Economic complexity theory uses math that is considered a bit “exotic” in economics, but is core in physics & computer science:

Eigenvectors.

So, let me explain how we can use this math to estimate the presence of an arbitrary set of economic factors. The economic complexity index or ECI is an estimate of the factors present in an economy built on a simple recursive observation.

An economy is as complex as the activities it can do, and an activity is as complex as the places that can do it.

Now that everyone is talking about the impact of AI on labor, here is a quick thread with some key findings from the broader literature.

TLDR; there is some evidence of both, AI enhancing and reducing gaps among low- and high-performers.

Pull the thread to learn more 🧵👇 A recent classic is the paper by Noy and Zhang in Science from 2023.

These scholars asked 100s of college educated professionals to complete writing tasks.

First, everyone took a pre-test, to asses their baseline. Then, they took a second test (randomized with or without AI).

How will the recently announced 25% tariffs on Mexican and Canadian products affect American businesses?

Mexico & Canada are key trading partners for the US. They represent 28% of US imports & 33% of US exports.

Here is a detailed thread on US-Mexico-Canada trade 🧵👇 /1
Mexico is by far the number one trade partner of the US.

It is well integrated into American value chains, especially in electronics and automotive.

This year the US imported from Mexico:

🚙 $31B Cars
💻 $28B Computers
🚚 $24B Delivery Trucks
/2

New Paper 📄 & thread 🧵!

How big are the digital exports of the U.S. compared to Europe?

Today, trade flows not only through container ports, but through routers. In this new @NatureComms paper, we introduce a method to estimate trade in digital products by combining machine learning methods with corporate revenue data.

There is a lot to unpack, so let’s dive in! /1
nature.com/articles/s4146… Digital product exports, such as purchasing a video streaming subscription from a foreign website, are notoriously hard to estimate because tech firms own foreign subsidiaries. Moreover, existing service trade statistics lack the fine granularity needed to track digital products. /2

**New Paper**
Economic complexity methods are popular tools in industrial policy. Yet, despite their widespread adoption, these methods are sometimes misunderstood. In this new paper, I explain & explore the policy implications of economic complexity. /1
https://t.co/BQ6xLOA8cMbuff.ly/3YjBFbh
First, why are economic complexity methods misunderstood?
A key part of the confusion comes from the predictive nature of these methods. The concept of relatedness, for instance, anticipates the probability that a country or region will succeed at an activity./2

What is intelligence?
And how is it different from problem solving?

These questions are central in our current discussion on AI & were debated passionately this week at the Santa Fe Institute’s conference on collective intelligence.

But what did we learn?

🧵 .. 1/N

First, a disclaimer. In this thread I will focus on one idea, not all the ideas discussed in the conference, and will obviate other aspects of intelligence (eg multidimensional intelligence), not because these are not important, but because I want to communicate one point.

AI hype is on full swing, to a large extent, because of language models.

But as a writer, I am not totally convinced about the “productivity boosts.”

You see, writing fulfills a dual purpose. On the one hand, we write to communicate. But on the other hand…. /1 we write to clarify our own ideas.

We write to learn in ways that cannot be accomplished by reading.

A big part of what motivates a writer to work on a book is knowing that at the end of the journey I’ll be a different person.

Can we improve economic complexity metrics by combining trade, patents, & research data?

In our latest paper with @ViktorStojkoski & @philippmkoch, we show that this is the case, but also, we explore why.

Learn about this in this thread 🧵👇/1

nature.com/articles/s4324… One frequent criticism of economic complexity metrics is that some countries, such as Mexico or Slovakia, rank too high while others, like Australia and New Zealand, rank too low. But is this a problem with trade data or with the method?

¿Cuanto litio exporta Chile? ¿Qué tan "lejos" esta Chile de las baterías? Acá un hilo con varios datos sobre el litio y sus derivados.

1. Chile es el principal exportador mundial de carbonato de litio. En el 2022, exportó USD 7600M de este producto
oec.world/en/profile/cou… 2. Este gran volumen exportador es un fenómeno reciente. Hasta finales del 2021 Chile exportaba ~USD 100M de carbonato de litio al mes. En Mayo del 2022 las exportaciones llegaron a USD 1400 millones mensuales! Hoy están alrededor de los USD 600M al mes.

What can LLMs teach us about economics?
Like everyone else, economists have been enjoying the foibles & virtues of large language models (LLMs). But can these models teach us something about the economy that we don’t already know?
weforum.org/agenda/2023/03… /1 I believe there is much that economists can learn, not by chatting with LLMs, but by deconstructing how they work. After all, LLMs are built on mathematical concepts that are powerful enough to simulate language. Maybe, these models work can become a new source of inspiration.

Tesla just announced they are building a new plant in Monterrey.

Could we have predicted this using AI?

The data tells us why Monterrey and Nuevo Leon are an excellent fit for Tesla.

Thread 🧵 /1
@elonmusk To do this, we will use datamexico.org, an official data distribution platform from the secretary of the economy, and tools from economic complexity, an academic field using machine learning to understand the evolution of economies.

What just happened in Chile? 🇨🇱

You may have read that on Sunday Chile rejected a new constitution on a referendum.

But how did we get here? And what happened in the days leading to the election?

A quick thread for an international audience. /1 🧵👇 First, some context.

In October 2019 massive mobilizations erupted in Chile. In response to spiraling chaos, politicians from various parties signed an agreement starting a new constitutional process
(Nov 15, 2019)./2

Why should we care about economic complexity? And why do complexity measures capture information that escapes measures of concentration such as entropy, Herfindahl-Hirschmann, or Gini?

A nerdy thread: 👇🧵/1 Measures of concentration (e.g. entropy, HH) are blind to composition. They cannot tell apart a country that exports 80% cars and 20% bananas from one that exports 80% bananas and 20% cars. /2

The “sentient AI” discussion not only shows a poor understanding of AI, but a poor understanding of basic philosophy, as it attributes sentience (ability to experience) to a behavior that is more related to agency (ability to do).

Let me explain.
/1 🧵 Since Aristotle, scholars have known of a basic model to understand minds. The model has 2 dimensions:

-Agency (ability to do) &
-Experience (ability to feel).

Babies 👶 are high experience (sentient) but low in agency (cannot do much or be responsible for their actions).

The Policy Implications of Economic Complexity.

Economic complexity methods are used frequently in international development. But the implications can be unclear. In this WP, I organize these policy implications into 4Ws: What, When, Where, & Who
arxiv.org/pdf/2205.02164…

1/12🧵 "What" methods are the most common implementation of economic complexity ideas. They focus on identifying target activities (products, industries, technologies), that a region can diversify into.

2020 Economic Complexity Index (ECI) numbers are out today!

How has economic complexity evolved during this decade?

In this thread, I discuss some key trends. /1

rankings: oec.world/en/rankings/ec…
summary: oec.world/en/blog/post/2…
#trade #development #economy #EconTwitter The Economic Complexity Index has become a popular indicator of economic capacity because of its ability to explain variations in economic growth, income inequality, & emissions (for a recent academic review see nature.com/articles/s4225…).

But the world is changing. /2

Want to read an insane tenure denial story? I was denied tenure at MIT in 2018. I was the only Hispanic faculty at my department, had 13k+ citations, 2 books, & papers in Nature, Science, & PNAS. I was never given a reason & the only letter I got told me to check the website./1🧵 After the denial I asked to meet my department head. He did not provide any reason other than “the decision came from above” & that there was nothing him or the department could do. So a few months later I asked to meet with him again & requested a meeting with the president.

For 10+ years I’ve been leading teams in academia & the private sector. When I started I was clueless, but I slowly picked up a few lessons that today I use to think about team leadership & management. Here is a list of three lessons I wish I had when I started🧵 /1 1. Manage around goals, not tasks.

This is a tough one for people making the transition. You probably earned the responsibility to manage a team because you were a good performer. But the ability to manage tasks that helped you perform well can backfire when you are a manager.

You published a paper and want to share the news with others. Should you:

A. Humble brag in a post saying something like: I am honored our paper is now out in Journal X.
Or
B. Write a thread communicating your results.

I am strongly in favor of B. Let me explain. 🧵/ In science, but also in the wider society, you can find cultures that discourage bragging. Unfortunately many scientists confuse communicating with self-promotion, and hence, censor by sharing their findings in ways that don’t communicate. /2

A question I am asked frequently after presenting "How Humans Judge Machines" is whether I think people will become more or less accepting of machines in the coming decades.

My answer often surprises some people in the audience... /1 My take is that people will not become more or less accepting of machines in the future, but instead, will become more polarized about it. /2

I don’t know if this is the meaning of life, but to me, it is damn close.
As I meet more people in their sunset years, & I imagine myself there, I see that some have achieved a zen-like quality that I hope one day I'll have. Those are people who can be truly happy for others. 🧵 No veil. No bullshit. Bliss for those living outside your own skin. I don’t think everyone gets there, but I think many do.