Skip to content

Crypto and the Conservation of Centralization

You can't decentralize the web. You can only kill old winners and pick new ones. Here's how.

Dror Poleg
7 min read
Crypto and the Conservation of Centralization

πŸš€ The January cohort of Hype-Free Crypto is open for enrollment. The November cohort sold out within a week. Click here for the full schedule and curriculum.


The web is broken. A handful of companies dominates it: Google (and Baidu) tracks all our queries, Facebook (and Tencent) monitors our social interactions, Twitter (and Weibo) decides what we're allowed to share, Amazon (and Alibaba) dominates retail, etc. Above these corporate giants, governments from Beijing to D.C. encroach on the free flow of information in the name of "social harmony" or "public health."

Crypto and blockchain-based applications aim to steer the web towards its original vision: an open network, based on public-domain protocols, controlled by no one. It promises to enable "decentralized" alternatives to the tech and government giants we all know and love.

This effort can be divided into two main fronts: Decentralized Utility and Decentralized Ownership.

Decentralized Utility aims to provide online services without relying on a centralized system. For example, instead of storing your files on server farms owned by Amazon (AWS) or Microsoft (Azure), you can store them on Arweave, Storj, Filecoin. The latter will keep your files encrypted on a network of computers governed by a protocol that cannot be stopped or altered by any individual entity.

Decentralized Ownership aims to share the ownership and governance of digital platforms with their users and stakeholders. Mirror, for example, enables writers to publish their content online, monetize it, as well as own a piece of the publishing platform itself and vote on how it is operated. Helium and Livepeer operate networks of wireless hotspots and video streaming infrastructure. These networks are maintained and secured by users who own specific tokens that compensate them for their services and enable them to participate in governance.

These crypto projects are still small and experimental. They point towards an alternative way of building, maintaining, and marketing the type of services that giant, centralized corporations currently provide.

But decentralizing one class of internet companies does not guarantee that a new class of centralized giants will not emerge in their stead. In fact, decentralizing power from one pair of hands is almost guaranteed to concentrate that power in another pair of hands.

A powerful theory and the history of the internet itself explain why. Let's start with the theory.

The Conservation of Attractive Profits

Clayton M. Christensen is best known for explaining how new companies with worse products can disrupt powerful incumbents. But another theory of his describes how power and profits shift within an industry. This theory also hints that when parts of a value chain become decentralized, other parts become more centralized.

In The Innovator's Solution, Christensen introduced "The Law of Conservation of Attractive Profits." I'll provide a simple explanation in a moment. But first, in Christensen's words:

"...the law of conservation of attractive profits states that in the value chain, there is a requisite juxtaposition of modular and interdependent architectures, and of reciprocal processes of commoditization and decommoditization, that exists in order to optimize the performance of what is not good enough. The law states that when modularity and commoditization cause attractive profits to disappear at one stage in the value chain, the opportunity to earn attractive profits with proprietary products will usually emerge at an adjacent stage."

In English, Christensen means that delivering a product (or service) relies on multiple components (or steps). Some of these components are available off-the-shelf and fit together easily. Other components need to be custom-made or combined using an additional custom component.

Consider a familiar example. Airbnb relies on an integrated digital layer and on a modular physical layer. Its website, online marketing, customer support, payment, and fraud-detection capabilities were developed or combined specifically for its own purposes. Meanwhile, the rooms and apartments it markets were developed by other people and Airbnb simply gets them "off-the-shelf." By doing, Airbnb assumes that these rooms meet some basic standards of safety and habitability.

Airbnb makes money by controlling the digital layer, even though its main business is providing access to physical spaces. This type of business would not have been possible 100 years ago. But it is possible now because (1) cities have codes that ensure all buildings meet a pretty high standard of safety and livability; (2) the online flow of information makes it possible for Airbnb and its customers to vet and trust properties even if they've never inspected them in person.

The emergence of building standards and the evolution of technology enables Airbnb to get its physical inventory "off the shelf." This stands in contrast to traditional hotels that had to build and control their own inventory to know that it is suitable for human habitation.

The emergence of building standards and the evolution of technology made it possible to run a de facto hotel without owning, building, or operating an actual hotel. It also shifted the way profits are made in the hotel industry. Building and operating a hotel are no longer the key activities that make a hotel company profitable. Instead, a hotel company must have an integrated online brand and marketing machine (and yes, many traditional hotel companies still make money, but to do so, they had to develop their own integrated digital channels).

To return to Christensen's terminology: standards and technology commoditized the physical components in the hotel value chain. As a result, profits shifted to those companies that integrated and controlled the digital components in the value chain.

The commoditization of the actual hotel did not eliminate all profits in the industry, they just shifted these profits to other activities. Hence, "The Law of Conservation of Attractive Profits." When some parts of the value chain become commoditized, other parts necessarily become more valuable. Β This brings us back to crypto.

The theory was developed in relation to the value chains of specific products, but it can loosely explain the evolution of whole industries and ecosystems. Let's look at the web itself.

Another Brick in the Wall

Christensen was a business school professor so his theory focuses on profits. But as Nathan Baschez pointed out, the theory is ultimately about power. To show what this means in practice, consider the evolution of the commercial internet.

In the pre-historic era, there was very little online content; accessing it was difficult, and participating in "online commerce" was an act of bravery. AOL and others created "walled gardens" that enabled people to access a curated version of the internet. Everything on AOL was curated and paid for. The sports content came from CBS, the news from ABC, and sending someone a bouquet had to go through 1-800-flowers. Social interactions were limited to other AOL users.

AOL charged an entrance fee to its "garden," and users paid because it was too difficult to figure out all the different steps on their own. It had the power to determine what people saw, whose content they consumed, and what goods they bought. In 2000, this power translated into billions in profits and a valuation of $224 billion.

But AOL's market power did not last long. More and more content was freely available online. New standards emerged to make it easy for people to browse websites, chat, and email each other. Many of the things AOL charged for were now available outside the walls of the garden. These activities became commoditized, and content became a commodity.

AOL's power eroded. In a growing web of millions of websites, the most valuable activity was no longer to tell people what to consume. Instead, it was to help people find what they were looking for. And so, Google emerged as the internet's primary gatekeeper β€” and moneymaker.

But Google's (relative) position did not last too long either. The internet has become more personal and more social. Individual people (as opposed to companies and specialized "webmasters") started to generate more content and wanted to interact with each other rather than simply "browse" static content. Power shifted towards Facebook and other platforms that facilitated such interaction and organized user-generated content in addictive algorithmic feeds.

There are other pieces to this story, but you get the idea. This brings us to our current era, where content creators are at the mercy of giant platforms, and a handful of trillion-dollar companies dominate the web.

The Conservation of Centralization

When one part of the web gets decentralized, another part necessarily gets more centralized. The trillion-dollar question is which companies will be the giants of the coming web3 era.

I'll have to revisit this topic to consider this question in more detail. But an initial way to approach is to divide it into two separate questions:

  1. What does blockchain commoditize? What does it make abundant?
  2. Who is in a position to become a gatekeeper in a world of such abundance?

The answer to the first question is that blockchain commoditizes investments by making everything ownable and investable. It creates an abundance of things that can be owned. It also creates an abundance of various resources by making them easier to share. Finally, it creates an abundance of human engagement – the ability to incentivize people to act in order to support different causes and projects.

Based on the above, the answer(s) to the second question is that the gatekeepers of web3 will be organizations that know how to allocate capital in a world of endless investment opportunities (A16Z is already working on that). They will also be organizations that can aggregate demand for various shared resources β€” just like Airbnb, but for many more things (some crypto projects are working on that, but the winner will likely be a consumer brand that makes these new technologies accessible to the masses). And finally, the giant gatekeeper of web3 will be organizations (and groups of individuals) that can create a compelling narrative and spark the imagination of large armies of fans. In a world where people can get paid to do anything, the biggest winners are those that make people part of something.

There is much more to unpack here, but I have to stop and ship this newsletter. I am still suffering from "newborn+toddler brain," but it was important for me to share these ideas while they are still raw.

πŸ™ If you enjoyed this piece, subscribe to my newsletter. If you're interested in crypto's impact on your business and career, join my upcoming Hype-Free Crypto course if you want to explore together.