In this episode I speak with David Berns, co-founder and CIO of Simplify ETFs and author of the book Modern Asset Allocation for Wealth Management.
Our conversation centers around the idea of what it means to build a portfolio for a human being. This concept arises both technically and philosophically in David’s work, where he emphasizes the importance of higher return moments in portfolio optimization, but goes about achieving this end through more holistic risk preference analysis.
David expands upon the ideas of risk aversion, loss aversion, reflection, and how both our personal balance sheets and our standard of living expectations impact the portfolio choices we should be making. While there is no straight forward prescription, David emphasizes that simply being aware of these different factors can help advisors select more appropriate portfolios. And, hopefully, as the toolkit of investment options expand, adopt exposures that can better shape investor return distributions.
I hope you enjoy my conversation with David Berns.
Corey Hoffstein 00:00
321 Let’s dance. Hello and welcome everyone. I’m Corey Hoffstein. And this is flirting with models the podcast that pulls back the curtain to discover the human factor behind the quantitative strategy.
Corey how Hoffstein is the co founder and chief investment officer of new found research due to industry regulations he will not discuss any of new found researches funds on this podcast all opinions expressed by podcast participants are solely their own opinion and do not reflect the opinion of newfound research. This podcast is for informational purposes only and should not be relied upon as a basis for investment decisions. Clients of newfound research may maintain positions and securities discussed in this podcast for more information is it think newfound.com.
Corey Hoffstein 00:50
This season is sponsored by simplify ETFs simplify seeks to help you modernize your portfolio with its innovative set of options based strategies. Full disclosure prior to simplify sponsoring the season, we had incorporated some of simplifies ETFs into our ETF model mandates here at New Found. If you’re interested in reading a brief case study about why and how visit simplified.us/flirting with models and stick around after the episode for an ongoing conversation about markets and convexity with the convexity Maven himself simplifies own Harley Bassman. In this episode, I speak with David Burns, co founder and CIO of simplify ETFs and the author of the book modern asset allocation for wealth management. Our conversation centers around the idea of what it means to build a portfolio for a human being. This concept arises both technically and philosophically in David’s work, where he emphasizes the importance of higher return moments in portfolio optimization, but goes about achieving this end through more holistic risk preference analysis. David expands upon the ideas of risk aversion, loss aversion reflection, and how both our personal balance sheets and our standard of living expectations impact the portfolio choices we should be making. While there is no straightforward prescription, David emphasizes that simply being aware of these different factors can help advisors select more appropriate portfolios. And hopefully, as the toolkit of investment options expands, adopt exposures that can better shape investor return distributions. I hope you enjoy my conversation with David Burns. David Burns, welcome to the show. excited to have you here for last episode of the season. This is it, it is oh, final guests coming in coming in strong. And I want to thank you. By the way, I should note, as I note in the intro, simplify has been a very generous sponsor this season. So thank you guys very much for that very excited to have you on.
David Berns 03:00
We genuinely love what you do for the financial community, educating helping everyone build better portfolios, so we really appreciate you. So we glad to sponsor you, man, we love it.
Corey Hoffstein 03:10
I appreciate that. And I should mention that you and I actually had this scheduled before simplify said that they would sponsor the season. So this is not like a quid pro quo. I’m very excited to have
David Berns 03:21
yours. Thank you for making that public. I wanted to make that very clear.
Corey Hoffstein 03:26
This is this was not a buy your way on I don’t do that on my show. So let’s start off with it’s 2019. And you publish a new book called Modern asset allocation for wealth management. Yeah, and I’ve tried writing a book, it is a complete pain, I have the utmost respect for anyone who gets a book out of them. Anyone who’s written a book will tell you, you don’t do it for the money. So what drove you to write this book? Yeah, so
David Berns 03:53
I spent my whole career building asset allocation systems for advisors, and just sort of reflecting on what’s out there and what the norm is and how to build portfolios, for individuals for institutions, even the stuff I had built previously, I just didn’t think that people had a really easy way to systematically build portfolios. How do you take a human being take their psyche, their subconscious, map it into a portfolio that respects some kind of risk profiling, like how much real estate should someone have based on sort of their subconscious? And then also, how do you systematically account for someone’s balance sheet? How can an advisor actually run an optimizer? optimizers are incredibly tricky, as you know. So there just wasn’t really solutions for advisors to build super customized portfolios, people, they just in general, take four model portfolios been some up in a bucket, and there are ways to do it. And that’s what I wanted to do is give people the tools to systematically build portfolios that really scratched well below this Rufus and got into some really interesting questions
Corey Hoffstein 05:02
right out of the gate in the book. One of the key points is that modern portfolio theory is really just mathematically it’s a second order approximation of sort of expected utility maximization. And a lot of your book, at least the initial chapters really hinged on the idea that to truly capture investor preferences, we actually need to include those third and fourth moments, right skew and kurtosis. Can you expand on that a little? Why is that so important for capturing maybe the full picture of someone’s utility?
David Berns 05:37
Yeah, I mean, it all stems from what we learned in Prospect Theory. That’s what it’s about before Prospect Theory, risk aversion was king, a power law utility was king. And then once Prospect Theory came along, we learned that human beings, their risk profile is not well defined by a single powerful utility with a singular parameter of risk aversion. And right when you start to incorporate prospect theory into the portfolio construction process, right, once you have a more advanced utility function and a power law utility, you start to have the potential for higher moments to become important. If you’re just doing a power law utility invariance, it doesn’t matter if your the assets in your portfolio have a lot of skew, kurtosis, Kolsky, whatever. But once you use a utility function that is sensitive to higher moments, like incorporating loss aversion, you will be sensitive to skew all of a sudden, that’s how you get to the point where you say, okay, higher moments matter is post Prospect Theory. And the entire world still lives in invariants land, whether they say it explicitly or not, everyone probably uses like a heuristic of five volatility buckets and an efficient frontier or even, you know, 6040 as like the middle vol bucket or whatever, very heuristic, end of the day, that comes from invariants. And so 95% plus of the world is somehow explicitly or not still just living in me various land. And they haven’t updated their systems to count Prospect Theory, which is from the 70s. So we’re good, almost 50 years behind.
Corey Hoffstein 07:16
One of the points that stuck out in your book to me is one of these. Wow, that’s really obvious, I can’t believe I’ve never thought of it is this idea that investors want to maximize their odd return moments, and minimize the even return moment, so they want a high expected return. And they want a high right skew, but low volatility low kurtosis, which to me was like, wow, that’s the most obvious thing I’ve ever heard. But I
David Berns 07:43
love it, that you isolate that it is like I’m with you. It’s, it’s one of the just more beautiful, simple as long as you’re you’re ignoring that reflection with just loss aversion added on, just yeah, I’m completely with you a deeply insightful and simple concept. And it really starts to just make the wheels wheel wheels turn right in your head, right? Because what you can do is, let’s say you’re doing manager due diligence, and you want to compare a manager to a benchmark, you can literally just compare how each if put into a portfolio would change the skew of the portfolio, and it’s a skew goes up. Everyone should prefer that just a fact, no matter how much loss aversion or not just pops out, like you said, so I think you can just turn it into really cool tools and analytics really, really quickly. Love that. I love that you isolated that.
Corey Hoffstein 08:36
Well, you didn’t hear my follow up question you might not love so much. Okay. So this is more of a philosophical question, which is the idea that if these are truly sort of general preferences that almost everyone holds these, and that certain asset classes allow for a better expression of higher moment trades? Wouldn’t we expect prices to ultimately clear at a level where sort of moment preferences offset each other? So for example, if the expected return would drop low enough, that we’re ultimately indifferent to one asset versus another that might have, you know, a higher expected return but lower skew?
David Berns 09:19
Yeah, it just ultimately depends on the agents in the market, how greedy they want to be on returns on the first moment. If everyone’s doing covered calls or selling puts, or if everyone’s excited to take on these units, risk premia, then yeah, you can really have the value of some of the higher moments be flipped and could long term add value. Oh, and by the way, they’re sort of continuous compounding effect with things like convexity that can improve your geometric return as opposed to arithmetic so complicated question, but I encourage people to read things like liquidity cascades and really do their homework on The markets and market structure and participants before they jump to a conclusion,
Corey Hoffstein 10:04
do you think it’s possible that with so much emphasis on portfolios that ultimately grew out of modern portfolio theory that skewness is actually potentially underpriced in the market? That there’s too much emphasis on that first moment and second moment?
David Berns 10:21
Yeah, exactly. Because it gets to the point we just covered because if you’re focusing just in the first few moments, there is no way to balance out that third moment, because everyone is just going after that first moment. Right. So you can just imagine, all this massive negative skew just coming more and more into the market, everyone’s in their median variance, looking at the first moment return in that process. And that Yeah, that’s exactly right. That’s spot on. I think that’s absolutely what we’re seeing. So let’s
Corey Hoffstein 10:47
get back to the book. Let’s dive back in, you propose an asset allocation framework that maximizes the utility function with not just one, but actually three dimensions of client risk preferences. So I was hoping you could walk us through what those dimensions are? And ultimately, why you think they’re important enough to complicate things.
David Berns 11:12
Yeah, great question. So everyone’s familiar with risk aversion. In general, it’s sort of the measure of the trade off between expected returns and variance, are you willing to take a very, you know, something with a lot of variance for little extra return. But like we mentioned earlier, when Prospect Theory came along, we realized that to build a portfolio properly, you need to account for someone’s loss aversion, which is just their asymmetric preference between gains and losses. That’s brand new, right. And then there’s also reflection, which is a pretty strange behavior, where when someone is in the loss domain, they’re willing to actually be risk seeking, as opposed to risk averse and utility function curves up. And I’ll give you one quick example of why these things are so important. If you aren’t an optimization with just someone’s risk aversion, and then you run an optimization with someone’s risk aversion as well as some moderate amount of loss aversion, you’ll generally see differences in the equity allocation of maybe 20 30%. So instead of a 6040, we might be looking at a 9010. Portfolio once you can philosophers, and it can get a lot worse. I mean, I was a moderate amount of loss aversion. So I think there’s a lot of liability if we as advisors are prescribing a prescription that is incredibly far off from what people deserve. And I think we saw a lot of that in 2008, how many advisors got fired lots of reasons, that could have happened, but I think one of the reasons is that there was portions of the population, right, not everyone has extreme loss aversion, for instance, but maybe five to 10% of the population does. And when those people had 40%, drawdowns in their portfolios, they probably fire their advisor, because really, they probably could only start with like a 20% drawdown or a 15% drawdown. So I think those are and also not just you, it’s not just advisors only fired, you want your client to have a good experience, you want them to be able to sleep at night. And so I think we should all be trying to think of diagnosing someone’s risk profile in as medical way as possible. And researchers have tried to give us the answers to that over the last 50 years with Prospect Theory and stuff like that. And we’re mostly ignoring it. But I think we can be a lot more medical and precise, and build portfolios that are way more suitable.
Corey Hoffstein 13:40
One of the things that comes out of a lot of research on investor behavior is that risk preferences seem to change over time. I think there’s a good reason risk preferences can change over time, right? Your personal balance sheet can change, you could change careers, you just had a different point in life. And so that’s, I think, sort of the expected reason things change, but you also find that it’s responsive to what’s happening in markets. Curious as to your thoughts of is that only because perhaps we’re not using a robust enough framework to really capture investor preferences. For example, if we’re only looking at risk aversion and not capturing loss aversion and reflection, that it’s going to look like risk aversion changes a lot. Whereas if we had a more holistic view, they’d actually potentially be more stable,
David Berns 14:33
potentially. And part of the reason I say that is because if someone has extreme loss aversion, if their loss aversion changes in time, a little bit around a pretty extreme number, the portfolio allocations aren’t going to change a lot. So it really depends on sort of where you are in terms of these risk profiles, these three dimensional risk profiles, and how sensitive the portfolios are to the actual levels. But at the same time, we expect all three of these parameters Prospect Theory to change circumstantially for people. So I actually think one of the huge value adds that advisors can provide their clients is not just a better understanding of who they are by actually diagnosing three dimensions of risk preferences, which is another key value proposition, but also monitoring it in time really carefully. I don’t think we do annual checkups for our medical health, I think we should be on a similar type of timeline with our risk profiling. And there are huge life changes, maybe you want to sort of retest as well. But it’s actually it’s not a nuisance. It’s a value prop for advisors. And it’s something clients will love clients love understanding themselves. Clients love doing a loss aversion questionnaire, and finding out about themselves. When you say, Hey, your loss aversion relates to the kind of jobs you like, or the kinds of iPhone cases you use on your phones or, and when you start to make those parallels and explain it. They genuinely enjoy it, they find, I think they appreciate it. And it helps to build a better connection with advisors. And then just keeping track of it in time is critical. And I think a welcome process for them.
Corey Hoffstein 16:18
So we’ve got these three measures of client risk, right, risk aversion, loss aversion, and reflection helped me tie these back to the return distribution moments. If a client has, say moderate risk aversion and high loss aversion, does that translate back to some expression for their preference around skewness, or kurtosis?
David Berns 16:41
Yes, so in general, more loss aversion you have, the more version, you’re going to have to negative skew. And in general, reflection kind of does the opposite. But the thing you have to remember is that having loss aversion will also give you a preference to not have volatility. So if you run an optimizer with loss aversion, for instance, it’s not just gonna go attack your negatively your highly negatively skewed assets, it’s also going to attack your highly volatile assets. So it’s a little tricky, little complicated. But if you just visualize it, loss aversion doesn’t like losses, you can kind of get there in your head and say, Okay, it’s probably not gonna like volatility to some degree, even though it’s a metric measure. But in general, more loss aversion, the more conservative the portfolio is going to be. And the more reflection you have, the more aggressive it’s going to be. So if you really want to stick with your buckets of five volatility stay with sort of a mean variance efficient frontier, your clients with more loss aversion should go to the more conservative side, and then your clients more flexion would go to the more aggressive side. So you just want to take those features, if you’re testing for those without running an optimizer, and you just want to fit them into the buckets. That’s how you do it.
Corey Hoffstein 18:00
For talking about these three risk preferences, anyone who’s taken a client risk questionnaire is probably somewhat aware of the types of questions that get asked for risk aversion. But I suspect many people haven’t seen the type of questions that might get asked for loss aversion or even reflection. Would you mind providing like an example of okay, here’s a question for risk aversion. Here’s how that question changes or the type of question we need to ask for loss aversion and then similarly for reflection. Yeah, so
David Berns 18:31
reflections? Little tricky. Let’s just do loss aversion reflection gets a little confusing, I probably wouldn’t get it right. But in the book, and at Portfolio designer, the software we build for it. We do it all sort of gambling lottery style questions. All right. So when you set it up like that, we’re trying to be very numerous, and really precisely map out mathematically the curvature of your utility function. So that’s why it’s all set up in this very numerous way. And in theory, because we’re being so explicit and numerate, precisely mapping to utility function, we should have what’s called high validity, right? In the psychometric testing world, we want very high validity of our questionnaire mapping to the thing we’re trying to discover. So let’s do the example of loss aversion. We can diagnose yours live here if you want. I flip a coin. Okay. And heads you lose $10. Tails, you make 20. All right, so lose 10 versus make 20. Do you accept that gamble or not? You don’t have to take it.
Corey Hoffstein 19:36
Yeah, I’d accept that gamble. Okay. All right, great.
David Berns 19:39
We just diagnose it, you don’t have extreme loss aversion. If you did not accept that gamble, you would be at the highest limit of loss aversion the way we’ve ended up in the book. So all right, let’s do this one out.
Corey Hoffstein 19:53
I hate interrupting in my podcast, but I have to interrupt you because it brings up a question for me which is good. For 10, or $20 isn’t really meaningful to me from a standard of living perspective. Yeah, I have to interrupt and say, if you had said 100,000 versus 200,000, I would suddenly say, Okay, I’ve got a different answer.
David Berns 20:16
Love, it feels like a settled question, but it’s a really good one. And I didn’t want to ask you like, what your annual income was on the pod. So but that’s actually how you would start the process off. You actually start with someone’s annual income, and then you size that that’s accordingly to exactly address this problem. Got it defaulted now, but okay. So let’s stick with that a bit. But it’s a great point. And it obviously, it’s very, it’s very critical. So. So let’s just go go again now. Right. So same question, lose 10. But now gain 15? Are you still interested in that? We can go quicker? Yeah. Yeah, probably lose 10? Get 11? Yeah,
Corey Hoffstein 20:58
I probably still take that. Alright, so
David Berns 20:59
you basically just shown that you have like no loss aversion, which doesn’t surprise me as a quant quants generally tend to be more robotic probabilistic calculators, and less emotional. So does that line up with what you assumed by yourself to have very minimal loss aversion?
Corey Hoffstein 21:18
Well, it’s funny, the last 110 I thought you’re gonna go 10 and 10. And I was thinking to myself, well, mathematically, there’s really no reason for me to take it. Yep. But if I was just sitting around with a friend, really bored, and we just decided to keep flipping coins and betting with each other, I could see myself doing it for the fun value.
David Berns 21:37
But what a minus 20,000 plus 20,000.
Corey Hoffstein 21:39
Probably not happening. Yeah, yeah. Yeah.
David Berns 21:43
So we’re isolating a really important feature. If you don’t phrase the problems, right, you will miss diagnose. You have to be incredibly careful. There’s a whole world of psychometric testing, like how do you think the SATs come about? I mean, this is not just some questions thrown together. Like there’s a whole field of psychometric testing, and you get into validity, reliability. And there’s way more potential problems than we’re talking about right now. am I offering this question as a one time gamble? Is it a repeated gamble get you into the whole world of Samuelson’s fallacy of large numbers like a 50 year research that’s gone on there? So there are lots of things embedded in this. And everyone took this risk questionnaire created by Grable in 2025 years ago, whatever it was, if you’ve read the papers on it, there are questions on risk aversion, there’s actually a section that’s called loss aversion. And there’s other types of risks. And they try to pin up the different textures, but there’s no connection to utility function. And there’s no realization and that simple questionnaire where you like sum up scores, that loss aversion can have a way bigger effect on the utility in the portfolio than risk aversion. So if you’re incredibly loss aversion, if you incredibly loss averse, simple questionnaire where you just sum up scores, you’ll be sort of binned in a group with the risk aversion questionnaires, and you won’t get this very nonlinear effect. And you can be very misdiagnosed. But there’s tons of things going on here like that. And I would just love the entire industry to take a step back and really ask if what we’re doing makes a lot of sense, and is really valid, right? Like, have you ever seen a questionnaire who where the questions change based off your income, everyone gets the same questionnaire. I mean, just just that alone, ask someone making $5 million a year, what they think about a $10 bet. It’s just it’s not relevant. It’s not a valid, not a valid question.
Corey Hoffstein 23:42
So let’s jump back into the technical for a second, we’ll jump back and forth, technical, philosophical. So modern portfolio theory would argue that what we should do is construct sort of the highest Sharpe portfolio and then lever it appropriately to meet our risk target, right. Similarly, under your framework, just expanding the moments, we could sort of create this four dimensional hypersurface find the point that maximizes utility and then lever that portfolio. You in the book sort of explicitly avoid this efficient hyperplane approach in lieu of a questionnaire that comes up with these sort of three risk preference diagnostics and use that in the utility function. And I’m curious as to what drove you down that path?
David Berns 24:33
I think it’s because of the way I wanted to keep optionality open for how we handled risk capacity. We haven’t talked about investors balance sheets yet. And in the book, there’s a systematic approach to taking someone’s goals and balance sheet taking their assets, taking their liabilities, and integrating that with the risk preferences. And once you start to either aggregate parameters, or sort of find one portfolio, and then kind of just lever it, and you’re losing information. So if you want to moderate reflection different than loss aversion based off risk capacity, you’ve lost that ability. So that’s the thing, we have a whole nother topic to talk about now on sort of how how to handle risk capacity and how to bring it in systematically. But that was another huge pet peeve of mine. And why I wanted to write the book was because people weren’t talking about risk capacity always at least a decade ago or two decades ago. And they surely if they were, they generally weren’t doing it in a really robust, systematic way. Right, it was more one off. And so I was deeply intrigued by the idea of doing something systematic, and giving advisors a robust process that they could trust.
Corey Hoffstein 25:54
Well, it’s like you were looking at my screen or something, because I was the next question I wanted to ask you was about this measure that you introduced in your client questionnaire about this, you know, about their balance sheet, right, you I think you call it the standard of living risk. And I was hoping you could expand on what that factor is, and ultimately, what you’re trying to capture with it. Yeah.
David Berns 26:15
So the whole idea here is basically to do an asset liability matching for an individual can do for institutions, but do for an individual. So you look at someone’s assets for all time going forward, you look at someone’s liabilities for all time going forward, you discounted all back to today, using a risk free rate. And based off that, you can see how much what we call discretionary wealth they have if they have lots of assets, and very minimal liabilities, the difference there is that discretionary wealth and have tons of discretionary wealth. And in that case, what I’ve proposed in the book is they can then express their risk profile as they please. Okay, it’s a luxury to express your risk preferences. At the end of the day, we need to have you surviving, you have goals you have retirement needs, you can only work for so long, that should be the first requirement. And then as you have if you can survive, and things are going well, then you can take on risk. But if not, let’s say you have no discretionary wealth, your assets, just meet your liabilities, just buy treasuries, and just get the job done. So in this approach, I’m proposing that we focus on 100% success rates in Monte Carlo simulations, as opposed to the industry norm, which is running Monte Carlo for this spin, I put you in with whatever your assets and liabilities are. And as long as you have like a 75% success rate in the Monte Carlo, you’re good. And that whole thing is premised on the idea that your goals and your expectations for life are flexible. If you’re planning on spending $40,000 a year to live in retirement, is that flexible? Or working till 60? Is that flexible? Can you work longer? I think that’s a deeply interesting social question. Social Experiment, I don’t think we’ve, we have date on it. Because we we shifted from defined benefit to defined contribution. And I don’t think we really understand yet how flexible people’s goals are. So does someone really want to be confronted in 10 years from now? Hey, look, we had assumptions, we had expected return assumptions for asset classes, we thought it’s gonna be 8% a year for equities, it’s been flat for 10 years, we have to have a really hard conversation right now. And you need to work five more years, or we need to cut your expected spending in half in retirement. We don’t know the results of that social experiment yet, but we will find out eventually. And I deeply believe that it’s a much more appropriate moral approach to say, let’s focus on 100%. And then slowly give various to the Monte Carlo simulation, right, give volatility to that multicolor cloud, as your discretionary wealth goes up.
Corey Hoffstein 29:09
One of the really interesting questions you posed to me during a prior conversation. And it was really meant as a rhetorical question was, what does it mean to build a portfolio for a human being? I thought that was a really profound, deeply interesting question. And while it was meant to be rhetorical, I’m just gonna flip it on you and ask you what does it mean to build a portfolio for human being?
David Berns 29:35
It’s a funny question. It’s like it’s deep and not deep. I think it’s two parts. I kind of just hit on it, right? Why do we invest? Why do we even invest? There’s a goal. So I think that has to be paramount. And that’s to some degrees because of the way people approach the success rate concept and the flexibility of your goals. I think that’s not being addressed properly. I think Monte Carlos and our way better than they were in the last year, a decade ago, two decades ago, I’m still critical on the success rate below 100% concept. But then also taking the best understanding of that we have today around human behavior on risky, risky views, risky propositions, which is prospect theory or some derivative of it, implementing that, as opposed to a simple mean, variance, power law construction, I think that should be done as well, I think those are the two things right, respecting someone’s goals first, and then really trying digging as deep as we can, around their risk preferences, and bring that into the portfolio. Even though a lot of times, it might not always, it’s not necessarily going to give you better expensive returns. That’s kind of a challenge for an advisor, it’s a principal agent problem. So I’m gonna give you something that respects loss aversion. And now you’re only going to make 4% a year instead of your buddy next door making eight, I have to have that conversation. So I get it. But it feels like feels like a more responsibility to fight that fight.
Corey Hoffstein 31:10
It strikes me not even just as the person next door, but the market at large. I mean, you have any conversation with any advisor, and they’ll tell you that tracking error is is a significant problem. They have clients who are in moderate to conservative portfolios complaining about why they’re not achieving what the s&p 500 is, I mean, it’s sort of the cliche complaint to hear, but that happens. And so I would imagine that when you start to introduce these higher moments, either explicitly or through the expression of these risk preferences, you might end up with some pretty weird looking portfolios, certainly not going to end up with something that probably looks like 60% stocks, 40% bonds, so correct me if I’m wrong. So how do you think behavioral tracking error comes into all this when the headline news is always about? This is what the Dow Jones or the s&p or the NASDAQ has done on the year?
David Berns 32:05
Yeah, it’s a great question. I think the most important thing to realize is that extreme parameters for loss aversion and reflection, definitely reflection, reflection is generally very mild, they’re very, there aren’t a lot of people with extreme reflection, small fraction of the population loss aversion, there’s pretty extreme, there’s just breaching loss aversion by 20 25% of the population. So I think the answer there is, the more extreme someone’s profile is, the more you have to just unleash yourself from the tracking error constraint. And you just have to tell them their diagnosis, you just have to be honest, be like, Look, you don’t have to work with me. And we’re going to have some tricky days ahead. But you have extreme loss aversion, and we need to talk about it. And that’s what I think the conversation should be like. But for all the other cases. So that’s maybe 80% of the population plus, I don’t think you’re going to be in overwhelming tracking error situations. So I think that’s a key component.
Corey Hoffstein 33:11
I want to talk about what happens when some of these parameters are potentially in conflict with one another. So the one that immediately jumps to mind perhaps is a scenario where risk tolerance and standard of living desires conflict, maybe someone has very high risk tolerance, but a standard of living that can’t support that risk tolerance, deal with those sorts of scenarios.
David Berns 33:38
Yeah, it’s the same type of issue that we just talked about. If an advisor wants the stand his ground and say, the SLR and your goal is critical, then that’s going to take priority. It’s just a conversation you have to have with the client, and ultimately, they can choose. But if it’s systematic, and it’s very clear what’s going on, I think it’s a relatively straightforward conversation to have. And generally, I think, a good one. Because, you know, clients know, when advisors aren’t doing much, they’re not stupid, they know it, they feel it. And all of these things we’re talking about, I think they feel that advisors are really digging in. And they can sense it. So I think it’s overall a good value prop.
Corey Hoffstein 34:27
What about risk aversion and loss aversion? I’m trying to imagine someone who is perhaps, risk seeking but highly loss averse, do you tend to find that these are actually correlated variables? Or can they really be independent in their diagnosis of someone’s behavior?
David Berns 34:44
Yeah, I think it’s generally a high level of correlation. But correlation is a population measure. And there are outliers. That’s another good point to kind of back to the question you asked earlier, because there’s a correlation. There. that sort of will minimize the tracking error of the loss of verse like the accounting philosophy version versus not, that tracking error will get minimized a bit because of the correlation. So if you’re coming from a highly risk averse portfolio, and then you transition someone into one that’s highly risk averse, also highly loss averse, there wouldn’t be as much tracking your as you might think. But there are outliers, I have a bunch of survey data, I can show you the scatterplot. And there are outliers to that it’s not perfect correlation across everyone the population.
Corey Hoffstein 35:29
So as we go to build portfolios that take into account these tire moments, or these multiple dimensions of client risk profiling, are there other asset classes or types of exposures that advisors should be thinking about, or investors themselves should be thinking about, to help better fit this sort of multi dimensional utility function? You know, I guess I’m thinking, yeah, do we have the tools to do it?
David Berns 35:57
That’s a beautiful question, skew kurtosis. These are third moment, fourth moment, and they’re cubed in the equations, and they just get their impact is just more and more muted as we go to higher moments. So if you aren’t an optimizer and you want an asset, to really to be introduced to the portfolio purely because of its skew, it has to have a very significant skew. And that has not been a focus historically. But now with the ability for betas to have a lot more derivatives, these things are on the table now. And that’s exactly what we’re focusing on simplify, I know you’ve greatly focused on that, as well, building things with a lot more, you know, modification of the SKU, taking negative skew distributions and making them more positive and stuff. It has to be significant, or it has to be priced really well. You can create convexity in a portfolio building blocking an asset, if you can do it, even if there’s marginal skew change if you have a positive expected return shift that would play out really well in a portfolio asset allocation. So it’s a wonderful question. And I love that people are tackling it, you know, for example, simplify. And it’s really fun to see how portfolios are evolving. And exactly what the trade offs are, you know, how much positive expected return shift do you expect, in a positive convexity, scenario, etc? complicated question, but we’re building the products, see how it all tests out of sample. I know you
Corey Hoffstein 37:22
build software, and you have a website that allows people to sort of try their hand at these risk questionnaires and build portfolios around it. This is something you’ve been doing for years now. What do you think the biggest surprise has been to you in building portfolios, maybe in comparison to your more standard off the shelf 6040 type models, when you start to take these things into account? What are the most surprising differentiators
David Berns 37:49
for me, and this might not be the where you might expect me to go with this. But for me, the biggest surprise is whenever you want to run an optimizer, I actually run one with some set of assets, you can’t have assets be very related, that will blow up your optimizer. And so as a part of this entire value proposition, a systematic process to diagnose utility function, bring in balance sheet and then actually turn an optimizer crank on the set of assets. You need your assets not to be very related. And so as a part of the book and the software, there are tools to diagnose assets, and make sure they’re not very related. And it’s shocking how many assets are really related. So there’s line item proliferation in portfolios and their portfolios out there with 50 line items. And it’s completely unnecessary. And I challenge everyone to kind of really think about what an asset is, what the value prop is, keep it simple. If you’re not really modifying return by being a performance asset, or volatility, like lowering volatility of being a diversifying asset, or raising skew from negative to positive, which is another form of diversifying asset. If you’re not doing one of these things, then what’s the value prop of the asset that to me is probably the biggest takeaway from this whole systematic process that we built.
Corey Hoffstein 39:16
So the last question that I’m asking everyone this season, David, is, and it’s interesting to get the perspective for me over time because I’ve been recording this season over several months, is around COVID. And the question is starting to see the light at the end of the tunnel, though, the Delta variant has maybe thrown a wrinkle into that and I know you just moved to Los Angeles where it seems like the indoor mask mandates may have just come back so so there you go, some speed bumps in the road, but it does seem like hopefully there is a light at the end of this tunnel. What are you most looking forward to as life gets back to normal?
David Berns 39:55
I’m excited for everyone to just be able to freely get out and about and do the things that they love doing that are just hard to do now, I think everyone has been frustrated with the lack of mobility. And people are social beings and they love to congregate and they love to explore. That’s, that’s human nature. And I just want everyone to be able to just get back to that.
Corey Hoffstein 40:18
Well, David, this has been fantastic. Thank you so much for joining me.
David Berns 40:21
Thanks a lot, man. Love it.
Corey Hoffstein 40:27
If you’re enjoying the season, please consider heading over to your favorite podcast platform and leaving us a rating or review and sharing us with friends or on social media. It helps new people find us and helps us grow. Finally, if you’d like to learn more about newfound research, our investment mandates mutual funds or associated ETFs. Please visit think newfound.com. And now welcome back to my ongoing conversation with Harley Bassman. Many people will remember credit default swaps as being the instrument that was at the center of the 2008 credit crisis. And so Harley, my question to you is what is a CDs payer option? And why do you believe that investors should have access to it in an ETF?
Harley Bassman 41:15
Oh, God didn’t get in trouble here and talk about guns don’t kill people, people kill people. CDs by themselves is not a bad instrument. If they are properly used, and properly managed, and properly regulated, I would lay a good portion of the debacle really starting, you know, two or three or four. But where else we blew up and I went no nine at the Feds feet at the government’s feet at not basically getting rid of Glass Steagall, when that happened, you basically allowed investment banks, you also become commercial banks, so they could borrow money at the government rate the FDIC insured deposits and invest those monies in speculative instruments, the Glass Steagall was still in place, I still be at Merrill Lynch, because they were too blown up CDS is just the same thing as a futures contract. But instead be based on gold, or on an interest rate, it’s based upon the spread of a basket of credit bonds. Nothing wrong with that. And it’s a very liquid market. And it’s a great way to do risk transfer, because trying to buy or sell 50 different bonds is very difficult. And you can’t short small issue bonds. So having this 50 or 100 bonds into a basket, agreeing about the basket, having a termination agreement everyone agrees with and then trading that around in a way where margin is properly collected, is great for the market, not always low risk transfer. It also gives the market information that spread widens or tightens, it tells you there’s more or less risk in the market. And people are trying to move it around. options on CDs is not much more than options on gold, or bonds or stocks, it’s an option on an index, what we’re going to do is the is the process where we’ve kind of broken out the professional is the agreement and allowed civilians access to this market, we’re gonna allow them to go and buy options on credit. And if you own a large portfolio of credit instruments, or if you have exposure to credit and some manufacturer form, you are buying insurance. Now, is that a good thing? The answer is yes. Will you make money? The answer is unclear. Remember, when you buy life insurance, you don’t win when you die. So therefore, I’m not terribly abused the idea that I want to women I buy insurance, what you should be doing is saying what is my total profile financial and otherwise? How would it be hurt or not hurt if there was some kind of financial thing that caused credit spreads to widen or defaults to increase? And what might be willing to pay to insure against that the same way you buy insurance for your house, or your car, or your life or anything else you have? This is a reasonable decision. If it’s at a fair price. We’re gonna go and buy options on a liquid instrument options that trade at a reasonably fair price because there are liquid options. There’s buyers and sellers, these are quite bespoke instruments, and therefore it probably reflects the fair market price for risk. So the question is not should you buy it, but how much you should buy. I have plenty of insurance for myself personally, for my family. You might get a higher or lower deductible depending on your situation financially, but everyone has insurance for something in life. Because remember, insurance companies when they’re selling insurance, they’re selling it to 1000s or millions of people, they will experience the average risk, the average return, they will own the distribution, you personally, you are a binary event, it happens to you or it doesn’t. Therefore, you can’t achieve the total profile, the total distribution of a risk, you’re a binary risk, there are distribution risks. Therefore, you have two different risk profiles. It’s possible actually likely that both sides can win because the seller and the buyer have different risk profiles. And if the price is in between those two, both sides can win. And so these are rather reasonable products and for us to offer it in an ETF is nothing short of genius.