An Interview With Wesley Gray on Value Investing
The following interview is extracted from my book, Breakthrough Strategies of Wall Street Traders. Similar to Market Wizards, within it I interview 17 remarkable traders and investors for their investing stories and unlike other investing books, get them to actually reveal the trading-investing rules and techniques they use as well as what they tried that didn't work. If you want to learn trading and investing, start with these systems.
Wesley Gray, fund manager and author of Quantitative Value, joined me to discuss value investing ideas that really work so that anyone can approach value investing properly. Because Wes has evaluated hundreds of investing ideas by computer I asked about popular investing notions you should avoid because they are absolute failures despite all the hype. We also discussed his research on quantitative momentum strategies that you can compare with the momentum strategies of Geoff Bysshe and Marvin Appel.
Eventually our conversation got around to the topic of mega-wealthy individuals and family offices that manage their own investment funds. Along these lines Wes revealed his FACTS formula for wealth accumulation that the richest billionaires follow to maximize their investment returns over time. His book, DIY Financial Adviser: A Simple Solution to Build and Protect Your Wealth, summarizes these principles as well as how you can protect and grow your assets using simple tested models that readily beat the experts.
The whole gist of our conversation re-emphasizes the common theme found throughout these interviews time and again, namely that you should use tested models to trade or invest and avoid taxes and fees whenever possible. As a professional researcher Wes has tested countless systems to discover their pros and cons, advantages and flaws and you can keep abreast of many of his trading ideas and research findings at AlphaArchitect.com.
Wes, what we really want to do is start off from the beginning. How did you get involved in trading and investing? What got you into this field and what are you doing now that you are offering to the public? Basically what got you started, what did you go through, where are you now?
I'll go way back. I actually grew up on a cattle ranch around Eagle, Colorado, which is near Vail for people who need a landmark. When I was there I used to raise steers and a bunch of different animals for 4-H. When I was eleven or twelve years old I had this Hereford steer named Big Red. I sold it and I actually ended up with a couple grand, which when you're a kid is a lot of money. My dad said, “You know, I'm a rancher. I don't know what the heck to do with money. Call up your grandmother.”
My grandmother from Southern California had been a real estate agent in Malibu in the ‘70s so obviously she had some money at the time. I called her up and I said, “Grandma Ginny, what do I do with this money? What do you recommend?” Literally she sent me a copy of Ben Graham's The Intelligent Investor. When I received it I actually read it.
I had always been into business and thinking economically even though I was just a kid. It just was always intuitive to me so I read Graham's book and I was like, “This makes so much sense. When you do investing it is just like buying businesses. You want to buy things that are cheap and not buy things that are expensive and you want to find things that are high quality, not junk.”
I also was just enamored with the idea that you could have money and it could make more money for you. That’s because my dad would pay me five cents to flip over each bail of hay when the wires weren't set right and when you work it basically just sucks, especially manual labor. I just thought it was really cool that you could have money and it could make more money for you. In any case, I initially got exposed to the whole Ben Graham value investment stuff super early on.
I went on through life and reached the age where I started getting actual money I could trade. When I was eighteen, instead of buying a mutual fund that my dad and my grandmother recommended or what have you I was like, “Okay, I can finally do this for myself.” I literally started the minute they would allow me to open a brokerage account. I was trying to be Warren Buffett picking stocks.
Actually, my first stock ever was a stock called Swisher Sweets, which made cigars. I bought it when it was around six bucks. I bought it because it had a high return on assets, return on capital for like ten years. It was selling at a P/E of 5 or 6 and of course, just dumb luck, I buy this thing at $6 and literally three or four months later it gets bought out for $9.50. I was like, “Oh, wow, this whole value investing thing is so easy. Just buy cheap high quality firms and you'll make 50% every three months.”
Now obviously later on in life I had to eat some humble pie, but that's what really got me triggered. I learned about value investing. I had great success right off the bat and I was like, “This is so easy that I’m going to be a millionaire.”
Anyways, I got really enamored with business finance. I ended up going to the Wharton School at the University of Pennsylvania, which is like the Mecca of finance, and was in the weeds there learning all that I could. Then along that line I started working with a professor there, Chris Geczy, and I ended up basically becoming the department programming data monkey and was doing really well with that. All the professors wanted me to do all their dirty work.
When it was coming up to when I was about to graduate they were like, “Hey, did you ever think about being a professor and doing your PhD? Oh, by the way, you should go to University of Chicago because that's where we all went.” I was like, “Well, I'm an undergrad. Don't you have to do a Masters degree or something like that?” They said, “No, no, we'll get you in. We know all the right people.” So I applied for grad school and ended up getting into the PhD in Finance program there at the University of Chicago.
I started learning more finance and got more buried in the space trying just to read and learn as much as I possibly could about value investing and accounting, but also about quantitative techniques and how to think about financial markets. Fast-forward eight years because I took a sabbatical and actually joined the Marine Corps for four years, which is another story, but I graduated and ended up becoming a professor at Drexel University. I went on the job market, but since my wife was born and raised in Philadelphia she was like, “Guess where you're going?” and I was like, “Okay, I'm going to Drexel,” which was in Philadelphia.
So we come out here and through this whole process I'm writing blogs about research I've been seeing or research that I've done personally and literally out of the blue I get a call from a billionaire because he read my blog. One thing led to another, I went up there and met him in New York, started doing consulting, then started doing more consulting and we basically became full-time consultants for him. We had been doing research on security selections specifically in value investment strategies.
Eventually I strong-armed him after some kicking and screaming. I said, “Hey, for you $20-30 million is like a basis point. For me, if I wanted to start an asset management business I could take my chump change and get some money from my dad but that's not going to be very credible.” We then basically cut a seed deal in 2012 with his family and we launched a managed account on this quantitative value algorithm, which is what my book Quantitative Value is based on. We have been doing that for almost three years now. We launched an ETF equivalent about seven or eight months ago. We started with around $20 million and now we have around $260 million under management and I guess the rest is history. Now we are rocking and rolling in the business.
From inception until now isn’t really enough time to show a clear differential or big track record difference between what you are doing and S&P 500 returns, but how is it doing so far?
We've been blessed. Since we started to manage accounts we are about 300 basis points a year over the index.
Is that for all three years?
No. Our CAGR (compound annual rate of return) is around 21% in change. The S&P value is what we use as our benchmark and is 18% in change so we have seen about 3% a year extra on average, but it's obviously been lumpy.
In the stub year 2012 we did really well, in 2013 we crushed it, in 2014 we got crushed, and then in 2015 so far we are basically in line with the index. The average CAGR spread is around 3%, but it has super high tracking error and is highly active value investing so it doesn't really track the S&P perfectly. Over that time period we have had a pretty lucky run, but it is long-term stuff that we are designed for.
Would you consider your method purely value investing without any momentum component?
It's pure value investing, but we do a lot of research in momentum investing as well. However, for this value system it is 100% a systematic way to exploit value investing as a pure play so there are no, at least directly, momentum elements involved in the quantitative value system.
We have another momentum specific system that just focuses on that aspect of the marketplace. With the two approaches we allow people to pick their poison as opposed to us mixing and mashing the two whereby no one then ever really knows what they are actually buying so this is a pure play value system.
How do those two differ and is the momentum or value system performing better? Also, traders and investors would be interested in knowing about your exact rules as well as what you evaluated that didn't work. I personally want people to know about that because there are a lot of crazy ideas out there and we need to tell people that some of these notions sound good but don’t work in reality.
We have been doing value investing for a long time. Momentum investing is something we have been doing research on for a long time. We actually just launched live in a managed account here about four or five months ago on some qualified money and then we have ETFs coming out on it here around October. The reason we have never launched momentum on actual capital is because all the money we deal with is insanely tax sensitive. For value investing - because the turnover requirements are much lower - you can do a lot more tax management. That is one of the reasons, frankly, that we have an ETF. It's basically to deal with the tax problems.
We have only been able to implement value strategies. As to the momentum strategies we have been doing a live track for literally four months now, but it’s in a qualified account that is not tax sensitive. And again, the reason for this is that momentum strategies - if you want them to actually work in expectation – need a lot of turnover and you have to do monthly rebalancing or rebalances every couple of months, but that's going to generate massive tax problems. All of the capital that we deal with hates taxes more than making money so we can’t do that without an ETF wrapper.
We can talk in great detail about our momentum system if you like, but it sounds like we should probably focus on the value system for now. Does that seem reasonable?
Yes, let's focus on that first. What have you arrived at? Let's also hear about what didn’t work, if possible, so we can we can pop some bubbles of wrong ideas.
The bottom line is what is value investing? It is trying to buy the cheapest, highest quality firms you can find, right? There is a great paper by the guys at AQR – it is called “Buffett's Alpha” - where they literally reverse engineer what Warren Buffett actually does. He can talk about what he says he does, but now with the benefit of hindsight we can reverse engineer all those trades and then quantify and identify what factors he's loading up on.
Basically, Buffett buys cheap, high quality firms and he uses some leverage.
So why would you ever do something different than buying cheap, high quality firms? You want it, right? That's the Ben Graham model. Even though Buffett doesn't necessarily say that explicitly, he'll say that you want to focus on quality and price simultaneously. The reality of what he actually does is that he buys firms that are cheap first, and then high quality meaning that quality is a secondary element after we have considered price. That's basically what our algorithm does and at a real high level we can go into the details on it.
Now, what are some things that don't work that may sound very similar to that approach?
As I mentioned, we buy the cheapest, highest quality firms. A lot of value investors listen to what Warren Buffett says, but don't know what he actually does. The classic anecdote is that Ben Graham was a guy who just bought cheap stocks, period. Then Warren Buffett apparently moved on beyond that and said, “Well, with Coca-Cola at 20 times earnings I'm willing to pay a little bit more because it's so high quality and that's still a value investment” whereas Graham might say, “Listen, I just buy cheap stuff, period.”
Now Warren Buffett talks about that, but again, empirically, that's not what he actually does and there is a reason for that. The minute you move out of the cheapest stock bin then you basically shoot yourself in the foot no matter what the quality component. An example would be the Magic Formula framework where it's like 50% quality weighted simultaneously with 50% price. You mix those two things where you want to try to find this maximized quotient between quality and price. It's a very intuitive algorithm, but the problem is that empirically it just doesn't work.
Why? Because by accident sometimes - because you have 50% of weighting related to quality - you might buy a stock that is just not absolutely cheap in the bottom decile. And again, it's just the evidence that the minute you move out of the cheap stock bin with your investing dollars is the minute you start losing expected return. I'd say that is a nuanced rule that people should focus on.
All right. What I now want to know are your measurements for what's cheap and what's quality. What are your actual algorithms, measuring sticks or sets of criteria for determining these things? We know Graham used a P/E to find cheap stocks but did you come up anything different and more importantly, what have you tested that didn't work so that we can avoid those traps?
What I'll do is walk you through every step of our process. Some of this may not be traditional value investing but this is what we do. Standing at a real high level you have to figure out where is your playground, so first identify your universe. In our case it is typically mid to large, liquid, tradable stocks. Obviously, if you're doing this for a PA and you didn't care about taxes too much then you may want to move down the spectrum and do small or micro stocks, assuming you could get liquidity, but we are talking about big liquid securities.
Then the second step in our process - after we've identified the playground and before we even get into the whole game of what's cheap and what's not - is that we are trying to avoid the proverbial “falling knife.” You don’t want a situation where you buy a cheap stock at five times P/E but earnings get cut in half, it stays at five times P/E and we just lost half our capital, right? We want to avoid that type of value trap if at all possible so the second stage in our process is literally a bunch of forensic accounting screens where we are trying to identify things like aggressive accounting, manipulation, fraud or potential bankrupt firms.
For instance, you never want to buy a cheap firm that has a high probability of going bankrupt and you really need to worry about this because a stock is cheap for a reason. It is either going to go up 100% or you are going to lose all your capital and you are just buying risk.
How do you do this forensic screening to avoid risk? Do you run all the stocks past an Altman credit score or something like that?
There are a lot of things you do, but I'll speak specifically to the Altman Z-score because that's a traditional financial distress metric. The Altman Z-score was actually written in an article in the ‘60s or ‘70s and it is still programmed up in Bloomberg. People still use it even though there has been subsequent literature which basically said (1) in that original paper he had look-ahead bias so the result doesn't count, and (2) people have corrected for that and showed that it just doesn't work.
Yet everyone loves to use Altman Z-score even though there is no empirical evidence that it is actually a good idea. The literature has moved well beyond that and I would say the best technology in financial distress or bankruptcy prediction is a paper called “In Search of Distress Risk” by some guys at Harvard where they basically say, “Listen, Altman’s Z-score was on the right path in the sense that clearly we want to look at some balance sheet type measures. This is because if you have a ton of debt or your current ratios are going terribly wrong then that could be an issue, but we also need to look at market indicators.” For example, a really highly volatile stock price in the short run or huge excess returns that are way out of line with the market are a bad sign.
The bottom line is that when you use balance sheet financial statement information combined with market information and then build those into a statistical model then they are much better predicting out of sample bankruptcy so we just used that algorithm. I can send you the paper but if you just Google “In Search of Distress Risk” - I think Campbell and Harvey and another guy whose name I can’t remember are the authors – that is the algorithm we use. It's like the Altman Z-score, but just actually one that in expectation works better.
In other words, you find large-caps or liquid stock issues and then you run them past a forensic accounting system to get rid of the potential losers. And then what do you do?
Now we are into the heart of value investing and this is, “How do you find the cheap stocks?” As you mentioned at the outset, finding cheap stocks can mean a lot of things to a lot of people. For example, we could do a P/E ratio. We could compute a book-to-market ratio. We could use free cash flow-to-enterprise value.
There are a hundred different ways in theory you could screen for “cheap” and you could also do CAPE (cyclically adjusted P/E) measures where you take an average of the past eight years of earnings for the P/E ratio. You can do composites where you average different metrics together like one-third price-to-sales, one-third price-to-book and one-third P/E. As you can imagine, there are just thousands of combinations.
Well, we are computer programmer guys so we just literally data mined that. We tested out every combination that you can ever imagine. In my book I think there are two chapters dedicated to just that. What we found was actually surprising because I was born on Graham and Dodd's Security Analysis. I was like, “Oh, doing a P/E ratio where the E (earnings) is an average of, say, the last seven to ten years (that's what Ben Graham said to use) is probably one of the better ideas. It makes intuitive sense.” There is even a paper in the academic literature where they do that analysis in UK markets and they show that that's actually true.
We did all of this in the U.S. and we basically found that in the end the most robust, reliable and simplest way as well to identify “cheap stocks” that tend to have the best out-of-sample performance is enterprise multiple, so some form of either EBIT over total price value or EBIT over enterprise value works best. At least in-sample over the past forty to fifty years that simple metric seems to be the best.
Didn't James O'Shaughnessy find the same thing? I know you know his work. Did you find it confirming yours?
O'Shaughnessy and those guys are big fans of the composite a little bit more as opposed to single ratios. They do identify that enterprise value ratio or enterprise multiples are certainly good.
I think that is their best single number one rating metric rather than P/E.
Yes, that's right. And they do a combination, right, like composites?
Which is fine. In research that we have not really ever published we wanted to understand and try to determine a great composite metric. I really like the composite idea intuitively because for example, take Google. For Google, why would you measure the company on price-to-book when the book value is the computers? Clearly, that is not the real book. The real book value is their human capital and everything so maybe you should use price-to-book, price-to-earnings, or price-to-sales and the composite is therefore going to tell you something better.
So what we did is say we are going to take a different track. We think business or investing is best when it's most business-like so who are the guys that buy businesses? Private equity guys. How do all my private equity buddies view business buying? They look at enterprise multiples because they have to buy the whole damn company. They have to buy the equity, the debt, and if there's any cash in the till they can throw it back in their pocket assuming they don't need it for working capital needs.
So what we did is we looked over time at the extreme cheapest securities based on price-to-earnings or cheap EBIT-to-total enterprise value or whatever, and then we compared amongst those names the propensity for buyout and take-over. Basically there is a much higher propensity for takeout amongst the cheapest enterprise value names as opposed to the cheapest, say, price-to-sales names or price-to-earnings names. This gave us a little more confidence that really what is driving the enterprise multiples edge is the fact that it is not just data mining, but rather it's probably due to the fact that a lot of times the names you buy in those buckets are names that private equity guys are also interested in buying because that is how they think about the world.
That's what I found out, too. You have to look at it as if you were sorting through companies and willing to buy the entire outfit. Whatever metric helps you do that works best. I also came to this conclusion exactly as you did figuring that this was what the private equity guys were looking at so therefore it made sense.
For the rest of the interview, pick up a copy of Breakthrough Strategies of Wall Street Traders (an average interview includes 20+ pages of techniques) on amazon.com.
An Interview with Wesley Gray on Value Investing
An Interview with Mike Messier on Catalyst Trading the Biotech Stocks
An Interview with Steve Hawkins on How to Trade Using Pete Steidlmayer's Market Profile
An Interview with Tim Sykes on Profitable Penny Stock Trading
An Interview with Bill Spetrino on Dividend Investing to Make Yourself Rich
An Interview with Charles Mizrahi on Benjamin Graham Value Investing
An Interview with Raymond Barros on Trading Stocks