Executive Summary

Natural gas prices rose dramatically in 2000 and have remained high through the first part of 2001. High prices have raised concerns about the longer term prospects for natural gas prices and their potential impact on consumers and on economic growth. Exacerbating those concerns are the current low levels of natural gas supply in storage. The central questions addressed in this report are “Why have natural gas prices risen so high and so quickly?” and “What is the outlook for the U.S. natural gas market in the short and mid-term?”

Natural gas represented 24 percent of the energy consumed and 27 percent¹ of the energy produced in the United States in 2000. The industrial sector was the largest user of natural gas—for cogeneration of electric power and as an industrial feedstock. In addition, natural gas is the largest energy source consumed in the residential sector and the fastest growing energy source for electricity generation.

In recent months, the high prices of natural gas used in the industrial, residential and commercial, and electricity generation sectors have caused exceptional public concern about the present and future operations of the natural gas industry and markets. The recent high prices have also prompted some policymakers to question whether natural gas can play a dominant role in fueling U.S. economic growth in the next 20 years. These concerns led Secretary Spencer Abraham to request that the Energy Information Administration (EIA) assess the recent trends in the U.S. natural gas market that led to high natural gas prices and evaluate the implications of those trends for the short- and mid-term outlook.

Why Have Natural Gas Prices Risen So High and So Quickly?

High natural gas prices, experienced in 2000 and expected to persist at least through 2001 and 2002, were caused by constrained domestic productive capacity² that resulted from a sustained period of relatively low oil and natural gas prices, followed by unusually high demand—the result of strong economic growth and an unusually warm summer and cold winter—and a poor storage position heading into the winter season (November 2000 through February 2001).

Low oil and natural gas prices for most of the decade before 2000 contributed to the limited natural gas productive capacity going into 2000. Annual average wellhead natural gas prices (in 1999 dollars) hovered between $1.61 per million Btu ($1.65 per thousand cubic feet) and $2.32 per million Btu ($2.38 per thousand cubic feet)³ through all of the 1990s, while crude oil prices (the composite refiners’ acquisition cost) ranged from $12.69 to $22.37 per barrel (excluding 1990). Oil and gas investments in exploration and production from 1990 through 1996 annually averaged $15 billion in real 1999 dollars, as compared with investments in excess of $30 billion annually (in 1999 dollars) before 1986. From 1986 to 1995, the average return on investment for major oil and gas companies⁴ ranged from 5.5 percent to 7.3 percent, except for 1990 when the return jumped to 9.5 percent as oil prices rose during the Persian Gulf war.⁵ These returns to investment were well under the range of 10.4 to 19.2 percent received between 1977 and 1985.

In 1996 and 1997, rising natural gas prices increased investment returns to over 10 percent, but in 1998, when natural gas prices fell below $2 per million Btu and oil prices were the lowest they had been in 25 years (in real terms), returns fell to 3.9 percent. Profits and returns on investments were considerably higher in 2000 as a result of the high oil and gas prices, and some of those revenues have been used to increase exploratory and developmental drilling. With the decline in industry investment and drilling during the 1990s, proved natural gas reserves declined from 169 trillion cubic feet at the end of 1990 to 164 trillion cubic feet at the start of 1999. More importantly, drilling levels were not sufficient to develop these reserves into increased productive capacity.

Except for 1994, when domestic production increased by 0.72 trillion cubic feet, annual U.S. production increased by less than 0.3 trillion cubic feet in every year during the 1990s. This was short of the average annual growth in demand during that period (Table ES1). In 2000, domestic gas production increased by 0.7 trillion cubic feet in response to the higher demand and higher wellhead prices.⁶ While this increase was sufficient to meet the major portion of demand growth seen in 2000, a large net drawdown of gas in storage and an increase in imports were also required to meet the remaining demand.

Sustained prices of about $2.25 per million Btu from 1994 to 1999 may have stimulated additional drilling and somewhat mitigated the tightened supply response that led to the jump in spot prices in 2000. For example, in 1996 natural gas prices rose by $0.60 to $2.21 per million Btu and the number of wells drilled increased by more than 10 percent over their 1995 level, and in 1997 prices rose by an additional $0.11 to $2.32 per million Btu and the number of wells increased by 26 percent above 1996.⁷ Gas drilling declined precipitously in 1991 and 1992, in 1994 and 1995, and again in 1999 as a result of relatively low prices, setting the stage for the tight supply situation that developed in 2000.

The prospects for adding significant amounts of new gas supplies from 2002 to 2005 look promising in view of expected natural gas prices. Natural Gas Week reports that U.S. contractors and service companies, pumped up by profits from current natural gas sales, “are flinging themselves into a headlong rush for rigs as the boom is beginning to take on fabled proportions.” First-quarter 2001 profits reported by Baker and Hughes rose by 600 percent over first-quarter 2000 profits, and Senior Vice President Andrew Sczescila predicted that 2001 would be the best year for service companies since 1981.⁸

Natural gas consumption increased by about 1 trillion cubic feet in 2000 because of strong economic growth and higher heating and cooling loads served by natural gas. As compared with 1.7-percent average annual growth in demand for natural gas from 1990 to 1999, demand jumped by 4.8 percent in 2000.⁹ More importantly, there was virtually no growth in gas consumption between 1996 and 1999, due in part to mild weather. Stronger demand was already evident in the spring, when natural gas demand would normally be expected to abate and prices to moderate significantly. Because natural gas prices began to rise in the spring of 2000, the refill of gas storage was slowed considerably as the industry waited for a possible return to lower prices. Gas storage injections were minimized as demand growth accelerated during the summer and gas acquisition costs escalated.

In the 6 weeks ending October 31, 2000, natural gas storage was aggressively filled to 2.7 trillion cubic feet. However, the additional demand for filling storage in the 6 weeks before winter only served to keep natural gas prices high, and the total amount of gas in storage at the start of the 2000-2001 heating season began at a 5-year low for that time of year. The storage situation was even worse for Southern California gas utilities served by El Paso Pipeline Company because of the rupture on the El Paso pipeline in New Mexico.¹⁰ Although interstate natural gas transmission capacity probably was adequate to meet normal peak demand with that pipeline in service, the pipeline rupture constrained California’s gas supply capacity. California’s environmental regulations on electricity generators also added to natural gas demand, because environmental emission allotments for other fuels were exhausted earlier in the year.

Thus, the U.S. natural gas market began the winter of 2000-2001 with high prices and a relatively weak storage position. Much colder than normal winter weather in November and December 2000 reduced gas stocks to such low levels that it raised concerns about possible supply shortfalls during peak periods for the remainder of the winter. The high natural gas demand and rapid gas stock drawdown strained U.S. productive capacity and drove up natural gas prices at the wellhead.

Although gas well completions have increased steadily since April 1999, production has not responded sufficiently to satisfy expanding market demand. The industry initially had to overcome the prior drilling slump associated with low natural gas prices. Despite this handicap, domestic production increased by about 0.7 trillion cubic feet in 2000, equivalent to about two-thirds of the increase in consumption from 1999 to 2000. Given an industry apparently pressing at the limits of its productive capacity, the higher demand did not bring about increased production, so prices rose higher.

Spot prices at Louisiana’s Henry Hub¹¹ (Figure ES1) were below $3 per million Btu until mid-April 2000, then broke the $4 barrier in late May as strong demand continued in the electricity generation sector. They remained above $5 per million Btu from September 2000 to February 2001 in response to aggressive filling of storage in the fall and later in response to high heating demand. The average wellhead price for the winter months, November 2000 through February 2001, was roughly 2.7 times higher than during the previous heating season, and the length of time for which spot gas prices have remained elevated is historically unprecedented.

At regional trading centers, average quarterly spot prices displayed unexpected price differentials from the average at the Henry Hub. The Henry Hub average rose from the third to the fourth quarter of 2000 and then changed little in the first quarter of 2001 (Table ES2). Although the pattern was similar for the regional trading centers, price differentials from the Henry Hub price varied significantly after the third quarter, especially for the California market. Because natural gas transmission rates are regulated, it appears that the significant variation in spot price differentials among the trading centers originated in the costs of unregulated bundled services (transmission plus fuel) provided by marketers.

Gas Pipeline and Distribution

Gas Distribution Systems and Intrastate Capacity

As the interstate and intrastate natural gas pipeline systems expand, LDCs may have to expand correspondingly. A substantial portion of the new pipeline capacity will provide additional delivery capacity to LDCs, which either are expanding their own capabilities to serve their existing service territories or are building new pipe segments to extend their systems into new neighborhoods or to serve new industrial or electric power customers.

LDCs continue to invest in new and replacement main and service lines and local compression facilities in order to satisfy the firm service requirements of their sales and transportation customers. According to the American Gas Association, construction projects by distribution companies totaled $9.7 billion (nominal) in 1998 and 1999, a 16-percent increase from $8.4 billion in 1996 and 1997.

An example of the market complications that can occur is provided by the recent developments in California. California is the Nation’s second-largest State market for natural gas and the tenth-largest producing State. In 1999, California’s natural gas demand (Table 2, Chapter 2) was met by 372 billion cubic feet from domestic production, 137 from storage and 1,800 billion cubic feet from interstate pipeline supplies, compared to an annual interstate delivery pipeline capacity of about 2.5 trillion cubic feet. On a peak-day basis,  interstate pipeline delivery capability into California is about 7 billion cubic feet per day while California’s ability to absorb natural gas  within the intrastate pipeline and distribution system (“take-away” capability) appears to be less, as low pressures and the inability to meet some interruptible gas load during peak periods indicate. Rapid gas demand and economic growth has evidently outstripped the rate of local infrastructure expansion and reinforcement required in some parts of California. Firm estimates await a more thorough investigation. However,  various sources have indicated the magnitude of the capacity shortfall likely is measured in hundreds of millions of cubic feet per day (MMcf/d).  One estimate for the total imbalance is about 300 MMcf/d.¹² However, estimates for specific border  crossings suggest a larger figure.

The two interstate crossings from Arizona into California are at a southern corridor crossing between Blythe, CA, and Ehrenberg, AZ, and at a more northern crossing between Needles, CA, and Topock, AZ.  Although the physical capability of the delivery point at Ehrenberg, AZ, could permit an estimated 1,410 MMcf/d to be delivered, the intrastate system can receive only 1,210 MMcf/d.¹³ The California Energy Commission estimated that the imbalance at the northern corridor crossing (Needles/Topock) is about 350 MMcf/d.¹⁴  These two estimates for the separate State border crossings combined indicate a potential shortfall in receipt capacity of 550 MMcf/d along the Arizona border.

Pipelines

The natural gas pipeline network has grown substantially since 1990, with more than 20 billion cubic feet per day of interregional capacity (a 27-percent increase) added through the end of 2000. The network has also become more interconnected, its routes more complex, and business operations more efficient. New types of facilities, such as market centers, and established operations, such as underground storage facilities, have become further integrated into the national pipeline grid, allowing the system to operate with greater flexibility and reliability. Except during periods of extreme weather conditions or disruptions caused by isolated pipeline outages, there has been no sustained disruption of the network since the mid-1970s.

Over the past 2 years, more than 60 natural gas pipeline construction projects (35 in 1999 and 28 in 2000) have been completed and placed in service in the United States.  These account for more than 12.3 billion cubic feet per day of new pipeline capacity, an increase of 15 percent over the capacity level in 1998.¹⁵ Since 1996, natural gas pipeline capacity has grown by more than 5 billion cubic feet per day annually in most years, totaling almost 30 billion cubic feet per day. Annual expenditures on pipeline development have exceeded $1.4 billion in most years.¹⁶

A major growth area in pipeline expansion during the past several years has been the import/export market for natural gas. Much of the pipeline construction of the past several years has been focused on expanding import capacity for Canadian gas into the U.S. Midwest and Northeast. The completion of the Maritimes and Northeast, Portland Gas Transmission, and Alliance Pipeline systems represented a 15-percent increase in overall natural gas import capacity since 1998: a 58-percent increase into the Central region (most destined for the Midwest) and a 23-percent increase into the Northeast. In addition, natural gas export capacity to Mexico has more than doubled since 1996. Export capacity to Mexico totaled 2.1 billion cubic feet per day at the end of 2000, compared with only 0.9 billion cubic feet per day in 1996.

Current pipeline capacity levels into the Midwest region were sufficient to meet 2000-2001 winter demand, even though the first 2 months of the heating season were colder than anticipated. Demand in the Midwest is still growing, however,¹⁷ and some of the capacity currently serving the region is expected to serve the Northeast in 2002. As a result, additional capacity to the Midwest region will be needed.

In most other parts of the country, immediate pipeline capacity limitations have not become apparent, although recent proposals to develop new pipeline capacity reflect a recognition that steady growth in natural gas demand is occurring. Florida, North Carolina, and South Carolina, for instance, have experienced significant growth in natural gas demand over the past decade, with sufficient additional pipeline capacity being installed to match the increase in demand.

What is the Outlook for the U.S. Natural Gas Market in the Short and Mid-Term?

Short-Term Outlook¹⁸

A major issue confronting the gas industry in 2001 will be the replenishment of gas storage to normal levels and the price implications of large net injections required during the April through October refill season. Given the low level of stocks at the end of the 2000-2001 heating season, net storage injections of about 2.0 trillion cubic feet will be required just to return to the level of 2.7 trillion cubic feet recorded for November 1, 2000. Total net storage injections during the 214-day fill period would need to be over 9 billion cubic feet per day or nearly 20 percent of daily gas deliveries to all consumers from April through October 2000, up from an average of 16 percent. The increased demand will continue to place upward pressure on natural gas prices in 2001.

Another issue will be the need to increase natural gas drilling and production. The cash flow from the sale of natural gas is an important determinant of drilling investments and has been a major factor in limiting increases in natural gas productive capacity, particularly from 1997 to 1999. Oil and gas investors do not initiate projects with long payback periods based on temporary price increases unless those prices are thought to be representative of a long-term market condition. Periodic downturns in the gas industry, such as in the 1984-89 and 1998-99 periods, trigger significant downsizing and cutbacks in spending for exploration and development of new gas sources. Reduced spending in these periods slows the construction of drilling rigs and other infrastructure needed to support future drilling, and results in downsizing and layoffs that reduce the industry’s ability to attract qualified new employees. In 2000, when the number of new gas well completions increased by almost 45 percent,¹⁹ gas production increased by an estimated 3.8 percent. The discrepancy reflects, in part, the lag in production following a shift in drilling (usually about 6 to 18 months) due to the time required to acquire necessary investment funds, install production equipment, and construct gathering lines in the field and pipelines needed for transportation.

Demand

The average growth rate for gas demand in the 2000-2002 time period is expected to be 3.4 percent per year, as compared with just 0.9 percent per year from 1994 to 1999. The next few years promise to provide an extraordinary boom in natural-gas-fired generating capacity additions, marked by the introduction into commercial service of about 22 gigawatts of new gas-fired capacity in 2000.²⁰ These additions contribute to expectations that natural gas will be the key fuel behind economic growth over the next few years.

Industrial Demand.²¹ Although natural gas prices remain high, industrial natural gas consumption is expected to increase by about 0.5 percent in 2001. In 2002, a strengthening recovery in natural-gas-intensive output (4.4 percent) and the prospect of lower average gas prices yields the expectation that industrial natural gas consumption will climb by about 2.5 percent.

Commercial Sector: Natural gas demand growth in the commercial sector in 2000²³ was more than 7 percent above the average annual rate observed during the 1986 to 1999 period and was generated by the combination of strong domestic economic growth and colder than normal weather. With normal weather assumed for the remainder of 2001 and 2002 and the growth rate for U.S. GDP expected to fall, gas consumption growth in 2001 is expected to slow to 3.5 percent. The combination of slower growth in commercial employment and output plus lower heating degree-days is expected to yield growth in natural gas for the commercial sector of about 1.1 percent in 2002.

Electricity Generation: The continuation of relatively high natural gas prices in 2001 points toward slower growth in demand for gas in the electricity generation sector. A rebound in economic growth and modestly declining gas prices result in renewed strength in expected growth in gas demand for electricity generation (12.4 percent) in 2002.

Prices

Given the outlook for robust growth in gas consumption over the next 2 years, prices at the wellhead will not soon be returning to the low $2 per million Btu experienced just a year ago. Although gas production and imports are expected to increase in the short term, gains in supply probably will not be enough to bring the wellhead price down significantly in the next 2 years.

Beyond the end of the 2000-2001 heating season, average wellhead prices are expected to decline somewhat, averaging near $4.38 per million Btu for the spring and summer. However, if the summer weather is unusually hot in regions that consume large quantities of gas-fired electricity (California and Texas, for example), injections into underground storage for the next winter could be strained, and prices could start rising more sharply and sooner than expected. For 2001, the annual average wellhead price is projected to be about $5.18 per thousand cubic feet ($4.85 per million Btu in 1999 dollars). In 2002 the storage situation is expected to improve modestly, and the average annual wellhead price is expected to decline to about $4.82 per thousand cubic feet ($4.43 per million Btu in 1999 dollars).

Mid-Term Outlook²⁴

The mid-term outlook for the U.S. natural gas market summarized in this report was developed from the Annual Energy Outlook 2001 (AEO2001), a mid-term annual energy-economy projection of U.S. energy markets developed using EIA’s National Energy Modeling System. The AEO2001 reference case assumes no change in current laws, regulations, or policies.

Because natural gas resources are expected to be adequate to meet future demand through 2020 and technological progress for exploration and development is expected to be sustained, natural gas prices are projected to return to a lower price path around 2005 and gradually increase to about $3.05 per million Btu in 2020. Advancing technologies are expected to offset some of the cost increases associated with harder-to-find natural gas pockets and smaller pools.

In the near term, however, natural gas prices are likely to be higher than projected in AEO2001. The higher near-term natural gas prices are expected to stimulate more non-gas-fired generation capacity between 2004 and 2010 than was anticipated in AEO2001. The expected surge in natural gas drilling activities, prompted by relatively high natural gas prices between 2000 and 2005, should add considerable natural gas productive capacity and increase proven reserves, making natural-gas-fired generating technology the preferred choice in the 2010-2020 time period.

Demand: In 2000, U.S. natural gas consumption of 22.8 trillion cubic feet accounted for almost 24 percent of domestic energy consumption. Natural gas consumption is expected to grow by 2.3 percent annually after 1999—faster than any other major fuel source—and is expected to reach 34.7 trillion cubic feet by 2020, mainly because of growth in natural-gas-fired electricity generation. More than half of the projected increase in consumption, which totals 13 trillion cubic feet, is expected in the electricity generation sector.

Supply and Prices: Domestic natural gas production is expected to increase more slowly than consumption over the forecast, from 19.3 trillion cubic feet in 2000 to 29.0 trillion cubic feet in 2020. Production over the forecast period is expected to total about 500 trillion cubic feet, or roughly 40 percent of the 1,281 trillion cubic feet of estimated recoverable resources as of the beginning of 1999. AEO2001 projects that the average wellhead price of natural gas produced between 1999 and 2020 will be less than $3.05 per million Btu (in 1999 dollars) over most of the forecast period. Like any commodity price, however, actual natural gas prices are likely to oscillate significantly around the trend line projected in AEO2001 as a result of business cycles in the industry, unusual seasonal temperature variations, or other special circumstances like pipeline ruptures—the kinds of events that have been experienced in the past 24 months.

Imports: Net natural gas imports are expected to grow in the forecast from 16 percent of total natural gas consumption in 1999 to 17 percent (5.8 trillion cubic feet), primarily from western Canada. Some new natural gas is also expected from Sable Island in the offshore Atlantic. Imports of liquefied natural gas (LNG) are expected to supply just 2 percent of U.S. natural gas consumption in the forecast, up from 0.6 percent percent in 2000.

Challenges Facing The Natural Gas Industry

Moderating the recurrence and severity of “boom and bust” cycles while meeting increasing demand at reasonable prices is one of the major challenges facing the U.S. natural gas industry today. The most serious short-term challenge is to increase production rapidly enough to satisfy natural gas demand at reasonable prices. This short-term challenge is inextricably woven into the investment cycles of the gas industry. Sustained high short-term natural gas prices can prompt significant new drilling investments and bring on new supply, but they can also prompt consumers to make potentially irreversible equipment investments and switch to lower cost fuel options. Both factors tend to put downward pressure on natural gas prices.

Attracting qualified personnel and natural gas drilling rig investments to meet expected demand growth is another challenge that may be difficult unless less risky but adequate long-term returns on investment can be achieved.  Recent events in the oil and gas industry have led some to question the industry’s ability to meet a projected 41-percent increase in domestic gas production by 2015. Periodic downturns in the gas industry, such as in the 1984-89 and 1998-99 periods, triggered significant downsizing and cutbacks in spending for exploration and development of new gas sources. Reduced spending slowed the construction of drilling rigs and other infrastructure needed to support future drilling, and continued downsizing and layoffs reduced the industry’s ability to attract qualified new employees.

Also, overcoming low production growth despite a large increase in well completions --because of smaller finds per well-- may be difficult to accomplish even with technological progress.  While the number of new gas well completions increased by almost 45 percent in 2000,²⁵ gas production increased by only 3.7 percent. The discrepancy reflects, in part, the lag in production following a shift in drilling (usually about 6 to 18 months).

Avoiding natural gas delivery system bottlenecks resulting from increased growth in natural gas demand over the past several years has increased utilization of pipelines and resulted in pressure for pipeline expansion in several areas of the country.²⁶ For instance, pipeline utilization levels in parts of the West (notably, pipelines delivering gas to the California market) have recently been well above 95 percent on a continuing basis. Further increases in demand could cause capacity bottlenecks to develop.²⁷ Growing gas service needs in the southern Nevada area also suggest the need for system expansion there.²⁸

Notes