Boom: Bubbles and the End of Stagnation

by Byrne Hobart

When investors debate the merits of an oil stock, the question might be whether a barrel of oil will sell for $70 or $50. When investors debated the merits of Uber, the question was whether the company would be a niche taxi service worth perhaps a billion dollars or a transportation revolution that would create and capture trillions in value. Because these visions of the future are so radically different, there is no middle ground.

Since there is no middle ground during an inflection bubble, prices are set by trades between the most optimistic investor and the most committed short seller.

This makes inflection bubbles much less leverage-driven. Instead, they are fueled by equity. As a result, inflection bubbles tend not to cause major financial crises. When the dot-com bubble collapsed, there were bankruptcies, layoffs, and recessions, but the financial system largely shrugged it off.

Founders who had been funded before their businesses were economically viable either found new models or lost their jobs, leaving a cohort of leaner and more profitable dot-coms.

The fundamental utility of inflection bubbles comes from their role as coordinating mechanisms. When one group makes investments predicated on a particular vision of the future, it reduces the risk for others seeking to build parts of that vision. For instance, the existence of internet service providers and search engines made e-commerce sites a better idea; e-commerce sites then encouraged more ad-dependent business models that could profit from directing consumers. Ad-dependent businesses then created more free content, which gave the ISPs a better product to sell. Each sector grew as part of a virtuous circle.

Companies and founders have an incentive not to scare people with their audacity. A firm might have a product idea that, if realized, would change the world, but the world is sometimes reluctant to be changed.

Microsoft used the internal slogan “A computer on every desk and in every home, running Microsoft software” early on (although it later abandoned the latter part of this dream—as it turns out, Microsoft and the Justice Department had a disagreement over whether Microsoft software should be running on every single computer).

What goes unstated is that Google also organizes information about who is interested in which products, information it uses to serve up targeted ads. If the goal were expressed as “Predict every thought that might eventually lead to a purchase decision,” Google would sound both greedy and dystopian.

The company’s engineers really do try to organize the world’s information, and Google Books and Google Scholar are not products that a company only focused on its bottom line would pursue. Still, businesses with lofty goals have bills to pay, and profits are a necessary element of their corporate mission. Public-relations decisions made in the service of profits may well have the effect of downplaying just how transformative a firm’s ideas truly are.

Some are in it because of their commitment to a vision, others because they want the opportunity to do unfettered research. Some want to climb to the top of a hierarchy and view the bubble as an opportunity to skip a longer slog, while others just want a paycheck.

An important differentiator is that participants in good bubbles tend to talk about the ways in which everyone’s behavior will be transformed in the future. In contrast, participants in bad bubbles often talk about how a new version of an existing product will be more widely available and slightly better than the previous one.

The organizations participating in an inflection bubble have another distinguishing characteristic: There’s some project for which working at that organization is the only plausible way to make it happen. This holds true for a company like SpaceX; there just aren’t many groups trying to get humans to Mars. But it doesn’t hold for a private equity firm. If one firm doesn’t buy that particular undervalued and slightly mismanaged company, another firm probably will, or the company’s better-run competitors will eventually beat it.

The inverse of FOMO is betting against bubbles. This approach has a mixed record. There are stories of people who correctly called the peak, but there are also plenty of people who placed their bets too early to make a profit. For example, one savvy fund manager put together a memo outlining the subprime bubble with the clever and memorable subtitle “A Home without Equity Is Just a Rental with Debt.” The fund manager was eventually proven right, but he wrote his memo in 2001, in the earliest stages of the bubble. Betting against the housing market at that point would merely bring years of pain.

Since good bubbles function by aligning people and capital to work on the right problem at the same time, FOMO helps ensure everyone gets on board at once.

If a bubble only excites people who are early in their careers, it won’t be able to marshal the institutional and financial resources necessary to reach fruition. If a bubble only grabs the attention of people who are late in their careers, it will end up constrained by existing paradigms. If a bubble excites speculators but not entrepreneurs, it will bid up assets without building anything. If a bubble only convinces founders to act, it will be starved for capital. All of these people need to participate in the bubble at the same time, and FOMO can bring them together.

Why does the PC industry today have open hardware standards and relatively closed software standards? In large part because the arrangement was good for Microsoft, which boosted it in the 1980s.

On the flip side, Mark Zuckerberg has lamented that Facebook was just late enough to mobile that it didn’t get to make decisions about how the platform would work, a privilege that went to Apple and Android instead. Part of the significance of Facebook’s rebranding as Meta Platforms is that the company is committed to ruling the next platform it runs on.

In the 1990s, some of the Atari Democrats persuaded their colleagues to give internet companies a safe harbor in the form of Section 230, which protected them from any liabilities related to content users posted on their sites. The provision made the internet something closer to an open forum than a traditional media operation that happened to be accessible via modems. Intentionally or not, this openness is now a defining feature of the web.

On the narrative side, calls to action like Amazon’s “Get big fast,” Facebook’s “Move fast and break things,” and Google’s “Don’t be evil” set a storyline for others to follow. These types of slogans can be an effective way to coordinate behavior because they’re a consistent reminder of a company’s priorities—including moving fast before opportunities slip by.

Slogans don’t have to be about values; they can also be about behaviors, like getting more people in the organization to think about problems in the same way and creating a healthy filter bubble in which everyone can cooperate.

Since bubbles are temporary phenomena with powerful long-term effects, you should fear missing out. Financial FOMO at its worst is the desire to make easy money at the expense of less well-informed people. But at its best, FOMO is the nagging suspicion that someone is building the future, and it could be you.

By 1998, Yahoo was the beneficiary of a de facto Ponzi scheme. Investors were excited about the internet. One reason they were excited was Yahoo’s revenue growth. So they invested in new internet startups. The startups then used the money to buy ads on Yahoo to get traffic. Which caused yet more revenue growth for Yahoo, and further convinced investors the internet was worth investing in. When I realized this one day, sitting in my cubicle, I jumped up like Archimedes in his bathtub, except instead of “Eureka!” I was shouting “Sell!”

But not every instance of investor validation results in disaster. This is often the case when the validation is the product of parallel innovation in two industries. Some of the most productive inflection bubbles seem to function as a pair of complementary bubbles, with each justifying the other’s existence.

For either thesis to be right, both had to be. Cars could only become a ubiquitous means of transportation if gas was readily available throughout the country. Otherwise, a car was just a fancy toy.

In general, the paired-bubble concept is a powerful one. As long as participants in each bubble face a lag between the decision to spend and the realization of results, they can leapfrog one another. In period 1, company A invests; in period 2, company B invests in response. In period 3, company A’s investment creates a broader market for company B’s product, which launches in period 4; this success rebounds to company A, encouraging it to invest in another round of spending, and the process repeats.

But building any individual part in isolation would be worthless except for demonstration purposes or to confirm theories. It took a megaproject to make nuclear weapons a reality. As a species of inflection bubble, a megaproject accomplishes a set of tasks in parallel that would never be accomplished serially.

But sometimes, at the intersection of finance and technology, when two industries producing complementary products embrace a shared delusion, the delusion becomes true.

One reason bubbles succeed is because they create an economic cluster that occurs in time rather than space. The existence of the bubble convinces people to work on specific projects right now rather than risk missing out.

And, like traditional innovation clusters, bubbles encourage deep specialization. You don’t have to build the full stack if you know that every layer is being worked on by someone else; instead, you can focus on the layer you know best. Finally, the bubble’s combination of long-term certainty (“This will happen”) and short-term radical uncertainty (“I have no idea how to make this work right now”) creates an environment that incentivizes information sharing.

A bubble participant is sacrificing some of their own autonomy, betting that the rest of the infrastructure they need will someday get built. At the same time, they’re performing the same role for someone else.

Financial bubbles emerge when perception and reality diverge. When this happens, one of two things eventually occurs: either perception moves closer to reality or reality bends in the direction of perception. In the former case, bubbles can result in spectacular crashes that annihilate value and wealth. In the latter, they serve as a necessary catalyst for massive technological acceleration, as some of the bubbles we document in this book demonstrate.

That possibility doesn’t eliminate disputes over resource allocation, but it makes them less existential—if you don’t get what you want this year, there will be more available next year.

In a no- or slow-growth environment, disputes involve not how to fairly divide the spoils of winning but who will lose out.

Under such circumstances, ambitious people will be attracted to jobs that involve zero-sum games: adversarial politics, irresponsible speculation, fraud, even revolutions and coups.

The more adverse the economic environment, the less positive-sum the thinking becomes. 151 Scarcity thinking kicks off a self-reinforcing doom loop, which results in more scarcity.

According to René Girard’s theory of mimetic desire, recently popularized among the Silicon Valley set by Peter Thiel, our wants tend to be borrowed from other people. We want not what we desire on our own but what we think other people desire.

But bubbles are also an escape from a broader mimetic trap: Instead of violently discharging mimetic tensions, bubbles channel the destructive mimetic dynamic into something productive and socially net positive. And since they represent the pursuit of something that has been visualized but not defined in its final form, direct mimicry is harder (except by joining the bubble).

A bubble also provides its participants with a narrative arc. It has a beginning, full of excitement and promise; a middle, comprising a series of difficult challenges; and an end, in which the original promise of the bubble has been partly refuted and partly fulfilled, at which point participants must find a new vision to pursue.

Intel used to send employees to an annual semiconductor conference that took place on the East Coast during the winter, in part to keep them up to speed on industry developments and in part to remind them that if they were tempted to leave for IBM or another East Coast competitor, they could say goodbye to sunny Santa Clara.

Private equity, like subprime lending, can create value—among other things, working at a private equity firm is high status, while being an executive at the type of company these firms buy is fairly low status. (One can view the private equity industry as a giant conspiracy to trick Harvard MBAs into proudly managing ball-bearing factories, janitorial supply companies, and nursing homes.)

But private equity doesn’t want to radically transform the business it acquires, and in one sense it can’t, since lenders want to lend against something real rather than something hypothetical.

The business can be turned into a better version of itself but not into something completely new. This is part of the function private equity serves: finding companies that are cheap because they could be run better, then running them better before selling them off. But i...

The scientists don’t know if it will work. They don’t even know if their goal is physically achievable. Just attempting it will require a $250 billion investment. That’s roughly the decision the US government faced in 1939 when confronted with the $2 billion price tag for the Manhattan Project.

Building it required an immense investment in untested, specialized physical infrastructure, and the request for that investment came in the middle of an economic depression and a world war.

The Manhattan Project ends up looking like a bubble: Physicists convinced one another that it was possible, then convinced politicians it was worth attempting. Substantial resources had to be marshaled in order to build the physical infrastructure necessary to construct a bomb, and this had to happen in parallel with the bomb’s design.

The scientists needed a new computer—a hybrid of machine and human operators (who were, in fact, called “computers”)—to approximate diffusion rates under conditions different from those in the theoretical models.

That risk was magnified by the nature of the Manhattan Project’s participants. The single most important secret in the United States was left in the hands of people whom the national elite were not particularly inclined to trust: academics who had flirted with the far left, assorted European refugees whose allegiances were unclear, and pure oddballs who thumbed their noses at authority, a trait that was tolerated at MIT and Berkeley but not exactly celebrated by the US Army.

Along with knowledge, talent, and money, the Manhattan Project also had a clear goal and a deadline. Early on, the government was concerned that Germany would build the bomb first.

The Manhattan Project was also bubble-like in that it featured high uncertainty, plenty of waste, many mistakes, and divergent concerns. Some participants were motivated by scientific curiosity, others by geopolitics, but the project focused all of them on the same endpoint.

The Manhattan Project itself assembled a critical mass of physicists, including three Nobel laureates and four others who would be future recipients of the prize. Gathering this team was possible thanks to a decade-long trend of physicists fleeing Europe and, to a lesser extent, the Soviet Union.