Carnival Corporation: the recovery is priced in, the re-rating isn't

The thesis in brief

Carnival’s operating recovery is largely done as a re-rating catalyst on its own. The equity already reflects what management itself called, on the Q2 earnings call, “our twelfth consecutive quarter of record net yields” — guidance still points to +3.2% net yield growth (current dollars) for FY2026 — and record customer deposits of $9.0bn as of 31 May 2026, an all-time high, up $450mm year-on-year. That resilience has held through a genuine exogenous shock: Gulf hostilities that began on 28 February 2026, whose demand impact on regional bookings is addressed in full in the Q3 setup section below.

Two figures in that paragraph carry more weight in this thesis than they might first appear to. Net yield is Carnival’s own core pricing-power metric — adjusted gross margin (total revenue less commissions, transportation, onboard-related costs and credit-card fees) divided by available lower berth days — built specifically to strip out the effect of the fleet simply getting bigger, so a +3.2% guide is a claim about getting more for each berth sailed, not about sailing more berths. Customer deposits matter for a different reason. Cruise passengers pay well ahead of sailing, so a deposit balance that keeps setting records is simultaneously a forward-demand signal — it only grows if new bookings are outrunning cancellations and refunds — and a source of cash Carnival holds, effectively interest-free, before the corresponding revenue is recognised. That second point is the one that connects to the credit story: deposit growth is one of the mechanisms behind the 3.1x net leverage improvement below, not just evidence that people still want to cruise.

What is not priced in, in the firm’s reading, is the credit re-rating — and the evidence for it is mechanical rather than narrative, not forecast. Two of three agencies now rate Carnival investment grade: Fitch since January 2025, S&P since 25 June 2026 — a crossing that mechanically released the collateral on Carnival’s 7.000% secured notes, which the company has since redeemed early. Moody’s is the outlier, at Ba1 with a positive outlook as of 31 July 2026 — one notch away — and it is the same agency whose own crossing, not the other two, triggered the actual multiple re-rating at Royal Caribbean. Net debt to adjusted EBITDA is 3.1x on management’s guided basis and falling. The full mechanics are in the credit-re-rating pillar below.

Despite that, Carnival trades at 8.4x EV/EBITDA against Royal Caribbean’s 14.9x — a gap of roughly 6.5 turns — while carrying comparable gross leverage on a like-for-like basis (3.4x Debt/EBITDA for Carnival against Royal Caribbean’s 3.3x). Norwegian, with materially worse leverage (5.9x), trades closer to Carnival (9.2x) than Royal Caribbean does — consistent with the market pricing some leverage information into the discount, just not enough of it against Carnival’s own recent trajectory. Royal Caribbean’s own multiple moved only once all three rating agencies had confirmed investment grade, not before. Carnival already holds two of the three, and the base case below models what happens if it clears the final notch the way Royal Caribbean did.

The firm is long Carnival on a 12-month horizon, and deliberately does not frame the base case as a bear/base/bull spread — the single biggest swing factor here is fuel price, which can move the base case in either direction just as quickly as it can confirm it, and a three-way scenario table tends to bury that rather than isolate it. Specifically, the base case Speculation applies Royal Caribbean’s own settled re-rating increment — the multiple gain that survived past its initial spike, excluding Royal Caribbean’s separate structural premium — to Carnival’s current multiple, for a target of $40.86, some 50% above the $27.27 close on 18 August 2026. The full derivation, the structural/catalyst split, and the realisation-sensitivity range (22% to 71% upside) are in the case-study box in the valuation section.

This is a conditional call, not a certainty dressed up as one. It is wrong if the multiple gap turns out to be more structural than this note assumes — Royal Caribbean’s premium brand mix and margin profile could be worth more of the remaining 4.5x than the structural/catalyst split above credits it for, permanently, regardless of what the rating agencies do. It is also wrong on a narrower, more specific point: Royal Caribbean’s own multiple did not move on its second rating crossing (S&P) — it moved on its third (Moody’s), detailed in the valuation section — and Carnival is at that same second-crossing stage now. If Carnival’s re-rating follows the same stage-by-stage pattern rather than just its eventual size, this base case has real weight riding on Moody’s own crossing landing inside the 12-month horizon, not just on the S&P milestone already achieved; a Ba1, positive-outlook rating one notch from investment grade makes that plausible but not guaranteed. It is wrong, separately, if the credit trajectory stalls entirely: a further shock that reverses the deleveraging path management has been guiding toward.

But the single largest swing factor, in either direction, over the 12-month horizon is fuel price — not a structural risk to the thesis so much as the variable most likely to move the base case up or down before the credit re-rating has time to play out. A favourable fuel-cost path compounds with the credit catalyst and pulls the re-rating forward; an unfavourable one, layered on top of the basis-widening already flagged for Q3, delays it or reopens the “is the discount structural” question this note is betting against. The two forward charts in this note size that swing and the credit trajectory on the same 12-month view, rather than folding them into a single bear/base/bull number.

Thesis pillars

Three claims sit under the summary above, and each carries its own evidence rather than borrowing weight from the others.

The operating recovery has not slowed into the back half of the year

Q2 FY2026 was not a soft comparison against a weak prior year. Revenue was $6.663bn, up 5.3% from $6.328bn, on capacity that grew only 1.5% — the difference is priced-in demand, not more berths sailed. Adjusted net income of $569mm was up more than 20% year-on-year, adjusted diluted EPS of $0.41 up more than 15%, and adjusted EBITDA of $1.582bn was itself a record quarter. Occupancy held at 104% for the second straight year, meaning the yield gain came from price and onboard spend rather than from selling more berths than the ship notionally holds. Management called it “our twelfth consecutive quarter of record net yields” on the earnings call — the same exact figure this note uses in the summary above, and the firm treats the company’s own stated figure as the sourced number throughout.

The more consequential evidence is forward-looking. Management said booking volumes and prices for 2027 sailings have been running ahead of prior-year levels “since March,” including a “substantial increase” in European bookings, and that the booking curve is “the furthest out on record.” Booked position for the second half of 2026 is also ahead of last year at what management called historically high prices in constant currency. None of that is guaranteed to convert at the guided rate — booking-curve commentary is inherently softer evidence than a reported quarter — but it is the same management that has now delivered twelve consecutive quarters of the pattern it is describing, which is the basis for treating the claim as more than promotional language Speculation.

Customer deposits close the loop between this pillar and the credit one below: the $9.0bn balance, up more than $450mm year-on-year, is both a demand signal — it only grows if new bookings outrun cancellations and refunds — and free financing collected before the corresponding revenue is recognised, one of the mechanisms behind the leverage improvement in the next pillar rather than a separate, unrelated data point.

The credit re-rating is mechanical, not promotional

Where the operating story rests partly on management’s own characterisation, the credit story rests on dated, third-party actions that have already occurred. Fitch upgraded Carnival to BBB- around 10 January 2025 — the company’s first investment-grade issuer rating. S&P followed on 25 June 2026, upgrading Carnival to BBB- from BB+ — the second. Under the indenture governing Carnival’s 7.000% first-priority senior secured notes due 2029, a second investment-grade rating mechanically released the collateral securing them; the notes became unsecured that same day, and the company redeemed all $500mm of them on 15 August 2026, at 103.50% of principal, ahead of their scheduled maturity. None of that required a forecast — it is a contractual trigger that already fired.

Moody’s is the outlier of the three, rating Carnival’s corporate family Ba1 with a positive outlook as of 31 July 2026 — one notch below investment grade — with the CFO having already flagged continued progress on the June earnings call before that action. That is not a coincidence worth glossing over: Moody’s played the identical role at Royal Caribbean, where it was the agency whose own crossing is dated to the specific quarter Royal Caribbean’s re-rating happened (detailed in the valuation section), not either of the earlier upgrades from the other two agencies. If Carnival’s re-rating is going to follow a similar path, Moody’s own crossing — not S&P’s, which the market has not yet rewarded — is the more likely trigger to watch for, and a positive outlook one notch from investment grade puts that plausibly within this note’s own 12-month horizon.

Net debt to adjusted EBITDA is 3.1x on management’s guided basis, more than half a point better than a year earlier, and falling. That deleveraging is the mechanical link between the two pillars above: an improving balance sheet is what let two of three agencies move, and record deposits are one of the reasons the balance sheet has room to improve without new equity or asset sales.

The valuation gap looks like Royal Caribbean’s own re-rating, one step behind

None of the above matters to the implied value unless the market treats it the way it treated a comparable set of facts at Royal Caribbean. Carnival trades at 8.4x EV/EBITDA against Royal Caribbean’s 14.9x and Norwegian’s 9.2x, despite carrying comparable gross leverage to Royal Caribbean (3.4x Debt/EBITDA against 3.3x) and materially better leverage than Norwegian (5.9x). Norwegian trading closer to Carnival than to Royal Caribbean is itself evidence the market is pricing leverage into the gap — just not enough of it against Carnival’s own recent trajectory, which is the mispricing this note is underwriting.

Royal Caribbean’s own history is the reason the firm is willing to treat that gap as partly closeable rather than permanent. Royal Caribbean averaged 12.8x EV/EBITDA in the five quarters before its own 2025–26 rating actions — a premium that predates any credit catalyst and that this note attributes to brand mix and margin, not credit, which is why it is deliberately excluded from Carnival’s base case. What moved on top of that structural premium was a further 2.9x, on average, once Royal Caribbean settled into its post-catalyst range — and Carnival, per the mechanical evidence above, has now cleared an analogous milestone of its own. The market’s response so far has been silence: Carnival’s multiple was essentially unchanged between its own second investment-grade rating on 25 June 2026 and the 18 August 2026 comp snapshot this note uses. That gap between a real credit event and an absent market reaction is the trade.

Q2 scorecard

The relevant benchmark for a scorecard is not Street consensus — the firm did not have access to a Q2-specific consensus build for this note — but the company’s own guidance for the quarter, issued alongside Q1 results in March 2026. Management said on the June call that results “exceeded our March guidance by $100 million”; the table below is what that refers to.

Cost, not yield, did most of the work. The yield beat is real but modest — 2.2% against a 2.0% guide, 0.2 points of upside. The larger variance is on the cost line: adjusted cruise costs ex-fuel grew 0.1% in constant currency against a 2.6% guide, a swing of roughly two and a half points that flows almost directly to the bottom line given Carnival’s operating leverage. Fuel came in a rounding error from guidance, $793 per metric tonne against $795 guided — notable mainly because it means the entire beat was self-inflicted execution, not a fuel tailwind the company got lucky on.

Capacity growth of 1.5% against a 1.9% guide is the one line that missed, and it is worth being precise about why the firm is not troubled by it. Fewer available berths than planned, with demand and pricing holding, mechanically helps both the yield and occupancy lines above — and the shortfall is consistent with, though not explicitly attributed by management to, the itinerary redeployments away from planned Gulf voyages described in the summary above. The firm has not found a management statement that draws this connection directly, so it is offered as a plausible read of the number, not a sourced fact Speculation.

Q3 setup & risk

The summary’s callout flagged the fuel-cost basis as the one risk that belongs above the fold rather than at the back of the note. This section is that finding in full.

The reason this section builds its own fuel-to-Brent model, rather than reading the guided fuel-cost figure on its own, is that a dollar number without a reference point cannot separate two very different risks: fuel costs rising because crude is expensive — a market-wide risk already in consensus — from fuel costs rising because the fuel-to-crude spread itself has widened, a narrower and less visible risk this note can size independently of where oil goes next. On the trend in the data below, that spread is more likely than not to have widened further by the time Q3 actually reports, not narrowed Speculation. That is not necessarily where it stays: crude supply from outside the Gulf has already been growing faster than expected this year, and a continuation of that trend could ease the basis just as it widened, into Q4 and year-end — a possibility, not a call, and one explored in full later in this section Speculation.

Q3 is guided softer than the quarter that just reported, on the metric this note leans on hardest. Net yield growth is guided at +1.2% in constant currency, against the +2.2% Carnival actually delivered in Q2 — a full point of deceleration in the core pricing-power metric. Adjusted cruise costs ex-fuel are guided up 2.8%, worse than either Q2’s 2.6% guide or its 0.1% actual. And fuel cost per metric tonne is guided at $812, up 33.8% year-on-year — the largest guided increase of any quarter in the past two fiscal years.

The question this note will not answer from Brent alone is whether that fuel guide is a crude-price story or something that has become structurally worse. Carnival does not hedge fuel — confirmed on both the Q1 and Q2 FY2026 calls — so realised cost should track spot purchases with a purchase-timing lag, not a hedge-smoothed blend, which makes it a cleaner read on the underlying market than a hedged peer’s cost line would be. Fitting Carnival’s own realised fuel cost against the Brent average for the same fiscal quarter across the seven clean pre-war quarters (Q3 FY2024 through Q1 FY2026: fuel $/mt = 275.2 + 4.74 × Brent $/bbl, r = 0.897) and checking the war-period quarters against that line gives a direct read on whether the basis has shifted, independent of where Brent itself sits.

Q2 FY2026 — Brent averaged $102.02/bbl for the quarter — ran 4.5% hot against the model’s $759 prediction, consistent with a purchase-timing lag cushioning the initial shock. Q3 is guided at $812 against a lower quarterly Brent average of $85.98 and a model prediction of $683 — an 18.9% premium to what the pre-war relationship would predict. The miss is getting larger, not smaller, as the war moves further into the rear-view mirror on the crude side.

That is not, on the evidence available, a Carnival-specific problem — the identical regime break shows up at Royal Caribbean and Norwegian despite materially different hedge books, and even Rotterdam and Houston bunker data, far from the physically-disrupted Gulf hubs, shows the same basis shift. The full peer and hub-level evidence is below; the short version is that this reads as a sector-wide bunker-market repricing, not a Carnival-specific hub or purchasing failure, which is why the firm treats it here as a real, current cost risk to size and watch rather than a reason to doubt management’s cost discipline elsewhere in this note.

Carnival’s own disclosure sizes this precisely: a 10% change in fuel cost per metric tonne moves adjusted net income by approximately $56mm in Q3 FY2026 alone, and by a further $102mm across the remainder of the year (fiscal Q4) — $158mm combined for the second half, and the base the chart below builds forward from. Grossed up to an EBITDA-equivalent at the company’s disclosed 4.0% effective tax rate and extrapolated at that same second-half average quarterly rate through FY2027, a sustained 10% surprise in either direction is worth roughly $353mm of cumulative EBITDA over the next 12 months. That is why this note treats fuel price, not a bear/bull spread, as the variable to watch: it is large enough on its own to move the base-case math in the valuation section by more than the gap between the realisation-sensitivity cases there.

That swing factor matters most in the second half specifically, because the second half is where it lands against decelerating, not accelerating, growth. Net yield growth is guided at only +1.2% in constant currency for Q3 — against the +2.2% Carnival actually delivered in Q2, and against the record-yield pace described in the summary above — and guided capacity growth for the quarter, 1.5%, holds at the same reduced pace Q2 actually delivered rather than reverting to the original 1.9% guide. On the figures Carnival itself has given, the second half is not shaped like a continuation of the first half’s record run; it is shaped like a step down in growth layered on top of a step up in cost.

In EPS terms, which is the more direct answer to how much this actually moves the number a shareholder sees: a 10% fuel-cost swing is worth about $0.04 per share in Q3 alone against a $1.35 guided quarter, and about $0.11 per share across the combined second half against $2.22 of guided full-year EPS — roughly 3% and 5% of the respective guides. That is before the basis-widening question below is even in the picture; it is simply the size of the lever, in the units that show up in an earnings release.

The 19% figure could also move the other way, and that possibility is not fully explored elsewhere in this note. The mechanism behind the basis widening reads as a supply constraint, not a demand shock: Gulf refinery shut-ins in Iraq, Saudi Arabia and the UAE cut regional refined-product exports and tightened low-sulphur bunker fuel markets broadly, a reading consistent with industry bunker-market reporting into August 2026 that cites “restricted Gulf supply, strained refining capacity and depleted inventories” rather than a spike in shipping demand. Two separate channels could ease that constraint, and both are already partly underway rather than purely hypothetical.

The first is non-Gulf crude supply itself: the IEA’s own May 2026 Oil Market Report shows non-Gulf crude growth revised up 540kb/d against its start-of-year forecast — led by the US (+320kb/d) and Brazil (+210kb/d) — with Atlantic Basin exports up 3.5mb/d since the war began. That is enough to cut the shortfall, not close it: global crude exports in April were still 6.6mb/d below their pre-war February level even after the offset, and the same report puts OPEC+ spare capacity outside the Gulf at a historic low of 170kb/d, with its own view that supply will “likely be slower to recover” than demand even once Gulf flows resume Speculation.

The second channel is the one already flagged by third-party bunker-market modelling — not Carnival’s own guidance — which carries an actual forecast for the basis narrowing: a path toward roughly $690 per metric tonne globally by Q4, explicitly conditional on Hormuz transit flows improving and regional inventories rebuilding Speculation. Neither channel had cleared its own condition as of the most current reporting available to this note: transit counts through the Strait fell week-on-week in mid-August (73 versus 91 the week before, per Lloyd’s List Intelligence), US-Iran diplomacy is reported to have “effectively collapsed,” and tanker markets are described as increasingly pricing in a prolonged disruption rather than a near-term resolution.

Applying the same disclosed linear sensitivity used above, closing the full $129-per-tonne Q3 basis gap already identified — not just a 10% swing — would be worth on the order of $0.06 to $0.07 per share in Q3 alone: an extrapolation beyond the sensitivity’s own tested range, and worth treating as illustrative rather than a second forecast Speculation. The honest summary is that there are two real, sourced paths to a positive surprise here, both tied to specific and trackable conditions, and neither was improving as of this note’s own working date.

Two things resolve this before the 12-month horizon is up. The clean test is the Q3 FY2026 print itself, expected late September or early October 2026 — the first quarter for which guidance becomes actual, and the point at which the 19% basis gap either narrows or is confirmed as a new, structurally higher cost regime.

Separately, and distinct from the cost channel entirely, the Gulf hostilities that began on 28 February 2026 carry an acute demand tail-risk, not a pricing one, layered on top of the standing, roughly 50bp yield-guidance “normalization” item tied to redeploying away from planned Gulf itineraries noted above. Management has described the demand impact as radiating in “concentric circles” from the conflict zone — Mediterranean bookings hit hardest, improving with distance — a useful frame for that risk, though it was articulated in the context of bookings rather than the fuel-cost channel this section is mainly about. The valuation section’s DCF cross-check prices this downside case directly, with no re-rating and no resolution of the basis question assumed — the basis for reading this uncertainty as a window for building the position rather than a reason to wait for it to clear Speculation.

Read together, the scorecard supports the operating-recovery pillar without needing to lean on it: the quarter that already happened beat guidance on the lines management controls directly (cost, and to a lesser extent yield), while the line most exposed to what happens next — full-year yield guidance — moved in the direction the firm’s own Q3 caveat would predict. That is the shape of evidence this note is comfortable with: real and delivered where it can be, openly conditional where it can’t.

Balance sheet & credit trajectory

The credit-rating actions described in the thesis pillars did not happen against a static balance sheet; they happened because the balance sheet underneath them kept improving, and that improvement is more durable evidence for the thesis than any single quarter’s leverage print. Total debt was $25,570mm gross ($24,889mm net of unamortised issuance costs) as of 31 May 2026, down from $27,383mm gross ($26,640mm net) a year earlier — a reduction funded from operating cash flow and deposit growth rather than new equity or asset sales, consistent with the customer-deposits mechanism described in the first thesis pillar.

That improvement shows up most clearly in the maturity ladder, which the firm reads as evidence against near-term refinancing risk rather than a source of it:

Less than 13% of the debt stack comes due across the remainder of FY2026 and all of FY2027 combined, and roughly 44% sits beyond FY2030 — a back-loaded profile that gives Carnival room to keep prioritising early redemption of its highest-coupon secured debt, as it did with the 7.000% notes in August, over refinancing under pressure Speculation.

The stack’s composition, as of the same 31 May 2026 filing, is a cleaner cut of the same total than the maturity ladder alone shows:

The $3,098mm secured subtotal is the same collateral package referenced above: all three tranches lost their security on 25 June 2026 and were brought down to Moody’s Ba1 accordingly on 31 July 2026, but only the $500mm 7.000% tranche has actually been redeemed so far — the remaining two secured-turned-unsecured tranches ($192mm due June 2027, $2,406mm due August 2028) are still outstanding, and would be the next logical candidates if Carnival keeps prioritising early redemption of its highest-coupon debt Speculation.

Liquidity carries the same message. Carnival held $2,243mm of cash and had $4.5bn available for borrowing under its revolving facility as of 31 May 2026 — combined headroom of roughly $6.7bn against a covenant floor of $1.5bn minimum liquidity, a cushion of roughly 4.5x the requirement. The company states plainly that “at May 31, 2026, we were in compliance with the applicable covenants under our debt agreements,” which also include a minimum interest-coverage ratio of 3.0x (adjusted EBITDA to net interest charges) and a debt-to-capital cap of 65%; the firm has not independently recomputed either ratio, since neither is disclosed as a standalone figure in the filing, but takes the compliance statement itself as sourced. Royal Caribbean’s own revolver, for comparison, was $6.0bn undrawn against an upsized $6.6bn facility as of the same quarter — comparable headroom at a comparable point in its own credit cycle, one more data point for reading Carnival’s balance sheet as investment-grade-shaped rather than merely investment-grade-rated.

The leverage trajectory behind all of this has been a multi-year pattern, not a single clean quarter. On the firm’s own quarterly build (an EV/EBITDA-basis proxy, distinct from management’s own guided figure and deliberately not blended with it), net debt to adjusted EBITDA has declined in every one of the last ten quarters on record, from a peak of 6.81x at fiscal Q4 2023 to 3.27x at the most recent quarter-end covered by the 10-Q (31 May 2026). Management’s own guided figure, on its own basis, puts the ratio at 3.1x as of the most recent update — a different basis, deliberately not blended with the proxy series, but pointing the same direction. That sustained pattern, not the redemption of any single bond, is what the implied value in the next section leans on when it assumes the deleveraging continues rather than stalls.

That assumption has a specific, sizeable stress test attached to it, and it is the same one flagged in the Q3 setup section: a sustained 10% unfavourable fuel-cost swing is worth roughly 5% of a full year’s guided EBITDA. Held everything else constant, an EBITDA impairment of that size would not reverse the deleveraging trend outright, but it would visibly slow it — the ratio has been improving at roughly 0.15x per quarter over the last four quarters on the firm’s own build, and a full year of that scale of fuel headwind would cost approximately one quarter’s worth of that progress Speculation. That matters beyond the balance sheet itself: it is the same leverage trajectory Moody’s would be watching ahead of its own crossing, which the valuation section treats as the more consequential of the two remaining rating catalysts. A fuel shock large enough to matter for EPS is also, mechanically, a fuel shock large enough to put time back on the clock for the credit story, not just the earnings line.

Valuation

The firm’s valuation method treats peer-multiple re-rating as the primary driver and a discounted cash flow as a cross-check, not the other way round — the credit and operating evidence above is a case about what Carnival’s multiple should do, and Royal Caribbean’s own completed transition is the closest available precedent for how far and how fast that kind of re-rating actually travels once it starts.

The Royal Caribbean precedent: what a settled credit re-rating actually looks like

Royal Caribbean’s own transition is now complete and dated, which is what makes it usable as a precedent rather than a story. Fitch already rated Royal Caribbean investment grade entering this window; S&P moved it to investment grade (BBB- from BB+) in early February 2025 — Royal Caribbean’s own second-of-three crossing, the same stage Carnival reached with its own S&P upgrade in June 2026; Moody’s completed the set in mid-May 2025, crossing it to investment grade as well (Baa3 from Ba1); Fitch then moved it a further notch within investment grade, to BBB, at a date the firm has not been able to confirm precisely and does not rely on here. Moody’s upgraded Royal Caribbean again, within investment grade, in early February 2026 (Baa2 from Baa3).

Royal Caribbean’s own EV/EBITDA multiple, on the firm’s own quarterly build, averaged 12.84x across the five quarters before either of its two investment-grade crossings (fiscal Q4 2023 through Q4 2024) — already well above Carnival’s current 8.39x, and before any credit catalyst had fired. The firm reads that gap as a structural premium — brand mix, margin profile — that predates the rating story and that a rating upgrade alone would not be expected to close; it is deliberately excluded from Carnival’s base case below.

What moved on top of that structural premium was the catalyst, and it is more precisely timed than “two of three agencies” suggests: Royal Caribbean’s multiple was 12.71x in the quarter its own S&P crossing landed (Q1 2025) — essentially flat to its 12.84x pre-catalyst average, not elevated by it — and only jumped to 16.77x, a 4.06x move in a single quarter, once Moody’s completed the set in Q2 2025. It then settled into a 14–16x range over the following year, averaging 15.77x across the five quarters since, below its 17.21x peak. That settled increment — +2.93x over Royal Caribbean’s own pre-catalyst average — is what survived past the initial spike, and it is the number this note applies to Carnival below.

The chart’s own most recent point (15.91x, June 2026) reads a touch above the 14.85x used everywhere else in this note for Royal Caribbean’s current multiple — the same proxy-versus-official basis gap already flagged for Carnival’s own leverage series, here on the EV/EBITDA side instead: this quarterly build uses a TTM operating-income-plus-D&A proxy for EBITDA, not the data provider’s own “Adjusted EBITDA,” and the two run about a point apart. The comp table elsewhere in this note uses the official 14.85x; this section’s own trajectory series is kept on its own proxy basis and not blended with it.

The base case: applying Royal Caribbean’s own pace to Carnival’s own multiple

Carnival passed an analogous milestone of its own on 25 June 2026 — its second investment-grade issuer rating, from S&P, detailed in the credit-re-rating pillar above. Unlike Royal Caribbean at the equivalent point in its own cycle, the market had not re-rated Carnival as of the 18 August 2026 comp snapshot this note uses: its multiple was essentially unchanged before and after the S&P action. The base case below is a forward call that it does, on a similar path and pace to Royal Caribbean’s own, within the next 12 months — subject to the timing caveat above Speculation.

The net debt figure in that build — $23,930mm, total debt of $26.17bn less cash of $2.24bn, both on the 18 August 2026 comp-snapshot basis — is deliberately the same-dated basis as the EV/EBITDA multiple being re-rated, and is not the $24,889mm figure used in the balance-sheet section above, which is the 10-Q’s own net-of-issuance-costs figure as of 31 May 2026. The two are about six weeks and a different accounting basis apart (carrying value net of unamortised issuance costs, versus total debt less cash); the valuation uses the snapshot basis because it is dated to match the multiple, not because the two figures disagree.

None of this is presented as a single-point certainty. The realisation-sensitivity range below shows how the implied value moves if Carnival’s own re-rating runs slower or faster than Royal Caribbean’s did — not as alternative bear/base/bull calls, but as supporting detail on the same base case:

The base-case implied value of $40.86 reflects a differentiated call on the pace and size of a re-rating that has, by Royal Caribbean’s own precedent, tended to run ahead of consensus once the credit milestone is confirmed rather than merely anticipated — while carrying the explicit condition, laid out above, that the precedent’s own sequencing puts real weight on a Moody’s action this note cannot schedule.

DCF cross-check: a floor, not a destination — and one that lands below spot

The re-rating call above is, by construction, a bet on what the market pays for comparable cash flows — so the firm runs a single-stage, FY2026E-as-base growing perpetuity alongside it as an orthogonal check: what Carnival’s own cash generation supports with no re-rating at all. That standalone floor is why the firm reads the fuel-basis uncertainty sized in the Q3 setup section as a case for building the position into near-term volatility rather than waiting for it to clear — the downside case does not depend on the catalyst landing on schedule Speculation. It is deliberately conservative — a single explicit year, not a multi-year build — and, as the result below shows, it comes in low enough that the reason why matters as much as the number itself.

Carnival’s raw beta of 2.34 is high enough on its own to be worth flagging Speculation — it reflects the stock’s own historically volatile pandemic-recovery period, not necessarily its forward-looking risk, which is why the Blume-adjusted figure (1.89, still elevated) is used for the base case rather than the raw number, with the raw-beta case shown separately as the more conservative bound.

A normalised check on the capex distortion. FY2026’s capex line blends two very different kinds of spending: recurring, non-newbuild capital expenditure, and lumpy, ship-delivery-driven newbuild spending that does not repeat every year at the same rate. The guided remainder-of-2026 figure is split cleanly between the two ($600mm newbuild, $1,300mm non-newbuild); H1’s actual $1,441mm is not split in the sources this note has, so it is left untouched rather than guessed at. Stripping only the guided H2 newbuild figure — the one piece this note can isolate with a sourced number rather than an assumption — and holding everything else in the build unchanged:

That single, sourced adjustment closes more than half the gap to spot on the adjusted-beta case — from -35.3% to -14.0% — and the sensitivity grid moves with it:

Unlike the as-guided grid, this one does clear spot at its more generous end — two of the twelve cells, $28.69 and $32.61 — confirming the capex base year really was doing a meaningful share of the original damage.

Risks & catalysts

The two largest risks to this thesis already got full treatment where they arose rather than being held for a back-page list: the Q3 fuel-cost basis in the Q3 setup section, and the Royal Caribbean precedent’s own second-versus-third-crossing timing in the valuation section. This section closes the note with the risks that have not yet had their own space, and the forward calendar of events that would confirm or move the base case.

The base case deliberately excludes Royal Caribbean’s own +4.45x structural premium, on the view that it reflects brand mix and margin profile Carnival is not assumed to close. The risk runs in the direction that assumption is wrong: Royal Caribbean traded above Carnival’s current multiple even in the five quarters before either of its own credit catalysts fired, which is the evidence for calling 4.45x of the 6.46x full gap structural rather than credit-related in the first place — but if the market ultimately decides brand and margin explain more of the gap than that, or all of it, the catalyst increment this note applies has less room to work with, or none Speculation.

Two of three rating agencies now rate Carnival investment grade, and the mechanical evidence for that trajectory continuing is real — but it runs in both directions. If the Q3 fuel-cost basis is confirmed as a structurally higher regime rather than a transitory shock, EBITDA growth slows, the deleveraging trajectory in the balance-sheet section stalls, and the credit momentum that got Carnival to two of three could stall with it, or reverse. That would remove the mechanism the base case leans on, not just delay it.

Carnival’s guidance embeds specific currency assumptions across the Australian dollar, Canadian dollar, euro and sterling; a material move in any of them shifts the current-dollar figures this note uses without reflecting any change in underlying operating performance — the gap between the current-dollar and constant-currency yield-guidance cuts already flagged in the Q2 scorecard section is a live example of this mechanism, not a hypothetical one.

The company’s own capex programme is newbuild-heavy through the DCF cross-check’s own base year, and that programme carries ordinary execution risk — shipyard delivery timing, cost overruns — that would show up first in the capex line the DCF is already sensitive to, before it showed up anywhere else in this note.

In April 2026, six purported class actions were brought against Carnival in the U.S. District Court for the Southern District of Florida over a data-security incident dated 14 April 2026, alleging negligence, breach of implied contract, invasion of privacy and unjust enrichment; the court consolidated the matters in May 2026. Carnival states it “believe[s] the outcome of this matter will not have a material impact on [its] consolidated financial statements” and has not disclosed a specific loss estimate. The firm has no basis to second-guess that assessment absent further disclosure, but includes the item here because it is real, ongoing, and sits entirely outside the operating and credit story the rest of this note is built on.

Finally, a standard due-diligence point this note has not built its call around: cruise demand is discretionary consumer spending, and a broader pullback in discretionary travel — from a weaker macro backdrop rather than anything specific to Carnival or the Gulf situation — would pressure the yield and booking-curve evidence in the thesis-pillars section regardless of how the credit and fuel stories above resolve.

None of these resolve the base case on their own; together, they are the calendar against which the thesis in the first section either keeps being confirmed or starts needing revision.