Quant Memo
Core

Natural Gas Markets and Seasonality

Natural gas is hard to store and hard to ship across oceans, so its price swings hard with the weather and the season in a way oil's global, tanker-borne market mostly doesn't.

Prerequisites: Commodity Futures Basics

Oil can be pumped in Texas, loaded on a tanker, and sold in Singapore three weeks later — it's a genuinely global market. Natural gas, historically, could not: without a pipeline or an expensive liquefaction plant, gas produced in one region mostly had to be consumed in that same region. That physical constraint is why natural gas prices behave so differently from oil prices — regionally fragmented, and swinging hard with the calendar and the weather.

Natural gas is a regional, seasonal market because it's expensive to store and historically hard to ship across oceans. Prices spike when cold weather meets tight storage, and the futures curve itself prices in the calendar months ahead of time.

Why storage drives everything

Gas demand for heating spikes in winter and (via air conditioning and power generation) again in summer, but production is roughly steady year-round. The gap is bridged by underground storage — depleted reservoirs and salt caverns injected with gas in the shoulder months (spring, fall) and withdrawn in winter. The Henry Hub benchmark in Louisiana is the US reference price, and its futures curve is famously seasonal: winter-month contracts routinely trade above summer-month contracts because that's when withdrawal risk (and price-spike risk) concentrates.

month of the year winter withdrawal summer injection winter again
Storage inventories fall through winter and rebuild through the shoulder seasons — the futures curve mirrors this pattern months in advance.

Reading the storage report

Every week, US inventory data shows the change in gas held in storage. A build smaller than the seasonal average (less gas going in than usual) signals tighter supply and tends to push prices up; a larger-than-average build signals looser supply and pushes prices down. Because the number is compared against a five-year average, not just the prior week, the surprise relative to expectations moves the market more than the raw figure does.

Worked example

Suppose the market expected a storage injection of 80 billion cubic feet (bcf) for the week, based on typical shoulder-season demand, but the actual report shows only 55 bcf injected — 25 bcf less gas went into storage than expected, because a cold snap pulled more gas into heating demand than forecast. Henry Hub front-month futures, trading at $2.80/MMBtu before the report, jump to $3.05/MMBtu afterward: the market re-prices for a tighter winter than it had assumed, an 8.9% move driven by a single weekly data release.

What this means in practice

Traders in gas markets watch weather forecasts as closely as they watch supply data, because a colder-than-normal forecast changes expected withdrawals directly. The seasonal shape of the futures curve — winter above summer — is itself tradable: a "widow-maker" spread trade (named for its history of blowing up traders) bets on the relationship between a specific winter month and the following March contract, right at the point where the market transitions from withdrawal season back to injection season.

Don't extrapolate oil-market intuition to gas. A supply glut in one region (say, Appalachia) doesn't necessarily depress prices in another (say, the Gulf Coast) if pipeline capacity between them is constrained — gas markets can stay regionally disconnected in ways a globally shippable commodity like oil rarely does.

Related concepts

Practice in interviews

Further reading

  • EIA, Natural Gas Weekly Update
  • CME Group, Henry Hub Natural Gas Futures
ShareTwitterLinkedIn