Mobile Edition
Print
Get the Mag
Weekly Updates1. INTRODUCTION
Natural gas markets have historically been regional with some linkage with international petroleum markets. However, as natural gas transportation capacity, including pipeline capacity and the capacity to produce, transport and receive liquefied natural gas (LNG), continues to expand, regional gas markets are increasingly becoming linked and moving toward a global market for natural gas. The linkages between natural gas and oil markets, and their future evolution, present a more complicated picture.
Oil and gas markets are linked in several ways. Natural gas competes with oil in several consumption sectors, making oil and gas substitutes. Natural gas liquids (NGLs) are produced alongside natural gas and are then stripped from the wet gas stream, with the NGLs sold into oil markets and the dry gas sold separately. The GTL process uses natural gas as a feedstock to produce low-sulfur diesel and other oil products. Further, oil and gas compete for exploration and drilling resources, and many current natural gas contracts base the contract price of natural gas on market oil prices. These last two linkages between oil and gas markets, however, are not accounted for in the INGM.
The INGM attempts to represent oil and gas competition in several demand sectors as well as the market affects of NGLs and GTLs. These multiple linkages between oil and gas markets lead to a complicated relationship between oil and gas prices in the model. Higher oil prices, generally, should stimulate increased consumption of natural gas in sectors where it can be substituted for oil. If this were the only way in which oil prices affected natural gas markets, one would expect the market price of gas to always rise in response to increased oil prices, as increased gas demand drives up the gas price and increases gas supply to meet the elevated demand, thus balancing the market. However, higher oil prices also lead to higher NGL prices and spur increased production of NGLs, and of their byproduct, dry natural gas.
Because of how NGLs link oil and gas markets, higher oil prices can lead natural gas prices to either increase or decrease depending on the regional gas supply/demand situation. If the elevated gas demand attributable to the higher oil prices is less than the increased gas supply attributable to the higher NGL prices, then for natural gas markets to balance, the gas price must come down to further spur demand. If the opposite is true, and the increases to gas demand are greater than to NGL and gas supply, then gas prices will rise in response to oil prices, as would generally be the case with substitute goods.
Higher oil prices can also spur increased demand for GTLs, thus elevating demand for natural gas as a feedstock for the process. This increased demand for natural gas can push up natural gas prices, leading to lower consumption in other sectors which then limits the upward pressure on gas prices, as consumers switch to competing fuels. However, the affect of GTLs on gas markets is limited by the extent to which global GTL capacity can reasonably expand.
GTL facilities require large upfront capital expenditures and potentially many years to recover the investment. Also, there are fewer commercial scale GTL facilities in the world and fewer under construction than there are LNG facilities, or other types of gas assets. This means there is less information on and greater risk surrounding what it takes to construct and bring online a commercial scale GTL facility. This also means there are fewer experienced personnel to complete such a job. Additionally, because of the significant energy losses in the GTL process, sustained low natural gas prices and high product prices are required to support the investment. The alternative GTL scenario is a response to these considerable risks and the significant affect of the GTL capacity expansion assumptions on the way the model responds to alternative oil price scenarios.
2. DESCRIPTION OF THE MODEL
The International Natural Gas Model (INGM) was developed to enhance the Energy Information Administration's (EIA) long-term assessment of world natural gas markets. The INGM simulates the natural gas and LNG markets from production, through processing and transportation, to end-use for 60 nodes which are detailed in Barden et. al, Table TA-1 and summarized in Table 1, below. The nodes are generally defined based on their current or potential significance in the world gas markets, as a major producer, consumer or transit location for natural gas. INGM uses a linear program (LP) to simulate gas markets with an objective of maximizing the cumulative discounted sum of producer and consumer surplus in developing the market equilibrium, capacity investment decisions, and capacity utilization (see Hogan, 2002). The INGM simulates demand for natural gas and utilization of the capacity in three seasons: Winter, Summer, and Spring/Fall. Utilization of capacity can vary between these seasons.
The key components of the model include natural gas and NGL resources by node, natural gas processing and transport capacity, and demand for natural gas and other fuels. Natural gas resources must be discovered and developed in INGM, creating reserves which then can be produced. The rate of discovery and development is a linear function of remaining economically recoverable resources, which is a function of the natural gas price and solved simultaneous with the natural gas prices. The percentage of economic resources that can be developed is constant over time. The base rate of production of reserves (the P/R ratio) is also constant over time but the model can increase the P/R ratio at a cost.
INGM accounts for removal of NGLs and the gas used in processing. The resulting dry gas can then be transported through a pipeline to a consuming node, converted to LNG, or consumed at the node. Natural gas that is transformed into LNG must, in an additional step, be transported by tanker along pre-defined tanker routes and then regasified for use or pipeline transport at the destination node. Dry natural gas can also be converted to a liquid fuel at a GTL plant. Finally, natural gas seasonal storage is simulated to address seasonal variations in regional production and consumption.
The capacity and utilization of the gas processing, LNG liquefaction, LNG regasification, tankers, pipelines, storage, and GTL facilities are all modeled similarly. Capacity must exist or be added before it can be utilized. Capacity additions incur initial development costs, annual fixed costs, and variable costs depending upon the utilization of the capacity. The period from the initial decision to add capacity until the first date of utilization of the capacity, is usually five to six years for land-based assets and four years for LNG ships but varies depending on the type of capacity. The investment and operating costs for LNG and GTL facilities are provided in Table 2. The utilization of the capacity will result in energy use/losses which vary considerably by the asset type and within the asset type (e.g. pipeline fuel use).
Capacity for any asset type will be added and utilized, endogenously, if the value of the capacity is equal to the discounted cost of constructing and operating the capacity, including the cost of the natural gas used as feedstock (GTL plants) or consumed by the asset. Once capacity is constructed, the utilization of the capacity depends on the variable operating costs including the cost of the natural gas fuel or feedstock. It is possible for capacity to be built, utilized for a number of years, and then left un-utilized due to changing economics.
Demand for NGLs and GTL products are assumed to be unconstrained globally but demand for natural gas in the power generation and end-use sectors are modeled using a logit formulation based on starting demand estimates provided by EIA from their WEPS+ modeling system. The INGM allows for interfuel competition using the following equation:
[S.sub.r,f,t] = [([P.sub.r,f,t] + [PA.sub.r,f]).sup.[alpha]] /[[SIGMA].sub.f], [([P.sub.r,f,t] + [PA.sub.r,f]).sup.[alpha]] (1)
Where [S.sub.r,f,t] is the share (fraction) of demand served by the fuel f in region r in year t, [P.sub.r,f,t] is the price of the fuel ($2006/GJ) for the region and year, [PA.sub.r] is a calibration variable for the region and fuel and reflects both the ability to use the fuel for the sector (e.g., natural gas in road transport) and the regional access to the fuel, and [alpha] is the price elasticity which takes the values of -0.88 for the residential sector, -1.22 for the commercial sector, -2.06 for the transport sector, -1.54 for industrial cogeneration, -0.92 for industrial feedstock uses, -0.66 for other industrial uses, and -2.00 for power generation sector. The high elasticity for the transport sector is offset by large regional price adjustments for natural gas for that sector.
3. DESCRIPTION OF THE CASES
The INGM reference case is based on the reference case in the EIA's International Energy Outlook 2008 (IEO2008), the U.S. Department of Energy's long-term assessment of world energy markets through 2030. The IEO2008 consumption projections were used as starting demand numbers for the INGM reference case. The world oil price path assumptions for the reference case are also based on the IEO2008 reference case. That said, the IEO2008 reference case oil price path was devised in mid-2007, when oil was trading at about $70 per barrel--substantially lower than the $140 per barrel observed only a year later. In the INGM reference case, the 2008 average nominal oil price is set at $100 per barrel, falling to $93 for 2009, and continuing to decline to $68 in 2016 before gradually increasing to a 2030 nominal price of $113 per barrel ($74 per barrel in real, inflation-adjusted 2006 U.S. dollars). The price trajectories for the three cases are shown in Figure 1 and in Barden et. al., Table TA-2.
FEED