Affirm Holdings
This is the archived full research. Read the story
A lender built into the checkout: shoppers split a purchase into fixed payments, merchants pay a fee for offering it, and Affirm prices every loan one purchase at a time.
The scores
Growth4/4 ↑
GMV +37% and revenue +32% in FY2026, organic, growing far faster than the e-commerce market it sells into, with top-partner concentration falling.
Business quality3/4 ↑
A real, quantified funding-cost edge built on data scale, but untested through a full credit cycle and facing a larger network and much larger rivals.
Financial position3/4 →
Cash-rich corporate balance sheet with modest, mostly-covered convert maturities, but origination growth depends on continuous outside capital-markets funding access.
Execution record3/4 ↑
Guidance delivered and raised through FY2026, first GAAP operating profit landed as flagged - but the profitable track record is under two years.
Per share capture3/4 →
Diluted shares +2.3% in FY2026 against 32% revenue growth, buybacks and note retirement paid in cash - but the 2029 converts could still add shares.
Potential2/4 →
A large e-commerce base with real but early growth vectors - Affirm Edge needs bank/credit-union adoption management itself calls slow.
Risk profile2/4 →
Two originating banks and two card-issuing banks concentrate a legal-structure risk that resolves in one step, layered on real Amazon/Shopify GMV concentration.
Valuation3/4 →
About 34.1 times earnings, 120% of an excess-return value near $56.85 at an 8% return: fair, inside the 20% margin.
Evidence quality3/4 →
Central claims are filed GAAP facts and the threat is externally corroborated by Klarna's own filing, but the normalized-earnings estimate rests on graded, unverified guidance.
If it works 0.4× · if it fails you keep about 41%.
The Verdict
We rate Affirm Neutral, on the edge. It lends at the checkout, letting shoppers split a purchase into fixed payments, and it is winning: purchases through it grew far faster than online shopping and it made its first full year of operating profit, so the business is solid. The price sits at the top of our band and counts on that first good year compounding for years, while the loan book has never lived through a real downturn and one partner brings a large share. As a bet on new countries and its bank product it is priced for success: shareholders keep about two-fifths if they fail.
The long caseShoppers are moving from credit cards to fixed instalments, at home and abroad; Affirm carries the product winning in America to new countries and to banks.
How we weighed it
The price against our value
Against it
Price at 120% of our value
The price sits at the top of our band, with the bear case far below.
The business
In its favour
Solid
Purchases through it grow far faster than online shopping, and it has turned an operating profit.
Growth against what the price assumes
In its favour
Growth is strong, and the next year's profit rate is guided above target.
What could go wrong
Against it
Its loans have not been tested in a downturn, one partner is a large share, two banks make its loans and convertible bonds may become shares.
Against competitors
Cuts both ways
No filed rival compares directly; its own past is mostly losses.
How its group is behaving
Cuts both ways
Fintech lenders move together on credit and rate news.
The latest news and report
In its favour
The last report beat its own profit guide and grew purchases fast, with no new proof on credit.
The evidence under this - 8 claims
The normalization assumes FY2026's operating income is a representative run rate; it is the company's first profitable fiscal year, so one quarter or year of margin softness or strength would move the normalized multiple by a large amount in either direction - this is exactly the kind of assumption A02 flags as high-sensitivity.
- 2026-10-02editedprice leg re-struck at the 2026-10-01 close of $70.43 (was $77.76) beside C29; the normalized EPS of about $1.02 is unchanged, so the multiple moves from about 76x to about 69x
- 2026-11-05editedFQ1 2027 print: normalized earnings re-struck on trailing pretax income of $614.7M; multiple about 60x at the 4 November close (was 69x)
- 2026-10-02editedvalid_through of 2026-09-30 passed; price, market capitalization and trailing P/E re-struck at the 2026-10-01 close of $70.43 (was $77.76 on 2026-08-28)
- 2026-11-05editedprice re-struck at the 2026-11-04 close of $75.59 (was $70.43 on 2026-10-01)
RLTC guided above its own stated midterm range is either durable operating leverage (consistent with C14) or a level that reverts toward 3.25%-4% as competition or funding costs normalize - management did not say which, and no filed figure yet exists to check FY2027 against.
- 2026-11-05editedFQ1 2027 shareholder letter: RLTC 4.30% of GMV, above the 4.16% guided; the first of the two quarters route P2 needs
- 2026-11-09editedFQ1 2027 call read: current-quarter growth and the raised growth floor added
The company
At an online checkout, Affirm offers to split the purchase into fixed payments: short plans at no interest, longer ones with simple interest and no late fees. Its card lets shoppers do the same after paying. Without it, the shopper reaches for a credit card or does not buy. Shoppers put $50.2B of purchases through it in the year to June 2026.
The evidence under this - 2 claims
How it makes money
Affirm is paid twice on many loans: a fee from the merchant, and interest from shoppers on the longer plans. Partner banks make the loans and investors fund them, so what Affirm keeps depends on how cheaply they lend. Management expects to keep 4.16% of every purchase this year after funding, losses and processing, above its own 3.25% to 4% target.
The evidence under this - 4 claims
- 2026-11-05editedFQ1 2027 10-Q and letter: funding costs grew 22% against GMV 34%, no longer a fifth the pace of revenue; the sequential cost of funds rose
- 2026-11-09editedFQ1 2027 call read: funding-cost explanation added (more than 70% of on-balance-sheet funding fixed rate)
RLTC guided above its own stated midterm range is either durable operating leverage (consistent with C14) or a level that reverts toward 3.25%-4% as competition or funding costs normalize - management did not say which, and no filed figure yet exists to check FY2027 against.
- 2026-11-05editedFQ1 2027 shareholder letter: RLTC 4.30% of GMV, above the 4.16% guided; the first of the two quarters route P2 needs
- 2026-11-09editedFQ1 2027 call read: current-quarter growth and the raised growth floor added
Where the money comes from, stream by stream
| Stream | Share of sales (FY2026) | How it earns it |
|---|---|---|
| Interest income | 48.1% | Simple interest consumers pay on interest-bearing installment loans Affirm holds, including Affirm Card pay-over-time balances. |
| Merchant network revenue | 27.0% | A fee charged to merchants as a percentage of each sale Affirm finances, larger on 0% APR loans. |
| Gain on sales of loans | 14.0% | The gain when Affirm sells loans to outside capital partners, through forward-flow sales and off-balance-sheet securitizations. |
| Card network revenue | 6.9% | Interchange on Affirm Card and virtual-card spending, earned through its card-issuing partner banks. |
| Servicing income | 4.1% | Fees for servicing the loans Affirm has sold to third-party owners. |
Who runs it
Max Levchin, a PayPal co-founder, started Affirm in 2012 and still runs it; Rob O'Hare has run finance since late 2024. Levchin owns about 9% of the company but, through shares carrying fifteen votes each, controls about 44% of the votes. He takes almost no salary and no bonus; his one big option award paid only if the share price climbed, and most of it lapsed unearned. The other executives' bonuses ride on growth, sales and adjusted profit, none on loan losses.
The evidence under this - 4 claims
Levchin sold 553,554 and 113,112 shares in January 2026 (C21), after this table's date, and filed further Form 4s in September 2026, so his stake is lower than shown here.
- Adjusted operating income excludes stock pay and depreciation, and no measure pays on credit losses, return on equity or per-share results.
- The fiscal 2025 revenue and adjusted operating income goals were beaten by 27% and 54%, so the plan paid near its 150% cap against goals the business cleared easily.
Why it wins, and who could beat it
Why customers come back
Every loan teaches the model. Affirm prices each purchase on data from about 553 million past loans, which management says lets it approve more shoppers at lower losses.
The investors who fund those loans appear to agree: funding costs rose 7% last year while purchases rose 37%. So the edge is cheaper money, not higher prices, and it has yet to be measured against Klarna, a larger rival that began filing public accounts only in 2025.
The evidence under this - 3 claims
- 2026-11-05editedFQ1 2027 10-Q and letter: funding costs grew 22% against GMV 34%, no longer a fifth the pace of revenue; the sequential cost of funds rose
- 2026-11-09editedFQ1 2027 call read: funding-cost explanation added (more than 70% of on-balance-sheet funding fixed rate)
The threat
The oldest rival is the credit card: banks such as Chase and Capital One fund it with cheap deposits and already own the customer. Klarna, Block's Afterpay and PayPal sell the same checkout choice, and Klarna's annual report names Affirm as a rival right back.
The evidence under this - 2 claims
The rivals: KLAR, KLAR, XYZ, PYPL, SYF, JPM, C, BAC, COF, BFH, AXP.
Who else does this, rival by rival
| Ticker | Rival | Where it competes | How far along | Priced against it |
|---|---|---|---|---|
| KLAR | Klarna Group plcPriced against it: Same product | Affirm in the United States and Canada: checkout loans and the Affirm Card | On sale | Same product |
| Klarna in the U.K. and Europe | Affirm's new markets and Affirm Edge: the U.K., Australia, continental Europe and bank partners | On sale | ||
| XYZ | Block, Inc. (Cash App / Afterpay)Priced against it: Small part of it | Affirm in the United States and Canada: checkout loans and the Affirm Card | On sale | Small part of it |
| PYPL | PayPal Holdings, Inc.Priced against it: Small part of it | Affirm in the United States and Canada: checkout loans and the Affirm Card | On sale | Small part of it |
| SYF · JPM · C · BAC · COF · BFH · AXP | Card-issuing banks (Synchrony, J.P. Morgan Chase, Citibank, Bank of America, Capital One, Bread Financial, American Express)Priced against it: Small part of it | Affirm in the United States and Canada: checkout loans and the Affirm Card | On sale | Small part of it |
How big it could get
How big this could get
Americans spent about $1.30 trillion online in the four quarters to June 2026, 8.4% more than a year before; Affirm's $50.2B sits inside that pool. Nobody measures how much of it could shift to fixed payments, so this is the size of the pool, not a target.
The evidence under this - 2 claims
Who has signed
Amazon and Shopify have signed agreements, but neither discloses a minimum volume, and partner deals like these usually run one to three years. Amazon alone is 22% of Affirm's purchases, and the top five partners 44%. Shopify also holds more than 10% of Affirm's votes.
The evidence under this - 2 claims
The results
The numbers
Sales grew 33% over the twelve months to Sep 2026, about the same pace as the 32% in the year to Jun 2026.
Operating margin is 12%, up from −27% in the year to Jun 2024.
Yearly sales growth has ranged from +18% to +55% over the last 5 years.
Each dollar put into the business earns 4.0% a year after tax, about where it was (−1.0%) in the year to Jun 2025.
It turned 18% of sales into free cash, $817M over the last twelve months.
It owes $8.18B more than the cash it holds, about 7.7 years of the cash its operations bring in.
The share count rose 25% in 4.3 years, and pay in shares costs 6.0% of sales.
All the numbers
Filed figures from the company's 10-K and 10-Q, read 2026-11-09. Fiscal years end in June.
| Jun 2022 | Jun 2023 | Jun 2024 | Jun 2025 | Jun 2026 | 12 mo to Sep 2026 | |
|---|---|---|---|---|---|---|
| Size and growth | ||||||
| Sales | $1.35B | $1.59B | $2.32B | $3.22B | $4.26B | $4.59B |
| Sales growth | +55% | +18% | +46% | +39% | +32% | +33% |
| Profit | ||||||
| Operating margin | −64% | −76% | −27% | −2.7% | 9.8% | 12% |
| Net margin | −52% | −62% | −22% | 1.6% | 45% | 44% |
| What growth costs, per sales dollar | ||||||
| Sales and marketing | 39% | 40% | 25% | 13% | 8.0% | 7.6% |
| General and admin | 43% | 37% | 23% | 17% | 14% | 13% |
| Pay in shares | 29% | 28% | 15% | 10.0% | 7.2% | 6.0% |
| Capital spending | 6.4% | 7.6% | 6.9% | 6.0% | 5.6% | 5.4% |
| Cash | ||||||
| Cash from operations | −$162M | $12.2M | $450M | $794M | $1.23B | $1.06B |
| Free cash flow | −$248M | −$109M | $291M | $602M | $993M | $817M |
| Balance sheet | ||||||
| Cash and short-term investments | $1.26B | $892M | $1.01B | $1.35B | $1.63B | $2.07B |
| Debt | - | - | $1.84B | $7.61B | $9.79B | $10.3B |
| Net cash | - | - | −$824M | −$6.26B | −$8.16B | −$8.18B |
| Returns and shares | ||||||
| Return on capital | - | - | - | −1.0% | 3.2% | 4.0% |
| Shares (diluted) | 282M | 295M | 310M | 341M | 349M | 352M |
What growth costs
Growing has made Affirm more profitable. The costs tied to each loan, such as losses, funding and processing, grow with sales, but overhead does not: marketing, administration and stock pay have shrunk sharply as a share of sales. Part of the marketing drop is partner warrants running out rather than cheaper selling, and some stock pay is now booked as software instead of expense.
The evidence under this - 3 claims
- Most of FY2026's sales and marketing fall is partner warrant expense running off, not cheaper selling: Amazon warrant expense fell $92.4 million and Shopify's $15.4 million, while marketing and promotion rose 36%.
- $179.8 million of FY2026 stock pay was capitalized into software rather than expensed, so the expensed stock-pay ratio flatters the true cost; total stock pay was $484.5 million.
- 2026-11-05editedFQ1 2027 10-Q and letter: funding costs grew 22% against GMV 34%, no longer a fifth the pace of revenue; the sequential cost of funds rose
- 2026-11-09editedFQ1 2027 call read: funding-cost explanation added (more than 70% of on-balance-sheet funding fixed rate)
The last report (5 Nov)
Quarter ended September 2026, filed figures.
In the quarter to September, Affirm's purchases grew 34% to $14.4B and revenue grew 35% to $1,262M. After the cost of money and loan losses it kept 4.30% of each purchase, above the 4.16% it guided for the year. Operating income nearly tripled to $187M.
Credit was the weaker part: late payments of 30 days or more rose to 2.9% from 2.55%. On the call, management said the new quarter began with purchases growing in the high 30s, and tied the higher loan-loss cost to approval tests it chose to run. That is management's account, not yet in the numbers.
What it did to the thesisIt passed the test: the profit rate beat its own guide and purchases grew far faster than online shopping. The call added a strong start to the next quarter but no new proof on credit. Our route needs the profit rate held for two more quarters, so Neutral holds.
The quarter in detail
Sales growth from the previous quarter, filed figures.
Earnings per share $0.47 against $0.38 expected.
The evidence under this - 4 claims
The normalization assumes FY2026's operating income is a representative run rate; it is the company's first profitable fiscal year, so one quarter or year of margin softness or strength would move the normalized multiple by a large amount in either direction - this is exactly the kind of assumption A02 flags as high-sensitivity.
- 2026-10-02editedprice leg re-struck at the 2026-10-01 close of $70.43 (was $77.76) beside C29; the normalized EPS of about $1.02 is unchanged, so the multiple moves from about 76x to about 69x
- 2026-11-05editedFQ1 2027 print: normalized earnings re-struck on trailing pretax income of $614.7M; multiple about 60x at the 4 November close (was 69x)
- 2026-11-05editedFQ1 2027 10-Q: 90-day past-due loans rose to about 0.78% from 0.67% and 30-day delinquencies rose 35 basis points on the quarter
- 2026-11-09editedFQ1 2027 call read: management's reasons for the higher provision and delinquencies added; the filed figures are unchanged
RLTC guided above its own stated midterm range is either durable operating leverage (consistent with C14) or a level that reverts toward 3.25%-4% as competition or funding costs normalize - management did not say which, and no filed figure yet exists to check FY2027 against.
- 2026-11-05editedFQ1 2027 shareholder letter: RLTC 4.30% of GMV, above the 4.16% guided; the first of the two quarters route P2 needs
- 2026-11-09editedFQ1 2027 call read: current-quarter growth and the raised growth floor added
What management did with the money
It has paid in cash, not shares: in fiscal 2025 it spent $1.0B retiring convertible bonds due in November 2026 and $250M buying back stock, and in fiscal 2026 only $26M more on those bonds. Diluted shares still rose 2.3% to 348.8 million, and $1.14B of convertible bonds remain that could still turn into shares.
The evidence under this - 2 claims
- 2026-10-04editedthe $1.0 billion note extinguishment and $250.0 million buyback were FY2025 (December 2024), not FY2026; FY2026 repurchased only $27.4 million of 2026 Notes for $25.8 million (FY2026 10-K cash flow statement, MD&A liquidity and debt note, S01)
- 2026-11-05editedFQ1 2027 10-Q cover page: 339.0M shares outstanding at 30 October (was about 337.2M at 30 June)
The price, and where we disagree
In short. Fair. The price is about 120% of what we think the business is worth, at an 8% yearly return.
Nothing forces the question today: the shares rose 8.7% in the weeks after the results. The test comes on 5 November with the first quarter of the new year. Does Affirm keep the 4.16% it promised, and do purchases keep growing far faster than online shopping?
In between, on 21 September, Affirm replaced the risk model the moat rests on, and published no figure on how well the new one works.
The evidence under this - 5 claims
RLTC guided above its own stated midterm range is either durable operating leverage (consistent with C14) or a level that reverts toward 3.25%-4% as competition or funding costs normalize - management did not say which, and no filed figure yet exists to check FY2027 against.
- 2026-11-05editedFQ1 2027 shareholder letter: RLTC 4.30% of GMV, above the 4.16% guided; the first of the two quarters route P2 needs
- 2026-11-09editedFQ1 2027 call read: current-quarter growth and the raised growth floor added
How we value it
- Method
- The owners' book plus the returns it earns above the rate
- Required return
- 8%
- Long-run growth
- 4%
- Moat
- An average moat
| Bear | Base | Bull | |
|---|---|---|---|
| Book growth, years 1 to 3 | 13.3%, 12.2%, 13.2% | 26.6%, 24.4%, 26.4% | 33.3%, 30.5%, 33% |
| Normal return on equity | 9.7% | 12.1% | 13.3% |
| Return on new capital after year ten | 8% | 12% | 15% |
| Value a share | $19.97 | $56.85 | $101 |
- Price
- $68.11
- Priced on
- 6 Nov
- Price as a share of the base value
- 120%
- Yearly return the price offers on the base case
- 7.5%
- Risk check: valued at 10%, the rate for a cyclical or indebted business
- $29.44
Where each figure comes from
| Figure | Source | Where exactly |
|---|---|---|
| The owners' book | Filed report | Quarterly report (10-Q) for the period to 30 Sep 2026, filed 5 Nov 2026; the balance sheet on that date. |
| Shares | Filed report | Read by hand from the filed statements; the period it was read from was not recorded. |
| Book growth, years 1 to 3 | Analysts' forecasts | FMP's consensus for the years to 30 Jun 2027, 30 Jun 2028 and 30 Jun 2029: the average of 19 to 23 analysts' revenue estimates; the owners' book grows on that path, held to 30% a year. FMP publishes how many analysts, not who they are. |
| Normal return on equity | Analysts' forecasts | FMP's consensus for the year to 30 Jun 2029, the average of 19 analysts' estimates: the return on equity of their last forecast year. FMP publishes how many analysts, not who they are. |
| Return on new capital after year ten | Our method's rule | Set by our reading of the moat: the 8% rate with none, 12% with an average one, 15% with a strong one; the bear case earns the rate. |
| The bear and bull cases | Our method's rule | Our method's spread: the bear case takes half the growth and a fifth off the margin, the bull a quarter more growth and a tenth more margin. |
| Price | Market price | FMP's closing price on 6 Nov 2026. |
What each case rests on
Bear caseEquity grows at half that path, 13%, 12% and 13%, the return on equity settles a fifth lower at 9.7%, and new equity earns only the 8% rate: the downturn test.
Base caseEquity grows 27%, 24% and 26%, the analysts' revenue path (C42) held to 30% a year, then fades toward 4%; the return on equity moves to 12.1%, the analysts' last forecast year.
Bull caseEquity grows a quarter faster, 33%, 30% and 33%, the return on equity reaches 13.3%, and new equity earns 15%.
What moves the base value
| Lower | Higher | |
|---|---|---|
| Growth in years 1 to 3, a quarter lower or higher | $47.68 | $68.30 |
| The normal return on equity, 15% lower or higher | $52.98 | $60.72 |
| Return on new capital after year ten, at the rate or one step up | $26.56 | $79.56 |
| The rate, a point higher or lower | $40.23 | $85.19 |
| Long-run growth a point lower | $49.59 | - |
Value a share as growth and margin change
| Normal return on equity ↓ · growth → | 50% | 75% | 100% | 125% | 150% |
|---|---|---|---|---|---|
| 9.7% | $37.10 | $43.54 | $51.69 | $61.91 | $74.65 |
| 10.9% | $38.76 | $45.61 | $54.27 | $65.11 | $78.58 |
| 12.1% | $40.41 | $47.68 | $56.85 | $68.30 | $82.52 |
| 13.3% | $42.06 | $49.76 | $59.43 | $71.50 | $86.45 |
| 14.5% | $43.72 | $51.83 | $62.01 | $74.69 | $90.38 |
Growth is shown as a share of the base case's. Green: the price is at or under 80% of that value. Red: at or over 120%.
How we read the multiple
On last year's reported profit the shares look cheap, at about 12.3 times. But $1.44B of last year's $1.93B profit was a one-off, non-cash tax credit.
Taxed at a normal rate instead:
- Before tax, Affirm earned $492.7M last year.
- Taxed at about 27.5%, in line with management's guide, that is about $1.02 a share.
- So buyers pay about 69 times profit: right if profit keeps compounding, too much if it merely holds or a downturn hits untested loans.
Against its own history. Its past is mostly losses, so there is no normal price for its profit to compare against. On sales and on what it owns, it sits near its usual level.
The evidence under this - 4 claims
- 2026-10-02editedvalid_through of 2026-09-30 passed; price, market capitalization and trailing P/E re-struck at the 2026-10-01 close of $70.43 (was $77.76 on 2026-08-28)
- 2026-11-05editedprice re-struck at the 2026-11-04 close of $75.59 (was $70.43 on 2026-10-01)
The normalization assumes FY2026's operating income is a representative run rate; it is the company's first profitable fiscal year, so one quarter or year of margin softness or strength would move the normalized multiple by a large amount in either direction - this is exactly the kind of assumption A02 flags as high-sensitivity.
- 2026-10-02editedprice leg re-struck at the 2026-10-01 close of $70.43 (was $77.76) beside C29; the normalized EPS of about $1.02 is unchanged, so the multiple moves from about 76x to about 69x
- 2026-11-05editedFQ1 2027 print: normalized earnings re-struck on trailing pretax income of $614.7M; multiple about 60x at the 4 November close (was 69x)
O'Hare explicitly flagged this as imprecise ('there's likely to be some volatility'), which is why C06 uses the range's midpoint rather than treating it as a point estimate.
Priced against its rivals
Only Klarna Group sells the same thing to the same buyers as a large part of what it does. Block, PayPal Holdings, Synchrony Financial and Bread Financial Holdings do too, but as a small part of much bigger companies, so their price says little about this one. A price comparison needs at least two, so the price is judged against its own history instead.
The bet: what it pays if it works
The bet is on Affirm's new markets and Affirm Edge: Britain, Australia, continental Europe, and a version of its product that banks offer on their own cards. Together they made almost nothing last year, so the American and Canadian business is counted separately, at the multiple this thesis would call fair.
If they grow many times over, the shares are still worth less than today, because the price already pays far more for them. If they never earn, shareholders keep about two-fifths.
| Affirm's new markets and Affirm Edge: the U.K., Australia, continental Europe and bank partners revenue in two to three years | Billions of dollars of yearly volume needed | At the stated multiple |
|---|---|---|
| $25.00M | about 0 at 100% kept | 0.4× |
| $100.00M (the big win) | about 1 at 100% kept | 0.4× |
| $300.00M | about 4 at 100% kept | 0.5× |
To pay 2× (a Fair bet) it would need about $6.78B of Affirm's new markets and Affirm Edge: the U.K., Australia, continental Europe and bank partners revenue (about 80 billions of dollars of yearly volume); to pay 5× (a Cheap bet), about $19.59B of Affirm's new markets and Affirm Edge: the U.K., Australia, continental Europe and bank partners revenue (about 230 billions of dollars of yearly volume), at the same multiple.
Each row is Affirm's new markets and Affirm Edge: the U.K., Australia, continental Europe and bank partners alone, plus the rest of the company (Affirm in the United States and Canada: checkout loans and the Affirm Card) counted at about $10.71B.
The steps it has to take
- Done U.K. launched and broadly available
- Done Limited launch in Australia
- In progress Affirm Edge live at its first bank partners due 2027-06
- Not started First launch in continental Europe due 2027-12
- Not started Revenue outside the U.S. and Canada passes one per cent of the total due 2028-06
The detail behind the label
The evidence under this - 4 claims
The note gives revenue by country and no profit, and the same filing's risk factors say expansion into new international geographies presents a variety of challenges and risks, so what the new markets cost is not disclosed.
An analyst on the same call described the international expansion as not yet overly meaningful to the year's revenue, and the chief executive gave no U.K. volume, customer or merchant count and no Edge partner's name.
The normalization assumes FY2026's operating income is a representative run rate; it is the company's first profitable fiscal year, so one quarter or year of margin softness or strength would move the normalized multiple by a large amount in either direction - this is exactly the kind of assumption A02 flags as high-sensitivity.
- 2026-10-02editedprice leg re-struck at the 2026-10-01 close of $70.43 (was $77.76) beside C29; the normalized EPS of about $1.02 is unchanged, so the multiple moves from about 76x to about 69x
- 2026-11-05editedFQ1 2027 print: normalized earnings re-struck on trailing pretax income of $614.7M; multiple about 60x at the 4 November close (was 69x)
Where we disagree with the market
A fast grower at a cheap price: 29 analysts say buy and none say sell, and the shares jumped on the results.
The cheap price is a one-off tax benefit. On normal earnings the shares already pay for years of compounding.
The market is reading the headline price-to-profit figure at face value; none of the coverage we read takes the tax benefit out.
The evidence against us is the business itself. If Affirm keeps 4.16% of each purchase and its volume keeps growing well above online shopping's pace, the compounding is happening and our caution is wrong.
What analysts say, and the evidence
The evidence under this - 7 claims
- 2026-10-02editedvalid_through of 2026-09-30 passed; re-sourced from the pack re-fetched 2026-10-02, which moved the split from 24/10/1 to 28/10/0
- 2026-11-11editedvalid_through of 2026-10-31 passed; re-sourced from the pack fetched 2026-11-11: the snapshot moved from 28/10/0 to 29/10/0 and the series held at 80.6% positive on 36 analysts.
- 2026-10-02editedvalid_through of 2026-09-30 passed; price, market capitalization and trailing P/E re-struck at the 2026-10-01 close of $70.43 (was $77.76 on 2026-08-28)
- 2026-11-05editedprice re-struck at the 2026-11-04 close of $75.59 (was $70.43 on 2026-10-01)
The normalization assumes FY2026's operating income is a representative run rate; it is the company's first profitable fiscal year, so one quarter or year of margin softness or strength would move the normalized multiple by a large amount in either direction - this is exactly the kind of assumption A02 flags as high-sensitivity.
- 2026-10-02editedprice leg re-struck at the 2026-10-01 close of $70.43 (was $77.76) beside C29; the normalized EPS of about $1.02 is unchanged, so the multiple moves from about 76x to about 69x
- 2026-11-05editedFQ1 2027 print: normalized earnings re-struck on trailing pretax income of $614.7M; multiple about 60x at the 4 November close (was 69x)
RLTC guided above its own stated midterm range is either durable operating leverage (consistent with C14) or a level that reverts toward 3.25%-4% as competition or funding costs normalize - management did not say which, and no filed figure yet exists to check FY2027 against.
- 2026-11-05editedFQ1 2027 shareholder letter: RLTC 4.30% of GMV, above the 4.16% guided; the first of the two quarters route P2 needs
- 2026-11-09editedFQ1 2027 call read: current-quarter growth and the raised growth floor added
What could go wrong, and how we would know
The risks
- An untested lender. Affirm holds $9.47B of loans and has never lived through a real credit downturn at this size.
- One partner is 22%. Amazon brings 22% of purchases on a contract with no disclosed minimum, though no single merchant is 10% of revenue.
- The partner-bank model. Two banks make nearly all its loans. If that model is ruled impermissible, Affirm could breach state lending laws.
- Bonds that become shares. $1.14B of convertible bonds could settle in new shares, spreading normal earnings more thinly.
The evidence under this - 6 claims
- 2026-11-05editedFQ1 2027 10-Q: 90-day past-due loans rose to about 0.78% from 0.67% and 30-day delinquencies rose 35 basis points on the quarter
- 2026-11-09editedFQ1 2027 call read: management's reasons for the higher provision and delinquencies added; the filed figures are unchanged
- 2026-10-04editedthe $1.0 billion note extinguishment and $250.0 million buyback were FY2025 (December 2024), not FY2026; FY2026 repurchased only $27.4 million of 2026 Notes for $25.8 million (FY2026 10-K cash flow statement, MD&A liquidity and debt note, S01)
- 2026-11-05editedFQ1 2027 10-Q cover page: 339.0M shares outstanding at 30 October (was about 337.2M at 30 June)
What we still do not know
- Whether next year's above-target margin is a new level or a funding window that will close. Management did not say.
- Whether Affirm's application for its own banking licence is approved, over opposition from industry and advocacy groups.
The evidence under this - 2 claims
RLTC guided above its own stated midterm range is either durable operating leverage (consistent with C14) or a level that reverts toward 3.25%-4% as competition or funding costs normalize - management did not say which, and no filed figure yet exists to check FY2027 against.
- 2026-11-05editedFQ1 2027 shareholder letter: RLTC 4.30% of GMV, above the 4.16% guided; the first of the two quarters route P2 needs
- 2026-11-09editedFQ1 2027 call read: current-quarter growth and the raised growth floor added
- 2026-11-09editedRe-sourced to the Q1 FY2027 10-Q; the 10-K's pending-with-opposition wording was not repeated in the section read, and no approval is reported.
What would change our view
- Neutral→AttractiveThe price falls 20% under fair value
It needs:
- ○The share price falls to about $45.48, 80% of what we think the business is worth.about $45.54 today, 41% below the last close
- Neutral→UnattractiveThe price runs 20% above fair value
It needs:
- ○The share price rises to about $68.22, 20% above what we think the business is worth.about $68.31 today, -12% above the last close
- Reopens the researchThe convertible settlement stops threatening the per-share line
It needs:
- ○Affirm shows the $1.14B of convertible bonds will be settled mostly in cash rather than new shares.checked at the next results, due in Feb 2027
We would also be wrong if
- Late payments rise well above today's roughly 0.7% of loans.
- Amazon or Shopify sends much less through Affirm, or ends the partnership.
- Operating margin fails to widen from this year's level through fiscal 2028.
What is coming
- 15 November 2026 2026 convertible notes mature: Whether the $221.3M is repaid from cash as expected; with conversion near $216 against a price near $70 it should settle without new shares.
Every dated event ahead
- 15 November 2026 2026 convertible notes mature: Whether the $221.3M is repaid from cash as expected; with conversion near $216 against a price near $70 it should settle without new shares.
- February 2027 Q2 FY2027 results, the holiday quarter: Holiday GMV, charge-offs and funding costs in the seasonally largest quarter, and whether FY2027 profit tracks the inflection the price assumes.
Since we wrote this
- 11 NovPrice movePrice near the top of our band; Neutral holdsThe shares sit about 116% of our value, four points under the line that would make them Unattractive. Insider sales were nearly all scheduled in advance.
- 9 NovEarningsThe call added high-30s growth; Neutral heldManagement said December-quarter growth started in the high 30s and tied the extra provision to its own experiments. The 4.16% test is still one quarter short, so the view held.
- 9 NovThesis reviewStill Neutral: priced at 120% of a value near $56.85Re-valued on its equity and the returns above 8% on it, the method drawn on every lender and insurer: the price sits inside the 20% margin of safety.
- 5 NovEarningsMargin beat its guide and profit nearly tripledRLTC reached 4.30% of GMV and operating income $187M. Route P2 needs a second quarter at 4.16% and a lower price, so the view held.
- 22 OctNewsAffirm becomes Sam's Club's exclusive pay-over-time provider, terms unread. View heldAffirm launches as exclusive pay-over-time provider at Sam's Club.
- 11 OctThesis reviewAffirm's new markets and Edge declared an early business, with a betThe owner's rule of 2026-10-11: every thesis names its lines, and an early line owes a bet. Affirm's expansion abroad and its bank product earn almost nothing yet.
- 5 OctThesis reviewThe long case now sits under the verdictAfter the five-name pilot, the owner asked for the long case on every thesis.
- 4 OctThesis reviewOwner's checklist: revenue streams, cost of growth, management, catalystsThe owner's checklist found the method silent on who runs the company, what growth costs, where the money comes from and what is dated ahead; this pass fills those four blocks from the proxy and the 10-K.
- 2 OctThesis reviewPrice and analyst split re-struck; view heldTwo perishable market claims expired. At $70.43 the shares cost about 69 times normal earnings, down from 76, still a price that assumes the first good year compounds.
- 21 SepNewsThe engine the moat rests on was replaced, and nothing yet says how well. View heldA new underwriting model announced, at the centre of the moat and with no figure attached.
- 30 AugThesis writtenNeutral, medium convictionWritten three days after Affirm's first full year of operating profit, when a one-off tax benefit made the shares look cheap.
Also checked 4 times with nothing new to act on.
Sources 18
Sources - 18
Every claim above cites one of these, or states the reasoning that derived it. The quote is frozen at the date it was read: a link can rot, and an agent fetch returns a reading of a page rather than the page. Backs counts the whole claim ledger, so it can name a claim this page does not surface.
- S01primarycompany filing2026-08-27Business description, competition risk factors, Affirm Bank charter application status, GMV/revenue by product, customer concentration, income statement, balance sheet, debt note, credit-quality tables, MD&A.backsC01C02C03C04C06C08C10C11C12C13C14C15C16C20C24C26C27C28C29C31C32C33C35C37C38C40
- S02primarycompany filing2026-08-27Confirms the Q4/FY2026 shareholder letter was furnished as Exhibit 99.1; the exhibit itself is image-heavy and did not yield readable text beyond its cover boilerplate, so this thesis relies on the 10-K's own statements for the underlying figures rather than the letter.read, but no claim rests on it
- S03primarytranscript2026-08-27Affirm Holdings Q4 FY2026 earnings call transcript (2026-08-27)CEO's 'most profitable quarter ever, even without the tax allowance release' framing, RLTC and FY2027 tax-rate guidance, credit-management commentary, Affirm Edge and international competitive color. The Q&A is a distinct origin from the prepared remarks even though both share this filing's independence group.backsC05C07C25C41
- S04primarycompany filing2026-02-03Confirms PayPal's own BNPL-adjacent lending (PPWC, PPBL, certain installment products) is also funded through a third-party 'partner institution' rather than PayPal's own deposits - the funding-model constraint AFRM discloses is shared by at least one large competitor, not unique to AFRM.backsC18
- S05primarycompany filing2026-02-26Klarna's own filing names 'payment networks such as Affirm, Block or PayPal' as direct competitors - external confirmation of the threat from the attacker's own side. Also gives Klarna's FY2025 GMV ($128 billion) and its own advertising-revenue scale ($190 million), used to assess whether Klarna is a product-matched, whole-company-comparable peer.backsC17C19
- S06primarygovernment data2026-08-30FRED series ECOMSA - U.S. Census Bureau E-Commerce Retail Sales (quarterly, seasonally adjusted) U.S. Census Bureau / Federal Reserve Bank of St. Louis (FRED)Trailing-four-quarter U.S. e-commerce retail sales matching AFRM's FY2026 (Jul 2025-Jun 2026) and FY2025 (Jul 2024-Jun 2025) fiscal windows, used as the base for the prize and the current-state growth-rate comparison.
2025-07-01,310817 | 2025-10-01,317999 | 2026-01-01,327861 | 2026-04-01,340244 | 2024-07-01,295249 | 2024-10-01,300329 | 2025-01-01,297701 | 2025-04-01,303284
backsC09 - S07secondaryvendor data2026-08-30AFRM research pack (market tier) - price, volatility, analyst split, insider filings, short interest, earnings historyLast close price and market cap, 20-day volatility, analyst consensus split, Form 4 insider activity, 8/8 EPS-beat history, technicals (returns, volume).backsC30
- S08primarygovernment data2026-08-30AFRM's short interest at the most recent FINRA settlement.
14,461,742 shares short, 5.1% of float, settled 2026-08-14, -6.0% vs prior settlement, 5.38 days to cover
backsC22 - S09secondaryvendor data2026-10-02AFRM research pack (market tier), re-fetched 2026-10-02 - close of 2026-10-01, analyst split, vendor market cap and trailing P/ELast completed close ($70.43 on 2026-10-01), vendor market capitalization ($23.59B), trailing P/E (12.33x), analyst consensus snapshot (28 buy, 10 hold, 0 sell) and the dated rating series (34 analysts, 79.4% positive).backsC29
- S10primarycompany filing2025-10-24Beneficial ownership and voting power as of September 30, 2025; executive officers and their tenure; the fiscal 2025 cash incentive plan's measures, goals and payout; the CEO's Value Creation Award terms. The fiscal 2026 proxy was not yet filed on 2026-10-04.backsC34C35C36C38
- S11secondaryvendor data2026-10-04Repo earnings calendar content/calendar/2026-10-04.json (FMP earnings rows, saved by the daily routine)The dated next print (2026-11-05).
"AFRM": {"archived": false, "next": {"date": "2026-11-05"
backsC39 - S12primarycompany filing2026-11-05FQ1 2027 GMV, revenue, RLTC, operating income, delinquencies, funding costs and fiscal 2027 outlookbacksC06C14C24C25
- S13primarycompany filing2026-11-05income statement, loans held for investment, past-due loans and shares outstanding at 30 September and 30 October 2026backsC06C13C14C24C27C28C29
- S14primarytranscript2026-11-05management's answers on growth, funding costs, the September provision and delinquencies
Even in the first month of Q2, the business is growing at a high 30s rate today. ... of the on-balance sheet funding, more than 70% of that funding was fixed rate in nature.
backsC14C24C25 - S15primarycompany filing2026-11-09shareholders' equity, goodwill and intangibles, preferred stock, diluted earnings per share and the share countbacksC43
- S16secondaryvendor data2026-11-09the analysts' average revenue and earnings per share for the next fiscal years, and the closebacksC42C43
- S17primarycompany filing2026-11-11Every open-market Form 4 row of the 365 days to 11 November 2026: 0 purchases and 19 sales of 1,119,005 shares, 17 of them (1,067,025 shares) flagged as made under Rule 10b5-1 plans and 2 (51,980 shares, the chief accounting officer in September 2026) not flagged.backsC21
- S18secondaryvendor data2026-11-11AFRM research pack (market tier), fetched 2026-11-11: analyst consensus snapshot and dated rating seriesAnalyst consensus snapshot (29 buy, 10 hold, 0 sell) and the dated rating series (36 analysts, 80.6% positive on 2026-11-01, unchanged from 2026-10-01).backsC23