IP as Ecosystem Lock
April 1, 2024
When Apple takes 30% of every App Store transaction, it isn't doing something unusual in business terms. It controls the platform that connects developers to iPhone users, so it sets the terms. Developers can accept them or build for a different platform. This is normal platform economics.
A patent works the same way. It gives the holder exclusive rights to a technology for a fixed period — typically 20 years — and everyone who wants to use that technology has to negotiate access on the holder's terms. The patent holder controls who can build on it, at what cost, and under what conditions.
That's ecosystem lock. The mechanism is legally enforced rather than technically enforced, but the structure is the same.
The Platform Parallel
AWS built a set of proprietary services — databases, machine learning infrastructure, networking tools — that are tightly integrated with each other and with the rest of AWS. Once a company builds on top of them, migrating away isn't just technically painful, it's expensive in time and engineering cost. The value of the platform grows with adoption, and adoption makes leaving harder.
Salesforce does something similar. The more a sales team lives inside Salesforce — their data, their workflows, their integrations with other tools — the more switching costs accumulate. The platform becomes infrastructure, and infrastructure is hard to replace.
In both cases, the lock isn't accidental. It's designed. The platform holder benefits from deep integration and makes it rational for builders to go deeper.
A patent creates the same dynamic without any product design required. The holder doesn't have to build a sticky platform. The law prevents anyone else from using the technology at all without a license. The barrier isn't migration cost or data gravity. It's legal exclusion.
What This Looks Like in Pharma
The COVID vaccine situation is a clear case study in what ecosystem lock looks like when the platform is medicine.
At the start of the pandemic, Moderna and Pfizer held the patents on the most effective mRNA vaccine technology. Qualified manufacturers existed in other parts of the world and were ready to produce doses. Demand was unmet. Supply was limited. The WHO asked both companies to contribute their vaccine IP to a shared technology pool so more manufacturers could produce doses.
Both companies declined.
The argument for holding the patents is recognizable from any platform context: we invested in building this, we should control who uses it and on what terms. The counterargument is that a platform monopoly over life-saving medicine has consequences that a platform monopoly over software does not.
South Africa eventually developed an alternative method for manufacturing the mRNA vaccine without the protected IP. It took two years.
Forking the Platform
In software, when a platform is too restrictive, builders sometimes fork it. An open-source project can be copied and modified. A standard can be replicated. The ecosystem isn't owned by anyone with legal force behind that ownership.
South Africa's path was a version of forking the platform. Developing new IP that achieves the same outcome without infringing on the existing patents. That's technically possible but enormously costly. Two years, large research investment, and the lag happened during a pandemic.
For developing countries more broadly, this is a recurring constraint. A company holds a patent on a drug, agricultural technology, or industrial process. The price point reflects what wealthy markets will pay. Local manufacturers can't produce a cheaper version without a license. A startup can't build an improved alternative if the core technology is blocked.
The platform analogy holds: if the App Store had no competitors and no regulatory pressure, and developers had no ability to distribute software outside of it, the terms Apple could impose would be different from what they are today. Patent holders often operate in exactly that position.
Why the Stakes Are Different
Platform lock in tech creates moats and generates returns for holders. It's a legitimate business strategy, debated mainly in antitrust contexts.
The same mechanism applied to essential medicines or food security technology doesn't stay in the domain of business strategy. Access to treatment becomes a function of licensing negotiations. A country's capacity to respond to a health crisis depends partly on whether patent holders choose to cooperate.
That's worth understanding as a business mechanism, separate from whether it's good policy. IP protection is a legally created platform monopoly. Like any platform monopoly, the holder extracts value from people who need access. Unlike most platforms, the thing being accessed is sometimes the difference between surviving a pandemic and not.
The geopolitics around this aren't simple. The U.S. tends to prefer bilateral negotiations with individual countries over multilateral forums like the UN on IP issues, partly because one-on-one negotiations preserve more leverage. That preference is itself a platform strategy: control the terms of engagement.
The same instinct that makes a tech company want to own the platform, control the API, and set the terms of access is the instinct that shapes how patent holders behave in health crises. Understanding it as a platform dynamic rather than just a legal or policy question makes the behavior more legible.