|Shares Out. (in M):||0||P/E|
|Market Cap (in $M):||29,600||P/FCF|
|Net Debt (in $M):||0||EBIT||0||0|
Burlington Northern Santa Fe is the second largest Class I railroad servicing the Western half of the
Brief History of Industry
In 1980, the Staggers Act was signed that largely deregulated the railroad industry since the passage of the 1887 Interstate Commerce Act. Not surprisingly, the industry underwent significant consolidation over the next 20 years bringing the number of class I railroads (a railroad company with over $320 million revenues) from 30 businesses to seven today. In the
Since 1980, the number of rail employees has been cut by over 50%, the miles of track by 40%, the number of freight cars in service by 25% and at the same time, the number of revenue ton miles operated by the rails has doubled. In 2004, demand started to outweigh the reduced supply of the railroad industry and the Rail companies finally gained pricing power of their customers.
A railroad company generates cash by charging customers such as utilities or trucking companies a fee per ton of load carried and uses cash to pay expenses of labor (mostly unionized), fuel, equipment rents, depreciation/replacement of railcars and track, and transportation services such as ramping and drayage.
The agricultural products segment generates approximately 17% of revenue for
Finally, BNI generates about 24% of revenue from the transport of industrial products. Of this 24%, revenue is comprised of building products (33%), construction products (32%), petroleum and chemical products (27%) and food and beverage (8%).
First, is the “hidden asset” in the form of below market contracts. Coal contracts are typically 10 years in length and since we are 3 years post the time during which they enjoyed pricing power, there are 6-7 years worth of coal contracts signed at a time when 1) supply exceeded demand and 2) UNP and BNI were engaged in price wars. Management estimates that these contracts are about 30% below market if re-signed today. Applying decently conservative estimates to the coal revenue streams, a full repricing will add approximately 20% of fully-taxed free cash flow to the current earnings streams. While this is relatively well known in the industry, another “hidden asset” lays in the fuel provisions in the contracts. Prior to 2004, many of the contracts did not contain the fuel pass thru that are included today. Diesel averaged around 65 cents per gallon versus the LTM $2.03 per gallon paid. Assuming simply a ‘mark to market’ of gas prices in historical contracts, BNI will add another 17.5% of fully-taxed (38%) free cash flow to current earnings. Over a 6 year time frame, repricings should add approximately 6% per year to a holder’s total return bringing us to 12% prior to any growth in GDP+inflation plus the earnings growth resulting from owning a non-replaceable asset that acts as a tax on commerce.
In addition the coal contracts, I think there lays significant hidden value in the intermodal business. Because I cannot find segment level numbers, consider simply that rails is a $50 billion per year industry. Trucking is a $600 billion per year industry. Now rails will never move things from Wal-mart’s distribution center to their stores, however there is significant upside as gas prices have increased for rails to take market share from the trucks in the medium hauls. Currently, the rule of thumb is less than 500 mile per haul goes to truck. However, implicit in that rule is a gas price which has tripled over the past years as we saw in the coal contracts. As gas prices increase, the “market” for intermodal actually increases exponentially. And so we are up to 6% cash yield + 6% from repricing plus GDP + inflation + outsized growth in intermodal business given its cost advantage over trucks.
And finally, to get a sense of the true upside in earnings power, I think a look at the replacement cost of this asset is far more useful. At the current price, you are effectively paying $1.35 million per mile of track. There are three costs to recreating a railroad: 1) cost of labor, 2) cost of material, 3) the “right of way” or real estate underneath the track. The cost of #1 and #2 today are approximately $2 million per mile or approximately 48% higher than you are paying for BNI. The cost of #3 is inestimable. Buffett often talks about a moat by illustrating the money it would cost to recreate Coca-Cola. This is better. You actually can recreate Coca-cola, you would just earn a miserable return on your capital that was required to do so. You can NOT recreate
A useful comp to consider is the situation with D,M&E who tried to build a new 262 mile track to haul coal in the
Everyone knows Buffett has been buying; the other important buyer is the Company itself. In 2004, the Company generated $782 million from cash flow from operations less investing. Of this, it spent $376 million repurchasing stock and $231 million in dividends. In 2005, the numbers were $586 of cash generated, $789 million spent buying back stock and $267 million on dividends. In 2006, BNI generated $1.022 billion in cash flow from operations less investing, and spent $1.040 returning cash to shareholders. Since 1997, the Company has repurchased over 160 million shares compared with 358 million fully diluted outstanding today. Even with these repurchases, the Company remains significantly underlevered today, particularly when you consider that the deferred tax liability on the balance sheet ($8.4 billion of $22.7 billion of total liabilities) will never have to be repaid.
When the recession fully rears its ugly head and the world realizes that
|Entry||12/05/2007 02:45 PM|
|Thanks for writing this up. I've been nosing around the railroad sector since Buffett announced this position.|
I wonder if you can provide some insight on some of the anecdotes I heard about this company. Buffett and Munger and many other observers have noted that the railroad industry has historically not been a very good one for shareholders. Fixed costs and debt were high, labor was unionized and relatively expensive, etc. However, something changed 4-5 years ago to propel earnings growth. Recent changes he mentioned were technical innovations leading to more efficiency, and the advent of double deck rail cars which increase volumes. He cited these as drivers of recently increased profits. I didn't see you mention these, can you comment? Clearly there has to be some source of substantial earnings growth for the valuation to make sense as a buy, and until recently earnings did not grow much. So something has changed in the past 4-5 years to allow this business to grow.
Aren't high oil and energy prices implicit to this thesis? Rail has a big advantage over trucking when diesel prices are high, so the converse must also be true (i.e. less advantage when diesel is cheap). BNI is also fairly exposed to coal. Though I believe it is unlikely, what happens if energy prices moderate?
This is kind of speculative, but do you have any thoughts on why the stock price has remained relatively flat even though everyone knows that Buffett has been buying this for 6 months? Or to put it another way, what is the bear or sell-side case?
Thanks for your points on the "hidden assets", I had not considered that before.
|Entry||12/05/2007 03:17 PM|
|Any thoughts on the STB reconsidering how it calculates cost of capital? I think once the contracts reprice, BNI's ROIC will be too high if the STB follows through with the proposed change, which might lead them to cap allowable rates. Any view on how this might play out?|
|Entry||12/05/2007 04:03 PM|
|There are various pieces of rail-related legislation winding through congress, some of which could put a lid on pricing power:|
There are other pieces of legislation that are favorable towards railroads.
What is your opinion about the net effect on the railroads of all this legislative activity?
|Subject||Speaking of Bear or sell-side|
|Entry||12/05/2007 06:20 PM|
|Speaking of Bear or sell-side case, I actually recommend taking a read of the bear stearns industry primer - that is one of the better compilations of industry history/information I've read. I only wish I had read it 3 or 4 years ago when he began writing them.|
To answer your first question, Charlie mentioned that specifically at the wesco meeting re: double stacking. When I spoke to a trucking company, he mentioned that double-stacking has been around for 20 years. Perhaps the advantage has just become more obvious as gas prices have increased over that time frame, which seems reasonable given you can have one train or 240 individual truck drivers. But I found it interesting that everyone quotes that and it's certainly not new.
I think the answer to your question is more simply the supply/demand. This is a classic commodity pricing model where supply finally meets demand (2004) after two decades of cuts, demand kicks up a little and you generate a ton of free cash flow. It's no different than what's going in the merchant power industry, except in this case, no one else can build a coal-fired plant nearby to add capacity to the grid. Intermodal has been fueled by rapid asian growth in the past 6 years which has pushed the west coast ports to capacity such that Slim is trying to build one in mexico. Industrial/building products is obvious and probably on the decline for a while as homebuilding slows.
The CFO did mention little things that were interesting at a recent conference - one being that their productivity has nearly tripled in test cases in their industrial business by simply giving a 'discount' to customers for unloading or loading a train in a tight time frame. So utilization has been up as well.
But overall, its a capacity issue.
I see your point, but I meant to imply that higher oil prices provide upside and not required. I can't find anyone willing to tell me the formula that goes into the "rule of thumb" of 500 miles, and so I don't know the breakeven point but the common sense rule would suggest to me that gas has to be nearly free before sending something from LA to Chicago is cheaper via a truck. I don't believe we're anywhere near the zone where the converse kicks in. If you find the answer out there, I'd love for you to share that one (ie the change in cost per mile given a change in price of gas)
Energy pricing moderating I'm not sure is an issue - it's why energy prices moderate. Electricity grows at about 2% per year and I don't think that will change. Electricity is priced on a dispatch curve where the marginal producer who can fire and earn zero gross margin to meet demand sets the market clearing price. Basically over the past 7 years, natural gas price has tripled to quadrupled while coal didn't. So coal plants are firing non-stop while the higher heat rate gas plants fire occassionally. The real issue to your question is what if gas prices get cut in half while coal prices increase which effectively shift the two on the dispatch curve given coal plants have a higher heat rate than gas plants. If that happened, generators would use a lot less coal and we'd have a problem. I don't know where gas prices are going, but I do know its more scarce than the 275 years or so of coal reserves that exist. I also know that 50% of electricity generated is via coal and 20% is via gas. And given the scarcity factor, even if gas prices dropped by 80% causing the switch in the dispatch curve, people would build a ton of gas plants (and if calpine financed them with debt again, they would go bankrupt again) and gas prices would end up back where they are. As much as people complain about the emissions, I just don't think coal-fired plants are going anywhere for a long time given the abundance of coal.
And to your last point, I think its simply the classic short term nature of the street. If you're a sell-side analyst, you work for people who have their performance measured annually (or less) and so you have to make calls regarding that time frame. And no one facing that incentive bias is going to suggest that freight volumes are going to be robust in the next year. The other "problem" is that I've learned that sell-side guys tend to determine the P/E multiple based on historical trading ranges which literally makes no sense. And so historically, the multiple has been lower, so WHY wouldnt it revert it to that (rather than thinking about what the multiple means??) And so you have near term headwinds in volumes and an above average multiple relative to historical = "hold" or "peer perform". Its not a reason to make an investment, but I actually think there will be multiple expansion as well once earnings are revealed in bad times and the pricing power to offset volume declines will be displayed. I haven't seen any of them discuss the under-levered balance sheet or the deferred tax liability either - will be interesting to see if they do a massive share repurchase (i'm sure buffett is whispering that) and multiple expansion along with that. You could have more than a double with the reprices in capital intensive, unionized railroad.
Another hidden asset is the international intermodal contracts FYI. They are one to five years, and though un-confirmed, I'm fairly certain theres one or two big ones being marked to market as we speak as well.
Hope that begins at least to answer your questions.
|Subject||I really dont to be honest. T|
|Entry||12/05/2007 06:32 PM|
|I really dont to be honest. That's actually the main risk to this investment which is ironic. If you make too much money it will be taken away. I got a little comfort reading some old articles and such I found regarding the safety concerns and unreliable nature, in terms of being on time, over the past 50 years. It's actually astounding the problems they caused both communities with accidents and problems to customers/end-users. This business is truly a tax on economic activity in the country. Almost everything you did today, you paid the railroad. And so with that in mind, reregulation (or so the rails obviously claim) will revert back to less capital being spent and less safety/less reliability etc.|
It's a fair argument and I like it better than the opposition which is 'they make too much money'. And as much as they are complaining about coal transportation costs, de-regulated power generators should be careful where they cast stones.
I also think it's interesting that we earn a decent (currently) return on capital at $1.35 million per mile, I'd love to see the projections (and pricing) on what DME submitted to the gov't for the federally subsidized loan that was going to earn them a return on 14 or 22 million per mile. I'm pretty sure its more.
So in a long-winded way, no I don't have any real thoughts on whether or not STB will change the calculation - or if re-regulation will or won't materialize. It's a risk to the upside for sure, but one worth taking at the current price I believe particularly since the current returns on invested capital are hardly worth complaining about (and this is the first time they are making any real money).
|Subject||Sorry - much like last reply,|
|Entry||12/05/2007 06:41 PM|
|Sorry - much like last reply, I don't have any particular or unusual insight into the legislative outlook. I think of it more like that pharmaceutical industry. There will always be people complaining, but the service they provide has positive effects on way too many people for politicians (who are funded by the companies and elected by those who receive the benefits) to ever really do anything about it.|
It's certainly a risk. But you're not paying 20 or 25x which would require significant growth for you to earn a decent return. You're paying 16.5x, and even less if you consider the coal, gas and intermodal re-pricing (and the incremental return provided by repurchasing shares at a lower implied multiple and the stock re-trading at 16.5 on the new earnings base).
|Entry||12/05/2007 06:50 PM|
|Thank you for very well thought out replies.|
It is interesting that Berkshire owns McLane yet invests in BNI. To some extent I am sure that they serve different markets. McLane seems primarily point-to-point while BNI is long haul. In long haul, trucking uses free highways, so is probably economical at some fuel prices, but recently they are suffering. Also some products are not feasible to ship by truck such as coal, and railroads with locations in the right places will have a permanent advantage. I am generally of the opinion that fuel and energy prices are high and are going to stay there. This should benefit the railroads.
I'd like to focus however on the recent trends towards earnings growth at BNI. At 16x earnings and 16x free cash flow, there has to be some pretty predictable/reliable earnings growth in the near future for this to work out. Here is where Buffett must have some insights, because it isn't obvious (to me) that the earnings growth of the past 5 years is going to continue for 5 more.
As others have pointed out, some portion of earnings growth is inherently offset by regulatory changes and caps, which is one of the problems with railroads as a long term investment. So there must be something else at work.
Reading about WAB helps to understand the technology and process changes at work in the rail industry. Not only technology changes improving efficiency, but process and fulfillment services similar to what UPS has started doing. It doesn't jump out at me how a rail company can help a client manage their inventory and fulfillment, but that would be an interesting trend.
But if it isn't this and it isn't increased volume from things like double-stack cars (which you say have been around for 20 years), what is driving the earnings growth exactly? One answer is the end of a period of price wars and consolidation which, like BUD, allows the dominant players to enjoy their pricing power and market position for a while until competitors make another run for their gold.
|Entry||12/05/2007 08:51 PM|
|very thoughtful - thank you.|
|Subject||yesterday's STB meeting a big|
|Entry||12/05/2007 08:54 PM|
|I suggest that those interested in the re-regulation debate listen to yesterday's hearing on the cost of capital. I thought the meeting was a big positive for railroads because it looks like the STB will at some point have a hearing on return on capital methods (which I think is far more important than the current cost of capital issue being debated). The arguments to move away from historic book value make a lot of sense (especially the testimony by Atticus Capital regarding marking assets to market under purchase accounting).|
|Entry||12/06/2007 08:30 AM|
|Here is part of CSX’s testimony (from two days ago) that talks about the merits of not using GAAP book values in order to determine revenue adequacy. If the STB changes it methodology on calculating return on capital, it will be a major positive for the rails… allowing them to raise prices substantially on regulated business:|
“A second bridge over the West Pascagoula River provides another example. Rebuilt following Hurricane Katrina for almost $15 million, that bridge was built in 1903 and expanded in 1917 for a total original cost of less than $130 thousand. There are over 15,000 active bridges, tunnels, and overpasses on the CSXT network. The above example highlights a collateral consequence of using historical costs instead of replacement costs as part of the revenue adequacy determination. The Board’s proposal to change the regulatory cost of capital methodology without considering the investment base to which cost of capital is applied will lead inevitably to constrained capital spending at a time when the U.S. is facing a transportation capacity crisis.”
|Entry||12/06/2007 12:34 PM|
|Thanks for the excellent write-up. I've been debating the merits of BNI vs. UNP - how did you choose BNI? It seems to me that with BNI you get more capacity for future volume growth and a lower multiple on consensus forecast vs. UNP's more significant opportunities for re-pricing and productivity. With UNP it seems as though you get more protection in a weak volume environment vs. BNI's long-term growth story. I think UNP's legacy of mismanagement and resulting lower ROC may also shield them from the regulation/capped pricing issue. Thanks in advance.|
|Subject||I'm sure UNP will be do well t|
|Entry||12/06/2007 01:27 PM|
|I'm sure UNP will be do well too but they are more leveraged to cyclical end users (BNI is best of all four by a decent margin) which I think is key to the downside protection. And when you talk to industry insiders/customers, they all told me that BNI was by far the best in terms of service in the intermodal business which I think will be the key growth driver to come. |
Additionally, UNP didn't seem to buy back any stock if I remember correctly. They paid down debt instead which is a fine use of cash, but I'd rather go with the guy who instead ISSUED debt to buy-in stock with a higher yield during the same time frame.
And BNI was cheaper. Though if I had gone with UNP, I'd be doing better!
Finally, and as technical as this sounds, UNP is based in Omaha. Berkshire is based in Omaha. If they aren't friends, they definitely know each other. And Buffett has since sold UNP to buy a ton more BNI, which I like better than the reverse.
I don't want to waiver from my steadfast "I dont know" re: the legislation. But i'm not sure i agree with your last statement. You might be right, but it's hard to imagine gov't regulating one guy in a two-man show.
|Entry||12/06/2007 01:52 PM|
|Great writeup - I was 2 days away from posting the same idea. I think you covered all the major points as BNI's attractiveness is its network infrastructure (moat), strong end markets, and shareholder friendly management. I think the big mis perception is that the business model is fundamentally different from 10 years ago. This difference relates to a focus on doing contracts based on returns on capital compared to network utilization. The effect is that contracts today are based on cost + margin + annual escalators + fuel surcharges. |
The significance of this structure becomes evident when one considers what effect the change in the way cost of capital is calculated.
To summarize the argument, the STB (main regulatory agency) has argued that they should change the way the cost of equity is calculated and use a CAPM approach rather than a DCF based approach. In doing so, the assumptions they have made reduces the cost of capital from north of 12% for the rails to around 7% last year. Although this is a ridiculous number to begin with, no one has articulated where this comes into play. In essence, this calculation is only used when a shipper contests a tariff rate (spot rate) and is used to determine what the revenue adequate number should be for a fair price. Thus, any signed contracts are not eligible for STB rulings and therefore for the majority of BNI's business pricing will not be affected regardless of the decision. Furthermore, revenue adequacy only relates to lines where there are no alternatives - obviously not the port of Los Angeles (UNP), coal from the PRB, etc. Therefore, although any regulation is a negative, the bottom line is that pricing should be stable going forward.
|Subject||BNI vs. UNP|
|Entry||12/06/2007 02:58 PM|
|Murphy -- you seem to have a good understanding of the merits of owning BNI vs. UNP... here are my thoughts (I own both).|
At the same p/e based on "normalized" earnings, I would rather own BNI. I believe BNI will grow faster (from a normal level) over the long-run due to (1) more intermodal exposure (the "transcon"), (2) more exposure to PNW exports of grain, and (3) a pure play in coal in the PRB (UNP has some non-PRB business).
That being said, UNP is under-earning versus BNI (due to more legacy contracts and poor operations) and UNP's operating margin should at some point be lower than BNI's (due to mix of business).
At these prices (UNP has moved up a lot recently), I think the merits or owning BNI or UNP are about the same.
|Entry||12/06/2007 03:29 PM|
|I meant to say that UNP's operating margin should be higher than BNI's at some point.|
|Entry||12/06/2007 04:11 PM|
|Thank you for the excellent writeup.|
Don't mean to go astray, but I have been looking at ACLI (leading player in inland U.S. barge shipping) and as this is in the same sandbox, I am interested in hearing anyone's thoughts on barging. See heffer504's writeup from a few years ago for background on ACLI and the industry.
When compared to trucking, barge is even further to the left on the slow/cheap/fuel efficient spectrum than rail. According to an ACLI presentation, cost iis 0.72 cents per ton/mile for barge, 2.24 cents ton/mile rail, 26.61 cents ton mile for truck, with a good chunk of that from fuel.
The situation shares some characteristics with rail:
Consolidation - The industry has consolidated heavily in the last five years, with the top 5 players in both liquid and dry cargo going frmo 32% to roughly 66% of industry.
Decreasing capacity - Due to a 25 year life-span on barges and a huge tax-break induced construction boom from 1979-1981, industry capacity is coming off line quickly, and the two builders are booked out for sevearl years. Rates have spiked in the past few years and the pricing on those is still working through.
Savvy ownership - In terms of ACLI, Zell owns 25% ($216m) and just upped his stake by $12m. Having taken the company into and out of bankruptcy, he knows it well.
I don't know the shipping sector all that well and am unsure to what degree the rate spike of the last few years was due to a one-time boom vs. more sustained based on longer-term favorable supply/demand conditions.
One final related0-topic note - RAIL (coal car construction) has gotten creamed lately - $400m market cap and $225 EV, ex-$50m pension liability.
|Subject||CSX vs. BNI?|
|Entry||12/10/2007 12:10 PM|
|Thanks for the write-up. I am not as familiar with the industry as you, but it seems to me that the overall thesis works for all four class-I railroads in the US. If so, could you talk a bit about CSX? This is one of the worst class-I railroads in North America (including the Canadian railroads) in terms of margins, safety records, and operating metrics. So it could be thought of as a turnaround play with a visible catalyst in place.|
This catalyst is Chris Hohn, from The Children’s Investment Fund. I believe they currently manage around 10bn EUR and the word is that close to 25% of their NAV is in US rails (CSX and UP, I believe). They have set up an activist site for CSX at www.strongercsx.com and have been exchanging sharply worded letters with CSX’s board of directors (Atticus’ presence at the STB hearings isn’t surprising given their close working relationship with TCI).
Hohn seems to think that CSX’s margins could improve dramatically, and he likes to look at Canadian National Railway as the “model” of what a modern railroad should look like (incidentally, at the 2007 Berkshire Hathaway meeting, Buffett pointed out that Bill Gates has owned Canadian National for years because he saw the thesis before Buffett, joking that “maybe we should have Bill run Berkshire’s investment portfolio”).
If Hohn is right and one can bring CSX up to CN’s margins, this could be a great investment. If he’s wrong, you still get contract price escalations and the benefits of the tailwinds this industry is currently experiencing.
CSX also counts Dan Loeb, Carl Icahn, and Atticus as shareholders. Any thoughts?