Block reward system

[creon]

After thinking a while about this topic I wanted to come out with this idea, especially now since the burning motion has passed. Right now the blockchain generates about 57,600 NSR every day, distributed among the staking clients w.r.t. to their staking weight. Right now almost none of these blocks contains a transaction besides the staking transaction, so the amount of NSR burned through staking is negligible.

I believe it is special to Nu that the main incentive to produce blocks is not to receive the block reward, but to be eligible to cast a vote which will have a positive long-term effect on the investment made (at least in the opinion of this shareholder).

Therefore I would like to ask you what you would think about either reducing the block reward to a very small amount (e.g. 1 NSR), or to remove it completely and instead to pay the transaction fees to the block finder instead of burning them. The latter option would ensure a fixed NSR supply.

[tomjoad]

It’s an interesting idea, but I think the rewards psychology underlying minting rewards is a powerful draw for all PoS currencies. The prospect of earning a reward at irregular intervals is really the basis of all sorts of gambling, from slot machines to MMORPGs. The small 40 NSR reward makes users feel good about securing the network. It also helps them rationalize the “risk” they feel in making those shares available (even if a properly encrypted and locked wallet is in no danger of theft).

It’s also worth debating whether the primary perceived reward is voting rights (as it might be for dedicated users who are passionate about enhancing the value of their investment by improving the Nu network) or if it is monetary gain (for users who view NuShares as a passive speculative investment). There will be people in both camps, although my instinct is that the second group is sizable. I think with the new burning mechanism that the concept of a fixed NSR supply has been overtaken by a system that adjusts the supply of NSR relative to the demand for NBT.

[creon]

It’s also worth debating whether the primary perceived reward is voting rights

Yes, I mean this would be answered by the corresponding motion of this protocol change. I just think that Nu has a non-altruistic incentive to stake that goes beyond the block reward. This makes Nu very special. We furthermore saw here that most blocks do contain recent votes.

In my opinion the whitepaper does not really describe how to deal with minting clients which solely stake for the purpose of the block reward. The voting system in its current state does not allow for neutral votes. So from the system design I think the monetary incentive is always assumed to be present in combination with a voting incentive.

[tomjoad]

In my opinion the whitepaper does not really describe how to deal with minting clients which solely stake for the purpose of the block reward. The voting system in its current state does not allow for neutral votes.

I hadn’t quite thought of the issue like that. So your argument is that if the minting reward was dropped to 0 NSR, we would still expect to see people minting for the privilege of voting. Doing so would still secure the network provided a high enough percentage of minters were motivated by voting rights, which I think is the case with the initial NSR distribution. Having a 0 NSR minting reward would also tend to self-exclude all those who are primarily motivated by monetary gain, which has the additional benefit of eliminating uninformed “no” votes.

I actually quite like the idea, even if it does force me into making a trade-off as an active minter. I would lose the 2% gain in NSR as a relative percentage of the network compared to those who don’t currently mint, but would be better assured that important votes can pass.

Perhaps the big question is how this would improve on data feeds that will automatically trigger in the client. I see the future of Nu as being one that has politicized data feeds that collect voters with similar economic beliefs. It will allow the network to more efficiently make decisions, provided the data feeds don’t reject all motions/custodians they didn’t introduce. As I’ve mentioned in the past, should that occur the Nu network would need to transition to fixed-schedule elections with a slate of motions, proposals, and custodians, and a “first-past-the-post” voting system. But, either way, I think data feeds are here to stay - and once enough quality feeds are automatically enabled in the client, the problem of “no” votes largely disappears.

I think your idea is worth discussing further though. I can definitely see the merits of it.

[creon]

I would lose the 2% gain in NSR as a relative percentage of the network compared to those who don’t currently mint, but would be better assured that important votes can pass.

There is no fixed inflation of 2% with the burning motion. The block reward creates about 21 million NSR per year. On the other hand the new motion allows us to arbitrarily adjust the NSR supply. So if you are minting 40 NSR today, and hold it for one year, you can impossibly know how much of the total supply those 40 NSR will be.

Maybe we decided to double the NSR supply, which would be an equivalent block reward of 20 NSR, or we burned half of the NSR, effectively doubling the block reward. With the burning motion the value of the block reward can only be defined in short term.

[Sentinelrv]

I was one of the first ones to receive NuShares, but due to some real life issues I had fallen extremely behind on this forum and had no idea what was going on. I wanted to mint with my NuShares because I knew I was losing money due to inflation. I was behind though, so I didn’t know what votes to set. Because of this, I refused to mint with blank votes because I knew it could negatively impact the network.

The continuous dilution of my NuShares however gave me a huge boost of motivation to read this entire forum as quickly as possible (about 1 month) so that I could get caught up to current events and start minting and voting along with everyone else. If that fear of dilution was not there, this motivation I had to get caught up wouldn’t have been as strong. It’s possible I might even still be sitting on the sidelines, but I doubt it.

Anyway, I think the minting reward is very important to ensure proper motivation. Without a personal financial consequence, complacency can set in and the willingness to let others do all the work. The detrimental effects not participating in voting decisions can have on the network may not show until later, but the dilution of NuShares is personal, painful and immediate.

[cryptog]

I was behind though, so I didn’t know what votes to set

You could use my feeds

[Sentinelrv]

Data feeds weren’t released until later unfortunately, plus I’d rather know things first hand before voting.

[creon]

The continuous dilution of my NuShares however gave me a huge boost of motivation

And you were right! Because the coin supply increased while your own balance was fixed. But now things are different.

The coin supply is now just a number that can be defined by the shareholders. To this number we arbitrarily define, we then add 21 million NSR per year, always, through the block reward. So when we decide to adjust the NSR supply we will fully calculate the block reward into the equation, and therefore eliminate its actual effect on the supply.

So the block reward is already just a number without any concrete long-term value and therefore it lost its original function in the economy since the burning motion passed.

[GreatScott]

To this number we arbitrarily define, we then add 21 million NSR per year, always, through the block reward.

It is my understanding that the block reward will be increased to keep a 2% inflation rate, though I was unable to find where I read this. I think it should also decrease to keep a 2% inflation rate in the event a large amount of NSR gets burned. Since lowering the block reward by 10 NSR is a large move from 40 NSR, I think the block reward should be dynamic based off of NSR moneysupply.

I do not like the idea of eliminating the block reward. It encourages shareholder participation and with null votes it is my understanding that non-voters will be able to subscribe to a null feed and not prevent motions from passing.

[creon]

It is my understanding that the block reward will be increased to keep a 2% inflation rate

The Nu system uses a constant block reward. It can only be changed through a hard fork, so we would need to hard fork each time be adjust the NSR balance. To ensure a 2% block reward without hard forking each time we would need to change the reward to a coin age based reward system, like PPC.

subscribe to a null feed

Hmm, how would such a null-feed be designed?

[GreatScott]

The Nu system uses a constant block reward

It turns out it was the draft motion for the burning that led me to believe the reward is not constant:

**Draft** motion for currency burning

The mint reward will be adjusted according to the total number of NuShares in the network. Specifically, each time the supply of NuShares reaches 50% and 100% of its orginal base (1 billion), the mint reward will also increase by 50% and 100%. For example, when NSR supply reaches 1.5 billion, the reward should be 60 NSR. At 2 billion, 80 NSR, at 3 billion, 120 NSR, at 4 billion, 160 NSR, and so on.

Hmm, how would such a null-feed be designed?

A data feed that aggregates all proposals and motions with an accompanying null vote.

[creon]

Ah thanks! I haven’t noticed this automatic adjustment, ok, this will give a very rough approximation of a 2% inflation, right.

A data feed that aggregates all proposals and motions with an accompanying null vote.

You mean if we implement the YES/NO/NULL voting system. This discussion is still ongoing and there are many open questions, like e.g. what happens to the block window of 10,000 blocks, the block maturity etc. Note that this also can not be applied to park rates.

[GreatScott]

This discussion is still ongoing and there are many open questions, like e.g. what happens to the block window of 10,000 blocks, the block maturity etc. Note that this also can not be applied to park rates.

I suppose if we did solve the 10,000 block dilemma it would also allow for park rate null vote without much trouble. The only solution I can think of requires ignoring blocks containing null votes from the 10,000 block window on a per motion basis effectively increasing the time to get something passed, but not affecting the amount of votes needed. But then even so, you are right this is all just a discussion at the moment.

I think we should first discuss how we handle when/if the NSR moneysupply were to fall. It is a new concept to Nu. Are we going to lower it to 20 NSR per block but only when there is 500 million NSR in circulation? It would be possible to make it adjust at the protocol level, right?

I am really hesitant to get rid of block rewards though, to be honest. I know I am not gaining anything by receiving them, but I think it is a psychological thing that encourages participation in securing the network. Keep in mind this is also the security of NuBits holders. If non-voters do not put up stake, it makes it easier to achieve 51%.