LINK: treasury releases ~12x what the Reserve absorbs. Is the Reserve real net demand?

Object of investigation

LINK (Chainlink, Ethereum): treasury releases via Binance deposit 0x9B1c...cDE9 vs Chainlink Reserve intake (inferred 0x9a70...9ec6) and the Community Staking Pool

Observation

Sourced facts (read times in evidence). Etherscan, ~20:48 UTC 2026-10-11: 0x9B1c...cDE9 carries the public tag 'Binance Deposit' and shows LINK-contract transactions on 2026-10-09 between 20:23 and 21:13 UTC (amounts not shown on the fetched page). 0x9a70...9ec6 holds 6,204,681.79 LINK (no name tag). The Chainlink Community Staking Pool holds 40,875,543.64 LINK. Hokanews (2026-10-03): the Chainlink Reserve added 462,730 LINK in August and 373,791 in September (cumulative 6,047,498), funded by offchain enterprise revenue and onchain usage via Payment Abstraction. Chainlink's Q1 2026 review: the Reserve added 1,473,379 LINK in Q1 (3.06M total); CCIP fee revenue rose 213% quarter on quarter. DefiLlama: circulating supply 766.85M of 1B; quarterly buybacks of $14.36M, $14.6M, $15.05M and $14.71M for Q4 2025 to Q3 2026; no burns.

Hypothesis

Interpretation, not established fact. The author's own tracing of Ethereum LINK transfers (a non-public indexed dataset, not verified here) found that round-balance non-circulating wallets release LINK in four batches a year, 70.0M LINK in each of the last two cycles, mostly via the Binance deposit address above. The Reserve took in about 5.54M LINK over the same 12 months, a ratio of about 12.6 to 1. The Community Staking Pool is full, with monthly inflows equal to outflows. About 45% of LINK entering the Reserve's payment converter (0x5680...171d) traces back to wallets funded by a monthly treasury drip, and about 51% to DEX and CoW buy executors. If correct, usage growth such as CCIP fees is not reaching LINK proportionally, and part of the Reserve's intake is recycled treasury LINK rather than open-market demand. The Reserve address is identified by matching Chainlink's published deposit amounts, which is an inference.

What would invalidate it

Invalidated if: (1) public tracing shows Reserve converter inflows come mostly (over 80%) from open-market purchases; (2) treasury releases are materially smaller than about 70M LINK a year or have been paused; (3) Reserve annual intake approaches the treasury release rate; (4) staking capacity expands with fee-funded rewards and shows net inflows; or (5) 0x9a70...9ec6 is shown not to be the Chainlink Reserve.

Published observations provide a starting point, not a verified conclusion.