Capital One's acquisition of Discover Financial Services

The deal in one paragraph

On February 19, 2024, Capital One announced it would acquire Discover Financial Services in an all stock deal valued at $35.3 billion, the largest bank acquisition to clear U.S. regulators since the 2008 financial crisis. Discover shareholders received 1.0192 Capital One shares for every share they held, worth about a 26.6% premium to Discover's last unaffected closing price of $110.49.

Once closed, Capital One shareholders owned roughly 60% of the combined company and Discover shareholders owned about 40%. The deal took fifteen months to close, not because the transaction itself was complicated, but because it needed sign off from the Federal Reserve, the OCC, and, informally, the Justice Department's antitrust division, against vocal opposition from progressive lawmakers.

It closed on May 18, 2025. What makes this one worth knowing cold: it is a case study in how bank mergers actually get approved, and in why a card company would rather own its payments network than rent one from Visa and Mastercard.

Why Capital One wanted Discover

Discover is a credit card issuer, the same business Capital One is already in, but it is also one of only four global payment networks, alongside Visa, Mastercard, and American Express. Visa and Mastercard are pure networks: they do not issue cards or lend money, they just route transactions between banks and merchants for a fee.

American Express and Discover are both issuer and network at once, touching a transaction twice, once as the bank extending credit, once as the network routing the payment.

Capital One had always been a Visa and Mastercard issuer, meaning some economics of every swipe flowed to whichever network processed it, not to Capital One. Buying Discover handed Capital One a functioning global network, roughly 70 million merchant acceptance points across more than 200 countries, and the ability to route its own volume over rails it owns rather than rents. That is vertical integration in its cleanest form: acquiring a supplier instead of continuing to pay it.

Scale reinforced the logic. Combined, Capital One and Discover control about 19% of the $1.3 trillion U.S. revolving consumer loan market, more than JPMorgan Chase or Citigroup individually, making the combined firm the largest credit card lender in the country by balances. Discover also brought a large, low cost deposit base and its own card brand, kept running alongside Capital One's.

Timing helped too: Discover had spent 2023 working through a self disclosed problem, having misclassified certain merchant accounts into a higher pricing tier for roughly sixteen years, an error regulators estimated cost merchants close to $1 billion and that drew a September 2023 consent order for consumer protection violations.

A distressed, cheaper Discover with a real network attached was a rare asset to find at all, let alone one buyable in stock.

Structure and terms

Two things about the structure matter for an interview. First, it is stock, not cash: Capital One issued new shares rather than raising debt, so the accretion or dilution math runs through the exchange ratio and each company's standalone earnings, not financing cost.

Second, at a 60/40 split, this reads close to a merger of equals in ownership terms even though Capital One is the acquirer and Discover disappears as a listed company. Large deal, done in stock, sub-30% premium, target shareholders keeping a big slice of the upside: that pattern recurs across bank M&A, because banks are capital constrained and cash purchases eat into regulatory capital ratios.

The regulatory fight (Fed, OCC, and the political scrutiny)

Bank mergers do not close the way ordinary corporate mergers do. Beyond the antitrust review any large deal faces, acquiring a bank holding company needs affirmative approval from the Federal Reserve, and, because Discover Bank is a nationally chartered bank, from the OCC too. Both must find the deal serves the public interest, not merely that it does not lessen competition, and both take public comment before ruling.

That comment period is where this deal became political. Senator Elizabeth Warren, joined by Representative Maxine Waters and other progressive lawmakers, wrote to regulators within days of the announcement asking them to block the deal, arguing it would push Capital One's share of the subprime and near prime credit card market above 30% and raise the Herfindahl Hirschman Index, a standard concentration measure, by roughly 400 points.

Comment volume ran heavy enough that the Federal Reserve held a public hearing on the deal in July 2024 rather than deciding on the written record alone.

The Justice Department's antitrust division reviews bank mergers informally without formally clearing the deal itself. An interim DOJ assessment prepared under the outgoing administration in January 2025 reportedly flagged concerns, including that Capital One could use Discover's debit network to sidestep interchange fee caps on large banks.

New antitrust leadership concluded in April 2025 there was not sufficient evidence to challenge the deal, clearing the way for the banking regulators to act.

The Fed and OCC approved the merger on April 18, 2025. The OCC's approval was conditional: Capital One had to submit, and get supervisory non-objection on, a plan addressing the root causes of Discover Bank's outstanding enforcement matters, meaning the 2023 misclassification fallout had to be formally remediated, not just disclosed.

Capital One also rolled out a five year, $265 billion community benefits plan, a commitment banks often pair with a large merger application. Warren was not satisfied; after approval she pressed the Justice Department to sue and unwind the deal. No suit followed, and it closed as approved.

How it played out

Fifteen months from signing to close is long next to a typical acquisition outside banking, but in line with other large bank mergers, most of which also cleared only after hearings and extended review. The deal never needed a court fight, a divestiture, or a restructured transaction, unlike headline antitrust battles in other industries.

If it comes up in your interview

Here is a version that runs 60 to 90 seconds. "Capital One announced an all stock acquisition of Discover in February 2024 for about $35.3 billion, at a roughly 27% premium, with Discover holders ending up owning 40% of the combined company.

The logic is vertical integration: Discover is one of four global payment networks, and owning it lets Capital One stop paying Visa and Mastercard on its own volume and instead use rails it owns, the model American Express already runs. It also made Capital One the largest credit card lender in the country by loan balances. The hard part was regulatory.

Bank mergers need Fed and OCC approval, not just antitrust clearance, and this one drew public opposition from Senator Warren over market concentration, plus a Discover specific issue, a multiyear merchant fee misclassification problem still being remediated during review.

Both regulators approved it in April 2025, with the OCC conditioning approval on a cleanup plan for Discover's problems, and the deal closed in May 2025, about fifteen months after signing."

Likely follow ups:

"Why pay in stock instead of cash?" Stock avoids raising debt or drawing down capital regulators want banks holding against their loan book, and lets Discover's shareholders share in the combined company's upside, which makes the deal easier to sell to Discover's board as a partnership rather than a sale.

"What is the difference between a card network and a card issuer?" The issuer lends the customer money and holds the credit risk, like Capital One. The network, like Visa or Mastercard, routes the transaction between merchant and cardholder banks for a fee. Discover and American Express are unusual because they do both jobs at once.

"Why does a 60/40 stock split matter for accretion and dilution?" With no cash or new debt, the math runs through the exchange ratio: whether Discover's earnings, converted at 1.0192 shares to 1, add more to combined earnings per share than the new shares dilute existing holders. A deal can be accretive even at a premium if the target's earnings yield exceeds the acquirer's implied cost of equity.

What this deal teaches

Three things carry over into other interviews. First, bank M&A runs on a different clock than ordinary M&A: a strategic buyer of a bank needs the Federal Reserve and, if a national charter is involved, the OCC, on top of any antitrust review, and both can hold hearings, take comment, and attach conditions, the way the OCC tied approval here to a Discover remediation plan. If asked what makes a bank deal harder to close than a normal acquisition, this is the answer.

Second, vertical integration into a payments network is a pattern worth having ready for any strategy question: a company decides a fee paid to a supplier, here the fee Capital One paid Visa and Mastercard, is worth eliminating by buying the supplier outright, provided the acquired capability is durable enough to matter. American Express is the existing example, Discover the newest.

Third, this is a clean exercise in all stock deal mechanics. When a target's shareholders take stock instead of cash, accretion or dilution flow through the exchange ratio rather than financing cost, and the resulting ownership split, 60/40 here, shows how much upside the seller kept instead of cashing out. Any interviewer testing whether you understand stock deals can use this one: work through what a 1.0192 exchange ratio implies about relative valuation, and you have demonstrated the concept.