Latest / Stock Picking with Fexingo: Individual Equities, Research, and Building a Concentrated Portfolio / How Meta Stock Won the Week After Earnings
Transcript
- Lucas: Meta shares jumped 11 and a half percent this week. That's the biggest single-week gain among megacap tech in a while, and it brings the stock to 669 dollars — a new high for the year. Luna: Eleven and a half percent in five days. That's not a grind higher — that's a rerating. What was the trigger? Lucas: Earnings, plain and simple. Meta reported its second-quarter results on Wednesday after the bell. Revenue came in at 42 point 3 billion — that's about 800 million above the consensus estimate. But the real story was operating income: 18 point 9 billion, a margin of nearly 45 percent. Luna: Forty-five percent operating margins. For a company that was spending like a tech startup two years ago on the metaverse. That's a dramatic shift. Lucas: Right. And the market is finally giving credit for the discipline. Mark Zuckerberg called this the 'year of efficiency' back in 2023 — it took a while to show up in the numbers, but the last three quarters have been clean. This quarter, the margin expansion came from headcount reduction and from ai driven ad targeting that increased conversion rates. Luna: So the AI spending — which was huge, I think capex this year is something like 40 billion — the market is now satisfied that it's yielding returns? Lucas: That's exactly the debate. Capital expenditure this year is guided to 38 to 40 billion, up from 28 billion last year. But management showed a slide on the earnings call: they claimed that the ai powered recommendation engine increased time spent on the platform by 8 percent year-over-year, and that drove ad inventory growth without adding people. That's the narrative that worked. Luna: And that's the difference between Meta and, say, Tesla this week, which fell almost 3 percent. Tesla also spends heavily on AI — but the payoff is less visible in the near-term financials. Lucas: Exactly. Meta's AI payoff is showing up in the core advertising business, which is already highly profitable. Tesla's AI bet is autonomous driving and robotics — much longer payback. So the market applies a different discount rate. Luna: Let's talk about the broader market context, because it matters. The S&P 500 is at 7575 — up half a percent on the week. The NASDAQ did a bit better, up 0.6 percent. But Meta's 11 percent gain is way outside that. It's a stock-specific story. Lucas: It is. And look at the other big tech names: Apple up 0.9 percent, Microsoft down 0.4 percent, Alphabet down 2.5 percent, Amazon up 0.5 percent, Nvidia up 7.9 percent. So the AI theme is still alive — Nvidia had a strong week too — but Meta was the standout. Luna: Nvidia up 7.9 percent — that's notable. Did they have a catalyst too? Lucas: Not a company-specific one. It was more of a macro rotation back into AI infrastructure names after a rough June. There was a headline earlier this week — Kevin Warsh, the Fed chair, named his task forces, and the market maybe read that as policy continuity. But honestly, Nvidia's move was mostly technical — it had sold off 15 percent in the prior month, and this week was a bounce. Luna: So for a stock picker listening right now: Meta is up 11 percent in a week. Do you chase it, or do you wait? Lucas: I think you have to look at the valuation. Meta trades at 28 times forward earnings now, up from about 24 before earnings. That's not crazy expensive for a company growing revenue at 18 percent and expanding margins. But the risk is that the AI capex cycle peaks next year — if returns on that spending start to decelerate, the multiple could contract. Luna: So the key metric to watch is return on invested capital for the AI projects. They gave some data this quarter — an 8 percent increase in time spent — but that's a lagging indicator. Lucas: Right. The leading indicator is ad pricing. If Meta can continue to charge more per ad because the targeting is better, that's a direct line to revenue growth without adding users. And user growth in the US and Europe is basically flat — it's all coming from Asia-Pacific and Rest of World. Those users are lower ARPU, but the ad tech is improving fast enough to make them more valuable over time. Luna: One thing I noticed in the data: the S&P 500 is at 7575, but the Dow is down 0.8 percent this week, and the Russell 2000 is down 1.1 percent. So the rally is still very narrow — it's large-cap tech pulling the market up, while small caps and value stocks lag. Lucas: That's been the story for most of 2026. The rotation out of megacap tech that some predicted in early summer hasn't really materialized. JPMorgan is down a bit, Chevron is up 4.9 percent — so energy is working — but the market isn't broadening out in a convincing way. Luna: For a concentrated portfolio, that raises a question: do you overweight the winners like Meta, or do you trim and rotate into the laggards? Lucas: I'd argue you trim if the position becomes too large. But the bar for trimming is high — if you sell Meta at 28 times earnings because it feels expensive, and it goes to 32 times because the business keeps accelerating, you've lost the upside. Better to set a price target based on your estimate of intrinsic value and only sell when the stock reaches that. Otherwise you're just trading on emotion. Luna: Speaking of which, I think this episode is worth a coffee to anyone who's been trying to figure out whether to hold or sell after a big move. If our conversation has helped you think about that decision differently, that's the link. Lucas: Yeah, buy me a coffee dot com slash fexingo. That keeps the show going and ad-free. We appreciate it. Luna: Alright, back to the stock. Lucas, what's your framework for deciding how much to invest in Meta at this point? Lucas: I use a simple rule: the position size should be proportional to your conviction that the intrinsic value is at least 30 percent above the current price. For Meta at 669, I'd want to believe the business is worth around 870 or more. That requires a discounted cash flow model where you assume 15 percent revenue growth for the next three years and then a gradual decline to 5 percent, plus a terminal margin of 40 percent. That gets you to about 850. Luna: So you're saying at 669, there's still room — but not a massive margin of safety. Lucas: Correct. I'd say it's fairly valued to slightly undervalued. But the risk is that if the economy slows and ad budgets get cut, that 15 percent growth assumption falls apart quickly. So I'd keep the position to maybe 5 percent of a concentrated portfolio — enough to matter, not enough to ruin you if it drops 30 percent. Luna: And you'd compare that to other opportunities. Would you rather own Meta at 28 times earnings or Apple at 32 times? Apple is up only 0.9 percent this week — it's steady but not exciting. Lucas: Apple is the tortoise, Meta is the hare. Apple has a services margin story that's incredibly durable — 74 percent gross margins on services now. But its growth is slower, maybe 6 to 8 percent. Meta has higher growth but more cyclical risk. For a picker, it depends on your time horizon. If you're looking three to five years out, I'd lean Meta. If you're looking for a ten-year compounder that never surprises you to the downside, Apple. Luna: And Nvidia? At 210 dollars, up 8 percent this week — that's a different beast. The valuation is higher, the volatility is higher, and the AI capex cycle is the only driver. Lucas: Nvidia is a trade, not an investment. I don't mean that as a put-down — it's a phenomenal company with a massive moat in GPUs. But the revenue is dependent on a few hyperscaler customers who can change their plans quickly. Meta itself is one of Nvidia's biggest customers. If Meta cuts its capex next year, Nvidia feels it. So for a concentrated portfolio, I'd rather own the customer than the supplier. Luna: That's an interesting lens. Meta is the customer, Nvidia is the supplier. And Meta's demand for Nvidia chips is what drives Nvidia's revenue. So if you think Meta is a good investment, you might indirectly be bullish on Nvidia — but you're taking on more risk. Lucas: Exactly. The supply chain is more volatile than the end market. I'd rather own the platform with the recurring ad revenue than the chip company that has to keep innovating every year to stay ahead. That said, Nvidia has surprised everyone for years. I'm not betting against it. Luna: Let's zoom out. The market is at all-time highs. The S&P 500 is 7575. The NASDAQ is 26282. Valuations are elevated by historical standards. How does that affect your stock picking approach? Lucas: It means you have to be more disciplined about the price you pay. In a cheaper market, you can buy good companies at 15 times earnings and let the multiple expand. Today, good companies are at 25 to 30 times. So the return has to come from earnings growth, not multiple expansion. That's harder — but it also means the stocks that do grow earnings will be rewarded disproportionately. Luna: And that's why Meta's earnings beat was so important. It confirmed that the growth story is real. Lucas: Right. The market is starved for growth. When a company delivers it, the stock moves big. That's the environment we're in. Luna: So for the concentrated portfolio: Meta is a hold or a modest buy. But keep an eye on the AI capex ROI. And compare it to other names like Apple for stability. Lucas: And one last thing: don't get anchored to the week's move. Meta was 315 dollars a year ago. It's doubled. That doesn't mean it's a sell. It means the thesis is working. Stick with it until the fundamentals change. Luna: That's a good note to end on. Next episode, we'll look at another stock that had a big week — maybe Chevron, up almost 5 percent. See if there's a similar framework. Lucas: Count on it. See you next time.