The curious oil and natural gas price differential

Something strange is happening in the natural gas and oil markets. The average real price of natural gas during the last 20 years has been about $3.80 per gigajoule, and the average real price of oil in that same period has been about $5.90 per gigajoule. During this time, the long-term ratio between the real price of oil and the real price of natural gas has constantly varied between 1 and 2 with the long-term average hovering at 1.6. Since January, however, that ratio has jumped to more than 4 to 1, which is unprecedented and has implications for long-term energy policy.

One short-term implication of the low price of natural gas has been low electricity costs across the United States, which is contributing to the slacking interest in alternative energy investment.”

Much has been written about the relationship between the price of natural gas and the price of oil. In 2007, for example, Stephen Brown and Mine Yücel from the Federal Reserve Bank of Dallas performed an excellent econometric analysis, which corrected for short-term supply and demand disruptions and showed that the price relationship is even stronger than indicated by simply tracking and comparing prices over time. In particular, they concluded, “An error-correction model that takes into account crude oil prices, weather, seasonality, and gas storage and gas production disruptions well explains natural gas prices. Moreover, the model shows that U.S. natural gas prices are related to those of crude oil with natural gas prices adjusting to changes in crude oil prices. The relationship has complex short-term dynamics, but is quite stable in the long run.”

Others have built fundamental models to relate the price of natural gas and the price of oil by analyzing the various end users that can switch relatively quickly between the two. A simple example of this is heating applications, since many residential, commercial, and industrial boilers can burn either natural gas or distillate fuel oil, which has historically been priced about 95 percent that of crude oil. In such cases, the premium spot price of distillate fuel oil should reflect the higher cost of getting natural gas from Henry Hub–a physical location where several U.S. pipelines meet and is used to set prices for natural gas futures contracts traded on the New York Mercantile Exchange–to the end user. That extra transportation cost has been estimated to be $0.50-$1 per gigajoule. Such fundamental models have done a decent job of explaining the long-term relationship between oil and natural gas prices; they do about as good as a simple linear regression. It is nearly impossible, however, to explain the current price anomaly between natural gas and oil with historical data. So, what’s going on?

In 2008, Terry Engelder, a geosciences professor at Penn State University, claimed that the Marcellus Shale, a geologic formation that extends from Tennessee to New York, would become one of the world’s “top super giant gas fields.” Engelder is widely considered a foremost expert on Appalachian geology, so his opinion is taken seriously. Once the media publicized his claims, a land rush for gas-production leases was on in the Appalachian Basin. Indeed, signing bonuses for gas production leases in the Marcellus Shale jumped from a few hundred dollars per acre to more than $2,000 per acre between 2005 and 2008.

The key to the Marcellus Shale gas deposits is that during the last 300 million years natural gas formed from buried organic matter, all the while expanding and causing vertical fractures in the overlying rock. After the excess pressure was relieved, the vertical fractures stopped propagating, trapping the gas in thin vertical columns that is difficult to access by traditional vertical drilling. Recently developed technologies, however, have enabled horizontally drilled wells to contact and reopen these fractures, enabling the economic recovery of gas from the shale. Over the last year, several wells drilled into the Marcellus Shale have reported initial production rates of more than 4,000 gigajoules of gas per day (1 megawatt-hour of electricity production requires 8-10 gigajoules of natural gas). At current prices, those wells would initially generate $12,000 per day in revenue. Engelder has estimated the formation’s ultimate yield by extrapolating these initial production rates along depletion curves from somewhat analogous geologies. His current estimate is 470 trillion gigajoules. Since the United States consumes about 22 trillion gigajoules of gas per year, then–assuming Engelder is correct–production from the Marcellus Shale could satisfy current U.S. demand for more than 20 years.

So there’s little mystery why natural gas prices are so low. During the last year, what may prove to be the biggest natural gas field in the country started producing, while simultaneously the global financial crisis sent the world into a recession that resulted in a dramatic decline in natural gas demand. The expectation that demand will decrease further and that supply will increase further has sent long-term gas prices hurdling downward.

But why has the ratio of the price of oil to the price of natural gas changed so dramatically? Specifically, why are natural gas prices not depressing oil prices? All of the models–both statistical and fundamental–indicate that the price of natural gas is closely related to the price of crude oil.

The most obvious explanation is the time lag in fuel switching. Some users can switch from oil to gas overnight, but for others, it takes more time. Since the ratio has trended up for the last three years, most of those who could easily switch probably have already done so. And thus, with the “quick” fuel-switching response more or less complete, additional supply of natural gas hasn’t been quickly met with increased demand.

There is, however, enormous potential for additional fuel switching. Nearly all consumers of distillate fuel oil could–in principle–switch to natural gas, and at the current energy-cost premium of oil relative to gas, there should be little doubt that making such a switch would be economical. In particular, about 37 percent of distillate fuel oil is used in stationary applications, which are relatively easy to convert to natural gas. It makes no sense, for example, for anybody to heat their home with oil at the current price ratio. And if the ratio holds for awhile, then I expect to see municipalities around the country start to invest in additional natural gas distribution. In addition, I expect to see a rebound in the building of natural gas-fired power plants–which had all but ceased a few years ago.

Furthermore, the remaining 63 percent of distillate fuel oil used in transportation is a decent candidate for switching to natural gas. Long-haul freight could cut fuel costs by more than 75 percent–at current prices–by converting trucks from diesel. Although the volumetric energy density of compressed natural gas is 25 percent that of diesel, at current price ratios the cost of bigger fuel tanks and engine retrofits will be paid for very quickly by the fuel savings. There simply is no way that trucking companies will leave that much money on the table.

Historically 1 joule of oil has been valued about 50 percent more than 1 joule of natural gas. That difference in valuation is a result of the intrinsic benefits–such as lower transportation and storage costs–of a fuel that is a liquid at surface conditions. Those intrinsic physical benefits won’t go away, but they won’t get better either. Thus, I expect the long-term price ratio of oil to gas will return to its historic levels of between one and two, which means either the price of oil will drop, the price of gas will increase, or both.

One short-term implication of the low price of natural gas has been low electricity costs across the United States, which is contributing to the slacking interest in alternative energy investment. (I will explore other implications of this phenomenon in my next column.) Such short-term planning is no different than in 2007 when the price of natural gas skyrocketed to nearly $13 per gigajoule and gas-fired power plant construction looked like utter folly. My advice to policy makers and industry: Look beyond short-term price fluctuations to the bigger picture. My advice to energy traders: Go short on oil and long on gas.