Written by: Boaz Sobrado, Forbes

Translated by: AididiaoJP, Foresight News

About 70 minutes before the voting ended, SGP-0002 was still struggling to get through. There was a shortage of about 58 million SOL on the books, and the deflation proposal was about to fail in Solana's first binding on-chain governance.

Then Kraken's major node changed its position.

On August 28, Solana validators finally passed SGP-0002, doubling the annual deflation rate from 15% to 30%. The nominal staking yield was placed on a clear downward track: expected to drop to 2.25% in about three years. At the time of writing, SOL was still around $102. The market did not interpret this as a day of plummeting prices, nor did it treat it as a token to speculate on due to reduced issuance. What was truly implemented was an issuance curve not yet encoded but already recorded in governance.

Passing by only 0.334 percentage points

This was the first binding proposal that truly passed under Solana's new governance framework, and it was also the first time validators agreed to cut issuance. The threshold was two-thirds. Ultimately, the approval rate was 67.001%, just 0.334 percentage points higher than 66.67%.

The vote count was clear. Approximately 176.29 million SOL in favor, about 66.19 million against, and about 20.63 million abstained. The snapshot included eligible staked SOL of approximately 433.49 million, with a participation rate of 60.7%, and 1,326 validators voted, setting a record for governance participation on that network. Calculating the support as a ratio of “in favor to in favor plus against,” the approval rate could reach about 72.7%; but the rules recognize the supermajority among those participating in the staking. By this standard, passing was as thin as a hair's width.

Mert Mumtaz, CEO of Helius, later described it plainly: about 500 calls were made over a few hours, "we finally gathered the votes in the last few seconds." In March 2025, a similar proposal SIMD-0228 only got about 61%, having failed once. The ability to pass this time did not rely on a sudden alignment of consensus but rather the last-hour vote changes.

Kraken initially suppressed the proposal, then helped it resurface

Kraken was not a side player in this vote tally. It had two nodes. The larger node had about 8.918 million SOL, which was 100% opposed that morning; at 10:37 UTC it changed to 90.34% in favor and 9.66% against, shifting about 8.1 million SOL from the opposition to the approval column. The smaller node had about 3.31 million SOL, consistently opposed at 100%.

From the same exchange, two nodes had two different attitudes. Co-CEO Arjun Sethi later simply stated, "Custodians should be pipelines, not loudspeakers."

This statement conveys restraint: clients' assets should not be used to vocalize the company's policy stance. However, on-chain records are more rigid. Estimates suggest that if the major node had consistently voted against, the approval rate would have dropped to about 63.9%, which would have been a direct failure. Galaxy also changed from near unanimous abstentions to partial support in the last hour. JitoSOL holders activated a mechanism to cover validator votes. Some analyses have noted: without this batch of cover votes, the proposal would not have passed.

The convenience of custodial staking was laid bare this time. Users placed SOL in the exchange, and the voting rights defaulted to following the validators. The cover mechanism exists, but it requires oversight and utilization. Those looking for convenience had almost no separate voice in this vote.

Deflation doubled, but not a halt to issuance tomorrow

SGP-0002 changed the speed of issuance decrease, not an immediate halt to the faucet. Solana's inflation would naturally decline each year, with an endpoint of a long-term floor of 1.5%. Originally, at a 15% annual deflation rate, it would take until 2032 to reach the line; after changing to 30%, the timeline is moved up to the first half of 2029, with the runway shrinking from around 5.7 years to about 2.8 years.

The proposal's authors are Helius's Lostin and 0xIchigo, corresponding to the technical proposal SIMD-0550. Model estimates suggest that over the next six years, about 18.9 million SOL will be issued less, roughly equivalent to 2.6% of the previously scheduled supply. Based on price estimations before and after voting, the scale of reduced issuance is about $1.5 billion to $2 billion. By June 2026, the network's annual inflation will be about 3.82%.

A point that is easy to confuse is: doubling the deflation rate does not mean an immediate halving of inflation. Currently, new SOL being issued is still on the books, just that the rate of decrease in issuance will be steeper in the upcoming years. The reduction applies to the incremental issuance over the next six years, not the already circulating supply.

SGP itself is merely directional authorization, not an automatic change to the chain. Developers still need to encode the new curve into SIMD-0550 and go through testing, client collaboration, and feature gate activation. Participants estimate that coordination among validators alone will take at least another four and a half months. On the evening of August 28, the daily issuance had not decreased.

Thinner yields are what the staking desks will face immediately

The current nominal staking yield is about 5.25%, with most coming from protocol inflation, supplemented by fees and MEV. According to calculations by 21Shares and the proposal itself, the first year under the new curve will yield about 4.34%, the second year about 3%, and the third year about 2.25%; if the old curve is maintained, the yield in the third year would still be about 3.52%.

This is clearer when projected to individual users. Staking 100 SOL now yields about 5.25 SOL per year; at the price of approximately $102.55 at the end of August, that comes to nearly $538. In three years, the yield will be about 2.25 SOL, approximately $231. The newly issued tokens decrease by about 57%. Non-staking holders, therefore, are less diluted—issuance inherently pulls a layer from every holder's pocket before replenishing the stakers.

The proposal also estimated a validator's accounting: in the first year, two nodes would face losses, and by the third year, it could reach 30 nodes. Validators like Figment and Everstake, who lean towards custodial and staking operations, opposing this was not surprising. The opposing vote from Figment was about 17.07 million SOL, one of the largest single opposing forces on the public record. For them, thinner yields will first impact node economics, then transmit to custodial institutions and exchange products.

About two-thirds of Solana's supply is in staking. By the end of Q2, the staked amount was about 427 million. In the same period, staking income was about $487 million, down 23% from Q1; issuance still accounted for over 98% of staking income, with Jito tips just a fraction. Cutting issuance makes this structure even more glaring. To maintain a high staking rate going forward, the network will need to rely more on real usage, fees, and MEV, rather than issuing tokens to support nodes.

That night, among three proposals, only two passed

In the same governance round, the destinies of the other two proposals were different. SGP-0001 "Solana Constitution" passed with about 86%, cementing the subsequent voting rules. SGP-0003 "Resource and Inclusion Fees" only received 53.9%, failing to reach two-thirds. It originally intended to separate fees into inclusion fees and resource fees, raising daily destruction from about 650 SOL to a maximum of about 9,000 SOL.

This was originally a "less issuance + more burning" combination. The reduced issuance passed, but the increased burning did not. The supply side only moved halfway. For those bullish on SOL's long-term scarcity, the story is not over; for those relying on burning to improve token economics, this missing half is critical.

Around $102, the market first noted the direction without prematurely realizing a reduced issuance

During the voting window, SOL spiked to about $103 intra-day. Subsequent reports mentioned prices near $106, a 24-hour drop of about 3%, with a weekly rise of about 10%. At the time Forbes was writing, the price was still around $102.

This position makes sense. Reduced issuance is a medium-term curve, not a switch to be turned off the next day; thinning staking yield is bad news for exchanges and nodes, but for non-staking holders, it may mean less dilution. Both long and short positions can coexist, making the market prone to horizontal movement. The real price test will come after SIMD-0550 is activated and the new curve begins its ascent.

For ETF, vault companies, and institutional allocations, an early tightening of supply may be a positive factor. For staking desks relying on yields just over 5%, product pricing, commissions, and node layouts will need a complete rework. Kraken itself is still selling SOL staking: flexible and fixed-term options of about three days coexist, with commissions not low. Customers seek convenience, with the exchange managing validators and voting for them. SGP-0002 documented the costs of this convenient setup in the public records.

Looking ahead to three things

First, when will the code turn? The passage of the proposal does not mean inflation has already changed. The activation date has not been set, and four and a half months is only the reported lower limit.

Second, can 2.25% retain stakers? The nominal yield dropping from just over 5% to just over 2% will determine who among liquid staking, institutional delegation, and small validators can’t sustain it, impacting the network's security budget reliability.

Third, how will the next key votes be cast? Coverage mechanisms, the role of custodians, and the silence of delegators will not automatically disappear because of Sethi’s single statement. The next votes involving fees, destruction, or inflation will first look at which major custodians take a stand, and then whether they will change their positions at the last minute.

The constitution passed, the deflation passed, but the burning did not. Solana cut issuance for the first time under the new governance, relying on a narrow margin of 0.334 percentage points and a major exchange's last-minute vote change. The 2.25% is not tomorrow's interest rate, but a direction for three years from now. SOL is still around $102. The direction has been noted, but the curve has yet to start turning.