Britain hands the Bank of England an innovation duty that stability still outranks

A secondary objective is a duty to consider, not a duty to deliver, and the government has been unusually plain about the escape hatch: the Bank does not have to support an innovation where doing so could undermine financial stability ^[5]. What is left with teeth is the reporting line, an annual account to Parliament of work done under the objective ^[6], which turns a run of private supervisory judgements into a document someone can be asked about in committee. Deputy Governor Sarah Breeden's welcome, framed as support for innovation "without compromising on financial stability" ^[8], reads like someone who has read the carve-out and is content with it.

The arithmetic that actually moves money is June's. Replacing a £20,000 individual holding cap and a £10m business cap with a single £40bn issuance limit for each systemic stablecoin ^[9] moved the binding constraint from the user's wallet to the issuer's balance sheet, and the scale of that move is the point: £40bn is 2 million people at the old individual maximum ^[19], or 4,000 firms at the old business one ^[20]. The limit applies only to coins HM Treasury formally recognises as systemically important ^[10], with everything else supervised by the FCA ^[12]; the cap only takes effect once an issuer reaches that status, not before it.

Then the reserve rule, which is where a sterling stablecoin business is priced. Up to 70% of backing may sit in short-term gilts, and the remaining 30% generally sits as non-interest-bearing deposits at the central bank ^[11]: at the cap, that is £28bn earning the front end and £12bn earning nothing ^[18], for a gross reserve yield of at most seven-tenths of the short-gilt yield before a single employee is paid ^[21]. The top line is set by a supervisory ratio, and the only variables the issuer controls are scale and cost.

Sequencing is the third constraint and the tightest. The authorisation window runs 151 days ^[22], existing anti-money-laundering registrations do not roll over into full authorisations ^[15], firms that miss the window may lose transitional arrangements ^[16], and roughly eight months separate the last permitted application from the mandatory start of the regime ^[23]. Legal and compliance budget committed to that file in that window is budget not committed to a rival jurisdiction's file in the same months.

This is probably wrong, but I read the innovation objective as domicile policy: City Minister Lucy Rigby's line about tokenisation and distributed ledger technology having "the potential to transform financial markets" ^[7] is what a Treasury says when it wants the issuance booked in London. The counter-thesis is the cheaper one, that a subordinate duty carrying a stability veto and a 30% cash drag amounts to a permission slip without an incentive attached, and reserves will be domiciled wherever they earn more. Two outcomes would separate them. If the amendments come out of a Parliament that may approve, reject or modify them ^[17] with the carve-out intact, and the first annual report catalogues consultations rather than approvals, the permission-slip reading wins; if Treasury recognises a systemic sterling coin and the £40bn cap starts to bite on a real book, the domicile reading does.