In one sentence: Phil has been happily buying clothes from Stitch Fix, so the two use it as a live example of how to research a company you like (use it, find the competitors, test the moat, check the track record, then rough out a price) and conclude, in the end, that a business this young is too hard to value.
Key ideas
- Why pros can't do this (the framing). Real value investing needs total control of your capital so you can wait. Buffett's line is that you should be graded on five years (20 quarters), not quarterly, and that is why he runs Berkshire (a company) rather than a fund that investors can pull money from. Individuals can wait. [00:05–04:00]
- Start from what you use. A product you use and like is a fine starting point for the Radar. Phil has kept every box from Stitch Fix, a personal styling service that sends clothes picked from a style profile. Buffett's Coke and Dairy Queen are the same instinct. [11:00–16:00, 25:00–26:00]
- Assume you are not the first. By the time a company goes public a niche is already known and others have jumped in (Trunk Club, Birchbox, meal kits). Look up who else does it; maybe one is better. [16:00–18:00, 27:00–29:30]
- A quick look at the moat types. Brand, switching, toll bridge, secrets, price, and "half a moat": network effects, a subset of switching that Phil doubts is a true moat. [19:00–20:30]
- Stitch Fix's possible moat is switching, but is it real? Phil's idea: the more you order, the better they know you, so leaving costs something. Danielle's counter: one bad box and the data feels worthless; signing up with a rival takes ten minutes. Brand moats are also hard: the brand must become the category (Band-Aid, Kleenex, Xerox). [21:00–25:00]
- Prefer companies that change slowly. Chewing gum, Coke, McDonald's and Chipotle don't have to be clever every season. A computer company must keep reinventing itself. The catch is that slow businesses seldom go on sale, so you wait for a recession. [21:30–23:00, 26:00–27:00]
- You need about ten years of track record. Ten years lets you see a recession, which is when weak balance sheets and weak competition show ("who's swimming naked when the tide goes out"). Phil admits he bought Chipotle after about seven and Google after about four or five years as exceptions. [30:00–32:00]
- Check the filings, not just the website. Search "investor relations" for a retailer's real site, then the latest 10-Q. A six-month cash flow statement is roughly doubled for a year, with a little extra for growth. [41:00–43:00]
- Rough owner earnings on the fly. Phil's pre-tax profit of about $36M plus $5M depreciation plus $11M payables change, less $8M of capital spending, comes to about $44M for six months, so about $90M a year. Times 10 gives about $900M, versus a market cap near $2.3B (about 97 million shares at about $24). It is a rough, same-day estimate. [44:00–49:00]
- A second lens. His margin-of-safety calculation, assuming an optimistic 20% growth for ten years, gave about $12 a share and a buy price of about $6, against $24. The 10 cap pointed to about $8–9 a share. Triangulate between methods. [52:00–54:00]
- Verdict: too hard. The real obstacle is not the price but whether the company will be more productive in ten years. Netflix and Amazon in the 1990s show how hard that is to see. Don't try hard questions. [49:30–52:00]
How it maps to RuleOne
- The /stock/TICKER/ page gives the long-run numbers Phil wants (ten years of history). A company with only one or two years of data simply has little to show, which is the point he makes about the toolbox.
- The screen is built for the slow-changing, long-record business Phil prefers, so young IPOs rarely pass it.
- The stock page's links to SEC EDGAR are where you pull the latest 10-Q for the newest cash flow numbers.
- The 10 cap and margin-of-safety figures are two of the price methods on the stock page. This episode shows why you compare more than one.
Buffett, Munger and Graham links
- Buffett on being judged over years not quarters: see the partnership letters and the Berkshire letters on the "institutional imperative" (1989 letter).
- "Who's swimming naked when the tide goes out": Buffett, 2001 Berkshire letter.
- Buffett saying he could not have predicted which early Amazon-like company would win is the circle-of-competence idea in 001.
Words to know
- Switching moat: customers stay because leaving is costly or painful.
- Market cap: share price times shares outstanding, the price of the whole company.
- 10-Q: the quarterly report filed with the SEC.
- Owner earnings: cash a business makes for its owners after the spending needed to keep it going.
Try this
Pick a product you use every week. On /stocks/ or /stock/TICKER/ check how many years of history exist. Then write three rivals and one sentence on why a customer would or wouldn't switch. If you can't write the sentence, you're done.
Check yourself
- Why does Phil want about ten years of history?
Answer
To see how the business behaves through a recession, when competition, debt and cash flow are exposed. - What is Phil's rough owner earnings method here?
Answer
Profit before tax, plus depreciation, plus working-capital changes, minus capital spending, annualised, times 10 for a price. - Why does Phil put Stitch Fix aside even though the price looks high only by a few measures?
Answer
He can't say with confidence it will be more productive in ten years, and a young company has no track record.
Short quotes
"You don't get to see who's swimming naked until the tide goes out." (Phil, quoting Buffett, ~31:00, auto-transcribed)