Dissecting one of Berkshire Hathaway's greatest purchases - BNSF

Disclaimer: This is a high level conceptual exercise and all figures are approximations using the CY2020 accounts.


In early 2010, Berkshire Hathaway finalised the acquisition of BNSF (Burlington Northern Santa Fe), the largest US based railroad for the consideration of $26 billion for the stock it did not already own. This equated to a total deal value of $34 billion.

The consideration of $26 billion was paid with ~$16 billion in cash and the balance in Berkshire stock; with around $8 billion in debt drawn to partial fund the cash consideration. Broadly, this structure was in line with Berkshire's target leverage of around 30% against assets.

Berkshire's operandi modus is to let competent management teams run their businesses but enforcing strict capital discipline with a strong focus on cash generation to the parent company. Optimal capital allocation decisions result in a large advantage over time, which Berkshire personifies in many ways. 

Without digging too deeply into the operational performance since acquisition, it is clear that Berkshire purchased BNSF at a cyclical low with extraordinary timing; just as the post-GFC recovery was gaining momentum. The growth in top-line revenue has been anemic, with a rise of only around 25% over the 10 year period since acquisition. However, cost control, capital allocation and market conditions have been sound and allowed Berkshire to harvest outsized returns from the investment.

Since acquisition, Berkshire has invested $41 billion in capital expenditure offset by depreciation of $20 billion. This implies that BNSF may have been either 1) under-investing in capital items prior to Berkshire's acquisition or; 2) Berkshire have invested heavily in capital items due to a strategic rationale. 

The latest Berkshire shareholder letter provides some clues: "Railroading is an outdoor sport, featuring mile-long trains obliged to reliably operate in both extreme cold and heat, as they all the while encounter every form of terrain from deserts to mountains. Massive flooding periodically occurs. BNSF owns 23,000 miles of track, spread throughout 28 states, and must spend whatever it takes to maximize safety and service throughout its vast system."

Whilst depreciation is a relatively stable portion of fixed assets; year-to-year capital expenditures can fluctuate more broadly depending on capital allocation decisions and any externalities that may impact the operations. In addition, given the relatively long-life nature of heavy fixed railway assets and the inexorable push of inflation over time; this effectively means that a dollar spent today, could mean over a dollar saved tomorrow in aggregate holistic costs.
Accordingly, we anticipate that BNSF will require a relatively lower capital requirement going forward, probably a figure closer to BNSF's depreciation and amortization expenses.

Incidently, this aggressive capital investment program has come about despite almost $42 billion in dividends paid to Berkshire over this time period. No doubt that a reasonable portion of this has come from levering up the balance sheet against new fixed assets against which Berkshire has greater certainty around their effective life. None-withstanding, this is a fantastic result against Berkshire's cost base of ~$30 billion. 

The largest component of value attributable to BNSF is the terminal value. The company's claimed equity value on its books is around $65 billion which looks to be under-valued on a standalone basis. For instance, BNSF's competitor, Union Pacific Corporation (UNP), is valued at around $154 billion at a capitalised dividend yield of approx 1.8%. 

I cannot say whether UNP has invested the minimum requisite capital expenditure necessary to keep its operations running smoothly, but I can say that BNSF's heavy capital investments provides it a good degree of certainty around its future capital requirements; so it's appropriate to use UNP's dividend yield as a basis for our standalone terminal value. 

For CY2020, a $5.8 billion dividend was paid by BNSF to Berkshire; we expect this to grow over time largely dependent on the sustaining capital requirements. Capitalising this dividend at a yield of 1.8%, provides us with a terminal value of $322 billion. If we are even more conservative and decide to use a 3% capitalisation figure, this provides us with a terminal value of around $193 billion. Both figures are vastly higher that BNSF's claimed book value of $65 billion.

Berkshire's overall return on BNSF has been over 28% per annum in our upper case and almost 23% in our lower case. It's actual return on equity is significantly higher; circa 32% and 27% in upper and lower cases.

These are extraordinary returns considering the size of the investment and truly demonstrate why the Berkshire team are considered the world's pre-eminent living investors.




Disclaimer: This article does not take into account your investment objectives, particular needs or financial situation; and should not be construed as advice in any way. The author holds no exposure to the stock discussed