The New Era of Global Energy Abitrage[1]

Roman Zytek
Washington, DC

Official forecasts of oil prices have proven unreliable in predicting oil prices (Horn 2004; Winebrake and Sakva 2006; EIA 2004-2014). This paper argues that the current decline in oil prices represents another leg in the process of restoring competitiveness of energy from crude oil in the energy market. The surge in the price of oil from 2005 to 2008 will go into history as another short-lived economic incident. The price surge created energy arbitrage opportunities which triggered three standard market responses: (1) efforts to substitute other energy -- coal, gas, and increasingly renewable sources -- for oil; (2) deployment of technologies to increase oil supply from regions outside of OPEC’s control; and (3) market integration within and across fuel classes. These three demand and supply responses are creating one, fully integrated -- geographically and across fuels – global market for energy. They are bringing the energy market back into long-term equilibrium and eliminating the abnormal rents earned by oil producers. Going forward, forecasters of energy prices should pay more attention to the competitive position of each energy fuel in the energy mix, and less attention to the marginal cost of production of the more expensive fuels at any point in time. 


Fuel Substitutability and Demand

Hydrocarbon fuels are bought primarily for their energy content. Demand for specific primary fuel is a function of its productivity and price relative to the productivity and prices of other fuels and energy forms (Bashmakov 2007). Historically, prices for oil, natural gas, and coal tended to oscillate within a narrow corridor that reflected the relative energy content of the fuels and the relative convenience of their use (Figure 1).


At the same time, on any particular day prices per unit of energy varied widely by fuel even in relatively well integrated geographical markets (Table 1).


Table 1. Energy Content and Energy Prices for Three Primary Fuels



Market price per standard unit (December 2, 2014)

Price per Mbtu






Natural gas (per 1 Mbtu; Hub Henry)




Oil (per barrel, WTI)




Source:  http://www.engineeringtoolbox.com/energy-content-d_868.html; DataStream; and author’s calculations.


Higher energy density in oil, and the ease of transport and storage, allows oil to command a price premium over coal and natural gas. Between 1994 and the end of 2004 energy from oil traded at an average price of about US$4 per billion British Thermal units (MBtu) compared to US$3.4 per MBtu of energy from gas. On average, daily oil price commanded a 30 percent price premium over gas during the period. Occasionally, energy from oil was cheaper than energy from gas. This price premium rose to an average of 60 percent in 2005-end of June 2008, and increased further to almost 300 percent after 2008, as natural gas prices plummeted with the advent of the shale gas revolution. The premium exceeded 1,000 percent in April 2012, when 1 MBtu from oil sold for almost US$22, while 1 MBtu from natural gas was available for just under US$2 in the U.S. market.  

Oil and its distillates have substitutes too, even in transport. When oil becomes unavailable or extremely expensive, given time, substitutes can be deployed effectively and in sufficient quantities.[2] During World War II, Germany, cut off from oil supplies, generated 92 percent of its aviation gasoline and half of its total petroleum from synthetic fuel (synfuel) made from coal. Years after the end of apartheid -- and related trade sanctions -- domestically made synfuel still accounts for about 30 percent of South Africa’s domestic oil consumption (U.S. Energy Information Agency 2008). In recent years, low cost natural gas and electricity have started to make inroads into passenger car transport, having already captured most of passenger railway transport in the past fifty years. Vehicle fleets using compressed natural gas (CNG), and hybrid and electric engines are moving from fringes to mainstream of the automotive market.[3],[4] Some mass-market car companies expect to stop producing cars powered by oil-based fuels by 2050.

Oil, natural gas, and coal are relatively close substitutes in electricity generation. Lower natural gas prices in the U.S. have already helped increase the share of natural gas in electricity generation. Moreover, the three primary fuels face increasing competition from other energy sources: nuclear power and its derivatives, such as wind, photovoltaic (PV), and the concentrated solar power (CSP).

Relatively high oil prices encouraged energy arbitrage in the past decade. The share of increasingly expensive oil in the world’s primary fuel mix fell from 37.6 percent in 2002 to 32.6 percent in 2014. The share of much cheaper coal rose from 25.5 percent in 2002 to 30.0 percent in 2014, notwithstanding some public concern about CO2 emissions and their possible climate impact. In the U.S., the shale gas revolution pushed the share of natural gas in the country’s energy mix up by 4.3 percentage points in the same period, mostly at the expense of coal and oil (BP 2004, 2015).

The ongoing gradual elimination of energy subsidies in many countries is reducing demand for oil (Sdralevich et al. 2014). The rise in the relative price of oil will slow demand growth for oil-based energy and support improvements in energy efficiency in countries that have subsidized energy in the past. Growth in demand for all energy could slow further while alternative sources of energy will become more competitive (Weiss et al. 2009, 2010).

Energy security concerns encourage innovation and diversification, even in the absence of price signals. The Capital Asset Pricing Model (CAPM) can explain investments in unconventional primary and alternative energy, such as solar and wind, even though such investment had been initially wasteful from a purely short-term economic point of view (Awerbuch and Berger 2003; Awerbuch and Sauter 2005).


Increasing Energy Supply

The rise in the price of oil stimulated the use of new technologies to open access to vast new supplies of natural gas and oil.[5] The commercialization of hydro fracturing and horizontal drilling technologies has led to a dramatic increase in economically viable and competitively supplied natural gas and oil from shale deposits outside OPEC (Kuuskraa 2010; Sandrea 2010; Snyder 2010). The rapid technological advances and competition continue to depress production costs, open new reserves to exploitation, and make long-term investments in transportation justifiable.[6] As a result, today’s annual hydrocarbon production is at the lowest level when measured in terms of production as a share of proven reserves (BP 2015).


Integrating the World’s Energy Market

The competitiveness of the supply side of the market for individual primary fuels has varied greatly over time and influenced specific fuel prices. In the mid-1860s, oil prices spiked due to the monopolistic structure of the transport sector. In 1973-1984 and from mid-2000s, OPEC’s dominance in the oil market resulted in large price increases in 1973-84 and again in 2003-14 (Table 2; and Dvir and Rogoff 2009). Until recently, the natural gas market has been highly segmented, and prices for piped natural gas were linked to oil prices (Okogu 2002). As a result, due to poor arbitrage opportunities for gas exporters, confined to fixed pipeline networks, natural gas prices could differ widely across suppliers and buyers, depending on their location and relative bargaining position. 


Table 2. Oil Price, 1861-2014

(In 2014 U.S. dollars; per barrel)



Standard deviation
















Jan-Nov 2015



Sources: BP Statistical Review of World Energy 2015; and author's calculations.


However, the gas market started to change in late 1990s. The establishment of the Henry Hub in 1988, and its NYMEX price benchmark in the United States in April 1990, created an independent natural gas market and a price setting mechanism that was not contractually related to the price of oil. The development of the LNG technology improved market power for isolated gas producers with access to international waters.[7] The large increase in gas reserves and expected output encouraged rapid expansion of transportation capacity for both piped gas and LNG (Figure 2).


As with oil, security concerns encourage investment in LNG infrastructure to access more costly imports in anticipation of possible supply disruptions from existing low cost suppliers.

The global gas market is integrating rapidly. The massive addition of pipelines to final customers, liquefaction facilities, LNG export and import port terminals, and LNG tanker capacity in the past decade have contributed to better integration of the global natural gas market, and opened up more arbitrage opportunities within the gas market and across fuels. LNG exports grew at an annual rate of almost 6½ percent in 2009-14, notwithstanding the overall weakness in global economic activity, and put downward pressure on gas prices exported via pipelines. As a result, the price differences across the world’s gas markets have started to narrow.


Policy Implications and Conclusions

The world has entered the period where natural gas, oil, and coal compete head on in a highly integrated global energy market. Given that today energy from oil is still about twice as expensive as energy from natural gas in at least one major, U.S. market, and given the high probability that natural gas prices in the U.S. market remain at about $3-$5 per MBtu, prudent corporate executives of oil companies and policymakers in oil exporting countries should prepare for oil prices to fall further from today’s $45-$50 per barrel range. The cost of energy from natural gas may soon determine the price of oil. Furthermore, the decline in oil prices will result in lower investment in high-cost oil fields, thus creating space for lower cost producers, in particular from the Middle East region, and improving global allocation of capital. At the same time, continued technological advances in shale oil and gas extraction are likely to limit OPEC’s ability to control prices, in particular given the relatively short time needed to restart production from more expensive shale deposits.

The coming age of total global energy arbitrage does not mean that prices will not be subject to short-term volatility and long-term cycles. Demand for oil will remain highly price inelastic in the short term. Prices will spike in response to occasional rises in political tensions in major oil and gas producing regions and regulatory actions that threaten to slow the natural market adjustment processes. In the long term, periods of exceptionally low prices will discourage investment in capacity and, over time, lead to cyclical price upswings.

The ongoing energy transition will continue to increase the share of electricity in the world’s energy mix, the most productive form of energy available. In an electricity-dominated energy world hydrocarbons will have a declining usefulness in energy generation. Further technological advances in nuclear power and energy storage, and to a lesser extent wind and solar power, will eventually reduce demand for primary energy from hydrocarbons.[8] This does not mean that oil, natural gas, and coal will not find buyers. It only means that the bulk of future demand for hydrocarbons will need to come from non-energy applications. Such a day is still far into the future, but policymakers in oil and gas exporting countries and regions should start preparing for it today by reviewing economic diversification strategies and management of their intergenerational sovereign wealth funds which could be soon called upon to finance the transition to the world of lower oil rents. In the meantime, forecasters of long-term oil prices may consider using the price of the lowest priced energy, measured in terms of dollars per MBtu, appropriately adjusted for the usual premiums or discounts applicable to the country’s own specific primary fuel export mix and geographic location, to project long-term prices for budget planning purposes.  




Alberth, Stephan, 2008, “Forecasting Technology Costs via the Experience Curve — Myth or Magic? Technological Forecasting and Social Change, Volume 75, Issue 7, September 2008, pages 952–983.

Awerbuch, S. and M. Berger, 2003, “EU Energy Diversity and Security: Applying Portfolio Theory to Electricity Planning and Policy-Making,” International Energy Agency, available at http://www.awerbuch.com.

Awerbuch, S. and R. Sauter, 2005, “Exploiting the Oil-GDP Effect to Support Renewables Deployment” The Freeman Centre, University of Sussex , Science and Technology Policy Research (SPRU), Paper No. 129, available at: http://www.sussex.ac.uk/spru/.

BP Statistical Review of World Energy, 2004 and 2015 editions, available at:   http://www.bp.com/en/global/corporate/about-bp/energy-economics/statistical-review-of-world-energy.html.

Bashmakov, Igor, 2007, “Three Laws of Energy Transitions,” Energy Policy 35, pages 3583–3594.

Dvir, Eyal, and Kenneth S. Rogoff, 2009, “Three Epochs of Oil,” NBER Working Paper No. 14927 (Cambridge, Massachusetts: National Bureau of Economic Research).

Gritsevskyi, A., and N. Nakicenovic, 2000, “Modeling Uncertainty of Induced Technological Change,” Energy Policy 28, pages 907-921.

Goldman Sachs Equity Research, “On the Shale Trail,” Fortnightly Thoughts, March 16, 2012.

Goldman Sachs Commodities Research, “Investor Returns will Survive the Productivity Comeback,” April 24, 2014.

Horn, Manfred, 2004, “OPEC’s Optimal Crude Oil Price,” Energy Policy, Vol. 32, pages 269−280.

Kuuskraa, Vello, 2010, “Gas Shales Drive the Unconventional Gas Revolution,” accessed on August 25th, 2015 at: http://csis.org/event/unconventional-gas-revolution-policy-strategic-and-market-implications.

Meade, Nigel, Towhidul Islam, 2006, “Modeling and Forecasting the Diffusion of Innovation—A 25-year Review,” International Journal of Forecasting, Vol. 22, pages 519-545.

Neij, Lena, 1997, “Use of Experience Curves to Analyze the Prospects for Diffusion and Adoption of Renewable Energy Technology,” Energy Policy, Vol. 23, No. 13, pages 1099-1107.

Neij, Lena, 2008, “Cost Development of Future Technologies for Power Generation—A Study Based on Experience Curves and Complementary Bottom-Up Assessments,” Energy Policy, Vol. 36, pages 2200-2211.

Okogu, Bright E., 2002, “Issues in Global Natural Gas: A Primer and Analysis,” IMF Working Paper 02/40.

Sandrea, Ivan, 2010, “Shale Gas Perspectives,” available at: http://csis.org/event/unconventional-gas-revolution-policy-strategic-and-market-implications.

Sark, van W.G.J.H.M., E.A. Alsema, 2010, “Potential Errors When Fitting Experience Curves by Means of Spreadsheet Software,” Energy Policy, Volume 38, Issue 11, pages 7508-7511.

Sdralevich, Carlo, Randa Sab, Younes Zouhar, and Giorgia Albertin, 2014, “Subsidy Reform in the Middle East and North Africa. Recent Progress and Challenges Ahead.” Available at: http://www.imf.org/external/pubs/ft/dp/2014/1403mcd.pdf.

Snyder, Jen, 2010, “Cost Declines Fueling the Shale Explosion,” presentation at the CSIS, March 9, 2010, available at http://csis.org/event/unconventional-gas-revolution-policy-strategic-and-market-implications.

U.S. Energy Information Agency, 2008, “South Africa,” Country Analysis Brief at www.eia.doe.gov.

Weiss, Martin, Martin Junginger, Martin K. Patel, Kornelis Blok, 2009, “A Review of Experience Curve Analyses For Energy Demand Technologies,” Technological Forecasting and Social Change, Volume 77, Issue 3, pages 411–428.

Weiss, Martin, Martin K. Patel, Martin Junginger, Kornelis Blok, 2010, “Analyzing Price and Efficiency Dynamics of Large Appliances with the Experience Curve Approach,” Energy Policy, Volume 38, Issue 2, pages 770–783.

Winebrake, James J., and Denys Sakva, 2006, “An Evaluation of Errors in U.S. Energy Forecasts: 1982−2003,” Energy Policy, Vol. 34, pages 3475−3483.



[1] The views expressed in this paper are those of the author. They do not necessarily represent or reflect the views of any of the institutions the author has been associated with or could be associated with today or in the future.

[3] See: International Association for Natural Gas Vehicles, http://www.iangv.org/ for discussion of CNG use in transport. See http://ycharts.com/companies/TSLA/market_cap; and http://ycharts.com/companies/GM/market_cap for relative market capitalization of major car companies.  (Accessed on December 14, 2014.)

[5] For more information on the technological developments in shale gas extraction, economics, and its global implications, see the proceedings of the 14th Annual Energy Policy Conference: The Unconventional Gas Revolution - Policy, Strategic and Market Implications, available at http://csis.org/event/unconventional-gas-revolution-policy-strategic-and-market-implications.

[6] See: Goldman Sachs (2012 and 2014). “First, shale technologies have already proved far more successful than their more conventional counterparts in keeping production costs low at the margin. For instance, breakeven costs for a significant proportion of shale oil producers are lower than many of the more conventional deep-water projects in the cost curve. Second, as producers learn more about optimizing the techniques associated with shale extraction (e.g., pad drilling, 3D seismic imaging), we have seen continuing improvements in the efficiency of the shale technologies themselves.” (Goldman Sachs 2014, page 15.)

[7] Cooling natural gas to -260° F/ -162.2° C changes it from a vapor into a liquid. This reduces the space natural gas occupies by more than 600 times, making it a practical size for storage and transportation. (http://www.ferc.gov/for-citizens/citizen-guides/lng.asp.)

[8] See Alberth (2008), Gritsevskyi and Nakicenovic (2000); Neij (1997 and 2008); Sark and Alsema (2010) for discussion of technological experience curves and how they can help inform long term forecasts of technological change. Meade and Islam (2006) review research on modeling diffusion of innovation, with an emphasis on their contribution to improving forecasting accuracy.


Click to view a printable version of this article.