Lido & stETH Weekly News Review — Week of August 31–September 6, 2026

1. The week in one arc

This edition covers August 31 – September 6, 2026 — seven dated stories, all sourced, nothing borrowed from outside the window. The arc is not a price story. It is a plumbing story: for the first time since this site launched, the network-level numbers moved more than the market did. A record 34.94% of ETH is now staked on the Sep 2 read (35.55% on our Sep 7 snapshot), the entry queue has swollen to 2.07M ETH with a 36-day wait, the exit queue is empty, and — the fact that makes the other three strange — staking pays 131 basis points less than a three-month Treasury bill. Meanwhile Ethereum's fee engine, the thing that was supposed to make staking yield competitive, burned 1.3% of what it issued over the trailing month. Prices did very little: ETH opened the window at $2,417.82 and closed it at $2,479.92, +2.6%. Everything that mattered happened underneath.

2. Sep 1: staking pays 2.61% against a 3.92% three-month bill — the carry gap, in official numbers

On September 1, US Treasury data put the three-month bill at 3.92%. The same week, validatorqueue.com put Ethereum staking APR at 2.61% across 904,541 active validators representing 34.94% of supply. The gap is 131 basis points — you can be paid more, in dollars, with no validator risk, no slashing risk and no smart-contract risk, than you are paid for securing Ethereum. FinanceFeeds framed the contradiction plainly, and it is the cleanest statement of the staking trade's current problem anyone has published: nobody queues to earn 131bp less than cash.

Source: US Treasury data for Sep 1 and validatorqueue.com for Sep 2, as compiled by FinanceFeeds (Sep 2) and republished by Gate and Cointelegraph.

Take: this is the number the whole "why stake at all" debate reduces to, and it is worth being precise about what it does and does not prove. It does not prove staking is uneconomic — holders accept a lower cash yield for ETH-upside participation and on-chain composability, and both are real. What it proves is that yield is not the reason capital is entering. Once you accept that, the next question becomes the interesting one: if not yield, then what? Item 3 supplies the answer, and it is not a comfortable one for the decentralisation thesis.

3. Sep 2: 2.07M ETH waits 36 days to get in; the exit queue is completely empty

Verified on September 2: the validator entry queue held 2,074,270 ETH — roughly $4.92 billion at that day's $2,373 spot — an intraday read; the same day's close was $2,417.92 — with an estimated 36-day wait to activation. The exit queue held exactly zero ETH. Not "small": zero. Staking participation stood at 34.94% of supply, and the same source noted that if the entire current entry queue processed, the staked ratio would rise to roughly 36.6%. Exit wait times were under three hours.

Source: validatorqueue.com and beaconcha.in, as reported Sep 2 by FinanceFeeds, Gate News and Cointelegraph.

Take: a zero exit queue is the single most bullish structural fact about staked ETH on the board, and it is being read backwards. It means not one staker — not institutions, not whale treasuries, not leveraged loopers — wanted out badly enough to queue, during a week when staking yielded less than cash. Combine it with a 36-day entry wait and you have a market with a one-way door and a line around the block to get through it. The competing explanations are testable: either holders are in it for something other than the coupon (consistent with the ETF and treasury-company flows carried in this week's data report), or the exit queue is artificially suppressed by re-staking and looping structures that never actually unwind. We can't yet distinguish them, and the distinction matters enormously — see this week's deep dive.

4. Sep 2: the fee engine — 1,160 ETH burned against 87,700 ETH issued in thirty days

The same Sep 2 reporting pulled the number that explains why staking yields keep falling: over the trailing 30 days Ethereum burned 1,160.53 ETH against roughly 87,700 ETH of issuance — the burn running at about 1.3% of issuance, against the 100%-plus implied by "ultrasound money." The 30-day average base fee sat at 0.179 gwei. Holding supply flat would require 13.384 gwei. Net supply growth is running at +0.86% a year. Ethereum L1 collected $10.4 million in fees over 30 days.

Source: ultrasound.money and DeFiLlama fee data, as reported Sep 2 by FinanceFeeds and Gate.

Take: read the two items together, because neither makes sense alone. Staking demand is at an all-time high and rising, while the network's ability to pay stakers from fees has fallen to roughly one-seventy-fifth of the level that would stabilise supply — 0.179 gwei against the 13.384 gwei needed. Every basis point of staking yield is now consensus-layer issuance; the execution layer contributes essentially nothing. That is why the yield compressed from 2.51% to the 2.2% area over seven months, and it is why item 6's upcoming gas-limit increase is a supply story rather than a demand story. Our founding deep dive predicted the direction; this is the first hard number that sizes the mechanism.

5. Sep 3–4: ETH's +4.9% day, and a window that went nowhere in the end

The window's largest single-session move came on September 4, when Ether closed at $2,507.65, +4.89% on the day, its high of the week. It did not hold: Sep 5 closed back at $2,456.08, and the window closed on Sep 6 at $2,479.92. The full daily path on the CoinGecko close series: $2,417.82 (Aug 31), $2,466.94 (Sep 1), $2,417.92 (Sep 2), $2,390.78 (Sep 3), $2,507.65 (Sep 4), $2,456.08 (Sep 5), $2,479.92 (Sep 6) — a +2.6% window whose entire range fits inside $2,391–$2,508.

Source: CoinGecko daily OHLC via The Defiant; Bitget printed $2,482.01 on Sep 6 16:48 UTC, within 0.1%.

Take: a +4.9% day that hands back 44% of the move in the very next session (and 24% of it by the window's close) is a liquidity event, not a trend, and the flat week it sits inside is the more informative fact. This is the second consecutive week in which the macro repricing dominated the headlines and the tape did almost nothing — which is what you'd expect if positioning had already been adjusted ahead of the September 15–16 FOMC. The interesting move this week was in the validator queue, not the price.

6. Sep 3–6: LDO gains 11%, its strongest week since July

LDO reversed last week's −12%: from a $0.3527 close on Aug 30 to $0.3917 on the tracker read, +11.1%, with the daily series showing the move concentrated in a +8.2% session on Sep 3 and follow-through of +3.1% on Sep 5 and +0.6% on Sep 6. Desk coverage attributed the push to a "September 2026 Lido Update" that reassessed the protocol's risk-and-reward profile for allocators, noting price had moved above its 20-, 50- and 200-day moving averages with resistance identified at $0.4203. Weekly range: $0.3447–$0.4050. Those three sessions compound to +12.2% on their own, so the remaining sessions of the stretch — Aug 31, Sep 1, Sep 2, Sep 4 and the two after the window — netted about −1% (DERIVED). The daily path and the +11.1% headline do reconcile.

Source: CoinLore (Sep 8 03:38 UTC, 7-day change +11.32%); daily series via BestCryptoChecker; technicals and the Lido Update via TradersUnion.

Take: note what was not cited: the desk itself flagged that no new products, regulatory developments or ecosystem events accompanied the move. An 11% week on a portfolio-review PDF, after a −12% week on a hawkish Fed — that is a token trading beta and narrative, not fundamentals, in both directions. We said last week that LDO "paid the market's beta bill"; this week it collected the rebate. Two-sided honesty means printing both with equal weight. The staking business underneath did not change in either week.

7. Sep 5–6: the peg, checked again — stETH $2,480.62 against ETH $2,479.92

At the window's close the two prices were within a dollar of each other — a 70-cent spread: stETH at $2,480.62 (Sep 5 read) against Ether's $2,479.92 Sep 6 close — a ratio of 1.0003, no discount in any whitelisted source. This is now the second consecutive week the peg has been tested against a macro-driven tape and the second week it has not moved. stETH's market capitalisation on that read was $23.97 billion across 9,661,935 stETH.

Source: crypto.news stETH market page (Sep 5); CoinGecko ETH close (Sep 6).

Take: two clean weeks is not a stress test — nothing this month has forced a leveraged unwind — and we will keep saying so until something actually breaks. But it is worth recording why it hasn't moved: redemption depth is adequate and the exit queue is empty, which means anyone who wanted to arb a discount had a working, fast, three-hour exit. Peg stability is a function of exit capacity, not of confidence. Right now exit capacity is abundant, and that is the honest reason the peg looks good.

8. Sep 6: the tape — BTC $79,992, ETH $2,503, and derivatives volume down 40%

The window closed with a broadly green but listless session: Bitcoin $79,991.96 (+0.2%), Ether $2,503.07 (+0.91%), Solana +3.61% to $106.92, XRP +0.86% to $1.42, and Polkadot the standout at +7.8% to $0.9872; BNB was the only notable decliner at −1.3% to $756.87. Every price in this section is an intraday snapshot at 18:00 IST, not a daily close — the CoinGecko Sep 6 close we use everywhere else on this page is $2,479.92, about 0.9% below the Ether print here. The more telling figures came from the plumbing: total crypto market cap $2.7094 trillion, ETH dominance 11.29% (up 0.13pp), BTC dominance 59.12% (down 0.46pp), derivatives 24h volume down 40.12% to $440.7B and stablecoin volume down 30.33%.

Source: Moneynomical with Yahoo Finance data (Sep 6, 18:00 IST); TokenPost Market (Sep 6, 16:00 KST) which printed BTC $79,778 and ETH $2,505.99.

Take: a 40% collapse in derivatives volume alongside a flat-to-up spot tape is the signature of positioning being taken off ahead of a known event, not of a market changing its mind. Everyone has decided to wait for September 15–16. For staked-asset holders the practical consequence is that the next two weeks of peg and yield data will be low-signal — thin derivatives means less forced-selling pressure, which is mildly protective, and also means any move that does come will be spot-led and therefore more durable.

9. Numbers of the week — every date inside the window

Eleven dated prints. Row 9 is the one measurement that extends past the window's final session — our sources publish no clean Sep 6 daily close for LDO, so we use the closest verifiable print and say so.

10. What we left out — and why

• The Tectonic exploit and the Cronos rollback. Both appear in the week's DeFi TVL wrap, but neither carries a date we could verify as falling inside Aug 31–Sep 6 from a whitelisted source. Until we can date them, they stay out of a column whose entire premise is dated events.
• Any "Lido crosses $30B" claim. Several content farms published this during the window. It is not supported by DefiLlama, lido.fi or any stETH supply read we can verify — those cluster at $21–24B. We do not launder an unverified number through a news slot.
• ETH at "$2,987" or "$3,100". One widely-circulated September blog put Ether above $3,000 during this window. No exchange, aggregator or market-wire print corroborates it; the daily closes are $2,391–$2,508. Rejected.
• Protocol state figures. Lido TVL, stETH supply, the yield corridor and the competitive board belong to this week's data report, where each carries a read date.

The week's three headline facts — 2.07M ETH queued 36 days to enter, an exit queue holding exactly zero, and staking paying 131 basis points less than Treasury bills — point in the same direction, which is what makes them jointly remarkable. If stakers were purely yield-maximizing, the negative carry would thin the entry queue; if entrants were driven by fear, exits would be long, not empty. The joint reading is that the marginal entrant is neither arbitrage capital nor frightened capital but conviction capital, or capital whose exit is structurally hedged elsewhere — and the queue's 36-day depth is the market's way of pricing patience. What would change the reading is as important: an exit queue that starts to fill while entries continue, or a queue-composition disclosure showing leveraged loops. Both are named in the watchlist precisely so the interpretation can be falsified rather than defended.

11. How this review is produced

Unchanged since the founding edition: a fixed source whitelist (Lido DAO official channels, Ethereum Foundation materials, beaconcha.in-class explorers, DefiLlama, Dune Analytics, CoinGecko/CoinMarketCap, validatorqueue.com, ultrasound.money, and named market media), then collect → de-duplicate → grade importance → classify → summarize → cross-verify → comment. Every story here is dated Aug 31–Sep 6, 2026, and any figure measured at the window's edge says so in its own line. When we get a number wrong, the correction is published alongside the record rather than edited into it — a practice this week's data report exercises twice, on ourselves.

FAQ

Related reading

• Lido & stETH Weekly Data Report — Week of August 31–September 6, 2026
• The Empty Exit Queue: 2.07M ETH Is Waiting 36 Days to Earn Less Than Cash