Reducing the Demand for Oil

Abstract

Efforts to fully price oil and mitigate climate change would reduce the demand for oil, which would reduce oil MENA’s ability to control oil prices and maintain a stable level of oil revenues. This chapter considers what would happen if the United States fully priced gasoline as a case study and if the world mitigated climate change by extracting a limited amount of oil over time.

Fully pricing gasoline. Currently, automobile users do not fully pay the cost of maintaining highways, the social cost of pollution and carbon, and the military cost of maintaining access to Middle East oil. We estimate that if these costs were added to the price of gasoline, it would increase by over $1.75 per gallon, and the quantity demanded for gasoline would be reduced by about 19%, which is approximately what the United States imports from the Middle East.

Mitigating climate change. To comply with the Intergovernmental Panel on Climate Change (IPCC) report and the Paris Agreement to limit global warming to 1.5–2° C, only a limited amount of oil can be extracted and burned. The problem becomes how to optimize the use of this limited amount of oil. Hotelling in 1931 concluded that the optimal rate of extraction of an exhaustible resource is such that the net price after extraction costs should rise at the rate of interest. However, the Hotelling optimum would require severe reductions in current and future consumption. Nevertheless, even if not optimal, we concluded that the value over time of a limited amount of oil can be increased significantly by reducing current consumption and gradually reducing future consumption.

A. Appendix: Estimating the Optimal Rate of Extraction of Oil Given the Limitations of the Paris Agreement and IPCC Report on Climate Change

1.1 A.1 Assumptions

In estimating the optimal extraction rate, we made the following assumptions.

Limited Resource

To achieve the goal to limit global warming to 2° C, much of the oil, coal, and gas must be left in the ground. A study (McGlade and Ekins 2015) estimated that 33% of the world’s known oil reserves²⁹ in 2010 needed to be left in the ground to limit global warming to no more than 2° C above pre-industrial times. We estimated in Table 6.9 that 587 billion barrels of oil can be burned globally after 2018 without causing global warming to exceed 2° C.

Table 6.9 Estimate of oil reserves that can be burned without causing global warming to exceed 2° C (McGlade and Ekins 2015; BP 2018; EIA 2019 STEO)

The IPCC report specified a mitigation path of the amount of energy in extra joules (EJ = 10¹⁸ joules) for each kind of energy for the years 2020, 2030, and 2050. The amounts for oil are shown in Fig. 6.6 and Table 6.10. Assuming a constant decrease in oil between those years, we estimate the total from 2020 to 2050 to be 4162 EJs or the equivalent of 689 billion barrels of crude oil.

Fig. 6.6

The IPCC mitigation path for world oil (IPCC 2018 path)

Table 6.10 Estimate of oil that can be burned on the IPCC mitigation path (IPCC 2018 path)

Time Preference

Since we are estimating optimal extraction from a social welfare point of view, we will use the social rate of time preference (the interest rate at which society is willing to forego current consumption for future consumption). We assume a real³⁰ rate of 3%.

It can be argued that it would also be appropriate to use the real rate of return that can be earned on investment in the analysis. If the resource can be extracted and the net proceeds can be invested and earn a higher rate of return than the increase in value of the resource if left in the ground, then it should be extracted and invested. In a perfectly competitive economy, the social rate of time preference and the marginal rate of return³¹ on investment should be equal. But because of imperfections in the market, such as taxes, the rate of return on investment is generally higher than the market interest rate and social rate of time preference. This higher rate of return on investment would imply that more of the resource should be extracted and would be the appropriate rate to use if the proceeds from extracting the resource were truly invested. If, however, the proceeds were used for consumption, then the rate of time preference would be the appropriate rate.

Demand

The world demand for oil in 2018 at an average price of $71 per barrel (EIA 2019 Brent) was about 36.5 billion barrels per year (EIA 2019 STEO). For each $10 increase in the price of oil, we estimate that the quantity demanded decreases slightly more than 1 billion barrels per year. We assume a price elasticity of demand³² of −0.2 at a price of $71, which means that a 10% increase in price would result in a 2% reduction in quantity demanded at that price. As the price increases, the price elasticity of demand increases as people are more sensitive to price at higher prices. At about $100 per barrel, the elasticity is −0.31, which is the long-run elasticity for gasoline estimated by a quantitative survey (Havranek et al. 2012). Figure 6.7 depicts the demand for oil assumed in this analysis.

Fig. 6.7

World oil demand used in this analysis

Cost of Extraction

Costs of extraction vary widely from a few dollars to over $90 per barrel. We assume a starting average cost of $44, which rises to $90 as the limited reserves are exhausted.³³ Generally, the oil with the lowest extraction costs is extracted first. Then as the price of oil rises, it becomes economically feasible to extract oil with higher extraction costs.

1.2 A.2 Analysis

While we do estimate what prices could be as a result of restricting oil consumption or imposing taxes, we do not project what prices will be in the future as a result of inflation. The analysis is done in real prices³⁴ in mostly 2018 dollars. Many values in this chapter were adjusted for inflation using the Consumer Price Index (CPI) from the Bureau of Labor Statistics (2019).

The price is the marginal value³⁵ of an additional barrel of oil, which decreases as the quantity increases. From the quantities demanded for each price or marginal value of oil, we can calculate the benefit of oil.

For simplification we assume that demand remains constant and only varies with price. We realize that high oil and gasoline prices alarm people and that there are other ways to reduce consumption of oil that many people think they would prefer even if they would be less economical. Furthermore, as economies adapt to using less oil, the demand may decrease due to factors other than price.

The cost of extraction per year is subtracted from the benefit to get the net benefit. Since the net benefit is in the future, it needs to be discounted to calculate the present value.³⁶ The present value of the net benefits for each year is summed for all years to get the cumulative present value of the net benefits of the limited burnable reserves. or the present value of the limited reserves (PVLR) for short.

The question is: what rate of extraction would maximize the present value of the limited reserves of 587 billion barrels of oil in the world as estimated from the study (McGlade and Ekins 2015) or 689 billion barrels as estimated from the IPCC report? We refer to the amount of oil that can be extracted and burned according to this climate limitation as “limited reserves,” which decline as the oil is extracted.

By increasing the net price by 3% per year, which reduces the quantity demanded, the present value of the limited reserves over time is increased. However, since the initial price of $71 may not be optimal, we calculated the PVLR for various initial prices.

Additional analyses were done with different extraction costs and demands for oil. The results were similar: by increasing the current price of oil and increasing the price of oil over time, the present value of limited reserves (PVLR) of oil can be increased. The PVLR can be increased even if the net prices are increased by other than 3% per year according to the Hotelling principle, although 3% is optimal for the optimal starting price of oil.

1.3 A.3 Results

1.3.1 A.3.1 Paris Agreement

Table 6.11 summarizes various scenarios of exhausting the limited reserves of 587 billion barrels. The baseline scenario starts with the average oil price in 2018 of $71 per barrel of oil, and this price increases only with the cost of extraction, which increases as the reserves are exhausted. Scenario A applies the Hotelling principle. By reducing the extraction rate so that the net price of oil increases by 3% per year, the present value of the cumulative net benefits (PVLR) over time is increased a little. Since the current price or consumption of oil may not be optimal, we calculated the PVLR for various initial prices and quantities as shown in the rest of the scenarios. The PVLR is maximized with an initial price of $154 per barrel.

Table 6.11 Exhaustion scenarios of limited oil reserves

Figure 6.8 shows some of these scenarios graphically. As the price of oil increases (shown by the black lines), the quantity demanded decreases as shown by the gray lines. Not only is PVLR increased by raising the initial price, the life³⁷ of limited reserves is also increased as shown in both Table 6.11 and Fig. 6.8. Increasing the initial price to $154 per barrel would give the world over twice as many years to adapt to alternatives before the limited burnable reserves would run out, as well as substantially more benefits as compared to an initial price of $71 per barrel. By starting with a large sacrifice early on, there are more benefits in the future.

Fig. 6.8

Estimated price and world demand for oil under various scenariosBlack lines and left axis: price per barrel of oil.Gray lines and right axis: world quantity of oil demanded—billions of barrels per year.Scenario A: dotted lines.Scenario B: thin lines.Scenario H, maximum PVLR: thick lines.

Raising the initial price to $154 per barrel³⁸ may seem like a lot. Moreover, there are likely to be delays in efforts to reduce the demand for oil. Therefore, we estimated other scenarios as shown in both Fig. 6.9 and Table 6.11. For comparison all these scenarios achieve approximately the same PVLR of about $104 trillion.

Fig. 6.9

Estimated price and world demand for oil under additional scenariosBlack lines and left axis: price per barrel of oil.Gray lines and right axis: world quantity of oil demanded (billions of barrels per year).Scenario C: solid lines.Scenario D: double lines.Scenario G: dashed lines.

Scenario D shows the effects of a delay of 6 years. In order to achieve the same PVLR, the price per barrel of oil must be raised in 2024 to $200. However, such a price rise does increase the life of the limited reserves. Scenario G starts at a more reasonable starting price of $101 per barrel and then raises the net price at a growth rate of 6% instead of 3%. While the initial price is lower, the price per barrel exceeds the price in Scenario C after 2030.

In the above analysis, we have assumed that demand does not change so that the only factor that affects the quantity demanded is price. Therefore, we have only used price to reduce quantity demanded and save oil for the future. In some of the scenarios, the price of oil in later years would be around $400 per barrel or about $10 per gallon of gasoline in the United States. Taxes required to raise the prices to those levels would unlikely be politically feasible in the United States. There are other ways to reduce demand to save oil for the future such as standards and regulation, although economists generally believe price is the most efficient way to save oil for the future.

Another way to view these scenarios is to consider the quantity or consumption of oil that would be necessary to follow the scenarios above. Then the black lines in the above graphs would indicate the marginal value of a barrel of oil, that is, what someone would be willing to pay for an additional barrel of oil if it were not available for less. The marginal value could be greater than its price if methods other than price are used to restrict consumption.³⁹ Applying the Hotelling principle means saving some oil for the future so that the marginal net value of oil rises at the rate of interest. The next subsection on the IPCC report will take this point of view in this appendix, while in Sect. 4.2 the same analysis is from the point of view of what prices are necessary to follow the mitigation pathways.

1.3.2 A.3.2 IPCC

The IPCC report (2018 path) estimated a low mitigation pathway for keeping global warming mostly within 1.5 °C, which is shown in Table 6.12 in the first row and Fig. 6.10 by the solid lines.

Table 6.12 IPCC and Hotelling mitigation pathways with marginal values (IPCC 2018 path)

Fig. 6.10

IPCC and Hotelling mitigation pathways with marginal values (IPCC 2018 path)Black lines and left axis: marginal value per barrel of oil.Gray lines and right axis: world quantity of oil demanded—billions of barrels per year.IPCC mitigation pathway: thick lines.Optimum Hotelling pathway: dashed lines.

Assuming the demand for oil remains the same as it was in 2018, the marginal value would increase from $71 to about $109 per barrel as a result of reducing consumption from 36.5 to 32.7 billion barrels (about 11%) in 2020. By following the mitigation path, the marginal value per barrel would rise to over $300 per barrel by 2050 as shown in Fig. 6.10. The total quantity of oil burned over the 30-year period is about 689 billion barrels.

If we stick to the IPCC mitigation path, the net marginal value would increase faster than 3%. If the real (above inflation) interest rate is 3%, then it would be more economical to save a little more of the oil for the future according to the Hotelling principle.⁴⁰ To get the maximum present value of the cumulative net benefits (PVLR) with the same total quantity of 689 billion barrel, the consumption would have to be reduced by about 20% in 2020 from 2018 consumption and continually reduced thereafter so that the net marginal value increased by 3% per year. This quantity of oil would last another 10 years until 2060 as shown by the dashed lines in Fig. 6.10.