Message Font: Serif | Sans-Serif
No. of Recommendations: 55
Their letter, is 112 pages long. If you’re like me you didn’t really read all of it. So, since a recent post from Commoncents33 brought this back to our attention, I went back to read it more carefully. And, to decipher the part about Berkshire, the pivot—immunization to the bear, I thought it might be helpful to carve out and post some of their points [including ** for highlight, and some comments] from this, their 2/14/19 writing:

Per Semper Augustus—
• Our discussion of return on equity earlier in the letter concluded that shareholder return falls short of return on equity. That’s certainly been the case for investors in the broad stock market and the S&P 500 over time, and it’s unfortunately the case for investors in most businesses. However, across our investment portfolio, Semper returns over time approach and replicate the underlying return on equity of our holdings. Bought right, we can and have earned considerably more.
• ** Berkshire is a case in point where, unless purchased badly or exceptionally well, most investors earn the returns approaching Berkshire’s return on equity over time and should continue to do so over many years to come. They have for the past 54 years.

General Reinsurance – The Pivot: Part Deux
The Berkshire cult knows the original textile business was lousy. Current management gained control of the company in 1965. The textile business was closed 20 years later, in 1985. Quickly recognizing the economics of the operation were going to consume capital instead of creating it, the company pivoted, diverting capital to insurance. National Indemnity, the insurance operation that today remains the crown jewel of the holding company, was acquired in 1967.

Without the pivot from textiles to insurance, Berkshire certainly wouldn’t exist today. Precisely three decades from the acquisition of National Indemnity, Berkshire pivoted again, this time by buying an insurance company which allowed it to shrink from insurance and the leveraged stock portfolio that created so much value for 30 years. ** We think the second pivot was every bit as important to the longevity of Berkshire as the first, marking the economically rewarding shift from insurance, primarily the investing that goes with it, to energy production and distribution, railroads, and a host of manufacturers, retailers and service oriented businesses. Our view is not universally shared.
. . .
The Berkshire Chairman’s letter in the following year’s 2016 annual report made this comment about the
General Re purchase:

Unfortunately, I followed the GEICO purchase by foolishly using Berkshire stock – a boatload of stock – to buy General Reinsurance in late 1998. After some early problems, General Re has become a fine insurance operation that we prize. It was, nevertheless, a terrible mistake on my part to issue 272,200 shares of Berkshire in buying General Re, an act that increased our outstanding shares by a whopping 21.8%. My error caused Berkshire shareholders to give far more than they received (a practice that – despite the Biblical endorsement – is far from blessed when you are buying businesses).

Here I [Semper] defend my initial summary.
• It was precisely the use of the “boatload” of Berkshire’s shares in the purchase of General Re that was so seminal and materially beneficial to Berkshire over the subsequent twenty years.
• Yes, those shares would now be worth $83 billion at year-end 2018. But it was the purchase of General Re’s investment assets, namely its 90% allocation to fixed-income, that marked Berkshire’s pivot away from a massively overvalued stock market, from its own massively overvalued stock portfolio, which alone was 15% larger than Berkshire’s entire book value, and from its business concentration in property casualty insurance and reinsurance.
• You might question the last part of that – the move from insurance, given that General Re, the acquiree, was, in fact, a reinsurer.
• Yes, but it wasn’t so much the insurance operation that was attractive – it was the ability to purchase a $25 billion investment portfolio, overwhelmingly bonds, using a stock trading at three times book value when fair value was half as much.
• It was the ability to shrink a stock portfolio from 115% of book value to only 69% and to do so by paying no capital gains taxes, then at a 35% corporate rate.
• Did management at Berkshire anticipate two decades of subpar stock market returns? Did they recognize that at three times book they possessed overvalued currency?
• An acknowledgment of that would suggest perhaps that Berkshire took advantage of General Re, and Berkshire would never utter that.
. . .
Action speaks louder than words. Data compels action. The market rose another 25% in 1998 and saw the intra-year azimuth of what we call the “New Nifty 50”, recast from 1972’s original version which held that you could own the top 50 stocks forever and make good returns. Then the market fell by half in 1973 and 1974 and was negative through the 1982 low, putting to bed the nifty Rip Van Winkle approach. Berkshire’s core equity holdings peaked in 1998 at prices approaching 40 times earnings.

Those that lived it will never forget the tech bubble. Most seem to have forgotten how expensive the New Nifty 50 were in 1998, two years prior to the peaking of the S&P 500 and most definitely the NASDAQ. 1998 was the peak of the blue chips. As an aside its very interesting to see how many high-quality, high- return on equity portfolios today are full of consumer staple and consumer discretionary names at prices that resemble those seen twenty years ago. The Berkshire portfolio at year-end 1998 [included high-PE stocks, table omitted]…
• Coca-Cola alone had risen more than ten-fold in just nine years since Berkshire’s 1988 purchase and from a $5.5 billion position in 1994 to $13.3 billion three years later.
• When stocks were 15% larger than book value at year-end 2017, Coca-Cola, at 36% of the stock portfolio, was 42% of Berkshire’s entire book value!
• The stock had compounded at 33.8% per year from year-end 1988, the year Berkshire first bought Coca-Cola to its 1998 high – right when Berkshire bought General Re.
• Since then, from year-end 1998 through year-end 2018, Coca-Cola compounded at 4.3% per year!
• Think about that, 33.8% for a decade and then 4.3% for the next two decades!
• The other amazing thing is how much two decades of middling gains can whittle down previously wonderful performance. For the 30 years that Berkshire has owned the stock the annual gain is now only 12.6% per year.
• It’s the same math that’s driven the 20.2% earned by the S&P 500 from the 1982 low to the 2000 peak and what’s now only an 11.6% annual gain for the entire period. Tell me again Berkshire didn’t know what it was doing…
. . .
Perhaps instead Berkshire could have just said, ** “When our stock trades at three times book, we spend it in deals, when it trades below 1.5 times, we use cash.” Hmm….
. . .
• How well has the stock portfolio performed for the last two decades? At year-end 1997, just before buying General Re, the portfolio had a market value of $36.2 billion, five times its $7.2 billion cost basis. Here at year-end 2018, the stock portfolio is valued at $182 billion (including Kraft Heinz) against a cost basis of $119 billion.
• ** Gone is the five-fold gain over cost basis.
• The portfolio is now 53% above cost and we calculate has earned 5.5% annually, only slightly trailing the S&P 500’s 5.6% annual gain.
. . .
• Given that Berkshire’s stock portfolio had compounded at a mid-20% rate for more than two decades, and then abruptly subsequently compounded at 5.5% for the next 20 years, immediately after Berkshire spent its shares to buy General Re and shrink its exposure to its own overvalued portfolio, ** we’d call the pivot genius, a masterstroke. Management won’t acknowledge it, though. Any inkling that General Re’s shareholders were “taken advantage of” wouldn’t be whispered.
o The reality is, General Re’s shareholders have been far better off with their investment in Berkshire. General Re is a better business inside Berkshire.
o If Berkshire pivoted from Coca-Cola, stocks and insurance, it was the pivot that allowed Berkshire’s shares to compound by 7.6% post-merger.
o Had Berkshire not pivoted, it would have remained more leveraged to its stockholdings, and subsequent results would have been dismal.
o We’d argue that General Re shareholders would have posted gains well below those realized by retaining the Berkshire shares received as currency in 1998.
o There’s even an argument to be made that General Re might have failed as a standalone business in 2008-2009. GE and their disastrous insurance reserving comes to mind.
• Berkshire paid $22 billion in stock for General Reinsurance, with $14.5 billion attributed to goodwill. The 272,200 shares of equivalent A shares that Berkshire used in the acquisition were priced at $80,882 per share, roughly three times its book value at March 31, 1998. At the time of the merger, we appraised Berkshire’s intrinsic worth at just half of that price per share. To our thinking, Berkshire had bought General Re for $11 billion, not the $22 billion purchase price when adjusting for the premium valuation in Berkshire’s currency.
• Prior to the merger, Berkshire’s $36.2 billion stock portfolio was 75% of $47.5 billion total investment securities and the 115% of total book value as mentioned earlier. When General Re’s nearly 90% allocation to fixed-income securities were added to the mix, Berkshire’s investments in bonds and cash increased by $21.2 billion – from $10.0 billion to $31.2 billion, shrinking the allocation to stocks to 50% of the total investment portfolio and 69% of Berkshire’s new book value.
• In addition to immediately shrinking stocks to 69% of book value, the General Re merger served to likewise lower total firm assets in stocks to 30% from 65%, right at the outset of the Berkshire stock portfolio heading into a two-decade slumber, working off the hangover of the boozy run-up during the late 1990’s. Since 1997, equities have averaged only 23% of firm assets, where for the prior more than two decades they averaged almost three times that amount, a period where wonderful investment returns, leveraged or not, were enjoyed.
• Berkshire also tripled the size of its insurance float by acquiring General Re. General Re had roughly $14.9 billion in float at the end of 1998, compared to Berkshire’s $7.4 billion. With the addition from General Re, Berkshire’s combined float ballooned to $22.7 billion and increased invested assets at Berkshire by about $25 billion. It was an astounding transaction, paying $22 billion in stock which was worth only about half that and adding $15 billion in float which financed an additional $25 billion in investment assets.
• General Re instantly became a better company inside Berkshire. Berkshire’s capital strength allows General Re to retain more of its reinsurance business. The business writes roughly equal amounts of property and casualty and life and health reinsurance globally. Prior to the merger, General Re had a stand-alone AAA credit rating, and without Berkshire’s diversity and surplus capital had to rely heavily on the retrocessional market, and even to turn away attractive business to keep volatility of earnings low.
• The wrong way to view General Re is in light of its premium volume being not much larger now than it was at the time of the 1998 merger. For most of the past two decades in the Berkshire fold, General Re’s annual premium volume averaged a little more than $6 billion annually and most of its profits over the years have been paid as dividends to Berkshire. The company has dealt with industry overcapacity by intentionally not growing. Our sense is the business is writing as much business as it profitably can. For the first time in years, volume is up recently, and is approaching $8 billion in premiums earned during 2018. Underwriting margins at General Re were consistently negative in the handful of decades leading up to the merger with combined ratios averaging 102% for every ten-year interval up to 50 years. By retaining more business, underwriting profits post-merger have been consistently profitable.
• The impact of the pivot on Berkshire’s health can’t be overstated. If the stock portfolio averaged 5.5% unleveraged for 20 years while Berkshire’s return on equity and annual gain in book value per share averaged 9.0%, then everything other than stocks and investments did better than 9.0%. Considerably better.
• Berkshire’s growth in book value averaged 28.6% over the 23 years ended 1997 (again, the return on equity). Equities averaged about 105% of book value, so equity returns were slightly leveraged. The stock picking alone produced terrific returns, even without leverage. Throw in the balance of investment assets, which were mostly fixed-income securities with modest cash equivalents also in the insurance companies and the business enjoyed leveraged returns with assets averaging 150% of book value. Most liabilities consisted of non-interest bearing float and deferred taxes against stock portfolio gains. It was an incredible ride, culminating with a parabolic ascent from 1994 to mid-1998...
Pull up the stock price charts of Coca-Cola, 42% of the 1998 stock portfolio, and American Express, Capital Cities/ABC (Disney by 1998), Freddie Mac, Gillette, Washington Post and Wells Fargo for that parabolic period. The P/E’s were in our table but the price charts from that era were incredible. ** Imagine you were the Chairman, with a deep understanding of value, watching your portfolio race ahead far faster and far longer than the underlying businesses are growing. What do you do? ** Sell and pay a 35% tax on gains that are more than five times your cost basis? Nope. You figure out a way to diversify out of stocks without paying the taxman. Ergo, enter General Re.

Perhaps instead of lamenting how much it cost in today’s Berkshire dollars to have bought General Re then, one should acknowledge what the purchase allowed the balance of Berkshire not only to do, but also what it was able to avoid. Further, instead of comparing General Re’s lack of success to the wonderful history of National Indemnity and eventually GEICO, perhaps consider those businesses were bought for different reasons in different eras. The same hand guided Berkshire’s stock portfolio and allocation to its book value for three decades before the General Re’s merger and for the two decades since. If the stock portfolio has earned a fraction of what it had under the same guiding hand in a different era, it’s logical to applaud everything else that was subsequently acquired and has contributed to the most recent 20-year run of 9% to 10% returns in a world where the stock market has done half that. Berkshire’s extraordinary success in stocks leading up to 1998 handicapped its subsequent performance. The company had the high- class problem of having made too much money too soon and played that hand beautifully. It’s worth applauding the pivot and the results both prior to and following it.

Let’s conclude this section with a recasting of the familiar first page of Berkshire’s Chairman’s letter updated each year. We show the three familiar columns that reflect change in book value per-share, change in stock price, and S&P 500 total return, all presented annually since 1965. We then supplement two extra columns for each original measure. For each metric, we add a compound annual growth rate for each year beginning in 1965, progressing forward, and a compound growth rate working backward. The backward working figures simply measure the one-year return, two-year average returns, three-year average returns and so forth using December 31, 2018 as the final year in each series. [see page 81 of attached.pdf]

** [Conclusion when the BRK.B stock-price was ~$200/share circa Feb-2019]:
• From the rightmost tables, the stock will return 11.2% per year if return on equity averages 8%, and will return 13.2% if return on equity averages 10%. Both earnings estimates are 1% higher than a year ago. The stock closed 2018 at 13.5 times estimated our calculation of normalized earnings, down from 15.4 times a year ago (using the updated tax assumptions we made for 2017).
• If the current multiple to earnings remains unchanged from today’s 13.5 a decade from now, you will earn Berkshire’s return on equity.
• If return on equity averages 8%, you will earn 8% absent any multiple expansion or contraction.
• If return on equity averages 10%, you will earn 10%.
• You can interpolate this scenario between the 13x and the 15x earnings columns in each table. An investor should be thinking in these terms.
• If average annual return on equity falls to 8% for the next decade instead of our 10% projection, and if the multiple to earnings contracts from 13.5x to 13x, you will earn a respectable 7.5% per year [worst case].
• That’s earning the return on equity minus the multiple contraction over a decade’s time. This is our worst-case assumption and exceeds a conservative estimate for returns in the broad stock market.
. . .
** Berkshire can on one hand be a beast of a company to get your mind around and at the same time be immensely understandable and predictable in terms of sustainable earning power.
. . .
For the year, we calculate Berkshire will show $5.7 billion pre-tax income and $5.5 billion in profit after- taxes. You read that right. Assuming a typical quarter on the operating front, ** Berkshire will show virtually the same amount in net income and in pre-tax income and a tax bill for the year of only $200 or so million, a tax rate of about 4%. Lordy.

As is always the case at Berkshire, the stated figures are always far from economic reality. On a GAAP basis, the stock market decline for the quarter and the year is “sheltered” by a reduction of the deferred tax liability that exists thanks to the stock portfolio trading at a gain above its cost basis. Gains and losses are offset by deferred taxes calculated at the 21% new federal rate. The $39.1 billion “loss” in the fourth quarter on the stock portfolio (excluding KHC) produced a tax “benefit” of $8.3 billion. The assumed $8.5 billion pre-tax income earned by Berkshire’s businesses will be taxed at approximately 18% for the quarter (and year). The rate is below the new federal 21% tax rate largely thanks to wind energy credits being paid to Berkshire to subsidize its large investments in wind power. The company had been reporting a 19% combined tax rate for the year, which is a combination of the tax rates across all the businesses and the deferred tax treatment at 21% of unrealized gains and losses in the investment portfolio. We’ll get to it in a bit but the cash taxes paid by Berkshire are even below this 19% number.

On the book value front, Berkshire’s $24.1 billion net loss for the fourth quarter will shrink book value by the same amount. The company had also spent approximately $928 million repurchasing shares through the third quarter (we expect and hope further shares were repurchased during the final quarter as the stock traded at attractive prices – we were a buyer if that means anything). Repurchases reduce shareholder’s equity by the dollar value of the purchase (offset a small bit by an amount representing capital in excess of par). Book value per share will decline slightly more due to the purchases having been made above book value. Throw in perhaps another negative $1 billion to $2 billion for transactions with noncontrolling interests and other comprehensive income and most of what had been a $27.3 billion gain in book value through September 30 for the year will have disappeared thanks to the 13% fourth quarter decline in the stock portfolio. If we learn the fourth quarter saw additional repurchases, any new stock market buys that declined by year-end, or any weakness in operating subsidiaries or insurance underwriting losses then the book value may have declined for the year. We’ve already quoted Tony Romo, but with the Super Bowl coming up, as Tony would say, “Oh boy, this is going to be close, Jim!” Good news for cheerleaders is that a flat book value will have beat the 4.4% decline in the S&P 500.

The truly better news, if you have read our letters for the last few years, ** you know there are enormous economic earnings earned at Berkshire that generally go unreported in the GAAP financial statements. A flat book value for a year necessarily implies better prospective returns to come both in intrinsic value and in the price of Berkshire’s shares.

Net Income Basis – 2018 Year-End Estimated (dollars in billions)

Pre-Tax Income After-Tax Net Income
Operating Groups
Berkshire Hathaway Energy $3.3 $3.8
BNSF 6.9 6.1
Manufacturing, Service and Retail 10.4 8.0
Finance and Financial Products 2.5 1.9
Operating Group Subtotal 23.1 19.8
Insurance Underwriting Normalized Gain 2.8 2.2
SAI Pension Expense -0.5 -0.4
Operating Group Plus Insurance Underwriting 25.4 21.6
Investment Income (Insurance and HoldCo) 16.6 15.5

Totals $42.0 $37.1

Cash Tax Rate 11.7%
Source: Semper Augustus

One note from the table. ** You correctly read the pre-tax and after-tax income figures for the Energy business [i.e., After-tax income is higher than pre-tax]. The company is receiving large tax credits for its investments in alternative wind (mostly) and solar energy. We further lower the tax bill for the ongoing use of accelerated depreciation on qualifying capital expenditures. You won’t see that portion of tax savings in the 10-K subsidiary presentation. We included an updated reconciliation between cash taxes and GAAP taxes in the appendix.

2018 Intrinsic Value by Market Cap and Per Share
Market Capitalization Price Per A Share Price Per B Share
Sum of the Parts Basis $659 billion $401,274 $268
GAAP Adjusted Financials 668 billion 406,754 271
Simple Price to GAAP Book Value 611 billion 372,046 248
Two-Pronged Approach (Ours) 672 billion 409,190 273
Simple Average 653 billion 397,316 265
Source: Semper Augustus

In most years, our methods used in estimating Berkshire’s intrinsic value produce similar results. Our primary process with Berkshire and throughout our portfolio usually involves a focus on sustainable economic profitability first and foremost and then secondarily the price we are willing to pay for those profits. With Berkshire, our Net Income Basis – GAAP Adjusted Financials approach drills down to our assessment of economic profits. Getting to the profit number requires several modifications to the published numbers and now yields more than $13 billion in annual earning power that you won’t find in the financial statements. At our estimate of fair value, the hidden earnings account for $234 billion of Berkshire’s intrinsic value.
. . .
Berkshire enjoyed a tremendous advance in earning power and intrinsic value, masked by the stock going nowhere during the year, by an obvious large decline in the stock portfolio, and because of financial statements that are now thoroughly incomprehensible to most readers. Previously they were simply incomprehensible. The business is reaping huge benefits from the 2017 tax code change.

Thanks to the falling stock portfolio and to large net purchases at low multiples during the year, the portfolio shifted from overvalued to undervalued. Most subsidiaries are enjoying record profits. We find Berkshire’s shares nearly as undervalued as they were at year-end 2015, when the stock had fallen 12.5% and clients were wondering if we had lost it, or if the folks in Omaha had.

Our 2015 letter was an attempt to allay those concerns. Whether the letter served its purpose, or the subsequent 50% gain both in Berkshire and in our stock portfolios, we can’t be sure. With Berkshire’s 2.8% gain in 2018 and the 1.4% decline in our stocks, we are thrilled to be sitting here staring at substantial upside again. It’s remarkable that at a time when so many assets are overvalued that we have been able to put together a portfolio of such high business quality and low price, with Berkshire as the cornerstone.
. . .
Intelligent investing can only be done with a constant appreciation of risk and a never-ending asking of the questions “Why?” and “What can go wrong?”. The theme of this year’s letter was Addicted to Loans. The ballooning debt bubble is front and center today.

While it’s been an obvious issue for years, the rising corporate component is of particular concern. Our process shuns leverage, and certainly excessive leverage. It’s not our game. Neither was investing in tech and internet names in the late 1990’s, but that didn’t mean we couldn’t see it and take requisite measures to avoid the places that would be affected by its fallout. It’s hard to know how the debt bubble will rectify itself. It could be slow or it could be quick.

One thing’s for sure, there won’t be many places to hide.
That allocation flows to places that employ maximum leverage, like private equity today, is mystifying. Allocators chase past returns, and the embedded risks in asset classes that thrive on the razors edge of bankruptcy now harvesting gargantuan flows that in no way can be put to economic good is comical, because we’ve seen this picture show before.
After a 10-year bull run, is the memo— this time is really not all that different?
I.e., Too much money chasing too few deals (per Howard Marks letter Sept-2018)
Or, even if economic history doesn’t really repeat itself—does it still rhyme?
Does Buffett’s reasoning for holding so much cash support a similar theme?
But, why not just throw down and buy-back BRK stock?
… Or, is Warren hedging Berkshire’s bets at the chance for better deals?

I think this letter supports our Chairman’s actions of late. Patience.
Print the post  


What was Your Dumbest Investment?
Share it with us -- and learn from others' stories of flubs.
When Life Gives You Lemons
We all have had hardships and made poor decisions. The important thing is how we respond and grow. Read the story of a Fool who started from nothing, and looks to gain everything.
Contact Us
Contact Customer Service and other Fool departments here.
Work for Fools?
Winner of the Washingtonian great places to work, and Glassdoor #1 Company to Work For 2015! Have access to all of TMF's online and email products for FREE, and be paid for your contributions to TMF! Click the link and start your Fool career.