Inflation hedges are cheap. The AI party won't last.

Transcript

Tom Inflation hedges at fifty-two-week lows, AI capex sailing past a trillion, and Korean stocks ripping eleven-point-eight percent in a single session—this tape is wild. Let's dig in, buddy.

Marie It's the New York Edition, July thirty-first. I'm Marie, joined by Tom and Gerald. Gerald, yesterday you were all over the Iran risk premium, and Shell's kept climbing, hasn't it?

Gerald Ha—fair play, it did. I mean, it's always nice when a buy call works for more than five minutes. But today the radar's lighting up with something else entirely—AI spending hitting a trillion, BP selling off its UK North Sea business, and Japan intervening in the yen all at once.

Tom The AI trade is on fire, Gerald! Korean equities up eleven-point-eight percent last session—that's not a rally, that's a launch sequence.

Gerald Yeah look, the momentum's ferocious, sure. But spending a trillion dollars on AI infrastructure when no one's asking the ROI question? That's the part that makes me nervous.

Marie Hold on—that's exactly my point. The FT's flagging it today: innovate-heavy portfolios need to brace, and inflation hedges are underowned. TIP at a fifty-two-week low, gold down five percent year to date. If that doesn't scream cheap, I don't know what does.

Tom But inflation's been a paper tiger for months, Marie. You buy TIPS now and you're just fighting the tape. Meanwhile, Nvidia's trading at fifteen times forward earnings—fifteen times, Gerald! After twelve hundred percent in a year. That's not a bubble, that's a bargain.

Gerald Alright, but hang on—twelve hundred percent return and a fifteen P-E sounds magical until the capex tap slows. The minute AI spending decelerates, that multiple shrinks fast. And Microsoft surged fifteen percent last session, sure, but Alphabet was down a percent. Not all AI spenders are equal.

Marie Right, and that's the dispersion they're not talking about. The trillion-dollar figure is a big round number, but are these investments actually generating returns? The press isn't asking that yet. It's the biggest gap in the coverage.

Tom Okay, but the mergers and acquisitions pipeline is at a record two-point-eight trillion globally—that's a massive fee tailwind for Goldman and Morgan Stanley. Goldman up four and a half percent last session, still eleven percent below its high. That's room to run.

Marie To be fair, Goldman and Morgan Stanley have rallied over forty percent from a year ago. Not exactly a hidden gem. But the deal flow is there, no argument.

Gerald And it's not just Wall Street banks. The AI arms race is driving Asian mergers and acquisitions too. Samsung's trading unit teaming up with a Japanese firm for a biomass power plant for AI demand—that's the kind of cross-border deal that keeps this going.

Tom Exactly.

Marie Hundred percent.

Gerald That's the whole story.

Tom And Nvidia's still the pick-and-shovel play. Every dollar of that capex eventually flows through their GPUs.

Gerald Yeah, but—Tom, remember when you said semis were cooked in Q2? I mean, you bounced back hard on this, but the capex tsunami can recede. That's the thing.

Tom That was then, buddy. This time it's different—it's always different until it isn't, right? But the data's on my side today: the Korea ETF up fifty-eight percent year to date.

Marie Tom, that's your 'I sounded the same about Cisco in 2020' moment. But let's pivot—Japan just intervened to buy yen, and it worked for about five minutes before the market faded it. The BOJ held rates, so the carry trade is still alive.

Gerald Alive and kicking, mate. The yen was up over two and a half percent in the session, but that's already being sold into. Short-term traders should watch this fade. Intervention without conviction is just a speed bump.

Marie And the equities market didn't care. Japanese stocks up four percent, riding the AI wave. So the currency headwind isn't hurting exporters—yet.

Tom For real? Even a yen intervention can't break the tech rally. That's the strength of this move.

Gerald Or the complacency. One of the two.

Marie Look, I'm going to push back on both of you. The FT piece is the most original take today: inflation hedges are cheap because no one's inflation-worried. But the minute a C P I print ticks up, TIP and gold rip. It's asymmetric, pure and simple.

Tom Okay, but what about the quake in Japan? Toyota and Honda could get supply-chain whacked, and they're already down double digits year to date.

Gerald Right, and on the flip side, rare-earth-free motors for German cars? BMW and Volkswagen are down thirty percent plus year to date, but that innovation is a structural cost-cutter. Might be a small tailwind.

Marie Sure, but rare earth miners are already forty percent below highs and down fourteen percent year to date—that substitution narrative is priced in. The quake disruption for auto makers is the near-term risk.

Tom But El Niño, Marie! Thai rice and Indian sugarcane showing stress—soft commodities could spike. The agriculture basket down two percent last week, sugar flat year to date. That weather premium hasn't been priced at all.

Gerald Oh, so now you're an agricultural commodities guy, Tom? What's next, dairy futures?

Tom Hey, I diversify. But seriously, it's another cheap hedge against a risk that's already showing early signs.

Marie And that's the point about hedging inflation: cheap, unloved, and the warning lights are flashing. The whole market is chasing AI and ignoring the risks.

Gerald Alright, but BP's North Sea sale—that's a value play. They're streamlining, stock's up twenty-five percent year to date. Ithaca's the likely buyer, already up forty-nine percent. The mergers and acquisitions premium might be baked in, but it shows energy is still being restructured.

Marie And the energy sector overall is up twenty-nine percent year to date. So even without a headline deal, the trend is your friend. Hold BP, watch Ithaca.

Tom Alright, let's synthesize: AI spending tsunami, mergers and acquisitions records, and cheap hedges. Our house view: TIP at a fifty-two-week low is the cleanest asymmetric bet. You get paid to wait, and if inflation even twitches, it skyrockets.

Gerald Exactly.

Marie That's it.

Tom Nailed it.

Gerald And on the growth side, selling some Nvidia strength to buy cheap out-of-the-money puts on the Nasdaq 100—that's a smart way to hedge against this tech concentration risk.

Marie Wait, but we haven't asked the most important question: is that trillion-dollar spend generating actual returns? The press isn't scrutinizing it, but we should. The silence is the biggest gap.

Tom Right, Microsoft's cloud growth suggests yes, but Alphabet's dip says not every AI bet is a winner. That differentiation matters—maybe we buy Nvidia and Microsoft, but be selective.

Gerald And that's the nuance: AI is real, but it's not a monolith. Same with energy divestiture: BP streamlining is good, but no deal price yet. Don't chase Ithaca at forty-nine percent up year to date.

Marie El Niño crops, auto disruptions—these are tail risks that the market isn't pricing. Wheat futures up twenty-three percent year to date but down two percent last week: a weather spike could be coming.

Tom So the playbook: buy cheap hedges, ride AI selectively, and keep some powder dry for the next inflation print. That's the show, folks.

Marie As always, none of this is investment advice. We're just looking at the signals and thinking out loud.

Gerald And if you're just finding us, hit follow on Spotify—or check investmentflash.com for the full digest with charts and sources.

Marie We're back tomorrow morning at ten a.m. London time for the Weekend Edition. See you then.