Would network restricted access and high fees be applied to NuShares transactions as well? Doing so may help prevent a full scale panic sale, preserving the value of NuShares until an auction takes place if it becomes necessary to conduct one. If it worked, it’s possible that many NuShares may be burned through transactions during this time while the fees are high, helping to counterbalance whatever NuShares have to be created in order to hold an auction.
Or, it might possibly exacerbate a NuShare panic sale depending on how long it takes for shareholders to vote up the transaction fees, for example a time window of days for people to sell before the fees are raised through protocol voting. It would also mean that larger shareholders would be voting to prevent smaller shareholders from transacting during this time, effectively holding them hostage until the crisis passes.
Another problem is that with NuShare transaction fees raised, how are we supposed to hold an auction? If the high fee only applies to NuBits, then people would have to purchase NuShares from us using an external asset like Bitcoin to avoid the NuBits fee, or maybe instead just prove that they burned the NuBits in order to receive the NuShares. If the NuShares fee is high though also, then it means we wouldn’t be able to distribute the NuShares to the auction winners until the NuShare transaction fees are lowered again, to prevent part of their shares from being burned in the transaction.
Very interesting idea. I think this would solidify the PR effect of Tier 7, because there would be equal sacrifice made by both NuShareholders and NuBits users in regards to their access to the network. I’d be interested in others’ thoughts on this, because my first instinct is that it is a good idea.
@tomjoad, you didn’t answer this question. Voting takes at least a week to respond. I think this is another reason to keep tier 6 as a buffer to buy time until tier 7 has geared up and network velocity is slowed down.
…is designed to incorporate a lower bound threshold, which triggers NuShare sales in order to refill tier 4 buy side funds when they get low. If @tomjoad’s tier 7 motion passed, then I imagine the following changes would need to be made:
The tier 4 threshold for NuShare buybacks would need to be raised from 15% of circulating NuBits to something higher in order to make up for the fact that we’ll no longer be strictly relying on NuShare sales to protect the peg, except for in an extreme emergency situation. This means more Bitcoin and other volatile external assets will need to be held in the tier 4 buy side reserve under FLOT and share buybacks will need to be reduced a certain degree in order to maintain this higher reserve.
Work will need to be done in order to figure out how to further decentralize our tier 4 reserve and limit its volatility to prevent network losses. There is a proposal I know of by @ttutdxhhere. If that won’t work, then we’ll need to think of other ones.
If this motion passes, basically we’ll need the following setup for liquidity operations:
Using ALPs and MLPs to provide decentralized liquidity to tiers 1-3.
Tier 4 buy side (Bitcoin) and sell side (NuBits) managed through FLOT.
A raised tier 4 threshold for NuShare buybacks to allow for a higher tier 4 reserve.
A lower bound tier 4 threshold when it is advised that shareholders raise parking rates.
An even lower bound tier 4 threshold when it is advised that shareholders activate tier 6 network restricted access.
NuShare sales can be decided in case the previous actions prove ineffective.
So basically we would need a higher buy side reserve for tier 4 as I already mentioned in order to make up for the absence of tier 6 NuShare sales. We would also need to decide on a certain threshold where it would be appropriate for shareholders to activate parking rates in case tier 4 gets mostly exhausted, maybe something like 5-10% of circulating NuBits as @masterOfDisaster suggested in his thread. Parking interest would in effect become the most used tool in case our tier 4 buy side reserve shrunk to a certain point.
As an example, tier 4 reserve drops below a certain defined threshold and NuShareholders raise parking rates. Buy side returns because of the synthetic demand brought by increased parking rates. When real demand returns again and FLOT is forced to bring NuBits to the sell side, tier 4 will be replenished with Bitcoin. When tier 4 buy side reserve rises above the required threshold again, shareholders lower parking rates.
If however parking rates prove to be ineffective and tier 4 buy side continues to shrink no matter the interest rates we offer, then an even lower bound threshold will need to be set where it is encouraged that shareholders activate network restricted access and higher fees. Then if NuShare sales are still required to soak up any remaining NuBits, shareholders will need to conduct an auction.
Note: I’m not saying I agree with any of this. I’m still interested to hear what others think about @tomjoad’s updated proposal. This is only the way I think liquidity operations should be setup in case this motion passed.
Just like to respond to this specifically as I think it is not only about PR. The notion of Stakeholders that they have skin in the game is likely to make them more careful in managing the peg in tier 1-4.
I think the biggest effort needs to be in preventing this event to happen in the first place by having solid tier 1-3 management and reserves which can be adjusted to the risk profile at hand.
This is also important and is an example of what I said above about a NuShare panic sale in case voting up higher NuShare fees takes a long time. The same can be applied to NuBits. If it takes an entire week to activate network restricted access and higher fees, it’s possible that the peg will be broken long before shareholders even have a chance to activate tier 6. If people see that shareholders are starting to vote up fees, then it could trigger a panic sale of NuBits on the market. The only way this could work is if we made it so that tier 6 could be activated on shorter notice, which means giving voting preference to larger shareholders over smaller ones.
How fast would voting need to be in order to activate tier 6 and head off a NuBits panic sale? Days? Hours? The shorter voting is on this, the more centralized tier 6 would become, the more vulnerable to attack the network is by an entity that bought up lots of shares.
I’m sorry, but I just don’t get the point of this at all. If you spike txn fees so no one uses the network, we stop making profit from fees. So you’re hoping to isolate economies to stop the peg from failing? Cause when you spike txn fees above the optimal point you start losing money not making it, so you must have some other effect in mind.
So Alice has 100 nbt on exchange. She’s just holding them as a hedge against btc. She notices the peg start failing, but she doesn’t panic because she knows Nu will sell nsr to take care of it. Instead, Nu institutes a 50% fee. Now, Alice gets weirded out: she can’t move her money off exchange anymore. She hears one little rumor about the exchange being insolvent and panics, selling the nbt for btc which she can tale off exchange without paying 50%. She dumps on a faltering peg.
Why are we trying to convert who Alice was (a trusting nbt holder) to who she became (a panic dumper)?
Profit = fee * #txn. A higher fee makes the first term higher (of course) but the second term lower. Therefore there is an optimal point that will make Nu the most profit. We need to just find this optimal point and not try to manipulate our customers.
I thought I answered it quite clearly. And yes, this tool would be more effective with a shorter block duration to set transaction fees, or, a future network state that has more rapid blocks than our current one minute spacing.
Likely because you haven’t taken the time to read my full argument, understand how this is an empirically proven method of macroeconomic asset management as per the IMF document I linked above, or thought about what a capital flow management measure is.
Nothing about what I’ve introduced here is controversial in economics. It is a proven fact this works. The only debate is whether its particular application makes sense within our network, due to technical factors.
If we’re willing to disregard proven economic ideas on a gut feeling alone, we’re no better than Bitcoin. I want to see arguments like “capital controls will not work on the Nu network because we have this particular design feature”, not “capital controls are dumb”. If we take that ideological stand, we will eventually be passed by another network that is willing to implement a full spectrum of tools in managing a fixed peg.
Lol, ok dude. Crypto capital control is not fiat capital control. We have no say over what goes on on crypto exchanges like governments do on currency exchanges. There are fundamental differences. You are taking a purely ideological stand that something has been tried and tested when in fact it has never been even attempted before.
And again, fiat capital control can take place in a matter of minutes or hours. Our ‘capital control’ relies on 50% shareholder agreement, something that takes weeks. Not only all that, but also US companies are forced to work with USD by the IRS. No one is forced to use NBT so when we institute high network fees people will simply sell their NBT for something with low network fees.
I love how I spelled these points out nicely in 5 bulletpoints a week ago that were never responded to, yet somehow I’m the one not paying attention. (And yes, I read the damn PDF)
This has been brought up by others. The basic question is how can Nu continue to function under such conditions? There are a lot of subtleties and issues here. This is by no means a show stopper, but it most certainly is something we need to think hard on before we accept such a philosophy.
This is a critical difference between Nu and Fiat capital controls. We have absolutely no control over the marketplace, where the transactions actually happen. If this were happening on a virtual exchange like BitAssets, the situation would be different.
Seriously, not an ounce of thought has been paid to this point.
This is a corollary to point 3. USD has a fundamental use: paying taxes. NBT has no fundamental use other than its ability to be transacted. By preventing all transactions on the Nu network, the currency becomes worthless as does the network. I realize that the idea is to make this a temporary thing to isolate economies that are doing poorly from ones that are doing well to stop the network from dragging the floating economies down with it. However, when our entire economy is a trade economy preventing all trade is a wonderful way to tank it everywhere all at once.
Point 46 in the PDF: Introducing outflow CFMs should always be part of a broader policy package that also includes macroeconomic, financial sector, and structural adjustment to address the fundamental causes of the crisis. They should not be used as a substitute for policy adjustment.
Restricted Network Access simply cannot right a peg that has broken. It can slow down the fall superficially, but as the network can respond with NSR on the same time scale as CFMs, the negative PR it generates outweighs the benefits. At least in my mind.
If you were talking 5 or 10% fees at worst to simply restrict arbitrage between exchanges, this would be a very different conversation.
—motion begin—
If buy side liquidity drops below 30%, shareholders agree to increase fees above 2.5% (or however makes sense with the implementation) and increase park rates to 2.5% or 5% (depending on whether parking is excluded from the fee or not). If liquidity goes below 20%, the fees and park rates should double. If it goes below 10%, they should double again.
—end motion—
You can see why i think it’s pointless to talk about this before we even understand how volume-dependent fees will be implemented.
This statement is an argument for restricted network access, to increase the friction within the system to make capital outflows more difficult. In a crisis we want to prevent NBT from being converted to USD at a rate that exceeds our liquidity operations. Tier 6 (Restricted Network Access) is a temporary application of major friction (which could be 10%, 20%, 50% tx fees, whatever shareholders decide is warranted for the magnitude of a crisis). This friction will allow shareholders to use Tier 7 (NSR sales) more effectively and with less dilution. Point 45 from the paper addressed the concept of capital outflows, which is the biggest threat to our peg:
In crisis situations, or when a crisis may be imminent, there could be a temporary role for the introduction of CFMs on outflows. For example, when countries face domestic or external shocks that are large relative to the ability of macroeconomic adjustment or financial sector policies alone to handle [e.g. our liquidity operations model], or when the size or duration of the shocks are highly uncertain, CFMs [i.e. Tier 6 (Restricted Network Access)]may help to prevent a free fall of the exchange rate and depletion of international reserves. When a crisis is considered imminent, CFMs may be desirable if they can help to prevent a full-blown crisis.
There’s no reason volume-dependent transaction fees couldn’t be structured in a way to have quicker response, whether through a decreased block voting window, or faster network blocks in a future state. Current network limitations shouldn’t influence our imagination of better tools.
Fair enough. Points 3, 4 and 5 appear to be the remaining points of contention. Here are my thoughts:
We have to imagine NuBits from the perspective of a widely-used product five or ten years from now. This means a state where 95%+ of all NBT in circulation are off-exchange in peoples wallets. Most of us don’t hold USD on Forex exchanges. You’re correct that if the majority of NBT are on exchange, the damage will be done regardless of what transaction fees we vote for. I don’t see exchanges holding mass NBT as a likely state a decade from now, especially with the development of decentralized exchanges like B&C Exchange.
I admit I don’t understand this point. It seems to be a general declaration that anonymity/security will suffer, but I don’t understand the context. Could you provide an example for my benefit? How would these aspects suffer relative to a normal transaction fee?
100% Tx fees are the most extreme example. Perhaps I didn’t explain it well, but this was my relevant paragraph to this point:
The concept may be easier to understand as @woodstockmerkle described it, as a technique to decrease the velocity of money. A punitive transaction fee will still allow users to transact on the network, but they will be paying a premium transaction fee price to do so. Shareholders could however set a 100% transaction fee (on both NBT and NSR as @Sentinelrv suggested) if they wanted to completely stop the flow of NBT through the blockchain (temporarily, of course) while they coordinated NuShares sales in a responsible way, as opposed to selling to whoever had available capital in the first hours of the crisis.
Absolutely not. Discussing new tools is always valuable for the network, even if we don’t have a set schedule for its use. We believe parking rates are valuable, even if we don’t know exactly what they will be set at.
It can slow down the fall entirely, allowing responsible Tier 7 (NSR Sales) to take place, as I posted in my updated content. Over a long enough time period, it could right the peg by destroying enough NBT. This is less desirable than a multi-pronged approach with Tier 7 (NSR sales) that would end the crisis quicker of course.
This tool is voluntarily applied, and powerful. I don’t see why shareholders are afraid to incorporate it into a liquidity model, because it will make Tier 7 (NSR sales) more effective. I will continue to debate its effectiveness until it is placed for hashing in two days. It may not pass immediately - or ever, depending on shareholder views - but I think there is value in signalling we are prepared to consider real-world economics in our currency management. I appreciate that you took the time to respond more fully.
This argument can just as easily be applied to NSR sales and buybacks, which can happen theoretically at the same speed as a block, meaning once per minute. I am unconvinced that capital controls can be applied faster than NSR sales on a fundamental level.
We’re skipping point 2 then? Alright, but this is one of my favorite points. It just further stresses how not-ready we are to talk about volume dependent fees this way.
I’d love to have this conversation five or ten years from now. For the time being, the shareholders are not informed enough to pass a motion about this. You can talk theoretically all you want, but you posted this thread as a draft.
The nature of the beast with volume-dependent fees is that you have to pay the fee to move money from one address to another no matter what. That means that if an address is compromised or a mixer is used the fee will end up being paid multiple times. This is an inconvenience with normal fees, though most would agree people have to pay for the added security. For exorbitant fees, it makes advanced security measures nigh impossible. And stuff like this is just completely thrown out:
The blockchain is our voice as shareholders, but it is a time sensitive record. If we make this declaration now, the assumption is that we intend to use it. However, we don’t. The only real tangible effect this motion has is 2 fold. Firstly, it renames the tiers just to make all our discussions that much more complicated. Secondly, implies an arbitrary 10% threshold without proper definition (10% of what? T1-3? No concern for T4?) to trigger a state that shareholders can’t yet apply.
I’d love to have this conversation in Nu 3.0, once we understand the network implications of volume-dependent fees.
@tomjoad, how do you feel about using NuShare sales in a limited fashion before activating tier 6 restricted network access? I’m thinking about what @Cybnate said below…
I believe we at least need a limited number of NuShares to act as a buffer before activating the new tier 6. For example, let’s say we have a supply of 822,144,717 NuShares as we do right now. Shareholders would create a buffer grant of NuShares for FLOT to hold, which would only be around 10% of the existing supply. So that 10% would mean FLOT would hold 82,214,471 NuShares in multisig as a backup buffer for the peg.
In the case that parking interest rates failed to stabilize the peg, this limited NuShare buffer would kick in and FLOT would sell these NuShares in order to buy and burn NuBits. If it became apparent that this limited NuShare buffer wasn’t going to be able to hold the peg, only then would shareholders activate tier 6 restricted network access and higher fees. This buffer would either be enough to stave off the NuBits selloff and protect the peg, or it would give shareholders enough time to activate restricted network access.
The 10% number is just an example and could be adjusted up or down depending on what shareholders felt was a safe number of NuShares for this buffer. Setting it up this way, tier 4 buy side and tier 5 parking rates aren’t the only things protecting the peg and using a limited buffer (instead of LOTS of NuShares) in the event tier 5 fails prevents the NuShares supply from being quadrupled. So the following liquidity model would take effect, altered from what I wrote in my reply above…
1. Tiers 1-3 decentralized liquidity provided by ALPs and MLPs. 2. Tier 4 buy side (Bitcoin) and sell side (NuBits) managed by FLOT. 3. A raised tier 4 threshold for NuShare buybacks to allow for a higher tier 4 reserve. 4. A lower bound tier 4 threshold when it is advised that shareholders activate tier 5 parking rates. 5. If tier 5 parking rates are having no effect and tier 4 buy side funds are still dropping, FLOT will use the limited number of NuShares granted to them to act as a buffer for the peg in case tier 4 buy side funds run out. 6. If it appears that the limited buffer of NuShares isn’t going to hold the peg, then shareholders will need to activate tier 6 restricted network access and higher fees. 7. If tier 6 stabilizes the peg, then shareholders will stop voting for higher fees and the network will return to normal. If tier 6 slows the panic selling, but doesn’t completely stop it, then shareholders can active tier 7 and sell even more NuShares to reduce the NuBits supply even further until the network stabilizes.
I’d like to hear what @Nagalim and @tomjoad think about adding in this limited NuShare buffer before restricted network access kicks in.
I’m sorry, I know I’m being Mr. Cynical all over this thread, but it seems like the only concrete statement that people are making here is:
Park rates are already chosen by consensus, nsr is already used to support the peg when T4 funds are low, and shareholders already know it’s probably a good idea to increase fees when we’re having trouble making ends meet. So ultimately, this thread is just about the T4 % threshold.
I would prefer, as others have suggested, increasing T4 reserves if that’s the case. Tier 4 should have enough reserves to cover the majority of minor demand crash scenarios, and to prevent against speculative attack. NSR sales are unlikely to be very effective in a severe crisis that tanks the value of NSR, especially if there is a time urgency that limits the amount of investors able to provide new capital.
Your proposed sequence of actions makes intuitive sense, except that those 82,000,000 NSR may have lost 90% of their value overnight if the crisis is severe (such as a government announcing its intentions to search for and prosecute NuShareholders, and to criminalize NuBits usage). Tier 4 reserves are more reliable, especially if methods are ever devised to privately hold stable USD among multisig groups. I appreciate that you are thinking through this problem with us.
I will be placing this up for voting within a few hours. Thank you everyone for the discussion; even if this doesn’t pass in the near future, we still have the beginnings of a debate for a mechanism when Nu 3.0 is released.
I can understand that you may have a unique perspective, from NSR buyback, in seeing what would be coming when an extreme adverse event drives down NSR prices. The capital control idea is interesting and I suggest you bring it up whenever there is supporting evidence, so that shareholders can see what you see as Nu grows, paving the way of a consensus. As a motion it is way too early now I think.
Based on feedback here I don’t expect it to pass anytime soon, but I do think there is value in signalling our willingness to consider new macroeconomic tools. Even if it only has 5% support for months, I’m okay with that. It’s a good thing that motions can be voted on forever!