AmEx, unfortunately, had a warehousing operation that was basically unrelated to its main business. Unless you consider that "warehousing" was a form of verification, just as paying off travelers checks.
Tino de Angelis, a criminal, stored large amounts of what purported to be "salad oil" in American Express warehouses. (The salad oil was only on top, most of the liquid was water.) He then used the resulting warehouse receipts as collateral for loans; banks accepted the American Express guarantee like few others. DeAngelis squandered the proceeds, and when the banks asked for the collateral, the fraud was exposed, to the tune of $160 million.
The value the stock was cut nearly in half, from around $60 to the mid $30s. American Express offered to settle for what it could pay: its book value, of $60 million. It could continue operations from ongoing cash flows.
After some wrangling, the banks accepted. Better to accept $60 million up front than to try for $160 milllion: 1) running the gauntlet of litigation over liability and 2) trying to enforce a judgment for a large amount against what would be a bankrupt company. And some of those banks now make money selling American Express travelers' checks.
The move went over well with Amex's customers (though not its shareholders). One of those customers, who became a shareholder, was Warren Buffett. He figured that $60 million represented a "fair" settlement, based on the company's "ability to pay," but that the company was worth a multiple of that (its former book value), and treated it as a "special dividend." Because of his correct judgment of the company's cash generation abilities, he made five times his money in five years off the depressed base.
Over twenty years later, in the Pennzoil-Texaco Oil case, Warren Buffett bought a slug of Texaco bonds at a discount. He premised this investment on a "reasonable" ($2-3) billion settlement. (This was opposed to a verdict of $7.5 billion plus $3 billion damages, plus interest.) When the actual settlement came in at the high end of Buffett's range, he made good money.
How does this apply to BP? The maximum liability figure I've seen is just over $100 billion. That's about $50-$60 billion for cleanup, with the remainder for enforcement and penalties.
BP could well end up paying up to $50 billion (or a little more), for cleanup. That would put BP America on the verge of bankruptcy. But it would probably balk paying the remainder out of the pockets of the parent.
I'm not a lawyer, and I'll admit that I don't know much about the law, but I do know that laws and rulings are nullified, or overturned with considerable frequency, in order to obtain a "fair" result.
Laws are enforceable only within the context of the community in which they are applied. And beyond $50 billion, that community would not be the United States. It would be the rest of the world, especially Great Britain.
So don't get panicked by the "maximum" liability amounts being thrown about for BP. Such litigation is (almost) never settled for "full" (inflated) value. Instead, I estimate "fair value" at $50 billion, plus or minus $15 billion. Within those (relatively large) parameters, the stock figures to be a good, though not superlative, investment.
On the other hand, don't put "too much" money into BP, or any other special situation. It's possible that the trial lawyers will take investors to the cleaners on any one special situation. But if you had a portfolio of ten or so such positions, eight or nine of them would probably do well.