[ACP-285] Reduce Minimum Consumption Rate

[meaghanfitzgerald]

This proposal reduces the MinConsumptionRate parameter governing Primary Network staking rewards from 10% to 7.5%. MinConsumptionRate sets the annualized reward rate for a validator whose stake duration approaches the minimum staking duration period. MaxConsumptionRate is not modified, leaving the maximum reward available to long-term validators unchanged. The change is intended to improve the long-term alignment between staking duration and network security and to redistribute rewards toward longer-duration participants.

[hinkavax]

Thanks @meaghanfitzgerald, the revision is a clear step forward. The 30-day ramp, grandfathering active stakes, and especially the explicit acknowledgment of risk to LSTs and "relevant strategies" all show the concern is being taken seriously.

We'd build on exactly that LST acknowledgment, with one structural point.

The ramp smooths when the cut lands, not where the 14-day rate ends up. After the ramp, a 14-day stake still settles at ~4.08% vs 5.36% today, a permanent −1.28pp that flows straight into LST yields, lending-market rates, and the spread that keeps looping positions solvent.

More importantly, the flat cut doesn't target the behavior ACP-273 actually enables. The new behavior is sub-14-day rolling stakes — nothing sits below 14 days today, since that's the current MinStakeDuration. Yet under the proposed cut, 48 hours (4.00%) and 14 days (4.08%) are still only ~0.08pp apart, so rolling 48h instead of committing to the structural minimum is barely penalized. The cut mostly lowers the whole floor, and its headline effect is widening the 14d→365d premium from 1.02 to 2.30pp.

A fixed 14-day breakpoint targets the regression directly:

• d ≥ 14 days: existing formula, completely unchanged.
• d < 14 days: steep interpolation from the 7.5% floor up to the existing 14-day rate.

(Pinned at a fixed 14 days — the pre-273 minimum — deliberately: since ACP-285 co-activates with ACP-273, a breakpoint defined on the live MinStakeDuration would collapse to 48h at activation and the steep segment would cover nothing.)

This produces a real ~1.2pp penalty between 48h (4.18%) and 14d (5.36%) — a far sharper disincentive against sub-14-day stakes than the flat cut's ~0.08pp — while preserving the 14-day rate entirely. It needs no transition ramp of its own, because it disturbs no existing stake: everything staked today is at or above 14 days and stays untouched.

It's also worth being concrete about who would actually occupy the sub-14-day range once ACP-273 opens it. The natural occupants are custodial and ETF stakers, Grayscale's GAVA and Bitwise's BAVA already hold and stake AVAX, whose binding constraint is redemption liquidity, not yield: staked AVAX they can't unwind quickly impairs their ability to meet redemptions. That cohort will gravitate to the shortest available duration and is largely indifferent to the rate, accepting a lower 48h yield as the price of near-daily liquidity. The steep sub-14-day segment therefore does double duty, a genuine disincentive for discretionary stakers who could commit longer, and an appropriate liquidity discount for the ETF cohort that they'll pay rather than be deterred by. The flat cut, by contrast, lowers the 14-day rate paid to yield-sensitive LSTs that feed DeFi: it lands on the cohort that minds and spares the one that doesn't.

On the duration premium: the breakpoint leaves 14d→365d at 1.02pp rather than widening it, and we'd argue that's correct. Widening that premium is the speculative lever, the proposal's own Goal 3 projects only a ~2-month shift in stake-weighted duration and frames it as an expected secondary effect, not a guaranteed one, with mixed evidence on whether large validators move at all. Sacrificing the structural 14-day rate (at the cost of real DeFi losses) to widen a premium the model says may not change behavior is a poor trade.

On inflation, the breakpoint still trims issuance to the sub-14-day cohort; it forgoes only the broad level reduction across the structural 14–365 day range. Weighing that honestly: the reduction the breakpoint gives up is diffuse (spread across all holders as marginally-lower dilution), modest, and reached only gradually, whereas the harm the flat cut imposes is concentrated on the full-security DeFi cohort, effectively certain, and amplified by second-order effects in looping and lending. We'd rather not fund a diffuse, gradual emissions reduction by compressing the rate paid to the validators who provide both security and DeFi liquidity, particularly when the breakpoint preserves most of the structural benefit and lands the cut on the cohort that accepts it.

Ask: adopt a fixed 14-day breakpoint, concentrating the gradient where ACP-273 creates the regression. If a curve change is out of scope for the Helicon timeline, extend the ramp from 30 to 60–90 days as an interim cushion.