Booking Holdings (BKNG): When the Hormuz Strait Hands You a 35% Drawdown in the World’s Best Travel Compounder
Growth, fair valuation, a category-defining brand, near-monopoly economics, and shareholder-friendly management – all the things I look for, now at the cheapest valuation in years thanks to the Strait of Hormuz.
By Philipp Haas | investresearch.net | May 19, 2026
Introduction & Investment Case
In this piece I want to walk you through one of my long-term favourites – a business that ticks essentially every box I care about: growth, a fair valuation, a dominant market position, world-class product quality, and a management team that has been quietly running one of the best buyback machines in the entire internet space. It is a name almost everyone knows but that sits a little in the shadow of the very biggest internet giants. The name, of course, is Booking.com – or rather its parent, Booking Holdings (NASDAQ: BKNG).
The reason I am writing this now is timing. As I type this on May 19, 2026, BKNG trades at $154.13 – down roughly 34% from its July 2025 all-time high of $233.58 and just a few percent above its 52-week low. A US-Israeli strike on Iranian nuclear infrastructure in late February tipped the Middle East into the most serious regional war of the decade. The Strait of Hormuz is effectively closed to most commercial shipping. Jet fuel spiked, airlines cut capacity, cancellations surged, and Booking’s Q1 2026 print on April 28 – which beat on every single line – came with a guidance cut. The stock has been sold relentlessly since.
My take in one paragraph: this is a textbook setup. A best-in-class, near-monopoly internet business with ~87% gross margin and a ~37% adjusted EBITDA margin, going through a temporary, identifiable, and reversible geopolitical shock – exactly the kind of moment that has historically been a gift for long-term holders. Applying my Fair PE / QFMA framework, I arrive at a Fair PE of 22x on 2028e earnings, implying roughly +24% per year over the next three yearsat the base case.
Now let me walk you through the work.
1. Product, Business Model, Brand & Moat
What Booking does
Booking is the world’s largest provider of online hotel bookings. If you have ever booked a hotel online, you almost certainly know Booking.com. As someone who used to run a hotel site back in the day (1000hotels.net – my wife runs the YouTube channel now, feel free to check it out if you love hotels), I can tell you from inside experience: of the world’s best 1,000 hotels, 99% were already on Booking.com years ago. They cover the market almost completely.
Yes – occasionally you’ll get a better rate going direct, the same way you used to with the hotel front desk, or by cross-checking on Trivago (which itself is an interesting name in the same neighbourhood). But for the vast majority of trips, Booking.com is the default. That is what category dominance looks like.
And it is not just hotels. The group operates six globally relevant brands:
Booking.com – the European-dominant accommodation crown jewel, ~2.5 million properties
Priceline – the US OTA with the iconic Negotiator brand (Randall Park just took the baton from William Shatner – yes, really)
Agoda – the Asia-Pacific specialist where local payments and language depth matter
KAYAK – flight metasearch and trip planning
OpenTable – restaurant reservations (great product, smaller coverage relative to hotels)
Rentalcars.com – car rental aggregation
How they make money
It’s beautifully simple. When you book a hotel through them, they take a ~10–15% commission. On a multi-thousand-euro trip, that adds up fast – and there is almost no cost of goods against it. Unlike e-commerce, where you have to actually buy, stock, and ship a product, Booking just operates a platform and runs marketing.
And here is the key shift: marketing costs are structurally declining as a share of bookings because more and more people book directly via the app instead of arriving through Google. Marketing expense as a percentage of gross bookings was 4.7% in Q3 2025 versus 5.0% the year before. The direct channel mix is now in the mid-60% range and growing. Mobile app penetration sits in the high-50% range on a trailing twelve-month basis, up from the mid-50s a year earlier. Every percentage point of “direct” is a structural margin gain forever, because it removes the Google tax. Incidentally, this is probably also one of the reasons Google itself never went fully into the travel-booking business: Booking was historically one of Google’s biggest customers, and Google has always preferred to sell marketing rather than be on the hook for the actual booking, customer service, multilingual support, refunds, and so on.
Three revenue streams, to be precise:
Agency model – list inventory, take a commission when it books. The high-margin core, especially in Europe.
Merchant model – collect payment upfront, own the customer end-to-end, capture FX and payment economics. Now ~68% of gross bookings.
Advertising / other – KAYAK and OpenTable advertising, similar in spirit to Amazon’s ad business. Still small but growing.
Brand
Booking.com is, in Europe, functionally a verb. “Book a hotel” and “use Booking” are essentially the same action for hundreds of millions of people. The 2025 numbers tell you the brand is still working: 2025 revenue of $26.9B (+13%), adjusted EBITDA of $9.9B (+20%), EBITDA margin expanded 193 bps to 36.9%, free cash flow $9.1B (+15%). These are not the financials of a brand under attack. They are the financials of a quasi-monopoly throwing off cash.
The Moat – and why I disagree with the AI bears
This is the question of 2026: will AI agents disintermediate the OTAs? My answer is no, for several reasons:
Supply, not search, is the moat. Booking has spent 25 years onboarding 2.5 million properties with API integrations, rate parity, instant confirmation, and a global customer-service layer in every major language. An AI agent that wants to actually transact either integrates with Booking’s API or replicates 25 years of supplier work. They are doing option (a).
Booking is the AI player, not the victim. Priceline’s “Penny” assistant now does end-to-end agentic booking – not just planning, actual transactions. On the Q1 2026 call, Glenn Fogel said users engaging with Penny show “a noticeable uplift” in conversion. CFO Ewout Steenbergen made the sharper point: most competitor AI tools “only provide planning” – Booking can actually close the loop.
Trust, language, and recourse. You’ll happily call a small hotel in Austria. You’ll be far less keen to call one in Vietnam in a language you don’t speak. The platform gives you transactional security, a written record, a refund channel, and multilingual support. That mattered before AI, and it will matter after.
They are absorbing the Airbnb threat, not losing to it. Yes, Airbnb has been the noisy newcomer in alternative accommodations. But Booking plays in all segments, including alternative stays – 38% of Booking.com’s room night mix is now alternative accommodations, with 8.8 million total alternative listings (+9% YoY in Q1 2026). They are quietly absorbing the category.
It is not winner-takes-all – but it absolutely is winner-takes-most.
2. Market: How Big, How Fast, How Politically Exposed
The travel market is enormous and structurally growing.
Global online travel market: ~$760B in 2026 → ~$1.4T by 2035 (~7.4% CAGR, per Global Market Insights).
OTA segment specifically: ~$107B by 2026 per Skift Research, ~7% growth.
Asia-Pacific: the structural engine – ~38% of OTA market share, 6.8% forecast CAGR through 2031. APAC international arrivals already recovered to 92.6% of 2019 levels by January 2026.
Three things to keep in mind
It grows structurally. Mobile penetration, the rising Asian middle class, Gen Z’s experience preference, and the offline-to-online migration all push the same way. Booking is gaining share in this growing pie – especially in the US, where room night growth has accelerated for four consecutive quarters into the low teens, and in Asia, where Agoda’s local playbook keeps winning. Add to that Covid effectively killing a chunk of the traditional travel agency business: more people now book their own multi-stop trips, and Booking is the natural default for that.
It is cyclical – but less than you’d think. Wars, recessions, oil spikes always hit travel in the short run. The pandemic was a once-in-a-century stress test for the industry. Booking’s operating profit went slightly negative in 2020 – but only slightly, and never threatened the business. The platform “breathes” with the cycle because the cost base is so light. The fact that an event of that magnitude didn’t break it tells you almost everything about durability.
Political exposure is real but localised. The Strait of Hormuz is the dominant near-term issue. Less discussed but more durable: Italy’s competition authority (AGCM) opened a formal investigation on April 22, 2026, into Booking’s Preferred Partner programmes – the third such EU probe in five years. Plus the Digital Markets Act classification in Europe is a long-term overhang. None of these have so far structurally broken the model, but they cost management attention and some margin.
3. Culture & Management
Glenn Fogel has been CEO since 2017 and at the company since 2000. He is not a founder-CEO in the Bezos mould – and he is probably not going to be remembered as a product innovator. But here is the single most important fact about him: he was the executive responsible for the Booking.com acquisition back in 2005, when the company was still known as Priceline. That acquisition is, without exaggeration, one of the top 10 M&A deals in tech history. Priceline paid roughly $135 million for what is now the dominant accommodation platform on the planet. Today the group does $186 billion in gross bookings annually. That is the kind of decision-making track record that buys you a lot of patience.
It’s also a uniquely European success story embedded inside a US-listed company – which makes sense, because travel is something Europeans simply do more of than Americans (more vacation days, more destinations within easy reach, and Europe is the world’s number-one tourist destination overall).
Some markers I look for and how Booking scores:
Long-term capital allocation: outstanding. Booking has been doing buybacks at scale longer than essentially any other internet company. Over the five years prior to the 25:1 split on April 6, 2026, the share count was reduced by ~21% – the company spent roughly $26 billion repurchasing its own shares. Q1 2026 alone delivered a record $3.6 billion buyback. This is organic growth plus buyback compounding. Many competitors only started serious buybacks recently, and at most of those the effect is partly offset by stock-based comp. At Booking the net reduction is real and ongoing.
Discipline through the cycle: The 2024 Transformation Program delivered $500–550M in annual run-rate savings, all being recycled into Connected Trip, AI, US, Asia, and OpenTable internationally.
AI restraint, not AI theatre: Penny is being rolled out via rigorous A/B testing and regulatory compliance. Fogel’s words on the Q1 call. Not vibes-based deployment. The right way.
One durable criticism: the buyback machine is so aggressive that shareholders’ equity is now in deficit (~–$8.7B by Q1 2026). This is what disciplined return-of-capital looks like in a business that genuinely cannot redeploy all its FCF productively – but it does make traditional ROE meaningless, and leverage is creeping up. Watch the interest coverage as new senior notes layer on.
4. Financials, Margins & Recent Developments
Free cash flow in 2025 hit $9.1B (+15%). Free cash flow yield at today’s market cap (~$119B) is roughly 7.6%. For a business growing high-single to low-double digits organically, that’s striking.
Q1 2026 – the quarter that crashed the stock
A beat on every single number. US room night growth in the low teens for the fourth consecutive quarter – genuinely impressive in such a mature market. Then management cut FY2026 guidance to reflect Middle East fallout: gross bookings up high-single to low-double, revenue up high-single, adj. EPS in the low-to-mid teens (~$10.82). Q2 specifically: room night growth of just 2–4%. March 2026 alone saw room night growth fall to just 1% as cancellations spiked.
A note on ROE
Booking’s stockholders’ equity is negative due to cumulative buybacks. Traditional ROE is mathematically undefined for this reason. The cleaner measure is return on invested capital, where Booking sits at ~93.6% LTM – one of the highest in the entire S&P 500. That number, more than any other, is the moat speaking.
Recent capital structure moves
In early May 2026 Booking issued $750M USD and €1.9B EUR senior unsecured notes (2030/2034/2039 maturities) – opportunistic, locking in tight EUR spreads and pre-funding more buybacks.
Connected Trip – the second S-curve
This is the most underappreciated piece of the story. Connected Trip transactions (customers booking more than one travel vertical) grew in the high teens YoY in Q1 2026 and are now a low double-digit share of Booking.com transactions. Flight tickets grew 44% YoY in 2025. As I’ve said for years – the dream is selling the whole trip from one platform: flights, hotels, cars, experiences, restaurants. We’re not there yet, but the curve is bending the right way. Honestly, at some point I wouldn’t even rule out a Tripadvisor-style acquisition for the experiences vertical, though I’d guess the regulatory bar is probably too high in today’s environment.
5. Valuation: My Fair PE / QFMA Framework
This is where I put my own framework to work. For those new to my Substack: my Fair PE / QFMA model assigns a target multiple based on quality, growth durability, margin profile, moat width, and management quality, then applies it to a normalised earnings number 2–3 years out.
For Booking Holdings, here is how I get to Fair PE = 22x:
Earnings path
Management’s 2026 guide implies adj. EPS of ~$10.80. Building out from there:
2026e EPS: ~$10.80 (matches guide)
2027e EPS: ~$12.30 (Middle East normalises, Connected Trip scales, mid-teens EPS growth, buyback contribution)
2028e EPS: ~$14.00 (mid-teens continued; base case, not the bull case)
Fair value
Fair PE 22 × 2028e EPS $14.00 = ~$308 per share
Against today’s $154, that’s roughly +24% per year over three years – a doubling if I’m directionally right. This is not a heroic forecast. It assumes:
Hormuz normalises sometime between H2 2026 and H1 2027 (consistent with management’s own planning assumption)
Connected Trip continues high-teens growth, slowly mixing up to a mid-teens share of transactions
Mobile direct mix grinds from high-50s toward 70%
EBITDA margins hold at ~37% (no further expansion assumed in my base case – that’s the upside)
No further multiple compression from AI fears
On consensus, BKNG trades at ~15.8x NTM P/E and ~12.1x NTM EV/EBITDA – a multi-year low for this business and below where lower-quality peers like Expedia trade. For a business with this margin profile and ROIC, that gap should not persist.
6. Risks & Why the Opportunity Exists
If the case is this clean, why is the stock $154 and not $230? Several reasons:
1. The Strait of Hormuz overhang. The market is treating a temporary fuel-price and travel-demand shock as if it will permanently reset Booking’s growth rate. It will not. Past shocks (2009 GFC, 2014 Crimea, 2016 Brexit, 2020 pandemic, 2022 Russia–Ukraine) have all turned out to be excellent entry points for high-quality travel businesses. Bookings compress, cancellations spike, demand snaps back – often above the prior trend.
2. AI disintermediation panic. Already addressed above. The short version: Booking owns the supply, owns the customer, has the buyback firepower to compound through any transition, and is itself the most aggressive AI deployer in the industry. The fear may be wrong, but as long as it persists, it caps the multiple – which is the gift.
3. Hotels pushing back on commission. Probably the most genuinely durable competitive risk. Hotels obviously don’t love paying 10–15%, and they’ll keep pushing direct channels via emails, loyalty programmes, and rate-club incentives. The dynamic has always existed – and for very long hotel stays it can absolutely make sense to call direct – but it’s a slow-grind risk, not an overnight one.
4. Regulatory risk in Europe. The Italy AGCM investigation, the DMA classification, and similar probes will cost some operating margin over time. Booking has a long history of navigating EU regulation – settlements, behavioural commitments, modest commission tweaks. Not structural impairment, but a recurring tax.
5. Negative equity / rising leverage. Aggressive buybacks have driven equity to ~–$8.7B. New senior notes issued in May 2026 add interest expense in a higher-rate environment. Not fundamental risk given the FCF profile, but a watchpoint.
8. Conclusion & Investment Case
The best opportunities in high-quality businesses come from temporary, identifiable, reversible problems. Booking in May 2026 has all three:
Temporary: Hormuz will reopen, or travel will route around it. The IEA already calling this “the most significant supply shock in oil market history” is exactly the kind of language that marks the peak of fear, not a structural change.
Identifiable: We can literally see the cancellation spike in the March 2026 data (room nights +1% vs. underlying +8%).
Reversible: Management has already signalled an H2 2026 recovery in its own planning assumption.
Catalysts for a rerate (12–24 months)
Middle East de-escalation / Hormuz reopening. Single biggest one. A ceasefire, Hormuz transit normalisation, or easing jet fuel removes the overhang within days.
Mobile app direct mix crossing 60% then 65%. Quiet, structural margin driver. Every point of “direct” is a permanent Google-tax reduction.
Connected Trip crossing mid-teens transaction share. Currently low double digits. The moment the market sees Connected Trip as a step-function in customer lifetime value rather than a pet project, the multiple re-expands.
AI agent integration and Penny conversion data. If Booking can show (and they will, on Q2 or Q3) that AI-assisted bookings have higher AOV and lower cancellation, the disintermediation thesis dies.
Sustained share buybacks at depressed prices. Every $3.6B retired at $154 is structurally accretive forever. Fogel is literally printing future EPS for shareholders patient enough to sit through this.
US share gains. Low-teens US room night growth for four straight quarters means Booking is taking share from Expedia and Airbnb in the largest, most mature travel market on earth. Most underappreciated part of the bull case.
Conflict of Interest Disclosure
Booking Holdings (BKNG) is a position in the cost-efficient Haas Invest4 Innovation Fund (invest4.net), the investment fund I manage, and is held in several of my Wikifolio portfolios. I personally also hold the stock. I may add to or trim the position at any time without prior notice. This article is written in my personal capacity as a publisher at investresearch.net and does not constitute investment advice.
Risk Disclaimer
This article reflects my personal opinions and analysis based on publicly available information as of May 19, 2026. It is not a recommendation to buy or sell any security. Investing in equities involves substantial risk, including total loss of capital. Past performance is not indicative of future results. Forward-looking statements – including valuation targets, growth forecasts, and EPS projections – are by their nature uncertain and may prove materially incorrect. Geopolitical events (including but not limited to the ongoing Middle East conflict), regulatory developments, foreign exchange fluctuations, changes in consumer behaviour, technological disruption (including AI-driven disintermediation), and broader macro conditions could materially affect Booking Holdings’ business and share price.
You should conduct your own due diligence and, where appropriate, consult a licensed financial advisor before making any investment decision. Do not invest more than you can afford to lose.
If you found this analysis useful, please consider subscribing to my Substack and sharing it with a fellow investor. For more research like this on global small- and mid-cap equities, visit investresearch.net and follow the work of the Haas Invest4 Innovation Fund at invest4.net.
– Philipp Haas
I want to tell you a story.
In late February 2026, a US-Israeli strike on Iranian nuclear infrastructure tipped the Middle East into the most serious regional war of the decade. Iran retaliated. The Strait of Hormuz – the artery through which roughly a fifth of the world’s oil moves – effectively closed to most commercial shipping. Jet fuel prices spiked, European carriers cut tens of thousands of “unprofitable” short-haul flights, and millions of travellers either cancelled trips to the region or rerouted around it.
If you were holding shares of Booking Holdings (NASDAQ: BKNG) on the day Glenn Fogel walked onto the Q1 2026 earnings call (April 28), you watched a company that beat on every single number – revenue, EBITDA, EPS, gross bookings – and then cut its full-year guidance because management was modelling the Hormuz fallout to persist through the end of June. The stock has been sold relentlessly since. As I write this on May 19, 2026, BKNG trades at $154.13, down roughly 34% from the July 2025 all-time high of $233.58 and sitting just a few percent above its 52-week low. The recent 25:1 stock split on April 6 is mostly cosmetic, but the rerating is not.
My investment case in one paragraph: Booking Holdings is the highest-quality online travel asset in the world, with a near-monopoly position in European accommodation, an ~87% gross margin, an adjusted EBITDA margin of 36.9% (Q4 2025) that is still expanding, a Connected Trip strategy that is finally compounding in the high-teens year-on-year, and a CEO who has just executed a record $3.6 billion buyback into his own pullback. The market is pricing BKNG as if the AI disintermediation thesis and the Middle East conflict are both permanent. Neither is. Applying my Fair PE / QFMA framework, I arrive at a Fair PE of 22x on normalised 2028 earnings, implying a fair value of roughly $293–308 per share and an annualised upside of ~24% per year over the next three years.
Full disclosure: BKNG is a meaningful position in the Haas Invest4 Innovation Fund and one of the names I have been quietly adding to during this drawdown.
Now, the work.
1. Product, Business Model, Brand & Moat
What Booking Holdings actually is
Most retail investors still think “Booking = hotels”. They are leaving 40% of the story on the table. Booking Holdings runs six globally distinct brands:
Booking.com – the crown jewel, dominant in European accommodation, with ~2.5 million properties spanning hotels and alternative stays (homes, apartments, B&Bs)
Priceline – US-focused OTA with the iconic Negotiator brand (now featuring Randall Park taking the baton from William Shatner – yes, really)
Agoda – the Asia-Pacific specialist, where local payment integration and language depth matter more than global reach
KAYAK – metasearch and trip planning
OpenTable – the dominant restaurant reservation platform in the US
Rentalcars.com – car rental aggregation
How they earn money
Two engines, increasingly merging:
Agency model – Booking lists inventory, the property collects payment, Booking takes a commission (historically the high-margin core, especially in Europe).
Merchant model – Booking collects the payment upfront, owns the customer relationship end-to-end, captures FX spreads and payment fees, and reinvests the float. This is now ~68% of gross bookings and the strategic direction of travel.
The merchant model is what enables the Connected Trip – the cross-sell of flights, accommodation, attractions, rental cars, and (soon) experiences inside a single booking. Connected Trip transactions grew in the high teens YoY in Q1 2026and now represent a low double-digit percentage of total Booking.com transactions. Flight tickets alone grew 44% YoY in 2025. This is the second S-curve, and it is finally bending upward.
Brand
Booking.com is one of the most-Googled commercial brands on earth. In Europe, “to book a hotel” is functionally a generic verb that means “to use Booking.com”. The mobile app is now where the action sits: app penetration reached the high-50% range on a TTM basis in Q1 2026, up from the mid-50s a year earlier. The direct channel mix (i.e. customers coming to Booking without going through Google) sits at a mid-60% range – this is the single most important metric for the long-term margin story and it keeps creeping up.
The Moat – and why I disagree with the AI bears
The bear case in 2026 has been simple: ChatGPT-style agents will disintermediate the OTAs. Why use Booking when an LLM can plan your trip?
I have spent a lot of time on this, and I think the bears are wrong, for three reasons:
Supply, not demand, is the moat. The bottleneck in travel is not “where do I plan?” – it is “what inventory can I actually book at a guaranteed rate with instant confirmation?” Booking has spent 25 years signing 2.5 million properties. An AI agent that wants to actually transact needs to either (a) integrate with Booking’s API or (b) replicate 25 years of supplier onboarding. Option (a) is what is happening.
Booking is the AI player, not the AI victim. Priceline’s “Penny” assistant now does end-to-end agentic booking– not just planning, but actual transaction. On the Q1 2026 call, Fogel explicitly noted that users engaging with Penny show “a noticeable uplift” in conversion. CFO Ewout Steenbergen drove the dagger in: most competitor AI tools “only provide planning” – Booking can actually close the loop. Booking.com is rolling out natural language search and smart filters across accommodations and cars.
The two-sided network effect compounds. More demand → more supplier interest → more inventory → more demand. The Genius loyalty programme deepens this with a third axis: switching cost for the high-frequency traveller.
Margins tell the same story. An 87% gross margin and ~37% EBITDA margin are not what a business under structural attack looks like. They are what a quasi-monopoly with operational leverage looks like.
2. Market: How Big, How Fast, How Politically Exposed
The global online travel market is a sprawling beast, and the various market reports throw around very different numbers depending on what they include. The cleanest framing I have:
Total online travel market: ~$760 billion in 2026, growing to ~$1.4 trillion by 2035 (~7.4% CAGR per Global Market Insights).
Online travel agency (OTA) segment specifically: ~$107 billion in OTA revenue by 2026 per Skift Research, growing roughly 7% per year.
Asia-Pacific is the structural growth engine – ~38% of the OTA market with a 6.8% forecast CAGR through 2031. Asia-Pacific international arrivals recovered to 92.6% of 2019 levels by January 2026.
Three things to understand about this market
It is structurally growing. Mobile penetration, rising middle-class travel in Asia, the experience economy among Gen Z, and the continuing offline-to-online shift in emerging markets all push in the same direction.
It is cyclical – but less than people think. Yes, travel demand wobbles with consumer confidence and oil prices. But the pandemic was an extreme test, and Booking emerged from it with higher market share and better margins than it went in with. Travel has proven, repeatedly, to be one of the most resilient discretionary categories. People give up restaurants, cars, and clothes before they give up holidays.
Political/regulatory exposure is real but localised. The Strait of Hormuz crisis is the obvious near-term issue. Less discussed but more durable: Italy’s competition authority (AGCM) opened a formal investigation on April 22, 2026, into Booking’s Preferred Partner programmes – the third such European probe in five years. Booking has navigated these before; it costs money and management attention but has not so far broken the business model. The DMA (Digital Markets Act) classification in Europe is the bigger long-term overhang and worth watching.
3. Culture & Management
Glenn Fogel has been CEO since 2017 and has been at the company since 2000. He is not a founder, but he has been in the trenches for a quarter-century, and his communication style on earnings calls has the dry, methodical quality of someone who has actually shipped product. The Q1 2026 call is a good example: he opened by acknowledging the human cost of the Middle East conflict before walking into the numbers, and he was direct about what the company can and cannot control.
A few markers I use to judge management quality – Booking scores well on most:
Long-term capital allocation: In Q1 2026 alone, the company repurchased a record $3.6 billion of its own stock. Over the long arc, share count reduction plus EPS growth has produced ~22% annualised returns over 25 years. That kind of compounding does not happen without consistent owner-orientation.
Transformation Program discipline: Booking has captured $500–550 million in annual run-rate savings from its 2024 efficiency program, and management has been transparent that these savings are being reinvested into Connected Trip, AI, the US market, Asia, and OpenTable’s international rollout – roughly $700 million of incremental investment in 2026, expected to deliver $400 million in incremental revenue.
AI restraint, not AI theatre. Penny is being scaled through “rigorous A/B testing and regulatory compliance” – Fogel’s words. Not vibes-based deployment. This is the right way to do it.
The one durable criticism: the buyback engine is so aggressive that shareholders’ equity is now in deficit territory (-$5.6B at end of 2025; -$8.7B by Q1 2026). I am not personally bothered by this – it is what disciplined return-of-capital looks like in a business that genuinely cannot redeploy all its FCF productively – but it does make traditional ROE meaningless, and it does mean leverage is creeping up. Watch the interest coverage.
4. Financials & Recent Developments
The long arc
YearRevenueAdj. EBITDAEBITDA MarginNet Income2021$10.96B––$1.17B2024$23.74B$8.25B35.0%$5.88B2025$26.92B (+13.4%)$9.9B (+20%)36.9%$5.40B (–8%, FX/transformation charges)
Free cash flow in 2025: $9.1 billion (+15% YoY). Free cash flow yield at today’s market cap: ~7.6%. For a business growing high single-digit to low double-digit organically, this is striking.
Q1 2026: the quarter that crashed the stock
MetricQ1 2026YoYRoom nights338M+6% (+8% ex-Middle East)Gross bookings$53.8B+15%Revenue$5.53B+16%Adjusted EBITDA$1.29B+19%Adjusted EPS$1.14+14%Share buybacks$3.6Brecord
Beats across the board. US room night growth accelerated to the low teens for the fourth consecutive quarter – a genuinely impressive milestone given how mature the US business is. Then management cut FY2026 guidance: gross bookings growth in the high-single to low-double digits, revenue growth high-single digits, adjusted EPS growth in the low-to-mid teens (~$10.82). Q2 specifically: room night growth of just 2–4%, with March 2026 having seen room night growth of only 1% as cancellations spiked.
ROE caveat: Stockholders’ equity is negative due to cumulative buybacks, so traditional ROE is not meaningful. The cleaner measure is return on invested capital, where Booking sits at ~93.6% LTM – one of the highest in the entire S&P 500. That number is the moat speaking.
Recent capital structure moves
In early May 2026, Booking issued $750 million in USD senior unsecured notes and €1.9 billion in EUR-denominated notes (2030, 2034, 2039 maturities). This is opportunistic – locking in EUR rates while spreads are tight and pre-funding more buybacks. It is also why some analysts are flagging rising leverage as a watchpoint.
5. Valuation: The Fair PE / QFMA Framework
This is where I put my own framework to work. For those new to my Substack: my Fair PE / QFMA model assigns a target multiple based on quality, growth durability, margin profile, moat width, and management quality, then applies it to a forward earnings number 2–3 years out and discounts back.
For Booking Holdings, here is how I get to Fair PE = 22x:
DriverBooking scoreContributionGross margin (~87%)Top decile+Adj. EBITDA margin (~37%, expanding)Top decile+ROIC (~93%)Outstanding++Organic revenue growth (high-single to low-double)SolidneutralCapital return disciplineOutstanding+AI disruption riskReal but overstated–CyclicalityModerate–Geopolitical/regulatoryCurrently elevated–Fair PE22x
Earnings path
Management’s 2026 guidance implies adjusted EPS of roughly $10.82 (pre-split equivalent ~$270; post 25:1 split this is the per-share economics that matter going forward). Applying my normalised growth assumptions:
2026e EPS: ~$10.80 (matches guide)
2027e EPS: ~$12.30 (Middle East normalising, Connected Trip continuing to scale, mid-teens EPS growth, buyback contribution)
2028e EPS: ~$14.00 (mid-teens continued; baseline scenario, not bull case)
Fair value
Fair PE 22 × 2028e EPS of $14.00 = $308 per share
Against today’s $154 price, that implies an annualised upside of approximately 24% per year over the next three years – or roughly a doubling if I am directionally right.
This is not a heroic forecast. It assumes:
Hormuz normalises sometime between H2 2026 and H1 2027 (consistent with management’s own planning assumption that the impact persists through end of June)
Connected Trip continues to grow in the high teens, gradually mixing up to a mid-teens share of transactions
Mobile app penetration continues its grind from high-50s toward 70%
EBITDA margins hold around 37% (no further expansion assumed in my base case – that would be the upside)
No further multiple compression from AI fears
On Bloomberg consensus, BKNG trades at ~15.8x NTM P/E and ~12.1x NTM EV/EBITDA – a multi-year low for this business and below where lower-quality peers like Expedia trade. For a business with this margin profile and ROIC, that gap should not persist.
6. Risks & Why the Opportunity Exists
If this is so obvious, why is the stock $154 instead of $230? Three reasons:
1. Macro overhang from the Strait of Hormuz. The market is treating a temporary fuel-price and travel-demand shock as if it will permanently reset Booking’s growth rate. It will not. Past shocks – the 2009 financial crisis, the 2014 Crimea annexation, the 2016 Brexit vote, the 2020 pandemic – have all turned out to be excellent entry points for high-quality travel businesses. The pattern is consistent: bookings compress, cancellations spike, then demand snaps back, often above the previous trend. My base case is that Hormuz follows the same playbook.
2. AI disintermediation panic. Investors have been bidding down the entire OTA complex on the assumption that LLM-powered agents will route around the OTAs. I addressed this above. The short version: Booking owns the supply, owns the customer, has the buyback firepower to compound through any transition, and is itself the most aggressive AI deployer in the industry. The fear may be wrong, but as long as it persists, it caps the multiple – which is the gift.
3. Regulatory risk in Europe (Italy AGCM, DMA, etc.). This is real and will cost some operating margin over time. But Booking has a long history of navigating EU regulation. Settlements, behavioural commitments, and modest commission adjustments are the historical pattern – not structural impairment.
4. Negative equity / rising leverage. Aggressive buybacks have driven shareholders’ equity to roughly –$8.7B. New senior notes added in May 2026 increase interest expense in a higher-rate environment. I do not view this as fundamental risk given the FCF profile, but it is a watchpoint, especially if travel demand surprises to the downside.
5. The flight vertical carries lower margins. As Connected Trip scales, mix shifts toward flights and attractions, which dilute the historical accommodation-heavy margin profile. Management acknowledges this in filings. The offset is volume and customer lifetime value – but the unit economics here need to be monitored.
7. Latest News & Earnings
A non-exhaustive list of what has happened since the Q1 2026 print on April 28:
April 28: Q1 beat across the board; FY2026 guidance cut; record $3.6B buyback announced
Early May: $750M USD + €1.9B EUR senior notes issued; rate-locked debt refinancing
May 4: Priceline launches Route 66 Negotiator campaign with Randall Park – early marketing for the US summer travel season
May 4: Morningstar published its “fair value $391” view (uncertainty rating: high) – arrived at via different methodology than mine but directionally supportive
May 5: Glenn Fogel scheduled to speak at the J.P. Morgan TMT conference
Throughout May: Multiple sell-side price target cuts – TD Cowen $230, DA Davidson $230, Cantor $175, Argus $205. Average target ~$226 vs. spot $154
May 12 (Skift): Editorial criticism of buyback-vs-strategic-investment balance
May 13: Stock down 3.45% on broader market weakness; volume picking up on the downside
On X/Twitter (which I monitor daily): sentiment is genuinely bifurcated. The “this is a layup” camp (Joe Frankenfield, FundamentalEdge, several of the European value accounts I follow) has been buying. The “AI will eat OTAs” camp has been louder. When the loud camp is wrong and the quiet camp is buying, that is usually when the asymmetry is worth taking.
8. Conclusion & Investment Case
I have been at this for long enough to know that the best investment opportunities in high-quality businesses come from temporary, identifiable, and reversible problems. Booking Holdings in May 2026 has all three:
Temporary: Hormuz will reopen, or travel patterns will route around it. The IEA called it the most significant supply shock in oil market history – which is exactly the kind of statement that marks the peak of fear, not the start of a structural change.
Identifiable: We can literally see the cancellation spike in the March 2026 numbers (room nights +1% vs. underlying +8%). We can quantify it.
Reversible: Management has already signalled an H2 2026 recovery in its own planning assumption.
Catalysts for a rerating (12–24 months)
Middle East de-escalation / Hormuz reopening. This is the single biggest one. A ceasefire announcement, a Hormuz transit normalisation, or even just easing jet fuel prices would remove the overhang within days. We saw something similar in 2022 with the Russia-Ukraine oil spike – the worst-case scenario priced in by markets ended up being far worse than what actually transpired.
Mobile app penetration crossing 60%, then 65%. Each percentage point of direct mix reduces Google performance-marketing dependence and expands the operating margin. This is a quiet, structural rerating driver.
Connected Trip mix crossing a mid-teens share of transactions. Currently low double digits. Once the market sees Connected Trip as a step-function in customer lifetime value rather than a pet project, the multiple should reflect the second S-curve.
AI agent integrations and Penny conversion data. If Booking can demonstrate (and they will, on the Q2 or Q3 call) that AI-assisted bookings have higher AOV, lower cancellation, and better retention, the disintermediation thesis dies and the multiple re-expands.
Sustained share buybacks at depressed prices. Every $3.6B of stock retired at $154 is structurally accretive forever. Fogel is essentially printing future EPS for shareholders patient enough to sit through this.
US share gains. US room night growth in the low teens for four consecutive quarters means Booking is taking share from Expedia and Airbnb in the largest, most mature travel market in the world. This is the most underappreciated part of the bull case.
My position
I own BKNG in the Haas Invest4 Innovation Fund (invest4.net) and have been adding through this drawdown. My 3-year fair value is ~$293–$308, implying ~24% annualised upside, with a margin of safety that I find rare in mega-cap quality. If the market is right and I am wrong about AI disintermediation, I expect to lose maybe 10–15% from here (the downside is well-supported by the buyback floor and the FCF yield). If I am right, I expect to roughly double my money over three years.
That is the kind of asymmetry I get out of bed for.
Conflict of Interest Disclosure
Booking Holdings (BKNG) is a portfolio position in the Haas Invest4 Innovation Fund (invest4.net), the cost-efficient innovation-focused investment fund I manage. I personally also hold the stock. I may add to or trim the position at any time without prior notice. The article is written in my personal capacity as a publisher at investresearch.net and does not constitute investment advice.
Risk Disclaimer
This article reflects my personal opinions and analysis based on publicly available information as of May 19, 2026. It is not a recommendation to buy or sell any security. Investing in equities involves substantial risk, including the risk of total loss of capital. Past performance is not indicative of future results. Forward-looking statements – including valuation targets, growth forecasts, and EPS projections – are by their nature uncertain and may prove materially incorrect. Geopolitical events (including but not limited to the ongoing Middle East conflict), regulatory developments, foreign exchange fluctuations, changes in consumer behaviour, technological disruption (including AI-driven disintermediation), and broader macro conditions could materially affect Booking Holdings’ business and share price.
You should conduct your own due diligence and, where appropriate, consult a licensed financial advisor before making any investment decision. Do not invest more than you can afford to lose.
If you found this analysis useful, please consider subscribing to my Substack and sharing it with a fellow investor. For more research like this on global small- and mid-cap equities, visit investresearch.net and follow the work of the Haas Invest4 Innovation Fund at invest4.net.
– Philipp Haas
You never mentioned anything about the rising disruption by Google VR and Houfy
Agree. well written