David Hay — Portfolio Update: UBER (Haymaker, Jun 2026)

One-line take: A portfolio update reaffirming UBER as a Buy (entered ~$84 in mid-January, now ~$70) — one of the world's great businesses (~$10B annual FCF, earnings +44% YoY) priced for existential threats it doesn't face. The core argument inverts the bear case: autonomous vehicles are a tailwind, not a threat — an idle self-driving car earns nothing, and Uber's demand network (202M MAUs, 40M daily trips) gives AV operators ~30% higher utilization, while removing the ~70%-of-fare driver cost lets Uber's take rate climb (~$20B incremental revenue at 20% AV penetration). Uber sits as a "toll booth" across 14+ AV partners (Waymo, WeRide, Pony.ai, Lucid-Nuro, Wayve, Waabi, Baidu) via Uber Autonomous Solutions. Valuation: 18x trailing FCF, PEG 0.64, SOTP DCF $110–130; 45 of 46 analysts Buy. Separately, he moves IBM to a Sell and off the Hold/Trim list — up ~50% since his Jan-2024 highlight, now its loftiest P/S & P/E in 25 years — to raise cash on "ominous" market signals and concentrate on best ideas. "Cash is definitely not trash."

1. Stocks & names mentioned

2. Talking points

Ticker	Name	Research	View	What he said	At
UBER	Uber Technologies	QT · SA · STK · FA	Positive	Maintaining Buy (entered ~$84, now ~$70). ~$10B FCF growing 42%, 18x trailing FCF, PEG 0.64; SOTP DCF $110–130. AVs are a tailwind not a threat — Uber's network gives AV operators ~30% higher utilization; as AVs remove the ~70%-of-fare driver cost, take rate rises (~$20B incremental revenue at 20% AV penetration). A "toll booth" across 14+ AV partners.	read
TSLA	Tesla	QT · SA · STK · FA	Neutral	The "Tesla marginalizes Uber" bear narrative: Tesla's AV service is still in testing with ~60 fatalities attributed to its autopilot vs zero driving deaths from Waymo tech failure — Hay thinks an Uber/Tesla cooperative relationship is more probable than displacement.	read
EXPE	Expedia Group	QT · SA · STK · FA	Neutral	April partnership lets Uber's 202M MAUs book hotels in-app — "the first tangible step toward a travel super-app" competing structurally with Amazon Prime.	read
NVDA	NVIDIA	QT · SA · STK · FA	Neutral	Partner in the Munich robotaxi test (with Uber and Autobrains) that went live this week — one data point on the widening AV footprint.	read
AMZN	Amazon.com	QT · SA · STK · FA	Neutral	The super-app comparison/competitor — Uber's Expedia tie-up is framed as the first move toward something that "competes structurally with Amazon Prime."	read
BIDU	Baidu	QT · SA · STK · FA	Neutral	One of the 14+ AV partners and a customer of Uber Autonomous Solutions' white-label suite — part of the "toll booth across all AV winners" thesis.	read
WRD	WeRide (ADR)	QT · SA · FA	Neutral	Uber's robotaxi-ops partner — launched Spain's first commercial robotaxi pilot in Madrid and runs the live Riyadh/Abu Dhabi ops; a flagship AV-partner relationship.	read
PONY	Pony.ai (ADR)	QT · SA · FA	Neutral	An AV partner and Uber Autonomous Solutions white-label customer — part of the 14-partner toll-booth network.	read
LCID	Lucid Group	QT · SA · STK · FA	Neutral	The Lucid-Nuro partnership is how Uber deploys its own AV supply in San Francisco — rebutting "Waymo kills Uber" by showing Uber also owns AV capacity, not just dispatch.	read
Waymo	Waymo (Alphabet unit)	—	Neutral	The "Waymo kills Uber" bear case rebutted: 500k weekly rides vs Uber's ~24M daily trips (>300:1; ~170:1 even at Waymo's 1M target), and Waymo's own data shows 30% higher utilization dispatched via Uber vs its standalone app. A lead AV partner, not a killer.	read
DHER	Delivery Hero (Frankfurt)	—	Neutral	The first bear narrative: Uber lifted its stake to 24.99% and explored a full takeover. Hay's counter — Delivery Hero's Talabat dominates food delivery across nine GCC countries, the same markets where Uber has Vision 2030 AV ops; owning the AV platform + the food network = a Gulf super-app.	read
IBM	International Business Machines	QT · SA · STK · FA	Negative	Moved to a Sell and pulled off the Hold/Trim list. Up ~50% since his Jan-29-2024 highlight (despite a 14% pullback off last week's spike), now its highest valuation in a decade — P/S and P/E loftiest in 25 years. Raising cash on "ominous" market signals + concentrating on best ideas; ~22x 2026E vs UBER's 18.6x.	read

Where we stand — Jan ~$84 entry, thesis intact

Recommended UBER in mid-January at ~$84 / a "very modest" 3.4x sales: one of the world's great businesses (~$10B annual FCF, earnings +44% YoY) priced as though it faced existential threats. "That perception hasn't really changed" — and the rebound "obviously hasn't happened yet," but he still strongly believes in both the company and the thesis.
Frames it with Druckenmiller (Morgan Stanley's Hard Lessons): "I like it when I have extreme conviction in a thesis and no one else believes it. It gives me even more conviction." That "perfectly sums up" his view on UBER.

Business momentum since January

Q4 2025 revenue +20% YoY, FCF +42%. Q1 2026: gross bookings +21% for the third straight quarter; CFO says earnings scale at >2x the topline growth rate.
46M Uber One members now drive half of all gross bookings, growing 55% annually. High-margin advertising crossed $2B annualized, +50%.
April Expedia partnership: 202M MAUs can book hotels in-app — "first tangible step toward a travel super-app" vs Amazon Prime. "Every financial metric that existed when we made the recommendation has improved."

Robotaxi expansions

Munich test with NVIDIA + Autobrains went live this week; WeRide + Uber launched Spain's first commercial robotaxi pilot in Madrid.
Autonomous operations now span Riyadh, Abu Dhabi, Austin, Atlanta and a growing list of international cities.

Bear narrative 1 — the Delivery Hero stake

Uber lifted its Delivery Hero stake to 24.99% and explored a full acquisition; the market prices it as a "dumb use of money."
Counter: Delivery Hero's Talabat dominates food delivery across nine GCC countries (UAE, Saudi, Kuwait, Qatar, Bahrain, Jordan, Iraq, Egypt, Oman) — the same markets where Uber has a signed Vision 2030 AV partnership (Saudi Transport General Authority) and live robotaxi ops in Riyadh/Abu Dhabi via WeRide. AV platform + dominant food network = a Gulf super-app.

Bear narrative 2 — "Waymo kills Uber"

Waymo's 500k weekly rides vs Uber's ~24M daily ride-hailing trips: >300:1 (still ~170:1 even at Waymo's 1M weekly target).
Waymo's own data: 30% higher utilization dispatched through Uber's network vs its standalone app. Uber also owns AV supply via the Lucid-Nuro partnership in San Francisco.

Bear narrative 3 — AI-cost discipline read as distress

Uber capped AI usage spending after exceeding budget and cut jobs in its People and Places division — Hay reads this as discipline, not distress, for a company generating $9.8B annual FCF.
(Related: the "Tesla dominates AI and marginalizes Uber" belief — Tesla's AV still in testing with ~60 autopilot-attributed fatalities vs zero from Waymo tech failure; a cooperative Uber/Tesla relationship seems more probable.)

The AV-as-accelerant take-rate math

AVs are framed as an existential threat; the operational data proves the opposite. An idle self-driving car earns nothing — utilization is the entire economics of the AV business, and Uber's demand network (202M MAUs, 40M daily trips) raises a Waymo vehicle's utilization 30%.
As AVs replace human drivers, the ~70%-of-fare driver compensation increasingly stays with Uber: on $193B annual gross bookings, moving 20% of trips to AVs at an 80% take rate (vs 28% now) adds ~$20B incremental annual revenue at full scale.

The "toll booth" + the training-data moat

Uber's $10B AV commitment spans 14+ partners (Waymo, WeRide, Pony.ai, Lucid-Nuro, Wayve, Waabi, Baidu) — "a toll booth on every autonomous trip."
Uber Autonomous Solutions (launched Feb 23, 2026) is a white-label suite — fleet management, proprietary training data from 40M daily trips, remote ops, AV insurance — for WeRide, Pony.ai, Wayve, Waabi, Baidu. The training data is a compounding asset.

Valuation

More attractive at ~$70/share: 18x trailing FCF, $9.8B FCF growing 42%, PEG 0.64. SOTP DCF fair value $110–130; ex-AV the core platform + advertising alone is ~$95–108 at conservative peer multiples.
45 of 46 analysts maintain Buy; consensus target $104.45; Morningstar DCF FV $97. Maintaining the Buy.

IBM — ring the register, raise cash

IBM is up ~50% since the Jan-29-2024 highlight (despite a 14% pullback off last week's spike) — now its highest valuation in a decade, P/S and P/E the loftiest in 25 years. Moved to a Sell and off the Hold/Trim list.
Two reasons: raise more cash given "ominous" signals from the overall market, and track fewer names / concentrate on best ideas. UBER at 18.6x 2026E vs IBM's 22x — "but cash is definitely not trash."

3. In plain English

UBER — Uber Technologies Positive

Everyone's afraid self-driving cars (robotaxis like Waymo) will put Uber out of business. Hay argues the opposite, and the key is one fact: a robotaxi only makes money while it's carrying a paying rider — an idle car earns nothing. Uber already has the riders — 202 million monthly users opening the app, 40 million trips a day. So a Waymo car that takes jobs from Uber's app stays busy about 30% more of the time than one waiting for fares in Waymo's own app. That makes Uber the most valuable thing to a robotaxi operator: the demand. So instead of competing with the AV companies, Uber signs deals with all of them — 14-plus partners (Waymo, WeRide, Pony.ai, Baidu and others) — and takes a cut of every autonomous ride. That's the "toll booth": Uber doesn't need to pick the winning robotaxi maker, it gets paid no matter who wins.

There's a second twist that actually makes Uber more profitable as robotaxis spread. Today, when you pay for an Uber, roughly 70% of the fare goes to the human driver and Uber keeps ~28% ("take rate"). Remove the driver and that 70% is up for grabs — Uber can keep far more of each fare. On Uber's ~$193B of annual bookings, shifting even 20% of trips to robotaxis could add roughly $20 billion of new revenue. So the thing the market fears (AVs) is actually the thing that could fatten Uber's margins.

Meanwhile the business is already humming — revenue up 20%, free cash flow up 42%, the subscription (Uber One) and advertising businesses growing fast — yet the stock has fallen from ~$84 to ~$70, which is only about 18x its cash flow and cheap relative to its growth (PEG 0.64). Hay's read: the market has misjudged the AV story, which is exactly why the stock is cheap, and that's why he's keeping his Buy. Almost every analyst (45 of 46) agrees it's a Buy, with fair-value estimates of $110–130.

IBM — International Business Machines Negative

Summary & excerpts derived from the paid Haymaker Substack post (saved text in transcript.txt) for personal study. Not investment advice. © Haymaker / David Hay for source material.