The Hubbert peak theory, also known as "peak oil", concerns the long-term rate of conventional petroleum and other fossil fuel extraction and depletion. It is named after American geophysicist Marion King Hubbert, who created a model of known reserves, and proposed, in 1956, in a paper he presented [1] at a meeting of the American Petroleum Institute, that oil production in the continental United States would peak between 1965 and 1970; and that world production would peak in 2000.

United States oil production peaked in 1971 [2], and has been decreasing since then. Global production did not peak in 2000. Supporters of peak theory suggest Hubbert's model did not account for the 1973 and 1979 Organization of the Petroleum Exporting Countries oil shocks, which effectively reduced global demand for oil, thus delaying the peak.

Given that petroleum is a non-renewable resource, it is inevitable that at some point there will be a similar peak in worldwide petroleum production. Hubbert's theory is that the same calculations that successfully predicted the peak in petroleum production in the USA would apply to other circumstances, such as the peak in worldwide petroleum production. Various estimates for the worldwide peak have been made by Hubbert and others, with some of these dates already having passed. This has led to criticism of the method and predictions made using the method.

Hubbert's peak theory is subject to continued discussion because of the potential effects of lowered oil production, and because of the ongoing debate over aspects of energy policy. Opinions on the effect of passing Hubbert's peak range from faith that the market economy will produce a solution, to predictions of doomsday scenarios of a global economy unable to meet its energy needs.

Some oil industry executives, economists, and analysts doubt that Hubbert's peak theory applies on a global scale. However, Chevron has launched the Will You Join Us? [3] ad campaign, seeking to inform the public to the possibility of petroleum depletion and encourage discussion. The campaign's website notes findings from the International Energy Agency's (IEA) World Energy Outlook 2004: "Fossil fuels currently supply most of the world’s energy, and are expected to continue to do so for the foreseeable future. While supplies are currently abundant, they won’t last forever. Oil production is in decline in 33 of the 48 largest oil producing countries, ..."

Few would disagree with the statement that fossil fuels are finite and that alternative energy sources must be found in the future. Most critics instead argue that the peak will not occur soon and that the form of the peak may be irregular and extended rather than a sharp logistic curve peak. Like any mathematical model, the accuracy of the prediction is dependent on the validity of the model and further to that, by the accuracy of the input data. If variables such as consumption are estimated incorrectly, then the formula will yield different results.

In 1971, Hubbert used high and low estimates of global oil reserve data to predict that global oil production would peak between 1995 and 2000. ASPO has calculated that the annual production peak of conventional crude oil was in early 2004. Events that occurred after Hubbert's prediction may have delayed the peak, especially the 1973 energy crisis, in which a decreased supply of oil resulted in a shortage, and ultimately less consumption. The 1979 energy crisis and 1990 spike in the price of oil due to the Gulf War have had similar, albeit less dramatic effects on supply. On the demand side, recessions in the early 1980s and '90s have decreased the demand and consumption of oil. All of these effects would theoretically delay peak oil.

The Association for the Study of Peak Oil and Gas (ASPO) was founded by the geologist Colin Campbell. Based on current information about known oil reserves, estimates of future discovery, growing oil demand, and available technology, the ASPO predicts that world oil production will peak around the year 2010. Natural gas is expected to peak anywhere from 2010 to 2020 (Bentley, 2002).

In 2004, 30 billion barrels of oil were consumed worldwide, while only eight billion barrels of new oil reserves were discovered. Huge, easily exploitable oil fields are most likely a thing of the past. In August 2005, the International Energy Agency reported annual global demand at 84.9 million barrels per day (mbd) which means over 31 billion barrels annually. This means consumption is now within 2 mbd of production. At any one time there are about 54 days of stock in the OECD system plus 37 days in emergency stockpiles.

The United States Geological Survey estimates [5] that there are enough petroleum reserves to continue current production rates for 50 to 100 years. That is countered by an important Saudi oil industry insider who says the American government's forecast for future oil supply is a "dangerous over-estimate."[6] Campbell argues that the USGS estimates are methodologically flawed (although he himself has admitted that he doesn't understand their methodology). One problem, for example, is that OPEC countries overestimate their reserves to get higher oil quotas and to avoid internal critique. Population and economic growth may lead to increased energy consumption in the future.

According to the Energy Information Administration of the United States Department of Energy, international reserve "estimates are based on non-technical considerations that support domestic supply growth to the levels necessary to meet projected demand levels. [emphasis added]" (Annual Energy Outlook 1998 With Projections to 2020[7]).

Professor Kenneth Deffeyes, author of "Hubbert's Peak" (ISBN 0691116253) and "Beyond Oil" (ISBN 0809029561), asserts that the peak was passed on Dec 16, 2005 [8]. He also asserts that the total of world oil is 2.013 Trillion Barrels.

In late 2005 as oil prices rise, greater attention is focused on Hubbert's theory and its potential implications. However, oil and gas prices are notoriously volatile and price increases have been caused by numerous other factors, though there is a general agreement that increased demand has been the major factor, with such increased demand bringing the Hubbert peak closer than would have been predicted otherwise. In June 2005, OPEC admitted that they would 'struggle' to pump enough oil to meet pricing pressures for the fourth quarter of the year. The summer and winter of 2005 brought oil prices to a new high. This may be a sign of increasing demand starting to outstrip supply or it may just be that the various geopolitical forces in the regions where oil is produced are limiting demand. One other explanation for the rising oil prices is that it is a sign of too much paper money and not too little oil. In this view, dramatically higher prices of all commodities and real estate indicates rising inflation. Evidence emerged in early 2006 that Kuwait had overestimated its oil reserves. Adjusting for the new figure reduced global reserves by 5%. If other oil producers have made similar errors, peak oil may be closer.