Proposal: Inflation Update 2.0

Overview

Our first inflation update was submitted on 11/19/2020 and went live on 12/3/2020, which lowered both inflation max and min by 5% (to 53% and 41% respectively) and changed the inflation rate change to 24% from 12%. This has resulted in a decrease of ~0.02% per day in the rate of inflation; as of 6/8/2021, actual inflation is currently 47.85% as of 6/24/2021.

Proposal of Changes

Our modeled token supply by 9/25/2021 (12 months after mainnet launch) is supposed to be ~149.3mm AKT and we’re currently projected to exceed that threshold. In order to ensure we follow our deflationary curve as closely as possible and stay true to our economics as our network has matured, we will be proposing our second inflation proposal to adjust the following parameters:

Inflation Rate Change: from 24% to 100%
Inflation Max: from 53% to 40%
Inflation Min: from 41% to 25%

These changes will put us back on track and start to decrease the growth rate of additional supply as deployments and usage continue to grow.

These parameter changes are expected to take place in mid July, accounting for 1 week of discussions and 2 weeks for the proposal voting period.

8 Likes

I’m in support of anything which strengthens the stability of the system. As Greg has mentioned before, the network is theoretically open to 51% attacks because our pricing is so low.

Lowering the inflation raises the probability of a rise in AKT price which, alongside benefiting long-term stakers, also secures the security of the system further.

Only comment I have is on the “Inflation Rate Change from 24 - 100”. Seems like this would be a “Deflation Rate Change”: as it increases deflation. Or perhaps I’m just misunderstanding what that term means and its a moot point.

In general, I believe I get the initiative, which is to slow the release of new supply, which will theoretically slow the order book dumping every time new rewards are claimed.

Please correct any part of the above if I misunderstand.

1 Like

We’ve analyzed the proposal and we believe it is a good decision to reduce the growth rate of added supply, first in order to be aligned with the original economics projection and secondly, we believe it is important to have a good balance between token value and inflation - introducing a high volume of new supply might have a negative impact on the token value. Therefore, Chainode Tech is in favor of reducing the inflation rate, thus providing more value to the existing tokens.

2 Likes

We support keeping the supply in line with planned levels, but require some more information to be able to make an accurate assessment whether the suggested parameters are suitable or not.

-Please detail your justification for inflation rate change of 100%.
-How much are we projected to exceed the 1 year supply by and what is the new estimate with the updated parameters?
-What staking % assumptions have you made in your above estimate?
-At current staking %, what would the new inflation rate be if this proposal is successful?
-Given these parameters, when would you expect the next proposal to be required in order to continue the down trend in accordance with the white paper?

Thanks!

2 Likes

As the proposal is now on chain, we have done some more research to understand the implications of the proposal and are sharing it here so that others can make an informed decision on their votes. Please correct anything we have wrong.

Inflation rate change
This variable has been discussed at length for the Cosmos Hub as part of Proposal 48 which passed a couple of weeks ago. The needs of the Cosmos Hub are different from Akash as they are sitting on the minimum inflation and are trying to protect against mass unstaking events by having the inflation rise, whereas Akash will be at the maximum inflation and would like it to gradually reduce (to help follow the intended curve in the whitepaper).

For those interested in learning more about the effect of the ‘inflation rate change’ variable on the inflation curve, you can check out the Cosmos Hub discussion here: Paramater Change Proposal: Inflation Rate Change - #8 by zaki_iqlusion - Governance - Cosmos Forum

In simple terms however, setting it to 100% just makes the inflation drop faster than it has been previously.

Inflation Min/Max
Inflation on Akash is currently ~47.73%. As this is 47.73% of all AKT, but only staked AKT receive the the reward, this means that APY is currently 47.73%/84.34% (staked ratio) = 56.59%.

If this proposal is successful, the max inflation will become 40% so there will be a step change of inflation from the current value of ~47.73% straight down to 40%. Assuming staked ratio doesn’t change, this means that APY will become 40%/84.34% = 47.43%. With compounding still resulting in a higher return.

Note: Akash network has a target staked ratio of 67% (e.g. we want at least 67% of all AKT to be staked) which means that if staked percentage is above that (as it is now at 84%) the inflation will try to be at the minimum inflation level (there is no need to give extra incentives for people to stake as we are already above the target). But as we will be sitting at the max inflation, the network will try to get down to the minimum as quickly as possible (depending on inflation rate change we discussed above).

@chenghiskhan How long will it take inflation to reach the Inflation Min of 25% assuming your choice of 100% inflation rate change and also assuming that the staked rate stays above 67%?

Although we’ve answered some of our own questions from the above post, it would be good to get a comment on the others.

Sorry if this detail is already posted somewhere but we weren’t able to find it.

Bliss Dynamics supports the proposal but reserves our vote until we hear more info on the tuning that has gone into the variables.

1 Like
  1. The proposed rate of change of 100% from 24% is meant to accelerate the rate at which the current inflation rate gets down to our intended target rate.
  2. 7/25/2021 target supply is 141,666,468 vs current supply ~146,000,000, so we’re definitely trending above.
    3/4. Staking assumptions. We’re leaving target bonded rate at 67%, so if actual bonded rate is higher, rates will trend towards the defined lower bound.
  3. We’re also pushing to build in the decay curve so that future proposals like this won’t be necessary b/c everything will be automatically adjust based on a function. This is probably an item we’ll complete this quarter and deploy in roughly in September
  4. Absolutely! We are trying to keep true to our white paper.
1 Like

Thanks @chenghiskhan, it sounds like you have put some thought into the new variables at least and we agree that the adjustment is necessary so Bliss Dynamics has voted yes.

Also, good to hear that you’re planning the new function to better manage it!

Thank you so much @KamuelBob for your detailed questions and input!
Your feedback and detailed consideration was amazing to behold and be a part of.

The next “inflation update” after this proposal will likely be the implementation of the decay curve itself, which we’ll definitely want to test in a more robust fashion. When that time comes, we look forward to having your voice and input once again!

1 Like

Hi @JoeMama

Strong points all around and I have nothing much to add other than helping to clarify the point around inflation rate of change:

The parameter “inflation rate of change” controls how quickly inflation can change within a 12 month period. Where it’s currently set (24%), we won’t be able to get the current inflation of ~47% down to the target of ~35-32% within the quarter since 24%/12 = a max change of only 2% per month. With a 100% range of change, 100%/12 = 8.33% per month, we can achieve our target inflation rate within ~2 months, which puts us squarely in target by the end of Q3.

Furthermore, since we’re currently over our projected AKT supply, a slightly steeper reduction in inflation can be accommodated to get back on our modeled supply trajectory. As a back-stop, an inflation min of 25% insures that no matter what our rate of change is, inflation will not fall below 25%.

By the end of Q3, our intention is to have a dynamic decay curve submitted for proposal to remove the need for constant tinkering.

Thank you for participating in this discussion and for your questions!

1 Like