Overview

Float

Documentation, v2.0.

The published essay, Why the biggest companies on earth can't be tokenized, and how Float lists them anyway, is still the argument. Nothing in the thesis has moved: 1,880 of the world's top 2,000 companies have no token on Solana, the reason is plumbing rather than demand, and the answer is to cross the wall once and hold the shares on the other side.

What has moved is the machine. The essay described a market that waited at a $5,000 milestone to open and a second at $10,000 to become spot. Building it taught us that both numbers were guesses at things the system can measure directly, and that the interesting problem was not the milestones but what happens between them. So we tweaked it. This is Float 2.0, and the differences are worth stating before anything else.

What Float is

Float lists the companies the tokenization wave skipped. Of the world's top 2,000 companies, 1,880 have no token on Solana. Nintendo, Tencent, Nestle, LVMH, SoftBank and Hermes are all in that group. The reason is plumbing rather than demand: buying a Tokyo or Hong Kong listed company from a US account means a foreign account, an FX leg, treaty paperwork and a custodian in the home market, and no consumer app is going to build that once per country.

Float crosses that wall once, holds the shares on the other side, and issues a token against them.

One fSHARE is one ordinary share of the underlying company, held in custody and owed to the holder. Not a basket, not a tracker, not an index: the specific security, in the specific quantity, in a specific account. The claim is redeemable, and selling an fSHARE back to the Desk burns it and releases the share behind it.

It carries no voting rights, which is true of every tokenized equity and is worth saying rather than implying.

How it works

The Desk is one vault of USDG standing on both sides of every listing. It quotes continuously against a reference price, which is what makes a Japanese company tradeable at three in the morning when Tokyo is shut and no pool exists to trade against:

buy  = mark * (1 + spread + impact + txFee)
sell = mark * (1 - spread - impact - txFee)

The spread is 30 basis points while the home exchange is open and 200 when it is closed. That widening is retained by the vault in full, and it is what pays for quoting a company whose own exchange is shut.

Impact scales with how far a trade pushes the book away from balance, so size pays for the risk it creates.

The reserve is five public numbers on ReserveBook.sol: what is outstanding, what custody actually holds, what has been ordered and paid for but not yet settled, and the ceilings those imply. Nothing about the backing exists off chain except the broker account itself. A market that cannot prove its cover becomes settle-only, which is the universal failure state here: existing positions can be closed, nothing new can be issued.

The launch funds the custody purchase from the first dollar rather than waiting at a milestone, thickening the market with every dollar after on a decaying ramp. A launched token graduates into two real Uniswap v4 pools, USDG to fSHARE to token, with 23.4% of supply seeded into the pool.

What an ADR is

The promise is one token, one ordinary share. The thing sitting in a US brokerage account is frequently not an ordinary share, and pretending otherwise is how a reserve quietly becomes fractional.

An American Depositary Receipt is a US-listed certificate issued by a depositary bank against home-market shares that bank holds through a local custodian. It trades in dollars, settles US-style, and clears through US infrastructure. That is its whole purpose: it is a wrapper that lets American plumbing hold a foreign asset.

It is a receipt, not the share. The depositary is a real counterparty in the chain. The receipt can be cancelled, the programme can be terminated, and the ratio can be changed. Sponsored means the company set the programme up itself, so there is one programme and real disclosure. Unsponsored means a depositary created it without the company's involvement, which is why several banks can run competing programmes on the same company and why the terms belong to the depositary. Most of Float's target names trade unsponsored on the US OTC market, Nintendo and Tencent among them.

How Float connects to them

One ADR is almost never one share. The depositary picks a ratio to put the receipt in a comfortable dollar price range, and it differs by name:

Get that number wrong and the reserve is wrong by exactly that factor. Ship 1.0 where the truth is 0.2 and the market runs a 5x fractional reserve while every on-chain view agrees with itself, because every view is derived from the same wrong constant.

We learned it the expensive way. A published fact sheet gave the NTDOY ratio as 1:8. Float derived it from live prices instead and got 1:0.25, a 32x discrepancy against the document, and the live price relationship was the correct one. So the rule is explicit: verify a ratio from prices, never from a fact sheet, and check it continuously, because a depositary can change the ratio on an unsponsored programme without asking anyone.

The custodian reports the units on its statement, whatever those units are, and the contract converts:

setCustodyUnit(assetId, sharesPerUnit, kind)   // 1e18 = one unit is one ordinary share
                                               // 2.5e17 = one unit is a quarter of a share
sharesHeld = unitsHeld * sharesPerUnit / 1e18

custodyKind is stored as a string, "ordinary" or "ADR" or "GDR", so the reserve page says plainly what is held rather than implying the ordinary share everywhere. A ratio change restates the existing position, so the book never keeps crediting an old conversion against shares it already holds.

Section 5 is the full mechanical account of how one token becomes one share, and section 6 is the whole of the ratio problem.

What changed since the essay

Everything else the essay promised is built and is the same: one real share per token, held in custody, proven on chain; a dealer quoting continuously against a live reference so the token trades around the clock; no Korean account, no FX leg, no treaty forms, no index classification standing between a person and a company.

Section 5 is the important one. It is the full mechanical account of how one token comes to be one share, and section 6 explains ADRs, because the ratio between what custody holds and what the token promises is where this gets genuinely difficult.