What Happened When Wall Street Trusted a Salad Oil Tycoon?

There's a storage unit with your name on the padlock, and a statement comes every month saying, it's all still in there. In 1963, Wall Street lent against a New Jersey tank farm on that kind of paper. The oil vanished, and the warehouse could never say how. That arrangement had been running since 1957. It came apart in one week of November, 1963, and the bill didn't land on the man who built it. A large share of it landed on people who had never bought a drop of soybean oil, and most of them never found out why their brokerage account stopped working. The story usually gets told like this. A commodities trader lied about how much oil he had, borrowed against the lie, and eventually somebody caught him. Tidy. It's also wrong in the part that matters. Look at the order of events in the court records. The account had run for six years. By the account given in court, AmeriExpress first learned there was trouble ahead during the week of the 18th of November, 1963. Allied crude vegetable oil refining filed for bankruptcy on the 19th, and the true size of the hole only became clear when somebody took inventory that weekend. Discovery arrived after the collapse, not before it. The warehouse company being paid to check hadn't produced it. Neither had the banks holding the receipts or the exporters or the brokers. What ended the arrangement was the price of soybean oil moving the wrong way. To see how something like that runs for years, you have to look at what a bank is actually holding when it lends against a commodity. It isn't the commodity. You can't hand a bank a lake of soybean oil. So the industry built a work ground called field warehousing. An independent warehouse company takes custody of your inventory where it already sits and writes a receipt. This much oil, this grade, in this tank. You take the receipt to a lender. The lender lends against the paper. From that point on, the oil is a story the paper tells, and the whole structure rests on one assumption that the company writing the paper has no reason to write it wrong. Follow one shipment. In the spring of 1963, Procter and Gamble sent just over 9 million pounds of fully refined soybean oil to a tank farm in Bayonne, New Jersey. 151 railroad tank cars of it. The warehouse company issued receipts saying it had the oil. Then, month after month, it sent statements confirming the oil was still sitting there. Those statements were drawn from the warehouse's own books. They kept going out until November. The tank farm belonged to Allied Crude Vegetable Oil Refining, run by a man named Anthony DeAngelis. Everyone called him Tino. Allied bought vegetable oil in enormous volume and shipped it overseas. And like every trader in that business, business owners are permanently short of cash. Receipts fixed that. Receipts went out to banks and exporters. Cash came back, and the cash went into soybean oil futures, bets that the price of the stuff in the tanks would rise, which would, helpfully, make the tanks worth more. Now, here's the part that stops you. The independent warehouse wasn't standing on its own ground. Allied leased those tanks and then subleased them to the warehouse company, a subsidiary of American Express. And that is how it became an independent warehouse. It rented its floor from the company whose inventory it was there to certify. And the receipts recorded the value of the contents as the depositor had stated it. So the certificates stood on the borrower's tanks and carried the borrower's numbers. None of this was concealed. It was written into the arrangement. By the autumn of 1963, warehouse receipts written by that operation for companies, DeAngelis Controlled, were circulating with stated values adding up to $136 million. Some were genuine. Some were forged. Because somebody had got hold of a receipt book. To a lender holding one, they looked identical. They were identical. Same form, same signature line, same famous name across the top. Then the price moved. Soybean oil fell that autumn. And Allied's futures positions threw off margin calls it couldn't meet. On the 19th of November, Allied filed for bankruptcy. The next day, Ira Haupt & Company, the brokerage carrying Allied's futures trades, was suspended from the New York Stock Exchange, insolvent. On the 21st, Procter & Gamble was told its oil couldn't be accounted for. On the 22nd, the president was shot in Dallas and the exchange closed. On the 25th, the stock exchange and a group of banks signed a deal to wind Haupt down and buy out the claims of its customers. There were about 20,000 of them. Ordinary accounts. Frozen. Belonging to people whose only link to a New Jersey tank farm was that their broker had a client there. Meanwhile, at Bayonne, people were opening tanks. Not reading a statement drawn from a ledger. Not copying a figure off a form. Opening the tanks and looking at what came out. What came out wasn't soybean oil. A court later described the structure as mostly acid soap stock, fish oil and water. As for where the soybean oil had gone, the warehouse company never said. Courts afterwards put the burden of explaining a loss on the warehouse and the explanation never came. A judge made the point plainly. A custodian can't claim it took proper care of something and also claim it has no idea what happened to it. Everything physically left in that tank farm when it was eventually sold came to $6 million. So $136 million of paper. $6 million of liquid. That's dense 22 claim tickets for every coat on the rack and every single person in the queue certain that theirs was the good one. Sorting it out took years. The warehousing subsidiary went into bankruptcy and American Express eventually settled with the claimants. In the litigation, one argument seriously advanced was that Procter & Gamble ought to accept its fair share of whatever was actually in the tanks. The court declined. DeAngelis pleaded guilty to carrying forged warehouse receipts across a state line plus conspiracy. In August of 1965, a judge stacked the counts and gave him 20 years. Not for the oil, which was never his to steal, and not for the losses, which were spread across too many balance sheets to pin anywhere. For the documents. Here's my opinion and I'll flag it as opinion. The con man is the least interesting thing in this story. Everyone agrees he was a crook, so there's nothing to argue about. What's worth arguing about is that field warehousing didn't break. It worked exactly as designed and the hole was in the design. If you P1 company to vouch for another company's assets and the vouching happens because the second company's property using values the second company supplies, you haven't bought verification. You've bought a document. And I don't think the lenders were fools either. That's the lazy read. They were being rational. Sending someone to Jersey to climb a tank costs money. Reading a receipt with a famous name on it costs nothing. Given how rarely the tanks are empty, the second one is the better trade almost every time. Almost. Which leaves the paradox the whole thing runs on. A receipt from a firm nobody trusted would have been checked. Someone felt would have driven out and dropped a line into the tank. A receipt from a firm everybody trusted didn't need checking and not needing to check was the entire reason for getting one. So the more solid a guarantee looks, the less anyone examines the thing behind it. Trust doesn't sit on top of the inspection. It replaces it. Which means the safest feeling collateral in any room is usually the collateral nobody in that room has actually seen.