PayPal Holdings

01 / The central investment debate

The opportunity: a still-profitable franchise valued at roughly 10× earnings can generate attractive per-share returns without returning to pandemic growth. The risk: growth in lower-margin processing disguises declining relevance in higher-value branded checkout.

Q2 revenue grew 4.8% and total payment volume 9.7%, but transaction-margin dollars rose only 1.5%. Branded online checkout grew just 2% FX-neutral; GAAP operating margin fell to 16.4%. More payment volume is not yet producing proportionately more profit.[S2][S3][S6]

The business generated $34.13bn trailing revenue, $4.90bn net income and $5.29 diluted EPS. From 2021–2025, revenue grew 6.9% annually, EPS 11.3% and headline free cash flow only 3.3%. Buybacks, not just operating growth, explain the per-share progress. At the reference price, equity value is approximately $46.03bn and conventional enterprise value $45.90bn.[S1][S2][S4][S5][M1]

Expectations: FY2026 company guidance calls for adjusted EPS around $5.38, adjusted FCF of at least $6bn and roughly $6bn of buybacks. The 5 August consensus snapshot estimated 2027 revenue of $36.18bn and adjusted EPS of $5.79; this is not company guidance or a live September consensus. Our valuation uses normalised earnings, not that adjusted-EPS series.[S3][S8][M1]

02 / Competitive position

PayPal has recognised brands, stored payment credentials, merchant acceptance and risk data. But consumers can choose another wallet and merchants can use several processors. It does not deserve Visa’s or Mastercard’s multiple simply because it operates in payments.[S4][M1]

Sources: company filings/releases and the original market snapshot.[S2][S10][S11][S12][M1] *Block’s GAAP earnings are depressed; gross-profit growth of 24.8% is more informative than bitcoin-inclusive revenue. Visa’s cash period is nine months; the others are six. Revenue definitions also differ.

Adyen and Stripe add competitive pressure through enterprise processing and developer distribution. Venmo offers PayPal a monetisation opportunity, but engagement is valuable only when revenue exceeds incentives, funding and credit costs. The base case assumes gradual improvement—not a new, unproven profit windfall.[S6][S13][S14][M1]

03 / Earnings are not distributable cash

The owner-cash bridge is $6.59bn headline FCF − $1.22bn full share-award costs − $0.65bn net loan funding = $4.71bn. The lower $4.11bn normalised starting cash flow avoids extrapolating favourable tax timing and unusually strong conversion.[S2][S4][M1]

Gross buybacks represented 13.1% of equity value, while Q2 average diluted shares fell 9.7% year-on-year. Gross spending overstates net retirement because employee awards create shares. Historical repurchases are not assumed sustainable: the base model funds $4.1–5.5bn annually from available cash, without new net debt. Buyback benefits already sit inside forecast EPS; they are not added again to returns.[S2][M1]

Balance-sheet constraint: $13.53bn of eligible corporate liquidity broadly offsets $13.40bn of funded debt; customer balances are not buyback resources. Capex is only 2.6% of trailing revenue, but credit funding, technology spending and compensation remain real costs. The DCF additionally reserves $3bn of operating cash.[S2][M1]

04 / Our five-year operating scenarios

The bear case assumes checkout erosion; the base assumes gradual repair; the bull requires successful monetisation and stronger margins. These are conditional paths, not probability bounds.[M1]

¹Includes newly retained corporate cash, but excludes dividends. Dividends accumulate without reinvestment. Owner-cash valuation is an alternative to earnings valuation, not an additional return.[M1]

The bear case is not a loss floor. It still assumes profitability and cash retention. A severe loss of checkout relevance, credit impairment or security event could produce materially worse results.

05 / The base-case path

Revenue grows 4.2% and total net income 6.9% annually, but EPS grows 14.5% because average diluted shares fall from 870m to 616.6m—about 6.7% per year. This substantial net retirement is central to the thesis, not a minor bonus.[M1]

The model allocates 85% of cash remaining after loan funding and dividends to buybacks. It allows for employee-share issuance and assumes repurchase prices rise 15% annually. Unspent cash accumulates separately and contributes $7.05 per share to terminal value; it cannot also be spent on unmodelled investment.[M1]

DCF cross-check: approximately $63 per share, using an 11.5% discount rate and 2.5% perpetual growth, with compensation and credit funding treated as costs. At 10–13% discount rates and 2–3% growth, the same forecast gives approximately $51–81. About 68% of enterprise value comes from the terminal period. This is a present-value estimate, unlike the $130 five-year terminal share value.[M1]

06 / Back-solving the 20% hurdle

At $53.81, required terminal wealth is $133.90 per initial share. Subtract base dividends of $3.45 and new corporate cash of $7.05, and the earnings-capitalised portion must reach approximately $123.40.[M1]

At 12×, the hurdle needs $10.28 EPS versus the base forecast of $10.24: almost no room for disappointment. With an unchanged starting P/E of 10.35×, it instead needs $11.93 EPS and a 22.1% operating margin, versus 19% in the base case.[M1]

The cash test is tougher: at 12× owner cash, a 20% return needs $6.34bn terminal owner cash, or 9.1% annual growth, versus $5.77bn in the base case.[M1]

20% entry-price ceilings: approximately $53.62 on earnings and $49.34 on owner cash, calculated as base terminal wealth ÷ 1.20⁵. A price near $45 provides more room for forecast error; a price decline caused by weaker fundamentals requires a new forecast, not blind use of these thresholds.[M1]

07 / What would confirm or break the thesis?

²Transaction-margin dollars excluding customer-balance interest. Compare consistent periods and definitions; a single severe event can justify an immediate reset.[S2][S3][S6][M1]

Catalysts: the next earnings release, 2027 guidance and evidence that the cost programme produces net profit after reinvestment. Management’s $1.5bn gross savings ambition is not $1.5bn of guaranteed additional earnings.[S3][S6]

The thesis breaks if branded checkout loses relevance faster than Venmo and processing can replace its profit, or if credit losses and dilution prevent cash-funded share retirement. A lower P/E does not protect against permanently lower earnings.

08 / Conclusion and limitations

At $53.81: the original model supports a plausible high-teens return, but 20% depends on execution and some valuation recovery. The strongest case is durable cash flow combined with net share retirement; the strongest objection is weak branded profit growth. Below roughly $49—and more meaningfully near $45—the same forecast offers a stronger cash-based margin for error.[M1]

Condensed from the 16 September 2026 research; original data, forecasts and valuation assumptions are unchanged. Financials run through June 2026; the consensus snapshot is dated 5 August. Forecasts use five fiscal years as holding-period proxies rather than a date-exact stub. Returns are USD, before tax, fees and AUD/USD effects, and are not guaranteed. Detailed forecasts, cash bridges, assumptions and sensitivity grids remain in the unchanged Excel workbook.