Synchrony Q3 2026 Earnings: Record $49.7B Volume, Net Income Falls 17% to $892M

On October 20, 2026, Synchrony Financial reported third-quarter net earnings of $892 million ($2.68 per diluted share), down 17% from $1.08 billion a year earlier — in the same quarter purchase volume hit a record $49.7 billion (+8%) and loan receivables grew 3% to $103.3 billion. The top of the funnel has never been stronger; the bottom line shrank because Synchrony shared the gains with its retail partners, rebuilt reserves against a prior-year release, and spent more to run the franchise — then handed $1.2 billion back to shareholders anyway. Our new Synchrony ledger models all of it — two annuals plus this quarter — in plain-text double entry.

The Headline Numbers​

Every figure below comes from the Form 10-Q and the earnings release (Exhibit 99.1 with full financial tables), which agree line for line:

Three things stand out. First, net interest income did its job: higher average loan receivables plus lower funding costs (deposit interest fell $45 million even as deposits grew $3.1 billion) lifted NII $89 million. Second, everything between NII and the bottom line moved the wrong way at once — partner payments, credit costs, loyalty costs and operating expenses all rose faster than revenue. Third, the share count did quiet work: net income fell 17% but EPS fell only 6%, because buybacks shrank shares outstanding 10% in nine months. The quarter is a course in where a lender's profit actually leaks.

Revenue Deep Dive​

Synchrony sells through five platforms, and the release gives purchase-volume and loan growth for each — a rare full split:

The mix tells the story the totals hide. Diversified & Value — the Walmart, Sam's Club and gas-station end of the book — carried both volume (+13% on partner expansion and higher gas prices) and loans (+8%). Home & Auto volume jumped 9% on new programs and furniture retailers, yet its loans grew only 2%: higher payment rates keep converting volume into interchange rather than balances. Lifestyle is the extreme case — volume +6%, loans −1%. Across platforms, interest and fees on loans grew only flat to 4% despite 8% more volume, partly because late-fee incidence fell. That is the private-label bargain in one line: the partners' customers spent, and Synchrony kept the smaller share as interest while the RSA line kept the partner share.

Management's demand language backs the volume print. CEO Brian Doubles called out "broad-based purchase volume growth across all five of our sales platforms" and said "spend reached a record third quarter level and increased across credit cohorts and generations, with the highest growth once again coming from our Generation Z and Millennial customers." Doubledigit out-of-partner growth in both discretionary and non-discretionary spend, "led by momentum in electronics, entertainment, gas stations and discount stores," confirms the strength is breadth, not one partner. What the release does not contain is any pricing, supply or cycle language — this is a lender's quarter, and the forward-looking sentences are all about credit quality and capital, not selling prices.

The Margin Story​

For a card lender the margin stack is net interest margin, the RSA share, credit cost and the efficiency ratio. Five quarters of the moving parts:

Net interest margin slipped 12 basis points year over year to 15.50% — still roughly double a money-center card margin, but compressing as lower late fees and mix dilute yield. The efficiency ratio jumped 330 basis points to 35.9%, mechanically: other expense (+9.9% on operational losses, technology investment and marketing) grew against revenue net of RSA that barely moved. Return on assets fell 70 basis points to 2.9% and return on equity 410 basis points to 21.0% — both still above the levels most banks report in good years, which says more about the starting point than the quarter.

Credit is the offsetting monochrome-to-color story. Net charge-offs fell to a "seasonal low" of 4.79% of average loans, down 37 basis points year over year; 30-day delinquencies eased 5 basis points to 4.34% and 90-day 6 basis points to 2.06%. The allowance coverage ratio drifted down to 10.01% from 10.35% a year ago. CFO Brian Wenzel tied the two halves together: "Credit performance remained strong during the quarter, with delinquency trends consistent with typical seasonality and net charge-offs declining to a seasonal low. The resulting improvement in program performance was shared through the RSA." Read that twice — better credit raised the partner payment. That is the flywheel and the treadmill of the RSA model in a single sentence.

The One Big Question: Where Did the Record Quarter Go?​

Net earnings fell $185 million while volume set a record. The bridge is exact and worth walking slowly:

Start with the $89 million of NII progress and watch it get spent three times over. The RSA took $72 million of it — program performance shared with partners, up 7%. The provision took $120 million more, but look inside that number: it is a $182 million reserve swing (a $30 million build this quarter versus a $152 million release last year) partly offset by $62 million of lower net charge-offs. Underlying credit improved; the accounting comparison worsened. Other expense took another $123 million, led by a $74 million jump in the opaque "other" line (operational losses) plus technology and marketing. Only taxes helped, $63 million lower on lower pre-tax income at a flat 24.5% rate.

The peer contrast sharpens the point. Capital One's Q3, reported the same day, booked a $225 million reserve release while Synchrony built $30 million — same consumer, opposite reserve direction. Capital One's NIM widened to 8.41% while Synchrony's slipped to 15.50%, though Synchrony's absolute margin remains nearly twice as high. Both lenders' efficiency ratios rose and both returned capital aggressively. The difference is structural, not cyclical: Synchrony's RSA means good credit performance is contractually shared, so the P&L keeps less of every benign quarter than a proprietary-network lender does. That is a feature in downturns — partners absorb pain symmetrically — and a governor in quarters like this one.

Tracking a $103 Billion Loan Book in Plain Text​

A card lender's balance sheet is where double-entry discipline pays most visibly: every dollar lent must reconcile to a dollar of deposits, securitization debt or equity, and the allowance must sit apart from the loans it reserves against. Our modeling follows how we model every company — revenue is total net revenue, the provision and the retailer share are the cost of revenue, and preferred stock folds into paid-in capital net of treasury stock. Here is the quarter's income-statement transaction, straight from the pushed ledger — income posts negative (credits), expenses positive (debits), and the whole thing sums to zero:

; FY2026Q3 Income Statement -- fiscal quarter ended September 30, 2026 (QTD)
; Check: -4914 + 1096 + 1266 + 1371 + 289 + 892 = 0
 
2026-09-30 * "Synchrony Financial" "FY2026Q3 Income Statement"
  Income:Revenue                              -4914 MUSD  ; total net revenue (NII 4,809: interest income 5,788 - interest expense 979; other income 105)
  Expenses:CostOfRevenue                        1096 MUSD  ; retailer share arrangements
  Expenses:CostOfRevenue                        1266 MUSD  ; provision for credit losses ($30M reserve build vs $152M release in Q3 2025)
  Expenses:OperatingExpenses                    1371 MUSD  ; total other expense (employee 507 + professional 234 + marketing 137 + info processing 260 + other 233)
  Expenses:IncomeTax                             289 MUSD  ; provision for income taxes
  Equity:Adjustments                             892 MUSD  ; net income offset (no NCI; preferred dividends 28 are a distribution)

The one balance-sheet number that tells the narrative is treasury stock: $21.7 billion, against $9.9 billion of paid-in capital — buybacks exceed everything shareholders ever paid in by $10.1 billion, which is why the equity account carries a debit balance:

2026-09-30 balance Equity:CommonStockAndAPIC                 10072 MUSD  ; preferred 1,716 + common 1 + APIC 9,912 - treasury 21,701 = (10,072): debit, buybacks exceed paid-in
2026-09-30 balance Assets:NonCurrent:LoanAllowance           -10341 MUSD  ; Allowance for credit losses (contra)

Shares outstanding fell from 347 million to 311 million in nine months. That $2.8 billion of 2026 buybacks is the reason a 17% earnings decline became a 6% EPS decline — the ledger shows the transfer from equity to shareholders line by line.

The Multi-Year Arc​

Two annuals and a quarter, with the 2024 gain broken out so the trend reads clean:

Read reported figures first: revenue peaked in 2024, net income is flat. Then adjust: 2024 holds the $1.1 billion pre-tax Pets Best sale gain ($802 million after tax), without which 2024 net income was $2,697 million — and 2025's $3,552 million is 32% growth on the clean base, driven by the provision falling $1.5 billion as post-pandemic delinquencies normalized. The loan book has hovered near $104 billion for two years while deposits fund 83% of it; the allowance coverage glides down as realized losses come in below reserved levels. This is a mature, self-funding book being harvested: earnings flat to growing, shares shrinking fast, capital returned at a pace ($1.2 billion this quarter, $4.6 billion of authorization left) that keeps CET1 drifting down from 14.3% toward 12.9%.

The Verdict: Bull vs. Bear​

Bull Case

• Record $49.7 billion purchase volume (+8%) with growth in all five platforms converts to loan and fee growth with a lag — the top of the funnel is the best leading indicator a card lender has.
• Credit is genuinely improving: charge-offs at a seasonal low of 4.79% (−37bp), both delinquency buckets down, coverage still above 10% — the $120 million provision increase is a reserve-comparison artifact, not deterioration.
• The buyback compounds relentlessly: shares outstanding down 10% in nine months, $4.6 billion of authorization remaining, book value per share still up 8% to $47.64.
• Deposit funding at 83% of the stack with deposit costs falling while balances grew $3.1 billion — pricing power on the liability side most banks would envy.
• Returns remain elite even in a "bad" quarter: 21.0% ROE and 25.2% ROTCE fund both the dividend and the buyback with room to spare.

Bear Case

• The RSA governor is structural: management's own framing admits better program performance flows to partners, so benign credit never fully reaches shareholders — Q3's $72 million RSA increase proves it.
• Reserve direction is a choice that flatters peers, not Synchrony: Capital One released $225 million while Synchrony built $30 million, and the $182 million year-over-year swing explains nearly the entire earnings decline.
• The efficiency ratio deteriorated 330bp to 35.9% on operational losses and tech spend with no offsetting revenue — expense discipline, not credit, is the line to watch now.
• CET1 fell from 14.3% to 12.9% in a year under the weight of capital returns; another year of $1.2-billion quarters narrows the buffer for the downturn the 10% allowance is priced for.
• The loan book is stagnant near $104 billion for two years while volume grows 8% — higher payment rates mean Synchrony originates activity it increasingly does not get to finance.

Our Take: This is a high-quality toll bridge having its tolls renegotiated in real time. The volume record, the improving credit and the 25% ROTCE say the franchise is as strong as ever; the RSA share, the reserve swing and the expense creep say shareholders capture less of each good quarter than the headline growth implies. At 12.9% CET1 with $4.6 billion of buybacks still authorized, management is betting the book's quality justifies harvesting rather than hoarding — and the allowance math agrees with them. Hold the shares for the compounding buyback; watch the RSA rate and the "other" expense line for the moment harvesting turns into skimping.