Inside the Investments of Warren Buffett: Twenty Cases

I have included Texas National Petroleum Company as an investment to look at because it is one of the few cases of a workout that is explicitly discussed by Warren Buffett in his annual letters to shareholders. In effect, this was a merger arbitrage situation, which is of course still an area of focus for many investment funds that dabble in special situations investing.

In 1964, Texas National Petroleum Company was a relatively small producer of oil that was in the process of being acquired by Union Oil of California. To be more precise, Union Oil of California had already announced a formal offer with specific details of terms, but this had yet to be accepted by Texas National Petroleum. Hence the deal was announced but not completed.

As with any merger arbitrage deal, there are three factors that investors examining the company at the time would have taken into consideration. First, they would want to know the specific terms of the offer, for example, at what price, in which form, etc. Second, they would want to know the timelines involved, such as the stage that the merger is at and how many months the deal is expected to take until it is completed. Third, investors would want to understand what the risks are of the deal falling apart. This collapse could be due to regulatory approvals needed, shareholder approvals from both the acquirer and the acquired, or other specific stipulations specified in the terms of the offer.

To set the historical context, mergers and acquisitions (M&A) in the oil and gas industry in the southern and midwestern United States were quite common during the 1960s when domestic oil production was still in full swing, so this deal would not have been completely unusual in its context.1 Certainly, there would have been other similar deals where precedence could be drawn and confidence gained in such transactions. In fact, Union Oil of California was no stranger to acquiring companies, having acquired Wooley Petroleum in 1959, and going on to merge in 1965 with Pure Oil Company in one of the largest mergers within the oil industry at that time.²

Coming back to the case of Texas National Petroleum, there were three outstanding classes of securities at the time the acquisition was announced. Research had to be done on the company disclosures, but such information would have been relatively easy for an investor to find. Buffett also includes the details in his discussion of the investment in his annual letter to clients.

First, there were bonds outstanding that had a coupon payment at an annualized rate of 6.5 percent of the face value of the bond. These bonds could be redeemed by the company for a value of $104.25, which was the plan at the close of this acquisition. Also, in April 1963 when the deal was announced, this was a precoupon date, so an investor would have likely expected to receive a coupon payment during the time of the workout. In total $6.5 million of these bonds were outstanding. Second, there was common stock; 3.7 million shares were outstanding, and the estimated price they were to fetch in the deal was $7.42 per share. Forty percent of this was owned by inside investors, with the balance being held by outside investors. Third, there were 650,000 warrants outstanding, which gave holders the option to purchase the common stock at $3.50 per share. This means that at the estimated deal price of $7.42 for the common share, the warrants had an estimated value of $3.92 in the deal.

Unlike most merger deals today, no exact closing date of the merger was formally announced, as far as I was able to determine, so it is not known when the acquisition was to be completed. To get an approximate time of the estimated completion date, however, two sources of information seem sensible to check into. The first source is disclosures from the parties involved, i.e., information given by either Texas National Petroleum or Union Oil of California. The other source is inferring from other similar deals at the time. Regarding the former, the management of Texas National Petroleum did provide some information. In his letter to shareholders Buffett discusses a conversation with the management of Texas Petroleum in which Buffett Partnership Limited pushed for a completion of the deal by August or September of 1963. If the end of September was the projected time for the completion of the deal, this would have meant that Buffett considered making the investment five months after the April announcement.

Although investors would normally seek to understand a business’s inherent quality, in this case, inherent quality would only be relevant if the deal fell through and investors became holders of the company on a stand-alone basis. A fundamental assessment of the business would sensibly have involved understanding the quality of the oil or ore assets. As this was clearly not the focal point of the investment case, it is sensible to go straight to the valuation of the deal. After the announcement, the prices for the three categories of securities would have been approximately as follows:3

If one assumes a five-month period until the deal closes and perhaps another month until payment is made, a six-month total for the investment period, then the estimated returns based on the offer price estimates would be:

(a) For the bonds, an investor would receive coupon payments that would be adjusted to an annualized return of 6.5 percent. If one assumes that the deal takes six months to complete, then the coupon payment would amount to $3.25 in total. In addition to this, an investor would expect a gain of $5.47 ($104.25−$98.78). Hence the total absolute return would be a gain of $8.72. As a percentage of the purchase price this would be about nine percent or 18 percent annualized. This looks fairly attractive, so if an investor is quite certain about the deal going through, that investor would likely buy the bonds.

(b and c) For the common equity and the warrants that convert into equity, the calculation is simple. If one expects to make about 11 percent, which is the spread between the offer price and the prevailing price, in six months one would expect a return of 22 percent annualized. This would be a very attractive return. If the deal closes faster than expected, one would earn an even better rate of return. This is also true if the offer price were raised. Conversely, if the deal took significantly longer to complete, the return would be poorer on an annualized basis.

Overall, the spread in the price of the offer and the current price certainly would have looked attractive. The only remaining issue to consider would have been the risk of the deal not materializing. To assess this risk, one might first consider the chance that shareholders, who had yet to pass the acquisition, would not approve the deal. As the management of Texas National Petroleum led this deal-making effort and itself owned 40 percent of the shares outstanding, one can quickly conclude that shareholder approval was fairly likely. In fact, as long as the deal price seemed even remotely fair, an investor could be quite certain that the deal would be approved as only another 10 percent of the remaining votes were necessary for approval. Buffett reached a similar conclusion on the risk of shareholder approval.

Furthermore, one can look at legal approvals and any antitrust issues that may be present. On these issues, it is not clear what the risk is, but the fact that many similar M&A deals were completed in the previous decade, where smaller oil exploration companies were consolidated by the larger ones, would suggest that this is a fairly straightforward case. Hence, while investors certainly would have checked with expert lawyers at the time, this did not appear to be a significant issue.

Regarding these concerns, Buffett did his research thoroughly; he gives us a full report on any potential legal risks as well as the progress of key legal developments.5 Specifically, Buffett details that title searches and legal opinions on the deal had passed with few issues and that the only major hurdle was that a tax ruling was necessary related to the University of Southern California, which had nonprofit status and was the holder of some production payments at the time. While this was an additional hurdle that might have delayed some processes, Buffett judged it not to be a threat to the overall deal because USC had suggested that it was willing even to waive its eleemosynary status to help complete the deal.

In this case, the potential rewards were clear. It would have been difficult for an individual investor to accurately assess the risk of the deal the way that Buffett had. In fact, I would have turned to lawyers in the field and conducted primary research. For a small investment fund, this is doable as such funds generally have access to a network of specialists. The individual investor, however, would have to make an extra effort. In any case, if one could have been sure about the risks, as seemed to be the case for Buffett, one could have expected to profit handsomely from this special situation investment.

In Buffett’s investment case, he ended up investing in all three classes of securities, amassing debentures with a total face value of $260K, 60,035 shares of common stock, and 83,200 warrants to purchase common stock. While the deal ended up taking a bit longer to complete than expected (payout for the bonds was in mid-November, with payouts for equities and warrants in installments in December and early the following year), the overall payoff was also slightly higher than originally calculated (about $7.59 per share rather than $7.42). In light of this, Buffett commented: “This illustrates the usual pattern: (1) the deals take longer than originally projected; and (2) the payouts tend to average a little better than estimates. With TNP it took a couple of extra months, and we received a couple of extra percent.”