Is PayPal Undervalued?

PayPal trades at $41. Five years ago it touched $300. The market has written it off as an ex-growth payments company being squeezed by Apple Pay, Shopify and every neobank with a checkout button. But a stock everyone hates is exactly where mispricing happens, so I built a full DCF with bear, base and bull scenarios, plus a trading comps cross-check, to test whether PayPal is genuinely cheap or just cheap-looking.

Most valuation write-ups show you outputs. This one shows you decisions. A DCF has three places to go wrong: how you define cash flow, how you forecast it, and how you discount it. I’ll take each in turn, including the choices that work against my own conclusion.

Step 1: Defining free cash flow honestly

The headline numbers in PayPal’s cash flow statement need two adjustments before they mean anything.

Stock-based compensation is a real cost. PayPal adds back roughly $1bn of SBC per year in operating cash flow because it is non-cash. But SBC transfers value from shareholders to employees through dilution just as surely as a cash salary would through the income statement. Treating it as free money is how software companies end up looking artificially cheap. I deduct it in full. For context, this single choice removes about $1bn per year from my FCF line relative to a naive read of the cash flow statement.

Buy Now Pay Later distorts the quarterly picture. PayPal originates BNPL receivables and periodically sells them to third parties. When a sale happens, operating cash flow jumps; when originations outpace sales, it sags. These swings are timing, not economics: $417m in Q1 2025, $817m in Q1 2026, roughly $1.3bn reconstructed across FY2025. I add the disclosed adjustments back so the model reflects underlying cash generation rather than the loan-sale calendar.

The resulting definition: adjusted FCF = operating cash flow, less capex, less SBC, plus the BNPL timing adjustment. On this basis, PayPal generated roughly $5.9bn of economic free cash flow in FY2025. Keep that number in mind. It matters for what comes next.

Step 2: Building the forecast (and choosing my bias deliberately)

This is the step most write-ups skip, and it is the step that determines the answer. A discount rate applied to invented cash flows is just precise nonsense, so here is exactly where my forecast numbers come from.

The mechanics. I take Q1 2026 reported actuals (operating cash flow of $1.13bn, capex of $231m, SBC of $259m), convert them to a monthly run-rate, and grow that run-rate forward at a small scenario-driven rate: flat in the bear case, 1% per quarter in the base, 2.5% per quarter in the bull. FY2027 then grows off the FY2026 total at a scenario rate, and FY2027 becomes the base for the terminal value. There is no revenue build, no margin bridge, no management guidance. The forecast is “what PayPal just did, continued forward.”

This construction has a known bias, and the bias is downward. Two reasons. First, Q1 is seasonally PayPal’s weakest cash quarter: Q1 2026 operating cash flow was $1.1bn against $2.4bn in Q4 2025. Flat-lining the weakest quarter across the year strips seasonality out of the model. Second, my forecast months exclude the BNPL adjustment entirely, even though I argued above that it represents real economics. I exclude it because I cannot forecast the timing of receivable sales with any precision, and a number I cannot defend does not belong in a forecast.

The combined effect: my FY2026 clean FCF forecast of roughly $3.4bn sits about 40% below the $5.9bn PayPal actually generated in FY2025. My terminal value, which carries most of the enterprise value, is built on that depressed base.

Why build it this way deliberately? Because the alternative is worse. If I anchored on the strong Q4, layered in assumed BNPL recoveries and added back seasonal uplift, every favourable assumption in the model would be mine, and the output would simply reflect my optimism dressed up in arithmetic. Anchoring on the worst recent quarter and excluding adjustments I cannot time means the valuation floor is built from the most pessimistic defensible reading of the actuals. If PayPal’s true sustainable run-rate sits anywhere between my $3.4bn and last year’s $5.9bn, every intrinsic value in this article is understated.

That is the right way round to be wrong. A model that needs generous inputs to show upside is a sales pitch. A model that shows upside despite inputs stacked against it is evidence.

A related choice: the short explicit horizon. I forecast explicitly for roughly 18 months before handing over to a terminal value. The textbook approach for a business in transition is a five-year fade. I kept it short on purpose: PayPal is mature, and for a mature business the valuation question is not “what does year seven look like” but “what is the sustainable run-rate of free cash flow, and what is that worth in perpetuity.” A long explicit period would just bury that question under a decade of spreadsheet columns filled with my own guesses. The trade-off is heavy sensitivity to terminal assumptions, which is why I flex them hard across scenarios and a full sensitivity grid rather than presenting one point estimate.

Step 3: The discount rate, built from first principles

With the cash flows defined and their construction transparent, the discount rate discussion now means something. I built WACC bottom-up:

• Risk-free rate: 3.85% (1-year UST).
• Expected market return: 11%, giving an equity risk premium of 7.15%. This is the binding assumption in the cost of equity, and it is deliberately demanding. Many practitioners use an ERP of 4.5% to 5.5%; mine effectively requires equities to clear a higher bar before anything looks cheap.
• Beta: 1.40, reflecting PayPal’s observed volatility versus the market. The stock has traded with more cyclicality than the payment networks, and the beta should say so.
• Cost of equity (CAPM): 13.86%. For perspective, that is a discount rate you might apply to a mid-cap turnaround, not a company generating billions in free cash flow. Again: demanding on purpose.
• After-tax cost of debt: 4.24% (5.10% pre-tax at a 16.8% effective tax rate).
• Weights at market values: 76.6% equity, 23.4% debt, giving a computed WACC of 11.61%.

Then, and this matters, I flex the discount rate by scenario rather than only flexing cash flows. The bear case adds 200bps for execution and competitive risk (applied WACC 13.61%); the bull case trims 50bps. Risk lives in both the numerator and the denominator, and a scenario framework that only stresses one of them is half a stress test.

Step 4: Three scenarios, three answers

Forecasts are monthly through 2026, then FY2027, then a Gordon Growth terminal value, all discounted at mid-period from June 2026.

Bear case: $32 per share, 22% downside. The already-depressed run-rate flat-lines with no recovery, FY2027 normalises only partially, terminal growth is effectively zero (0.05%), and WACC carries the extra 200bps. Note carefully what this scenario actually is: PayPal’s worst recent quarter, extended forever, discounted punitively. The downside does not come from modelling further business deterioration. It comes from applying a near-14% discount rate and zero perpetual growth to a base that is already conservative.

Base case: $47 per share, 15% upside. FCF grows 1% per quarter off the depressed run-rate, FY2027 grows 5%, terminal growth is 2.5% (roughly in line with long-run nominal GDP, which is where a mature payments company should land), WACC is the computed 11.61%. This is a stabilisation case, not a turnaround case. No reacceleration, no margin story, no BNPL recovery in the forecast months.

Bull case: $56 per share, 36% upside. Quarterly growth of 2.5%, FY2027 up 10%, terminal growth 3%, WACC 11.07%. This is the world where branded checkout holds share, Venmo monetises properly and margins expand. It is the only scenario that requires believing in execution, and even it keeps terminal growth at nominal-GDP levels rather than assuming PayPal outgrows the economy forever.

The sensitivity grid makes the asymmetry explicit. Flexing WACC from 8.2% to 13.7% and terminal growth from 1.5% to 3.5% around the base case cash flows, intrinsic value spans roughly $36 to $88. To justify today’s $41.22 you need a discount rate above 12% combined with perpetual growth below 2%. The market is pricing PayPal as simultaneously riskier than its own capital structure implies and permanently ex-growth. Either belief alone is defensible. Holding both at once is where I think the mispricing sits.

Step 5: The comps cross-check

A DCF is one method with stacked assumptions, so I cross-checked against payments peers: Block, Adyen, Fiserv and Global Payments. Visa and Mastercard are shown in the workbook for context but excluded from the benchmark, because the networks earn a structural quality premium (four-party model, near-zero marginal cost, regulatory moat) that PayPal cannot claim. Including them would flatter the comparison; excluding them is the honest cut.

The numbers:

• PayPal trades at 7.2x P/E against a peer median of 16.6x
• PayPal trades at 5.8x EV/EBITDA against a peer median of 7.8x
• The peer median EV/EBITDA multiple implies roughly $56 per share
• The peer median P/E implies roughly $94 per share, which I treat as an upper bound rather than a target, since earnings multiples flatter PayPal’s mix relative to EV-based measures

Two independent methods, built from different inputs, land in the same place: the DCF base case says $47, the EV/EBITDA comp says $56, and the current price sits below both. A profitable business generating billions in free cash flow trading at 7x earnings is priced for structural decline. The whole question is whether decline is the right base case, and my forecast construction already assumes most of the decline has happened.

What would change my mind

A thesis without a kill switch is a belief, not a position. Three things would make me re-cut this bearishly:

1. Clean FY2026 FCF landing below $3.4bn. My forecast is already built on the weakest quarter with no seasonal recovery. If actuals undershoot even that, the run-rate decline is structural, not seasonal, and the bear case becomes the base case.
2. Branded checkout share losses accelerating rather than stabilising. The entire stabilisation premise rests on PayPal’s button holding its place at checkout.
3. BNPL adjustments turning out to be credit problems rather than timing. If receivable sales slow because buyers are worried about the book, my add-back logic inverts from conservative to generous.

Verdict

The base case DCF says PayPal is worth around $47 against a $41 price. The comps independently say $56 on EV/EBITDA. Both methods point the same direction: the market is treating the bear case as the central case.

The honest counterpoint is that the bear case is plausible. Checkout competition is real, and my own model shows clean FY2026 cash generation running well below FY2025. This is not a table-pounding buy. It is a stock where the forecast itself is anchored on the weakest recent quarter, with SBC fully deducted, BNPL upside excluded from every forecast period and double-digit discount rates throughout, and the model still produces upside, confirmed by a second method built from entirely different inputs. If FY2026 cash generation lands anywhere near FY2025’s $5.9bn rather than my $3.4bn forecast, the upside is materially larger than stated.

That is what undervaluation usually looks like in practice: unloved, not undiscovered.

Bottom line: modestly undervalued, with the margin of safety resting on FCF stabilisation rather than recovery. I would size the position accordingly.

Here is the spreadsheet that contains all my analysis: https://docs.google.com/spreadsheets/d/17ChtSmTrqkWA30wGFjIKOadPydwdyMzysmhhmKS9T8U/edit?usp=sharing

This is analysis, not investment advice. All figures from PayPal filings and market data as of June 2026. The full model, with every formula visible, is linked below.