Episode 275: Live from Future Proof 2023: Decoding Financial Decision-Making with Hal Hershfield

October 19, 2023 Angelica Montagano

In this episode, we welcome back Hal Hershfield, Associate Professor of Marketing and Behavioral Decision Making at UCLA Anderson School of Management. Hal is renowned for his pioneering work in understanding how individuals make financial decisions, and he shares invaluable insights that can help us navigate the complexities of financial planning. In our conversation, live from Future Proof, we explore the intersection of behavioural economics, financial decision-making, and the potential for AI to enhance financial advisory services through the lens of Hal’s latest research findings. We explore framing insurance decisions, the impact of generative AI on financial choices, and the often-overlooked realm of end-of-life decisions. Discover why the key to success lies in understanding different consumer segments, how advisors can optimize the frequency of client meetings, and how clients and advisors should be working together. We also unpack the importance of personalized decisions, the value of a decision-making journal, the framework for making the right financial choice, and much more. Tune in to gain valuable insights into behavioural economics, consumer preferences, and the evolving financial planning landscape with Hal Hershfield!

See this content in the original post

Key Points From This Episode:

(0:02:41) Hal shares his motivation for writing the paper and why the topic of financial decision-making is so vital to understand.

(0:04:28) An overview of our current understanding of financial decision-making and interesting findings from the latest work on the subject.

(0:09:00) How to leverage the current knowledge of financial decision-making to your benefit.

(0:10:27) Opportunities for the industry to improve, both in academia and industry.

(0:15:09) Characterizing the framework for conceptualizing financial decisions, from decision-making to the consequences.

(0:18:13) The biggest gaps and opportunities for future research and the value of writing and maintaining a decision journal.

(0:22:33) The potential of AI to influence financial decision-making, and an example of an exciting use-case.

(0:26:31) Exploring the role of human financial advisors in an AI-dominated world.

(0:29:56) Insights into the steps for a client and advisory firm to work together effectively.

(0:34:07) What area of research in behavioural finance excites Hal the most.

(0:36:23) Bridging the gap between industry and academia.

Read The Transcript:

Ben Felix: This is the Rational Reminder Podcast, a weekly reality check on sensible investing and financial decision-making from two Canadians. We are hosted by me, Benjamin Felix and Cameron Passmore, Portfolio Managers at PWL Capital.

Nick Magiulli: We have a great podcast up today. One of my favourites. I love being on earlier this year. It's the Rational Reminder Podcast with Ben Felix, Cameron Passmore, and their guest, Hal Hershfield.

Cameron Passmore: Everybody, good morning. Thanks for coming out. Super cool to be here. I want to thank John and the whole team and Advisor Circle for impeccable organization, and for inviting us. We're super excited to be here. Got a bunch of Canadians. Great to see you all here. Hope everybody got their bucket hat. So, we're going to try to do a clean recording here with Ben's traditional introduction, as you know, we’ll do a clean cut. Hopefully, it goes clean all the way through. How about we go with that, Ben? Want to start the intro, then we'll kick it off?

Ben Felix: Yes. Sounds good. All right. This is the Rational Reminder Podcast, a weekly reality check on sensible investing and financial decision-making from two Canadians. We are hosted by me, Benjamin Felix and Cameron Passmore, Portfolio Managers at PWL Capital.

Cameron Passmore: Welcome to Episode 273. This is a special event for us. We are live at Future Proof which is an incredible festival of professionals in the wealth management industry, down here in Huntington Beach. It is an unbelievable venue. So, we've never done a live recording before in front of an audience. And someone commented on Twitter this morning, Ben, you and I haven't recorded together since the pandemic. So, since March of 2020.

Ben Felix: Yes, that’s right.

Cameron Passmore: As listeners know, we've announced this for a while now. We have super special guest joining us this morning, third-time guest, Hal.

Hal Hershfield: Yes, that’s right. Third time.

Cameron Passmore: So, Hal Hershfield is with us. Hal joined us in 256 earlier this summer after his fabulous book, Your Future Self was released. And he was also here in Episode 141.

Ben Felix: Yes. Hal is a Professor of Marketing Behavioural Decision Making and Psychology at UCLA. His research sits at the intersection of psychology and economics, and examines the way that we can improve our long-term decisions. Hal got his Ph.D. from Stanford University. He published in all the top academic journals, and he has just written a review paper that I don't think many people have seen

Hal Hershfield: Yes. It was really just the authors and you guys, at this point.

Ben Felix: Yes. So, we've got some very fresh research that we're going to talk about with Hal today on consumer financial decision-making. So, with that, I think we can go ahead and start talking about this research.

***

Ben Felix: Hal, in your review paper, how do you define financial decision-making?

Hal Hershfield: First off, thanks so much for having me on again. I feel very lucky to be here again.

Cameron Passmore: Well, you're a good friend of the firm. A good of ours. It's awesome to have you.

Hal Hershfield: I appreciate that.

Cameron Passmore: And the commute wasn't too long.

Hal Hershfield: That's true. So, this is a paper that I wrote with Abby Sussman at Booth and Oded Netzer at Columbia. We were editors on this special issue on Consumer Financial Decision-Making, which I know sounds super exciting to everybody. But part of the issue, we decided, let’s write a review paper about financial decision-making. We said, “Well, we first probably should define what that means” which was a lot harder than we realized, going into it. Abby was the one who really nailed down the definition. I think it's something like the accumulation and use of resources over time, as reflected in behaviour and choices. That really encompasses a lot there. But the earlier definitions were really broad, didn't really tell you what it was about, and so we had to nail it down to something.

Cameron Passmore: So, why is the study of consumer and financial decision-making important? I mean, this podcast is all about decision-making.

Hal Hershfield: Yes. I mean, to some extent, we're preaching to the choir here. I mean, I think you guys know this. But in the academic world, I think people are starting to realize, look, money is at the heart of so many other decisions, and so many other outcomes. It touches on so many things. To be able to understand what are the antecedents of how people make decisions? What are the moderators? What are the influences on them? What are the outcomes? How do people think about these choices? And how do they change? And how can we change them? I mean, right away, you realize how important it is and how crucial it is to know this, not just for the sake of money itself, but for everything else.

Ben Felix: So, in the paper, you look at where we've been in this field of research and where we're going. Can you talk about the main topics that have been studied?

Hal Hershfield: Sure. So, what we did was first, gather all the articles we could. We had several research assistants that comb through the journals. We actually use ChatGPT to help us figure out what's actually a financial decision-making article, which was hard in itself because we had to define it. It didn't count some of Abby's articles. Then, we realized, “Maybe we need to redefine the definition here”, so we ran what's called a topic modelling analysis.

It's basically using AI to say okay, what are the different buckets or machine learning? What are the different buckets here? It comes down to five buckets, really. It's saving and spending. It's how do people make trade-offs between money and other things. It also looks at budgeting decisions. We also look at things like negative outcomes, fraud, and economic inequality, and whatnot. Then, the final one was really like looking at biases, which, of course, is something that’s been studied a lot in behavioural economics,

Cameron Passmore: What are some of the most interesting findings that you guys came up with?

Hal Hershfield: In looking through the papers, we've all got our special preferences here. One of the things to note is that there's been so much time spent on saving and spending, and not as much on a lot of these other topics. So, I think academics know a lot about what makes people save or not. But then, all these other topics are really kind of open. I'll tell you some of my favourite newer findings from the last couple of years.

Actually, one of the papers that was in this special journal, it's by Scott Rick and Jenny Olson, and it was about financial decision-making within couples. I think this is an area that on the academic side, we study so much of these individual decisions, and yet, I think everybody in the advisory world knows, rarely are these decisions made as an individual. So, one of the things that Scott and Jenny looked at was, what's the interplay? Or what matters more for couples’ decisions? How much people actually know, or how much people think they know. It turns out what matters more, if you look at the stats, what actually has a greater impact on decisions is the partner that thinks they know the most, and that may be related to how much they actually know. But it has a unique influence, more so than what they objectively know, is what they think they know.

Cameron Passmore: Okay. So, you're sitting down with a couple of dinners some night, what would be your advice to them on that specific topic?

Hal Hershfield: This is actually a really good question, because part of the big picture here is, well, how do you translate this over? I think making this explicit. When I say this, I mean, making the conversation about what do you actually know and what do you think you know? Explicit is important, because it turns out with financial decision-making with couples, oftentimes, one partner ends up doing the decisions, not because there was an explicit conversation where someone says, “I'll do this.” But they just start doing it.

Cameron Passmore: So, you're basically saying, validating, if you know what you think you know.

Hal Hershfield: I would say like almost auditing it. Look, there are tests you can do this with. But I mean – and this is tricky because you don't want to embarrass somebody.

Cameron Passmore: We’re all busy, right? Because so many people, “Okay, you do finance. I’ll take care of something else.” So, it gets done, but you never often don't take the time to audit whether you're competent at that.

Hal Hershfield: Absolutely. And you could imagine that you could maximize. You could do better, right? Just because of the status quo or inertia, it doesn't mean that's how it has to be within a couple.

Cameron Passmore: That is so fascinating.

Hal Hershfield: Yes, so I think that research, and I should plug it. Scott Rick’s got a book coming out in January about financial decision-making in couples. And I think that research is just so interesting. Jenny Olson’s really one of the forefronts of it too.

One of the other areas that I'm really interested in is work done by Wendy De La Rosa at Wharton and Steph Tully down at USC. You know that I really like it, if I'm going to plug somebody over at USC. She was also my former student. They do great work. And one of the things that they've looked at is how payment frequency impacts spending decisions. You think about the model of advising, the model of studying spending, so much of it's based on this antiquated payment system of you get paid monthly, or bimonthly, or whatever it may be. Now, there is so much of a rise of contract workers, and I know that's not necessarily always the people who are getting financial advice, but they make up a huge part of the economy.

Now, payments are much more frequent. And one of the things that they're finding is that people who get paid more frequently spend more frequently, in part, and that mechanism there is that if I get money more often, I feel that money is more certain, and I have greater subjective wealth. I think I have more money to spend, that’s why I spend it more, which can have negative impacts.

Ben Felix: That is fascinating. So, when you take interesting findings like that, how do you think decision-makers or their financial advisors can be using that information to improve their decision quality?

Hal Hershfield: To me, so much of this is about disseminating the research out there on the advisor side, and I'm sure we'll talk about it. There's a difference between disseminating information and actually operating on it and acting on it. But I just highlighted a couple of findings. There's a lot of other ones I'm happy to go into. But some of this stuff, if you think about acting on it, well, how often do we ask the question of how often are you paid? And how often do we ask the question, subjective perceptions of wealth. Because I think it's really easy to look on paper and look at the numbers and look at the objectives. But what about the subjective?

Ben Felix: That's really interesting. So, there's an element of just knowing what questions to be asking clients.

Hal Hershfield: Absolutely. What more might you uncover if you start on this subjective side? In addition, I'm not saying replace the standard objective metrics, but let's add on some subjective parts too.

Cameron Passmore: But can you imagine being paid annually, for example? And to divvy up your annual salary?

Hal Hershfield: That's wild. I mean, well, look, you know what, to some extent, when I was a grad student, I mean, I think it was like blocked into semesters or quarters. But I had to be really forward-thinking because it wasn't that much money to then eke out over the coming months.

Cameron Passmore: Okay. So, we're here a Future Proof down in Huntington Beach. There's 3,000 other financial professionals here. This theme of helping people make better decisions and have a better experience is certainly top of mind down here. So, where do you think the industry is lagging in this area?

Hal Hershfield: I have to be careful, right? I don't want to say like, “Here's where you're lagging.” But okay, if I were to actually think about that, one area where I think industry lags academia, is the use of experimentation. So, academics, I think, really know how to run a randomized controlled trial. They know how to do what industry calls AB tests, and how to isolate things and randomize. The reason that we do that so that we can get clean insights and know what worked and why it worked. I totally recognize. Sometimes in industry, you don't necessarily need to know why something worked. You just want to know that it works, right? But then there's a downside of not knowing if it really works, but thinking you know that it works just based on intuition, right?

So, I think, one space where academia can really inform industry is to say, “Well, how can we go about testing better?” That also is hard sometimes, especially if you have maybe have a lot of money under management, but not as many clients. So, it's hard to do clean tests with not a lot of people. And also, sometimes we may not want to test, because there's some risk involved there. But it's short-term risk for long-term gain.

Cameron Passmore: What kind of test might we do?

Hal Hershfield: You're supposed to figure that out, Cameron. I mean, the way I see things, every aspect is testable. Down from maybe on the most surface level, the marketing angle, and even there, I would say, let's move away from the one size fits all, but pay attention to nuance. What messages work better for different clients across the lifespan and so on? But then also, could you play around with the way you bring up certain topics? Is there one way to approach a topic and another way, and let's see what works better? Also, by the way, this is where I'm coming at it, observing the industry, but not being in it. So, if you all say to me, “Well, nice idea, but it's not going to work.” That's fine. I'll take that criticism. But I'd love to see more of that actually happen at firms, and maybe some firms are doing that.

Cameron Passmore: You're talking about taking our client experience, documenting how clients react to different prompts or whatever, and then basically using that to do research.

Hal Hershfield: Yes. I mean, look, if I can get some research out of it, that would be cool, too. But if it also makes an impact – I mean, I should flip that around. First and foremost, on your side, I think you wanted to make an impact. If there's research that comes out of it, or “thought leadership”, that's also great.

Ben Felix: What about the flip side? Where do you think academia is lagging, what's happening in industry?

Hal Hershfield: To me, one of the big things that really stands out is academic study, these more frequent decisions, spending decisions. Sometimes the saving decision isn't as frequent, but it's, are you going to contribute or not? What's your portfolio allocation? What we have very little purchase on is big, transformative decisions. How do consumers and clients make decisions about education? How do they make it about mortgages and how much house to buy and end-of-life decisions and all these big ones? I think academics can only speculate about these things. And advisors have a lot – you have all the experience there, and I don't see as much conversation between the academic side on the advisory side on how these sorts of big transformative decisions actually play out, and I would love to know more about that.

Cameron Passmore: Do you have any thoughts on how we as practitioners can get investors, the public, just to think more about the decision, as opposed to being on autopilot?

Hal Hershfield: It's funny, because the first question I'd want to know is, do people want to think about the decision? So, I don't know. This is where I approach a lot of these research questions is like, just because we want to do it, what do people actually want to do? Then, you go back to what I was saying earlier about the one-size-fits-all, versus the more segmented approach and my bet, and I would put money on it, is that there's clients who would love to think more about the decision and others who would say, “There's so many other things I want to think about. You think about the decision. Put me on autopilot.”

Cameron Passmore: But I'm saying the decision to work with someone to help you make better decisions just to get out of the inertia of not making a decision about their financial future.

Hal Hershfield: Sorry, I misunderstood, which was the decision –

Cameron Passmore: The initial decision to get off of autopilot and work with an independent advisor.

Hal Hershfield: This is a question that I'm deeply curious about. I always think about the advisory world, like the medical field where nobody questions the need to go to a doctor. There's somebody who thinks about health all the time. They have some ideas about what may be best for you. Well, I guess some people question it. But the majority would say, “Yes, that's where I would go to for my advice.” Then, the financial advisory world is very different, where I'd say, “Well, I can kind of do this. I know this and I can talk to my butcher or whoever it is.” So, that's not an answer, more to say, it's a question I would –

Ben Felix: Love to know.

Hal Hershfield: – love to know the answer to. Yes.

Ben Felix: Can you describe from the paper the framework that you guys came up with for conceptualizing financial decisions?

Hal Hershfield: Yes. So again, we are taping this live. I want to be sure that I don't bore anybody but talking about an academic framework. But I'll do it really quickly. The way we think about a financial decision, and Abby, Oded, and I went back and forth, and back and forth on this. But if you had the history of the PowerPoints that we did, it would look ridiculous. But essentially, we boiled it down to there's the consumer. There's the decision itself, and there's the overall context. Then, you say all these things change over the lifespan, from childhood to adolescence, to young adulthood, to middle age, and toward the, I’ll call it, not the end of life, the latter third if you will. But part of doing that allowed us to then say, “Let's group the research that's been done and let's figure out where we need to go there.” But I particularly like that sort of bucketing of the financial decision-making space.

Cameron Passmore: So, within this framework, Hal, where do you think financial advice is most useful?

Hal Hershfield: I would have to think it's at the decision itself, right? But it this is a tricky question, because if an advisor would have just pay attention to the decision itself, the aspects of the decision, how consumers think about it, and this entails literacy, and how people understand interest, and so on, and so on. But if an advisor were to just focus on that, and miss the consumer side and the context side, then it's not that important anymore, because they've missed sort of the big picture. So, I mean, I'd say, start with the consumer, and maybe people know this in the industry. But I think recognizing the interplay between all three parts is what probably is most important there.

Cameron Passmore: Do you think the industry in simple terms should go further up the food chain, like away from the product, away from the retirement projection, and go more upstream into the decision process, just in general?

Hal Hershfield: So, my take, and I should be careful to say this. This is just my take, and it's not based on – there's no empirical research I could point to about this, and I would love to sort of do it is to say, I'm not sure they need to go away from a product. But my take is that the product itself is table stakes. Everybody assumes that. But then to go I mean, I would actually say deeper upstream and deeper downstream. My guess is that that's what gets an edge from the advisor perspective for certain clients, right? I assume there's somebody who goes there and says, “I just want to talk to you about the decision itself and the financial aspects.”

Ben Felix: So, what do you mean by downstream?

Hal Hershfield: I would say downstream is the consequences that arise from the decisions. So, call it the more holistic approach. And researchers have been talking about how do we define a wellbeing? What do we mean by that, purpose, happiness, meaning? The typical topics that are brought up in business schools, right? But at the end of the day, I would argue everybody in the advisory world eventually cares about that, because that's what money is leading to, but to have a better purchase on that, and understand those connections from the decision downstream, I think could actually deepen the knowledge that we have about the decision itself and help clients make better decisions.

Ben Felix: So, knowing what's in the existing literature on this topic, where do you think are the biggest gaps and opportunities for future work?

Hal Hershfield: We need to know a lot more about decisions across time. A lot of the research is kind of focused on the middle of the lifespan, and there is some academic work on financial decision-making as people get older, but it's not that deep. Of course, you look at what's the average age of clients, and it's older than what academics are studying, right? So, I think we need to know a lot more about that space. I also think that we could really deepen our knowledge about budgeting.

This sounds funny, right? Because I think academics think about budgeting and they think about literally making a budget and what that entails and who does it, and how strict should it be, and how flexible should it be. I've been talking to advisors, you say, “That's not really how we think about a budget” right? So, Abby has done some incredible work on budgets, along with Chuck Howard, in case folks are interested. But to know what's the actual use case of budgets, not just at the lower end of the income spectrum, but at the middle, and the sort of mass affluent end, I think that would be really useful to know. And then maybe, if I could give one more area for future work, it's how consumers think about insurance.

There's a product that we all have, we all think about on some level, and I don't think enough is done right now on that. In fact, I keep mentioning her, but Abby was so involved and so integral to this. Obviously, she and I have a call later today, the day of we're recording this, about what research we might be doing in this space of insurance, because I think there's not enough academic insight there. There's someone like earthquake insurance, but not on the financial side.

Ben Felix: Can you talk about what kind of questions you'd be wanting to answer on insurance?

Hal Hershfield: I mean, there's got to be some low-hanging fruit there. One of which is what's the best way to frame it, to get people to think about it in ways that make them more amenable to it when it's something that's right for them, when it's a good fit. I’d also want to understand what are the barriers, right? Where is somebody hesitant to jump in, especially if this is one side of the whole wealth equation? We spend so much time on the accumulation side. Researchers do at least, that this is the protection side, and I don't think we know as much about it. So, I'd want to know, well, how do I frame it better if it's something that they should be getting?

Cameron Passmore: How do you know if you're making the right decision around insurance? Are you properly trained? As opposed to the stigma, “I don't like life insurance.”

Hal Hershfield: Right. Exactly. I mean, so much of that world, of course, has been stigmatized by this smaller segment that might not be peddling products that are good for consumers. But then there's so much else that is good, right? But if consumers just hear about that one end, then that automatic visceral reaction. I mean, obviously, first and foremost, could be like, don't call it insurance. But eventually, you have to.

Cameron Passmore: You mentioned time. I'm curious, we once talked about having a decision journal for decisions you make in financial planning over time and going back and reviewing it. Do you see value in an exercise like that?

Hal Hershfield: Yes. I mean, part of the value of that would be taking some of the emotion away from past decisions, and looking into what I've done. The value to that if it's done frequently enough, and I think that I have to stress that, is that you can start to see where there are intersections between what are the contextual things that have been happening in my life, or your client's life that led to certain decisions or certain fears or whatnot, because then you can start to anticipate, “Look, the context is happening. How should I react to a given decision or not?”

Cameron Passmore: Oh, interesting. So, put more colour around the decision, so it’s not just the decision, but the environment in which you made that decision.

Hal Hershfield: I think that's right. Now, of course, this is going to be a function of how frequently some of the – we're talking abstractly here. But how frequently are the decisions being made? We've talked about this offline about budgeting, and even the wealthiest clients sometimes can use help there, right? And those can be frequent spending decisions. I might not think I'm spending as much as I am and then I'd want to look and say, “Well, what was leading to me going down the path of spending more than I meant to?” Or something like that.

Ben Felix: You talk in that paper about the potential impact of generative AI? How do you think that might affect the way people make financial decisions?

Hal Hershfield: It's a requirement to speak about AI right now.

Cameron Passmore: Here, this year, absolutely.

Hal Hershfield: I mean, look, you can look at this from the client side and the firm side, right? I'm really interested in this from the client side. So, one of the things we know from financial decision-making research, and I'm sure many of you, many folks have heard of this, maybe not. Almost 10 years ago, there's this great paper by Fernandez and Lynch and others. John Lynch is one of the –he was the senior author on it – looking at the impact of financial education. So, this depressing finding from that, and I know, there's a lot of debate about this, but my read on that paper is that the depressing finding is that financial education has very little to no impact on consumers, once you partial out the impact of self-selection. The people who want to learn, financial education may be meaningful, but they might have learned anyway.

The one case where it seems to really matter is what we call just-in-time education, right? So, if you're 20, and you're learning about mortgages, that's not going to matter. But if I'm trying to figure out interest rates, and I'm debating different products, that's when I really want to learn about it. Here's where I think Generative AI could be so useful. If you couple it with a machine learning approach of saying, “What's the digital footprint that’s leading to somebody wanting to make one of these X , Y , or Z decisions? Now, let's inject that with A, B, or C education.” And of course, you can think about the can of worms that this could open up. Who's providing that education? Where's it coming from? And generative AI, it's not brilliant, right? It's only based on what else is out there. But if you could have a version of it that's really smart, and also recognizes different segments, man, that can be really useful to a consumer, in the moment of needing to make a decision. That's at least my take on the consumer side. You guys probably have more to say about on the firm side.

Ben Felix: Well, just think about your book, Your Future Self. What could AI help if you want to be something and 20, 30 years, tell me what do I have to do now to get there?

Hal Hershfield: This is where I need to say, I finished the book before ChatGPT was a thing. Just in case you're wondering. But absolutely. I mean, to think about the possible future. I'm actually working with some folks at the MIT Media Lab, and one of the things, a new project we have is looking at possible future selves and showing a consumer what those might look like, and what that life would be like. This is really in the nascent stages. I can look at it now and say, imagine even in two or three years looking back on this, like how novice this is. But I can imagine really deepening the ability to talk to clients and consumers about what possible future paths would look like.

Ben Felix: Do you see a world where it could analyze the information that's out there, the research that's out there, but also analyze you and your profiles, perhaps your social media profiles, to learn about you and the research and come back with a future sell strategy?

Hal Hershfield: I mean, look, to some extent that's already being done. There's a former postdoc from UCLA, Poruz Khambatta, and he's done this great work using machine learning to say, “Let's identify what your preferences might be.” He did this in a really specific use case of looking at recommending articles to you. To some extend, I mean, this is what's done. When you go online, or Netflix or anything like that, it's recommending content to you based on what other people like you consume, right? So, there's no reason to think that we couldn't do the same thing for life paths, but we balk at it. Because we say, “Well, I'm unique. No one's like me.” But of course, there's a lot of people. I mean, you guys are literally wearing the same shirt. There's a lot of people like you, right?

So, when we start thinking about the use of that technology to make some predictions, I would put a bet that there might be certain consumers who would almost be more willing to take that advice than if somebody like an older person told me like, “Here's a good career path for you.” What do they know? Oh, well, computer? They know a lot more.

Ben Felix: That's very interesting. One of the obvious questions that I think follows from that, given our audience and where we are, is, what role do you see for human financial advisors in a world where machine learning and Generative AI are able to give some level of advice?

Hal Hershfield: That, just done, right? I mean, you know what this reminds me of, Ben, is go, what would it be? Ten, 11, I don't know, 12 years ago? I remember when you started to see the rise of robo-advisors. I remember having a very similar conversation on a stage and somebody says, “What role do financial advisors play in it?” I wasn't so sure then. And then only a few years went by when you realized it's not a role where you're replacing the advisor, but it's complimentary. I think you look at what some of the firms are doing, where you say, well, there's a role for the robo-advisor, and then there's a role for the financial advisor, and that goes from everything from explaining concepts to the emotional side, right?

Look, we could come back in two years, and I could be wildly wrong, and the emotional intelligence of AI could go through the roof and I really trust it. But I don't see that happening anytime soon. And I think the right advisor will know how to use the tools in conjunction with that emotional appreciation, or this focus on wellbeing, and put that combination into place. So, I, if anything, see it as almost sharpening, deepening, enhancing. Maybe I'm optimistic.

Ben Felix: I don't know. We've played with training an AI chatbot on our content, and then it'll spit out answers that have come from past podcasts interviews, or our written content. So, I think the ability to have the information or the content to train an AI is a big part of that whole thing.

Hal Hershfield: Well, I mean, that's right. It's only as good as the input into it. You guys, I can say this, I don't think I'm kissing up here. But you guys have good content. So, if you put that prior content in, it could help generate questions. Maybe you work off of that. Here's another good use case. One of my friends, Todd Rogers has this new book out. Cameron, I think we were talking about it, Writing for Busy Readers.

Cameron Passmore: It’s a great book.

Hal Hershfield: It's an amazing book, right? And it's just all research-back strategies of how to put communication out there so that people listen, which is like, half the battle. When I think about this, I think the longest emails I ever get are from students who don't have an appreciation of how long it takes me to read. I'm not saying, “Oh, I'm so important.” But it's like, I know that what they're trying to do is be respectful and write a really detailed message. And the most respectful thing that could do is make it really short. The book goes through all these strategies that Todd and his co-author have determined based on research. He actually trained like a conversational AI agent, on the principles from the book. And now I can pop in an email and say, “Make this better based on the book’s principles.”

Now, I told him, “Don't release that before people buy the book, because who's going to buy the book?” But it's great. I've played around with it, and that's the type of thing where I could say, “Man, how amazing is that?” Now, people actually listen to what I'm trying to say to them. Like, if only I could use that at home with my kids, like, just filter – okay, that was a dumb joke. But we'll keep going.

Cameron Passmore: I have a practical question for you.

Hal Hershfield: Okay. Thank you.

Cameron Passmore: So, what do you think is the best way for a client to work with an advisory firm? I'm looking for advice for both sides of this. In terms of cadence and meetings, relationship with technology, and enabled support. It strikes me that meeting once or twice a year might not be optimal for effective long-term decision-making.

Hal Hershfield: So, here's another case. I don't mean to keep coming back to this, but this would be another case where segments matter. One of the – I'll get to this question in a second. But one of the things that we focused on in our review paper was that so much of the research that's been done thus far, really has been on the broader consumer. There's a great new paper by Chris Bryan and David Yeager, where they say, “The heterogeneity revolution is coming”, which is a great title for a paper that basically says, there are different consumers out there. My MBAs hate it when I say, they asked me a question, I say, “It depends.” And I'll do that now a little bit, too. But like, here's a place where I'd love to know what the data would say on like, which clients do best with the once or twice-a-year meeting? Which clients do best with a monthly check-in, whether sometimes it's on email, or text, or whatever, and sometimes it's in person or whatnot. But I have to think there's colour there and individual differences.

Cameron Passmore: How could you find that out?

Hal Hershfield: So, what would be the experiment –

Cameron Passmore: Like, you’re coming as a client. You and your wife coming as a client. What can we do? Or what could the consumer do to understand what works best for them? As opposed to just being preference. I'm talking about outcome, success-driven decision.

Hal Hershfield: I would say, here's where behaviour would trump any stated preference. So, you could start by asking, “How much would you prefer that we contact you?” You could ask them that. But another version is, you could play around, and if I'm reaching out to have a meeting, and they're saying, “I'm good.” That probably tells you something. But if I'm reaching out to have a meeting, and there's contact in between those meetings, maybe that's a client that we want to ramp up the meetings for. I mean, this is like a, at the risk of being labelled a dumb insight, I would question how many people are actually looking at that, I don't know, if you want to call it AB testing or whatnot, to see. I mean, I think, I'm like introspecting.

When I get a call from my guy, every couple of months, and I often like, I shouldn't say this on air, but sometimes I dodge it, because I'm like, “I'm good right now.” I don't want to do the thing and look into this, and I know, I'm going to have to up my life insurance or something like that. But maybe it will actually be better for me to talk more. I don't know. But that's the behaviour, we're suggesting. Maybe I don't want to meet as much.

Ben Felix: I wonder if there are prompts that advisors can be using to identify when a client is making a transformative decision and engage then.

Hal Hershfield: That would be fantastic. Look, I mean, we know there's already a model for this, right? I mean, Target did this for years. I mean, they probably still do it. I'm sure in big companies do. The story about target using data to identify when someone is pregnant, right? I mean, they're doing it because they know that's a transformative decision. When do people make choices about switching brands, or adopting a new brand is when they go through that, when they become a parent, when they go through divorce, or marriage, or whatever it might be, and that kind of blew up. The story goes, they sent messages to a teenager about a new crib and diapers and the dad said, “Why are you sending this?” And they said, “Well, it's just an algorithm.” And then he called back and said, “Actually, the algorithm was right. She’s pregnant.”

So, I think, you don't have enough data at a firm, right? The big stores have reams and reams of data. And this is what we mean when we talk about big data analytics to then make some of those predictions, and they're never perfect, but they're saying better than chance, I can predict if you're going through a big change. I do wonder, though, if that's something that you guys can start picking up on, and maybe you do already, as well?

Ben Felix: Yes. We don't know. But I think getting connected to client’s bank accounts, or at least having insight into what's going on at a transactional level is probably a –

Hal Hershfield: I mean, if you told me – look, I'd be a skeptic at first, if I was a client and said, “Why do you need access to my credit card account?” But if you were to say to me, “Look, we're going to analyze it.” And now maybe you do have the data. If you have enough clients and enough credit card transactions over time, then I can start saying, “Well, what are some of the patterns that are happening here?” And even within client, you might say, “I noticed that you have a real ramp-up spending around April.” “Oh, is that right? I could have anticipated December. But April, interesting. Why is that? What's happening? Can we plan for that?” That's not a transformative jump or life decision. But it is still something that could be happening with some frequency or enough frequency to speak to it.

Cameron Passmore: So, what area of research are you most excited about now in this whole field?

Hal Hershfield: What am I most excited about right now? Some of the stuff that I'm starting to work — I mentioned the really early stage, looking into insurance. Some of the other stuff I'm really interested in, nobody ever says I'm really excited about the end of life. But I'm really excited about end-of-life decisions. How do consumers or clients think about wills? How do they think about advanced directives? What's the conversation? What's the messaging that may work best in those spaces?

A couple of years ago, some of my students and I partnered with UCLA Health, and we were trying to put out messages out there. This was just in pilot testing. Not with real healthcare patients. But just putting messages out there about, what might move the needle when it comes to advanced directives? And nothing was moving the needle. I think part of the thing there is that, that's a space where nobody is going to respond to like a quick message, because it's not had a quick decision. So, I like to better understand, what's the deeper conversation? How can I analyze like the ebb and flow of that conversation and what predicts whether someone's going to go ahead and make the will that they need to make, and also updated over time? Who does that? Can you tell? I'm getting a little excited about that. But that's not something we know a lot about right now, and I'd love to know more about it.

Cameron Passmore: What about decisions in there about how much to donate, for example, just like the foundation giving to kids?

Hal Hershfield: I mean, there's all sorts of questions there about sort of intergenerational connections, right? So, you said there's the better to give with a warm hand than a cold one. There's the, what do I leave to my kids? Or do I not want to leave to my kids? That's not just like a simple like, I like them or don't. There's research showing that millionaires who make their money are happier than millionaires who inherit it. I'm oversimplifying. It's not to say that, “Oh, if you want a happy kid, don't give them money.” Of course. But those are decisions that are not ones that we want to think about. People don't want to think about that stuff, because it involves so much uncertainty, and it involves dying, and conflict between siblings and kids, and all that stuff. But if we could try to get more of a framework, in a more systemized way of talking about it, and we were chatting about that earlier. I think there could be so much value there.

Ben Felix: We talked earlier about the connection between industry and academia. What do you think financial advisors should be most excited about in this area of research?

Hal Hershfield: Oh, yes. I would love them to be excited about sort of operationalizing concepts, and that's a fancy way to say, or not fancy, but maybe that's a too complicated way to say, “I think a lot of advisors know about the concepts.” Most advisors now know about system one and system two thinking and behavioural economics and Kahneman, Tversky, and all that stuff. But then think about how to put that into practice is something I would love more people to get excited about.

But then topic-wise, I mean, maybe it's because I just said it, but also, the end-of-life decisions, and decumulation decisions. That's not something we've talked about much, but I'm doing some work in that space. Suzanne Shu is one of the big proponents of that work over at Cornell. She was my former colleague at UCLA. I think we don't know that much about when people claim or we know about when they claim, but what are the thoughts that go into when people claim their social security benefits here in the US? Then, also decisions about how do I decumulate how much do I spend per year? And those old heuristics may not really do the – 4% heuristics may not do the –

Ben Felix: Do you have that? You've done that research?

Hal Hershfield: Yes. We have done some of that research and we have a paper that just came out, this is led by Adam Greenberg over at Bocconi University looking at what are some of the interventions that can impact people's likelihood to want to claim later? And what are the individual differences that impact those likelihood to claim? But that's all hypothetical, because we can't – the Social Security Administration didn't let us change the messages that got sent out to consumers, despite us asking them.

Ben Felix: What were the findings though? What influences how late people take their –

Hal Hershfield: There's a lot that was there, but one of the findings of that paper was to talk about the future spending power of the benefits you could get, and so, how much you have and how little you have can make an impact on wanting to claim later. Oh, that's just like a teaser there. But we tested like 13 different interventions, and we did it over and over to make sure what we were finding works.

Ben Felix: Really interesting.

Cameron Passmore: Super interesting. Hal, this has been awesome. I think you’re the perfect guest for this event. Live from Huntington Beach at Future Proof.

Hal Hershfield: Thanks so much guys.

Cameron Passmore: Thanks for coming. Thanks for listening.

Is there an error in the transcript? Let us know! Email us at info@rationalreminder.ca.

Be sure to add the episode number for reference.