Surfing the Right S-Curve

Plus! Google: Benefits and Retention; Tail Risk on Trial; Policy Homogeneity; Procurement; Density Redistribution; Crisis Hormesis; The Other Derivatives Narrative

Welcome back to The Diff. Here are the subscribers-only posts you missed this week:

Support The Diff and subscribe now!

This week’s Q&A will be today at 2pm Eastern/11am Pacific.  This week, the call will be open to all readers, not just paying  subscribers. Topics include shareholder votes, TikTok and Tech  Nationalism, Covid, S-Curves, and more.

This is the once-a-week free edition of The Diff, the newsletter  about inflections in finance and technology. The free edition goes out  to  15,163 subscribers, up 298 week-over-week.

In this issue:

Surfing the Right S-Curve

Take anything that’s subject to selection pressure, and track how it  grows in a new environment over time. You’ll get a graph like this:

It starts out small, adapts to the environment, grows a lot, and  peaks when it’s perfectly optimized for its current environment.

Usually, if you wait long enough, you get a graph like this:

That’s a graph of something that’s gotten good at exploiting all of  the inputs it needs, and either a) it accidentally depletes one of those  inputs, b) the mix of inputs changes, or c) the phenomenon in question  turns out to be a good input for some other phenomenon adapted to  defeating it. i.e. It becomes part of the “environment” for something else that quickly turns it into a source of growth.

That curve is deliberately vague, because it can describe so many things:

The label of the Y-axis varies: biomass, Google Ngram share, salary,  population, votes, unit sales, market value. The threats vary, too: a  novel pathogen, a change in fashion, skill obsolescence, internal  frictions, newer technology, or stagnation.

Some of these phenomena are fundamentally mindless: a colony of  bacteria doesn’t wonder what it will do when it reaches the edge of the  petri dish and starts running low on agar. Members of a political  movement think, but the movement itself is dumber; it can easily peter  out when the people who are incentivized to join are low-status and the  ones tempted to leave are high-status (or if it takes power and its policies are disastrous).[1] Any given technology doesn’t  have a will of its own, although it sometimes looks that way in the  short term. You can anthropomorphize them the same way you can say a  given gene “wants” to spread as far as possible; it’s a thing without a  will, but selection effects mean it acts the way a willful thing would  act, in the short term.

But for others, there’s intelligence and planning at work. The curve  is not a constant. Some people look at where their career is going and  see that they’re approaching a dead end. And companies, especially,  thrive when they identify the S-curve they’re on and choose which one  they’d like to be on instead. Many people reach a point in their work  where the S-curve inflects: every project they finish gets them skills  and referrals that they can apply to doing an even better project after.  That’s an enjoyable situation, at least while it lasts, but empirically  it doesn’t. Wage growth is fastest at the start of a career, and slows  monotonically throughout—one study has real wages rising 3.3% annually from ages 25 to 34, but actually dropping slightly from the mid-40s onward.[2]

Greasing skids, Pulling up Ladders

One important element of growth cycles is that they can lower or  raise the barrier to copies. A single success in an industry makes  competitors more fundable, and offers a playbook for new entrants.  Sometimes, all a novel company achieves is a proof-of-concept:  that there’s a buyer for a particular kind of product, even if the  product itself is not that great. So follow-on entrants can improve the  original version, pay closer attention to how high the S-curve goes, and  achieve much better returns.

The Blackberry followed this arc: it basically created a market for  an always-connected device built for people who always needed access to  email. Which meant it created an expectation that any professional was  available for a quick one-line email on an hour’s notice during their  waking hours. And that made a market for a better-designed, more flexible device like the iPhone.

Facebook and Google made it possible for other companies to acquire a  large audience, but charged a high price, and had durable scale and  network effects that made a direct challenge difficult. Even while they  made direct competitors tough—it’s hard to build a general search engine  acquiring customers via AdWords, and doable but expensive to build a social network audience by advertising on social networks—they made building a search- or social-adjacent business easier.  The combination of greasing the skids and pulling up the ladder means  that big advertising platforms make it compelling to build complementary  products: if there’s a set of Google searches that are poorly-served by  the Internet’s existing content, Google is a viable channel for  marketing a site that serves that demand; Facebook makes it so that any  business that could benefit from hyper-targeted ads will get most of its  benefit from Facebook. Amazon is designed so third-party sellers can make good money offering anything that Amazon shoppers look for but can’t find on Amazon.

Big platforms build potential S-curves for other companies to ride.  Google was an affordable sales channel, for paid and organic search.  This has created numerous businesses, and enhanced others, but some  companies either structured themselves around eternally cheap ads or  built themselves around parts of the Google algorithm and layout that  weren’t permanent. If Google’s traffic is cheap, and a business can  resell it for more, that’s a profit opportunity—for them, but also for  Google, which can cut out the middleman.

And in some cases, an incumbent manages to restructure a business,  capture the highest-margin piece, and commoditize the company that  helped make their business possible. Since failure has so many causes,  it can’t be directly attributed to competition or commoditization, but  they’re frequent culprits:

If every company is trying to optimize its resource consumption to  maximize output, it’s going to grow until some constraint slows it down.  The bigger the company, though, the more it reshapes the entire  environment; Google made some businesses possible, and since those  businesses could grow their audience at Google scale, it made them grow  far faster. Platforms are naturally allergic to high-margin hypergrowth  by their customers—if it’s scalable and profitable, it’s probably an  arbitrage, and owning a platform means having an eventual monopoly on  all arbitrages that platform enables. In IBM’s case, they helped make  Microsoft ubiquitous, and Microsoft’s ubiquity helped materially erode  IBM’s importance.

Bottom of the Curve

If you’re aiming for maximum success, why not pick the steepest S-curve you can find?

Why not, indeed?

If Facebook had started by trying to dominate the world, rather than  Harvard, they would have had a much slower growth path. Any random  person who joined the site wouldn’t have a single friend—so they  wouldn’t stick around long enough to receive their first friend request.  By starting with an audience of a few thousand people, Facebook was  able to reach network density fast. They expanded gradually, and at  times with trepidation, to other Ivy schools, then to other colleges,  then to high schools, then to everyone. They jumped from one S-curve to  another:

This is a point Reid Hoffman makes in Blitzscaling: network effects are nice when a company grows, but they’re terrible  when it’s subscale. When nobody used Airbnb, nobody had any reason to  use Airbnb, and the founders had to hustle and growth-hack their way to  viability, one city at a time. The Airbnb network effect had brief  flashes of unsustainable usage, during a design conference and the 2008 Democratic National Convention, but the actual network effect took a  long time to work. A network effects-driven growth strategy grows on its  own once it gets rolling, but the beginning is more prosaic:

Marketplaces are so hard to get rolling that you should  expect to take heroic measures at first. In Airbnb’s case, these  consisted of going door to door in New York, recruiting new users and  helping existing ones improve their listings.

One reading of the Airbnb story is that they converged on the right  S-curve. Their first S-curve, renting three airbeds to attendees of a  design conference, was too shallow. Market saturation was a few  transactions away. The next S-curve, a global product, was too steep.  But the New York S-curve was just the right slope to be achievable and  worth trying. (And “New York” might be too broad: going door-to-door  means picking out neighborhoods, since NYC’s 3.2m households represent  too many doors for even a very energetic founder to knock on.) But  success in New York let them shift to another S-curve, dominating  quirky-but-affordable travel. And from there, they could advance to  their current steeper but more rewarding S-curve.

The most successful companies make a habit of 1) surfing a rewarding  S-curve, and 2) planning ahead so they can shift to the next one.

It’s the signature of every great company that, if it occurs to you  to compete with them, you realize you’re a few years behind. Earlier  this year, I looked into the secondary market for shares in private  companies, and after lots of research and due diligence, realized that  Carta had quietly gotten a five-year head start. The S-curve shift is a  particularly valuable strategy, because it combines planning with  misdirection. A company’s initial S-curve makes them look like a trivial  product aiming for a small market—the kind of company that can easily be  dismissed as unambitious. And the shift to a new curve diverts resources  from other projects, so superficially it looks like an unambitious  company that’s also slowing down.

One indication that a company is at the top of the curve is that its  decisions are less existential but much harder. When a business finds a  working growth strategy, scaling that growth is nontrivial, but  ultimately involves optimizing for just one metric. When growth stalls,  it involves endless complex tradeoffs. Facebook, for example, needs to  grow users and to add engineers. Its users (on the core app, in the US)  are right-of-center, and its engineering staff is left-of-center, so  every decision with political implications is a choice that trades off  between two critical metrics.

No one has composed a good corporate Siddhartha  guiding CEOs through the process of the growth-to-maturity transition.  And the biggest US tech companies seem to have an instinct for  continuously scouting out newer and better S-curves, or at least  diversifying from one huge curve into dozens of smaller ones that add up  to a larger overall market. The usual way the transition works is that  the founding CEO retires, or otherwise leaves the company. Switching  from a startup to a growth company is rewarding, both financially and  morally, but going from growth to stability is wrenching, even if the  money’s still good.

It’s a necessary process, though. Since the industrial revolution, US growth has been eerily stable  over long periods, but the nature of that growth has completely  changed. Economic growth is one S-curve stacked on top of another, ad  infinitum, and as the economy gets bigger and more complex, the law of  large numbers gets stronger. Inside that fitted-exponential-hugging  trendline is an endless series of stories of daring, ambition, hubris,  and stagnation, all occurring at different timescales. For the most  successful people and companies, success is a matter of identifying when  the plot will get depressing, and switching stories before it does.

[1] Like a social network, a political movement has to start out  obscure-but-very-cool; ideally it’s invite-only, and there’s no good way  to request an invite. To grow, it has to manage the glide path from  elite/exclusive to boring/ubiquitous.

[2] Older people often see their peak net worth growth a bit later,  but that’s because of compounding capital, not pure wages. And while  there are some investors who put up great numbers late in their careers,  the usual rule of thumb is that late-life record net worth is achieved  through a slowly declining growth rate on top of a continuously  compounding base.


Google: Benefits and Retention

Google has a new program to help employees pay student loans.  This is a trivial cost for them on the surface, but illustrates one of  the levers large companies pull to reduce employee defections: the more  their benefits are comprehensive, and simplify employees' lives, the  more logistically challenging it is to quit. Google looks great on a  résumé, which is another way of saying that any college graduate vetted  by Google is immediately a target for recruiters.

It’s basically the HR equivalent of Zoom’s strategy of leasing  teleconferencing hardware to customers: they want to make leaving  complicated and inconvenient enough that it’s not worth the effort.

Tail Risk on Trial

There are three broad schools of thought around how investors can  avoid being caught in a market crash: they can time the market (which is  hard, since, by definition, the average investor can’t do it); they can  diversify into lower-risk assets or anticorrelated assets (which, with  most such assets yielding close to zero, is challenging for other  reasons), or they can use a tail-risk fund. The pitch for tail-risk  funds is that they can structure a set of bets on a market crash, such  that an investor who owns mostly equities is protected during a crisis,  and can even buy more stocks at low valuations. But it’s a tough sell.

Normally, financial products get sold to institutions based on some  kind of backtest. But for crisis insurance, the problem is that every crisis changes the way insurance gets priced.  After 2008, many asset managers wanted to buy tail-risk hedges, and as a  consequence, the price of tail risk insurance was inflated to ludicrous  levels. The crash of 1987 made some options traders rich, but  permanently raised the price of bets on extreme price moves; the trade  worked, one time. In a sense, a tail-risk fund’s backtest consists of  its founders' knowledge of market structure and derivatives, and their  personality type; the risk/reward of the trade itself always has a  sample size of 0 when the tail-risk plan is put in place, and 1 once its history is irrelevant.

A Theory of Policy Homogeneity

Robin Hanson asks why regulatory choices are correlated across nations.  This is an important question, because it lends itself to very positive  or negative answers. The positive answer is that we’re at the end of  history: every country that’s sophisticated enough to create nuclear  bombs, human clones, or prediction markets is also smart enough to  regulate them in similar ways. Alternatively, elites could be  conformist, and peer pressure emanating from New York, Davos, Basel, and  Brussels could nudge everyone to fall in line.

In the latter model, a rise in nationalism has positive effects on  the flow of ideas that partly mitigate its negative effects on the free flow of  people and goods: if more countries adopt different ideas about what  kinds of research are acceptable and what kinds could be banned, it’s a  sort of implicit global federalism that allows the best regulatory  regimes to demonstrate their benefits. The object-level effects very  much depend on whether this results in more pharmaceutical trials or  more nuclear proliferation, though.


Anduril has been selected to help build the Advanced Battle Management System, a new software program for coordinating different parts of the Air Force and Space Force in combat.

Defense is an industry where the bull case on startups is that the incumbents are terrible (and the budgets are big—US defense spending was $686bn last year):

One unique aspect of Anduril is that while they are a  defense contractor, they actually take on all the research and  development (R&D) risk themselves, before selling to  government. This is a significant shift in the way things have “always  been done” because the U.S. government –taxpayers — take on all the  funding and R&D risk in the form of “cost-plus” contracting, where  contractors are paid a guaranteed profit, regardless of cost. With  Anduril, the U.S. government not only saves this money but diversifies  its portfolio of the best defense technologies.

The bear case on defense tech is that it’s the worst of enterprise  software: selling to a buyer who isn’t the end user, and who is more  sensitive to upfront specifications than to whether the product itself  is useful. A shift to new vendors like Anduril implies that the  sales-over-product and specs-over-outcomes approach is eroding.

Density Redistribution

One model of cities, which I’ve written about before,  is that their density and high prices compound the network effects that  cause those prices in the first place. Covid has redistributed some of  the beneficiaries of those effects to less dense places; it’s hard to  make friends via Zoom, but relatively easier to maintain them. Japan is  now trying to accelerate this,  by offering a $9,000/worker subsidy to people who relocate from Tokyo.  The ostensible goal is to encourage tech companies in rural areas, but  that’s a hard sell: without a dense cluster of similar companies, the  Internet rather than the real world will be the default place for these  workers to network. And with costs outside Tokyo already lower, the  extra money won’t turn into consumption, either. So this policy ends up  sounding like a lot of Japan’s fiscal policy since the start of their  Lost Decade: the government deficit-spends, which puts money in people’s  pockets, but those people end up saving the money instead of spending  it. The net result is a scarier deficit that doesn’t lead to inflation  or growth.

Crisis Hormesis

Bloomberg profiles  Andreas Lehnert, the Fed’s “disaster junkie.” He says the Fed has  contingency plans for all sorts of unlikely disasters, and was able to  dust one of quickly in the early days of Covid, and put it into action.  This sensitivity to risk came about in an unhappy way, though. In 2005,  Lehnert reported:

“Institutions with large amounts of mortgage credit risk  on their portfolios are well-positioned to handle severe losses,”  Lehnert said then. “Neither borrowers nor lenders appear particularly  shaky. Indeed, the evidence points in the opposite direction: Borrowers  have large equity cushions, interest-only mortgages are not an  especially sinister development, and financial institutions are quite  healthy.”

Some institutions respond to crises by getting good at  crisis-management, others in a more destructive way. It’s unclear  exactly what the dividing line is; since the Fed did well in this  crisis, you can tell a story about how long employee tenure and a  tolerance for understandable mistakes helps them build knowledge over  time. On the other hand, if the economy were doing worse right now, a  profile of the same economist might focus on the fact that he dismissed  the risk of a mortgage crisis and didn’t get fired.

Careers often follow a Joseph Campbell-style Hero’s Journey,  and the most effective people seem to be the ones who get through their  career nadir early enough that the damage they’re responsible for is  limited but the lessons are memorable.

The Other Derivatives Narrative

There are two classic genres of derivatives disaster stories: in one,  a sophisticated institutional investor makes a large bet that, for very  various academically interesting reasons, blows up on them in  spectacular fashion. The other is that a ruthless financial institution  sells trades to investors who don’t quite understand them, whether those  investors are individuals or smaller, less sophisticated institutions.  SocGen is going through a third, less well-known story ($, FT):  they built an equity derivatives empire, selling risk permutations to  everyday investors. Because they were slicing up risks and selling only  the most attractive bits, they ended up warehousing some odds and ends,  and those odds and ends did poorly.

(Other sources give more details: due to the quirks of retail demand  and hedging ability, SocGen and other European banks ended up owning  large positions in equity dividend futures. These generally track the  broader market, but a short/sharp recession can blow up the trade in  hard-to-hedge ways.)

One comment on the FT article makes a useful point about potentially toxic derivatives products:

Back-testing optimisation: products were sold on the  basis of their backtest (almost) exclusively. Therefore, the basket of  stocks/indices was carefully selected and optimised to look good. Stocks  were swapped around to find the best-looking possible combination (e.g.  suitable stock trends), which essentially also resulted in the biggest  gap between historical parameters (e.g. realised volatility and  correlation) and implied ones (implied volatility, correlation, dividend  yield). Basically, historical parameters are what is shown in the  backtest, implied parameters are the cost of hedging.

In other words, part of the structured products model was to toss  many coins many times in a row, and then sell Lucky Nickels at a  premium.