💡 5 Quick Questions for ... physicist César Hidalgo on economic complexity
"An economy like that of San Francisco should be able to support a city of the size of Shenzhen, with 23 million people. But the Bay Area today is barely able to house seven to eight million people."
I’ve written before about how economies can be thought of as network-based supercomputers that physically rearrange atoms — what some academics mean by the term “information” — to generate value. As I wrote last May:
Imagine an $18.7 million Bugatti La Voiture Noire — perhaps the most expensive car in the world — crashing into a wall at something approaching its top speed of 255 mph. Although I’m not much of a car guy, I would think this marvel of automobile engineering would be little more than 4,400 pounds of carbon-fiber scrap. Yet moments before the spectacular impact, it would have been worth roughly $4,250 a pound. The moment after, pretty much nothing per pound. So what changed? Why is the pre-crash Bugatti worth infinitely more than the post-crash Bugatti? They both contain the same number of atoms, after all. The amount of matter in one is the same as the other. The only real difference between the two vehicles — and this is the important thing — is how that matter is arranged.
When a car crashes or sand is turned into a microchip, there’s been a change in information. In the first case, the change has led to less complexity, in the second case more. And value is embedded in complexity. All the value in the pre-crash Bugatti was stored in the complex arrangement of its atoms, not in the atoms themselves. Research by physicist (and friend of Faster, Please!) César Hidalgo and Harvard University economist Ricardo Hausmann shows a relationship between an economy’s complexity and its GDP growth. But the US has been slipping in the national economic complexity rankings, from number five in 2000 to number nine today.
As we look at America’s economic prospects, it’s worth keeping in mind that a more connected economy is a key ingredient for a prosperous economy over the long run. Here are 5 Quick Questions (plus three more because I love my subscribers) that I recently asked Hidalgo (from whom I nicked and riffed the above auto example).
César Hidalgo is the director of the Center for Collective Learning at the Artificial and Natural Intelligence Institute of the University of Toulouse. Hidalgo known for his contributions to economic complexity, data visualization, and applied artificial intelligence. He is the author of The Atlas of Economic Complexity, Why Information Grows, and How Humans Judge Machines.
1/ You define economic complexity as "the use of network science and machine learning techniques to explain, predict, and advice changes in economic structures." Can you give an example of what economic complexity looks like in practice?
Today, economic complexity is used to guide industrial policy around the world. This is because it involves methods that are good at predicting the activities that an economy will enter in the future and at explaining international variations in economic growth, income inequality, and emissions. For instance, you can predict the probability that a city starts patenting in a new technology or that a country starts exporting a new product. That is the idea of "relatedness." These methods hit a trifecta of outcomes (inclusive, green, growth) while also providing a means to navigate changes in economic structure. That's why they are increasingly used in dashboards, for both understanding where an economy is and thinking strategically about how to develop it. For example, the secretary of the economy of Mexico recently released a new economic complexity dashboard in Data Mexico. Economic complexity methods are also used today in reports informing smart specialization strategies in Europe. In practice, it is a modern way to understand where an economic structure is, where it can go, and what are the implications of these changes.
Keep reading with a 7-day free trial
Subscribe to Faster, Please! to keep reading this post and get 7 days of free access to the full post archives.