While oil and gas prices now are falling after the latest experience with fuel-price volatility, the Global Energy Decision fuels team is focused on modeling an integrated world-wide system of fuel relationships encompassing crude oil, natural gas, coal, and increasingly, synfuels to help our clients assess the implications of fuel-price swings on their businesses. Let’s look at the potential impacts and implications of this growing reliance on liquefied natural gas (LNG) for North America’s power-generation demand.
Though adequate supplies of the fuel commodities are not in doubt, deliverability constraints on gas pipelines generally, railroads from coal basins, and pipelines from oil production facilities are increasingly troublesome, adding to market-price volatility of fuel commodities. With imported Canadian pipeline gas supplies (the United States’ traditional source of incremental supply) projected to decrease to provide for development of Alberta’s tar sands, LNG assumes a prominent role in planning for gas-demand growth.
New supplies of imported LNG are among the positive developments on the near-term horizon, offering the prospect of “floating pipelines” to boost gas-supply sufficiency. But these floating pipelines, depending on their level of contractual commitment, can change course to seek more attractive markets, and, as price chasers, uncommitted supplies of LNG introduce a whole new dynamic to fuel-price analysis.
Several factors determine the role and impact of LNG in the North American power market:
• Allocation and commitment of liquefaction and shipping capacity by the LNG producing world;
• Location and building of new regasification capacity in North America;
• Interchangeability of LNG treatment in FERC and state regulatory venues;
• Competition by LNG with conventional gas storage fields;
• Growth of an OPEC-like cartel for LNG; and
• Impact of LNG as a price taker, not a price maker, among fuels for power generation.
By 2029, our modeling suggests that North America may be importing more than 44 Bcf/d of LNG—a significant increase from the 2 Bcf/d of projected LNG imports during 2005. This rise in U.S. demand for LNG will require a corresponding increase in global liquefaction capacity. This capacity expansion will be spread out across various countries, including some that are already players in LNG exporting, like Trinidad and Tobago, Iran, and Russia.
The costs of both liquefaction and regasification facilities have declined significantly over the last decade. Costs of liquefaction have decreased by 35 to 50 percent and plant capital costs have declined from $500 per ton of annual liquefaction capacity to under $200 per ton.1 Despite the reduction to costs, meeting the demand needs of the North American market will require a considerable investment.
Based on current costs, the total necessary investment in new liquefaction capacity from now until 2020 to meet import demand into the North American market may reach $90 billion. Liquefaction investment between now and 2010 for North American deliveries is estimated at $20 billion. Post-2010 investment could be as high as $17 billion to $22 billion in today’s dollars.
Beyond liquefaction, a major investment in the LNG tanker fleet also will be required to accommodate LNG demand growth. Currently, approximately 159 tankers are involved in global LNG trade. To meet the 20 Bcf/d of expected North American LNG demand by 2020, more than 200 additional LNG tankers must be added to the existing fleet dedicated to North America.
In 2005, five regasification terminals were operating in the United States with a combined peak-withdrawal capacity of 4.2 Bcf/d, and expansion plans are under way for another 1.8 Bcf/d of peak capacity in 2006, for a total withdrawal rate of 6 Bcf/d. During 2005, this regasification capacity had a 50-percent use rate.
In addition, there are plans for 41 LNG import terminals that could increase maximum deliverability capacity to more than 48 Bcf/d. Twelve of those 41 terminals, with a total peak of 16 Bcf/d, have received final regulatory approval in Canada, the United States, and Mexico. Two-thirds of the maximum withdrawal capacity is in the Gulf Coast regions and will compete with flows from conventional storage, both salt cavern and depleted reservoirs.
In the near- to mid-term period, Global Energy’s long-term forecasts include the five operational U.S. facilities with expansions, two additional Gulf of Mexico offshore facilities, plus one Bahamas plant, one in Baja, a Nova Scotia facility, and three more in Freeport, Cameron, and Sabine. Throughput at these facilities is expected to rise significantly over the coming seven years.
Global Energy’s forecasted LNG imports could require regasification investments of more than $10 billion over the forecast horizon with half of the new regasification facilities likely to be located along the Gulf of Mexico (GOM) Coast, which is attractive for a number of reasons:
1. Relative ease of siting a new LNG facility in GOM compared with the East or West Coasts;
2. GOM already is industrialized and industrial friendly with low opposition; and
3. Abundance of natural-gas pipelines and infrastructure already built and in place helps significantly lower the cost of new projects.
These LNG facilities and their cryogenic storage with a projected 16 Bcf/d withdrawal capacity by 2020 will compete with conventional onshore gas- storage facilities including the Starks facility with a projected withdrawal capacity of 1.2 Bcf/d by 2009.
In addition, a smaller number of new facilities likely will be built in high demand locations like the Northeast and Southern California. Two good examples already are under construction—the Bear Head LNG facility in Nova Scotia, and the Energia Costa Azul LNG facility in Baja California, Mexico. They are located to serve both domestic and United States demand. Bear Head will transport gas into the supply constrained New England market, while Energia Costa Azul will deliver gas to Southern California. By siting facilities close to the U.S. border, these projects overcame the regulatory burden and public opposition that plagued U.S. projects.
Global Energy has modeled all the greenfield and brownfield expansion LNG projects currently under construction. LNG regas capacity is expected to reach 13 Bcf/d by 2009. However, capacity utilization at these facilities is forecast at approximately 50 percent of the maximum sendout rate. Our analysis delays adding LNG plants and expansion projects that are not already under construction (or operating) until 2012. This removes several proposed projects and one expansion plant with earlier announced online dates. LNG provided approximately 2.3 percent (1.35 Bcf/d) total domestic demand in 2005, and is projected to provide approximately 12.5 percent (8 Bcf/d) by 2025. Liquefaction capacity under contract for deliveries to the United States is reported to be approximately 8 Bcf/d by 2009, and indicates a continued low-utilization rate for regasification capacity.
Regasified LNG from cryogenic storage will compete with both salt-cavern and depleted reservoir storage for conventional supplies. LNG price structure and deliverability capacity will affect the value of existing conventional gas storage. LNG should improve the value of conventional depleted storage reservoirs by using them as restorage sites. Continuous regasification of LNG, which is controlled by limited cryogenic capacity and shipping schedules, will fill these storage sites to capture the intrinsic value related to seasonal-price spreads.
Existing gas storage facilities likely will face greater competition. The Energy Policy Act of 2005 gives FERC the right to grant market-based rate treatment for new natural-gas storage capacity. This will increase competition among storage facilities and lower the entry cost for new storage capacity owners to penetrate the market due to reduced levels of regulation and support for market-based rates. Most of these facilities will be located near or around existing pipeline header systems to take advantage of the pipeline connectivity. Most (55 percent) of them are also expected to be constructed from depleted oil and gas fields.
We also expect new LNG receiving terminals will include cryogenic storage tanks with 0.5 to 1 Bcf/d withdrawal rates for five to 10 consecutive days competing with existing on-shore storage. Underground storage facilities need certain geological formations and characteristics, but the LNG storage tanks sit above ground and do not require permissive geology. If situated in the same markets served by conventional storage, they directly will compete by causing the demand for existing storage to decrease due to more winter peak-day supplies and cryogenic storage.
Almost all presently planned LNG requires cryogenic storage. The amount is typically five to 10 consecutive days of maximum revaporization (withdrawal) rate and directly is tied to the vessel’s shipping schedule (generally 22 to 32 days depending on shipping distance from the liquefaction facility). Only offshore facilities that provide on-ship regasification and injection of gas into subsea salt caverns will have no cryogenic storage.
Most LNG supplies have been used for peak-shaving and load following, functions that compete directly with conventional storage, especially market-area storage. Current long-term LNG contracts have been focused on base-load supplies with moderate take-or-pay flexibility in order to follow seasonal load. Such base-load supply is likely to have some need for conventional storage (probably from depleted reservoirs after regasification). Such need could include blending/dilution of high-Btu gas if quality standards for transportation between cryogenic and conventional storage are maintained.
In addition, the Hackberry ruling by FERC exempts such facilities from open access and non-discriminatory operations, which has resulted in the utilization of cryogenic storage to avoid receipt of spot cargoes of LNG. Loss of spot LNG will reduce the demand for conventional storage of regasified LNG, and will increase the competition between cryogenic gas and conventional storage gas to capture short-term peaking demand. An offsetting factor might be that less spot base-load LNG gas will be competing for seasonal price differentials with conventional storage gas.
On a full-cycle cost basis (accounting for exploration, development, and production costs), LNG costs delivered to the regasification facility, also called the commodity, insurance and freight (CIF) cost, from many liquefaction sources presently are competitive with existing production from many traditional North American supply basins. Global Energy research indicates that regasified LNG, even from the most expensive and distant supply regions, has the potential to displace incremental Gulf of Mexico onshore and offshore production on both a full-cycle replacement-cost and on a marginal-cost basis.
Our research indicates that incremental indigenous GOM non-associated gas has a full-cycle replacement cost of $2.75 to $3.75/MMBtu—well below current market price. In comparison, regasified LNG has a full-cycle replacement cost ranging from $2.25/MMBtu to $2.75/MMBtu for the Caribbean to Gulf of Mexico, to $3.35/MMBtu to $4.15/MMBtu from Australia, Middle East, or Norway to GOM, based on wellhead net-back prices of $0.30/ MMBtu to$1.30/MMBtu.
Marginal cost is defined as the total full-cycle recovery cost (export wellhead to long-haul import pipeline) minus fixed capital cost (return on equity and cost of debt). The marginal cost of in-digenous gas from the GOM generally is 65 percent of the total cost into the interstate pipeline, compared with regasified LNG supplies at 50 percent. Marginal costs for the GOM are $1.80/ MMBtu to $2.50/MMBtu, compared with regasified LNG at approximately $0.90/MMBtu from the Caribbean to $1.75/MMBtu from Australia.
The market price of LNG is a contractually determined surrogate for the energy cost for the import country, except for Europe and the United States, with existing international trade on pipeline gas. The market price is tied to the competing power fuels, usually oil (both crude and products) and now coal, as seen by recent Indian and Chinese transactions. For the United States, liquefied gas will compete directly with indigenous pipeline gas. Even long-term contracts for base-load LNG supply have been tied to the short-term NYMEX market. A traditional long-term price structure based on discrete investments for upstream infrastructure has not yet been offered to U.S. “anchor” customers.
One misperception is that at some price level, LNG will flood into the North American gas market and set a cap on indigenous pipeline gas. This price level is based on the fact that the marginal cost—and for many of the downstream U.S. markets—the full-cycle cost of regasified LNG is less than into-pipeline marginal cost for incremental gas. This perception lacks an appreciation of the difference between cost-of-service and net-back pricing, as well as the role of competition in disciplining prices, particularly the asymmetric application of market competition for LNG.
The issue is not whether increased LNG imports into the United States will moderate gas prices. They obviously will as an incremental supply, but in net-back pricing, LNG is a “price taker,” and as such could establish a “floor” for pipeline gas, perhaps as a discount to NYMEX. How much lower will it be than the then-current cost for the next increment of indigenous gas, and when will that floor be effective?
After Elba Island shut down (because Sonatach’s long-term contract price at $2.00/MMBtu was greater than the current spot market for indigenous gas), most LNG sales have been short term (usually less than one year) and based on the spot market for indigenous gas (NYMEX Henry Hub and futures).
Never has a flood of LNG resulted in a ceiling or cap price for indigenous gas. Last year, LNG provided only 2 percent of total domestic demand, approximately half as downstream peak or swing supply. In North America, the distinction between net-back and cost-of-service pricing essentially has disappeared because of the highly competitive nature of its pipeline gas supply. North American gas supplies are priced as a competitive commodity; LNG supplies are not. The relatively low delivered and regasified cost of LNG—and more important, lower marginal cost as compared with GOM production—means that LNG sellers are likely to be price takers rather than price makers. Therefore, Global Energy’s forecast for imported LNG is infra-marginal; that is, LNG will be a price taker. LNG will be landed and regasified at or just below the marginal cost of incremental indigenous gas. Dedicated liquefaction and shipping capacity with contract prices linked to regionally competitive fuels will continue to set imported LNG prices.
Clarification: In the printed version of this article, Table 3 omitted Weaver’s Cove Energy. The table has been updated in this online version.