By Grant Mercer
There is a substantial risk of loss in trading commodity futures and options products. Losses in excess of your initial investment may occur. Past performance is not necessarily indicative of future results.
In July 2008, a Tulsa-based pipeline and storage company with thousands of miles of pipe, dozens of terminals, and a real, operating physical business filed for bankruptcy — not because its pipelines failed, but because its trading book did. SemGroup's paper oil positions lost roughly $3.2 billion, and the margin calls arrived faster than the company could answer them. A physical empire was bankrupted by a derivatives desk.
The episode is the purest illustration of a boundary every commodity business must police: the line between hedging the physical business and speculating alongside it. When that line dissolves, the trading book can become larger than the company — and the company becomes the trading book's hostage.
What happened: the setup
SemGroup LP was, on paper, a midstream energy company: crude oil pipelines, storage terminals, and transportation assets, primarily in the mid-continent United States. It bought, stored, and moved physical crude for refiners and producers — the kind of business presumed insulated from the price swings of the commodity it handled. That presumption was wrong.
Alongside the physical business, SemGroup maintained an oil trading operation — futures on NYMEX and over-the-counter derivatives — that was described as hedging the company's physical exposure. In a midstream business, some trading is legitimate and necessary: a company holding physical crude in storage faces price risk between purchase and sale, and futures positions offset that risk. But SemGroup's book grew far beyond any plausible hedging need. Public reporting after the collapse indicated that the firm's trading positions were enormous relative to its physical volumes — the paper business had outgrown the physical business it was supposed to serve.
The positions were also directionally exposed. As crude oil climbed through the first half of 2008 — driven by tight global fundamentals, a weak dollar, and speculative momentum that would carry WTI to nearly $147 a barrel in July — SemGroup's book was positioned the wrong way. The company's trading operation was effectively short crude oil into one of the steepest oil rallies in history. Whether the book began as a hedge that drifted into speculation or was speculative from the start was litigated for years; what is established is the scale — paper positions large enough to destroy the physical company.
Compounding the problem, the trading was reportedly directed at the very top — the chief executive was personally involved in the trading decisions — so the normal corporate checks on a trading desk, independent risk management and enforceable position limits, were absent or overruled. When the person taking the risk is also the person who approves the risk, the control structure is decorative.
The mechanism: margin calls as a liquidity death spiral
The mechanism that killed SemGroup was not the trading loss itself. It was the margin call — the daily cash demand that converts a paper loss into a liquidity crisis.
Futures positions are marked to market every day. When a position moves against you, the clearinghouse demands variation margin — cash, posted that day, to cover the loss. There is no grace period, no negotiation, no "wait and see." A short oil position during the 2008 rally generated enormous daily margin calls: every dollar crude rose multiplied across every contract, every day, in cash, immediately.
SemGroup faced a second, quieter drain: its lenders watched the trading losses mount and began pulling back. Credit facilities were the lifeline for a trading book hemorrhaging cash, and as the banks tightened, the firm entered the classic liquidity death spiral — margin calls demanding cash, cash drawn from credit lines, credit lines withdrawn because of the margin calls, each turn requiring more cash the company did not have.
In July 2008 the company disclosed trading losses of approximately $3.2 billion — a figure that dwarfed its equity and rendered it insolvent. On July 22, 2008, SemGroup LP filed for Chapter 11 bankruptcy protection in Delaware; the stock collapsed and equity holders were effectively wiped out. A company built over years of acquiring physical assets was felled in weeks by positions that existed only on screens.
The warning signs
As with the other disasters in this book, the warning signs were available to anyone who knew where to look — and to anyone inside the company willing to ask uncomfortable questions:
The book dwarfed the business. The central warning sign was proportion: a trading operation whose notional size vastly exceeded the physical volumes it could plausibly be hedging. In a legitimate commercial hedging program, the paper positions track the physical exposure — barrels in tanks, barrels in transit, contracted purchases and sales. When the paper book is multiples of the physical book, the excess is not hedging. It is speculation wearing a hedge's clothing, and it should be sized, capitalized, and risk-managed as speculation.
Direction into a historic rally. Crude oil's 2008 ascent was one of the most visible trends in financial history — front-page news for months before the July peak. A short crude position into that rally was not a subtle risk; it was a loud one. Trend persistence is one of the market's most reliable features, and fighting a historic trend with leveraged size is a bet that the turn comes before the margin call. It usually doesn't.
Control failures at the top. The reported involvement of the chief executive in directing trades is a governance red flag of the first order. Sound trading operations separate risk-taking from risk oversight: the desk takes positions, and an independent risk function — reporting outside the desk's chain of command — enforces limits. When the same person sets the strategy, takes the positions, and adjudicates the limits, there are no limits. Investors and lenders evaluating a commodity company should ask who runs the trading book and who can overrule it.
Opacity. SemGroup's trading positions were not transparent to its investors, its lenders, or arguably its own board in anything like real time. A physical company with a large derivatives book that does not disclose the book's size, direction, and margin requirements is asking its stakeholders to underwrite risks they cannot see. The inability to see the book is the warning sign.
Leverage on leverage. SemGroup had grown aggressively through debt-financed acquisitions of physical assets, leaving the company with a leveraged balance sheet before the trading losses began. A leveraged company with a speculative trading book has no shock absorber: the balance sheet cannot fund the margin calls, and the margin calls destroy the balance sheet. Each form of leverage was manageable in isolation; together, they were fatal.
The aftermath
The bankruptcy filing in July 2008 triggered the standard machinery of corporate collapse, accelerated by the commodity context. The chief executive was dismissed. The company's physical assets — the pipelines, terminals, and storage facilities that had been the ostensible business all along — were sold off to new owners through the bankruptcy process, with major midstream operators acquiring the pieces. The trading book was unwound. Creditors received partial recoveries; equity holders received effectively nothing.
The legal aftermath stretched for years, as the bankruptcy trustee and creditors pursued claims arising from the trading losses. For the energy industry, SemGroup joined a small, infamous club of physical companies destroyed by their paper books — a club whose membership includes Metallgesellschaft, and whose lesson the industry keeps having to relearn.
The structural consequence was a renewed focus on the boundary between commercial hedging and proprietary trading inside non-financial companies: lenders asked harder questions about borrowers' derivatives books, and boards were reminded that a trading desk inside a physical company needs the same independent risk infrastructure as one anywhere else — and that the chief executive's personal involvement in trading is a governance failure, not engagement.
What the episode teaches
Hedging and speculation are separated by proportion, not by label. Calling a book "hedging" does not make it so. The test is arithmetic: do the paper positions correspond, in size and direction, to the physical exposures of the business? Any excess is speculation, and speculation inside a physical company must be capitalized and risk-managed as its own business — because the market will treat it as one.
Margin calls transmit market risk into corporate survival risk. A trading loss on paper is a P&L event; a margin call is a cash event. For a leveraged company, the difference is everything. Risk management that tracks only the size of potential losses, and not the timing of the cash demands those losses generate, is measuring the wrong thing. The question is not "how much could we lose?" but "how much cash could we be required to post, and how fast, and where would it come from?"
The person taking the risk cannot be the person approving it. Independent risk oversight is not bureaucracy; it is the mechanism by which organizations survive their own convictions. When the chief executive directs the trading book, the organization's most powerful voice is also its most concentrated risk — and there is no one left to say stop.
If you cannot see the book, you cannot underwrite the company. This lesson runs in both directions. Investors and lenders must demand transparency into the derivatives books of commodity companies — size, direction, margin requirements, stress scenarios. And companies must provide it, because opacity is itself priced: when the book finally becomes visible, it is always at the worst possible moment.
Physical assets do not immunize against paper risk. SemGroup's pipelines and terminals were real, valuable, and ultimately sold to real buyers. None of that mattered to the company's survival, because the trading book's margin calls arrived in cash and arrived first. A physical business is not a hedge against a derivatives book; the derivatives book is a claim on the physical business's cash. Size the paper so the physical can survive it.
There is a substantial risk of loss in trading commodity futures and options products. Losses in excess of your initial investment may occur. Past performance is not necessarily indicative of future results.