Why Decentralisation Is a Bug, Not a Feature, When You're Trying to Build Money
There is a particular kind of engineering hubris that shows up when a technology is powerful enough to feel like it should solve everything. The internet did it with publishing, social connection, and eventually democracy itself. Machine learning is doing it now with reasoning. And for the better part of fifteen years, cryptographic consensus mechanisms have been doing it with money.
The argument always sounds clean in a whitepaper. Money is inefficient because it relies on trusted intermediaries. Trusted intermediaries are corruptible. Therefore, remove the intermediaries. Ship the trust directly into the protocol. Problem solved.
But this framing contains a category error that no amount of elegant cryptography can fix.
Money Is Not a Protocol. It Is a Social Fact.
When you hold a dollar, or a euro, or any sovereign-backed currency, what you are actually holding is a claim. Not on gold, not on a commodity, but on the collective productive capacity of an economy, backed by a state's legal and coercive apparatus. The value is not intrinsic. It is institutionalised.
This is what economists mean when they describe money as a "unit of account" and a "store of value." Those are not just functional properties. They are social properties. They require that a sufficiently large and stable group of people agree, continuously, that the thing holds meaning. Governments enforce this agreement through legal tender laws, tax obligations, and the monopoly on legitimate currency issuance. Central banks manage it through monetary policy.
A decentralised asset, by design, cannot do any of that. It has no issuer, no backstop, no mechanism to absorb economic shocks, and no authority to compel its use. These are not implementation details waiting to be solved in a future protocol upgrade. They are structural absences that follow directly from what decentralisation means.
The Narrative Drift Problem
The European Central Bank has documented something interesting in its analysis of crypto-asset markets: Bitcoin and its successors have cycled through distinct identity claims over time. Digital cash. Inflation hedge. Store of value. Base layer for DeFi. Gateway to Web3.
From a product perspective, you might read this as iteration. From a monetary theory perspective, it looks like something else: an asset that has never found a stable function, constantly re-narrating itself to attract the next cohort of buyers.
This matters because money derives much of its power from stability of expectation. The dollar is useful partly because everyone expects it to still be the dollar tomorrow. When an asset's core value proposition shifts every two to three years depending on macro conditions and community sentiment, it is not behaving like money. It is behaving like a speculative instrument that borrows monetary vocabulary when convenient.
CBDCs Are a Different Problem Entirely
Here is where the conversation usually collapses into noise. Critics of crypto often get lumped in with critics of Central Bank Digital Currencies, and proponents of CBDCs get accused of defending the traditional financial system against innovation. But these are separate debates.
A CBDC does not attempt to replace sovereign money. It digitises sovereign money. The distinction is architecturally significant.
Consider a rough mental model of what that means:
// Crypto token model
Token.transfer(from: alice, to: bob, amount: 1.0)
// Settlement is final. No issuer. No recourse. No policy lever.
// CBDC model (simplified)
DigitalFiat.transfer(from: alice, to: bob, amount: 1.0, authorisedBy: centralBank)
// Issuer maintains policy control. Programmable compliance hooks.
// The money is still sovereign money, just natively digital.
The engineering challenges in these two cases are genuinely different. A CBDC requires building privacy-preserving architectures that still satisfy AML obligations, designing programmable money without creating surveillance infrastructure, and handling massive transaction throughput at a national scale. These are hard, interesting, and worth solving.
Crypto's challenge, by contrast, is not primarily technical. You can solve throughput with better consensus mechanisms. You can solve interoperability with bridges and standards. But you cannot solve the absence of a social contract with a smarter contract.
What Actually Gets Built in the Meantime
None of this means the cryptographic primitives are useless. Zero-knowledge proofs, threshold signatures, and decentralised key management are genuinely powerful tools. They are finding real homes in identity systems, supply chain verification, and privacy-preserving computation. The infrastructure that emerged from fifteen years of crypto development will outlast the monetary claims made on its behalf.
The mistake was always framing the technology as a monetary replacement rather than a cryptographic primitive. Hammers are not failed cars.
The practical takeaway for engineers working anywhere near this space: separate the cryptographic stack from the monetary claims being made on top of it. The former is solid and worth building on. The latter is a sociological and political problem that a distributed ledger cannot resolve, regardless of how elegant the consensus algorithm is.
Top comments (0)