Berkshire Hathaway
More Thoughts

Format for Printing

Format for printing

Request Reprints


By gdefelice
November 8, 2005

Posts selected for this feature rarely stand alone. They are usually a part of an ongoing thread, and are out of context when presented here. The material should be read in that light. How are these posts selected? Click here to find out and nominate a post yourself!

Here are some more extended comments and opinions for those interested.

As an inveterate bull on Berkshire, I realize I expose myself to accusations that I belong to cult of Berkshire. Nevertheless, I remain firmly bullish on Berkshire, its business and its valuation. I should disclose that I continue to maintain a very, very large percent of my portfolio in the stock.

From 30,000 feet, it is obvious to me that Berkshire is tremendously undervalued for anyone running a DCF model for the next 20 years. If fair value is the price at which one could buy Berkshire and receive a 7.5% CAGR for 20 years into the future, I think Berkshire would be fairly valued today at between $125,000 and $135,000 per A share. I think that valuation band increases each day.

In order to arrive at this valuation, one must look ahead. However, looking back over the last couple of years should also convince investors that the road ahead just may be enormously profitable. That Berkshire has endured the last two catastrophe years so well is a testament to its remarkable structure.

Berkshire is in an ideal position to benefit from rising insurance rates, as it still has a virtually bottomless well of capital. Berkshire has just reserved about $3 billion (pre-tax) for the largest single catastrophe in history (on an absolute and inflation adjusted basis). Given the nature of insurance payouts on cats like this (spread out over time) and the tax benefits, Berkshire can easily pay out this hit from the pre-tax earnings of its operating subsidiaries. This is remarkable. To my knowledge, there is no other insurance company in the world with a structure like this. They will likely not have to use a dollar of their cash or a dollar of their investment income (much less the principal of the bonds or the stock holdings) to pay these claims.

The anti-cultists claim that the Berkophiles conclude that no catastrophe losses are good for Berkshire and that huge catastrophe losses are good for Berkshire. They cite these responses from the cultists as if it is proof in itself that the cultists are mad. In fact, though, I think the Berkophiles have it right. No losses are good for obvious reasons and in years when little or no losses occur, one might argue that Berkshire should have written much more business. Losses are good because, since Berkshire has this enormous well of capital and earnings power, it is always in a position to write more profitable business when premiums are high.

The fact that places like Florida, as an example, continue to build high value structures in hurricane prone areas should be comfort to owners of the most well capitalized reinsurer in the world and not a source of concern. The continue growth in the value of property that needs to be insured means that Berkshire's revenues from this business will grow almost as far into the future as insured values grow. Moreover, when one factors in what inflation (even if building was constant) does to insured values on property (or what it does to insured values for casualty, for that matter), one can see that as long as Berkshire remains extraordinarily well capitalized, it will always be in a position to benefit from the rise in property valuations. This is part of the great growth story.

Berkshire's float now stands just under $50 billion. I believe it is likely that this float � on a net basis -- will cost nothing this year. This is astounding. One could argue that GEICO is carrying the load. However, even if one backs GEICO out of the equation, Berkshire after-tax cost of float would be well below rates for risk-free money � the metric that Buffett says must be used to determine if the underwriting business has value. (Note, of course, even if float cost 10% this year � with the biggest mega-cat ever, it would be folly to necessarily conclude that Berkshire's super-cat reinsurance business was a lemon. Happily, it isn't even necessary to refute such a claim).

If one backs out the $907 million in underwriting gains from the first 9 months of the year from GEICO (such that GEICO's combined would then have been 100), after tax losses from the remaining segments of Berkshire's insurance business would have been approximately $1.07 billion. Assume float has averaged about $38 billion this year (if you back out about $7 or $8 billion of float from GEICO) and that's a cost of float for the first nine-months of about 2.8%. This strikes me as a pretty conservative approach.

This simply is not supposed to happen when there is a $60 to $70 billion catastrophe.

Berkshire's insurance businesses are simply beautiful to any red-blooded capitalist.

I have little doubt that on a longer term basis this float will continue to power Berkshire's value for at least two decades. Float now stands at almost 55% of book value. Berkshire is highly leveraged but this leverage has traditionally come extraordinarily cheaply. Berkshire continues to borrow for less than the United States Government. Book value is about $90 billion and tangible book below $70. Total assets approach $200 billion and MidAmerican's debt (which will never be direct obligation of Berkshire) has yet to be consolidated.

Speaking of MidAmerican, we got some more color on what is likely to happen as a result of MidAm's consolidation onto Berkshire's books. We also got some color on what will happen with PacifiCorp (assuming it goes through, of course).

From the latest Q:

The Energy Policy Act of 2005 ("Act") was signed into law on August 8, 2005 and includes the repeal of the Public Utility Holding Company Act of 1935 ("PUHCA"), effective in the first quarter of 2006. Berkshire intends to convert its convertible preferred stock investment to common stock upon the effective date (six months after the Act was signed into law) of the repeal of PUHCA and after all regulatory approvals are obtained. Upon conversion, Berkshire will possess approximately 83.8% of the voting common stock interest and economic interests in MidAmerican. Thus, the accounts of MidAmerican are presently expected to be consolidated in Berkshire's Consolidated Financial Statements no later than the first quarter of 2006. However, there will be no changes in MidAmerican's operations, management or capital structure as a result of the consolidation of MidAmerican. Specifically, MidAmerican's debt is currently not guaranteed by Berkshire and the Company has no intention of guaranteeing any debt or obligations of MidAmerican in the future.

Equity in earnings of MidAmerican includes Berkshire's proportionate share (83.8%) of MidAmerican's undistributed net earnings reduced by deferred taxes on such undistributed earnings in accordance with SFAS 109, reflecting Berkshire's expectation that such deferred taxes will be payable as a consequence of dividends from MidAmerican. However, upon conversion of Berkshire's convertible preferred stock to voting common stock, Berkshire will include MidAmerican in its consolidated Federal Income Tax Return and income taxes will not be payable as a consequence of dividends from MidAmerican. No dividends from MidAmerican are likely for some time. Berkshire's share of MidAmerican's interest expense (after-tax) on Berkshire's investments in MidAmerican's trust preferred (debt) securities has been eliminated.

First, note that our actual economic interest in MidAm is almost 84%. Personally, I had been confused about what our actual economic interest was � I thought it might have been only 80.5%. This was a nice surprise.

As well, there is a little more color on the reporting of MidAm's actual owner earnings enjoyed by Berkshire. The implication of the above is that a deferred tax is being set aside against the future dividends that MidAm will one day pay Berkshire. I don't know what the rate would be on the tax � presumably 35%. Therefore, Berkshire is enjoying another interest free loan from the gov't for as long as these dividends are not paid to Berkshire. Berkshire's reported profits from MidAm do not include the "value" of this interest free loan. It looks like Berkshire will make, after-taxes, approximately $500 million at MidAm this year. I believe this includes both its after-tax economic interest in MidAm and its after-tax profits on its 11% trust preferred securities.

Here is more detail the likely structure of the PacifiCorp deal:

In May 2005, MidAmerican reached a definitive agreement with Scottish Power plc to acquire its indirect subsidiary, PacifiCorp, a regulated electric utility providing service to 1.6 million customers in California, Idaho, Oregon, Utah, Washington and Wyoming. MidAmerican will purchase all of the outstanding shares of PacifiCorp common stock for approximately $5.1 billion in cash. It is currently expected that MidAmerican will issue $3.4 billion of additional capital stock to Berkshire. The proceeds from the issuance of the capital stock by MidAmerican along with proceeds from the planned issuance by MidAmerican of $1.7 billion of long-term debt or other securities will be used to fund the purchase. The acquisition is subject to customary closing conditions, including the approval of the transaction by the shareholders of Scottish Power plc, which was received on July 22, 2005, and the receipt of required state and federal approvals. The transaction is expected to be completed in March 2006.

If I recall correctly, the original deal called for a $5.1 billion in cash to Scottish Power plus the assumption of $4.3 billion in debt. So, MidAm will likely borrow an additional $1.7 billion in debt. So, there will be about $3.4 billion from Berkshire and a total debt of about $6 billion. I am not clear if this $3.4 billion will be in the form of 11% trust preferred securities or if Berkshire's economic interest in MidAm will rise substantially due to the additional capital stock. In either case, Berkshire should profit handsomely as compared to rates on cash.

What is pretty clear to me is that, after the PacifiCorp deal closes (and perhaps already), MidAmerican will be Berkshire's most profitable non-insurance operating subsidiary and Berkshire's 83.8% economic interest might provide close to $1 billion annually pre-tax for Berkshire by the end of next year.

Revenues from all the major non-insurance operating subsidiaries are up against the first nine months of last year. Moreover, pre-tax earnings are up in every category (including "other businesses") except for Flight Services. In a few cases, these rises are substantial. I expect that, as increased input costs � in the form of raw materials and especially energy and oil � are ultimately passed on to buyers, Berkshire will continue to enjoy higher earnings from these subsidiaries. Because Berkshire's operating subsidiaries have moats, they can ultimately pass on increased costs to their customers. That is, in my opinion, an important reason that a moat has value.

I have seen people comment that in a recession the earnings of Berkshire's operating subsidiaries will "get crushed" like everyone else. While I readily concede that Berkshire's operating subs. will not thrive in a recession, Buffett has plainly stated over the years that he buys businesses with consistent earning power. Berkshire's operating subs. should fare much better than average in downturns just as they usually fare worse than average in upturns. As well, Berkshire's major equity investees are also businesses that have proven to be pretty recession "proof". In fact, many are referred to as "defensive" by the Street. Coke, Procter & Gamble, American Express, Budweiser, etc. are not businesses that should suffer mightily in a consumer recession as none of them sell expensive goods.

Moreover, given Berkshire's cash hoard, any recession should provide an opportunity to put the capital to work � we should be so lucky.

If no opportunities arrive, Buffett has suggested that by, approximately, May of 2007 � roughly 18 months from now � "something" will be done with all the extra capital.

Berkshire is so overcapitalized that an enormous one-time dividend could be paid and, in my opinion, little would be done to even alter Berkshire's capital position. I asked earlier today about the fact that Berkshire has an insurance surplus of $48 billion. Yet, excluding the huge earning power of the non insurance operating subs. -- which I think will approach $5.5 billion pre-tax next year � Berkshire has just under $40 billion in cash, $45 billion in equities, just under $25 billion in bonds, almost $2 billion in additional investments, and another $4 billion in equity still remaining in the Financial Products division. Presumably, insurance regulations simply do not count much of these investments when calculating surplus. It is conceivable that they count much of it as a fraction of its value and thus a big payout would reduce the total value. I don't know.

In any case, since the largest super-cat. ever can be paid out from the earnings of the operating subsidiaries, there is little doubt that Berkshire has far, far more capital than it needs. And, while the cash is likely to provide about $2 + billion pre-tax next year, even if every dollar was paid out as a dividend, there would still be $25 billion in bonds (likely providing over $1.5 + billion pre-tax) plus another $600 or $700 million in dividends (at least) from the equity holdings.

Let's assume that nothing changes next year except that the BHRG and Gen Re have no underwriting losses and GEICO produces underwriting gains of only $500 million.

We should see at least $4 billion in investment income from our cash, bonds and stocks. Add $500 million from GEICO and the insurance business produces $4.5 billion pre-tax. Of course, this doesn't include any appreciation from the equity holdings, which I think should grow about 7% per year (after dividends) from these prices on a 10 to 20 year basis. That again is pretty conservative given that the major positions have gone nowhere for half a decade even while their collective earnings have risen materially. Should our stocks ever rise again, a 7% annualized gain (after dividends) amounts to another $3 + billion per year before tax (which I think is accrued at something like 13% or 17%) in just year 1. This is always ignored when analyzing Berkshire because it has been so long since the big positions have gone anywhere. I think such an assumption suffers from a major case of rear-view forecasting. These major equity positions have grown at decent rates for many, many, many decades. Check Valueline or any other reputable source to see the consistent growth in earnings and dividends. In addition, some of the smaller positions continue to grow substantially in value � namely Moody's, Petrochina, White Mountains, USG, etc.

The non-insurance operating businesses which should include MidAmerican might produce $5.5 billion pre-tax next year. With little capital investment, I think these earnings will grow at 5 to 7% per year. They just perform remarkably.

There is little reason that, if one calculates even modest returns on the current equity portfolio, Berkshire isn't actually earning well over $10 billion after-tax on a real, owner earnings basis. I feel that one must look beyond GAAP to arrive at this conclusion.

I think that 13.5 times earnings for this business is cheap. Should Berkshire jettison some of its cash and/or find a home for more of it, the PE should drop. The core part of Berkshire is extraordinary. Eventually, I think the market will see it that way too.


Become a Complete Fool
Join the best community on the web! Becoming a full member of the Fool Community is easy, takes just a minute, and is very inexpensive.