World oil prices in AEO2005 are set in an environment where the members of OPEC are assumed to act as the dominant producers, with lower production costs than other supply regions or countries. Non-OPEC oil producers are assumed to behave competitively, producing as much oil as they can profitability extract at the market price for oil. As a result, the OPEC member countries will be able effectively to set the price of oil when they can act in concert by varying their aggregate production. Alternatively, OPEC members could target a fixed level of production and let the world market determine the price. 

The behavior and ability of OPEC member countries to set the price of oil will be influenced by many factors about which there is considerable uncertainty. These factors include the forces that will drive world oil demand, such as the rate of economic growth in the developed and developing world and the degree to which oil demand is linked to economic growth. The behavior of each major non-OPEC producer and changes in technologies that use or find and extract oil also will be important. Each of these factors will also be influenced by the market strategy that the OPEC members choose for OPEC in the aggregate or for themselves. For example, a strategy targeting relatively low prices and high market share would reduce the risk that new oil conservation or development technologies might be developed. It also would reduce the incentive for individual OPEC members to exceed their output quotas and reduce the risk that world economic growth might be slowed. With such a strategy, OPEC members would face little risk of losing market power, but their revenues and profits would be relatively low. 

Conversely, if OPEC members jointly limited production to maintain high prices and low market share, new oil conservation or exploration and production technologies might be developed. Such a strategy would also increase the incentive for individual OPEC members to exceed their output quotas, cause importing countries to enact oil consumption reduction policies, and increase the likelihood that world economic growth would be slowed. While this strategy could result in relatively high revenues and profits in the short term, it would also be a relatively high-risk strategy. 

Approach 

The AEO develops world oil price scenarios through an iterative process that examines the reasonableness of candidate oil price paths and their impacts on world oil supply and demand. The AEO process also considers the stated OPEC basket price target range, as well as ongoing discussions among OPEC members regarding possible changes to it. 

The AEO2005 reference case assumes a moderate market strategy between low-price, low-risk market share maximization and high-price, high-risk profit maximization. Alternative cases, in which different oil market behaviors are assumed, are also considered in AEO2005, including the October oil futures case, high A and B world oil price cases, and a low world oil price case. As with all of the projections in AEO2005, the oil price forecasts do not represent an assessment of what will happen, but rather, an assessment of what might happen under various scenarios. Higher or lower price paths are possible, and short-term price volatility in oil markets, which AEO scenarios do not attempt to model, is likely to continue. 

World Oil Demand. Key inputs for projecting world oil demand—for example, the worldwide demand for various energy services (heating, cooling, transportation, etc.)—are estimated using EIA’s System for Analysis of Global Energy Markets (SAGE) [74]. SAGE is an integrated set of regional models that provides a technology-rich basis for estimating regional energy supply and demand. For each region, estimates of end-use energy service demands (e.g., car, commercial truck, and heavy truck road travel; residential lighting; steam heat requirements in the paper industry; etc.) are developed on the basis of economic and demographic projections. Projections of energy demand are estimated on the basis of each region’s existing energy use patterns, the existing stock of energy-using equipment, and the characteristics of available new technologies, as well as new sources of primary energy supply. 

While oil products are used for many energy services (i.e., heating, steam generation, electricity generation, etc.) and as industrial feedstocks, the major use of petroleum products is for transportation. As a result, the worldwide demand for transportation services is the key driver for oil demand. In turn, the demand for transportation services in the various regions and countries represented in SAGE is driven by the projected level of income per capita, complemented by other important region-specific factors, such as the state of the transportation infrastructure. For the industrialized countries with well-developed transportation networks, demand for transportation services is influenced primarily by projected income levels and lifestyles; for developing countries, the lack of transportation infrastructure can be a significant constraint. 

Table 16 summarizes by region and country the projected average annual growth rates for real GDP and oil demand, and the resulting oil intensity, in the AEO2005 reference case from 2003 to 2025 [75]. The table also shows region and country shares of world GDP and oil demand in 2003 and 2025. As shown, total world GDP is projected to grow at an average annual rate of 3.1 percent, with the developing and former Soviet Union (FSU) countries generally projected to grow at higher rates, while the industrialized countries generally grow at slower rates. Total world oil demand is projected to grow more slowly, at 1.9 percent annually. World oil intensity declines by 1.2 percent per year. 

Because of the differences in projected growth rates for GDP and oil demand, the developing countries are expected to play a growing role in the world economy and oil markets. In 2003, the industrialized countries accounted for 77 percent of world GDP and 57 percent of total world oil consumption. It is projected that in 2025 real GDP in industrialized countries will account for 68 percent of world GDP and 48 percent of total oil demand. In contrast, developing countries are projected to account for 28 percent of world GDP in 2025, up from 20 percent in 2003. Similarly, oil demand in developing countries is projected to account for 45 percent of world oil demand in 2025, up from 36 percent in 2003. 

The projected growing role of the developing countries in the world economy and oil markets makes understanding the impact of economic growth on oil demand critically important. The sensitivity of oil demand to income is often characterized by what economists refer to as the income elasticity of demand, defined as the percentage change in oil demand with respect to the percentage change in real income. A rough approximation of the relative sizes of income elasticities for the different countries and regions represented in SAGE can be calculated from Table 16 by dividing the 2003 to 2025 average annual growth in oil demand by the average annual growth in real GDP. This calculation yields an income elasticity of demand of approximately 0.6 for the developing countries, compared with 0.4 for the industrialized countries [76]. 

The implication that oil demand in developing countries will be more responsive to changes in economic and income growth is consistent with research, but there is a great deal of uncertainty about the level of response. The response of oil demand to income growth and changes in oil prices has been examined in a number of empirical studies. The estimates of income elasticities in those studies vary widely, depending on the time period under study, the groups of countries considered, and the econometric specifications used [77]. Although the empirical evidence is not conclusive, and the magnitude of income elasticity estimates varies widely, most studies have found that developing countries generally have higher income elasticities than the industrialized economies. 

Studies have shown both greater and smaller responses in developing countries than is reflected in SAGE. For example, Gately and Huntington found that the income elasticity of demand for oil in developing countries ranged from 0.5 to 1.0, depending on the groups of developing countries being considered [78]. The Gately and Huntington study, as well as most other empirical studies, used historical data and employed a single-equation reduced-form framework relating oil demand changes to changes in income, or income per capita, and oil prices in various lag formulations. 

Such formulations may not fully capture the changes that have occurred in world economies or technologies in recent years, nor reflect how these changes might affect the future. For example, in an era of increased globalization and rapid technology transfer across countries, empirical estimates derived from historical data and simplified model formulations may not fully capture the more rapid transfer of new, efficient technologies from the industrialized countries to the developing countries that is likely to occur in the future. In contrast, the inferred income elasticities approximated in this report are based on projections coming from a structural model that explicitly incorporates the technical and cost relationships projected to exist between energy service demands by end-use sectors and the supply of energy. The model also represents region-specific factors that may encourage or inhibit demand for oil, such as transportation infrastructure constraints that are likely to arise as developing economies grow. One key assumption is that vehicles sold in both developing and industrialized countries in the future will be more fuel efficient than they were in the past. 

World Oil Supply. Once oil demand has been estimated by region and country, the levels of regional non-OPEC conventional and nonconventional oil production are developed to be consistent with the assumed world oil price path and assumptions regarding proved oil reserves, undiscovered oil, and reserve growth. The gap between projected world oil consumption and non-OPEC oil production determines the call on OPEC producers. Production from individual OPEC suppliers is estimated based on information regarding proved reserves, project development schedules, long-term development plans, and production economics in each country or region. Production capacity estimates reflect both projected levels of supply and historical utilization rates. Several Persian Gulf OPEC producers, including Saudi Arabia, Kuwait, and the United Arab Emirates, are assumed to have production capacity utilization rates of 90 to 95 percent, while non-OPEC producers are assumed to use all of their capacity. Other OPEC producers are assumed to fall between these extremes. 

The growth in non-OPEC oil supplies has played a significant role in the erosion of OPEC’s market share over the past three decades, as non-OPEC supply has become increasingly diverse. North America dominated growth in non-OPEC supply in the early 1970s, the North Sea and Mexico evolved as major producers in the 1980s, and much of the new production since the 1990s has come from Latin America, West Africa, and the former Soviet Union. Non-OPEC supply from proved reserves is expected to increase steadily from 48.8 million barrels per day in 2003 to 65.0 million barrels per day in 2025 in the reference case. 

The expectation in the late 1980s and early 1990s was that non-OPEC production in the longer term would stagnate or decline gradually in response to resource constraints. The relatively low cost of developing oil resources in OPEC countries (especially those in the Persian Gulf region) was considered such an overwhelming advantage that non-OPEC production potential was viewed with considerable pessimism. In actuality, however, despite several periods of relatively low prices, non-OPEC production has risen every year since 1993, growing by more than 8.2 million barrels per day between 1993 and 2003. Three factors are generally given credit for the impressive resiliency of non-OPEC production: development of new exploration and production technologies, efforts by the oil industry to reduce costs, and efforts by governments in non-OPEC countries to promote exploration and development by encouraging outside investors with attractive financial terms. 

It is expected that oil prices will remain high enough that non-OPEC producers will be able to continue to increase output profitably, producing an additional 6.8 million barrels per day by 2010 in the reference case when compared with 2003. Much of the increased non-OPEC production is expected to come from Africa and Central and South America. 

No one doubts that fossil fuels are subject to depletion and that depletion leads to scarcity, which in turn leads to higher prices; however, there are many resources that are not heavily exploited because they cannot be produced economically at low prices and with existing technologies. With higher prices, the development of such resources could become profitable. Ultimately, a combination of escalating prices and technological enhancements can make more resources economical. Much of the pessimism about oil resources has been focused entirely on conventional resources. However, there are substantial nonconventional resources, including production from oil sands, ultra-heavy oils, gas-to-liquids technologies, coal-to-liquids technologies, biofuel technologies, and shale oil, which can serve as a buffer against prolonged periods of very high oil prices. Total nonconventional liquids production in 2025 is projected to be 5.7 million barrels per day in the reference case, up from 1.8 million barrels per day in 2003. 

figure data

figure data

Comparison of Projections 

The world oil price cases in AEO2005 are designed to address the uncertainty about the market behavior of OPEC. They are not intended to span the full range of possible outcomes. The cases are defined as follows: 

• Reference case. Prices in 2010 are projected to be about $10 per barrel lower than current prices (2003 dollars) as both OPEC and non-OPEC producers add new production capacity over the next 5 years. After 2010, oil prices are projected to rise by about 1.3 percent per year, to more than $30 per barrel in 2025. 
• October oil futures case. Prices in the near term rise through 2005, and then resume a growth trend similar to the reference case. The results of this case, which are similar to the reference case in the long term, are compared with the reference case results in the text box on "The October oil futures case". 
• High A world oil price case. Prices are projected to remain at about $34 per barrel through 2015 and then increase on average by 1.4 percent per year, to more than $39 per barrel in 2025. 
• High B world oil price case. Projected prices continue to increase through 2005 to $44 per barrel, fall to $37 in 2010, and rise to $48 per barrel in 2025. 
• Low world oil price case. Prices are projected to decline from their high in 2004 to $21 per barrel in 2009 and to remain at that level out to 2025. 

World oil price projections in the five cases are shown in Figure 12. A detailed tabular summary and comparison of each of the oil price cases with the reference case is provided in Appendixes C and D. 

Reference World Oil Price Case. In the reference case, the assumption is that the OPEC members will continue to demonstrate a disciplined production approach that reflects a strategy of price defense in which the larger producers are willing to increase or decrease production levels to maintain fairly stable prices (in real dollar terms) to discourage the development of alternative crude oil supplies or energy sources, allow for continued robust worldwide economic growth, and maintain compliance with quotas, particularly by smaller OPEC producers. It is also assumed that OPEC producers will achieve sufficient oil revenues to expand production capacity enough to keep prices in a range of $27 to $30 per barrel in 2003 dollars, near the high end of the current OPEC price target range. Their current level of proven reserves (870 billion barrels) is sufficient to meet the implied production levels.

In the medium term, there is enough resource potential in non-OPEC countries to allow non-OPEC oil production to continue growing. Over the longer term, it is estimated that it will be harder for non-OPEC producers to continue to increase production. Assuming reference case prices, the search for alternatives and unconventional liquids will be limited, while demand will continue to grow. Therefore, OPEC members will have to make up the production difference (Figure 13). To satisfy the remaining global demand for oil at the given reference case prices, OPEC production will have to increase from 30.6 million barrels per day to 55.1 million barrels per day, an average annual increase in production of 2.7 percent. This is projected to result in an increase in OPEC’s market share from 39 percent in 2003 to 46 percent in 2025, as cheaper sources of non-OPEC oil are depleted.

Table 17 summarizes the main features of the reference case in terms of cumulative production volumes, cumulative revenues, and the sum of the discounted cumulative revenues (at a 5-percent discount rate) from 2003 to 2025 [79]. The OPEC and non-OPEC countries are aggregated by major regions.

The reasoning behind the assumed prices and production patterns in the reference case can be questioned. If OPEC members have sufficient market power and cohesiveness to set world prices, why would they not try to set higher oil prices? If OPEC comprised a group of producer countries with similar oil reserves, resource depletion time horizons, geopolitical concerns, and no fear of alternatives to oil at higher prices, then a more limited production strategy that maximizes economic profits in the short to medium term would appear more plausible. In the absence of these conditions, however, and given the difficulty of enforcing tight production goals to limit output, a reasonable strategy is to maintain stable prices that discourage oil alternatives while limiting the risk that member countries will exceed their quotas.

Another issue is whether OPEC members will be able to finance the investments needed to expand their output as projected in the reference case. While some OPEC producer countries are currently closed to foreign involvement in the exploration and development of oil resources, it is expected that they will be able to attract foreign capital, if needed, while retaining sovereignty over their energy resources. The markets for financial capital have provided sufficient resources in similar situations in the past, especially when there are strong incentives from both the demand and supply sides. The current experience of China, which did not attract much financial capital in the past, is an example of what can happen with the appropriate economic incentives or when the motivations are strong. Other historical examples include the flow of foreign capital to Latin America in the 1980s and East Asia in the 1990s.

There are also factors that may encourage countries in the Middle East to open up their energy sectors to foreign participation in one form or another. For example, Saudi Arabia, for some time now, has been lobbying to gain admission to the World Trade Organization. One of the conditions that Saudi Arabia needs to fulfill to gain entry is to open up its economy, especially its financial markets. The opening up of the United Arab Emirates to foreign financial capital and its creation of an export trade zone provide another example of how the economic environment can change.

High A World Oil Price Case. In the high A world oil price case, the OPEC countries in aggregate are assumed to maintain a relatively constant share of the world oil market. There are a number of ways that a constant market share for the OPEC countries might result over the projection period. First, more cohesion among OPEC members could begin to place greater emphasis on short-term profit maximization, with more control on member output, as might occur if a mechanism were devised to enable stricter enforcement of quotas. This cohesion might be reinforced by a perception that the incremental non-OPEC oil resource development costs are quite high and that the resource base is limited, and thus that there is less risk from non-OPEC producers in the long term. Second, some large producer countries in OPEC might not be able to finance sufficient development and enlargement of productive capacity because of competing social infrastructure demands on government budgets.

In this case, the world oil price would tend to reflect the projected incremental cost of non-OPEC oil and rise faster than in the reference case—from about $28 per barrel in 2003 to more than $39 per barrel in 2025 in real terms, an average annual increase of 1.6 percent from 2003 to 2025. As a result of higher world oil prices, world oil demand in 2025 is projected to be lower in the high A world oil price case than in the reference case (115 million barrels per day and 120 million barrels per day, respectively). Table 18 summarizes the main features of the high A world oil price case.

For OPEC members, cumulative production of almost 280 billion barrels in the high A world oil price case is projected to bring in $9.9 trillion (in 2003 dollars), as compared with cumulative production of 343 billion barrels and revenues of $9.7 trillion in the reference case. Although the high A world oil price case appears to be more attractive to OPEC producers than the reference case in terms of economic profits, the sustainability of the higher prices over the projection period is uncertain. Higher prices would create greater incentive for OPEC countries to exceed quotas, greater likelihood of increased conventional and unconventional oil production in non-OPEC countries, and greater possibility of increased conservation measures in oil-consuming countries, induced both by higher prices and by public policy measures.

While the AEO cases are developed under the assumption of unchanged policy in consuming countries, major oil exporters may expect that higher prices would spur policy responses in oil-importing nations. Based on these considerations, economically rational producers would be likely to apply higher discount rates when evaluating the revenue stream associated with the high A world oil price case than that associated with the reference case. Taking this difference into account, key OPEC producers might accept the reference price case.

High B World Oil Price Case. There is a great deal of uncertainty about the size and availability of crude oil resources, particularly conventional resources, the adequacy of investment capital, and geopolitical trends. While the high A world oil price case tries to reflect the uncertainty in some of these variables, some analysts argue that the higher prices seen in recent years will be sustained and represent a fundamental change in the market. The high B world oil price case was completed to evaluate the impact of world oil prices that remain close to current levels for the foreseeable future.

The high B world oil price case assumes a continued rise in prices through 2005, followed by a gradual decline to 2010 and then strong increases through 2025. The near-term prices reflect the trends observed in oil futures on the NYMEX for WTI during October 2004, where crude oil futures prices exceeded 2004 levels in 2005 before falling back somewhat, but to levels well above those projected in the AEO2005 reference case. The world oil price in the high B case is assumed to be $2 higher than in the reference case in 2004, or $37 per barrel, to grow to about $44 per barrel in 2005 before falling to $37 in 2010, and then to rise to $48 per barrel in 2025, compared with $30 in the reference case and $39 in the high A world oil price case.

The high B world oil price case reflects an assumption that OPEC producers will be less able or willing to expand their productive capacity and that their output growth will be constrained considerably (Table 19). As a result, the OPEC members are projected to lose market share over time, in contrast to the high A world oil price case, where their market share remains constant over time. OPEC member country production is projected to grow from 30.6 million barrels per day in 2003 to 36.6 million barrels per day in 2025, compared with 55.1 million barrels per day in the reference case and 42.4 million barrels per day in the high A world oil price case. The worldwide impacts on energy supply in the high B case are more uncertain because of limited experience with sustained periods of high world oil prices. Nevertheless, roughly one-half of the difference between OPEC member country production in the reference and high B world oil price cases is projected to be made up for by non-OPEC countries (Figure 14). The remaining difference reflects the reduction in oil demand resulting from higher prices, as well as increased production of synthetic oil from coal and natural gas and nonconventional liquids.

figure data

Undiscounted cumulative revenues from OPEC member country production in the high B world oil price case exceed those in the reference and high A world oil price cases, despite lower production; however, the high B case is projected to result in significant impacts on world energy demand and alternative sources of supply, including increased production from synthetic fuels. In addition, strong cohesiveness among OPEC members would be required to maintain the strict production quotas implicit in the high B case. As a result, the uncertainty and risk associated with this case for individual OPEC members suggest that a higher rate is appropriate for discounting the projected revenue stream.

The projections in the high B world oil price and reference cases are compared in the text on "Comparison of projections in the reference and high B world oil price cases". It is important to stress the uncertainties and limitations of this case. The market conditions in the high B world oil price case fall outside the range of experience best represented in NEMS. In particular, some of the modeling uncertainties and limitations about the case are as follows:

• The level of economic production of oil from both conventional sources and unconventional sources (such as oil sands) is subject to considerable uncertainty, particularly with sustained oil prices at much higher levels than in the reference case.
• The effects of global competition for natural gas through pipelines, LNG, and gas-to-liquids (GTL) are highly uncertain in an environment of high sustained oil prices. For example, stranded gas (gas production at sites without access to pipelines) that might otherwise be economical to export as LNG could potentially become economical to process as GTL. These impacts on world natural gas supply cannot be evaluated endogenously with the present versions of EIA’s U.S. and global energy models; however, an adjustment to the assumed cost profile of LNG imports to the United States has been incorporated to reflect the potential market impact. As model development is able to continue, additional analytical capability in this area would be a high priority.
• Prospects for synthetic petroleum—GTL and coal-to-liquids (CTL) may be constrained by plant siting issues that have not been investigated, such as waste disposal and limited water supplies.
• The worldwide economic and political response to a regime of prolonged high oil prices is uncertain, as is the long-term effect on domestic economic growth.
• EIA’s modeling of petroleum consumption reflects observed patterns of use and consumer preferences, as well as existing and foreseeable technologies. Consumer and manufacturer behavior in the face of sustained high oil prices may depart from the patterns on which the model is based. For example, there could be shifts to smaller, more efficient vehicles, more penetration of alternative-fuel vehicles, and a shift in the demand for vehicular travel to other travel modes, such as from truck to rail freight.
• High world oil prices and high natural gas prices may spur unforeseen technological innovation and adoption, but quantifying these possibilities remains a challenge.

Low World Oil Price Case. The low world oil price case reflects a future market where all oil production becomes more competitive and plentiful. There are several ways in which this could come about. First, the OPEC countries could become less cohesive, with each producer attempting to sell as much of its productive capacity as the market will allow. In this sense, the low world oil price case is exactly the opposite of the high A world oil price case. Another possibility would be a decline in the costs of non-OPEC oil production or the viable development of competitive alternatives. To forestall the penetration of alternatives and other sources of competition, OPEC would lower its price band and increase production.

The world oil price (in 2003 dollars) is projected to decline from about $28 per barrel in 2003 to $21 per barrel in 2009 in the low world oil price case, and to stay at that level through 2025. As a result of increased competition between OPEC members or a conscious attempt to increase market share, the market share of OPEC’s member countries increases from 39 percent in 2003 to 51 percent in 2025. Within OPEC, nearly all producers, except for Indonesia, which has limited remaining resources, are projected to increase production at an average annual rate of 3 percent or higher over the 2003 to 2025 period. The average annual growth in production by OPEC members over the same period is 3.5 percent. The low world oil prices in this case cause world oil demand to increase from 80 million barrels per day in 2003 to 128 million barrels per day in 2025, an average annual increase of 2.2 percent.

Given the projected state of technology, projected reserves, and their relatively higher cost structures, non-OPEC producers would be expected to increase output at a slower rate in the low world oil price case than in the reference case (Figure 14). Starting from a production level of 49 million barrels per day in 2003, non-OPEC oil output is projected to grow at an average annual rate of 1.1 percent in the low price case, to 62 million barrels per day in 2025. Table 20 summarizes the main features of the low world oil price case.

The low oil price case is the most favorable of the AEO2005 oil price cases in terms of economic welfare, because the world oil price is projected to be closer to its marginal cost. It is less favorable, however, from the producers’ point of view. Relative to the reference case, OPEC members would end up producing 11 percent more oil over the 2003 to 2025 period and earning roughly 11 percent less in cumulative revenues. Further, with a decline in oil prices there would be less exploration activity at the margin, a tendency for more cohesion in OPEC, and lower penetration of alternative fuels.

Notes and Sources

Contact: Nassir Khilji
Phone: 202-586-1294
E-mail: nassir.khilji@eia.doe.gov