Introduction

Patrick: [00:01:21] My guest today is Dan Egan, who is the Managing Director of Behavioral Finance and Investing at Betterment. If that job title sounds wide-ranging, it's because it is. Dan has his hands in most of the ways that Betterment interacts with its clients and how it invests their money. This is one very interested and smart guy who is clearly passionate about helping investors make better decisions. We talk about science fiction, automation, investor behavior, and how Betterment tries to solve problems beyond the automated asset allocation engine that is their bread and butter. You can find show notes at investorfieldguide.com/egan, E-G-A-N. Now, please enjoy my conversation with Dan Egan. Thank you very much for doing this with me. I try to start somewhere interesting every week so it's not just about business. Well, start with sci-fi.

Dan: [00:02:11] Excellent

What Makes Great Science Fiction?

Patrick: [00:02:12] We met in person for the first time maybe a month or so ago. Both of us were speaking at the Evidence-Based Investing Conference, and we got to talking about sci-fi. So I'd love for you to share maybe your own origin story for how you developed the passion for reading sci-fi novels. Then we'll get into some favorites and some more specific questions from there.

Dan: [00:02:33] Definitely. I was trying to think back on this. I think the first thing I ever read which might've been considered sci-fi was actually a Kurt Vonnegut short story called Harrison Bergeron, I believe. It's actually just this social fiction thing about: what if we got into a world where everybody was fundamentally equal? There's this person called the Handicapper General, where if you're unusually beautiful, you have to wear a mask, or if you're unusually strong, you have to carry around these weights. So it's this pursuit of everybody actually being completely equal to each other and the dystopia that comes along with that, the lack of vibrancy and diversity that you get from a society that allows people to be different and really enjoy each other. So I think that was probably the first one, and that very simple what-if way of thinking, like, "What if we changed this? How can we think about what's really going on that are unspoken assumptions about how we work or how society works?" was what kept pulling me down into it. I think there's a lot of stuff that's considered sci-fi, but when you look at... One of the core things for good sci-fi for me is that we're going to either talk about real science or we're going to change something about the way things work that's going to tell us about ourselves. A lot of sci-fi is just about space operas and various things, which are really entertaining and really easy to read but don't play that what-if game quite so much. And I think Ursula K. Le Guin wrote a beautiful introduction to one of her books that says, "The job of this is to really think about if we change that there are two genders. What if there's only one, or what if you can be both genders? What if you can be immortal?" All of these what-if questions about how we change society, really make you think, and I love that.

Patrick: [00:04:29] This is already a great example of why sci-fi's so neat because I had never heard of the Vonnegut story, but have you heard of John Rawls, the moral philosopher John Rawls?

Dan: [00:04:37] Absolutely.

Patrick: [00:04:37] So it made me think of him, where his question that he poses, which is such an interesting one, is: what kind of system of government or society would you design if you knew you had to live in it, but you had no idea where you'd be born? You could be born at the absolute bottom of the absolute top. What would give the fairest shake no matter what? which is such an interesting thought experiment. And it seems like sci-fi, good sci-fi, is great storytelling married to systematic thinking. So these writers want to understand the levers that make not only people but institutions and countries and nations tick and then peel something away and make you think about it in a different fashion.

Dan: [00:05:21] Definitely, absolutely.

Patrick: [00:05:22] Can you give you... I don't know, as many as you want because there are so many, some of your most favorite or most formative and maybe even dive into some of the ideas, the reasons why you like them so much?

Dan: [00:05:35] Definitely. One of the ones that I read recently that I really, really loved was in this book, Your Life and People Like You, I think it was, which's a book of short stories. One of the things I've learned recently is that I love short stories. I love long-form, but at the same time, there is a book Sum, just S-U-M, it's like a bathroom book. All of these are 10-page short stories. So Your Life and People Like You has a story in it where basically, by default, people have these very easy, noninvasive procedure is done to them that makes them not realize when people are attractive. You can still identify people. You just don't have that lizard brain reaction of, "You're attractive, and you're not attractive." Most people go through life having attraction be a non-physical thing, and this is pervasive. It's not just about who you end up hitting on at a bar. It's who gets hired in a job or who you're friends with and all these things. And when children come of age in this book, they can have that turned off. They are given the choice at that point to decide if they want to have this visceral feedback loop about what other people look like. It's told from a couple of different viewpoints, where there's a young woman coming of age, and recently, her boyfriend who she had loved broke up with her, and she realizes that she's quite attractive. And she's like, "Well, wait, no. He should really like me. He hasn't turned this thing off." And he's less attractive than her, and he realizes that he doesn't like this feeling of, number one, realizing that he's unattractive but also knowing why other people are judging him like this. So great example of like, "Let's take one thing, the pervasive effect of attractiveness across our society, and let's turn it off, or let's give people the option of turning it off." The other cool thing about it that I think about it, obviously from a behavioral perspective, the idea of defaults. By default, naturally, we're going to notice attractiveness. We're going to notice symmetry in people's faces, all this stuff. By default, we're able to get pregnant. So what if we turn these on things off, and we flip them and said, "The default is not that thing. You have to choose to do that thing." Both people, before they get pregnant, they have to actually say, I don't know, "I'm going to flip this switch, and now I can actually get pregnant," both from the man's side and the woman's side. "I'm going to turn on this thing that allows me to see other people as being attractive. Think about all of the little consequences that bleed across society by making that one little change. It's really quite remarkable when you realize how much of your own life is influenced in tiny, tiny ways by these automatic defaults that you didn't even think about.

Patrick: [00:08:12] So it seems like some of the books you love are... both ideas you mentioned so far, identifying customs or conventions and not so much flying in the face of them but just asking the question: what if it was different? What if the defaults were different?

Dan: [00:08:29] Absolutely.

Patrick: [00:08:29] What about the big names? You have the Philip K. Dicks, the Isaac Asimov's. Are you a fan of some of the bigger works?

Dan: [00:08:39] Definitely. I actually went after, at the conference you mentioned, Ubik went and read it.

Patrick: [00:08:46] Yep, crazy one.

Dan: [00:08:46] And it really reminded me of...

Patrick: [00:08:48] What'd you think?

Dan: [00:08:49] It was really good. It's funny how, if you were to watch Star Wars... I'm 36. I remember being, I don't know, like eight or nine and watching Star Wars and being like, "This is the most amazing thing now." I kind of feel like kids today, if they watch Star Wars, they're like, "The special effects are kind of janky." So I read it, and I was like, "This reminds me of, number one, the end of Inception," the movie where he spins a top and you don't know. You don't know if it's going to fall over. For reference... I don't know. Should I spoil it for everybody?

Patrick: [00:09:21] It's a hard book to spoil just because it's so out there and odd, and no one's actually going to read it because it's the most esoteric sci-fi book.

Dan: [00:09:28] Good point.

Patrick: [00:09:28] But maybe don't spoil it. But you can maybe explain what the Inception idea is.

Dan: [00:09:32] I mean, I think the fundamental thing is that there's this existential question that the book ends with about: are we what we think we are? And it just cliff-hangs you like the end of Inception, with the top, it's like, "Is he still dreaming?" You don't know. That brought me back to actually... and that's funny that I'm now in movies, to The Matrix, where the original movie, you got the traitor with the little goatee. All bad guys always have facial hair. And he's sitting with the Agent Smith thing eating a steak, and he's like, "I know this isn't real, but it tastes good, and I want to be rich when I come back," and this voluntary choice of saying, "I will choose to be in something that I know is not real because I know it will feel good."

Patrick: [00:10:16] If you had to pick three books to dip their toes into the water, someone that's never read a sci-fi book before, what three books would you prescribe, where the idea is, if you finish these three books and you don't like them, then you've learned something, which is that you're not going to be a sci-fi fan? But if you do like them, they're going to be three good books that give you a sense of what's available out there, what's possible if you continue to explore the genre.

Dan: [00:10:43] One author who I think is worth trying out at least one is Ian M. Banks. He wrote a whole series of sci-fi novels placed very far into the future. He's one of the first people... Lots of sci-fi authors do this. It's one of the best plot devices ever, is the idea of time, about: what if you can live longer? I think Interstellar did a great job of it, where they say, "If some of your family starts traveling close to the speed of light, they're going to age differently than you. How is that going to change your family structure? And how does it change the way that we think about... In Star Wars and all these things, people are firing lasers at each other. If all these ships are moving at the speed of light, they're just going to blow right past each other and never even see each other. So Banks is good for what I consider space opera but dealing with the physics of the situation. The Three-Body Problem is this book by... I'm going to mangle the name, Cixin Liu, C-I-X-I-N, L-I-U, who's a science fiction writer that's coming out of China right now and has-

Patrick: [00:11:47] It's awesome. I'm reading it right now on your recommendation.

Dan: [00:11:50] Yeah, absolutely one of the best that I've read recently. It kind of deals with, actually, the game theory of being an intelligent species in the universe. One of the things that evolution has kind of brought us to is this idea of individual selection, that is my genes that need to replicate, and there's no such thing as group selection, where we all have to band together to actually fight something. And, all of a sudden, when you're on a cosmological scale, your entire species needs to survive, and there's another intelligent species out there. This comes through in things like Ender's Game and everything. How do we get over nationalistic tendencies and tribal tendencies and everything? For better or for worse, it seems like the one way we can imagine doing that is if we define something so incredibly alien that it scares us more than each other.

Behavioral Economics

Patrick: [00:12:31] Yeah. Yeah, band together. Well, a great couple of books to check out. I'll link to a few more in the show notes, as well. Now we can pivot to your day job. You're a behavioral economist, I guess, by trade, which is, for me, such a fascinating and continues to be an emergent field, now obviously popularized all over the place. Michael Lewis has a new book out about the birth of this field of thought and inquiry, and I'm curious what that means day-to-day for you. We've talked about it offline, but I think it'd be interesting for people out there that, within this huge field that's become so important that we're not these automatons, that we're people and that we behave with biases, how that translates into what you're trying to do to help your clients.

Dan: [00:13:23] I think one of the best parts about it is that I am effectively thinking about how to build something for myself and other people like me. There's sometimes a feeling of relationship and financial advice where's there's an expert who knows more than you and you come to them because of the asymmetry in knowledge. And what I do here is I'm effectively saying, "I am this flawed human being. I know where I go wrong. I know how I wish things were done better," and taking a very specific set of academic work, usually, that's been done that I can effectively free-write on and applying it to make people better investors. I'm thinking about what triggers people to do well. What are the fundamental predictors of success in terms of savings and tax and fees and market timing and so on? If I wanted to design a system that made me a better investor with knowledge of all of those things, and then everybody else gets it for free, I'm going to be looking at things like an investor dashboard is, like a heads-up display on a fighter jet. You would never put an odometer that says, "You've come 2,000 miles." And yet, if you log into most financial dashboards, they're going to say what your returns have been over the past however long a period, which is not useful information. It doesn't tell you what turns you need to make in the future, upcoming decisions you have to do, and so on. So a lot of it comes to the design of what information should be right in somebody's face and how you portray it, what's it's going to trigger in them, what emotions, what thought processes, and having a view about... We want them to end up in a good place. How do we design for success?

If I wanted to design a system that made me a better investor with knowledge of all of those things, and then everybody else gets it for free, I'm going to be looking at things like an investor dashboard is, like a heads-up display on a fighter jet. You would never put an odometer that says, "You've come 2,000 miles." And yet, if you log into most financial dashboards, they're going to say what your returns have been over the past however long a period, which is not useful information.

Patrick: [00:15:03] It seems like it is something that would touch every aspect of the interface, marketing, the brand, the product. Is that true? Do you have your hands in all those areas?

Dan: [00:15:15] Almost all of them. I like to think about that... One of the better ways is helping people design the systems, the process, and so on so that it's not that other people have to know what the purpose of behavioral economics or finance is but rather that they're just experiencing it seamlessly. One of the good examples that we've had is in the hiring process. Hiring, obviously, in finance and tech, has historically had some effectively profiling problems. Even for myself, when I'm hiring people, I don't really care about what their name is, what gender they are. So I want to set up an initial application process that really is good at filtering out people who are not qualified and who I don't want and lets everybody else through. So what I started doing on my own team, we were just getting too many applications for analysts, is I said we're going to have a normal application process, and we're going to add on two questions, one which is a super simple coding question that, if you know how to code, you're going to be like, "This is fun. I'm going to knock this out in 5 to 10 minutes, no biggie," and another one, which is just, "Do you have a view on a somewhat complex diversification question?" just to show that you've thought about it and this is something in the back of your head. The purpose of those is not only to test the knowledge and natural passion of these people. Are they coming to the table with the skills necessary? But are they okay just investing those extra whatever it is, 10 or 20 minutes to answer those questions? As soon as you do that, finished application rates drop to about 5% of what they had previously been, and I'm only dealing with 5% of absolutely fantastic, high-caliber candidates. So take that process, have two questions that basically are going to naturally identify who's going to be a great fit for this. That's your filter, and you should interview almost everybody who gets through that filter because they've shown that they're going to be good at their job.

Patrick: [00:17:01] The interesting term for this, or the popular term, is this idea of nudges, that if you can sort of set defaults or screens, you can get a long way from what would otherwise be a real mess. So I want to come back to that dashboard idea, the investor dashboard. First, how often do Betterment clients, on average, check their account?

Dan: [00:17:23] Too often. There is no... And I think part of that is we've made it-

Patrick: [00:17:28] Is it daily? On average, is it monthly? What's the frequency?

Dan: [00:17:31] I don't know off the top of my head. I would put it probably somewhere between weekly and biweekly. I think, when you look at the numbers, there's a little bit of strangeness to it in that we're growing very quickly. So people check at an astonishing rate in the first 90 days. Some of that's actually good. They're setting up their account. They need to log in. They need to set up goals. They need to understand the way the system works. And part of it is even just a basic auditing procedure: "Are they doing what I think they're doing? Let me just double-check and make sure. When they're saying they're investing in 12 ETFs in fractional shares, are they actually doing that? Okay, they're actually doing that." So it tends to trail off significantly after that 90-day period. It nudges up a little bit. We generally go from about... I believe it is 50 in 10,000 customers logging in during normal market periods to about 100 per 10,000 customers logging in during things like Brexit or the election when they have some stimulus that they're worried about. And it is definitely higher on mobile. Mobile phones are, without a doubt, the most dangerous environment to be making investing decisions in because it's so small, it's so contextless, and for better or for worse, you can be in a bathroom at a bar after just hearing a tip from some guy, and you can actually make the decision without any thought at that point in time. So, making a mobile experience that is on par in terms of nudges with a larger web-based experience has been one of the most interesting challenges.

How to Stop Making Common Investment Mistakes

Patrick: [00:18:58] Really interesting, and I love this history of the transaction part of buying and selling securities, where the costs and arduousness of it used to be much more cumbersome. It was a whole process to even buy a single share. And now obviously it's incredibly easy. Now, it's much, much cheaper to do it now, which I guess is good, I think is good. But maybe's it's not good because everything we know about average holding periods and over-trading and some of these things that have actually gotten worse as the tools themselves have gotten better. So how do you think about that? How can you, as a behavioral economist, try to start to chip away at the biggest problem in investing, which is this behavior gap, this behavior problem? I'm fascinated by this. We can spend as much time here as you want. I'd love to hear examples, tangible examples of things that you have, systems that you've set up, nudges that you've introduced, suggestions you have for the future for how we can get people to stop making all these same stupid, common mistakes.

Dan: [00:20:06] Absolutely. I think you touched on a great point. I think financial instruments and financial costs have gone down so much that I think about it like a power tool. We used to have maybe an axe. An axe is a dead simple to move. You can still have an accident with it. But now we have chainsaws, power saws that could really do some damage, and you need to design them very thoughtfully with things like dead man switches and safeties and various things like that, or sometimes just that it's not the right tool for the job and making it very clear. One of the most common things that I see as a mistake, there are a couple of them, is what I call prohibition. So I'm just not going to let you do that. That doesn't work, and I think we saw that in the 1920s. The response that has been more intelligent to say, "We're going to try and figure out how to mediate and kind of reduce the extremity of this kind of behavior." The way my parents said it when I was about 16 or 17, they're like, "We know you're going to drink. Just do it here because that way you're not out driving around, and you're not at greater risk for making mistakes." Don't prohibit it. You can't stop people and say, "You're just not allowed to do that." You have to allow them to do it but really give them salient options that they can go to instead or things that are going to stop them on their track. They can do it, that is not going to stop them, but they're going to make them think about it differently.

A great example of this is a service we have called Tax Impact Preview. So, when a customer comes in in a taxable account and goes to change their asset allocation, say from 90% stocks down to 0% stocks because they're worried about the election or whatever, it's going to, in realtime, calculate exactly how much tax we estimate they're going to pay at the end of the year because of this transaction. This is done in realtime. We don't necessarily know what their end-of-year tax situation is going to be, but this is the impact of this decision. And generally speaking... We split-test this. We actually run tests to see whether or not we're doing the right things for our customers. It reduced allocation changes by somewhere around 70%. You can either think about this in a classical rational economist way, that people just didn't realize the effect that these decisions would have in the next year, and we've just given them the relevant information and helped them make a better decision. Or you can think about it in what's called a counter-biasing format, which is that people really hate taxes, and by putting up some tax figure, it has a disproportionate effect upon their reaction to it. Generally speaking, it seems like it's the latter because, number one, the average tax impact that we showed people was a little bit less than $1. So, if you're really worried about markets tanking, you should not be worried about paying $1 in tax next April. The other thing that was very interesting about it... This isn't perfectly defined, but we looked at where people were from and mapped where they lived back to the vote count in the last presidential election, so the second Obama administration. We said, "How strongly Democrat or Republican was it?" We'll view you as being randomly drawn from this county and look at how sensitive you are to tax-impact preview. People who tend to be from very liberal Democratic areas tended to be far less sensitive to tax than people who came from very conservative areas, so not perfect. We don't actually know that the customer was or wasn't fiscally conservative or anything, but it's a good indicator, and that people really dislike tax in an almost irrational way. This is not just a mathematical decision. There's actually preferences, and we're stopping them in their tracks from making a market timing decision, that they're worried about the market going down, with a definite, certain tax decision where they say, "I'm definitely going to pay that."

Patrick: [00:23:46] Do they have to click some sort of button? How is this served up to them? Is it a gate they have to get through, or is it just like, "By the way, here's the number," and then execute?

Dan: [00:23:58] One of the things I love about working at Betterment, we have great product designers who think through these issues very well. And we tend to take a conservative, iterative approach. So the very first time that we launched this, you had to click a button. You had to opt in to see this. When we saw how effective it was, they actually changed and said, "Wait, it looks like it's doing the right thing." One of the things that we were worried about it backfiring on is withdrawals. So, if you need to pay for your car to be fixed, you should take the money out. The tax is not that big a problem. Don't put it on a credit card. Actually, pay for it. So we were looking at: is this also influencing people's withdrawals in any negative fashion? The answer was no. Okay. It's doing the right thing, and it's not doing the bad thing. Once we had confidence in that, the product designer said, "Okay, we're going to make it non-optional. Everybody is going to see this if you are doing something that has a potential tax impact." So we were able to iterate through and get comfortable with the idea that we were going to be kind of forcefully nudging people to think about these things, but only because we'd studied it along the way.

Patrick: [00:24:59] This made me pull up on my computer a little sentence. There's a book called The One Sentence Persuasion Course. I'm sorry for the author because I'm going to read the sentence. It's a pretty short book. But it says, "People will do anything for those who encourage their dreams, justify their failures, allay their fears, confirm their suspicions, and help them throw rocks at their enemies." So this tax thing as an enemy is an interesting analog, that people are so averse to it that you're able to... I mean, 70% is a massive reduction because one of the criticisms I hear most often from random people I'm talking to is that, in the robot-advising world, it's going to be too easy to make... Just like buying and selling is so much easier and cheaper now, it's going to be too easy to make asset allocation changes that reflect behavioral mistakes. Are there other things that you've tried or that you see beyond... The tax one seems brilliant. But what else do you do? What are some other examples?

Dan: [00:26:01] They're even really small things. So it used to be that, when you logged in, on your Summary page, losses and gains were coded as bright green and bright orange, classic, things that you see every day. And I fought for quite a while to say, "Listen, can we just make them less emotionally laden. They're a number. There's a positive number and a negative number." We took those off, and there were people who were a little bit grumpy about it, but generally speaking, I have the benefit right now of being able to say, "We're either building for the only 220,000 customers that Betterment has right now or building for the half a million or one million or two million that we're going to have in the future. And I want to get this right from the ground level up." So that's a good example of something that's reduced it over time. If you go to a Performance page, we now have the ability to, if you add in the S&P 500 as a benchmark index to your portfolio, automatically, content talking about the pitfalls of using that as the benchmark goes up on that screen and encourage you to read them. And I think this is one of the interesting things about the general behavioral thing, is how much of it is education. All the evidence points to the fact that, if you're going to do education, it needs to be right point-in-time. What's the question on somebody's mind? What are they thinking about? You have to deliver it super surgically in order for it to be effective. Otherwise, don't worry about education. Just do nudges and various other design elements. I think one of the misunderstandings, also, that people have about Betterment is who our customers are, so this... when a big downturn comes. So here's a genuine question for you because I love these criticisms because I have no idea what they actually mean. What's a big downturn?

Patrick: [00:27:40] I would probably say it's something like 30% or more.

Dan: [00:27:44] Right. So, over the period that I've been at Betterment, we've experienced... I think it's like three 10% drawdowns, so "correction," and one 20% drawdown. I think it actually topped out at like 19.6% down in the S&P 500. These are the kind of little false alarms or whiplash-y type drawdowns that I think actually are what hurt people's long-term returns because they see the 10 or 20% drawdown and like, "Oh, this might be the 30 or 40% drawdown. I need to get out now." Since I've been here, there have been absolutely the most perfect, classic whipsaws where it's a drawdown, and then it rallies, a drawdown, and then it rallies. So, if people haven't learned over the past three or four years about head fake type drawdowns, they haven't really been paying attention. But I actually am a bit sanguine that, when there's a bigger drawdown, we are going to be fine. I think there's a lot of self-selection that works very well for us. We have a younger customer base who are generally in the accumulation phase, and they tend to regard drawdowns as a buying opportunity. If you're 45, you still have 20 years left until retirement, you see a big drawdown, and you say, "Well, I have 20 years. I'm going to top it up. I've got a bit of spare cash on the side." We tend to self-select people who understand a passive index-tracking philosophy. So they naturally are not thinking that they're smarter by reacting to market movements. They're going to tend to stick through it. We have a lot of tax features that make drawdowns a little bit like lemonade out of lemons, tax-loss harvesting, for example. You get to look at this and be like, "Oh, well, at least I'm going to have some tax offsets this year." I think the thing that is really interesting from a public point of view is, say Betterment has half a million customers, if one basis point of them is really unhappy, so 1% of 1% of that entire customer is unhappy, you're going to hear about it. They're going to be on Twitter. They're going to be on various other places being publicly grumpy about it, and you're really going to miss this amazing silent majority of people who are like, "I'm perfectly cool. I think it's fine." So I think that there's a kind of asymmetry of public exposure that we have right now where we got a great customer base who are bought into what we're doing, who genuinely don't seem to react negatively, but you're going to hear about the very loud minority.

We have a younger customer base who are generally in the accumulation phase, and they tend to regard drawdowns as a buying opportunity. If you're 45, you still have 20 years left until retirement, you see a big drawdown, and you say, "Well, I have 20 years. I'm going to top it up. I've got a bit of spare cash on the side."

Patrick: [00:29:59] Do you think that when Betterment gets to... Let's say you get to two million customers, right? That's so many people that, at the 30, 40, 50% bear that we see next, they're not going to be any different because people don't sell at the bottom. Markets bottom because people sell, right?

Dan: [00:30:16] Yeah, yeah, yeah.

Patrick: [00:30:17] I think people get that backward. It just seems hard that, once you get to a sufficiently large size, and the average account size hopefully grows as people continue into their prime earning years, seems like it would take a herculean behavioral win for you to keep people from making the same mistakes that everyone else makes, right?

Dan: [00:30:39] I like to call that job security. I think that there is genuinely a ton to do. One of the best parts of my job is that we have these ideas about things we can do. And I get to call up some of the leading academics in the space and say, "What do you think about this?" It's not like we're Panglossian about whether or not this happens. It's something that is on our minds. To the point of how we do things or how I think you kind of deal with habits and so on, we are never making decisions about what to do in a downturn during a downturn. We have always set out a systematic plan of if-this-then-that type flows. What if this happens? Then here's what we're going to do. Here's how we're going to pull the trigger on various things so that, during some kind of a crisis, we're just executing on a plan. So we have those plans, and we're always adding to them and learning from the last one. Every single downturn, we have run some kind of an experiment to test whether or not we are doing things effectively.

Patrick: [00:31:34] Can you give me an example?

Dan: [00:31:35] So, very early on, I think it was late... When was the taper tantrum, late 2013?

Patrick: [00:31:41] 2013, yeah, or '14 maybe.

Dan: [00:31:47] There was media coverage. It was a classic, "This could be bad. What's going to happen next?" type scenario. Conventional wisdom is that you should reach out to your customers before they reach out to you. So we had heard this. I had heard this from my previous roles. So we said, "Okay. We're going to send an email, a timely email telling people what's going on and why and so on." There were actually four different conditions for that email. Number one was a control, which was actually probably about 50 or 60% of the population because we weren't sure that this email was going to be effective. We wanted to hold them back because you can always email a control group after you've done it if one of those treatment emails is really good. And then we had three or four different treatment emails that either focused on what's going on in the markets or a very positive, person congratulatory issue that you haven't done anything yet, but you seem to be acting rationally. This is what's called generally a growth mindset psychology. Tell somebody that they are the good that you want them to be, and they'll feel very motivated to fulfill that implicit compliment. And then a couple of other very normal... One of them's like, "Don't do anything. Stay the course," what I think of as a very negative frame of what somebody can do. And what we found was that, compared to the control group, the people who we sent the email to just did more in general. They were more likely to deposit, as well, but they were more likely to change their allocation or withdrawal. That, to me, indicated there's this targeting issue. Most people weren't concerned about what was going on. They might not have even been paying attention to it. But we were the ones to initiate the conversation about something that maybe they should be anxious about, and we kind of caused... We had some false positives in our treatment arm. So we switched from it being a shotgun blast thing out to all of our customers to when somebody logs in during some kind of market stress or market crisis moment, we're going to message them right then. It's going to be part of that initial log-in experience, is, "Here's the message about what's going on right now. And you're doing the right thing. Don't worry about it." So, in that case, we were able to go from a high false-positive rate to basically nearly a 100% surgical targeting of people who are going to be stressed and are at risk of doing something.

How Dan Egan Got Into Behavioral Economics

Patrick: [00:34:03] Yeah, amazing the leverage that you can get through technology solutions. I have not, frankly, made up my mind on this, on robot versus a really skilled advisor. I know the power of relationships. So I just deeply believe that there is some value in that if it's done well, but I can't quantify that. And I don't think there are good studies or it would even be possible to run a good study as to whether or not one is better than the other. But given, obviously, I know where you stand on the issue, I'd be curious to hear some of your investment philosophy roots, so how you got to where you are today, which I think it's fair to categorize you as maybe not an efficient markets thinker as a behavioral economist, but as someone who believes in asset allocation, in low cost probably above all else. I would just be curious to hear where you started and how you got to where you are today.

Dan: [00:35:00] Yeah. I grew up in Philadelphia, and I probably had heard the name Vanguard and John Bogle long before most people really even should have, for better or for worse. But, yeah, I really came into it from a classical economics education. And you keep hearing these sort of, "The market is efficient." I don't know that I ever really bought into it. I'm kind of a, I don't know, semi-weak form... The market is sufficiently efficient that I don't really need to care about its inefficiencies. But what I think I came into it is I started looking at it from my point of view about where I have leverage. And I mean a classic: what are these things where I can affect a small change on my side, and there's a huge change and a huge impact with high certainty on an outcome in the future? These are things that have high certainty. I tend to not think low-cost. I tend to think of net value. But if I'm getting the exact same thing for a lower cost, that's pretty much a guaranteed improvement in my outcome. So I try to find things, taxes are another one, that is pretty darn predictable. There's something that really influences normal investors like you and me, that all of the people who you tend to think of as being the great investors who work with charities and endowments and various things, they don't have to worry about this. This isn't part of the optimization process, looking at liability identification. So, through the course of my life, I had to buy a wedding ring. I had to put a down payment on a house. Now I've got a kid. I have to be realistic about... There are these things in the future that I'm pretty much going to have to spend time on and money on. And the better I am at identifying and planning for them further in advance, the more immunized I'm going to be from that kind of liabilities. The second is genuinely trying to remove myself from execution and management as much as possible, and this gets into things like habits but also automation, where I want to sit down and say, "What do I believe? What are my principles about how I should do this?" And I want to design a system that automatically does it for me. I want to be at arm's length from the execution. If you think about it as even a tactical trader, you say, "Here's my hypothesis. Here are my entry points. Here are my exit points. Now, I don't want to be involved with actually doing those trades. I've spelled out what should happen, and somebody else should just go and do it." And I think that that's a very strong form of the behavioral perspective of ... people are so fallible in certain execution settings that they should never do them. They're in a hot emotional state. One of my, I don't know, favorite and scariest studies on this is on judges granting people parole or probation. It's basically like, "When did they last eat?" If you're going to go for parole at like 4:30, just don't. There's no chance of you getting it. This is such a serious decision. There should be some way of saying, "The way the judge will make a decision is... Here are the factors. Here's how you weight them," and it should not involve when they last ate. So I try and take myself out of the doing as much as possible and instead design the thought process or flow diagram or decision structure so that it can be executed properly.

Patrick: [00:38:15] That was one of those studies that knocked my socks off, too, so much so that I was very skeptical of it. I wanted to see what the sample size. I actually found it through Daniel Kahneman, a pretty big name talking about this study. And the fact the decisions the judge was making were parole decisions, which is crazy, right? They're deciding whether or not someone's going to go free.

Dan: [00:38:38] I know.

Patrick: [00:38:39] And it is just a... Every time they eat, the rate goes up to whatever it is, 60-something, and it goes down pretty much in a linear line to close to zero until right before their next meal.

Dan: [00:38:50] It's amazing, zero.

Patrick: [00:38:50] It's like, "Oh, my god. This is...

Dan: [00:38:51] Yep. They found similar things, this is even scarier, with doctors. Something that I joke about but I actually think is really true, if you're going to the hospital or something, bring chocolate because doctors, the high-end diagnostic ability goes way up if you just literally give them a piece of chocolate and wait 5 or 10 minutes. If you've got something seriously wrong, just give your doctor a chocolate bar, chat for a little while, and then get into the nitty-gritty, and they'll be actually better at diagnosing you.

Automation in the Information Age

Patrick: [00:39:15] This is going to be a little tangent, but you're exactly the person whose opinion I want on this. So we talked about sci-fi. So, obviously, you think about the future. We're talking about the fallibility of human decision-making. I am curious to hear how worried you are about this coming wave of automation, where we're in this kind of Information Age now. It used to be that we were automating mostly repeatable, mechanical jobs, and now there are computers that read x-rays better than radiologists. There are all of these not just skill but problems like the two you just mentioned with judges and doctors, two positions that most parents would be thrilled if their child ended up as a judge or a doctor, right?

Dan: [00:40:04] Definitely.

Patrick: [00:40:04] That's at the top of society, and it seems as though big aspects of their jobs are susceptible to a more repeatable, automated system that split tests, that learns, that's automated. So what do you think about all this? It's an open question. We don't know what's going to happen, but I'm sure you've thought about this. I'm curious to hear your thoughts.

Dan: [00:40:25] I'm incredibly, incredibly worried, and incredibly, incredibly excited at exactly the same time. The optimist in me says, "As long as we nail education, this will be fine. As long as everybody feels like they're a part of the system, and I think we worry... Sci-fi, there's this phrase that Ian M. Banks wrote in one novel that has always stuck with me, "Money is a symptom of poverty." Truly rich cultures and societies don't need money because of the level of scarcity of resources, they don't use tokens to exchange things to get things done. Maybe it's pride and prestige. Maybe it's some sort of vocational pull that leads you to be artisanal in one way or to do incredible plays. I don't know what it is, but there's this idea that, at some point, if we've automated a lot of the stuff that, to be honest, throughout history, we wanted to automate... Nobody wants to be a mine worker, a coal mine worker. Nobody wants to do incredibly dangerous labor that might end up with their wife being a widow and having an orphan. So, once we take away those bad things, what are we as people in terms of our vocations and jobs? That's where I think the scary part is for me, is that a job isn't just money. A job is something that you love doing. It's pride. It's your place in a community. And I think we do need to think very carefully about ensuring that the vast majority of every society feels like they're contributing, like they have a purpose. That's going to be a tough transition. I think, for a long time, it's been very clear about... My grandfather worked as a steelworker. He would tell stories about people. Literally, molten steel would go on your arm. It's gone. But there's also a pride that it was tough work, and you were making good money. And it was dangerous, and you were manly and all this sort of stuff. So I think that we need to identify not only that more people are educated in such a way that they can partake and contribute to the higher-end things... I think defining the goal of AI or whatever it is, that's still a thing. AI is a little bit like a dog. It doesn't just make the right decision. You have to train it what the right decision is, and then it can take over and identify the patterns and everything. So I think that it's still up to us to figure out what good is and how most of society can take part in it and be proud of being part of society. If that's the case, we can end up in the Star Trek world where there's no money, but people, they love prestige and science and exploring and talking to each other. Otherwise, we're going to end up in some very, very unhappy society that's incredibly unequal not because people are greedy but because the laws of scale and physics are so stacked against them. I think about this in terms of Mark Zuckerberg. There's no way to build Facebook and have it scale, and there's one network. He's going to end up rich. That's just the function. You could tax him all you want. I don't think most people want taxes as a handout. I don't think redistribution is really the solution to that problem. You need to actually help other people say, "Here's the job that I'm going to be proud of now."

Patrick: [00:43:41] It's interesting that some of the... Zuckerberg's a great example, but there are elements of what's going on that just magnifies these inequality problems. And I have no idea what we're going to do about it because, on the one hand, if you are confiscatory and you redistribute... Let's say his tax rate's 98% or something like that. It's going to, in some ways, discourage risk-taking of the sort that he did, maybe, maybe not. We don't know. I think we know that he's a passionate guy, and maybe he would've done it no matter what. But it seems like such an interesting problem that, especially when you're thinking about your own career, when I think about my career or a loved one's career, what to tell them to focus on. And one of the litmus tests I always go for is: is there something that you don't like about your job that's repeatable that you have to do over and over again and you don't like it? A lot of things you'll do over and over again, you like them a lot. That's great. But if you have to do something over and over again and you don't like it, there's a very good chance that that thing is going to be cheaply handled by technology, and you should just get rid of it.

Dan: [00:44:46] There's a great business case for it: I'll pay for somebody else to do it. It should go away eventually. It's funny. So, on your way here, you asked people for questions, and somebody asked about my decision to come to Betterment. I was at Barclays Wealth for years. Granted, I had to wear a suit every day, but it was a stable job. I was very comfortable there. It was actually this issue of money versus happiness that made me come here. I was making... I don't know if you hear about these above $70,000 a year, happiness doesn't increase. I live in New York City. So maybe it's a little bit higher, but I was up around there, and I was spending 40, 50 hours a week in a job that was maybe a 5 out of 10 on my happiness scale. And here I have the opportunity to definitely take a huge pay cut but to jump to somewhere which possibly, not guaranteed at all, would be a 9 out of 10 or a 10 out of 10 on the happiness scale. So I really like strongly progressive tax rates. I actually wrote something about this a little while ago because I think they encourage really good behavior, and they're not actually going to stymie any kind of passion or creative impulse. If you're motivated by money, you're going to be pretty crap at that, anyway, unless it's something that's just making money like finance. If you're an entrepreneur or designer or something else, you are just motivated to get out of bed and do that thing, anyway. I think that you could tax me significantly more, and unless you start taxing the enjoyment that I get from coming in and doing something that I feel is good and that's intellectually challenging, you're not going to stop me from doing it.

Betterment and Moats

Patrick: [00:46:23] To that end, what is the current area that you're working on here at Betterment that has you most interested, that has you jumping out of bed? What problem is it that you've identified that you're trying to solve for clients?

Dan: [00:46:36] To your point, earlier I touched on a large number of things, and that diversity of projects is probably the overarching thing. How do we personalize and tailor portfolios, investment solutions, advice not only to different people but different people who are different in a circumstantial way? Betterment announced a deal with Uber drivers where they would subsidize their retirement account, and they really strongly encouraged Uber drivers to take up an investment account with us, an IRA. We also have Betterment for Business, the 401k line of business. Again, on the retail side, you're predominantly getting people who have chosen you. Either they have a recommendation for a friend, that's our highest acquisition channel, or they've done the research themselves and they really believe in what we're doing. So that self-selection's really worked for us very well. In the Betterment for Business line, you have people who...their employer chose you. So, dealing with that slightly different population and understanding how they're going to approach you is kind of an interesting change. All through this, people want the right level of control. We're very aware that people want to, "Oh, you have too much international," or, "Can I do socially-responsible investing with you?" Right now, we're a little bit Henry Ford saying, "You can have whatever color car you want as long as it's black." It's not that we're like, "Black is definitively the second derivative optimal, there's no way that anything better." We're just like, "It's a little bit hard to flex infinitely to accommodate everything that people want." So I think that idea of: 'what kind of variation in advice do you need, genuinely?" I live in New York City. Maybe I should be in a New York municipal bond fund. That's a form of personalization that an advisor would recommend. On the other hand, I want some good pharma stocks. I don't know so much about that. That's more speculative. So that kind of tension between personalization meaning good personalization and personalization meaning speculative personalization is one of the kinds of things that we're thinking about here, and also expanding the breadth of our services into different areas, so state retirement plans that are sort of subsidized that people, that's their best guess for it. It's kind of interesting. I did my health care enrollment recently and ended up really going deep into HSAs and looking at this and being like, "Well, for me, I'm a healthy, young guy. This is basically a triple-tax-free investment account that I can put money into. Maybe I should be focused on doing stuff like that." The last thing that's kind of interesting is, coming back to the idea of costs just keep coming down, what's the future of funds and of Betterment? If we keep growing at the rate that we're growing, we are going to become very big in the fund marketplace. And scale brings problems of its own. But if somebody really wants a socially-responsible investment portfolio, that means different things to everybody. The only real way to do that is to actually cut out the middleman and just start designing indices that are different for every individual person. Then we happen to have an omnibus structure that allows us to trade fractional shares at zero cost. We could basically build personalized indices for every goal that you happen to have, whatever it may happen to be, and manage them very cost-effectively and tax-efficiently.

Patrick: [00:50:04] Meaning get around the fund structure entirely, the ETF and mutual funds structure, and be effectively building security-by-security indexes for each person. Is that what you're saying?

Dan: [00:50:14] Correct. Yep.

Patrick: [00:50:15] Yeah, it seems like if costs come down enough, and obviously there are still tax advantages embedded in the ETF structure. We don't know if they'll last or what they'll look like in the future, but the ability, through any kind of transaction, to kind of step your basis up through the time when you trade is a huge advantage.

Dan: [00:50:32] Definitely.

Patrick: [00:50:35] I'm fascinated, by this idea that, in the future, it will just be individuals who directly own individual securities via Betterment or some other platform, and it's completely custom-tailored. So, in that world, how much would you really let people tailor it, given your investment beliefs and philosophy that some form of efficient markets, markets are efficient enough, people shouldn't design a biotech index for themselves? So how do you think about that? Where would the line be?

Dan: [00:51:08] The general principle is: don't let people hurt themselves really badly, I think. Definitely, again, it's not prohibiting because, again, you're just going to go do it somewhere else. If we say you absolutely can't, you go, "Okay. Well, Robinhood will let me. I'm going to go do that." So it has to be some modicum of, "We're going to let you do enough that you feel you're getting that juice. We're going to let you overweight biotech, but you're not allowed to have more than, I don't know, 20% of your portfolio in it. I don't know where those numbers or lines are specifically, but the point is to try and flex as much as possible. I think there will eventually be some people who really fundamentally, "I just want to trade currencies." Those exist. That is not going to ever be Betterment's competitive advantage. There are people who have been in the DIY brokerage world for a lot longer who are a lot better at that. That's not novel or innovative. That's not going to actually give us incredible growth or market share. We keep looking at: what are the things that are not zero-sum trading things? Where can we expand our services that are going to help more people who never had access to that kind of service before? We recently launched an asset location service, so something that used to have, at least to me, a very complicated thing that generally an advisor had to look after and run tax numbers on. I know I've looked at the tools that some of our competitors have, and it was like, if you were an independent financial advisor, you had to enter all of this tax information yourself. It was up to you to say what the after-tax return of one asset class was versus another. So I really like, for better or for worse, I don't know what you'd call this, the TV model of innovation, where there are some things that are genuinely novel that we can do. But a lot of what we're doing is saying, "Here's something that we know is good, and we're just going to take it and productionalize it so that it's low cost for everybody." But when that happens, it's like a light bulb going on. We say, "Everybody gets tax-loss harvesting. Boom. Do you have $1,000? You can get tax-loss harvesting. Do you have an IRA and a 401k and a taxable account? You can have an asset location." I think that commoditization of the doing of the algorithm rather than coming up with the algorithm. We're going to just keep pushing on that across the board.

Patrick: [00:53:18] So that last philosophy of just give it to everyone may be one of the answers to this question. But if you think about all the different variables and dimensions of companies like yours, I'm curious in which area you think Betterment's moat, so to speak, is widest, meaning: what aspect of... And obviously, if we do this again in a year, there's going to be so much different about this space. It's just growing so quickly, and the technology is changing so fast: asset location, tax-loss harvesting, customization, all these things. But right now, and hopefully, this will persist into the future, where do you think your moat is widest?

Dan: [00:53:56] I'm going to share something with you that's almost inside information just because nobody else knows about it. It's the people who work here. So I joined. I was about employee number 21, and after working there for a few months, I was like, "This is just a crazy group of people." The people who do coding are really good. The people who do design are really good. So there's a little bit of like, "Oh, gee, how are we going to scale a company to do this stuff?" Betterment's now 220 people, and I have the exact same feeling every day. One of our lawyers who's an incredibly deep ERISA lawyer guy. He actually helped design the DOL rule. He gave them some feedback on it. He's a great guy. He loves his job. He's going to keep pushing on it. The software engineers that I know here, again, they're so good that they go in and they say, "This could be better. We're going to rip out this entire piece of software, rewrite it in a new language because it'll perform so much better." About two weeks ago, one of our engineers, I don't know, figured out some way how to use AWS better that reduced our costs by 50%. He was just like, "We can do better than this. We can keep doing it." And that's pervasive. Our customer service team is just amazing. They genuinely are really trying to help people. I think recently Consumer Reports put us, Vanguard, and one other person as the best customer service in financial services. So I don't know the people at any of these other companies. It's not necessarily that, but I have been struck by the fact that everybody here is passionate, works really hard, loves what they do, and works well with each other. And there's very little politics and very little bullshit. That's hard to compete against. If there's one person who you don't want to compete against, it's somebody who enjoys what they do more than you and does it with people that they enjoy doing it with.

If there's one person who you don't want to compete against, it's somebody who enjoys what they do more than you and does it with people that they enjoy doing it with.

Patrick: [00:55:42] That last point's a really good one. I'm curious, how does that scale? You were employee 21. You're at 10 times that now. Beyond the, I think, great but for obvious characteristics of passion, love what they're doing, work hard, are there other things that you are screening for as you look for new people that maybe people out there could think about for themselves for their own businesses that aren't just that kind of platitude-level stuff?

Dan: [00:56:12] Yeah. We recently went through this, and the things that we definitely screen for, number one, is horsepower. Number two is, I don't exactly know how to put it, but some kind of learning flexibility, openness. One of the questions I used to ask during interviews was: tell me about a time where you failed, and just tell me what happened. The question is actually to see how comfortable they are with admitting that they failed about something and what they learned from it and how they moved on from it. I'm not trying to make you feel bad, but failure is part of what happens. It's more important that you're really okay with it and that you can talk about how you learn from it. So that openness or comfort learning and being wrong is one of the things that I think we definitely select for, and I think just a desire to do well, to make something better. Every once in a while, I get asked one of these, like, "Where do you think the S&P's going to be?" blah, blah, blah. And I'm like, "I have no idea." But I keep coming back to... It's a quote from Abraham Lincoln, "The best way to predict the future is to build it." That's how I can see things happening, is that we have people coming who are like, "Okay, this is underserved, and there's a way we can do this better. I'm going to go in, and I'm pushing for us to make it better in this marketing way." One other thing that I think has done very well is that, and this is... I give huge credit to Jon Stein and a lot of the management team here and the board of directors. This is a very open, friendly place to work. If I had been in the elevator with the CEO at my previous company, I probably would've tried to figure out how to not talk to him because I would've felt like you could only do something wrong. Here, the CEO, whoever it is, you can joke with them. They're perfectly friendly people. You can bring ideas up with them, and it really does feel like a very horizontal culture where you can speak up and have your ideas considered. And it feels like a good family. We recently had not only Thanksgiving here but also our holiday party. The holiday party was much less tame than the Thanksgiving party. But the Thanksgiving party, we had seven babies this year. We're kind of all growing up together and supporting each other not just in the work but in our actual lives. And that comfort and close-knit-ness is something that I haven't felt anywhere else that I've ever worked.

Patrick: [00:58:45] Tell me about your most individual memorable day here.

Dan: [00:58:49] Oof. It's not a good indicator if you can't remember one, huh?

Patrick: [00:58:53] Take your time.

Dan: [00:58:55] Actually, I think it was when we moved into this building. I might we might've been around 40 or 50 people at that time, and we needed to move from our old office space to our new office space. I didn't know what to expect. Are we going to hire a moving company or whatever? They did part of it, and then, I don't know, a week or two in advance, they were like, "Everybody's got to be here on Saturday. We got to put desks together and put chairs together." So you rock in on a Saturday, and there are pizza and beers, and people are putting together chairs and doing all this stuff. And there's this moment of like, "I want to be here. I want to continue to make this for a while. There's a community, and I like these people. They're good people, and we have the ability to genuinely build something unique or new that you're going to be proud of." And it wasn't... We were literally putting together chairs, but it was still one of my fondest memories of realizing this was a company that I wanted to be at for a while.

Patrick: [01:00:00] Then my favorite question and one that I ask every single person, which is: what is the kindest thing that anyone's ever done for you?

Dan: [01:00:07] I'm going to admit something. I prepared for this one.

Patrick: [01:00:11] Yeah, I know. I'm going to need to switch it up. People know my tactics.

Dan: [01:00:15] There are two that are definitely true. The first one, which is a little bit obvious but it's worth saying, is my wife. My wife's British. We met in London, and she not only said it was okay for us... She was a diplomat. She was able to move around the world a little bit easier than a lot of people were, but she said, "Yeah, let's move to New York. Let's do that for a few years. This is about your career. I'm comfortable doing that for a little while." Then, when I asked after a year... We moved to New York. A year later, I'm like, "Hey, could I take a giant pay cut and go do this startup thing with some kids?" Once again, we thought through it. We really worked on it. My wife and I have the best arguments in the world. They're genuinely really good. It's one of the best things I love about our relationship, is that we don't agree on everything, but when we disagree, I come out of it feeling like we made a better decision because we disagreed on it. So definitely her for giving me some leeway to try and do that. To be clear, I'm absolutely going to pay her back for that at some point in time when it's like, "Now it's your turn. Where do you want to go? What do you want to do? I'm the trailing spouse. Let's go do your dreams." Then the other one that actually, I don't think it would've been thought about as a kindness at the time, but it's really important to me. When I was in sixth or seventh grade, for some reason, there was a long period where I didn't have any good way of getting home after school. And there was this math teacher of mine who would give me rides home from school, and I really appreciated it. He was a good guy, and he would make sure, on rainy days, that I was able to get home rather than walk. One day, I hadn't done my homework, and I realized it at the beginning of the class. I did it through the course of the class, and I handed it in at the end of the class, saying, "Oh, I forgot it. It was in my backpack," or something. And he got really upset with me. He said, "I saw you do your homework through this." And he said, "I'm just really disappointed. I've done good things for you, and you've disappointed me by lying to me now." That was one of those moments where it just hits you. It's not even shaming. It's just like, "I need to be a better person." And that moment, in terms of doing what's right, being honest about it, that's stuck with me for a long time. I don't think he was trying to be kind at the time, but that line in the sand, it's been very good for me.

Advice To Young People Interested In Markets

Patrick: [01:02:44] That's amazing. Usually, I close on that one, but I thought of another one, which I think you'll have a good answer to, which is, for younger people out there, whether it's coming out of college or a graduate degree program, that are really interested in markets, one of the things that I'm struck by is... and I've started teaching classes, and I really enjoy meeting students... that there are fewer and fewer people, for the obvious reasons, that are interested in things like asset management. But for someone out there that is passionate about the market, just wants to get into this world, what advice would you have for them? Given the secular trends that we have towards index and low cost and away from security selection, things like that, what would you tell someone that came to you and said, "Where should I go? Tell me what to do, and I'll do it"?

Dan: [01:03:34] I would say always double down on whatever is most interesting to you. I remember a game theory professor in college saying this thing, "Never compete against somebody who enjoys what they're doing more than you because they'll always beat you. You want to be on the opposite side of that." You talk about how much you read. I think read on whatever's most interesting to you. If a book is boring you, you don't have to feel like it's virtuous because whatever that is, it's going to interest somebody else, and they're going to be really good at it. You put it down and go off into something else that really interests you. One of the ways I kind of... I had no idea I would end up working in finance at all. If you had asked my parents, they would've laughed at the idea that I would be where I am today. And I got into it very back-ended, through psychology and then behavioral economics in psychology. It was a big letdown to me to realize that I was born in 1981, all the papers that I thought would be groundbreaking had been written in the 1970s by Danny Kahneman and so on, and being like, "Okay. Maybe I'm going to need to switch up my career path here a little bit." But then really got into learning about the social construct of markets and the history of them, like the capital ideas and Against the Gods books that talk about the history of financial markets.

Patrick: [01:04:47] Right, the risk is only a few hundred years old as an idea, right? Crazy stuff, yeah.

Dan: [01:04:49] Absolutely. If that stuff, if you're reading about it, and you churn through those books, whatever books you're just burning through, just keep looking at that and trying to figure out how you can work in that area. If you had asked, I don't know, 10 years ago, "Can somebody have a career in behavioral finance at a robot-advisor?" people would've gone, "What? What? What? What? What?" So that's the only other thing, is it is worth taking shots that you think... things don't exist but should. My job, it's not like Jon Stein came out and said, "What I need is a Director of Behavioral Finance." I was a Betterment customer. I kept looking and being like, "This could be better. This could be better. We can do more here." And I sent him an email with six different things he could do to improve things, and he said, "Let's have a coffee." People and things like that are just way too risk-averse. If there is something that you love and you want to see in the world, take the shot of learning a ton about it and finding other people who can work with you to make it your job. You'll be happy. You'll be making something that you love. It's the best setup ever.

Patrick: [01:05:55] Probably my favorite line of the whole thing, "Double down on what's interesting." So we'll end there. It was a blast getting to know you the last couple of weeks and months. I really appreciate your time today.

Dan: [01:06:04] It's been fantastic. Thank you.

Created by @ekmokaya