Measures of Petroleum Dependence and Vulnerability in OECD Countries
this article was originally published in
Copyright 2003 A. F. Alhajji and James L. Williams
A. F. Alhajji and James L. Williams¨
F. Alhajji, PhD.
The recent strike of oil workers in Venezuela, the war with Iraq, and violence in Nigeria have reduced production, increased oil prices and raised concerns of sufficient oil supplies in the near future. Such concerns, accompanied with high natural gas prices and the impending discussion of new energy policy in the US, raised the issues of energy security, petroleum dependence, and petroleum vulnerability, in various consuming countries, to the extent that the International Energy Agency (IEA) made a presentation regarding its response in case of disruption. The effectiveness of OPEC production increases, the Saudis utilization of their excess capacity, the IEA’s commitment to releasing oil from public oil stocks such as the Strategic Petroleum Reserves (SPR), and the recent cooperation between IEA and OPEC to avert an energy crisis and lessen the impact of any petroleum shortages will be instrumental to any future discussions of energy security and energy policy in the oil consuming and producing countries. In addition, they will determine the fate and the level of future cooperation between petroleum consuming and producing countries.
This article presents a historical overview of various measures of petroleum
dependency and vulnerability in OECD countries. It intends to assess
the current security of petroleum supplies and the possibility of petroleum
shortages in the near future.
This study utilized data from the EIA, the IEA, and other sources. Inconsistency
among various data sets is common in this type of literature; however,
nothing of significance to our findings. IEA data are in metric tons and
EIA data are in barrels. In our presentation we rely upon ratios minimizing
the effect of the utilization of two different units. For those not familiar
with the problem we just note that when crude oil is processed in a refinery
there is a volumetric increase (barrels) in the output due to cracking
but a decrease in the weight (metric tons) because of fuel use. Conversion
is difficult because of the various petroleum products that can be obtained
from a barrel of oil. In addition, we have encountered three more
problems, first, only the IEA adjusted its data for new OECD membership,
while other organizations did not make the adjustment. Second, only
quarterly and yearly data are revised, leaving the monthly data without
revision. Third, petroleum definition differs from data set to another.
Most data sets exclude some petroleum products from the data, a procedure
that exacerbates inconstancy among data sets.
II.Measures of Petroleum Dependence
Several measures of dependency on petroleum have been introduced in the literature and will be covered by this study. Various countries within OECD view import dependence differently from the way it is viewed in the US. While some politicians consider increasing petroleum imports a threat to national security, many experts believe that the level of imports has no significant impact on energy security, or even national security. The issue becomes a problem when import vulnerability increases as petroleum imports rise. This occurs when consuming countries increase the share of petroleum imports from politically unstable areas of the world.
This study uses four measures of petroleum dependence:
(1) petroleum imports as a percentage of total petroleum consumption,
(2) the number of days total petroleum stocks cover petroleum imports,
(3) the number of days total stocks cover consumption, and
(4) the percentage of petroleum in total energy consumption.
1.Petroleum Imports as a Percentage of Total Petroleum Supply
Figure (1) shows petroleum imports as a percentage of total petroleum supply (inland deliveries) in the OECD countries. The Figure indicates that while this percentage declined in the 1970s and early 1980s, it has been relatively stable since the 1991 Gulf War at about 60%. The decline in this percentage from 85% in the early 1970s to about 50% in the mid 1980s resulted from the increase in OECD’s petroleum production during that period especially from Alaska and North Sea and a price induced decline in demand. The degree of import dependence as percentage of consumption increased from about 50% in the mid 1980s to 60% in the 1990s because of higher economic growth and lower oil prices on the demand side and declining US production on the supply side. Much of this increase in dependence on petroleum imports came from the US. However, OECD dependence on petroleum imports has declined throughout 2002 to reach 51% in the fourth quarter. This probably understates fourth quarter dependence due to declining inventories replacing imports especially in the case of the US.
The degree of import dependence among OECD countries differs from country
to country. Data indicates that some countries, such as Japan, Sweden,
and South Korea, import all their petroleum from foreign oil producers
while other countries, such as the Netherlands and Ireland, depend relatively
less on petroleum imports. However, while most of the rhetoric regarding
dependence on foreign oil imports comes from the US, among the group of
net importers the US depends the least on foreign petroleum imports; only
55% of total consumption is imported.
Percentage of Net Petroleum Imports Relative to Total Petroleum Supply
Figure (2) shows the level of import dependence in the US from 1973 to present. Figures (1) and (2) indicate that while total OECD dependence remained relatively constant in the last decade, US dependence grew rapidly, indicating that the dependence of OECD countries, excluding the US, has decreased while US dependence has increased.
The US dependency on foreign oil has increased steadily since 1986 as shown in Figure (2). The US dependence on foreign oil reached record highs in the last two years declining slightly after September 11. Major changes in US imports are usually related to changes in the US economy and the US oil production.
Dependence on imports grew from 35% in 1973 to 42% in 1978. As a result of increased production from Alaska and decline in US demand, dependence on imports fell to 27% in 1985. With the oil price collapse in 1986, dependence, fueled by increasing consumption and decreasing production, once again resumed its upward path to 55% in 2001. The impact of higher prices and the lingering aftermath of September 11 led to a modest reduction to 53% in 2002.
Should the US worry about petroleum import dependence? Among the OECD countries
the US is unique in several respects. It is the only OECD member country
with significant production which is also a net importer. It is also unique
in that among the net importers it has the lowest dependence in terms of
net imports as a percent of consumption but it also has the highest absolute
level of imports. The geo-political and economic interests and commitments
raise the level of concern by US policy makers about dependence on imported
US Import Dependency versus Oil Prices, 1973-2003
Source: EIA, 2003
2.Days of Total Stocks Coverage of Imports
Figure (3) shows days of imports coverage by total OECD stocks, which include commercial and government-controlled stocks like the SPR. The number of days of stock coverage of imports increased drastically in the early 1980s as OECD countries took various measures in the aftermath of the two energy crises to enhance their energy security. However, OECD stock coverage of imports has been decreasing steadily since its peak of 192 days in the first quarter of 1983. Since that peak, OECD stocks coverage of imports reached its lowest level of 130 days at the beginning of 2001. Nevertheless, the number of days increased since the end of 2001 because of slowdown in economic growth, which slowed imports, and because of the build up in the US SPR. Given the slow economic growth since 2001 relative to 2000, the number of days of stock coverage of imports is still low by historical standards. In fact, it is similar to that of middle of 1980 when oil prices reached record levels and total stocks covered only about 135 days of imports.
However, while the current average for all OECD members is only 135 days, many countries are below this average.
Days of Total Stocks Coverage of Imports
The situation is relatively more severe in the US, the largest economy in the world. Commercial oil stocks in the US are at their lowest level in twenty-seven years. Today, US total petroleum inventories, which include commercial and SPR stocks, are as low, in terms of days of coverage, as they were during the first and the second energy crisis, as shown in Figure (4). Current commercial inventories are near the level at which spot shortages can occur. Since early December, lower imports have resulted in inventories declining at an average rate of slightly under one million barrels per day. During this period U.S. production and imports were about 5 percent lower than consumption. Figure (4) indicates that the US is in no better condition to cope with a supply interruption than it was at the time of the 1973 Oil Embargo, the 1979 Iranian Revolution, or the 1980 Iraq-Iran War.
Although the SPR level has been increasing since President Bush ordered its refilling after September 11, the decline in commercial stocks is greater than the increase in the SPR and the capacity of the SPR and commercial stocks to deal with a crisis is less than before the refilling program began. An SPR release coinciding with the beginning of conflict in Iraq was anticipated by many. Increased production by several OPEC member countries is largely responsible for postponing or eliminating any release.
Moreover, the premature release of SPR can jeopardize US national security in case of continued political problems in the oil producing countries and weakens US ability to respond to real shortages. Premature release of SPR stocks can also reduce incentives for private companies to maintain a stock cushion thereby further increasing US vulnerability to supply interruptions. The impact of any SPR release on oil prices depends on the size of the release and the language of the White House announcement. We believe that an announcement declaring a continuous drawdown in installments will have a greater effect on prices than an announcement declaring the drawdown of the same quantity in a very short period of time. Reports that Saudi Arabia has 50 million barrels in storage to help avert any shortfall may have contributed to the US decision not to release any crude from the SPR at this time.
US Total Stocks/US Net Petroleum Imports
3.Days of Stocks Coverage of Total Consumption
Figure (5) shows the number of days that OECD inventory can cover its petroleum consumption. This shows how long OECD countries can sustain a complete halt of world production and exports, including their own. The general trend indicates an increase from about 70 days in the 1970s to approximately 90 days in the early 1980s. However, while the number of days stabilized at about 87 days in the second half of the 1980s, the ability of OECD stocks to cover consumption has been declining since 1991 and reached 77 days at the end of 2002, a situation similar to that of the second energy crisis at the end of 1979.
Despite the fact that Figure (5) shows coverage of approximately 77 days of consumption, the number of days varies from country to country greatly, as indicated in Table (1), which ranks OECD countries based on the number of days that their stocks covered consumption at the end of 2002. While Japan’s stocks can cover 122 days of its consumption, the UK stocks cover 63 days and South Korea’s stocks cover only 37 days.
It should be noted that in actual practice the number of days are much lower as certain minimal stock levels are necessary for the operation of the system system. Commercial inventories of crude oil in the US are currently near that level and would therefore not contribute substantially in an emergency. A superior measure would be the days’ coverage above minimal operating levels but the determination of those levels on a country-by-country basis with accuracy over time would be a daunting task. The US estimates the Lower Operational Inventory (LOI) of total crude and petroleum products at 862 million barrels. As of March 14, 2003, US inventories, including the SPR, were 1,481 million barrels of which 882 million were commercial stocks. Adjusting for the LOI total stocks available including the SPR are 619 million. Excluding the SPR, commercial inventory has an excess of 20 million barrels or about one day of U.S. consumption. In reality the inventory coverage in the US is little more than half of what would be indicated by this measure.
OECD Inventory Days of Consumption
Stocks Coverage of Days of Consumption, October 2002
Source: IEA, 2003
One needs to bear in mind that there is unlikely to be a complete cut-off, but rather a supply squeeze. A strong supply squeeze would tend to lead to reactive measures: first panic with all that it entails, but then also reduction of consumption by rationing, higher prices and substitution of other sources of energy. So the stocks might have more of a psychological panic-reducing function. In reality, supplies would last longer than the measure indicates. The reader may wish to investigate the 1979-1981 period for an example of panic purchases on the part of refiners.
4.Percentage of Petroleum in Total Energy Consumption
The share of petroleum in total energy consumption in OECD countries has decreased substantially after the two energy shocks in the 1970s. The share of petroleum has declined from 55.27% in 1973 to 43.42% in 1985. However, the decline in oil prices halted the rapid decline in the share, which reached 42.82% in 1990.
Petroleum share in the total energy consumption in OECD countries has been relatively stable in the last decade at approximately 42%. However, the sharp increase in oil prices in 2000 may have caused some fuel switching that led to a slight decline in petroleum share by about 0.5%. The current share of petroleum in the total energy consumption stands at 41.6% for OECD and it is slightly lower in OECD Europe at 41%.
Figure (6) illustrates the trend of petroleum share in total energy consumption in OECD countries. It shows that this share is sensitive to oil prices. Extended periods of high prices led to fuel switching and conservation. These changes tend to be more or less permanent, which will explain the continual decline in the petroleum share.
OECD Petroleum Share in Total Energy Supply (%)
Source: IEA, 2003
Although OECD Pacific countries have made substantial progress in reducing their dependence on oil from 73% in 1973 to 57.18% of total energy supply in 1990, they still depend more on petroleum than the rest of OECD members do. This is the only group that made significant progress in reducing their dependence on petroleum in the 1990s.
While the share of petroleum in energy consumption stayed relatively the same in the last few years, the share of petroleum in the US total energy consumption has increased. Figure (7) shows the US petroleum share in total energy consumption. It demonstrates how the share of petroleum declined steadily from 47.5% in 1977 to 38% in 1995. However, that percentage has been increasing since 1995 but decreased slightly in 2000 because of higher oil prices. US dependence on petroleum has increased in recent years from 38% of total energy consumption in 1995 to around 40% in early 2003.
This measure, the share of petroleum in total energy consumption in OECD countries, indicates three possible areas of concern regarding the extent to which petroleum influences energy security: the inability of OECD countries to reduce their dependence on oil in recent years, the large share of petroleum in the energy consumption of OECD members in the Pacific, and the increase in the petroleum share in the US in the last few years.
US Petroleum Consumption Share in Total Energy Consumption (%)
III.Measures of Import Vulnerability
Many experts argue that the degree of dependence has no impact on energy security as long as foreign oil is imported form secure sources. However, if the degree of dependence on non-secure sources increases, energy security would be in jeopardy. In this case, vulnerability would increase and economic and national security of individual OECD countries would be compromised. Data indicates that OECD countries’ vulnerability, especially the US, is at a historic high, higher than at any of the preceding energy crises. We will use the following three measures to assess OECD vulnerability to supply disruption:
1.Percentage of OECD Imports From the Top Five Supplying Countries
We look at the percentage of imports from the top five suppliers as a measure of the supply vulnerability to an interruption by one or more key suppliers. This is an important measure of OECD countries’ vulnerability to supply disruption because it shows the high level of OECD import concentration by importing from few suppliers. The top five suppliers to OECD countries in the last few years have been Saudi Arabia, the former USSR, Norway, Venezuela, and Mexico. Only Norway and Mexico are OECD members and Saudi Arabia and Venezuela are OPEC members.
Data indicates that OECD countries’ vulnerability to a supply shock has increased recently as the dependence on the top five petroleum exporters has increased from 44.7% of total imports in 2000 to 48.5% of total imports in the third quarter of 2002. Given the recent strike in Venezuela, the invasion of Iraq, and the unrest in Nigeria, the level of import dependence on few countries have increased in recent weeks and may have reached record levels.
Since most petroleum trade is regional, the exposure to vulnerability among OECD members differs from region to region. Although the US has a unique location between two OECD petroleum exporting countries, Canada and Mexico, The problems in Venezuela underline the US vulnerability to interruption from a major supplier.
Figure (8) indicates that the US vulnerability to supply disruption has increased to historic levels recently as the percentage of its petroleum imports from its top five suppliers increased from 62% in 2001 to 76% in the first ten months of 2002. Comparing this number to those of previous energy crises, we find that it is higher than the 63.8% and 53.4% concentration ratios in the first and second energy shocks respectively.
The percentage of US Petroleum Imports from the Top 5 Exporters
2.OPEC share of world petroleum supply
Some experts view the share of the petroleum market controlled by OPEC as a measure of vulnerability. Production cuts to increase oil prices and various problems in OPEC countries make OECD countries more vulnerable to a supply shock similar to the two energy shocks in the 1970s. From the OECD point of view, the smaller the share of OPEC, the better.
Figure (9) plots OPEC crude oil production share in world crude oil production against oil prices. OPEC’s share declined from 56% of world oil production in 1973 to about 28% at the end of 1985. However, it has increased steadily between 1986 and 1993, and it increased again in 1997 and 1998. OPEC’s market share reached its most recent peak in 2000, the highest since 1980. Production cuts by OPEC for the last two years decreased its share to about 40%.
Recent events should increase concern among importers of dependence on OPEC. Venezuela, a dependable source of crude even in WWII, has had severe problems and is recovering, but the political problems and potential for another interruption still exist. Iraq crude is off the market for an indeterminate period of time. Conflict in Nigeria has significantly reduced its output.
In addition, the recent expansion of production in other OPEC countries at the expense of excess capacity makes OECD countries more vulnerable to a supply shock.
Figure (10) presents the world’s historical production capacity versus oil prices. It indicates that the world excess capacity today is at its lowest level in the last 30 years. It is even lower than the excess capacity during the first and the second energy crises, and lower than that of 1991, when Kuwaiti and Iraqi production were taken off the market because of the Gulf war. Similar to the energy crises in 1973 and 1979, current excess capacity exists only in very few OPEC countries, primarily in Saudi Arabia. The Saudi ability to expand capacity by another 500,000 barrels would not help the US in case of immediate oil shortages because it takes a long time to get that oil to the US for two reasons. First, it requires sixty to ninety days to bring that capacity on line. Second, it takes four to six weeks to get that oil to the US. In addition, some experts believe that OPEC capacity statistics as reported by the US Department of Energy (DOE) are overestimated. In fact, the DOE in recent reports revised downward some of its estimates of excess capacity in Kuwait and Indonesia.
In conclusion, world excess capacity is at a record low, the capacity is concentrated in a few OPEC member countries, and it takes time to get the oil to the US the worlds largest consumer. These three reasons can exacerbate the shortages in the US and create a potential for crisis.
OPEC Crude Oil Production as a Percentage of Total World Crude Production
Source: EIA, 2003
The World Petroleum Excess Capacity
Source: EIA and IEA, 2003
3.The Persian Gulf Share of World Petroleum Supply
Another important measure of vulnerability is the share of world crude coming from the Gulf region. The Gulf region has been viewed historically as a politically unstable area and crude oil imports from the Gulf are viewed, especially by the US, as unsecured imports. Incidents in the Gulf region led to the three energy crisis in 1973, 1979, and 1990-1991. From the OECD point of view, a smaller share of Gulf production means lower vulnerability. However, as discussed earlier, the share of the Gulf in total world production works both ways, it reduces and increases vulnerability at the same time. When the Gulf has a lower share it usually means large excess capacity, which, in turn, reduces vulnerability.
Figure (11) shows the share of Persian Gulf crude oil production as a percentage of world production. The share of the Gulf nations declined substantially in the first half of the 1980s as Saudi Arabia made a gradual 8 million b/d production cut to maintain prices. The share of the Gulf increased after 1986 only to decrease suddenly in 1990 because of the Iraqi invasion of Kuwait and the war that followed. The share of the Gulf nations rebounded after the war and stabilized at about 28%. Changes in the Gulf’s share in the last three years are the result of OPEC’s frequent adjustment of members’ quotas. Figure (11) illustrates that with higher oil prices in 2000, the share of the Gulf nations in world production reached 30%, the highest in 10 years.
The Persian Gulf Crude Oil Production as a Percentage of World Crude Production
Several measures have been used to assess the energy security of OECD countries. These measures indicate that OECD in general and some members in particular are more susceptible to energy crisis than in previous years. Measures of dependency indicate that OECD members are still heavily dependent on oil imports. They also indicate that OECD countries have not made a significant reduction in dependence on imports since the mid 1980s and they continue to import 60% of their total petroleum consumption. Although the level of dependence on oil imports in the US is lower than all net-importing OECD countries, the US has experienced a sharp growth in petroleum imports in recent years to the extent that US dependency reached an all time high early in 2002. OECD petroleum stocks have been declining, especially in recent months. Historical data indicates that the number of days of total stocks coverage of imports have been declining since 1983. The number declined from 192 days in the third quarter in 1983 to about 135 days recently. The situation is even worse in the US where the number of days that stocks cover imports is similar to those during the energy crises of 1973 and 1979. In addition, OECD stocks coverage of total petroleum consumption has declined recently to about 77 days of consumption, which is equal to what prevailed during energy crisis of 1979. The OECD petroleum share in total energy supply also reflects dependence on petroleum. According to this measure, the share of petroleum has been relatively constant for some time. In recent years; however, it has increased in the US. In addition, this measure indicates that while OECD Pacific countries have lowered their dependence on oil petroleum substantially since the early 1970s, the share of petroleum is still very high at approximately 50% compared to 30% and 40% for the other OECD members.
Measures of petroleum import vulnerability indicate that OECD vulnerability to supply interruption has increased recently as the imports from the top five suppliers increased from 44.7% of total petroleum imports to 48.5%. The US is the most vulnerable among all OECD members based on this measure as its dependence on the top five suppliers increased from 62% in 2001 to 76% in 2002. With the war in Iraq and problems in Nigeria, OECD vulnerability has increased. Recent data indicates that OPEC excess capacity is at its lowest level in the last 30 years.
The possibility of energy crisis in the near future is greater than previous years. All measures point to the fact that OECD dependency and vulnerability are very high. Although the measures indicate different levels of dependency and vulnerability, all OECD countries will be affected by an energy crisis. When oil prices increase, they increase world-wide, and everyone is affected. The global nature of financial and spot markets make a shortage in one region felt throughout the world. However, petroleum prices may increase in some countries more than others for a short period of time. Theoretically, a country may suffer during the period needed to transfer oil from other markets to this particular country.
The use of the SPR or government controlled stocks to lessen the impact of an energy crisis is subject to debate. We believe that there are certain conditions that must exist to make an SPR release successful. However, the premature release of the SPR may exacerbate an energy crisis as it depletes the stocks while shortages still exist. It leads to lower prices and, therefore, increased consumption.
Dependency and vulnerability measures indicate that energy security can be enhanced by energy diversification, petroleum imports diversification, and better cooperation between consuming and producing countries. Although any shortage in the world will lead to higher oil prices worldwide, countries that are more diversified will be less affected by any petroleum shortage. In addition, imports diversification will lower the relative impact of supply disruption on most countries.
Cooperation is needed to stabilize oil prices and to avoid massive swings in upstream investment. The coincident problems in Venezuela, Nigeria and Iraq underscore the need for excess world capacity. Furthermore, they raise concerns that most of that capacity has historically been concentrated in a single country.
On another front, we have limited our use of data to that which is available
to the general public. In so doing, we became acutely aware of the restriction
imposed upon unfunded academic researchers and students who desire to analyze
energy supply, demand and inventories. As many significant contributions
to the understanding of markets come from the IEA and EIA, a secondary
result of our work is a policy recommendation to the IEA member countries.
The development and dissemination of detailed historical data for member
countries at the country level with open access to the public following
the model of the EIA would lead to a better understanding of energy markets.
It would also be useful if the data were published in barrels and metric
Petroleum: A broadly defined class of liquid hydrocarbon mixtures. Included are crude oil, lease condensate, unfinished oils, refined products obtained from the processing of crude oil, and natural gas plant liquids. Note: Volumes of finished petroleum products include non-hydrocarbon compounds, such as additives and detergents, after they have been blended into the products. Lease condensate: A mixture consisting primarily of pentanes and heavier hydrocarbons which is recovered as a liquid from natural gas in lease separation facilities. This category excludes natural gas plant liquids, such as butane and propane, which are recovered at downstream natural gas processing plants or facilities. Natural gas plant liquids: Those hydrocarbons in natural gas that are separated as liquids at downstream gas processing plants or at fractionating and cycling plants. Data on lease condensate are excluded. Products obtained include liquefied petroleum gases and pentanes plus. On the supply side in the statistics a series may include crude oil, crude oil and lease condensate, or crude oil, lease condensate and natural gas liquids. Lease condensate in gas liquid separated at the lease i.e. near the well. As opposed to plant liquids which are separated at a gas plant which removes enough liquid from the gas so that the BTU content is in the standard range, normally about 1 MMBTU per 1 MCF. Source EIA and WTRG Economics.
 For more information, see Williams and Alhajji “The Coming Energy Crisis?”, The Oil and Gas Journal, Vol. 101, No. 5, February 3, 2003.