The Diff - Longreads + Open Thread

Longreads

• Brian Potter on how China's development since the late 1970s parallels the US's economic growth in the late 19th century. China grew faster, but there was a lot more catch-up growth available. One great point Potter makes is that the US and China had similar immigration-driven tailwinds to growth—in the US case, from international immigration, and in China's, from people leaving the countryside in order to get manufacturing jobs in cities. For both countries, having an effectively infinite supply of unskilled labor was exactly what they needed. That commonality is especially striking given how different they were: US economic growth was accompanied by a rise in state capacity (in the minimal sense that the government had to be big enough to give away land to railroads and impose high protective tariffs on American manufacturers). In China's case, it was a shift in how that state capacity was deployed: from being mostly repressive to being focused on identifying and scaling local reforms once it was clear that they worked.
• Tyler Cowen interviews Christopher Kirchhoff on drone warfare, hypersonics, big defense contractors, and more. A good piece about a broken system that is actually getting incrementally improved, though those increments are necessarily small given that the overall US defense budget is over $800bn. One of the problems he highlights is more financial: defense procurement takes time, and venture investors want to back companies that can show demonstrable revenue growth. These two approaches are incompatible, and while investors are getting a bit better at handling this ambiguity (one which biotech VCs have dealt with just fine for decades), there turns out to be a small loophole in procurement rules that allows the government to move faster than usual—and they've moved $70bn through this loophole since its discovery.
• Bloomberg profiles Pepijn Van der Stap, a Dutch hacker who was prolific as both a white-hat security professional and as a hoarder of stolen data who periodically extorted his victims. This is an interesting one because the personality type that's good at identifying vulnerabilities in security is very close to the personality type that's going to exploit them. It is, for very good reasons, hard to know how many people with these traits end up having a grand time using this skill full-time on behalf of their home country's government, but that seems like the most socially optimal place for them to be. (At least from the perspective of people who broadly agree with said government's aims.) Keeping very busy is sometimes the only way to stay out of trouble.
• Brad Setser has a good piece on how globalization has fared since the rise of tariffs and industrial policy since the late 2010s. It's a fun piece at the intersection of economics and accounting—specifically the way accounting rules distort economic incentives and encourage companies to realize profits in low-tax jurisdictions, whether on paper or by moving some of their business offshore. Measuring the extent of this is also an accounting exercise, and there's some nice forensic work. (It's yet another weird use case for the "risk factors" section of 10-Ks: companies will disclose which countries' tax law changes might hurt them, which tells you a bit about which low-tax jurisdictions they depend on.) The piece also talks a bit about how China's growth has shifted back towards being export-driven, but increasingly relies on local supply chains rather than just performing low value-added work with expensive components and equipment acquired elsewhere.
• Nilay Patel interviews Duolingo CEO Luis von Ahn. The single best part: Patel asks if there's a conflict between building an app for engagement and building an app for education. And when von Ahn is asked how he manages that, his answer is "Very easily. Always go with engagement." There's an obvious business case, but von Ahn's argument is that the main reason people don't learn languages is that they give up too early, so an app that's a fraction of maximum efficiency, but that keeps people logging in every day, beats one that has optimal pedagogy, but isn't very fun. At least, that's true if the app is consumer-facing and targets a large market—in any case where someone is taking the "fun" side of the fun-versus-education tradeoff, there's room for a less fun and more educational service doing the same thing at a higher price point for people who genuinely care about the outcome.
• In Capital Gains this week, we look at companies whose core competency is not their main source of profits, like car dealers who make more than 100% of their profits from financing, or grocers who make all of their money selling prime shelf space. These kinds of models turn out to have some deep commonalities relating to which products are intrinsically commoditized and what businesses do to differentiate them regardless.

Books

10½ Lessons from Experience: Perspectives on Fund Management is a delightful little book from someone who's been a participant in, and driver of, the hedge fund business's shift from cottage industry to institution. So it's a book about surviving cycles and epicycles: not only has the way to make money picking stocks changed over Marshall's career, but the way that skill gets monetized has also changed.

Some chapters are close to conventional wisdom. Did you know that systematic investing is more rigorous than discretionary stock-picking, but that systematic investors tend to mess up when the world changes and they're trained on the wrong distribution? Yes, you did, and so did I, and so do many people. And yet, turning this truism into alpha is nontrivial, and big market turns are often associated with systematic investors blowing up. (They have, mercifully, dropped the 80s through early 2000s tendency of quoting moves in standard deviations when the whole reason they're explaining themselves is that asset price returns are not normally distributed.)

The book closes out with Enoch Powell's famous line: "All political lives, unless they are cut off in midstream at a happy juncture, end in failure, because that is the nature of politics and of human affairs." Which naturally applies perfectly well to fund managers. Since returns are a function of skill and luck, any given snapshot of high-returning fund managers will be a relatively lucky part of the population, and for some of them, luck will have offset negative skill. The result is that dollar-weighted returns are worse than time-weighted returns, i.e. people get assets because of a good track record, but that track record gets worse as their asset base rises, and luck doesn't persist forever. It's good to be realistic about this. It's hard to control your own skill level, but it's possible to retain some awareness that at least some of your outcomes were luck, too.

Open Thread

• Drop in any links or comments of interest to Diff readers.
• Are there any other good memoirs by people who kept working in the same job over a lifetime even as the nature of that job completely changed? Yes, I’m eagerly awaiting Bill Gates’.

Diff Jobs

Companies in the Diff network are actively looking for talent. See a sampling of current open roles below:

• An AI startup building tools to help automate compliance for companies in highly regulated industries is looking for legal engineers with financial regulatory experience (SEC, FINRA marketing review, Reg Z, UDAAP. JD required; top law firm experience preferred. (NYC)
• A hyper-growth startup that’s turning customers’ sales and marketing data into revenue is looking for a product engineer with a track record of building, shipping, and owning customer delivery at high velocity. (NYC)
• A well funded startup founded by two SpaceX engineers that’s building the software stack for hardware companies is looking for a staff product manager with 5+ years experience building and managing data-intensive products. (LA, Hybrid)
• Ex-Ramp founder and team are hiring a high energy, junior full-stack engineer to help build the automation layer for the US healthcare payor-provider eco-system. (NYC)
• A well-funded startup that’s building the universal electronic cash system by taking stablecoins from edge cases to the mainstream is looking for a senior full-stack engineer. Experience with information dense front-ends is a strong plus. (NYC, London, Singapore)

Even if you don't see an exact match for your skills and interests right now, we're happy to talk early so we can let you know if a good opportunity comes up.

If you’re at a company that's looking for talent, we should talk! Diff Jobs works with companies across fintech, hard tech, consumer software, enterprise software, and other areas—any company where finding unusually effective people is a top priority.