Petroleum Refining: Economic Performance and Challenges for the Future

Order Code RL32248
Petroleum Refining:
Economic Performance and
Challenges for the Future
Updated June 3, 2008
Robert Pirog
Specialist in Energy Economics
Resources, Science, and Industry Division

---

Petroleum Refining:
Economic Performance and
Challenges for the Future
Summary
The petroleum refining industry provides products that are critical to the
functioning of the economy. Virtually all transportation, land, sea, and air, is fueled
by products that are refined from crude oil. Industrial, residential, and commercial
activities, as well as electricity generation, use petroleum-based products. Along with
volatile changes in crude oil prices, the industry has faced evolving health, safety,
and environmental requirements which have changed and multiplied product
specifications and required capital investment in refineries.
Since the late 1990s, the industry has undergone significant structural change
which might alter its profitability requirements, its ability to provide stable product
volumes to the consuming market, and its ability to adapt to current and future
environmental requirements.
Two significant structural changes characterize the industry. Mergers,
acquisitions, and joint ventures have changed the ownership profile of the industry,
altering concentration patterns both regionally and nationally. A change in the
business model from an integrated component, to a stand-alone profit center, has
focused attention on earning competitive profit rates at each stage in the production
chain. Evidence suggests that the new market structure and business model might
demand better economic performance from the industry. Regulatory compliance to
meet congressionally mandated environmental standards, both on refined products
and refinery sites, requires substantial capital investment by refiners, and has resulted
in reduced profitability, according to the Energy Information Administration (EIA).
To the extent that continued capacity expansion and technological investments are
reduced, or not undertaken, because of low historical rates of return (even though
recent returns are higher) U.S. dependence on imported refined products might
increase, or product markets could be disrupted by shortages and price spikes.
As the 110th Congress considers energy legislation, it is likely to be concerned
with domestic energy security and market stability, issues linked to the performance
of the petroleum refining industry. Increased imports of refined products, particularly
motor gasoline, combined with growing imports of crude oil, could make the United
States increasingly vulnerable to shocks originating in the world oil market.
Importing motor gasoline into the United States in appropriate volumes may become
increasingly difficult because of the unavailability of world supplies consistent with
U.S. fuel specification requirements.
This report will be updated.

---

Contents
Refining Economics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
Defining Profitability . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
Performance of the Refining Industry . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
Changing Structure of U.S. Refining Industry . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
Challenges Facing the Refining Industry . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18
Environmental Requirements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18
Capital Investment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19
Production and Growth . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22
Imports . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23
Permits and Regulatory Uncertainty . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25
Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26
List of Figures
Figure 1. U.S. Refining Capital Expenditures for FRS Companies, 1988-2001 21
List of Tables
Table 1. U.S. Real Price Spread Between Light Crude Oil and Heavy Crude Oil,
1983-2007 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
Table 2. U.S. Real Price Spread Between Motor Gasoline and Residual Fuel Oil,
1983-2007 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
Table 3. Refining Margins of U.S. Major Oil Refiners, 1993-2006 . . . . . . . . . . . 7
Table 4. U.S. Refinery Output,1991-2006 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
Table 5. Domestic Refining/Marketing Return on Investment for FRS
Companies, 1991-2006 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
Table 6. Refinery Capacity and Utilization, 1991-2006 . . . . . . . . . . . . . . . . . . . 11
Table 7. U.S. Refining/Marketing Net Income, 1995-2005 . . . . . . . . . . . . . . . . 12
Table 8. Top Twenty U.S. Refiners, 1980-2005 . . . . . . . . . . . . . . . . . . . . . . . . . 14
Table 9. Refining Concentration by PADD . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16
Table 10. Imports of Petroleum Products, 1992-2006 . . . . . . . . . . . . . . . . . . . . 24

---

Petroleum Refining:
Economic Performance and
Challenges for the Future
The petroleum refining industry processes crude oil and produces a variety of
products that are used in the transportation, residential, commercial, and industrial
sectors of the economy. In 2007, over two thirds of refinery output went to
transportation uses, nearly a quarter went to industrial uses, and the remainder was
used in residences, commercial activities, and electricity generation. The
transportation sector remains the most heavily dependent on petroleum, drawing over
95% of its fuel needs from refineries.
Because the refining industry plays such a key role in providing energy for the
U.S. economy, its structure and economic condition are matters of national interest.
In recent years, the industry has undergone significant change. The traditional
industry model, based on ownership by vertically integrated oil companies with
profitability viewed within the context of a linked supply chain, has been altered by
companies and joint ventures whose primary business is refining. Increasingly, the
business model for these firms, as well as the integrated oil companies, is the stand-
alone profit center. Refiners now must earn market rates of return for investors, as
well as returns sufficient to make investments in expansion, technological
improvements, possible business restructuring, and to meet environmental
regulations, both with respect to refined product specifications and refinery site
operations and expansion.

Should the industry fail to meet these market-based performance standards,
policy makers may be faced with the prospect of weighing the costs and benefits of
decisions taken to mitigate supply and price volatility, especially in gasoline
markets. If the industry determines that it is unable to invest sufficiently to meet
projected demand increases, additional supplies of refined products will likely be
imported. If imports were to increase significantly, policy makers might need to
prioritize the needs of the local markets within the context of national energy security
goals. The ability of the refining industry to meets its economic challenges will
likely determine, in part, the nature of the energy challenges facing Congress. This
report seeks to describe the recent performance of the refining industry, to evaluate
the structural changes that are occurring in the industry, and to analyze the nature and
effects of the challenges confronting the industry.

---

CRS-2
Refining Economics
When considering the economic performance of a petroleum refiner, several
characteristics should be noted. First, the production process is capital intensive, and
technological improvements are embodied through investments in the basic refinery.
As a result, any circumstance which affects the availability, or allocation, of capital
investment can have significant structural effects on the firm and the industry.
Second, the major variable cost in the production process is crude oil, the price of
which is determined in the world petroleum market. Third, while labor and other
variable costs are relatively small fractions of total costs, a great deal of attention is
given to minimizing them in the interests of enhancing profit margins. Finally,
regulatory costs, both capital and operational, required to meet congressionally
mandated environmental standards on both refined products and refinery sites,
increase as new requirements are phased in.

Evaluating the economic performance of the petroleum refining industry is
complicated by the fact that many refineries can use crude oil of lower quality as an
input, while others cannot. Crude oil can be of lighter or heavier density, as well as
having a higher or lower sulfur content. The density of the crude oil is important
because, in general, a lighter crude oil input yields a lighter product mix. A lighter
product mix is important because lighter products are generally in higher demand,
and yield higher prices for the refiner.
Does this mean, then, that refineries that are able should always use the lightest
crude oil available? Not necessarily, if the crude oil market pricing mechanism is
working well. The market compensates for differences in the quality of crude oil by
a price differential, the light-heavy price spread. However, at any given time, the
actual, specific spread value for any set of crude oils is also influenced by relative
availability on the world market as well as the location of the oil. As a result, the
value of the spread changes, and, at any given time, the purchase of either light or
heavy crude might be warranted by economic conditions, if technologically feasible
for the refiner.
Table 1 shows the behavior of the light-heavy price spread, over time, measured
in constant dollars. Following a historic peak in the late 1970s, the spread dropped
sharply with the general collapse in the price of oil in the mid-1980s. In the early
1990s, the spread increased relative to the low values of the late 1980s, and, as a
result, during the 1990s, U.S. refiners invested heavily in technologies that allowed
the use of heavier crude oil inputs. These investments led to stronger demand for
heavy crude oil, effectively raising its price and reducing the magnitude of the
spread, as well as the profit potential of those investments. During the recent period
of increased crude oil prices, from 2004 through 2007, the spread between light and
heavy oils has again increased, reaching record levels in 2006. By January 2008, the
spread had increased to $18.79, another record. In the longer term, it may well be
that the relative quantities of light and heavy crude oils available on the market will
continue to shift toward the heavier oils. This shift, if it continues, might still
validate the financial investments made by refiners in the 1990s and encourage
additional investment in refinery enhancement.

---

CRS-3
Table 1. U.S. Real Price Spread Between Light Crude Oil and
Heavy Crude Oil, 1983-2007
(Dollars Per Barrel)
Year
Spread
Year
Spread
1983
8.10
1996
5.10
1984
3.97
1997
6.43
1985
3.40
1998
6.19
1986
4.07
1999
5.08
1987
2.43
2000
7.61
1988
3.49
2001
8.83
1989
4.23
2002
5.37
1990
6.59
2003
7.03
1991
9.09
2004
10.63
1992
8.06
2005
14.23
1993
5.78
2006
14.94
1994
4.49
2007
14.62
1995
3.47
Source: Energy Information Administration, Petroleum Marketing Monthly, April 2008, Table 24,
p. 49.
Further complications in refining economics arise from the technology of
refining. Simple refineries, those whose technology is limited to a basic distillation
process, produce low yields of light products. An output of 20 percent gasoline, 30
percent middle distillates and 50 percent heavy residuals, based on an input of
Arabian Light crude oil, might be typical. With an investment in the most
sophisticated refinery technologies, the output mix might change to as much as 60
percent gasoline, 35 percent middle distillates and 5 percent heavy residuals.
In 2007, motor gasoline earned an average refiner price of $2.18 per gallon
excluding taxes, while residual fuel oil brought an average refiner price of $1.35 per
gallon, net of taxes. This 83 cent per gallon price differential represents a premium
of almost 61% for motor gasoline compared with residual fuel oil.1 Table 2 traces
the value of the price differential between motor gasoline and heavy residual fuel oil.
Since the mid 1980s, while the differential has shown little trend, it has shown
significant volatility, this can affect expected profitability.
1 Energy Information Administration, “Petroleum Marketing Monthly”, March 2007, Tables
35, 42. Calculation of averages by the author.

---

CRS-4
Table 2. U.S. Real Price Spread Between Motor Gasoline and
Residual Fuel Oil, 1983-2007
(Cents per gallon)
Year
Spread
Year
Spread
1983
27.3
1996
29.3
1984
17.8
1997
31.3
1985
25.8
1998
24.6
1986
22.6
1999
29.1
1987
20.4
2000
39.7
1988
27.7
2001
41.0
1989
29.4
2002
29.8
1990
37.2
2003
34.1
1991
38.5
2004
60.7
1992
36.9
2005
70.0
1993
33.3
2006
83.3
1994
28.2
2007
83.2
1995
26.3
Source: Energy Information Administration, Petroleum Marketing Monthly, April 2008, Table 4, p.
13.
The most advantageous market position for a complex refinery that has invested
in the capability to produce a light product mix from a heavy crude input is a large
price spread between light and heavy crude and also a large spread between light and
heavy products. In that environment, similar to that observed in 2007, a refiner can
buy heavy crude to minimize direct input costs and sell a light product mix at
relatively high prices to enhance the gross margin per barrel.
The potential for economic gain represented by these two price spreads form the
incentive for investing in more complex processing units in the refining process.
While these units enhance economic performance, they require substantial capital
investment, usually billions of dollars. In 2005, refinery capital expenditures by the
Financial Reporting System (FRS) companies in the United States totaled $15 billion,
up from $8.1 billion in 2004, likely reflecting the high returns earned by refiners in
2004 as well as other factors.2 However, in 2006 refinery investment declined to $11
billion, a decline of 27%, when compared to 2005. Additionally, $5.7 billion was
spent by the FRS companies in foreign refinery capital expenditures in 2006, an
2 Energy Information Administration, Performance Profiles of Major Energy Producers
2006, December 2007, Table 16, Data and Charts.

---

CRS-5
increase of 88%.3 Not all of this amount was for technological improvements or
capacity expansion; environmental based investment is also included.4
An additional factor in refining economics is crude oil sulfur content, defined
to also include the presence of heavy metals and contaminants. Low sulfur crude oil
is said to be “sweet” while crude oil with a high sulfur content is said to be “sour.”
Low sulfur crude oil sells at a premium compared to higher sulfur crude oil, because
substantial investment must be made at the refinery to reduce the sulfur content of
the product mix. These two factors, crude oil density and sulfur content, taken
together, determine the quality differential in crude oil prices. At any point in time,
refiners strive to run an optimal mix of crude oils through their refineries. The
optimal mix depends on the state of the refineries’ equipment and technology, the
desired output product mix, and the price spreads on available crude oils and
products. In the longer term, the refiner must decide whether to continue to invest
in new technologies as they become available, based on expected future values of
these factors.
Defining Profitability
Measures of economic performance in the refining industry usually begin with
the gross margin. The gross margin is defined as the difference between the
wholesale composite product price and the composite refiner acquisition cost of
crude oil. Both the product price and the acquisition cost of crude oil must be
composites, or weighted averages, because they reflect the multi-product nature of
refinery output as well as the multi-grade character of the crude oils used as refinery
inputs. The gross margin is a simple, first approach to refinery profitability. It is
computed as the total revenues from product sales minus the cost of the largest single
input in refinery operations, crude oil. However, other inputs are used and they too
generate costs, which leads to the net margin. The net margin is defined as the gross
margin minus petroleum product marketing costs, internal energy costs and other
operating costs. Both margins are usually expressed on a per barrel basis.
Over time, crude oil prices have been volatile, reflecting international political
and economic events. Both margins normally respond inversely to movements in the
price of crude oil. Other things held equal, an increase in crude oil prices would
reduce a positive margin and a decline in the price of oil would expand the margin.
Of course, other things are rarely equal. Changes in the price of crude oil may be
passed on to consumers of petroleum products. However, refiners’ ability to pass on
price increases and maintain sales volumes may not be equal for each component of
the product mix produced by the refinery. As a result, in practice, margins may
3 Energy Information Administration, Performance Profiles of Major Energy Producers,
2006, December 2007, Table 16, Data and Charts.
4 The FRS companies are those major energy companies that report to the Energy
Information Administration’s Financial Reporting System (FRS). To be included as a major
energy company the firm must satisfy at least one of the following: control at least 1% of
U.S. crude oil production or reserves, control at least 1% of U.S. natural gas production or
reserves, or control at least 1% of U.S. crude oil distillation capacity or product sales. In
2001 the FRS companies share of refined product output was 85%.

---

CRS-6
increase or decrease when the price of oil increases. Additionally, in some cases, the
event which accounted for the change in crude oil prices might itself alter specific
product demand levels. For example, when expectations concerning the onset of war
in Iraq drove up crude oil prices, they also reduced the demand for air travel, and
hence, jet fuel. Lower demand for air travel made it difficult for refiners to pass on
cost increases of crude oil while maintaining sales volumes for jet fuel.
As a result of these factors, margins are not stable, or even predictable, during
periods of crude oil price volatility. The outcome tends to depend on the nature and
magnitude of the crude oil price change as well as the level of demand in the product
markets. Over time, persistently low margins may have important consequences for
the industry. Low margins may lead to reduced refinery investment which ultimately
can lead to constrained domestic capacity and higher product prices. But low
margins also put pressure on refiners to reduce operating costs and spread fixed costs
over larger production runs in an effort to enhance margins.
The relationships between crude oil quality, product price mix, and
technological improvement make capital investment management important to the
refining industry. In refining, there are several competing demands for investment
funds. Capacity expansion is necessary to keep pace with growing demand. Siting
a new refinery is a long and expensive process. As a result, virtually all U.S.
capacity expansion in the last decade or more has come from enhancing and
modifying existing refineries. Technological investments can improve refinery
economics by allowing refiners to use cheaper heavy, sour, crude oils as inputs and
still produce a light, high value mix of products. Finally, investment must be
undertaken to keep both the refinery site and the products it produces in compliance
with evolving environmental standards. The issues surrounding refinery investment
are covered later in this report.
The most widely used measure of economic performance in the refining industry
is return on investment (ROI).5 As with the gross and net margins, assessing this
measure requires care. First, refining returns are usually reported together with
returns from marketing, largely determined by motor vehicle gasoline retail sales.
Marketing net income can also be affected by credit card business performance and
convenience store profits, neither of which is directly associated with the core
business of refining petroleum products. Second, depending on accounting
procedures and transfer pricing rules, an integrated oil company might report profit
as earned in various business activities of the company, contrasted to a company
whose only business is refining. Third, depreciation reduces the net value of assets
over time creating an upward bias in ROI over time for any specified level of net
income. Fourth, business strategy decisions, such as whether to distribute products
at the retail level through company-owned service stations, or through franchises, can
alter the measured ROI by affecting the invested capital base.
5 The EIA in compiling FRS data defines return on investment as net income earned by the
U.S. marketing/refining line of business (excluding unallocated items, mainly interest
expense) as a percentage of net fixed assets involved in U.S. refining and marketing.

---

CRS-7
Performance of the Refining Industry
Table 3 provides historical data on refining margins for U.S. major refining
firms in the FRS survey. Net margins have generally increased since the late 1990s,
except for a sharp decline in 2002. The relationship between the net and gross
margins was variable. For most of the year-to-year changes, the two margins track
each other, but in 1992-1993, and again in 1997-1998, and again in 2003-2004, the
gross margin decreased while the net margin increased. Net margins recovered in
2003 from the poor performance of 2002. Refining margins reached record highs in
2005 and remained high into 2007.
Table 3. Refining Margins of U.S. Major Oil Refiners, 1993-2006
(inflation adjusted dollars per barrel)
Year
Gross Margin
Net Margin
1993
8.63
0.91
1994
7.46
0.89
1995
6.77
0.60
1996
7.80
1.05
1997
8.01
1.74
1998
7.08
1.77
1999
6.30
1.27
2000
7.95
2.51
2001
8.60
2.99
2002
6.89
0.21
2003
8.36
2.18
2004
8.05
2.56
2005
10.18
3.62
2006
12.10
5.29
Source: Energy Information Administration, The U.S. Petroleum and Gasoline Marketing Industry,
Table 5, updated August 2004, and Performance Profiles of Major Energy Producers 2006, Table
14, Data and Charts. Margin data are expressed in terms of 2005 dollars (price adjusted) and on the
basis of dollars per barrel, where a barrel equals 42 gallons.
Two factors, which reflect refinery economics, emerge from the data underlying
these margins. First, general operating costs declined by about 50 percent over the
period. The refining industry has exerted significant effort in controlling cost and
enhancing efficiency. Second, energy costs related to the production process within
the refineries have declined. This is the result of significant effort by the refining
industry to harness waste energy and apply it to productive purpose in the interest of
reducing costs. For most of the period, the combination of volatile prices for crude
oil and refined products offset the increases in efficiency implemented at refineries
yielding low and volatile margins. After 2002, the financial performance of the
industry was enhanced by the rising petroleum product prices in the wake of the war

---

CRS-8
in Iraq, demand growth in China and other consuming nations, and the shortages that
resulted from refinery closures in the aftermath of hurricane Katrina.
Table 4 shows the mix of U.S. refinery outputs through 2006. Over-all output
increased, but not uniformly across the product lines. Motor gasoline output
increased through 2006, but was supplemented with imports of both finished gasoline
as well as blend stocks, to meet demand. Distillate fuel output, including diesel fuel
and home heating oil, also increased. Jet fuel output was a relatively stable
component of output, while residual fuel oils, the heavy component of the product
mix, declined.
Table 4. U.S. Refinery Output,1991-2006
(millions of barrels per day)
Motor
Distillate
Residual
Other
Year
Gasoline
Jet Fuel
Fuel Oil
Fuel Oil
Products
Total
1991
6.97
1.44
2.96
0.93
2.96
15.26
1992
7.05
1.40
2.97
0.89
3.09
15.40
1993
7.30
1.42
3.13
0.83
3.11
15.79
1994
7.18
1.45
3.20
0.83
3.13
15.79
1995
7.46
1.42
3.15
0.78
3.96
15.99
1996
7.56
1.51
3.32
0.73
3.20
16.32
1997
7.74
1.55
3.39
0.71
3.37
16.76
1998
7.89
1.53
3.42
0.76
3.43
17.03
1999
7.93
1.56
3.40
0.70
3.40
16.99
2000
7.95
1.61
3.58
0.70
3.40
17.24
2001
8.02
1.53
3.69
0.72
3.32
17.28
2002
8.18
1.51
3.59
0.60
4.59
17.27
2003
8.19
1.49
3.71
0.66
3.44
17.49
2004
8.26
1.55
3.81
0.65
3.54
17.81
2005
8.31
1.55
3.95
0.63
3.45
17.71
2006
8.32
1.48
4.04
0.63
3.43
17.91
Source: Energy Information Administration, Annual Energy Review 2006, Table 5.8, p. 139.
In 2004, refiners experienced a market characterized by increasing demand for
most products and increasing prices for refined products, along with rising prices for
crude oil. The potential for instability in crude oil supplies began to play a role in
late 2003, and affected the market even more in 2004, through 2006, continuing into
2008. Crude oil prices, as measured by the composite refiner acquisition cost, began
2004 at $30.93 and ended the year at $36.60, with an average price of $36.98,
reflecting concerns of the effects of war with Iraq, growing world demand, and

---

CRS-9
instability in other oil producing parts of the world.6 While motor gasoline demand
for 2004 increased by about 1%, the price of gasoline increased by about 24%. Jet
fuel demand increased by about 4% and the price of jet fuel rose by almost 38%.
Distillates demand increased by about 3% while the price increased by about 24%.
Residual demand decreased by about 1%, but the average price increased by about
6%. Overall, refined product revenues for the FRS companies increased by about
26% from 2004 to 2005 and total refined product costs rose by almost 21%.
Data for 2005 and 2006 suggest a much stronger financial performance for the
refining industry. Crude oil prices, based on the refiner’s average composite
acquisition cost, averaged over $50 per barrel in 2005, and over $60 per barrel in
2006. This led to dramatic spikes in motor gasoline prices and other petroleum
product prices. The airline industry has continued to show weak demand for jet fuel,
but the cold winter periods increased the demand for heating oil, and diesel fuel
demand continued to increase. In the short term, the demand for most petroleum
products appears to be relatively inelastic with respect to price, meaning that the
rising cost of crude oil can be passed on to consumers with little loss in sales volume
yielding high net income for refiners when oil prices are high.7

Table 5 provides an overview of profitability for the refining and marketing
sectors of the FRS’s companies, measured by ROI. The values for ROI are
complicated by the joint reporting of refining and marketing as noted earlier in this
report. These two sectors have different capital intensities and different age profiles
for their capital stocks leading to differing values of invested capital, net of
depreciation. The data shown in Table 5 are somewhat at odds with other measures
of industry activity, especially those related to capital investment. Even though ROI
has become negative twice in the data set, production rates and industry capacity
continue to expand, suggesting that an incentive for capital investment exists. How
might this behavior be explained? For the integrated oil companies, even though
profit rates might have been low and volatile in refining and marketing, it is possible
that the firm as a whole made adequate returns from other parts of the business,
notably, the production of crude oil. However, this explanation is not consistent with
the profit center business model the industry appears to be adopting. For the non-
integrated oil refiners who constitute a larger component of the industry since the
wave of merger and acquisition activity of the late 1990s, this explanation is not
applicable. The data in the Table 5 indicate that in 2000, ROI in domestic refining
and marketing was 9.6%, rising to 14.5% for 2001, an increase in the rate of return
of over 50%. The declining margins of 2002 left the ROI in the negative range for
that year, the first negative returns since 1992. For 2003, the FRS companies
6 Petroleum price data from the Energy Information Administration, Petroleum Supply
Monthly, Table 1, Crude Oil Prices, November, 2003. The prices reported by the EIA in this
table are the composite refiner acquisition price of crude oil which is a weighted average of
the prices refiners paid for domestic as well as imported crude oil.
7 Economists define demand to be inelastic when a specified change in price, say a 1%
increase, yields a decline in quantity demanded of less than one percent. Petroleum product
demand is more elastic in the longer term, as consumers are more able to adjust their
decisions to the higher prices.

---

CRS-10
domestic ROI from refining and marketing rose to 9.3%, setting the stage for the
even higher returns earned in 2004 through 2006.8
Table 5. Domestic Refining/Marketing Return on Investment for
FRS Companies, 1991-2006
Percent Return
Year
on Investment
1991
1.95 %
1992
-0.44 %
1993
3.38 %
1994
3.56 %
1995
1.00 %
1996
4.36 %
1997
6.59 %
1998
7.90 %
1999
6.54 %
2000
9.64 %
2001
14.46%
2002
-1.70%
2003
9.30%
2004
18.60 %
2005
23.50 %
2006
25.60%
Source: Energy Information Administration, Performance Profiles of the Major Energy Producers
2006, December 2007, Figure 6, Data and Charts.
The American Petroleum Institute computes statistics comparing the
profitability, based on ROI, in oil and gas production and refining and marketing, to
the returns earned by the Standard and Poor’s (S&P) Industrials, which they use as
a proxy for average returns to American industry. The American Petroleum Institute
found that the 2001 performance of the refining/marketing sector, 14.4%, represented
one of the few times that this sector outperformed both the S&P Industrials as well
as the oil and gas production sector. In 2001, the S&P Industrials earned 7% and oil
and gas production earned 13.1%. The averages over the past five and ten years tell
a different story, however. Ten year average returns for the S&P Industrials were
11.7% and the five year average was 14.3%. These figures compare with
refining/marketing average returns of 5.4% and 9.1% over the same periods.
8 Energy Information Administration, Performance Profiles of Major Energy Producers,
March 2005, Table B.8, p. 75.

---

CRS-11
Comparable average rates of return for oil and gas production were 8.4% and 10.3%.9
Table 6 shows that production capacity and capacity utilization are less volatile than
the returns earned by the industry and show relatively steady incremental increases
over the period.
Table 6. Refinery Capacity and Utilization, 1991-2006
Operable Refineries
Capacity
Utilization
Year
Number
(thousand of barrels per day)
(percent)
1991
202
15,707
86.0
1992
199
15,460
87.9
1993
187
15,143
91.5
1994
179
15,150
92.6
1995
175
15,346
92.0
1996
170
15,239
94.1
1997
164
15,594
95.2
1998
163
15,802
95.6
1999
159
16,282
95.6
2000
158
16,512
92.6
2001
155
16,582
92.6
2002
153
16,744
90.2
2003
149
16,748
92.6
2004
149
16,794
93.0
2005
148
17,196
90.6
2006
149
17,339
89.7
Source: Adapted from Energy Information Administration, Annual Energy Review, 2006,
Table 5.9, p. 141.
Domestic refining capacity has generally increased incrementally since 1993,
with total growth of about 15%. Refinery inputs have increased by approximately
11% over the period, with the slack being picked up by capacity utilization, which
ranged from a low of 87.9% to a high of 95.6% during the period, as well as refinery
gain. High capacity utilization rates leave a slim margin available to meet any
increase in demand, raising, at least the potential, of market disruptions, either
shortages or price spikes, in the retail market. High utilization rates also increase
potential system unreliability due to stress, damage, and difficulty in scheduling
down time for maintenance, repairs, and investment activities.
9 Michael A. Lobue, Challenges Facing U.S. Refiners, Presentation at NEMS/AEO
Conference, March 18, 2003.

---

CRS-12
While the picture of low and volatile returns might be considered somewhat
inconsistent with steady capacity growth and high capacity utilization rates, one
measure of return, refining/marketing net income, might be helpful in understanding
the data. Table 7 shows net income data for the FRS companies from 1995 to 2006.
Table 7. U.S. Refining/Marketing Net Income, 1995-2005
(million dollars)
Yearly Growth
Year
Net Income
(percent)
1995
508
1996
2,251
343
1997
3,106
38
1998
5,932
91
1999
4,883
-18
2000
7,659
57
2001
11,951
56
2002
-1,350
-119
2003
7,434
650
2004
15,197
105
2005
20,963
38
2006
24,313
16
Source: Energy Information Administration, Performance Profiles of the Major Energy Producers,
2005 and 2006.
Net income increased every year except 1999 and 2002. The year-to-year
changes in net income indicate substantial volatility; however, most of it was
favorable to the industry. While the change from 1995 to 1996 was a gain of over
340%, the change from 2001 to 2002 represented a decline of almost 120%,
associated with a $1.3 billion loss for the sector. The total number of dollars earned
rose from $508 million in 1995 to $11.9 billion in 2001, a twenty fold increase,
before turning into a loss in 2002. This increase in cash flow from the refining sector
could account for much of the interest the industry has shown in acquiring, investing
in, and operating refineries in the face of low rates of return on investment. Net
income declined sharply in 2002, falling to a loss of $1.35 billion for the FRS major
energy firms. Data for 2003 show a recovery in net income, with $10.2 billion being
earned from refining and marketing by the FRS firms. By 2006, net income from
refining and marketing had more than tripled compared to 2003.
Changing Structure of U.S. Refining Industry
The refining industry grew from 1973 to 1981. In 1973, there were 268
refineries with a capacity of 13.64 million barrels per day in the U.S., and by 1981,
there were 324 refineries with a capacity of 18.62 million barrels per day. As shown

---

CRS-13
in Table 6, by 2006 the number of refineries had declined to 149 with a capacity of
17.3 million barrels per day. Over the period 1981 to 2006, the number of refineries
declined by over 50% while the total capacity of the sector declined by about 8%.
Ownership of refining assets has changed as well. Table 8 shows the top twenty
U.S. refiners, rated by capacity. Several observations can be made about this table.
First, in 1980, ownership of refineries was largely held by the major, integrated U.S.
oil companies. In 2006, significant foreign ownership existed, including British
Petroleum, PLC, the fourth largest refiner, and PDV America, a subsidiary of the
Venezuelan national oil company. Second, the period since 1981 has seen significant
merger and acquisition activity within the U.S. oil industry as oil companies sought
to acquire expanded crude oil reserves as well as other assets through acquisition and
merger with other firms. As a result of mergers, acquisitions, and corporate strategy,
the ownership of many specific refineries has changed hands, even though the total
capacity of the corporate entity has remained relatively stable. For example, the
Bayway refinery in Linden, New Jersey, with a current capacity of 238,000 barrels
per day was owned by ExxonMobil, but is now owned by ConocoPhillips; Conoco
and Phillips Petroleum were separate companies in 2001. Third, the industry now
has significant capacity operated by firms that are not integrated oil companies at all,
but specialize in only a part of the petroleum supply chain. Fourth, specific joint
venture organizations, (e.g., Motiva Enterprises) have been created to engage
specifically in refining. Joint ventures have become common because they allow for
significant cost and risk reductions through asset sharing and rationalization, but do
not require the financial and institutional stresses of a more direct union through
merger or acquisition.

---

CRS-14
Table 8. Top Twenty U.S. Refiners, 1980-2005
(thousand barrels per day)
1980
2006
Company
Capacity
Company
Capacity
Exxon
1,557
Valero
2,219
Standard of California 1,383
ExxonMobil
1,862
Standard of Illinois
1,238
ConocoPhillips
1,778
Shell
1,123
BP PLC
1,460
Texaco
1,059
Chevron 1,011
Gulf
949
Marathon Oil
974
Mobil
835
Sunoco
903
Atlantic Richfield
811
PDV America
812
Sun
521
Koch Industries
777
Marathon 513
Motiva
Enterprises
762
Union
490
Tesoro Corp
567
Ashland
462
Royal Dutch Shell
532
Standard of Ohio
452
WRB Refining LLC
452
Conoco
448
Deer Park Refining
333
Phillips
397
Lyondell Chemical
270
Cities Service
291
Total SA
232
Union Pacific
283
Chalmette Refining
193
Coastal
278
Sinclair Oil Corp
161
Getty
261
Frontier Oil Refining
154
Tosco
260
Murphy Oil Corp
154
Source: American Petroleum Institute, Basic Petroleum Data Book, Section VIII, Table 11, October
2007.
The characteristics of the refining industry suggest that significant structural
change has occurred over the past twenty years. Much of that change can be
attributed to the forces of economic rationalization rather than to economic decline.
However, even if it can be determined that the industrial structure today is stronger
than that of the past, it still may be that the industry is not fully capable of meeting
the challenges of the future.
One of the most obvious structural changes undergone by the industry is the
number and size of available refineries. Clearly the number of refineries has sharply
decreased. Three factors play a role in explaining the decline: the end of government
regulation, technological developments, and economic reality.

The Emergency Petroleum Allocation Act of 1973 (P.L. 93-159) attempted to
ensure an equitable distribution of petroleum products, establish equitable prices, and
preserve the independent sector of the oil industry at a time when oil market
relationships were in disarray after the oil embargo of 1973. Through its Supplier-
Purchaser Rule, the Buy-Sell Program, and the Crude Oil Entitlement Program, all
of which tried to protect small, independent refiners, the act had the effect of

---

CRS-15
increasing the incentive to build and own small refineries, many of which were
inefficient. Between 1973 and 1981 there was a net expansion of 43 refineries (75
newly constructed, 32 old units closed). All but two of the new refineries had an
input capacity of less than 50,000 barrels per day, which was the average refinery size
in 1973. The average capacity of the new refineries that came on line over the period
was 14,900 barrels per day. The utilization rate of the refinery sector declined
during the expansion phase. In 1973, the capacity utilization rate was 93.9%, but by
1981 it had fallen to 68.6%. The industry was characterized by idle capacity, too
many producers, and many refineries too small to benefit from economies of scale.
In addition, many of the small refineries were very simple distillation facilities
unsuited to produce the product slate required by the market and environmental
requirements. As a result, a good portion of the output of these facilities was only
suited to serve as feedstock for larger, technically sophisticated refineries. President
Reagan ordered the elimination of the allocation programs that favored the small
refiners in 1981, and since then the industry has responded to market forces.10

Technological and economic rationales for the decline in the number of
refineries are linked. The market for petroleum products has become heavily skewed
toward a lighter product mix, mainly gasoline, to serve demand from the
transportation sector. Lighter product mixes are primarily produced by adding
additional technical processing units downstream from the basic distillation
procedure. The technology and economics of the industry suggest that investment
in those downstream facilities is only rational for large scale plants, due to
significant economies of scale. The net effect of these factors may be that the smaller
number of larger refineries operating in 2006 are economically and technologically
better prepared to meet the requirements of the market than the earlier structure, even
with far fewer refineries and a smaller total capacity.
Although foreign ownership of refineries on U.S. soil has increased, this may
be the result of a continuing process of internationalization of the oil industry. U.S.
major oil companies have long had a presence in all aspects of the industry around
the world. They have brought quality products as well as technological expertise and
investment to many countries. Today, with the largest oil companies even more
international in character than in the past, the United States is receiving a share of the
products, technology and investment that U.S. oil companies provided, and continue
to provide, around the world. A threat to U.S. supply security related to foreign
ownership might be possible if the owner of the refineries was controlled by foreign
governments with oil supply as well as hostile intentions to the United States, but this
does not seem to be the case in the present pattern of ownership.
Beyond the number of firms and their productive capacity, and the extent of
foreign ownership, another important part of the structure of the refining industry is
the degree of concentration. Concentration refers to the proportion of the total
market accounted for by the largest firms. In petroleum refining, concentration can
be determined on the national level and on the regional level. The results are quite
different. The largest national refining companies are not necessarily the most
10 Energy Information Administration, The U.S. Petroleum Refining Industry in the 1980’s,
October, 1990, p. 6.

---

CRS-16
important refiners in any given region, and the most important regional refiners are
not necessarily the national leaders.11 Table 9 shows these relationships for the
Petroleum Administration Defense Districts (PADDs).12
Table 9. Refining Concentration by PADD
(percentages)
Top 5 Refiners Share of Capacity by PADD
US
PADD1
PADD2
PADD3
PADD4
PADD5
1/1/06
49
87
63
48
51
71
1/1/02
44
90
56
53
62
69
1/1/96
32
71
48
40
59
61
National Top 5 Refiners Share of Capacity by PADD
US
PADD1
PADD2
PADD3
PADD4
PADD5
1/1/06
49
49
47
52
27
57
1/1/02
44
27
33
39
20
56
1/1/96
32
19
28
25
27
28
Note: PADD 1 includes Connecticut, Delaware, District of Columbia, Florida, Georgia, Maine,
Maryland, Massachusetts, New Hampshire, New Jersey, New York, North Carolina, Pennsylvania,
Rhode Island, South Carolina, Vermont, Virginia, and West Virginia. PADD 2 includes Illinois,
Indiana, Iowa, Kansas, Kentucky, Michigan, Minnesota, Missouri, Nebraska, North Dakota, Ohio,
Oklahoma, South Dakota, Tennessee, and Wisconsin. PADD3 includes Alabama, Arkansas,
Louisiana, Mississippi, New Mexico, and Texas. PADD4 includes Colorado, Idaho, Montana, Utah,
and Wyoming. PADD5 includes Alaska, Arizona, California, Hawaii, Nevada, Oregon, and
Washington.
Source: “Refining Concentration and Industry Dynamics,” Petroleum Industry Research Foundation,
Inc., N.Y., April 2002. p. 2-4, Energy Information Administration, [http://www.eia.doe.gov].
The table shows that, on the national level, the share of the industry held by the
largest five firms has increased from 32% in 1996 to 44% in 2002 and to 49% by
2006. However, the top five firms were different in 2006 than in 1996. The top
section of the table shows that concentration is generally higher on a regional level
than on the national level. The bottom part of the table reveals, however, that the
firms that dominate refining at the national level are not necessarily the same firms
that hold the largest shares of refining capacity in the PADDs.
PADD 3, which is the Gulf Coast, has the lowest regional concentration, 48%.
This region also has the second largest share held by the leading national refiners,
and the share of the national leaders has increased by 27% since 1996. However,
PADD 3 production is far greater than regional demand and pipelines connect the
major Gulf Coast refiners to other major consumption markets. As a result, refiners
11 The information on industry concentration in the PADDs draws from a report by the
Petroleum Industry Research Foundation, Inc., Refining Concentration and Industry
Dynamics, April 2002.
12 PADD is an acronym for Petroleum Administration for Defense Districts. The 50 states
and the District of Columbia were divided into five geographic districts in 1950.

---

CRS-17
in this region compete not only among themselves to satisfy local demand, but
against refiners in other PADDs.
PADD 5, the West Coast, presents unique features. The share held by the top
five regional refiners is over 70%, and the share held by the largest national refiners
is 57%. Several features also isolate the region from the rest of the nation. First,
there is no major finished product pipeline between PADD 5 and the rest of the
country. This effectively isolates the market from competition from other areas,
especially PADD 3, with its large refineries that compete in other regions of the
nation. Second, PADD 5 is characterized by the CARB (California Air Resources
Board) gasoline, a unique, state-mandated formulation which adds to the relative
isolation of this market by making gasoline from other areas generally unsuitable in
large portions of PADD 5. As a result, the Federal Trade Commission has been
especially active in requiring divestiture during mergers and acquisitions in this
PADD to assure continued competition. For example, both the Exxon Mobil and the
Chevron Texaco mergers required divestiture of PADD 5 refining assets by at least
one of the merging firms. Many other mergers, acquisitions, and joint ventures in
PADD 5 only achieved Federal Trade Commission approval after divestiture. In
some cases, these actions brought new competitors into the region, as when BP
acquired ARCO’s refineries.
PADD 1, the East Coast, could be characterized as an open market. While it is
connected, via pipeline, to the large refineries on the Gulf Coast, it is also accessible
to imports from overseas. Within the region the largest five national refiners have
a 49% share of the market, while the leading refiner, Sunoco, has approximately a
5% share of the national market.
PADD 4, the Rocky Mountain area, is again, similar to PADD 5, an isolated
market, but in this case the top five regional refiners hold a 51% share while the
national leaders hold a 27% share.13 PADD 2, the Midwest, has had a relatively
stable pattern of market share, and it shares with PADD 1 the characteristic of being
an open market. It is tied by pipeline to PADD 3, which supplies the region with
approximately 20% of its supplies.
In summary, even though there has been significant merger and acquisition
activity in the refining sector, the data for the period 1996 to 2001 do not indicate
sharply different patterns of industry concentration at the national or regional levels,
while concentration has increased from 2002 to 2006. Because of the economic
rationalization and concentration on core competencies that these mergers and
acquisitions seem to imply, the firms that remain in the industry might well be better
positioned for market competition.
13 As a result of the Conoco Phillips merger the share of Conoco, which was the largest
regional refiner, grew to 25% from 21%. This caused the top five share to rise to 66%.

---

CRS-18
Challenges Facing the Refining Industry
Historically, the major oil companies treated the refining activity as an
integrated part of a production stream that ran from exploration to final retail sale of
petroleum products. One implication of this business model was that the refining
sector was not necessarily considered to be a stand-alone profit center. Refinery
profits or losses could be integrated with other parts of the business. This model has
changed in the 2000s to one requiring each component of the supply chain to pull
its own weight, or generate sufficient profitability, to satisfy investors’ requirements
for return on invested capital. This changed business model results, at least partly,
from the entrance into the refining market of large, independent companies whose
major interest in the petroleum industry is downstream refining and marketing.14
Given the place of refined products in the economy, earning sufficient return on
invested capital in the refining industry should be possible. After all, the industry’s
output is largely characterized by products that are essential to modern life: motor
gasoline, diesel fuel, jet fuel, and other products whose demand has generally been
growing. The industry has also made significant progress in controlling operating
costs, although the most significant cost, that of crude oil, is beyond its control. The
basic challenge to the industry is to maintain a competitive ROI at the same time that
adequate capital is allocated to technological improvements, refinery expansion is
undertaken to meet forecasted increases in demand, and investments to keep refinery
products and sites consistent with environmental laws and regulations are made.15
If the refining industry is successful in meeting this challenge, a more stable market
will likely prevail. If industry is unsuccessful in meeting this challenge the market
is more likely to experience instability, characterized by supply disruptions and price
spikes, as well as an increasing dependence on foreign supplies of refined products.
Environmental Requirements
From a refiner’s point of view, most environmental requirements, whether they
affect product specification or site compliance, affect business operations in a similar
manner. They require capital investment in additional stages, or technical processes,
to be added to the refinery. The investment process begins with certification and
permit approval with regulatory agencies and, ultimately, moves to design and
construction at the refinery to implement the new processes. Some of the current,
or expected, product specification requirements that are likely to affect refiners over
the next several years include reduced sulfur content in gasoline and diesel fuels,
reductions in smog-forming compounds released during handling (vapor pressure),
reduced smog-forming emissions from vehicles (reformulated gasoline), reduced
toxics and chemical exposure during handling and storage (methyl tertiary butyl
ether (MTBE) bans), improved engine performance (drivability index), and the use
14 D.J. Peterson and Sergej Mahnovski, New Forces at Work in Refining, Industry Views os
Critical Business and Operations Trends, RAND Science and Technology, 2003, p. 14.
15 The effect of environmental requirements on industry profitability is analyzed in Energy
Information Administration, The Impact of Environmental Compliance Costs on U.S.
Refining Profitability 1995-2001, May 2003.

---

CRS-19
of non-petroleum based feedstocks (ethanol).16 The goal in this report is not to
evaluate the net benefit of any of these requirements, or environmental site
requirements which form an additional set of restrictions, but to recognize that they
all have a similar impact on the refining industry: they require an allocation of
resources to capital investment that does not result in lower costs or increased output,
and are, therefore, likely to affect the ability of the industry to meet economic
performance expectations.
Capital Investment
Capital expenditures in the refining industry serve a variety of purposes. First,
they may augment capacity or upgrade technological capability, either through the
construction of new refineries or through the modification of existing refineries.
Second, they may be required to allow both, or either, the produced products or the
site itself to meet environmental requirements. Third, capital expenditures may be
used for mergers and acquisitions.
As discussed earlier in this report, there has been no significant investment in
new refineries in the United States in the past twenty five years. However,
investment has taken place to enhance the capacity of existing facilities. Investment
in capacity expansion and technology is expected to generate profits enhancing the
net margin and yielding a positive ROI. Environmental investments are different.
They are required to allow the refinery to stay in operation, but they yield little or no
direct financial return for the refiner. They are either absorbed from profits or passed
on to the consumer, or both. Since they increase the capital value of the refinery, but
yield no net revenue, they can reduce the overall ROI, depending on cost pass-
through to the consumer.

Capital expenditure for merger and acquisition has dominated the investment
picture in the refining industry since the late 1990s. These expenditures are made for
a variety of reasons, from entering or expanding a firm’s presence in the industry, to
rationalizing refining strategy, to production and marketing goals, or by legal
directive as in conjunction with the stipulations to complete a merger or acquisition.
Firms expect these expenditures also to be profitable, at least in the longer term.
Even though refining and marketing profitability has grown with the oil price
increases since 2004, the profit and cash rich position of integrated oil companies and
domestic refining companies may not translate into capital investment in the refining
sector. The choice to invest in a new refinery is actually a decision to produce an
additional barrel of refined petroleum products for the U.S. market. Growth in
product demand, especially gasoline, suggests that the demand for the additional
product is fairly certain.
The question remains; what is the cheapest way to provide that extra product to
the market? Four possible answers exist. First, a company could decide to expand
the capacity of an existing refinery. Second, a company could build a new refinery.
Third, a company could decide to import the extra production into the United States.
16 Ibid., p. 22.

---

CRS-20
Fourth, a company could acquire the refinery of another company. The fundamental
economics, as well as the market and regulatory environments, suggest that, all other
things equal, construction of new refineries is not likely to be the chosen alternative.
Over the past two decades refiners have chosen to expand existing refineries in
preference to new construction. The reasons are part economic and part regulatory.
Economic rationales favor expansion over new construction, because the fixed cost
associated with a new facility can be avoided and economies of scale at the expanded
facility can be realized. Satisfying regulatory requirements for an expansion at an
existing facility are likely to be less costly, less time consuming, and less subject to
challenge by local groups compared to a new facility.
Given the economics of the oil market, it may remain cheaper to import extra
petroleum products than to produce them in the United States. Excess capacity for
gasoline that meets U.S. specifications is available in Europe and other parts of the
world. Similarly, it may also be cheaper for any given company to expand its
refining capacity by purchasing assets from other companies. While this strategy
may improve the supply position of the acquiring firm, it does little to improve the
supply position of the U.S. market as a whole.
Special circumstances may justify construction of a new facility. Such a facility
is planned for construction in Arizona.17 In the case of the Arizona facility, the
refinery is positioned to supply products to the persistently tight southern California
market, where high margins might be earned, using Mexican crude oil as an input.
Although the permitting process for this facility is nearing completion, it has been in
process for over five years.
Another new proposed U.S. refinery project is to be located in South Dakota,
and produce ultra low sulfur gasoline and diesel fuel. The source of crude oil would
be from Canadian oil sands, and the refinery is expected to use about 400,000 b/d of
crude oil. The cost of the refinery has been estimated at between $8 and $10
billion.18
17 Sarah Reynolds, East County Oil Refinery Will Move, The Yuma Sun, February 8, 2008.
Available at [http://www.arizonacleanfuels.com/2008/020508_YS.htm].
18 New York Times, Refinery Considered for South Dakota, June 14, 2007. Available at
[http://www.nytimes.com/2007/06/14/business/14oil.html].

---

CRS-21
Figure 1. U.S. Refining Capital Expenditures for FRS Companies,
1988-2001
Source: Energy Information Administration, The Impact of Environmental Compliance Costs on U.S.
Refining Profitability, May 2003, Table 6, p. 10.
*Mergers and Acquisitions.

Figure 1 shows the behavior of refining capital investment by the FRS
companies since the late 1980s, isolating both environmental and merger and
acquisition based expenditures. The effect of environmental requirements,
specifically the Clean Air Act Amendments of 1990 (P.L. 101-549) on capital
expenditures are shown in the figure. In the late 1980s capital expenditures for
environmental purposes averaged $0.5 billion annually in inflation adjusted dollars.
To meet the oxygenated gasoline, reduced sulfur diesel fuel, and reformulated
gasoline mandates of P.L. 101-549, investment peaked at $2.7 billion in 1992 and
remained above $2 billion for the following two years, in inflation adjusted dollars.
According to the EIA, these environmental investments made substantial
contributions to the reduced ROI of the refining industry during the period 1991 to
1995. The EIA has determined that ROI was reduced by 69% over the period as a
result of the implementation of Federal environmental statutes and/or the Clean Air
Act Amendments. Both Phase I and Phase II of the complex emission regulations for
reformulated gasoline, in 1998 and 2000 respectively, caused increases in capital
expenditures in anticipation of their effective dates. The EIA determined that ROI
was reduced by 42% from 1996 to 2001 as a result of environmentally based capital
expenditures.19
The challenge for the industry is how it responds to the next round of required
environmental investments given this experience. Estimates forwarded in the 2003
19 The Energy Information Administration, The Impact of Environmental Compliance Costs
on U.S. Refining Profitability 1995-2001, May, 2003, p. 1. [http://www.eia.doe.gov/emeu
/perfpro/ref_pi2/refpi2.pdf]

---

CRS-22
RAND study of refiners are $4-13 billion to meet the on-road ultra low sulfur diesel
requirements by 2006, and perhaps $10-15 billion additional to meet other
mandates.20 The American Petroleum Institute sees the total cost of meeting the 2006
on-road ultra low sulfur diesel requirements at $8 billion. It sees an additional $8
billion investment to meet gasoline sulfur reduction requirements.21 The range of
investment cited is large and significant uncertainty remains concerning actual
values. If accurate, there could be some concern that environmentally based claims
on capital expenditure might crowd out other investments. In that case, investment
in capacity expansion might not be funded, even though potentially profitable. That
lack of investment could drive up capacity utilization rates even higher which would
likely increase the fragility of the market and make it more open to price volatility
and quantity disruption.
The strength of the crowding-out argument diminishes when the rapid increase
in capital expenditure shown in Figure 1 for mergers and acquisitions from 1999 to
2001 is considered. Clearly, the ability to raise capital, either in the form of debt or
equity, must have been available to the industry to fund the approximately threefold
increase in investment from 1999 to 2001, much of it to finance mergers and
acquisitions. The ability of the industry to manage and balance its capital investment
budget between competing claims may well be an important determinant of how well
the industry is able to meet increasing demand with new, environmentally compatible
capacity.
Production and Growth
In its reference case forecast, the EIA projects that refined products consumed
will increase from 20.7 million barrels per day in 2004 to 22.7 million barrels per day
in 2015 and 26.9 million barrels per day in 2030.22 This growth implies an average
annual growth rate of 1.1% over the period. Another way of looking at the projected
growth is to note that if it were to be met by domestic refining capacity, it would
require capacity increments of approximately 238,000 barrels per day, per year. This
would be equivalent to the production capacity of one new, large, refinery. An
additional, similarly sized refinery would have to be added to the capacity base each
year to meet projected market requirements. Alternatively, capacity expansion based
on technological improvements would have to continue at a high rate for the 26-year
period.
Set against this forecast of increasing demand is the incrementally increasing
capacity of the industry, as well as the declining number of refineries. The last new
refinery to open in the United States was the Petro Star refinery in Valdez, Alaska,
in 1993, which had an original capacity of 38,000 barrels per day and was a simple
refinery. The last refinery of significant capacity (200,000 barrels per day, or more)
20 D.J. Peterson and Sergej Mahnovski, New Forces at Work in Refining, Industry Views of
Critical Business and Operations Trends, RAND Science and Technology, 2003, p. 64.
21 Michael A. Lobue, Challenges Facing U.S. Refiners,NEMS/AEO Conference, March 18,
2003.
22 Energy Information Administration, Annual Energy Outlook, 2007, Table A11, p. 156.

---

CRS-23
to open in the United States was the Marathon refinery in Garyville, Louisiana,
which opened in 1977, more than 30 years ago.23 With this historical record,
construction of a new 400,000 barrel-per-day capacity refinery per year to meet
projected demand growth would appear to be a major challenge.
Several factors come into play in the refining industry’s plan to meet forecasted
market requirements. Even though no new refineries were constructed during the
1990s, and the number of refineries declined, refinery capacity increased on average
1.5% per year. If projected into the future, this growth is sufficient to meet the EIA
long term forecast of demand increases for refined products. These increases in
capacity are due to investment in new processing units, marginal expansion (known
as capacity creep) obtained through bottleneck removal, and more intensive use of
existing capacity by lengthening time between maintenance and overhaul. This type
of expansion ultimately could encounter diminishing returns, although the record of
the industry suggests that even more capacity might be squeezed out of the existing
stock of refineries.
Another key factor in maintaining production is capacity utilization. Capacity
utilization rates remain high, approximately 93% in 2003, and almost 91% in 2002.
The capacity utilization rate has remained above 90% every year since 1993. Excess
capacity is a luxury no refiner seems anxious to support, because of its effect on
profitability. This view is borne out by executives who participated in the RAND
study.24
Imports
With capacity utilization rates near a maximum, no new refinery construction
likely, and growth through capacity creep less than the growth in EIA projected
demand, imports of petroleum products would seem to play an increasingly important
role in the supply of refined products. Refined product imports, mostly from Canada,
the Caribbean and Europe have been an important component of the total supply
equation for some years. In 2006, petroleum product imports of all types, totaled
about 3.5 million barrels per day, which amounted to approximately 17% of total
product supplied to the U.S. market. Import dependence on refined products has
averaged approximately 6% of total product supplied since 1992. However, the
percentage has increased in each of the past six years. Most observers feel that there
is sufficient short run capacity in the world market to expand U.S. imports, with the
expansion of specific blends of motor gasoline for specific regional or state markets
posing the greatest potential challenge. Table 10 shows data from 1992 to 2006 for
total petroleum product imports, as well as the most significant sector, motor
gasoline, and blending components.
23 Data provided by the Energy Information Administration, Office of Oil and Gas,
Petroleum Division by personal communication, September 12, 2003.
24 D.J. Peterson and Sergej Mahnovski, New Forces at Work in Refining, Industry Views of
Critical Business and Operations Trends, RAND Science and Technology, 2003, p. xv.

---

CRS-24
Table 10. Imports of Petroleum Products, 1992-2006
(thousands of barrels per day)
All Products,
Net of
Gasoline
All
Gasoline, and
Motor
Blending
plus
Product
Blending
Year
Gasoline
Components
Blending
Total
Components
Components
1992
294
41
355
1,805
1,470
1993
247
27
274
1,833
1,559
1994
356
20
376
1,933
1,557
1995
265
48
313
1,605
1,292
1996
336
166
502
1,971
1,469
1997
309
200
509
1,936
1,427
1998
311
209
520
2,002
1,482
1999
382
217
599
2,122
1,523
2000
427
223
650
2,389
1,739
2001
454
298
752
2,543
1,791
2002
498
311
815
2,370
1,496
2003
518
367
885
2,599
1,714
2004
496
451
947
3,057
2,110
2005
604
494
1,098
3,471
2,373
2006
477
669
1,146
3,517
2,371
Source: Energy Information Administration, Annual Energy Review 2006, Table 5.3, p.129,
and computations by the author.
The table shows that imports of motor gasoline and blending components began
a more or less steady upward movement in 1996. Imports of these two items
increased by 60% in 1996 compared to 1995, and, since then, have increased by
another 128%. Over the period covered in Table 10, the increase in motor gasoline
and blending components accounts for about 80% of the increase in the imports of
all petroleum products. The total increase in petroleum product imports without
motor gasoline and blending components is less than 61% over the period, while the
increase in motor gasoline and blending components is 223%. Imported petroleum
products other than gasoline and gasoline blending components include a wide
variety of distillate fuel oils, jet fuel, liquefied petroleum gases, residual fuel oil,
unfinished oils and other products.
Several difficulties could emerge as increased reliance is placed on imports.
The first issue is sourcing. If worldwide economic growth continues to remain
sluggish, an excess of world refining capacity could persist and the U.S. market
might well continue to find product available. European sources accounted for over
50% of motor vehicle gasoline and 40% of gasoline blending component imports in
2007. Continued availability of European stocks at volumes comparable or greater

---

CRS-25
than these levels may not be sustainable. Europe is undergoing a transformation of
its transportation fuels mix. While its refinery capacity is largely oriented toward
gasoline production due to past investment decisions, the European vehicle fleet is
rapidly moving toward diesel fuel. The implication is that, until refinery conversion
is completed, European refiners will have an excess supply of gasoline to sell on the
world market. As a result, the long term availability of European supplies of gasoline
may diminish as the U.S. demand for gasoline continues to grow. Slow growth in the
Asian economies, excluding China, continues to hold back the growth of demand for
gasoline in the region. Before the financial crisis in 1997, Asian gasoline growth
rates were the highest in the world. Until economic recovery is complete and
demand growth comparable to pre-crisis levels returns, the Asian market is also
likely to have excess refining capacity. Near term sourcing of imports then, from
Europe as well as Asia, seems relatively secure. In the longer term structural change
in the European fuel mix and the recovery of Asian demand growth may limit
available imports.
The second factor influencing the viability of increased reliance on imports is
the segmented nature of the U.S. gasoline market. Gasoline in the United States is
no longer a fungible, or easily transferable, product. Differing U.S., regional, and
state air quality standards with respect to fuel specifications have produced a market
where shortages may appear because gasoline of a particular specification is tight,
even though gasoline of other specifications might be available. As the large U.S.
market is divided into smaller sub-markets that demand fuels not required in other
regions, or other parts of the world, it might become difficult to find refiners on the
world market willing to customize production to satisfy particular U.S. demands. The
result can be that spot shortages and price spikes, similar to those that occurred in
Chicago in 2002 in conjunction with the elimination of MTBE as a gasoline additive.
Availability, as well as price considerations, suggest that if gasoline were a
standardized commodity rather than a specialized good, it would be less likely to be
linked to local market instability.
A final factor regarding imports has to do with long and short run matching of
demand and supply. The EIA forecast indicates a long term need for imported
refined products, mainly gasoline and blending components. Recently, there has
been a growing market in very short run “target of opportunity” cargoes of refined
products. That is, an almost speculative market has developed in which a tanker
might load a cargo without a firm buyer at a port where gasoline is available at an
attractive price. The owner of the cargo then directs the tanker to the port at which
the best price might be earned. On the one hand, this type of activity represents
market forces directing supply to areas with high demand. On the other hand, the
process may represent a somewhat opportunistic, and uncertain way of securing
supplies. It is not hard to imagine a scenario where this type of supply dependency
might periodically lead to market disruptions complete with shortages and price
spikes.
Permits and Regulatory Uncertainty
An important concern of the refining industry is largely administrative and
regulatory. As noted earlier in this report, the refining industry is working within a
long term program of implementing congressionally mandated environmental

---

CRS-26
standards. Some see the regulated environment as an opportunity. Some participants
in the RAND study noted that regulations formed a barrier to entry into the market.
The result might be higher profits, for which they cite California refiners who are
faced with the most stringent set of federal, state, and local environmental regulations
in the nation, and yet operate the most profitable refineries in the nation.25 In general,
however, the RAND study reports significant concern in the industry regarding the
administration and implementation of the regulatory process.26
According to the RAND study, the best case regulatory process from the
industry’s point of view is one that minimizes uncertainty and maximizes flexibility.
The industry’s preference is to see a regulatory roadmap laid out which covered an
adequately long planning horizon, say a decade. They would like to see a plan where
no regulations contradict or compete with one another. They would like to see
phasing in of regulations in a manner that minimizes pressure put on construction or
investment timetables and plans. They would like to see a smooth, timely permitting
process in which administrative delays are minimized. Industry would also prefer to
have flexibility in how, technologically, to meet environmental standards.27
Conclusions
The petroleum refining industry forms a critical infrastructure in the U.S.
economy. It provides products which allow America to move, whether by air, land,
or sea, but it must accomplish this while maintaining national and regional
environmental standards. The industry has demonstrated that this is possible. There
has been minimal market disruption that can be traced to the refining sector, over the
past twenty years. However, the EIA Annual Energy Forecast 2006 suggests that the
industry is unlikely to be able to maintain this performance over the planning period
to 2030 without the domestic market becoming significantly more dependent on
imported motor gasolines and blending components. Increased import dependency
implies the potential for both supply disruption as well as national energy security
concerns.
The EIA has found that the legacy of past environmental requirements on the
industry have been substantially reduced ROI. The industry has been able to absorb
these lower rates of return in the past largely because of the ownership structure of
the industry and the associated business model. A new market structure and business
model might implicitly require better economic performance from the industry. To
the extent that the performance of ROI negatively affects decisions concerning
continued capacity expansion and technological investment in the industry, the effect
of the structural changes in the industry might have national energy policy
implications.
25 D.J. Peterson and Sergej Mahnovski, New Forces at Work in Refining, RAND Science and
Technology, p. 85.
26 Ibid., p. 70-79.
27 Ibid.

---