Longreads
- Richard Hanania has an interesting pair of posts on why many US institutions seem more politically liberal/progressive and when this accelerated. Note that these are positive, not normative, claims: he's asking what's going on, not whether or not it's a good thing. With many institutions, there's a formal org chart and then a set of informal mechanisms that decide what will actually get done. In Hanania's model, conservatives invest less energy in politics’ informal mechanisms, in part because they care less. In a way, it's a reflection of both sides' fundamental worldviews. If your view is that a) the world can be changed in significant ways, and b) institutions like government and business will have to change, too, you'll naturally invest effort in making that happen. If your view is that social change is a poor risk/reward, and that the government doesn't do a great job of it, you're only energized insofar as the other side is winning. So a 50/50 split in viewpoint can mean an 80/20 split in what actually gets done. This can run in the other direction, too; if big business is uniformly conservative, but the electorate is 50/50, there's a concentrated economic interest in preserving the status quo and a more diffuse interest in changing it, so outcomes in that scenario might lean more conservative than polling numbers would suggest. Matt Levine makes an adjacent point here, when he notes that the opposite of an ESG fund is just a regular fund, not an anti-ESG fund. Anti-woke investing is niche, but un-woke investing is the default.
- Brian Lui has a piece I wish I'd written on upside decay. It ties together the histories of companies, the histories of countries, and a model for understanding prosocial interactions. Strongly recommended.
- Building in the US is expensive, but are we an outlier because of expensive labor, expensive materials, or something else? This piece makes a good data-driven argument for a specific "something else."
- Tyler Cowen interviews Daniel Carpenter: a good piece to read if you're disappointed in how healthcare regulators responded to Covid-19 (as I am). Carpenter has some useful thoughts on the constraints they operate under. I'm not persuaded by all of the claims Carpenter makes, but he does put the FDA's and CDC's behavior in better context. The asymmetry remains, though: the reported four-day weekend the agency took in November 2020 is the kind of thing that ought not to be repeated.
- Ken Norton interviews Michael Siliski on building products at Stripe. Particularly worth reading is the "shaping" discussion; it's an API-first way to think about what problem is actually getting solved.
Books
- Amazon Unbound by Brad Stone is a sequel to his Everything Store. In one sense, it's just a few more years of relentless iteration (and, yes, more scandalous stories about various people's personal lives than the last one). The main difference is that Amazon's expansion now seems to happen in multiple dimensions. The Amazon of The Everything Store was a mostly horizontal company, that connected shoppers to merchants online and had a growing sideline in selling its computing platform. The modern Amazon experiments with physical retail, operates a much more extensive logistics network, and, because of its scale and economies of scale, increasingly sets the de facto US minimum wage.
Open Thread
- What kinds of skills generalize across many domains? A common thread in stories about investors is that they were very good at poker or bridge, and I've previously noted that getting good at video games or arbitrage is unusually common in the early careers of successful people. Which other skills generalize like this?
- As always, feel free to drop in any interesting long-form links, or good open questions that would be of interest to Diff readers.