My last newsletter stated that there are three ways to start a city: be a government, start a religion, or have an economic reason. I would like to add a fourth: be a large multi-national company. I would not be surprised if over the next decade we see new city developments launched by established companies.
Multi-national companies are comparable to countries in several respects concerning new city development. First, multi-national companies have deep pockets. Microsoft, Berkshire Hathaway, Alphabet, and Apple all have over $100 billion on hand. Even companies without cash on hand typically have access to cheaper interest rates and larger credit lines than small companies. Second, multi-nationals have the capacity to overcome the first mover challenge in new city development. Multi-nationals can build a factory, creating the first few thousand jobs necessary to create the beginning of agglomeration for a new city.
Of course, nation states have fewer budget constraints and more ability to relocate people, however, multi-nationals occupy a nice middle ground between nation states and younger firms in both dimensions.
There are a handful of historical examples of companies leading new city developments. In India, Tata built Jamshedpur around a steel plant. Today Jamshedpur is the only 1 million+ person city in India without a municipal corporation. While Jamshedpur can largely be considered a success, Fordlandia wasn’t. Henry Ford wanted to control the supply of rubber for his auto factories, so he attempted to build a new city in the Brazilian Amazon. The scheme quickly fell apart and is largely deserted today.
We have already seen several multi-nationals move in the direction of city development. Sidewalk Labs started by “thinking about [a city] from the Internet up” before it decided to redevelop a brownfield site in Toronto, later bowing out of the project. Y Combinator announced that they were going to build a new city in 2016, but there hasn’t been much public info since. Toyota recently joined with Panasonic to develop a city, though at 175 acres it’s closer to a town.
The high profile failure of Sidewalk Labs in Toronto has likely scared multi-nationals, but as the years progress and the memory fades, I expect some to try again.
While tech companies dominate the stock market and tend to have the most cash on hand, manufacturing companies like Tesla are better positioned to develop cities. One of the primary challenges of building a city is overcoming the coordination challenge of getting the first pool of residents to move to the city. No one wants to move to an area where a city is going the be built, they want all the city amenities to be there and ready.
Tech companies tend to have few employees compared to manufacturing companies. Tesla, for example, has four Gigafactories planned. The original Gigafactory in Nevada employs around 7,000 people. It’s easy to imagine Tesla planning a new Gigafactory on a greenfield site and planning a small town of 25,000 people around it.
The challenge is that to build a city requires a different way of thinking than most multi-nationals are used to. Tesla, for example, is a company built on engineering. They need good designs and good planning to mass produce cars. Cities, on the other hand, are more organic. You can’t build a city by planning out every single building. Instead, you need to establish the parameters and let order emerge.
It is important to distinguish between a city and the company towns of yore. Company towns were built on resource extraction. Because of economies of scale, they would often supply everything needed to live in the otherwise inhospitable region, including stores and schools. There was no broader economic activity.
Cities, by definition, have a self-sustaining economy. There need to be prices for land to allocate it to its highest valued use. There needs to be a labor market with multiple firms bidding for services. There must be an entire interdependent ecosystem of business, people, and civic organizations. It is unclear if multi-nationals would be able to understand this dynamic sufficiently well to develop a city.
The City as a Service
Byrne Hobart wrote an interesting essay on the challenges of the hardware as a service model. A SaaS company can generate revenue on its core product for an extended period of time. A hardware product, on the other hand, needs to continuously iterate to improve and expand its market.
Cities can fit within the ‘as a service’ framework. Imagine a new city development built by a single developer. The developer either leases the land or leases the building. Either way, the developer retains ownership. Instead of creating a one off revenue stream from the sale, the developer has a continuous revenue stream from rent.
The service part comes in because the developer needs to continue to make the real estate attractive for renters. There needs to be public services, roads, water, electricity, sanitation, safety, etc. This allows for the optimization of the asset, as the developer presumably will not get everything right on the first try, but can continue to iterate in hopes of better services increasing rents.
Of course, the challenge for this model is the up front capital cost required to build a city. The infrastructure could cost hundreds of millions of dollars, if not more. Depending on the availability of capital this could create a strong incentive to make early sales of land.
If this sounds farfetched, it’s basically the model of a shopping mall. The mall provides public goods like open spaces, lighting, and security, in exchange for rents from stores that want to locate in the mall. Whether the model can scale up to a city is a different question.
Kanye West is building a city on a Haitian island. His partner is Shervin Pishevar, who is Chairman of Hyperloop and was on the board of Uber. Despite few details, the project looks more promising than Akon City. In Kanye’s Joe Rogan interview he describes giving current inhabitants of the island equity in the city so they can catch the upside. A version of this worked in Shenzhen were villages were turned into corporations allowing the existing residents to benefit substantially from the subsequent increases in land values.
Starlink ‘recognizes Mars as a free planet‘. The first Terra Nullius in centuries. It will be interesting to see how it develops. My guess is a private company, likely SpaceX as the governing body, sanctioned by the reigning hegemon on earth. This is approximately how many colonies were founded in the early modern period.
Mike Ogunji, a college student, argues for charter cities in the Cincinnati Republic. The themes he puts forward are very similar to the themes of CCI. It’s wonderful to see the ideas spread.
Cul de Sac, a startup that’s developing carless neighborhoods is profiled in the New York Times. While not a charter city per se, reducing the dominance of cars will go a long way towards improving cities.
Lastly, an Oxford blog asks, ‘Is the charter cities moment here?’ They even cite a previous newsletter. So send to your friends who want the scoop months before Oxford! In all honesty, I expected it to take 2-3 years for the broader conversation to shift. While an Oxford Blog is hardly that end, it is a promising step forward!
Thank you for reading.
Founder and Executive Director, Charter Cities Institute