As I mentioned yesterday, there wasn’t anything that amazing and new in the annual letter of Berkshire Hathaway. Lots of people found things to comment on, and there is always something true to be reminded of by Buffett, but there was little that was new. Tonight, I want to focus on a few new things, most of which was buried in the insurance section of the annual report.
Before I get to that, I do want to point out that Buffett historically has favored businesses that don’t require a lot of capital investment. That way the earnings are free to be reinvested as he see fit. He also appreciates having moats, because of the added pricing power it avails his businesses. Most of his older moats depend on intellectual property, few competitors, established brand, etc. Burlington Northern definitely has little direct competition, but it does face national regulation, and dissatisfaction of clients if services can’t be provided in a timely and safe manner.
Thus the newer challenge of BRK: having to fund significant capital projects that don’t add a new subsidiary, may increase capacity a little, but are really just the price you have to pay to stay in the game. From page 4 of the Annual Letter (page 6 of the Annual Report PDF):
Our bad news from 2014 comes from our group of five as well and is unrelated to earnings. During the year, BNSF disappointed many of its customers. These shippers depend on us, and service failures can badly hurt their businesses.
BNSF is, by far, Berkshire’s most important non-insurance subsidiary and, to improve its performance, we will spend $6 billion on plant and equipment in 2015. That sum is nearly 50% more than any other railroad has spent in a single year and is a truly extraordinary amount, whether compared to revenues, earnings or depreciation charges.
There’s more said about it on pages 94-95 of the annual report, but it is reflective of BRK becoming a more asset-heavy company that requires significant maintenance capital investment. Not that Buffett is short of cash by any means, but less will be available for the “elephant gun.”
Now for more arcane stuff. There are lots of people who write about Buffett and BRK, but I think I am one of the few that goes after the insurance issues. I asked Alice Schroeder (no slouch on insurance) once if she thought there was a book to be written on Buffett the insurance CEO. Her comment to me was “Maybe one good long-form article, but not a book.” She’s probably right, though I think I have at least 10,000 words on the topic so far.
Here are two articles of mine for background on some of the issues involved here:
Here’s the main upshot: reserving is probably getting less conservative at BRK. Incurred losses recorded during the year from prior accident years is rising. Over the last three years it would be -$2.1B, -$1.8B, and now for 2014 -$1.4B. (See page 69 of the annual report.) Over the last three years, the amount of reserves from prior years deemed to be in excess of what was needed has fallen, even as gross reserves have risen. In 2012, the amount of prior year reserves released as a proportion of gross reserves was greater than 3%. In 2014, it was less than 2%.
In addition to that, in general, the reserves that were released were mostly shorter-tailed reserves, while longer-tailed reserves like asbestos were strengthened. In general, when longer-tailed lines of business are strengthened in one year, there is a tendency for them to be strengthened in future years. It is very difficult to get ahead of the curve. Buffett and BRK could surprise me here, but delays in informing about shifts in claim exposure are a part of longer-tailed lines of insurance, and difficult to estimate. As I have said before, reserving for these lines of business is a “dark art.”
From page 91 of the annual report:
In 2014, we increased estimated ultimate liabilities for contracts written in prior years by approximately $825 million, substantially all of which was recorded in the fourth quarter. In the fourth quarter of 2014, we increased ultimate liability estimates on remaining asbestos claims and re-estimated the timing of future payments of such liabilities as a result of actuarial analysis. The increase in ultimate liabilities, net of related deferred charge adjustments, produced incremental pre-tax underwriting losses in the fourth quarter of approximately $500 million.
This was the only significant area of reserve strengthening for BRK. Other lines released prior year reserves, though many released less than last year.
There were a few comments on insurance profitability. In addition to asbestos, workers’ compensation lost money. Property-catastrophe made a lot of money because there were no significant catastrophes in 2014, but rates are presently inadequate there, and BRK is likely to write less of it in 2015.
My concern for BRK is that they are slowly running out of profitable places to write insurance, which reduces BRK’s profitability, and reduces the float that can be used to finance other businesses.
Maybe BRK can find other squishy liabilities to use to create float cheaply. They certainly have a lot of deferred tax liabilities (page 71). Maybe Buffett could find a clever way to fund pensions or structured settlements inexpensively. Time to have Ajit Jain put on his thinking cap, and think outside the box.
Or maybe not. Buffett is not quite to the end of his “low cost of informal borrowing” gambit yet, but he is getting close. Maybe it is time to borrow at the holding company while long-term rates are low. Oh wait, he already does that for the finance subsidiary.
From an earnings growth standpoint, there was nothing that amazing about the earnings in 2014. A few new subsidiaries like NV Energy added earnings, but existing subsidiaries’ earnings were flattish. Comprehensive income was considerably lower because of the lesser degree of unrealized appreciation on portfolio holdings.
On net, it was a subpar year for Berkshire Hathaway. The annual letter provided a lot of flash and dazzle, but 2014 was not a lot to write home about, and limits to the BRK business model with respect to float are becoming more visible.
Full disclosure: long BRK/B for myself and clients, for now