While Montier makes a good (and unoriginal) point, I'm not convinced by his argument. The first example: IBM and JnJ are in wildly different industries, compet

While Montier makes a good (and unoriginal) point, I'm not convinced by his argument. The first example: IBM and JnJ are in wildly different industries, competing in different markets against different competitors employing vastly different strategies. To use an arbitrary time period (who is to say IBM’s returns 10 years from now won’t trounce JnJ’s), and employ a metric (shareholder returns) that is often out of whack with reality (as Montier himself would acknowledge) isn’t persuasive enough for me. The second macro example of comparing returns between two totally different time periods also isn't exactly apples to apples either. And also completely ignores the effects of hundreds of other potentially significant factors (interest rates, gdp levels, global trade etc) This article seems to be a strong case of confirmation bias on his part. Ironical given that he's written a popular book on behavioral investing. (And yes this comment can also be viewed as suffering from a similar bias) Yes SVM has pitfalls, yes focusing on the customer should be a high priority, however the agency problems Montier describes (management extracting undue value) are solved by neither. Borrowing from churchill(?) — SVM might be the dumbest idea in how we organize our public markets, except for all the others.