In the summers of 2007 and 2008—when talking about oil at $150 per barrel was a real thing—Warren Buffett sunk $7 billion from Berkshire’s Omaha Treasury into Houston’s Conoco Phillips, the third largest American integrated oil company at the time. It was, at the time, the largest amount of money that Warren Buffett had sunk into a publicly traded business. Those large $10+ billion stakes in American Express and Coca-Cola are largely the result of capital appreciation, with Buffett only sinking $1.28 billion into American Express and $1.29 billion into Coca-Cola, respectively. Buffett’s commitment to Wells Fargo ($11.8 billion of Berkshire’s money invested) and to IBM (at least $11.6 billion and counting of Berkshire’s money invested) have been larger investments in recent times, and full-business acquisitions like Heinz and Gen Re also consumed more of Berkshire’s cash than the Conoco investment.
Buffett bought his Conoco stock at a price of $85 per share and sold shortly thereafter. Someone might be tempted to think, “Gee, the price of the stock is currently $67 per share—good thing he got out of there when he did.” That analysis would not be correct because it ignores the effects of dividends and the Phillips 66 spinoff. Phillips 66 is at $69 per share, and given that you collected half a share of Phillips 66 for every share of Conoco you owned, the value is $67+$39.50 plus dividends. In other words, it is as if the stock is trading at $106.50 plus Conoco pays out a $2.92 dividend and Phillips 66 pays out a $2 dividend (but really the effect is halved when you adjust it) so you have to add up all the 2008, 2009, 2010, 2011, 2012, 2013, 2014, and 2015 payments as well.
In short, I’m not convinced that Buffett’s decision to sell Conoco stock going into the financial crisis was the correct thing to do: He bought at the high of a commodities cycle, and given that this measuring period involves a low of the commodities cycle, it is still impressive to me that we are talking about a compounding rate in the 8% annual range if you assumed full reinvestment over the time frame.
Given that the timing of the purchase price through our current measuring period does oil stock investing no favors (because we are comparing oil’s returns at high valuation through a period of lowish valuations), it tells you something about the compounding engines of energy giants of this caliber that you’re still building wealth at a decent rate even over these comparison periods.
That is my interlude way of saying that just because Warren Buffett decides to sell something does not mean that it is a bad investment going forward. He sold Disney. He sold Guinness (now Diageo). He sold McDonald’s. These are all companies that proved very lucrative if you invested on the day that Buffett left the company. Heck, he even sold Disney and American Express in the 1960s, something that would have made one of the richest men anyway even if he just sat on either of those two blocks of stock and held on to it.
Now that we have gone full circle, we talk about the report that Warren Buffett has sold ExxonMobil. First of all, we should examine the basis by which people have reached that conclusion. Media observers and investors looked at Berkshire’s 13F filing, noted the absence of ExxonMobil, and presumed that Buffett sold because Exxon was not listed as an ownership position. This does not, in fact, guarantee that Buffett sold Berkshire’s position in Exxon stock. It could mean that he is increasing Berkshire’s position in Exxon and has received special clearance from the SEC to stealthily build up his position beyond prying years.
This is a customary practice to give widely mimicked investors the opportunity to buy as much of a given stock before the rest of the market participants learn and possibly bid up the stock price. Or, Buffett could have simply sold the Exxon stake. The point is that no one knows, and that is by design.
It is important to make decisions based on what makes sense to *you*, and what fits your understanding. I have no doubt that if someone went to Computershare, clicked here at: “Welcome To ExxonMobil Shareholder Services At Investor Centre”, and arrange to have a couple hundred dollars regularly used to purchase shares of Exxon, you will own an unfettered compounding machine to the tune of 8-12% annually, with the catch being that you won’t be able to predict those returns on a year by year basis.
By unfettered, I mean there are no fees dragging you down. There is no 1% charge taking $1,000 on every $100,000 in net worth that you have. There are no deductions every time you make a purchase. The transaction costs on the purchase side are $0, and that lets you put your money directly to work in a highly productive asset that hasn’t missed a dividend payment since its formal incorporation in 1882.
It sells $382 billion worth of oils, natural gas, and chemicals per year. It is replacing its reserves at a 120% rate, with anything above 100% meaning that you are acquiring oil and natural gas to drill at a faster rate than you are actually drilling it. That is saying something, given that Exxon refines 4.6 million barrels per day and engages in 5.9 million barrels per day in product sales. There are $25 billion barrels of oil and oil equivalents in reserve. Cash flow per share has increased 13% annually over the past decade, even with lower price. That’s because the company buys back absurd blocks of stock, turning 6.9 billion shares in 2000 into 4.2 billion today (the actual buyback is more impressive given that Exxon had to issue some shares during this time to do things like acquire XTO Energy).
In short, I have no idea if Buffett sold his Exxon stake or is adding more to it, and my secondary point is that it is irrelevant to your analysis. Not just as a matter of logic, but also as an appeal to authority—with the exception of Fannie Mae, large American companies that Buffett sold in the past two decades have still gone on to deliver great returns. Exxon’s cash flow is enormous, the reserves are outstanding, the debt is small for a company of its size (about $21 billion or equal to about eight months profit), the production is high, and the dividend raise strike goes back over three decades and the consecutive dividend payment streak goes back to the late 1800s. It matters what you think of those facts, not what you speculate based on a supremely successful Omaha investor’s opportunity costs. I choose to follow Buffett often because the logic is persuasive, not because of the fact that he said it or did it.