This was not a great book, because Miles is an unabashed Berkshire fanboy. The chapters quickly take on a repetitive feel — the first parts are always about the history of the business and the manager, but quickly give way to a lot of backslapping and gushing about Buffett and the awesomeness of being a part of Berkshire. That isn’t to say that Buffett and Berkshire aren’t great, but all the backslapping gets old after maybe five chapters.
I’m giving the book three stars because it’s still useful. I think three things stand out:
First, there aren’t that many books that profile Berkshire’s managers. You’re kinda stuck with this one, since so many of the CEOs are such low profile individuals.
Second, Buffett has a more fluid view of moats than I expected. If the business has good financials, is a leader in its market (by some definition of market) and is run by a top tier team, Buffett is perfectly happy to buy the company, even if its moats are mediocre (but still existent!). This was a surprise! Buffett essentially assumes that a top tier operational team that has been in charge of the business for decades is likely to maintain a small but meaningful competitive advantage in its (admittedly mature) market, even without structural advantages. And when he misjudges the management, he is basically stuck with the company, since his reputation is that he rarely meddles with management.
(To his credit, Buffett has only made this mistake ... twice? Not very often, for sure.)
Third, the book is useful as a source of clues as to how Buffett runs Berkshire in a decentralised manner. The real trick is to a) align incentives so that individual managers are incentivised to pass on excess cash to HQ; their bonuses are tied to ROIC, not pure growth. (Which also means they are incentivised to keep capital within the business, if they think they can generate more returns that way). Money may be borrowed from the HQ, but with an internal interest rate. Miles doesn’t mention how this interest rate is set (presumably equivalent to the risk-free rate? I don’t know. Update 2023: I realise they’ve talked about this in their annual shareholder meetings multiple times over the years: it’s 15-20% depending on the situation, and always custom to the specific ask).
Berkshire’s incentive system is helped by the fact that b) Buffett ensures that most owner-management teams keep 20% of the business; Berkshire owns 80% when dealing with family-owned businesses. It is also interesting that c) Buffett prefers to use cash to pay for companies, instead of Berkshire stock, even though stock purchases defer taxes for the seller. But d) when there is stock ownership as part of incentives, Berkshire gets managers to buy stock at market prices and never issues an option plan, for (and Munger and Buffett’s views on this are well known) option pools tend to be a hidden cost to shareholders. Finally, e) there are zero synergies between Berkshire subsidiaries — because (or so Buffett+Munger say) they’re too difficult to do. Miles notes that various CEOs he’s interviewed have reached out to the others within the conglomerate to talk and learn, but none have done anything substantive together.