100% USD Reserves Offers Zero Benefit In terms of peg stability

Here, I use a mathematical model to show that USD reserves are completely useless in a decentralized system. The model shows that nubits are backed by nushares alone, regardless of reserve holdings. All other forms of backing are impossible.
I also explain that the only way of achieving stability in a decentralized system is capping the ratio of value in nubits to value in nushares.
Finally, I resume my argument for negative interest rates.

Comments welcome.


Fascinating. This analysis goes one step further than your previous papers and genuinely makes me think we should be considering this no-reserve model backed by NSR.

In particular:

In the absence of third-party legal enforcement, solvency is not a very meaningful condition. Since shareholders are not bound by law to repurchase nubits, shareholders could allow the peg to break even when they have the financial capacity to support it. In my view, we should only expect nubits repurchases to occur when repurchases align with shareholders’ economic interests. Rather than vote for repurchases, shareholders could just as easily vote to abandon a 100% reserve system and distribute reserve funds as dividends. If we allow shareholders to vote on the use of reserve funds and assume that they vote according to self-interest, the argument that a 100% reserve offers support for the peg falls apart.

resonates with me completely. Custodial funds held at a regulated cryptocurrency bank might be able to have certain conditions placed upon them mandating peg support, but it introduces the third-party trust you mention in the first part of this analysis. I think right now the altruistic nature of our majority shareholders would dictate that the peg is supported even if not perfectly economically rational (because of guiding philosophies behind the project), but it’s not a reliable condition to hold constant with further decentralization.

For people reading this who don’t want to wade into the heavy-duty economics, it is worth reading the “Implications” section at the very end on Page 6.

I will go on the record as saying we should give very serious consideration to the following three key recommendations (as I understand them), unless a similarly-detailed quantitative defense can be made as to why our current system offers more stability, decentralization, or utility.

  1. The quantity of NBT should be capped at a set ratio with the intangible value of NSR. An added benefit of this is that USD reserves will no longer need to be maintained, eliminating costly custodial fees and theft risk.
  2. The system should be monetized primarily through user fees of services rendered, not the sale of the asset (NBT) itself. Fees generated enhance the intangible value of NSR that support the peg in recommendation 1, allowing for a greater supply of NBT to be in circulation.
  3. The current system of Tx fees (and upcoming variable Tx fees) should be replaced by negative interest rates.

Did I get all of this correct @Benjamin? I’m very glad you posted this. There are a substantial amount of debates that can be had with this analysis, and the end result (assuming rational shareholder behaviour) will be a stronger Nu product.


I agree that concept of leverage, or the ratio of the value of all NuShares vs. the value of all NuBits in circulation is an important metric that reflects the health of the peg. I advocate tracking this ratio and making it public.


It’s overwhelming to see that what so far has only be a gut feeling for me can be explained in a mathematical way.
Thumbs up for making clear that the peg stability is ultimately depending on the value of NuShares.


Whether we have fractional reserve and what that fraction is depends the value of NuShares + reserves (such as USD, BTC and PPC) / NuBits in circulation. For example, right now NuShares are worth about 3,000,000 NBT, reserves are approximately 600,000 NBT (total of BTC and PPC held by KTm, Julie Miller and myself), and I believe outstanding NBT is around 500,000. 3,600,000 / 500,000 = 7.2. Assuming these numbers are accurate (they are not) we have 7.2 times as many assets as liabilities. Only if this ratio drops below 1 is there a fractional reserve. Fractional reserve doesn’t imply insolvency, but it does imply a vulnerability to it in the case that too many NuBits are sold at once.

The above analysis presumes the existence of currency burning. I also agree with Benjamin that non-NuShare reserves (such as BTC held by custodians) are not of the same quality as NuShares, because it cannot be ensured they will be used to support the peg. Therefore, given the numbers I presented above, there is reasonable debate about whether the ratio of reserves to liabilities is 6 or 7.2, or somewhere in between.

Reserves are useful for providing immediate liquidity. It is worth noting that I have worked hard to encourage liquidity provider custodians to provide their own funds, which in no sense would be considered reserves, which if consistently done would put us in the position of providing liquidity without the use of reserves.

We ought to be able to present the kind of numbers I presented above. This will take the coordination of any custodian who still has possession of NuBits received via custodial grant as well as info about the remaining proceeds of undistributed NuShare sales. Because I do not have current information from other custodians, the numbers I mentioned are only estimates used to explain the significance of this type of information.


Under the no-reserve model, are any dividends paid? If not, what gives NSR its value?

Imagine there is market demand for 100M of NBT. Does this imply any market demand for NSR under this model?

The white paper clearly suggested that under the 0% reserve, at the end of the custodial contract period, all the USD$ in the custodian buy wall will be sold for Peercoins to be paid as dividends, while all the NuBits left unsold in the sell wall will be destroyed.

I guess this will increase the NuShares value a lot more than the 10% dividends scheme or just waiting the reserve ratio to exceed 100%, as long as the peg is maintained.

1 Like

I like what you suggest @Benjamin , and i agree that NuBits savers will hurt the system, but i am not sure yet about the effect of Negative interest rates on the NuBits demand and NuShares price.

Any way i think that we are not only providing the peg service but also facilitating money transfer and commerce , so removing transaction fees at all will also hurt the system and facilitates getting around the negative interest fees for free.

May be we can keep the 0.01 NBT transaction fees and apply a reasonable negative interest rate.

guys, i haven’t understand the consept of negative interest! is it analogous with freicoin’s demurage?

A rising price of NSR gives value as well. Only I don’t know what happens when large amounts of NSR are sold to people who might not always be aware of the responsibility that comes with holding an asset like NSR.
As this is would lead to a fundamental discussion about the drawbacks and benefits of distributing NSR wider, I keep focussed on answering the question:
Nu doesn’t need dividends to give NSR value. In a free market the value is defined by supply and demand. The more successful Nu is, the more desire for holding NSR arises. As the supply is limited this results in rising prices.

1 Like

Why would any one buy NSR if he doesn’t expect dividends ? or use enough NSR to vote on a huge NuBits grant as a fraud .

He may just speculate with NuShares, but where does the demand come in the first place if there is no benefits on holding NSR ?

one benefit is the minting for sure :wink:

Yes minting is profitable if you can sell it, why would i buy from you, planning to mint and sell again? this is a meaningless infinite loop, there have to be a value for NSR other than selling, to create the demand needed for this speculating and minting schemes.

1 Like

Coinmarketcap indicates a total supply of more than 2m nbts --> http://coinmarketcap.com/currencies/nubits/
How much of that total supply has been actually sold? 600k nbts?

Any way, what i like about the 0% reserve scheme, that it facilitates pegging Nu products to anything.
So we can peg the NuBits to an anti-inflationary basket of goods, thus when we introduce a 1.5% Negative interest, customers will still find it a reasonable store of value.

Does this mean we may be unable to meet the NuBits demand if NuShares value lagged behind for some time ? wouldn’t this cause NuBits price to increase and lose the peg temporarily?

1.The quantity of NBT should be capped at a set ratio with the intangible value of NSR. An added benefit of this is that USD reserves will no longer need to be maintained, eliminating costly custodial fees and theft risk.

Yes, that is correct. Though I would go a step further than this.
I propose that a specific target for the ratio of value in NBT to ratio of value in NSR be adopted. This target should only be adjusted at rare intervals, say once every 6 months.

This target would be a second policy objective in addition to maintaining the peg.
One could achieve both objectives using a mixture of nushares burning and adjustments to the interest rate. Nushares burning is used to maintain the peg.

We can think of the interest rate as a fee we charge to ration storage of USD value in nubits.
If we have a ratio of value in nubits to value in nushares that is below the target, then we have unused space available to store USD value. In this case, we do not need to ration this space using a negative interest rate. Instead, we can offer nubits’ holders a positive interest rate as an adoption incentive. The positive interest rate simultaneously decreases the value of nushares and increases demand for nubits. This brings the ratio back to the target level.

If we have a ratio of value in nubits to value in nushares that is above the target, then we have exceeded our capacity to store USD value. This places the system under risk. In this case, we need to ration storage capacity using a negative interest rate. The negative interest rate simultaneously increases the value of nushares and decreases demand for nubits. Again, this brings the ratio back to the target level.

1 Like

I think you may be missing the point I’m trying to make. Shareholder assets don’t help if shareholders can’t be expected to spend them to support the peg.
Using the numbers you provide
Intangible Value in Nushares = Market Cap of Nushares - Reserves
= 3,000,000 - 600,000
= 2,400,000 NBT
This measures the value that shareholders would be willing to spend to support the peg.
Holding more reserves doesn’t increase this in any way.

The market will provide liquidity if you get the peg correct. That is, if you ensure that the expected value of a nubit is $1, then people will buy up nubits if the price falls below $1 and sell nubits if the price rises above $1. You don’t need to provide liquidity yourself to sustain the peg.
In the long-term, you can allow the market to step in and fulfill this function.

1 Like

One of the points here is that nubits can never be competitive as a long-term store of value. In the short-term, we can share some of the gains through growth in the network with nubits holders.
In the long-term, the target market must be people who would be willing to pay negative interest for the privilege of holding nubits. These would be people who store money in the system temporarily in order to facilitate txns.

1 Like

Yes. I don’t mean that we should eliminate txn fees entirely, just that we should not expect them to be the primary source of revenue in the long-term.

1 Like

negative interest ? That’s even more inflation than USD, crazy.

NSR holders choose to let pegging broken? insane.

benjamin, have you checked your logic before calculation?