we are effectively risk free. The reason people will be interested in these loans is because they need funds in the short term but don’t want to let go of their crypto. Say someone holds Bitcoin and believes it will go up a lot long term, however his car broke down and he needs 2000 USD for a few months. He could sell some of his BTC and repair his car but he would miss out on a rise in BTC value if it occurs. Instead he could park the BTC with us get the money he needs and pay us back later, knowing his BTC are safe and if they go up in value he won’t have to buy back later. Even if he never manages to pay us back if his BTC go’s up by 30% he only loses the USD amount he loaned + interest not the whole collateral.

Now I see this working in the following way, we have customer X who wants to lend say 5000 USD, we ask him to put up collateral of 200% value, so he is required to deposit 10k USD in crypto assets. We could accept any assets we want but I’d say we should aim for established assets that have liquid markets (I’m thinking BTC, LTC, PPC, NMC and maybe ETH, NSR, BKS). Now when person X has deposited his collateral in a multisig address controlled by a Nu multisig team akin to FLOT (I would vote for creating a new team) we transfer the Nubits to him. We charge 1% interest a month (or 0.25% a week) over the lend Nubits, the person has the option of paying us the interest when the loan ends or we subtract the interest from his collateral upon returning the Nubits.

These loans would be indefinitely prolonged, meaning the person can hold onto the Nubits as long as he wants as long as the collateral value – interest stays sufficient. The question is at what point do we start invalidating the loans? My opinion is when the value of the collateral drops to 130% of the loaned amount (in the above example that would mean if the collateral value drops to 6500 USD) we issue a warning that at the ending of the next month the loan is invalidated and we sell the collateral. If the value of the collateral drops to 110% (5500 USD in above example) we instantly sell the collateral and the loan is considered settled. The interest also counts towards this “margin” call so to speak so if the collateral remains stable after the first month the collateral is devalued to 200-1=199% (since 1% is our interest).

If we do it in this way we are protected against counterparty risk and also the lender to a certain extend since he doesn’t have to trust a single Nu entity but his funds are held in a multisig address.

To get this running we would need a Nu multisig team akin to FLOT that can hold a large amount of different crypto currencies in multisig accounts to secure the collateral.
We would also need an market executioner who can sell assets on loans that are being margin called or on interest that is charged.
And we would need a bookkeeper (we could also do this on the forum) that keeps track of all the loans and the collateral value vs the loaned value. I feel this last step we could automate to a large extend with a smart script. It would work something like this:

Row 1 I Row 2 I Row 3 I Row 4 I Row 5

Loaned amount I Colleteral I Colleteral USD value I Interest I Margin level

5000 NBT I 24 BTC I 10004 USD I 150 USD I 197,08%

In order to automate this we would need to pull the USD value of the coins we have as collateral out of some sort of API and put them into the spreadsheet. We could also manually do this of course.

The numbers I have chosen are of course arbitrary, we could also require 180% or 250% collateral and have margin calls at 130% instead of 110%. The numbers are up for debate as well as which coins we accept for collateral. However I think the method is sound and could be very profitable for Nu, if we aren’t able to automate this we could also say that the executioners get a % of the profits, I’m sure many would be interested in performing this service for us.

6 Likes

Thank you very much for explaining a scenario, which has been mentioned numerous times before, but never explained in such a detailed way.

Except for things you already mention, we need to consider

  • minimum amount that can be loaned at a certain interest rate

  • too small amounts create too much efforts compared to the reward

  • the interest rate needs to be adjusted accordingly, if the amounts are smaller, because the efforts for the multisig team stay more or less the same - unless you have lower multisig levels for smaller amounts

  • introducing a minimum loaning period helps here

  • the amount of collateral and the margin call levels should be relating to the liquidity of the collateral and the fluctuation

  • the less liquid and the more volatile, the bigger the collateral and the higher the margin call level

  • the more liquid and the less volatile, the smaller the reuired collateral and the lower the margin call level

  • the loaned amount can play a role as well - the more collateral you need to liquidate in a crashing market, the higher the amount of collateral and/or margin call level needs to be if you don’t want to sell at a loss

  • automation of important parts / part 1 - the financial part

  • ideally the collateral is put on order (by bots that automatically adjust them together with the fluctuation of the collateral) at the margin call level to make sure Nu doesn’t miss the point of time to sell it

    • doing that at centralized exchanges poses a total loss risk and it can’t be done with multisig
    • BCE will be a boon here, but will take some time until it’s available; still placing the full collateral at margin call level at all times should be considered as an option to protect Nu
  • automation of important parts / part 2 - the informatory part

  • a framework that informs the takers of loans would be reuired to avoid missing the point of time to inform them; developing that might be complex and costly

  • doing it the other way round and telling people that the collateral will be sold, if condition x (e.g. exchange rate below y for time z) is met, saves Nu from the hassle of informing the burrower.
    This way Nu isn’t being reponsible for having missed such an information
    Besides: how will you be able to verify that you really informed the customer?
    I sense a lot of trouble from this field, if we decide that Nu needs to inform the burrower of the liquidation of the collateral

  • automation of important parts / part 3 - offering or negotiating conditions for loans

  • Nu can either negotiate loans individually, which creates a lot of efforts or

  • provide information about the conditions along the lines of the book keeping

  • by the latter way Nu can offer “different products” and the customer can choose

  • that is much like at traditional banks, except for the transparency;
    you won’t be able to negotiate interest rates at banks freely; you will always be served with the fitting product according to a matrix.
    The difference is that banks normally don’t show you all available products, but only those which are a match for you after they gathered information from you.
    Nu would make it the other way round and just offer products like (just random examples…)

Product A
Loaned amount | Collateral | Collateral USD value | Interest | Margin level |Minimum runtime
2500 NBT      | 12 BTC     | 5000 USD            | 1.5% p.m. | 120 %        | 6 months
|-------------------------------------------------------------------------------------------
Product B
Loaned amount | Collateral | Collateral USD value | Interest | Margin level |Minimum runtime
5000 NBT      | 24 BTC     | 10000 USD           | 1.25% p.m.| 130 %        | 3 months
|-------------------------------------------------------------------------------------------
Product C
Loaned amount | Collateral | Collateral USD value | Interest | Margin level |Minimum runtime
10000 NBT     | 48 BTC     | 20000 USD           | 1% p.m.  | 145 %        | 2 months

##Let’s get this started!

2 Likes

What would be better in order to increase adoption and liquidity for Nu?

The funds being provided by Nu DAC (another “Tier” required?) or by Nubits holders (creation of a P2P Decentralized Lending Platform?) whose NBTs goes to a multisig Pool (like Nulagoon)?

This is an interesting and simple idea for holding collateral. But would it be really cost effective?

Thanks for putting it together. My main concern is that hte cost of signers, market executioners, and book keepers would easily eat any profit away unless there is a very big demand. I suspect in the end only a centralized business (like nuLagoon) is able to run the whole thing efficiently enough to make money. I would still support a private nubit lending business as it increases nubit demand.

[quote=“masterOfDisaster, post:2, topic:3637, full:true”]

  • minimum amount that can be loaned at a certain interest rate
  • too small amounts create too much efforts compared to the reward
  • the interest rate needs to be adjusted accordingly, if the amounts are smaller, because the efforts for the multisig team stay more or less the same - unless you have lower multisig levels for smaller amounts
  • introducing a minimum loaning period helps here[/quote]

Yes we need a minimum loan amount and also declining interest rates for larger loans. So for example loans above 10k USD value get a 1% interest rate and our lowest loans (lets say 1000 USD) get a 5% interest rate. Loaning out 100 USD for 1 month a month would not be worth the trouble. Also this would increase incentive for lenders to go for larger loans, meaning less work and more revenue for us (similarly to how exchange offer lower rates for larger clients).

[quote]* the amount of collateral and the margin call levels should be relating to the liquidity of the collateral and the fluctuation

  • the less liquid and the more volatile, the bigger the collateral and the higher the margin call level
  • the more liquid and the less volatile, the smaller the reuired collateral and the lower the margin call level
  • the loaned amount can play a role as well - the more collateral you need to liquidate in a crashing market, the higher the amount of collateral and/or margin call level needs to be if you don’t want to sell at a loss[/quote]

Yes for every coin we accept as collateral we need to think of a different risk profile accounting for liquidity and volatility. Specific margin call levels and collateral required would be different for every coin we would accept, or we could group them into different risk category’s. I suspect this will be the main brunt of the work in setting up and maintaining this service.

[quote] * automation of important parts / part 1 - the financial part

  • ideally the collateral is put on order (by bots that automatically adjust them together with the fluctuation of the collateral) at the margin call level to make sure Nu doesn’t miss the point of time to sell it
    • doing that at centralized exchanges poses a total loss risk and it can’t be done with multisig
    • BCE will be a boon here, but will take some time until it’s available; still placing the full collateral at margin call level at all times should be considered as an option to protect Nu[/quote]

We could automate this with bots to a large extend I think, although my preference would be to set up margin large enough to only occasionally have liquidations. It also depends on how many loans we have running. If the service gains a lot of traction additional automation will most likely be required due to increased load. However increased traction should also mean good revenue, it should be able to pay for its own devilment and maintenance costs with this revenue.

[quote] * automation of important parts / part 2 - the informatory part

  • a framework that informs the takers of loans would be reuired to avoid missing the point of time to inform them; developing that might be complex and costly
  • doing it the other way round and telling people that the collateral will be sold, if condition x (e.g. exchange rate below y for time z) is met, saves Nu from the hassle of informing the burrower.
    This way Nu isn’t being reponsible for having missed such an information
    Besides: how will you be able to verify that you really informed the customer?
    I sense a lot of trouble from this field, if we decide that Nu needs to inform the burrower of the liquidation of the collateral[/quote]

The margin call levels should be explicitly communicated so that customers should always know when they get liquidated themselves. However, I would advocate a sort of messaging system in which we notify customers in case their margin drops to “dangerous” levels and to inform them of liquidation if it occurs. It would be required to have a contact email to acquire a loan. Seeing as the idea is to set up wide margin levels liquidations shouldn’t occur often so this should be manageable by hand.

Yes!

We could incentive this by letting the operators have a part in the profits. The operation should be able to pay for itself + generate revenue for Nu.

I think 3 active shareholders would be enough to get this running, if there is enough demand for such a service it could become very lucrative. Say we get an average interest rate of 1.25% and manage to get out 150k USD in loans. That’s 1875 USD in revenue, should be more then enough to pay for the hours of the operators + a fat check for Nu.

I had the same feeling: I wonder if only a centralized or P2P decentralized lending platform would be able to make it through…[quote=“mhps, post:4, topic:3637”]
I would still support a private nubit lending business as it increases nubit demand.
[/quote]

Me too!

A centralised business would be hindered by the BitLicense. Even several Bitcoin/P2P lenders are withdrawing from the US market because of potential legal issues with this. Having said that there is life outside the USA :slight_smile:

1 Like

Since the person who gets nubits loans is very likely to spend it as fiat, say to pay a car repair shop, deep liquidity of NBT/FIAT pair is important, even crucial. @ronny’s nubit-backed debit card could also help but it is a centralized solution that needs extra trust.

1 Like

To narrow down, whether it’s possible to offer loans with an “LPG” (Loan Processing Group)?

We could try to define (words in bold letters)
Loaned amount | Collateral | Collateral USD value/Collateral overbooking rate | Interest | Margin level |Minimum runtime
to find out at what minimum rewards per contract it’s profitable for Nu to offer loans (LPG needs to be paid with a yet unknown fee as well).

And we need to try to guesstimate/calculate/define (words in bold letters)
Loaned amount | Collateral | Collateral USD value/Collateral overbooking rate | Interest | Margin level |Minimum runtime
what risk is associated with such a loan.

Interest plays a role in both parts of the evaluation.
If we can define conditions under which Nu can offer such a product (loan) profitably, while it’s considered possible to really deliver (considering the collateral risk) at these conditions, we can move on.

There might be a gap between the interest rate evaluated as

  • interest rate for product profitability in terms of effort that needs to be compensated and
  • interest rate for product profitability in terms of risk that needs to be compensated

If this gap can’t be closed, Nu can’t (yet) offer this type of loan. At least the type of collateral needs to be exchanged then.
Ways to close the gap are

  • increase the “collateral overbooking rate”
  • increase the margin level

This sounds a little bit abstract, but I want to know whether this is the right way to start shaping different (hopefully profitable) types of loans.

Have you checked the lending service of Poloniex?
Although they don’t support NBT yet.
Is this centralized service close in what you are thinking?

It’s a good start.

  • illiquid collateral asset is easy to have its market price manipulated
  • people don’t trust Nu enough to let Nu have the collaterals for a long period of time
  • people borrow (printed) nubits to park and defeat the purpose of parking.
1 Like

Came across this just now: https://lendroid.com/

It’s a more refined / automated version of the idea behind NuPawn.

Nu could have had it all revenue, stable currency and a decentralized exchange what a waste…

2 Likes