Boosting Our Berkshire Valuation

We recently increased our fair value estimate for

We use a sum-of-the-parts methodology to value Berkshire. We believe the company's insurance operations are worth $83,100 ($55) per Class A (Class B) share, down 15% from our previous valuation because of lowered near- to medium-term projections for underwriting and investment results, as well as the extraction of capital to be used to complete the Precision Castparts deal. Berkshire's investments in Kraft Heinz KHC and Precision Castparts PCP (which we value separately from the insurance operations) are worth $14,300 ($10) and $20,600 ($14) per Class A (Class B) share, respectively.

Our estimate for Burlington Northern Santa Fe declined 5% to $58,300 ($39) per Class A (Class B) share because of a reduction in our near-term assumptions for railcar volume and pricing. We did increase our fair value estimate for Berkshire's energy and manufacturing, service, and retailing operations about 4% each to $22,100 ($15) and $57,800 ($39) per Class A (Class B) share, respectively, due to better-than-expected performance from both units. Our estimate for the company's finance and financial product division also increased slightly to $8,800 ($6) per Class A (Class B) share.

Not Expecting Much from Geico, Reinsurance Looking more closely at Berkshire's insurance operations, we've updated our forecasts for each of the firm's four subsidiaries--Geico, General Re, Berkshire Hathaway Reinsurance Group, and Berkshire Hathaway Primary Group. Geico's earned premiums and underwriting profits have traditionally been more consistent because auto insurance is a frequency-driven business. That said, the company has been on a tear the past several years, generating double-digit earned premium growth. Coming into 2015, we had thought that much of the firm's above-average premium growth was driven by Geico's desire to overtake Allstate ALL as the second-largest domestic provider of private auto insurance by premiums written (which it succeeded in doing in late 2013).

However, claims expense has also increased, a sign to us that Geico has been underwriting more aggressively than it perhaps should have been in some markets, undercutting its competitors and underpricing policies relative to their corresponding risk levels. As a result, the firm has seen an uptick in its loss ratio from 74.4% at the end of 2010 to 83.6% during the second quarter of this year, as it experienced increases in claims frequencies and severities in several of its major coverages. This has been prolonged by an improving U.S. economy and lower fuel prices, which has led to more drivers being on the road more often. An overlooked contributor to the uptick in accidents over the past five-plus years, at least in Warren Buffett's view, has been an increase in distracted drivers using smartphones to talk, text, or even watch videos while driving.

While Geico has not been alone in experiencing an uptick in its loss ratio on its auto insurance lines, with Allstate also being affected during the first half of 2015, Progressive PGR seems to have been more immune to the issue--perhaps a sign that its Snapshot telematics program is providing it with better information for its underwriting efforts than its peers. As long as Geico trails the industry and its closest rival from a loss ratio perspective, there will be calls for it to adopt some of the technologies that have allowed Progressive to generate one of the lowest loss ratios in the industry. We don't believe Geico's reticence in pursing telematics is a big issue yet, as the firm has traditionally overcome a fairly high loss ratio through scaling its underwriting expenses, but it seems to currently have some issues with adverse selection.

Geico had talked earlier this year about implementing premium rate increases to offset the increases it has seen in both the frequency and severity of customer claims, but these changes (which include tightening up the firm's underwriting standards and taking pricing actions) have taken somewhat longer to work their way through the system than we had expected. As such, we've pushed out our expected recovery in the firm's combined ratio (which was 98.1% during the first half of 2015) to the end of 2016 (as opposed to the end of this year). The downside to any recovery in the firm's combined ratio is that earned premium growth (at least on a volume basis) will tick down in the near to medium term. Combining this with the fact that many of Geico's competitors having stepped up their own efforts to become more price-competitive with Geico's low-cost offerings, we expect earned premiums to grow at only a high-single-digit rate during 2015-19, down from the double-digit rate of 2013-14 but more in line with what Geico was generating during 2008-12.

We're also not expecting much in the way of growth from Berkshire's reinsurance operations, which continue to struggle with a weak pricing environment driven by excess capacity that has been driven by the entrance of new players (including hedge funds), as well as the growth of alternative capital (such as catastrophe bonds) and the lack of large-scale catastrophes (which makes it harder for reinsurers to justify raising their rates). While Buffett believes reinsurance could be an unattractive business for Berkshire over the next decade, the company--unlike many of its peers in the business--has the luxury of walking away from reinsurance underwriting when an appropriate premium cannot be obtained, with both General Re and BHRG showing no real inclination to pursue growth in the current pricing environment. That's not to say that the company won't do business, though, as its reinsurance arms have had a knack for finding unusual opportunities that bring in a fair amount of float and provide Berkshire with adequate returns for the risks they are covering.

In this type of operating environment, reinsurers that have more diverse insurance portfolios and serve multiple distribution channels are more likely to offset the negatives in the market, and they should be able to target profitable opportunities as they arise. We think General Re and BHRG fall into this category. That said, Berkshire's reinsurance operations are also dealing with a number of runoff contracts in the near term, so we see the ability of the two firms to generate float being somewhat limited the next few years. With about three fourths of Berkshire's overall float coming from its two reinsurance arms--General Re (23%) and BHRG (51%)--we believe further increases in the overall level of the company's insurance float (which stood at $85.1 billion at the end of the second quarter) will be more difficult to come by.

For General Re, which reported its first decline in earned premiums in two years during the first half of 2015, we forecast low-single-digit annual declines in earned premiums over the next five years (down from flattish earned premium growth in our previous forecast). We also expect the firm's combined ratio, which has deteriorated to 98.0% this year (from 93.8% in the year-ago period) to eventually recover enough to put its average combined ratio over the next five years at 94.9% (the same level seen during 2010-14). BHRG's results have been a bit more difficult to forecast, given the impact that the runoff of the Swiss Re contract has had on earned premiums and profitability the past five years. Much as with General Re, we're not expecting much premium growth during the next five years at BHRG (which has made a name for itself underwriting policies with risks that most other reinsurers are unwilling or unable to take), with top-line growth being similar to General Re's--albeit with a bit more lumpiness in the annual results. We do expect profitability to improve over the 98.6% combined ratio the firm reported during the first half of 2015, with BHRG's profitability overall eventually settling down below the 96.4% seen during 2010-14.

We think a far more telling signal of the level of weakness that Berkshire expects from its reinsurance operations is the fact that Ajit Jain, who has been primarily responsible for BHRG the past three decades, has turned most of his attention to Berkshire's latest endeavor in insurance underwriting--Berkshire Hathaway Specialty Insurance--which was formed in June 2013 and resides within BHPG. While not quite as homogeneous as Berkshire's three other insurance subsidiaries, being a conglomeration of multiple insurance operations that offer coverage as varied as workers' compensation and commercial auto and property coverage, BHPG has been its fastest growing, with the best combined ratio and the strongest float growth over the past 5-, 10-, and 15-year time frames. Past results for BHPG include both acquired and organic growth, something that we expect to continue in the years ahead, with the division posting double-digit earned premium growth during 2015-19 as a result. We also expect the division's combined ratio to range between 85% and 90% during the next five years, allowing it to maintain its ranking as Berkshire's most profitable insurance subsidiary.

Combining our projections for the future underwriting results of the different insurance operations with a slightly better environment for investment income (as interest rates eventually rise in the near to medium term), we arrive at a fair value estimate of $83,100 ($55) per Class A (Class B) share. While our estimate of the overall value of Berkshire's insurance operations is down 15% from our previous valuation of $97,200 ($65) per Class A (Class B) share, this was due not only to the weaker expectations we've built into our valuation model for the different insurance arms, but the extraction of capital from the insurance operations to be used to complete the Precision Castparts deal. We think Kraft Heinz and Precision Castparts, which we value separately from the insurance operations despite being held on their books (primarily because we continue to have financial statements to work with), are worth $14,300 ($10) and $20,600 ($14) per Class A (Class B) share, respectively.

Volume Glut Dings BNSF As for Berkshire's noninsurance operations, the fact that both Burlington Northern Santa Fe and Berkshire Hathaway Energy file quarterly and annual reports with the Securities and Exchange Commission, despite being completely enveloped in the insurer's holding company structure, allows us to put together more robust forecasts for their operations. We've lowered our fair value estimate for BNSF to $58,300 ($39) per Class A (Class B) share from $61,500 ($41) due to a reduction in our near-term assumptions for railcar volume and pricing. After posting stronger results during the first quarter, BNSF took a step back in the second quarter, with revenue declining 6% year over year and pretax earnings increasing just 4%. On a year-to-date basis, revenue was down 2%, but pretax earnings were up 21%, with the year-ago period being affected by poor weather conditions in much of BNSF's territory that affected railcar volume and drove costs up meaningfully during the first quarter of 2014. Lower average revenue per car/unit (due primarily to lower fuel surcharges year over year) and relatively flat volumes (due primarily to softening demand for energy-related products) were responsible for the decline in year-to-date revenue.

While we view the current volume glut, which has also affected most of BNSF's peers, to be a temporary glitch in the company's long-term performance, we have revised our near- to medium-term outlook for volume, which contributed to most of the change in our fair value estimate. For 2015, in particular, we see overall volume declining around 2%, down from our previous forecast of flat to slightly positive growth in carloads hauled. Weaker near-term demand for coal (which has fallen off a cliff this year thanks to high stockpiles from mild weather and ongoing gas conversions at utilities and other customers) and Bakken shale crude oil (which is down because of the significant drop in crude oil prices over the past year) should keep volume growth under wraps next year as well. We expect things to normalize some after 2016, with volume growth returning to 1%-2% annually. Our new forecast has total unit volume increasing at a 0.9% compound annual growth rate during the next five calendar years (down from 1.5% previously). While this compares negatively with the positive 4.1% rate of volume growth BNSF generated on average during 2010-14, it should be noted that volume during that time was recovering off recessionary lows.

We also envision pricing being less robust overall for BNSF, with revenue per unit increasing at a 1.6% CAGR during 2015-19 (down from 3.2% previously). Pricing is expected to be down this year for each of BNSF's segments, with the exception being agriculture (where we see pricing rising slightly year over year), as overall revenue per unit is affected by weaker demand in some segments and lower fuel surcharges overall. That said, we still believe the pricing gains achieved during the remaining years of our five-year forecast should be enough to outpace rail cost inflation, which should contribute to ongoing improvements in profitability. We continue to believe that BNSF's operating ratio will improve from 69.8% during 2014 to 60.0% by the end of our five-year forecast period, putting the railroad's profitability more on par with its most comparable peer, Union Pacific UNP, which we project to have a 59.0% operating ratio by the end of 2019.

BHE Benefiting From Acquisitions We value Berkshire's share of Berkshire Hathaway Energy at $22,100 ($15) per Class A (Class B) share, up from $21,200 ($14) per Class A (Class B) share previously. While BHE came into 2015 with slightly higher hurdles, having benefited from the addition of NV Energy last year as well as solid results from its other regulated utilities, the firm has seen some benefit this year from the AltaLink purchase (which closed in December 2014) and ongoing acquisitions and better operating performance from its real estate arm, Berkshire Hathaway HomeServices. With BHE's year-to-date top-line growth at 5% at the end of the second quarter, the company is at the lower end of our targeted range of mid- to high-single-digit revenue growth for the full year. However, we expect the firm to generate high-single-digit top-line growth in the back half of 2015, which should push full-year revenue growth closer to the midrange of our forecast. This should improve profitability as well, with pretax operating margins increasing 100 basis points year over year to more than 16%.

We continue to expect BHE's U.S. regulated utilities--PacifiCorp, MidAmerican Energy, and NV Energy--to receive constructive rate-case outcomes and earn somewhere close to their current allowable returns on equity. This would leave annual revenue growth for these businesses in the 2%-3% range over the next five years. For Northern Powergrid, we expect ongoing price reviews to allow the firm to generate mid-single-digit revenue growth over the next five years. This is similar to the rate of growth we expect for the firm's pipeline and renewables businesses, which assumes the near-term prospects for additional renewables projects coming on line will be limited by the expiration of federal tax credits. On a consolidated basis, we expect annual EBITDA growth of 6.7% over the next five years, with EBITDA margins around 40%. Our assumptions about revenue and profitability for BHE's regulated utilities put our value for these operations at 5.5 times our 2015 EBITDA estimate, which is basically in line with our valuations for similar high-quality utilities with favorable regulatory structures and above-average growth opportunities. Our valuation also implies that BHE's pipeline group is worth 10 times EBITDA (on an enterprise value/EBITDA basis), in line with peer multiples and indicative of the higher returns that pipelines are able to realize.

Other Divisions Looking Brighter We expect Berkshire's manufacturing, service, and retail operations (excluding Precision Castparts) to generate revenue growth above the midrange of our mid- to high-single-digit target during 2015-19, slightly better than our previous forecast. While our long-term projection models past behavior by including bolt-on acquisitions to our future revenue forecasts, should Berkshire's acquisition activity be higher (lower) than we forecast, it would have a positive (negative) impact on our valuation for the segment. We see operating margins (excluding Precision Castparts) to expand some 50 basis points over the 7.0% the segment reported in 2014. When combined with the higher revenue growth expectations we have for the segment, our estimate of the value of Berkshire's manufacturing, service, and retail operations has increased to $57,800 ($39) per Class A (Class B) share from $55,200 ($37). At some point, we will roll Precision Castparts (which we expect Berkshire to add to the manufacturing, service, and retail segment for reporting purposes) into our valuation, but as long as we have financial data to work with, we'll continue to value it separately, much as we did with Lubrizol during the first couple of years after that firm was acquired in 2011.

As for the company's finance and financial product division, which includes Clayton Homes (manufactured housing and finance), CORT Business Services (furniture rental), Marmon (railcar and other transportation equipment manufacturing, repair and leasing), and XTRA (over-the-road trailer leasing), we've increased our estimate of the value of this unit to $8,800 ($6) per Class A (Class B) share from $8,700 ($6). We continue to expect revenue to grow 6% per year on average during 2015-19. While this is better than the low-single-digit rate of growth produced annually by Berkshire's finance and financial product segment during the past decade, we should note that our forecast has revenue growth starting out at a high-single-digit rate this year, then trending down to a mid-single-digit rate longer term. We continue to expect pretax margins for the finance and financial product segment to remain around 28% over the next five years, primarily because of the addition of Marmon's higher-margin transportation manufacturing, repair, and leasing businesses to the overall mix.

We Still See $10 Billion-Plus in Excess Cash by End of 3Q Berkshire's book value per Class A equivalent share at the end of the second quarter was $149,735. We expect it to expand to $156,460 at the end of the third quarter and $160,372 at the end of the year. This would infer a 6%-plus year-over-year base rate of growth for Berkshire's book value per share during the third and fourth quarters. It also includes the impact of the insurer having to true up the cost basis of its investment in Heinz, which was used as part of the purchase price for the Kraft Heinz deal. Berkshire has said it expects to book a noncash gain of around $7 billion during the current quarter to represent the increased value of the company's stake in Heinz that was merged into Kraft Heinz. We expect part of this gain to be booked as a deferred tax liability, with the remainder going toward shareholders' equity. Assuming an effective tax rate of 30%, Berkshire would book a $4.9 billion increase in shareholder's equity, equal to an increase in book value per Class A share of $2,982.

While the company closed the second quarter with $66.6 billion in cash on its books, a fair amount of that capital has already been spoken for. Buffett has been fairly explicit about his desire to keep around $20 billion in cash on hand as a backstop for the insurance business. We believe the rest of the firm's operations are also likely to require at least 2% of annual revenue on hand as operating cash, leaving Berkshire with $43.4 billion in dry powder that could be used toward acquisitions and other investments. Buffett already used $5.3 billion in early July to complete the Kraft Heinz deal and is slated to use another $22.4 billion to fund the Precision Castparts acquisition (expected to close in the first quarter of 2016). This would reduce Berkshire's excess cash balance to $15.7 billion. While the firm has been aggressive about buying shares of Phillips 66 PSX and other stocks during the third quarter, we expect it to still have more than $10 billion in excess cash on its books at the end of the third quarter.

Although the company did not buy back any shares during the first half of 2015 (and has not bought any since the fourth quarter of 2012), its stock is getting fairly close to the 1.2 times book value threshold that Buffett set out for share repurchases in the third quarter of 2011. Based on the firm's book value of $149,735 ($100) per Class A (Class B) share at the end of the second quarter, Berkshire should be willing to buy back stock at prices up to $179,682 ($120) per Class A (Class B) share, implying a floor on the company's common stock that is about 9% below where Berkshire's shares are trading. Assuming the company's stock price continues to trade at around the same levels, that threshold would be even higher by the end of the year, where (based on our calculations) Berkshire should be willing to buy back stock at prices up to $192,446 ($128) per Class A (Class B) share.