TON618 CapitalEquity Valuation · Update
As of 3 November 2026
v1.0
Equity Valuation · Fintech · XYZ · 3 November 2026

Block (XYZ) — The Margin Arrived, and So Did the Bill

Block now guides a 32% adjusted operating margin for 2027, above July's best case, and the legal charge July called one-time has become a $764 million accrual for a Justice Department anti-money-laundering case. Its credit gauges point opposite ways because the loans got longer. We value the stock at $91.17, from $83 in July, against $76.00 at the close before the release; the next session closed at $87.63.

TON618 Capital Research · As of 3 November 2026 · Download PDF · Download model

20%July base20%FY2529%Q3-2632%2027 guideADJUSTED OPERATING MARGIN ON GROSS PROFITDOJ ACCRUAL, $M240Q1526Q2764Q3The margin arrived. The bill came with it.Block guides a 32% margin for 2027, the level July thoughtmight come in its best case. The legal charge July calledone-time grew in three steps to $764 million.Margins are the company’s adjusted figure, which leaves outlegal and restructuring charges; §2 puts them back.THE PRINT · THE MARGIN · THE BILL · THE CREDIT · THE VALUE · THE RECEIPT · THE TRIPWIRES

In July we called Block the most misread name in the crypto complex and hung the call on one question: was the tripled loss line timing or trouble? Four months on, three things have moved. The margin arrived early: the company now expects to earn in 2027 the margin on gross profit that July thought possible only in its best case. The bill arrived too: the litigation charge July set aside as one-time was the first accrual for a federal anti-money-laundering case, now $764 million and not yet final or paid. And the credit gauges now disagree, for a reason that turns out to be the length of a loan. This note sorts out how much of each is real, then values the business by its parts.

The questions

What this note answers

  1. What did Block earn, and why do GAAP and adjusted profit differ by half a billion dollars?
  2. What is the 32% margin guided for 2027, and how much of it lasts?
  3. Is the legal charge one-time, and what does it cost holders?
  4. Why do the two credit gauges point opposite ways, and what reading would settle it?
  5. What is Block worth now, what moved the number from $83, and what does $76.00 require?
  6. What would change our view?
$91fair value a share · bear $47 / base $97 / bull $124 at 25 / 50 / 25
The view. We value Block at $91.17 a share, from $83 in July, 20% above the $76.00 close of 3 November, the session before the results traded. The next session closed at $87.63, 15.3% higher, and touched $88.51, a 52-week high; that is one data point, not an input, and it leaves the stock 4% below our value. The quarter's data alone, on July's method, give $81.36; the move comes from the margin. Fed into July's frame, the 2027 outlook would say $117.46. We value the business by its parts instead and take back three things that frame left out: the charges the adjusted figure excludes, which have run every year since 2023; the equity the growing loan book ties up; and a cost of equity of 13.0%, argued from Block's own risk. The price requires gross profit to grow 5.7% a year for a decade at our margins, against at least 15% guided for 2027, or margins about 5.0 points lower than ours in every year. Our largest single judgment is that the cost of equity falls to 10.5% as the business matures: held at 13.0% throughout, the value is $74.09, below the 3 November close, so that fall carries more than the whole gap (§5). Credit is the open condition: losses per dollar lent are rising by design, inside the company's targets so far, and fourth-quarter write-offs on the season's best loans will show whether they stay there.
31.5%adjusted operating margin guided for 2027 (the company rounds to 32%), against July's 20% base and 24% bull; Q3 printed 29%, and GAAP 13%
$764Maccrued for the DOJ's Bank Secrecy Act case, built from $240M in Q1; accrued legal contingencies $981M, from $173M in December
14.54%reserve coverage on consumer loans, down from 14.73%; at June's loan length it would be 15.9%, and reserves per dollar lent rose to 5.03%
614.8Mdiluted shares, against July's 585M forward; no shares repurchased in Q3, $4.6B still authorized

Terms used throughout, defined once. Gross profit: revenue less transaction, bitcoin and lending costs; Block's own top line, since bitcoin sales are booked gross at a sliver of margin. It is struck before credit losses. Adjusted operating income: the company's figure, operating income before contingencies, restructuring and other charges and before amortization of acquired intangibles, but after stock pay. Recurring charges: the contingencies, restructuring and other charges the adjusted figure leaves out, excluding the DOJ accrual and the February layoffs. Coverage: the reserve (allowance) for credit losses on consumer loans held for investment over their amortized cost. 60–89 days: consumer loans 60 to 89 days past due. Originations: loans made in the quarter; for credit ratios we use "purchases and originations of loans originally classified as held for investment" from the cash-flow reconciliation. Variable profit margin: the company's measure, gross profit less processing costs and risk losses, before financing, over originations. DPA: deferred prosecution agreement. Cost of equity: the return holders require; WACC: the blended cost of equity and debt.

The case for the price

The case for $76, in the skeptic's terms

The case against paying more than $76.00 is not that Block is a bad business. It is that the margin story is told in a figure that excludes the costs that keep appearing, on a lending engine whose losses are rising, in a stock that moves more than twice as much as the market.

  1. The adjusted margin leaves out a bill that comes every year. Contingencies, restructuring and other charges were 3.2% of gross profit in 2023, 3.4% in 2024 and 1.6% in 2025. In 2026 they are 4.2% even after setting aside the DOJ accrual and the layoffs. The DOJ matter is not a stray lawsuit: the charge it will defer is that Block "failed to maintain an effective anti-money laundering program," which is a failure of how Cash App was run. A business whose adjusted figure has left out charges in every year it has reported it should be valued on what it keeps, not on the 32%.
  2. Growth is coming from the riskiest products. Financial solutions, mostly Cash App Borrow, is 41% of gross profit and grew 32%. The company moved Borrow from four-week loans at a 5% fee to six-week loans at 7.5%, with a target loss of 4.2% of each loan against 2.9% before, and Afterpay Post-Purchase targets 4.3%. The 60–89-day delinquency rate rose this quarter, net write-offs over originations have risen three quarters running, and the targets are portfolio averages for customers past their first month, not results.
  3. The market prices it as a high-risk equity. Block's 60-month beta to the S&P 500 is 2.56, with a standard error of 0.34, and its 36-month beta is 2.47. Blended toward peers, the 60-month read is 2.29, a 15.3% cost of equity, at which our own model gives $83.02. Nor need the rate fall as the business matures: financial solutions, mostly lending, is now 41% of gross profit, the newer loan products are young. Held at our own 13.0% for every year, the model gives $74.09.
  4. The comparable trades at eleven times. PayPal, the mature version of this business, traded at 10.8 times its 2025 earnings at the same close. Applied to our base case's 2027 after-tax operating profit, PayPal's multiple gives Block $59.78. The $76.00 close was 14.7 times that profit and 18.3 times the company's own adjusted earnings guide for 2026.
  5. Growth is slowing where it matters. Gross profit growth is guided to +15% in the fourth quarter against +22% in the third. Primary banking actives, the customers who run their pay through Cash App, were 9.7 million, 9.4 million and 9.4 million over the last three quarters; inflows per transacting active grew 6% (per primary banking active, 8%).
  6. The cash is not being returned. Block bought back $636 million of stock in the first quarter, about $65 million in the second and none in the third, with $7.9 billion of cash. Whatever the reason, holders are not getting the buyback July's 585 million forward shares assumed.

This note tests the first four claims and the fifth's arithmetic, and sizes and tracks the sixth (§3, §7). It sets out what the margin is after the recurring charges (§2), what the DOJ case costs and whether the other charges recur (§3), what the credit ratios show once loan length and product mix are counted (§4), and what the beta and the multiples imply (§5). It cannot test, inside a ten-year model, whether regulators will limit the fees on short-term consumer loans, what conditions a final DPA will impose beyond the payment, or when the next recession reaches Cash App's customers. Our bear case carries a credit turn in 2027 at $47.12; it does not carry a rule change that reprices Borrow.

The print

The quarter, and why GAAP and adjusted profit differ

Gross profit was $3.235 billion, up 22% and 3.4% above the $3.13 billion guided in August. Cash App's was $2.07 billion (+27%) and Square's $1.16 billion (+14%). By type, commerce enablement, the fees on Square payments, Cash App Card and buy-now-pay-later, was $1.851 billion and accelerated to +19% from +15% and +18%; financial solutions, mostly Borrow, was $1.332 billion and slowed to +32% from +55% and +43%; bitcoin was $52 million, down 50% after Block cut a fee on some bitcoin purchases.

Exhibit 1 — Gross profit by type, five quarters

$0B$1B$2B$3B$2,662M+18%+11%+34%Q3-25$2,872M+24%+11%+51%Q4-25$2,909M+27%+15%+55%Q1-26$3,166M+25%+18%+43%Q2-26$3,235M+22%+19%+32%Q3-26Commerce enablementFinancial solutionsBitcoin ecosystem

How to read it. Millions of dollars of gross profit by quarter, Q3 2025 to Q3 2026, stacked by type: black is commerce enablement, gold financial solutions, light grey bitcoin. The figure above each bar is the total and its growth on a year earlier; the small figures beside the bar are each type's own growth. The one thing it shows: lending's growth came down from +55% in the first quarter to +32%, about where it was a year earlier (+34%), while the fee business accelerated from +11% to +19%; lending still grew faster in every quarter, so its share of gross profit rose from about 38% to 41%.

Square's payment volume grew 14%, the fastest since early 2023: 10.4% in the US and 26% abroad. Cash App had 60 million monthly transacting actives (+4%), 28 million of them using the Cash App Card. Consumer lending originations were $19.0 billion, up 40% on a year and 0.5% on the quarter: Block held Borrow limits flat while it moved the product to six-week loans, which it finished in September, and originations per customer fell 7% in the quarter. The average Borrow limit is $300 against a $1,500 maximum.

Operating income was $428 million, 13% of gross profit and up 5%; adjusted operating income was $923 million, 29% and up 92%. The $495 million between them is mostly $435 million of contingencies, restructuring and other charges, after $365 million in the second quarter and $743 million in the first. Net income to common holders was $566 million, or $0.92 a diluted share, and it includes a $230 million gain on Block's bitcoin, 37.5% of pre-tax income. Adjusted earnings per share were $1.13. Credit costs, transaction, loan and consumer receivable losses, were $602 million, up 66%, or 18.6% of gross profit.

The guide was raised: fourth-quarter gross profit of $3.315 billion (+15%) and adjusted operating income of $1.005 billion (30%); for 2026, $12.626 billion (+22%), $3.519 billion (28%) and adjusted earnings of $4.15 a share (+75%). For 2027 the company gave a preliminary outlook: gross profit up at least 15% to $14.519 billion and adjusted operating income up 30% to $4.575 billion, about 32% of gross profit. Management said "our 2027 growth guidance does not rely on those initiatives," meaning, on the call, Cash App Score, Builderlab and Squaremates; the letter says the same of Neighborhoods and AI sold to other companies. We give them no value either.

The margin

The margin: what 32% is, and how much of it lasts

Bridging the definitions first. July's base case put the normal margin at 20% of gross profit, which is what the adjusted figure was in 2025 ($2.08 billion on $10.36 billion). The two measures are the same: both exclude the one-time charges and both carry stock pay. So the guide and July's base case can be compared directly, and the guide is twelve points higher, three years earlier than July's 24% best case.

Exhibit 2 — The margin three ways, and the 2027 guide

−10%0%10%20%30%40%29%22%13%Q3-25Q4-25Q1-26Q2-26Q3-2631.5%2027 guideJuly's base case, 20%Adjusted operating income (company)Adjusted, less charges outside the DOJ accrual and layoffsGAAP operating income

How to read it. Operating margin as a percent of gross profit by quarter, Q3 2025 to Q3 2026. Gold is the company's adjusted operating income; black is the same figure less the contingencies, restructuring and other charges it excludes, adding back only the DOJ accrual and the February layoffs; grey is GAAP operating income. The gold dash at the right is the 2027 guide, and the dotted line July's 20% base case. The one thing it shows: on the company's figure the margin has risen every quarter; once the charges that recur are taken out, it has risen far less, and it fell in the third quarter.

Where it came from. Operating costs outside credit losses, which we can measure as non-GAAP operating expenses less transaction, loan and consumer receivable losses, were $1,832 million in the third quarter of 2025 and $1,723 million this quarter, down 6% while gross profit grew 22%. In February Block cut more than 40% of staff, in a plan that concluded in the second quarter, and expects $800–900 million of annual savings. Those costs have held flat since: $1,694 million, $1,729 million and $1,723 million in the three quarters of 2026. Some is regrowing in other lines: software and cloud fees rose $94 million in the quarter, partly AI tools, and sales and marketing grew 15%. The 2027 outlook implies that 56% of each added dollar of gross profit reaches adjusted operating income, above the 50% the company set at its investor day for 2028. That is a fixed-cost business growing into a smaller cost base, and it can run for a few years. It cannot run forever, and the company says it will "reinvest profit upside" in Neighborhoods and AI.

The fourth quarter's +15%. It is mostly the comparison. Third-quarter growth of 22% came on 18% a year earlier, a two-year total of 40 points; the fourth quarter's 15% comes on 24%, 39 points. The fourth quarter laps the launch of Cash App Green, and Borrow limits were held flat through the migration. Block beat its third-quarter guide by 3.4% and its adjusted operating income guide by 5.5%. We read the +15% as a cautious guide on a hard comparison, not a slowdown in the business, and our base case grows gross profit 15.9% in 2027, a little above the guide.

Where it lands. We set the terminal margin from a mature peer, not from the guide. Block's gross profit is struck before credit losses; PayPal's revenue less transaction expense and transaction and credit losses, which we derive from its 10-K and call its transaction margin, is struck after them. On a matched basis, adjusted operating income over gross profit less losses was 35% in the third quarter and is about 34% on the 2026 guide. PayPal's GAAP operating income was 36.7%, 36.3% and 39.2% of its transaction margin in 2023–25, 41.4% in 2025 before restructuring, on $15.5 billion of transaction margin. Our base case lifts Block's adjusted margin to 34% in 2029–30, two points above the 2027 guide but below the 35–37% that the company's own 50%-plus incremental margins would give by 2029, because management says it will reinvest upside. It lands at 32% in 2036: 39% of gross profit less losses at this year's loss rate, and 37% after the recurring charges (§3), inside PayPal's range. The bear lands at 25% (31% matched) and the bull at 35% (43%), above PayPal's best year, which is the bull's claim that a company that cut 40% of its staff and automates its engineering can run leaner than PayPal does.

The bill

The bill: one matter, three times the size

July read the first quarter's operating loss as largely non-operating, and the $253.8 million legal accrual inside it as one-time. The filings since show what it was. The DOJ opened its inquiry after a short seller's report in March 2023, presented terms in March 2026, and in October reached an agreement in principle to resolve it through a three-year deferred prosecution agreement on a charge that Block "failed to maintain an effective anti-money laundering program in violation of the Bank Secrecy Act." The gross monetary component is about $964 million and the net payment about $764 million, after credit for earlier regulatory settlements. Block accrued $240 million in the first quarter, $526 million by the second and $764 million by the third. The agreement still needs final terms and court approval, and the 10-Q says losses beyond the accrual are reasonably possible. The SEC closed its parallel investigation in March without recommending action.

Exhibit 3 — The charges the adjusted figure leaves out, 2023 to 2026

$0M$400M$800M$1,200M$1,600M$240M3.2% of GP2023$302M3.4% of GP2024$169M1.6% of GP2025$1,543M4.2% of GP2026 (9 mo.)Other contingencies, restructuring and chargesFebruary layoffs (ex stock pay)DOJ accrual

How to read it. Contingencies, restructuring and other charges in millions of dollars by year, 2023 to the first nine months of 2026, stacked: black is everything except the two 2026 items, grey the February layoffs (excluding their stock pay), gold the DOJ accrual. The percent beside each black segment is that segment over the year's gross profit. The one thing it shows: the DOJ case is one matter, but the black segment has been there every year, at 1.6% to 4.2% of gross profit.

One matter, as July said, but three times the size, and not outside the business. The DOJ accounts for $764 million of the $981 million of accrued legal contingencies at September, against $173 million in December. A failure of an anti-money-laundering program is a failure of how Cash App onboarded and monitored customers, which is an operating failure even if the payment happens once. We score July's read as reversed in part (§6).

What else sits in the bill. The other $217 million of accrued contingencies is not itemized. Known matters: a securities class action covering 2020–24 purchasers, where the court denied Block's motion to dismiss in January and the accrual is "not material" but a larger loss is reasonably possible; derivative suits that the defendants agreed in October to settle, with $150 million paid by insurers to Block less a fee award; a multistate settlement in July for an immaterial amount; and $114 million of San Francisco gross-receipts tax already paid and carried as a deposit. Outside the DOJ accrual and the layoffs, contingencies, restructuring and other charges were $389 million in nine months, 4.2% of gross profit, against 1.6–3.4% in 2023–25.

What we do with it. The accrued claims are unpaid, so we deduct them from equity: $764 million for the DOJ and $217 million for the rest, with no tax shield on the DOJ payment, our assumption (the second-quarter 10-Q describes its litigation and regulatory accruals as non-deductible; the third is silent). We add $112 million for the insurers' payment, assuming a 25% fee award. If the DPA is finalized at the gross $964 million, fair value is $90.85; a further $500 million for the class action gives $90.36.

The charges that recur. Because they have run every year, the model deducts 3% of gross profit through 2029, the term of the DPA, falling to 2% from 2032. The step down is a judgment, not a measurement: 2026 carries the monitoring and remediation years of a federal case and an open class action, and once both are behind it we expect the charges to run between the 1.6% of 2025 and the 3.2–3.4% of 2023–24; 2% is our point in that range. Taking the adjusted figure at face value gives $98.03; 3% throughout, $88.77; 4% throughout, $85.72.

The buyback. Block repurchased 4.2 million shares in January, 5.5 million in February, 1.0 million in March and 0.9 million across April to June, and did not repurchase any shares in the third quarter, with $4.6 billion still authorized. It repaid $1.575 billion of notes in May and June and put $1.4 billion into growing its loan book over the year. The company gave no reason for the pause and no analyst asked. The taper began the month the DOJ presented its terms; we note the timing and draw no conclusion from it. We value the stock on 614.8 million diluted shares, the third quarter's count, with stock pay charged as an expense in the margin, so the count does not grow and no buyback is assumed. A buyback at fair value adds nothing; spending the $4.6 billion at $76.00 would lift fair value to $92.83. The gap between July's 585 million shares and 614.8 million is worth −$4.14 a share on July's own method.

The credit

The credit: two gauges, one longer loan

July set two tells: reserve coverage on consumer loans, and the 60–89-day delinquency rate. This quarter they point opposite ways. Coverage fell to 14.54% from 14.73%; 60–89 days rose to 4.13% from 3.90%, still below the 4.66% of March. The book grew 10.1% in the quarter, to $3.94 billion. Read alone, falling coverage on rising delinquency would mean reserves are not keeping up. The boring checks come first.

The book grew on longer loans, not more lending. Held-for-investment originations rose 1.5% in the quarter, and consumer lending originations as a whole 0.5%. Divide the book by the quarter's originations and the loan length rises from 29.0 days in June to 31.8 in September, as the six-week Borrow loans replaced four-week ones. A longer loan sits on the balance sheet longer, so the same lending makes a larger book and a lower coverage ratio. At June's loan length coverage would be 15.9%. Measured against what was lent, reserves rose from 4.09% of the quarter's originations in March to 4.70% in June and 5.03% in September.

Losses per dollar lent are rising by design. The company published its target economics by product this quarter.

ProductRisk lossAverage durationVariable profit marginAnnualized return
Borrow 4-week2.9%21 days2.0%35%
Borrow 6-week4.2%28 days2.8%37%
Afterpay Post-Purchase4.3%25 days2.7%39%
Afterpay Pay-in-41.0%22 days1.8%30%
Afterpay Pay Monthly5.0%110 days5.2%17%

The six-week loan charges 7.5% against 5%, targets a 4.2% loss against 2.9%, and earns 2.8% of each loan after losses and processing against 2.0%. The annualized return is that margin times the number of times the money turns over in a year, 2.8% × 365 / 28 = 37%, against 35% for the four-week loan, so most of the higher margin is the longer loan, not a better one. Afterpay Post-Purchase, which more than tripled, targets a 4.3% loss. As the mix moves to these products, losses per dollar lent should rise toward about 4%. Reserves per dollar lent and net write-offs over the prior quarter's originations (2.39%, 2.89% and 3.11%) have risen every quarter; provisions went from 3.14% to 3.57% of the quarter's originations and eased to 3.47% in the third. One caution on levels: the originations line covers all loans held for investment, including commercial and other loans, which are 22% of that book but, being longer loans, a much smaller share of what is lent; we estimate about 5%. The consumer ratios are therefore about 5% higher than shown, net write-offs about 3.3% rather than 3.1%, and we compare them with the targets as a trend first and a level second. 60–89 days over the prior quarter's originations ran 1.21%, 1.54%, 1.31% and 1.45%: up this quarter, below March.

Exhibit 4 — The credit gauges, on the book and on the lending

0%5%10%15%20%On the book (stock)14.54%4.13%Q4-25Q1-26Q2-26Q3-260.0%1.5%3.0%4.5%6.0%On the lending (flow)5.03%3.11%1.45%Q4-25Q1-26Q2-26Q3-26Gold: reserve (allowance) · Ink: 60–89 days past due · Grey: net write-offs. Solid: over the loan book. Dashed: over originations (prior quarter for write-offs and 60–89 days).

How to read it. Cash App consumer loans held for investment, four quarter-ends, December 2025 to September 2026, in percent. Left, over the loan book: gold is reserve coverage, black 60–89 days past due. Right, over originations: gold is the reserve over the quarter's originations, grey net write-offs over the prior quarter's originations, black 60–89 days over the prior quarter's originations. The one thing it shows: on the book, coverage fell; measured against what was lent, reserves and write-offs have risen every quarter, which is what a move to longer, higher-fee, higher-loss loans should do.

Seasonality. The company's chart of Borrow returns by month of origination, for customers past their third month, shows the weakest loans are made in March to May, at roughly 22–25% annualized in 2025, and the strongest in September, October and January, at 36–41%. The 2026 cohorts with actual results, January to May, ran above the 2025 cohorts in every month but January. Borrow loans are written off at 90 days past due, so most third-quarter write-offs should come from loans made in the second quarter, the seasonally weakest; that is our inference, as the filings give no split by vintage. August and September are the season's best months, and the company projects them above 40%; those projections include its own underwriting model's forecasts, not results.

Our read. Nothing in the quarter shows losses running above what the products are priced for: net write-offs of about 3.1–3.3% of originations sit below the 4.2–4.3% targets of the products growing fastest (the targets are product loss rates for customers past their first month, so the comparison is approximate), and the company says variable profit margin was in line with a year ago and up modestly on the quarter. Nor does anything show they will not: the six-week loans have not been through a full season, and the targets are portfolio averages for customers past their first loan. The single reading that would settle it is net write-offs on consumer loans in the fourth quarter over third-quarter originations, in the 10-K in late February. At 3.2% or less the season's best loans are behaving as projected. At 4.0% or more they already sit near the ceiling of the new products' targets, and we would move the bear case's weight from 25% to 35% (fair value $84.84).

How long lending can compound. Borrow's average limit is $300 against a $1,500 maximum, and the company's six-week buy-now-pay-later products already go to $3,000; it "expect[s] to increase average limits over the next several years." A 50% rise in the average limit over four years is about 10% a year; with low-single-digit customer growth and new products that gives 11–15% a year. We set a four-year stage, 2027–30, at 15.5%, 13%, 11% and 9% in the base case, then a fade to 4% by 2036. Square Loans grew 27% a year for a decade at its 4% loss target, the company's own base rate, which we do not assume repeats. The funding has changed too: warehouse lines are 66% drawn ($1.54 billion of $2.35 billion, drawn plus available), and Square Financial Services began issuing brokered certificates of deposit this quarter, $381 million at 4.09%, beside $521 million of customer deposits. We treat those lines and deposits as operating funding of the loan book and charge the rest of its growth to equity: of the $7.49 billion lending book, $2.45 billion is funded that way and 67% by Block itself. The base case charges 60% of each year's book growth to equity.

The value

What it is worth

The method. July multiplied 2027 earnings by 15, 21 or 26. We value the business by its parts instead, because three things that matter now sit outside one multiple: the loan book that grows with lending and needs equity, the charges the adjusted figure leaves out, and a balance sheet that now carries a $764 million unpaid claim. We discount after-tax operating profit less the equity that loan growth needs, by scenario, mid-year from 3 November 2026, with the fourth quarter of 2026 as a stub, stock pay expensed, and the balance-sheet items below added once. We use the tech stack's method, not the financials stack's: the lending book is $7.49 billion against $22.9 billion of equity, the consumer loans in it turn over about every 32 days, and 57% of gross profit is commerce enablement, mostly fees on payments and cards. Block reports no regulatory capital ratio for the group, so the book's capital is charged as reinvestment in the cash flows rather than valued as a bank's equity.

Gross profit by line. Square grows with its payment volume; Cash App commerce with card and buy-now-pay-later attach; Cash App financial solutions with Borrow's limits and eligibility; bitcoin and other at a small rate. Each stage's length comes from what drives it.

How long each part of the business grows faster than normal.

BusinessStageYearsWhat drives itWhat sets the length
SquareDistribution build3New sellers from self-onboarding, field sales and 400+ ISO partners; international GPV +26%; more than half the US addressable market still on legacy systems (company estimate)Three years, 2027-29: the company guides GPV acceleration for a third year in 2027 and reports five-quarter paybacks on new-seller spending; management said on the call that the ISO channel went from zero to 400 partners in the past year, so the channel's ramp, not the market, sets the length; after 2029 the base assumes Toast/Clover-style competition caps share gains
SquareShare maturity7GPV growth converges on nominal US consumer spending plus slow international share gainsSeven years, 2030-36, a straight-line fade from 11.5% to the 4% terminal rate; Fiserv's merchant business, the mature comparison, grows low-to-mid single digits
Cash App commerce (card and BNPL)BNPL attach3Afterpay on Cash App Card and Post-Purchase: 10% of card and Pay actives use BNPL, with about 3x the gross profit of those who do not; monetization rate 1.56% to 1.74% in a yearThree years, 2027-29, a judgment: BNPL attach is 10% of card and Pay actives and monetization rose 18bp in a year; at that pace attach roughly doubles in about three years, after which card economics set the growth; actives grow only low single digits, so attach, not users, sets the length
Cash App commerce (card and BNPL)Fade to card economics7Volume growth with spending once attach saturatesSeven years, 2030-36, straight line to 4%
Cash App financial solutions (Borrow)Limits and eligibility4Average Borrow limit $300 against a $1,500 maximum (BNPL products already go to $3,000); six-week loans at a 7.5% fee; Cash App Score sold to other lendersFour years, 2027-30: management expects to raise average limits 'over the next several years'; a 50% rise in the average limit over four years (~10% a year) on low-single-digit customer growth gives ~11-15% a year; Square Loans held a 4% loss target while compounding 27% a year for a decade, the company's own base rate, which we do not assume repeats
Cash App financial solutions (Borrow)Fade to consumer-credit growth6Loan growth with income of the customer baseSix years, 2031-36, straight line to 4%
Gross-profit growth202720282029203020322036
Square, bull16%15%14%12%9.3%4.0%
Square, base14%13%12%10%8.3%4.0%
Square, bear9.0%7.0%6.0%5.6%4.7%3.0%
Cash App commerce, bull23%19%16%13%10%4.0%
Cash App commerce, base19%15%12%11%8.6%4.0%
Cash App commerce, bear8.0%5.0%5.0%4.7%4.1%3.0%
Cash App financial solutions, bull22%19%16%13%10%4.0%
Cash App financial solutions, base16%13%11%9.0%7.3%4.0%
Cash App financial solutions, bear−8.0%−2.0%4.0%3.9%3.6%3.0%

The three cases. The base grows gross profit 15.9% in 2027 to $14.63 billion and to $29.6 billion by 2036. The bull (19.8%, $35.5 billion) raises Borrow limits toward the buy-now-pay-later ceiling and lands the margin at 35%. The bear (3.1%, $18.5 billion) is a credit turn in 2027: losses on the newer products run above target, Block cuts limits, financial solutions falls 8%, Square slows with small-business spending, and the margin gives back six points. Terminal growth is 4.0% in the base and bull and 3.0% in the bear, from the Federal Reserve's longer-run projections of 16 September 2026 (medians: 2.0% real growth, 2.0% inflation).

Bear (25%)Base (50%)Bull (25%)
Gross profit 2027 (growth)$13.01B (+3.1%)$14.63B (+15.9%)$15.12B (+19.8%)
Gross profit 2030$14.8B$20.3B$23.0B
Gross profit 2036$18.5B$29.6B$35.5B
Adjusted margin 2027 / peak / 203622.0% / 24% / 25%31.5% / 34% / 32%33.0% / 37% / 35%
Less recurring charges, 2027 → 2032 on3.0% → 2.0%3.0% → 2.0%3.0% → 2.0%
After-tax operating profit 2027$1.88B$3.17B$3.45B
Equity share of loan-book growth70%60%55%
Terminal growth / return on new capital3% / 15%4% / 30%4% / 40%
Terminal value, share of operating value45%55%58%
Value a share$47.12$96.64$124.30
The weights (unchanged at 25 / 50 / 25, argued on their own). July's weights were set when the credit question was open. It is still open: this quarter shows losses rising by design and inside the targets, not a breach and not a clean bill. A tilt toward the bear needs a result above the targets; a tilt toward the bull needs a season of the six-week loans at or below them. Neither has happened, so the base stays the central case. §7 sets both readings: bear to 35%, or bull to 30% with the bear at 20%. Weights of 35/45/20 give $84.84; 20/50/30, $95.03.

The balance sheet. Net corporate cash is now positive: Block repaid $1.575 billion of notes this year and holds $8.4 billion of cash and marketable securities against $5.775 billion of notes. Two adjustments July did not make. Customers payable exceed customer funds and settlements receivable by $889 million, which is sellers' money held in corporate cash (risk holds, unlinked accounts), so we deduct it. And $1.31 billion of deferred tax assets will lower cash taxes, which the model charges at a full 24%; we add them at present value.

Item$ million$ a share
Cash and marketable securities less $5.775 billion of notes2,668+4.34
Customer balances funded from corporate cash−889−1.45
DOJ accrual (net payment in the agreement in principle)−764−1.24
Other accrued legal matters−217−0.35
Insurers' derivative payment, less an assumed 25% fee award112+0.18
Bitcoin, 9,237 at $76,934711+1.16
Other equity stakes105+0.17
Deferred tax assets, present value916+1.49
Total2,642+4.30

The cost of equity: 13.0% to 2029, 10.5% by 2036 (−$10.10). July stated no rate; its 21 times 2027 earnings implies about 8.1% at 4% growth. Block's own beta is high and precise: 2.56 over 60 months (standard error 0.34), 2.47 over 36, 1.95 over 24 and 1.34 over the last 12. Blended toward the median of PayPal, Fiserv, Toast, Affirm and SoFi it is 2.29, a 15.3% cost of equity on the 4.95% 10-year Treasury and a 4.5% equity risk premium. We do not use that read alone, for two reasons. The 60-month window was measured while adjusted operating income ran 4–20% of gross profit; at 28–32% the same swing in gross profit moves profit about a third less (gross profit was 5.0 times adjusted operating income in 2025 and is guided at 3.2 times for 2027), which scales 2.56 to about 1.64. And weighting peers by Block's mix, 57% commerce enablement, mostly payment and card fees, at the 1.38 median of PayPal, Fiserv and Toast, and 41% financial solutions, mostly lending, at the 1.83 median of Affirm, SoFi, Capital One and Synchrony, gives about 1.57, or 12.0%; buy-now-pay-later sits in commerce enablement, so that read understates the credit share. We set 13.0%, a beta of about 1.79: above both business-based reads, because the newer loans carry more risk than the old mix, and below the 24-month read. It converges in a straight line from 2030 to 10.5% by 2036, a beta of 1.24, the median of PayPal, Fiserv, Capital One and Synchrony. At 12.0% fair value is $95.05; at the 24-month read's 13.7%, $88.58; at 15.3%, $83.02; a terminal rate of 11.1% gives $86.01.

The fall to 10.5% carries more than the whole gap to the price. Held at 13.0% for every year, fair value is $74.09; 12.8% held flat gives the $76.00 close, and 11.6% the $87.63 close of 4 November. We set the fall from the business, not the calendar. Payment and card fees earn like PayPal and Fiserv; the lending earns like Capital One and Synchrony, whose betas carry the leverage of deposit-funded banks, which Block's mostly equity-funded loan book does not. The operating-leverage read above, about 1.6 at the margins Block now earns, already sits closer to the end point than to the 60-month beta, and the shorter windows are lower too: 1.95 over 24 months and 1.34 over 12. Against that, financial solutions, mostly lending, has risen to 41% of gross profit and the newer products are young. The end point is the median of four peers, and one swap moves it: with Toast in place of Fiserv the median is 1.38, an 11.1% terminal rate, $86.01.

From $83 to $91.17. We first re-ran July's model, which reproduces $83.42, then applied one change at a time, so the lines sum to the move.

Exhibit 5 — From July's $83 to $91.17, one change at a time

$0$20$40$60$80$100$120$140$83July−1.1Grossprofit+4.6Balancesheet−1.5Legalclaims−4.1Sharecount+36.12027outlook−3.1Stagedcash flows−7.6Recurringcharges−5.6Lendingcapital+0.1Cash andtax assets−10.1Cost ofequity+0.0Weightsoutlookmethodrateweights$91Nowclose $76.00Fair value, $ a share; each bar is one change applied to the running value, the quarter's data first

How to read it. Fair value in dollars a share. The black bar is July's value; each bar after it is one change applied to the running value, gold for a rise and grey for a fall, labeled with its effect; the gold bar at the end is today's value. Dashed verticals separate the quarter's data, the 2027 outlook on July's frame, the method changes, the cost of equity and the weights; the dashed horizontal is the $76.00 close. The one thing it shows: the data alone barely move July's value; the margin outlook adds about 44% to it, and the method and the cost of equity take back about 73% of that.

ChangeKind$ a shareFair value after
Gross profit: trailing twelve months to September, July's rates held, FY2027 kept as the yeardata-1.0582.36
Balance sheet: net cash after $1.575bn of notes repaid, bitcoin at the 3 Nov spotdata+4.6386.99
Legal: the DOJ accrual and other accrued matters, less the insurers' paymentdata-1.4885.51
Share count: 585M forward to 614.8M diluteddata-4.1481.36
The 2027 outlook on July's frame: each case's 2027 gross profit and adjusted margin, July's multiplesassumption+36.10117.46
Method: staged cash flows by business line replace a 2027 earnings multiple (at the 10.5% terminal rate throughout)method-3.10114.36
Recurring charges the adjusted figure excludes: 3% of gross profit to 2029, 2% from 2032method-7.58106.79
Lending capital: growth in the loan book funded by equitymethod-5.64101.14
Customer balances funded from corporate cash (-) and deferred tax assets at present value (+)method+0.14101.28
Cost of equity: 13.0% to 2029, straight line to 10.5% by 2036cost-of-equity-10.1091.17
Scenario weights: 25/50/25 kept, argued on their ownweighting+0.0091.17
Fair value 83.42 → 91.17+7.7591.17

Each row applies one more change to the running value, in the order shown. The first four rows are the quarter's data on July's method; a different order would change the split, not the total. July's published $83 reproduces at 83.42.

In short: the quarter's data on July's method give $81.36; the 2027 outlook on July's frame adds +$36.10; valuing the parts, the recurring charges, the loan book's equity and the cost of equity take most of that back. July's 21 times on our base case's 2027 profit would give $112.58, and that multiple implies perpetual growth of about 6% at our terminal rate, above the economy's; we show it as a cross-check, not a weight. PayPal's 10.8 times, on the same arithmetic (reinvestment held at the terminal 30% return on new capital), implies about 2.4% perpetual growth from 2027, below the economy's: it prices Block as a business that stops growing after next year, which is not what its guide or its last five quarters show.

Return on capital. We measure it on the line the model values: adjusted operating income less the recurring charges, taxed at 24%, over equity plus notes less cash and bitcoin. Customer funds are not capital: they sit outside cash and are matched by what Block owes customers. The loan book is, because Block funds most of it; the warehouse lines and deposits that fund the rest are operating funding, and their interest runs through operating costs. On that basis the return is 10.0% against a WACC of 12.1%, up from 6.4% a year ago: the regime is recovering.

Exhibit 6 — Return on capital against its cost, twelve quarters

0%4%8%12%16%WACC 12.1%cost of equity 13.0% (reference)Mar '24Sep '24Mar '25Sep '25Mar '26Sep '2610.0%New capital63% earned30% assumedReturn on capital, trailing twelve months: adjusted operating income less recurring charges, taxed at 24%, over average capital

How to read it. Trailing-twelve-month adjusted operating income less the recurring charges (the DOJ accrual and the layoffs added back), taxed at 24%, over average capital (equity plus notes less cash and bitcoin), December 2023 to September 2026, in percent. The dashed line and shaded band mark the 12.1% WACC; the dotted line is the 13.0% cost of equity, a reference. The two bars at the right, on their own scale, compare the return earned on capital added across the window with the return on new capital the base terminal value assumes. The one thing it shows: the return has climbed every quarter since early 2025 and is about two points short of the WACC.

The WACC sits 0.9 points under the cost of equity because notes are 11% of capital at market value and Block holds more cash than notes. The return stays below it because $13.1 billion of goodwill and acquired intangibles, mostly the Afterpay purchase, sits in capital; without it the return is 28.7%. On the company's adjusted figure it would be 11.7%, and with every charge counted 5.7%. The base case crosses the WACC in 2027. Over the window, profit rose about $1.9 billion on $3.1 billion of added capital, 63% on new capital. That figure overstates what new capital earns: most of the gain came from cutting costs once, which lifted the return on capital Block already had. The base case's 30% rests instead on the return without goodwill, 28.7%, because the capital growth needs from here is mostly the loan book.

What the price requires. At $76.00, with the base case's margins and rates, the price takes gross profit growth of 5.7% a year for ten years, against at least 15% guided for 2027; a base case growing only 4% from next year would be worth $68.45. Or a base margin landing at 22.5% instead of 32%. Or margins about 5.0 points lower than ours in every year and case; the $87.63 close of 4 November needs about 1.2 points. Or a cost of equity held flat at 12.8%; the rate to 2029 alone would have to reach 17.5%, because those years carry little of the value. Inputs can also move together: a 15.3% cost of equity to 2029, charges at 4% and margins 2 points lower give $72.56; with the 24-month beta's 13.7% in place of 15.3%, $77.38. Terminal value is 55% of the base case's operating value (45% bear, 58% bull). The $76.00 close was 14.7 times our base case's 2027 after-tax operating profit of $5.16 a share; our value is 17.7 times; PayPal trades at 10.8 times trailing earnings.

The receipt

The receipt: July's tests and claims, scored

July set two credit tests and made claims this print can check. Both tests are retired: neither fired, and the ratio they read has changed meaning now that loans are longer.

holds reversed or superseded not fired, not met or re-tested
NOT FIREDBorrow reserve coverage past the mid-teens on a flat or falling book, and/or 60-89 days past due above 4.66%

Coverage 14.54% (from 14.73%), under the ~15% bar, on a book that grew 10.1% in the quarter, so neither condition is met; 60-89 days past due 4.13%, under the 4.66% of March but up from 3.90%, so the caveat metric rose this quarter. The book grew on longer loans, not more lending: originations rose 1.5% while days outstanding went from 29.0 to 31.8. At June's days outstanding, coverage would read 15.9%, and reserves per dollar lent rose from 4.70% to 5.03%. Retired; replaced by xq3-net-writeoffs and xq3-dpd-flow, which read the flow, not the stock.

NOT METTwo consecutive quarters of stable-or-falling coverage and receding 60-89 days, with Square near +9% and ARPU closing the bank gap

Coverage: June up (13.18% to 14.73%), September down (14.54%). 60-89 days: June down (4.66% to 3.90%), September up (4.13%). No two consecutive quarters had both. Square: +13% and +14% against the ~9% bar, beaten. ARPU: Cash App gross profit per active $130 on the trailing year against $107 for 2025 (+21%), but primary banking actives went 9.7M, 9.4M, 9.4M and inflows per transacting active grew 6%, so the gain came from lending and BNPL, not from banking. The trigger's $110-130 did not arrive by its route; fair value moved to $91.17 through the margin outlook instead, with the credit legs still open. Retired.

HOLDS, SMALLERJuly: bitcoin is ~4% of gross profit and fair value barely moves with it

Bitcoin ecosystem gross profit was $52M of $3,235M in Q3 (1.6%) after a deliberate fee cut, 2.5% on the trailing year. The 9,237 BTC are $711M at the 3 Nov spot, $1.16 a share. The quarter's $230M bitcoin gain was 37.5% of pre-tax income, which is why GAAP earnings are not the measure.

HOLDS, REFINEDJuly: value gross profit, not revenue

Still right about revenue (bitcoin pass-through is 27% of Q3 revenue). Refined: gross profit is struck before credit losses, which took 18.6% of Q3 gross profit. Margins here are compared with peers on gross profit less losses (PayPal's transaction margin is struck after them).

SUPERSEDEDJuly: the tripled loss line is CECL front-loading on +175% origination growth

Originations grew 1.5% in the quarter, yet provisions were still 3.47% of the quarter's originations. Timing no longer explains the loss line: the company moved Borrow to six-week loans with a 4.2% target risk loss against 2.9% for four-week loans, and Post-Purchase targets 4.3%. Higher losses are now the design, priced in a 7.5% fee.

REVERSED IN PARTJuly: the Q1 operating loss is largely non-operating; the $253.8M legal accrual is one-time

The Q1 accrual was the first of three for one matter: the DOJ's Bank Secrecy Act charge, $240M, then $526M, then $764M (agreement in principle, October; a three-year deferred prosecution). One matter, as July said, but three times the size, and an anti-money-laundering failure is an operating failure of Cash App's controls. Charges outside the DOJ and the layoffs still ran $389M in nine months (4.2% of gross profit), against 1.6-3.4% in 2023-25, so the model deducts recurring charges the adjusted figure leaves out.

RE-TESTEDSynthesis: XYZ ~$83 vs $78.72 (fair); BTC beta ~0.41; bitcoin ~4% of gross profit

Fair value now $91.17 against $76.00; 90-day beta to bitcoin 0.28 (2026-06-26 to 2026-11-03); bitcoin 2.5% of trailing gross profit. The synthesis's ranking (Block the least crypto-coupled of the three) holds and the coupling fell further.

The tripwires

What carries the story to the next update

3.2% / 4.0%net write-offs on consumer loans in Q4 over Q3 originations: at or below the first the season's best loans behave as projected; at or above the second the bear weight goes to 35%

Each tripwire has its reading rule set now, so the next update scores it rather than reinterprets it. Effects are on fair value at today's other inputs.

  1. Net write-offs on consumer loans, Q4-26, over Q3 originations. Q3 read 3.11% on June's originations, which carried the seasonally weakest Borrow cohorts. August and September are seasonally the strongest months, and the company projects them above 40% annualized. <= 3.2%: consistent, no change. 3.2-4.0%: hold, read Q1-27. >= 4.0%: the season's best cohorts already sit near the six-week products' 4.2-4.3% target ceiling; bear weight 25% -> 35% ($84.84). Reading: 10-K (late Feb 2027), Note 'Customer loans': (write-offs - recoveries) on consumer loans held for investment / Q3-26 'purchases and originations of loans originally classified as held for investment' ($11,396M).
  2. 60-89 days past due over prior-quarter originations, Q1-27. Q1-26 read 1.54% on a mostly four-week book. The six-week products target 45% more risk loss (4.2% vs 2.9%), so up to ~2.2% is the design. > 2.2%: losses outrun the price of the loans; bear weight to 35% ($84.84). < 1.6%, with Q4 net write-offs at or below 3.2% (xq3-net-writeoffs): the six-week loans have been through a season at or below target; bull weight 25% -> 30% and bear 25% -> 20% ($95.03). Reading: Q1-27 10-Q (May 2027): consumer loans 60-89 days past due / Q4-26 held-for-investment originations.
  3. Final DOJ deferred prosecution agreement. Final at a net $764M: no change. Each $100M above: -$0.16 a share (gross $964M: $90.85). A monitor or added compliance spend named in the agreement: recurring charges 3% -> 4% through 2029 (4% throughout: $85.72). No final agreement by the Q1-27 10-Q, or DOJ action: bear weight to 35%. Reading: 8-K or 10-K: final DPA terms and court approval.
  4. Buyback resumption after the DOJ resolution. Repurchases tapered from the month the DOJ presented terms and stopped through the talks; the company gave no reason (January 4.2M shares, February 5.5M, March 1.0M, April-June 0.9M, July-September none). Resumption of $500M+ in a quarter below fair value: the $4.6bn authorization spent at $76.00 would lift fair value to $92.83. No repurchase in the two quarters after a final DPA: read the cash as needed for lending; equity share of book growth 60% -> 100% ($87.83). Reading: 10-K / 10-Q Item 2 issuer purchases.
  5. February 2027 guidance for 2027. Adjusted operating margin guided below 30% (preliminary 32%) or gross profit below +13%: base margin path down 2 points ($85.08). Margin 33% or more with gross profit +15% or more: up 2 points ($97.27). Reading: Q4-26 shareholder letter (24 Feb 2027).
  6. Charges outside the DOJ accrual and the layoffs, FY2026. Nine months ran 4.2%. Full year above 4%: recurring path to 4% ($85.72). Below 2%: 2% path ($91.81). Reading: 10-K reconciliation: contingencies, restructuring and other charges, less the DOJ accrual and Workforce Plan charges, over FY26 gross profit.
  7. 24-month beta at the next update. 24-month beta 1.95 now. Below 1.6: forecast-year rate 13.0% -> 12.0% ($95.05). Above 2.3: 15.3% ($83.02). Reading: capm.py / beta_windows.py at the Q4 print (monthly returns on SPY).
  8. Does the risk come down as the business matures (the fall from 13.0% to 10.5%). Fair value rests on the cost of equity falling from 13.0% to 10.5% by 2036; held at 13.0% it is $74.09, and a flat 12.8% gives the 3 Nov close. The 24-month window to June 2027 is measured mostly at the post-layoff margins (24-month beta 1.95 now, SE 0.54, with 2025 months at ~20% margins in it). Above 2.0: the rate stays at 13.0% through 2036 ($74.09). Below 1.6: the fall stands. Between: no change; read again at the Q4-27 update. Reading: beta_windows.py at the Q2-27 update: 24-month beta on monthly returns vs SPY, window to June 2027.
  9. Price against the case values. Close below the bear $47.12 or above the bull $124.30: re-underwrite the cases before the next print. A close at or above $91.17 for 20 sessions: the price sits at fair value; no change to the model. Reading: settled close (Robinhood close.price / Schwab closePrice).

What would change our view

We would move the bull path toward the base if the evidence tracks it: 2027 guided in February at a 33% adjusted margin or more with gross profit +15% or more; fourth-quarter net write-offs at or below 3.2% of third-quarter originations and the first quarter's 60–89 days below 1.6% of fourth-quarter originations; a final DPA at or below $764 million with no monitor; and buybacks resumed. That path is worth $124.30, and its inputs then become the base. The mirror moves it the other way: a 2027 margin guided below 30% or gross profit below +13%; fourth-quarter net write-offs at 4.0% or more, or first-quarter 60–89 days above 2.2% of originations; a DPA above the accrual or with a monitor, or none by May; charges outside the DOJ above 4% of 2026 gross profit. Any one of these moves its own input as §7 sets out; together they move the bear weight to 35% and the margin down two points, toward $85.08 and below, and the bear case is worth $47.12. The price is right, and we would say so, if the margin path sits about 5.0 points below ours in every year, which means a 2027 margin guided in the mid-to-high 20s rather than above 30%, or if Block's risk does not come down as the business matures: a cost of equity of 12.8% held in every year gives the price. The margin would show in February's guide. The risk would show in the 24-month beta at the August 2027 update, a window measured mostly at the new margins: above 2.0, the rate would stay at 13.0% through 2036, worth $74.09; below 1.6, the fall stands.

The call

The call

Block's margin arrived three years ahead of July's base case, and its bill turned out three times larger and closer to the core of the business than July read it. We value the stock at $91.17, from $83, against $76.00 before the results and $87.63 the day after. The value is a fixed-cost business growing into a smaller cost base, landed at PayPal's mature margin on a matched basis, less the charges that recur, the equity the loan book needs and the DOJ payment, discounted at 13% falling to 10.5% as the business matures. That fall is the judgment the call rests on: held at 13%, the value is $74.09, below $76.00. The price asks for gross profit growth of 5.7% a year, about a third of what the company guides for 2027. Credit is the condition: losses per dollar lent are rising toward the new products' targets by design, and the fourth-quarter write-offs on the season's best loans, reported in February, are the reading that would move our weights.

Sources

Sources & Method

Filings. Block Form 8-K Exhibit 99.1, Q3 2026 shareholder letter (acc. 0001628280-26-069513, filed 3 November 2026): income statement, balance sheet, cash flows, segment and type tables, operating metrics, reconciliations, guidance and the 2027 preliminary outlook. Q3 2026 Form 10-Q (acc. 0001628280-26-069524, filed 3 November): Notes 3–19, including customer loans and receivables, accrued legal contingencies, brokered certificates of deposit, bitcoin, indebtedness and warehouse facilities, repurchases, the DOJ, class-action and derivative matters (Note 16), restructuring (Note 19) and $2.2 billion of cloud commitments ($475 million in 2027, about 3% of 2027 gross profit; $392 million after 2030), spending the margin path already carries. Q1 and Q2 2026 10-Qs (0001628280-26-032200, 0001628280-26-053368) for the credit series, the $240 million and $526 million DOJ accruals and monthly repurchases; FY2025 10-K (0001628280-26-012254); Q3 2025 10-Q; shareholder letters from November 2023 to August 2026 for adjusted operating income and charges by quarter. PayPal FY2025 10-K (0001633917-26-000024) for revenue, transaction expense, transaction and credit losses and operating income, from which we derive its transaction margin.

Company materials. Block Q3 2026 investor presentation (target unit economics by product, p. 13, read from the page image; Borrow returns by origination month, p. 12) and prepared remarks. A draft OCC charter application for Builders Bank & Trust, a proposed national trust bank for digital-asset custody, dated 4 September 2026 and posted on Block's investor site, is noted and given no value.

The call. 3 November 2026, 5:00 PM ET. Quotes are taken word for word from two independent transcriptions, Motley Fool and Benzinga (via Webull), and each quoted sentence appears in both (archived in data/tells/transcript.md); quotes from the letter and 10-Q are from the filings. No analyst asked about the DOJ, the legal accruals, delinquencies or the buyback.

Market data. XYZ settled close 3 November 2026 ($76.00), confirmed against two brokerage feeds; the 4 November close ($87.63, high $88.51) and the 5 November close ($84.04) are reported, not inputs. FRED DGS10 (4.95% on 3 November) and Coinbase bitcoin price ($76,934 on 3 November). Month-end closes for XYZ, SPY, PayPal, Fiserv, Toast, Affirm, SoFi, Capital One and Synchrony, split-adjusted, from the brokerage feed, for the betas; PayPal's trailing multiple is its 3 November close over its FY2025 earnings per share ($5.41) from the same feed. Ninety-day beta of XYZ to bitcoin 0.28, from daily closes.

Method. Free cash flow to equity by scenario, weighted 25/50/25: gross profit by line from the house 2026 base (Q1–Q3 reported, Q4 the company's guide split by line), adjusted operating margin by scenario, less recurring charges of 3% of gross profit through 2029 falling to 2% from 2032, taxed at 24%, less the equity share of loan-book growth (60% base, 70% bear, 55% bull). Mid-year discounting from 3 November 2026; cost of equity 13.0% to 2029, straight line to 10.5% by 2036; terminal value by the Gordon formula on 2037 after-tax operating profit, less reinvestment of growth over the return on new capital (30% base, 15% bear, 40% bull). Balance-sheet items as tabled, on 614.8 million diluted shares. July's model is re-run in data/july_engine/ and attributed one change at a time. Credit ratios use consumer loans held for investment over the cash-flow line for held-for-investment originations, which also includes commercial and other loans (22% of that book, an estimated 5% of originations).

Limitations. Fourth-quarter 2026 figures by line are house estimates on the company's guide. Growth paths, margins, stage lengths, recurring charges, the equity share of loan growth and the fee-award assumption are house judgments, shown with their effects. The unit economics are company targets, not results, and the cohort returns include the company's own projections. The DOJ agreement is not final. The beta reads are statistical estimates with wide error bands.

Disclosures

Information only. TON618 Capital. This report is for information purposes only. Nothing here is an offer to sell or a solicitation of an offer to buy any security, fund interest, or digital asset, and nothing here is personalized investment advice or a recommendation regarding any instrument.

Publisher's exclusion. All research is published solely as general, impersonal information of regular circulation. It is not tailored to the objectives or circumstances of any individual and is not issued in connection with compensation from any client. The Fund has no clients and distributes all research free of charge. On that basis it publishes in reliance on the publisher's exclusion from the definition of "investment adviser" under the Investment Advisers Act of 1940 (§202(a)(11)(D); cf. Lowe v. SEC, 472 U.S. 181 (1985)).

Registration & conflicts. TON618 Capital is not registered as an investment adviser or broker-dealer in any capacity. The Fund is a Bitcoin fund and may hold or transact in the securities or digital assets it discusses; material conflicts are disclosed where they exist. Ownership: the Fund holds no position, long or short, in XYZ or in any other security named in this report as of the report date, and has no economic interest in their price. The Fund is long bitcoin; Block holds 9,237 bitcoin, about $1.16 of this report's fair value a share. Compensation: the Fund received no compensation from any party in connection with this report and charges nothing for it.

Use of AI. Artificial intelligence is used in the creation of this research. All methodology and data integrity are reviewed and approved before publication by TON618 Capital's Chief Investment Officer, Keyth Beck; errors may nonetheless occur, and readers should verify independently.

CFA. This report was prepared to align with CFA Institute analytical standards (methodology only). CFA® and Chartered Financial Analyst® are registered trademarks owned by CFA Institute. That reference describes the analytical framework applied; it does not imply the report was prepared, reviewed, or authored by a CFA charterholder, and the report is not issued, reviewed, endorsed, certified, or approved by — nor affiliated with — CFA Institute.

Risk & feedback. Past performance is not indicative of future results. Digital assets and equities are volatile and may result in total loss of capital. Corrections and feedback are welcome — please direct them to CIO Keyth Beck at keyth@ton618capital.com. Version 1.0.