OSB Is Buying Back Its Own Stock Below Book. The Margin Outlook Says Whether It Works

The headline reads like bookkeeping. "Transaction in Own Shares" is the boilerplate a listed company files whenever it buys back a few thousand of its own shares, and OSB Group's version lists 120,542 shares picked up over four trading days at around 510 pence each. But behind that routine daily disclosure is a fact worth slowing down for: OSB, the United Kingdom's largest independent buy-to-let lender, is in the middle of a £100 million buyback that cancels its own stock at roughly 12% below what it says those shares are worth on the balance sheet. For a US investor thinking about a UK bank, the question is whether that gap is an opportunity or a sign the market knows something the books don't.

A lender that charges mainstream banks don't

OSB — formerly OneSavings Bank — is a specialist mortgage lender. It writes buy-to-let loans, plus residential, bridging, commercial and semi-commercial mortgages, largely for borrowers the high-street banks "cannot or will not" serve: professional landlords, portfolios held through limited companies, houses in multiple occupation. It funds that lending through its own retail savings brands, Kent Reliance and Charter Savings Bank, rather than relying on wholesale money. That structure is why the stock behaves less like a commodity bank and more like a fee-stable lender with a commodity-adjacent tail: the loan book was £26.3 billion at mid-2026, and the weighted-average loan-to-value on the portfolio sits at a conservative 67%.

The business metrics are not those of a bank in trouble. Return on tangible equity was 13.7% for 2025, eased to 13.3% in the first half of 2026, arrears of three months or more improved to 1.6%, and the loan-loss charge ran at just 12 basis points. The CET1 capital ratio stood at 15.2% — well above the 13% to 13.5% the company targets — with tangible net asset value of 579 pence per share at the end of 2025.

Where the market and the books diverge

Here is the crack the buyback sits on. In August 2026 the shares fell hard — roughly 13% in one day to the bottom of their 52-week range — after OSB cut its net interest margin outlook for 2026. Net interest margin had slipped to 223 basis points in the first half, and management guided the full year down to 215–220, trimming the return on tangible equity target to about 12.5%, because competitive retail deposit funding now costs more than the bank assumed. Margin compression is the one number that genuinely matters to a lender's earnings.

But the market sold off a margin projection, not a broken credit book. The cash flow is still there: first-half profit before tax of £187 million, down 3%, but earnings per share up 3% because the share count keeps shrinking. Credit is stable, capital is high, and the deposits funding the book are sticky. In the persona of an investor who starts from operating cash flow and survival before cheapness, this is the useful observation: the downgrade touched earnings growth, not solvency or asset quality.

That is the backdrop against which the buyback makes sense. Management had already run one £100 million buyback, completing it in March 2026 at an average 527.94 pence per share, and immediately launched a second, running to March 2027 through broker Jefferies, with the shares bought back to be cancelled rather than parked in treasury. Cancellation is the meaningful detail: it permanently shrinks the share count, mechanically lifting per-share book value, earnings, and the dividend that follows. Roughly £70 million of the current programme had been spent by early August, and the daily filings show the machine still turning at about 510 pence against 579 pence of book value. Buy a pound of assets for 88p, cancel the ticket, and the assets left are worth more per share.

The arithmetic of returning capital

The attraction for income and value investors is straightforward. The 2025 total dividend of 35.3 pence — up 5%, with the 2026 interim raised another 5% — works out to roughly a 7% yield on the current price. Add the buyback and the cash being pushed back to shareholders approaches 14% a year before any earnings growth. Over the past 18 months the group has returned £360 million to owners, and more than £1.3 billion since 2019. At about seven times forward earnings and under tangible book, the stock is priced as if that 13%-odd return on equity will not hold.

The honest caveat is that the buyback is a signal, not a guarantee. Buying back stock below book is accretive only if the underlying returns survive, and the return on equity is exactly the line management just shaved. If deposit-funding costs keep running ahead of assumptions — the bank is now assuming retail funding at SONIA plus 30 to 40 basis points, with no normalization within 2026 — the margin and the return on equity stay under pressure, and the discount to book will look less like a bargain and more like the market pre-pricing a lower-return bank. A cheap multiple is not an opportunity until durability is established; here durability rests on one variable.

For a US retail investor, the practical gate to this is a London listing (or the OSBGF over-the-counter line), so it is an international exposure with currency and settlement friction. The thesis OSB's own checkbook is testing comes down to a single question: is the margin cut a one-time funding-cost shock that fades as back-book loans roll off, or a new equilibrium where deposit competition permanently lowers the return on that 26-billion-pound book? The daily "Transaction in Own Shares" filings are management's way of placing a bet on the former at 88 cents on the dollar. The credit book and the capital ratio support that bet; the margin trajectory is what will decide whether it pays off.