向投资传奇学习以迎战2020年（富达官方实录）

TRANSCRIPT

Learn from investing legends for 2020

Justin Carbonneau: Hello, everyone. I’m excited and I feel very privileged to be

presenting to you on today’s webinar. My name is Justin Carbonneau, vice

president at Validea. In today’s presentation, “Learning from Investing

Legends for 2020 and Beyond,” we’ll discuss the philosophies and disclosed

strategies of great investors and other demonstratable stock selection

methods. In order to maximize our time today and help you get the most out

of this, I want to jump right into the presentation.

So, here is the big idea of today’s presentation in 163 words. I want to read it

to you so it’s very clear of the challenge, but also the opportunity in today’s

market for the active stock investor and what I hope we will accomplish in

today’s presentation. “Holding individual stocks remains a vital piece of long-

term wealth creation. Sourcing and analyzing specific ideas, however, can be

time-consuming, risky & complicated. With over 15,000 stocks, mutual funds

and ETFs in the U.S. alone, you could spend a lifetime analyzing the set of

investable ideas. In today's fast-paced world, many people lack the time to

even sit down for dinner, so finding efficient methods for stock analysis and

idea generation is essential for today’s active stock investor. So, how do you

effectively and systematically identify and research good opportunities? We

1

believe proven, time-tested methodologies from legendary investors like

Warren Buffett, Peter Lynch, Ben Graham and others with demonstrable

approaches extracted from publicly disclosed writings provide a useful

framework for sourcing and analyzing stocks that can play an important role in

long-term wealth accumulation. The following presentation is dedicated to

how we leverage and emulate these strategies to find sound investments

ideas and what active investors like yourself can learn from them.” So, you can

see where we are going in today’s presentation and how we’ll try to tackle that

big idea.

Here’s an overview of the presentation: we’ll get some opening remarks from

a legendary stock picker. I’ll talk about how we go about identifying these

guru strategies that we model at Validea. I’ll tell you who the gurus and the

strategies are, and what they’re based on. Then, I’ll give you an overview of

five different approaches across investing styles. We’ll look at strategy, give

you the evidence and the source behind the strategy model. And then we’ll

establish the investment thesis for each one in today’s market. I’ll walk you

through the detailed investment methodologies of these five, and we’ll look at

some concrete, stock-specific examples. I’ll conclude with a few key lessons

and other tips that I hope can help you become a more successful investor and

helping your stock selection process and strategy.

2

OK, so, on December 21st in Barron’s Magazine, they had two features with

Fidelity’s former legendary fund manager, Peter Lynch. The pieces were full of

knowledge and investing wisdom. What I wanted to do is highlight a few

things that I thought were highly relevant to today’s presentation. In the first

piece, which was titled “Master Stockpicker, Peter Lynch: If You Only Invest in

an Index, You’ll Never Beat It,” the author wrote, “Lynch demystified investing.

He emphasized searching for companies that could deliver earnings growth 20

to 50 percent. He espoused a PEG ratio, a company’s price earning’s ratio

divided by its long-term growth rate. A PEG of less than one means the stock

is worth a closer look. Lynch believed that individual investors have an edge

over fund managers because they could spot trends before a stock became

popular. In the second piece, where it’s an interview that Lynch did, the title of

that piece was “Peter Lynch Draws on 50 Years of Stock-Picking to Find

Growth Opportunities in Today’s Market,” Lynch says, “if you’re going to

invest, you have to follow certain rules. The one thing I want everybody who is

buying individual stocks to get is they have to understand the story, the five

reasons they think something’s going right for the company. If you can’t

convince an eight-year-old why you own this thing, you probably shouldn’t

own it. Don’t invest in a company before you look at the financials. If you

made it through fifth grade, you can handle the math.” So, these concepts,

owning individual stocks, looking at the fundamentals and valuations, having a

3

set of rules, even knowing the story behind the stocks you own, these are all

important, and to a large extent, will relate to our discussion today.

So, the first question is, “How do we go about identifying successful investors

and other investment strategies?” At Validea, we look at really three things

when sourcing one. One, we want to see a long-term track record of market

up performance, either in the real world or back tested. The strategy needs to

be publicly disclosed in books, academic papers, or other sources. And then

then number three, the methodology needs to be able to be run largely

quantitatively via computer programming and modeling, which means looking

at things like fundamentals and pricing in order to do the analysis. Before I tell

you who the gurus are, let me just read you this important disclaimer: “None of

the individuals mentioned in today’s presentation endorse any of the

strategies or our implementation of these strategies. Validea’s models are

extracted from publicly available sources (a book or research paper by or

about each of the individuals mentioned) and are not meant to represent the

‘gurus’ or individuals themselves in any way. All of the strategies are

quantitative in nature and the gurus themselves may have changed, altered or

never followed these exact strategies. The information presented in today’s

presentation is intended to give you an understanding of active and

quantitative stock selection strategies only. In addition, individual stock

4

selection can be very risky and none of the material discussed in his

presentation should be construed as investment advice or as individual stock

recommendations.”

So, since 2003, we’ve been building computerized investment strategies by or

about the following individuals and investors. We capture these strategies and

model portfolios in a quantitative stock analysis system. I’m going to quickly

walk you through who these individuals are and the source that we base our

models on. I’m going to go from left to right and top to bottom, and I’m going

to do this quickly because there’s a lot of people to get through.

So, we run a model based on Peter Lynch’s “growth at a reasonable price”

investing approach, which he outlined in his book, One Up on Wall Street.

Peter Lynch, as many know, is one of the most successful fund managers of all

time. We run a contrarian investment strategy based on David Dreman’s

approach he outlined in Contrarian Investment Strategies. There’s a strategy

based on John Neff’s value screen, which he outlined in John Neff on

Investing. Some may know that Neff managed the Windsor Fund for a decade

and was able to beat the market by about three percent annually over a very

long period of time. We have a value investing model based on Ben Graham’s

deep value approach he outlined in The Intelligent Investor. Graham is known

5

as the father of value investing and had a huge influence on the greatest

investor in the world, Warren Buffett. We run a model based on Ken Fisher’s

approach that he discussed in Super Stocks, where he popularized the price-

to-sales ratio. The magic formula by Joel Greenblatt, which we extracted from

his book, The Little Book that Beats the Market. This is a combination of

Graham’s value approach and Buffett’s focus on quality. It brings these two

concepts together in one simple systematic model. The cancel method, which

was developed by William O’Neil and written about in his book, How to Make

Money in Stocks, a model that seeks to replicate Warren Buffet’s strategy by

looking for high-quality blue-chip stocks that are attractively priced. We base

our Buffett model on the book Buffettology. We’ll actually walk through that in

a few minutes. A growth investing model based on Martin Zweig in Winning

on Wall Street. At one point, Zweig had the number one ranked investment

newsletter based on risk adjusted returns according to the Hulbert Financial

Digest. A model based on the research done by Meb Faber in his book,

Shareholder Yield. Shareholder Yield captures all the ways a company can

return cash to shareholders. It’s become a much more popular metric over the

last 10 years. Multiple strategies based on work and back testing of James

O’Shaughnessy in What Works on Wall Street. What Works on Wall Street is

the seminal book in terms of simple systematic, quantitative strategies and the

ones that have worked best over the long term? We’ll look at one of the

6

O’Shaughnessy models in a few minutes. A small cap growth investor model

based on the Motley Fool Investment guide written by Tom and David

Gardner, the founders of Motley Fool. Another deep value model extracted

from a recent book, The Acquirers Multiple, written by Tobias Carlisle. Wayne

Thorp, an analyst at AAII, he wrote about and tested an earnings revision

model. The name of that research report is How to Profit from Revisions in

Analysts’ Earnings Estimates. There’s a growth investing strategy based on the

academic research of Partha Mohanram. The name of that paper is Separating

Winners from Losers Among Low Book-to-Market Stocks Using Financial

Statement Analysis. Another academic-based model from Joseph Piotroski.

That’s Value Strategy. A strategy using Patrick O’Shaughnessy’s research and

long-term testing from his book, Millennial Money. A dual momentum

approach modelled after the academic study Twin Momentum: Fundamental

Trends Matter by Dashan Huang. That strategy combines both price

momentum and fundamental momentum together. A pure momentum model

extracted from Wes Gray’s book, Quantitative Momentum. That is a strategy

we’ll look at in a few minutes. And then lastly, a multi-factor and low volatility

method developed by Pim van Vliet and written about in his book, High

Returns from Low Risk and following research paper. We’ll also look at that

model in some detail.

7

So, there’s really two points I want to mention here. One is that every one of

these strategies I just mentioned can be read about in books and/or academic

papers. These sources exist for you to reference and learn from. You can buy

the books on Amazon pretty much with like one click, and all of the research

papers that I mentioned are free for download on SSRM. The second thing is

we do our best to implement the models as they were outlined in these

sources, so we’re not trying to reinvent the wheel here. What we’re trying to

do -- and hopefully I’ll show you today -- is that parts of these stock selection

approaches can be used to help analyze stocks and find new investment ideas

with a little bit of effort. With each investment model, they’re differen… so, the

first grouping is what I would consider value-focused models. These are

models that are seeking stocks that trade at a discount based on earnings,

book value, or some other value metric. Most of if not all couple the value

criteria with other fundamental factors. The next is what I would call our value

in quality category. These are models that emphasize both value and quality

characteristics. So, the idea here is buying good companies at reasonable

prices a la the Warren Buffett style of investing. There’s another category that I

call quality and yield. This is a bit unique, but basically there’s a quality

component to it, and then it couples that with other fundamental criteria that

are important in stock selection. We then have our growth or growth at a

reasonable price grouping. These are strategies that look to identify firms that

8

are growth-like companies that have other attractive fundamentals or

characteristics that make them the type of growth stocks you want to be

looking at. And then there’s the momentum models, which are models that

reward companies with strong price strength. In most cases, momentum is

combined with other fundamental factors, but not in all cases. So, what we’re

going to do is we’re going to look at one model in each category in detail, and

I’ll walk you through that. But what I also thought was important, and I wanted

to emphasize, was that for investors looking to utilize these strategies, I believe

it’s important to find strategies that align with what you believe in. So, if you’re

a big believer in value investing, you probably want to be not going into the

growth or momentum type of models. Or if you highly value dividends, you

may want to focus on more of the quality and yield type of strategies. Like

Warren Buffett and Charlie Munger like to say, “Stay within your circle of

competence. Stay focused on what you know and what you believe in, and

don’t go too far outside that circle.”

So, there’s two ways, I think, that investors can utilize strategies like this. The

first is an individual stock analysis so you can utilize the fundamental criteria in

these strategies and apply it to individual stock research. By learning about

the strategies, you’ll be able to understand the key, financials, valuation

metrics, and technical variables that have been deemed important and robust.

9

You can then start to incorporate more rules and thresholds in your own

process as you familiarize yourself with each of these strategies. You can also

use strategies like this for investment idea generation. So rather than creating

a screen from scratch, you can extract the fundamental criteria outlined and

tested by these investors and screen for new stock ideas. Screens, obviously,

as many know, provide a nice, efficient way to source new investment ideas

and you can screen for or screen through hundreds or thousands of stocks all

at once. What you can then do, if you feel comfortable enough, is you can

integrate your own fundamental variables and further refine and sort the list of

stocks.

So, here are the five models we’re going to walk through today. We’re going

to start with the value model based on James O’Shaughnessy’s value

composite approach. We’ll then look at the Warren Buffett based model,

based on the book Buffettology. This will be our value and quality model.

Under the quality and yield category, we’ll look at the Pim van Vliet multi-

factor and conservative stock selection method. In the growth camp, we’ll

look at Partha Mohanram’s research on how to identify growth stocks that

have the potential to go on and perform well. And for momentum, we’ll look

at the method developed by Wes Gray and his book Quantitative Momentum.

Now, I understand that not all of these are necessarily helpful names, but with

10

each of the models, what we’re going to do is we’re going to walk through

them. I’m going to give you the key fundamental investing insight of what I

hope you can remember and walk away from this presentation with. What

we’ll then do for each model is we’ll establish what that key insight is. I’ll give

you a brief strategy overview. I’ll disclose the source and the long-term

evidence behind the strategy. And then I’ll provide you what I think could be a

reasonable investment thesis for this type of strategy in 2020. We’ll walk

through a subset of the models in detail and I’ll provide concrete examples

using a specific stock and the fundamentals to demonstrate how these

strategies actually work.

So, the first model we’re going to look at is the value composite model, which

is based on the research and writing of James O’Shaughnessy in What Works

on Wall Street. In What Works on Wall Street, O’Shaughnessy back-tested 45

years of stock market data from the comprehensive Standard & Poor’s

Compustat database to find out which quantitative strategies have worked

over the years and which haven’t. The original publication of the book was

1996, so quite a long time ago. But since the, O’Shaughnessy has continued

to publish new editions with updated strategies and longer-term data. I really

think if anyone’s interest in sort of these type of quantitative models, this is

really a must-have book. It’s packed full of great long-term data on factors and

11

how these factors worked when combined together. When most investors

hear the word “value investing,” many think we’re looking for stocks with low

PE ratios, low price-to-book, or price-to-sales ratios, and that can certainly be a

start, but what O’Shaughnessy’s research showed is that you can improve the

results of the values factor by using multiple value metric simultaneously. So

rather than just one, we want to use more than one. We want to combine

them together. So, the strategy starts out by filtering out the lowest quality

stocks in the market. It then selects the least expensive stocks remaining using

a value composite ranking, which I’ll explain to you. The source and the

evidence, as I explained, is based on What Works on Wall Street, where

O’Shaughnessy tested this approach back to 1964. So, from 1964 to 2009, his

value composite portfolio returned 18 percent in the back test versus 11.2

percent for the all stock universe. On Validea, we also track a very focused,

value-composite model portfolio, and since ’08, which is how far we’re able to

go back and track it, the portfolio is actually one of our better-performing value

models and is up 18 percent annually over that time. The investment thesis

here, I think, for this is that many value stocks and value strategies have

struggled over the last 10 years. Even though the strategy we base and we run

based on the value composite has done well, many value strategies and stocks

have not. But we also know that there’s a cyclicality between value over

growth and growth over value, and since the late summer, early fall of 2019,

12

we’ve actually started to see better performance out of the value cohort of

stocks, and so if that continues to materialize, I think this type of strategy could

be a beneficiary of that turn in value.

So, kind of moving into the strategy now, the first part of the value composite

method is to remove the junkiest stocks. So, what we want to do here is we

want to remove the stocks that could be value tracks, stocks that look like

they’re value, but they’re permanently impaired for one reason or another. In

order to do this, we layer in a number of checks, including declining expected

future earnings, cashflow that’s not keeping up with earnings, high levels of

debt, and very low relative strength. From there, the bottom 10 percent,

based on our combined quality score are not eligible to go onto the next step.

Those stocks that remain are then looked at through multiple lenses of value.

So, this model is looking at the price to book, the price to sales, price to

earnings, price to cashflow, and an EV/EBITDA, and then it does integrate

shareholder yield to improve the performance of the strategy. And

O’Shaughnessy points out in his book, when you look at these value factors

over time, the best performing value factor changes, and one year it might be

the price-to-sales ratio, in another period it might be enterprise EV/EBITDA.

So, the combined value composite type of model manages this jockeying by

scoring stocks through this series of distinct value criteria. You’ll also notice,

13

and this is I believe very important, that the different financial statements are

represented here in these ratios. So, the balance sheet via price-to-book,

income statement via price-to-sales and price-to-earnings, cashflow, enterprise

value, EV/EBITDA, and price-to-cashflow. Shareholder yield was included

because it helped improve the performance of the strategy, as I mentioned.

The way that these factors can be used in terms of screening and building a

focus investible set of stocks is you basically score -- you create a composite

score. So, what you do is you average the score, you order rank all of the

value metrics -- so price-to-book, price-to-sales -- you can create a composite

score. And then what you do is you only want to select the best one to two

percent of stocks that score highest according to that composite score. So,

you’re really trying to focus in on the very best opportunities that get the best

scores on average across all of those value metrics.

Let’s look at an example. We’re going to look at MetLife here very quickly.

MetLife is in the 58th percentile of our database in quality. So, it meets that

quality screen, so we can move on to the next part of the strategy. We then

look at price-to-book, price-to-sales, PE, price-to-cashflow, and price

EV/EBITDA, shareholder yield. From there, we get a combined ranking based

on all of the factors above. That’s essentially what a composite ranking is, and

14

when we do that, MetLife is actually in the first percentile of all stocks in our

value stock universe.

Moving into the next category, we are going to look at our implementation of

the Warren Buffett-based model, which we based on the book Buffettology.

Of course, Warren Buffett has never really fully disclosed his exact investment

approach, and he probably doesn’t have an exact formulaic investment

approach. But if we did try to emulate and capture it, how might we do that?

We know whether he’s buying public stocks or companies outright. Buffett is

really trying to buy great companies at a reasonable and fair price with the

hope that these will be long-term, compounding machines. If we were to try

to capture the concept of great companies quantitatively, how might we go

about doing that? How could we uncover companies with the characteristics

that have competitive moats around their business? In order to find those

companies -- and I think this is the key insight I’d like you to take from this

model -- is that you can look at the long-term consistency in growth and

earnings along with the long-term returns on equity and capital to try to help

us find companies that have competitive moats around their business. So, to

start out, this strategy looks for firms with long-term predictable profitability

and low debt relative to profits that trade at reasonable valuations. The model

goes back as far as a decade to look into a company’s history to uncover those

15

stocks with competitive moats around their business. What is the source and

evidence for this? Well, since 1965, Buffett has one of the most phenomenal

track records ever generated, compounded at 20.5 percent over the last 53

years. Interestingly enough, our Buffett-inspired strategy actually scored

Apple 100 percent in 2015, which was two full years before Buffet and

Berkshire were buying the stock. As many people may know, Apple is now

Berkshire’s largest public holding. Our Buffett-based model that we run,

which is the optimal strategy on Validea in 2019 was up 36 percent. Berkshire

Hathaway itself, that stock, didn’t have a great year last year, and I’m not

saying this model is a substitute whatsoever for Berkshire Hathaway holders,

but still there was a big difference in terms of what our strategy was able to

pick up using this Buffett model and the performance of Berkshire Hathaway in

2019.

The investment thesis, you could say that using a strategy like this, you’re

buying high quality stocks at a discount, and it gives you a less volatile way to

access the outperformance of value stocks over time, and maybe if you’re

lucky -- which maybe we got lucky with Apple. There was certainly some luck

in there, but the system was picking it up. Maybe fallowing a strategy like this

will allow you to find Buffett-like stocks before he actually buys in.

16

So now, let’s look at the actual model in some detail. The first step, as I

mentioned, Buffett wants to see consistent long-term predictable earnings.

So, the model goes back 10 years to see if profits are growing over time and

consistent. Over many 10-year periods, you’ll see economic expansions and

contractions, so we’ll give you a sense of how a company’s profits react to

economic downturns and competition. By looking at 10 years worth of

earnings, it helps give Buffett confidence in the future earnings power of the

company. One reason Buffett avoided technology stocks up until recently is

that he didn’t fully understand them, but also that their earnings were highly

variable, and what he likes to see is more consistent earnings. Next, wants to

see higher-than-average long-term returns on equity and capital. So, the

model looks at 10 years of ROE and ROC or return on assets for financials. If a

company has long-term, above-average return on equity and capital, this

indicates that there’s a competitive moat to some extent around the business.

It could be a strong brand name. They could be a low-cost producer. They

could be a toll bridge type company. But something about what they do and

the company has protected profits and allowed them to get above-average

returns. Buffett also doesn’t like too much debt. He wants to see the earnings

can be paid off in a reasonable amount of time. So, if debt can be paid off

with five years worth of earnings, that’s good. The model favors that.

Businesses that have to invest too much in capital expenditures aren’t as

17

appealing as those that don’t, so as a result, this model rewards positive free

cash generation. Share buybacks are another positive. So, the model wants to

see that companies are buying back shares at hopefully fair and reasonable

prices. It then looks at retained earnings, and it wants to see that retained

earnings have contributed to profits. So, the model we run takes the total and

retained earnings over the past 10 years and compares it to the total gain and

EPS over that same period. And then to determine if the price is right, using

two different methods, the ROA and the EPS method, and it’s trying to

calculate the expected future return. And what this model essentially looks for

is stocks that are going to have the potential to compound out at at least 12

percent annually over time.

So, here’s an example of the Buffett model, and I’m going to stick with Apple

because our model still rates Apple, actually, very highly. I don’t think it’s 100

percent. It might be like 93 percent. It’s on a scale of zero to 100 percent that

we score each stock. So, in this case, we can see here’s the 10 years of

earnings for Apple. The model does allow for a dip in earnings, and there was

a recent dip, and that’s OK. The growth rate is still really, really good, and

based on the average three-, four-, and five-year EPS growth rate, it’s about 10

percent. But it does look at the full decade, the last decade’s worth of

earnings, and it wants to see growing, consistent earnings. It then goes down

18

to say, “OK, is there a moat around the business looking at return on equity

and return on capital,” which are well above what the average is in the stock

market, which is around maybe 12 percent or 13 percent. I mean, we’re

talking 49 percent, 28 percent on ROE and ROC respectively here, so that’s

really good. It then says, you know, “can earnings pay down debt?” So, Apple

has 92 billion in debt, and it’s generating 53 billion a year in earnings, so that

can be paid down in less than two years, which is very good. It’s generating

positive free cashflow. That’s a positive. Shares outstanding has actually

fallen, so Apple has reduced the share count from 5.5 billion to 4.5 billion over

the last five years. And then the question is, “Has management done a good

job in generating returns on retained earnings?” And based on our analysis

and what’s captured in this model, looks like management has shown it can

generate a 16.4 percent return on retained earnings. And then lastly, using

those two methods I talked about before, it says, you know, “What type of

return could I expect given these past fundamentals, and growth rates, and

earnings expectations in the future?” And our model is predicting a 13.7

percent return on Apple, which is enough for it to pass this strategy very

strongly. So that’s the Buffett strategy in some detail.

The next model that I’d like to walk you through is in our quality and yield

category, and what we’re going to do is we’re going to look at a model that

19

rewards low volatility and conservative stock. So, this is interesting. The

fundamental insight here is that by combining low volatility stocks with other

things like momentum and high shareholder yield or net payout yield, you can

actually get a more robust, consistent investment strategy that still exhibits

that performance potential. So, the strategy overview is this model seeks out

low volatility stocks that have strong momentum and high payout yields. The

source and the evidence behind this strategy -- this was developed by a

European money manager. His name is Pim van Vliet. He wrote a book called

High Returns from Low Risk: A Remarkable Stock Market Paradox and a follow-

on research paper, “The Conservative Formula: Quantitative Investment Made

Easy.” That is available for download on SSRN, where you can read more

about the model. But in van Vliet’s book and his follow on study, he was able

to back-test this approach and develop 100 conservative stock model

portfolio, and he was able to test it from 1929 to 2016, and his back-tested

results showed a 15.1 percent annualized return versus 9.3 percent for the

market in the back test. We also run a more focused model portfolio based on

this approach on Validea. Since 2009, our model portfolio has produced a

hypothetical return of 16.7 percent per year versus 13.2 percent per year for

the market. I think the investment thesis behind this one is, you know,

contrary to belief, research shows that investing in low volatility stocks can

enhance returns in addition to reducing risk, and I say we’ve been -- stocks

20

have not -- well, they were volatility at the end of 2018. I mean, the volatility

has sort of been muted lately, but we know as investors that can change at any

point in time, and these types of conservative stock I think should be less

volatile than the market given their characteristics. So, if volatility picks up,

these are the types of stocks that should be beneficiaries.

So, let’s walk through the model in a little bit of detail here. So, the first thing

we need to do is we want to segment out the largest stocks in the market,

because that’s where the most conservative are. So, what the model just

simply does is it screens on market cap and it wants to have only the thousand

largest names based on market cap to be eligible for our selection universe.

What it then does is it wants to identify the stocks with the lowest amount of

volatility. So, to do this, we look at the standard deviation of returns. With this

standard deviation, stocks that actually are in the lowest 50 percent based on

their standard deviation or returns are the ones that make it to the next stage.

From there, it wants to see stocks are exhibiting price strength relative to other

stocks. So, there’s an intermediate momentum factor here. That specific

criteria is a 12 minus one-month momentum criteria. We’re going to talk

about that in a little bit, but that’s the actual investment criteria used. It sorts

on that. And then it also sorts on net payout yield. So, it’s not just looking at

dividend yield here. It’s looking at dividend yield plus share buyback yield.

21

Those are two ways that a company can return cash to shareholders, and it

sorts on that as well. So, you’re getting two shorts based on -- you’re getting

one based on momentum, one based on net payout yield. What happens

then is there’s a combined rank. So again, it’s this combined ranking concept.

Once we have the combined rank, we can see what stocks are in the top one

to two percent of all stocks in the universe, and from there, those are your top

performers for further investigation. So, here’s an example of a stock that

passes very strongly: Fortune Brands and Home Security (sic). The company’s

market cap is nine billion. It places it in the top one thousand stocks in our

universe. The stock’s three-year standard deviation is 23.7 percent versus the

market median of 31.8 percent. We then rank the stock based on a 12-month

minus one-month momentum, which is 68.1. It places it in the tenth percentile

of all stocks in the database. We then rank on net payout yield, which is 4.3

percent. This ranks in the 17th percentile of all stocks in the database. The

combined ranking puts fortune brand home security in the top one to two

percent--

We’re going to now move into the growth investing part of the categories

here, and we’re going to talk about a model based on Partha Mohanram. And

what this model is looking to do is looking to identify growth stocks that have

the potential for continued growth. So, the key investing insight here is that

22

although value stocks in the group tend to outperform growth stocks over long

periods of time, the best-performing individual stocks come from the growth

segment of the market. So how do we identify these? At a high level, the

strategy looks for low book-to-market stocks that exhibit characteristics via

their fundamentals and trends in those fundamentals that are associated with

sustained future growth. As I mentioned, the model is based on Mohanram’s

academic research. Partha Mohanram’s a professor at the University of

Toronto. He wrote a paper, Separating Winners from Losers Among Low Book-

to-Market Stocks Using Financial Statement Analysis. This is also available for

download on SSRN. In his paper and through his test, Mohanram was able to

construct a back test of the model from 1979 to 2001, so over a 22-year period.

The model showed a return of 20.3 percent versus 11.3 percent for the market.

Since 2006, we’ve been tracking this model as well. It’s one of the very best

performers on Validea. I think it’s benefited to a large extent from the

outperformance of growth stocks over the past 10 years or so. But in our

testing and our model’s performance, the strategy has produced a 15.3

percent per year return versus 6.8 percent for the market over that time frame.

The investment thesis that expensive stocks generally underperform the

market, but by applying a series of fundamental strength and reinvestment

criteria tests, you can identify outperformers among that expensive group. As

I mentioned, you know, growth has been a good place to be in this market

23

environment, and I think should that trend continue, which it looks like it is

continuing here early in 2020 -- value has tailed off a little bit and growth has

kind of picked up here -- this model looks poised to continue the benefit from

that.

So, Mohanram developed a scoring system called the “g-score.” And what

effectively that does is it scores stocks based on profitability, cashflow factors,

and reinvestment criteria. So, the g-score is an eight-point scale that is applied

to the 20 percent of stocks in the market with highest price-to-book ratios. So,

his initial cut of the universe was looking for the most expensive stock based

on price-to-book, which is the inverse of book-to-market. From there, the

investment criteria used by Mohanram can be categorized into three different

buckets: profitability and cashflows, business and earnings durability, and

reinvestment in the business. The first criteria in the model is to identify those

stocks that are the most expensive. So basically, what this says is it says it’s

looking for the 20 percent of stocks in the bottom of the book-to-market ratio,

which is, remember, just the inverse of price-to-book. So, it starts out by

looking at the most expensive stocks first. From there, we want to step into

the profitability and cashflow criteria. So, this model looks at return on assets.

It wants to see return on assets above the industry median. Cashflow from

operations to assets above the industry average, and then cashflow from

24

operations to assets versus return on assets above the industry average. One

of the things that’s unique about this model is the focus on comparing and

rating companies against their peer group. So Mohanram realized that stocks

in different industries have different characteristics and dynamics. And so, to

handle this, the model really only compares companies against their industry

peers. The next part of the model is to reward consistency. So, what this is

doing, it wants to look for and identify companies that exhibit consistent

growth and are more likely to sustain that growth in the future. So first it looks

at the variance in the firm’s return on assets. It wants to see that below the

industry average, and then it wants to see variance in year-over-year sales also

below the industry average. Once we have that, we can look to see is the

company investing in the future? Are they investing in areas that can help

generate future growth? So here we’re looking at RND, CAPX, and

advertising. For example, the model wants to see RND to assets, CAPX to

assets, and advertising to assets above the industry average.

So, here’s an example of Intuit, the tax and accounting software maker. It

scores in the lowest 20 percent of stocks based on the book-to-market, which

means it’s in the 20 percent. That’s most expensive. So that’s indicating that

it’s got an expensive valuation, and so therefore, it’s probably a growth-like

company. We’ve been looking at return-on-assets. Return-on-assets, with

25

24.4 percent. That’s above the industry average. Cashflow from operations is

26.6 percent, which is above the industry average. And then it wants to see

the cashflow from operations to assets is higher than return on assets. So

that’s a positive. We then look at the variance in return on assets. That’s

below the industry average. The sales variance is also below the industry

average. And then it looks at the advertising-to-assets, CAPX-to-assets, and

RND-to-assets. And all of those things are scored based on the g-score. So,

Intuit is one of only four stocks that actually gets 100 percent according to this

model. So that’s an interesting growth model, and I think there’s some factors

in there, there’s some uniqueness in there that hopefully you can learn from

and maybe even utilize in your own investment research.

The last model we’re going to look at is we’re going to look at a way to identify

sort of the purest, best-performing momentum stocks. This is probably the

simplest in that it only uses two criteria, but it’s also a model that some

investors may not be able to really wrap their head around because it involves

buying stock that have already gone up a lot in price. This model was

extracted from the book Quantitative Momentum, which was written by

Wesley Gray and Jack Vogel. They are the founders of Alpha Architect, a

quantitative investment research firm. The key investing insight here is that if

you’re going to use momentum as part of your selection process, it’s better to

26

look for stocks with consistent momentum. So for example, if you have two

stocks that are both up 60 percent over the past 12 months, the stock that

produced that four-to-five percent return month in and month out over the last

year is more attractive versus a stock that is up 80 percent and then down 20

percent in the last two months. So, the model doesn’t look at anything other

than price performance of stocks and the consistency. So specifically, we’re

looking at the one-year momentum excluding the most recent month. That’s

the momentum factor criteria. And then what we want to do is we want to find

the stocks with the least amount of volatility within that high momentum

group. The source and the evidence, as I mentioned, is Quantitative

Momentum. In their book, they were able to back test the strategy based on --

they went back to 1927. So, from 1927 to 2014, the strategy produced a

simulated return of 15.8 percent versus 9.2 percent for the S&P 500. And on

the Validea site, we run a very focused version of our version of Quantitative

Momentum model that we’ve been able to implement, and in 2019, it was our

best-performing model on Validea. It was up 48 percent momentum. Had a

good year last year. The investment thesis behind this, I think, is that

momentum is a robust factor. It’s been proven academically. And it’s also

been pretty strong in recent years. And so, I think if you’re going to utilize

momentum, utilizing a clean strategy and looking at these two variables could

be useful in today’s market environment. So, to start, this strategy looks at

27

one-year momentum, as I mentioned, and it excludes the most recent month.

So academic research has indicated that stock momentum tends to persist in

the intermediate term but mean reverts in the short-term and the long-term.

In this particular strategy, what we want to find is the top 10 percent of stocks

using this 12-month minus one-month momentum criteria. What we then

want to do is we want to reward the stocks that have the most consistent

momentum. So, in order to calculate this, we take the percentage of days in

the previous year that a stock has produced a negative return, and we subtract

the percentage of days a stock has produced a positive return, and then we

multiply by the 12 minus one-month momentum. The lower number gives us

the better, more favorable result. Stocks are then ranked, and the top one-to-

two percent of stocks are the best according to this strategy. So again, it’s

another order rank ranking system, looking for the stock in that momentum,

strong momentum camp with the most consistency.

Here we have AppFolio. As I mentioned, this is a pretty -- there’s not many

criteria. So, the 12 minus one-month momentum is 93.8 percent. It ranks in

the sixth percentile of all stocks in our database. The stock is then ranked by

return consistency, and based on that ranking, this stock is in the second

percentile of all stocks in our universe.

28

So, that is the overview of the models. I just want to end real quickly with five

tips that I think are important sort of to bookend this with that I think can help

you become a more successful investor if you choose to utilize strategies like

this. And the first is really just a rehash of what Lynch said and what we’ve

pretty much talked about this entire presentation, which is look at the

financials and valuations. By doing this, it will help remove some of the

emotional biases that we have, I think, when it comes to buying and selling

stocks. There’s plenty of great tools out there, including the tools on

Fidelity.com that do that. If you do start to utilize systematic strategies, I think

it’s important to find something you understand, believe in, and fits you as an

investor. We sort of talked about that in that slide with the categorizing of

strategies. And I think if you can do that, you’ll have a better chance to stick

with things when things get tough. Another sort of tip I thought might be

helpful, and some of you probably already do this, but consider using a

consensus of screens if you use screens. If you see a stock that pops up on

different types of screens, that means it’s scoring highly based on multiple

different things. And so, I think if you see that consensus, it could be a good

indicator for the security if it’s popping up on multiple screens simultaneously.

The next point, which is number four, fundamental stock selection and

modeling is far from perfect. I think if you’re going to do stuff like this, you’ve

got to understand the realities of using these models. So, there’s two points

29

here: one, we track model portfolios based on all these approaches. And I can

tell you with certainly, none of them work all the time. There are actually

plenty of periods of underperformance, and in some cases, it can be multiple

years. So, these strategies are susceptible to periods of underperformance

just like anything else. The other datapoint I just want to share with you is

even the best models -- we track what we call the “accuracy rating,” which is

how many positions these strategies get right. So even the best performing

models, they hit about 60 percent accuracy. So that means out of ten

positions, six positions are winners, but four end up being losers. The simple

fact is none of these come anywhere close to batting 100 percent accuracy

and finding all winners. And then lastly, there’s a cyclicality in quantitative

stock selection models just like in styles, and in factors, and in probably a lot of

the mutual funds that you hold and buy. The last five and 10 years of the

market has seen many trends, growth stocks over value, mega cap stocks have

performed extremely well. U.S. has dominated international. And it probably

seems obvious, but these trends also find their way into influencing models

like the ones I’ve talked to you about and the ones that we run. But the other

important point to remember there is that because these things come in

cycles, just because something has worked over the last five years doesn’t

mean it’s going to be what works best over the next five or 10 years.

30

So, with that, I’m going to conclude my formal part of the presentation. I hope

you found this educational and I left you with something you can use and

incorporate in your stock selection process to help make it more effective and

efficient for you. Thank you very much.

END OF AUDIO FILE

31