Metallgesellschaft: The $1.5 Billion Hedge That Worked Too Well

The Setup: Long-Term Supply, Short-Term Hedges

Metallgesellschaft AG, one of Germany’s largest industrial conglomerates, entered long-term fixed-price oil supply contracts with customers—commitments extending 5 to 10 years into the future. To hedge the resulting price exposure, its U.S. subsidiary MGRM implemented a “stack-and-roll” strategy: hedging the entire long-term commitment with short-dated (nearby month) futures contracts, then rolling the position forward each month.

The mathematics of this approach is sound. The optimal hedging function $h(t)$ for short-term futures covering long-term commitments satisfies a differential equation with specific boundary conditions. When oil prices rose, the hedge produced profits. When they fell, the hedge produced losses—but these losses were offset by gains on the long-term supply contracts. The problem was not the mathematics.

The Liquidity Trap

The stack-and-roll strategy requires rolling massive futures positions monthly. When oil prices declined in 1993, MGRM faced enormous margin calls on its futures positions. While the long-term supply contracts were theoretically gaining in value (they could now be fulfilled at lower cost), those gains were unrealized and illiquid—locked in 5-to-10-year customer contracts that could not be monetized.

The losses on the futures positions were real cash outflows. The gains on the supply contracts were accounting entries. MGRM needed cash today to maintain a hedge that would pay off over years. When the parent company’s supervisory board lost confidence, they liquidated the hedge—crystallizing a $1.5 billion loss.

Was the Hedge Wrong?

The academic literature remains divided. One school (Culp and Miller, 1995) argues that the hedge was fundamentally sound and the parent company’s decision to liquidate was the error. Another school argues that the basis risk and rollover costs made the strategy untenable from the start.

The mathematical analysis supports both perspectives simultaneously: the hedge was sound in expectation, and it was operationally untenable given MGRM’s liquidity constraints. This is precisely the lesson—hedging is not purely a mathematical problem. It is a mathematical problem embedded in an institutional liquidity framework.

A hedge that is theoretically perfect but generates margin calls you cannot meet is not a hedge—it is a leveraged position with a story attached.

Modern Lessons for Risk Managers

Metallgesellschaft remains required reading for every risk management curriculum, and its lessons apply directly to contemporary situations:

• Basis risk in rolling hedges. Every stack-and-roll strategy carries basis risk—the risk that the relationship between nearby and deferred contracts changes. In contango markets, the rolling cost alone can erode the hedge value.
• Liquidity of the hedge must match liquidity of the exposure. If the hedged exposure is illiquid (long-term contracts, real assets, infrastructure concessions), a daily-margined hedge creates a structural liquidity mismatch.
• Governance structures must understand hedge mechanics. MGRM’s parent board liquidated the position because they did not understand that the futures losses were offset by unrealized gains on the supply contracts. Governance systems that cannot evaluate hedge effectiveness will make pro-cyclical decisions.
• Mark-to-market is not the same as economic loss. The margin calls reflected mark-to-market losses, not realized economic losses. The distinction between accounting and economic hedging remains one of the most important concepts in institutional risk management.