Listen on:

Factors that lead to portfolio outperformance.

Join SmartBe Investments Chief Strategy Officer Mr. Gavin Graham for a broad and simplified perspective on “factor” investing with Brandon Koepke, PM and Chief Technology Officer at Alpha Architect. Brandon discusses his unique background that lead him from Disney and Amazon to Alpha Architect; the evidence in favour of the “value” and “momentum” factor, and; the particular timely opportunity for “value”. All this and more in this segment of the Gavin Graham Show.

About the Speakers


Brandon Koepke, CTO and Senior Portfolio Manager

Alpha Architect

Mr. Koepke is a full-stack developer and currently serves as the Chief Technology

Officer and Senior Portfolio Manager for ETF Architect. He has fifteen years of

experience working with technology ranging from small businesses to large

corporations including Disney and Amazon. Mr. Koepke is a CFA® charterholder,

has a BSc in Computer Science specializing in Software Engineering at the

University of Calgary, and a BComm in Finance from the Haskayne School of


About Gavin

Mr. Graham joins SmartBe with over 30 years of experience in Canadian and international markets. A graduate of Magdalen College at the University of Oxford, Mr. Graham has directed investment strategy at numerous institutions, such as the Guardian Group of Funds in the UK and BMO Asset Management in Canada. In addition to his position at SmartBe, he serves as a contributing editor to The Income Investor, a respected online Canadian investment journal. The Gavin Graham Show is a platform for Gavin to share his wealth of knowledge with the SmartBe community, as well as engage in stimulating conversations with peers and thought leaders.


[00:00:00] Gavin: Hello, I'm Gavin Graham. And this is The Gavin Graham Show sponsored by SmartBe Investments. Today we're very pleased to welcome a Brandon Koepke from Alpha Architect, which is based in Philadelphia. Which is a leader in factor investing, particularly the factors that very extensive academic research have indicated consistently add value over time and help to up form index.

Thanks very much for coming in, Brandon.

[00:00:28] Brandon: Thank you for having me.

[00:00:29] Gavin: A little bit about yourself. Firstly, a Canadian who's been working south of the border, but has now decided to return home. Very sensibly of course. If you could tell us a little bit about your background and how you, first of all on the academic side, what got you into working in finance and then how your career path progressed from there?

[00:00:45] Brandon: Well my background, may be a little bit unusual. So I actually started more on the computer science side, double majored at the University of Calgary, Computer Science and Finance. Always wanted to do something that was in both. And started working first at Solium Financial, which kind of checked both those boxes.

They were an employee stock option plan management system. Ended up at Disney Interactive helping to write computer games, which was pretty interesting. From there I moved on to Amazon, working on a team there called Rack Retirement. It's very interesting work there. Met a lot of smart people. From there I was actually working on my CFA at the time. Always wanted to get back into finance and met Wes Gray sort of by chance and actually through an early version of SmartBe. And they were looking for someone who could fill a portfolio management role as well as handle their it CTO software development needs.

So I checked both those boxes.

[00:01:43] Gavin: Where did you Wes?

[00:01:44] Brandon: Oh, that would've been I guess would've been 2017 the first time that I met him. I'd actually been invested in their funds personally since about 2013, 2014 I guess. Right from the start pretty well.

[00:01:55] Gavin: Right. You were a believer from the start. Incidentally you said you were developing video games for Disney.

You didn't have anything to do with Kingdom Hearts did you?

[00:02:01] Brandon: I didn't, but I did work with the same studio that was developing that. Yeah.

[00:02:06] Gavin: No, it's my youngest daughter's favorite game and I actually got taken through several levels. She's way better at it than me. I understand that you had a particularly wonderful reward for saving Amazon an enormous amount of money. Would you like to tell us a little bit about that story?

[00:02:18] Brandon: Yes. Yes. So that was Rack Retirement team on Amazon Web Services, AWS. What is Rack Retirement? What the heck am I talking about? If you ever seen on television, you see these massive, massive server racks. Think about what a data center looks like.

Right? A big warehouse. Great big computers in there, but there's sort of a logistical problem. You have existing old computer infrastructure that you wanna get out of that data center and bring in newer infrastructure that you can bill out at a higher rate. So there's this optimization problem, you know, which racks do we "quote" retire, disable, and remove from the data center and bring in new racks? And there's an optimization problem. On the finance side we actually work very closely with the finance management team. They gave us some predictions on what they were worth. We use those as well as a forecast of demand to figure out which rack families, which particular racks, were ideal to retire.

And then we actually built a workflow system that basically did everything. It migrated any existing customers off the racks, shut everything down, sent the alert to let the team know "come get it it's ready," and bring in the new racks.

[00:03:29] Gavin: And that saved Mr. Bezos quite a lot of money?

[00:03:31] Brandon: Yeah. So we could actually quantify it, which was interesting.

So the first year that we completed that program, we saved roughly $250 million, and the subsequent year... this is a team of eight of us. So pretty small team. And of course I don't wanna downplay too much. There were lots of other teams involved here and people physically moving the racks and other things.

We built on a lot of other infrastructure, but the second year I'm told saved somewhere between $700 and $900 million.

[00:03:58] Gavin: And your reward for this?

[00:03:59] Brandon: And our reward, well actually so we got recognized at the all hands conference, which was good. But our prize, if you will, was a inflatable Nerf football. One for the team of eight.

[00:04:10] Gavin: No bonuses, no large stock options.

[00:04:13] Brandon: Nothing that I saw.

[00:04:14] Gavin: Then that's perhaps, without wanting to be unfair, possibly one of the reasons why Mr. Bezos continues to be one of the wealthiest people around.

[00:04:21] Brandon: I think that's fair.

[00:04:23] Gavin: So Alpha Architect we've had many of their staff here to tell us about the way that it works. Could you very briefly summarize why Alpha Architect thinks that Value and Momentum are the factors which consistently deliver outperformance over time?

[00:04:37] Brandon: So we wanted to find factors that work across, essentially, any market. Like we're only interested in factors that are extremely robust. Both Value and Momentum, it doesn't matter where you look, you know, domestic, international, emerging markets. So across all different countries, stocks, bonds, equities, commodities, currencies.

Basically anywhere that you look, you can find that Value is gonna be there and even out of sample and over very long time horizons. So we've done back tests back 200 years as early as we can get the data. Everywhere we look Value shows up. It's the same story with Momentum. It's essentially everywhere and Momentum is an even broader class of assets because, you know, certain things you think about, like for instance crypto, right?

What's the fundamentals behind crypto investing?

[00:05:32] Gavin: I'm sorry. Did you say fundamentals in crypto? Do continue.

[00:05:35] Brandon: And there's tons of asset classes that are like that, but Momentum basically works there as well. I mean, it works quite literally everywhere.

[00:05:42] Gavin: And I'm sorry, this is Price Momentum, correct?

[00:05:44] Brandon: So I'm talking about cross sectional Momentum. So within an asset class, other people you'll hear the term Momentum. Momentum Trend is what I would say. So Time Series Momentum is what we would call that. Time Series Momentum is where you're trying to figure out, do we go long, or flat, or potentially short a particular asset.

So relative to its past performance, is it performing well or not? Whereas this is using the same sort of algorithm, but within an asset class.

[00:06:13] Gavin: So that's what you call Cross Sectional.

[00:06:15] Brandon: Yeah. So, for instance, if you're doing Cross Sectional Momentum on Canadian securities you buy the securities that had the highest Momentum. Relative Momentum compared to all the other Canadian equity securities.

[00:06:28] Gavin: As you say, they have been going, getting on for a decade now,. How much in the way of assets do Wes and his team look after?

[00:06:34] Brandon: It's around a billion and a half. I mean with with the recent–

[00:06:38] Gavin: U.S.

[00:06:38] Brandon: U.S. With the reason market route I haven't been checking as frequently.

[00:06:41] Gavin: Yeah. Well, again, we're not asking to three decimal places.

It's simply to give a feel. It's obviously been sufficiently attractive and sufficiently well performing to have got over a billion dollars into it which is pretty remarkable given it doesn't have a big institutional backing. And it's not a major, it's a portion of a larger financial conglomerate, correct?

[00:07:01] Brandon: Yeah. Well, we started basically direct to consumer. That's how the firm got started. Actually Wes Gray wrote this Alpha Architect blog and basically just wrote on all this quant factor geek stuff that, and I mean that's what attracted me to the firm. Which might say something about me. And more recently we've been looking more institutional, we're starting to get more larger asset managers that are coming on. Prior to that it was mostly clients who were really sold on the idea of Value Momentum, and they really got it and they love this sort of deep quantitative stuff. They love reading all the stuff that we have. So then branching off trying to do more on the institutional side it's trickier because we make these premiums or we expect in the long run and make premiums off Value Momentum because they do have periods where they don't perform the same as a broad market index.

They'll underperform. If people don't understand it, they'll come in, they buy the asset, they hold it. Then they have loss relative to the broad market, not necessarily a net loss, just they didn't do as well as the S&P 500, for instance. And they sell. That doesn't do us any good because they're in, they're out.

What was the fees for us? It is not worthwhile for us to, to be doing that. It's sort of performance chasing. And that client got hurt. And we really, really don't want that to happen. So we really try to make sure we educate clients before they enter the product and that they're in it for the long haul to make sure they actually harvest those expected premiums.

[00:08:26] Gavin: Now that brings up a very interesting point, and I know that you've been doing some presentations with us at SmartBe to actually tell the story about why those periods of underperformance for value assets, particularly, might be setting up a major opportunity to get some serious relative outperformance.

So as you say, there are periods when these factors are underperforming the broader market. Could you tell us where you think we are as far as value, as opposed to the broader market or indeed, to other styles of investing?

[00:08:56] Brandon: It seems to not matter what fundamental factor we look at. If it's fundamental based, so book to market, sales to price, earnings to price, IBIT to enterprise value. Across the spectrum, if you're measuring on fundamental value of cheapness, Value looks very cheap. There's almost sort of like how cheap is Value compared to it historic average, relative to the broad market. Looking from 1990 until today, you sort of see two very large peaks where Value gets extremely cheap.

So one of those, probably first one that comes to mind for people, is gonna be the tech bubble. You see 99th percentile value spread. So the spread between how cheap value is relative to the broad market. And, of course, subsequent outperformance of value after that. In the U.S. was close to 30%. Compounded annually for five years straight, from the start of 2000 through the end of 2004.

The second time that we see this in the data is 2008. So not quite as extreme as what happened during the internet bubble. But still very significant Value spread, you know, in the 90th percentile, 95 depending on what measure you're looking at. And then more recently, and again amazing performance from 2008 up until the end of 2012 and into 2013. Value just has a significant rally. We see the same exact setup today. And actually, depending on what multiple you look at, it's even more extreme than what we saw during the internet bubble. Well what's interesting about this time it's more broad base. So the internet bubble it was solely the technology stocks that were very expensive.

But if you look today It's actually impacting a wide variety of assets. So I know we had a discussion earlier today about some of the rates, like some rates are actually very expensive on a yield basis. Like extremely expensive. And, to your point, they've got fixed rates, but if they're only paying 2% cap rate any sort of decline in the principle value of their holdings could be very significant. Financial technology is a big one. Like firm holdings is one name that is just, they may not be solvent. It's hard to say, right? You look at these other securities. Coinbase in the crypto space, unfortunately selling off very significantly. They're almost a value security, but we may have to watch for financial distress there. Utilities. Everywhere!

[00:11:19] Gavin: As you said, it's not merely the technology stocks in the way that it was in the internet bubble. There's a whole bunch of stuff that's pretty expensive, which means that value is actually even a better potential protector than it was in 2000 in terms of the ability to deliver relative outperformance going forward.

[00:11:34] Brandon: Well, I'm personally betting on it. So I certainly hope that's gonna work out. Yeah.

[00:11:39] Gavin: But again, it seems it was '99, 2000, 2008, 2009, and here we are in 2021, 2022. These periods of extreme measurement come along, maybe once a decade. Would it be fair to say that, you know, if you had a long bull market in a number of assets that investors become somewhat complacent and become indifferent valuation, and quite frankly are, you know, just buying it because it's been going up and expected to keep going up?

[00:12:06] Brandon: I would definitely say that's fair. There's almost like a siren song as you get value under performance, right?

The media chime just starts to get louder and louder. Value is dead. It just doesn't work anymore. You know, it's a new era. All these...

[00:12:20] Gavin: It's different this time.

[00:12:21] Brandon: Yeah. Yeah. Warren Buffet is old hat. You know, the 2000 bubble same sort of deal. You saw recently in Barons. They'll do a cover with buffet on the front page, you know, Has Warren Buffet Lost His Touch?

They change it slightly, but same energy.

[00:12:36] Gavin: It is interesting to see the same patterns of behavior occur. And maybe that's just those who don't remember the past are condemned to repeat it. At the moment then you would feel that the, just talking about the indices rather than specific products. The value index, but whether it's in the U.S. or in Canada. I know you have an international one as well in the U.S., but that's not available here in Canada. But they are the outperformance they've been demonstrating over the last year, year and a half is likely ... first of all, how large has that been? And secondly, you feel it's likely to continue?

[00:13:07] Brandon: Depending on which index you're looking at it could be more significant than 10 on a spread basis. But largely that's because growth is just having a horrific time. So value is sort of more flat, or up until very recently it seems like value is pulling back a little bit. But maybe taking one step back there. We've had a lot of advisors that have been coming to us and saying after that recent sort of value outperformance that happened late 2021, and now a little bit more going on this year, they come to us and saying, you know, have I missed it?

Like, did I miss the value run? You know, I gotta wait for it to come around. Even looking today, we're still above the 95th percentile for cheapness. So it did come down. Like you saw a very precipitous fall from this 99 percentile, but we're still a very long way from normal.

Now that isn't to say that it, you know, spreads couldn't widen again. It definitely could happen, but it certainly seems primed to outperform. Like value value it looks very cheap and very compelling.

[00:14:07] Gavin: And again, looking at that one would have to say if one was going to use somewhat hackneyed baseball analogy, we're what second or third inning rather, you know, we have a long way to go yet in the value bull market. The value outperformance.

[00:14:22] Brandon: Yeah, we're barely off the plate.

[00:14:25] Gavin: Yes. Well, of course it depends which umpire is calling balls and strikes as we saw recently. But in that case you would obviously feel that that would be a sensible place for people to put some of their money. The actual methodology for the construction of the indices, you are a very concentrated portfolio, correct?

[00:14:41] Brandon: Yes.

[00:14:42] Gavin: So what 50 stocks in the U.S.?

[00:14:44] Brandon: 50 in the U.S. We actually used to do 40 but recently broadened it to 50. And I believe the Canadian index is 20 names.

[00:14:51] Gavin: And that's all equally weighted?

[00:14:52] Brandon: Equally weighted. Yes.

[00:14:53] Gavin: And then rebalanced every three months with whatever names the model has produced.

[00:14:57] Brandon: Rebalanced quarterly. Yes. And there's strong evidence and that's actually kind of goes back to our original discussion about why we believe in the Value and Momentum premiums. The more frequently you rebalance them, the higher your premiums are. That's why we're quarterly. On international, at least in Value, we do six months because trading costs are increased and we wanna make sure the premium is high enough to overcome those. But domestically we do quarterly.

[00:15:22] Gavin: And there's the people who go, oh, you've got a very high turnover, you know? That's all trading costs, spreads, whatever. But, as you say, the evidence would be that in fact it generates more premium and that's more than sufficient to overcome any drag. Although it's interesting that point on trading costs internationally, because obviously there are some particular wrinkles of stamp duty in the UK, for example, which make it more expensive.

Therefore the presumption is that it doesn't look anything like the broader index. You are genuinely going to get diversification by putting a small portion of your portfolio into these factor funds.

[00:15:53] Brandon: So it's actually a very difficult thing to quantify. So we had to build an in-house software system to do this.

I was just trying to explain this to a client is challenging. So we built these visual factor tools show you exactly, you know, what's the overlap between our portfolio and others. But the single measure that we tend to quote is active share. And what we do there is we actually look through the underlying portfolio holdings and we compare them to every other fund that we can get data for.

So we have the entire ETF universe in both the us and Canada. And so for instance, if you pull up the SSPY Fund.

[00:16:26] Gavin: Which is the the spider on the S&P 500,.

[00:16:28] Brandon: It has 0% active share. Like there's a million different funds out there that exactly track the S&P index. It just has no active share at all.

It exactly tracks. And there's lots of other funds. That's why those are so cheap, right? I mean, you can get SSPY for six basis points. Right? And there's futures products that track the same sort of thing. So there's lots of ways to get access that beta cheaply. If you look at our indexes, we're consistently above 90%.

Usually even when we need to start rebalancing the fund, even then we're still way above 90%. Usually above 95% active share. So we have extremely low overlap with other funds.

[00:17:06] Gavin: Which makes you a great diversifier because of that lack of overlap, you're actually gonna reduce the overall volatility of your portfolio by introducing Alpha Architect funds.

[00:17:15] Brandon: Well, I can tell you that we'll be very different. Very, very different. With certainty.

[00:17:21] Gavin: And again, that's back to your point about people who buy in are not realizing that there is gonna be tracking error and probably fairly substantial tracking error because you look very different from the underlying index.

[00:17:32] Brandon: Exactly. That's, I mean, that's our game. We are the tracking error team.

[00:17:38] Gavin: Actually that that's a very good way of putting it. And again, people's tolerance for that. And you have a chart, I think, in your most recent presentation, showing those periods of relative under deployments and you have risk of being fired or the arrow's pointing to when the the differential performance is the greatest. Would you say that's maybe one of the reasons why it is sometimes a difficult sale? Because there is career risk there. If you actually do this and it doesn't work within a time scale, then the person doing it could be vulnerable.

[00:18:05] Brandon: Yeah. I mean, it's extremely difficult and that's why we started off direct to consumer because we knew that we had to sell individuals directly. We had to make sure that they understood what the strategy did. Because if you're an advisor owning these funds and you have a year where your fund is down 5% relative to the S&P that would be mild tracking error.

Right? And it can go the other way too. Right? You can be up 10 and the market's down five, like it's just expect to be different. You know, imagine being an advisor, sitting there holding these and your client says, why did it do that? And you hadn't adequately explained to them in advance how different this fund was.

Right? What are they gonna do? Well, first they're gonna try and get outta the fund. And then if they don't, they're gonna fire you. And if you tell them you do somehow convince them to stay in it. And three years later, they still don't understand what it does, three years later it's still down, you're for sure out. They're moving their money somewhere else.

And that's lose-lose, right? So we really try to avoid doing that. Make sure everyone fully understands. So we've got tons of materials that explain exactly how all this stuff works. I mean, we're quite literally open book. Like if you wanted to implement these exact strategies yourself, Wes Gray, Jack Vogel, quantitative value, quantitative momentum, you can buy the books.

Check out formulas. You gotta know quite a bit of math and stats. For me, it'll keep me awake, but for some people it sends them to sleep, puts them right to sleep. So even if you don't enjoy it, good sleep aid.

[00:19:27] Gavin: But nonetheless, as you say, it is transparent and you know what you are getting, and therefore you ought to be aware of that before you buy these. Because if not, then you will suffer periods of major tracking error.

And, Saul's Law says that that will be . Negative to start with. Do you have an emerging market product at the moment or are you thinking of doing that?

[00:19:46] Brandon: We're actually also an ETF white label provider. First of all, what do I mean by that? Well, we'll actually launch funds for other individuals or companies who want to launch an ETF.

So they'll come to us and they say, here's my strategy. Here's my idea. Is it possible to do this in the ETF wrapper? We figure out how to do it, if it can be done. If it can't we make the modifications to make it work. And then they send us names and waits and their job is send us their next constituents, we'll trade the portfolio, go sell.

So our first fund doing that was per toll and she launched an emerging markets fund that I think is very, very compelling.

[00:20:25] Gavin: Is that the Freedom Fund?

[00:20:26] Brandon: Freedom Index. Yes. Yeah. Very interesting product because her whole thing is freedom waiting. So, I'm gonna butcher the exact index that she's tracking...

[00:20:34] Gavin: Governance, democracy, human rights, reluctance to invade neighbors without provocation.

[00:20:40] Brandon: Exactly. Yeah. Yeah. So obviously Russia is out.

[00:20:42] Gavin: China is not included.

[00:20:44] Brandon: And so she was smelling like a rose lately, which has been nice.

[00:20:47] Gavin: So therefore, there's an example of you putting something together, add to the specifications, given you by the fund manager who has a good idea, but doesn't have necessarily the resources to implement.

How would you go about doing this and then, as you say, white labeling. So that's an emerging market fund. Have you had any other interesting ideas you can talk about in terms of people approaching you? Are there any other slightly more esoteric or is it very much generally just sticking with the cooking? The value of momentum?

[00:21:12] Brandon: Oh, we see everything under the sun. If it's out there. I mean, we've got crypto indexes that they're trying to buy crypto miners, for instance. We have some portfolios that track sort of gen Z based securities. We're all, absolutely all over the place. If you can put it in the ETF wrapper we're at least seeing it.

[00:21:31] Gavin: Which is good. And again, it's an interesting additional income stream. An additional example of the flexibility that the resources that you have, the methodology that you have, can be useful. Finishing up here, Brandon, is there any one sort of overarching message that you'd like to leave the listeners to the podcast with, apart from they should be following your footsteps and buying value?

[00:21:51] Brandon: Well, I would just say if you have a very long time horizon, and you have the right temperament, you can sort of put your portfolio in a box and just forget about it. Ideally 10, 20 years. If you can have that kind of time horizon, you know, both value and momentum historically have significantly outperformed the market and momentum diversifies value in this sort of wonderful way.

At least historically. And so if you've got a long time horizon, these are very compelling factors to be looking at right now.

[00:22:19] Gavin: So in fact, if you had an educational savings plan for your children, who've just been born and are gonna be going off to college in 18 or 20 years, this would be the ideal thing to put in it because they don't care about volatility.

They don't care about tracking error. What they want is to have some growth in their assets so that they can pay those extremely reasonable university tuition fees, or perhaps not by the time they get there.

[00:22:39] Brandon: Mm-hmm. That certainly seems like a good time.

[00:22:41] Gavin: Excellent. Well, that's been really helpful and thank you for some of those additional insights, because as you know, with our longstanding relationship at SmartBe with Alpha Architect we had the opportunity of having Wes, and Jack, and now yourself to tell us, you know, about the product and how it's working.

But I think it would probably be fair to say that the stress on value as something to look at is a relatively recent development, perhaps because of the degree of cheapness. I know that, Wes, perhaps, was initially reluctant to say a great deal about what ought you to be doing, but that certainly when he was on here in February he was saying, yeah actually it does look to be a really good time to be looking at value.

So it's reassuring to have that reinforced. Thank you very much, indeed.

[00:23:20] Brandon: Thank you.

[00:23:26] Gavin: Thank you very much for listening to The Gavin Graham Show sponsored by SmartBe Investments. If you would like to learn more about the subjects discussed today, please go to our website, at or @smartbeinvestments on any social media platform.

[00:23:50] Disclaimer: The information contained in this podcast is: (i) for Canadian residents only; (ii) for informational purposes only; and (iii) not intended to be, nor shall it be interpreted as, a recommendation or the provision of advice (including without limitation financial, investing, sales, marketing, legal, accounting, or tax advice), and shall not be relied upon by you in that regard. SmartBe Investments Inc. cannot guarantee the accuracy of any information contained in this podcast that is derived from third-party sources, although we believe those sources to be reliable.

SmartBe Investments Inc. and its affiliates may receive compensation derived from investments in, or sales of, its products, or both. SmartBe Investments Inc.’s products described in this podcast may not be applicable or available with respect to all investors, and may change without notice.

Past performance does not guarantee future comparable results. Exchange-traded funds (ETFs) are not guaranteed; their values change frequently and past performance may not be repeated. Investors will pay management fees and expenses, but they will not pay commissions or trailing commissions, and they may experience a gain or loss from their investment in or sale of an ETF.

You may request a prospectus [for an ETF] from your financial advisor by visiting, or by calling SmartBe Investments Inc. 403-930-8693. The prospectus includes investment objectives, risks, fees, expenses and other information that you should consider carefully before investing.  

Please consult a lawyer or tax professional regarding your specific legal or tax situation, respectively.