Longreads

• Daniel Oberhaus and Caleb Watney on prospects for geothermal in the US, with an emphasis on what policy changes can be enacted right now to improve them. Geothermal is especially interesting from a political standpoint, because one of the obstacles to renewable energy is the large number of people who specialize professionally in digging very deep holes in the ground to extract energy. If those skills can be redeployed, vested interests matter less.

• Louise Matsakis, Meaghan Tobin and Wency Chen of Rest of World on fast fashion powerhouse Shein. In fashion one of the challenges seems to be aligning production cycles with hype cycles, or vice-versa; Shein has specialized in responding to Instagram/TikTok trends quickly at a small scale, then growing fast once there's demand.

• Amy Feldman at Forbes on Robotics firm Symbotic, which was incubated as part of a grocery wholesaler and is now valued at $5.5bn. The record for incubated businesses is mixed, partly because the incentives are so hard to manage: the new business takes a disproportionate amount of money and attention, which means that it generally looks bad on paper. Family-controlled businesses are one way around this. (Mitchell Energy & Development was a publicly-traded example of this: dual-class stock allowed the company to invest heavily in fracking in order to improve its natural gas business—but also allowed the CEO of a natural gas company to build a massive real estate development.)

• Tyler Cowen interviews Ray Dalio. It's striking how much of investing is pattern-matching, and how the more important an event is, the smaller the historical sample size. Dalio's depression studies were a small-N example, and one might characterize Taleb's framework as looking at some N=1 situations in order to study the category of "N=0 so far" ones.

• Tim Danton of ITPro on the story of the IBM PC (via Abnormal Returns). It really was an incredible deal for Microsoft, but it's worth considering what a bold bet it was for IBM, too. IBM was willing to make a pivot to a new product category and a new way of building products, and while the results mostly benefited other PC manufacturers and Microsoft, failing to make that bet might have been equally dangerous on a different time scale.

Books

• Cable Cowboy: A broadly favorable book detailing the rise of John Malone through the early 2000s—convenient, because the early 2000s are about when lots of media coverage went online, so it's easier to follow the story manually from there. The book is a mix of looks at individual operating problems and at the grand strategy of spinoffs and other value-realizing measures. It's favorable in terms of what it concludes, but balanced in terms of what facts it addresses (i.e. the book will detail something TCI did but say Malone didn't approve, or will note that anybody could have read the 700-page Liberty spinoff prospectus and figured out it was designed to benefit buyers).

• King of Content: If you're interested in stories about modern media moguls but prefer something much more depressing, considering this book instead of Cable Cowboy. The King of Content tells the story of Sumner Redstone, and the book gradually transitions from being about how to make a fortune to being a bleak story about family conflict. One of the underrated determinants of success is mere willfulness, which Redstone had in spades—he survived a hotel fire in 1979 by hanging out of the window until his wrist had been almost burned through. That same willpower, directed towards staying in control of a company, at the cost of other relationships, well past the point where that made sense.

Open Thread

• Drop in any links that would be of interest to Diff readers.

• Since this week's posts were all about well-known conglomerates with long track records, let's reverse that: are there any companies with newly-acquisitive managers who are trying to build such track records today? Private equity competition has made this harder, but maybe there are industries where it still works.

Leave a comment

A Word From Our Sponsors

Here's a dirty secret: part of equity research consists of being one of the world's best-paid data-entry professionals. It's a pain—and a rite of passage—to build a financial model by painstakingly transcribing information from 10-Qs, 10-Ks, presentations, and transcripts. Or, at least, it was: Daloopa uses machine learning and human validation to automatically parse financial statements and other disclosures, creating a continuously-updated, detailed, and accurate model.

If you've ever fired up Excel at 8pm and realized you'll be doing ctrl-c alt-tab alt-e-es-v until well past midnight, you owe it to yourself to check this out.