- Patrick McKenzie's Bits About Money remains indispensable: this is a great writeup on buy-now-pay-later, explaining its core economics (and, importantly, why people who are likely to opine on the viability of Affirm and Afterpay do not overlap with the demographic for whom those sites make a big difference). In very broad terms there are only two mistakes you can make about lending. The first is making loans that don't get paid back, and the second is thinking that the only thing that matters for a loan's economics is whether or not it gets paid back.
- Jacob Bernstein in the NYT on Ron Perlman's business empire. New financial products like junk bonds can create fortunes from people who learn how to use them (the short version: one thing that kills value investors' returns is time, and if they can get enough leverage to acquire an entire company immediately rather than buying the stock and waiting, that's an opportunity to exploit big mispricings). In some cases, that means using a novel product to launch a career, but in other cases there's a one-time benefit followed by stagnation. I had some interactions with a corner of Perlman's business empire a long time ago; I bought shares of M&F Worldwide in 2002 or so, but unfortunately got out at some point before it went up 10x.
- Jesse Drucker and Maureen Farrell of the NYT on the qualified small business stock tax break, which is generous enough that it looks like a misprint: it allows profits from investing in some small businesses to be realized tax free up to either 10x the purchase price or $10m. This is one of the more surprising tax breaks out there; it's regressive given that most people investing in businesses are already well-off. On the other hand, it's great for economic mobility since the cases in which it applies are ones where the business founder started out with a small company and ended up selling a bigger one.
- Jamie Catherwood in Compound's The Archive on business clusters. He uses the example of late 19th-century Cleveland, where an early success in the lighting business created a self-sustaining community of tech companies. These clusters can be created (as 19th century Cleveland demonstrates), but they don't last forever (see 21st-century Cleveland).
- In the spirit of longreads, why not make Manhattan longer and wider? Expensive land makes cities specialize, which can increase wealth—it prices out some activities but means that the density of people engaged in other ones is higher, so it helps form economic clusters. But that's an inflexible model; the more expensive a city is, the more real estate prices end up being more costly for projects with a distant or uncertain payoff. It’s worked before.
- Asian Godfathers: Joe Studwell's How Asia Works is a great guide to what made Northeast Asia so prosperous in the last half of the twentieth century, and this book is a prequel of sorts talking about what kept other parts of Asia so poor. His main thesis is that key industries in the region got captured by financial interests—the "godfathers"—who extracted economic rents but didn't produce much wealth. The book is a bit more abrasive than How Asia Works, but it's a very good look at the flipside of fortunes built on banking monopolies, gambling licenses, palm oil plantations, and the like.
- Drop in any links that might be of interest to Diff readers.
- Are there good N-of-1 examples of companies or countries that either a) had a totally unique growth model, or b) used a growth model that failed in nearly every other case, but worked for them?
Diff Jobs is our recruiting service that matches Diff readers to interesting roles at companies looking to hire them. Some positions we're working on right now:
- Chief of Staff for a Series B edtech startup—a great role for anyone who wants to get a founder's-eye-view of rapid growth.
- Sales/business development roles in multiple industries, including crypto (US, remote).
- A head of security research at a crypto firm (US, remote).
- An ML Operations / Infrastructure Engineer at a firm that helps detect and stop fraud (Remote, Central Europe Time ideal).