Solana's first binding on-chain governance vote closes at approximately 15:30 UTC on Thursday, August 27, 2026. Validators and delegators are deciding three questions simultaneously: whether to ratify a written network constitution, whether to double the rate at which SOL inflation decays, and whether to rewrite transaction fees so that a far larger share of SOL is destroyed rather than paid out. If all three pass, issuance falls by roughly 19 million tokens over six years, daily burns rise about fourteenfold, and nominal staking yield compresses from 5.84% toward 2.25% inside three years.
For a network governed until now by rough consensus among core development teams, this is a structural change rather than a routine parameter tweak. The ballot is stake-weighted, the passage thresholds are explicit and published in advance, and the result binds the roadmap. It is also moving quickly enough that most SOL holders will read about the outcome rather than participate in it.
This article sets out exactly what is on the ballot, what each outcome does to token supply and to staking income, who is arguing against it and what they stand to lose, and what a delegator can still do before the epoch closes. Every figure below is attributed to a named source with a date, and the closing section documents which numbers are contested and how they were checked.
What Is Actually on the Ballot
Three proposals are being voted on independently. This matters more than most coverage conveys: a split outcome is entirely possible, and any one proposal can pass while another fails.
The Three Proposals at a Glance
SGP-0001 - The Solana Constitution. Ratifies a written constitution and formally activates the on-chain governance program that subsequent proposals will run through.
SGP-0002 - Double Disinflation. Raises the annual disinflation rate from 15% to 30%, accelerating how quickly SOL inflation decays toward its 1.5% terminal rate.
SGP-0003 - Resource and Inclusion Fee. Splits the base transaction fee into a fixed inclusion fee paid to block leaders and a resource fee, tied to compute actually consumed, that is burned in full.
SGP-0001: The Constitution
The first proposal is procedural, and it is the one that gives the other two their force. It ratifies a written constitution for the network and switches on the on-chain governance program through which future proposals will be submitted, debated and decided.
Until now, protocol changes on Solana moved through the SIMD process, a design-document workflow closer to Ethereum's EIPs than to a formal vote. Rough consensus among core teams and validator operators determined what shipped. SGP-0001 replaces that convention with a codified procedure. Whatever one thinks of the specific parameters in the other two proposals, this is the change that determines how every subsequent parameter gets set.
SGP-0002: Double Disinflation
Solana's inflation schedule does not run at a fixed rate. It starts high and decays by a set percentage each year until it reaches a 1.5% terminal floor, where it remains permanently. SGP-0002 does not cut inflation on any given day. It doubles the annual decay rate from 15% to 30%, which compresses the entire remaining schedule.
Under the current schedule, SOL reaches the 1.5% floor around 2032. Under the proposed one, it arrives around 2029 - roughly 2.8 years instead of 5.7, according to the modelling attached to the original SIMD-0550 design (Solana Compass, August 2026).
SGP-0003: Resource and Inclusion Fee
The third proposal restructures how transactions are priced. Today a base fee is charged per signature, largely irrespective of how much computational work a transaction demands. SGP-0003 splits that into two components: a fixed inclusion fee that goes to the block leader for the service of including the transaction, and a resource fee that scales with the compute units actually consumed and is burned in full.
The design intent is to make wasteful transactions expensive without taxing efficient ones. It is directionally similar to Ethereum's EIP-1559, which also burns a usage-linked fee component, but the mechanism differs: EIP-1559 burns a congestion-priced base fee, while SGP-0003 prices and burns against measured resource consumption.
Why the Proposal Names Keep Changing
The naming has caused genuine confusion, and it matters for anyone searching for primary sources. These proposals were designed and debated as SIMD-0550 and SIMD-0553. On the ballot they appear as SGP-0002 and SGP-0003 respectively. Earlier reporting also referenced them bundled as SGP-0003 in a package described as covering both measures.
Both naming sets point to the same underlying designs. If a source cites SIMD-0550, it is discussing the disinflation change now on the ballot as SGP-0002.
How the Vote Actually Works
Two Thresholds, Both Required
Passage requires two separate conditions to clear, and either one can sink a proposal on its own.
First, participation: at least one-third of active stake must cast a vote. Second, approval: two-thirds of decisive stake - votes cast For or Against - must be in favour. Abstentions are counted toward the participation threshold but excluded from the supermajority calculation (Crypto Times, August 24, 2026).
The practical consequence is that a proposal can attract overwhelming approval and still fail. A 95% For ratio on a quarter of network stake does not pass. A 70% For ratio on 40% of stake does.
The Delegator Override Most Coverage Skips
The mechanism that matters most to ordinary holders receives the least attention. Delegators are not bound by their validator's position. A staker can submit an independent on-chain ballot that overrides the validator's default vote, and can do so without undelegating their SOL.
This is the only action a retail SOL holder can take that affects the outcome, and the window closes when the epoch does. If your validator has taken a position you disagree with - or has not voted at all - an override is available until roughly 15:30 UTC on Thursday.
Where the Tally Stood
Early figures point to strong support, with the important caveat that they are partial and drawn from a small number of very large voters.
As of August 23, 2026, SGP-0002 stood at approximately 5.27 million SOL in favour against 547,019 SOL opposed, with zero abstentions recorded across 24 votes. That put the For side at roughly 90.6% of decisive stake at that moment (CryptoSlate, August 23, 2026).
Twenty-four votes is not a representative sample of a network with hundreds of validators. Treat this as an interim signal, not a forecast.
What Double Disinflation Does to Staking Yield
The Issuance Math
Compressing the disinflation schedule reduces cumulative issuance. Over a six-year horizon, the change would result in nearly 19 million fewer SOL being created than under the current schedule (crypto.news, August 2026).
Staking rewards on Solana are paid primarily from that issuance. Reducing issuance therefore reduces nominal staking rewards. This is not a side effect of the proposal; it is the mechanism by which it works.
The Yield Schedule
The modelling attached to SIMD-0550 sets out the trajectory for nominal staking yield:
| Period | Nominal staking yield | Change vs today |
|---|---|---|
| Current | 5.84% | - |
| Year 1 | 4.34% | -1.50 pp |
| Year 2 | 3.00% | -2.84 pp |
| Year 3 | 2.25% | -3.59 pp |
Source: SIMD-0550 modelling, as reported by CryptoSlate (August 23, 2026). The "change vs today" column is arithmetic applied to the published figures.
What the Model Does Not Include
These numbers describe an issuance schedule, not a promised return, and the assumptions materially affect them.
The projections assume 68% staking participation across the network. They exclude validator commissions, which typically take a percentage off the top. They exclude MEV revenue, priority fees, and any other yield source a validator or liquid staking protocol passes through. And they assume participation holds steady - if stakers exit as nominal yield compresses, the remaining stakers receive a larger share of a smaller pool, which partially offsets the decline.
Real income will differ, in both directions, depending on the validator and on how the network responds.
The 14x Burn, Explained
How Resource Fees Work
SGP-0003 approaches the supply question from the demand side rather than the issuance side. Instead of every transaction paying a similar base fee, transactions pay for the compute they actually consume, and that resource component is destroyed rather than paid to validators.
Critically, the burn scales with network activity. Where SGP-0002 follows a fixed schedule regardless of usage, SGP-0003 burns more when the network is busier. The two proposals are complementary but independent, which is precisely why they are being voted on separately.
The Burn Numbers
Anza, the core protocol development firm, estimated the effect at recent activity levels:
| Measure | Current | Under SGP-0003 |
|---|---|---|
| Daily SOL burned | ~648 SOL | 7,500-9,000 SOL |
| Daily USD value | ~$47,000 | up to ~$657,000 |
| Multiple | - | ~14x |
Source: Anza estimates, as reported by CoinDesk (August 4, 2026). USD figures depend on the SOL price at the time of the estimate and will move with it.
For scale, roughly $657,000 per day annualises to around $240 million in destroyed supply - meaningful against a market capitalisation in the mid-$50 billions, but not transformative on its own. The burn matters more as a structural change to how fees flow than as a near-term supply shock.
Who Opposes It, and What They Stand to Lose
Solana Company's Position
Public opposition to SGP-0002 has come from a notable direction. Solana Company, the Nasdaq-listed treasury firm, has stated its opposition to the disinflation proposal. Its reasoning centres on institutional predictability:
Predictable inflation and staking yield help institutions model returns and adopt SOL.
The argument is coherent on its own terms. Institutional allocators do build models around expected yield, and changing the schedule mid-stream invalidates those models. Treasury companies that have raised capital against projected staking income have a legitimate planning interest in stability.
The Revenue Concentration
That position is worth reading alongside the company's own filed numbers. In its Q2 2026 filing, Solana Company reported $2.512 million of $2.526 million in total revenue - 99.4% - derived from staking. Its validator cluster launched in July 2026, with roughly 500,000 SOL delegated to it by the filing date (CryptoSlate, August 23, 2026).
A proposal that compresses staking yield compresses very nearly the entirety of that firm's revenue line. Both the stated argument and the financial exposure are matters of public record, and readers can weigh them together. Nothing here implies the argument is made in bad faith - parties with the most at stake are often the ones who examine a proposal most carefully. It does mean the position should be read as an interested one.
The Backdrop: Record ETF Inflows
The Numbers
This vote arrives during the strongest recorded period of institutional demand for SOL.
| Metric | Value | Context |
|---|---|---|
| Cumulative net inflows | $1.22 billion | Record for US spot SOL ETFs |
| Single-day inflow (Mon) | $33.5 million | Largest single day of 2026 |
| Inflow streak | 5 sessions | Consecutive net-positive days |
| Trading volume | $166.8 million | Highest since October 2025 |
| Bitwise BSOL holdings | $948.2 million | ~80% of all US spot SOL ETF capital |
| BSOL volume, Aug 24 | $108 million | Record day for any SOL ETF |
Source: CoinDesk (August 25, 2026).
A Correction Worth Noting
Several outlets have reported that Bitwise's BSOL surpassed $1 billion in inflows. On the most recent available figures it stands at $948.2 million - approaching that mark, not past it. We have used the reported figure rather than the rounded claim.
Why BSOL Dominates
One fund holding roughly four-fifths of a category is unusual, and the likeliest explanation is structural: BSOL passes a portion of the staking rewards earned on its SOL holdings through to investors. For a traditional allocator, that turns SOL exposure from a pure price bet into a yield-bearing position inside a regulated wrapper.
The Exposure Nobody Is Pricing
That same mechanic creates a direct link between this week's vote and the ETF flows. If SGP-0002 passes and nominal staking yield compresses from 5.84% toward 2.25% over three years, the yield component of BSOL's proposition compresses with it.
The fund does not disappear, and price exposure is unaffected. But the specific feature that plausibly explains its 80% share becomes materially less attractive over the medium term. This connection has gone largely unremarked in coverage of both stories, and it is worth watching as institutional adoption of Solana matures.
The Backdrop: The August 12 Near-Halt
What Happened
Two weeks before the vote opened, Solana came closer to halting transaction finality than at any point since February 2024.
At 07:46 UTC on August 12, 2026, a routing failure at hosting provider Teraswitch knocked roughly 90 validators offline for approximately 33 minutes, across 12 sites in Europe and Asia-Pacific. At peak, 28.83% of staked SOL went delinquent against the 33.34% threshold at which Solana stops finalizing transactions - a margin of 4.5 percentage points, or about 86% of the way to a finality halt (Solana Compass, August 12, 2026).
Separate reporting counted 102 of 699 validators as having stopped voting during the incident (crypto.news, August 2026). The two counts measure slightly different things - validators unreachable versus validators not casting votes - and both are cited here rather than reconciled.
| Measure | Value |
|---|---|
| Incident start | August 12, 2026, 07:46 UTC |
| Duration | ~33 minutes |
| Validators offline | ~90 (102 of 699 stopped voting) |
| Sites affected | 12, across Europe and Asia-Pacific |
| Peak stake delinquent | 28.83% |
| Finality halt threshold | 33.34% |
| Margin | 4.5 percentage points |
| Staking rewards missed | ~333 SOL |
| Blocks halted | None |
The Root Cause, in Plain Terms
The failure was a routing misconfiguration, not an attack or a consensus bug.
Teraswitch advertised a default route from its Miami site without the accompanying route attributes. A route reflector in Amsterdam propagated that route onward to sites across Europe and Asia. Edge routers at those sites treated it as locally originated and preferred it over the valid default route. Teraswitch's own core network then rejected the route as invalid. The result was 12 sites with a preferred route that led nowhere and no usable fallback.
Triton One, whose infrastructure was affected, said it would publish further detail after receiving Teraswitch's postmortem and completing its own review. Block production continued throughout, finality was never interrupted, and affected validators collectively missed around 333 SOL in staking rewards, which validator bond programs are expected to cover.
What It Says About Validator Concentration
The instructive point is not that Solana broke, because it did not. The consensus mechanism performed as designed, absorbed the loss of nearly a third of voting stake, and kept producing blocks.
The point is that more than a quarter of all staked SOL sat behind a single connectivity provider, and that this was discovered through a misconfigured BGP announcement rather than through a deliberate attack. Solana's Tower BFT consensus requires more than 66.67% of staked SOL to participate actively for transactions to finalize. Consensus upgrades such as the Alpenglow finality upgrade change how quickly the network reaches agreement; they do not change where validators are physically hosted or whose fibre they share.
Neither proposal on Thursday's ballot addresses hosting concentration. That is not a criticism of the proposals - it is simply a gap worth naming, because the governance narrative and the infrastructure narrative are running in parallel this month without intersecting.
What Happens After 15:30 UTC Thursday
Nothing Flips Immediately
These are stake-weighted signaling votes. Passage authorises implementation; it does not execute it. The fee restructuring in particular would ship in a future client release rather than at epoch close, and the disinflation change alters a schedule rather than triggering a one-time event.
Three Branches
| Outcome | What follows |
|---|---|
| All three pass | Constitution activates, on-chain governance becomes the standing process, supply schedule compresses, fee rewrite enters the release pipeline. |
| Split result | Each proposal stands alone. SGP-0002 can fail while SGP-0003 passes, or the reverse. A constitution can be ratified without either economic change. |
| Quorum missed | Nothing passes, regardless of how lopsided the For and Against tallies appear. Participation below one-third of active stake voids the result. |
What to Watch
Participation, not sentiment, is the number that decides this. Given an early tally running above 90% approval on the disinflation proposal, the approval threshold looks comfortable. The one-third participation floor is the live question, and whether large treasury holders and major staking pools turn out will determine both figures.
For price context: SOL traded at $97.00 with a market capitalisation of $56.64 billion and $2.90 billion in 24-hour volume as of 21:01 UTC on August 26, 2026, up 0.34% on the day (CoinGecko). That is a snapshot taken before the vote closed and should be read as one.
What This Means If You Hold SOL
If You Stake
You bear the direct cost. Nominal yield compresses on the schedule above, and if you stake through a liquid staking protocol or a yield-passing ETF, that compression reaches you as well. The offsetting benefit is reduced dilution of the tokens you already hold. Whether that trade is favourable depends on your time horizon and on whether you are accumulating or living off the yield.
The actionable step: check your validator's declared position and submit an override ballot if you disagree. That option expires with the epoch.
If You Hold Without Staking
You receive the dilution benefit without paying the yield cost, which makes both proposals straightforwardly favourable from your position. It is worth recognising that this is precisely the transfer the proposals create - from stakers to non-stakers - and that stakers are the ones securing the network.
If You Trade
The supply effects are real but slow. Nineteen million fewer tokens over six years and roughly $240 million per year in additional burns are structural changes, not catalysts. The nearer-term variable is whether a quorum failure or a surprise result changes sentiment on Thursday afternoon, and whether ETF inflows continue at their current pace once the yield outlook shifts.
The Bottom Line
- Three independent proposals close at approximately 15:30 UTC on Thursday, August 27: a network constitution, a doubled disinflation rate, and a fee rewrite that burns resource fees in full.
- Stakers absorb the cost. Nominal yield falls from 5.84% toward 2.25% within three years under the SGP-0002 model, assuming 68% network participation and excluding commissions, MEV and priority fees.
- Supply tightens on two fronts. Roughly 19 million fewer SOL issued over six years from disinflation, plus daily burns rising from about $47,000 to as much as $657,000 from the fee change.
- Quorum is the real question. One-third participation and a two-thirds supermajority are both required, and abstentions count toward the first but not the second.
- The loudest opposition is an interested party. Solana Company opposes the disinflation proposal and derives 99.4% of its revenue from staking. Both facts are on the public record.
- Delegators can override their validator without undelegating, until the epoch closes.
Sources and Methodology
How These Figures Were Checked
Every statistic in this article is attributed inline to a named publication with a publication date. Where two credible sources disagreed, both figures are presented rather than one being silently selected. Price data was retrieved directly from CoinGecko's public API at 21:01 UTC on August 26, 2026, and is timestamped in the text rather than described as "current".
Known Discrepancies in Circulating Coverage
Several errors are circulating widely enough to be worth documenting:
| Circulating claim | What the sources actually show |
|---|---|
| Bitwise BSOL surpassed $1 billion in inflows | $948.2 million as of August 25, 2026 - approaching, not past |
| The outage occurred on August 13 | August 12, 2026, at 07:46 UTC |
| The vote deadline was August 18 | August 18 was the deadline for the 15% stake threshold to trigger a formal vote. The formal vote closes August 27. |
| ~90 validators vs 102 of 699 validators offline | Both figures are reported by credible sources measuring different conditions. Both are cited above. |
| The proposals are SIMD-0550 and SIMD-0553 | Correct as design documents. On the ballot they are SGP-0002 and SGP-0003. |
Primary and Secondary Sources
- Solana Developers - voting window and epoch timing announcement (August 24, 2026)
- Crypto Times - quorum and supermajority thresholds (August 24, 2026)
- CryptoSlate - yield projections, early tally, Solana Company filing figures (August 23, 2026)
- CoinDesk - Anza burn estimates (August 4, 2026)
- CoinDesk - ETF inflow and volume data (August 25, 2026)
- Solana Compass - Teraswitch incident detail and stake figures (August 12, 2026)
- crypto.news - validator counts during the incident (August 2026)
- crypto.news - issuance reduction over six years (August 2026)
- Solana Compass - disinflation timeline modelling (August 2026)
- CoinGecko - SOL price, market capitalisation and volume (retrieved 21:01 UTC, August 26, 2026)
Disclosure and Limitations
This article is news analysis, not investment advice. It contains no price prediction and no recommendation to buy, sell or stake. The yield projections are model outputs published alongside the proposal, not guarantees, and their stated assumptions are reproduced above so readers can judge them.
The vote was still open at the time of publication. We will update this article with the final tally and outcome once epoch 1023 closes.