Ethereum supply on exchanges falls to 2016 levels — and institutions are ‘quietly’ scooping up

ambcryptoPublished on 2025-12-17Last updated on 2025-12-17

Abstract

Ethereum's exchange supply has plummeted to its lowest level since 2016, signaling a historically tight supply environment. Concurrently, institutional and corporate entities are accumulating ETH at an accelerated pace, with 27 public companies and government-linked entities now holding nearly 6 million ETH valued at $17.7 billion. This represents almost 5% of the total supply. Key drivers for the supply squeeze include staking, which has locked over 37 million ETH, absorption of liquidity by Layer 2 networks, and a shift toward long-term holding. This combination of dwindling sell-side supply and rising institutional demand creates the potential for a significant supply shock, a condition that has historically preceded strong upward price movements for Ethereum.

Ethereum is entering one of its tightest supply eras ever, with new data showing that exchange balances have dropped to their lowest point since 2016, just as corporate and institutional entities increase their ETH holdings at the fastest pace in years.

Ethereum’s exchange supply ratio has fallen to 0.137, according to CryptoQuant — a level last seen in the network’s earliest days.

In previous cycles, similar supply squeezes have preceded major price expansions, as less ETH available on exchanges reduces immediate sell-side pressure and signals growing conviction among long-term holders.

Supply leaves exchanges as institutions accumulate Ethereum

New data reveals a clear counter-trend: while ETH on exchanges continues to contract, entities holding ETH in treasury are steadily rising.

Figures from Coingecko show:

  • 27 public companies and government-linked entities now hold ETH
  • Combined holdings total 5,961,187 ETH
  • Treasury ETH is valued at $17.7bn, up nearly 50% from the previous reporting period
  • Treasury ownership accounts for 4.94% of all ETH

The list includes U.S.-listed firms such as Tom Lee’s BitMine Immersion, SharpLink, Coinbase Global, and others.

Notably, BitMine Immersion added 407,331 ETH in the last 30 days alone — one of the most aggressive accumulation streaks by a public entity in ETH’s history.

This expansion of corporate ETH reserves adds a layer of structural demand that did not meaningfully exist in previous cycles.

Why Ethereum’s supply is tightening

Multiple forces are contributing to the decline in exchange balances:

  • Staking: Nearly 37 million ETH remains locked in validators
  • L2 ecosystems: Base, Arbitrum, Optimism, and others continue absorbing ETH liquidity
  • Treasury adoption: Corporates increasingly view ETH as an operational and strategic asset
  • Long-term holding behavior: Investors are withdrawing to self-custody rather than actively trading

With sell-side supply dwindling and institutional absorption rising, Ethereum appears to be entering a low-liquid, high-demand environment — one that historically precedes strong upward volatility.

What this could mean for ETH’s price

Ethereum recently traded around $2,900, stabilizing after a choppy few weeks. While short-term price action remains tied to broader market sentiment, the supply structure is shifting beneath the surface.

If exchange balances continue to fall and treasury accumulation remains steady, Ethereum could face a classic “supply shock” scenario — where even moderate demand triggers outsized upside.

Final Thoughts

  • Ethereum is experiencing its tightest exchange-supply conditions since 2016, setting the stage for a potential supply squeeze.
  • Institutional accumulation of nearly 6 million ETH adds strong long-term support and introduces a new demand engine not present in earlier cycles.

Related Reads

OpenAI Post-Training Engineer Weng Jiayi Proposes a New Paradigm Hypothesis for Agentic AI

OpenAI engineer Weng Jiayi's "Heuristic Learning" experiments propose a new paradigm for Agentic AI, suggesting that intelligent agents can improve not just by training neural networks, but also by autonomously writing and refining code based on environmental feedback. In the experiment, a coding agent (powered by Codex) was tasked with developing and maintaining a programmatic strategy for the Atari game Breakout. Starting from a basic prompt, the agent iteratively wrote code, ran the game, analyzed logs and video replays to identify failures, and then modified the code. Through this engineering loop of "code-run-debug-update," it evolved a pure Python heuristic strategy that achieved a perfect score of 864 in Breakout and performed competitively with deep reinforcement learning (RL) algorithms in MuJoCo control tasks like Ant and HalfCheetah. This approach, termed Heuristic Learning (HL), contrasts with Deep RL. In HL, experience is captured in readable, modifiable code, tests, logs, and configurations—a software system—rather than being encoded solely into opaque neural network weights. This offers potential advantages in explainability, auditability for safety-critical applications, easier integration of regression tests to combat catastrophic forgetting, and more efficient sample use in early learning stages, as demonstrated in broader tests on 57 Atari games. However, the blog acknowledges clear limitations. Programmatic strategies struggle with tasks requiring long-horizon planning or complex perception (e.g., Montezuma's Revenge), areas where neural networks excel. The future vision is a hybrid architecture: specialized neural networks for fast perception (System 1), HL systems for rules, safety, and local recovery (also System 1), and LLM agents providing high-level feedback and learning from the HL system's data (System 2). The core proposition is that in the era of capable coding agents, a significant portion of an AI's learned experience could be maintained as an auditable, evolving software system.

marsbit20m ago

OpenAI Post-Training Engineer Weng Jiayi Proposes a New Paradigm Hypothesis for Agentic AI

marsbit20m ago

Your Claude Will Dream Tonight, Don't Disturb It

This article explores the recent phenomenon of AI companies increasingly using anthropomorphic language—like "thinking," "memory," "hallucination," and now "dreaming"—to describe machine learning processes. Focusing on Anthropic's newly announced "Dreaming" feature for its Claude Agent platform, the piece explains that this function is essentially an automated, offline batch processing of an agent's operational logs. It analyzes past task sessions to identify patterns, optimize future actions, and consolidate learnings into a persistent memory system, akin to a form of reinforcement learning and self-correction. The article draws parallels to similar features in other AI agent systems like Hermes Agent and OpenClaw, which also implement mechanisms for reviewing historical data, extracting reusable "skills," and strengthening long-term memory. It notes a key difference from human dreaming: these AI "dreams" still consume computational resources and user tokens. Further context is provided by discussing the technical challenges of managing AI "memory" or context, highlighting the computational expense of large context windows and innovations like Subquadratic's new model claiming drastically longer contexts. The core critique argues that this strategic use of human-centric vocabulary does more than market products; it subtly reshapes user perception. By framing algorithms with terms associated with consciousness, companies blur the line between tool and autonomous entity. This linguistic shift can influence user expectations, tolerance for errors, and even perceptions of responsibility when systems fail, potentially diverting scrutiny from the companies and engineers behind the technology. The article concludes by speculating that terms like "daydreaming" for predictive task simulation might be next, continuing this trend of embedding the idea of an "inner life" into computational processes.

marsbit22m ago

Your Claude Will Dream Tonight, Don't Disturb It

marsbit22m ago

Duan Yongping's Bottom-Fishing in CoreWeave Is Turning into a Battleground for Bulls and Bears

CoreWeave's Q1 2026 earnings report has intensified the ongoing bull-bear battle over the AI infrastructure stock. While revenue doubled year-over-year to $2.08B and the firm's remaining performance obligation (RPO) surged to nearly $100B, its net loss more than doubled to $740M. The critical point of contention is profitability: while adjusted EBITDA margin was a robust 56%, the adjusted operating margin collapsed to just 1% due to soaring infrastructure and sales costs. A weaker-than-expected Q2 revenue guidance further triggered an 11.4% single-day stock drop. The bull thesis hinges on CoreWeave's massive order backlog, deep strategic ties with NVIDIA (as a customer, investor, and key supplier), and a diversified client base now including Anthropic and Meta. The bear case focuses on the "scale at all costs" model, where expanding revenue leads to wider losses, ballooning debt ($25B), and massive capital expenditures ($6.8B in Q1). Insider selling by executives contrasts with a notable new investor: Chinese investor Duan Yongping initiated a small position (0.12% of his portfolio) in Q4 2025 near the stock's lows. The coming Q2 report is seen as a key test for management's promise of a profit margin recovery.

marsbit28m ago

Duan Yongping's Bottom-Fishing in CoreWeave Is Turning into a Battleground for Bulls and Bears

marsbit28m ago

China's Version of 'Tech Burning Man' Debuts in Shanghai, muShanghai Creates a Global Geek 'Pop-up City'

From May 10 to June 6, 2026, the inaugural muShanghai "Pop-up City" experiment, dubbed China's version of a "tech Burning Man," launched in Shanghai. Co-organized by the international open-source community The Mu and the Hongqiao Alibaba Center, this 28-day event aimed to build a global "parallel city" for geeks, attracting over 800 participants from more than 50 countries, including former OpenAI engineers and startup founders. The program featured four themed weeks: AI Week, Biotech Week, Robotics Week, and Culture Week, hosting nearly 100 sessions like ClawCon 2026. Activities ranged from discussions on AI safety and consumer apps to robot battles and cyberpunk culture, culminating in large outdoor "Innovator Marketplaces." A core principle was "Build in Public," encouraging open sharing of projects and progress. Hongqiao Alibaba Center served as the co-host and primary venue, positioning itself as a first-stop hub for international talent in China. The event marks a significant step for The Mu community, which has previously organized similar pop-up cities in Argentina and San Francisco, in bringing its model of immersive, collaborative innovation to China. It aims to be a key window for global科创 (scientific and technological innovation) exchange.

marsbit28m ago

China's Version of 'Tech Burning Man' Debuts in Shanghai, muShanghai Creates a Global Geek 'Pop-up City'

marsbit28m ago

CLARITY Act: Banking Trade Groups Push For Yield Agreement Revision – Details

US banking trade groups are urging revisions to the stablecoin yield compromise in the upcoming CLARITY Act ahead of a key committee markup. The Act currently aims to ban all passive, deposit-like interest on stablecoins to prevent competition with traditional bank savings, while allowing rewards tied to active uses like staking or transactions. In a letter, groups including the American Banking Association and Bank Policy Institute proposed stricter language to eliminate perceived loopholes for passive yield and prevent deposit flight from banks. However, these efforts are reportedly viewed as minor by some lawmakers. The Senate Banking Committee is scheduled to mark up the bill on May 14, a critical step before it can advance through Congress.

bitcoinist13h ago

CLARITY Act: Banking Trade Groups Push For Yield Agreement Revision – Details

bitcoinist13h ago

Trading

Spot
Futures

Hot Articles

Discussions

Welcome to the HTX Community. Here, you can stay informed about the latest platform developments and gain access to professional market insights. Users' opinions on the price of ETH (ETH) are presented below.

活动图片

Top Questions