Chapters
Shownotes Transcript
Ryan Reynolds here for, I guess, my 100th Mint commercial. No, no, no, no, no, no, no, no, no. I mean, honestly, when I started this, I thought I'd only have to do like four of these. I mean, it's unlimited premium wireless for $15 a month. How are there still people paying two or three times that much? I'm sorry, I shouldn't be victim blaming here. Give it a try at mintmobile.com slash save whenever you're ready. For
This episode is brought to you by Shopify. Whether you're selling a little or a lot...
Shopify helps you do your thing, however you cha-ching. From the launch your online shop stage, all the way to the we just hit a million orders stage. No matter what stage you're in, Shopify's there to help you grow. Sign up for a $1 per month trial period at shopify.com slash special offer, all lowercase. That's shopify.com slash special offer.
Do you remember a first moment where you thought there is just something wrong with the way the field of economics is explaining our world? Yeah, I was actually, I remember I was on the Connecticut Turnpike with my brother-in-law, whose name is Cookie Gibb.
And he described the efficient market theory to me, that investors are rational. And so investors are processing the information and making the perfect decisions that will make the market efficient. And I just said, well, that's got to be wrong. I just don't believe that can be strictly true. You know, it just didn't sound plausible to me.
Driving down that turnpike with Cookie in the mid-80s, Doan Farmer was far from the first person to challenge this idea of rational people making rational decisions. I mean, economists, I have to say, to make it clear, mainstream economists are also challenging assumptions of rationality.
For decades, behavioral economics has been drawing from psychology and real-life experiments to better understand the irrational decisions people make. But they haven't been able to apply what they've learned on a grand scale, like to understand the whole economy. It's something that's very much up in the air. There's no agreed-on method for doing that yet.
So how do you study a messy, complex economy full of messy, complex people? I'm trying very hard to be nice. I don't want to criticize economists, but I just think we can do better, or at least we need to explore alternatives.
Doan started his career as a scientist, a physicist, and he thinks that that gives him a different perspective. I decided to combine my previous background in chaos and apply it to economics and finance. And so I've been doing that ever since. So this week in our series on economics, Doan Farmer tries to make a new kind of sense of our messy economy.
using the same tools he used back when he was a physicist. The tools of chaos. He's now the director of complexity economics at Oxford and has an upcoming book called Making Sense of Chaos. Can he help us understand the impact of tiny, irrational human decisions on the whole big economy? I'm Meredith Hodnot, and this is Unexplainable.
I'm here to talk about the economy. What, like it's hard? You may be wondering who got all this money. Nobody knows. The money's not here. Well, your money's in Joe's house and Mrs. Maitland's house and a hundred others. It's just money. It's made up. It doesn't exist. It's not real. It's not alive. If you believed. So tell me about the first time you thought about chaos. Yeah, I do remember.
I was having dinner at the commune I lived in during graduate school. This was in 1977 by the UC Santa Cruz campus. That night, Doan's friend Rob burst into the house. And was very excited and was telling us about
this new thing he just found out about, which at the time wasn't called chaos. There was something called a strange attractor. After dinner, everybody rushed up to the physics department and robbed Station Wagon. They crowded around an old analog computer. Now, this wasn't a computer like we think of today. There was no screen with text, no keyboard. There was just a small display with a little green dot and a bunch of knobs on the side.
Rob began dialing in equations that described different types of movement. He had a little knob he could turn, and as he changed that knob, he would start it at an initial point and then let it go. The first equation Rob programmed was a very simple movement, a movement like pulling a spring with a weight on the end. One equation with just two variables, how springy the spring is and how far you pull it.
The dot on the screen bounced up and down, but eventually fizzled out and stopped moving. It would just go to a point. It would just damp out. Like, you know, if you have a mass on a spring and you pull the mass, the mass on the spring will oscillate and it will come back to rest. Like there's an equilibrium there and it would just sort of rest at one spot. Yeah. This equation had something called a fixed point attractor.
Meaning that no matter how springy the spring is or how far you pull it, in the long term, it's attracted to the same fixed point.
Then Rob dialed in a second set of equations, the strange kind that he was so excited to show everyone. So it was a very simple set of equations that were discovered by Edward Lorenz, a meteorologist. Lorenz was trying to describe the movement of water being heated from below and cooled from above, because he wanted to understand how air currents might create turbulence.
So Rob set the initial starting point and released the little green dot. The dot danced around the screen, looping back and forth, making the same pattern over and over again. A pattern that looked a bit like a figure eight or maybe butterfly wings. The pattern the dot was making was clearly organized. At first it retraced its steps, but then all on its own,
It started wandering, still in this figure-eight pattern, but one time it'd loop in a little closer, and the next it'd loop out a little further. I was blown away. I went, wow, this is incredible. Instead of a fixed point, this attractor was strange. And we could even feed the signal into a speaker and make sounds and listen to it. Grrrr!
It would sound like a storm, like the wind blowing through the trees during a storm. It wasn't completely random. It wasn't like white noise, but it had that character that the wind through the trees has when a storm is going on. What was going through your mind as you were listening to these tones? I mean, I'd been educated as a physicist, and so I was used to motions that were regular, very predictable. And so...
Having the realization that the world could be unpredictable was kind of mind-blowing. You start something that's very orderly and yet you get disorder out the other end. So I wanted to understand how that worked. In science, chaos isn't like chaotic evil chaos. It's a pattern that at first seems predictable, but over time it changes and becomes harder and harder to predict.
Little things, tiny little disturbances in the pattern blow up into big deals. Take, for example, the planets. The planets move around the solar system following a pretty simple set of rules. Newton's laws, F equals ma plus Newton's law of gravity. You put those two together. Now we have a rule for how the planets are going to move. Saturn isn't jumping around the solar system randomly. It's following these rules.
So if I measure where the planets are right now... I can predict the future motion of those planets. But with chaos, teeny tiny differences between our measurements and reality grow bigger and bigger and bigger over time, leading to a dramatically different future than we predicted. In celestial mechanics, you can predict pretty well up till about 20 million years into the future and then just can't go there anymore because of chaos.
Yeah, it's a pretty long runway, though. It's a pretty long runway for celestial mechanics. For other things, it can be a very short runway. For the weather, it's about two weeks. The weather is just more complicated. There are more variables. Temperature, humidity, air pressure, wind speed, cloud cover, and literally billions more. Meteorologists try to measure all the variables they possibly can.
But every measurement is an opportunity to get things just a little bit wrong, maybe imperceptibly wrong. And those errors grow exponentially over time. Because of chaos, tomorrow's forecast is clear, but there's a horizon we just can't see beyond. Saying that, you know, two weeks from now it's going to be raining rather than sunny, that's very hard to do because of chaos.
Doan is trying to use the same tools scientists have been using to improve predictions about the weather, to improve predictions about the economy. Answering questions like, when will a recession hit? When will inflation rise? Because once we know that, we can try and avoid it.
How would you articulate your argument for looking at the economy through chaos? Well, chaos enters when you ask the question of what causes the economy to change. What causes business cycles like booms and busts, bubbles and crashes? According to Doan, a lot of mainstream economic models, the ones based on rationality,
are mostly focused on forces changing the economy from the outside. Well, like the COVID pandemic would be a good example. The COVID pandemic meant a lot of people couldn't go to work and it meant a lot of people weren't going to nightclubs or they weren't flying as much. The economy got hit by COVID, so it was an external shock. By their very nature, external shocks are totally random and unpredictable.
They can happen anytime, anyplace. I mean, I think of Zeus sitting up on Mount Olympus, hurling lightning bolts down at mortals below. In these models, economic change happens in the economy because random, completely unpredictable events keep knocking it off balance. But there's another kind of economic change, one that isn't so well represented in mainstream economic models.
Change coming from inside the economy itself. What we would call endogenous oscillation. Endogenous oscillation. Meaning things move all by themselves. This type of internal economic change could look like a bubble or a crash that was caused by people making bad decisions or taking bad measurements.
So, for example, in the crisis of 2008 was driven from inside the economy. It was caused by collective bad decision making by lots and lots of people, by the banks that were using too much leverage, by the banks that were lending with less collateral. So there was a housing bubble that then collapsed. These are all things that happened from inside the economy.
The 2008 financial crisis wasn't an unpredictable external shock like COVID. The crash happened because of systems within the economy itself, like credit, banking, insurance, these systems that were all interconnected. Let me maybe think about it a different way. This is an experiment that the listeners can do at their home. Amazing. Take a pole like a broomstick. Okay. And
Rest it in your palm and try and balance it vertically. Now, think about that pole as the economy and think about your hand as what we're doing to guide the economy. So the pole starts to fall. I move my hand. So the pole becomes straighter. It starts to fall a different direction. I move my hand. So there's this interaction between my pole and my hand. Right.
You got to kind of wiggle around trying to balance. Now, and we can see that, in fact, you can even show mathematically that that angle now oscillates chaotically. So the economy is oscillating because of my imperfect ability to balance the pole. Okay. I'm just not a perfect pole balancer. I don't steer the economy just right. We collectively don't steer the economy in the right way. We collectively are like the person trying to balance the pole.
And so, Doan argues that this internal, chaotic motion is at the heart of a lot of economic business cycles. Every one of our measurement errors, our irrational decisions, our imperfect predictions add up, millions and millions of them, and together build to the next economic bubble or crash. If we better understood how these forces influence the economy,
Doan thinks that we could use the tools of chaos to forecast the next recession. Like we forecast the weather. But to do that, he has to make a giant simulation of the whole world. Support for Unexplainable comes from Greenlight.
If you're a parent, teaching your kids about money early can help set them up for all kinds of milestones. Their first purchase, their first job, their first car. Not for a while, but still good habits. Greenlight is a debit card and money app made for families. Parents can send money to their kids and keep an eye on their spending and saving. And kids and teens can learn how to use their money better.
Oda Sham is my colleague here at Vox, and she gave Greenlight a spin. I would definitely recommend the Greenlight card for other families. And like, even as I am with my 10-year-old and his friends, I talk about it all the time because it's made our lives so much easier. Millions of parents and kids are learning about money on Greenlight. Sign up for Greenlight today and get your first month free when you go to greenlight.com slash unexplainable.
That's greenlight.com slash unexplainable to try Greenlight for free. Greenlight.com slash unexplainable. Support for Unexplainable comes from listening. You probably like listening to things, right? What if you could listen to more things like blogs or newsletters or academic studies?
Listening is an app that can turn any text into high quality audio that sounds like this. Whoa, whoa, whoa, whoa. I was just reading a paper about how intermittent access to beer promotes binge-like drinking in adolescent rats, but not adult rats. It seemed like it was going to be a cool paper, but all of a sudden I'm in an ad. Um, okay. Yeah.
So if you want to know more about binge drinking in rats and you really hate using your eyes to get information, check out the listening app. I put a scientific paper in there and I could listen back to it. I could change the playback speed. The whole thing was super easy. The listening app might just transform how you consume reading material.
This episode is brought to you by Shopify. Whether you're selling a little or a lot...
Shopify helps you do your thing, however you cha-ching. From the launch your online shop stage, all the way to the we just hit a million orders stage. No matter what stage you're in, Shopify's there to help you grow. Sign up for a $1 per month trial period at shopify.com slash special offer, all lowercase. That's shopify.com slash special offer. Chaos Control!
Doe and Farmer isn't the first researcher to think that the economy might be chaotic. Mainstream economists had this hypothesis already 40 years ago. But back in those days, computers were about a billionth as powerful as they are now. So the computer is the key tool that's enabling us to do this now. That's why the time is right. That's why it couldn't be done before.
The only way to prove the economy is chaotic is to build a digital simulation of it in a computer with lots and lots of data from the real world, and then try and make accurate predictions for how the economy changes. So, Doan is building simulations of the economy, kind of like a giant game of The Sims. Every day, these sim agents wake up...
And they start making decisions. Decisions about what to buy and what to build. But it's really important that these agents are not perfectly rational. The decisions they make aren't always the best. You might run your credit card up to a silly level where you're losing a lot of money. Or you might stuff all your money under your mattress, which might also not be rational.
Doan uses information from all kinds of data, from the census or household surveys, rental prices or demographics, and turns them into algorithms describing how people make decisions. Like, when this happens, this type of person does X and that kind of person does Y. It's the same for businesses and banks and governments, all making imperfect decisions.
We can run simulations with millions of agents with lots of institutional realism. And then he just lets it play out. Millions of agents making millions of decisions. Day after day after simulated day.
Then we see in our models, under at least some circumstances, we see these spontaneous oscillations of the economy. We see growth and decline, or we see financial crashes, as we do in the real world.
You could never make a true simulation of the whole world. So, like, what is the complexity that you're choosing to incorporate? Yeah, we are trying to make simulations of the whole world. That's a goal of ours. But there is always a decision about what you put in the model and what you leave out. Right. So, you always have to make a decision about what's key.
We have to be careful we don't get the model so cluttered up we don't understand what it's doing. What does that mean? Meaning you have to spend a lot of work dissecting your model, putting stethoscopes inside it so you get information about what works and what doesn't. You have to spend a lot of work so it's not a black box so you can see inside it. So the more complex you make the simulation, the harder it is to understand how it's coming up with what it's coming up with. That's right. That's right.
One thing mainstream economists don't like is that we don't use equations as much. Doan says mainstream economists like to use mathematical models to develop generalizations about the economy. Equations like when unemployment goes up, inflation goes down. Equations that are super useful when everything behaves as expected. If you have an equation, you can use that to explain something and really understand what it does.
And it's harder to get explanations out of a simulation. But my bigger concern is, is the model right? I'd still rather, at the end of the day, have a model that more faithfully captures the world as it is and gives better predictions and that ultimately is going to give me more reliable explanations. These models let economic phenomena emerge from the bottom up.
Doan's not trying to describe the economy with math. He's trying to describe people with math and then see what economy they build. So the main point of the book is that we can make better predictions about the economy by simulating it rather than by using the mathematical equations that mainstream economists use.
And that this is particularly useful when we want to deal with things like climate change, where the world is really out of equilibrium, where things are changing in a significant way. What do we still not understand about using this approach? Oh, there are many things we don't understand. One of the hardest problems is...
with making agent-based models, particularly as they get more complicated, is to get things right. Are we modeling the economy as it is? So this is like the building your sims to make sure that they're as lifelike as possible? Exactly right. And we're making progress, they're getting better and better, but there's still an enormous amount left to be done. By building these simulations, Doan is trying to recreate the chaotic heart of the economy.
Us. Our tiny imperfect decisions on a complex global scale. And if we are the source of chaos in the economy, maybe we can understand it better. Maybe we can even steer it. And that's a pretty special opportunity. Because most of the time, chaos is beyond our control.
It's the force that limits how far we can see into the future. It's the small moments that make big changes and set us on a different path. You know, in my life, I had the good fortune to encounter somebody when I was a teenager who had a huge impact on who I've become. I had the horrible misfortune to have my oldest son die in a sledding accident. And, you know, if he hit a pole, if he'd been six inches to the side, he would be alive.
Very small things can have big impacts on your life. And we have to accept that we can't do anything about it. But we can try and steer a better path with the things we can control. And the study of chaos is learning to be able to tell the difference. Yeah, between the things you can change and the things you can't and having the wisdom to know the difference. If you want to learn more about Doan's work, you should check out his book, Making Sense of Chaos.
And if you missed it earlier this month, you should listen to the rest of our Unexplainable Economy series. This episode was produced by me, Meredith Hodnot. I also manage the show. We had editing from Marianne McKeown with help from Jorge Just. Mixing and sound design from Christian Ayala. Scoring from Noam Hassenfeld. Fact-checking from Melissa Hirsch. And production support from Mandy Nguyen.
Bird helped with research on this story, and then she thanked the platypuses and told them that there was more work to be done. There were more beaked animals that needed to join their side. Next stop, Tortopolis, the capital of the tortoise empire. Special thanks to Dorothy Nichols, Rob Byers, and thanks, as always, to Mr. Brian Resnick for co-founding our show.
If you have any questions or thoughts about this episode, you can email us at unexplainable at Vox.com. You can also support the show and all of Vox's journalism by joining our membership program today. You can go to Vox.com slash members to sign up. And if you really love supporting us, leave us a nice rating or review wherever you listen. It really helps us out a lot.
Unexplainable is part of the Vox Media Podcast Network, and we will be back in your feed next week.
Life is full of complicated questions. I want to know how to tell if my dentist is scamming me. What age is it appropriate or legal to leave your kid at home? From the silly to the serious and even the controversial. Can I say something that will probably just get me canceled? I'm John Cullen Hill.
And I'm hosting a new podcast at Vox that'll be your go-to hotline for answers to the questions you don't know how to answer. Email a voice memo to askvox at vox.com or call 1-800-618-3545. I promise you it's better than asking ChatGPT.
