Standing In The Shoes Of The Oracle From Omaha: American Express

Most of us are familiar with the Salad Oil scandal of 1963 and the repercussions that followed. For the sake of the few that may not be aware of the Salad Oil scandal, I will give a brief background belowbefore trying to think like Warren Buffett.

Background of the Scandal

There was a fellow by the name of Anthony De Angelis who had started a company in 1955 to exploit the U.S Government’s Food and Peace program that aimed at selling surplus food to Europe for cheap. The company’s name was Allied Crude Vegetable Oil Refining Co. and De Angelis had been successful (for a scam artist). By 1962 he was a large player in the market and thought he could make more money by attempting to corner the market.

Cornering the market consisted of attempting to buy up all, or most of the current capacities supply (through futures), essentially creating an artificial demand, in turn driving prices of both current soybean oil inventories up (that he owned) and the future contracts (that he bought), for a large profit. De Angelis planned (and succeeded) to also use these inflated inventory values to further finance additional companies, through collateralized loans (against his current inventory).

American Express around the same time was entering into the warehousing business and had a specialized unit that made loans to businesses based on inventories. Essentially, De Angelis took these receipts from American Express to a bank or broker and had them discounted for cash. He soon realized this was a very easy way to make money and began falsifying inventory, using seawater.

American express was also duped when inspecting the warehouse. Elementary science teaches us oil is more viscous than water thus it sits on top of the water. When the inspectors checked the tanks their instruments only inspected the top of the tanks and some tanks also had pipes to transfer the oil between one another.

In the end things blew up, but not before De Angelis could secure loans from 51 businesses (ranging from P&G to Bank of America). After logistical mistakes and rumors of bribery, American Express was eventually tipped off. Over the weekend of November 15-18th, 1963 the salad oil scandal broke as American Express realized that Allied Crude did not have the reported 150 million in vegetable oil but instead had only 6 million. On November 19th, Allied Crude reported bankruptcy as the soybean futures crashed erasing the remaining value of the loans.

[Soybean futures below]

Unknown Warren Buffett (TradesPortfolio) Enters the Scene

The following Friday, Nov. 22, 1963 President Kennedy was assassinated and the market fell 3% but rebounded the following week. American Express stock was not so lucky, and was caught in a free fall that consisted of the shares falling from $65 a share in November 1963 to $37 a share in January 1964 or a 43% decline in less than 90 days. Buffett enters and begins buying shares of American Express between 1964 and 1966, eventually spending 13 million or 40% of the partnerships money. Buffett, by simply reading the 1963 and 1964 annual reports would have got a clear indication that American Express was actually not legally on the hook for the receipts but “morally obliged” to make good on the warehouse receipts.

[Dow Jones during Salad Oil Scandal and JFK Assassination]

In the 1964 annual report (link at the bottom) on pages 18-19 it is explained that the payment will not exceed 45 million and that certain insurance companies were actually subject to litigation to make good on their policies to pay for the losses, possibly further lowering the eventual payout. It is also explained and shown that this mishap did not material impact the business, as their main lines of business hit all time highs a year after the scandal broke.

[American Express Share Price 1959-1973, split adjusted]

Why Did Buffett Buy…? So Much? 

Now for the more exhilarating part, trying to understand what Buffett was thinking while he was buying and the valuation behind it. Lets estimate the average price of common shares for WB was $50. Combining that estimate with approximately 4.461 million outstanding shares (1964 annual report) we find an estimated market capitalization of 223 million.

Doing some other quick back-of-the-envelope calculations using the 1964 annual report we can compute the following:

  • Buffett paid roughly 17.79x, 1964 earnings and 2x, 1994 revenue.
  • Traveler Cheques and Travelers Letter of Credit had 10-year CAGR of 7%
  • Customer Deposits and Credit Balances had CAGR of 4.5%
  • 10-year revenue had CAGR of 10.89%
  • 10-year net income had CAGR of 8.76%
  • 2.8% dividend yield
  • 10x, 1964 EBIT and 16.9% operating margins

We get the general idea at $50 a share it was roughly fairly valued or a little undervalued depending on the valuation methodology (23% lower than where it was before the scandal broke). Why would Buffett put 40% of the partnerships money into something that was not trading at a large discount, at least a 50% discount to intrinsic value? American Express is the investment I presume, that Warren Buffett (TradesPortfolio) discovered the powers of the float, as he went on a “float” buying spree shortly after.

We see that American Express had 263.8 million in cash, 515.6 million in security investments at market value and a few other operational assets.

On the liabilities side we see two large numbers that stick out like a soar thumb, Travelers Cheques and Customer Deposits and Credit Balances of 525 million and 387.6 million respectively.

Now relating the idea back to the idea of floats and competitive advantages, if the brand of American Express is trust and loyalty, and that brand is not tainted, then we can assume the growth will continue in the float, bottom line and top line.

Lets assume 5.95% CAGR on the combined liabilities (using 0.58 and 0.48 weighting on growth outlined above). Now are the “liabilities” truly liabilities? Well what if the liabilities that come due are offset by the new liabilities that are taken on, what we will call using Peter to pay Paul. As long as there is no “run” on American Express over trust issues, we could assume that this is nearly 1 billion dollars of encumbered value or borrowed for free, which in turn we can invest in securities and earn a spread. American Express sold pieces of paper in exchange for actual money, sometimes the paper was not redeemed for long periods of time, sometimes never.

Reading the annual report we also find out that $45 million would be the maximum American Express would be willing to pay even though it was not legally obligated. We see that that can be easily covered with liquid assets.

Well when we look at it from that standpoint, what price would you pay for a perpetually growing zero-coupon bond? Close to nothing would be my bet, hence the free (or almost free) cost of an enduring float.

Well when that debt is eliminated from the equation, we end up with 983 million in shareholder equity or a 78% discount from intrinsic value (using a 223 million market cap).

Now lets assign some values to the outcome probabilities, over a course of 5 years.

5% chance of 80% capital impairment
40% chance of 100% gain
25% chance of 150% gain
20% chance of 200% gain
10% chance of 300% gain (98% of intrinsic value is achieved)

143% expected return or 28.6% annually for the next 5-years.

Things ended up better for Buffett who realized a gain of 154% over 3-4 years, buying at an estimated price of $50 a share and selling at an estimated share price of $126.5 turning an initial investment of 13 million into 33 million. This ended up being a 26-36% CAGR investment depending on n=3 or n=4.

Selling Too Early….. A Mistake Even the Best Make 

In the end Warren sold his investment too early and ended up buying it back at much higher prices missing out on a large portion of the returns and the steady stream of dividends to come. What he learned though from this experience likely became invaluable to him, as he learned how inefficient the market truly could be during unfavourable news, the power of a brand and enduring float, and the power of buying and holding great companies. After painfully selling his first investment early at 11 years old, cities service preferred stock soared and it happened again over the years with various other holdings, including American Express.

Warren came to the following realizations:

When we own portions of outstanding businesses with outstanding managements, our favorite holding period is forever.  We are just the opposite of those who hurry to sell and book profits when companies perform well but who tenaciously hang on to businesses that disappoint. Peter Lynch aptly likens such behavior to cutting the flowers and watering the weeds.

You should be fully aware of one attitude Charlie and I share that hurts our financial performance: We are very reluctant to sell sub-par businesses as long as we expect them to generate at least some cash and as long as we feel good about their managers and labor relations. Gin rummy managerial behavior (discard your least promising business at each turn) is not our style. We would rather have our overall results penalized a bit than engage in that kind of behavior.” 

He later bought American Express preferred stock in 1992 but sold out before the next reporting period. It was not until 1994 that Warren repurchased his shares of American Express, for the 3rd time in his life; he had the following to say.

“Our history with American Express goes way back and, in fact, fits the pattern of my pulling current investment decisions out of past associations. My American Express history includes a couple of episodes:

In the mid-1960’s, just after the stock was battered by the company’s infamous salad-oil scandal, we put about 40% of Buffett Partnership Ltd.’s capital into the stock – the largest investment the partnership had ever made.  I should add that this commitment gave us over 5% ownership in Amex at a cost of $13 million.

As I write this, we own just under 10%, which has cost us $1.36 billion.  (Amex earned $12.5 million in 1964 and $1.4 billion in 1994.) My history with Amex’s IDS unit, which today contributes about a third of the earnings of the company, goes back even further.

I first purchased stock in IDS in 1953 when it was growing rapidly and selling at a price-earnings ratio of only 3.  (There was a lot of low-hanging fruit in those days.)  I even produced a long report – do I ever write a short one? – on the company that I sold for $1 through an ad in the Wall Street Journal.

(Note: IDS was acquired by American Express in 1984 and spun into a stand-alone company, Ameriprise Financial. Ticker: AMP)

Obviously American Express and IDS (recently renamed American Express Financial Advisors) are far different operations today from what they were then.  Nevertheless, I find that a long-term familiarity with a company and its products is often helpful in evaluating it.”

The last part above is very crucial, as many of his investments embody his personal nostalgia. The next famous investment, which I will cover in my next post, is Walt Disney, after Warren met him personally. I highly recommend taking a look through the 1964 annual report posted below. You will find things have clearly changed, as the pages went from 32 in 1964 to 188 in 2012, or a 3.75% CAGR. Enjoy!

American Express 1964 Annual Report



2 thoughts on “Standing In The Shoes Of The Oracle From Omaha: American Express

  1. Hi Tannor

    Great post! I learned quite a lot. I just had a quick question. How did you calculate the intrinsic value of $983 mm?


    • I hope this helps as the question was asked as well as talked about below and in the comments section of the article.

      Workhorse_investor – 4 weeks ago
      Regarding floats, would you differentiate customer deposits from other forms of float (ie insurance premiums)? Deposits are not completely “free” for banks/financial services companies as they are interest bearing: today, AXP has about $40Billion worth and I imagine it pays out interest for most of those deposits–a paltry 0.85% for a savings account currently, but likely higher rates for the brokered deposits.
      Let me know what you think and keep up the articles.
      TannorPTannorP premium member – 4 weeks ago – Edit

      I do definitely differentiate between various “floats,” particularly by how enduring it is, the cost of obtaining the float and what is done with it. The customer deposits would be interest bearing while the traveler cheques would not (now-a-days TCs are not very important). It looks like American Express pays 1.3% on average (2012, 10-K) on their customer deposits of 40B. Taking that and loaning it to card members at 7.5% and 9.1% produces a spread of 6.2% and 7.8%, as you would have already known. Another way of phrasing this is that the customer deposits cost AP -6.2 to -7.8%, well below the risk free rate.

      Now if the input of this spread is enduring and continually growing (identifying enduring is the tricky part) we can assume that this is an unencumbered source of value or invisible equity. Essentially new customer deposits or inflows will offset redemptions or outflows as most customer deposits are revolving in nature (I do not take my money and deposit it at a new bank or financial institution each year). So this “short-term” debt is not short-term at all, especially if it is growing, it can be used to finance business operations.

      “Net cash used in financing activities of $0.7 billion decreased $7.2 billion compared to $7.9 billion in 2010, due to increases in customer deposits and issuances of long-term debt during 2011 as compared to 2010, partially offset by increases in principal payments on long-term debt and repurchases of common shares and a decrease in short-term borrowings in 2011.” (2.3 Billion was customer deposit in-flows in 2012)

      Would you like to collect a spread of 5-8% on money that you pay nothing for?
      Is this something you would rather have or live without?

      Well this is one of the main reasons financial institutions aggressively promote and market various accounts, with the sole purpose of you depositing your money with them. Customer deposits fuel loan books.
      Insurance is different and depends on the long-term combined ratio, or the cost of the float and risk management. But most importantly I look at the three sources of financing: Debt, Equity or Float, float being the best alternative until the cost of equity or the cost of debt is exceeded. Competitors drive the cost of floats up. Monitoring the float could give you insight into operations and how effectively the company is managing market share.

      To answer your question, no the float is not free in most cases but if it is allocated effectively, is enduring and cost less than the risk free rate, it may be a competitive advantage.

      Note: In 1964 (non-interest bearing) TCs were a much larger component of APs balance sheet and business. Hope that helps and thanks for the question.

      20punches20punches premium member – 4 weeks ago
      Great article, Tannor. Didn’t know when Buffett purchased AMEX, it was 17 times earnings. That’s very interesting.
      Quick question on the following statement you made:
      “Well when that debt is eliminated from the equation, we end up with 983 million in shareholder equity or a 78% discount from intrinsic value (using a 223 million market cap).”
      In this context, what equation are you refering to and how did you end up with 983 million in shareholder equity?
      TannorPTannorP premium member – 4 weeks ago – Edit
      Thanks Ning,


The 17x is based off 1964 earnings and a subjective estimate of the market cap and buy price. You could also end up with 15-18x depending on the numbers used for the numerator and denominator. EBIT would be much better for a financial companies.

      The 983 million is (Shareholder Equity + TCs + Customer Deposits)

      I am unsure when it comes to customer deposits if they are a short-term liability as they are revolving in nature and usually growing, although they are interest bearing and have no maturity. The TCs are non-interest bearing, revolving and growing, essentially zero coupon bonds with a far off maturity. Essentially I am characterizing TCs and Customer Deposits as invisible liabilities and removed them. We could also remove only TCs and end up with 595 million equity.


Lets say we can earn 7% and pay 1% to obtain the billion. This is $60 million net a year produced from 0 equity. It is like a ponzi because the liability will never come due (as long as there is not a run) but produces cash flow for investors in the mean time. This type of business is worth a lot more alive than dead and the unencumbered source of value it produces. I am not sure if this is the correct way to think about it and would be happy to hear what others have to say.



      Josh ZachariahJosh Zachariah premium member – 4 weeks ago
      Great article Tanner and thanks for linking the AmEx report. But I think you’re mistaken on the issue of TCs and equity. Whether or not the TCs and deposits pay interest they’re still a claim on the business. For the sake of investors, debt whether it pays interest or not is senior to equity and in the event of bankruptcy those claimants will get paid first. Bondholders, depositors and TC holders have a specific claim on the business, shareholders are the residuals. Further, to lump the depositors and travelors checks holders with equity holders would be giving a false impression on the amount of leverage.
      I recall a couple years back a related case with Circuity City, Borders or some otherretailer that had gone bankrupt. The issue was the matter of gift cards (store credit). Bond holders thought holders of gift cards should be treated equally with senior creditors. In fact the court ruled that gift card holders would get paid before the bondholders which naturally makes sense. Imagine for a moment these gift card holders being treated on par with bond holders or for that matter the same as equity holders.
      AmEx’s solvency didn’t seem threatened in this case, but the company would certainly never consider their traveler’s check holders and deposits as the same as shareholder capital, nor would investors
      TannorPTannorP premium member – 3 weeks ago – Edit
      Thanks for the kind remarks and great comment Josh, 

      I definitely agree with what you are saying and am unsure how various accounts that are labeled liabilities but could be characterized as assets or some form thereof (while the company is solvent) should be labeled or valued. I understand it is not truly equity but neither is indefinitely deferred taxes, but one could view it as a portion of government equity, until closed.

      I cannot understand why Buffett would put 40% of his partnership into American Express under those valuation metrics unless there is something hidden within the liabilities (i.e. TCs and customer deposits).

      I had no clue about the claims on store credit during bankruptcy but do remember reading in an annual report, I think Sherwin-Williams or Wal-Mart, that after 4-5 years of unclaimed deposits it is very probable or very likely (I can’t remember the exact statistic or language) that the deposits would never be claimed. These claims go on to be stated as a short-term liability indefinitely although they may never be claimed. Accounting for the time value of money, each $1 given to me today should be worth at least $4 within 15 years, thus each $1 has produced $3 of equity while remaining as a $1 short-term liability. Could we not rephrase using revolving and growing TCs and deposits? (albeit when the business is solvent)

      How would we treat such an account?

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s