Grab Holdings is negotiating to buy a majority stake in Atome Financial in a deal that would value the Singapore buy-now-pay-later lender at more than $2 billion, according to people familiar with the matter, a price that hinges on whether the buyer can keep the seller's bank syndicate intact.
"Grab is buying a loan book and a licence, not a brand," said Tom Brennan, an analyst who covered deal flow at Dealogic before moving into journalism. "The multiple only holds if the syndicate that funds Atome's receivables rolls over at the same spread after control changes."
The talks, first reported by Bloomberg on Sept. 10, are ongoing and no final agreement has been reached, the people said, asking not to be identified because the discussions are private. Atome is the consumer-finance arm of Advance Intelligence Group, whose investors include SoftBank Vision Fund II, Warburg Pincus, Northstar Group and Singapore state investor EDBI. Grab, Atome and SoftBank have not commented publicly.
Atome's economics explain the interest. The unit posted operating revenue of $236 million in 2024 with gross merchandise value above $2 billion, its first full year of profit, and Advance Intelligence has guided to roughly $470 million of revenue for 2025 — an 80% increase. The parent has projected annualized net income of about $800 million and annualized GMV above $6 billion for 2026. Atome Card issuance in the Philippines alone has passed 3 million, and the brand operates in Singapore, Malaysia and the Philippines.
The funding stack is the part that matters to a buyer. In January, Atome expanded a syndicated loan to $345 million from $200 million in 2024, with HSBC as structuring arranger, DBS leading, and Mitsui Sumitomo, Cathay United Bank, Fubon and SPD Bank participating. In May, Advance Intelligence subscribed to about 38 million shares at $3.90 each, injecting $149 million. Total group funding has passed $700 million, including an $80 million round in 2023 and a 2021 raise that pushed the parent's valuation above $2 billion.
Grab's credit book is already the constraint
Grab's own financial-services line is still losing money, which makes the price tag the central question rather than the strategic logic. Adjusted EBITDA losses in the segment were $17 million in the first quarter of 2026, against a loan balance of $1.44 billion — nearly double the $625 million a year earlier. Management has said it aims to break even in the second half of the year.
That trajectory cuts both ways. A larger loan book is the fastest route to the interest-spread revenue Grab wants; it is also the fastest route to a bigger provisioning line if regional credit quality deteriorates. Grab shares closed at $3.04 on Sept. 9, down 6.46% on the session, giving the company a market capitalization of about $12.4 billion after a 39% decline this year. The stock trades on Nasdaq under the ticker GRAB.
The deal would be Grab's third sizeable acquisition this year. It bought U.S. advisory platform Stash Financial for $425 million and took over Delivery Hero's Foodpanda Taiwan operations for $600 million. It has also agreed to acquire Validus' Singapore business from GXS Bank, and has held intermittent merger talks with Indonesia's GoTo — a process that has repeatedly stalled over Indonesian state-shareholder objections.
What the $2 billion anchor depends on
The reported valuation is a private-market marker, not a settled price. Advance Intelligence's last round valued the group above $2 billion in 2021, and the current talks apply a similar figure to the consumer-finance unit alone — a step up that assumes Atome's profitability holds and that its wholesale funding costs stay near current levels. If SoftBank and other existing shareholders sell a majority stake, the reference point for the business shifts from private equity marks to the credit quality of a consolidated Grab balance sheet.
Regulatory review is the second variable. A controlling stake in a licensed lender operating across Singapore, Malaysia, the Philippines and Indonesia — Atome's Indonesian digital-loan business trades as Kredit Pintar — would require approvals in each jurisdiction, and consumer-credit concentration would draw scrutiny of how much of the region's installment lending sits inside one super-app. The deal structure, including whether Grab pays cash or stock and at what premium to the last round, has not been disclosed.
The strategic logic runs both directions. Grab gets ready-made risk models, merchant fee income and a revolving loan portfolio rather than building a BNPL brand from scratch; Atome gets access to Grab's ride-hailing, delivery and wallet users, shifting customer acquisition away from e-commerce advertising. Comparable moves include Sea Group's use of Shopee traffic to support SPayLater and Ant Group's conversion of Alipay scenarios into Huabei credit.
For Grab holders, the near-term question is dilution and capital deployment against a segment that has not yet broken even. For the region's independent BNPL lenders, a completed deal would remove one of the few scaled, profitable standalone players from the field and leave the remainder competing for funding against a listed balance sheet. The next hard data point is Grab's third-quarter results, when management would have to quantify any financing commitment and the timeline for closing.
This article is for informational purposes only and does not constitute investment advice.
