Why Solana Staking Yields Have Not Changed After SGP-0002

TL;DR. SGP-0002 passed on August 28, 2026 with 67.001% For, clearing the two-thirds bar by 0.334 points. It is a governance mandate for SIMD-0550 (double annual disinflation from 15% to 30%), not a live mainnet APR cut. Epoch rewards still follow the existing 15% disinflation schedule. Before yields can move, SIMD-0607 must replace floating-point inflation math with deterministic integers, Agave and Firedancer must align, and validators must schedule the SIMD-0550 feature gate at an epoch boundary. No public activation date exists. If you stake natively or hold an LST, nothing changed in your last payout. The gate is still closed.

SGP-0002 made headlines. Staking dashboards did not blink.

Solana's first economic governance win approved a faster path to lower inflation. Headlines and social posts sometimes read as if staking yields already stepped down. They have not. The passed proposal is permission to change the schedule, not the schedule itself.

Here is what SGP-0002 did, what SIMD-0607 is blocking, and what to watch before your APY path moves.

---

What SGP-0002 did (and what it did not)

On August 28, 2026, SGP-0002 passed with 67.001% For, 0.334 percentage points above the 66.667% supermajority threshold.

SGP-0002 is a governance mandate for SIMD-0550. If activated, SIMD-0550 would double the annual disinflation rate from 15% to 30%. Modeling attached to the proposal projects roughly 18.9 million fewer SOL issued over six years and a 1.5% terminal inflation floor around 2029 instead of roughly 2032. The terminal rate itself does not change. Only the speed to get there does.

What SGP-0002 did not do: flip a switch on mainnet. It is not a live APR cut. Your epoch rewards still follow the existing 15% disinflation schedule until the activation chain completes.

For the full ballot context, including how JitoSOL staker override tipped the margin, see our first SGP ballot recap.

---

What SIMD-0607 is

Before SIMD-0550 can activate, SIMD-0607 must clear review and ship.

SIMD-0607 is titled Remove Floating Point from Runtime. Status: Draft. Created August 21, 2026. Authors: Tao Zhu (Anza) and Tom Pointon (Firedancer). It extends SIMD-0194 and SIMD-0550.

The problem is consensus math, not politics. Inflation rewards and rent parameters affect capitalization, account state, and bank hashes. Agave and Firedancer must produce bit-for-bit identical reward math on every validator. Floating-point / IEEE-754 arithmetic cannot guarantee that across hardware and compilers. SIMD-0607 replaces those operations with deterministic integer math so both clients hash the same state.

That makes SIMD-0607 a prerequisite gate, not a side quest. Faster disinflation is approved in principle. It cannot go live until reward calculation is safe for dual-client consensus.

---

Why integer math gates the yield change

Think of it as a bank reconciliation problem at network scale.

Every epoch, Solana computes inflation rewards, updates stake accounts, and produces a bank hash that every validator must match. If Agave on one machine and Firedancer on another round a floating-point intermediate differently, you do not get a polite warning. You get consensus failure.

SIMD-0550 changes how fast inflation falls. SIMD-0607 changes how that math runs so two independent clients agree on every lamport. You cannot schedule the rate change until the math is deterministic.

Implementation status as of early September 2026:

Do not treat draft PRs as shipped code. The gate is still upstream.

---

The activation chain (checklist)

Solana Compass summarized Anza's post-vote sequencing on August 29, 2026. The order matters:

1. SIMD-0607 merges and clears review.
2. Client release ships with the integer-math changes. Compass cited Agave v4.4 as the target release line.
3. Feature gates for SIMD-0607 and SIMD-0550 are scheduled across the validator set.
4. Disinflation rate changes at an epoch boundary when the SIMD-0550 gate flips.

There is no public activation date for SIMD-0550. Say that plainly. Do not budget on a calendar that has not been published.

Until step 4, current epoch rewards follow the existing 15% disinflation schedule. That is true whether you stake natively, delegate to a validator, or hold a liquid staking token.

For the projected APY path once activation arrives, revisit our staking rewards and disinflation guide. Some September 2026 coverage framed staking yields as already heading to roughly 2.25%. That describes the long-run path after activation, not today's payout.

---

What native stakers should watch

Your last epoch payout is unchanged. SGP-0002 set direction. It did not rewrite this week's inflation rate.

Watch the implementation chain, not the vote tally. SIMD-0607 merge status, Agave release notes, and Firedancer alignment matter more than re-reading the August 28 result.

Do not assume immediate yield compression. Budget on the current schedule until gate activation is announced. Then update projections from official release communications, not from headline APY figures.

Governance still matters for the next fight. SGP-0001 ratified the Constitution. SGP-0002 proved a close economic proposal can pass. Your override vote remains relevant. Our governance and staker voting guide covers who controls the ballot for native stake.

If you are earning on Hubra, the mechanics are the same inflation curve as anywhere else on Solana. The difference is visibility: you can track positions while the schedule stays on its pre-gate path at hubra.app.

---

What LST holders should watch

Same curve, same gate. JitoSOL, mSOL, and other major LSTs earn the same underlying inflation rewards as native stake. When SIMD-0550 eventually activates, liquid positions move with it. There is no separate yield schedule for LSTs, and nothing has changed yet.

Governance is the Hubra angle. On SGP-0002, Solana Compass credited JitoSOL staker override with tipping the vote into passage. Pool operators cast the default ballot. Delegators who overrode carried the margin. That pattern will repeat on future proposals.

Do not rotate LSTs because SGP-0002 passed. Do know how your pool handles governance: default operator vote, override availability, and public statements on upcoming SIMDs.

---

What to watch next

1. SIMD-0607 progress. Draft status, Agave PR #14940 updates, and Firedancer review. This is the bottleneck.
2. Agave v4.4 (or successor) release notes. Integer-math landing in a shipped client is the next hard milestone after merge.
3. Feature gate announcements for SIMD-0550. That is when the 15% to 30% disinflation change actually schedules. Until then, yields stay on the pre-vote path.
4. The next SGP. Constitution rules are set. Stakers proved they can swing a close vote. Override remains live for the next ballot.

---

This article is for information only and is not financial advice. Vote tallies, sequencing, and implementation status are cited from Solana Compass, the official SIMD-0607 markdown on GitHub (solana-foundation/solana-improvement-documents), Agave PR #14940, and our prior SGP ballot recap as noted in the text.