[Discussion] Tier 7 liquidity: Volume-dependent transaction fees and "Restricted Network Access" status

[tomjoad]

@woodstockmerkle’s post made me think about this topic more, and I’m now convinced that Tier 7 is a new form of capital control for digital currencies that hasn’t been discussed among cryptocurrency communities, as far as I’m aware. I know shareholders are busy with finalizing FLOT and liquidity operations, but this topic needs to be addressed in the coming months.

The definition: > Capital controls are residency-based measures such as transaction taxes…

So, the Nu network should realize it has already inadvertently pioneered yet another form of digital monetary policy (the ability to institute capital controls) through variable or volume-dependent transaction fees. This is a topic that needs to be addressed eventually, because it has the possibility of alienating hardcore cryptocurrency enthusiasts. The Nu network should either pass a motion declaring that it will never use Tier 7, or else begin to formulate the conditions under which Tier 7 could be applied. To choose the first option would be to increase the odds of permanent peg collapse however, and so I favor the second option for the reasons that follow.

The International Monetary Fund (IMF) has published an extensive series of articles and papers on capital inflows and outflows. In the case of Nu, our largest risk comes from capital outflows (ie. NBT being sold for fiat). One excellent IMF paper that is relevant to our network is here: http://www.imf.org/external/np/pp/eng/2012/111412.pdf beginning on page 25.

42. Capital outflows that are large, sustained, or sudden can pose significant policy challenges. Some capital outflows are a natural consequence of openness, as foreign investors recoup their investments and domestic investors diversify their portfolios and expand business operations abroad. Outflows can, however, become disruptive in some circumstances, which are usually associated with economic crises. Such disruptive outflows can be driven by domestic factors but they can also be driven by international factors such as global risk appetite, liquidity, interest rates, and global growth, and by contagion effects through trade and financial linkages and investor confidence. **Disruptive outflows can lead to reserve depletion, currency collapse, financial system stress, and output losses. Even short of crisis, large or sustained outflows can pose challenges through their effects on exchange rates, availability of financing, and interest rates.**
    

45. In crisis situations, or **when a crisis may be imminent, there could be a temporary role for the introduction of Capital Flow Management Measures (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, or when the size or duration of the shocks are highly uncertain, CFMs 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.
    

In this case, our primary CFM would be Tier 7 transaction fees.

51. **CFMs in response to disruptive outflows, while they need to be comprehensive, should be temporary.** The right time to lift outflow CFMs will depend on specific country circumstances. In general, outflow CFMs should be lifted when macroeconomic stability, particularly with respect to the exchange rate, debt sustainability, and financial stability are restored, confidence is regained in domestic assets, access is resumed to international capital markets, and foreign reserves climb above critical levels
    

This supports my assertion in the OP that transaction fee hikes should only be in place until equilibrium is restored.

Overall, I think there is a place for Tier 7 liquidity as a last line of defense against currency collapse. The IMF report seems to support limited use of CFM’s like Tier 7 when necessary. Shareholders should begin to discuss under what circumstances other CFM’s may be applied, according to IMF (or other economic expert) guidelines.

[mhps]

It’s rational to have insurance for raining days. But it’s like talking about a baby’s unemployment plan – it’s applicable time is so far in the future that no one for sure knows what the baby will have grown up to before the time comes.

[tomjoad]

It’s rational to have insurance for raining days. But it’s like talking about a baby’s unemployment plan – it’s applicable time is so far in the future that no one for sure knows what the baby will have grown up to before the time comes.

In 14 months of the network we’ve already accessed Tier 6 once during a crisis. It’s not unrealistic that Tier 7 could be required in our next year of operations if an exchange like Poloniex failed.

It’s also possible that shareholders may even prefer Tier 7 over Tier 6 in some circumstances, because it preserves the rarity of their shares. We should consider this topic while we are experiencing success, rather than waiting for a crisis to emerge.

[luxiang7890]

Must consider, the more the baby, the more experience not from the wind and waves, the crisis appear at any time. Not only to consider, in order to anchor the $1, should be in the 7 set a lot of money.

[mhps]

I am not sure which is better, a broken peg or a unbroken peg that costs an arm and a leg to use. The free market would prefer the former. Mental experiment -

If I know T7 is going to be used what if I offer a crypto token that is pegged to 1 NBT, backed by 100% NBT reserve (proceeds from selling the token), and only cost 1/5 or 1/2 T7 fee ?

[tomjoad]

I am not sure which is better, a broken peg or a unbroken peg that costs an arm and a leg to use. The free market would prefer the former. Mental experiment -

This is what I hope more shareholders will discuss here, and I plan on keeping this discussion bumped until more input is provided.

Tier 7’s definition could have an impact on other tiers as well. I’ve always thought Tier 6 is inherently flawed, because we will be attempting to sell an asset (NSR) that derives the majority of its value from the performance of another asset (NBT) when that secondary asset (NBT) is failing. It will lead to a tremendous amount of NSR dilution if the emergency is severe enough. That makes the asset (NSR) undesirable to many people for long-term speculation.

With Tier 7, we impose capital controls on the network for a brief period of time. Our shares retain their scarcity, which should make them much more valuable in the long-run. One of Peercoin’s problems from a marketing perspective is that there is no cap on the number of PPC that can be in circulation. I remember many Bitcoin purists disregarding Peercoin for that reason.

I would even go so far as to say that if the network eliminated Tier 6 and replaced it with this “restricted network access” concept that our NSR would immediately become more valuable. There is always a looming threat that NSR will be diluted in an emergency; with Tier 7, we can have a constantly decreasing supply of NSR forever. That would be very attractive to speculators.

Tier 7 makes more sense to me than Tier 6. In retrospect, I’ve felt this way since the beginning, as shown in a quote from the original NBT burning discussion thread:

**Draft** motion for currency burning

I still struggle a bit with how this burn proposal will work in practice. For the purposes of simplicity I’m imagining a single-currency system like the one that currently exists. When this retirement scenario is reached, it implies that the demand peak of the individual currency has definitively passed. In a scenario where demand is decreasing and is forecasted to never again reach a new peak, the value of NuShares should be equal to its intangible assets (whatever the intrinsic value of its branding elements, the number of existing users as defined by Metcalfe’s Law, etc.). There shouldn’t be any additional value present in the price because future dividends would be expected to be zero.

In that scenario, it should already be very easy for a speculator to buy a large amount of NuShares on the open market for cheap. In a situation with very cheap market-rates for NSR, this would lead to very large NSR/NBT burn exchange rates being offered, to the point that very few NBT would be purchased. This is the same final outcome that is present in our current design.

Tier 6 is flawed but easy to digest. Tier 7 requires us to think of Nu as a true currency that uses the same tools (capital controls) all real currencies do in times of crisis. I think we should eliminate Tier 6 because it will require mass dilutions at large enough scales, and because it will make NuShares worth much more if there is no possibility of dilution.

What do other people think? I may introduce a motion because I feel very strongly that Tier 7 is a more effective economic tool than Tier 6.

[Nagalim]

sell an asset (NSR) that derives the majority of its value from the performance of another asset (NBT) when that secondary asset (NBT) is failing

I basically agree with you as far as T6>T4 has been implemented (using BTC as a middle man). However, true T6 should be implemented more in a way like seeded auctions, where we trade NSR for NBT directly. In that case, you can see that a failing NBT actually causes the NSR/NBT price to go up rather than down (anti-blackswan). Still, you make a very good point that a dropping NSR/BTC price is indicative of failure in the long term and will force Nu to go through some bloody shrinking of operations. T7, as you put it, is the cure by providing real long-term profits to keep people’s spirits up.

[tomjoad]

However, true T6 should be implemented more in a way like seeded auctions, where we trade NSR for NBT directly. In that case, you can see that a failing NBT actually causes the NSR/NBT price to go up rather than down (anti-blackswan).

I’ve largely avoided commenting on seeded auctions because I don’t understand this logic. Could you explain further? That statement would appear to be in direct opposition to my statement:

In a scenario where demand is decreasing and is forecasted to never again reach a new peak, the value of NuShares should be equal to its intangible assets (whatever the intrinsic value of its branding elements, the number of existing users as defined by Metcalfe’s Law, etc.). There shouldn’t be any additional value present in the price because future dividends would be expected to be zero.

If it is common knowledge the price of NuShares will benefit from increased NuBits demand, I don’t see how decreased NBT demand can also increase the price of NSR. A failing NBT peg would cause the NSR/NBT price to go up if, and only if, the valuation from NuShares was independent of NuBits demand. You cannot reliably back an asset with a derivative of itself.

T7, as you put it, is the cure by providing real long-term profits to keep people’s spirits up.

Tier 7 has nothing to do with profits. It is a form of capital control that organizations like the IMF recommend using to mitigate currency collapses in times of demand decrease. Normal volume-dependent transaction fees (say 1%) are a form of profits; Tier 7 would exist when those fees are pushed to an extreme level that make transactions undesirable for a portion of users.

[Nagalim]

the price of NuShares will benefit from increased NuBits demand

There are two scales to consider, the short term and the long term. On the short term what we have to be concerned with is an NBT peg break causing an NSR black swan event. From the perspective of the auction, both the NBT and the NSR price would dive at the same time, meaning the NSR/NBT ratio would stay the same, or at least be higher than the NSR/USD ratio. On the long term we regain the peg but put continual stress on NSR and everything you say is true. However, in the long term our system is able to pass motions and adjust the Txn fee accordingly.

Tier 7 has nothing to do with profits.

I’m confused by what you mean by capital controls other than volume-dependent transaction fees (which you say are profits and therefore not T7). Can you give an example?

[tomjoad]

I’m confused by what you mean by capital controls other than volume-dependent transaction fees (which you say are profits and therefore not T7). Can you give an example?

Normal volume-dependent transaction fees (say 1%) are a form of profits; Tier 7 would exist when those fees are pushed to an extreme level that make transactions undesirable for a portion of users.

Volume-dependent (or any variable) transaction fees can be used as a tool for profit (at perhaps 1% Tx fee per transaction) or extended to a form of capital control (at 20%, 40%, 50% Tx fee per transaction). The percentage transaction fee that regular profits turn into capital controls is a matter of opinion, but you could perhaps argue that capital controls are in place once our transaction fee is set above our competitors.

This is why I am so enthusiastic about this concept:

Tier 6 requires us to sell an asset (NSR) at the bottom of its valuation to support NBT, requiring greater dilution. It requires us to trust a multisig group. It may require multiple auctions and multiple custodial grants to “mop up” all the excess NBT in a severe crisis, creating a disincentive to purchase NSR when the crisis first begins. Even in good times, Tier 6 threatens the perceived scarcity of NuShares because any investor knows the NSR supply could quadruple overnight. It is slow, it is centralized, and it is inefficient.

Tier 7 can be applied in whatever block duration it takes to adjust a transaction fee. It has no scalability limits, in that we could set up to a 100% transaction fee if needed. It requires no centralized trust of a multisig group. It improves the perceived scarcity of NuShares, which will increase the number of users advocating for our network. And, it is a tried and tested macroeconomic tool for currencies, just as interest rates (Tier 5) are. It is fast, it is decentralized, and it is efficient.

Unless I hear a compelling reason I will be introducing a motion soon to re-define Tier 6 as the application of “restricted network access” status through increased transaction fees. The motion will call for the burning of all NSR held by FLOT. The old Tier 6 has served its purpose as a final peg defense, but I believe a superior option is now available to our network.

[Nagalim]

Unless I hear a compelling reason I will be introducing a motion soon to re-define Tier 6 as the application of “restricted network access” status through increased transaction fees. The motion will call for the burning of all NSR held by FLOT. The old Tier 6 has served its purpose as a final peg defense, but I believe a superior option is now available to our network.

So you propose to increase our static fee in response to a peg break scenario? The response to our product losing value is to reduce its utility? That seems like a death sentence. To burn our NSR to try to achieve this is a step backwards from overall confidence in the peg. There is no way I would vote for such a motion given the details in this thread.

[tomjoad]

There is no way I would vote for such a motion given the details in this thread.

As is your right, of course. Major changes are always controversial.

I would be interested in hearing more logic-based arguments against the case I’ve presented however, as I suspect you’re glossing over the materials I’ve introduced in favor of an emotional defense. What I’m proposing is exactly what professional economists would recommend if they ran a digital currency.

What specifically do you disagree with in my quote below? I’m interested in understanding your objections better.

Tier 6 requires us to sell an asset (NSR) at the bottom of its valuation to support NBT, requiring greater dilution. It requires us to trust a multisig group. It may require multiple auctions and multiple custodial grants to “mop up” all the excess NBT in a severe crisis, creating a disincentive to purchase NSR when the crisis first begins. Even in good times, Tier 6 threatens the perceived scarcity of NuShares because any investor knows the NSR supply could quadruple overnight. It is slow, it is centralized, and it is inefficient.

Tier 7 can be applied in whatever block duration it takes to adjust a transaction fee. It has no scalability limits, in that we could set up to a 100% transaction fee if needed. It requires no centralized trust of a multisig group. It improves the perceived scarcity of NuShares, which will increase the number of users advocating for our network. And, it is a tried and tested macroeconomic tool for currencies, just as interest rates (Tier 5) are. It is fast, it is decentralized, and it is efficient.

[Nagalim]

Sorry I have so many objections I’m having a hard time articulating.

1. We aren’t ready. Static txn fees simply cannot make us the $$ we need. You’ll need to see fees like 100 nbt, long passing the point of being able to buy coffee with what we are calling a currency.

2. It breaks some things. I’ll give you the fact that pools can burn as much nbt as they want in operation without significantly affecting the reward system (I.e. Nu can just give more nbt to the operator). However, the trusted address holders would have a much larger risk if they needed to hold a vastly larger amount of funds to do their job.

3. Transactions on the blockchain are not transactions on exchange. If I can’t withdraw my nbt without paying a huge fee, you know what I’ll do? Sell it.

4. Anonymity and security pays a huge price when you can’t freely move outputs around.

5. Maybe I’m confused. If we spiked the fee to 100%, wouldn’t nbt (and therefore nsr) become worthless?

[Cybnate]

Questions for NuShareholders:

1. Because shareholders should only consider this action after believing Tier 5 and Tier 6 liquidity will fail, should we classify this defense mechanism as Tier 7 liquidity, or as something else outside the liquidity operations framework? In my view this is an “active” tier that requires shareholder action, similar to Tiers 1 through 6.

2. How should the proportionality constant be set in regards to buy-side/sell-side liquidity ratios? I am interested in feedback from @woodstockmerkle, @nagalim, and others who took an interest in forecasting transaction fee rates in the volume-dependent transaction fee thread.

3. How should Nu categorize this mechanism? I have chosen to define it as a state of “restricted network access”. Users may still manually trade their NBT to other (very) trusted users through private keys, but our network would not facilitate the trade over the blockchain unless a premium is paid during the emergency period.

All feedback is welcome. In contrast to my discussion on Tier 0 liquidity, I am more convinced this belongs in the liquidity operations framework because of its requirements for explicit shareholder action.

1. Had to think about it. The tier model confused me at times as it consists of funds and mechanisms/policies. I had to readjust my view of it to make sense of it. The tier model is really a peg defense mechanism. In that context tier 7 as the ultimate mechanism makes sense.

2. Refrain from answering this on. Don’t have a strong view and there are others who have a better understanding and more experience with this.

3. I wouldn’t call it restricted access, capital control as you mention later is better. We should live in a free world, however scarcity always comes at a premium price. The proposal to slow velocity as capital control appears to be widely adopted in the financial world. The 100%+ reserve model is too expensive and involves a lot of ‘dead’ capital. I think the tier 7 policy should go together with say a 33%-50% reserve model (in tier 4 or 6?) to be able to manage the smaller bumps in the road without pain. Application of tier 7 will hurt (and cost) and therefore should only be used for the serious potholes in the road which threatens to stop the network in its tracks (losing the peg) when the reserves (shock absorbers) can’t take any more.

I don’t think it is too early to think about it as @MHPS raised and start preparing for it, see reasons below. We have seen a sudden increase in demand, it is not unusual that this temporarily turns around again. The market may test the peg sustainability before taking stronger positions.

Exchanges had issues, too. Withdrawal transactions wouldn’t go through. NuDroid had a centralized solution. I prefer to see technical problems properly worked out before pushing variable fee further.

I’ve seen the same turmoil. It was certainly not only NuDroid. The CoinOmi App and exchanges also had problems. Exchanges mostly work with fixed fees. When the fee is changing often exchanges won’t be happy as they would loose money or users won’t be happy because of excess fees until the exchanges have changed their process/wallets. We need to make sure the Apps, community/exchanges are prepared and take into account that funds and time may be required to implement this.

So you propose to increase our static fee in response to a peg break scenario? The response to our product losing value is to reduce its utility? That seems like a death sentence.

Selling NSR when it is not in demand is also a death sentence. We are talking about already having a serious problem and trying to come up with the best solution given the circumstances. I tend to prefer the capital control model. The challenge is to have adequate reserves. The tier 7 policy should only be executed to cover those unknown risks. The other tiers should cover the known risk to an affordable extent.

The fees should be adjusted to the amount. Exchanging 1 NBT might only cost 0.25 in a crises, but exchanging 10,000 NBT might cost 2,500 NBT and increasing exponentially.

Eventually we will be paying for going cheap on risks in the short term.

[tomjoad]

I think the tier 7 policy should go together with say a 33%-50% reserve model (in tier 4 or 6?) to be able to manage the smaller bumps in the road without pain.

This is a great observation. If we are uncertain about predicting future demand and would like more reserves, I would prefer increasing Tier 4 and eliminating Tier 6. One could argue that increased BTC exposure in Tier 4 is undesirable, but it is still better than selling NSR at the absolute bottom of its worth in Tier 6.

Increased Tier 4 reserves, and a capital control tier as a last resort are superior to our current design, in my opinion.

[masterOfDisaster]

One could argue that increased BTC exposure in Tier 4 is undesirable, but it is still better than selling NSR at the absolute bottom of its worth in Tier 6.

That’s why I was arguing for refilling T4 buy side with T6 NSR sales before T4 buy side is completely depleted.
Commenting the rest of the T7 idea needs some more time. I haven’t thought it through thoroughly enough.
I like the proposal, the discussion about it, although I’m not sure what position I will have once I’ve established a position.

[tomjoad]

That’s why I was arguing for refilling T4 buy side with T6 NSR sales before T4 buy side is completely depleted.

The better approach would be to simply hold more BTC in T4 from NBT sales. So, a custodian sells NBT for BTC, and FLOT keeps a higher percentage of NBT that are not used for development, buybacks, or dividends. This would support our goal of increasing the perceived scarcity of NSR. What do you think?

[masterOfDisaster]

withhold more BTC

means more volatility risk.
Investing BTC in NSR buybacks increases the scarcity while decreasing BTC vlatility risks. But this possibly pushes the price of NSR far north and reduces the effectivity of the buybacks.
Unless there are effective means to reduce BTC volatility risks (PPC, clever ways to make use of USD, etc.) I don’t feel comfortable with increasing T4 size.
Nu already pays a big price for compensating liquidity operations.
Increasing T4 size increases the risk for BTC volatility.

I still haven’t thought it to the end.
But I fear using T7 without having used all appropriate lower layers before damages Nu or rather NBT more than using T4-6 first.
At least until we reach the point that T7 needs to be activated as well, because T4-6 are not sufficient.

I like a model that involves T4, 5, 6 and ultimately 7.
Use T4 to fill T1-3.
Use T5 or/and T6 if T4 runs low (using T5 makes only sense if the buy side runs low)
Use T7 if the effect of T5-6 isn’t sufficient.

[tomjoad]

means more volatility risk.

Cybnate summarized my points well here:

Selling NSR when it is not in demand is also a death sentence. We are talking about already having a serious problem and trying to come up with the best solution given the circumstances. I tend to prefer the capital control model. The challenge is to have adequate reserves. The tier 7 policy should only be executed to cover those unknown risks. The other tiers should cover the known risk to an affordable extent.

Tier 6 requires us to sell an asset (NSR) at the bottom of its valuation to support NBT, requiring greater dilution.

One could argue that increased BTC exposure in Tier 4 is undesirable, but it is still better than selling NSR at the absolute bottom of its worth in Tier 6.

[masterOfDisaster]

Selling NSR when it is not in demand is also a death sentence.

One could argue that increased BTC exposure in Tier 4 is undesirable, but it is still better than selling NSR at the absolute bottom of its worth in Tier 6.

That’s why I’m all for selling NSR before the situation is dire
Vote for this, if you want to have NSR sold before NSR reach the bottom of their worth - which might be close to emptied T4 buy side with pegs that can’t be kept.
Better have T4 refilled way before that happens

I really need to think more about it, before I’m in the position to really contribute to the discussion.
My initial impression still is:
I think it’s good to have more tools/tiers to deal with situations.
I prefer to exhaust/max out tiers before the next higher tier gets involved.
Anything else will be hard to handle. How else do you know when the time is right?

It’s one of the problems we experience with the current liquidity operations on T1-3. T2 and T3 are not properly involved. The result is a mess.
I wouldn’t want to expand that mess to higher tiers as well.