BRK IV per Whitney Tilson

Stocks A to Z
1

“Note that I subtract all of the underwriting and investment profits from Berkshire’s massive insurance operations but add back a rough estimate of the average insurance underwriting profits over the past decade ($1.4 billion annually).

This results in $17,877 in adjusted pretax earnings per share in 2021, to which I apply a multiple of 11 times to arrive at a value for the operating businesses of $197,000 per share.

Now add the $330,000 in cash and investments per share to arrive at a total intrinsic value of $527,000 per A-share (or $351 per B-share).

With the A-shares closing yesterday at $476,000, that means the stock is trading 10% below my estimate of its intrinsic value.

My friend and former partner, Glenn Tongue, has come up with a second methodology:

“There is another viable way to look at Berkshire’s valuation by looking at the economic earnings of the stock portfolio rather than the accounting treatment.

In his annual letter, Buffett notes that while Berkshire owns 5.6% of Apple (AAPL), the only thing that appears in Berkshire’s income statement related to this stake is the $786 million in dividends Berkshire received from Apple last year. But in reality, Berkshire owns 5.6% of Apple’s total net income, equal to $5.6 billion of what Buffett calls “look-through earnings.”

Applying the same calculation to all of Berkshire’s stock holdings (including Apple) results in look-through earnings of at least $15 billion.

Adding this to Berkshire’s after-tax operating earnings of $27.5 billion yields total economic earnings of $42.5 billion.

If we compare this to Berkshire’s current market cap of $702 billion, less roughly $100 billion of excess cash, we can see that Berkshire is currently trading at a price-to-earnings (P/E) ratio of 14.2 times trailing earnings.

Given that the S&P 500 is currently trading at 23.8 times trailing earnings, Berkshire’s multiple is downright cheap, especially given that it’s a far-above-average collection of businesses.”

Thank you, Glenn!

In summary, Berkshire Hathaway is incredibly safe and growing at a healthy rate. And its stock, any way you look at it, is undervalued – enough that the world’s greatest investor is buying it back in size. What’s not to like?”

17 Likes
2

first time I have seen an IV estimate from him that is within 10%.

of course, that is narrowing today.

2 Likes
3

Re: Glenn Tongue’s methodology quoted in the original post -

Adding this to Berkshire’s after-tax operating earnings of $27.5 billion yields total economic earnings of $42.5 billion.

Doesn’t the $27.5 billion of operating earnings include dividend income we are now counting twice if we go and add “look through earnings” ?? We would be counting the apple dividend received and our share of apple’s net income in that version.

3 Likes
4

Doesn’t the $27.5 billion of operating earnings include dividend income we are now counting twice if we go and add “look through earnings” ??

It probably depends who has done the calculations.

Personally, I count all investment related things completely separately from all operating related things.
For example, I have recent after-tax operating earnings (which excludes dividends) at about $23bn
and total after-tax investee earnings (which include the dividends before they’re paid out) at about $18.5bn.
That’s $41.5bn, not much lower than their $42.5bn, especially since

  • I think they’re talking about the year ahead and I used trailing year operating earnings, and
  • I didn’t include any interest income.
    Forward year operating earnings would be expected to be somewhat higher than my $23bn figure. Maybe pencil in a rise of inflation + 7%?

So, offhand, the $42.5bn OP figure seems very plausible, no matter how he arrived at it.
At today’s price that equates to a P/E of 16.7, if you do NOT do his adjustment for the cash pile.

On a per-share basis the rate of increase is different due to share buybacks.

Jim

9 Likes
5

52 week high at the open?

6

The $500k magnet is pulling the largest lime-green bubble to the top !

https://www.chartfleau.com/bubblechart

ciao

1 Like
7

Hate to say this… but calculating “look through earnings” while valuing Berkshire seems silly.

For all practical purposes AXP, etc., own and control those earnings. Theoretically, as a part owner, BRK owns a portion of those earnings. But since BRK can’t do anything to get those earnings in hand and control how they are used, it seems like a waste of time, IMHO.

Please show me where I am wrong.

An owner in a small business can actually do an “owner’s draw” but an owner of publically owned shares cannot.

Sincerely,

jan

:^|

8

For all practical purposes AXP, etc., own and control those earnings. Theoretically, as a part
owner, BRK owns a portion of those earnings. But since BRK can’t do anything to get those earnings
in hand and control how they are used, it seems like a waste of time, IMHO.
Please show me where I am wrong.

The idea of look-through earnings is valid. The only real distinction is timing.

If a firm you own makes a buck in profits and pays it out as a dividend, you get the buck.
If they keep it as retained earnings, you don’t…at that time.
But statistically, and on average, that firm will invest that money at a reasonable rate of return* and the value of the share will go up, then the price of the share will go up.

Even if the earnings just stack up as cash, a company with an extra $10 of cash per share on its books will trade higher than the same company without that cash.
Probably around $10 per share higher, as a guess.
The more earnings they make, the higher the stock price will rise over time.

A shareholder tends to make more money over time with each retained dollar than with each dividend dollar, on average.
But the main point is that the shareholder DOES get the benefit of all the earnings one way or another…just with uncertain timing.

So, if you’re doing any sort of valuation methodology that does not include stock price appreciation, you should include the value of the look through earnings.
Or count the average stock price appreciation number and ignore the look through earnings.
Not both of course, but not neither either.

Jim

  • A return of about 12% is pretty typical for an incrementally retained dollar by a big US firm.
    Median trailing full year ROE among the Russell 1000 is 12.6% at the moment…a pretty typical year in that sense.
19 Likes
9

Chris Bloomstran recently had look through earnings at $48 Billion compared to the $42.5 Billion from Glenn Tongue notes above.

Does anyone know if there was an obvious reason for the gap?

I have another question. I like the simplicity of a multiple of owner earnings. I struggle to conceptualise how to value the insurance business. Is the float based valuation essentially saying we have float that gets replaced and is therefore worth the value of float because it’s like our money, in the sense it doesn’t get repaid. But then I think why would that be valuable if it doesn’t make a return sitting in cash at almost zero rates. That said, maybe I can understand that while the cash related to float, is not ‘currently’ earning a return, it’s what happens between now and judgement day that matters. So I see ‘potential’ value in that.

It just seems more intuitive to me to value the insurance business on a multiple of underwriting gains average over say 5 years plus investment returns on cash (however small) and allow the multiple to reflect two important future potential positives. Grow in float and higher returns from float.

I’m confused. Please help.

10

Excuse me, I’d like to recommend Mungo’s comment twice, please.

5 Likes
11

<<Chris Bloomstran recently had look through earnings at $48 Billion compared to the $42.5 Billion from Glenn Tongue notes above.>>

Either number seems too high. BRK’s stock portfolio is only about $350 billions. $42.5 billions of earning would imply a PE Of 8.3

12

Perhaps the numbers refer to look through earnings + operate earnings.

13

Yes both numbers are system wide earnings. The gap might be underwriting profits which are more open to interpretation. With averages, current year or exclude completely all options. It just seems like a big gap. I prefer Bloomstran’s number.

14

Chris Bloomstran recently had look through earnings at $48 Billion compared to the $42.5 Billion from Glenn Tongue notes above.

Bloomstran makes up some of his earnings, i.e. he assumes some of the cash will be put into businesses generating earnings.

Counting the chickens and all that.

4 Likes
15

I have another question. I like the simplicity of a multiple of owner earnings. I struggle to
conceptualise how to value the insurance business. Is the float based valuation essentially saying
we have float that gets replaced and is therefore worth the value of float because it’s like our
money, in the sense it doesn’t get repaid. But then I think why would that be valuable if it doesn’t
make a return sitting in cash at almost zero rates. That said, maybe I can understand that while the
cash related to float, is not ‘currently’ earning a return, it’s what happens between now and
judgement day that matters. So I see ‘potential’ value in that.

There are various approaches.
The arguments are never ending because two things are certainly true:

(1) The float is worth something. We invest a lot of it in things like Amex and pocket decades of dividends.
That has to be worth something.

(2) The float is NOT worth its full value to us as shareholders, even if it is perpetual.
The assets backing up those liabilities will always, of necessity, include some investments with zero real return
(cash and short term bills) that are of no use to us compared to a “normal” unconstrained investment portfolio the same size.
Sometimes big claims have to be paid out on short notice.

So, it’s necessary to pick a number between 0% and 100% for what it’s worth.
After much fiddling, I finally came to the conclusion that 70% was a good enough number.
i.e., I assume that $100bn of float at Berkshire has about the same very long term earning power value (and therefore value to shareholders) as a $70bn normal unconstrained investment portfolio.
Maybe you want to pick 50% or 80% or whatever, to taste. But I find 70% looks good.
Note that this is NOT the same as saying that the average size of the cash pile will average 30% of float over time.

The way I do it is simple:
I add up all investment assets. Cash, stocks, bonds.
(the occasional cyclical adjustment if a big position is extremely cheap or extremely expensive)
Then I subtract 30% of float. Other than that I ignore the float liability.
To me, that figure is the “good enough” value of all of the firm’s investments.

Note, this corresponds to a very old rule-of-thumb valuation methodology for Berkshire:
It’s worth book value per share plus 70% of float per share. (or some small multiple of that total–maybe 1.2?)
Given the simplicity, it’s not a bad rule.

Jim

12 Likes
16

For all practical purposes AXP, etc., own and control those earnings. Theoretically, as a part owner, BRK owns a portion of those earnings. But since BRK can’t do anything to get those earnings in hand and control how they are used, it seems like a waste of time, IMHO.

Please show me where I am wrong.

Well, there is a certain logic to that. But, if the logic applies to BRK’s investment in AXP, it also applies to you when you are doing your BRK analysis. You personally can’t do anything with BRK’s earnings or control how they are used any more than BRK can do anything with AXP’s earnings (actually, I dare say that BRK’s influence on AXP is probably considerably more significant than your influence on BRK). If you follow your own logic, the only value that you would assign to BRK is for the dividends that it doesn’t send to you because you cannot directly control its earnings or assets.

In reality, earnings that we personally cannot control and assets that we personally cannot allocate clearly still have economic value. Your valuation of those will be driven by your confidence in management. If you believe that AXP’s management (or BRK’s for that matter) will use those earnings in a value-destructive manner, you need to haircut them. But if you believe they’ll use them in a value-neutral manner or a value accretive manner they are definitely of considerable economic benefit to you.

SJ

10 Likes
17

A lot of thought goes into ‘how to value float’ as a sort of static liability/asset - but how do we value the part where we bring in an average of $9 Billion per year in new float - new cash in the door. I’ve heard Buffett many years ago include this growth in float in Free Cash Flow. It certainly is new cash in the door that gets invested, however conservatively. In my view, this $9 Billion in annual float growth is pretty significant.

4 Likes
18

Thanks to SJumper and Jim for chiming in.

The comparison to my personal “look thru earnings” in BRK was especially clarifying, because it has never occurred to me to calculate that as part of our real current holdings.

That being said, if any of you tally your liquid net worth based on intrinsic value of each position and the “look through earnings” of each position, it would be interesting to hear about that.

Do any of you guys do that?

Seriously, it is so easy to just dump market prices of common stocks into a spreadsheet to estimate one’s net worth.

:^)

19

“That being said, if any of you tally your liquid net worth based on intrinsic value of each position and the “look through earnings” of each position, it would be interesting to hear about that.”

I calculate my share of the normalized earnings power of the various businesses (stocks) I own. I’ve been a bit lazy but ideally, I would also calculate how that earnings power changes from year to year, especially since I’m still in the accumulation stage.

1 Like
20

I do a similar estimate of current earning power based on the last 5 years. Since all earnings ultimately end up in Shareholders’ equity for Berkshire and so do the stock prices for the investees which in turn reflect their underlying earnings over time, the long term change in shareholders’ equity provides a pretty reasonable sense of the current earning power being demonstrated.

The last 2,3,4 and 5 year average annual gains in shareholders’ equity are ( in billions)

40704 52,499 39475.75 44825.8

At the current share count ( as of most recent filings), this corresponds to a range of between
18.34 to 23.71 per share earning power. Averaged out a number of 19 or 20 as an estimate of current per share earning power seems more than reasonable and this is rought in the 41-42 billion range as estimated by others.

Note that this method uses no assumptions about future use of cash or growth etc ( which Bloomstran tends to be a bit gung ho on) but is a conservative estimate of current power based purely on the recent past,

8 Likes