[Draft] Motion to adjust the dividend distribution structure of custodian grant

[KTm]

Shareholders:

I believe strongly in the Nu network and in the prospects that it has to change the financial landscape for millions of people. Tempering this optimism is the knowledge that any missteps now, while the network is small, will have potentially catastrophic outcomes.

The community has engaged in and continues to have many excellent conversations about how best to grow and protect the network. A lot of those have focused on the ability for the network to scale appropriately and what to do with profits generated from the sale of NBT in the form of dividends.

I’ve struggled with the situation that I find myself in as an elected custodian. The proposal that I drafted, and that was passed by the Nu Shareholders was flawed in it’s conditions and execution of dividends. With the information that we have now about the state of the network that we find ourselves in and of the dynamics of the markets that we operate on, I believe it to be irresponsible to blindly go forward with the sales-based dividend distribution structure.

I am not, however, suggesting that shareholders not be paid out dividends on the full amount that was granted to me – on the contrary, I’d like nothing better than to be able to fully meet the agreement. What I’d like to do, however, is propose a modification to the structure that is weighted towards protecting the network and allowing it the liquidity to develop and scale out.

Please review my draft motion and let me know if what I’ve introduced meets the needs of the Nu network’s NBT users and of the NSR holders. If the feedback is good, I’ll be able to quickly formalize this motion and put it to vote – if there are many questions, I’ll work with the community to mold the motion content to best fit.

Thank you for the consideration and your on-going support.

//KTm

---

///// BEGIN MOTION (DRAFT)

Proposal Summary:

The successful passage of this motion will alter the contract between the custodian of Grant B7mmVdVQ1SNNcT9zuQRK1B3Cbvo8vHeoB1 (“Grant KTM” - 1,800,000 NBT + fees) and the Nu Shareholders regarding how dividends on the sale of granted NBT will be paid out.

The proposed structure of dividend payouts will adjust from the current “period sales goal” system of variable length and fixed percentage, to that of a monthly, fixed percentage system based on the total amount of buy-side liquidity available to the Nu network.

Rationale for Proposal

Many assumptions were made when Grant KTM was initially introduced by Kiara Tamm and passed by the Nu Shareholders. Two important assumptions turned out to be incorrect:

• The belief that in addition to Grant KTM, the Nu network would have access to a significant amount of liquidity from LPCs. This additional buy-side support would be a backstop that would allow the payment of dividends to shareholders on the schedule as proposed;

• That the main market that Grant KTM would be operating on would be NBT/USD, rather than various NBT/crypto markets.

The inability to forecast those two conditions has made it both difficult and dangerous for Grant KTM to be operated in accordance with all of the conditions of the original proposal. This motion is an attempt to course-correct the grant’s operations and provide shareholders with their expected dividends, in full, while simultaneously attempting to continue to stabilize the Nu liquidity pool as the network grows.

Like any successful venture, growing the business is the surest way to support the distribution of profits to shareholders. Simplifying the dividend distribution process and making it predictable to all shareholders provides an improvement to a system that has proven to be extremely variable and needlessly complex and difficult to manage.

By tying dividend distributions to the size and success of the Nu network rather than to the previous schedule, it is the intent of this proposal to gracefully and appropriately compensate shareholders for the expansion of the network. At the same time, should there be slow or stagnant growth, the custodian is able to protecting the Nu network’s NBT holders’ interests by making sure that a steady supply of buy-side liquidity is available for NBT repurchase and that it isn’t dangerously reduced by contractually obligated dividend distribution schedules.

Adjusted Operation Details

If passed, the dividend payout schedule previously approved will be modified as such:

1. The “Period 001” dividend’s remainder (20,000 NBT of PPC) will be paid immediately to satisfy the previously voted on motion. The completion of this distribution will satisfy the original “Period 001” dividend target of 70,000 NBT. Once this has been paid, “Period 001” will be the last of “sales goal”-based dividend cycles.
2. Moving forward, the custodian will pay dividends every thirty (30) days, with the first distribution to commence on [date and time to be determined].
3. The amount of the dividend to be paid out will be proportionate to [ex. 5%, final TBD] of the average amount of liquidity available to the Nu network on the buy-side during the thirty (30) days preceding the date of the distribution.
4. Dividends will be distributed to shares held as of 00:01 GMT on the day of the distribution.
5. Should the calculated amount of dividend distribution exceed the available amount of buy-side liquidity that Grant KTM controls, sell-side NBT will be used to purchase the remainder from other Nu custodian’s buy-side holdings at the then-established market rate.
6. Once the funds held by Grant KTM have been exhausted through dividend distributions the operations will be considered concluded and a post-operations report will be prepared for the Nu Shareholder.

Example Period Dividends

Time Period: 15-JAN to 14-FEB
Avg. Buy-side Liquidity: 180,000 NBT
Dividend (at 5%): 9,000 NBT

Time Period: 14-FEB to 16-MAR
Avg. Buy-side Liquidity: 291,000 NBT
Dividend (at 5%): 14,550 NBT

Time Period: 16-MAR to 15-APR
Avg. Buy-side Liquidity: 400,000 NBT
Dividend (at 5%): 20,000 NBT

Time Period: 15-APR to 15-MAY
Avg. Buy-side Liquidity: 700,000 NBT
Dividend (at 5%): 35,000 NBT

///// END MOTION (DRAFT)

[mhps]

This is certainly a huge step in the right direction than just paying from a fixed part of the grant money.

A better step is paying in proportional to the increase of the buy-side liquidity in the period. No increase no dividends.

An even better step is paying from profits generated from exchange spread, tx fee, etc.

[tomjoad]

I think this proposal is a much-needed adjustment to the dividend schedule. As Jordan Lee mentioned, this proposed dividend schedule adjustment introduces additional counterparty risk for the benefit of a stronger short-term peg. I think it is a trade-off that is worth making at this point given @KTm’s honesty and transparency.

While I have faith that the design of NBT will eventually allow stability in the absence of USD reserves through NBT-for-NSR burning, this proposal is an important intermediate step that needs to occur to protect the quality of our offering.

I will be voting for this as soon as a motion is finalized.

[Yurizhai]

Love it. Not only is it the ‘safer’ route in upholding the peg, it gives shareholders a guaranteed dividend every month which will surely increase morale and interest in Nu.

[desrever]

Very good proposal.

Just a couple of questions :

fixed percentage system based on the total amount of buy-side liquidity available to the Nu network.

Very good choice, I think this value is a good proxy to estimate Nu wealth status in real time.

Now you are the only dividend custodian, but I believe we should edit the motion text to make space for potential-future dividends custodians. What if another dividend custodian gets elected? I believe it should affect the % somehow, probably on a case by case basis. I would add just a bullet point that specifies that in the event of a new dividend custodian joining the network, the percentage can be reviewed to match the specific goal declared above.

Dividends will be distributed to shares held as of 00:01 GMT on the day of the distribution.

While scheduled payouts can be simple and good to manage, they can also give rooms to speculation on PPC and NSR price, imho. While I generally do not mind other people taking advantage of it, I fear the existence of some possible attack where knowing the exact time of dividends paying can be used harming the Nu network in some way.

Maybe saying that dividends will be paid every 30 days with a ± 5 days random offset, will help making such attacks (if they exist) harder to perform.

EDIT: I would probably also add another bulletpoint that allows KTm to skip one or more payment dividends in case of emergency (for example the unlikely scenario where buy liquidity is highly needed to maintain the peg and paying the dividends could harm the pegging capabilities of the network)

[seki]

i agree

[Ben]

First off, I like it, a lot. Now, my feedback (which turned out to be quite lengthy):

The amount of the dividend to be paid out will be proportionate to [ex. 5%, final TBD] of the average amount of liquidity available to the Nu network on the buy-side during the thirty (30) days preceding the date of the distribution.

A better step is paying in proportional to the increase of the buy-side liquidity in the period. No increase no dividends.

I’m torn between these two variations of the approach, but after reading @mhps’s take, I think it makes a lot of sense. It’s going to be less per dividend for the shareholders but that isn’t necessarily a bad thing. What it also does it correctly adapt to changing market conditions. For example (using your numbers above as a base):

// Proposed model

Time Period: 15-JAN to 14-FEB
Avg. Buy-side Liquidity: 180,000 NBT
Dividend (at 5%): 9,000 NBT

Time Period: 14-FEB to 16-MAR
Avg. Buy-side Liquidity: 291,000 NBT
Dividend (at 5%): 14,550 NBT

Time Period: 16-MAR to 15-APR
Avg. Buy-side Liquidity: 400,000 NBT
Dividend (at 5%): 20,000 NBT

Total dividend for 90 day period (as proposed): 43,550 NBT

// Using the increase in liquidity

Time Period: 15-JAN to 14-FEB
Avg. Buy-side Liquidity: 180,000 NBT
Dividend: 0 NBT
// ^ No baseline to set the difference from because this is the first month

Time Period: 14-FEB to 16-MAR
Avg. Buy-side Liquidity: 291,000 NBT
Liquidity Increase (month-over-month): 191,000 NBT
Dividend (at 5%): 9,550 NBT

Time Period: 16-MAR to 15-APR
Avg. Buy-side Liquidity: 400,000 NBT
Liquidity Increase (month-over-month): 109,000 NBT
Dividend (at 5%): 5,450 NBT

Total dividend for 90 day period (using liquidity delta): 15,000 NBT

Here’s what that would look like if in the “16-MAR to 15-APR” period the network’s liquidity went down on average compared to the previous 30 days:

// Using the increase in liquidity

Time Period: 15-JAN to 14-FEB
Avg. Buy-side Liquidity: 180,000 NBT
Dividend: 0 NBT
// ^ No baseline to set the difference from because this is the first month

Time Period: 14-FEB to 16-MAR
Avg. Buy-side Liquidity: 291,000 NBT
Liquidity Increase (month-over-month): 191,000 NBT
Dividend (at 5%): 9,550 NBT

Time Period: 16-MAR to 15-APR
Avg. Buy-side Liquidity: 290,000 NBT
Liquidity Increase (month-over-month): -1,000 NBT
Dividend (at 5%): 0 NBT

Total dividend for 90 day period (using liquidity delta): 9,550 NBT

While the second model does reduce dividends, I see two major benefits with it:

1. It reflects a better connection between the performance of the Nu network and profits that are extracted from it
2. When a month has been “bad” (negative liquidity), dividends would not be required and that (presumably) needed liquidity would be retained by the network to back its NBT holders.
3. If the network is stagnant, buy-side funds stay there to provide reserves for NBT holders to exit the market if needed. If the network is growing and more grants and custodians are coming online the total impact that @KTm’s grant has as a percentage of the total reduces accordingly. Eventually this grant will be diluted to a large degree by the size of the market, rendering the impact that it has on total operations small.

My vote is to update the draft to use the second option. Ok, moving on to the next item of business, if others (and @KTm) agree in going with that plan there is a section that will need to be updated.

Moving forward, the custodian will pay dividends every thirty (30) days, with the first distribution to commence on [date and time to be determined].

There is no guarentee that a dividend would be paid (if the delta between average liquidity from period to period decreased), so you’d want to modify this item to read something like:

Moving forward, the custodian will pay dividends every thirty (30) days if the average available liquidity on the buy-side has increased from the previous 30 day period.

Someone else may have a better wording for that, when I re-read it is clear to me, but maybe not to others.

Dividends will be distributed to shares held as of 00:01 GMT on the day of the distribution.

Won’t you need time to convert holdings into peercoin? In the previous proposal you indicated a 72 hour period. This may fit nicely with @desrever’s suggestion:

While scheduled payouts can be simple and good to manage, they can also give rooms to speculation on PPC and NSR price, imho. While I generally do not mind other people taking advantage of it, I fear the existence of some possible attack where knowing the exact time of dividends paying can be used harming the Nu network in some way.

Maybe saying that dividends will be paid every 30 days with a ± 5 days random offset, will help making such attacks (if they exist) harder to perform.

If you allow yourself up to 120 hours to process the data, acquire the peercoins, and then announce the dividend amount and market price for PPC, it may be easier to manage (and less likely for someone to game). I still agree with your previous policy of issuing dividends based on holding NSR at a certain time in the past, so you could still use that “36 hours prior to the announcement of the dividend” plan. Does anyone else have a recommendation for a better / simplier / more standard way of handling it?

Next:

Should the calculated amount of dividend distribution exceed the available amount of buy-side liquidity that Grant KTM controls, sell-side NBT will be used to purchase the remainder from other Nu custodian’s buy-side holdings at the then-established market rate.

Will you use PPC held under your custodianship first to meet a dividend obligation?

EDIT: I would probably also add another bulletpoint that allows KTm to skip one or more payment dividends in case of emergency (for example the unlikely scenario where buy liquidity is highly needed to maintain the peg and paying the dividends could harm the pegging capabilities of the network)

I agree with @desrever, give yourself a “way out” to respond to external and unexpected situations…just do a good job of making sure that the NSR shareholders are well aware of the situation and the reason that you are invoking a discretionary holdback.

Now you are the only dividend custodian, but I believe we should edit the motion text to make space for potential-future dividends custodians. What if another dividend custodian gets elected? I believe it should affect the % somehow, probably on a case by case basis. I would add just a bullet point that specifies that in the event of a new dividend custodian joining the network, the percentage can be reviewed to match the specific goal declared above.

This is a good point. You should update the motion text to indicate that when new dividend custodians are put into action that the percentage used to calculate the dividend will be reviewed for modification. We should have enough time to consider the “big picture” before you and the other custodian(‘s)(s’) operations run into one another, so on a case by case basis that value can be updated to be balance the needs of the shareholders vs. the needs of the NBT holders.

Otherwise, beyond those items, the last things I’d like you to consider are:

1. There’s no end-date set for this grant’s operation. Not that you really had one in the previous proposal, but this could mean that you’ll be operating the grant into the foreseeable future. How does this mesh with your costs and does it affect your personal calculations regarding fees?
2. Where do you plan on getting the data to support the average liquidity for the network? Will a new tool need to be developed or is there already something in production that can be retro-fit to work for you? How will this information be conveyed to the Nu shareholders so that they can independently verify the dividend calculations?

[KTm]

Thank you all for the feedback. I will take it under review and update the text for the final motion with the modifications that I agree with.

The suggestion to move to a dividend paid out based on the change in liquidity over consecutive thirty day periods is a well considered and is in line with how a traditional company (or, in the future, a DAO) would handle their payouts. Effectively, if the previous quarter (in our case, previous thirty days) was under-performing there are no profits to distribute.

I will also update the language of the distribution schedule to reflect that a dividend isn’t ensured every thirty days but only when the conditions have been met (positive delta between periods). I will also take the suggestion under consideration to include a provision for discretionary withholding (with a valid and publicly communicated reason).

While scheduled payouts can be simple and good to manage, they can also give rooms to speculation on PPC and NSR price, imho. While I generally do not mind other people taking advantage of it, I fear the existence of some possible attack where knowing the exact time of dividends paying can be used harming the Nu network in some way.

I’d like to hear suggestions from the community for how this should be handled. Normally there are four important dates when a dividend is declared by a traditional company:

1. Declaration date: The date when the announcement that a dividend will be issued is made,
2. Holder of record date: The date that a share must be in your possession to be eligible for part of the distribution,
3. Payment date: Self-explanatory, but the date that the distribution is sent out to the eligible shareholders,
4. Ex-dividend date: The date that the company officially trades with a market capitalization that has been reduced by the amount of the dividend paid.

1, 2, and 3 are all relevant to Nu grants that issue dividends. Technically 4 is as well, but for all practical purposes, NBT dividends behave differently in their effect on NSR pricing than they would if it was the Nu NSR capitalization that was being reduced.

What structure makes the most sense to the community? What risks are there if the distributions are made on a set schedule? Would this not behave similar to how traditional companies act when they issue quarterly dividends?

Will you use PPC held under your custodianship first to meet a dividend obligation?

Yes. If this results in lower than required buy-side liquidity on the NBT/PPC trading pair, I will use available buy-side reserves from other markets (as long as the balance is maintained) to make this work.

You should update the motion text to indicate that when new dividend custodians are put into action that the percentage used to calculate the dividend will be reviewed for modification.

I would like to get @jmiller’s opinion on this. She is the first custodian that I would need to coordinate distribution percentages with.

There’s no end-date set for this grant’s operation. Not that you really had one in the previous proposal, but this could mean that you’ll be operating the grant into the foreseeable future. How does this mesh with your costs and does it affect your personal calculations regarding fees?

For now it is expected that costs for the next year will be manageable within the original grant’s fee. Later this year I will evaluate the situation and review the operations to see if this is still the case. If additional fees are required in the future for any reason, a new grant proposal will be submitted to the Shareholders and they can decide how we will move forward.

Where do you plan on getting the data to support the average liquidity for the network? Will a new tool need to be developed or is there already something in production that can be retro-fit to work for you? How will this information be conveyed to the Nu shareholders so that they can independently verify the dividend calculations?

I will communicate with the Nu product team (including you) to make this determination. The information is there (using the getliquidityinfo RPC command), it just need captured and then presented somewhere. My goal will be to have a solution in place before motion voting begins, or if not implemented, at least the requirements and tools to use defined with a date to have it ready.

@woolly_sammoth and I have already started talking about a tool he developed, https://bitbucket.org/mj2p/nu-data, which looks like it may be a starting point.

---

If anyone else has feedback, now is the time to offer it, before I finalize the motion. I would like to get submit it for Shareholder review ASAP.

//KTm

[KTm]

Shareholders:

If there is no additional feedback, I will move forward with drafting the final text for the motion. I would prefer to hear from more of you,

especially concerning the timing of dividend payouts,

but I will not indefinitely hold this motion. If I’m able to get to a point that the tools and data are totally available to Nu Shareholders to use to identify what is going on with the Grant’s operations, I expect that NSR exchange rates will “price in” upcoming dividend distributions appropriately. Coupled with using the average liquidity of the preceding thirty day period, it will be difficult to game either the market or the price of NSR.

I may also be revealing my naiveté; if there are attack vectors that have not been considered that will be detrimental to the Grant’s operations, Nu Shareholders, or the Nu network itself, I NEED TO BE MADE AWARE OF THEM so they can be mitigated to the best of my abilities.

I light of yesterday’s discussion about devaluation, I plan to include a a section in the motion that describes discretion that I will use when deciding to issue dividends if the price of holdings has declined due to external factors, even when the amount of liquidity available in the market has increased period-over-period. A decision like this would not be made in a vacuum. Plenty of public discussion would be held if the situation appeared to be likely for an upcoming dividend period.

As a companion to that last point, I also plan on including a provision that would allow me to pay out dividends at a higher-than-expected rate in times of externally-influenced holdings appreciation. By doing this, I can disseminate profits to the Nu Shareholders at the time that they have occurred rather than holding on to them and risking a reversal.

//KTm

[cryptog]

To me it sounds good too. Will vote for the upcoming motion at this stage.

[rodchi]

I just read the draft of this motion and I like it very much. Mainstream people are not buying Bitcoin because the volatily is too much for most to handle. A monthly dividend would make NuShares very attractive, and it could very well attract mainstream (e.g. Wall Street) investment. NuBits will also benefit. People will invest in NuShares and use NuBits to completely transactions. I expect to see more merchants adopting NuShares.

I will vote “Yes” to the final version of this draft.

[Cybnate]

To me it sounds good too. Will vote for the upcoming motion at this stage.

Count me in, had to sit on it for a while, especially on the schedules. But I’m good with it now.

[JordanLee]

On occasion, BTC received from the sale of NuShares has been used increase the amount of buy side liquidity (by purchasing NuBits on the open market with them). When this occurs, it seems like it shouldn’t contribute to dividend distribution.

[desrever]

if there are attack vectors that have not been considered that will be detrimental to the Grant’s operations, Nu Shareholders, or the Nu network itself, I NEED TO BE MADE AWARE OF THEM so they can be mitigated to the best of my abilities.

Not anything that I am aware of. This has not been done before, at least in crypto, so I guess we will need to go empirical.

When I mentioned possible future attacks, I simply wanted to make sure we do our best to shape the contract in such a way that allows custodians to limit potential damages.

No-one raised concerns of particular attacks made possible by knowing payout schedule, except from me. So we can comfortably say that I was making some premature optimization. We will probably need to make this discussion one year from now, with 12 payments on our shoulders and some data to talk about.

[pennybreaker]

On occasion, BTC received from the sale of NuShares has been used increase the amount of buy side liquidity (by purchasing NuBits on the open market with them). When this occurs, it seems like it shouldn’t contribute to dividend distribution.

I’m not sure if I agree with this. If shareholders wish to exclude these funds from dividend distribution, It seems that this type of balancing should be coordinated with the custodian and occur off of exchange for proper accounting to take place. There is also a loss to the shareholder, in fees, when this happens on exchange.

[mhps]

proper accounting

I agree. We should avoid using accounting equivalent of global variables and goto statements.

[pennybreaker]

KTm:
The amount of the dividend to be paid out will be proportionate to [ex. 5%, final TBD] of the average amount of liquidity available to the Nu network on the buy-side during the thirty (30) days preceding the date of the distribution.

mhps:
A better step is paying in proportional to the increase of the buy-side liquidity in the period. No increase no dividends.

I’m torn between these two variations of the approach, but after reading @mhps’s take, I think it makes a lot of sense. It’s going to be less per dividend for the shareholders but that isn’t necessarily a bad thing. What it also does it correctly adapt to changing market conditions.

Kiara and Jamie are losing their worth due to vulnerabilities explained here and here. Also, buy-side liquidity can change drastically from moment to moment. Unless additional liquidity is continually added to the network, they could end up in an endless cycle of never meeting a buy-side liquidity percentage based dividend payment structure due to continual losses and a moment to moment coin flip of which side the most liquidity will be.

I believe that it makes more sense to base the dividend payment structure on total network liquidity (buy+sell), so that it is always based on actual network growth, and so that the shareholders (the community) has a chance to earn some of what will otherwise end up in the hands of currency traders. The community will have a strong incentive to keep the liquidity coming so that they may continue receiving dividends, and by keeping more of these funds within the community we will help pave a path to new liquidity provider proposals.

Besides, who says that dividends have to come out of the buy wall? We’re moving away from sales based dividends, right? Why don’t we always take dividends out of heavier wall?

[cryptog]

@JordanLee I have noticed several times that the buy side was was well kept at a high level at times where BTC was losing value rapidly. This is impressive. But there is a scenario in which if BTC continues to lose value with no recovery, it could deplete the reserves to zeo. Do you have confidence that Nu has enough reserves even in that worst-case scenario?