In the above paper by Frazzini, Kabiller, Pederson they reference another paper called quality minus junk by Assnes Frazzini and pederson.
It is a quality formula and consists of:
Profitabilty:
Gross profits over assests
Return on Equity
return on assets
Cashflow over assets.
Gross Margin
Fraction of earnings composed of accruals.
Growth:
The five year growth in profitability.
Safety:
Low beta
low idiosyncratic volatility
low z score
low leverage
low z score
roe volatility
Payout:
Equity issuance
Debt issuance
total net payout over profit.
Most of the safety and growth ones seem to be very hard to put into p123.
As to inputting factors into p123, it’s important that we refrain from allowing ourselves to get paralyzed over expectations for getting it “right.”
For one thing, a very important aspect of Buffet’s alpha was something that had nothing to do with stock factors; it was his ability to leverage up 1.6 to 1 with pretty-much none of the downside that normally accompanies leverage: margin costs (his margin “loans” amounted to nearly zero-cost funds supplied by his insurance float) and absolute freedom from concern about margin calls (again, his own insurance units were the “lender” so his so-called margin was not subject to the usual collateral requirements.
As to stock factors, bear in mind Buffett did not use quantitative platforms. The only z-scores he thought about probably involved aesthetic comparison of zebras at the Omaha zoo. And his attention to Beta was pretty much focused on making fun of MBA-types who spend a lot of time calculating and using it. Everything you read about a Buffett “factor” (including what you see in the Buffett models I’ve posted on p123) is a creation of the author, not Buffett. Such factors and combination of Buffeett factors are inspired by Buffett but Buffett himself has nothing to do with their articulation and use. So if you encounter a Buffett factor in this paper or anyplace else that you are unable to precisely implement on p123, then focus on the spirit of the law (or, more importantly, spirit of the combination of laws) and come up with your own detailed factors. The Buffett models I posted on p123 are not offered as Buffett; they’re offered as one person’s way (me) of applying Buffet’s ideas on p123. I encourage you to use it as an example and come up with your own interpretation.
Finally, the ideas “betting against beta” and “quality versus junk” are academician’s ideas based on their interpretation of what they see when they look at what Buffett has done. I very much like that research theme; i think those folks have done terrific work. But here again, focus on the themes. If you can’t precisely replicate their details, then come up with your own details.
Working with Buffett ideas, whether drawn from the Buffett’s Alpha paper, an one or more of the Buffett books on the market, or the source (BRK’s annual Chairman’s Letters) is a wonderful way to hone your p123 skills; to learn to adapt great fundamental idea in terms of p123 language. And of course, it’s a just-plain way to come up with some good investable models. And if you want to talk about this as you go along, I hope you’ll post on the forums.
Buffet was/is a quant. He admitted it in one of his partnership letters, I don’t remember which. Him and Ben/Walter, etc used to go through the S&P stock guide looking for “net-net” stocks.
He should be in politics, very good at saying something but doing the other. ie “Derivatives are financial weapons of mass distruction”, while he’s been using them all along. Same with leverage, Don’t believe a word he says.
Even after he shut the partnerships he would have to have some sort of quantitative system going.(Charlie’s mental models maybe).
People wouldn’t think so highly of him if he just came out and said something like “here are my 20 simple quantitative rules for picking a stock”, etc.
I think you’ll find all successful investors have some sort of quant model going on in the background. They just couldn’t keep the performance up year after year without it. It may be really basic but I still think it’s there.
He might be a quant, but I think he applies his quant rules discreationary and also uses a lot of Qualitative criteria (scalable Business model, good Management…)