Share buybacks: Buffett's price rule, not a schedule

September 20, 2026 · 5 min · Free, with full transcript

Greg Abel's appointment as Berkshire Hathaway chairman brought a wave of commentary on Warren Buffett's buyback record. This episode argues the common reading is backwards. Buffett was not a great buyer of his own stock, he was a disciplined refuser, and the discipline lived in a price rule: no repurchases before 2011, then a 1.2 times book value ceiling, then a valuation test that produced about $27 billion of buybacks in 2021 and zero for three straight quarters in the second half of 2024. You leave able to separate the mechanics of a repurchase from its value, read earnings per share against buyback yield and treasury bill returns, weigh the empire building counter argument, and set a written price above which you will not buy a single share.

Listen to the podcast

4 min

Chapters

Key takeaways

  • Set a written price ceiling for your own buybacks before the pressure to act arrives.
  • Judge repurchases on what you paid versus the business's worth, not on quarterly consistency.
  • Read earnings per share alongside buyback yield and the return on cash held in treasury bills.
  • Treat buying nothing for several quarters as a legitimate allocation decision, as Berkshire did in late 2024.
  • Weigh the honest counter argument that a predictable buyback guards against vanity acquisitions.
Read the full transcript

Host:Leaders' insights. Five minutes on Buffett's exit proves the conventional CFO wisdom on buybacks needs rethinking. Berkshire Hathaway named Greg Abel chairman three weeks ago, and every finance blog on earth ran the same headline about Buffett's buyback genius.

Expert:You think they read it backwards? Almost entirely backwards. The story everyone tells is that Buffett was a great buyer of his own stock. He barely bought any for most of his

Host:career. Define the thing first. When a company does a buyback, what actually happens to my money?

Expert:A buyback, the company using cash to purchase its own shares in the open market and cancel them, leaves fewer slices of the pie. Same profit, fewer slices. So each remaining share owns a bigger piece. That's the mechanic. The question nobody asks is whether the slices were cheap when you bought them. And that's where the conventional CFO wisdom goes wrong. The chief financial officer's default religion is return cash to shareholders. Buyback every quarter, brag about it on the earnings call, put it in the deck. It becomes a habit, like a gym membership you pay for and

Host:never use. Let me pull up the actual artifact then. Berkshire's own repurchase policy. Walk me

Expert:through it. Berkshire didn't even allow buybacks until 2011. The first rule was buy only below 1.2 times book value. Book value being roughly what the company's assets are worth minus its debts.

Host:That number was a hard ceiling. So the discipline was a price rule, not a schedule.

Expert:Exactly right. And that's the piece the imitators dropped. In 2018, Buffett scrapped the 1.2 ceiling but kept the spirit. Buy only when the stock trades below what he thought the business was actually worth. In 2021, Berkshire spent about $27 billion on buybacks. Then he slowed to a crawl. And by the second half of 2024, he bought zero, zero for three straight

Host:quarters. Zero while sitting on how much cash? Over $300 billion in cash and treasury bills by

Expert:early 2025. The takeaway most CFO steal is buy your stock. The actual lesson is refuse to buy when it's expensive, even if your pockets are bursting. That's the part that it's uncomfortable,

Host:though. A CFO who does nothing for three quarters looks asleep at the wheel. And that's the whole

Expert:trap. Doing nothing is a decision. And it's often the expensive looking one that's actually cheap. Compare it to what a lot of firms did. IBM bought back roughly $125 billion of stock over about a decade while its business shrank. Buying an ever larger share of a smaller cake.

Host:So the metric everyone quotes earnings per share hides the crime. Earnings per share,

Expert:the profit divided across all those slices, goes up mechanically when you shrink the slice count. A CFO can juice that number while destroying value because they paid $40 for something worth $25. The dashboard says up and to the right. The bank account says otherwise. Tear the metric apart

Host:for me. What works about earnings per share? What doesn't? What works? It tells an owner how much

Expert:profit backs each share, which matters. What fails? It's silent on price. It never asks, did we overpay? You need a second line next to it. Buyback yield versus the return you'd have earned leaving the cash alone in treasury bills, which right now pay you something real for doing

Host:nothing. Give me the honest counter argument. Isn't a steady buyback better than a management team that hoards cash and buys a bad company instead? Fair. And that's genuinely the strongest

Expert:defense. A predictable buyback is a guardrail against empire building. The CFO who'd rather spend $8 billion on a vanity acquisition? Buffett's answer was to hold cash and wait, which only works if you have the temperament to be mocked for a year. Most executives don't.

Host:So the real Buffett's skill wasn't buying. It was not buying.

Expert:It was patience with a spreadsheet attached. He'd tell you the buyback is a tool, not a virtue.

Host:Give me the one thing a CFO listening should do Monday morning. Write a hard price rule for

Expert:your own buybacks. Wait. A specific value above which you will not buy a single share. And put it in writing before the pressure hits. If you can't name that number, you're not allocating capital. You're just spending it. What we read for this one,

Host:Financial Times Accounting Today, CFO Dive, CNBC Finance. We'll stop there. The whole CFO track in order is waiting at MBA-training.com.

Read and go deeper