High PoS interest rates are harmless and would benefit security

General
1

In discussions about minting incentives, I keep reading that 1% of interest is not much of an incentive to mint, but that a higher rate would be detrimental, as it would generate too much inflation.

At first, as I was mentally nodding to these statements and didn’t pay much attention to them.

But I then realized that they are actually extremely misleading: high interests rates would not be an issue at all. Of course, they would result in a faster increase of the monetary supply. But this is a false issue: your stash would be increasing as fast. Your hotdog could be worth twice as much, but it wouldn’t be an issue if you earned twice as much and had twice as much savings. It’s basically the same as stating prices in mPPC: numbers get 1000 times as big, but you still pay the same thing.

So nominal prices (numbers of PPC) go up, but in the end you are not losing any purchasing power. Unless you don’t mint, and that’s the point.
If you normalized all prices to account for the inflation of the money supply, then you would see that minters get richer and non-minters get poorer, as relative wealth is slowly redistributed towards minters.

In a nutshell, the misconception is that inflation results in loss of purchasing power. This is not the case with the inflation caused by PoS interests, because interests are distributed equitably, relatively to what each person owns, assuming everybody is minting. So the people who don’t mint are the only ones to suffer from high interest rates. Conversely, it’s the people who do not mint that allow minters to get relatively richer: if everybody mints, no one gets relatively richer by minting.

With the current situation (1% interest, 10% of coins used in PoS minting), minters get 1% richer every year while the total supplies increases by 0.1% (from PoS). In relative terms, minters get 0.9% effective interest ((1+0.01)/(1+0.010.1)-1), and non-minters take a 0.1% hit only (1-1/(1+0.010.1)). Not much of an incentive.
Let’s increase the interest rate. If the interest was 10% and 10% of coins were used for minting, then the effective interest rate would be 8.9%, and the effective penalty for not minting would be 1%. Then people would undoubtedly start minting more.
If the interest was 10% and 50% of coins were used for minting, then the effective interest rate would be 4.8%, and the effective penalty for not minting would be 4.8% too.
If the interest was 10% and 90% of coins were used for minting, then the effective interest rate would be 0.8%, and the effective penalty for not minting would be 8.2%.

I saw people considering to have adaptive interests rates, depending on how much coins are used in PoS, to maintain a certain level of security. Given that the more people mint and the higher the interest rate is, the harsher the effective penalty for not minting is, I suggest that the protocols continuously adjusts the interest rate to aim for an effective 5% penalty for not minting, with the goal of improving the security of the network.

2

i like your statements.
i agree

3

This analysis seems correct to me. However, I think the 1% was chosen in order to diminish the effect of compounding interest.

Imagine if the growth were 100% per year. Minters would compete fiercely for blocks in order to maximize compounding interest. This might increase the incentive to mint on multiple chains at once in order to maximize the chance of finding a block in the accepted chain, leading to an increased number of orphaned blocks.

Just a thought. I could be completely wrong.

4

This may not be technically possible, because you cannot tell if dormant coins will suddenly mint and received a large reward. The only way you can be sure that coins will not mint is when their coindays are destroyed by a transaction.

5

[quote=“Chronos, post:3, topic:2621”]This analysis seems correct to me. However, I think the 1% was chosen in order to diminish the effect of compounding interest.

Imagine if the growth were 100% per year. Minters would compete fiercely for blocks in order to maximize compounding interest. This might increase the incentive to mint on multiple chains at once in order to maximize the chance of finding a block in the accepted chain, leading to an increased number of orphaned blocks.

Just a thought. I could be completely wrong.[/quote]

Compound interests could be an issue, but it is easily solved: for the moment, interests are proportional to the number of blocks since the last time the coins were transacted, which is the source of the issue.

We just have to change it to be applied per block (e.g.: if the interest is 1% per block or per 10 min, then the interest for one hour would be 1.01^6-1=6.2%). This ensures that you always get compound interests.

Well, the target would be computed based on a moving average of past data. If you want to minimize potential deviations from the prediction, I also had an idea to encourage regular minting by capping minting rewards per block (Cryptoblog - notícias sobre bitcoin e criptomoedas!).

6

with the mechanism when more users are minting, higher posreward, with a maximum of 5%, i think the maximum will be reach rather quickly , dont know how this will invluence the incentive

7

[quote=“Chronos, post:3, topic:2621”]This analysis seems correct to me. However, I think the 1% was chosen in order to diminish the effect of compounding interest.

Imagine if the growth were 100% per year. Minters would compete fiercely for blocks in order to maximize compounding interest. This might increase the incentive to mint on multiple chains at once in order to maximize the chance of finding a block in the accepted chain, leading to an increased number of orphaned blocks.

Just a thought. I could be completely wrong.[/quote]
I agree with that. It only works in a fair way if almost everyone has the opportunity to mint on a regular base. Blocks will get more scarce when interest rates go up, leaving only very large stakeholders with the ability to mint and leaving the rest with high inflation. I think it is very important to find the balance between the rewards and blocks available (the scarce good). As this demand will be variable over time (competition or other external factors) adjusting the rewards will be a key instrument to keep the network attractive for more than only the savings of a few.

8

IF your right, start your own coin with those parameters.

Coins and parameters are much like the navier stokes equations, where the boundary conditions can have very different outcomes.

The only reason people are not minting in Peercoin is they are waiting for the cold locked minting feature when that arrives there will be a massive increase in minting.

PeerCoins boundary conditions have been specifically chose for high value transaction.

It would be counterproductive to change the interest rate at this time.

9

[quote=“jubalix, post:8, topic:2621”]IF your right, start your own coin with those parameters.

Coins and parameters are much like the navier stokes equations, where the boundary conditions can have very different outcomes.

The only reason people are not minting in Peercoin is they are waiting for the cold locked minting feature when that arrives there will be a massive increase in minting.

PeerCoins boundary conditions have been specifically chose for high value transaction.

It would be counterproductive to change the interest rate at this time.[/quote]
I guess you are replying to OP, but I wished I was as confident as you regarding the uptake of cold minting. I think you underestimate the laziness factor. And if you are right we get indeed into the situation that only a few are able to mint and compete for blocks fiercely. In both cases you should probably adjust the reward either upwards or downwards depending on demand.

Only pooling would take out the laziness factor, but that is very undesirable from a security perspective.

10

I actually suggest the opposite: when fewer users are minting, higher PoS rewards.
The 5% limit is on the effective penalty, not the interest rate. Another way to say it is that interest rate is adjusted so that PoS leads to a 5% annual growth of the money supply, regardless of how many users are minting (users who skip minting opportunities forego their interests).

Expected result:

[ul][li]If few users are minting, the interest rate will be high, and it will incite other users to start minting.[/li]
[li]As minters join in, interest rate goes down to maintain the 5% growth, until minters stop joining.[/li]
[li]When minters stop joining, the minters’ share of the money supply will still keep growing, as they are the only ones who get interests, while the non-minters’ share shrinks. This increases the security of the blockchain.[/li]
[li]A bigger and bigger part of the money supply will be used for minting. The network will detect it and reduce the interest rate to maintain a 5% growth.[/li][/ul]

11

I actually suggest the opposite: when fewer users are minting, higher PoS rewards.
The 5% limit is on the effective penalty, not the interest rate. Another way to say it is that interest rate is adjusted so that PoS leads to a 5% annual growth of the money supply, regardless of how many users are minting (users who skip minting opportunities forego their interests).

Expected result:

[ul][li]If few users are minting, the interest rate will be high, and it will incite other users to start minting.[/li]
[li]As minters join in, interest rate goes down to maintain the 5% growth, until minters stop joining.[/li]
[li]When minters stop joining, the minters’ share of the money supply will still keep growing, as they are the only ones who get interests, while the non-minters’ share shrinks. This increases the security of the blockchain.[/li]
[li]A bigger and bigger part of the money supply will be used for minting. The network will detect it and reduce the interest rate to maintain a 5% growth.[/li][/ul][/quote]

aha, ok, I think I have to see things more from the perspective of the network instead of the user also, is often confusing for me.
what you would want, is constant annual growth X (5%, or whatever),
Xtotalavailablecoins - totalavailablecoins = number of peercoins to be added N , (annually)
52560 blocks / year

ok , so you will get ((X*total available PPC) - total available PPC)/52560 = posreward, which isnt related anymore to your own coins in terms of percentage,

#need to get the average coins minting over the past some blocks (say 2016 or whatever) to get some moving average

ok so the pos difficulty can be used to determine the number of minters more or less
to keep the inflation stable, you need to adjust the block reward, if 10% of peercoins is minting reward would be 10 times as much

, probably possible to write down in one neat looking formula… , just some thoughts anyway

edit: i think you could make variable posrewards which adds incentive, at which percentage it should be , the global inflationrate, i dont know, could still be 1% possible

it would just be difficulty based posreward, same as how pow works now (im also ignoring pow completely in my example btw),

12

The problem with this analysis is that it assumes a closed system. In reality the new coins will be sold on the market for BTC and fiat to buy other things or rebalance portfolio. Sandwiches and electricities are not prices in PPC. When most people are hoarding, the price of the coin will be decided by a few who trade. The price of PPC will go down in terms of BTC and fiat as a result.

Another thing is those who keep coins in circulation will get penalized due to the 30day waiting period. This encouragement of hoarding will not only directly hurt the spenders, but also creates a image that this is hoarder’s coin for the rich to get richer by doing no work.

Read this where people talk about investing high “interest rate” coins. It’s not working out. None of the coins with sky-high POS reward is doing well.

That said I am not sure the 1% reward is spot on. It certainly seems low if looked at as the sole incentive to minters because most owners of PPC are not minting. So I am for a method to reward those who mint often, but not to cause arms race (a form of POW).

13

I personally think the reward rate should be just enough to keep the network secure, but not more.

It is a problem that minters are rare but we need to analyze this carefully.

The user base is still quite small so naturally there could not be too many minters. I agree that cold storage minting will boost the number of minters significantly but the more important factor is the size of user base. If minting is safe and low cost, people WILL mint. Cryptocurrency will not prevail without some kind of ideology education and culture. For peercoin, it should be ‘for the sake of the whole network, let’s mint!’

How to keep minting from becoming a centralized business is a potentially serious matter that the whole community should keep an eye on.

14

@irritant
It seems a little too complicated. I was rather thinking of something like this (note: it is a little different from what I described in the original post).
You would look at the blocks of the last 6 months or so, add up the coin-days destroyed in PoS, and compare it to the total amount of coin-days generated by the network during the same time:
fractionMinting = SUM_6months(CDD_PoS) / ((totalSupply_now - totalSupply_6monthsAgo)/2 * 6 months)
You then calculate the interest rate to get a 5% growth:
interestRate * fractionMinting * totalSupply_now = 0.05 * totalSupply_now
After rearranging:
interestRate = 0.05 / fractionMinting
With the modification for compound interests, you have to calculate the interest rate per block.
(1 + ratePerBlock)^52560 = 1 + interestRate
raterPerBlock = (1 + interestRate)^(1/52560) - 1
You can then calculate interests when someone mints:
reward = coins * (1 + ratePerBlock)^ageInBlocks
But I think it would then be better to cap ageInBlocks, to penalize coins that were offline for too long.
cap = ageInBlocks * (averageCDDtoMint + 2 * stDevCDDtoMint) / CDD_newBlock
averageCDDtoMint and stDevCDDtoMint can be calculated formally, but I currently don’t know exactly how PoS scores are calculated.

[quote=“mhps, post:12, topic:2621”]The problem with this analysis is that it assumes a closed system. In reality the new coins will be sold on the market for BTC and fiat to buy other things or rebalance portfolio. Sandwiches and electricities are not prices in PPC. When most people are hoarding, the price of the coin will be decided by a few who trade. The price of PPC will go down in terms of BTC and fiat as a result.

Another thing is those who keep coins in circulation will get penalized due to the 30day waiting period. This encouragement of hoarding will not only directly hurt the spenders, but also creates a image that this is hoarder’s coin for the rich to get richer by doing no work.[/quote]

The price per PPC will go down, but it will be countered by the fact that everybody owns more PPC. It might not be very practical to have varying prices, but we could normalize them by using a unit derived from the fraction of the total supply you own.
And it is a closed system: peercoins change hands, but they are still part of the network. Buyers will have a choice to mint or not as well.

Regarding the 30 day waiting period: this is completely true and I didn’t think of that in this context. The issue is already there in the current Peercoin implementation, though not as exacerbated. I think it needs to be fixed.
I’ll admit to not having a ready-made solution, though I played with the idea of paying interests as change when coins are transacted. To prevent this from being abused, we could limit it to small transactions (in terms of CDDs). Splitting big amounts into many small amounts would incur high transaction fees when combined. But that’s just an idea, I haven’t checked how viable it actually is.

15

It’s a closed system if you never spend or trade the coins for something else – then a coin is always a coin. But when you start to spend or trade, a coin will be worth less than a coin after more coins are created with less work.

16

Ah, you mean there is an influx of new coins? Sure, but it does not really change anything if the new coins are distributed as the older coins, does it?

17

I think mhps is making a very good point. Inflation may not be a big concern for hoarders holding large stakes indefinitely and minting every 30 days, but high inflation is bad if we want to promote Peercoin as a medium of exchange. I think the concept of “just enough” is very important when it comes to incentivizing minting. If Peercoin is ever to become the “backbone” of any real economy it will need to provide liquidity as well as storage of value.

18

i think 1% is just right, as you don’t want to punish too much ppl that decide not to mint (pure investors). there is no reason why everyone has to mint. if the reward were like 5% it would become a coin where either you mint or you sell the coin so that you don’t have to pay for the inflation. with 1% it’s acceptable for a pure investor not to mine, lose the 1% but still hold the coin. when cold minting will be out many more ppl will mint anyway.

19

I don’t have anything against pure investors per se, but I think it is also important to realize that this behavior weakens the security of the network. Pulling coins out of the minting cycle makes it easier for anyone to set up an attack.

20

in one way i like the idea of 5% “dividends” as a incentive for investors to stick with a coin and stabilize the market value. its a common way of assuring loyalty for nowadays stores of value, namely blue chips on the stock market.

on the other hand, the relatively low 1% may not trigger greed and its not worth to try to cheat the system. its just enough to give an extra good feeling to the ones that want to support the network anyways and invest 50€ in a raspberry pie and subsequently 10€ in electricity per year.

this bare foot movement character is really charming and may become handy when the stakeholders of the currently fashionable IPO based PoS coins are starting to do stupid things out of greed. i cannot imagine examples of how this might look but people get very creative when it comes to maximize profit. with 1% interest its even more difficult to find a way to play the algorithm in a way that provides you any substantial profits.