Been thinking about the "comprehensive income"
I suggest a different definition of "true" earnings
that I think reflects the Berkshire reality a lot
1) Take regular earnings of operating
2) Add underwriting gain
3) Add income on fixed
4) Add realized and unrealized
appreciation of non-fixed-income securities IN EXCESS of S&P
I think that this measure would more or less
properly reflect the value that Berkshire
What do you think?
I agree with everything said in this
But, one caveat which I believe a lot of people
overlook, is that BRK is an insurance company, unique in
its supercat exposure, and WILL eventually have to
pay up. Put that in your Intrinsic Value
calculations! THEN tell me how over/undervalued BRK is, because
I must admit I have no clue how to figure that one
Anyway, how would you like to be that Ajit (sp?) dude??
Trying to figure out the actuarials involved with
insuring multi-billion dollar catastrophes would give me a
Good Luck to all
You believe that BRK's underlying businesses,
like See's or 8% of Coke, would be cheaper if you
bought those companies directly than if you bought BRK
stock. However, that assumption is generally not true.
By buying BRK (when it itself is undervalued), you
are able to buy KO, G, etc. for below their
individual market prices. It's also an ongoing mystery in
finance circles why mutual funds often sell below the
collective value of their stock holdings, but it happens. As
recently as 12/31/97, BRK stock was undervalued. It's
"investments per share" was $38,000 and the intrinsic value of
its wholly-owned subsidiaries was about $20,000, but
BRK's market price at the time was only $46,000. You
could have bought it then and gotten all the stocks in
Buffett's portfolio on the cheap.
In addition, if
you buy BRK stock, you benefit from Buffett's ability
to obtain businesses (whether in whole or in part)
AT A DISCOUNT. Maybe you have the same ability as
Buffett to know when a business is selling at a discount
or not, but most people do not. For example, Buffett
is buying GRN now because he believes he can
purchase it at a discount to underlying value (he's paying
only $20 bil. for $24 bil. in floate). However, the
average investor does not have the ability to determine
whether a potential acquisition is selling at a discount
or not. Buffett has the patience to wait until a
stock he likes get extremely undervalued.
true Buffett's reputation adds something to BRK's
stock price. But, as Buffett and Graham have said,
markets are efficient in the long run (but not
necessarily in the short run), so even BRK stock will be
correctly valued by the market over time.
with your assertion that buying BRK reduces
diversification. Quite to the contrary, I don't think you could
get more diversified than owning World Book, Dexter
Shoes, Dairy Queen, etc. (all different types of
businesses) at the same time, which owning BRK stock amounts
BRK seems as I am now view it is a mixed mutual
fund. I think I would rather own the underlying
companies that are not wholly owned by BRK investing the
difference in those stocks directly. The consequences of
investing in BRK is that it reduces diversification and I
am paying a fee for WEB's stock picking. I am also
stuck with dogs among the gems. What I am interested in
doing now is determining what premium I am paying to
WEB for partial holdings as a percentage of BRK's
pricing versus investing directly in the underlying
securities while adding a few of my own. Has anyone here
done that calculation?
This is not to say that
BRK has not been good. It has been great.
I think Buffett calculates BRK's intrinsic value
in two ways, one a short-cut way and the other the
more comprehensive way.
The short-cut way is to
define the intrinsic value of BRK's investments in KO,
G, etc. as merely their market price and to then
come up with an "investments per share" figure for
BRK. In the last 3 Letters to Shareholders, Buffett
has been providing a table showing BRK's "investments
per share" and Buffett asserts that this figure is
the first component of BRK's intrinsic value (the
other component being the intrinsic value of BRK's
wholly-owned subsidiaries, determined by discounting their
aggregate earnings and earnings growth rate).
more comprehensive way that Buffett determines BRK's
intrinsic value is to look at not only the earnings of
BRK's wholly-owned subsidiaries (and discounting them)
but also to discount the "look-through earnings" of
BRK's holdings in KO, G, etc. Since Berkshire actually
owns 8% of Coke's outstanding stock, Buffett believes
Berkshire (and its shareholders) have a claim to 8% of
Coke's earnings -- just as if BRK owned 100% of Coke,
BRK (and its shareholders) would have a claim to 100%
of Coke's earnings. The SEC merely has a rule that
says if you own less than 20% of a company's stock,
you can only claim the dividends on that 20% as your
own income and not 20% of all the earnings outright.
But in the case of Berkshire, even if it owns less
than 20% of Coke, 8% of Coke (worth about $16 bil.) is
still a huge amount of Coke stock, so Buffett has a
case when he says Berkshire can properly claim 8% of
Coke's earnings as its own earnings. Just imagine if BRK
merged with Coke someday (i.e., owned 100% of Coke). BRK
(and its shareholders) would have a rightful claim to
100% of Coke's earnings. The same is true if BRK only
owns 8% of Coke -- BRK's shareholders have a rightful
claim to (and actually own) 8% of Coke's earnings. It's
a transitive relationship: if BRK shareholders own
BRK (true) and BRK owns 8% of Coke (also true), then
it must be the case that BRK shareholders own 8% of
you don't have to discount the Ko, G etc.. all
you have to do is discount BRk's earnings out 5-10
years add them up and add the book value or investments
per share (this will give u a higher value) and that
is the intrinsic value
You are correct, in that some advertising costs
are if fact capitalized and amortized over a deemed
useful life. However, I'm not sure whether this would
ever be material enough to factor into the owner's
earnings computation. It would depend on the company and
its industry I guess.
requires a company to disclose its policy with respect to
advertising costs (ie..expensed as incurred or capitalized
and amortized). This was promulgated because of the
disparity of methods used by different companies (even in
the same industry). This is most often presented in
the significant accounting policies footnote.
Thanks for a very clear explanation.
add a very minor point, I think that sometimes
certain advertising expenses may be capitalized, for
example Account Acquisition Costs and so on. (Is that
But generally, you are right.
that's what buffett talked about: if you add back
depreciation, then you have to subtract cash outlays that you
capitalize (eg, Capital Expenditures).
I think your definition of capital
expenditures is too comprehensive. Capital expenditures by
definition are purchases of plant and equipment.
Advertising, maintenance and R&D are clearly ongoing,
recurring operating expenses already reflected in net
income. Your definition and formula would in essence
deduct these items twice.
Mr. Buffett's premise
is that if you are going to add back depreciation (a
noncash expense) to net income then you must also factor
in the average annual expeditures for plant and
equipment to maintain a company's long-term competitive
position. Many analysts add back depreciation but do not
allow for the fact that ongoing cash outlays for new
equipment will also be required.
The cash flow
statement reports annual capital expenditures in the
"investing activities" section. Presumably, an investor
would be interested in the average of this line item
over a period of years.