VersaBank: The Growth Is Real, the Margin Isn't — a ~30x Bank Stock Priced for Perfection

VersaBank isn't the bank you picture when you read the word. It has no branches to speak of and doesn't court depositors on street corners. Its engine is the Structured Receivable Program (SRP): it buys the point-of-sale loans that other lenders originate — a store-brand credit card for a contractor's heating-and-cooling install, a vendor's equipment-finance contract — buying the receivables at a discount and funding them on its own balance sheet. Put simply, it is a lender to other lenders, built on the idea that financing thin-margin, low-risk receivables through a cost-light digital operation adds up.

That is the story investors have been paying for, and after the bank's fiscal third-quarter report on September 3 the stock jumped about 9% to $22.41, within reach of its 52-week high. The headline that moved it: total assets crossed C$7 billion. Underneath that single line is a quarter that genuinely supports the bull case — and a margin that doesn't. The gap between the two is the whole question here.

The quarter that made the growth real

The numbers, all in Canadian dollars, are record and accelerating. Revenue rose 23% from a year earlier to C$38.8 million, helped above all by the U.S. business, where revenue climbed 199% and net income surged 803% to C$3.9 million. Adjusted (core) net income rose 27% to C$12.3 million, or C$0.38 a share. Total assets hit C$6.9 billion at quarter-end, up 26% year over year, and ballooned to about C$7.2 billion in the weeks after — the source of that $7-billion headline.

The U.S. SRP book is the reason this is now a growth story rather than a slow lender. It stood at US$793 million at quarter-end, and management laid out a fiscal 2027 target of adding at least US$3 billion in new U.S. fundings — more than C$4 billion, or roughly a 60% increase in credit assets. A signed partnership with an ECN Capital subsidiary is expected to contribute at least US$300 million a year, with a path beyond US$500 million, and the bank just launched "Real-Time SRP," an AI-enabled product that funds individual loans in hours instead of warehousing them for days.

By the reported numbers, the machine works. The reported net-income figure of C$10.1 million (up 53%) flatters it — that total is after C$3.1 million of one-time reorganization and software-write-off costs, and the adjusted figure that strips those out is the cleaner read. Fine. The acceleration itself is not the problem.

Why the growth is expensive to produce

Here is the catch an investor should pause on: this bank grows by stacking more low-yielding loans onto a balance sheet that is funded largely with brokered and time deposits, and that funding machine is getting more expensive just as the loan mix gets cheaper. The net interest margin — the spread between what the loans earn and what the deposits cost, the fundamental profit machine of any bank — fell to 2.44% on credit assets, down 11 basis points from a year ago and 27 basis points just from the prior quarter. Management says the margin should recover to roughly 2.3% in the current quarter; the recovery matters, because at 2.4% this is already an extraordinarily thin spread for a lender.

Two things follow. First, the 27% growth in core earnings is real but leverage-funded: with a margin near 2.4%, even a huge asset base yields modest profit, and returns on equity run in the low teens at best. Second, the quarter's clean numbers went slightly the wrong way relative to expectations — adjusted EPS of C$0.38 came in about a cent below consensus, and revenue missed by roughly C$1.25 million.

Now weigh that against the price. The stock is up about 50% year to date and trades at roughly 33 times trailing and 30 times forward earnings, at about 1.8 times book value, with a dividend yield near 0.3% — a rounding error, not an income case. For scale: a large North American bank routinely trades near 10 to 12 times earnings. VersaBank is instead priced like a high-growth fintech — its ~33x trailing multiple sits essentially alongside SoFi's ~37x. The market is treating this as a software-like growth story, not a bank.

What would change the call

That is the crux: the growth narrative is evidenced, but through the margin lens the market has paid for a lot of it. This reads less as an obvious buy and more as a "watch the proof" situation, where the next two to four quarters decide whether the multiple is earned.

The proof points sit right on the surface of this quarter. First, does the margin actually recover to about 2.3% as liquidity normalizes and deposit costs ease — or does the SRP mix keep squeezing it? Second, does the pipeline convert: ECN Capital and Real-Time SRP scaling, alongside the reported US$720 million of year-to-date fundings, need to beat the ~US$3 billion fiscal 2027 target with revenue that outpaces the balance-sheet growth. Third, does the noise clear — the C$4 million of additional reorganization costs management expects in the current quarter, the completion of the shift to a U.S. bank holding company (scheduled for around the end of October, after a shareholder vote on September 16), and the eventual exit of the money-losing cybersecurity unit.

The one structural number to keep watching is the capital side: management says it has no immediate need to raise equity if the risk weighting on SRP assets drops to 20% under a Basel III change targeted for mid-2027 — relief that would let it fund far more growth off the same capital. That, more than any single quarter, is what could make the current multiple look cheap in hindsight.

None of this calls the company a bad business; the SRP model is real, growing, and demonstrably profitable. But separating the company from the stock is exactly what the multiple demands you do. At ~30 times forward earnings for a lender whose profit machine narrowed this quarter, the market has already paid for the success. The disciplined move is not to chase the headline asset milestone, but to wait for the margin and the funding pipeline to prove they can carry a ~30x price — and to treat a slip in either as a reason the rich valuation is borrowed against earnings it hasn't yet produced.