bio: cepr.org
Stephen P. Utkus is the head of the Vanguard Center for Investor Research. The Center conducts and sponsors research on investor behavior and decision-making. Mr. Utkus’s personal research interests include retirement and aging, behavioral economics and household finance. Mr. Utkus received a Bachelor of Science degree from MIT and an MBA from The Wharton School of the University of Pennsylvania. He is a member of the advisory board of the Wharton Pension Research Council, a Visiting Scholar at The Wharton School of the University of Pennsylvania, a member of the advisory board of the Consumer Finance Institute at the Federal Reserve Bank of Philadelphia, and a member of the Board of Trustees of the Employee Benefit Research Institute in Washington, D.C.
What shapes investor beliefs and behaviours? Steve Utkus, a leading expert in investor behaviour and former Global Head of Investor and Retirement Research at Vanguard, brings decades of groundbreaking insights to this episode. Drawing on exclusive access to Vanguard’s anonymized client data and investor surveys, Steve uncovers the intricate links between what people believe and how they invest. In the first half, he reveals surprising findings from his research into investor beliefs and portfolio decisions. The second half dives into the impact of financial advisors, both human and robotic, on improving investor outcomes. Steve’s reflections, enriched by years of collaboration with academic leaders and personal conversations with Vanguard founder Jack Bogle, offer a rare window into the world of data-driven financial research. Join us today for this fascinating conversation as we unpack fresh perspectives on investor behaviour and the evolving role of financial advice!
Key Points From This Episode:
(0:00:20) The importance of understanding investor belief, Steve’s unique approach to studying it, and the benefits of using survey data.
(0:08:37) Understanding the effects of individual beliefs on portfolio equity shares.
(0:13:40) How equity sensitivity varies with things like trading frequency and how observed sensitivity compares with predictions of an asset pricing model.
(0:17:27) The variation of beliefs across different groups and the strong effect of being a pessimist, optimist, or having a middle-of-the-road perspective.
(0:21:29) Investor cash flow expectations, how it affects stock return expectations, and how it aligns with models of equilibrium.
(0:24:35) The impact of stock market disaster expectations on future stock returns and the effect of COVID-19 on investor expectations.
(0:33:37) ESG investing motives, portfolio impact, and the role of financial returns.
(0:38:35) Unpacking the impact of robo-advisors on previously DIY investors and who benefits.
(0:45:21) Pros and cons of human financial advisors: the needs they satisfy over robo-advisors.
(0:53:12) How unadvised investors' needs differ from those who get financial advice.
(0:54:04) What determines how much value investors place on financial advice and how they think about the trade-offs between fees and the value of advice.
(01:00:00) Reasons traditionally-advised people give for not switching to robo-advising.
(01:03:15) Having a relationship with a good advisor: how it impacts investor behaviour through poor market performance periods, and the importance of frequent quality communication.
(01:13:07) The key attributes of a high-retention advisor and what they should be focusing on.
(01:19:02) Success, retirement, timing, and knowing when to leave, according to Steve.
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, Chief Investment Officer at PWL Capital, and Mark McGrath, Associate Portfolio Manager at PWL Capital.
Mark McGrath: Well done, episode 336 today.
Ben Felix: That's it. Today, we're joined by Steve Utkus, who is fascinating. We covered a lot of ground with him. We spent the first half of the episode talking about his survey-based research on investor beliefs. Real quick background, we'll talk more about his bio in a second, but Steve spent three decades working at Vanguard. Two of those three decades were in a research capacity. He's at Vanguard doing research. He's got a bunch of friends and connections in the academic community. Like you talked about, he's known many of our past Rational Reminder guests. It was fun for him to listen to them on the podcast, stuff like that. But he knows all these people.
He's also got access when he was at Vanguard to Vanguard's data. So he's done all this really interesting research that you couldn't do without being in that situation because they have the ability to do stuff like run frequent investor surveys that Vanguard clients are willing to answer, and they have access to account-level data of those same Vanguard customers, which they can anonymize but then do research on. The beliefs-based research is about what do investors say they believe, and then how does that relate to how they behave in their portfolios. We spent the first half talking about that.
The second half of the episode, we got into financial advice, and it's a similar kind of idea. He has access to this incredible data, and they're able to look at what impact robo-advisors and human financial advisors have on investor behaviour. What are your thoughts, Mark?
Mark McGrath: Oh, it was great. Like you said, we covered a lot, but it was a great conversation. I learned a ton. I have some personal takeaways even from the latter half of the episode and great guy, great conversation overall. I really enjoyed it.
Ben Felix: His answer to the success question was really, really good, really practical and good. Steve is currently a visiting scholar with the Pension Research Council of the Wharton School of the University of Pennsylvania and a fellow at the Center for Financial Markets and Policy at Georgetown University. He previously was Global Head of Investor and Retirement Research at Vanguard. His current research interests include retirement security and aging, household finance and macro finance, which we talked about most of those things. He's got a Bachelor of Science degree from MIT and an MBA from the Wharton School at University of Pennsylvania.
As I mentioned before, he was at Vanguard for three decades. He alluded a few times to his many conversations with Bogle and reviewing Bogle's books before they came out, which was pretty cool. He spent two of those decades doing research. I think it was a pretty enlightening conversation on multiple fronts. Anything else to add, Mark?
Mark McGrath: No. No. It was fantastic.
Ben Felix: All right. Let's go to our episode with Steve Utkus.
***
Ben Felix: Steve Utkus, welcome to the Rational Reminder Podcast.
Steve Utkus: It's great to be here. Thank you for having me.
Ben Felix: Very excited to be talking to you. Let's start this off. Why is it important to understand the beliefs of investors?
Steve Utkus: As an industry, we are awash with pest performance data about markets, products, portfolios. It inundates us. But when have you actually asked your clients what they actually believe about the future? Yet if you think about it, if you do a financial forecast for a client, if you do a wealth forecast, you obviously have capital markets assumptions which are about the future. In academia, all those financial formulas, even just a simple Sharpe model, that's all based on expected returns, not historic realized returns. But I bet you no one in a finance class ever said to you, “And here's a time series of expected returns, so you know what we're talking about, as opposed to realized returns.”
What my co-authors and I decided to do is actually build a time series and not just a one-off. There are actually a lot of surveys out there. There's one from Duke. There's one from the American Association. Individual investors, if you're in the institutional market, there are lots of institutional consultants who ask you once a year, “What do you think the 10-year return on global stocks will be?” But we thought we'd bring that home to a real live time series with a group of consequential investors buying whole Vanguard investors.
I have to give credit to my co-authors who are Matteo Maggiori at Stanford, Johannes Stroebel at NYU, Stefano Giglio at Yale because they really came up with the idea. But once they approached me with it, I was really excited about it because, actually, for years I'd worked in this marketplace and never really thought about what do actually investors believe about the future. That's why we're excited to study it.
Mark McGrath: Steve, how do you describe your unique empirical approach to studying beliefs?
Steve Utkus: Well, it's survey-based. As I said, there are some surveys out there, but they don't get a lot of press because they are episodic. This is going to be bi-weekly. It's going to be very simple. You could do this on your phone in five minutes. We're going to ask short questions about short and long-run stock market returns, interest rates, economic growth.
We also had a really interesting, innovative way of asking people to take 100 points and build their own personal variance of the one-year stock returns. What are the chances that stocks will do worse than minus 30%, minus 30 to 10, 10 to 30, 30 to 40, and 40 and above? You actually had to sprinkle your 100 points across this distribution, and we got these personalized estimates of variance, which is kind of hard to believe that people would be willing to do this, to be honest with you.
The other thing that was kind of interesting about our survey is we never asked people to drop out. If we sent you three or four emails and you never responded, we pulled you out. There's a Vanguard client service principle here, which is not to annoy people too much with market research. You might be a serial responder, respond to every survey. You might drop in and drop out. If you dropped out, we would then replace you with someone else statistically similar. So we've got this time series component that we really want to see variation across time within a person. That was really exciting. Those were sort of particular innovations.
Of course, then we married it in an anonymous way, in a data secure way. All the data was adding on Vanguard servers, totally secure, totally anonymous. We actually merged it with portfolio data to see what people were actually doing with their holdings.
Ben Felix: That part is crazy. You've got Vanguard customers. You're asking them for pretty high-frequency data about their expectations, and then you can actually see how that relates to what they're doing in their Vanguard accounts.
Steve Utkus: That's right. That's actually a really good point I forgot to mention is that one thing that’s true that this is a Vanguard-only sample. It's mostly what are called retail direct clients, clients who go online and manage their accounts. It has a sampling of 401(k) participants as well. But Vanguard is the ultimate strategic asset allocation firm. You trade in response to market information. You might trade in response to your changing personal circumstances, changing risk tolerance, changing tax status, family status, whatever. But Vanguard is the ultimate strategic asset allocation firm.
The way to think about these results is not all investors. Particularly, it's not active traders, but it's about a very consequential group, which are buy-and-hold investors. That's why we position it that way.
Ben Felix: Can you talk about the benefits of using survey data?
Steve Utkus: Well, it's the only way to get into people's heads. With the explosion of big data coming off of any kind of system, you can do a lot of analysis about people's behaviour. But you can never really understand the rationale behind that behaviour. What's great about survey data is you're going to ask people, "What's the rationale? What's their thinking?" Of course, you frame it in particular ways with particular types of questions you ask. But that's why this marriage really of survey data and administrative data is particularly powerful in financial services. It's a big trend in empirical research in household finance. This paper is sort of household finance, but it's also a bit of macro finance, so it has to deal with broader questions about evaluations of asset classes and expected returns on asset classes. But this marriage of the two is really quite powerful.
Mark McGrath: Overall, how informative is survey data about individuals' portfolio decisions?
Steve Utkus: We go to the administrative data for the portfolio decisions. That's one of the things we do, which is actually quite beneficial. If you ask people, “Did you trade last month,” they roll their eyes. Maybe I did. Maybe I did. Or we asked them what their equity allocation was of their Vanguard, IRA, or taxable brokerage account, they’d probably make up a number, and there would be big error term in it. But this way, we could actually go to the administrative data to get the actual holdings and behaviour and the intention in the survey data.
Ben Felix: What effect do individual beliefs have on portfolio equity shares?
Steve Utkus: We were struggling with how to organize the results, and we just decided to come up with five big facts about investor beliefs and how it effects portfolio behaviour. Before I dive into the first one, which is how does it affect portfolio allocation shares, I just thought I'd give for the benefit of the listeners some idea of the numbers we gather. What are our expectations? I'm talking about the minute in abstract.
This is from the August 2024, the most recent wave of the Vanguard survey. If anyone's interested, just go to a search engine. Type in Vanguard expectations survey, and it'll pop up. Just to give you an idea, these are averages across about 2,000 respondents. Stocks, one-year expected return of 6.2%. That was the mean expected return. By the way, two years ago, that was in the fours. One-year returns are extremely volatile based on market conditions. They go up and down.
The 10-year return was 7.9% annualized, so almost 8%. That is the highest it's ever been since we started in 2017, so it's way up, and it's been drifting higher. But this is, again, August 2024. I have no explanation for that, by the way, other than it's too early for the election. It's too early for the election post bubble. I don't know what was going on. Is it a political explanation? Is it the economic news? Is it investor expectations catching up with, “Oh, gosh. The market’s up 20%. I can't really believe in short and low rates of return.” I don't know.
Economic growth, one of the highest numbers ever at three-year expected economic growth rate of 3.5 % real, and the 10-year number, first time it approached 4%. So there was this really interesting peak in August 2024 that we have to analyze. But the other thing we ask people for the estimated variance, the lower side of that distribution, what's the chance that stocks will lose more than 30% is what we call the probability of a stock market disaster. In August, it was 4.4%, but it had been two years earlier as high as 8%. That was actually a peak. It's very interesting. That number moves around a bit. It's quite interesting. That gives you a flavor for the numbers.
Now, what's fact one? Fact one is when I become more optimistic or less optimistic about equities, my portfolio allocation barely moves. Specifically, if I think stocks are going to yield one percentage point more over the coming year, I'll only change my equity allocation by 0.7%, less than 1%. This is negligible. It suggests a very sluggish behavioural response to changing expectations.
The other interesting thing is when we did this, we said, “Well, if your portfolio allocation doesn't really vary based on changes in your expected returns, how does it change based on that variance we asked you to construct?” If the variance got bigger, it had no effect. Basic finance says I should demand a higher expected return in response to variance, and there's no relationship. What?
The downside risk had a huge influence. If I was more concerned about downside risk, I put less inequities. If I was less concerned about downside risk, a disaster, I was more willing to put inequities. You might notice this from client discussions. Clients respond when thinking about portfolio allocations. They tend to respond more to the disaster scenario than some average theory about variants. I don't know if you see that.
Ben Felix: Oh, yes. For sure.
Mark McGrath: Well, it's classic prospect theory, isn't it? Loss aversion, probably.
Steve Utkus: It is. We didn't really trumpet that much, but that's actually quite a striking finding that portfolio allocations are based on downside risk. It's an extreme downside risk, like the chance of a disaster rather than chance of a loss.
Ben Felix: We did an episode recently, Mark and I, where we talked about a 2020 paper from Lubos Pastor, Political Cycles and Stock Returns. Are you familiar with it?
Steve Utkus: I've heard of it, but I haven't read it.
Ben Felix: It's basically when risk aversion is high, expected stock returns tend to be high, and democratic presidents tend to be elected. The headline empirical fact that they're discussing in the paper is that pretty much all of the equity risk premium from 1927 to 2015 over three-month treasury bills has been to under democratic presidents. The theory behind it is, well, risk aversion is high when Democrats get elected, and risk aversion is low when Republicans get elected. Obviously, we just had a Republican get elected. Based on the numbers that you just talked about, people seem to have really high expectations and low assessments of crash probability, which just seems very interesting.
Steve Utkus: This was in the field mid-August, so that's post-Democratic convention.
Ben Felix: Markets have only gone up and up and up since then. I would guess that those numbers have probably only gotten higher if you were to do the survey again now.
Steve Utkus: Well, let's stay tuned. It'll be out shortly.
Ben Felix: Okay, that's good.
Mark McGrath: Super interesting.
Steve Utkus: Yes. It was in the field just before the election.
Mark McGrath: How does equity sensitivity to beliefs vary with things like wealth, attention, trading frequency, and confidence?
Steve Utkus: These probably won't surprise you that if you're wealthier, you're more likely to reshape your portfolio in response to your expectations maybe because you have more wealth at stake. You want to have a greater alignment between what you believe and what's in your portfolio. That's at least the hypothesis. This idea, attention, I don't know how familiar your listeners are to this term, financial attention. That's the time that you spend actually looking at your account online.
In fact, I did a paper about this years ago. It's just so fascinating. When markets are up, people look. When markets are down, people don't look. But in general, in financial attention here, the more you attend to your account digitally, the more likely you are to move in response to change beliefs. Same thing, what else do we say? Trading, frequency, confidence; they're all the indicators that you're more actively engaged, and therefore more likely to align your portfolio more quickly at the margin to your changing beliefs. Either going up or going down.
Ben Felix: How does the observed sensitivity of equity shares to beliefs, which you said is pretty low, compare to the predictions of an asset pricing model?
Steve Utkus: I am with very esteemed macro finance guys. This is a very important point from a macro finance point of view, which is that there's a standard, frictionless Robert Merton model. If your expectations in that model went up by one percent, in other words, you say you think stocks are going to earn six percent over the next year, and you went to seven, you should raise your equity allocations by five, six, seven percentage points.
To be honest with you, if we were thinking about this from a portfolio management perspective, we’d probably agree with that, too, that if you really want to have skin in the game, you have to move your portfolio by at least five percentage points. That's what the Merton model predicts. But I think what we point out in the paper, and I think that it's intuitively whatever you understand, is first of all, there's this major behavioural constraint, the cognitive bandwidth. You have a life. It's full of things. Over here is managing your portfolio and trying to respond to that change in belief.
Then there's also the issue of capital gains tax and taxable accounts. Then there's also just the administrative time. Oh, I have to get to a computer. I don't want to do it on my app because I want to do it on computer. But I'm busy at work, and I don't want to do it at work and sort of administrative burden. There are all these frictions, and that's why this result suggests for academic finance that it's really these friction models that we have to pay attention to when we think about investor behaviour.
Mark McGrath: How do beliefs affect individuals' trading behaviour then?
Steve Utkus: That’s fact two. Fact two is – so if that's how they allocate, how do they trade? Well, it turns out our sample is about three and a half years because every other month. It's not that great. It's not like a 10-year data set, but it's still a substantial period of time. The answer is it doesn't affect trading at all. There's no relationship between levels of trading. It doesn't cause people to trade more.
But if your belief has changed, you're more likely to change the trade direction and magnitude in the direction that you're changing belief. If you're more likely to think stocks are going to go up, you're more likely to move more money into stocks. The direction will be, of course, into stocks as opposed to away from stocks. That sort of affects magnitude and direction of trading. Well, you don't see people trading at higher rates because their beliefs are not more positive. You don't see them trading at lower rates. They seem to be trading episodically related not to their changing beliefs. That might change if you had a 10-year data set. Maybe if you had a more active trading population as well.
Ben Felix: What about belief heterogeneity? Are beliefs shared by similar demographic groups, or are they unique to individuals?
Steve Utkus: We only have a limited number of demographic factors in our data, and there's not really a strong relationship with demographics. There are some minor ones. But the most interesting, and this is our fact three. We’re working through the five facts. The most interesting thing here and actually I think this is the one most relevant for practitioners is that at least half of your belief is determined by whether your prior belief, whether you're a pessimist, whether you're an optimist, whether you're in the middle of the road. There's just persistence in beliefs. In other words, it's not like this month and this way to the survey. I'm a real gung ho for the stock market. Two months later, I'm a real pessimist. Two months later, I'm in a middle of the roader. It's like the optimists are cruising along being optimistic, and the pessimists are cruising along being pessimistic.
This is why I actually think an interesting exercise for advisors to consider. I've had this discussion when this paper, some of the initial results first came out with some advisor groups that worked with Vanguard, is actually it'd be interesting to talk to clients. Get them to tell you what their beliefs are. Then look at the Vanguard distribution and say they're a pessimist, they're an optimist, they're somewhere in the middle. It might help to have better performance-related conversations, but also might lead to better conversations about expectations for the future.
There's something here that might be worth providing. Do you guys ask about expected returns of clients?
Ben Felix: We do not.
Steve Utkus: Yes, no one does.
Ben Felix: We tell clients what our expected returns are, yes. But that's a really interesting idea because it is really the difference between the client's expectation and the actual outcome that they care about.
Steve Utkus: It's really hard. Before I started this project, I might have said a lot of people base it on the old Ibbotson numbers because people published charts with 10 or 12, depending on whether they're doing geometric maintenance. People would say 10 to 12, 10 to 12, and maybe they were more. But if you think about it, for the average client, whether they're self-directed or advised, where do they get that information from but historic data. But I don't know that people actually could articulate it. I don't think I could have given you a specific number until I had to fill out our survey questionnaire.
Ben Felix: I would probably guess something like 10%. If you just look online at people talking about expected returns and what they use in their financial plan, a lot of people say 10%, which I think is crazy.
Steve Utkus: Crazy. But that, by the way, adds an expectation for globally diversified stock index. Even that seems crazy in today’s valuation polls.
Ben Felix: I've done a video, probably a podcast episode on this too. Is 10 % a reasonable expectation for stock returns? Then you go and look at historical US stock returns. It's like one relatively recent chunk of US historical returns are around 10% a year. But if you go back further, or if you extend to other countries, or if you look at expected returns for US stocks, it falls apart pretty quickly.
Mark McGrath: Those are index returns too. That just seems you actually capture those returns, and there's fees involved.
Ben Felix: Yes, true.
Mark McGrath: And people just like round numbers, so 10 % sounds nice.
Steve Utkus: I've always been a pessimist anyway. I'm still stuck at six.
Mark McGrath: It's a really interesting point you make, though. Ben, you've worked with clients for a long time. If you've worked for any amount of time with clients, you can kind of tell if they're a perpetual pessimist or an optimist. So I think it's a really interesting way to talk to them about their expectations.
Steve Utkus: I think Vanguard has an investment strategy that publishes in annual forecasts. When we first did this, the numbers, they were all single digit, those 10 years and one-year returns on stocks that we thought these millions of people can't be reading research published from the investment strategy, no matter how much we were proud of it. How did they develop these low single-digit expectations?
Then I was listening recently. I think it was Goldman out of New York issued a big paper on US. It might have been US or global. It was four percent nominal, one percent real. There is this information that's filtering out to investors. That's their view. But that was very interesting. That, of course, maybe caught my attention, maybe your attention because we're in the field. It's that kind of information that perhaps is shaping these single-digit numbers, even when we blow through 20% a year.
Mark McGrath: What effect do investors' cash flow expectations have on their expectations of future stock returns?
Steve Utkus: As you can imagine, any of the classic stock prices are a function of expected future dividends, but we weren't about to ask people about expected future dividends on the stock markets. It's kind of your question. Only a few people know about. We asked these a proxy for cash flows. We asked growth rates in the economy. We asked about real growth rates, which is itself a challenge. What you saw, this is fact four, there's a pretty clear alignment between cash flows and expected returns. That's encouraging. In other words, the people who are more optimistic about the economy or more optimistic about stock market returns, people are more pessimistic about the economy or more pessimistic about returns.
There is a general alignment, which was nice to see from an academic perspective and from a – it actually showed a coherent response by investors, something they really understood that if you were gung-ho about the economy, you should be gung-ho about stocks as well. These consistent beliefs.
Ben Felix: That one lines up with the theory.
Steve Utkus: Well, it does. But the takeaway for theory if you want to go down a little bit of academic rabbit hole for a moment. A lot of models of equilibrium and value of stock prices are based on a single representative agent that looks and sees expected future dividends and has expectations for future stock market returns. A way to think about this, and we do this in the papers, we use this so-called Campbell-Shiller decomposition. It's pretty well-known that the logarithm of the price-dividend ratio of the stock market, which we can all observe, is equal to the expected future change in dividends, minus the change in the discount rates or expected returns.
In this model, in the model we were just talking about, there are optimists, and the optimists believe they're going to be great dividends. But they also believe in high returns, and the returns discount the dividends, the price-dividend ratio in the market. Then the pessimists, they believe in low cash flow, but they also believe in low returns. That's the discounting factor on their cash flows. The result of that is they both arrive at the single discounted rate in the marketplace. Now, we're not talking about a representative agent pricing securities, but we're talking about groups of people with different beliefs, beliefs about cash flows and beliefs about discount rates. That's actually a much more robust theory.
I think if you had to ask our academic friends, one of the major contributions is to say this is the path forward for academic development, thinking about the overall valuation of stock prices. It's not single representative agent model, but these tiers of people with different beliefs all competing. We all observe the same price. Market closes on a specific price at a specific day, and we have some general view of aggregate dividends to get a price-dividend ratio today. But it's where we disagree is the other two components.
Mark McGrath: Very interesting. How do investors' subjective probabilities of stock market disasters affect their expectations for future stock returns?
Steve Utkus: This is my favourite belief because of all that stock market disasters. I'm the guy who in 2024, when everyone says market’s up, whatever it is throughout the year, I always say, "And what do you think the chances next year will be down 35, 40%?" I'm always thinking about that. I think I'm one of those pessimistic groups, slightly more concerned about location. We asked about this probability of greater than 30% drops. No surprise, if I believe there's a greater chance of a stock market disaster, I believe in lower overall stock market returns. Bigger stock market disaster, lower over stock market returns. Small stock market disaster, higher overall returns. That's what we found.
Ben Felix: How does that one line up with theory?
Steve Utkus: That’s the other aha moment from this paper from an academic finance perspective, which is the theory is if I believe there is a greater probability of a disaster, I should demand a higher expected return. That's what we've all been taught. That's standard models. But, in fact, this model is there's a group of people who have had this distribution. Big disaster probability, mean shifted left, people with this probability distribution. Small disaster probability, mean shifted right. It's a very different model than if there's this big stock market disaster. My mean should be way over here to compensate me for it.
That's a very interesting example of where the beliefs did not line up with the classic explanation for why you should demand a higher stock market return as an investor. That's where I think some of these facts we organize, some of them are really just takeaways for the empirical finance community to think about the next generation of models when you think about risk and return and overall valuation levels in the market. But as you can see, some of these are pretty profound ideas. You don't allocate based on the variance or the portfolio. Allocate based on the downside risk. If there's a greater probability of a disaster, you demand less, and you expect less.
Ben Felix: Right.
Mark McGrath: Speaking of stock market disasters, how did investor expectations about economic growth and stock returns change during the COVID-19 stock market crash?
Steve Utkus: Here we were going along, administering our survey. It was February 2020, and we had our regular administrative meeting, and we organized it. It was in the field. Then the whole world fell apart. One of the great benefits of this survey, and I have to thank our marketing research partners, they were so nimble about this, is they said, “Well.” They had worked with us on this process. “Let's do a special survey right in the middle of the market collapse." It turned out stocks peaked early February globally, and we were in the field two weeks later, and then stocks went like this. We went into the field off-cycle in March and then in April, and we did this analysis.
I was looking at some of the numbers. They're not too surprising. We have February to April with a data point in the middle. No surprise, expected returns fell. But surprisingly, we're still positive. Until the crash, they were hovering. Investors were saying, "We think the market's going to do five, six percent a year." Two months later, they were one to two percent a year. It actually fell a little bit late. They didn't go negative, which I found interesting. Obviously, there were plenty of people saying negative returns, but the mean did not go negative. The probability of disaster, let me look at my notes, doubled, no surprise. People were more worried about a potential disaster.
Ten-year expected returns rose slightly. This is April. I think they drifted down a little bit later in the year. But as of April, people were thinking, and this is not going to affect 10-year returns. In fact, it might be slightly more positive because prices have fallen, so expected returns should rise slightly over the next decade. The same thing sort of happened in the economy. There was a slowdown in expectation for economic growth, but nothing like what happened, say, during a global financial crisis when rolling three-year GDP was like 0.3%. The mean was still around two. I think it was 2.2.
There wasn't anticipation of sort of major economic contraction, even though, in fact, there was. But perhaps unanticipated because of no one really realized what was going to happen. But at least through that period, it was really a downward shift in short-term expectations, not long-term.
Ben Felix: Yes. That's really interesting that short-term went down, long-term went up a little bit. Like they believed in momentum but also understood that expected returns increased.
Steve Utkus: But not too much because I would have thought, given the size of the stock market crash through April, I think it was down 50%, 60%. It was pretty sizeable and –
Ben Felix: It was big. Yes.
Steve Utkus: Yes. Numbers should have gone up more.
Ben Felix: What impact did pre-crash, so before the crash, optimism or pessimism? We talked about those steady states that people in your sample have. What impact did that have on whether people sold out of equities after the crash?
Steve Utkus: Roughly speaking, there wasn't a lot of selling actually. This is a general rule in buying and holding investors, even based on a severe market shock like this. 70% of investors didn't do anything, 30% did. The optimists were more willing to trade than the pessimists, but the optimists also had a bit more inequities anyway. They had higher 60% allocation, average of equity allocations, as I recall.
For example, of the optimists who traded, okay, so that's roughly 30% of them, they shifted about 4% of their portfolio on average. Clearly, some moved more. Some moved even smaller amounts. The pessimists shifted to 2%, so it was almost – even among the traders, the shifts were, I don't know, what you just described as microscopic.
Ben Felix: Yes. Small, for sure.
Steve Utkus: Yes. Small, for sure. It was almost like, okay, the market's declining. I think I might sell out a bit among the traders. Again, the 30% who were trading.
Mark McGrath: They were trading away from equities in both camps?
Steve Utkus: And both sides, yes. Now, I will say this. That's the net movement. Gosh, I don't have the gross movement in front of me, and I think it's in a spreadsheet somewhere. I don't know that we published it. But we know from prior experience that although the net movement was away from equities among all traders, there was a group of people who were trading into equities. This was the net movement of optimists away from equities, but there probably were some optimists who were buying as well. It's important to realize markets are two ways. There's always people on the other side of the trade.
Mark McGrath: What do you think are the lessons for investors in how pre-crash beliefs affect selling behaviour after the crash?
Steve Utkus: Well, a lesson from this crash was don't do anything.
Ben Felix: Yes.
Mark McGrath: It was so quick.
Steve Utkus: It's a lesson for most crashes. The other lesson I would take away from this, if you're going to do anything, don't do too much of it. This is actually a long-standing belief. I remember about 25 years ago when I was reviewing first drafts of Jack Bogle's Bogle and Mutual Funds book. There's a chapter on asset allocation, and I had been doing some work on that for some reason, and I've read the chapter. Even I think Jack at the time was writing it like, “If you're going to trade, don't trade too much.” This idea of swapping in and out of the market based on tactical beliefs was sort of reborn at Vanguard.
I think you see that sort of reflected in the messaging to clients. If you're going to trade, make small trades. I still believe that personally as well. It's otherwise too damaging. By the way, the optimists who oversold made a mistake, obviously.
Ben Felix: The optimists reduced their equity on net by four percent and pessimists by two percent?
Steve Utkus: Yes.
Ben Felix: The optimists reduced their equity by more, but they had a higher equity allocation to start, though.
Steve Utkus: That's right. Again, underlying that, and I wish I had the numbers, but there's some group of optimists who are increasing, which means that the optimists who were decreasing we're doing more than four percent on average. Let's say that was 10%. Let's just make that 10% for the moment, just making that up. They then subsequently in April traded more back into equities.
Mark McGrath: I did exactly that. I think I was 80/20 in my portfolio, 80% equity, 20% fixed income at the time, not across my whole portfolio, but in one of my accounts. I managed to bottom-tick the crash and went from 80/20 to 100% equity, just dumb luck. But I literally bought them, ticked it to the day, and I've been 100% equity ever since.
Steve Utkus: Congratulations.
Mark McGrath: Well, thank you. I didn't really do much at any long term, but it was just a funny thing. You talked about traders, optimists who were moving the other way. It's just funny that I happened to have been one of them by luck.
Steve Utkus: There's a senior thought leader within Vanguard at the time, I'm not going to give his name, who did exactly the same thing and crowed about it for weeks after.
Mark McGrath: I don't think I told anybody, but yes.
Steve Utkus: We're like, “This is not the advice we get in the class. We say stay the course.” He said yes, but no one said over invest inequities when they're down. He said we're silent on that question.
Ben Felix: That's too funny. I was already a hundred percent equities. I had nowhere to go, except using leverage, which I did not want to do.
Mark McGrath: I was going to say that's the next move, I guess.
Ben Felix: Yes.
Steve Utkus: Well, based on that earlier discussion we had prior to the podcast, you're going to stay that way the rest of your life. You won't have any choice to make.
Ben Felix: Listeners will know we're talking about the Scott Cederburg paper, which I've reviewed the updated draft of. It's not out yet. But by the time this episode comes out, it will be out, and everyone will have been very excited to read it. Okay, I want to move on to your other survey paper on ESG. What survey motivations do people give for investing in ESG assets?
Steve Utkus: We were using the same bi-monthly survey infrastructure, but we decided we'd add a few questions because it was a hot topic in 2021. By the way, I do have to thank Xiao Xu at Vanguard and Joe Tan at Yale. They also helped on this project. They were part of this effort. We decided to add these questions. Even today, when I watched the discussion about ESG, no one cites any data. I wish they would cite this paper. But actual investors believe it's all about extreme points of view.
As you imagine, in any marketplace, there are two points of view, if not more, on any particular issue. We added these questions on ESG. First of all, we thought to ask, are ESG factors important or not in your investment decision-making? It was about 50 /50 split; 45% not at all, 55% said could be partially or somewhat or fully. Among the people who believed in ESG, it was very interesting. The dominant reason was the ethical aspects do the right thing. The second dominant reason was climate change. Forget the S and the G. It's all about E, and not just all E, but climate change E.
Then there's a small group of people who said, “Oh, it's just a way to make money. The reason why I'm investing in ESG,” and they had higher expected returns. The other two groups that do the right thing, and the climate change people actually thought there was a penalty associated. On average, it worked out to be one percent a year return associated with doing the right thing or hedging against climate risk. But there was a group of people who, of that 55%, that small group, who believes that you're going to do better in the long run. All those stories about stranded assets are going to come down to haunt all those stocks, and you're going to be on the right side of the return distribution.
Ben Felix: That's really interesting because that seems like the two bigger groups, they have a taste for ESG and therefore should expect lower returns, which they do. Then the smaller group, maybe they have a hedging motive. They think climate's not properly priced, so therefore they expect higher returns.
Steve Utkus: They do the right thing and the climate people. Both think there's a penalty. There's actually a group of people say, "I'm investing in ESG just because. My main reason is because I think it's going to have higher expected return.” It was a third group. The key point I continue to make when I ever offer this question is to observe that there are two sides of the trade. There's some people who think it's important. Some people think it's not. Otherwise, you wouldn't expect that. There are heterogeneity in beliefs, as we were saying. It's not all about necessarily doing the right thing. For some, it's just pure climate edge. That's the second most important.
Mark McGrath: How are ESG motivations related to portfolio behaviour?
Steve Utkus: First of all, there's a big caveat here. We only had around four percent of holdings in ESG products because they were just being launched at Vanguard on the platform. As you know, when you launch a product, the group of investors who are self-directed, it could take decades for it to be incorporated into their portfolios as they have new money cash coming in as they reconsider their portfolio allocations.
As a general rule, if you had any of these particular reasons, you're concerned about climate risk, you're more likely to be concerned about doing the right thing, you're more likely to invest in ESG products. You saw that even in this small sample. I thought that was quite interesting. It is a small sample because most ESG investors, when you looked at the construction of portfolios, are using it as a minor portfolio holding. You rarely saw someone who, say, had it in an ESG, all ESG stock portfolio.
Ben Felix: You talked about the lower expectations that those two groups have. How important are financial returns to the investors with ethical and hedging motives?
Steve Utkus: That's the interesting thing. If you take people who have non-pecuniary motors, and among those groups, on average, they thought they would do not as well in ESG. But within that group, there were some people who thought they would do better based on their expected returns. We asked them in this survey because it was about expectations. We asked them about the expectation for ESG investments versus we had already asked them about the stock market, so we could compare the two. It turns out the more optimistic you were about the returned prospects, the more you put in ESG.
I'm concerned about climate risk. I don't think ESG is going to do that well. I'd allocate a certain amount. If I'm concerned about climate risk, but I think ESG is going to beat other stocks, and then put a lot more money into it. Even among ESG investors, there's a differential based on financial expectations, which I found fascinating. It's a curious wrinkle.
Ben Felix: That really is fascinating. People care about this stuff, but they also care about their financial returns.
Steve Utkus: Right. Conditional on believing in it. It gets back to this idea we had about heterogeneity. There's the optimists and the pessimists. There are ESG. When I talk to people about this, I do try to clarify where they stand because there are ESG investors who believe it's a penalty. It's a cost, and they are willing to assume a cost of X. But then there are also people who are ESG enthusiasts who believe it's a positive benefit, a positive benefit, Y. You have to really differentiate those two groups and understand their motivations because they're going to do something completely different when it comes to portfolio allocation. Of course, the optimists, they're really going to ramp up their interest, and they're going to be big proponents.
Ben Felix: We went through three of your papers on investor beliefs.
Steve Utkus: It was a lot.
Ben Felix: Yes, it's a lot. You're going to shift now to your papers on financial advice where there are, I think, even more papers.
Steve Utkus: On to three.
Ben Felix: Same numbers before, three.
Steve Utkus: The magic three.
Ben Felix: I'm super excited for these ones. I found all these papers to be really interesting. The first one, I think this one's published in the Journal of Financial Economics on the effect of robo-advice. Your perspective from Vanguard is just such an interesting place to do this research from. How impactful have robo-advisors been on the portfolios of previously DIY investors?
Steve Utkus: Well, actually, pretty profound. Before I get to that, I just want to say it's a project with my longstanding collaborator, Alberto Rossi at Georgetown. Alberto and I were actually not sitting far from where I'm speaking. We were at the Wharton School at a seminar. He came and said, "Steve, I have an idea for a project." Here we are many years later.
At the time, Vanguard was beginning to roll out what is a hybrid robo-advisory service. It's not actually the largest robo-advisory service, but it’s high as an important distinction. It has a human element to it. The investment process is highly automated. The methodology is highly automated. You can customize it. An advisor can customize it, but it's a structured customization. An advisor just can't arbitrarily pick something based on their own personal beliefs, but on this structured methodology.
It varied a lot, but the best way to think about it is it fixed the common portfolio errors that you see most retail investors making. The biggest one being too much caps. Sitting on cash, uninvested, regardless of risk aversion. Just sitting there. Being undiversified internationally. There were some people who were very well diversified internationally, but that was not the norm. Not thinking intentionally about portfolio equity risk share.
What's interesting is, on average, the portfolio equity risk share pre and post-DIY, say, advised investors, didn't shift that much. But individually, there are a lot of people whose share went up and went down because they hadn't actually gone through any structured process. They ended up where they were, not based on a sense of what level of risk am I willing to take with my portfolio. But they really ended up there almost by accident.
The other thing was active-passive mix. There was a strong emphasis on passive with active as a supplemental strategy, but we had DIY clients who just, for example, put a whole bunch of money into a small-cap active fund with no real sense of why they're allocating that much money to small cap or to active versus an active-passive blend or anything like that. That's pretty much what the investment advisory methodology did.
And the way to think about this is any financial advisor has an efficient frontier that they believe in. And luckily, for Vanguard, the efficient frontier we were offering people is a beautifully efficient empirical in academic finance terms, efficient frontier. So we were able to measure improvements in Sharpe ratios across the sample. And on average, Sharpe ratios rose. But it's through all of fixing all of these portfolio construction errors and moving people closer to a very efficient frontier. That's really the effect of the robo-advisor. And that's what you'd expect any advisor, but this was a particularly efficient frontier.
Ben Felix: Seems really impactful. It seems like that robo-advisors have probably been a pretty good positive innovation for previously DIY investors.
Steve Utkus: Not everyone though. Interestingly, there's a group of people, and they were the people more likely to leave the service over time. They sort of came by, got a portfolio checkup, said, "You read Bogle on mutual funds. You're very well-organized." And then they went on their way.
By and large, one of the things that we found encouraging, and actually the business leaders of the service found encouraging, is that the people who needed it most are the people who stuck around. Because it was the people who made the largest errors who were more likely to say, "I really perceive the service to be valuable." Now they have very high retention rates, but the people who did a really good job, more often they were more willing to say, "Okay, I don't really need an advisor. I'll come back in a couple of years and get another portfolio check." But most of the clients who stayed, they were fixing really significant individual portfolio errors.
Mark McGrath: I think you more or less just answered that. But what type of investor benefits the most then from switching to a robo-advisor?
Steve Utkus: The ones who make most mistakes. And the highest Sharpe ratio gains were among the people who were doing all those problems that we were just mentioning. People struggle with the technology of portfolio construction. I know for us a lot of it's just intuitive and because we are used to recommending portfolios to clients and thinking about portfolio risk and how to put the pieces together. But I think that's really the real advantage where this comes into play.
And particularly, remember, as clients get new money or they have to withdraw money, half these clients are still having inflows because they're working. Other clients were spending from their portfolios, rebalancing their strategy, maintaining their strategy in the face of market stresses. All those reasons really were the opportunities for having benefit from the service.
Ben Felix: How does switching to a robot advisor affect investors' time use?
Steve Utkus: I assume you guys think about this a little bit. I mean, one of the advantages of advisors is labor specialization. Your clients don't have to do it on their own. We estimate you save about six hours of monitoring and managing your portfolio a year in this particular setting where you switch from being DIY to having the Vanguard advisor.
The interesting thing is this isn't the time spent – for example, if you're a DIY investor, you have to spend a considerable amount of time learning and building sort of an endowment of financial knowledge. What we're measuring here is just the time needed to manage a portfolio on an ongoing basis, but not necessarily the hours that you need to read or engage yourself in personal development efforts to build your financial knowledge.
I think that's why when six hours seems on one hand, a lot of time to be spending managing your portfolio in a given year. But on the other hand, that's all it is. It's not any of the information you need to gather to develop points of view, to develop information about how you should allocate, or change, or trade your portfolio.
Ben Felix: And it sounds like a lot of the people in this sample maybe should have been spending more time building the endowment because they had significant portfolio improvements.
Steve Utkus: That's right. For example, you see this I'm sure in your own business, that the cash problem is partly a knowledge problem. What do I do with new cash? And how do I allocate it? But it's also an inertia problem. I have to get to a computer. I have to spend time. I have to think. I have to execute a trade. I have to follow up. Did it go through correctly? Is it the right amount? Those are a lot of daunting questions for someone who's trying to run another part of their life. That's where the advantage of the advisor comes in.
Ben Felix: I think people delay based on stuff like, "I'm going to wait until the election's over." Or, "I'm going to wait until whatever."
Steve Utkus: There's always going to be election. There's always going to be a war. There's always going to be a recession or a boom.
Mark McGrath: So if robo-advisors can automate the portfolio management side of things, then what needs are human financial advisors satisfying?
Steve Utkus: This was the second paper that Alberto and I did. There was a Vanguard survey that had gone out, and we decided to write an academic version of it to look at the non-pecuniary factors involved in selecting. Why do I use a financial advisor? So presumably, one is to manage my money. Maybe two is to give me advice on other financial matters in my life. But three, what are the other non-financial, non-portfolio manager-related, what I call non-pecuniary factors?
From the survey, it's interesting, you wouldn't I'm surprised by this. It's pretty basic. Number one, peace of mind. This whole notion that I'm better off and I sleep better at night because I have someone in charge of my money. Who is this expertise question? If I need to talk to somebody, there's someone who's a financial expert who can answer my questions. I see this, by the way, with my friends who have financial advisors, that they really do rely on them as a point – and that's a sole point of view, I would say. They actually solicit input from lots of different parties, but they actually rely on that person as a source of financial information.
Then the third is this delegation issue, bringing up time and delegating to someone who's professionally qualified. Those are standard reasons. It wouldn't surprise you that those are common reasons clients. This was sort of a separate paper about some colleagues of mine, Cindy Pagliaro and Anna Madamba. They took the same survey and they deconstructed it. They tried to estimate the total relationship value due to emotional factors versus financial factors. And they came up with a 50-50 split.
Clients assigned value to over 20-some activities that advisors did for them in the survey. And they used the same survey data we did. And they partitioned that into the value derived from hard things like managing a portfolio and the other stuff for personal coaching, interaction with the advisor, and trustworthiness of the advisor and the value from that. They signed 50 % to these. I call them social psychological factors or emotional factors, which I thought was quite interesting.
Ben Felix: What was the title of that paper?
Steve Utkus: I don't remember at the top of my head, but I'll find it for you. But it's the same data set, but they just did this estimate. This is relevant. You had Juhani Linnainmaa on here, who was talking about all the results. I think this is one of the interesting challenges for academic research and portfolio construction. Because this work is great showing that the portfolio management side, portfolio management pillar is not on average adding value relative to fixed charge and then advisors sort of put everyone in the same portfolio regardless, et cetera, et cetera.
But that does raise this question about what are these other attributes, the sorts of value? Is this the cost drag of paying for this? Or is this because people aren't aware of the cost drag here? And maybe they would demand more value from this. They understood that. I do think this research is all pointing to this question of the non-pecuniary benefits of financial advice, financial advisers, which I think is really important and how to measure it. I think it's going to be the next wave of challenge in the industry.
Ben Felix: Yeah. I know in one of Juhani's papers, I don't know if it's in a published paper or if it's in one of their working papers, but they make that comment of, "We see this advised clients are underperforming the market," which either implies that they're making a mistake by paying for advice or that they're deriving a pretty significant amount of value from the service.
Steve Utkus: Or both.
Ben Felix: Or a combination. That's probably closer to where the answer is. You mentioned some of the top needs that caused people to want an advisor. Peace of mind being right up there. I got a real quick story. We recently had a conversation with a client who is fine financially. And they told us that, for them, it was completely life-changing to have a budget, which we helped them create, and to have a financial planning model with the Monte Carlo simulation that based on their budget shows that they're going to be okay. They're going to be fine. They're quite wealthy. But I just thought such a simple thing for them to say that was life-changing. I thought that was fascinating and it fits with the peace of mind idea. What needs are not important? What needs do people not find important from an advisor?
Steve Utkus: Well, let me respond to your scenario actually a little bit. In this paper, we had actually all these very granular needs, like the transparency of fees, portfolio management and allocation, tax planning, budgeting, some broader financial planning considerations, and even interpersonal issues like a person I can trust, a person I can have a connection with. Notice those are two different, very distinct questions. And one of my favourite was a person who can resolve disputes between me and my partner or spouse, which you know is with married couples or unmarried couples, is a really important problem for financial advice.
When we grouped them up into five high-level categories, this one called trust and transparency. But the second one was exactly what you were saying, which was we called it self-improvement, which was a terrible term. But what it meant was I got my life organized. I have a plan. I am successful because I have a plan. Someone's in charge. We fixed problems and we're moving forward. And I have a vision going forward.
When you talk in general about research about financial advisors, there's a lot of focus on behavioural coaching during market downturns, which I think is a really important service feature. We've done that to great an extent because there are all these other attributes that I just listed that come into play.
At the time we were doing this research, one of our colleagues working on this had just hired a trainer at the gym and said, "It's the same kind of concept. I'm getting a plan. I'm executing on the plan." It's not necessarily I'm losing weight and I look great or I'm not afraid. It's actually I have a plan for my physical fitness. There's a sense of ownership, and pride, and success that comes with that. And the other thing besides – trust, self-improvement, and then this knowledge – having access to knowledge. Getting more knowledge about my bears. Knowing more about the markets.
The thing that surprised us most, and this gets back what we were talking about with Juhani's paper, which is the investment component of the value chain was the least important to traditional clients who have advisors. In other words, portfolio construction, rebalancing, tax-loss harvesting, performance, relative performance.
First of all, obviously, I think she'd report on performance. She'd report on fees. And people should understand that and know that because that's an important benchmark. We have to do that from a regulatory and fiduciary perspective. But reminding clients of these other attributes are really quite valuable because they are valuable to clients. And it's important to know that people devalue the investment performance relative to the marketing pitch that an advisor gives to a client. There's a lot to be learned there from a very practical point of view about how you pitch things to clients and also what clients derive ultimate satisfaction from.
Mark McGrath: it's so interesting that organizational piece is huge. Just getting organized with your finances is such a massive catalyst towards financial success. And on the other hand, rarely do I meet somebody who's very well organized financially, who's not also well on track to meet or exceed their financial goals. Just a balance sheet and an income statement, honestly, for most people can be a massive, massive catalyst. And it can remove such a huge mental burden as well, that it often becomes the thing from which they can leap forward, at least anecdotally in my experience.
Steve Utkus: All I would say is that the survey data of large groups of traditionally advised clients, as well as other types of advised clients, like people who use robo-advice services, they also have this broad sense of self-improvement, self-organization, spyingly taking charge of my finances, which is quite distinct from coaching me during market downturns, which is distinct from helping me and my spouse or partner resolve disputes about money. There's a rich territory here that goes beyond just behavioural coaching and market downturns.
Mark McGrath: So how are unadvised investors' needs different from those that work with a robo or traditional advisor?
Steve Utkus: As you'd expect, unadvised people are less interested in trust because there's no one that trusts there. They're executing on their own. They're also less focused on having these knowledge attributes, having someone to talk to because there's no one to talk to. They're really doing all that on their own.
I will say this, one of the things we did not ask, but we thought about it afterwards is, for do-it-yourself investors, there's a big question about trust in the institution they're dealing with. So let's say you're a U.S. investor and you're dealing with E-trade, or Vanguard, or Fidelity, or whoever, and you're self-directed, there are still these trust elements. But it's not the trust in an advisor, but it's more institutional brand trust. A company I can work with successfully, that sort of thing. So I just want to make that point. When I say trust, I mean, having a person you can trust. Obviously, it's not relevant because, of course, they trust themselves.
Ben Felix: What determines how much value investors place on advice?
Steve Utkus: If you ask now, where do they assign the most value? It turns out this trust category. Trust is a broad category. It's not only a person that I can trust, but it's also a transparency on fees and knowing what you get for value for the money. The overall perceived value of the relationship comes from that category.
And again, as we said, it's really that. It's this self-improvement and this knowledge category. All these three categories are the ones that drive this perceived value. If you want to increase clients' perceived value, those are really the top three you need to focus on. And in the paper, we have the itemized questions. You could see, for example, under trust, it's not simply about the person you can trust, but it's also about transparency and understanding how people are getting paid, even though people don't understand that. I'm going to talk about that. And I know it a little bit. It's a broader category than this "person you can trust." That's the category label.
Mark McGrath: How do investors think about the trade-offs between the fees that they pay and the value of advice?
Steve Utkus: I don't know if you have this experience. But literally, every survey I've done in my career on fees, no one understands at all. And that's the conclusion that we make from this survey. Years ago, I sat down, I was having coffee with two friends of mine who are entrepreneurs, small business entrepreneurs, quite successful. They had a portfolio being run by a family member, a nephew, who worked for a big, very reputable U.S. brokerage firm. But it was very much clearly organized along an RIA model. Wasn't a transaction-based compensation.
But I asked him, I said, "So just out of curiosity," this was way back when I was doing some research on fees before this project, and I said, "How's your nephew get paid? Well, I'm sure – no, he probably doesn't get paid. Does he get paid? Who pays him?" Back then, we're talking multi-million-dollars portfolio. He surely must get paid." And I said, "I think he does too. I know he gets paid. He gets salary. But how does it come out of your money?" And they said, "That's a good question."
Now, these were college-educated, very intelligent and aware. You can't say, "Oh, it's a lack of schooling." They're very talented business people. They knew where every penny went in their business." That's the extent of the challenge. So what we saw in this survey, clients seemed to have a sense of how they were getting paid, we thought. So half said AUM. And then smaller fraction said transactions, flat dollar. I'm not sure. We thought, okay, that seems low for AUM fees. But let's go with that thread. Let's ask another question. Quantity-wise, how much are you paying a year? 70% refused to answer the question. 30% who tried said the mean fees were 6.7% a year. Not 67 basis points. 670 basis points.
Some people said I pay one basis point. Some people say I pay 10 basis points. And we thought, "Well, there are a few people in this survey who are doing robo. But even the robo stuff was a little bit higher than that." Some clients said 90% a year.
Ben Felix: What?
Steve Utkus: That's why the mean got to be six. We should have been a trimmed mean. That then led us to question whether they really understood the question in front of that. What I would conclude from this, if you're a regulator, or an industry participant, or you're a client, people really do need help understanding how advisors are getting paid from this. This is a broad-based national survey. We're not picking on any specific firm. It really does highlight the need for a very simple straightforward. We get paid this way and roughly this amount a year. And simple basic information so people can really know what they're paying.
Ben Felix: I've definitely seen the same thing. We are super transparent about fees. And it's disclosed multiple times to clients when we're onboarding them. But if you ask a client how much they pay us, in a lot of cases, they don't know.
Mark McGrath: Maybe we should increase our fees to 90%. They're getting a ton of value from those relationships.
Steve Utkus: Or at least 670 basis points. Clearly, some of this is a percentage problem. A lot of people struggle with percentages. And obviously, if you paid 95% fees, you wouldn't have any money next year, virtually no money. Certainly, two years, you'd have zero. The second thing is my suspicion is that because people switch advisors once a decade, if that, maybe more frequently seven years. Maybe less frequently every 15 years. And if they do, that's when they think about prices. There's no frequency and transparency. But it does say people do need help.
Ben Felix: Would correcting investors' miscalculations about what they're paying for advice change their decision to pay for it?
Steve Utkus: That's what we'd like to conclude. You think there'd be a revolution in the marketplace if people knew what they were paying? There are two arguments against that. One is that people don't value many of the financial aspects. They value the non-pecuniary aspects. We really don't know the price. What would you pay for having a good relationship with a person you can trust who is open, and easy, and accessible to talk to? As an expert you can speak to who help organize your finances? We don't really know the price of that.
I think you would find that for the people who would say, "Gosh, this is a lot of money. And I need more justification for value," you'd be also surprised by the number of people who said, "Wow, I only pay that to get that value." I really don't think we have a very clear indication. We do know that among the 30% who gave us fee data, we do know that regardless of what fees their level of trust was unrelated to the fee level. In this very small sample, granted 70% gave us no fee data, but this 30%, at least it suggests there isn't a very clear indication of how it would work out. I'm actually in favour of the disclosure, just because people should know. It's not really clear which direction it will go.
Mark McGrath: I think I know how you're going to answer this, but what reasons do traditionally-advised people in your sample give for not switching to robo-advising?
Steve Utkus: Well, it's the human element. It's definitely someone to talk to. Separately, as part of the survey after we asked about all these value questions, we did ask this very specific question about, among robo-advised people, why'd you choose it? And among non-robo-advised, why did you consider switching? And they said, "Look, it's really having someone to talk to. I don't want an algorithm in charge of my money. I want a person in charge." And number three was "I don't trust technology," which is what led to our third paper on algorithmic aversion.
We were presenting results we were just talking about earlier from the Vanguard DIY to advised clients at a seminar. And [inaudible 1:00:40], Imperial College, discussed it. And he said, "There's an underlying theme here among traditional advisors, but even among people who choose to switch to a robo-advisor." This idea that you might be averse to technology being in charge. That's the biggest factor for why.
Broadly speaking, traditionally-advised clients are used to that human contact. They're not used to the machines being in charge. I think that's going to change. Technological adoption is a generational thing, and I think younger people are much more comfortable relying on algorithms to suggest whether it's streaming options, or healthcare choices, or debt choices, or cars you should buy, or how to buy cars. There's this whole technological revolution that's being led by younger people. It's only a matter of time, I think, that there'll be greater and greater reliance on algorithms. But as a general, if you ask traditional-advised clients today, it's this lack of algorithms, the preference for human contact.
Ben Felix: Algorithms could get more human in the way that they interface with people, too. Like with AI, LLM-type communication, it feels a lot more natural, but I don't know. I've been thinking a lot about AI. One of the things that I still find is that when I have expertise in something and I start trying to get information from ChatGPT, for example, about that thing, whether it's an academic paper that I'm familiar with or a more general concept, I find it still makes errors that are really hard to identify unless you already have significant expertise. Andrew Chen, who's a past guest on this podcast, he tweeted a while ago that as AIs get better, AI problems are just becoming human problems.
Steve Utkus: That's a good one.
Ben Felix: You have to trust AI in the same way that you have to trust a person. It'd be interesting to see how all that stuff changes people's perception of algorithms and technology.
Steve Utkus: Yeah, I found that to be the case too. For another paper – I'm working on a paper on health span. Unrelated topic. I do some retirement stuff, aging, with a friend of mine, Olivia Mitchell at Wharton. And we were both like, "Okay, let's enlist ChatGPT as a research system." I said, "Very useful."
Ben Felix: I find the same thing. It's scary. If I'm asking you to help me interpret a paper that I'm already very familiar with, I'll find glaring errors that you wouldn't know unless you're already familiar with paper. So now I'm scared to ask you about a paper I'm not already familiar with. And it's like, "What's the point?"
Steve Utkus: It's very good at writing a very nice – the French term is précis. But these summaries that are these beautiful bullet point summaries. For example, I think they would do a decent summary of this paper we were just discussing. Oh, trust isn't important. They sort of ended with that.
Ben Felix: What effect does the relationship with a good, and you can define good, advisor have on investor behaviour through periods of poor market performance?
Steve Utkus: So this is a segue to this third paper with Tarun and Ansgar Walther, one of his colleagues, and Fiona Greig, as well as Alberto. Fiona Greig of Vanguard and Alberto Georgetown. We sort of took this DAS environment where people were assigned robo-advice and we began to study the impact of the advisor, the so-called behavioural coaching question, but using this robo-advice setting. And we didn't really have a measure for the quality of an advisor. We just said, "Who are the people who managed clients? That's sort of a measure. We could have obviously gone back and tried to get attributes of the individuals. We could have actually gone through the client and video chat transcripts to see how good they were. But there were all kind of daunting problems with that. But we thought we're going to pick this one thing, which is who are the advisors who keep more clients in the service? How do they behave during periods of market turbulence?
Well, this was a really striking thing. They do the same that poor advisors do during periods of market turbulence. In other words, the good quality and the poor quality advisors both reach out, both do handholding, both talk about long-term investing. The big difference is that the high-quality investors talked to clients more before the market turbulence, during periods of benign markets or rising markets.
Basically, what it suggests is a spillover effect. The more time you spend building a client relationship and there is no market turbulence, the more willing that client is, say, to trust you during periods of market downturns, which I thought was actually a very pretty and practical tool to think about from an advice perspective. If you're running an advisory business and you're worried about, "Oh, if there's a market downturn, how many clients might leave us because they're disappointed?" The answer is, "Don't sit around and prepare better bear market communication strategies and tools for your advisors to reach out. Start building a relationship now." Because it's that relationship that's going to spill over into the bear market.
Mark McGrath: So it's largely about communication. Was there anything else that changed with, call it, lower-quality advisors?
Steve Utkus: Very exactly. In a mirror image, they were less likely to reach out. And interestingly, you might find this fascinating because you're in a business like this. In benign market conditions, they had similar levels of communication with clients except the client communications were driven by the client reaching out. If your business model is don't call the clients when times are good because they don't want to hear from us, the answer from here is they're going to trust you more in the market downturn and they're more likely to stay with you and not be disappointed.
And it was really kind of interesting. We were able to deconstruct how much of this had to do with the quality of the advisor persuading me that the robo-advice was doing a good job managing my money. We saw that if you were with a low-quality advisor, as a client, you spent more time on the website trying to understand parts of the website that explain how the methodology works. Whereas if you were with a high-quality advisor, you never went online.
The two elements here are, number one, do proactive outreach to clients. But be a good, I want to say salesman, I don't mean selling, but I mean promoting or communicating. A good educator about the investment process that we use to manage your money during good and bad times. I know it sounds like really basic information. It seems very intuitive. This is empirically verified that if you do proactive reach out and you do a better job explaining what you're doing, you're going to retain clients longer.
Mark McGrath: It's the frequency and quality of the communication, essentially.
Steve Utkus: That's right. And the talent that the person has in explaining why your money is being managed in a sound way. And that's what I found really interesting is we see the clients of the low-quality advisors going on trying to figure out how their money is being managed. Not just looking at their accounts, but looking at the explanations for how we manage money on your behalf. And when they left the service, because they had lower retention in exit interviews, they cited dissatisfaction with the methodology. So they weren't just dissatisfied with the advisor. They were dissatisfied with the robo-advice methodology.
Mark McGrath: Was it their understanding of that methodology or lack of understanding of the methodology?
Steve Utkus: Their lack of understanding, exactly. They just trusted more. They found it less trustworthy. Getting back to the point, even this very successful hybrid robo-advisor service, the quality advisor really stood out as a critical success factor.
Ben Felix: We did a talk for a small group of clients recently, and we just did basic stuff. We talked about the origins of asset pricing models, and market efficiency, and whatever, index funds, challenges with managers outperforming, lack of persistence, basic stuff. We use index funds and dimensional funds primarily to build portfolios for clients. We talked a little bit about that and why we think that makes sense. Like I said, just basic stuff. And afterwards, the feedback was like, "That was such incredible information. I feel so much better of what you guys are doing now." It's interesting how I would call it basic. But then to a client, it's really important information to help them understand what it is that we're doing and why. That all really resonated.
Mark McGrath: Well, that's just it, Ben. It's basic to us because we've been in it for a long time and it's second nature. But to the general public, that's really, really deep, fascinating information, I think.
Steve Utkus: And really, helping understand the black box. How’s my money's being managed? There's a lesson here for robo-advice, which is you have to explain what the machine is doing on your behalf and on behalf of the advisor. But the broader investment for non-traditionally advised or traditionally advised clients or people with less automated processes is still this communication about the investment methodology that matters.
I don't know how you think about this, but honestly, the more I talk to RIAs, most of them do have some automated processes anyway. Even "traditional shops". The strategic allocation process is all standardized. The customization process, there's 15 levers you can flip, but you can't flip 20. There's a standardization that seems to be occurring. I just think longer term, more and more of different aspects of the investment process. Even like rebalancing strategy. It's not going to be sort of a, "Hey, we should take some money off the table," or this sort of casual conversation with a client and then trying to figure out the tax piece. It's going to be much more rigorous and much more automated with the advisor communicating what the methodology is saying rather than doing the math.
Ben Felix: Our business on portfolio management is hyper-automated on rebalancing, on asset allocation models. Yeah, I totally agree with that. In this sample, we talked about helping them get through market crashes. What other benefits are investors in this sample getting from the human advisors?
Steve Utkus: We did some really interesting calculations. We built this really fancy model, which you haven't asked me any questions about, which is fine. Because it's too complicated. Discussed today about algorithmic aversion among investors and clients. That model allowed us to come up with these really fascinating estimates, which blew my mind. A high-quality advisor in our model is worth a 30-basis point increase in lifetime total return. That seems pretty shocking.
Another way to put that is, if I'm with a high-quality advisor and someone said, "Oh, Mark's my high-quality advisor, but Mark's too busy, so we're going to sign you to Ben. And he's kind of low-quality." I, as a client, would ask for 10% increase in my current wealth to switch to the low-quality advisor. And if I'm running a firm from a profitability perspective, it's a 25 basis point increase and my gross profits and my sort of corporate surplus from that relationship by having a high-quality advisor.
Depending on which way you look at it, whether it's the client perspective or the business perspective, this whole question of high-quality advice versus low-quality advice is really at the heart of the economics of the business and benefits to clients. That's why to me, this paper we did, it's kind of interesting academic, but it just scratches the surface. So now, I think there's a real opportunity for academia and for industry to go deep on the determinants of high-quality advice. And maybe it is running seminars on manager persistence and capital market efficiency and all that sort of stuff. It could be because it could be that's the sort of information about the investment process that builds trust, and transparency, and strong relationships. There's real money to be made here for the client and for the firm by thinking more about quality of the advisor relationship.
Ben Felix: This podcast, I think it's another example. We get a lot of new clients from the podcast. We have a lot of clients that listen to the podcast and it's much more scalable and systematic than doing in-person seminars. Maybe it's not quite the same, but it has some of the similar effects.
You mentioned your model being too complicated and I didn't ask about it. Yes, because it was too complicated and I was reading it and I was like, "I don't even know what to ask." Those numbers are crazy. Can you give some intuition about where that's coming from?
Steve Utkus: Well, it's literally coming from this – in a world in which I'm algorithmic averse, a high-quality advisor persuades me that the advisory methodology that is accompanying us along the process of managing our money is, in fact, valuable and worth following. Therefore, I stay with the advisor. It's really a client retention effect, but it's the effect of high turnover. If I had said it the other way, there are two firms, one firm with high client turnover, one firm with low client turnover. You'd know economics of the high turnover client are terrible relative to the economics of the low turnover firm. It's that kind of idea. Going back to the survey work with Alberto, what elements of trust are we? What elements of self -organization are we driving? What elements of those needs are driving the value? I think there's a lot of work to be done on this.
Mark McGrath: You talked about communication, but what are the attributes that define a high-retention advisor in your research?
Steve Utkus: This is where I think there's this real opportunity to make it tangible for individuals. In our research, can you imagine going to a firm and saying, "Okay, Ben, I'd like to see your 10-year client retention numbers compared to everyone else in the firm. And if you're not at the top half, I'm not taking you." The metric that we had came up with, it's obviously very imperfect. I think that's the real opportunity here for firms to figure out what are the drivers of advisor quality and what are the characteristics that we want to lead with clients?
That's actually where we recognize in this paper, we're sort of at the edge of what we could do because we couldn't get into the life histories of our advisors, their characteristics, simple demographic, their psychological characteristics. I think if I were a client, I'd probably look for all those attributes that we talked about earlier. Someone you trust, someone you can work with, someone you have a communication with, someone who's transparent with you, who he feels transparent with you. That's what you really have to rely on. But ideally, you'd ask about 10-year client retention rates.
Ben Felix: You could probably ask about that. We'd give people those numbers. We track them.
Mark McGrath: I don't think I've ever been asked that. I get asked for referrals, direct testimonials. I get asked if they can talk to one of my other clients directly, but not necessarily. I don't think I've ever been asked about retention rates. It's really interesting.
Ben Felix: When people ask for referrals, I say, "We can do that." We've got clients who are willing to talk to you. But I'm going to pick the best ones that are going to give the best. I don't know how valuable this is to you. If you want to talk to someone, that's fine. But I don't know. I'm always a bit hesitant to do that.
We talked about AI and where technology is going and stuff like that. What are your results in these papers suggest about the role of human financial advice in an increasingly automated world?
Steve Utkus: It's the same thing we talked about earlier. Behind the scenes, traditional advisors have a lot of automated processes, and it's only going to become more automated. The example of these robo-advisory services and hybrid robo-advisory services, it's interesting who uses them, by the way. If you look at broad industry data, this may be five years old or so. I haven't really looked at this in a while. But it's young people are interested in a new thing on the block. But it's also very affluent, older investors who are experimenting to see, "Oh, I'll take 10% of my portfolio and give it to a robot-advisor and see what it does with it compared to what my traditional advisor is doing."
So there's going to be this gradual technological adoption. And I'm with you that there's going to be this human interface. Because even if there's this next generation of quality, right now, the AI technology is just too erratic and hallucinates too much. I could see it as an augmentation to a very domain-specific digital advisor. I'm going to create a human-like figure who's gonna talk to you rather than the digital advisor that does everything through the CX interface that the digital advisor you're on. But I can't see it replacing traditional for a while. But I do see this world in which the human component becomes the explicator, the explainer of what the machine is doing.
Ben Felix: There's some research on the impact of AI on knowledge workers that kind of suggests that it brings the lowest-quality ones up closer to the average. It doesn't necessarily make the best ones better. But it could have the same effect on financial advice, where a very mediocre financial advisor gets a lot better because maybe they're getting help or prompts from an AI assistant. That could happen.
Steve Utkus: Yes, and the AI assistant who reads this paper will say, "Call up the guy who hasn't reached out to your client. You actually need to reach out to these three clients proactively this week. Because come to my ex-market downturn, you're going to lose out."
Mark McGrath: What do you think financial advisors should be focusing on then in their practices to be as valuable to their clients as possible?
Steve Utkus: This is me speculating more as a manager and hire of people and a person who managed different parts of businesses over the years. But I think the human interpersonal skill side is absolutely critical. And it used to be that, of course, the investment business meant sales skills. And it still does in a lot of corners. But now I think the leading advisors, its empathy, clarity of communications, reading the client, find a partner for comprehension and understanding. It really is these soft, emotional, intelligent skills, I think, are definitely what I would invest in if I were, say, running an advisory firm today. I would hire for that as well as invest in it.
I definitely think this idea of the persuasiveness of the investment approach. If Mark is really that good at explaining the investment process and ben isn't, we should get Ben learning from Mark and we should be distilling the best explication, the best explainer of our investment process. We should make that more universal.
And then the other things I think would just be basic building block things. First of all, this general financial knowledge question. You see that in all these surveys, that after trust and transparency, the issue is having access to an expert. The depth and roundness of the training for the individual. I do think with AI though, and especially in larger firms, it's not about you knowing everything, you the personal advisor. It's about you being seen by the client as a person who can get the information that I need. Not immediately at your fingertips, but through other channels or through an AI chat box that you then filter to make sure it's correct.
The fourth thing I would say based on the whole body of research is this transparency around fees and value added. I think you just need to be absolutely transparent in dollar terms and percentage terms about how much people are paying. But I also think you don't just leave it at that. You actually have to talk about your value added in terms of capabilities and not just the numerical we rebalance your portfolio, "We did this. We did that." Not the sort of transactional piece, but also the emotional planning, organizing piece. I think those are pretty basic ideas that come out of this research that seem pretty fundamental to the future of financial advisors.
Ben Felix: Really deep, interesting stuff. All right, Steve, our final question for you. How do you define success in your life?
Steve Utkus: I'm actually going to answer different questions slightly, which is how do I define success in a particular point in life, which is transitioning from work to retirement, which I did four years ago. That's a very specific thing that I have lots of knowledge in. I have really three lessons I would share with your audience. The first one is timing and knowing when to leave. I remember a colleague of mine, I was giving actually a going away retirement speech for him years ago, about a decade ago. I talked to him and he said, "The Critical thing in timing is knowing when to leave the party. You want to leave the party when it's in full swing and everyone's having a good time. You don't want to hang on until three o'clock in the morning."
That's my first lesson is people contemplate this – for you guys, a future transition to retirement is really thinking carefully about the timing and leaving the party when it's in full swing. The second is this idea, particularly important successful attributes, is defining a life independent of your prior work. It doesn't mean necessarily not doing things. For example, I'm doing this work, which I really enjoy. I keep abreast of financial information. I'm engaged with some academic research projects, as I mentioned. I'm working on some stuff where I'm actually on the client side. But you have to define a life that is independent of your work. The work elements can be part of it and they can continue, but it has to be independent.
And then the third element of success I would point to the people I see who do this particularly well is dealing with health and medical issues which arise as you age, with grace and perseverance. We've all seen these ads, happy couple walking in the sunset on a beach and it's all about financial planning. But I don't know. He might have cancer and he might have heart disease. And that whole health issue is often hidden in discussing the aging process.
I have a personal motto I use from Jack Bogle, who always used to say to the Vanguard crew, which is the term used for Vanguard employees, he quoted this old English expression, "Press on regardless." And so, as you age and you move into this third phase of life, pressing on regardless, that's my motto not just the health issues, but just this new phase of life.
Ben Felix: That was an incredible answer to the question, Steve.
Steve Utkus: And it was relevant. I mean, it's directly from my experience in the past couple of years.
Ben Felix: All right, Steve, this has been an incredible conversation. We covered a lot of ground. Your research is super interesting and relevant to our audience. So we really appreciate you coming on the podcast.
Steve Utkus: Well, I really appreciate being here. And I really enjoyed dipping into all the history, the extensive history of your podcast. It's really such terrific information available. And it actually changed some financial aspects of my financial plan based on your podcast, which I find exceptionally valuable.
Ben Felix: Super cool to hear that. Very cool. And we do appreciate it.
Steve Utkus: Thanks for having me.
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.
Participate in our Community Discussion about this Episode:
Books From Today’s Episode:
Bogle on Mutual Funds — https://www.amazon.com/Bogle-Mutual-Funds-Perspectives-Intelligent/dp/111908833X
Papers From Today’s Episode:
‘Political Cycles and Stock Returns’ — https://papers.ssrn.com/sol3/papers.cfm?abstract_id=2920401
‘The Diversification and Welfare Effects of Robo-advising’ — https://www.sciencedirect.com/science/article/abs/pii/S0304405X24000928
‘Who Benefits from Robo-advising? Evidence from Machine Learning’ — https://papers.ssrn.com/sol3/papers.cfm?abstract_id=3552671
‘Human Financial Advice in the Age of Automation’ — https://papers.ssrn.com/sol3/papers.cfm?abstract_id=4301514
‘The Misguided Beliefs of Financial Advisors’ — https://papers.ssrn.com/sol3/papers.cfm?abstract_id=3101426
Links From Today’s Episode:
Meet with PWL Capital: https://calendly.com/d/3vm-t2j-h3p
Rational Reminder on iTunes — https://itunes.apple.com/ca/podcast/the-rational-reminder-podcast/id1426530582.
Rational Reminder Website — https://rationalreminder.ca/
Rational Reminder on Instagram — https://www.instagram.com/rationalreminder/
Rational Reminder on X — https://x.com/RationalRemind
Rational Reminder on TikTok — www.tiktok.com/@rationalreminder
Rational Reminder on YouTube — https://www.youtube.com/channel/
Rational Reminder Email — info@rationalreminder.ca
Benjamin Felix — https://pwlcapital.com/our-team/
Benjamin on X — https://x.com/benjaminwfelix
Benjamin on LinkedIn — https://www.linkedin.com/in/benjaminwfelix/
Cameron Passmore — https://pwlcapital.com/our-team/
Cameron on X — https://x.com/CameronPassmore
Cameron on LinkedIn — https://www.linkedin.com/in/cameronpassmore/
Mark McGrath on LinkedIn — https://www.linkedin.com/in/markmcgrathcfp/
Mark McGrath on X — https://x.com/MarkMcGrathCFP
Steve Utkus on LinkedIn — https://www.linkedin.com/in/steveutkus/