No. of Recommendations: 12
We’ve discussed this ad nauseum but the comparisons with Berkshire over decades using the P/BV yardstick at a valuation metric are apples to oranges.
The theory behind this is certainly sound. But the fact remains that SO FAR, a simple multiple of book still gives almost precisely the same "fair value" levels as very much more sophisticated measures of value. There is, as you note, no theoretical reason that this must be so, and good reasons to believe that the two should diverge over time--they just haven't yet. So far, "about 1.5 times book" has almost exactly the same meaning in terms of valuation level as it has for many many years, at least well within the margin of error that it always had. In fact, depending on the time frame you're looking at, the fair P/B is a hair *lower* than it was some years back.
The amount of value arising from the operating subsidiaries versus investments per share is certainly a big factor, but I haven't found that the fraction has really budged (net) in 20 years.
Jim