Joe: What if the people in charge of the economy, the ones at the central banks and treasury departments, are navigating with a map that's not just old, but dangerously wrong? A map that shows a flat earth when we're living in a world of mountains and valleys. Lewis: That’s a terrifying thought. It’s like giving a 16th-century sea chart to a modern container ship captain and saying, "Good luck, don't hit any continents we didn't know about." What makes the map so wrong? Joe: That's the explosive premise at the heart of Making Sense of Chaos: A Better Economics for a Better World by J. Doyne Farmer. He argues that the fundamental models of economics that we've relied on for a century are built on elegant, but flawed, assumptions that simply don't match reality. Lewis: And Farmer isn't your typical economist, right? This is a guy who started his career building a wearable computer to beat casinos at roulette and then founded one of the world's first automated hedge funds. He's a physicist who looks at the economy and sees a system ripe for a scientific revolution. Joe: Exactly. He’s an outsider who came in and asked the simple, almost childlike questions that the insiders had stopped asking. And the answers he found suggest we need to rethink economics from the ground up. Lewis: I love that. The brilliant troublemaker. So, if the old map is useless, we need to start by understanding why it's leading us off a cliff.

Joe: Precisely. And there’s no better example of this than the 2008 global financial crisis. Farmer points out that in the years leading up to it, the top economists at the New York Federal Reserve were actually worried. They asked themselves, "What's the worst that could happen if the housing market crashes?" Lewis: A very sensible question to ask. What did their super-sophisticated models say? Joe: They ran the simulation. They modeled a 20% drop in average house prices, which at the time seemed like a catastrophic scenario. Their main model, the FRB/US, crunched the numbers and came back with an answer: it would be a minor blip. A small, manageable recession. Lewis: Wow. And in reality, what happened? Joe: Well, house prices dropped by about 23%, and the global economy nearly vaporized. Five million people in the US alone lost their jobs, unemployment hit 10%, and the economy lost something like $10 trillion. The President of the European Central Bank at the time, Jean-Claude Trichet, later confessed, "In the face of the crisis, we felt abandoned by conventional tools." Lewis: Abandoned. That’s a chilling word from someone in that position. So, what was so fundamentally wrong with their models? Were they just missing a variable? Joe: It was deeper than that. The models were built on a few core assumptions that Farmer calls the "holy trinity" of standard economics. First, they assume that people are perfectly rational agents—what they call Homo economicus. We all have perfect information and make perfectly optimal decisions to maximize our own happiness or profit. Lewis: Okay, I can already see the problem there. I can't even rationally decide what to have for lunch, let alone optimize my entire life's utility. So it's like a blueprint for a car that assumes the driver is a flawless supercomputer. Joe: A perfect analogy. The second assumption is equilibrium. The models assume the economy is always in, or quickly returning to, a state of perfect balance where supply equals demand. It's a neat, tidy, and stable world. Lewis: But the real world is never in equilibrium. It's messy, it's constantly changing. It’s more like a chaotic weather system than a balanced scale. Joe: That's exactly the metaphor Farmer uses. And the third assumption is that all major changes in the economy are caused by external shocks. A war, a pandemic, a sudden oil crisis. The model itself has no internal engine for change. It’s like a rocking horse; it only moves if you hit it with a stick. Lewis: So, the 2008 crisis, in their view, must have been some huge, unforeseeable external shock? Joe: That’s the problem. There wasn't one. The crash was generated from within the system. It was a cascade of bad loans, interconnected banks, and flawed risk models that created a self-reinforcing death spiral. The standard economic map had no way of seeing it coming because, according to its own rules, it couldn't happen. The map didn't show mountains, so it couldn't predict you'd crash into one. Lewis: That is profoundly unsettling. It’s not just that the map was wrong; it was fundamentally incapable of being right. If the old map is that broken, what does Farmer’s new map of the world even look like?

Joe: This is where it gets really fascinating. Farmer's new map comes from a completely different field: complexity science. Instead of seeing the economy as a simple, predictable machine, he sees it as a complex adaptive system. Think less like a clock and more like a rainforest or a weather system. Lewis: A rainforest. I like that. It's teeming with life, there are predators and prey, there's cooperation and competition, and it's constantly evolving. It feels much closer to the truth. Joe: Exactly. And to understand how this works, we have to go back to that wild story of him beating roulette. In the 70s, he and his friends were physics grad students, and they got it in their heads that roulette wasn't truly random. They believed it was a physical system, governed by the laws of motion. Lewis: A very physics-student thing to believe. How did they even begin to test that? Joe: They bought a roulette wheel and installed it in their house. They spent months solving the equations of motion for the ball and the wheel. Then, they built the world's first wearable computer. It was hidden in a shoe and under an armpit. One person would click switches in their shoe to time the velocity of the ball and the wheel. The computer would calculate a prediction, and then send a signal to an accomplice who would place the bets. Lewis: You're kidding me. This is like a low-budget, high-IQ Ocean's Eleven. Did it work? Joe: It worked spectacularly. They had a 20 percent edge over the house, which is astronomical. They consistently won. The only reason they didn't get fabulously rich was because the hardware kept failing and they were, justifiably, afraid of getting their kneecaps broken. Lewis: That’s incredible. But what does beating roulette have to do with fixing the global economy? Joe: It's the core philosophical insight. What appears to be random is often just a complex system whose underlying dynamics we don't yet understand. If you can understand the physics of the system, you can make predictions. Farmer argues the economy is the same. It looks random, but it has its own physics, its own internal dynamics. Lewis: So, complexity economics is about finding the 'physics' of the economy? Joe: In a way, yes. And it leads to some mind-blowing ideas. For instance, Farmer and his colleagues started applying concepts from biology and ecology. They looked at the economy's production network—the web of industries buying from and selling to each other—and treated it like a food web. Lewis: A food web? Like, lions eat gazelles, gazelles eat grass? Joe: Precisely. In an ecosystem, you can calculate a 'trophic level' for each species—how many steps it is from the primary producers, like plants. They did the same for the economy. A primary industry like mining would be at the bottom. A car manufacturer, which buys steel, which is made from iron ore, would be at a higher trophic level. Lewis: Wait, you're telling me the same math that describes a food web can predict whether China's economy will grow faster than the US? That's wild. Joe: It sounds wild, but their research found a stunning correlation. Countries with a higher average trophic level—meaning more complex, longer supply chains—tend to have higher GDP growth. It’s because innovation cascades down the chain. An improvement in chip manufacturing benefits the laptop maker, the car maker, the airplane maker. A more complex economic ecosystem is a more innovative one. Lewis: My mind is a little bit blown. So we're moving from a model of a single, rational 'Robinson Crusoe' agent to an entire, messy, interconnected jungle. This is a fundamentally different way of seeing the world. Joe: It is. It’s a world of emergence, where the whole is truly different from the sum of its parts. It’s not about one rational agent; it's about the collective, often irrational, behavior of millions of interacting agents. Lewis: This is fascinating theory, but does it actually work in the real world? Can you make actual predictions with this stuff that are better than the old, broken models?

Joe: That is the billion-dollar question, and the answer is a resounding yes. This is where the revolution gets real. The book gives the ultimate proof-of-principle story, and it happened during the most uncertain economic moment of our lifetimes: the start of the COVID-19 pandemic. Lewis: March 2020. I remember it well. Nobody had a clue what was about to happen to the economy. Joe: Exactly. Governments were flying blind. They had to decide between locking down to save lives or staying open to save the economy. The standard models were useless because they couldn't handle a simultaneous supply shock—people can't go to work—and a demand shock—people can't go out and spend money. Lewis: So what did Farmer's team do? Joe: He mobilized his small group of students and postdocs at Oxford. They worked around the clock, fueled by adrenaline and takeout. Instead of using a top-down, aggregate model, they built an agent-based model from the bottom up. They looked at the US and UK economies, industry by industry, and asked two simple questions: How much will demand for this industry's products fall? And how much will its ability to supply those products fall due to lockdowns and sickness? Lewis: A 'digital twin' of the economy, piece by piece. How did they get the data? Joe: They were incredibly resourceful. They used occupational data to figure out which jobs could be done from home. They even found a list from the Italian government of which industries were deemed 'non-essential' to predict which sectors would be shut down. They pulled it all together into a dynamic model that showed how these shocks would ripple through the supply chain. Lewis: Okay, the moment of truth. What did their model predict? Joe: In May 2020, they sent their paper to UK government officials. Their model predicted that the UK's GDP would contract by 21.5% in the second quarter. At the time, the median forecast from major financial firms was a 16.6% contraction. The Bank of England was even more pessimistic, predicting a 30% crash. Lewis: And the actual number? Joe: The actual contraction was 22.1%. They were astonishingly, almost perfectly, accurate. They outperformed every major bank and government institution. Lewis: Whoa. So they built a 'digital twin' of the economy that was more accurate than the official government models during a once-in-a-century crisis? That's not just a lucky guess. Joe: It's proof that the approach works. By modeling the economy as the complex, interconnected network it actually is, you get a much clearer picture. And this is the same logic they're now applying to the biggest challenge of all: climate change. Lewis: Right, where the standard models have also been famously unhelpful. William Nordhaus won a Nobel Prize for a model that basically suggested we should aim for a 3.4°C warming because the cost of acting faster was too high. Joe: Exactly. But Farmer's team, using their data-driven approach to technology forecasting, found the complete opposite. They looked at the historical data for renewables like solar and wind. And they saw something unprecedented. Lewis: What was that? Joe: The costs were dropping exponentially, year after year, while deployment was growing exponentially. This combination has never happened for any other energy source. When you plug those real-world trends into a model, you get a shocking result. Lewis: Let me guess. It's not as expensive as we thought. Joe: It's better than that. A rapid green-energy transition, one that keeps us closer to the 1.5°C target, will most likely save us money. Tens of trillions of dollars, even before you account for the damages from climate change. It turns the entire cost-benefit debate on its head. Acting fast isn't a cost; it's an investment with a massive return. Lewis: That’s a game-changer. So the old models, by ignoring the real-world dynamics of technological progress, were not only wrong, they were pointing us in the exact opposite, and far more expensive, direction. Joe: Dangerously wrong, just like we said at the beginning.

Joe: So, the book's ultimate message is that we're flying blind, but we don't have to be. The tools of standard economics are like trying to predict a hurricane with a barometer and a wet finger in the wind. Complexity economics, with its agent-based models and data-driven approach, is like the modern satellite and computer simulations that actually show you where the storm is going. Lewis: It's a call to build a more 'conscious civilization,' as Farmer puts it. One that can actually see itself, understand its own internal weather patterns, and predict its own turbulence. It’s about moving from superstition to science. Joe: And it’s not just about avoiding crises. It's about steering toward a better future. A future with less inequality, a more stable financial system, and a healthier planet. The models show it's possible. Lewis: It makes you wonder, what other 'flat earth' maps are we still using in our own lives or in society? Where else are we relying on overly simple models for deeply complex realities? Joe: It's a powerful question. And it’s one that goes far beyond economics. We'd love to hear what our listeners think. What's a system you've always felt was more chaotic and complex than the simple explanations we're given? Join the conversation on our socials. Lewis: This is Aibrary, signing off.