Broken Segues and Segways: From Chips to Dips in the Market

Show Notes

Dan and I opened on a piece of news that just made me happy: TSMC's first fab in Arizona is reportedly pulling chip yields about four percentage points higher than the comparable plants back in Taiwan. These aren't the easy chips, either — it's leading-edge work, with the first fab on the four-nanometer process and a second fab eventually pushing into the three and two-nanometer range, where you're basically counting atoms. The fact that they can scale that kind of knowledge across an ocean is genuinely impressive to me, and there's room for something like six fabs on that site. Naturally Dan asked the question I hadn't thought of — where does the water come from? So we did what we do and asked the interwebs, and it turns out they pull from the Colorado River through four big underground pipes, plus the Salt and Verde Rivers and a little Phoenix groundwater. At full tilt the first fab runs around four and three-quarter million gallons a day, but they recycle about sixty-five percent of it, which makes the number a lot less scary than the headline would have you believe.

That bled into a long stretch on a book I blazed through on my commute, the abridged Neil Postman, Amusing Ourselves to Death. The whole premise is that contrast between Orwell and Huxley — control by constriction versus control by drowning you in pleasure — and Postman's bet, written back in nineteen eighty-five, was that television was carrying us toward the Huxley version. We chewed on his arc from print to teletype to the newscast, how information got chopped into smaller and smaller chunks until you end up informed but not learned. Dan and I kept landing on the same word, which Dan supplied: inoculation. How do you come through an education system able to consume misinformation and disinformation and still find the actual truth? It's information versus synthesis, a theme we keep circling back to — we need a lot more people who can synthesize and a lot fewer who can only repeat. The funny part is Postman closes by waving off the personal computer entirely, which from here looks like the one thing that maybe answers some of his own complaints.

From there it got personal in the best way. I told Dan about Steamboat, a hotel room with actual cable, and the basket full of tears when the commercials wouldn't let my kid watch her program. That's the generational tell — my dad grew up with three channels and sometimes just static, I grew up rationing whatever was on the four networks after school, and this generation simply cannot compute that what you want isn't available the instant you want it. We both admitted YouTube has quietly become our television, and that we've gotten to the point where a single commercial break sends us up off the couch looking for a chore. I'll pay fifteen bucks to rent something on Amazon before I'll sit through ads now, which is equal parts privilege and broken attention span. We even wandered into how hunter-gatherers spend their time — a lot of rest, a lot of squatting instead of decimating our posterior chains in office chairs — before catching ourselves.

Then came the broken segue that gave the episode its name, and the real meat: news you can actually act on. I pulled up the Shiller PE — the cyclically adjusted price-to-earnings ratio that smooths for inflation — and we were sitting right around thirty-seven. I'd run the table version and found we've been above that level less than two percent of the time in the whole history of the index, the only real company being the dot-com bubble and late twenty twenty-one. At a thirty-seven multiple you'd better hope earnings climb, because otherwise that's roughly thirty-seven years to get paid back. Dan and I both said it at the same time: the trade is SPY puts, or get surgical and underweight the highest-PE names while keeping the true rocket ships — the Nvidias whose earnings have actually kept pace. We talked about Nvidia maybe doubling its forecast, though I told Dan flat out not to quote me on the exact figures, and good thing, because they were loose.

On the other side of the ledger we looked at debt versus equity — Howard Marks' tidy point that those are really the only two asset classes — and the ten-year-minus-two-year curve finally un-inverting after the longest inversion on record. That's sometimes, not always, a recession flag. I floated high-yield bonds as a place to hide, and Dan immediately and correctly poked the hole: if you expect a recession you'd expect more corporate defaults, and sure enough, when we pulled up HYG it tracked the S&P almost perfectly through two thousand eight. Not the defensive play I wanted it to be. So we drifted into stories instead of math — Boeing as a misbalanced, financially-engineered organization that's let R and D wither, the kind of national-security company that, like TSMC or old Chrysler, can't really be allowed to fail, which makes its bonds interesting even when its stock isn't. We never did place the paper-trade bet, but Dan summed the whole thing up about as cleanly as it can be summed.

The whole bet is whether it's the P or the E — and that, as Dan said, is the entire point of a ratio.

— Sean