In one sentence: Phil adds Shiller's cyclically adjusted P/E to the Wilshire/GDP ratio as evidence that the market is priced far above history, warns about "bargains" in such a market, and uses buybacks and debt-funded deals as tests of whether management allocates capital well.
Key ideas
- Recap. Wilshire/GDP is around 170%, over double the 80% level Phil calls a good deal, so finding values means stretching the numbers. Danielle's point: wishing a company into your price range is the risk. [00:00–02:00]
- Be extra careful with "on sale" stocks. In an expensive market, ask who is selling to you: Phil says about 85% of volume is professionals, who have reasons. Danielle counters that event-driven drops happen in any market and that the long horizon is the edge; Phil agrees but says check for a terminal problem. [05:00–10:00]
- A falling price is not a bargain. Phil's example is Boeing (after the 737 MAX trouble and a CEO change), down from about $450 to about $330 but not yet near the $250–280 he would want. Price alone means "what someone paid", like an 80% off sticker. [10:30–13:00]
- Shiller P/E (CAPE). Price divided by ten years of average inflation-adjusted earnings; Shiller won the Nobel for related work. The long-run average is about 15 to 16, below 10 has been a great time to buy, above 24 or 25 is dangerous. Phil's data: 5 in 1921, about 30 in 1929, over 40 in 1999, about 28 in 2008, and 31 now. [14:00–19:30]
- Superstars quit at the 1999 peak. Phil recalls investors such as Julian Robertson sitting in cash and closing their fund, and Buffett holding cash and buying back Berkshire stock. [19:30–21:30]
- Plain P/E too. The S&P 500's ordinary P/E of 24.2 is higher than all but a few years (2000, 2008 and one early year), so the "CAPE is distorted" objection doesn't rescue it. [21:00–22:30]
- Inductive risk. Phil's turkey/chicken story: a pattern holding so far doesn't prove it will continue. He says history only supports waiting, and he still can't say when. [23:00–24:00]
- Buybacks. Companies buying their own shares prop up the price. Instead of capex (the real investment) firms repurchased shares after the tax cuts. Buying back stock above its value spends "my dollar to buy fifty cents of value"; Phil cites IBM's years of repurchases and says to check the figures yourself. His opinion. [24:00–31:00]
- Other allocation tests. Debt-funded acquisitions (Dollar Tree's purchase of Family Dollar is Phil's example), and paying with overpriced stock. Companies borrow at roughly three-year terms and face refinancing risk, unlike a 30-year mortgage. [30:00–34:30]
- Allocation as a screen idea. Phil says he wants capital allocation judged in the toolset: low debt, high and rising ROE/ROIC, and sensible buybacks. He refers to the Jacob Taylor interview and book. [26:30–35:00]
- Use the numbers as comfort. Danielle finds that knowing the market is expensive explains why she can't find much. Her response is patience and adding names to the wish list. [35:00–36:30]
How it maps to RuleOne
- The Big Five numbers and ROIC on the stock pages (/stock/TICKER/) cover the debt and returns part of management quality; buyback timing is a manual check against the fair-value estimate.
- Share count over time on a stock page shows whether buybacks actually shrank the count.
- The watchlist idea maps to your saved list on /stocks/.
Buffett, Munger and Graham links
- Buffett's rule that buybacks make sense only below intrinsic value is in his 1984 and 2011 letters (the 2011 letter states it explicitly).
- Graham and Dodd's use of average earnings over several years (Security Analysis) is the root of Shiller's ten-year smoothing, though Shiller is not a value investor by trade.
- Munger on management who buy at silly prices appears throughout Berkshire meeting transcripts; I don't quote him.
Words to know
- CAPE / Shiller P/E: price over ten-year average inflation-adjusted earnings.
- Stock buyback: a company buying its own shares.
- Capital allocation: how management spends cash: capex, acquisitions, dividends, buybacks or debt paydown.
Try this
On a stock page like /stock/IBM/ look at five years of share count and free cash flow. Did the buybacks reduce shares, and was the price paid below your fair value?
Check yourself
- Why is a price drop of 40% not enough to call a stock cheap?
Answer
Price only says what someone paid before; value depends on the business's cash flows. - When are buybacks good allocation?
Answer
When the stock trades well below intrinsic value; paying above value wastes shareholder money. - What risk does corporate debt add to an acquisition?
Answer
Short maturities mean refinancing risk if the company can't repay.
Short quotes
"Price doesn't mean value. It means what somebody paid." (Phil, ~12:00, auto-transcribed)