Introducing a SIMD to reduce Solana inflation

[tjain-mcc]

SIMD ###: Introducing a Programmatic, Market-Based Emission Mechanism Based on Staking Participation Rate

Authors: Tushar Jain, Vishal Kankani

#Summary

This is the first of two SIMDs intended to make Solana emissions more market oriented. This SIMD proposes a market-based mechanism to dynamically determine Solana emissions.

#Motivation

As Solana matures, stakers increasingly earn SOL through mechanisms like MEV. This income stream reduces the network's historical exclusive reliance on token emissions to attract stake and security. According to Blockworks (https://solana.blockworksresearch.com/), in Q4 2024 MEV, as measured by Jito Tips, was approximately $430M (2.1M SOL),representing massive quarter-over-quarter growth. In Q3 Jito Tips were approximately $86M (562k SOL), Q2 was approximately $117M (747k SOL), and Q1 was approximately $42M (300k SOL).

Given the level of economic activity the network has achieved and the subsequent revenue earned by stakers from MEV, now is a good time to revisit the network’s emission mechanism and evolve it from a fixed-schedule mechanism to a programmatic, market-driven mechanism.

The purpose of token emissions in Proof of Stake (PoS) networks is to attract stakers and validators to secure the network. Therefore, the most efficient amount of token issuance is the lowest rate possible necessary to secure the network.

Solana’s current emission mechanism is a fixed, time-based formula that was activated on epoch 150, a year after genesis on February 10, 2021. The mechanism is not aware of network activity, nor does it incorporate that to determine the emission rate. Simply put, it’s “dumb emissions.” Given Solana’s thriving economic activity, it makes sense to evolve the network’s monetary policy with “smart emissions.”

There are two major implications of Smart Emissions:

Smart Emissions dynamically incentivizes participation when stake drops to secure the network.
Smart Emissions minimize SOL issuance to the Minimum Necessary Amount (MNA) to secure the network.

This is good for the Solana network and network stakers for four reasons:

High inflation can lead to more centralized ownership. To illustrate the point, imagine a network with an exceedingly high inflation rate of 10,000%. People who do not stake are diluted and lose ~99% of their network ownership every year to stakers. The higher the inflation rate, the more network ownership is concentrated in stakers’ hands after compounding for years.

Reducing inflation spurs SOL usage in DeFi, which is ultimately good for the applications and stimulates new protocol development. Additionally, a high staking rate can be viewed as unhealthy for new DeFi protocols, since it means the implied hurdle rate is the inflation cost. Lowering the “risk free” inflation rate creates stimulative conditions and allows new protocols to grow.

If Smart Emissions function as designed, they will systematically reduce selling pressure as long as staking participation remains adequate. The inevitable side effect and primary downside to high token inflation is increased selling pressure. This is because some stakers in different jurisdictions have taken the interpretation that staking creates ordinary income, and therefore they must sell a portion of their staking rewards to pay taxes. This selling is a significant detriment to the network and does not benefit the network in any way.

In markets, sometimes perception is as important as reality. While SOL inflation is technically not cost to the network, others think it is, and that belief overall has a negative impact on the network. Inflation causes long-term, continual downward price pressure that negatively distorts the market’s price signal and hinders fair price comparison. To use an analogy from traditional financial markets, PoS inflation is equivalent to a publicly listed company doing a small share split every two days.

Historically, issuance curves have remained static due to Bitcoin’s immutability ethos—a “Bitcoin Hangover” so to speak. While immutability suits Bitcoin’s mission to become digital gold, it doesn’t map to Solana’s mission to synchronize the world’s state at light speed.

In summary, the current Solana emissions schedule is suboptimal given the current level of activity and fees on the network because it emits more SOL than is necessary to secure the network. An issuance curve set by diktat is not the right long-term approach for Solana. Markets are the best mechanism in the world to determine prices, and therefore, they should be used to determine Solana’s emissions.

#Detailed Design

###Five variables drive Solana’s staking market:
Yield for stakers (y)
Issuance Rate (i) - SOL emitted
SOL staked (N)
MEV in SOL terms (MEV)
Validator commissions (c)

These variables are mathematically related:

y = ((i+MEV)/N)*(1-c)

Currently, the network has a fixed issuance rate (i) while the number of SOL staked (N) fluctuates based on market conditions. MEV also fluctuates based on market conditions.

When considering new models for issuance, this relationship is critical.

##Proposed Design:
Programmatic, Market-Based Emission Mechanism Based on Staking Participation Rate

A dynamic, market-based rate can be determined using the following factors:

The Staking Participation Rate (s = SOL staked / Total SOL in existence) should be based on what is needed for consensus safety.
The network should reduce issuance if the staking participation rate is higher than the target rate and increase issuance if it is lower.
There should be a ceiling on the inflation rate as a protection mechanism.

We imagine the Target Staking Participation Rate (T) as a governable variable and recommend a target staking participation rate of 50% for the following reasons:
Beyond 67% incremental staked SOL does not add any incremental security guarantees because a supermajority of all SOL has voted on any given block and a long range attack is impossible. This “excess stake” explicitly inhibits network economic activity and hampers growth.
Below 33%, we potentially risk network safety because a supermajority of all SOL has explicitly not voted on any given block and this opens the edge case possibility of long range attacks.

It also proposes the following bounds for the issuance rate:
Upper Bound: The current Solana issuance curve (decreasing at a rate of 15% per year and will stop decreasing once nominal inflation is 1.5%).
Lower Bound: 0%

Increases or decreases in inflation should be proportional to the magnitude of the difference between the actual staking participation rate and the target rate (for example, 50% as per this proposal).

This approach would allow for a more dynamic response to fluctuations in staking participation. By aligning inflation adjustments with the actual deviation, network issuance better reflects the network’s real-time economic and security conditions.

Inflation adjustment function:

Δi= k * Δs

Δi = Inflation change for the new epoch
k = Speed Co-efficient
Δs = Staking Participation Rate (s) at the start of epoch – Target Staking Participation Rate (T)

inew = max (0%, min (current issuance curve, ilast + Δi)

ilast = Inflation in the last epoch
inew = Inflation in the new epoch
current issuance curve = inflation defined by current Solana issuance curve

This proposal sets k = 0.05 per annum. So, for each extra percentage point higher/lower in staking participation rate, inflation would come down/go up by 0.05% p.a. in the next epoch. With the current staking participation rate of ~70%, the network would see a reduction of inflation of 1% p.a. in the next epoch. On the other hand, with a hypothetical staking participation rate of say 40%, the network would see an increase of inflation of 0.5% p.a. in the next epoch.

The max function ensures that inflation is at least zero, and the min function ensures that the inflation does not rise above the current issuance curve.
This design offers several key benefits:
Consensus Safety: Adjusting inflation based on staking participation ensures sufficient validator incentives to maintain network security, prioritizing consensus safety.
Market-Based Flexibility: The model adapts to the network's economic activity, making it more responsive to changing market conditions. It’s possible to imagine a future where stakers are earning enough from MEV that no SOL emissions are necessary.
Validator Retention: It accommodates Solana-aligned validators who are willing to stake even with lower emissions, recognizing that they can earn more through MEV in higher economic activity ecosystems.
This dynamic approach balances the need for a secure, decentralized network with the flexibility to thrive in a competitive market.
##Alternatives Considered

###Alternative Design 1: Pick another fixed curve
A simple alternative would be to adjust the issuance rate to a fixed number, determined by community inputs. However, this approach presents several risks:
Lack of Market Mechanisms: Setting a fixed rate ignores the dynamics of free markets and the network’s real-time economic conditions.
Arbitrary Adjustments: Using another arbitrary number risks undermining the integrity of the system and may lead to decisions that are disconnected from the network’s needs.
Erosion of Trust: Relying on fixed adjustments could erode trust in the community’s decision-making process, especially if future changes seem disconnected from market realities.
Compromised Consensus Safety: A fixed issuance rate, especially in uncharted territory, could undermine consensus safety, as it would not be dynamically tied to staking participation or broader network health.
###Alternative Design 2: Fix Target Staking Yield
MEV has become a significant revenue source for stakers. One can consider changing the issuance rate by factoring in MEV tips, maintaining the same target yield as the original curve but offsetting it by the 30-day moving average of MEV tips.
New Issuance Rate (i) = Target Staking Yield − 30-day moving average of MEV tips
MEV tips reflect real revenue for validators and stakers, allowing the system to adjust to market conditions:
Hot Markets: Higher MEV tips allow for lower emissions.
Cold Markets: Increased emissions compensate validators, maintaining network security.
This approach is inspired by central bank monetary policy, adjusting inflation based on economic conditions.
But the big challenge with this design is that it incentivizes MEV payments to move out of sight of the tracking mechanism, thereby rendering the design completely ineffective.
For an abundance of clarity, we are not proposing any design which requires measuring MEV payments.

##Impact

Implemented thoughtfully, this design could have a major positive economic impact on the overall health of the Solana economy.

##Security Considerations

Targeting a staking participation rate of 50% ensures sufficient stake for consensus safety while maintaining the network’s security and decentralization.

Below 33%, we potentially risk network safety because a supermajority of all SOL has explicitly not voted on any given block and this opens the edge case possibility of long range attacks. It is important to note that these long range attacks are entirely theoretical and we have not seen one in practice. There are other mechanisms in Solana to protect against long range attacks.

This proposal is the first in a series of steps to make Solana’s consensus more secure and economics more market driven. The successor to this proposal is another SIMD that introduces the concept of long-term staking, which seeks to improve network security. The option to unstake SOL on a relatively short notice (i.e., a short cool down period) poses a potential risk to networks’ stability and safety, particularly in extreme circumstances where a significant amount of SOL is unstaked within a brief timeframe. The combination of these two SIMDs address these concerns while improving network security and economic activity.

[cfl0ws]

Thanks for this and I look forward to providing more input soon. In the meantime, would you like to introduce and discuss the proposal during our next Solana Community Validator call? It's happening a week from today on 23 Jan @ 1900 UTC. I think it would generate a lot of constructive discussion. If you are interested and available, let me know and I'll add it to the agenda.

[cfl0ws]

Thanks for joining the call. As we were discussing this during the call, the Livepeer network came to mind, as it has an adjustable inflation rate based on staking ratio. I couldn't remember what the target/trigger ratio was. When I looked it up, the ratio turns out to also be 50%.

I realize there are vast differences between Livepeer and Solana, first and foremost being that Solana validator revenue streams are becoming increasingly diverse and LIvepeer has struggled to generate user fees. However, I still thought it could be an interesting reference example.

1 - A brief, explanation of the inflation methodology, particularly what happens when the ratio hits 50% here

2 - 50% ratio met here

3 - A proposal to increase inflation again, after the 50% ratio was met here

4 - And a more recent conversation about changing the ratio here

[cfl0ws]

I'm starting a new thread for more substantive comments, having read through this idea in its entirety. Thanks again for starting the discussion.

Generally I feel supportive of this idea. Basing inflation on market dynamics, rather than a single, somewhat arbitrary and dated, variable, sounds beneficial.

I also now have a better understanding of how the 50% threshold was chosen and the logic resonates with me. Based on my read of the proposal, 67% provides "maximum useful" security of the network. If that interpretation is accurate, I'd suggest we start with a target ratio of 67%, rather than the proposed 50%.

I also still find myself a bit hooked on the idea of setting the target as a staking ratio, rather than dollar value, since ultimately, economic security is valued in dollars, yet recognize the difficulty in setting a dollar value in this situation.

The proposed curve feels like a reasonable starting point to me. A scenario we'd want to avoid is one in which whales may choose to unstake part of their holdings to lower the ratio, because the total rewards at a higher inflation rate on a lower amount staked is greater than total rewards at a lower inflation rate on a higher amount stakes. I don't think following the current curve would result in that happening, however might be something to confirm.

A few additional thoughts -

• When speaking about additional validator inflows, it might make sense to include or at least mention priority fees alongside MEV.

• I agree that high inflation (or any inflation really) contributes to centralization via a "rich get richer" scenario, as I wrote about going on four years ago here. That said, reducing inflation may be a step toward addressing the issue. However MEV and other inflows contribute to the same problem, as they're all generally proportional to total stake weight.

• Given that most validators, at least the named validators, retain only a very small percentage of inflationary rewards, this idea seems very relevant to delegators. However, delegators have no direct voice in the evolving governance process as it is currently defined. If this idea comes to a vote, I believe the validator community should make a conscious effort to include delegators in the process to the extent possible. One idea that we tried on a previous vote, for example, was signaling our vote intention prior to voting. This approach allows delegators to provide input to the decision, as well as a chance to delegate to/from a validator prior to the vote happening.

[ghostlabs-x]

I just want to start by noting from an economic point of view a dynamic market based emissions schedule is a sensible mechanism to drive toward, so no debate on the premise.

The main thesis below - are there variables missing from this model based on how the current implementation operates? And what goals are driving these changes that should be measured to determine success and/or recalibration?

1. The 'Stake Weight' component impacts individual validator relevancy - and ultimately total number of validators able to operate successfully. [Validator Participation Rate]
2. Prior to Q4 last year, SOL issuance averaged 33% of Operator Payouts - i.e. subsidies. 'What does good look like' when these subsidies are removed?
3. 'SOL issuance' reflects inflation, 'SOL burn' reflects dynamic deflation. Both exist however only one is factored into this model stoking a potential disconnect in the overall economics. What are the impacts of this? (of course, all of this is goal dependent, and to me the primary variable missing in the proposed model is a decentralization coefficient. If that goal doesn't matter, then most likely this network will continue to operate just fine with lower overall validator participants)

For the sake of discussion and to avoid repeating what has already been laid out above/elsewhere, I’ll make a few succinct points that I would be interested to see community views or debate on as I am still trying to formulate my own informed view on this one:

1. If a transition to a dynamic model occurs, is the proposed method of anchoring ‘Staking Participation’ the optimal approach? Part of the rationale provided for this change hinges on substantial increases in REV as the source of income to Validators to operate a globally distributed transaction network. Over the course of 2024 average REV contributed only 22% of the network TEV. Q4 was an impactful operating period contributing to a 5.6% increase alone (an outlier in the dataset) and seemingly a heavily weighted assumption of the rationale being REV will continue to sustain or maintain a relative growth rate. Should REV factor in to this model as a more intrinsic input toward bottom-line network operator revenue growth? If “MEV” = “REV” perhaps its a terminology alignment, but the point remains - should the ‘Issuance Rate’ be tied to a function of actual network revenue growth to offset current subsidization rates?
   [Referenced Blockworks data. For consideration, ‘Operator Payouts’ do not reflect the variable expenses of running this global network, i.e. is it profitable/sustainable? What is the economic timing to reduce subsidization of the network? Separate discussion points below]

2. What is the definition of “Securing the Network”? Number of Operators? Enough diversified stake to represent the common interest? Given blockchain networks have other dimensions beyond ‘continued operability’, should these metrics be evaluated within the economic equation. In theory, if only 3 independent equally staked validators remained the network would still be secure and operational, until…?

3. Is there any analysis to assess the elasticity of the current stake distribution? {Price (growth) vs Rate of Return (income)} + [some coefficient of decentralization] - assuming that’s important in this new model

4. A big theme of discussion around this proposal as some have defined, “Death of the Small Validator”. Markets are dynamic, volatile and chase returns. Generally capital seeks to optimize return, and in liquid markets will move quickly. Running a successful validator has been posed in many cases as a ‘Full-Time Job’. Part of that is technical, another part is attracting “inelastic” stake. The magnitude of short-term or drawn out shocks to markets can be impactful. This is more of a fundamental consideration pointing back to some of the factors to the equation as mentioned above. How important is this one?
   A random example that never leaves me, 2009 financial collapse led to much greater centralization of the ‘global financial transaction network’ - blockchains and market dynamics solve this (in this case if it is profitable to be an operator in the network).

5. Could a mathematical "Short Squeeze" occur where stake reduces sufficiently enough that most small validators can no longer operate profitably (even with Foundation delegated stake), then the formula triggers higher inflation again compounding into a smaller set of the largest validators? Or a game theory attack - keep your yield percentage higher than everyone else (even at a temporary loss), attract the stake, grow bigger amongst a shrinking operator network - when satisfied, reprice.

From my position I raise these points as additional considerations for what the appropriate inputs of this new economic ‘staking/inflation’ model might be. This is still unchartered territory, overall structure and timing are important to ensure a stable foundation for sustainable long term network growth. Does this formula consider all of the variables in alignment with the overall goals of this new ‘Global Network State’?

(Obviously I admit I don't have the answers to these questions but I see this as a much bigger formula than just an Inflation Rate when there are many more dimensional factors at play requiring consideration)