SoFi (SOFI): The Bank Charter Is Finally Paying Off

Fifteen million eight hundred thousand people now hold a SoFi account, up 35% from a year earlier, and the company just closed its eleventh consecutive profitable quarter. Neither number was true four years ago. What changed in between was not a product launch anyone remembers; it was a piece of paper. On January 18, 2022, the OCC and the Federal Reserve approved SoFi’s application to become a bank holding company. The acquisition that made it official closed a couple of weeks later, in February 2022. Most coverage at the time treated that approval as a regulatory footnote buried under the SPAC-era noise around the stock. It was not a footnote.

Four years is not long for a bank charter to prove itself. It has been long enough.

Before the charter, SoFi funded the loans it originated mostly through warehouse credit lines borrowed from other banks: expensive and capital-intensive, with a middleman’s margin sitting between SoFi and its own loan book. A bank charter let it fund loans with deposits gathered directly through its own products instead, at a lower cost, with more control over the balance sheet, and with the option to hold loans rather than needing to offload every one immediately to free up capacity. My thesis is straightforward: that funding advantage is now visible in the actual numbers rather than in a narrative about future potential. Deposits and net income both hit records in the June 2026 quarter, and the open question is no longer whether the charter works but whether the market has noticed yet.

The charter turned funding cost into an edge

Deposits at SoFi Bank grew by $5.3 billion in the quarter ended June 30, 2026, reaching $45.5 billion, funded in large part by a high-yield savings product that undercuts what most brick-and-mortar banks pay savers. Every dollar sitting in that deposit base is a dollar SoFi is not borrowing at a worse rate somewhere else, and the effect compounds: more deposits mean cheaper funding, cheaper funding means SoFi can originate loans other lenders would have to pass on, and that flows straight into the profit line rather than getting eaten by financing cost. Total net revenue for the quarter reached $1.22 billion, a record for the company, with adjusted net revenue up 40% year over year. GAAP net income came in at $156.6 million, adjusted EBITDA rose 44% to $358 million at a 30% margin, and this was, again, the eleventh straight quarter of profitability rather than a one-time beat. A single strong quarter is a data point. Eleven in a row is a pattern.

Three businesses, and lending still carries the load

SoFi reports results across three segments: lending, financial services, and a technology platform business licensed to outside banks. Lending is still the one that produces most of the actual profit dollars, built from personal loans, student loan refinancing, and home loans. Financial services gets talked about far less, and I think that is a mistake, because it is where the deposit and member growth actually lives: checking and savings accounts, the SoFi credit card, and the brokerage and robo-advisor products that keep a member’s money inside the ecosystem instead of just passing through it once for a loan. A member who only ever takes out one loan is worth a fraction of what a member who also parks a paycheck and a savings balance at SoFi is worth, and financial services is the segment built specifically to convert the first kind of customer into the second.

What I find more interesting than either of those is what SoFi has built alongside them: a loan platform business that originates loans for outside institutional partners in exchange for fee income, without SoFi having to hold all of that credit risk on its own balance sheet. In March 2026, SoFi announced new agreements in that platform business totaling more than $3.6 billion, building on over $10 billion in commitments it had already secured in 2025. That is a capital-light way to keep growing loan volume without growing the balance-sheet risk that comes with it, and it is the kind of de-risking move that tends to get ignored by a market still pricing SoFi as a single-product lender.

That same profit-over-volume discipline is playing out at PayPal, where a new CEO inherited a business that had already started choosing margin over headline volume before he arrived. SoFi’s version of that trade is a bank charter instead of a metric change, but the logic is identical: growth that does not clear a profitability bar is no longer worth chasing for its own sake.

Charge-offs are falling as the book grows

Fast growth at a lender is supposed to raise one obvious question: is underwriting quietly getting looser to produce it? SoFi originated a record $10.7 billion in personal loans during the June 2026 quarter, and if that volume were being bought with worse credit standards, the loss numbers would already be showing it. They are doing the opposite. The company’s total net charge-off ratio across all loans came in at 1.81% for the quarter, down from 2.12% in the same quarter a year earlier. On personal loans specifically, the all-in annualized charge-off rate, including the effect of loan sales, was about 2.62%; stripped of late-stage delinquency sales, the underlying rate improved by roughly 80 basis points year over year to near 3.7%. Better credit performance alongside record origination volume is the specific combination bears usually bet against, on the theory that fast growth eventually forces underwriting standards to slip. In SoFi’s reported numbers, through the middle of 2026, it has not happened yet.

Rate sensitivity cuts both ways here

Because SoFi now funds itself like a bank, it is also exposed to rate policy like one, which is not a small detail. When funding costs fall, SoFi’s net interest margin tends to widen, and when the Fed moves the other way, that margin compresses, the same mechanic I looked at when a surprise Fed rate hike hit bank stocks broadly and the same tension Bank of America investors are implicitly betting on with every rate decision. SoFi is smaller and newer to the banking business than either of those names, which cuts two ways: less of a legacy cost base to defend, but also less of a cushion if a rate cycle turns against deposit-funded lenders all at once. I read the current trend as favorable, deposit costs have been falling faster than loan yields have compressed, but that is a trend that can reverse inside two or three quarters if the rate path changes.

Metric	Value (Q2 2026 unless noted)	Context
Share price	~$17 (Sept 22, 2026)	trailing P/E roughly 36x
Forward P/E	roughly 22-24x	market cap ~$22 billion
Net revenue	$1.22B, record	adjusted net revenue +40% YoY
GAAP net income	$156.6 million	11th straight profitable quarter
Adjusted EBITDA	$358M, +44% YoY	30% margin
Total deposits	$45.5B	+$5.3B in the quarter
Members	15.8 million	+35% YoY, +1.1M added in Q2
Loan platform agreements	$3.6B+ new (Mar 2026)	on top of $10B+ in 2025

A trailing multiple that still reads like a story stock

Here is where I get less enthusiastic. SoFi trades around $17 a share as I write this, at roughly 36 times trailing earnings and something closer to 22 to 24 times forward estimates depending on which consensus number you use. That is not a bank multiple; regional and money-center banks mostly trade in the single digits to low teens on earnings, because the market treats deposit-funded lenders as slow, cyclical, and capital-heavy. SoFi’s multiple says investors are still pricing member growth and platform optionality the way they would a software company, not a bank. I think that premium is partly earned, member growth of 35% is not something regional banks post, but it also means the stock has very little room for a quarter that merely meets expectations instead of beating them.

Where this bet could still break

My genuine uncertainty is whether SoFi’s member growth rate holds once the easiest converts, people already dissatisfied with a traditional bank, have been signed up. Growth this fast rarely stays this fast forever, and a deceleration toward, say, half the current pace would force a real rerating even if profitability keeps climbing, because so much of today’s multiple assumes the growth line stays steep. The specific condition that would prove me wrong on the bullish case: if credit quality in the personal loan book deteriorates in a slowing economy at the same time member growth cools, the two effects would compound rather than offset, and a stock priced for a software company’s growth would need to reprice toward a bank’s multiple all at once. I don’t see that combination in the numbers reported so far. It is the scenario I would watch for before adding to a position rather than after.

The next number I would wait for is member growth in the December 2026 quarter. Anything still above 30% year over year would say this is a structural shift in how people choose a primary bank, not a pandemic-era fintech boom running on fumes. A drop toward the high teens would say the easy growth phase is ending, and the multiple should start acting like it.

Analysis and opinion only, not investment advice. Figures come from SoFi’s Q2 2026 earnings release and its filings on SEC EDGAR, and from its investor relations site; valuation multiples are approximate and were checked on September 22, 2026.