Clare Flynn Levy: Measuring Decision-Making Skill in Investing

Summary

Enterprising Investor Podcast Summary
Episode: Clare Flynn Levy: Measuring Decision-Making Skill in Investing
Host: Mike Wahlberg
Guest: Clare Flynn Levy, Founder and CEO of Essentia Analytics
Release Date: November 1, 2024

---

Introduction

In this episode of Enterprising Investor, the flagship podcast of the CFA Institute, host Mike Wahlberg engages in a profound discussion with Clare Flynn Levy, the founder and CEO of Essentia Analytics. Essentia Analytics specializes in behavioral analytics for professional investors and capital allocators. The focal point of their conversation revolves around Essentia's innovative tool—the Behavioral Alpha Score—and its implications for measuring and enhancing decision-making skills in investment management.

---

Understanding the Behavioral Alpha Score

Defining Behavioral Alpha

Clare Flynn Levy introduces the Behavioral Alpha Score as a pioneering metric designed to quantify an investor's decision-making prowess. Drawing from her decade-long journey as an active equity fund manager, Clare sought a systematic way to analyze and improve her own investment decisions. This endeavor culminated in the creation of the Behavioral Alpha Score.

"It is a measure of decision making skill. We call it behavioral alpha because in this day and age, alpha is very hard to come by harder than ever... you can generate sustainable alpha by making better decisions than other people make because humans are biased and flawed."
— Clare Flynn Levy [01:17]

The Genesis of Behavioral Alpha

Clare emphasizes that traditional sources of alpha, such as superior information or exceptional intelligence, are no longer reliable or sustainable. Instead, the Behavioral Alpha Score focuses on enhancing decision-making processes to achieve consistent outperformance.

---

Study Findings: Behavioral Alpha and Performance

Research Objective

Clare and her team conducted a white paper study to investigate whether the Behavioral Alpha Score could predict investment performance better than past performance alone.

Methodology

• Data Analyzed: 123 unique long-only equity portfolios.
• Time Frame: Daily holdings data from 2013 to 2023, encompassing 532 portfolio years.
• Primary Question: Is the Behavioral Alpha Score predictive of future performance?

Key Findings

1. Past Performance Non-Predictive:
   The study confirmed that past performance does not predict future performance. This aligns with traditional financial wisdom and serves as a foundational validation for introducing new predictive metrics.

   "Past performance was not predictive of future performance."
   — Clare Flynn Levy [03:51]

2. Behavioral Alpha Predictive Within the Same Year:
   The Behavioral Alpha Score showed a predictive relationship with performance within the same year. Managers scoring above 50 were more likely to outperform their benchmarks within that year.

   "Managers with scores over 50 were one and a half times as likely to outperform their own benchmark in the subsequent year as managers with score under 50."
   — Clare Flynn Levy [05:00]

Implications

While the score's predictive power within a single year is valuable, Делare highlights the potential for broader applications, suggesting future research could explore longer-term predictability.

---

Enhancing Traditional Models: Explanatory Power of Behavioral Alpha

Integration with Factor Models

Clare discusses how the Behavioral Alpha Score complements traditional factor models like the Fama-French five-factor and three-factor models. The score adds explanatory value to these established frameworks, providing a more comprehensive understanding of a manager's ability to generate alpha through decision-making.

"What we're saying is this actually can quantify some portion of that alpha piece so you can hone in that much better."
— Clare Flynn Levy [08:10]

Fund Manager Execution

Clare underscores that while stock selection is a critical component, the execution of these selections—how decisions are made and implemented—is equally vital. The Behavioral Alpha Score aims to quantify and improve this execution aspect, often categorized under the elusive "alpha."

---

Decomposing Investment Decisions

Seven Decision Types

Essentia Analytics breaks down an investment episode into seven distinct decision types to assess decision-making skills thoroughly:

1. Picking Decisions:
   Evaluates the difference between holding a specific stock versus holding the index.

2. Sizing Decisions:
   Assesses how the portfolio weight deviates from an equally weighted benchmark.

3. Entry Timing Decisions:
   Analyzes the timing of initial stock purchases and their efficacy over a one-month window.

4. Scaling In:
   Looks at the speed and efficiency of building a position after an initial entry.

5. Exit Timing:
   Examines when a manager decides to sell a position and the effectiveness of that timing.

6. Scaling Out:
   Evaluates the process and strategy behind reducing a position size.

7. Size Adjusting:
   Involves adding or trimming positions and assessing the value added through these adjustments.

Evaluating Decisions

Each decision type is tested to determine if it adds value compared to a baseline scenario, such as holding an equally weighted portfolio or following an average daily close strategy.

"It's about is this person adding value or destroying value? You know, if you want to look at did they pick the right stocks... is that actually adding value?"
— Clare Flynn Levy [17:00]

Equal Weighting in Scoring

All seven decision types are equally weighted in calculating the Behavioral Alpha Score. This approach ensures a balanced assessment, regardless of a manager's specific strategy or decision-making frequency.

---

Accounting for Liquidity Constraints

Challenges in Scaling Decisions

Clare addresses the potential impact of liquidity constraints on scaling decisions. Larger managers might face challenges in scaling in and out due to the size of their positions and market liquidity.

Empirical Findings

Contrary to common assumptions, the study found that poor exit timing and scaling out were not predominantly caused by liquidity issues. Instead, these pitfalls often stemmed from behavioral biases like loss aversion.

"Portfolio managers with scores under 50, and that maps to loss aversion behavior... that's where you'll get the most."
— Clare Flynn Levy [14:24]

Behavioral Biases Over Liquidity

Clare emphasizes that behavioral factors, such as emotional attachments to stocks or hesitancy to admit errors, play a more significant role in poor decision-making than liquidity constraints.

---

Applications for Fund Selection

Current Usage Among Allocators

Currently, the Behavioral Alpha Score is primarily utilized by institutional investors managing segregated accounts with access to daily holdings data. These allocators incorporate the score into their manager selection and ongoing monitoring processes.

Path to Mainstream Adoption

Clare reveals Essentia’s ambition to scale the Behavioral Alpha Score across all market segments, including broader audiences lacking access to daily holdings data. Upcoming partnerships aim to integrate monthly holdings data from active ETFs, making the score accessible to a wider range of investors.

"Our goal with this is to make it mainstream and to make the behavioral alpha score something that everybody looks at, just like you look at past performance."
— Clare Flynn Levy [19:36]

Implications for Managers

As data transparency increases, managers must adapt to the enhanced scrutiny of their decision-making processes. A good Behavioral Alpha Score can serve as a testament to a manager’s skill, while a poor score necessitates strategic improvements.

---

Enhancing Transparency Through Data Availability

Rise of Active ETFs

The proliferation of active ETFs publishing daily holdings significantly improves data accessibility. Essentia can now analyze data from over 200 ETFs, facilitating the Behavioral Alpha Score’s application without relying solely on private data sources.

"If you're the manager, you need to be able to explain yourself one way or the other and brag about yourself if your score is really good."
— Clare Flynn Levy [21:34]

Future Prospects

Clare anticipates that as Behavioral Alpha Scores become more widespread, they will become a standard metric alongside traditional performance indicators, fostering greater transparency and accountability in investment management.

---

Personal Insights and Closing Thoughts

Clare’s Industry Journey

Clare shares a personal anecdote about her first job in the industry, emphasizing the value of record-keeping and data analysis from the outset of one’s career.

"I wish I had kept better records... capturing more data about what I was thinking and doing."
— Clare Flynn Levy [23:40]

Advice to Aspiring Investors

Her key piece of advice to those entering the investment field is to maintain meticulous records of their investment decisions and thought processes. This practice not only aids personal growth but also provides valuable data for future analysis and improvement.

"Do it in a structured data format that you can then analyze and not letting that data go."
— Clare Flynn Levy [23:40]

---

Conclusion

Clare Flynn Levy's insights into the Behavioral Alpha Score offer a groundbreaking approach to quantifying and enhancing investment decision-making skills. By systematically analyzing seven critical decision types, Essentia Analytics provides a robust tool for managers and allocators to assess and improve performance beyond traditional metrics. As data transparency increases with the rise of active ETFs, the Behavioral Alpha Score is poised to become an essential component of investment management, fostering a more disciplined and analytically driven industry.

For more information, listeners are encouraged to explore Essentia Analytics' resources, including their white paper "Actions Speak Louder Than Past Performance," available on Essentia’s website and SSRN.

---

Notable Quotes:

• Clare Flynn Levy [01:17]: "It is a measure of decision making skill. We call it behavioral alpha because in this day and age, alpha is very hard to come by harder than ever..."

• Clare Flynn Levy [03:51]: "Past performance was not predictive of future performance."

• Clare Flynn Levy [08:10]: "What we're saying is this actually can quantify some portion of that alpha piece so you can hone in that much better."

• Clare Flynn Levy [14:24]: "Portfolio managers with scores under 50, and that maps to loss aversion behavior..."

• Clare Flynn Levy [19:36]: "Our goal with this is to make it mainstream and to make the behavioral alpha score something that everybody looks at, just like you look at past performance."

• Clare Flynn Levy [21:34]: "If you're the manager, you need to be able to explain yourself one way or the other and brag about yourself if your score is really good."

• Clare Flynn Levy [23:40]: "I wish I had kept better records... capturing more data about what I was thinking and doing."

---

This comprehensive summary captures the essence of Clare Flynn Levy’s expertise and the innovative approach Essentia Analytics brings to the investment management industry through the Behavioral Alpha Score.

Transcript

[A]

Get ready for CFA Institute Live 2025 with free power packed webinars designed to inspire and inform. Led by industry experts, these concise sessions tackle game changing topics like how AI is transforming investment strategies in emerging markets. Don't just show up, show up ready. Gain the insights you need to drive meaningful conversations this May in Chicago. Claim your spot [e-mail] Foreign.

[B]

Welcome to the Enterprising Investor, the flagship investment podcast for CFA Institute. I'm Mike Wahlberg and I'm joined today by Claire Flynn Levy, founder and CEO of Essentia Analytics, a firm that provides behavioral analytics services to professional investors and capital allocators. At the core of Essentia's offering is something they calculate called an investor's behavioral alpha score, which analyzes elements of an investor's process and quantifies their decision making skills. In a recent white paper, Claire and her colleagues set out to test whether investment decision making skill was predictive of outperformance. And we're going to find out today. Welcome to the show, Claire.

[A]

Thanks for having me.

[B]

So, to start us off, maybe describe what a behavioral alpha score is.

[A]

Sure. Well, I suppose it helps to start off by explaining where it's coming from. So the background here is that I was once an active equity fund manager myself and I wanted to continuously improve as a fund manager and continuously make better decisions if possible, and so set out to try to analyze my own decision making. And through that journey that unfolded over the next decade, we arrived at what we call the behavioral alpha score. It is a measure of decision making skill. We call it behavioral alpha because in this day and age, alpha is very hard to come by harder than ever. You can't find it by being smarter than other people, really. I mean, not in a sustainable way. You can't generate sustainable alpha through having better information than other people, that's for sure. So what can you do? Well, you can generate sustainable alpha by making better decisions than other people make because humans are biased and flawed. And we all have read about this and we are all aware of the field of behavioral finance, but until now there hasn't been a way of measuring decision making skill and therefore improving it. And that is what the behavioral alpha score does.

[B]

Okay, so not to ruin the ending here, but in terms of the study that you conducted looking at the impact of these decisions, what were your high level findings from it?

[A]

Well, so what we did in the study was take 123 unique long only equity portfolios and use data between daily holdings data from those portfolios between 2013 and 2023. So altogether that was 532 portfolio years of data. And the question we set out to answer was, is the behavioral alpha score as a measure of decision making skill predictive of performance? To start with, what we wanted to check was, are we sure that past performance is not predictive of performance? Because if that's the case, you know, all bets are off. There's no need for this.

[B]

All our compliance listeners ears are perking up right now. They're all freaking out about the footnotes it bears.

[A]

You know, let's just check in with that because it's important statistically. And we did that. So the answer to that, not surprisingly, was no. Past performance was not predictive of future performance. What was? So the behavioral alpha score was predictive of future performance, but only within the same year. So if you said like, is the behavioral office score a crystal ball that will tell you exactly who's going to perform well within the same year? You might call it that. But that's not really the time frame that people care about. The question is like, what about further out then this year?

[B]

Yeah. So there was a correlation between detected high skill in that one year time period and high performance that year.

[A]

Exactly. And it's like, well, that's not surprising. That's not really telling you anything that exciting. But the exciting part came when we looked at a comparison of managers with a score over 50. So the behavioral alpha score 50 is effectively what it would have been achieved by chance. So that's the sort of baseline. What we found was that managers with scores over 50, and this is looking at the past three years of their decision making, and it's looking at the P and L generated by their decisions compared with the P and L generated by their benchmark. So it's all done in relative terms. And it's saying, okay, if we look at all your decisions over the last three years and the value they added in relative return terms, did you add value through your decision making? And a manager who has a score over 50 did, and a manager with a score under 50 did not. So what we found in the research was that managers with scores over 50 were one and a half times as likely to outperform their own benchmark in the subsequent year as managers with score under 50. So practically speaking, while it's not a crystal ball, it is saying you could compare two managers on this basis and have a better sense of which one is likely is more likely to outperform, which is a very valuable thing to be able to do if you're an allocator or a manager selection person.

[B]

For sure. For sure. And not to skim over the first point you made there, which is that it did have explanatory power in the first place. Right. Apart from forecasting future potential outcomes. Can you talk a bit about, about that? Just a little bit of the detail around kind of what tests you did, the factor tests, factor models that you applied to it?

[A]

Yeah, I mean, so, so then the question was, would this be additive to the model of somebody who is selecting managers based on, you know, traditional Fama French factors and you know, the sorts of stats that we've, we've all been learning at university all these years. And you know, there are, there are some large firms that have very sophisticated models at this point that are being used to predict which managers are more likely to outperform. So would this be additive to that? And what we found was yes, it would. So we looked at whether it added explanatory value in the context of the Fama French five factor model as well as a three factor model, but in both cases the answer was yes. What it's saying to me anyway is that as a fund manager, part of what you're doing is picking the right stocks. Right. That's half the battle is the idea, but the other half is the execution. And the stocks you pick are reflected in risk factors as, you know, as defined by Obama French. But what you did, the execution piece, is just sort of lumped together in alpha. And that's the sort of the part that everybody's trying to find and put their finger on. And what we're saying is this actually can quantify some portion of that alpha piece so you can hone in that much better.

[B]

Can you push in a bit on what those decisions were? Because I know you have seven that you looked at. I'm just curious if you could talk a bit about how finely you cut this when you were looking at individual decisions that make up skill.

[A]

Sure. Well, so what we did, and this is what we do day to day at Essentia analytics, is we take historical holdings data, daily holdings data, and we assemble that data into what we call investment episodes. So if you think about this like sports analytics, okay, an investment episode is like a game and every game has a game tape. And we do not in the behavioral alpha score and not in this piece of research, but separate to that, we do a lot of analysis about episodes. And what do your good episodes have in common and what did your bad episodes have in common? But then what we do is break down the episode into its constituent decision types. And the sort of first principles basis of all of this is a fund manager's job is to make decisions. And if you're the fund manager, it should be a goal to make the best quality decisions that you can, as often as you can. And in fact, invest as much of your energy as possible into making the decisions where you have the highest probability of making a very good decision. And invest as little energy as possible into making decisions where you know, historically you've done a bad job. But not all decisions are created equal. So what we're doing is decomposing each episode into seven different decision types. And at the most basic level, you've got the picking decision. Okay, that's the difference between the outcome of having held one share of the stock that you held and having held one share of the index. So very simple, but it's about isolating just that decision to hold that stock. Then you get into sizing decisions. If you were not making a sizing decision, you would just hold an equally weighted portfolio. So every day that you're not holding an equally weighted portfolio, you're making sizing decisions wherever you're not holding the equal weights. So we look at those and that's by comparing the portfolio every day with an equally weighted portfolio. Then we look at entry timing decisions and that's about, you know, identifying the point where the first share was bought and then looking, in this case, one month into the future and saying, would you been better off just taking the average daily close over that one month window? Was there really any value to add in by this? Then we look at scaling in. So the timing of the day you bought the first stock is one thing, but then how quickly do you ramp it up and how quickly do you build that position? We call that scaling in. And again we're looking at, well, what else could you have done? Let's just take the window over which you did your scaling in and say, what if you just set an algorithm to get you the average daily close? So it's a bit like in each case throwing a dart at a dartboard. What was what you did better than throwing a dart at a dartboard? And so, yeah, maybe yes and maybe no.

[B]

How do you account for liquidity constraints on that? Or do you, I mean, because certainly, obviously bigger managers are going to have to be a little bit more slow and thoughtful on scaling into decisions. Is there a way to account for that?

[A]

Yeah, yeah, you can sort, you know, it all depends on the window that you're using the type of strategy that you're looking at, we work in great detail with managers of all different types of equity strategy. And some of them don't scale in. They just go straight to size. Some of them take a really long time to get in. Sometimes that's driven by liquidity. Usually it's not. That's just a matter of style and how their organization behaves. Sometimes it's more on the way out. When you're scaling out, that liquidity can rear its head. And, you know, you can see that people's, people's exit timing. So I mentioned entry timing. We also look at exit timing. I mentioned scaling in. We also look at scaling out. When you look at the scaling out and exit timing, those tend to be the skills that people struggle with the most. That's where you'll get the most. Portfolio managers with scores under 50, and that maps to loss aversion behavior. And, you know, all the biases that we've all read about that are to do with what happens when things are going down or when you're losing money. So it's all very familiar. But if you were running a small cap strategy or an emerging market strategy or something, you know, where the liquidity is tight, you might say, oh, I have a poor exit timing and scaling out score. That's because of the liquidity. It must be. Maybe that's a hypothesis. So what we'll say is, you know, it doesn't matter to me either way. So let's have a look, let's see. You know, if that, if it is, then you can quantify the cost of the strategy in that sense, you know, or if this is about the fact that your fund is so big you can't move in and out with ease, it's telling you something about the cost of, of letting the fund get that big. But actually, you'd be surprised. Sometimes people think liquidity is what's causing it. And actually when you look at the, the analysis by, you know, the percentage of average daily volume that you're holding in the stock. So if you divided up all of your episodes by what was the max average daily volume that we held in that stock over its life and then said, all right, are the ones where we did poor exits the ones where we were the most illiquid? It's usually not the case.

[B]

Yeah, people fall in love with their stocks, right. Or they don't want to admit an error. I mean, like you say, it's a, it's a classic behavioral error. Right.

[A]

And sometimes it's about, you know, I don't want to upset the analyst, I don't want to undermine this person. You know, there will be institutional reasons that people, you know, take longer over these decisions than they otherwise would. And you can show that back to them and then you can help them come up with ways of making that decision more efficiently without it being so personal. It's just a matter of when the stock behaves like this. Here are the rules that we implement. We ask ourselves this question, we ask ourselves that question and then based on the answers to these questions, we either buy it or we sell it. And that helps a lot. And then you can measure whether the exit timing and scaling out are improving, which is always very gratifying.

[B]

So within those seven decisions, then effectively it sounds like each one you have some kind of a test that determines whether it was a good or a bad decision. So you've got, did the stock beat the market? Did, did your position? And if it beat the market, didn't you mean these are the classic attributions, things you look at, right? If you, if it beat the market and you own more than then index, then, then that was a good decision. How do you weight those, those decisions in the actual calculation of the score? Are they all equally weighted in terms of like adding up to what kind of score you get? Or is stock picking more important? Is, is overall weighting more important? Or like, how does that work?

[A]

We've, we talked about it a lot because, you know, different people have different views and instinctively you might say, well, picking should be the most important one or maybe picking and sizing, but actually we don't weight them. So they each have an equal weight. And it's on the basis that if you're making, you know, if you're a very low turnover manager and most of the decisions you're making are picking decisions and then every day you hold the stock, you're re picking it effectively, but that's kind of all you're doing. You don't have a lot of entries, you don't have a lot of exits. The other, the seventh one, if anybody was counting that I didn't mention, was size adjusting. And that's about adding in, trimming. But, you know, say you don't do much of that. Then you know, out of the n number of decisions, more of them are going to be picking decisions, you know, proportionately than would be the case for somebody who's doing a lot of adding and trimming. But then you would say, well, the person who's doing a Lot of adding and trimming. Ultimately this is about like, is the, is this person adding value or destroying value? You know, if you want to look at did they pick the right stocks? There are lots of analytics you can do specifically about that. But is all the energy that they're expending and maybe fees that they're paying on transaction costs and all the rest, is that actually adding value? And if the answer is yes, great, you know, that's everyone's a winner and that's all we need to know. If the answer is know, then that's a conversation that needs to be had.

[B]

So if I understand it correctly, the paper, the scope of the paper defined an episode as one year. So you're looking at kind of what happened in terms of the decisions that the PM made in that year around, you know, the holdings in their portfolio. But as, as you know, often these theses play out over three or four years. So any, any given year, maybe certain theses bearing fruit, certain theses burning out, certain stocks being added as new ideas. So is there an appetite to tackle a longer term study to test the value added over a longer period? Or how do you think about that?

[A]

Well, it's actually in the study, it's looking at three years overall. So you're taking data for three years.

[B]

Okay, so I misunderstood.

[A]

Okay, but I mean, it's a fair point because going into the future, all it's saying is we can tell you which manager is more likely to outperform next year. And if you're an allocator and you say, well, I'm choosing managers for the long term, where does that leave me? Fair enough, it's a good point. That's for us to explore with the next research. But in the meantime, I would say to that manager, well, yeah, you invest for the long term, but every year you need to re underwrite that investment and revisit. Do I think this guy's going to, or this woman is going to keep outperforming for me next year as well as, you know, five years down the road. And is there any reason to believe that that story is deteriorating or that, you know, their ability to outperform for me has changed? And this is what this data can tell you. It'll either give you reassurance, pause.

[B]

That's actually one thing that I wanted to talk to you about and that's this idea of its applications for fund selection because as you know, all too common, unfortunately, especially among retail investors, to pick last year's winners. And even in an institutional setting, you Know, you have the benefit of consultants who vet managers based on things like culture, adherence to style, persistence of performance, stability of the team, ownership structure, and on and on. It seems to me that for the layman or even the professional, being able to quantify a manager skill would be hugely helpful. How do you see this metric being applied across market segments?

[A]

So amongst the allocator community, it is still early days for this. So we have been working with allocators for a few years now, but that has been exclusively those who have segregated accounts and therefore have access to the daily holdings data of the underlying managers and so can do this sort of analysis. And they use that in manager selection as part of their due diligence process. These would be more institutional investors and then also as an ongoing part of how they monitor those managers, they sit down with the manager quarterly and they bring an essential report with them and they ask questions or share thoughts based on what's in that report. But what's happening increasingly, our goal with this is to make it mainstream and to make the behavioral alpha score something that everybody looks at, just like you look at past performance, which is not predictive, as we've established, but, you know, you haven't necessarily had anything else to look at. And you can do qualitative research and try and gauge all those things that you listed about the team and the process and all that kind of stuff, but there's not been anything better to go on quantitatively. And now that there is, it means that even if you don't have access to daily holdings data, if you're invested in pulled vehicles, you will soon be able to see a behavioral alpha score for any manager based on their monthly holdings data. And that's through a partnership that I'm not quite yet at liberty to announce, but it's coming soon. And I guess the message to managers in that is this is happening whether you like it or not. You know, this data is. The world of data is moving very quickly and the data is already out there.

[B]

So especially with ETFs publishing, you know, the rise of active ETFs publishing their holdings daily. Right. That's a different world than we lived in 20 or 40 years ago.

[A]

I mean, that makes it way easier because then we can literally, we already have data on, think over 200 ETFs in our, in our behavioral alpha benchmark. And that's just public data. So, you know, if you're the manager, you need to be able to explain yourself one way or the other and brag about yourself if your score is really good. If it's not, at least understand it and be able to speak to it. Because allocators aren't necessarily going to make a decision to fire you based on your behavioral office score, I hope. I'd like to think they would give.

[B]

You a chance or hire you.

[A]

Yeah, I mean, but I'd like to think that if they're thinking about hiring you that they'll use this as an extra substantiation of their thesis and it should validate some of the stuff they found in their more qualitative research about the firm and the process. But if your score's not good and you're looking to retain assets, which a lot of managers are, then it's about being able to explain yourself and explain that you understand because you now have this level of visibility, you understand what you can do better to get or do differently to get a better result. And now you're going to. And allocators do cut you slack if you have a plan, just like anybody in life cuts you slack if you have a plan. Now you've got to follow through with the plan. But you know, it's, everybody benefits in the end from being able to see this data. It's just the manager needs to sort of wrap their head around the fact that yeah, it's a different level of transparency than existed before. That's the way of the world.

[B]

So unfortunately we're coming to the end of our chat today. Claire, I'm going to hit you with the two parter to wrap up the interview here. And what was your first job in the industry? And if you could go back and take yourself for coffee on your first day, what key piece of advice would you offer yourself?

[A]

Well, my first job in the industry. This sounds like a cliche, but I swear to you it's true. Was in the mailroom at Gabelli Funds, which is a mutual fund company in the US and I was stuffing paper prospectuses in envelopes because, you know, this was a long time ago, I'm not going to betray my age here, but long time ago. And yeah, it was literally stuffing little brochures into envelopes and sending them out to people from, from a fund management company. But from there worked my way into answering the phone and from there into data entry and from there into the front office and the rest is history. But what would I have, what would I have told myself when I, when I first started out? I think I wish I had kept better records, I think. And I would give this advice to, to any person starting out in the industry when I was doing PA investing because that's all I was doing when I was working in the mailroom. I didn't, you know, I did start investing for myself, but in a very small way. I wish I had captured more data about what I was thinking and doing and whether that's keeping a trading journal or better yet, doing it in a structured data format that you can then analyze and not letting that data go. You know, when people if you're going to change firm or you're going to move to a different role, see if you can negotiate taking that data with you because it is valuable and down the road it would be useful to be able to look back for amusing if nothing else. So I wish I had done that.

[B]

I've been speaking today with Claire Flynn Levy, founder and CEO of Ascensia analytics and co author of Actions Speak Louder Than Past Performance, the relationship between professional investors decision making skill and portfolio returns that can be found on Essentia's website. And for the academics in the crowd, it's on ssrn. Thanks so much for coming on the show today, Claire.

[A]

Thanks for having me.

[B]

I'm Mike Wahlberg and this is B, the enterprising investor.