Part III – Decentralisation Under Fire: The Quiet Re-Centralisation of Digital Finance

(The Great Unravelling – Part 3 of 3)

Prologue

This is the concluding part of The Great Unravelling series — my ongoing, evidence-based look at measurable distortions inside digital finance. In Part I, I examined the structural weaknesses of bridges and wrapped assets. In Part II, I focused on liquidity illusions and why market capitalisation often masks fragility.

Part III examines the most sensitive topic of all: control. Not in a conspiratorial sense, but in the measurable, structural sense — how power within digital finance has quietly condensed into a handful of nodes, custodians, infrastructures, and regulators.

Decentralisation was supposed to prevent this. Yet the evidence shows a different reality emerging right beneath the surface.

1. The Decentralisation Promise — And the Reality Beneath It

The early promise of blockchain systems was simple:

distributed validation,
censorship resistance,
and economic autonomy without reliance on central operators.

But systems do not remain idealistic as they scale. Economic incentives, infrastructure costs, and regulatory thresholds shape them over time — sometimes more strongly than the underlying protocol rules.

Across 2023–2025, publicly available analytics from platforms such as Messari, CoinMetrics, Glassnode, the Bank for International Settlements (BIS), and major custodial disclosures all point in the same direction: the concentration of power is increasing, not decreasing.

And it isn’t happening because anyone planned it. It is happening because systems naturally follow the path of least resistance.

2. Validator Concentration: The New Power Centres

Proof-of-Stake (PoS) was meant to democratise validation. But the actual distribution of control tells a different story.

Ethereum — the clearest example

On-chain data and staking dashboards make several facts clear:

Lido, Coinbase, and Binance collectively control a very large share of staked ETH, regularly exceeding half of the total stake.
The top set of validators routinely accounts for the majority of attestation participation.
A significant portion of validator infrastructure runs on cloud providers such as AWS, Google Cloud, and Hetzner, as visible in network telemetry and operator disclosures.

These patterns are visible on tools such as Rated.Network, Ethereum staking explorers, and validator distribution dashboards. In Proof-of-Work, mining pools consolidated. In Proof-of-Stake, staking has consolidated. The mechanism changed — the outcome did not.

Why this erodes decentralisation

When a small number of entities produce most blocks and attestations:

coordinated downtime becomes a systemic risk rather than an isolated incident,
governance signalling can be passively shaped,
regulatory or jurisdictional pressure has fewer, more identifiable targets,
correlated slashing or misconfiguration events can destabilise an entire network,
and neutrality becomes dependent on institutional policy rather than protocol design.

This is not theoretical. It is routine in 2025.

3. Custodial Gravity — Where the Assets Really Are

It is difficult to claim decentralisation when control over private keys is centralised.

Public custody disclosures and on-chain clustering analysis show that:

Coinbase Custody, Anchorage, BitGo, Binance Custody and major ETF custodians collectively hold a very large proportion of circulating BTC and ETH on behalf of clients.
Bitcoin ETFs now concentrate substantial BTC exposure under a small number of custodial entities, based on publicly reported fund holdings.
Exchange internalisation means a meaningful share of “liquidity” never touches public order books at all.

On-chain data from Glassnode confirms that most BTC remains dormant for long periods, and when it moves, it often moves between centralised custodial clusters.

The concentration is measurable:

fewer than a few dozen custodial operators effectively control the majority of accessible BTC liquidity,
and a similarly small set of entities controls a large fraction of ETH liquidity.

This directly contradicts the decentralisation narrative that the industry still presents publicly.

4. Real-Time Evidence: What Bitcoin’s Decline Since October 1 Reveals

Here, the trilogy meets the present moment.

Since October 1, Bitcoin’s price has retreated sharply. The reasons are not mysterious; they are supported by observable data and consistent with the structural weaknesses identified in Part I and Part II.

(1) Macro tightening

U.S. Treasury yields have moved to elevated levels.
The U.S. Dollar Index has strengthened.
Risk assets broadly have faced pressure as global liquidity tightened.

These trends are visible in Treasury auction data, central-bank communications, and futures positioning.

(2) ETF outflows

Bitcoin ETF flows are transparent and publicly reported. During this period:

several major ETFs have shown net outflows,
and in-kind redemptions through custodians reduce accessible tradable liquidity, even if underlying assets remain in custody.

(3) Miner pressure

Hashprice — the revenue per unit of hash power — has been at relatively depressed levels, putting pressure on miners. Public wallet data shows periods of increased miner selling as operations adjust to revenue and cost constraints.

(4) Declining spot liquidity

This ties directly back to Part II. Order-book depth data from providers such as Kaiko shows that spot liquidity on many major exchanges has thinned, especially within tight price bands. When liquidity falls, the same amount of selling produces larger price movements.

(5) Custodial concentration

When most BTC sits in ETF vaults, exchange wallets, and institutional custody, less supply is actively available to form real-time bids. That reduces the resilience of spot markets under stress.

In short, Bitcoin’s recent decline is not unusual. It is a predictable consequence of concentrated custody, thinning liquidity, macro tightening, and leverage unwinds — all of which reinforce the broader patterns outlined in this series.

5. Regulatory Forces Quietly Pulling Systems Toward Centralisation

As decentralised ecosystems grow, they attract regulatory frameworks built for traditional finance. These frameworks rarely treat decentralisation as a feature; they treat it as a compliance variable.

Examples include:

FATF Travel Rule — requiring entity attribution and information exchange above certain thresholds, effectively pushing intermediaries to centralise monitoring.
MiCA (EU) — introducing the concept of “responsible entities” for crypto-asset issuance and service provision, which often assumes a central operator.
SEC and CFTC enforcement in the U.S. — driving platforms toward custodial, centralised models to fit within registration or exemption frameworks.
FinCEN rulemaking and guidance — broadening the definition of “control” over digital assets in ways that frequently point back to identifiable intermediaries.

The net effect is clear: decentralised systems either adapt to centralised compliance structures, or they are gradually excluded from regulated financial channels. In practice, this means more power is concentrated in those entities capable of absorbing regulatory obligations at scale.

6. Economic Gravity — Why Decentralisation Contracts Over Time

Technology does not remain decentralised by default. Economies follow power laws — liquidity, attention, and infrastructure pool where friction is lowest.

Across history, this pattern repeats:

banking moved from local banks to national and then global institutions,
telecom networks consolidated into a small number of carriers,
internet access consolidated into a few major ISPs and backbone providers,
cloud computing consolidated into a handful of hyperscale providers.

Digital finance is subject to the same forces.

The cycle is familiar:

Experimentation phase — highly decentralised, many participants.
Adoption phase — competition grows, but liquidity starts to pool.
Maturity phase — consolidation around dominant providers and venues.
Regulatory phase — formal frameworks arrive, further reinforcing centralisation.

Blockchains are not exempt. They operate inside economic and regulatory environments that naturally encourage concentration over time.

7. The Emerging Alternatives (Without Naming Names)

At this point, the implications for future architectures become clear. Any system that intends to avoid the failures observed in today’s digital-asset markets will need to remove, by design, the structural weaknesses highlighted in Parts I–III.

That implies several technical and economic requirements:

No reliance on cross-chain bridges as a core mechanism for value transfer, to avoid wrapped IOUs, bridge exploits, and fragmented liquidity.
No dependence on external oracles for core economic truth, to preserve determinism and minimise trust in off-chain actors.
No reliance on liquidity depth for basic system stability, so that the architecture does not fail simply because free-market liquidity contracts.
No validator concentration risk where a small number of operators can meaningfully control consensus or execution.
Deterministic economic models that do not depend on speculative markets to sustain internal balance.
Self-validating and self-accounting frameworks that minimise dependence on external infrastructure (indexers, relayers, centralised cloud hosting).

These are not promotional criteria; they are structural requirements derived from observing where current systems are failing. Architectures that satisfy them will be better positioned to survive stress and scrutiny, regardless of branding or narrative.

8. Historical Parallels — The Cycle of Re-Centralisation

Every technological revolution begins with decentralisation and ends with concentration, unless its design makes concentration impossible.

History offers several clear parallels:

Early banking was fragmented before central banks emerged as anchors.
Early internet connectivity involved many small providers before backbone carriers and major ISPs dominated traffic.
Early file-sharing was peer-to-peer before streaming platforms centralised distribution.

Digital assets are tracing a similar path. The tools are new, but the underlying dynamics are familiar: innovation → proliferation → consolidation → regulation → redesign.

Decentralisation survives long term only when its architecture leaves little room for centralisation to re-emerge. Otherwise, economic gravity and regulatory pressure do the work.

9. The Road Ahead — Completing the Unravelling

This trilogy has highlighted three major fault lines:

Part I — wrapped assets exposed fragility: bridges and tokenised IOUs introduced artificial value and new systemic risks.
Part II — liquidity illusions distorted valuation: market caps grew while real liquidity thinned and order-book depth declined.
Part III — control has quietly re-centralised: validators, custodians, infrastructure providers, and regulators now shape market outcomes more than protocol ideals.

None of this is speculative. Each observation is grounded in public data, on-chain analytics, or official disclosures. Anyone can verify the underlying evidence.

What comes next?

If digital finance is to evolve beyond these recurring cycles, its foundations must be redesigned around:

deterministic rather than probabilistic economics,
self-sovereign computation and settlement,
value that is not a wrapped representation of something elsewhere,
execution that does not depend on deep speculative liquidity to remain functional,
and decentralisation that emerges from structure, not from branding.

This is not a prediction; it is a recognition of historical pattern. The next systems that succeed will not be the loudest or the most heavily marketed. They will be the ones that are architecturally immune to the weaknesses identified in this series.

The future will not belong to whoever has the biggest market cap. It will belong to whoever has the fewest points of failure.

That, in essence, is the conclusion of The Great Unravelling.

References (selected)

Messari — market structure and exchange-volume integrity reports (2024–2025)
CoinMetrics — network and validator-concentration analytics
Glassnode — on-chain liquidity, holder distribution, and miner-behaviour analytics
Kaiko — market-depth and order-book-liquidity reports
BIS — Quarterly Review publications on crypto-asset systemic risk
IMF — Global Financial Stability Report sections on digital assets
Lido, Coinbase, Binance, and other staking dashboards — Ethereum validator distribution
Bitcoin ETF issuer disclosures — reported AUM and custody structures