Proposal for Adjusting Polkadot's Inflation System: Reducing Issuance and Complexity

Governance
26

Hi @alice_und_bob

Thanks for your comments and thanks for putting forward an alternative (apart from different parameters). This helps people to make up their mind.

Let me first say that I don’t have an opinion right now on changing the spending cycle but it appears to me a bit off-topic, because I don’t see how it is correlated with any change to the inflation system. This might be better discussed somewhere else.

Regarding your proposal on a fixed token issuance

While I am not strictly opposing the idea, I am not convinced how this makes things easier to understand than a percentage-based inflation relative to total issuance. The latter guarantees a stable APY for stakers (at least within the staking rate, but that’d affect your model, too). And this is the metric that nominators are focusing on and from where the system derives (some) of the economic security (by paying validators). Having a fixed token issuance would reduce total inflation and APY over time. It might take some time for this to become meaningful (but then also for it to become effective?), but eventually the fixed token inflow would be dwarfed by the total issuance and we’d need to transition to a new system.

I’d further argue that the simplified model I am proposing is making it already much more intuitive and it’s not an argument to come up with an even simpler model (while I stated above that I even disagree its simpler). I also don’t think it’s fair to assume that people are too stupid to understand it. “Memeability” in that regard is not more important than having a stable security budget.

As I said, I don’t strictly oppose the idea, but I don’t see how it gives us any benefits. If you want to make your proposal more convincing, it would good to provide some justification for the initial parameters (total inflow) and make some projections about the future. This should be done within the framework that we are discussing (i.e., treasury inflow, staker APY, and total inflation). For example, your proposed “25M DOT per Year” is difficult to compare to my proposal on a glance, but translates to 1.64M DOT per 24 days.

A last note: I completely disagree that we should divert any funds from coretime sales to the Treasury. I am sure you are familiar with my reasoning, but for others interested, here is the approved RFC that is stating some arguments.

In my opinion, we should strive towards a system where we have “inflation” in terms of new token issuance to pay for public goods (i.e., Treasury and Security), but have a counterweight to it by burning coretime. Hopefully, and that takes time to see, we can scale the available blockspace so much that individual coretime is affordable but collectively accrues sufficient burns.

3 Likes
27

In my mind, this proposal is too radical and we’d be changing things too much, too quickly. It’s a complex system where many actors and mechanisms are intertwined.

Further, I don’t think we need to “force healthy spending of treasury” by capping the inflow as radically as you propose it. Yes, we’ve seen excessive spending recently. BUT, the spending came from many years of accruing funds, where we’ve seen very little outflows. The reasons were simply that Gov1 had large barriers to entry that got removed by OpenGov and that the Treasury was less known/used as viable funding source.

That means, even with my proposal of 1.15M DOT, the spending must already become much more efficient and conservative, simply because the Treasury is depleting otherwise. The 1.15M DOT is “smoothing” the outflows towards the average, accounting for the years with low outflow. I fear that people subconsciously make the error to attribute the current large spends to inflow rather than savings.

It remains a delicate trade-off between having a potent Treasury and not overburden every token holder with the minting of new DOTs for it. I’d still argue that the biggest DAO in the world needs to be potent enough to fund substantial endeavours that ultimately lead to its success. Optimising spending is super important for that, but I feel there is a lot of movement in the community in that regard.

3 Likes
28

Re: stability

Yes, proportional inflation gives the same APR towards infinity, but I would say a fixed inflation gives a stable enough APR over the foreseeable future.

As an example, let’s compare 10% inflation per year to a fixed 140m DOT inflation (10% of the current supply) per year:

image
image903×499 13.8 KB

We can see that the inflation reduces from 10% to 5% in 10 years, which is a very long timeframe to adapt if something turns out to be running bad. It is also progressively changing the environment, so that everything can be observed and reacted to.

Re: Are people too stupid to understand that exponential inflation is okay?

After having spent what feels like hundreds of hours on Twitter arguing with people about Polkadot tokenomics, I can tell you that: yes, on average people are not educated enough (or sometimes too stupid) to understand that exponential inflation is okay. They do not understand that it is a simple shift in purchasing power. They are burned by our generational pain of getting rekt from TradGov policies. They understand the Bitcoin narrative of “capped supply gud”. They don’t understand the dynamics and effects.

If you look at the numbers above, there is no risk to the security budget. Polkadot is paying 1 billion USD per year for security! It could easily afford to pay 500 million or 300 million or maybe even 100 million at its current state. So the security of the network would not at all at risk if we reduce the security budget by 50% over 10 years. And that isn’t even considering that we have to assume that Polkadot will become much more successful (DOT exchange rate aka security budget goes up (sorry mods)) in the future; OR become irrelevant anyways.

Both systems are equally sound, but one has a better chance to be accepted by people than the other.

Re: Comparing systems

You can achieve both outcomes with both systems, it’s just a matter of mapping params and updating them regularly via OpenGov. I think the actual parameters are a second step that will be negotiated via OpenGov anyways and where we can only guess now. Our work right now is to discuss what possible constraints we might want to see fulfilled and build the right parameters and mechanics for it.

Within the current system (and your improvements), we could achieve linear inflation by updating max_inflation once a year and setting it gradually lower. Within the system that I suggest, we could have exponential inflation by bumping the fixed inflation up every year. So both systems can represent both approaches.

It is more of an ideological question if we want to conform to the realities of how people comprehend economics or continue to do it as we did.

2 Likes
29

Very good point from Tommi to redirect Coretime revenue to the Treasury to reduce dependance on inflation :ok_hand:

1 Like
30

Thank you for your detailed reply. While I understand the concerns raised by parachain teams, I believe we need to approach this issue with caution and consider some key factors:

Firstly, it’s important to note that the entire cryptocurrency market has experienced similar price movements during this bear cycle, regardless of individual project inflation rates or staking rewards. This suggests that Polkadot’s high APY may not be directly responsible for its price performance or parachain competitiveness.

Secondly, the proposed reductions in APY lack a clear scientific or economic basis. Whether we’re considering a eg 1%, 2%, or 5% reduction, there’s no evident data or model demonstrating how these specific changes would impact our ecosystem’s health or DOT’s value. Without this analysis, such adjustments risk being perceived as arbitrary measures rather than sound economic policy.

Moreover, we should consider that many investors, particularly institutional ones, view staking rewards as analogous to dividends in traditional finance. This perspective has been a key driver of institutional interest in proof-of-stake networks like Ethereum. In fact, Polkadot currently offers the highest rewards of among major PoS networks. This is a significant competitive advantage that we haven’t fully capitalized on in our marketing to investors. Reducing our APY could diminish this unique selling point and potentially drive institutional interest towards other networks.

Let’s focus on enhancing Polkadot’s fundamental value proposition, improving parachain interoperability, and driving real-world adoption. Additionally, we should better market our high staking rewards as a competitive advantage to attract more institutional investors. These efforts are likely to have a more significant and sustainable impact on our ecosystem’s success than adjusting inflation rates based on short-term market sentiments.​​​​​​​​​​​​​​​​

1 Like
31
  • :100:
    That’s the 1st step to implement asap
32

As you say and with the limited information that we have now about the future, it’s more an ideological question whether we’d want to keep a %-inflation on total issuance or a fixed token supply. This also fuels some of additional arguments against a schedule:

  1. The initiative to reduce inflation is driven by the sentiment of many ecosystem members that feel it would help their applications / outreach / and the general ecosystem now. Therefore, we should be more focused on what the inflation should be in the immediate future rather than implementing a system with projections towards 10 years. We can always reduce it to 5% if we feel it then to be appropriate, or we can lower it even further. There is no point in committing to a schedule when OpenGov can always change it and the future is uncertain (we have little experience with coretime burns).
  2. People that prefer to have a lower inflation would benefit from lowering it sooner rather than later.
  3. The projections you give (which are on 10% not 8%) are only the worst case assuming no burning of tokens. As I mentioned, I think should strive towards a system with significant counterweight to inflation, i.e., burning tokens.

Edit: Maybe you want to specify all parameters that you think should be used?

33

This proposal is very sensible. I agree with Tommi here, after spending a lot of time on socials I get the feeling we need a system that can be better accepted by the token holders. I know it sounds populist, but we need to accept the fact that the current inflation system is:

  • to difficult to understand and explain
  • continuously attacked and creating negative narrative (and I do not believe any narrative is good, at least not in this context)

Jonas’s proposal makes the system much less complex, which is good. However, we need to be careful and “smart” and change the current negative narrative around DOT inflation.

I think the inflation reduction system must have some predictability. It worked very well with Bitcoin, and I think we should consider it. The linear inflation system provides exactly that: inflation decreases over time. It is “memable” and creates a positive social narrative every year. Solana has done exactly that → https://solana.com/staking#what-will-the-inflation-rate-be, a simple system that anyone can understand, predictable and cyclical like Bitcoin, and with inflation reduction as a long term goal.

The decrease in inflation will be barely noticeable to token holders IF treasury funds are spent to add value to Polkadot and we work together to make this ecosystem fun for users to use and devs to build on.

1 Like
35

I’m surprised such changes are being introduced into the next release without a mandate from open governance.

No change can be introduced without a mandate from OpenGov. Every runtime upgrade requires a referendum. Note, that the changes in the next runtime upgrade will not change anything to the inflation model except that it becomes easily configurable through OpenGov. But if you oppose even that, feel free to vote against the runtime upgrade once it’s referendum is up.

1 Like
36

I like it, but to me it would be more complete if all the tokens the treasury receives for the transactions it currently receives + those from Coretime when it becomes active were burned.

This way the treasury would only receive 580K DOT per era and that’s all it would have to spend

37

I agree.

38

By the way, if I remember correctly, the conclusion of the discussion about changing the inflation, which was held some time ago, was that we should at least wait and see how the (initially slight) deflationary instruments of coretime selling affect the overall situation. What happened to this idea?

39

Incentives are misaligned between ecosystem actors, mainly caused by Treasury spending.

Jonas, you wrote all of that, but you have yet to explain, in simple terms, why does the Treasury need that much income? Everyone keeps saying that Treasury needs to be healthily funded in order to continue to be able to built the ecosystem (something that I bought into for several years), however… in practice that’s just not true. The Treasury is not used to build up the ecosystem… so why keep funding it so much?

Based on the Treasury spending habits, I see no valid reason to funnel more than 1%
of Era protocol rewards into it. Like… 0 logical business reasons. And NONE OF YOU can explain WHAT THE BENEFIT IS.

I see no reason for the Polkadot community to move forward with this proposal. It’s badly written from an analytical perspective of the entire ecosystem and it is also badly timed.

Scrap this and go back to the drawing board.

2 Likes
40

The problem with Polkadot is that most people making decisions here are too smart , and you think everyone is smart and mathematically skilled as you. Overcomplicating things is also one of the pain.

The concepts of Polkadot need to be much simpler to be accepted by the market, developers, and businesses.

  • Reduce inflation - GOOD

  • Linear inflation - GOOD

  • Limited Token Supply - GOOD

  • Burning part of transaction fees and Coretime fees - GOOD

  • Support for independent validators on unique locations - GOOD

  • More utility cases for DOT - GOOD

  • Increased OpenGov activity and rewards for active participants - GOOD

  • Big treasury spending during a bear market- BAD

  • Grifters and treasury milking - BAD.

  • Shame, disrespect and consequences for those who cheat and lie to community, even big players. - GOOD

I suggest formulating a message to the community in such a simple way that it will be understood and accepted.

4 Likes
41

A correct and simple summary

1 Like
42

Agree on all points!
I like the idea of the rewards for active participants in OpenGov, because most holders just ignore voting on proposals.

2 Likes
43

The right decision, I fully support it!

1 Like
44

@jonas and others
1.
1.115M DOT inflow to treasury every 24d would be without the option to burn the unused amount, as it is now?
I don’t have a very concrete idea of ​​what this amount is needed for but it seems high to me. (Haven’t we already collected a large amount of stablecoins in the treasury via Hydradx over the last six months in order to be able to support salaries in a weak market situation?)
2.
What exactly do we hope to achieve by reducing inflation from 10% to 8%? (Please explain in less theoretical terms than so far.) And in what time frame should we achieve the desired effects? Would it be possible to make the treasury and its capital liable in some way if the effect does not occur as desired?
3.
In principle, I think 15% treasury inflow is (far) too high in relation to the staking rewards issued. After all, it is these 15% that work against the token holders (and their rewards) in an open market. A hefty amount.
4.
It is claimed that the various security functions that we currently have are no longer needed. Why exactly? Because we have a consistently high staking rate? What would happen if, in a bull market with a 2d unbonding time, a situation arose in which a very large number of token holders suddenly parted with their tokens at the same time for understandable reasons?

1 Like
45

I’ll likely get my post deleted, but echoing earlier comments about dumbing down the proposal to make it more attractive is paramount.

My two cents as a non-savvy peasant investor in DOT:

My only considerations are increased value through either token value growth or as holding DOT for dividend income. My continued holding in DOT is an extension of my belief in the project. However, any action that does not lead to either of those two outcomes (price growth or stable dividend income) is detrimental to investors holdings and likely results in further erosion among the investor base. This will ostensibly lead to transitioning of investor funds from DOT to elsewhere.

Without DOT making it clear that their proposal benefits DOT holders through fundamental token upward momentum (another way to say price appreciation), then the proposal to lessen inflation and increase funds further to the Treasury comes off as selfishly at the expense of DOT holders many of whom, including myself, have given up on potential gains elsewhere to invest in DOT.

It remains fundamentally unclear why the Treasury requires so much DOT to function. The external perception among DOT holders (non-developers) is Treasury has a spend problem through poor governance controls as evidenced by recent marketing projects for whom value was less than clear.

If the proposal to cap inflation and thus stakers APY is to proceed, then yourselves need to provide a much clearer benefit to holders beyond dangling two day unstaking. Otherwise, please be do not be surprised if there’s a reduction in DOT holders over time.

3 Likes
46

It would also be interesting to hear from validators whether they would feel compelled to increase their commission on nominators due to the reduction in inflation in order to be able to continue working safely.
That would of course be another factor that would contribute to the game.

3 Likes