Mt. Gox, FTX, and What Survives

Late 2022 will be remembered, in retrospect, as the second crypto collapse. The first — Mt. Gox in 2014 — was contained to early adopters and exchange counterparties; most institutional capital had not yet arrived to be incinerated. The second is different. FTX’s bankruptcy in November takes down a market-making layer that had become structurally entwined with the rest of the ecosystem. The question is not whether crypto survives the collapse; that question is now decisively settled. It does not, in the form most institutional allocators were holding it. The harder and more useful question is what does survive.

Two things crypto has been

The crypto ecosystem of 2017–2021 was always two things layered on top of each other. The first was a speculative asset class — coins, tokens, NFTs, leverage — that absorbed enormous flows and produced the headline returns and the headline collapses. The second was an infrastructure thesis: that decentralised settlement, programmable money, and verifiable digital ownership were durable primitives, regardless of which speculative wrapper rode on top of them at any given moment.

The collapses we are watching this year — Terra/Luna, Three Arrows, Celsius, BlockFi, FTX — are exclusively a story of the first layer. The infrastructure layer has not collapsed. Bitcoin transactions clear. Ethereum completed the Merge in September with no observable disruption. Stablecoin float — the working capital of crypto’s actual commerce — has compressed but not broken. Layer-2 scaling solutions continue to ship. The distinction between speculation and infrastructure has rarely been clearer.

What 2014 taught

We took our first foundational positions in Bitcoin in 2014, in the aftermath of Mt. Gox. The view then was simple: a payment-and-settlement primitive resistant to single points of failure has structural value, and the collapse of one exchange does not tell you anything material about the primitive itself. That distinction held. Bitcoin and Ethereum compounded — not in a straight line, with several intermediate drawdowns — but compound they did. The speculative wrappers around them did not.

The same distinction holds in 2022. Most of what is collapsing this autumn is the wrapper. The primitive layer is doing what it was designed to do.

Where the infrastructure layer points

Three primitives look durable beyond the current cycle.

Stablecoins. Roughly $130B in stablecoin float currently moves cross-border value at speeds and costs traditional payment rails cannot match. The total addressable market is not crypto speculation; it is global remittance and merchant settlement. The category survives the regulatory tightening that is now arriving; in some jurisdictions it accelerates.

Programmable money. Smart contracts as a settlement substrate — for derivatives, escrow, recurring obligations — are deployable today. The institutional adoption curve is slow but the architectural argument is unchanged. The category does not depend on token prices.

Settlement layers. Bitcoin and Ethereum, post-Merge, function as final-settlement layers that increasingly resemble what their original architects described. The compound value of being a credibly neutral settlement primitive across an increasingly fragmented financial system rises, not falls, as that fragmentation continues.

Reading

What 2022 is teaching is the same lesson 2001 taught about the internet: the speculative cycle is not the technology cycle, and conflating them generates poor capital allocation in both directions. Most of what is being written off this winter was always going to be written off. Most of what is quietly continuing to be built — the actual settlement infrastructure of decentralised systems — will look in five years much as the surviving infrastructure of the dot-com era looked in 2005.

The question for capital is not whether crypto recovers. It is which parts were ever the actual asset.