Oil Market Manipulation: The Dark Pools Regulators Still Can’t See#

In the summer of 2008, Robert McCullough — an energy consultant working out of Portland, Oregon, known for his methodical, numbers-first approach — started pulling apart trading patterns in the crude oil futures market. What he found didn’t look like a bubble. Not in the usual sense. Bubbles are crowd phenomena: they emerge from the collective behaviour of thousands of actors, none of whom set out to create one. What McCullough’s analysis pointed to was something far more disturbing. The trading patterns he uncovered weren’t consistent with speculative herding. They were consistent with a small number of very large players deliberately concentrating market power.

He was careful with his language. He didn’t use the word “manipulation.” In commodity regulation, that word carries legal weight — specific definitions, specific burdens of proof that no outside analyst can shoulder. But the patterns he documented — concentrated positions, coordinated timing, anomalous price swings during thinly traded hours — forced a question the entire speculation debate had managed to sidestep: what if the problem wasn’t just systemic? What if it was intentional?

From Passive to Active#

Everything we’ve built to this point has been, at its core, a structural argument. The paper barrel engine manufactured shadow prices. The infiltration pipeline pumped speculative capital into futures markets. The narrative shield made inflated prices look normal. These were systemic mechanisms — architectural features of a market that had been financialised well beyond anything its designers imagined. No conspiracy needed. No villain required. The machine produced the outcome on its own.

That structural account has a real strength: explanatory simplicity. It explains the bubble without requiring anyone to have planned it. But it also has a blind spot — it may be giving too much benefit of the doubt.

The structural story assumes everyone involved was acting in good faith. That swap dealers were genuinely hedging. That index funds were passively allocating capital. That investment banks were simply offering services. It assumes the machine ran itself, and that nobody had a hand on the throttle.

McCullough’s evidence didn’t conclusively prove otherwise. But it cracked open a door that couldn’t easily be shut again. If a single entity held positions large enough to move the price of a globally traded commodity — and that same entity published research forecasts that shifted market sentiment — and the trading happened in over-the-counter markets where nobody could see it — then the preconditions for deliberate price manipulation weren’t hypothetical. They were baked into the market’s own architecture.

The Three Conditions#

Market manipulation, in any commodity, needs three conditions to exist simultaneously. Understanding them matters, because the oil market of 2007–2008 checked every box.

First: capability. A single actor has to be able to build positions large enough to materially move prices. In a properly regulated market with effective position limits, that’s hard to pull off. But the CFTC’s “bona fide hedging” exemptions had gutted position limits for swap dealers. One swap dealer could — and did — hold positions that made every physical oil company’s book look tiny by comparison. The capability condition wasn’t just met. It was structurally guaranteed.

Second: motive. The actor has to benefit from the price movement it can cause. When investment banks were simultaneously publishing oil price forecasts (shaping sentiment), running swap desks (earning fees as commodity allocations rose), and maintaining proprietary trading books (profiting from directional bets) — the alignment of incentives wasn’t exactly subtle. In the language of criminal law, it was a textbook case of means, motive, and opportunity living under the same roof.

Third: concealment. The manipulation has to be difficult or impossible to detect. And here, the over-the-counter derivatives market offered something close to perfect cover. OTC trades were bilateral, private, unreported. No central clearinghouse recorded them. No public database tracked them. The CFTC’s surveillance powers reached only as far as exchange-traded futures — the visible tip of an iceberg whose underwater mass was, by 2008, estimated to be several times larger than what showed above the surface.

If a tree falls in a forest and nobody hears it, did it make a sound? If manipulation happens in a market where nobody can see it, does it exist? The philosophical question had a very practical answer: the absence of evidence, in a market deliberately built to prevent evidence from being collected, is not evidence of absence.

The Unobservable Domain#

The OTC market’s opacity wasn’t some accidental oversight. It was a fiercely guarded privilege. When Brooksley Born, then CFTC chairwoman, tried in 1998 to extend regulatory oversight to over-the-counter derivatives, she was overruled by a coalition that included the Treasury Secretary, the Fed Chairman, and the SEC Chairman. The Commodity Futures Modernisation Act of 2000 sealed the deal, explicitly exempting OTC derivatives from CFTC jurisdiction.

What this produced was a regulatory architecture that was, from a surveillance standpoint, intentionally blind. The CFTC could see exchange-traded futures. It couldn’t see the vastly larger OTC market that was increasingly setting prices. It was like giving a police department jurisdiction over a well-lit main street while explicitly barring them from patrolling the back alleys where the real crimes happened.

By 2008, the scale of this blind spot was staggering. Estimates of OTC energy derivatives’ notional value ran into the trillions. The CFTC’s publicly available data — the Commitments of Traders reports, the supplemental swap dealer data — captured maybe a quarter of total market activity. The rest existed in what you might generously call the unobservable domain. The same structural opacity that sheltered this invisible mass continues to raise alarms on Wall Street itself — Goldman Sachs’s oil trading operations have recently drawn fresh scrutiny over dark pool activity, with critics at Bloomberg warning that vast swaths of petroleum derivatives still change hands beyond any regulator’s line of sight.

Even now, the blind spot hasn’t been fully closed. Despite post-crisis reforms, more than forty percent of oil-related derivatives trading still takes place outside the perimeter of regulated exchanges, under disclosure rules that remain substantially weaker than those governing exchange-traded instruments. A recent Reuters analysis of the CFTC’s own reform agenda confirmed as much: the regulatory gaps in OTC oil derivatives persist, and the agency’s surveillance tools still cannot reach the bulk of the market they are supposed to police. The dark matter of the oil derivatives universe has been partially illuminated. But significant portions remain in shadow.

The Rabbit Hole#

McCullough’s work, and the broader concerns it raised, never resulted in manipulation charges. No smoking gun was found — which is exactly what you’d expect in a market engineered to prevent smoking guns from being found. Whether deliberate manipulation contributed to the 2008 oil price spike remains, in the strictest evidential sense, an open question.

That the question endures isn’t merely historical. When Iran’s missile strikes disrupted supply routes in early May 2026, oil prices swung violently — and market surveillance bodies flagged anomalous trading patterns during the chaos, patterns uncomfortably reminiscent of the coordinated activity McCullough had documented nearly two decades earlier. The investigation is ongoing, the outcome uncertain. But the structural conditions that made exploitation possible in 2008 have not been dismantled.

But that unanswered question is itself the point. A market where the three preconditions for manipulation are structurally embedded — capability, motive, concealment — is a market that can’t credibly claim to be manipulation-free. It can only claim that it doesn’t know. And a regulator that can’t tell the difference between “no manipulation occurred” and “we lack the tools to detect manipulation” is a regulator that has failed at its most basic job.

This is the darker portent the title promises. The speculation debate, for all its heat, played out on relatively comfortable ground: were market structures allowing passive speculative flows to inflate prices? The answer, as the preceding chapters have argued, was yes. But underneath that question lurked a far more unsettling one: were some participants actively exploiting those structures for private gain? The market’s own opacity guaranteed the question could be asked — but never definitively answered.

What we’re left with isn’t a conclusion. It’s a warning. The shadow oil price isn’t merely the product of impersonal market forces. It exists in a market whose architecture — whether by design or by neglect — provides the perfect environment for exploitation. Whether anyone took advantage of it is a question the available evidence can’t resolve. That the evidence can’t resolve it is, perhaps, the darkest portent of all.