The Wonderful Curse


Production constraints and demand pressures mean high gas prices are here to stay.

Fortnightly Magazine - December 2007

Volatility in energy prices is both a scary and wonderful thing. It brings risks that must be managed under uncertain future conditions. It also brings opportunities to profit from price movement and competitive market advantages exploited through strategy, skill and luck. Just how good the outcome of such volatility can be depends on how well each market participant studies the fundamentals, manages uncertainty and remains flexible.

With natural-gas supply and demand nearly in balance, gas prices and volatility levels have remained tenaciously steep by most historical measures since early 2003. The horrific hurricane damage sustained in 2005 in the Gulf of Mexico added further stress to domestic natural-gas and oil supply infrastructure that is not quite yet back to normal. U.S. crude oil production in 2006 averaged about 5.1 million bbl/day, down slightly from 2005 levels as a result of the hurricanes. And offshore gas production averaged 7.8 Bcf/day in 2006, down nearly 20 percent from mid-2005 levels, although some is undoubtedly due to gas-deliverability depletion. Since that time, gas prices have retreated but still remain well above long-run supply cost.

Current high gas prices reflect several factors that have converged into the “perfect storm.”

First, the high cost of replacing natural-gas production across all basins has raised the price floor. The gradual reduction in supply from conventional gas basins and the steady increase from unconventional basins, coupled with increased imports of liquefied natural gas (LNG), has exacerbated the situation. Absent immediate alternative sources of supply, these price levels will persist for some time.

Second, persistently high crude prices (in part due to increased global turmoil) have strengthened the “crude sympathy” that exists between the two commodities. High oil prices “allow” gas prices to rise due to competitive fuel switching. Also, the petroleum supply industry tends to favor oil development over gas when prices rise, because oil development costs less and leads to production sooner than gas development does. And due to cold weather, a record amount of natural gas was withdrawn from storage in February, dropping inventories below the five-year maximum for the first time in more than a year.

Oil and gas price moderation likely will occur after several years, but the actual timing and extent are still subject to large amounts of uncertainty. In particular, price declines are not expected until significant new sources of supply materialize in North America.

For some time, Global Energy has considered the impending increase in LNG supply as one sign that moderation eventually would occur, and the recent startup of the Gulf Gateway Energy Bridge and Altamira projects bolstered this opinion. Additionally, significant new LNG-import capacity is now under construction (such as Canaport in New Brunswick), as is more drilling for “unconventional” gas. These projects are scheduled to enter service in time to moderate North American gas prices by the end of the decade.

In addition, pipeline imports from Canada are expected to continue declining due to their increased domestic demand and supply moderation (i.e., disappointing production in the North Atlantic/Canada Basin). Therefore, the days of $2 or $3 gas, as the market experienced during the supply glut of the 1990s, are long gone. However, the 1.8 Bcf/day Rockies Express Pipeline is scheduled to go into service January 2008, relieving the congestion of relatively inexpensive Rocky Mountain gas and allowing flows from Colorado and Wyoming eastward to Illinois, Ohio and western Pennsylvania. Of course, basis differentials will change markedly.

Following the NYMEX forward curve, Henry Hub natural-gas prices for 2008 likely will average $7.99/MMBtu, compared with $8.07 in the previous Global Energy natural-gas reference case (down $0.08 and 1 percent). For 2009, the Henry Hub price is expected to be $8.00/MMBtu (up $0.73 and 10 percent). However, overall the current reference-case view predicts just slightly to moderately higher gas prices than forecasted last spring, about 4.5 percent for the 25-year forecast period from 2008 to 2032. Forecasted gas prices do not sag as low following the high “prompt” years as previously forecast, reflecting the impact of greater costs from increasing percentages of gas from unconventional sources; greater perception of natural gas being “greener” than other options and thus able to contribute to global-warming solutions; and the reality of a continuing tight supply/demand situation in the United States.

In the longer term, the landed price of LNG remains uncertain given the vagaries of the world market, available LNG supply, political unrest, and cartel formation; however, its cost of production is well understood and very attractive at today’s market price. Other potential new supplies in North America (and the uncertainties associated with them) create additional uncertainty for the price of natural gas. Many of these factors can be quantified through stochastic volatility analysis.

Gas Goes Global

One conclusion from the analysis, which is possibly the most important point, is that increasing reliance on LNG will transform the continental gas market into a global gas market during the next 10 years. This is expected to affect market prices, industry financial performance, and capital deployment. For example, although LNG imports totaled just over 0.5 Tcf in 2006, by 2030 LNG likely will supply 8.4 Tcf of the total U.S. gas supply requirement of 32.7 Tcf, up more than 15-fold.

Another force already present is the massive buildup of gas-fired electric generation capacity of recent years. Since the late 1990s, more than 200,000 MW of new combined-cycle, gas-fired power plants have entered the North American markets. Thus far, the financial and operating performance of these plants has been disappointing as a result of the massive overbuild of capacity witnessed in many regions. During the next 10 years, capacity utilization of these plants is projected to grow by nearly 40 percent over present levels. Given these combined-cycle plants are operating already, and long lead times and other financing and permitting hurdles exist for building alternative resources such as coal or nuclear, gas-fuel demand growth can be characterized as “predetermined”—at least through the 2012 to 2015 time frame. This will apply continuous pressure on gas markets and fuel suppliers. It also will further squeeze industrial gas consumers, who are relatively price sensitive, resulting in only flat industrial gas demand during the forecast period.

The view presented above remains the baseline scenario; however, in the past two years, utilities and independent power producers (IPPs) have been increasingly proposing alternative sources of generation. What was certainly the base-case view one or two years ago, is now being viewed with diminished certainty: The record high gas fuel costs, strong industrial demand destruction, global and political uncertainty of supply, and the development plans noted above will require close scrutiny in the coming years. High fuel prices already have begun taking their toll on the industry, encouraging growth in proposals for alternative generation supply—especially coal, nuclear, and renewables. As a result, natural-gas suppliers—and LNG suppliers in particular—might find themselves out of the market, along with generators holding large amounts of gas-fired capacity.

Although we certainly live in interesting times when it comes to gas prices and volatility, gas prices always have been relatively volatile, driven primarily by unexpected weather events. From mid-1985 to mid-1993, the EIA’s survey of monthly average wellhead gas prices averaged $1.78/Mcf. From Order 636 in 1993, which opened up the interstate gas price network and increased competition, gas prices remained mostly in the $2 to $3/MMBtu range, until the 2000s, when the decade-long drilling recession ended and supply/demand was more or less balanced.

The historical record of gas prices at the Henry Hub—North America’s main gas-trading hub and delivery point of the NYMEX futures market—shows key events, primarily weather-driven, affected prices over the last 10 years (see Figure 3). Most prominent was the effect of Hurricanes Katrina and Rita on market prices and the impact from cold spells pushing Henry Hub prices into the $8 to $9/MMBtu range. Most recently, record cold in January and February 2007 caused a bump in prices, followed by a collapse due to high LNG imports and high storage levels.

Changing Gas Sources

Several market forces have upset conventional views of the natural-gas market in North America. The market is transforming from a continental gas market, mostly disconnected from world LNG trade, to a more integrated global gas market with increasing dependence on various global LNG suppliers. This transformation has begun, in part, due to rising supply cost options, conventional reserve depletion, and because of the impending growth in gas demand for electric-power generation.

By 2020, more than 21 percent of U.S. gas supply will be sourced from LNG, with less than 10 percent coming from pipeline imports. By 2030, LNG supply will increase to nearly 26 percent of total requirements while less than 8 percent will come from North American pipeline imports (see Figure 4).

By 2011, North America is expected to overtake Europe, currently the second largest global importer of LNG. In that year, LNG imports likely will exceed 10.3 Bcf/d to the United States, Mexico, and Canada. While delays in building new LNG facilities already under construction could slow down this growth, North American LNG trade will increase substantially (see Figure 5).

Given expectations for strong gas demand through the forecast horizon, the supply scenario is affected by the Gas Exporting Countries Forum (GECF), a supply cartel similar to OPEC. The idea of a real gas OPEC was first floated in 2002 by President Vladimir Putin of Russia and backed by Kazakh President Nursultan Nazarbaev. From March through April 2005, the GECF, with headquarters in Algeria, met to discuss establishing a “fair price” for the international trade of LNG. And in January 2007, Russia (the world’s largest gas exporter and largest holder of gas reserves) signed a cooperation agreement with Algeria (Europe’s largest LNG supplier) in Tehran. The countries that comprise GECF hold nearly 73 percent of worldwide gas reserves and 41 percent of production.

Within North America, production and supply sources compete directly with LNG and imported pipeline gas. Canadian net exports will remain relatively flat until later in the forecast period, when Alaskan gas routed through Canada begins to flow. Several U.S. basins exhibit production declines, while others grow—notably the Rockies. The forecast for offshore Gulf of Mexico production shows a near continuous decline. In part this is one of the key reasons given by project sponsors for developing new LNG regasification terminals along the Gulf Coast. Building back deliverability means the existing network of offshore, onshore, and interstate pipelines will remain well utilized in the future.

Because onshore and offshore Gulf production represents nearly 36 percent of current domestic production, replacing part of that decline from other sources will be challenging. Furthermore, many of the offshore Gulf conventional wells lie in deep water and experience exponential decline rates, some with a first-year decline equaling 50 percent of the initial production rate. Thus, offshore Gulf production is relatively expensive.

Generation Fuel Demand

Since 1997, the overall gas consumption for U.S. end-use sectors—residential, commercial, industrial, and electric generators—has declined by 1.9 Bcf/d from 57 to 55.1 Bcf/d. Residential and commercial consumption combined, referred to as core demand, has remained relatively flat, while electric generation and industrial consumption materially have changed. Electric generation demand has grown by about 6 Bcf/d, a nearly 54 percent rise from 1997. However, the much more price-sensitive industrial-sector gas demand has fallen 22.5 percent from 23.3 to 18.0 Bcf/d (see Figure 6). One very bright spot, however, is the rapid growth in gas usage due to ethanol production.

Persistently high natural-gas market prices have meant price-sensitive gas load, especially industrial consumers for both energy and feedstock, have reduced (or curtailed) demand. Many feedstock users producing nitrogen and urea as well as other large petrochemical consumers cannot compete globally in many instances with competitors whose gas feedstock prices are less than $1/MMBtu, as is available elsewhere. This resulted in significant demand destruction between 1997 and 2006. The initial price shock, measured between 1997 through 2001, resulted in an initial 3 Bcf/d response. However, industrial gas demand has fallen further in the last two years. Overall industrial demand destruction now approaches 5.25 Bcf/d as hedges unwind and belief in the permanence of higher prices works its way through industrial users and industries move overseas (see Figure 7).

Annual consumer or “end use” natural-gas consumption in the United States might increase from roughly 22 Tcf in 2006 to about 32.7 Tcf by 2030, a 2 percent annual increase. Accommodating this sizable increase will require significant investments in natural-gas pipeline and storage infrastructure. The need is compounded further as LNG likely will come ashore in areas not currently able to accommodate and transport large gas volumes without significant new infrastructure.

Industrial demand slides from first place into second by 2012 (see Figure 8). This occurs in part because of the continued pressure price-sensitive industrial gas consumers will experience under forecasted gas prices, tempered by increased demand for natural gas by the ethanol production industry.

Continued demand growth for generation will place constant pressure on the supply industry’s ability to find and develop new sources of natural gas. Electric generators tend to be far less price sensitive than many industrial consumers. Fuel is only one of several components to the price of delivered power, so higher fuel prices will have a less than proportional impact on power prices.

Much of the gas demand growth in this sector is caused by the delayed impact of the electric-power overbuild. Currently, many power markets are overbuilt significantly with combined-cycle, gas-fired power plants. However, these plants increasingly will be used during the coming 5- to 10-year time frame due to continued electric-load growth, retirements of older existing plants, and new environmental restrictions (e.g., NOX, SO2, CO2, Hg), which will increase the cost of generation for some solid fuel and oil-fired generators.

For example, SO2 and NOX allowance prices were relatively very high from January 2005 through March 2007 (see Figure 9). During this period, older vintage residual oil-fired steam-turbine plant was significantly penalized in relation to a gas-fired steam plant of similar vintage. The penalty (an increased SO2/NOX adder in EPA state implementation plan [SIP] call states and just SO2 in other states) during this period varied between $0.30/dth and nearly $1.00/dth—a hefty amount. The increased penalty for burning residual fuel oil during 2006 (not just 1 percent sulfur but for other qualities such as 0.3 percent, 0.5 percent, 0.7 percent and 2 percent as well) in part caused greatly increased natural-gas usage for electric generation that year, taking over the number-two spot from nuclear power. In the future, CO2 adders likely will affect coal and oil plants similarly, being at a proportional disadvantage compared to natural-gas units.

Enjoy Scarcity While It Lasts

The market psychology has been extremely bullish for prices in recent months, as witnessed by the NYMEX gas forward strip as of mid-March 2007. The entire summer of 2007 NYMEX was more than $8/dth, with December 2007 through March 2008 prices near $10/dth. The run-up to record high world oil prices since 2003 boosted confidence in this view, as did the perception the Gulf of Mexico production region could be vulnerable to repeated hurricane damage. Analysis of the current period and the previous transforming market period confirmed:

• Crude-price Sympathy: A projected average 7.5-to-1 WTI/HH price ratio during this period maintains much of the increased value of gas relative to oil gained in the prior period;

• Security Premium: Petroleum prices include a premium for increased world tension and geopolitical security risks;

• The China Factor: Strong demand growth in several large developing countries, especially China and India, are driving higher global prices of hydrocarbon fuels, including crude-oil products, LNG and coal;

• E&P Inflation: Gas finding and development (F&D) costs are rising across North America;

• Gulf Depletion: Drilling in the Gulf of Mexico is declining despite record high prices; and

• Gas-Turbine Glut: The current generator overbuild is now resulting in predetermined increases in electric-power fuel demand.

The combination of these market factors has resulted in a permanent rise in gas prices, rather than prices spiking up and then returning to lower, long-run equilibrium levels. Crude-price sympathy, security and scarcity premiums, and a very bullish market perception have caused Henry Hub prices to trade well above long-run replacement cost. The price forecast for the next 48-month period takes these factors into account.

Recent oil and gas profitability is well above long-term industry norms, which supports the view that gas prices have risen significantly faster than increases in finding and developing costs. Oil and gas producers might be expected to continue increasing their capital-spending programs in the coming years in an attempt to grow production volumes. That growth will be tempered by several factors, namely: barriers to entry; the long lead times necessary to bring new gas reserves into production; and the high decline rate of both current producing wells and of expected lower-quality new wells. As a result, actual growth in production, if at all, likely will be gradual at 1 to 2 percent per year.