DR - Distributed reserves

Looks like I removed the marrow of you proposal by my (overly) simplifying it :wink:

Are we still going to get this side discussion moved to its own thread? It would be great to have some brand new eyes on it. I’d move it myself, but unfortunately I don’t have any mod powers in the Nu forum like at Peercointalk.

1 Like

@ttutdxh, you have your own thread now for your idea.

i think this way can avoid the move price of BTC very well

Thanks @CoinGame for splitting the topic.


I have added a title, lets hope it clears things up a bit.

Let’s continue with the discussion, any new comments on this?
I find that the model is difficult to explain, what would you change about it?

A lot of discussion have taken place in @Dhume NuSafe proposal about what I think are two very similar methods, what do you guys think?

I am going to continue @Dhume discussion at the NuSafe post here, because we went off topic there.

This is not pointless, the point is to prove you own the money and secure a collateral at the same time. It is weird, yes, but it does what is supposed to do.

We only have reserves so we can buy when there is a lot of selling (demand peaks). Right now we could put up all reserves and T4 money in USD buy walls, do a buyback and burn of all the NBT we can. Then we run out of funds, we relay there will be then a lot of demand, and we don’t have to buy, because we have cut a slice of the necessary NBT quantity for current usage and demand force us to print new NBT. If that is not the case, and demand is not there, there are more NBT than needed and the whole systems collapses, even if T4 reserves prevent that happening soon, they will eventually go to 0 if the demand is not there.
We kept reserves so we can use them to smooth demand peaks, and cover temporal demand decreases without breaking the peg for a second.

No, the point is that users buy it for 1 USD and sell them for 1 USD. Nu leaves offer and demand cancel each other, and only acts on demand peaks.

If it is Nu who always sells them for 1 USD and buys for 1 USD we are basically Theter without being backed by a full reserve.

[quote=“ttutdxh, post:29, topic:3163, full:true”]
This is not pointless, the point is to prove you own the money and secure a collateral at the same time. It is weird, yes, but it does what is supposed to do.[/quote]

We already have a way to prove someone owns the money by using our ALP software. Your proposal is essentially nothing more than a less refined inefficient form of liquidity provision. I’m sorry but it doesn’t make any sense at all.

[quote=“ttutdxh, post:29, topic:3163, full:true”]
We only have reserves so we can buy when there is a lot of selling (demand peaks). Right now we could put up all reserves and T4 money in USD buy walls, do a buyback and burn of all the NBT we can. Then we run out of funds, we relay there will be then a lot of demand, and we don’t have to buy, because we have cut a slice of the necessary NBT quantity for current usage and demand force us to print new NBT. If that is not the case, and demand is not there, there are more NBT than needed and the whole systems collapses, even if T4 reserves prevent that happening soon, they will eventually go to 0 if the demand is not there.
We kept reserves so we can use them to smooth demand peaks, and cover temporal demand decreases without breaking the peg for a second.[/quote]

I just feel your reasoning is based on a misunderstanding of our tier system and the way Nu works. We have tier 1-3 to actively engaging in buying and selling Nubit at 1 dollar each (albeit with a tiny spread). In general the money used to do this is privately owned by individuals who are willing to use their funds to protect our peg and earn an interest rate over the amount of funds they provide (depending on pool etcetc).
The ALP software allows them to do deliver this service without your plan to buy and burn Nubit to prove they own the money since the software checks if they actually put up buy/sell orders and grants them interest over verified orders. This is how we work essentially our peg is run without using Nu funds other than the interest we pay to our liquidity providers, giving them incentive to use their funds in Nu benefit.

Now T4 funds (aka reserve) as meant to balance the underlying tier 1-3 in case it gets out of balance. Essentially when there is an increased demand for Nubits tiers 1-3 will start running out of Nubits (and thus have excess of BTC) to sell and request to buy some Nubits from T4 with BTC to restock. On the other hand when the demand for Nubits decreases and there is more sell pressure tier 1-3 start running out of BTC (and thus have excess of NBT) and request to sell some Nubits to T4 for BTC from our T4 reserves to restock on BTC.

Essentially there is a balance between buy and sell side in tier 1-3 and T4 is there to rebalance incase the scales are being tipped to either sell or buy side. This is the reason we have a T4 reserve of BTC, in order to balance tier 1-3 when they get out of balance.

Yes that is why we have trustless liquidity provision aka users of Nu. So we don’t actively buy up and sell NBT for 1 dollar each but pay others, the liquidity providers, to do so for us. For this service they receive interest, a fee over the amount of liquidity they provide. I don’t know exactly how Theter works but yes we are not backed by a full reserve available in T4, we are backed by NSR. In case a huge decline of demand for Nubits occurred and we ran out of T4 reserves we would have to resort to auctioning NSR to raise additional funds to protect our peg. That is how Nu works.

Now I honestly have a hard time explaining why your proposal doesn’t work since I feel I’ve already done so, but essentially your trying to buy up Nubits (to burn them) that are not being sold. What you are proposing is essentially what happens when demand for Nu decreases thus leading us to have to rebalance T1-3 with our T4 buy side funds. This cannot be done unless we cancel our ALP’s and remove the pools and start using T4 funds to buy up Nubit on the exchanges ourselves.

I would like to hear why is it inefficient, a example with numbers about expenses or whatever you consider inefficient, and why it does not solve the problem, with numbers. We are going too deep in terminology, and too light on practice.

Feel free to make a copy of my version an correct the mistakes/inefficiencies/negative results.

I’m afraid I can’t explain it any better than I already did. Maybe someone else can shed a light on this cause I really don’t know how to be more eloquent about this.

the main point is that: we want to avoid the move price of BTC? am i right?

Yes, eventually having a way to store USD might lead to other usages.

NU hold BTC , and others short BTC at USD market. maybe a good idea.

Continuing the discussion from here

@Dhume do you think that allowing buy side liquidity go way bigger (like 150k, vs 70k sell), and not enforcing the 40% on the sell side like this, would solve DR collateral risk completely? That modification should not create any problem.

Are you referring to the scenario where DR is funded by “new” money entering Nu (so external USD flowing in) or the scenario where we are talking about stray Nubits are already out in the wild (so a decline in outstanding Nubits) being burned to accommodate DR?

Both, because T4 will ultimately kick in in the event of a liquidity unbalance further than 40%, thus risking collateral not being able to be converted back to USD into the buy wall.

Alright, well my take on it is that stepping away from enforcing a 40% would help in the case of new money flowing in. Since that new money is now being buffered in T1 instead of being rebalanced by T4 and thus ending up as excess funds (since we now have a contract instead of reserve funds) and being spend.

In the case of outstanding Nubits funding DR and thus being burned the situation stays the same. Since we then have an excess of funds in T4 since part of those reserved funds are now replaced with the DR “contract” and thus promoting us to use anything over 15% for buybacks and other goals.

A better solution in my opinion is changing our reserve structure. So we want a 15% hard reserve, why not try and add an additional 10% reserve on top of that in the form of DR contracts. Since DR contracts are very cheap (0,5% annually right?) they would be very suited for this. If they default no biggie, we have a 1% profit (the collateral) and still 15% reserved in other means to maintain the peg. Of course tweaking the numbers like a 10% hard reserve and a 10% DR reserve would work as well. It’s a matter of preference.

The nature of DR means it’s very cheap for Nu and even comes with a potential profit in case a default occurs, but it comes with higher likelihood of defaulting. Which is okay for a first line of peg reserve but less suited for a do or die reserve.

I don’t follow, in that case T4 is reduced/converted to DR because of the BTC payment in step 2.

Edit: Ok I think I got what you are trying to say: Burning NBT reduces circulating NBT and thus reduces T4 15% equivalent creating an excess because the 15% is now less money.
That should not happen because those burned NBT collateral should be accounted together with circulating NBT, even thought they are not, so the 15% of T4 does not change.

My apologies my comment is not always valid, its only valid if the “outstanding” nubits are not used in T 1-3. So in case they are Nubits held in a private wallet not doing anything this should not affect tier 1-3. In this case the contractor is payed back for his burned Nubits with BTC currently in T4, this leads to a decline of T4 funds held by Nu in favor of a contract between Nu and the contractor.

There is a small effect on T4 requirement since this is a decline in outstanding nubits due to the burning and thus requiring a slight reduction in the 15% we require in T4 (this effect is small but makes examples far more complicating so I’ve ignored it :wink: )

If the Nubits are currently held by someone involved in Liquidity provision in tier 1-3, this will remove Nubits from the sell side and thus create a similar imbalance as when they are bought from this sell side and burned, however now the imbalance is only a reduction in sell side funds not an increase in buy side BTC (since they are not bought from the sell side but removed by the contractor).

Edit: [quote=“ttutdxh, post:41, topic:3163, full:true”]Edit: Ok I think I got what you are trying to say: Burning NBT reduces circulating NBT and thus reduces T4 15% equivalent creating an excess because the 15% is now less money.
That should not happen because those burned NBT collateral should be accounted together with circulating NBT, even thought they are not, so the 15% of T4 does not change.[/quote]

Ok, but that changes a lot since then it’s like @mhps said and they are not actually “burned” or at least they are still being counted in the money supply.

So letting buy side run free solves the model for you?

Well it creates a sort of buffer that is desirable and makes DR more attractive. The only problem is that comes at a tradeoff of having less sell side liquidity at hand. Unless we increase total liquidity across exchanges, but this of course would cost us more interest.

Letting buy side grow instead of rebalancing with T4 means we can buffer the burned Nubit in our buy side walls, however this is naturally limited to how much the buy side can hold until the imbalance becomes to impractical. Aka a 10k sell side with 190k buy side would be problematic. But I like the idea, but unless we want to pay additional interest over additional liquidity there is a limiting factor.