China Comment

Energy, Environment, and Economy

Modeling China’s Crude Demand

Models employed by IEA, the Chinese Government, and the United States Department of Energy estimate, quantify, and predict China’s future energy demands. The estimates range from lowball numbers to significant over-estimates. 2008’s wild swings in commodity prices demonstrated the difficulty of predicting international energy markets. Exponential energy growth models significantly overestimated China’s demand, and a 1996 model failed to predict China’s torrid pace of growth. The methodology behind the World Economic Outlook’s (WEO) predictions have closely tracked reality for the past four years. The WEO provides perhaps the most useful framework for understanding China’s future crude oil demand.

China’s year 2008 crude oil production, at 189 million tons, was the country’s highest ever. China’s oil imports also soared to record levels, at over 164.51 million tonnes imported through 2008’s first eleven months (AFP). Still, China’s oil demand appears to have slowed in 2008’s fourth quarter. Oil is currently 21 percent of China’s energy mix. Yearly, China’s oil imports increase by 5-10 percent. Increasingly, oil gains in importance as part of China’s energy security strategy. Correct models of China’s future energy demands may make China’s future economic and political goals more understandable. Below, three energy demand models are examined and are followed by a data-table.

Exponential Energy Growth Model (Ramirez, Alvarez and Rodriguez- 2005)

China’s future energy demand has been difficult to predict. Models, such as Ramirez, Alvarez and Rodriguez’s (below) depend on exponential growth patterned on prior returns. An exponential growth graph fails to capture nuances of human nature and account for plateauing of demand after an optimum supply-level and equilibrium price is reached. At some point, every society becomes “built out” and its growth slows. Accompanying the reduced GDP growth, crude oil demand likewise declines. Ultimately, exponential growth predictions that do not incorporate an eventual plateau will tend to over-estimate growth.

Exponential models of increased oil demand help provide some answer to the mystery of this summer’s rise to $150/barrel oil. Ramirez’s 2005 prediction of Chinese energy demands in 2010 requires an equilibrium crude price of greater than $120 bbl. Under the Ramirez scenario, China’s requirements rise from 7.6 million barrels a day to 11 million barrels of oil a day in 2010. Even with consumption at 7.6 million barrels a day, China still ranks second highest in world demand to US consumption of 20.68 million barrels a day. If China’s middle class, its economy, and its car culture threaten to grow to US, Japanese, or European levels of consumption per capita, then the world’s oil production infrastructure will be put under a mighty strain.

Moderate Growth Model (China Energy Strategy Study- 1996/CSIS 2003)

In 2003 the Center for Strategic and International Studies (CSIS), presented and commented on the 1996 China Energy Strategy Study’s predictions for China’s energy demand growth. The China Energy Strategy Study significantly underestimated future target figures. The Energy Strategy Study’s predicted energy demands were made before China joined the WTO. The predictions appear to have expected 6-8 percent GDP yearly growth. In contrast, the years 2003-2008 posted near-consistent yearly double digit GDP growth.

The Most Applicable Model for the Near-Term Future (WEO- 2004)

Of the three models for Chinese oil growth, the one with the best track record appears to be the 2004 World Economic Outlook prediction. Although the WEO revises its estimates slightly higher each year, the 2004 and 2006 WEO methodology gives the best model of future Chinese crude oil industry growth.

The WEO’s “reference scenario” approach closely tracks the actual evolution of Chinese crude oil demand. In contrast, Ramirez’s 2005 exponential energy growth model significantly overestimates actual requirements since China’s hyper-expansion slowed. The underestimating 1996 China Energy Strategy Study  presumed China would benefit relatively less from international integration and trade than was ultimately achieved.

Chinese crude oil demand also appears to be best tracked by the WEO’s “reference scenario.”  WEO projections will likely not underestimate Chinese crude demand since demand is somewhat tied to China’s economic success. Additionally, China’s future economic growth will be tempered by environmental externalities. In 2004, China’s government-created Green GDP measurement demonstrated environmental externalities cost three percent on potential GDP growth. The percentage-cost of environmental degradation is expected to rise as pollution’s harmful effects accumulate and China’s energy demands increase. State policies attempt to mitigate the problems of environmental externalities, but environmentally-friendly policies such as increasing automobile taxes and vehicle fuel-economy standards depress demand growth and discourage consumption.

The WEO’s “reference scenario” also likely does not overestimate China’s crude energy demand. Although the global automobile industry is currently in a retraction period, the annual arrival of hundreds of thousands of Chinese to the middle class presages a continued demand for automobiles and petroleum. Although US car sales plunged an average of 25 percent each month from September through November; Chinese automobile sales in November were only 3.2 percent off from the previous month (China Daily). Still, a sharper drop in Chinese automobile sales could be coming. To prevent that drop, the Chinese government offers valuable incentives for trading in cars (China Daily). Chinese GDP however, is still likely to grow at 6-8 percent in 2009, and as long as oil remains below $120 a barrel, there will be a significant number of citizens who reach the wealth threshold where it makes sense to own and purchase gas for a private car.

– Past experience is no guarantee of future returns, as world stock market investors learned to their chagrin in 2008. The WEO estimates appear to offer good insight into the future, but has the paradigm changed? (Feel free to respond in the comments.)


Actual crude oil demand and future projections given by various sources, including the above-mentioned models.

* Note on converted numbers: There may be some variations between actual amounts and reported amounts. Some “actual amounts” such as the BP and CBC data differ slightly, but by no more than 15 million barrels in any year. In all cases below, I have converted barrels/day to m. tonnes of crude/year. I identify where I convert numbers.

Total Crude Demand (Estimates and Actual)
2050– 520 m. tonnes of crude (1996 Prediction- Barring aggressive adoption of renewable energy or a decline in Chinese standards of living, this is vastly underestimated.)

2030- 746 m. tonnes of crude (WEO 2006, p. 86)
2030– 683 m. tonnes of crude (IEA WEO 2004 as reported by Dr. Tang (see sources).)

2020– 599.7 m. tonnes of crude (Converted from numbers of the DOE IEO 2005 and 2005 East-West Center report in Lieberthal)
2020– 585 m. tonnes of crude (Converted from numbers of the IEE Japan 2004 report in Lieberthal.)
2020- 565.75 m. tonnes of crude (Converted from the IEA WEO 2004 report in Lieberthal)
2020– 563 m. tonnes of crude (3.3% annual growth rate 2011-2020) (April 2008 Xinhua prediction)
2020– 320 m. tonnes of crude (1996 Prediction- Notably equal to 2004 actual data.)

2015– 487.6 m. tonnes of crude (WEO 2006, p. 86]

2010– 536.55 m. tonnes of crude (converted from “Short-term predictability of crude oil markets: A detrended fluctuation analysis approach” by Ramirez, Alvarez, and Rodriguez doi:10.1016/j.eneco.2008.05.006, Energy Economics 30 (2008) (p 2654). Based on 1984-2005 exponential growth activity)
2010– 448.2 m. tonnes of crude (Converted from numbers of the DOE IEO 2005 report in Lieberthal)
2010– 419.39 m. tonnes of crude (Converted from numbers of the East-West Center 2005 report in Lieberthal)
2010– 409.6 m. tonnes of crude [WEO 2006, p86.]
2010– 407 m. tonnes of crude (4.5% annual growth rate 2007-2010) (April 2008 Xinhua prediction)
2010– 385 m. tonnes of crude (Converted from numbers of the IEA WEO 2004 report in Lieberthal)
2010– 355 m. tonnes of crude (Converted from numbers of the IEE Japan 2004 report in Lieberthal; Notably equivalent to 2008 actual data.)
2010– 260 m. tonnes of crude (1996 Prediction- Notably equivalent to 2002 actual data.)

2009409.6 m. tonnes of crude (Converted from EIA numbers).

2008– ~354+ m. tonnes
2007– ~345 m. tonnes
2004– 323.39 m. tonnes. converted. (CBC) & BP 2005.
2003- 270.47 m. tonnes. converted. (CBC).
2002- 251.49 m. tonnes. converted. (CBC).
2001- 239.8 m. tonnes. converted. (CBC) .
2000– 233.97 m. tonnes. converted. (CBC) & BP 2005.

Domestic Supply
2050– 80 m. tonnes of crude. (1996 Prediction, perhaps lowballed.)

2030– 136.5 m. tonnes of crude. (WEO 2006, p. 92)

2020– 180 m. tonnes of crude. (1996 Prediction, notably equivalent to 2006 actual data.)

2010– 165 m. tonnes of crude. (1996 Prediction, notably prior to the Bohai Bay discovery, and  equivalent to 2004 actual data)
2010– 185.3 m. tonnes of crude. (WEO 2006, p. 92)

2008– 189 m. tonnes of crude (Up 1.6%) [Xinhua Jan 2008]
2007– 186.7 m. tonnes of crude (Up 1.6%) [China Daily Jan 2008]
2005– 175.5 m. tonnes of crude (WEO 2006, p. 92).
2000– ~155 m. tonnes of crude

2050- 440 m. tonnes of crude. (1996 Prediction- frightening energy security situation for Beijing if this is underestimated.)

2020- 429.24 m. tonnes of crude. (2005 DOE IEO estimate in Lieberthal. Converted.)
2020- 346 m. tonnes of crude. (2004 IEA WEO estimate in Lieberthal. Converted.)
2020- 140 m. tonnes of crude. (1996 Prediction- Vastly Underestimated)

2010- 224 m. tonnes of crude. (2006 WEO Prediction, after I run the numbers.)
2010- 95 m. tonnes of crude. (1996 Prediction- Vastly Underestimated)

2008- 164.51 million tonnes (In 1st 11 months- 9.5 percent growth; (Reuters Dec. 2008) Note: Q1 and Q2 posted an 11 percent growth at 90.53 m. tonnes)
2007- 159.28 m. tonnes of crude imported [China Daily Jan 2008]
2004– 120 m. tonnes of crude imported. [China Daily Nov 2005]
2000- ~50 m. tonnes (IPR Strategic Business Information Database)

China’s Oil Demand Deficit

Year Demand Deficit (Imports)
2000 233.97 m. tonnes. ~50 m. tonnes
2004 323.39 m. tonnes. 120 m. tonnes
2008 354+ m. tonnes. 164.51 m. tonnes.
2010 (Xinhua & WEO predictions) 407 m. tonnes. 210-230 m. tonnes (surpasses 50%)
2020 (Xinhua & WEO prediction) 563 m. tonnes. 340-400 m. tonnes (around 75%)

(*”PetroChina, like many European oil companies, measures its output in tonnes instead of the US standard of barrels.”)

Other Sources:
– Gill, Bates and Matthew Oresman. “China’s New Journey to the West: China’ Emergence in Central Asia and Implications for U.S. Interests: A Report of the CSIS Freeman Chair in China Studies.” Center for Strategic and International Studies. Washington, D.C. August 2003. 25. [Analyzing the 1996 China Energy Strategy Study].

– “China says 2008 crude production up 1.6 pct: state media” AFP. December 2008.

– Lieberthal, Kenneth and Mikkal Herberg. “China’s Search for Energy Security: Implications for U.S. Policy.” NBR Analysis (National Bureau of Asian Research) April, 2006. 12.
– Tang, James. “With the Grain or Against the Grain? Energy Security and Chinese Foreign Policy in the Hu Jinato Era.” October 2006.

6 January, 2009 - Posted by | China Energy, China Future | , ,

1 Comment »

  1. […] China Comment – Modeling China’s Crude Demand – An assessment of the various models used to predict China’s future energy demand […]

    Pingback by CER links: Phelps, Obama, A- and H-share parity - China Economic Review | 8 January, 2009

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: