Longreads + Open Thread

Longreads

• Misha Saul reviews the book Exactly, a history of precision in manufacturing in Kvetch. The book (and the review) are a love letter to both the process and philosophy behind taking extreme care when making things that are extremely small. The tech industry is entirely downstream from this process; faster software hasn't driven the same kinds of improvements as smaller transistors. (And a fun detail: there's a discussion of a venture round from 1935, which led to the process of growing jet engine turbine blades from a crystal. Every business, technology, and discipline turns out to be older than it first seems!)
• Patrick Radden Keefe has a wonderful profile of art dealer Larry Gagosian in The New Yorker. Everyone quoted in this article is doing their best to be extremely quotable, and the result does not disappoint. Gagosian is a bit like Michael O'Leary of Ryanair, in that he joined a sometimes-glamorous industry but was comparatively indifferent to the glamor, and really just wanted to run a profitable business. (And there's a callback to another fun business story: Gagosian inherited some clients through his relationship with Leo Castelli, who was a dominant art dealer in a previous era; it's a bit like Ken Griffin's Citadel being seeded by Ed Thorp, who also shared some of his old financial records.) Fans of market structure will also be delighted by the many stories of trades that are available in an opaque market where conflicts of interest, insider trading, and price manipulation are all either legal or rarely prosecuted.
• Rachel Browne in The Walrus profiles a direct mail scammer who made upwards of $20m annually selling the services of a psychic who may or may not have been aware of any of this. Like many such stories, it's grim: this kind of pitch works best on people who are both financially desperate and naive. It's also an interesting tale of compartmentalization, both on the business side (lots of shell companies and careful outsourcing) and morally (the defense was that customers should have known it was a form of entertainment).
• A guest post in Slime Mold Time Mold reviews the literature and argues that attention spans have probably been decreasing but that it's hard to tell for sure. This seems like an important issue! It's adjacent to the "burden of knowledge" problem. In the latter, the issue is that as a field gets more advanced, it takes more time to get to the cutting edge, which means fewer years of intellectual peak performance advancing the state of the art. There are some topics that are very hard to figure out without sustained attention, and if we're collectively getting worse at this, we'll lose the ability to accomplish some things entirely. (On the other hand, we know more about pharmacological means of extending attention spans, at least temporarily, so perhaps there's a semi-happy ending after all.)
• This 1997 piece by David Banks asks why some cities and periods are incredibly productive, like Renaissance Florence, Periclean Athens, or Elizabethan London. Many plausible theories are considered and discarded, but a few survive: artistic and scientific golden ages are likely when there's social mobility, a recent military victory, and decent education, ideally involving tutors. (The postwar US hits several of these, but not all of them.) The cities turn out to be a Schelling Point, which exaggerates the concentration of accomplished people in a specific location because it's where all of the accomplished people want to go. A good research question here is how much of this can be deliberately engineered, whether at a large scale by a country or at a smaller one with other institutions.
• This week's Capital Gains asks how to choose a discount rate. Discount rates can arbitrarily turn a bad investment into a good one, just by lowering the required rate of return, and they're somewhat subjective, which makes them a tempting thing to tweak. Ultimately, choosing a discount rate is more about the process of building a model of an investment's underlying economics; ideally, the precise details of the valuation are irrelevant because the thesis is so obviously true.

Books

• The Millionaires' Factory: The Inside Story of How Macquarie Bank Became a Global Giant. Australian investment bank Macquarie has the impressive two-part achievement of 1) beating the market over long periods (since its IPO, shares have returned 17% annualized, compared to 4% for Australia's ASX 200 index), and 2) not blowing up (yet). They did this partly through the standard financial story of being more entrepreneurial than average in an industry that deregulated and globalized over their tenure, and more specifically by being early to infrastructure investing. Infrastructure is a unique asset class with an extremely long duration: the book references 99-year leases, for example. And these assets have a lower failure rate than companies. Over very long periods, their valuation is heavily dependent on demographics. So in a way, the story of Macquarie is analogous to that of the people who got stupendously rich managing bond funds or macro-but-mostly-bond funds; it takes skill and effort to stand out in a competitive field, but if you choose a field that has multi-decade secular tailwinds, the results are impressive.

Open Thread

• Drop in any links or comments of interest to Diff readers.
• Are golden ages like the ones discussed in the Banks link similar to "mafias" like PayPal/Tiger Management/Drexel alumni? (One possibility is that military victories fill the same role as company collapses/acquisitions—they suddenly shake loose a lot of talent that was previously focused on a different task, but only after solidifying their social bonds and giving everyone a demonstration that they can accomplish a lot if they work hard.)
• And a side note: we're going to do a piece on QSBS exemption, and would be delighted to talk to a tax professional to make sure we get the details right.

Reader Feedback

In last week's Longreads, Calvin McCarter raises a fun question:

There was an interesting conversation on Twitter today about what adoption rate says (or does not say) about the quality of a technology: https://twitter.com/charleskfisher/status/1682832279284899845. Francois Chollet remarked on the slow rollout of self-driving AIs since 2016, as evidence that they were not better than humans in 2016. Charles Fisher responded that this is not sufficient evidence, due to the inherently slow pace of tech adoption. This makes me curious whether we should expect tech adoption rates to follow a super-exponential distribution, such as the "unreliable friend distribution", or the exponential distribution with its memoryless property? For example, does each passing year since 2008 where blockchains are not widely used for non-speculative purposes provide evidence against them ever being used for non-speculative purposes? How many technologies have had slow beginnings and then eventually taken off? One might might point to the failure of IBM's Watson, followed by the success of ChatGPT, but I'd argue that these were fundamentally different technologies.

There are some cases where a technology exists for a long time before it finds a use case with positive feedback loops, and then grows fast once those are in place: using computers to design better computers is one example, using the output from crude oil to 1) fuel cars, 2) build roads, 3) increase the efficiency of oil exploration, and 4) sequester CO2 by using it to pump more oil might be another. And sometimes, adoption is a bet on experience curves: if you expect a product to get much cheaper with scale, you can underwrite what looks like overinvestment in it.

Diff Jobs

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

• A well funded seed stage startup founded by former SpaceX engineers is building software tools for hardware engineering. They're looking for a full stack engineer interested in developing highly scalable mission-critical tools for satellites, rockets, and other complex machines. (Los Angeles)
• A fintech startup that lets investors trade any theme as if there were an ETF for it is looking for a senior backend engineer. (NYC)
• A crypto proprietary trading firm is actively seeking systematic-oriented traders with crypto experience—ideally someone with experience across a variety of exchanges and tokens. (Remote)
• A vertically integrated PE-backed cannabis company is looking for a data analyst with visualization experience. Excel wizards encouraged to reach out. (Little Rock, AR—no remote, but relocation assistance is possible)
• A company that helps investors use alternative data to make better decisions is looking for early-career data scientists and business analysts. (Remote)

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.