So What is BRK Really Worth? (Part 2)

Let's look at the basic model for valuing Berkshire Hathaway.  It's the model that is endorsed by Buffett himself in his annual reports.

BRK basically breaks down into two basic components: 

1.     An insurance company
2.     A bunch of wholly owned operating businesses.

In what is usually called the two column valuation method, the insurance business is assumed to be worth the total investments held in the insurance segment.  Since they do at least break even on underwriting insurance (float cost of zero or better), then the profits that the insurance company makes will simply be the investment returns on their investment portfolio.  Therefore, the insurance business is simply worth whatever investments it owns.

I will look at that in more detail in a later post, but for now let's just take that as a reasonable assumption.  Buffett is no promoter-type; he wouldn't advocate something that wasn't reasonable if not conservative.

So the insurance segment is valued simply as total investments per share held.

The other operating business segment is valued simply at some multiple of the pretax profits of the operating business; this includes all businesses that BRK owns outside of the insurance business and is not included in the total investments in the insurance segment (which is cash, bonds, equities and other securities).

So the instrinsic value of BRK is calculated simply by adding 1. the total investments per share and 2.  some multiple of the pretax profits of the operating businesses.

From the 2010 annual report, we see that investments per share representing the insurance operations was $94,730/share (per A-share).

Also, the pretax operating earnings per A-share was $5,926/share.  Just as a quick check, I looked at what the net (after-tax) profit was of the non-insurance businesses and that came to $3,964/share, or around 67% of the pretax number.  Some people use a 35% tax rate to adjust the pretax earnings number to get a quick estimate of the net figure and we see here that this is not far off.

What is the Intrinsic Value of BRK Using the Two Column Method?
With the above figures, it's easy to calculate the intrinsic value of BRK.  First, the investments per share is worth just that, investments per share.  That is:

$94,730/share

To calculate what the non-insurance operating businesses are worth, we have to find a multiple to earnings.  Since people look a lot at p/e ratios and p/e ratios are based on net earnings, let's use the net earnings per share figure above, which is $3,964/share.  The long term market p/e and recent p/e ratio too of the stock market happens to be around 15x p/e, so let's say BRK's operating businesses are at least as good as the average company.  So using 15x p/e gives the non-insurance operating business the following value:

$3,964 x 15x = $59,460/share

So summing the two gives us an intrinsic value of: $154,190/share

With BRK shares now trading in the $110,000/share - $120,000 per share range, we see why many BRK fans see it as undervalued and attractive; it is trading at a 20-30% discount to it's intrinsic value.

But Here's the Problem
OK, so that's an overstatement.  This is not really a problem, but just something that I would like to point out.  We take it for granted that the investments per share are actually worth investments per share.  Well, of course it is.

But we have to look at what's inside of that.  I think some people assume that because Warren Buffett is investing those assets, that the return is going to be pretty high.  Others may assume that most of that is in equities so it can do well in a bull market.

This is true to an extent.  Buffett will outdo most others in investing the insurance assets.  But as we have seen in the other post, we know that BRK will not invest all investments into equities; they will own a lot of that in cash and bonds, and in this low rate environment, if it continues, will lower expected returns on this "investments per share" portion of BRK's intrinsic value.

The total investments in the insurance segment of BRK was $137 billion and that broke down as follows:

Cash:      $25 billion
Bonds:    $33 billion
Equities: $60 billion
Other:     $19 billion

We know that bond yields are as low as 2% out to the ten year, and cash rates are basically zero.  The "other" section has special securities, often with a 10% coupon (like the Bank of America preferreds, previously the GE and GS preferreds with 10% coupons etc...), so I will assume the "other" earns 10% pretax.

The equities is the stock portfolio.  Let's assume a generous 10% return on equities over time.  Since BRK only owns the largest of large cap stocks, some may argue that 10% is way too high.  Many are expecting 6-7% returns for large company stocks.

But let's just use 10% for now.  What is the blended, expected return on this total portfolio?

Cash return:   $0
Bonds:           $660 million  (2% yield)
Equities:        $6 billion (10% return)
Other:            $1.9 billion (10% coupons)
Total:             $8.6 billion

Total percentage return:  $8.6 billion / $137 billion = 6.3%

If you lower the equities expected return to 7%, then the total portfolio expected return drops to 4.9% or so pretax.

So if you value BRK using total investments per share, then that portion of the stock will return to you only something like 4.9%-6.3% pretax over time.

This is certainly not a bad return given that bonds yield only 2%, but I don't know that the average BRK investor expects such a modest return on the investment portfolio managed by Buffett himself.

Buffett and Munger will tell you that there is nothing wrong with 5-6% returns, and to expect more is folly in this low rate environment.   Fair enough.

But equity investors like to invest in businesses with high returns on equity.  With lower interest rates reducing the return on equity of insurance and other financial companies, I don't think it's wrong to value them a bit lower. 

Again, this is only relevant and an issue when low interest rates may persist for a long time like in Japan, and I am more and more leaning towards that scenario (this is the reason why I am doing this whole exercise in the first place).

If equity investors demand 10% return on their investments, which to me is not that unreasonable even now, then it would not surprise me in the least if the market didn't give full credit to the total investments per share at BRK (for example, the market may demand a discount to the portfolio such that the portfolio expected return does go to 10%.  In this case, a discount of as much as 50% would not be too surprising; with a 50% haircut on the investments per share, the BRK total intrinsic value would fall to $106,825/share.  This means if you paid this price, you could expect a 10% return on the investments per share portion of BRK and pay 15x p/e on the operating business).

However, all of this analysis is based on a static analysis.  Another factor that will *increase* the expected return over time will be growth in float.  As float increases over time, the investments per share will also increase on top of whatever is earned in terms of capital gains, interest and dividends.

Float has grown tremendously in the past at BRK, but Munger has argued that it would be unreasonable to assume that float can keep growing as they get bigger.

In the next few posts, I will take a look at the book value of BRK by segment, the return on equity (ROE) by segment and see if it is reasonable to assume that BRK trade at over book value per share.  I myself find 1.5x book value a reasonable valuation of BRK based on some of the popular, basic models.

But I want to sort of look under the hood to see where this ROE is coming from, which segments should be valued above book value etc...