World crude production: Bad statistics produce poor conclusions

When supply and demand statistics didn’t balance, a missing barrel category was fabricated to explain the difference. However, a close inspection of the figures reveals serious errors in the original data and a precarious balance between supply and demand

George S. Littell, Groppe, Long & Littell, Houston

hrough a long bear market, the oil industry has continued to keep track of prices. However, it has ceased to monitor the volumes of production and consumption. By default, the Oil Market Report, published monthly by the International Energy Agency (IEA) in Paris, has become the source of oil balances, not only for Wall Street but for the industry itself.

Over time, IEA’s oil balances have become inaccurate to the point of being misleading. From the actual statistics, it is easy to conclude that:

• There were no "production cuts" after OPEC met in March 1998 and cuts of less than 0.3 million bpd after it met in June.
• Increasing OPEC crude oil production in the first two quarters of 1998 — most of it coming from Iraq — was the cause of the collapse in oil prices.
• IEA’s "missing barrels" — the difference between calculated and observed inventories — are the sum of all the errors in statistics for deliveries and production.

There were no legal exports of crude oil by Iraq from August 1990 through November 1996. As Iraq returned to the oil market in 1997 and 1998, stable prices required that increases in its production be offset, to some degree, by cuts in production by other OPEC members. Confusion about the numbers was a factor in their failure to do that. IEA is not solely responsible for that confusion, since the OPEC members could, and did, read similarly inflated numbers for their production of crude oil in the industry’s trade press.

Basic Concepts

The production of oil has three categories:

• Crude oil — A mixture of hydrocarbons that exist in liquid phase at reservoir conditions of temperature and pressure.
• Condensate — Hydrocarbons that are in a gaseous state at reservoir conditions, but condense to the liquid state at surface standard conditions.
• Gas liquids — Ethane, propane, butanes and natural gasoline are components of natural gas along with methane, its primary component. They are separated from the methane in processing plants using various technologies.

Refineries convert crude oil, condensate and gas liquids into useful products such as gasoline, naphtha, jet fuel, diesel, residual fuel oil, lubricants, wax, asphalt, coke, etc. In addition to the international trade in crude oil, condensate and gas liquids, there is extensive commerce in refined products.

The simple basis for oil balances is to begin with production, i.e., the Production Basis. In the Production Basis, deliveries for a country or area are obtained by the formula:

Deliveries = Production + Net Trade + Change in Stocks

The more difficult method is to begin with deliveries of refined products and gas liquids, or the Deliveries Basis. The formula for determining balance is:

Deliveries = Production + Net Trade
+ Change in Stocks + Processing Gain

Aside from the small amounts of hydrogen that may be added, the refining processes do not change the weight of the hydrocarbons refined. Many of the processes (cracking, reforming, etc.) result in products that are less dense than the input. The volume of products from a refinery — not their weight — typically exceeds the volume of the input of crude oil, condensate, etc. by 1% to 6%, depending upon the combination of processes employed. The difference between output and input volumes is termed Processing Gain.

Although oil is sold by volume, the data is reported either as a volume — gallons, barrels, cubic meters, etc. — or as a calculated weight (usually metric tons). It requires time to collect and audit the data. For members of the Organisation for Economic Cooperation and Development (OECD), statistics by weight for the first three quarters of 1998 are available in the IEA publication, Oil, Gas, Coal, & Electricity: Quarterly Statistics. In the U.S., the Energy Information Administration (EIA) has published statistics by volume for all of 1998. Comprehensive statistics by weight for the entire world are complied by the UN. They have been published for the years ending with 1995.

IEA Oil Balances

For 1997 and 1998, oil balances according to the Oil Market Report compare with the deliveries basis and production basis as shown in Table 1.

In concept, IEA uses the deliveries basis. Because of its low figure for processing gains — 1.57 million bpd in 1997 versus the real 2.73 million bpd — IEA has to overstate production and / or understate deliveries in order to reach a balance. In practice, it does both, but still ends without a balance. The calculated increase in stocks serves as a balancing item. The "missing barrels" are those in excess of stocks reported by the OECD countries.

Import Statistics

An oil balance is, of course, an exercise in tracking a physical flow — where the oil comes from and where it goes. The best way to construct it is to begin with the statistics for oil imports by the OECD countries which are shown in Table 2.

Members of OECD account for 85% or so of the world trade in oil. Their statistics do not support the idea that OPEC production increased significantly in the first quarter of 1998, and then was reduced by large "production cuts." Instead, they are compatible with the estimates shown in Table 3.

When the OPEC ministers met in Jakarta, Indonesia, late in November 1997, the belief that its members were already producing an average 27.2 million bpd of crude oil — versus quotas for 25.033 million bpd — was a major factor in the decision to raise the quotas to 27.5 million bpd for the first quarter of 1998. In March 1998, the "production cuts" were from an estimate of 28.7 million bpd for crude oil production in February 1998. No real oil was involved since the production figure from which cuts were to be made was an estimate that was inflated. All most of the members planned to do was to improve reporting of actual volumes by IEA and the trade press.

Fraud and Confusion

In field operations, it is common for condensate, natural gasoline and even some butanes to be blended with crude oil for transportation in pipe lines and ships. The resulting commercial crude oil is not crude oil as produced.

In OPEC’s system, quotas apply to crude oil, not to condensate and gas liquids. It is not difficult to overstate total oil production by counting condensate and some of the gas liquids twice — once as a part of crude oil and again as gas liquids. The trade press and IEA are generally innocent of the distinctions. IEA, in particular, is under continuous pressure to overstate total production in order to make up for its low estimate of processing gains.

OPEC members of do not provide many statistics for production. Where there is no disclosure, it is reasonable to expect fraud. Production figures from Venezuela, Algeria and Nigeria usually are considered suspect.

Venezuela. OPEC adopted a formal definition of condensate in November 1988 after a long dispute over an attempt by Venezuela to classify a substantial amount of its production as condensate rather than crude oil. The specified test procedures expect that condensate has an API gravity of more than 50° API (8.088 bbl per metric ton).

In the process, OPEC did not adopt a formal definition of crude oil. Although Venezuela does not produce much condensate, it is a large producer of heavy crude with an API gravity less than 10.5°. Asphalt (4.5° API) is a solid at standard conditions. Up to 10.5° API, hydrocarbon mixtures require some sort of diluent to maintain a liquid state at standard conditions. In Canada, the diluent for bitumen is natural gasoline. A technical advance in Venezuela was the development of orimulsion, which uses water as a diluent.

For 1997, Venezuela’s production can be broken down as in Table 4.

Whether or not bitumen is crude oil can be debated — with Venezuela on one side of the argument and the rest of OPEC on the other. Omitting production from fields operated by companies other than Petroleos de Venezuela was not an accident.

In 1998, production of crude oil — including bitumen — compared with OPEC quotas about as shown in Table 5.

At OPEC’s meeting in March 1998, it was easy for Venezuela to agree to 3.170 million bpd for the second quarter because that quota was more than its capacity to produce crude oil by any definition. The agreement to produce 2.845 million bpd in the second half of 1998 was simply disregarded.

Algeria. Foreign oil companies operating in Algeria also have established crude oil production in recent years. Volumes reported to OPEC can be reconciled with total production as listed in Table 6.

Nigeria. The Nigerian national oil company does not operate fields. Production of crude oil in recent years can be reconciled with volumes reported to OPEC as shown in Table 7.

Mobil, the operator of the Oso gas / condensate field, regularly reports condensate as part of its crude oil production in Nigeria. Oso accounted for most of the 0.096 million bpd of condensate produced in Nigeria during 1998. Trade press and IEA figures for Nigeria are from a more-or-less complete survey of the foreign operators and include condensate as a result.

Condensate and foreign operators are a hazard to accurate reporting in both Indonesia and Qatar as well. The other problems in OPEC are Iran and Saudi Arabia.

Iran. In addition to only applying to crude oil, the OPEC quotas limit supply rather than production. The concept of supply was invented during the war between Iran and Iraq when Saudi Arabia built up large inventories of crude oil outside its borders. It argued successfully that the oil required did not affect the market and therefore should not count against its quota.

The OPEC definition of supply is:

Supply = Exports + Domestic Consumption
+ Withdrawals from Stocks Outside National Borders

In the early days of that war, the refinery at Abadan was partially destroyed. Its loss made Iran a net importer of refined products — a unique circumstance in OPEC — until the new refinery at Bandar Abbas finally went into full operation during the third quarter of 1998.

In 1997, Iran’s production compared with its OPEC quota and volume reported as shown in Table 8.

Indonesia is an importer of crude oil, but carefully excludes those volumes from its internal consumption. Since Iran does not exclude its imports of products, the volumes reported to OPEC, and its OPEC quota, have been higher than its actual production of crude oil for years

One virtue of a quota for Iran, more than its capacity, is that it makes up for some or all of the production in excess of quotas by Nigeria, Venezuela and others.

Iran’s agreement at the June 1998 OPEC meeting for a quota of 3.318 million bpd in the second half of 1998 — roughly its capacity — was a departure from usual practice. Iran has reverted to its usual policy in 1999.

Saudi Arabia. In addition to being the largest producer in OPEC, Saudi Arabia also has the largest refining capacity and burns some crude oil directly because of limitations to its supply of natural gas. The internal system in Saudi Arabia is complex. The general formula for its operation is:

Exports = Production – Internal Use
(including stock changes)

With constant production, seasonal changes in consumption, refinery turnarounds, etc., result in changes in the volume of exports. Estimates in the trade press for Saudi Arabia production follow the formula:

Production = Exports + Internal Consumption

With internal consumption considered constant — which it never is.

Whether or not Saudi Arabia is complying with its OPEC quota depends on what one thinks of its practice with Abu Saafa field. Abu Saafa was discovered in 1963 on the maritime boundary between Bahrain and Saudi Arabia. The 1972 treaty for its development made Saudi Aramco the operator and provided for an equal division of revenue. In 1992, the treaty was amended to give Bahrain the revenue from 100 million bpd of the 145 million bpd of production from Abu Saafa. As of April 1996, it was further amended to have all the output marketed by Bahrain National Oil Company.

It is certainly laudable for Saudi Arabia to help Bahrain. Whether or not Abu Saafa should be excluded from Saudi Arabian production for OPEC purposes is another question that has never been considered at an OPEC meeting. Saudi Arabia’s unilateral change in definition may have encouraged Venezuela in its deliberations concerning heavy crude.

Conclusions

There has certainly been rampant fraud in OPEC statistics, but it has been almost entirely an effort to understate production of crude oil. Over time, IEA and the trade press have swung the opposite way — systematically overstating crude oil production.

During 1998, confusion about the facts cost the OPEC members a great deal of money. Whether future statistics will be more accurate — no further additions to the "missing barrels" — remains to be seen.

The author

George Littell has been a partner of the firm of Groppe, Long & Littell since 1976. Successor to a firm founded in 1955, Groppe, Long & Littell has its office in Houston.

Copyright © 1999 World Oil
Copyright © 1999 Gulf Publishing Company