How gas supply and price disruptions now outweigh oil imports as the nation's real energy problem.
During much of this past year, we have seen concern rising among government officials, members of Congress, media representatives, and financial and economic analysts about escalating crude oil prices. The price increases stemmed in part from earlier actions taken by 10 of the eleven members (i.e., excluding Iraq) of the Organization of Petroleum Exporting Countries (OPEC), to reduce supply to boost price levels that were believed to be unsustainably low. However, relatively little attention has been paid to a potentially more severe problem-the consequences of the natural gas drilling slump in the Unites States during the two prior years.
This drilling slump in 1998 and 1999 has caused the rapid run-up of natural gas prices seen of late. It also seems to have reduced deliverability in the lower 48 states, even though the most recent (December 1996) forecast by the Energy Information Administration (EIA) might have suggested otherwise.
The EIA had projected a lower-48 wellhead productive capacity of about 70 billion cubic feet (Bcf) per day, compared to actual dry gas production of 52 Bcf per day.1 Instead, rather surprisingly, the slump in gas drilling led to a rate of replacement of gas reserves in 1998 of only 83 percent. That low rate occurred even though the average active gas rig count for the year was near its high since 1990, and even though there were 12,106 gas well completions-the highest during the period 1990-1999 .
Then came 1999, when the much lower counts for active gas rigs and gas well completions were predicted to lead to even more disappointing results. However, as is shown in Table 1, reserve replacement was at a 10-year high-118 percent. That figure was reached because of record net revisions and adjustments, which offset the lowest total discoveries since 1994.2 In fact, this aberration may explain the gas deliverability problems that have occurred, and which likely will grow worse, because of the less-than-required rate of refill of seasonal underground storage that was to have been achieved by the end of the 2000 reinjection period on Oct. 31.
As of early October, refill was projected to total only about 2,700 Bcf of working gas, compared to the generally accepted minimum of 3,000 Bcf. The resulting gas price escalations will affect not only current gas users, including heating customers (who are far more numerous than oil heating customers). They also could have an impact on electric power supplies and prices, because nearly all urgently needed new generating capacity is gas-fired.
In short, the current preoccupation with high crude oil and refined products (gasoline, distillate fuel oil, etc.) prices diverts attention from other energy problems. At least as serious today is the tightness of natural gas supplies, caused by the drilling slump of 1998-1999, which has led to unprecedented price escalations and a substantial increase in longer-term price expectations.
Of course, it is true that crude oil acquisition costs for U.S. refiners rose from the unsustainable lows of $10 per barrel (bbl) seen in late 1998 and early 1999, to the low to middle $30s at their highs during fall 2000. In reality, however, the oil price increase at its most recent peak was still no higher than the increase in nominal (i.e., not deflated) terms experienced after the outbreak of the Iran-Iraq war in the fall of 1980. (And in deflated terms, the recent highs in crude oil prices were well below this earlier peak, although close to the fly-up following the invasion of Kuwait by Iraq in August 1990.)
By contrast, natural gas markets hold potentially wider ramifications. Gas supply and price disruptions affect not only gas users, but may delay urgently needed additions to electric generating capacity, for which gas-fired turbine systems are the only practical option, for economic and environmental reasons.
At Henry Hub, La., cash market prices for natural gas increased from roughly $2.50 per million Btu during mid-September to mid-October 1999, to more than $5.00 per million Btu during mid-September to mid-October 2000. The partially deregulated electricity market is also in turmoil, with firm on-peak wholesale prices remaining well above $100 per megawatt-hour (MWh) in a number of U.S. regions during the past three summers and reaching astronomical levels in some markets, such as Southern California. This development has precipitated controversial price caps in the $250 to $1,000 per megawatt-hour range by the new, federally regulated regional transmission organizations to protect consumers.
Overall, the gas industry must develop the infrastructure (including pipelines, storage, and new drilling technology) to serve what is likely to be a 25 percent to 30 percent increase of gas consumption just for power generation over the next 15 to 20 years. Nevertheless, in spite of the current supply and price problems, this growing reliance on natural gas is fully justified. Natural gas resources in the lower 48 states are more than adequate to make it the logical transition fuel to a sustainable energy system over the next 50 years or so, as long as policymakers will focus on the real energy problem.
The Oil Situation
The U.S. economy and individual consumers have had ample time to adjust to oil price fluctuations within the recent range. Nevertheless, we must expect continued volatility caused in part by OPEC actions.
In addition, there is the continuing threat of irrational action by Iraq, including cuts in its 2.5 million to 3.0 million bbl per day of production, or even renewed hostile actions against Kuwait. The current conflict between Israel and the Palestinian Authority and, possibly, also Lebanon and Syria, adds another potentially destabilizing factor-one that would have much more serious consequences than in the past. The reason is that spare productive capacity margins (except for the leading producer, Saudi Arabia) are low. A major influx of capital is needed to meet world demand, which has now risen to 68 million bbl per day of crude oil and 77 million bbl per day of total liquid fuels (including natural gas liquids, processing gains, and various additives).
By contrast, one positive element may develop from the slow recovery of production from within the former Soviet Union (FSU), but certainly not to the pre-breakup peak of 12 million bbl per day. Fortunately, North Sea production also continues to confound predictions of imminent decline and, in fact, has risen somewhat. At the same time however, OPEC production (excluding Iraq) still seems to linger only moderately above the June 2000 quota of 25.4 million bbl per day, in spite of unilateral increases by Saudi Arabia and the involuntary price band mechanism adopted in June 2000. This mechanism called for an increase of 500,000 bbl per day if the OPEC crude basket price stayed above $28 per bbl for more than 20 days.
Nevertheless, according to the EIA, average OPEC output during the first six months of 2000 was 28.3 million bbl per day (including Iraq) and 25.8 million bbl per day (excluding Iraq).3 Those figures are a little higher than the equivalent 1999 output reported by the EIA, but well below the 1998 output. Moreover, these data seem to disagree somewhat with other analyses that show OPEC output prior to its September 10-11, 2000 meeting in Vienna has been nearer 25.9 million bbl per day (excluding Iraq), or roughly 500,000 bbl per day above the quota. In fact, this variance may explain why the OPEC decision to increase production by 800,000 bbl per day effective Oct. 1, reached at its meeting in Vienna, had little impact on oil prices, as the decision may have been seen by the market only as largely validating current overproduction.
All in all, U.S. consumers so far have been able to deal with the escalation of gasoline and diesel and jet fuel prices. However, the predicted fly-up in distillate fuel oil prices for residential users into the $2 per gallon range during the winter of 2000-2001 has caused a very strong reaction and government intervention.
On Sept. 22, it was announced that 30 million barrels of crude oil would be released from the U.S. Strategic Petroleum Reserve (SPR), which now contains about 570 million barrels. This release is only a swap, however, to be replaced by 31.5 million a year later. The release was to take place this month at a rate of 1 million bbl per day. Already, by early October, the release had temporarily reduced crude oil prices somewhat. However, it is unlikely to produce any significant increase in the supply of distillate fuel oil and other premium distillates (gasoline and diesel and jet fuel), because U.S. refining capacity is nearly fully utilized and stocks are extremely low. Any increase in heating oil prices will, of course, be compounded by the unavoidable increase in natural gas prices to what will undoubtedly be record levels on a nominal basis.
Natural Gas: Now Linked With Power Markets
By contrast with crude oil, the current price and supply situation for natural gas appears more likely to have repercussions beyond the boundaries of gas markets, including the large number of politically sensitive heating customers.
The extent of these problems can be seen in gas prices in the NYMEX futures market (based on the Henry Hub, La., cash market). As of October, the price ranged as high as $5.50 to $5.60 per million Btu for December 2000 and January 2001. Those levels are roughly double what they were a year ago. In fact, there is only moderate relief even as far out as the winter of 2002-2003, when these futures prices are still about $4 per million Btu. This trend is not surprising, in spite of the dramatic increase in the active gas rig count from an average of 496 in 1999, to 810 in September 2000,4 because gas price expectations reflect the lag-time between drilling activities and actual production.
The only near-term solution to the electric power supply and price problems is more electric generating capacity. What is required is more than 100 gigawatts (GW) of gas-fired, simple- and combined-cycle combustion turbines. Merchant plant investors and other independent power producers would build these plants at strategic locations where there are constraints in grid-carrying capacity and nearby to natural gas pipelines and storage supplies. Through 2020, the EIA projects a total of about 260 GW of such new capacity, even without a decline in the more than 300 GW of coal-fired capacity and a decline of nuclear power capacity of only 40 GW.5 The huge orders seen so far from turbine manufacturers certainly support the high expectations for this strategy.
Over the longer term, smaller natural-gas-fired distributed generation and cogeneration options (i.e., in the single-digit kilowatt to 25-MW range) are expected to make a significant contribution to solving the power reliability and price-spike problems. However, at present, they still have relatively high investment costs and are disadvantaged by delivered natural gas prices substantially higher than those paid by larger power generators. For example, according to the EIA, even in the still "normal" year 1999, residential users paid an average of $6.62 per thousand cubic feet (Mcf) and commercial users $5.27 per thousand cubic feet, vs. $2.62 for electric utilities and $3.04 per thousand cubic feet for industrial users.6 (One Mcf corresponds to roughly 1 million Btu-actually 1.03 million Btu.)
This higher cost paid by retail gas customers could be especially problematical for early market penetration of consumer-owned generation options such as fuel cell systems and microturbine generators.7 It seems unlikely that commercial fuel cell systems, which of necessity must include a natural gas reforming and purification step, will be available at less than $1,500 per kilowatt in the foreseeable future. Their electric efficiency, because of the reforming step, is only about 40 percent (lower heating value basis). This corresponds to a heatrate (higher heating value basis) of about 9,500 Btu per kilowatt-hour. (The lower heating value of natural gas-reflecting no condensation of the water produced during combustion-is about 90 percent of the higher heating value on which gas prices and heatrates are based.) Just the gas cost differential of $4 per thousand cubic feet would therefore translate into a power cost disadvantage of nearly 4 cents per kilowatt-hour. Microturbines, which have more attractive installed costs ($600 to $700 per kilowatt), have an even lower electric efficiency of 30 percent, or a heatrate of 12,600 Btu per kilowatt-hour. Cogeneration can only partially offset these disadvantages compared to the installed cost of simple turbine generators ($300 to $400 per kilowatt) and of combined-cycle generators ($450 to $500 per kilowatt), and their heatrates (which run from 9,500 to 10,000 Btu per kilowatt-hour, and from 6,300 to 6,700 Btu per kilowatt-hour, respectively). Of course, these larger modular generation options also may require additional investments in electric grid capacity, unlike the smaller distributed generation technologies.
Scenario I: Gas Turbines as Peaking Plants
Let us do a "thought experiment" on how many hours per year a simple-cycle combustion turbine "peaker" costing $350 per kilowatt would have to operate to be profitable at a capital cost recovery factor of 15 percent per year and at power prices of $100, $250, $500, and $1,000 per megawatt-hour. Let us also assume a reasonably conservative levelized natural gas price of $4 per million Btu, a heatrate of 10,000 Btu per kilowatt-hour (38 percent efficiency, lower heating value basis), and 0.5 cents per kilowatt-hour ($5 per megawatt-hour) non-fuel operating and maintenance costs. Thus, we have fuel costs of $40 per megawatt-hour, annual capital recovery costs of $52,500 per megawatt of capacity, and O&M costs of $5 per megawatt-hour.
For example, if the peaker operated for 1,000 hours per year, or at an operating factor of 11.4 percent, generating 1,000 MWh per megawatt of capacity, the capital recovery cost would be $52.50 per megawatt-hour. Adding in fuel costs of $40 per megawatt-hour and O&M costs of $5 per megawatt-hour would yield a total of $97.50 per megawatt-hour in fixed and variable costs. In this scenario, a power price of $100 per megawatt-hour would be quite consistent with past practice of expecting about a 10 percent to15 percent annual load factor for peakers (operating during 10 percent to 15 percent of the 8,760 hours of nominal annual availability).
But consider a now-not-so-unusual, firm on-peak price index of $1,000 per megawatt-hour occurring during short periods in the summer in some of the market hubs. In this case, the required operating time to make such a peaker investment profitable shrinks to 55 hours per year as shown below.
$1000.00/MWh Less $ 45.00/MWh fuel and O&M costs $ 955.00/MWh available for capital recovery $52,500/MW-year = 55 hours/year $955/MWh
What is a reasonable strategy in this market environment for gas combustion turbines?
On one hand, it is doubtful whether any anyone would invest in peaking capacity to satisfy demand at some of the astronomical prices (up to $9,999 per megawatt-hour) experienced briefly in some markets. On the other hand, capacity designed to meet the current price caps of $250 to $500 per megawatt-hour could be profitable in areas where such fly-ups become endemic in the summer for just a few hundred hours of annual operation, as shown in Table 2. Thus, it might be justifiable to install gas-fired peakers in such areas, assuming that the needed gas supplies will available. However, while higher gas prices-say, up to $5 to $6 per million Btu-would not invalidate this conclusion, the assumption of gas availability in such substantial quantities for limited periods of time in the summer might not be realistic.
For example, a typical 150-MW, simple-cycle combustion turbine plant with a heatrate of 10,000 Btu per kilowatt-hour would require 36,000 Mcf of gas per day. Even for a large pipeline with 1 Bcf per day throughput, this required gas supply would correspond to 3.6 percent of the total. Consider that there might be several such plants needing this much gas in a relatively small area, and it is apparent that counting on such large volumes of reliable supply for relatively short periods at reasonable costs could create large contracting problems. Therefore, while this "thought experiment" is illuminating, it may not offer a practical response to even endemic price fly-ups into the range of $250-plus per megawatt-hour lasting for only a few days per year.
Scenario II: Combined-Cycle Turbines for Base Load
A more reasonable approach would be to plan to use simple-cycle combustion turbine capacity for more extended periods of firm on-peak prices of $100 per megawatt-hour (10 cents per kilowatt-hour) or more-i.e., in excess of 1,000 hours per year. In fact, some of the most advanced (aeroderivative) combustion turbine systems have electric efficiencies as high as 42 percent (on a lower heating value basis), or a heatrate of 9,000 Btu per kilowatt-hour. Their installed cost will be higher than that of the less-efficient systems, but they are clearly suitable for more extended operation.
Overall, however, for intermediate and baseload service, combined-cycle modules in which about two-thirds of the power is generated by one or more combustion turbines and the remainder in a steam turbine are most attractive. The steam turbine is operated on the output of a heat-recovery boiler, which utilizes the 900¼-1,000¼F flue gas from the combustion turbine(s). The economics of a typical 250-MW module with an installed cost of $500 per kilowatt in a baseload mode (85 percent annual operating factor) are shown in Table 3. A levelized gas price of $4 per million Btu is again assumed. Of course, only a year ago we assumed gas prices in the $2.50 per million Btu range and profitable operation at 3 cents per kilowatt-hour ($30 per megawatt-hour) compared to about 4 cents today. The electric efficiency of such combined-cycle plants has now risen to 57 percent to 60 percent (heatrates of 6,700 and 6,300 Btu per kilowatt-hour, respectively), which makes these plants somewhat less sensitive to gas price escalations.
So far it appears that the combined-cycle option makes up at least half of the large orders for delivery of turbine systems. However, even if gas prices do not temporarily derail the boom in turbine system capacity additions, the basic supply problem creates some risk. Construction of 100 GW of combined-cycle capacity over the next 15 years for baseload operation would increase gas consumption by nearly 5 trillion cubic feet (Tcf) per year. As recently as 1999, the EIA had projected an increase of 160 MW between 1997 and 2015, although in its 2000 projection, EIA reduced this to 105 GW between 1998 and 2015.8
In any event, the close relationship between resolving the current U.S. gas supply problem and solving the power supply and reliability problem is evident. There is, of course, plenty of justification for optimism. The independent and major producer segments are responding quite rationally to price expectations nearer to $4 per million Btu over the next several years, rather than the relatively unanimous predictions of forecasters as recently as December 1999 and January 2000 of wellhead prices in the range of $2.00 to $2.50 per million Btu. There is certainly no basic resource problem. Those gas reserves and resources in the lower 48 states that are technically recoverable total somewhere between 1,500 and 1,700 Tcf. It only takes continuing technology advances and fewer restrictions on where one can drill for these resources to make a major portion economically recoverable at prices that still make natural gas-fired power generation attractive. Substantial investments in more pipeline and storage capacity also are needed to ensure reliability of gas service for all users. Obviously, power prices from gas-fired generation cannot compete with the variable costs in the 1 to 2 cents per kilowatt-hour range of existing, substantially depreciated coal-fired and nuclear capacity. However, judging from the relatively high prices paid by independent power producers for this capacity as it is being divested by utilities, expectations are still high for meeting most of the new capacity needs with gas-fired turbine plants. That expectation remains despite the substantially higher variable costs of such plants (3 to 4 cents per kilowatt-hour, assuming a gas price of $4 per million Btu). In addition to the highest gas drilling rates since the late 1980s, other positive developments are substantial projected increases in Canadian and liquefied natural gas imports, currently at about 3.5 Tcf per year.
Lessons for Policymakers
Policy and other decision-makers must balance their current preoccupation with what has been a long-term problem of oil import dependence and the resulting large price fluctuations, with greater emphasis on the closely interrelated natural gas and power supply problems.
In particular, that means accepting the idea that price and supply disruptions in natural gas play represent the real energy problem facing the United States. This increased importance stems in part from partial electricity deregulation, which has shrunk generation reserve margins and has lowered incentives for investment in generation and transmission facilities.
Today it appears doubtful that there will be any significant investments in new coal-fired steam-electric plants (which now supply over half of U.S. power requirements), or in the still very costly "clean coal" technologies (which do not adequately address the problem of emissions of carbon dioxide and greenhouse gases). There are certainly no near-term prospects for construction of new nuclear power capacity, which still represents nearly 100 GW of the 800 GW of total utility and non-utility U.S. generating capacity. About 40 GW of this nuclear capacity is nominally slated for retirement by 2020, even though it produces no air pollutants and greenhouse gases.
Therefore, there is only one practical solution to growing regional shortages of power that have led to unacceptable periodic fly-ups in spot-market prices to levels of $250 to $1,000 per megawatt-hour and even higher. It is the construction by 2020 of a total of as much as 300 GW of gas-fired simple combustion turbine systems to meet peak and some intermediate load requirements and of highly efficient (up to 60 percent) combined-cycle, combustion/steam turbine systems for intermediate and baseload requirements.
This new capacity may require up to 7 Tcf of additional annual gas supply, compared to current annual consumption of about 22 Tcf. Such an increase entirely is feasible, although not at the $2 to $3 per million Btu wellhead price levels in constant dollars projected over the next 15 to 20 years until quite recently. The challenge is to ensure that this supply will materialize at prices that still make gas the logical fuel for new generating capacity and replacement of any existing coal-fired capacity, which cannot meet tightening environmental standards at acceptable costs.
1 "Natural Gas Productive Capacity for the Lower 48 States 1985 through 1997," Energy Information Administration, Document No. DOE/EIA-0542(97), December 1996.
2 "Advance Summary, U.S. Crude Oil, Natural Gas, and Natural Gas Liquids Reserves-1999 Annual Report," Energy Information Administration, Document No. DOE/EIA-0216(99) Advance Summary, September 2000.
3 "Monthly Energy Review," Energy Information Administration, Document No. DOE/EIA-0035(2000/09), September 2000. ("Monthly Energy Review," EIA.)
4 "Natural Gas Week," Vol. 16, No. 41, p. 13, Oct. 9, 2000.
5 "Annual Energy Outlook 2000," Energy Information Administration, Document No. DOE/EIA-0383(2000), December 1999. ("Annual Energy Outlook 2000," EIA.)
6 "Monthly Energy Review," EIA.
7 Linden, Henry R., "Fuel for Thought: Some Questions on the Future of Gas-Fired Generation," Public Utilities Fortnightly, Vol. 37, No. 22, pp. 26-35, December 1999.
8 "Annual Energy Outlook 2000," EIA. "Annual Energy Outlook 1999," Energy Information Administration, Document No. DOE/EIA-0383(99), December 1998.
Articles found on this page are available to Internet subscribers only. For more information about obtaining a username and password, please call our Customer Service Department at 1-800-368-5001.