OBSERVATIONS ON PETROLEUM PRODUCT SUPPLY · 2006-02-28 · observations on petroleum product supply...

OBSERVATIONSON PETROLEUM

PRODUCT SUPPLYA SUPPLEMENT TO THE NPC REPORTS:

U.S. PETROLEUM PRODUCT SUPPLY–INVENTORY DYNAMICS, 1998

AND

U.S. PETROLEUM REFINING–ASSURING THE ADEQUACY

AND AFFORDABILITY

OF CLEANER FUELS, 2000

NATIONAL PETROLEUM COUNCIL • DECEMBER 2004

recycledpaper

NATIONAL PETROLEUM COUNCIL1625 K Street, NW

Suite 600Washington, DC 20006

OBSERVATIO

NS O

N PETRO

LEUM PRO

DUCT SUPPLY

NATIO

NA

L PETROLEUM

COUN

CIL • DECEM

BER 2004




AND


AND AFFORDABILITY


NATIONAL PETROLEUM COUNCIL • DECEMBER 2004Bobby S. Shackouls, Chair

NATIONAL PETROLEUM COUNCIL

Bobby S. Shackouls, ChairLee R. Raymond, Vice Chair

Marshall W. Nichols, Executive Director

U.S. DEPARTMENT OF ENERGY

Spencer Abraham, Secretary

The National Petroleum Council is a federaladvisory committee to the Secretary of Energy.

The sole purpose of the National Petroleum Council is to advise, inform, and make recommendations

to the Secretary of Energy on any matter requested by the Secretary

relating to oil and natural gas or to the oil and gas industries.

All Rights ReservedLibrary of Congress Catalog Card Number: 2004113953

© National Petroleum Council 2004Printed in the United States of America

Preface.................................................................................................................................................... 1

Study Request.................................................................................................................................. 1

Study Organization ......................................................................................................................... 1

Study Approach ............................................................................................................................... 2

Issues Not Analyzed......................................................................................................................... 2

Study Report.................................................................................................................................... 3

Executive Summary of Findings and Recommendations.................................................................. 5

Refining Capacity and Import Availability Findings...................................................................... 5

Capacity Growth ....................................................................................................................... 5

Ultra Low Sulfur Diesel ............................................................................................................ 6

Inventory Findings .......................................................................................................................... 6

Inventory Trends....................................................................................................................... 6

Lower Operational Inventory ................................................................................................... 7

Price Volatility........................................................................................................................... 7

Relationship of Inventories to Price ........................................................................................ 7

Strategic Petroleum Reserve.................................................................................................... 8

Recommendations........................................................................................................................... 8

New Source Review................................................................................................................... 8

National Ambient Air Quality Standards ................................................................................. 8

Implementation of Ultra Low Sulfur Diesel Regulations ....................................................... 8

National Energy Legislation..................................................................................................... 9

Sound Science, Cost Effectiveness, and Energy Analysis ....................................................... 9

Permitting ................................................................................................................................. 9

Depreciation Schedule Adjustment ....................................................................................... 10

Fuel Waivers and Enforcement Discretion............................................................................ 10

Alternative Fuels ..................................................................................................................... 10

TABLE OF CONTENTS i

TABLE OFCONTENTS

Distillation and Driveability Index ......................................................................................... 10

Site Security............................................................................................................................ 10

2005 Domestic Capacity ......................................................................................................... 10

Part I: Refining Capacity and Product Import Availability ............................................................ I-1

Observations on Light Product Supply and Demand Since the 2000 NPC Refining Study...... I-2

Domestic Light Product Production ..................................................................................... I-2

Refinery Distillation and Conversion Capacity Changes ............................................... I-2

Light Product Yield Changes .......................................................................................... I-3

Capacity Utilization Changes .......................................................................................... I-5

Recent Import Trends ............................................................................................................ I-6

Jet Fuel Domestic Production and Imports Respond to Demand Changes.................. I-7

Increased Distillate Imports Tied to Canadian Refinery Expansion.............................. I-7

Gasoline Imports from Europe Increasing..................................................................... I-9

New York and Connecticut MTBE Bans Illustrate the Market at Work...................... I-14

Factors That Potentially Impact the Growth in Domestic Product Producibility................... I-14

Refining Industry Financial Performance........................................................................... I-14

Capital and Other Resource Constraints............................................................................. I-16

Ever More Restrictive Product Quality Requirements ....................................................... I-18

Increasing Project Costs and Uncertainty from Environmental Regulations................... I-18

Waivers Can Have a Negative Impact on the Investment for Clean Fuels ........................ I-19

Environmental Operating Restrictions ............................................................................... I-19

Facility Security Environment Could Potentially Impact Operations............................... I-19

Grassroots Construction vs. Refinery Expansions .................................................................... I-19

Near-Term Outlook..................................................................................................................... I-22

No Serious Supply Issues Anticipated with Low Sulfur Gasoline Implementation.......... I-22

Boutique Fuels Fragment Markets and Reduce Supply Efficiency.................................... I-23

Significant Supply Concerns Exist for Ultra Low Sulfur Highway Diesel Implementation.................................................................................................................... I-24

Recommendations ...................................................................................................................... I-25

New Source Review.............................................................................................................. I-26

National Ambient Air Quality Standards............................................................................. I-26

Implementation of Ultra Low Sulfur Diesel Regulations................................................... I-26

National Energy Legislation ................................................................................................ I-26

Sound Science, Cost Effectiveness, and Energy Analysis................................................... I-27

Permitting ............................................................................................................................ I-27

Depreciation Schedule Adjustment..................................................................................... I-27

Fuel Waivers and Enforcement Discretion ......................................................................... I-28

ii OBSERVATIONS ON PETROLEUM PRODUCT SUPPLY

Alternative Fuels .................................................................................................................. I-28

Distillation and Driveability Index ...................................................................................... I-28

Site Security ......................................................................................................................... I-28

Reducing Distribution Incidents Would Improve Supply Reliability ................................ I-28

Access to Ports ..................................................................................................................... I-28

2005 Domestic Capacity....................................................................................................... I-29

Longer-Term (to 2010) Observations......................................................................................... I-29

Part II: Observations on U.S. Petroleum Inventories .................................................................. II-1

Role of Inventory......................................................................................................................... II-1

Inventory as an Operational Necessity................................................................................. II-1

Inventory as a Component of Supply................................................................................... II-2

Inventory as a Financial Opportunity .................................................................................. II-3

Inventory as a Cost of Doing Business ................................................................................ II-3

Inventory as Protection Against a Worldwide Emergency.................................................. II-3

Supply/Demand Imbalances ................................................................................................. II-4

Imbalances Sometimes Coincide with Industry Maintenance ..................................... II-4

Unanticipated Events ..................................................................................................... II-4

Changes in Fuels Quality Specifications Can Impact Inventories ............................... II-5

Inventory Trends and Lower Operational Inventory ................................................................. II-5

Crude Oil ............................................................................................................................... II-6

Motor Gasoline...................................................................................................................... II-6

Distillate ................................................................................................................................ II-8

Lower Operational Inventory ............................................................................................... II-9

Prior Studies Have Been Unable to Define an Inventory Level Where Supply Problems Begin ...................................................................................... II-9

LOI – One of Many Measures for Assessing Inventory Adequacy................................. II-9

Crude Oil LOI Recommended to be a Range of 260-270 Million Barrels.................... II-9

Data Suggest the 1998 Analysis of Product LOI is Still Valid .................................... II-10

The Importance of High Quality Inventory Data........................................................ II-10

Assessment of Product Price Volatility..................................................................................... II-10

Increase in Global Crude Oil Price Level........................................................................... II-11

Increase in Overall Retail Gasoline Price Level................................................................. II-11

Crude Oil Prices are Still the Main Driver of Product Prices........................................... II-13

Relationship of Crude Oil to Spot Product Prices ...................................................... II-14

Relationship of Spot to Retail Product Prices ............................................................ II-14

Product Price Volatility ...................................................................................................... II-15

Price Events.................................................................................................................. II-15

Volatility........................................................................................................................ II-15

TABLE OF CONTENTS iii

Future Product Price Volatility .......................................................................................... II-17

Factors That Potentially Increase Volatility ................................................................ II-17

Factors That Potentially Decrease Volatility ............................................................... II-17

Assessment of Inventories and Price ........................................................................................ II-17

Inventories, Prices, and Rational Market Behavior........................................................... II-18

Relationship between Inventories and Prices.................................................................... II-18

Gasoline Inventories and Gasoline Price Spreads ............................................................. II-19

Inventory Levels and Future Price Spreads ..................................................................... II-19

Inventory Levels Influence the Forward Price Spread...................................................... II-21

Price Expectations Influence Inventories .......................................................................... II-21

Appendices

Appendix A: Request Letter and Description of the NPC.......................................................... A-1

Appendix B: Study Group Rosters ............................................................................................. B-1

Appendix C: 2000 Refining Study Recommendations and Status............................................ C-1

Appendix D: Cumulative Regulatory Impacts on Refineries, 2000-2008.................................. D-1

Appendix E: New Source Review (NSR) Reform Background.................................................. E-1

iv OBSERVATIONS ON PETROLEUM PRODUCT SUPPLY

Study Request

In his address to the National Petroleum Council(NPC) at its June 22, 2004 meeting, Secretary ofEnergy Spencer Abraham expressed concern aboutthe adequacy of U.S. refining capacity and petrole-um inventory levels, stated that he would request astudy on these subjects, and noted that he wouldlike the Council’s advice by September 30, 2004.1

By letter dated July 16, 2004, Secretary Abrahamformally requested the National Petroleum Councilto undertake a new study on refining and inventoryissues that would provide advice on the challengesrelated to petroleum refining and product supply inthe United States. Specifically, the Secretary’s letterstates:

Accordingly, I request that the NationalPetroleum Council identify the factors that willimpact the refining and distribution industry’sability to meet future product demand, andreport on potential near-term options to meetdemand for transportation fuels and heating oilover the next year. Additionally, I ask that theCouncil reexamine its 1998 advice on loweroperational inventory levels for crude oil andpetroleum products.

Items to consider include the current andfuture demand for refined products, domesticcapacity to meet this demand, potential barri-ers to efficient markets, the influence of petro-leum product supply on price, industry actions

to meet environmental requirements, and thecapital investment and other factors that willdrive supply growth. Additionally, I wouldappreciate the Council’s insights on how refin-ing capacity, inventories, and demand patternsoutside the United States may impact meetingthe consumer demand for refined petroleumdomestically.

(Appendix A contains the complete text of theSecretary’s request letter and a description of theNPC.)

As noted in the Secretary’s request, inventoryissues were most recently addressed by the Councilin its 1998 report, U.S. Petroleum Product Supply—Inventory Dynamics. That report provided adviceon the interrelationships between inventories andprices and estimated lower operational inventory(LOI) levels for crude oil and light petroleum prod-ucts. Similarly, refining issues were last addressedby the Council in its 2000 report, U.S. PetroleumRefining—Assuring the Adequacy and Affordabilityof Cleaner Fuels. That report provided insights tohelp ensure a reliable supply of light petroleumproducts to consumers. Those reports had notionaltime frames of 2002 and 2005, respectively.

Study Organization

In response to the Secretary’s request, theCouncil established a Committee on Refining andInventory Issues to undertake a new study on thesetopics and to supervise the preparation of a draftreport for the Council’s consideration. The Councilalso established a Refining Subcommittee and anInventory Subcommittee to assist the Committee inconducting the study.

PREFACE 1

PREFACE

1 The Council’s study was conducted during August and September,but due to unavoidable conflicts this report was not presented to theSecretary until December 1, 2004.

Lee R. Raymond, Chairman and Chief ExecutiveOfficer, Exxon Mobil Corporation, served as theCommittee’s Cochair for Refining; James J. Mulva,Chairman of the Board and Chief Executive Officer,ConocoPhillips, served as the Committee’s Cochairfor Inventory; and David K. Garman, Acting Under Secretary of Energy, served as theCommittee’s Government Cochair. Donald H.Daigle, Vice President, Refining, ExxonMobilRefining and Supply Company, chaired the RefiningSubcommittee; Philip L. Frederickson, ExecutiveVice President, Commercial, ConocoPhillips,chaired the Inventory Subcommittee; and Mark R.Maddox, Acting Assistant Secretary, Fossil Energy,U.S. Department of Energy, served as GovernmentCochair of both Subcommittees.

The members of the Committee and its Sub-committees were drawn from the NPC member-ship, NPC member and other organizations, andgovernment. These study participants representedbroad and diverse interests including large andsmall petroleum refiners, transporters, marketers,and consumers as well as providers of financial andconsultant services. They brought their substan-tial individual experiences to the collective effortof producing this report. (Appendix B containsrosters of the study’s Committee and Subcommit-tees.)

Study Approach

Given the interrelated nature of refining andinventory issues, the Council determined that asingle study would be the most efficient way to pre-pare a response to the Secretary’s request foradvice. Also given the study’s limited time frame,the Council determined that the scope of the studywould be to review and supplement the most recentNPC reports on these issues and would be qualita-tive rather than quantitative in its approach. Inspite of this qualitative nature, the study results aresoundly based in the depth of experience of theindividual study participants. Broadly, the newstudy:• Performed qualitative reviews of the NPC’s 1998

inventory study and the 2000 refining study, andreaffirmed or modified recommendations asappropriate:

– Reviewed and validated major assumptions

– Assessed the government’s response to recom-mendations

– Focused on the remainder of this decade.

• Developed consensus on additional observations/recommendations based on industry experiencesince the 1998 and 2000 studies.

Specifically, the refining issues analysis:

• Assessed the 2000 refining study, as describedabove

• Evaluated refinery capacity growth versusdemand since 2000

• Considered the most recent EIA demand forecast

• Identified factors affecting domestic refinerycapacity expansions for the remainder of thisdecade, including notional grassroots refineryeconomics

• Identified factors affecting availability of importsto augment domestic supplies for the remainderof this decade

• Identified potential near-term governmentactions that could affect refinery capacity andimport availability over the next year

• Identified how refining capacity outside theUnited States might affect industry’s ability tomeet domestic demand.

Specifically, the inventory issues analysis:

• Assessed the 1998 inventory study, as describedabove

• Identified the role of inventories in the supply system

• Identified inventory trends and evaluated LOI lev-els for crude oil and major petroleum products

• Assessed product price volatility

• Evaluated the 1998 study’s conclusion regardinginventories’ relationship to price.

Issues Not Analyzed

Given the limited time frame in which to com-plete the study effort, the Council and theDepartment of Energy agreed on a targeted scope ofissues to be addressed. The Council recognized thatthis approach might preclude some issues, whichcould be important in public policy formation, from

2 OBSERVATIONS ON PETROLEUM PRODUCT SUPPLY

being adequately addressed and analyzed. That thisstudy does not include quantitative analyses of theseissues is a reflection of time and data constraints,and inferences should not be drawn on indicationsof their importance.

Specifically, the refining and inventory issuesanalyses used publicly available forecasts and did not:

• Provide a quantitative analysis or forecast ofdemand for crude oil and refined petroleum prod-ucts, domestic or worldwide

• Provide a quantitative analysis or forecast ofdomestic or worldwide refining capacity andcapability to meet changing product specifica-tions

• Provide a quantitative analysis or forecast of theavailability of imports into the United States

• Assess any additional infrastructure requirements(ports, tanks, pipelines, etc.) necessary to handleincreased petroleum imports

• Assess any national security implications of thepotential for increased reliance on imports

• Assess the potential for public policy actions toreduce fuel demand to worsen the investment cli-mate for expansion of domestic refining

• Reassess 2000 refining study production cost esti-mates

• Forecast prices or the potential price effects ofactions, but did make qualitative assessments offactors that impact price volatility.

Study Report

Results of the 2004 NPC review of refining andinventory issues are intended as an update and sup-plement to the 1998 and 2000 NPC reports on these

subjects and are presented in this report volume asfollows:

• Executive Summary of Findings andRecommendations presents observations onpetroleum product supply and an integratedresponse to the Secretary’s request for advice onboth refining and inventory issues. It providesinsights on petroleum market dynamics as well asadvice on actions that can be taken by industryand government to ensure adequate and reliablesupplies of petroleum products to meet the ener-gy and environmental requirements of Americanconsumers. It includes an overview of the study’sinterrelated refining and inventory analyses andrecommendations.

• Part I, Refining Capacity and Product ImportAvailability, contains the results of the RefiningSubcommittee’s review and update of the 2000NPC refining report. This part of the report pro-vides support for the findings and recommenda-tions presented in the Executive Summary. Itincludes discussions of the items outlined in therefining issues analysis part of the StudyApproach section above.

• Part II, Observations on U.S. PetroleumInventories, contains the results of the InventorySubcommittee’s review and update of the 1998NPC inventory report. This part of the report pro-vides support for the findings and recommenda-tions presented in the Executive Summary. Itincludes discussions of the items outlined in theinventory issues analysis part of the StudyApproach section above.

• Appendices contain the Secretary’s study requestletter, a description of the NPC, rosters of thestudy’s Committee and Subcommittees, and sup-porting materials for the report.

PREFACE 3

This supplemental report focuses on majorlight petroleum product (gasoline, jet fuel,heating oil, and diesel) supply and petroleum

inventories in the United States. It is important tounderstand this market in the context of the largerflexible and responsive global petroleum market.Delivery of petroleum products to the consumerinvolves multiple separate activities, includingrefining, transportation, and storage of petroleum.Many different competitors participate in this sup-ply chain. Some are integrated throughout thechain while others specialize only in certain seg-ments. Competition in the global marketplacedrives adoption of efficient strategies, includingthose related to inventory management and refiningoptimization.

In the near term, the NPC does not foresee signif-icant hurdles to the availability of gasoline and heat-ing oil supplies to meet consumer demand.However, there are concerns about meeting UltraLow Sulfur Diesel (ULSD) demand during the tran-sition to the 15 ppm maximum sulfur specificationbeginning in mid-2006.

The demand for light products in the UnitedStates is expected to continue to grow. Demand willbe met with a combination of domestic productionand imports. Imports have been growing for thelast several years, and imports are expected to con-tinue to be an economic component of U.S. supply.The amount of future demand growth supplied fromdomestic refineries will be dependent on several fac-tors, including domestic investment decisions byindividual companies. This report includes recom-mendations for government actions that wouldavoid impeding domestic refining capacity growth

and improve the investment climate for domesticrefining.

Companies continually strive to economicallyoptimize their operations while reliably meetingconsumer demand. Inventories are an integral partof the supply system and act as an interface betweenthe various segments of the supply chain. The com-petitive nature of the industry drives companies tominimize working capital, of which inventory is acomponent, while ensuring reliable supply systems.Ultimately, consumers benefit from efforts toreduce petroleum supply costs.

The petroleum markets respond to supply/demand changes with price movements that providethe incentive to increase or decrease supply to cor-rect any imbalance. This is an integral part of nor-mal and effective market operation. Through theindividual responses of various companies to theseprice movements, the petroleum industry as awhole reacts quickly and effectively to maintain thesupply and demand balance in response to changesin local, regional, or global market conditions.Although the U.S. supply system is very complex, itis robust and has the flexibility to adjust to supplydisturbances. Even major supply disturbances aretypically rebalanced within a short period of time.Market mechanisms provide the fastest and mostefficient response to supply disturbances.

Refining Capacity and Import Availability FindingsCapacity Growth

U.S. refining capacity has grown since the 2000NPC refining study, but the rate of U.S. refining

EXECUTIVE SUMMARY OF FINDINGS AND RECOMMENDATIONS 5

EXECUTIVE SUMMARY OF FINDINGS AND RECOMMENDATIONS

capacity growth has slowed in recent years.Domestic light product production has not keptpace with demand growth. Some of the factors thatimpede domestic capacity growth include:

• Investment economics, reflecting historical refin-ing and marketing returns that are lower thanother segments of the oil business and lower thanthe average of the S&P 500 companies

• An uncertain regulatory environment, resultingfrom issues such as New Source Review (NSR)enforcement, National Ambient Air QualityStandards (NAAQS) implementation schedules,and lack of a defined mechanism for the grantingof waivers

• Resources used for regulatory-driven refineryprojects.

The NPC expects that increases in domestic pro-duction will come from continued expansion ofexisting refineries, because expansion at existingsites is generally more economic than new refineryconstruction. Even the very recent improvement inrefinery profitability does not appear to be sufficientto create an economic environment conducive tobuilding new grassroots refineries. Recent increasesin gasoline imports suggest that the economic andregulatory climate for investment in domestic refin-ing capacity has not supported capacity expansionand utilization equal to the growth in gasolinedemand.

Product imports are expected to continue to be aneconomic component of U.S. supply. The volume ofimports in the future will depend upon a number offactors, including domestic demand growth anddomestic refining capacity growth, as well as sup-ply/demand factors outside the United States thataffect the economics of imports versus domesticrefining.

Emissions reductions and an improved environ-ment benefit society in many ways. As they have inthe past, U.S. refining industry participants expectto continue to devote significant resources to envi-ronmental improvement, including cleaner fuelsproduction and reduction of emissions from sta-tionary sources. However, the magnitude anduncertainty of environmental requirements andtheir enforcement increases cost and adverselyaffects domestic refinery investment. Foreign com-

petitors that are not subject to these additionalcosts and uncertainties may have a competitiveadvantage, resulting in reduced domestic capacitygrowth.

Ultra Low Sulfur Diesel

The NPC believes that the transition period forULSD is likely to be more difficult and longer thanhistorically associated with major product specifi-cation changes. This is due to the difficulty antic-ipated in maintaining and assuring the specifiedsulfur level and needed volumes during distribu-tion from refineries to the ultimate consumer.Enforcement of the 15 ppm maximum sulfurretail cap without an adequate tolerance for testreproducibility could result in large quantities ofdiesel being disqualified as ULSD for supply toconsumers. In addition, pipeline companies couldset a very low sulfur requirement at the refinerygate because of contamination concerns in thedistribution system, which would reduce refineryproduction capability. It is uncertain whetherdomestic refinery production will be of sufficientvolume and low enough sulfur content to over-come anticipated distribution issues. The NPCdoes not expect that imports of ULSD will be wide-ly available to make up for the downgrade duringdistribution or reduced refinery production.Consequently, there is the potential for significantsupply disruptions.

Inventory FindingsInventory Trends

Crude oil inventory has continued the slow down-ward trend noted in the 1998 study. This trend,which is likely to continue, is attributed to deliverysystem efficiency improvements and decliningdomestic crude oil production. With decliningdomestic production, imports have increased.However, imports in transit are not counted in U.S.inventory data.

The long, slow decline of gasoline inventory pri-marily associated with finished gasoline inventoryat terminals that was observed in the 1998 invento-ry study is no longer apparent. Distillate (heatingoil and diesel) inventory has remained essentiallyflat through both the previous and current studyperiods. The NPC believes further efficiencies have


taken place allowing for lower inventory levels, butthese efficiencies have been offset by increases inproduct demand and the number of different fuelspecifications, which have increased the need forinventories.

Lower Operational Inventory

As discussed in the 1998 NPC inventory study,U.S. petroleum inventories respond to both marketand infrastructure changes in the supply system. Asignificant part of petroleum inventory is requiredto operate the product and crude supply systemsand is not readily available to meet demand. In the1998 study, the NPC redefined these inventories as“lower operational inventory” (LOI). The NPCdefined LOI as the lower end of the demonstratedoperating inventory range updated for known anddefinable changes in the petroleum delivery system.The LOI was introduced in the 1998 study in orderto move away from the concept that there is somedefinable inventory level where supply system relia-bility becomes of greater concern. While generallynot used by industry, the NPC recognizes LOI as agauge to help the government assess current inven-tory levels.

Based on the observed crude oil inventory trends,the NPC concludes that the crude oil LOI should bea range of 260 to 270 million barrels, compared tothe 1998 study conclusion of 270 million barrels.Since the 1998 study, crude oil inventory has beenas low as 260 million barrels with no impact oncrude oil supply to U.S. refineries. However, inSeptember 2004, Hurricane Ivan had a significantimpact on offshore oil platforms, pipeline move-ments, and oil imports. This created localized sup-ply disruptions at a few refineries even thoughcrude oil inventories were slightly above 270 mil-lion barrels. This reinforces the concept that theLOI is only one indicator of adequacy of supply andtherefore a crude oil LOI range is recommended,rather than a single value, to better represent thedegree of accuracy associated with the LOI method-ology.

No change is appropriate at this time in the LOIsfor gasoline or distillate fuel oil. Given the shorttime frame of this study, the potential impact of reg-ulatory changes in diesel sulfur content on distillateinventory was not studied.

Price Volatility

Global crude oil prices continue to be the primarydriver of product price levels. Even though thenumber and magnitude of product price up-ticks(increases of greater than 10% or more versus prioryear period) has increased since 1997, most of theseevents were driven by events in the global crude oilmarket. Retail price changes continue to lag behindspot price changes, which has the effect of dampen-ing and delaying price swings at the retail level.

Product and crude oil price levels and volatilityhave increased since the previous study, whichfocused on a time period of relative calm in oil mar-kets (1992-1997). In the 1998 to 2004 time period,crude oil price volatility peaked in 1998, while gaso-line price volatility peaked in early 2002. What hasoccurred since 2002 is an upward movement inproduct prices in line with an upward movement inglobal crude oil prices. Retail gasoline prices, how-ever, have been observed to be less volatile thancrude oil prices.

These conclusions are based on analysis of U.S.national data. Consumers at a local level may besubject to more or less volatility than the nationalaverage as a result of local factors that are not cap-tured by this analysis.

Relationship of Inventories to Price

As addressed in the 1998 NPC inventory study, theexpectation that inventories influence prices isbased on the economic assumption that pricesreflect the current supply/demand balance and thatinventories provide a measure – however imperfect– of the changing balance between supply anddemand. Inventories are a result of supply anddemand fundamentals. Any factor that serves as ameasure of the short-term supply/demand balancewould be expected to influence prices.

Statistical analysis of the simple relationshipbetween inventories and prices finds only a modestcorrelation. This conclusion is indicative of the factthat the interaction of inventories and prices iscomplex. Inventories are an imperfect measure ofthe supply/demand balance, and prices for crude oiland petroleum products are influenced by manyother factors in addition to inventories. Whenpetroleum inventory data are made public, it can


potentially have a short-lived effect on petroleumprices, including futures prices. This appears par-ticularly true when the inventory data deviate frommarket expectations.

Strategic Petroleum Reserve

The United States’ Strategic Petroleum Reserve isthe largest government crude oil stockpile in theworld. The NPC remains strongly supportive ofholding these inventories, but they should only beused during significant crude oil supply disruptionsthat threaten the system’s ability to meet domesticdemand.

While it is recognized that there is a strategicheating oil reserve in the Northeast, the concept ofa products strategic reserve was discussed as part ofthis supplemental report and the NPC does notbelieve such reserves are appropriate for the UnitedStates.

Recommendations

The NPC provides the following recommenda-tions to help ensure a reliable supply of light petro-leum products to the U.S. consumer. These recom-mendations are aimed at avoiding hindrance ofrefining capacity expansion, improving the envi-ronment for investment in domestic refining andlogistics capability, and allowing the current supplysystem to continue to operate efficiently. Allowingthe market to work efficiently will benefit the cus-tomer as the market provides the fastest and mostefficient response to supply disturbances.

The recommendations of the 2000 NPC refiningstudy remain applicable and should be implement-ed. A summary of those recommendations and thecurrent status are included in Appendix C.

New Source Review

Immediate implementation of comprehensiveNSR reform is a very important policy step neededto improve the climate for investment in domesticrefinery expansion. The NSR reforms promulgatedby the Administration, including the EquipmentReplacement Rule currently under judicial review,should be implemented as soon as possible.Attempts to delay or overturn the reforms should bevigorously opposed. Additional NSR reform propos-

als regarding de-bottlenecking and project aggrega-tion should be issued and finalized.

National Ambient Air Quality Standards

The U.S. Environmental Protection Agency(EPA) should revise the NAAQS compliance dead-lines and procedures to take full advantage ofemissions reduction benefits from current regula-tory programs such as cleaner fuels/engines andreduction of regional emissions transport. As cur-rently structured, attainment deadlines precedethe benefits that will be achieved from emissionsreductions already planned. Thus, even thoughprograms are already being implemented to pro-vide emissions benefits, states with non-attain-ment areas will be required to pursue additionalcostly controls that might otherwise not be neededand might not be deliverable in the time framecurrently required.

The current deadlines could result in:

• Requirements for additional emissions offsets forany refinery modifications, reducing the econom-ic attractiveness of investment in refinery capaci-ty expansion

• Additional investment in stationary source con-trols at refineries, reducing the overall profitabil-ity and viability of domestic refining versusimports

• Additional requirements for boutique fuels,reducing the efficiency of the distribution systemand increasing the potential for supply disrup-tions.

These requirements would be disincentives toexpansion of domestic refining capacity. If thestates were given sufficient time to allow emissionsbenefits of clean fuels/engine programs and region-al transport regulations to be considered in attain-ment demonstrations, the adverse impact of theseregulations on domestic refining capacity would begreatly reduced.

Implementation of Ultra Low Sulfur DieselRegulations

Although the timing and specification level of theULSD regulations do not follow the NPC’s 2000 rec-ommendations, the timing requirement has beenfinalized and should not be changed this close to the


implementation date, since refiners are alreadymaking investments to comply.

To reduce the potential for supply disruptions,EPA should work with the Department of Energy(DOE) and the various fuel supply industries to con-sider emerging information about the behavior ofULSD moving through the entire distribution sys-tem and to consider how to achieve the goals of theprogram while recognizing distribution systemrealities.

EPA’s current testing tolerance for ULSD shouldbe adjusted to reflect the reproducibility of the teststhat will be available for regulatory compliance;otherwise, enforcement actions based on testinginaccuracy may result in disruption to the supplysystem.

National Energy Legislation

The NPC recommends passage of national energylegislation as embodied in the 108th Congress con-ference report on HR.6 as the vehicle with the high-est probability of obtaining prompt action on thereformulated gasoline (RFG) oxygenate, oxygenateliability, and boutique fuel issues. While clearly acompromise, the package will help remove some ofthe uncertainty around the future of the domesticrefining industry.

• Oxygenate Liability. Congress should enact lim-ited liability protections against defective productclaims involving methyl tertiary butyl ether(MTBE) and other federally required additives.This action would eliminate only defective prod-uct claims that penalize fuel manufacturers formeeting the Clean Air Act requirements.Negligence and other traditional causes of actionfor MTBE cleanup would be unaffected.

• Boutique Fuels. Requests for specialty fuels for-mulations, whether driven by NAAQS or other-wise, should be approved only where such pro-grams are necessary and cost-effective relative toother emissions reduction options. Proliferationof boutique fuels has fragmented the market,increasing the potential for supply disruptionsand price volatility. While the industry has beenable to adapt to the current slate of boutique fuelswithout significant supply disruptions, continuedproliferation would substantially increase the riskof supply disruption and price volatility. Imple-

mentation of state and local fuels programs,including any actions on MTBE, should be coor-dinated to avoid hindering operation of the distri-bution system and should provide sufficient leadtime to implement any necessary refining anddistribution changes. Repeal of the 2% oxygena-tion requirement for RFG could eliminate muchof the incentive for boutique fuel proliferation.

DOE and EPA should conduct a joint study of theboutique fuel issue, with participation by thestakeholders. This study should provide impor-tant information on the impact of boutique fuelson fuel production and distribution.

Sound Science, Cost Effectiveness,and Energy Analysis

The 2000 NPC refining report recommended that:“Regulations should be based on sound science andthorough analysis of cost effectiveness.”

Executive Order 13211, signed by President Bushin 2001, requires agencies to prepare a “Statementof Energy Effects” including impacts on energy sup-ply, distribution, and use, when undertaking regula-tory actions. The NPC recommends that ExecutiveOrder 13211 be made law and strictly enforced.

The NPC is not suggesting elimination or roll-back of environmental requirements, but ratherthat the cost analysis of proposed regulationsshould include a thorough analysis of energy supplyeffects from production to end-use. Examples ofregulations that the NPC does not believe reflect athorough analysis of the energy supply effectsinclude ULSD and NAAQS regulations. As a result,implementation of these regulations may imposeunintended costs without commensurate benefit.

Regulatory cost/benefit analysis should be per-formed on an incremental basis, to ensure that thecost of each required increment is justified. Using atotal and average analysis may result in adoptingemissions reduction increments that are not costeffective.

Permitting

Streamlining the permitting process would helpimprove the environment for domestic refiningcapacity investment. Some activities are current-ly underway to review processes and identify


streamlining opportunities; these activitiesshould include industry and other stakeholders.Streamlining should provide for expeditious over-all review and a clearly defined process for obtain-ing a permit, with agency roles and responsibili-ties well-defined and specific deadlines for makingpermit decisions.

Depreciation Schedule Adjustment

Adjusting the depreciation schedule for all refin-ing equipment to five years from the current tenyears, consistent with the treatment of similarprocess equipment in other manufacturing indus-tries, would have a positive impact on expansioninvestment economics. This action would reducethe capital recovery period for investment in refin-ing equipment, helping to offset the historically lowreturns in the refining/marketing business thathave hindered investment in capacity expansion.

The depreciation adjustment should be applied toall new domestic refining investment. Attempts toapply revised treatment to some small sub-segmentof investment may have the perverse effect of reduc-ing the incentive for more significant additions inbase capacity. The depreciation schedules for petro-leum pipelines and storage facilities should be sim-ilarly reduced.

Fuel Waivers and Enforcement Discretion

Use of exemptions, exceptions, and waivers shouldbe limited to serious supply disruptions that affectend users’ ability to obtain petroleum products.States have sought and EPA has repeatedly consid-ered and/or granted enforcement discretion, and thishas increased uncertainty in the marketplace. EPAshould issue a definitive variance procedure forallowing non-complying fuel to be sold in the mar-ketplace. Proposed guidance on waivers has beenrecently released by EPA as a first step in this process.

Alternative Fuels

Mandates or subsidies for alternative fuelsincrease uncertainty and reduce the incentive forinvestment in additional domestic petroleum refin-ing capacity. Therefore, these mandates and subsi-dies may not reduce petroleum product imports asintended and could increase the cost to consumers.

Distillation and Driveability Index

The 2000 NPC refining report recommended thatthe Driveability Index not be changed without thor-ough additional analysis. To date, EPA has resistedautomakers’ calls for a reduction in DriveabilityIndex, or a change to Distillation Index (DriveabilityIndex plus an ethanol adjustment). EPA shouldcontinue this position. A reduction in Driveabilityor a change to Distillation Index could result in asignificant reduction in domestic refinery gasolineproducibility.

Site Security

Site security enhancement should remain anindustry responsibility with ongoing risk assess-ment coordinated with the Department ofHomeland Security, which should retain the leadfederal coordination role. Refining industry participants are committed to keeping their facilities secure from threats of violence and ter-rorism. Refiners have expended substantialresources to enhance security and expect to con-tinue to do so. There are proposals being dis-cussed that include provisions for refining tech-nology changes and criminal liability. In theopinion of the NPC, these provisions do not pro-vide an additional security benefit but have thepotential to negatively impact light product pro-duction capability.

2005 Domestic Capacity

These recommendations are aimed at avoidinghindrance of capacity expansion, improving theenvironment for investment in domestic refiningand logistics capability, and allowing the currentsupply system to continue to operate efficiently.These recommendations should be implemented assoon as practical.

Major refinery modifications can take four ormore years lead time for all the activities necessaryfor implementation. Due to this long lead time, thecapacity that will be available in 2005 is the result ofregulatory actions and investment decisions overthe last several years. The NPC has not identifiedany government actions that could significantlyincrease the domestic refining capacity available for 2005.


The downstream oil business is very competi-tive in the United States, and both refinersand marketers continuously look for the low-

est cost means of supplying their customers. Thedemand for light products in the United States isexpected to continue to grow. Demand will be metwith a combination of domestic production andimports. For the last several years, imports havebeen growing. Likewise, domestic refining capacityhas been growing and is expected to continue togrow. Product imports are expected to continue tobe an economic component of domestic supply,because the United States is part of a flexible andresponsive worldwide petroleum market. Theamount of future demand growth supplied fromdomestic refineries will be dependent on domesticinvestment decisions by individual companies, aswell as the economics of domestic production ver-sus imports.1

One of the factors considered in evaluatingpotential distillation or conversion capacity expan-sion projects is the economic return under expect-ed future market conditions. Since imported prod-uct has been and is expected to be an ongoingcomponent of domestic supply, the relative eco-nomics of domestic production versus importsaffects investment decisions by individual compa-nies for domestic refining. Recent increases ingasoline imports suggest that the economic and

regulatory climate for investment in domesticcapacity has been such that individual domesticrefiners have been unable to justify sufficientamounts of distillation and conversion expansionsto satisfy all the growth in light product demand.This report includes recommendations for govern-ment actions that would avoid impeding domesticrefining capacity growth and improve the invest-ment climate.

This section discusses U.S. light product petrole-um supply (gasoline, jet fuel, and distillates – bothdiesel and heating oil) in six segments:

• Observations on supply and demand since the2000 NPC refining study, including domesticcapacity and import trends

• Factors that can potentially impact the growth indomestic refining capacity

• A generic assessment of the economics of con-structing a new grassroots refinery

• The outlook for the near term, covering observa-tions on low sulfur gasoline and ultra low sulfurdiesel implementation

• Recommendations that could reduce hindrancesand improve the climate for investment in domes-tic capacity

• Brief comments on potential issues for the longerterm.

Most of the data presented in this section includeannual data through 2003. While partial year datafor 2004 are available in some cases, the effort nec-essary to adequately annualize the information forvalid comparison was beyond the scope of thisstudy.

PART I – REFINING CAPACITY AND PRODUCT IMPORT AVAILABILITY I-1

PART II

REFINING CAPACITY AND PRODUCT IMPORT AVAILABILITY

1 The NPC did not develop a quantitative forecast for imports in thelimited time frame of this study. The Energy InformationAdministration’s 2004 Annual Energy Outlook forecasts that netpetroleum product imports into the United States will increase byabout 500 thousand barrels per day from 2003 to 2010. Thisincrease represents about 20% of the 2.5 million barrels per dayincrease in domestic gasoline, jet fuel, and distillate demand fore-cast over that same period.

Observations on Light Product Supply and Demand Since the 2000 NPC Refining Study

U.S. light product supply is driven by demand andis provided from both domestic refinery productionand imports. Since the 2000 NPC refining study,domestic supply and demand for light products hasbeen slowly increasing. Figure I-1 shows domesticproduction of gasoline, jet fuel, and distillates aswell as the total net imports of all these products.2A slowly growing overall trend is apparent fordomestic production. While imports remain a smallportion of domestic supply overall, the percentagegrowth in imports has been significant over the lastfew years. As discussed later in this section, the netimport increase is primarily a result of increasedgasoline imports from Europe.

Domestic Light Product Production

Increases in domestic light product productionresult from a combination of changes in both distil-

lation and conversion capacity, the yields of prod-ucts from crude oil, and refinery capacity utiliza-tion. The increase in domestic production of lightproducts since the 2000 NPC refining study shownin Figure I-1 has resulted from capacity expansionsand improvements in yield, which more than offseta slight decline in average utilization. Trends incapacity, yield, and capacity utilization are discussedin more detail in the following sections.

Refinery Distillation and Conversion Capacity Changes

Domestic atmospheric crude oil distillationcapacity has been steadily increasing since 1996, asshown in Figure I-2. However, the rate of capacityaddition has slowed since the 2000 NPC refiningstudy. Factors that have affected the rate of capaci-ty increase are discussed later in this report.

U.S. refinery capability is commonly assessedusing atmospheric distillation capacity. This is asomewhat simplistic approach since, depending onthe characteristics of the specific crude oil used,only 30 to 50% of an atmospheric distillation facili-ty’s output can be made directly into light products.U.S. demand is mostly light products and thereforethe heavier materials produced from atmospheric

I-2 OBSERVATIONS ON PETROLEUM PRODUCT SUPPLY

0

2

4

6

8

10

12

14

16

1996 1997 1998 1999 2000 2001 2002 2003

MIL

LIO

N B

AR

RE

LS P

ER

DA

Y

YEAR

Figure I-1. U.S. Light Product Supply

GASOLINE

DISTILLATES

NET IMPORTS

JET FUEL

Source: Energy Information Administration.

Figure I-1. U.S. Light Product Supply

2 Unless otherwise indicated, data are from U.S. Energy InformationAdministration or International Energy Agency Reports.

distillation must be processed further in a variety ofconversion units to produce additional light prod-ucts for the market.

Conversion units manufacture light petroleumproducts from heavier boiling material present incrude oil. Conversion capacity increases are illus-trated in Figure I-3 as a ratio to 1996 capacity. Fluidcatalytic cracking (FCC) capacity, which mainly addsto gasoline yield, has grown at about the same per-cent as distillation capacity. Both hydrocracking andcoking have grown more than distillation capacityon a percentage basis, with coking capacity growingthe most. Hydrocracking provides the most flexibil-ity to vary the product slate but is the highest costconversion capacity. Coking capacity allows refinersto reduce the yield of heavy fuel and/or process heav-ier crude oils. However, the coking process produceslow quality products that require a significantamount of further processing before their quality issuitable for use as light petroleum product.3

Light Product Yield Changes

The relative amount of light products producedper barrel of atmospheric distillation input isreferred to as yield and has been slowly increasingover time as shown in Figure I-4. This increaseindicates that the industry’s overall capability toconvert crude oil to light products has more thankept pace with distillation capacity increases. Thisresults because expansions of conversion facilitieshave been faster than the increased production ofheavier materials from atmospheric distillationcapacity additions.

Overall yields of distillate and lighter productswere about 86% at the time of the last study. Sincethen they have risen to an average of 87% over thelast three years. Data for 2002 and 2003 both appearinconsistent with the general trend. Given the timeavailable for this study, a detailed analysis of therecent data could not be performed.

While overall yield has been improving, there aresome factors that can offset potential yield improve-ment. Individual refiners have been and continueto process lower quality (heavier and higher sulfur)


Figure I-2. U.S. Atmospheric Distillation Capacity

16

17

18

1996 1997 1998 1999 2000 2001 2002 2003

MIL

LIO

N B

AR

RE

LS P

ER

ST

RE

AM

DA

Y

Note: Data for each year are those reported for January 1 of the following year.Source: Energy Information Administration.

1997 capacity was not reported.Value estimated by averaging 1996 and 1998 data.

YEAR

Figure I-2. U.S. Atmospheric Distillation Capacity

3 For a detailed discussion of refinery processes, see Appendix C in theNPC’s 2000 refining study, U.S. Petroleum Refining—Assuring theAdequacy and Affordability of Cleaner Fuels.


Figure I-3. U.S. Refinery Capacity Relative to 1996 (Barrels per Stream Day)

0.90

0.95

1.00

1.05

1.10

1.15

1.20

1.25

1996 1997 1998 1999 2000 2001 2002 2003

CA

PA

CIT

Y R

AT

IO, 1

996

= 1

DISTILLATION

CATALYTIC CRACKING

HYDROCRACKING

RESID DESTRUCTION

Note: Data for each year are those reported for January 1 of the following year.


1997 capacity was not reported.Value estimated by averaging 1996 and 1998 data.

YEAR

Figure I-3. U.S. Refinery Capacity Relative to 1996 (Barrels per Stream Day)

Figure I-4. U.S. Refinery Light Product Yield

85

86

87

88

89

1996 1997 1998 1999 2000 2001 2002 2003


YEAR

PE

RC

EN

T

Figure I-4. U.S. Refinery Light Product Yield

crude oil in response to economic drivers. Whenrefiners add heavy oil processing capacity such ascokers and reduce the production of residual fuel,light product production will increase per barrel ofcrude oil input. However, if heavy oil processing isadded primarily to process lower quality crude oiland results in no net residual fuel destruction, lightproduct production per barrel of crude oil input willnot increase. In the past, refiners have been addingheavy oil processing both to reduce heavy fuel oilproduction and to increase their ability to run heav-ier crude oils. Current U.S. domestic refinery fueloil yield is less than 5%, therefore there is limitedopportunity to further increase yield by convertingheavy fuel oil.

Utilization rates and seasonal factors also affectyield. When refineries are run at very high through-put, such as during the high-demand summermonths, on average they can suffer some light prod-uct yield decline. In the summer, the yield of gaso-line goes down as shown in Figure I-5 even thoughtotal gasoline production may be higher. This resultsprimarily because some lighter refinery streams can-not be blended into the summertime finished gaso-line pool due to vapor pressure restrictions.

Switching from methyl tertiary butyl ether(MTBE) to ethanol as an oxygenate in gasolineincreases emissions of volatile organic compounds(VOCs) unless the hydrocarbon portion of the gaso-line blend is modified. During summer gasolineproduction, refiners must remove approximately5% of other light gasoline components to compen-sate for the Reid Vapor Pressure (RVP) increasefrom ethanol blending as opposed to MTBE. Someof the materials removed from gasoline during thesummer are stored and used during the winter andsome are used in other ways and lost to the gasolinepool. In either case, they are not usable to producegasoline in the summer.

Capacity Utilization Changes

At the time of the 2000 NPC refining study, U.S.refineries had operated at record high atmosphericdistillation capacity utilization for the previous twoyears, averaging between 88 and 89% on an annualbasis and rising to above 91% on average during thesummer months (Figure I-6). Distillation capacityutilization has declined somewhat since that peak tomore historical levels. The record low refining andmarketing financial return in 2002 was likely a factor


Figure I-5. U.S. Refinery Monthly Light Product Yield

78

80

82

84

86

88

90

92

1996 1997 1998 1999 2000 2001 2002 2003


PERC

ENT

YEAR2004

Figure I-5. U.S. Refinery Monthly Light Product Yield

behind reduced utilization that year. Utilizationincreased in 2003 and domestic gasoline productionwas at record levels for the first six months of 2004.Domestic crude oil runs for May and June of 2004 setnew monthly records.

Due to the limited time available, a quantitativeanalysis of the reasons for capacity utilizationchanges was not performed as part of this study.Capacity utilization can vary for a number of rea-sons:

• Refining facilities periodically need to be shutdown for planned and unplanned maintenance.When a crude oil distillation unit is shut down, nocrude oil is processed. When other refinery facil-ities are shut down for maintenance, distillationutilization may also have to be reduced to containthe unfinished products from distillation.

• Regulatory requirements for industry-wideinvestment in a narrow time frame, such as forlow sulfur gasoline, can tend to group refineryshutdowns for maintenance and modificationtogether and result in lower capacity utilizationfor that period than the historical trend.

• Economics of incremental operation of domesticrefining compared with the cost of importedproduct supplies will affect utilization. If incre-mental crude oil runs do not recover their cost inthe market, domestic refinery capacity will not befully utilized.

• Limitations on capacity use can be imposed byenvironmental permit requirements.

Even though utilization in the years since the2000 NPC refinery study was lower than the recordyears in 1997 and 1998, light product productionhas increased as the result of capacity additions andthe higher yields achieved.

Recent Import Trends

Imports are not new to the U.S. supply system, asthe U.S. has been importing products routinelysince World War II. The U.S. is a net importer(imports minus exports) of all three light products– gasoline, distillates, and jet fuel. Imports haveprovided an economic balance between domesticrefinery production and U.S. consumer demand andhave increased over the last decade. The U.S. is part


Figure I-6. U.S. Atmospheric Distillation Stream Day Capacity Factors

83

84

85

86

87

88

89

90

91

92

1996 1997 1998 1999 2000 2001 2002 2003

ANNUAL

SUMMER


YEAR

PE

RC

EN

T

1997 capacity was not reported.Capacity factor based on estimated capacity data.

Figure I-6. U.S. Atmospheric Distillation Stream Day Utilization

of a global petroleum products market. If it is moreeconomical to produce a product outside the U.S.and import it into the U.S., lower prices result thanif the product were produced in less economic oper-ation of domestic facilities.

The economic flow of oil products into a countrycan be impeded by government policy. Commonlyused programs designed for this purpose have takenthe form of restrictive import tariffs or import quotasystems. During the 1950s and 1960s, the UnitedStates employed a quota system to limit the flow ofinternationally produced products into the U.S.Currently the U.S. does not have a program in placethat restricts the flow of oil products to U.S. mar-kets. Consequently, the U.S. is part of the interna-tional oil market, and oil product imports that sup-ply U.S. markets are provided on an economic basis,resulting in lower cost products to U.S. consumers.

Jet Fuel Domestic Production and Imports Respond to Demand Changes

Figure I-7 shows domestic production, demandand, by difference, net imports of jet fuel. Importedjet fuel increased slowly between 1995 and 2000.The events surrounding September 11, 2001 result-

ed in a sharp reduction in U.S. jet fuel demand anddecreases in both U.S. production and imports.These jet fuel data illustrate the economic selectionof supply sources into the marketplace. Domesticproduction of jet fuel in 2000 essentially equaledU.S. demand for jet fuel in 2002 and exceeded 2003demand. Both domestic production and importsdeclined in response to lower domestic demand in2003, but some imports remained even thoughdomestic refining had demonstrated capacity toproduce the entire domestic demand volume. Thisresponse indicates that the U.S. supply systemrebalanced in an economic fashion.

Increased Distillate Imports Tied to Canadian Refinery Expansion

Figure I-8 shows that between 1996 and 2000, netimports of distillate increased from relatively smallvolumes to more than 100 thousand barrels per day.Since 2001, distillate imports have remained atabout 200 thousand barrels per day, varying season-ally with the demand for heating oil.

Distillate net imports by quality are shown inFigure I-9. Lower sulfur distillate imports (500ppm sulfur maximum) have been in the range of


1,000

1,100

1,200

1,300

1,400

1,500

1,600

1,700

1,800

1995 1996 1997 1998 1999 2000 2001 2002 2003

DE

MA

ND

AN

D P

RO

DU

CT

ION

(Tho

usan

d B

arre

ls P

er D

ay)

0

50

100

150

200

250

300

350

400

NE

T IM

PO

RT

S(T

hous

and

Bar

rels

Per

Day

)


YEAR

Figure I-7. U.S. Jet Fuel Production, Demand, and Net Imports

NET IMPORTS

DOMESTIC REFINERY PRODUCTION

DEMAND

Figure I-7. U.S. Jet Fuel Production, Demand, and Net Imports


Figure I-8. U.S. Distillate Production, Demand, and Net Imports

2,600

2,800

3,000

3,200

3,400

3,600

3,800

4,000

1995 1996 1997 1998 1999 2000 2001 2002 20030

100

200

300

400

500

600

700


YEAR

DE

MA

ND

AN

D P

RO

DU

CT

ION

(Tho

usan

d B

arre

ls P

er D

ay)

NE

T IM

PO

RT

S(T

hous

and

Bar

rels

Per

Day

)

NET IMPORTS

DOMESTIC REFINERY PRODUCTION

DEMAND

Figure I-8. U.S. Distillate Production, Demand, and Net Imports

Figure I-9. U.S. Net Imports of Low Sulfur Diesel and High Sulfur Distillate

-40

-20

0

20

40

60

80

100

120

140

160

1995 1996 1997 1998 1999 2000 2001 2002 2003

<500 ppm SULFUR >= 500 ppm SULFUR


YEAR

TH

OU

SA

ND

BA

RR

ELS

PE

R D

AY

Figure I-9. U.S. Net Imports of Low Sulfur Diesel and High Sulfur Distillate

80 to 100 thousand barrels per day since 1998. Ingeneral, distillates in this sulfur range appear inthe marketplace as highway diesel fuel. Imports ofhigher sulfur distillates, which are normally soldas heating oil or non-road diesel, have increasedsince 2000. This increase has been primarily fromCanada into PADD I. This observation is consis-tent with a major expansion to a MaritimeProvinces Canadian refinery and appears to resultfrom a logistics advantage of this supply source toNortheast markets.

Gasoline Imports from Europe Increasing

Net gasoline imports including blendstocks havesteadily increased, reaching 730 thousand barrelsper day in 2003, as shown in Figure I-10. Gasolineimports, the largest product import volume into theU.S., have grown steadily. In 2002 and 2003, domes-tic production of gasoline leveled off at about 8.2 million barrels per day while gasoline demandcontinued to grow.

The majority of the U.S. gasoline imports, about87% in 2003, are received into PADD I (Figure I-11).Overall they account for about 25% of the gasolinesupply into this area (Figure I-12).

Between 1999 and 2003, net gasoline importsgrew from about 6% to about 8% of domestic prod-uct demand. As shown in Figure I-13, the majorityof gasoline imports are from the Atlantic Basin,with Canada, the Virgin Islands, and WesternEurope accounting for about two-thirds of the vol-ume. The remaining imports come primarily fromLatin America and Eastern Europe, with some gaso-line being imported to the West Coast from the AsiaPacific area. Since 2000, imports from Venezuelahave declined. Venezuela reported a major mainte-nance period in 2002 and a strike in 2003. Importsfrom other Latin American countries such as Braziland Argentina increased. There was also a signifi-cant increase in gasoline imports from Canadabeginning in 2000 as a result of the EasternCanadian refinery expansion mentioned earlier.

A major growing supply source for gasolineimports into the United States is Western Europe.The Western European highway fuels market hasbeen impacted by the dieselization of the region’svehicle fleet resulting from preferential taxationfavoring diesel-fueled vehicles. Highway dieseldemand has been steadily displacing gasolinedemand in Western Europe, as shown in Figure I-14.


Figure I-10. U.S. Gasoline Production, Demand, and Net Imports

6,000

6,500

7,000

7,500

8,000

8,500

9,000

1995 1996 1997 1998 1999 2000 2001 2002 20030

200

400

600

800

1,000

1,200

1,400

1,600


YEAR

DE

MA

ND

AN

D P

RO

DU

CT

ION

(Tho

usan

d B

arre

ls P

er D

ay)

NE

T IM

PO

RT

S(T

hous

and

Bar

rels

Per

Day

)DOMESTIC REFINERY

PRODUCTION

DEMAND

GASOLINE (INCL. BLENDSTOCK)NET IMPORTS

Figure I-10. U.S. Gasoline Production, Demand, and Net Imports


Figure I-11. U.S. Gasoline Import DistributionYear Average 2003

PADD I

OTHERPADDs


Figure I-11. U.S. Gasoline Import Distribution–Year Average 2003

Figure I-12. PADD I Supply

SUPPLY FROMOTHER PADDs

49%

LOCALPRODUCTION

27%

NET IMPORTS24%


Figure I-12. PADD I Gasoline Supply

Figure I-13. U.S. Gasoline Imports by Country

0

100

200

300

400

500

600

700

800

900

1,000

1995 1996 1997 1998 1999 2000 2001 2002 2003

Source: Energy Information Administration.YEAR

TH

OU

SA

ND

BA

RR

ELS

PE

R D

AY

OTHER

EASTERN EUROPE

WESTERN EUROPE

CANADA

OTHER LATIN AMERICA

VENEZUELA

VIRGIN ISLANDS

Figure I-13. U.S. Gasoline Imports by Country

The diesel demand increase is being offset to someextent by a decrease in heating oil (gasoil) demandresulting from fuel switching, as indicated in FigureI-15.

Although diesel demand has been increasing,gasoline values in western Europe are still abovediesel. Hence, there have been few gasoline-to-distillate refinery conversion projects in Europeanrefineries. As gasoline demand in local marketsdeclines, Western European refiners have the poten-tial to produce more gasoline than needed locally.

Figure I-16 indicates that market behavior in theMediterranean is consistent with the demand trendsin Western Europe. Between 1996 and 1999,Mediterranean gasoline value versus distillateincreased as specifications for gasoline were tight-ened significantly (lower vapor pressure, sulfur,benzene, and aromatics). Following 1999, the rela-tive value of gasoline to diesel has declined asexpected from the increase in diesel demand.

Contrary to Europe, in the United States bothgasoline and distillate consumption have been

increasing. Since 2000, the result of these differingmarket conditions has been a steady increase in therelative price of gasoline in New York Harbor versusRotterdam, as indicated in Figure I-17.

As shown in Figure I-18, the Rotterdam gasoline-distillate price differentials behaved similarly tothe Mediterranean market until 2001. Since then,the differential has been increasing. This suggeststhat gasoline in Rotterdam is now being valuedinto New York Harbor as opposed to being valuedinto the local market. Historically, theMediterranean market has been less impacted byconditions in U.S. markets than the Rotterdammarket as a result of the difference in transporta-tion costs to the U.S. East Coast from the twoareas. This price trend in Rotterdam is consistentwith the increase in Western European gasolineexports to the U.S.

The recent increase in gasoline imports fromEurope reflects the changing economics of AtlanticBasin supply as refiners seek to move material tothe highest value markets. Given the expecteddemand trends in Europe, European refiners are


Figure I-14. Western European Gasoline and Diesel Demand

1,800

2,000

2,200

2,400

2,600

2,800

3,000

3,200

3,400

1995 1996 1997 1998 1999 2000 2001 2002 2003

Source: International Energy Agency.

GASOLINE

DIESEL

TH

OU

SA

ND

BA

RR

ELS

PE

R D

AY

YEAR

Figure I-14. Western European Gasoline and Diesel Demand


Figure I-15. Western European Distillate Demand

0

500

1,000

1,500

2,000

2,500

3,000

3,500

1995 1996 1997 1998 1999 2000 2001 2002 2003

DIESEL

OTHER GASOIL


TH

OU

SA

ND

BA

RR

ELS

PE

R D

AY

YEAR

Figure I-15. Western European Distillate Demand

Figure I-16. Mediterranean Gasoline to Gasoil Spreads

-0.50

0

0.50

1.00

1.50

2.00

2.50

3.00

3.50

4.00

1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003


YEAR

NO

MIN

AL

DO

LLA

RS

PE

R B

AR

RE

L

Figure I-16. Mediterranean Gasoline to Gasoil Spreads


-0.50

0.00

0.50

1.00

1.50

2.00

1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003


NO

MIN

AL

DO

LLA

RS

PE

R B

AR

RE

L

YEAR

Figure I-17. Gasoline Differential – New York Harbor to Rotterdam

Figure I-17. Gasoline Differential – New York Harbor versus Rotterdam

Figure I-18. Rotterdam Gasoline to Gasoil Spread

-2.00

-1.50

-1.00

-0.50

0.00

0.50

1.00

1.50

2.00

1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003


YEAR

NO

MIN

AL

DO

LLA

RS

PE

R B

AR

RE

L

Figure I-18. Rotterdam Gasoline to Gasoil Spreads

likely to be able to produce additional quantities ofgasoline surplus to the local markets for severalyears. The U.S. East Coast is a logical outlet forthese suppliers. The ultimate quantity that moveswill be dependent on the relative cost of theimports compared with incremental domestic pro-duction, as well as the gasoline value in otherpotential markets.

New York and Connecticut MTBE Bans Illustrate the Market at Work

As of January 2004, the states of New York andConnecticut banned MTBE. MTBE is blended atrefineries to meet the government-mandated oxy-gen content requirements of reformulated motorgasoline (RFG). The primary alternative to MTBEfor meeting these requirements is ethanol. UnlikeMTBE, ethanol must be blended in terminalsbecause its affinity for water makes it impractical tomove ethanol-blended gasoline in the petroleumproduct pipeline network. The hydrocarbon mix-ture that can be blended with ethanol to produceRFG is called “reformulated blendstock for oxy-genate blending,” or RBOB.

Venezuela has been a historical supplier of RFGto the East Coast. Initially in 2004, they reported-ly experienced difficulty in producing RBOB, andtheir supplies to the U.S. declined. However, othersupply sources such as Western Europe respondedand there was no supply disruption observable tothe consumer. Recently, imports from Venezuelahave begun to increase, suggesting that theirissues limiting RBOB production have beenresolved and Venezuela is able to compete in themarket. This observation indicates that AtlanticBasin suppliers of gasoline have the flexibility torespond to changing U.S. market requirements. Itis also illustrative of market interactions and sup-ply adjustments during periods of fuel specificationtransitions.

However, it is important to recognize that theMTBE bans in New York and Connecticut created aboutique fuel in that area, necessitating a new seg-regation in the distribution system and new sourcesof supply. While industry adapted and the new fuelhas been delivered without disruption of supply, thepotential for supply disruption and resulting pricevolatility has increased.

Factors That Potentially Impact the Growth in Domestic ProductProducibility

Refining Industry Financial Performance

The petroleum industry is very competitive andthe domestic downstream (refining and marketing)business has on average had financial results belowboth other segments of the oil business and otherindustries. The U.S. Energy Information Adminis-tration (EIA) collects financial and operating datafrom a sample of U.S. energy companies representa-tive of the U.S. energy industry as a whole. The mostrecent report titled Performance Profiles of MajorEnergy Producers 2002 was published in February2004 and covered data through 2002. In 2002, theEIA’s Financial Reporting System (FRS) data samplecovered companies that accounted for over 80% ofthe refinery product output of all U.S. refining facil-ities. The data sample ranged over the entire spec-trum of downstream companies. The FRS data aregenerally accepted as representative of results thatwould be expected from both the U.S. total anddownstream petroleum industry.

Figure I-19 shows FRS petroleum industry returnon equity compared with the average return onequity of the S&P 500 companies through 2002, themost recent data available from EIA.4 While thedata vary considerably from year to year, over theentire 1981-2002 period, the average for the FRScompanies was slightly below the S&P 500 compa-nies’ average, at 11.3% and 12.2% respectively.Contrary to many perceptions, returns in the U.S.petroleum business are, on average, lower thanthose of other U.S. industries.

A historical perspective on U.S. downstream petro-leum profitability can be seen on Figure I-20.Returns in this segment of the business are lowerthan the other segments of the oil business. Totalpetroleum industry return on capital employed aver-aged 7.7% across the entire period, with return oncapital employed in the refining and marketing seg-ment averaging only 5.3%. The same trend is evi-dent in the more recent period (since 1990) asshown on the graph. The low returns in the refining


4 More recent data on refinery gasoline margins are presented inFigure I-25 later in this section.


0

5

10

15

20

25

1981 1983 1985 1987 1989 1991 1993 1995 1997 1999 2001

YEAR

PE

RC

EN

T


Figure I-19. Return on Equity for U.S. FRS Companies and the Average of S&P 500 Companies

TOTAL PETROLEUM

AVERAGE OF S&P 500 COMPANIES

FRS COMPANIES

Figure I-19. Return on Equity for U.S. FRS Companies and the Average of S&P 500 Companies

-5

0

5

10

15

20

1977 1979 1981 1983 1985 1987 1989 1991 1993 1995 1997 1999 2001

Percent

U.S. Petroleum

U.S. Refining & Marketing


Refining &Marketing5.0%

AVERAGE1990–2003

Total Petroleum7.1%

YEAR

Figure I-20. U.S. Petroleum Industry Return on Capital Employed

-5

0

5

10

15

20

1981 1983 1985 1987 1989 1991 1993 1995 1997 1999 2001

PE

RC

EN

T

TOTAL PETROLEUM

U.S. REFINING & MARKETING


Refining &Marketing5.0%

AVERAGE1990–2002

Total Petroleum7.1%

YEAR



and marketing segment result from the highly com-petitive nature of the business as well as the signifi-cant amount of regulatory driven investments thattend to capture little to no return in the market.5

For FRS companies, operating costs have beengenerally declining since 1981, as seen in Figure I-21. Prior to 1981, the petroleum industry operat-ed under federal price control regulation. Since theprice regulation era, U.S. downstream operatingcosts have declined in response to competitive fac-tors. Gross margins (product selling price minuscrude oil price) closely followed the operating costpattern over the time period. Net margins (grossmargin minus operating cost) have exhibited a flattrend.

These data illustrate that the cost efficiencies(lower operating costs) captured as a result of compe-tition within the industry have essentially all been

passed to the market, as evidenced by the flat netmargin trend. The significant savings realized overthe past 20 years have been reflected in lower pricesin the market rather than increased returns in theindustry.

Decisions by individual refiners to invest indomestic refining capacity are complex and dependon expectations of return on investment. Refineryassets are long-lived as evidenced by the fact that alloperating U.S. refineries are several decades old.For investment to occur, an individual refiner mustreasonably expect investment in capacity to provideat least the same rate of return as other investmentoptions over the life of the asset. While each com-pany has its own perspective for future market con-ditions and will make its own independent decisionson investment and operation, the long-term historyof below-average rates of returns remains a factor tobe considered.

Capital and Other Resource Constraints

As they have done in the past, U.S. refining indus-try participants expect to continue to devote signif-icant resources to environmental improvement,


0

2

4

6

8

10

12

14

16

18

1981 1983 1985 1987 1989 1991 1993 1995 1997 1999 2001

Source: Energy Information Administration.YEAR

2002

DO

LLA

RS

PE

R B

AR

RE

L

Figure I-21. U.S. Refining and Marketing Margins (Product Sold Basis)

OPERATING COST

GROSS MARGIN

NET MARGIN

Figure I-21. U.S. Refining and Marketing Margins (Product Sold Basis)

5 In its estimates of low sulfur gasoline and ultra low sulfur dieselcosts, EPA estimated a 7% before-tax capital amortization forinvestment costs, which provides an after-tax discounted cash flowreturn of 1-4% depending on the useful life assumption – beloweven the below-average return for petroleum refining and market-ing assets per EIA’s FRS (see 66 FR 5093 and 69 FR 39107).

including cleaner fuels production and reduction ofstationary emissions. Figure I-22 shows U.S. refin-ing and marketing investments in total and for envi-ronmental compliance through 2002 in constant2004 dollars. Data for total expenditures are report-ed by the Oil & Gas Journal, and environmentaldata by API.6 Since 1999, overall expenditures havebeen slowly growing, reaching over $8 billion in2003, the highest level since 1995. Expenditures inthe “base,” however, have been fairly flat over thistime period, with environmental investmentsaccounting for most of the increase.

Figure 1-22 indicates that the refining and mar-keting industry has continued to invest in the basebusiness in spite of the relatively low overall returndata discussed previously. However, the figure alsoshows that investment in the base business can bereduced when expenditures for environmental facil-ities increase significantly. This is particularly evi-dent during the early 1990s and in 2002 when significant expenditures for gasoline qualityimprovements occurred. This observation is likely

a result of a combination of practical constraints onthe ability to maintain base business investmentswhen faced with a significant increase in invest-ments to meet environmental standards. Thedomestic refining industry is comprised of manydiverse companies and resource constraints arelikely to vary widely among individual companies.

Capital needs in refining compete with capitalneeds in other petroleum segments as well as capi-tal needs of other industries. Uncertainty aroundthe current and future climate for the industry cannot only impact internal capital allocation deci-sions, but can also affect external evaluations suchas bond ratings. These external evaluations can sig-nificantly affect an individual company’s access tocapital. While no two industry participants are like-ly to access capital funding in exactly the same way,no participant has unlimited access to resources.7Thus, it should be expected that spending on


Figure I-22. U.S. Refining and Marketing Investments

0

2,000

4,000

6,000

8,000

10,000

12,000

14,000

1985 1987 1989 1991 1993 1995 1997 1999 2001

BASE

Sources: Oil & Gas Journal and American Petroleum Institute.YEAR

MIL

LIO

N 2

004

DO

LLA

RS

ENVIRONMENTAL

0

2,000

4,000

6,000

8,000

10,000

12,000

14,000

1985 1987 1989 1991 1993 1995 1997 1999 2001 2003

ENVIRONMENTAL

BASE

YEAR

MIL

LIO

N 2

004

DO

LLA

RS

Figure I-22. U.S. Refining and Marketing Investments

6 American Petroleum Institute, U.S. Oil and Natural Gas Industry’sEnvironmental Expenditures, 1993-2002, January 28, 2004.

7 For additional discussion of the range of considerations for capitalacquisition and allocation among companies, see the reportRefining Capacity – Challenges and Opportunities Facing the U.S.Industry by Kummins, Yaccobucci, and Parker of the CRS, August2004, pages 6-9 in particular. The NPC does not endorse the entirecontents of this report.

mandatory environmental projects can detract frominvestments in the base business that provide capac-ity growth, yield flexibility and reliability improve-ments.

Large capital projects are typically designed andengineered by a combination of internal companyresources and third-party companies. There are anumber of these engineering companies, ranging insize from very small to those capable of handlingprojects of several billion dollars. This segment ofthe project business has in the past demonstratedthe capability to expand with workload and has notbeen a constraint to domestic refining capital proj-ect implementation. However, before these entitiesare engaged, a development effort is required todetermine what facilities should be built. Whilethere are many different approaches to this phase ofa capital project, the evaluation of the multitude ofoptions with respect to costs, risks, impact on refin-ery operations, etc., is a very complex task. Thisphase of a project usually requires a significantamount of effort from highly skilled internal com-pany individuals with detailed knowledge of the spe-cific facility. These resources are not easily lever-aged with additional resources when projectworkload is high. Thus, when environmentalinvestment is high, the ability to develop projectsthat might lead to throughput or yield improve-ments can be limited by the availability of theseinternal company human resources.

Ever More Restrictive Product QualityRequirements

Each refinery is unique, with a collection of facil-ities that have evolved over the years in response toboth market and regulatory changes. The operationof these facilities is routinely optimized in responseto the market. As product specifications becomemore restrictive, the ability to utilize all the flexibil-ity and volume capacity inherent in the refining sys-tem can be diminished. For example, regulationssuch as the gasoline vapor pressure restrictions inthe summer mentioned previously reduce the abili-ty of the industry to maximize volumetric output inresponse to market demands. While individuallymany of these impacts are small, collectively theycan be significant, particularly during periods ofhigh utilization. Although additional capital may bespent to recapture this lost capacity, it can be diffi-

cult to generate a return on an investment that onlyprovides short-term or seasonal capacity since thetime frame for capital recovery is limited.

Increasing Project Costs and Uncertainty from Environmental Regulations

Compliance with increasingly restrictive environ-mental regulations has been a way of life in therefining business for decades, and emissions reduc-tions and an improved environment benefit societyin many ways.8 However, the magnitude of environ-mental requirements and the uncertainty aroundtheir enforcement can have a significant impact onrefinery investment. If these costs and uncertain-ties are not faced to the same extent by foreign com-petitors, the domestic industry can become lesscompetitive and capacity growth could be reduced.

A multitude of issues around environmental com-pliance, such as the history of reinterpretation ofpast environmental permit rulings led by EPA’s NSRenforcement initiative, environmental justice litiga-tion, the uncertainty of NSR reform, and NAAQsimplementation schedules with questionable feasi-bility, has resulted in a very uncertain climate forthe implementation of new refinery projects. A sig-nificant factor affecting investment decisions is thelevel of uncertainty about the future. While thisuncertainty cannot be eliminated, it can be affectedby government action or, in some cases, inaction.

For decisions requiring large amounts of capital,increased uncertainty tends to stop, minimize, ordelay investment in order to reduce risk. The long-lived nature and high capital cost of refining assetsresult in a long payout period. Awaiting resolutionof uncertainty by delaying investment and even tak-ing a short-term economic loss may be a moreattractive alternative than early investment inequipment that may prove not to be optimum forthe long term. Uncertainty is increased when regu-latory requirements are not based on sound scienceor thorough analysis of cost effectiveness. Suchregulations face a higher likelihood of later chal-lenge and change, increasing the uncertainty of theoutcome and therefore increasing the risks associat-ed with investment decisions. The recent confusion


8 Refer to Appendix D for a summary of recent and pending regula-tory challenges facing the domestic refining industry.

about the status of RFG requirements for the BatonRouge and Atlanta areas highlights the increaseduncertainty of environmental requirements withquestionable benefits.

Waivers Can Have a Negative Impact on the Investment for Clean Fuels

Another uncertainty in the market is the poten-tial for waivers or enforcement relaxation of fuelspecifications in response to political pressure dur-ing price events. As the risk of waivers increases,the potential incentive to invest to produce cleanfuels will be reduced. States have sought and EPAhas repeatedly considered and sometimes grantedenforcement discretion, and this has increased mar-ket uncertainty. Use of exemptions, exceptions, andwaivers should be limited to serious supply disrup-tion situations that affect end-users’ ability toobtain petroleum products, and the circumstancesof their potential use should be clearly establishedin advance.

Environmental Operating Restrictions

Refineries today are subject to increasingly morestringent permit and public relations constraints.For example, it is no longer acceptable to flare forany significant length of time. New requirementscontinue to be imposed such as continuous emis-sions monitoring and the NOx ppm cap for FCCregenerators. In some areas, State ImplementationPlans are more restrictive than federal require-ments, further limiting operational flexibility.While industry participants will comply with thesenew requirements, as operations become more con-strained both the effective capacity and the flexibil-ity of existing facilities to respond to unexpectedevents can be reduced.

Facility Security Environment Could PotentiallyImpact Operations

Refining industry participants are committed tokeeping their facilities secure from threats of vio-lence or terrorism. Companies have been heavilyengaged since before September 11, 2001 in main-taining and enhancing facility security. Through apublic/private partnership approach, industry isworking with many federal, state, and local agenciesto address security issues. These agencies includethe Department of Homeland Security (including

U.S. Secret Service, Transportation SecurityAdministration, and U.S. Coast Guard), the CIA, theFBI, the Department of Transportation, theDepartment of Energy, and the Department ofDefense. Individual refineries also work closelywith various state and local emergency responseand law enforcement officers to ensure understand-ing and coordination of security plans.

Refiners have expended substantial resources,both human and financial, to enhance physical andcyber security. Expectations are that investments infacility security will continue to be necessary aspotential threats are identified and assessed and assecurity regulations such as the U.S. Coast Guardregulations implementing the Maritime Transpor-tation Security Act are promulgated. In addition,industry expects to continue to work with theDepartment of Homeland Security and the otheragencies listed above in the public/private partner-ship to address security issues in the future. Whileresources have been consumed in implementingthese changes, to-date security issues have had lit-tle, if any, detrimental impact on operations.However, there are proposals being discussed thatinclude provisions for refining technology changesand criminal liability. In the opinion of the NPC,these proposals do not provide additional securitybenefits but could significantly reduce refinery pro-duction.

Grassroots Construction vs. Refinery Expansions

As discussed previously, domestic refinery capaci-ty has continued to grow over time. However, asshown on Figure I-23, the number of refineries hascontinued to decline. No refinery capacity has beenbuilt on a new site (grassroots) in the U.S. since themid-1970s.

The absence of grassroots domestic refinery con-struction indicates the relative attractiveness ofexpanding or acquiring existing refineries com-pared with building a new refinery. Refinery con-struction at a new site would require very long plan-ning, site selection, permitting, financing, andconstruction lead times. Expansion, acquisition,and increased utilization of existing refineries aretypically more feasible, quicker, and less costlyoptions to increase light product production,


although the lead times for modification of existingrefineries can also be substantial.

The economics of constructing a grassrootsrefinery would be dependent on numerous factorsincluding the location, size, crude oil source, andthe product slate produced. For illustrative pur-poses, the following analysis provides a high levelgeneric comparison derived from public data ofthe cost of grassroots construction compared withthe cost of capacity additions through acquisi-tions.

Information published in the Oil & Gas Journal(March 19, 2001) by the consulting firm Turner,Mason & Company estimates the grassroots con-struction costs for a hypothetical large-scale, high-conversion refinery, designed to manufacture lowsulfur fuels for the U.S. market. The refinery wouldbe situated on the U.S. Gulf Coast, a location thathas generally favorable construction costs relativeto other domestic markets. The estimated con-struction cost for such a refinery is between $2.5 and $3.0 billion. Considering the additionalhurdles beyond construction costs that a U.S. grass-

roots refinery project would face, such as obtainingsite and environmental permits, the estimateappears to be somewhat low.

Equivalent Distillation Capacity (EDC) is a con-cept that is widely applied in the refining industry toperform comparisons between refineries of differentscale and sophistication. Based on analysis of theTurner Mason estimate, the cost of the hypotheticalGulf Coast refinery would be within the range of$1,050/EDC to $1,275/EDC in 2004 dollars.

Refinery acquisition values can provide a measureof the value of domestic refining assets from boththe buyer’s and seller’s perspective. The followingdata summarize refinery acquisition prices from alldomestic refinery sales since 1998 that satisfiedthree conditions:

• Transaction price is public record

• Refining assets were primarily in the fuels busi-ness as opposed to asphalt or lubricants

• The primary assets included in the transactionwere individual refineries, as opposed to company-


Figure I-23. U.S. Refining Capacity and Number of Refineries

0

5

10

15

20

1950 1955 1960 1965 1970 1975 1980 1985 1990 1995 20000

50

100

150

200

250

300

350

400

Note: Data for each year are those reported for January 1 of the following year.Source: Energy Information Administration.

YEAR

NU

MB

ER

OF

RE

FIN

ER

IES

MIL

LIO

N B

AR

RE

LS P

ER

DA

Y

OPERATING REFINERIESOPERABLE CAPACITY

Figure I-23. U.S. Refining Capacity and Number of Refineries

wide transactions or portfolios comprised of bothrefining and non-refining assets.

There were 25 transactions that met these condi-tions. In 2004 dollars on an EDC weighted basis,the average sales price was $320/EDC. A compari-son of grassroots construction costs with refineryacquisition price is shown in Figure I-24.

This analysis suggests that current industry par-ticipants have valued U.S. refining assets at aboutone-fourth to one-third the cost of grassroots con-struction. This would appear to explain why somecompanies seeking to add domestic capacity havepursued acquisition rather than grassroots con-struction. The number of transactions in recenthistory indicates that a viable market for refiningassets does exist.

As shown in Figure I-25, refinery gasoline mar-gins have varied considerably over time. Marginsfor 2003 and 2004 to-date average about 25% abovethe average for the years 1998-2003. Assuming thatthe refinery sales history presented above has gen-erally reflected margin expectations based on thelonger-term averages, even the recent increase in

margins would not appear to be sufficient to justifynew grassroots refinery construction if translatedinto long-term future expectations.

The continuing expansion of existing refineriesin this environment indicates that there have beenrefinery expansion opportunities available at lowercost than grassroots refinery construction.Capacity expansion projects can have more favor-able economics than grassroots projects, becausethey can often use the existing refinery site, utili-ty, and support units and can generally be imple-mented in a shorter time frame. Often, expansionprojects can also be implemented with limitedadditional fixed operating costs in contrast withgrassroots projects that bear a full fixed cost over-head.

The analysis above is a very generalized assess-ment. In reality, actual decisions about investingin a grassroots refinery, acquiring assets throughacquisition, or expanding capacity at an existingsite would be based on an individual company’sassessment of the market and economics includ-ing the particulars of the investment being con-sidered.


0

200

400

600

800

1,000

1,200

1,400

GRASSROOTS REFINERY CONSTRUCTION REFINERY ACQUISITION

2004

DO

LLA

RS

PE

R E

DC

BA

RR

EL

MINIMUM MAXIMUM AVERAGE

Note: EDC = Equivalent Distillation Capacity.Source: Derived from public data as described in text.

Figure I-24. U.S. Grassroots Refinery Construction Costs vs. Refinery Acquisition Price

Figure I-24. U.S. Grassroots Refinery Construction Costs vs. Refinery Acquisition Price

Near-Term Outlook

No Serious Supply Issues Anticipated with Low Sulfur Gasoline Implementation

The National Petroleum Council does not expectserious supply issues to arise associated with imple-mentation of low sulfur gasoline through 2006.Individual companies have made independent deci-sions regarding investment in lower sulfur gasolineproduction, including timing, location, capacity,technology, and design assumptions for invest-ments. No independent public source is available toprovide aggregate data on these gasoline desulfur-ization investments, nor has EPA tracked theprogress of refiners in achieving compliance withthe new rules as it has done with the ULSD imple-mentation. Therefore, the NPC must rely on gener-al industry knowledge and limited public data toassess the likelihood that the demand for low sulfurgasoline will be met through implementation in2006. Based on the information available, the NPCbelieves that current capacity for gasoline desulfur-ization along with announced additions of gasolinedesulfurization capacity and the expected availabili-

ty of imports will provide for continued domesticsupply in the near term.

The NPC did not develop a demand forecast aspart of this update activity. The NPC did review themost recent EIA demand forecast as part of thisactivity. The EIA forecast has domestic gasolinedemand increasing about 2% per year in the nearterm.

Recent performance in meeting the early phase-in of the low sulfur gasoline rule can be used as aguide to expected future performance. In the 2000NPC refining study, the average gasoline contentwas estimated at 340 ppm in the U.S. Beginning inJanuary 2004, the cap on gasoline sulfur contentwas lowered to 300 ppm, with a corporate average of120 ppm. Industry participants produced gasolinemeeting these more stringent requirements atrecord total volumes for the first six months of2004.

Further evidence of the ability to meet gasolinesulfur regulations in the near term is provided byreviewing imports and the availability of gasoline


Figure I-25. U.S. Refinery Gasoline Margins

0

5

10

15

20

25

30

35

40

45

50

1995 1996 1997 1998 1999 2000 2001 2002 2003 2004

1998-2003 AVERAGE MARGIN2003-2004 AVERAGE MARGIN

YEAR

CE

NT

S P

ER

GA

LLO

N (

2004

$)

Note: Gasoline Margin = Difference in U.S. refiner average price realizations for gasoline minus actual average crude oil acquisition price.


Figure I-25. U.S. Refinery Gasoline Margins

sulfur allotments, which allow for the production orimportation of gasoline with higher sulfur than thecorporate average but less than the 300 ppm cap.Allotments are generated by reducing the averagesulfur content of gasoline below the prescribed cor-porate average of 120 ppm. In early 2004, EPA con-templated a relaxation of the average sulfur require-ments for importers to allow additional importsafter a small number of importers claimed theycould not meet the average 120 ppm sulfur standardfor 2004 because sulfur allotments were not avail-able. Following a review, EPA determined that allot-ments were available to allow compliance and deter-mined that such a relaxation was not warranted.Through the first six months of 2004, imports ofgasoline were at record levels, and gasoline sulfurallotments remained available for purchase. Whilethis does not provide definitive information on theflow of allotments or the absolute sulfur levels ofdomestic production or imports, it does indicatethat current domestic production in combinationwith imports is capable of achieving market demandvolumes while meeting specifications, and thatsome capacity exists to reduce sulfur below 2004levels.

Gasoline imports are expected to continue to bean economic supply source for the U.S. in the near-term future. Primary sources of gasoline imports tothe U.S. should be able to meet U.S. Tier 2 gasolinespecifications. Canada and Western Europe bothhave mandated reductions in sulfur similar to thosebeing mandated in the U.S. in approximately thesame time frame. The Virgin Islands depends on theU.S. for the placement of its production andVenezuela has publicly stated it is investing in itsrefineries in order to continue to be a supplier ofgasoline to the U.S. market. This being said, theU.S. may well lose some of its historical sources ofsupply including Eastern Europe and parts of LatinAmerica. These supply sources will likely not beable to meet U.S. specifications without significantinvestment in the near term, and may choose not toinvest to comply with U.S. specifications if othermarkets with growing demand are available as aneconomic outlet for their oil products.

The European Union has implemented regula-tions that are aimed at a reduction in emissions ofglobal warming gases. Processing changes that maybe required in European refineries to enable com-

pliance with the global warming gas emissions reg-ulations may cause an increase in the cost of pro-ducing oil products in European refineries relativeto today. Depending upon market forces, this mayresult in upward pressure on the cost of importedgasoline from Europe to the United States.

The NPC is optimistic that gasoline supplies willbe sustained through the implementation of the lowsulfur gasoline rules in 2006. Some localized sup-ply disruptions may occur due to operational diffi-culties, but these are not expected to be of signifi-cant duration.9 There are a number of pendingregulatory and legislative issues discussed in theRecommendations section of this report that ifresolved would remove some uncertainty from themarket and improve the climate for domestic refin-ery investment, further assuring this optimism.

Boutique Fuels Fragment Markets and Reduce Supply Efficiency

As boutique fuels proliferate, the marketplacebecomes even more fragmented, increasing thepotential for supply disruptions and price volatility.This occurs because boutique fuel requirementshinder the supply flexibility inherent in the complexpetroleum product distribution system. Highlybalkanized fuel requirements restrict potentialalternative supply sources that can be used to bal-ance demand. For example, when New York andConnecticut implemented MTBE bans effectiveJanuary 1, 2004, they could no longer be suppliedwith MTBE-based RFG used throughout the rest ofthe Northeast. This required additional segrega-tions and realignment of supplies. While industryresponded successfully, preventing disruption ofsupplies to consumers, the system overall was andcontinues to be exposed to a higher probability ofbreakdown.

The NPC is not prepared to recommend changesin existing boutique fuel requirements; however,requests for additional specialty formulations, forexample those under consideration in Detroit,should be approved only where necessary and costeffective relative to other emissions reductionoptions. The uncertainty generated by situations


9 Refer to the 2000 NPC refining study for a detailed discussion of theoperational issues that may be associated with implementation ofmore restrictive product specifications.

such as RFG implementation issues in Baton Rougeand Atlanta further complicate supply planning andhinder efficient delivery of fuels to the market.

In addition to being a driver for boutique fuels,the proliferation of MTBE bans results in a net lossof domestic gasoline producibility. With the RFGoxygenate requirement in place, removing MTBErequires substitution of ethanol. Base gasolinevolatility must be reduced to accommodateethanol’s higher blending volatility. This reductionis accomplished by removing light material fromgasoline. As recommended in the 2000 NPC refin-ing report, adequate lead time should be provided toallow all the activities necessary to enable refineryand distribution system modifications to respond tosignificant product quality changes such as MTBEbans.

Significant Supply Concerns Exist for Ultra LowSulfur Highway Diesel Implementation

The NPC has significant concerns that the cur-rent Ultra Low Sulfur Highway Diesel (ULSD) regu-lations have the potential to result in serious andprolonged supply problems. The current regula-tions require earlier implementation and lower sul-fur levels than those evaluated in the 2000 NPCrefining study, raising additional supply concernsbeyond those posed in that study. EPA regulationsrequire ULSD in mid-2006, whereas the 2000 NPCrefining study recommended implementation ofULSD in mid-2007 to avoid overlap with the Tier 2sulfur gasoline investments. The ULSD rule alsorequires a 15 ppm sulfur cap for ULSD, rather thanthe 30 ppm sulfur refinery average evaluated in the2000 NPC refining study. Further, EPA is requiringsulfur enforcement at retail, versus the refinery-gate enforcement strategy that was recommendedin the 2000 NPC refining study and has been usedsuccessfully for the RFG and Tier 2 gasoline sulfurprograms.

The EPA Highway Diesel Fuel 2003 Pre-Compliance Reports survey suggested that refin-ers’ plans should provide sufficient productioncapacity of diesel at or under 15 ppm sulfur tomeet demand, provided the 80/20 phase-in optionworks and downgrade of ULSD in the distributionsystem is minimal. The results of the 2004 EPApre-compliance reports were released just at the

time this report was drafted, and they indicated avery slight reduction in the anticipated productionof 15 ppm maximum sulfur highway diesel versusthe 2003 report. In the 2004 report, EPA summa-rizes: “Hence, it appears that the refining industryas a whole is adequately planning for projectedhighway diesel demand through 2010.”10

The NPC has serious concerns about the feasibil-ity of ULSD movement through the distribution sys-tem of pipelines, vessels, terminals, and trucks thatwill still move substantial volumes of higher sulfurjet fuel, non-road diesel, and heating oil.Significant losses and downgrades during deliveryfrom refineries to the final consumer could result inapparently adequate supplies at refineries beingunable to meet demand at retail. Trial pipeline ship-ments of ULSD suggest the potential for significantvolume loss of on-specification ULSD, potentially inexcess of 10% in each transportation segment, dur-ing distribution even allowing for a significant sul-fur increase. Quantitative data were just becomingavailable at the time of this update.

The ULSD regulation provides for sale of limitedvolumes of ULSD with sulfur greater than 15 ppm atretail (but less than 500 ppm), provided it is segre-gated separately from sales of under-15 ppm sulfurULSD. EPA is relying on this provision (“80/20phase-in”) to conclude that refineries will produceadequate ULSD to meet demand in 2006. However,there are no public evidence or announcements ofdistribution system preparation for segregation ofan additional diesel product required to utilize the80/20 phase-in option and to sell volumes of ULSDwith sulfur slightly above 15 ppm. The combinationof a short four-year window for the 80/20 phase-inand the small 20% volume offers little or no oppor-tunity to payout any new facilities to segregate anadditional diesel product. Without advance prepa-ration in the distribution and retail system, down-graded ULSD cannot be sold to highway diesel cus-tomers. Further challenging ULSD distribution isthe fact that EPA’s testing tolerance is inconsistentwith the sulfur test precision. EPA regulationsallow only a 2 ppm tolerance for testing variabilityat retail whereas the best current sulfur test meth-ods exhibit reproducibility in the 4 to 8 ppm range.


10 U.S. Environmental Protection Agency, Summary and Analysis ofthe Highway Diesel Fuel 2004 Pre-Compliance Reports.

Enforcement of the retail cap without an adequatetolerance for test reproducibility could result inlarge quantities of diesel being disqualified as ULSDfor supply to consumers. Alternatively, distributioncompanies could use the current sulfur test statisti-cal reproducibility and the potential for downstreamcontamination to establish a refinery gate sulfurspecification below 10 ppm. This is particularlytrue for long and complex distribution movements,such as those from the Gulf Coast to the East Coastand Midwest. Some pipelines are reportedly consid-ering very stringent refinery gate requirements;numbers as low as 5 ppm sulfur have been men-tioned. Design should have been completed andconstruction should be underway for most refineryprojects to produce ULSD to meet the 2006 imple-mentation date. While no public information existson design targets for desulfurization projects, theNPC has concerns that actual release specificationsmay be significantly more restrictive than refineryproject design. If this is the case and desulfuriza-tion projects are forced to operate at lower sulfurlevels than design, then actual ULSD productioncould be substantially below design rates.

The NPC believes that it is unlikely that importswill be available to balance any significant domesticproduction and delivery shortfall that could occurwhen ULSD is implemented. As discussed earlier,the U.S. has not historically imported significantvolumes of low sulfur highway diesel (less than 100thousand barrels per day). While Canada and theVirgin Islands will likely maintain their historicalvolumes of highway diesel imports to the U.S. forthe same reasons driving gasoline imports, it isunlikely that there will be any other reliable sourceof ULSD imports to the U.S. As mentioned previ-ously, Western Europe diesel demand continues toincrease and acquiring diesel from Europe could beproblematic. In fact, Europe might compete withdomestic markets for ULSD, though higherEuropean cetane requirements could be a barrier.In addition, ultra low sulfur specifications are notscheduled to be fully implemented in WesternEurope until 2009.

Although the timing and specification level of theULSD regulations do not follow the NPC’s 2000refining study recommendations, the timingrequirement has been finalized and should not bechanged this close to the implementation date,

because refiners are already making investments tocomply.

However, EPA’s current testing tolerance forULSD should be adjusted to reflect the reproducibil-ity of the tests that will be available for regulatorycompliance. Otherwise, some on-specificationbatches of ULSD may be ruled non-compliant sole-ly as a result of testing inaccuracy. This has thepotential to significantly reduce ULSD supplies, par-ticularly in the transition period.

EPA should work with DOE and the various fuelsupply industries to consider emerging informationabout the behavior of ULSD moving through theentire distribution system and to consider how toachieve the goals of the program while recognizingdistribution system behavior. A phased-in distribu-tion tolerance could ease the transition supply con-cern.

In addition to the concerns with the implementa-tion of ULSD, other potential diesel supply issuesexist. Reducing sulfur to very low levels reducesdiesel lubricity, and ASTM has adopted a diesellubricity specification for highway diesel as ofJanuary 1, 2005. However, recently concerns havearisen with lubricity additive trailback into subse-quent products in pipelines, particularly jet fuel, ifthe additive is injected at the refinery. Variouspipelines have proposed conflicting refinery gatelubricity additive requirements, and provision forterminal addition of lubricity additive may be prob-lematic for the January 1 timing. Industry partici-pants are actively working this problem, but if suc-cessful resolution is not promptly achieved,localized supply issues with jet fuel or highwaydiesel may occur in early 2005. Another issue is theavailability of additized 15 ppm sulfur No. 1 diesel toblend into the diesel pool to meet cold flow proper-ties in the winter. This issue is unresolved and war-rants further study.

Recommendations

The NPC provides the following recommenda-tions to help ensure a reliable supply of light petro-leum products to the U.S. consumer. These rec-ommendations are aimed at avoiding hindrance ofcapacity expansion, improving the environmentfor investment in domestic refining and logistics


capability, and allowing the current supply systemto continue to operate efficiently. Allowing themarket to work efficiently will benefit the cus-tomer as the market provides the fastest and mostefficient response to supply disturbances.

The recommendations of the 2000 NPC refiningstudy remain applicable and should be implement-ed. A summary of those recommendations and thecurrent status is included in Appendix C.

New Source Review

Immediate implementation of comprehensiveNSR reform is a very important policy step neededto improve the climate for investment in domesticrefinery expansion. The reforms promulgated bythe Bush Administration, including the EquipmentReplacement Rule currently under judicial review,should be implemented as soon as possible.Attempts to delay or overturn the reforms should bevigorously opposed. Additional reform proposalsregarding de-bottlenecking and project aggregationshould be issued and finalized. Additional back-ground on NSR is included in Appendix E.

National Ambient Air Quality Standards

EPA should revise the NAAQS compliance dead-lines and procedures to take full advantage of emis-sions reduction benefits from current regulatoryprograms such as cleaner fuels/engines and reduc-tion of regional emissions transport. As currentlystructured, attainment deadlines precede the bene-fits that will be achieved from emissions reductionsalready planned. Thus even though programs arealready being implemented to provide emissionsbenefits, states with non-attainment areas will berequired to pursue additional costly controls thatmight otherwise not be needed and might not bedeliverable in the time frame currently required.

The current deadlines could result in:

• Requirements for additional emissions offsets forany refinery modifications, reducing the econom-ic attractiveness of investment in refinery capaci-ty expansion

• Additional investment in stationary source con-trols at refineries, reducing the overall profitabil-ity and viability of domestic refining versusimports

• Additional requirements for boutique fuels,reducing the efficiency of the distribution systemand increasing the potential for supply disrup-tions.

These requirements would be disincentives toexpansion of domestic refining capacity. If thestates were given sufficient time to allow emissionsbenefits of clean fuels/engine programs and region-al transport regulations to be considered in attain-ment demonstrations, the adverse impact of theseregulations on domestic refining capacity would begreatly reduced.

Implementation of Ultra Low Sulfur DieselRegulations

Although the timing and specification level of theULSD regulations do not follow NPC’s 2000 recom-mendations, the timing requirement has been final-ized and should not be changed this close to theimplementation date, because refiners are alreadymaking investments to comply.

To reduce the potential for supply disruptions,EPA should work with DOE and the various fuelsupply industries to consider emerging informationabout the behavior of ULSD moving through theentire distribution system and to consider how toachieve the goals of the program while recognizingdistribution system realities. EPA sponsored aworkshop on this subject in November 2004.

EPA’s current testing tolerance for ULSD shouldbe adjusted to reflect the reproducibility of the teststhat will be available for regulatory compliance;otherwise, enforcement actions based on testinginaccuracy may result in disruption to the supplysystem.

National Energy Legislation

The NPC recommends passage of national energylegislation as embodied in the 108th Congress con-ference report on HR.6 as the vehicle with the high-est probability of obtaining prompt action on theRFG oxygenate, oxygenate liability, and boutiquefuel issues. While clearly a compromise, the pack-age will help remove some of the uncertaintyaround the future of the domestic refining industry.

• Oxygenate Liability. Congress should approvelimited liability protections for defective product


claims involving MTBE and other federallyrequired additives. This action would eliminateonly defective product claims that penalize fuelmanufacturers for meeting the Clean Air Actrequirements. Negligence and other traditionalcauses of action for MTBE cleanup would be unaf-fected.

• Boutique Fuels. Requests for specialty fuels for-mulations, whether driven by NAAQS or other-wise, should be approved only where such pro-grams are necessary and cost-effective relative toother emissions reduction options. Proliferationof boutique fuels has fragmented the market,increasing the potential for supply disruptionsand price volatility. While the industry has beenable to adapt to the current slate of boutique fuelswithout significant supply disruptions, continuedproliferation will substantially increase the risk of supply disruption and price volatility.Implementation of state and local fuels programs,including any actions on MTBE, should be coor-dinated to avoid hindering operation of the distri-bution system, and should provide sufficient leadtime to implement any necessary refining anddistribution changes.

DOE and EPA should conduct a joint study of theboutique fuel issue, with participation by allstakeholders. This study should provide impor-tant information on the impact of boutique fuelson fuel production and distribution.

Sound Science, Cost Effectiveness,and Energy Analysis

The 2000 NPC refining report recommended that:“Regulations should be based on sound science andthorough analysis of cost effectiveness.”

Executive Order 13211, signed by President Bushin 2001, requires agencies to prepare a “Statementof Energy Effects” including impacts on energy sup-ply, distribution, and use, when undertaking regula-tory actions. The NPC recommends that ExecutiveOrder 13211 be made law and strictly enforced.

The NPC is not suggesting elimination or roll-back of environmental requirements, but ratherthat the cost analysis of proposed regulationsshould include a thorough analysis of energy supplyeffects from production to end-use. Examples ofregulations that the NPC does not believe reflect a

thorough analysis of the energy supply effectsinclude off-road ULSD and NAAQS regulations. Asa result, implementation of these regulations mayimpose unintended costs without commensuratebenefit.

Regulatory cost/benefit analysis should also beperformed on an incremental basis, to ensure thateach required increment is cost justified. Using atotal and average analysis may result in adoptingemissions reduction increments that are not costeffective.

Permitting

Streamlining the permitting process would helpimprove the environment for domestic refiningcapacity investment. Some activities are currentlyunderway to review processes and identify stream-lining opportunities; these activities should includeindustry and other stakeholders. Streamliningshould provide for expeditious overall review andhave a clearly defined process for obtaining a per-mit, with agency roles and responsibilities well-defined and specific deadlines for making permitdecisions.

While the permitting process can be time con-suming and burdensome, it is not the only hin-drance to additional domestic refining capacityexpansion. Streamlining the permitting processalone may not result in a significant increase in therate of domestic capacity expansion.

The 2000 NPC refining report identified environ-mental justice challenges to permit requests as apotential issue and recommended that EPA define aclear process for dealing with these challenges.While environmental justice challenges have notbeen widespread since the 2000 report, there havebeen a few instances where environmental justicechallenges resulted in abandoning projects. Thepotential for more widespread challenges remain,and the NPC reaffirms its 2000 recommendationthat EPA define a clear procedure for dealing withenvironmental justice challenges.

Depreciation Schedule Adjustment

Adjusting the depreciation schedule for all refin-ing equipment to five years from the current tenyears, consistent with the treatment of similar


process equipment in other manufacturing indus-tries, would have a positive impact on expansioninvestment economics. This action would reducethe capital recovery period for investment in refin-ing equipment, helping to offset the historically lowreturns in the refining/marketing business thathave hindered investment in capacity expansion.Depending on the specifics of the project, thisdepreciation change should increase the return oninvestment by about one percentage point.

The depreciation adjustment should be applied toall new domestic refining investment. Attempts toapply revised treatment to some small sub-segmentof investment may have the perverse effect of reduc-ing the incentive for more significant additions inbase capacity. The depreciation schedules for petro-leum pipelines and for storage facilities should besimilarly reduced.

Fuel Waivers and Enforcement Discretion

Use of exemptions, exceptions, and waiversshould be limited to serious supply disruption situ-ations that affect end-users’ ability to obtain petro-leum products. States have sought and EPA hasrepeatedly considered and/or granted enforcementdiscretion, and this has increased market uncertain-ty. EPA should issue a definitive variance procedurefor allowing non-complying fuel to be sold in themarketplace. Proposed guidance on waivers hasbeen recently released by EPA as a first step in thisprocess.

Alternative Fuels

Mandates or subsidies for alternative fuelsincrease the uncertainty and reduce the incentivefor investment in additional domestic petroleumrefining capacity. Therefore, these subsidies maynot reduce petroleum product imports as intendedand could increase the cost to consumers.

Distillation and Driveability Index

The 2000 NPC refining report recommended thatthe Driveability Index not be changed without thor-ough additional analysis. To date, EPA has resistedautomakers’ calls for a reduction in DriveabilityIndex, or a change to Distillation Index (DriveabilityIndex plus an ethanol adjustment). EPA shouldcontinue this position. A reduction in Driveability

or a change to Distillation Index could result in asignificant reduction in domestic refinery gasolineproducibility.

Site Security

Site security enhancement should remain anindustry responsibility with ongoing risk assess-ment coordinated with the Department ofHomeland Security, which should retain the leadfederal coordination role. Refining industry partic-ipants are committed to keeping their facilitiessecure from threats of violence and terrorism.Refiners have expended substantial resources toenhance security and expect to continue to do so.There are proposals being discussed that includeprovisions for refining technology changes andcriminal liability. In the opinion of the NPC, theseprovisions do not provide an additional securitybenefit but have the potential to negatively impactlight product production capability.

Reducing Distribution Incidents Would ImproveSupply Reliability

The “one call” system to avoid digging incidentsshould be made national and aggressively promotedand enforced. Damage to underground pipelines bythird-party digging has been a source of distributiondisruption. Broadening the one-call system toinclude government entities and strictly enforcing itsuse will help reduce the potential for these incidents.

Access to Ports

Efficient access to ports is necessary for receipt ofboth imported crude oil and petroleum products.Sufficient funding should be provided to increase ormaintain existing depth in waterways serving keycrude oil and light product import facilities. Thedredging permit process for individual terminalsites should be streamlined and adequate designat-ed disposal sites should be provided for the dredgingmaterials that are removed during the dredgingprocess.

During periods of significant supply issues,increased waterborne movements of petroleum maybe desirable. The current process for obtaining awaiver of the Jones Act should be evaluated andclarified so that definitive rules are in place shouldthis need occur.


2005 Domestic Capacity

These recommendations are aimed at avoidinghindrance of capacity expansion, improving theenvironment for investment in domestic refiningand logistics capability, and allowing the currentsupply system to continue to operate efficiently.These recommendations should be implemented assoon as practical to begin taking effect.

Major refinery modifications can take four ormore years lead time for all the activities necessaryfor implementation. Due to this lead time, thecapacity that will be available in 2005 is the result ofregulatory actions and investment decisions overthe last several years. The NPC has not identifiedany government actions that could significantlyincrease domestic refining capacity available for2005.

Longer-Term (to 2010) Observations

Historically, large petroleum markets have beenprimarily served by refineries with a logisticaladvantage to the market. This occurs as a result ofthe relative cost advantage of transporting crudeoil long distances as opposed to transporting amultitude of products. While worldwide refiningcapacity has historically been available to meetproduct demand, products flow to and from differ-ent demand areas as the market and refining capa-bility changes in response to economic realities.Worldwide product demand growth is expected tobe greatest in the developing countries of the FarEast such as China and India. It is possible thatrefining capacity growth in those areas may notkeep pace with demand and that some of the

sources of current U.S. oil product imports willfind it more attractive to serve markets other thanthe United States. Potentially this could alter U.S.refinery expansion economics such that domesticrefining capability will grow to economicallyreplace imports. Conversely, there might be over-expansion of refining capacity in growth marketsin other parts of the world that could provide eco-nomic oil import supplies for the U.S. market-place. Barring any artificial barriers to world tradein oil products, market economics and local con-struction and operating costs will dictate the eco-nomic location for new refinery capability.

It is still too early to determine how the reductionin non-road diesel sulfur in 2007 and 2010 will pro-ceed. Any problems that do arise related to theimplementation of 15 ppm non-road diesel sulfurwould likely overlap into the highway ULSD supply.Each individual refiner will continue to independ-ently evaluate when to upgrade/invest for addition-al diesel desulfurization capacity to meet the 500ppm non-road diesel rule in 2007 and the totaldiesel sulfur pool at 15 ppm in 2010. Incremental15 ppm diesel production costs are likely to rise asdesulfurization of increasingly more difficultstreams is required.

While there is uncertainty surrounding theimpacts of implementation of the various diesel sul-fur rules, heating oil supplies should not be nega-tively impacted. However, a potential wildcard con-tributing to the uncertainty in the heating oilsupply outlook is whether individual states willimpose lower heating oil sulfur requirements in thistime frame.


Role of Inventory

This supplemental report focuses on major lightpetroleum products and crude oil inventories in theUnited States, but it is important to understand theseinventories and their relationship to the light petro-leum product market in the context of the largerglobal petroleum market. Production and delivery ofpetroleum products involves at least 14 separateactivities, as shown in Figure II-1. This figure illus-trates the various components of the global petrole-um supply chain, and the points at which inventoriesoccur. Many unique competitors participate in thissupply chain. Some are integrated throughout thechain while others only specialize in certain seg-ments. Competition in the global marketplace drivesadoption of the most efficient strategies, includingthose related to inventory management.

Inventory is held at many points in the globalsupply chain and plays several roles. The focal pointof this study is based on data collected from the EIA– major light petroleum products and crude oil inthe primary system in the United States. The pri-mary system is one component of global inventoriesand contains about half of the total U.S. inventory ofmajor light petroleum products. Although move-ment and processing of a specific crude oil can takemonths due to the geography and complexity of thesupply chain, in reality products are continuouslybeing shipped from over 140 U.S. refineries andarriving at more than 1,500 points of terminal dis-tribution. This continuous process allows promptreallocation of products to meet fluctuations inlocal or regional needs.

Products move from primary storage into second-ary storage, which includes distributor, retail sta-

tion, and industrial and commercial inventories.Product from secondary storage is transferred to theconsumer or end-user, making up the third inven-tory category called tertiary storage.

In facilitating the operation of this supply chain,inventory plays several roles: operational necessity,component of supply, financial opportunity, andprotection against a worldwide emergency. Whileinventories are a requirement for reliable supply,they also represent a cost of doing business.

Inventory as an Operational Necessity

The flow of crude oil and products around theworld is not always constant, homogenous, or unin-terrupted, due in part to the batch nature of move-ments in ships, barges, pipelines, railcars, andtrucks, as well as the real world variability associat-ed with these transportation modes.

Delivery of crude oil and products to the cus-tomer is a complex process. The movement ofcrude oil from its reservoir through the supplychain to become a finished product can takemonths due to the vast distances in which crudeoil and products normally travel. Crude oilreserves are generally not located near refineries,and refineries can be far removed from consumers.Significant quantities of crude oil and petroleumproducts are moved in large ships (some of whichhold the equivalent of one-quarter of the U.S. dailygasoline consumption) at speeds around 15 milesper hour. Pipelines flow at the pace of a fast walk,3 to 6 miles per hour. Even inside the UnitedStates, crude oil produced in Texas requires overtwo months before arriving as product on the U.S.East Coast. In addition to the relatively slow

PART II – OBSERVATIONS ON U.S. PETROLEUM INVENTORIES II-1

PART III

OBSERVATIONS ONU.S. PETROLEUM INVENTORIES

movement of oil, the batched nature of oil move-ments (crude oils from the well to the refinery andthe myriad of products from the refinery to termi-nals and ultimately to the consumer) creates addi-tional logistical complexity.

Inventory used as an operational necessityincludes tank bottoms, pipeline fill, in-transitinventory, and working inventory. Pipelines mustbe full in order to push crude oil and products totheir delivery points. Many tanks must have a cer-tain amount of crude oil or products always intheir bottom in order to receive or deliver. Thisinventory is normally not available to meetdemand because it is required to maintain a steadyoperation and it provides an interface betweeneach segment of the industry’s supply chain – pro-duction, transportation, refining, product distri-bution, and marketing. This allows the differentrates of flow in different parts of the supply systemto be balanced.

As the industry has achieved higher throughputrates while adding limited new inventory capacity,the improved operational efficiency is reflected inthe reduction in the “days of supply of inventory.”While a reduction in this number is reflective ofimprovements in efficiency, it does not reflect alower level of supply reliability.

Inventory as a Component of Supply

Another role of inventory is that of a componentof supply. This inventory is produced and stored inorder to meet expected future demand. Productdemand has seasonal and regionally specific char-acteristics. Off-season storage of inventory mayserve as an economic method of supplying futuredemand versus a more timely purchase or produc-tion increase. In the United States, heating oil isthe product that has historically seen the largestseasonal inventory builds to supply periods of peakdemand. Even though U.S. refinery capacity is

II-2 OBSERVATIONS ON PETROLEUM PRODUCT SUPPLY

Crude Oil Production

Lease Storage

Refinery Crude Oil Storage

Refining: Distillation

Refining: Cracking, Coking, etc.

Secondary Storage

Refinery Product Storage

Intermediate Product Storage

Tertiary Storage

GL

OB

AL

PE

TR

OL

EU

M S

UP

PLY

CH

AIN

CR

UD

E O

ILR

EF

ININ

GP

RO

DU

CT

S

Figure II-1. Global Petroleum Supply Chain

Product Distribution TerminalsPlus Shipment to Commercial Customers

Product ShipmentPipelines, Tankers, Barges, Tank Cars, Tank Trucks

Refinery Intermediate Storage

Shipments to RefineriesPipelines, Tankers, Barges, Tank Cars, Tank Trucks

Reserves, Natural Storage

Prim

ary Storage

Figure II-1. Global Petroleum Supply Chain

more than adequate to meet peak demand forheating oil and other distillate products, differingseasonal production economics have encouragedseasonal inventory builds. These seasonal buildshave been diminishing in recent years, whiledemand has been met by higher and more flexiblerefinery output, and increased imports. Compa-nies also build inventories in preparation forplanned maintenance in the production, refining,and logistical systems. To ensure that customerneeds are met, additional inventories are normallyheld as protection against variability in the ele-ments of the supply chain as well as in customerdemand. The quantity and use of these inventoriesvary and are subject to individual company operat-ing philosophies.

Inventory as a Financial Opportunity

Changes in industry aggregate inventories can-not be interpreted as changes in the adequacy ofsupply. Physical inventories can also be used toimprove economic performance. Inventory heldfor this purpose is referred to as “discretionary”because it is in excess of the level necessary foroperational efficiency. This capability is, to varyingdegrees, available in all segments of the supplychain: production, refining, terminaling, commer-cial end-use, and consumers. In the primary sec-tor, these inventories appear predominantly in ter-minals. Most companies actively manage theirphysical product inventory in response to econom-ic incentives. Companies may manage their indi-vidual level of discretionary inventories based ontheir assessment of future market conditions.There may also be incentives based on the futuresmarket that influence the level of discretionaryinventories. For example, if futures prices fordelivery in coming months are higher than theprice today, an incentive may exist to build inven-tory because it may be worth sufficiently more laterto cover the carrying cost and financial risks.Under this scenario, the market is said to be in con-tango. Conversely, if the futures prices for deliveryin coming months are lower than the price today,companies may have an incentive to draw discre-tionary stocks and keep them low while maximiz-ing sales. Under this scenario, the market is said tobe backwardated. Regardless of these financialdrivers, companies strive to hold sufficient invento-ries to operate reliably and meet customer demand.

Inventory as a Cost of Doing Business

The industry is very competitive, which driveseach company to continually strive to financiallyoptimize operations while providing adequate supplyto meet expected customer demand. Achieving boththese aims continually and successfully is a complexplanning task for an operating company, requiringcontinuous coordination between the various seg-ments of the supply chain – crude oil supply, refin-ing, transportation, and final product delivery.Crude oil has to be supplied in adequate volumes andin a reliable and timely manner for refineries toprocess it into finished products. Refineries thenhave to run reliably to process the crude oil to meetlocal or regional product demand and the distribu-tion systems need to have the capability to deliverproduct to market. Failure to plan for adequateinventories to meet an individual company’s require-ments results in a competitive disadvantage throughthe eventual loss of product sales and margins thatcould have been gained from those sales.

Inventory is considered working capital and, assuch, is a cost of doing business. Carrying moreinventory than required to operate the system mayincrease costs unnecessarily, unless there is a finan-cial incentive to do so. Conversely, carrying too lit-tle inventory increases the risk that the system maynot be able to meet customers’ demand, resulting ina lost competitive advantage. The competitivenature of the industry is such that companies try tominimize working capital (or costs) consistent withoperating reliable supply systems. Ultimately, cus-tomers benefit from efforts to reduce petroleumsupply costs, as evidenced in Figure I-21.

Inventory as Protection Against a Worldwide Emergency

The United States’ Strategic Petroleum Reserve isthe largest government crude oil stockpile in theworld. It was started in 1975 as part of the EnergyPolicy and Conservation Act. The Strategic Petro-leum Reserve currently contains around 670 mil-lion barrels of crude oil (August 2004), accountingfor about 40 percent of the primary U.S. petroleuminventory. The NPC remains strongly supportive ofholding these inventories for use only during signif-icant crude oil supply disruptions that threaten thesystem’s ability to meet domestic demand.


Although there is a strategic heating oil reserve inthe Northeast, the concept of a products strategicreserve was discussed as part of this update and theNPC does not believe such reserves are appropriatefor the United States. There isn’t the same necessi-ty of a products reserve as there is for the crude oilreserve given that light product imports representonly 7% of light product demand, while crude oilimports represent about 60% of U.S. crude oil sup-plies. There are a number of factors that make theholding and management of a strategic productsreserve complex and impractical. For example, thevery large number of boutique gasoline formula-tions would likely require strategic storage at mul-tiple regional, state, and/or local locations. Thisconcept is further complicated by the requirementof seasonal inventory turnovers to meet gasolinespecifications and product degradation experiencedwhen held in storage for a lengthy period. It is alsounlikely that there would be sufficient product ofthe right specification in the right location to behelpful during a supply disruption. Additionally, inthe face of a supply/demand disruption, the pres-ence of a products reserve could create uncertaintyabout when or whether it would be drawn down.These uncertainties would dampen the marketresponse to correct the supply/demand imbalance.

Supply/Demand Imbalances

Imbalances Sometimes Coincide with Industry Maintenance

As part of good operating practices, companiesperiodically shut down facilities for planned mainte-nance. Planning for major maintenance activities isa lengthy (typically in excess of 12 months) andcomplex task, which requires significant planningand lead time to establish required activities,acquire resources and equipment to carry out thoseactivities, and secure alternate product supply tocover customer demand, including adequate inven-tory. Planned maintenance is necessary for a vari-ety of reasons, such as equipment inspection, repair,and improvements, equipment replacement whenrequired, and ultimately safe operation. As shut-down activities are typically high cost and result inlost production, the industry has a strong financialincentive to extend periods between maintenanceand expedite shutdown activities as quickly as possi-ble, consistent with safe operating practice.However, planned maintenance almost always has

lower costs than emergency maintenance, so refin-ers have a practical limit on how long they can oper-ate between planned maintenance periods.

Maintenance activities on large assets, such asrefineries, is typically planned for either early springor late fall. There are primarily two reasons why theindustry selects these periods. Firstly, typicalweather patterns during these periods are less like-ly to interrupt (and hence extend) maintenanceactivities and secondly, historical demand and mar-gins are typically lower outside of the summer driv-ing season (minimizing the operating and financialimpact of the maintenance activities).

The nature of planned maintenance is such that itis extremely costly to make anything but a minorchange in the start date. This is because commit-ments for manpower and materials have been madewell in advance. As a result of committing to shut-down dates, industry maintenance activities some-times coincide with periods when there are signifi-cant unplanned supply and demand disturbances.

Unanticipated Events

Even with the best planning around maintenanceactivities, factors outside the control of operatingcompanies may lead to disturbances in thesupply/demand balance. Market dynamics and cus-tomer demand are unknown. If local or regionalproduct demand increases unexpectedly when anoperating company is committed to perform main-tenance activities, there may be a resultant tempo-rary imbalance in supply and demand. Similarly,there are times when unplanned events occur thatcause a temporary disturbance to supply. While thepublic may be aware of a few major unplannedevents each year, such as refinery fires, pipelinebreaks, etc., there are in fact many events that theindustry successfully responds to. Most of these dis-turbances are minor or moderate in nature but ifmultiple events occur simultaneously, a larger sup-ply disturbance can result.

Petroleum markets respond to supply/demandchanges with price movements that provide theincentive to increase or decrease supply to correctany imbalance. This is an integral part of normaland effective market operation. Through the indi-vidual response of various companies to these pricemovements, the petroleum industry as a whole


reacts quickly and effectively to maintain the sup-ply/demand balance in response to changes in local,regional, or global market conditions. Although theU.S. supply system is very complex, it is robust andhas demonstrated flexibility to adjust to significantsupply disturbances. Even during major supply dis-turbances, the supply/demand balance is typicallyrestored within a short period of time. Given thiscapability, the fastest and most efficient response tosupply disturbances is via market mechanismsdescribed above.

Changes in Fuels Quality Specifications Can Impact Inventories

The promulgation of federal and state mandatedchanges to fuel specifications is having an impacton inventories. Examples are:

• Lower inventory levels during spring gasolinevapor pressure transition. Federal fuels regula-tions require that gasoline have reduced vaporpressure in summer. The industry strives tomake the seasonal (winter to summer) transition,from high to low vapor pressure gasoline produc-tion, at as low a cost as possible. High Reid VaporPressure (RVP) gasoline inventories are reduced,to remove as much of the winter season gasolinestocks as possible, and then re-built with productthat meets the lower summer vapor pressurerequirements. During the period of reducedinventory, the system is at increased risk of notbeing able to respond as quickly if there is anyevent that causes a supply and demand imbal-ance.

• Boutique fuels. Fuels specifications establishedindependently by states and municipalities resultin the need to produce, transport, store, and deliv-er to consumers an increasing number of “bou-tique” fuels. In addition, the new gasoline formu-lation can disrupt normal supply processes becauseof reduced product fungibility. One example of thisis the MTBE ban imposed by the states of NewYork, Connecticut, and California in 2004.

• Ultra Low Sulfur Diesel. Beginning in June2006, highway diesel sulfur specifications arerequired to be lowered from 500 ppm to 15 ppm.The inventory-related impact of this specificationchange is not well understood at this time. Thereduced diesel sulfur specification will result inmore complex handling issues in the product dis-

tribution system to deal with the potential formore off-specification product.

Inventory Trends and Lower Operational Inventory

As discussed in the 1998 NPC InventoryDynamics study, U.S. petroleum inventoriesrespond to both market and infrastructurechanges in the supply system. This section reviewscrude oil and petroleum product inventory datasince the 1998 study and evaluates the lower oper-ational inventory (LOI) levels defined by the NPCin the 1998 study. The LOI, as defined in the 1998study, is the lower end of the demonstrated operat-ing inventory range updated for known and defin-able changes in the petroleum delivery system.While generally not used by industry, the NPC rec-ognizes the LOI as a gauge to help the governmentassess current inventory levels. Several factorsimpact the LOI level of crude oil and petroleumproducts. Factors that act to decrease LOI areimproved efficiency of the delivery systems andreduced domestic production. Increased demandand product specification changes may act toincrease the LOI.

Based on the observed crude oil inventory trends,the NPC concludes that crude oil LOI should be arange of 260 to 270 million barrels, compared tothe 1998 study conclusion of 270 million barrels.Since the 1998 study, crude oil inventory has beenas low as 260 million barrels a few times with noimpact on crude oil supply to U.S. refiners.However, in September 2004, Hurricane Ivan had asignificant impact on offshore oil platforms andpipeline movements. This created localized supplydisruptions at a few refineries even though crudeoil inventories were slightly above 270 million bar-rels. There were no supply issues prior toHurricane Ivan. Hurricane Ivan substantiallyreduced crude oil imports and shut in Gulf ofMexico production simultaneously. Peak disrup-tion resulted in a reduction of about 60% of theGulf Region crude oil supply, and as of earlyOctober, around 30% of Gulf Coast crude oil pro-duction remained shut in. As a result, a few refin-ers requested crude oil loans from the StrategicPetroleum Reserve to support crude oil runs attheir Gulf Coast refineries. All of these refineriesneeded low sulfur crude oil, which has been in tight


supply globally. The supply disruption is a reflectionof a few refineries in the region being impacted by anextreme disruption to local low sulfur crude oil sup-plies and their lack of ability to substitute lowerquality, higher sulfur crude oils. A crude oil LOIrange is recommended, rather than a single value, tobetter represent the degree of accuracy associatedwith the LOI methodology. Delivery system efficien-cy improvements and declining domestic crude oilproduction have resulted in reduced crude oil inven-tory levels. Continued improvements to pipelineand other infrastructure, and technologicalimprovements are likely to continue the trendtowards lower operating inventory in the future.

The situation brought about by Hurricane Ivan isonly one example of many different situations thatcould impact the LOI for crude oil. Using a singlenumber or even a range for LOI is not likely to coverall situations where apparent shortages of crude oilmay result. Since crude oil quality varies signifi-cantly from one supply point to another and refiner-ies are designed to operate using crude oil of givenqualities, sudden changes in crude oil supply couldcause a reduction in refinery throughput eventhough there appears to be ample volumes of crudeoil available. The same situation could occur withproducts because broad product categories likegasoline and distillate include a number of specificproduct types, all of which are not produced at everyrefinery. It is possible to have a product inventoryabove its LOI and experience an apparent shortagein a given location because product of a specificquality is not available.

Review of petroleum product inventory data sincethe 1998 study suggests that no appreciable changehas occurred in the overall gasoline and distillate(heating oil and diesel) inventory levels. The NPCbelieves that inventory changes from improvementsin system efficiency have been offset by increases inproduct demand and the number of different fuelspecifications. Therefore, product LOIs were not stud-ied in detail and no change is recommended at thistime in the LOIs for those products. It may be appro-priate to have an LOI range for both gasoline and dis-tillate. However, this was not analyzed as part of thisupdate given that there is no recommended change tothe LOIs. Similar to crude oil, it would be expectedthat lower operational inventories would continue todecline as the product supply system evolves. Given

the short time frame of this study, the potentialimpact of impending regulatory changes in diesel sul-fur content on distillate inventory was not studied.

Crude Oil

Figure II-2 shows the monthly closing inventorylevels of crude oil from 1973 to July 2004. Dataavailable for the 1998 study ended at mid-year 1998.New data and the LOI inventory level developed dur-ing that study are shown on the right hand side ofthe figure. Crude oil inventory has continued itsslow downward trend and has reached new lowssince the 1998 study. The prior lowest observedcrude oil level was 284 million barrels in December1996. Since then, month-end crude oil inventoryhas been around the 270 million barrel range sev-eral times, with the lowest observed value of 268million barrels in December 2003. In addition,weekly data (which are based on surveys and are notas accurate as monthly data) approached the 260million barrel level several times in early 2004. Nocrude oil shortfalls or supply issues were observedwith inventory in this range.

The crude oil delivery system and its inventoryrequirement is fundamentally different than for fin-ished products as a result of the difference in thenature of the end-users. The product delivery sys-tem has been built up over a number of years to pro-vide relatively small amounts of product to millionsof customers in thousands of locations. Over time,the evolution of the product delivery system haslargely been focused on efficient delivery of increas-ing amounts of product to an expanding customerbase. Crude oil, however, flows to relatively fewrefiners that have been decreasing in number overtime. In addition, as domestic production declinesand imports grow, an increasing amount of crudeoil inventory is in transit, which is not captured inthe inventory data. The very competitive nature ofthe petroleum business coupled with continued effi-ciency improvements and crude oil increasinglysupplied from offshore sources will likely drivedomestic crude oil inventories lower over time.

Motor Gasoline

Figure II-3 shows the monthly closing inventoryof total U.S. motor gasoline over the same period asfor crude oil. The long, slow decline of gasolineinventory primarily associated with finished gasoline



Figure II-2. U.S. Commercial Crude Oil Inventory Profile

200

250

300

350

400

450

1973 1977 1981 1985 1989 1993 1997 2001

MIL

LIO

N B

AR

RE

LS

MOI

Source of Data: Energy Information Administration and National Petroleum Council.

Notes: Excludes inventories in the Strategic Petroleum Reserve. Minimum Operating Inventories estimated in previous NPC studies.

LOI

YEAR

Figure II-2. U.S. Commercial Crude Oil Inventory Profile

MIL

LIO

N B

AR

RE

LS

Source of Data: Energy Information Administration and National Petroleum Council.Note: Minimum Operating Inventories estimated in previous NPC studies.

150

175

200

225

250

275

300

1973 1977 1981 1985 1989 1993 1997 2001

MOI LOI

YEAR

Figure II-3. U.S. Gasoline Inventory Profile

inventory at terminals that was observed in the 1998inventory study is no longer apparent. The NPCbelieves that while additional system efficiencieshave occurred, including those related to mergersand acquisitions, they have been offset by increasesin product demand and the number of different fuelspecifications.

While inventories continue to fluctuate seasonal-ly, the average level has remained flat since the laststudy. In addition, the low inventory level reachedduring the mid-1998 to 2004 time period was nolower than the minimum reached during the previ-ous study’s time period (1992 to mid-1998). Sincethe last study, inventories approached the LOI onlyonce, reaching 188 million barrels at the end ofOctober 2000. This was essentially the same as theprior minimum level observed in August 1997.

As discussed in the 1998 study, oxygenates are animportant part of the U.S. motor gasoline supplysystem and should be considered when evaluatingthe capability of the supply system to respond tomarket forces. Since the 1998 study, the role of oxy-genates has become more complex with MTBE bansin several states and the continued oxygenate

requirement in reformulated gasoline. As a result,ethanol markets have expanded beyond the Midwestand ethanol is now used for gasoline production inboth the Northeast and California. The transporta-tion and inventory management systems for ethanolare largely out of the control of petroleum industrycompanies and the ethanol industry’s capability tomeet demand is beyond the scope of this study.

Distillate

Figure II-4 shows the monthly closing invento-ries of distillate since 1973. Distilllate inventory hasremained essentially flat, albeit with seasonal fluc-tuations, through both the previous and currentstudy periods. The lowest observed monthly closinginventory of about 90 million barrels occurred inMarch 1996. Since that time, distillate inventorieshave been in the 95 to 100 million barrels range afew times but have not approached the prior low.

The distillate market is approaching a period ofsignificant change as new highway and off-highwaydiesel sulfur regulations begin to impact the mar-ket. The industry is addressing issues around thetransportation and delivery, such as quality control


50

100

150

200

250

300

MIL

LIO

N B

AR

RE

LS

Note: Minimum Operating Inventories estimated in previous NPC studies.Source of Data: Energy Information Administration and National Petroleum Council.

YEAR1973 1977 1981 1985 1989 1993 1997 2001

MOI LOI

Figure II-4. U.S. Distillate Inventory Profile

and product batch interface containment, of theselower sulfur products. At this time it is not clearwhat impact, if any, these regulations will have ondistillate inventories.

Lower Operational Inventory

Prior Studies Have Been Unable to Define anInventory Level Where Supply Problems Begin

It is widely understood that a significant part ofreported petroleum inventories are required tooperate the product and crude oil supply systemsand are not readily available to meet demand. Theseinventories were identified as “operational” invento-ries in the 1998 study. The NPC has performed sev-eral studies since the early 1970s to define theseinventory levels, originally defined as “minimumoperating inventory” and subsequently modified as“lower operational inventory” in the 1998 inventorystudy. The redefinition was developed to move awayfrom the concept that there is some definable inven-tory level where supply system reliability becomesof greater concern. While these levels have beenreviewed several times, in retrospect, the “mini-mum” operating inventory levels identified havebeen broached several times with no visible impacton available supplies.

LOI – One of Many Measures for Assessing Inventory Adequacy

LOI is a measure defined to indicate there may bediminished inventory-related supply flexibility, andit is developed from observed inventory levels.However, the LOI definition specifically highlightsthat the supply system is not necessarily approach-ing problems when inventories are at low levels andthat many other considerations need to be takeninto account.

Although the NPC has studied and defined loweroperational inventories a number of times, in prac-tice these data are not used by individual compa-nies when making supply decisions. Reliable prod-uct supply to U.S. consumers requires delivery ofabout 20 million barrels per day of a variety ofpetroleum products to thousands of locations.Over time, the system has evolved to a complex,interrelated delivery system comprised of ships,barges, rail, refineries, pipelines, terminals, andlocal outlets. This delivery system, from crude oilto end-user, has proven numerous times that it has

the inherent flexibility to respond to day-to-daymarket fluctuations as well as significant one timeevents. While there is some apparent comfortwhen inventories are perceived to be high, in reali-ty these inventories are seldom in the right place,at the right time, or of the right quality, when prob-lems occur. Similarly, when inventories approachlow levels relative to an historical perspective, thesupply system is not approaching some failurepoint. Rather, the inventory level is reflective of anoverall supply/demand balance where supply hasbeen lagging demand.

Crude Oil LOI Recommended to be a Range of 260-270 Million Barrels

As previously mentioned, crude oil inventory hasbeen below the LOI of 270 million barrels definedin the 1998 NPC study a few times with no impacton crude oil supply to U.S. refineries. A briefanalysis of the minimum observed inventory ineach of the reported crude oil inventory categoriesalong with the observed simultaneous minimumssuggests that PADD V in-transit inventory is about5 million barrels below the level used in determin-ing the prior crude oil LOI. This reduction is adirect result of the decline in Alaskan crude oilproduction. The data also indicate that theobserved lower inventory level in pipelines and ter-minals in PADDs I - IV is about 10 million barrelsbelow the data used in the 1998 study offset byabout a 5 million barrel increase in the minimumobserved level at PADD I - IV refineries. Noattempt was made to analyze the apparent causesof these changes given the study’s time constraintsbut they are consistent with the trend towardsincreased offshore crude oil supply coupled withcontinued efficiency improvements in the crude oillogistics system.

The NPC recommends that the LOI for crude oilbe reduced to a range of 260 to 270 million bar-rels, supported by the following factors: theobserved crude oil inventory has been as low as260 million barrels, apparent regional structuralchange has reduced inventory by 10 million bar-rels, and the expectation that there will be contin-ued cost pressure to reduce operational invento-ries. A crude oil LOI range is recommended,rather than a single value, to better represent thedegree of accuracy associated with the LOI meth-odology.


Data Suggest the 1998 Analysis of Product LOI is Still Valid

Given the short time available for this supple-mental report, a detailed assessment of all the vari-ous segments of reported product inventory datawas not possible. For gasoline, the overall data sug-gest that no significant changes have occurred since1998 and the NPC recommends no change to the185 million barrels LOI. For distillate, given thelack of apparent trend since the last study, the NPCrecommends no change to the 85 million barrelsLOI. As with crude oil, it may be appropriate tohave an LOI range for both gasoline and distillates.However, this was not analyzed as part of thisupdate. Market pressures for continued efficiencywill likely continue to drive operating inventories tolower levels. It should be expected that lower oper-ational inventories may continue to decline as theproduct supply system evolves, unless offset byother structural changes. Also, given the short timeframe of this study, the potential impact of impend-ing regulatory changes in diesel sulfur content ondistillate inventory was not studied.

The Importance of High Quality Inventory Data

The U.S. government maintains a system withinthe Energy Information Administration for gather-ing and disseminating information on U.S. petrole-um industry statistics. Petroleum data and informa-tion are also available from the American PetroleumInstitute and other private organizations. Thesedata and information are important for industry inmaking business decisions and for government inmaking informed policy decisions affecting the U.S.petroleum industry. U.S. petroleum information isgenerally considered to be among the best in theworld. In order to be useful, petroleum data must betimely, accurate, and relevant.

Every effort should be made by the EIA, workingwith industry, to ensure that the data are accurate,timely, relevant, and meet the highest quality stan-dards. Therefore, from time to time, the EIA andits U.S. petroleum industry counterparts shouldreview the data that are collected and subsequentlyreported to be certain the data meet the currentneeds. The NPC recommends that appropriateresources be made available to the EIA to supportcontinued efforts to work with industry to improvedata quality.

Assessment of Product Price Volatility

The 1998 study on U.S. Petroleum ProductSupply—Inventory Dynamics concluded that overthe time period of 1998-2002, the petroleum supplysystem balancing mechanisms available to respondto market events would not appreciably change.Thus, the frequency or magnitude of significant(non-crude oil related) upwards retail price moveswould not likely increase in the 1998-2002 timeperiod. Additionally, the 1998 study concluded thatsignificant price excursions of major light petrole-um products in the United States would continue tobe driven primarily by movements in the globalprice of crude oil.

Data analysis for this update, covering the 1998-2004 time period, supports most but not all of theconclusions of the 1998 study and also offers addi-tional insights. Many of the primary conclusions ofthe 1998 study remain valid:

• Global crude oil prices continue to be the primarydriver of product price levels. Even though thenumber and magnitude of product price up-ticks(increases of greater than 10% or more versusprior year period) has increased since 1997, mostof these events are driven by events in the globalcrude oil market.

• Retail price changes continue to lag behind spotprice changes, which has the effect of dampeningand delaying price swings at the retail level.

However, product and crude oil price levels andvolatility have increased since the previous study,which focused on a time period of relative calm in oilmarkets (1992-1997). In the 1998-2004 time period,crude oil price volatility peaked in 1998, while gaso-line price volatility peaked in early 2002. Marketprices since 2002 reflect an upward movement inproduct prices in line with an upward movement inglobal crude oil prices. Retail gasoline prices, how-ever, have been observed to be less volatile thancrude oil prices, as identified in the last study.

These conclusions are based on analysis ofnational data. Consumers at a local level may besubject to more or less volatility than the nationalaverage as a result of local factors that are not cap-tured by this analysis. This study primarily focusedon gasoline price volatility because it is the productwith the largest domestic demand and it is of the


greatest national concern. The study also focusedon the U.S. market, while crude oil markets, and tosome extent product markets, are global, and assuch are influenced by global fundamentals.

Increase in Global Crude Oil Price Level

Since the previous study was completed in 1998,there has been a significant increase in global crudeoil price levels, which has impacted prices across thevalue chain. As shown in Figure II-5, the globalcrude oil price collapsed (as evidenced by West TexasIntermediate) from an average price of over $20 perbarrel to a low of $12 per barrel in February 1999,and over the course of the following five years roseto the level of about $45 per barrel in August 2004.

In 2004, crude oil prices were driven by robustglobal oil demand growth, resulting from theglobal and U.S. economic recovery and stronggrowth in China. Global oil demand growth thisyear was high relative to the weak growth ratesexperienced in three of the last four years. Thisstrong demand growth is occurring at a time whenOPEC has little spare oil production capacity. Theresulting very tight supply/demand balance hasbeen exacerbated by market concerns about

increased geopolitical risk in a number of oil-pro-ducing countries.

Increase in Overall Retail Gasoline Price Level

As shown in Figure II-6, U.S. retail average gaso-line prices have also increased since 1998, followingcrude oil prices. Retail gasoline prices were partic-ularly strong in the first half of 2004 due to an ele-vated spot gasoline-to-crude oil price spread, whichadded to the crude oil price increase. Figure II-7shows that the spot gasoline-to-crude oil pricespread was elevated earlier this year, but by August2004 it had retreated to significantly lower levels.

There were some exceptional factors in the firsthalf of 2004 that were responsible for the tempo-rary up-tick in the gasoline-to-crude oil pricespread. After several years of weak growth, U.S. oildemand in the first half of 2004 grew at a rate of2.3% versus the first half of 2003 due to strongeconomic growth, as shown in Figure II-8.Temporary supply constraints associated withproduct specification changes also contributed tothe up-tick in the gasoline-to-crude oil spread inthe first half of 2004. Required and extendedrefinery turnarounds in 2004 in preparation for


10

15

20

25

30

35

40

45

50

1987 1989 1991 1993 1995 1997 1999 2001 2003

2004

DO

LLA

RS

PE

R B

AR

RE

L

Note: Monthly averages from daily data.Sources: Energy Information Administration (WTI, Spot Price (FOB) @ Cushing, OK ($/bbl)).

YEAR

Figure II-5. West Texas Intermediate Crude Oil Price


80

100

120

140

160

180

200

AUG.1990

AUG.1992

AUG.1994

AUG.1996

AUG.1998

AUG.2000

AUG.2002

AUG.2004

2004

CE

NT

S P

ER

GA

LLO

N


Figure II-6. Weekly Regular Conventional Retail Gasoline Price (including taxes)

YEAR

Figure II-6. Weekly Regular Conventional Retail Gasoline Price (including taxes)

0

2

4

6

8

10

12

1982 1984 1986 1988 1990 1992 1994 1996 1998 2000 2002 2004

GASOLINE SPREAD

DISTILLATE SPREAD

2004

DO

LLA

RS

PE

R B

AR

RE

L

Source: Platts Mid pipeline for products, WTI @ Cushing; 1982-84, WTI estimated from DOE Texas price. 2004 = year-to-date average through August.

AUG. 2004AVERAGEGASOLINE

YEAR

Figure II-7. U.S. Gulf Coast Gasoline and Distillate Spreads vs. West Texas Intermediate Crude Oil

the next phase of tighter gasoline sulfur specifica-tions may also have contributed to tightness ingasoline supplies.

Occasionally there are instances when productsupply/demand fundamentals influence the globalcrude oil market. There may have been such aninstance in the first half of 2004, when it appearedthat the tighter U.S. gasoline balance was con-tributing to the strength in the global crude oilprice. However, other factors were primarilyresponsible for higher crude oil prices during thistime period, such as strong global oil demand andgeopolitical supply risk. Strong U.S. gasolineprices in the first half of 2004, and the need foradditional gasoline supplies, may have con-tributed to the widening of the differential in theprice of light, low sulfur crude oils that maximizegasoline yield, like West Texas Intermediate, andthe price of heavier, higher sulfur crude oils thathave a lower gasoline yield. This type of incidenttends to equilibrate over time as thesupply/demand balance in product markets isrestored.

Crude Oil Prices are Still the Main Driver of Product Prices

Crude oil prices continue to be the main driverof product prices, as was concluded in the previ-ous study. A recent report by the Federal TradeCommission indicated that changes in crude oilprices have accounted for approximately 85% ofthe increases and decreases in U.S. motor gaso-line prices over the past two decades.1 A majorreason for this is that the cost of crude oil repre-sents a substantial portion of retail productprices. In July 2004, the U.S. Department ofEnergy estimated that crude oil price representednearly 60% of U.S. retail gasoline prices, exclud-ing taxes.

This sub-section examines the relationships ofcrude oil and spot product prices, and of spot andretail prices.


16

17

18

19

20

21

1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004

MIL

LIO

N B

AR

RE

LS

PE

R D

AY

Note: 2004 based on EIA Short-Term Energy Outlook, September 2004.Source: Energy Information Administration.

Figure II-8. U.S. Total Product Demand

YEAR

Figure II-8. U.S.Total Product Demand

1 Federal Trade Commission Bureau of Economics, “The PetroleumIndustry: Mergers, Structural Change, and Antitrust Enforcement,”August 2004, page 1.

Relationship of Crude Oil to Spot Product Prices

Figure II-9 demonstrates that gasoline and distil-late spot product prices tend to track crude oilprices. There are a number of publicly availableprice quotes for the various crude oil and productprices that provide a framework for all transactions.Spot prices are wholesale prices for physical deliveryof the crude or product set at a limited number ofindustry transfer points, such as a location on apipeline or at a harbor.

Spot product prices tend to follow crude oil pricessince the cost of crude oil is a large part of the costof products, and both crude oil and products aredriven by the same global market fundamentals.There is little to no lag in the response of spot prod-uct prices to crude oil price movements, as productmarkets have become more global and commodi-tized and more responsive to national and globalfundamentals.

There are still a number of factors that may resultin periodic dislocations between crude oil and prod-uct prices, such as seasonal demand, a tight productsupply/demand balance, and product specificationchanges. As previously described, there were a

number of exceptional factors (e.g., strong demandgrowth, product specification changes, etc.) in thefirst half of 2004 that resulted in an up-tick in thegasoline-to-crude oil price spread. However, thespread had fallen off substantially by the latter halfof the year.

Relationship of Spot to Retail Product Prices

Retail prices typically lag spot prices in partbecause some product goes through a succession ofresales by any combination of traders, jobbers, les-see dealers, or independent marketers. Figure II-10shows that retail prices lag spot gasoline prices onthe East Coast of the United States.

A U.S. Department of Energy study2 on gasolineprice pass-through from the spot to retail levellast year determined that significant changes inspot prices appear to show up in retail prices withsome time delay, and somewhat dampened. Thestudy estimated that 50% of the price pass-through from the spot to the retail market on a


1995 1997 1998 1999 2000 2001 2002 2003 20041996

Source: Energy Information Administration (USGC for products, Cushing for WTI, daily data averaged weekly 1995-present).

WTI

DISTILLATE

GASOLINE

YEAR

0

10

20

30

40

50

60

NO

MIN

AL

DO

LLA

RS

PE

R B

AR

RE

L

Figure II-9. Relationship Between West Texas Intermediate Crude Oil Prices and Spot Gasoline and Distillate Prices

2 U.S. Department of Energy, Energy Information Administration,Michael Burdette and John Zyren, “Gasoline Price Pass-through,”January 2003.

U.S. average basis occurs within two weeks and80% within four weeks.

Product Price Volatility

Price Events

As a means to evaluate retail price movement,the previous study focused on upside price eventsof significant magnitude to have generated publicconcern in the past. The study identified monthlyprice increases in crude oil, retail gasoline, andretail distillate prices in excess of 10% above theprior year.

Figure II-11 shows the instances of price increasesin excess of 10% above the prior year for crude oil,retail gasoline, and retail distillate. Non-crude oilrelated upward retail price movements tend to bedriven by an infrequent large event or a confluence ofsmaller events in the same direction. This figureshows that the number and magnitude of productprice up-ticks has increased since 1997, and that theyare mostly driven by events in the global crude oilmarket.

Volatility

Price changes play an important role in markets,providing incentives to producers and consumers toadjust supply and demand and keep the market inbalance. However, when prices rise rapidly, whatev-er the starting point, it is recognized that con-sumers of gasoline and home heating oil can beconcerned. This update is focused on the petroleumsupply system and does not address the implicationsof price volatility to consumers.

The standard statistical definition of volatility isdeviation around a mean. This study analyzedvolatility based on a 52-week moving average.Using a 52-week moving average reduces the impactof seasonality and is more indicative of a longer-term trend. Comparing daily percentage changes tothe average percentage change over the periodattempts to isolate true volatility from the underly-ing upwards trend in price.

Figure II-12 shows that both crude oil and gaso-line price volatilities increased in the period from1998 to 2004 relative to the period of focus for the


Figure II-10. Retail Prices Lag Spot Gasoline Prices

20

40

60

80

100

120

140

160

2000 2001 2002 2003 2004

NO

MIN

AL

SP

OT

CE

NT

PE

R G

ALL

ON

80

100

120

140

160

180

200

220

NO

MIN

AL

RE

TAIL

CE

NT

PE

R G

ALL

ON

(IN

CLU

DIN

G T

AX

ES

)

Note: Reformulated gasoline; Spot in NY Harbor, Retail for PADD I includes federal, state, and local taxes.Source: Energy Information Administration.

SPOT

RETAIL

YEAR

LAG BETWEEN RETAIL & SPOT REFORMULATED GASOLINE PRICE – PADD I (EAST COAST)

Figure II-10. Retail Prices Lag Spot Gasoline Prices


0

20

40

60

80

100

120

140

160

180

1973 1977 1981 1985 1989 1993 1997 2001 2005

Gulf War

PE

RC

EN

T

YEAR

Figure II-11. Instances of >10% Change in Price vs. Prior Year – Crude Oil and Products

Source of Data: Derived from Energy Information Administration Data. (Note: Distillate series not identical to 1997 study basis.)

IranianCrisis Price Recovery

from 7/86($11.26 Nominal)

Price Recoveryfrom 11/88

($12.63 Nominal)

1996Events

Price Recoveryfrom 12/93

($12.51 Nominal)

Price recoveryafter ‘98collapse

Run upprior to

Iraq war &Venezuela

strike

Arab OilEmbargo

CRUDE OIL RETAIL GASOLINE RETAIL DISTILLATE

Figure II-11. Instances of >10% Change in Price vs. Prior Year – Crude Oil and Products

0

10

20

30

40

50

60

1988 1990 1992 1994 1996 1998 2000 2002 2004YEAR

CRUDE OIL

AN

NU

ALI

ZE

D P

RIC

E V

OLA

TIL

ITY

(%

)–

52-W

EE

K R

OLL

ING

AV

ER

AG

E

Sources: Energy Information Administration; Oil and Gas Journal for retail gasoline prior to 1992.

RETAIL GASOLINE

Figure II-12. Retail Gasoline and West Texas Intermediate Crude Oil Price Volatility

previous study (1992-1997), which was one of relative market calm. Crude oil price volatilitypeaked in 1998, while gasoline price volatilitypeaked in early 2002. What has probably occurredsince 2002 is an upward movement in productprices in line with an upward movement in globalcrude oil prices. Retail gasoline prices are alsoshown to be less volatile than crude oil prices.

Future Product Price Volatility

It was beyond the scope of this report to devel-op a new supply/demand forecast or provide aforecast of future product price volatility.Furthermore, future product price volatility isinherently unknown since it is dependent uponfuture events that impact the global crude oilmarket. Some of the factors that could increaseor decrease product price volatility in the futureare identified below.

Factors That Potentially Increase Volatility

• Political events causing disruptions in oil-produc-ing countries.

• Greater proportion of global demand comingfrom developing countries with more volatileeconomies than OECD countries.

• Oil demand becoming more inelastic as oil isbacked out of the power and industrial sectors(elastic sectors), and income levels supportgreater use of oil in the transportation sector(inelastic sector).

• Lengthening and greater complexity of supplychains, including the possible diversion ofimports to other countries if they are short prod-uct and have higher prices than the U.S.

• Refining investments have long lead times.Uncertainty over regulations can delay invest-ment decisions and permitting processes can addto investment lead times. Both of these factorswill slow the industry’s response to bringing addi-tional supplies to the market.

• Continued specification changes, with highervolatility during transition periods.3

• Boutique fuels within the U.S. reduce the fungi-bility of products.4 This phenomenon is compli-cated by international fuel specifications notkeeping pace with changes in the U.S.

Factors That Potentially Decrease Volatility

• Common fuel specifications, increasing fungibili-ty of products

– Regions and states within the U.S. standardiz-ing specifications

– Non-OECD countries increasing stringency ofspecifications to be closer to OECD countryspecifications

• Removal of government barriers to industryresponsiveness

– Removal of barriers to permitting refineryexpansions, pipelines, and storage facilities

– Adjusting the depreciation schedule for equip-ment so that it is consistent with the treatmentin other manufacturing industries

– Allowing for adequate lead times in regulatorychanges

– Providing certainty about regulations, withoutrule changes midstream

• Improvements in industry response time

– Greater percentage of inventory on the waterallows greater mobility and flexibility for theproduct to go where it is most needed, althoughproduct on the water could be diverted to othercountries as previously specified.

Assessment of Inventories and Price

As addressed in the previous NPC inventory study,the expectation that inventories influence prices isbased on the economic assumption that pricesreflect the current supply/demand balance and thatinventories provide a measure, however imperfect,of the changing balance between supply anddemand. Inventories are a result of supply anddemand fundamentals. Any factor that serves as ameasure of the short-term supply/demand balancewould be expected to influence prices.


3 Office of Oil and Gas, Energy Information Administration, U.S.Department of Energy, 2003 California Gasoline Price Study:Preliminary Findings, May 2003.

4 Office of Oil and Gas, Energy Information Administration, U.S.Department of Energy, Gasoline Type Proliferation and PriceVolatility, September 2002.

However, statistical analysis of the simple relation-ship between inventories and prices or inventorychanges and price changes finds only a modest cor-relation. This conclusion is indicative of the factthat the interaction of inventories and prices is com-plex. Inventories are an imperfect measure of thesupply/demand balance, and prices for crude oil andpetroleum products are influenced by many factorsin addition to inventories.

When petroleum inventory data are made public,they can potentially have a short-lived effect onpetroleum prices, including futures prices. Thisappears particularly true when the inventory datadeviate from market expectations, in which case themarket reacts by bidding prices up or down on thebasis of the new information.

The previous NPC study also discussed the role offorward markets and time spreads in inventory deci-sions. Inventory levels may influence the shape ofthe forward price curve. As inventory levels rise, thespot price tends to fall relative to futures prices dueto perceptions of an oversupply. Conversely, asinventory levels fall, the spot price tends to rise rela-tive to the futures prices due to perceptions of cur-rent market tightness. In addition, the forward pricecurve can provide economic incentive or disincen-tive to companies to hold discretionary inventory, asthe forward price curve represents future priceexpectations. However, regardless of these factors,individual companies strive to hold sufficient opera-tional inventory to meet customer demand.

These conclusions are based on national indica-tors. The NPC recognizes that consumers at a locallevel may be subject to different forces and volatili-ty as a result of local conditions that are not cap-tured by this analysis.

Inventories, Prices, and Rational MarketBehavior

The policy community’s interest in inventory lev-els may stem from the expectation that when petro-leum inventory data is made public, it can potential-ly have a short-lived effect on petroleum prices,including futures prices. If inventory reports devi-ate significantly from market expectations, the mar-ket typically responds as expected by bidding pricesup or down on the basis of the new information con-veyed in the inventory data. Market participants

react to inventory signals because they are per-ceived as a measure of the short-term balancebetween supply and demand.

If product inventories are significantly belowexpectations, this signal of tightness in the balancemay lead participants to bid spot prices higher. Anincrease in prices should spur incremental produc-tion by progressively exceeding the economicthresholds required to deliver incremental supplies.These new supplies may come from incrementalproduct imports or increased output from domesticrefineries, as progressively higher-cost marginalproduction steps are implemented. These new sup-plies serve to restore the supply/demand balanceand ultimately cause prices to decline.

In a similar fashion, a report of higher-than-expected inventory levels would typically be inter-preted as a signal of looseness, which may causeprices to decline. As prices decline, marginal supplysources become increasingly uneconomic, and mar-ginal production increases are eventually reversed.With supply reduced, excess inventories decline, andsupply and demand are brought back into balance.

Relationship between Inventories and Prices

The NPC found only a modest correlation in thesimple relationship between current inventory lev-els and prices. It must also be understood that cor-relations do not necessarily imply causality. Inaddition, this correlation was not sufficient by itselfto use as a predictive tool. This modest relationshipindicates that current inventories are only animperfect measure of the supply/demand balance,and that many other factors impact prices. The U.S.Department of Energy found a slightly strongerrelationship between inventories and crude oil pricewhen they included total crude oil and productOECD inventories (not just U.S.), used relativeinventories as measured by actual inventories ver-sus “normal” inventories defined by historical sea-sonal movements and general trends, and includedrecent inventory trends (not just the current level).5

As shown in Figure II-13, prices and inventoriesonly loosely track each other, reflecting the weak


5 Michael Ye, John Zyren, and Joanne Shore, “A Forecasting Model forMonthly Crude Oil Spot Prices,” U.S. Department of Energy, July 15, 2003.

correlation. However, as indicated by theDepartment of Energy’s work, the relationshipbetween inventories and prices is complex.

Gasoline Inventories and Gasoline Price Spreads

Gasoline inventories are weakly related, if at all,to gasoline-to-crude oil price spreads (spot gasolineprice minus crude oil price). Various analysts havecompared gasoline inventory levels versus a “nor-mal” level (defined as previous five-year average) tothe gasoline-to-crude oil price spread. The theory isthat when gasoline inventories are in surplus, thegasoline-to-crude oil price spread will be low, andwhen gasoline inventories are lower than normal orin deficit, the gasoline-to-crude oil price spread willbe high. That is because inventories are a yardstickfor the short-term gasoline supply/demand balance.However, as shown in Figure II-14, the correlationbetween inventories and the price spread is weak.

This is not intended to imply that gasoline inven-tory levels relative to “normal” levels (previous five-year average) have no influence on gasoline-to-crude oil price spreads, but rather that there are

other factors that overwhelm the influence of inven-tories – whether inventories are measured inabsolute terms or relative to “normal levels.” Pricesultimately reflect the combined influences of allavailable market information, including inventorylevels, but also many other market factors, such asexpectations of future supply and demand.

Inventory Levels and Future Price Spreads

While inventories are weakly correlated to cur-rent prices, they may influence the shape of thefutures price curve or the spread between futuresprices and current prices. Futures markets, like theNew York Mercantile Exchange, allow buyers andsellers to conduct transactions for a limited numberof crude oil and petroleum products for delivery at aspecified time in the future at a specific location andprice. At any given time, prices are established bythis open and continuous auction for current andfuture trading months. This “strip” of future pricesis referred to as the forward price curve. The shapeof the forward price curve refers to the relationshipbetween outer month futures prices and the currentmonth price.


Sources: Energy Information Administration. Monthly data for inventories; Prices are monthly averages of daily data.June, July, and August estimated from weekly data for inventories.

TO

TAL

U.S

. CR

UD

E O

IL IN

VE

NT

OR

IES

(MIL

LIO

N B

AR

RE

LS)

[RE

VE

RS

E S

CA

LE]

WT

I CR

UD

E O

IL P

RIC

E (

$/B

BL)

240

260

280

300

320

340

360

380

4001987 1989 1991 1993 1995 1997 1999 2001 2003

10

15

20

25

30

35

40

45

TOTAL U.S. CRUDE OILINVENTORIES

WTI CRUDE OILPRICE ($/BBL)

YEAR

Figure II-13. Total U.S. Crude Oil Inventories (excluding SPR)vs. West Texas Intermediate Crude Oil Price

Figure II-13. Total U.S. Crude Oil Inventories (excluding SPR) vs. West Texas Intermediate Crude Oil Price

Figure II-15 shows the nature of the relationshipbetween inventories and the forward price spread,defined as a futures price minus the current price.Inventory, as it reflects the supply/ demand balance,

is one of the factors that can influence the forwardprice spread. In addition, the forward price spread,which represents future price expectations, mayalso influence company inventory-holding behavior.


FINISHED MOTOR GASOLINE INVENTORY VARIANCE

YEARSource: Energy Information Administration. June, July, and August estimated from weekly data.

-30

-25

-20

-15

-10

-5

5

10

151995 1996 1997 1998 1999 2000 2001 2002 2003 2004

- 6

-3

0

3

6

9

12

15

18

Figure II-14. Inventory Variance from Five-Year Average vs. Gasoline Price Spread

FIN

ISH

ED

MO

TO

R G

AS

OLI

NE

INV

EN

TO

RY

VA

RIA

NC

E F

RO

M P

RE

VIO

US

5-Y

EA

R A

VE

RA

GE

(MIL

LIO

N B

AR

RE

LS)

[RE

VE

RS

E S

CA

LE]

GA

SO

LIN

E P

RIC

E S

PR

EA

D (

$/B

BL)

[UN

LEA

DE

D G

AS

OLI

NE

LE

SS

WT

I CU

SH

ING

]

GASOLINE PRICESPREAD ($/BBL)

0

Figure II-14. Inventory Variance from Five-Year Average vs. Gasoline Price Spread

INVENTORIES

SUPPLY DEMAND

FORWARD SPREAD:DIFFERENCE BETWEEN CURRENT

AND FUTURES PRICES

CURRENTPRICE

FUTURESPRICE

Figure II-15. Interplay between Inventories and Futures Markets

Inventory Levels Influence the Forward PriceSpread

The relationship between commodity inventoriesand commodity price spreads has often beendescribed in the economic literature.6 The empiri-cal work demonstrates that there is a relationshipbetween inventories and the shape of the forwardprice curve. Inventories are positively correlatedwith the forward price spread.

When futures prices are higher than current spotprices, the market is said to be in contango.Conversely, when prices for future delivery arelower than spot prices for current delivery, the mar-ket is said to be in backwardation. Figures II-16 andII-17 clarify the features of a contango and backwar-dated market.

As shown in Figure II-18, low crude oil invento-ries tend to be associated with a backwardated for-ward price (futures price below current price) andhigh inventories tend to be associated with con-

tango in forward markets (futures price above cur-rent price).

Low inventories tend to be associated with back-wardated futures prices because they may be indi-cating present market tightness, which causes cur-rent spot prices to rise relative to future prices.Conversely, high inventories may be indicating anoversupplied market, causing current spot prices tofall relative to future prices.

Price Expectations Influence Inventories

As explained in the previous NPC study, future priceexpectations can affect company decisions regardinginventories. If the contango in the forward curve(futures price above current spot price) is sufficient tocover the storage costs and financial risk associatedwith holding inventory, there is an economic incen-tive to build stocks. Conversely, backwardated prices(futures prices below current spot price) may providean economic disincentive to hold inventories greaterthan those needed to meet supply requirements.Backwardation in the forward price curve indicatesthat prices are expected to fall and that the inventorieswill be worth less in the future. Regardless of thesefactors, individual companies strive to hold sufficientoperational inventory to meet customer demand.


SPOT 1 2 3 4 5 6 7 8 9 10 11 12

PR

ICE

(D

OLL

AR

S P

ER

BA

RR

EL)

FORWARDSPREAD

FUTURES PRICE

MONTHS TO CONTRACT EXPIRATION

Figure II-16. Contango Market – A market condition in which a futures priceis higher in the more distant delivery months than in the near delivery months

CURRENT OR “SPOT” PRICE

Figure II-16. Contango Market – A market condition in which a futures price is higher in the more distant delivery months than in the near delivery months.

6 Jeffrey C. Williams, “The Economic Function of Futures Markets,”Cambridge, England: Cambridge University Press, 1986. Michael J.Brennan, “The Supply of Storage,” American Economic Review 57,No. 1 (1958), pp. 50-72. Holbert Working, “The Theory of the Priceof Storage,” American Economic Review 48 (1949), pp. 1254-1262.


SPOT 1 2 3 4 5 6 7 8 9 10 11 12

MONTHS TO CONTRACT EXPIRATION

PR

ICE

(D

OLL

AR

S P

ER

BA

RR

EL)

FUTURES PRICE

Figure II-17. Backwardated Market – A market condition in which a futures priceis lower in the more distant delivery months than in the near delivery months.

CURRENT OR “SPOT” PRICE

FORWARDSPREAD

Figure II-17. Backwardated Market – A market condition in which a futures priceis lower in the more distant delivery months than in the near delivery months.

250

270

290

310

330

350

370

1992 1994 1996 1998 2000 2002 2004-$8.00

-$6.00

-$4.00

-$2.00

$0.00

$2.00

$4.00

$6.00

CR

UD

E O

IL IN

VE

NT

OR

IES

(M

ILLI

ON

BA

RR

ELS

)

CONTANGO

FORWARDSPREAD

Sources: NYMEX as reported by Logical Information Machines (LIM); Energy Information Administration.June, July, and August estimated from weekly data.

CRUDE OIL INVENTORIES

BACKWARDATION

YEAR

FO

RW

AR

D S

PR

EA

D (

$/B

BL)

[FU

TU

RE

S P

RIC

E (

12T

H M

O.)

– S

PO

T]

Figure II-18. U.S. Crude Oil Inventories and Forward Price Spreads

APPENDICES

APPENDIX A A-1

APPENDIX A

A-2 OBSERVATIONS ON PETROLEUM PRODUCT SUPPLY

BACKGROUND INFORMATION ON THE NATIONAL PETROLEUM COUNCIL

In May 1946, the President stated in a letter to the Secretary of the Interior that he had been impressed by the contribu-tion made through government/industry cooperation to the success of the World War II petroleum program. He felt thatit would be beneficial if this close relationship were to be continued and suggested that the Secretary of the Interior estab-lish an industry organization to advise the Secretary on oil and natural gas matters.

Pursuant to this request, Interior Secretary J. A. Krug established the National Petroleum Council (NPC) on June 18,1946. In October 1977, the Department of Energy was established and the Council was transferred to the new department.

The purpose of the NPC is solely to advise, inform, and make recommendations to the Secretary of Energy on any mat-ter requested by the Secretary, relating to oil and natural gas or the oil and gas industries. Matters that the Secretary wouldlike to have considered by the Council are submitted in the form of a letter outlining the nature and scope of the study.The Council reserves the right to decide whether it will consider any matter referred to it.

Examples of studies undertaken by the NPC at the request of the Secretary include:

• U.S. Arctic Oil & Gas (198l)

• Environmental Conservation – The Oil & Gas Industries (1982)

• Third World Petroleum Development: A Statement of Principles (1982)

• Petroleum Inventories and Storage Capacity (1983, 1984)

• Enhanced Oil Recovery (1984)

• The Strategic Petroleum Reserve (1984)

• U.S. Petroleum Refining (1986)

• Factors Affecting U.S. Oil & Gas Outlook (1987)

• Integrating R&D Efforts (1988)

• Petroleum Storage & Transportation (1989)

• Industry Assistance to Government – Methods for Providing Petroleum Industry Expertise During Emergencies(1991)

• Short-Term Petroleum Outlook – An Examination of Issues and Projections (1991)

• Petroleum Refining in the 1990s – Meeting the Challenges of the Clean Air Act (1991)

• The Potential for Natural Gas in the United States (1992)

• U.S. Petroleum Refining – Meeting Requirements for Cleaner Fuels and Refineries (1993)

• The Oil Pollution Act of 1990: Issues and Solutions (1994)

• Marginal Wells (1994)

• Research, Development, and Demonstration Needs of the Oil and Gas Industry (1995)

• Future Issues – A View of U.S. Oil & Natural Gas to 2020 (1995)

• Issues for Interagency Consideration – A Supplement to the NPC's Report: Future Issues –A View of U.S. Oil & Natural Gas to 2020 (1996)

• U.S. Petroleum Product Supply – Inventory Dynamics (1998)

• Meeting the Challenges of the Nation's Growing Natural Gas Demand (1999)

• U.S. Petroleum Refining – Assuring the Adequacy and Affordability of Cleaner Fuels (2000)

• Securing Oil and Natural Gas Infrastructures in the New Economy (2001)

• Balancing Natural Gas Policy – Fueling the Demands of a Growing Economy (2003).

The NPC does not concern itself with trade practices, nor does it engage in any of the usual trade association activities.The Council is subject to the provisions of the Federal Advisory Committee Act of 1972.

Members of the National Petroleum Council are appointed by the Secretary of Energy and represent all segments of theoil and gas industries and related interests. The NPC is headed by a Chair and a Vice Chair, who are elected by the Council.The Council is supported entirely by voluntary contributions from its members.

APPENDIX A A-3

Philip C. AckermanChairman, President and Chief Executive OfficerNational Fuel Gas Company

Jacob AdamsPresident/Chief Executive OfficerArctic Slope Regional Corporation

George A. Alcorn, Sr.PresidentAlcorn Exploration, Inc.

Conrad K. AllenPresidentNational Association of Black Geologists

and Geophysicists

Paul M. AndersonChairman and Chief Executive OfficerDuke Energy Corporation

Robert O. AndersonRoswell, New Mexico

Thurmon M. AndressManaging DirectorBreitBurn Energy LP

Philip F. AnschutzChairman and Chief Executive OfficerThe Anschutz Corporation

Gregory L. ArmstrongChairman and Chief Executive OfficerPlains All American Pipeline, L.P.

Robert G. ArmstrongPresidentArmstrong Energy Corporation

Greg A. ArnoldPresident and Chief Executive OfficerTruman Arnold Companies

Ralph E. BaileyChairman and Chief Executive OfficerAmerican Bailey Inc.

Robert W. BestChairman of the Board, President

and Chief Executive OfficerAtmos Energy Corporation

Alan L. BoeckmannChairman and Chief Executive OfficerFluor Corporation

Donald T. BollingerChairman of the Board and Chief Executive OfficerBollinger Shipyards, Inc.

John F. BookoutHouston, Texas

Wayne H. BrunettiChairman, President and Chief Executive OfficerXcel Energy Inc.

Philip J. BurguieresChief Executive OfficerEMC Holdings, L.L.C.

Victor A. BurkChairmanOil & Gas DivisionDeloitte & Touche LLP

Frank M. Burke, Jr.Chairman and Chief Executive OfficerBurke, Mayborn Company, Ltd.

Karl R. ButlerPresident and Chief Executive OfficerICC Energy Corporation

Thos. E. CappsChairman, President and Chief Executive OfficerDominion



MEMBERSHIP

2004/2005

Robert B. CatellChairman and Chief Executive OfficerKeySpan

Clarence P. Cazalot, Jr.President and Chief Executive OfficerMarathon Oil Company

H. Craig ClarkPresident and Chief Executive OfficerForest Oil Corporation

Luke R. CorbettChairman and Chief Executive OfficerKerr-McGee Corporation

Gregory L. CraigChief Executive OfficerCook Inlet Energy Supply L.L.C.

William A. CustardPresident and Chief Executive OfficerDallas Production, Inc.

Charles D. DavidsonChairman, President and Chief Executive OfficerNoble Energy, Inc.

Claiborne P. DemingPresident and Chief Executive OfficerMurphy Oil Corporation

Cortlandt S. DietlerChairman of the BoardTransMontaigne Inc.

David F. DornDenver, Colorado

Laurence M. DownesChairman and Chief Executive OfficerNew Jersey Resources

W. Byron DunnPresident and Chief Executive OfficerLone Star Steel Company

Daniel C. EckermannPresident and Chief Executive OfficerLeTourneau, Inc.

Lynn Laverty ElsenhansPresident Shell Oil Company

James W. EmisonChairman and Chief Executive OfficerWestern Petroleum Company

Ronald A. EricksonChief Executive OfficerHoliday Companies

Sheldon R. EriksonChairman of the Board, President

and Chief Executive OfficerCooper Cameron Corporation

Stephen E. EwingPresident and Chief Operating OfficerDTE Energy Gas

John G. FarbesPresidentBig Lake Corporation

Claire Scobee FarleyChief Executive OfficerRandall & Dewey, Inc.

G. Steven FarrisPresident and Chief Executive OfficerApache Corporation

William L. FisherBarrow Chair in Mineral Resources

Department of Geological Sciences andDirector of the Jackson School of GeoscienceUniversity of Texas

James C. FloresChairman of the BoardPlains Resources Inc.

Eric O. FornellManaging Director and Group ExecutiveGlobal Natural Resources GroupJ. P. Morgan Securities Inc.

APPENDIX A A-5


Joe B. FosterNon-executive ChairmanNewfield Exploration Company

Robert W. FriVisiting ScholarResources For the Future Inc.

James A. GibbsChairmanFive States Energy Company

John D. GiglioExecutive DirectorNational Association of

State Energy Officials

Lawrence J. GoldsteinPresidentPetroleum Industry Research

Foundation, Inc.

Charles W. GoodyearChief Executive OfficerBHP Billiton Plc

Andrew GouldChairman and Chief Executive OfficerSchlumberger Limited

S. Diane GrahamChair and Chief Executive OfficerSTRATCO, Inc. and Ecopath

Patrick J. F. GrattonPresidentAmerican Association of

Petroleum Geologists

William E. GreeheyChairman of the Board

and Chief Executive OfficerValero Energy Corporation

James T. HackettPresident and Chief Executive OfficerAnadarko Petroleum Corporation

Frederic C. HamiltonChairman and Chief Executive OfficerThe Hamilton Companies LLC

Christine HansenExecutive DirectorInterstate Oil and Gas Compact Commission

Lewis Hay, IIIChairman, President and Chief Executive OfficerFPL Group

John B. HessChairman, President and Chief Executive OfficerAmerada Hess Corporation

Jack D. HightowerPresident and Chief Executive OfficerCelero Energy LLC

Roy M. HuffingtonChairman of the Board and

Chief Executive OfficerRoy M. Huffington, Inc.

Dudley J. HughesPresidentHughes South Corporation

Ray L. HuntChief Executive OfficerHunt Oil Company

Hillard G. HuntingtonExecutive DirectorEnergy Modeling ForumStanford University

John R. HurdGeneral PartnerHurd Enterprises, Ltd.

Ray R. IraniChairman and Chief Executive OfficerOccidental Petroleum Corporation

Eugene M. IsenbergChairman and Chief Executive OfficerNabors Industries, Inc.



Mark A. JacksonSenior Vice President and Chief Financial OfficerNoble Corporation

Frank JohnsonFormer Chairman of the Board of DirectorsAmerican Association of Blacks in Energy

A. V. Jones, Jr.ChairmanVan Operating, Ltd.

Jon Rex JonesChairmanJones Management Corp.

Jerry D. JordanPresidentJordan Energy Inc.

Fred C. JulanderPresidentJulander Energy Company

John A. KanebChairman and Chief Executive OfficerGulf Oil Limited Partnership

W. Robert KeatingCommissionerDepartment of Telecommunications

and EnergyCommonwealth of Massachusetts

James W. KeyesPresident and Chief Executive Officer7-Eleven, Inc.

Richard D. KinderChairman and Chief Executive OfficerKinder Morgan Energy Partners, L.P.

Harold N. KvislePresident and Chief Executive OfficerTransCanada Corporation

Susan M. LandonPetroleum GeologistDenver, Colorado

Stephen D. LaytonPresident E&B Natural Resources

Virginia B. LazenbyChairman and Chief Executive OfficerBretagne G.P.

Joseph Lee IIIPresident and Chief Executive OfficerMemphis Light, Gas & Water Division

David J. LesarChairman of the Board, President

and Chief Executive OfficerHalliburton Company

Michael C. LinnPresidentLinn Energy, LLC

Daniel H. LopezPresidentNew Mexico Institute of Mining and Technology

Thomas E. LoveChairman and Chief Executive OfficerLove's Country Stores, Inc.

William D. McCabeDirectorNative American Programs and ResourcesDistributed Generation Systems Incorporated

Aubrey K. McClendonChairman of the Board and Chief Executive OfficerChesapeake Energy Corporation

W. Gary McGilvrayPresident and Chief Executive OfficerDeGolyer and MacNaughton

Cary M. MaguirePresidentMaguire Oil Company

Steven J. MalcolmPresident and Chief Executive OfficerThe Williams Companies, Inc.

APPENDIX A A-7


Charles J. MankinDirectorOklahoma Geological Survey

Timothy M. MarquezChairman and Chief Executive OfficerVenoco, Inc.

Donald L. MasonCommissionerThe Public Utilities Commission of Ohio

Frederick R. MayerChairmanCaptiva Resources, Inc.

F. H. MerelliChairman, President and Chief Executive OfficerCimarex Energy Co.

C. John MillerChief Executive OfficerMiller Energy, Inc.

David B. MillerSenior Managing DirectorEnCap Investments L.P.

Merrill A. Miller, Jr.Chairman, President and Chief Executive OfficerNational Oilwell, Inc.

Michael G. MorrisChairman, President and Chief Executive OfficerAmerican Electric Power Co., Inc.

Robert A. MosbacherChairmanMosbacher Energy Company

James J. MulvaChairman of the Board and Chief Executive OfficerConocoPhillips

John Thomas MunroPresidentMunro Petroleum & Terminal Corporation

David L. MurfinPresidentMurfin Drilling Co., Inc.

Mark B. MurphyPresidentStrata Production Company

William C. Myler, Jr.PresidentThe Muskegon Development Company

Gary L. NealeChairman, President and Chief Executive OfficerNiSource Inc.

J. Larry NicholsChairman of the Board and Chief Executive OfficerDevon Energy Corporation

John W. B. NorthingtonPrincipalThomas Advisors Inc.

Erle NyeChairman of the Board and Chief ExecutiveTXU Corp.

Scott OganFormer ChairmanThe Energy Council

Christine J. OlsonChairman and Chief Executive OfficerS. W. Jack Drilling Company

David J. O'ReillyChairman of the Board and Chief Executive OfficerChevronTexaco Corporation

C. R. PalmerChairman EmeritusRowan Companies, Inc.

Mark G. PapaChairman and Chief Executive OfficerEOG Resources, Inc.



Paul H. ParkerVice PresidentCenter for Resource Management

Robert L. Parker, Sr.Chairman of the BoardParker Drilling Company

A. Glenn PattersonPresident and Chief Operating OfficerPatterson-UTI Energy Inc.

Ross J. PillariPresidentBP America Inc.

L. Frank PittsOwnerPitts Energy Group

Keith O. RattieChairman, President and Chief Executive OfficerQuestar Corporation

Lee R. RaymondChairman and Chief Executive OfficerExxon Mobil Corporation

John G. RicePresident and Chief Executive OfficerGE Energy

Corbin J. Robertson, Jr.PresidentQuintana Minerals Corporation

Robert E. RoseChairman of the BoardGlobalSantaFe Corporation

Henry A. Rosenberg, Jr.Chairman of the BoardRosemore, Inc.

Mark A. RubinExecutive DirectorSociety of Petroleum Engineers

Robert SantistevanExecutive DirectorSouthern Ute Indian Tribe Growth Fund

S. Scott SewellPresidentDelta Energy Management, Inc.

Bobby S. ShackoulsChairman, President and Chief Executive OfficerBurlington Resources Inc.

Scott D. SheffieldChairman, President and Chief Executive OfficerPioneer Natural Resources Company

Matthew R. SimmonsChairman and Chief Executive OfficerSimmons and Company International

Sam R. SimonPresident and Chief Executive OfficerAtlas Oil Company

Bob R. SimpsonChairman and Chief Executive OfficerXTO Energy Inc.

Robert D. SomervilleChairman and Chief Executive OfficerAmerican Bureau of Shipping &

Affiliated Companies

Joel V. StaffChairman and Chief Executive OfficerReliant Resources, Inc.

Mitchell SteinhauseChairman of the BoardNew York Mercantile Exchange

Charles C. Stephenson, Jr.Chairman, President and Chief Executive OfficerVintage Petroleum, Inc.

J. W. StewartChairman, President and Chief Executive OfficerBJ Services Company

APPENDIX A A-9


Richard H. StraeterPresidentContinental Resources of Illinois, Inc.

Diemer TruePartnerTrue Companies, Inc.

H. A. True, IIIPartnerTrue Oil Company

William P. UttPresident and Chief Executive OfficerTractebel North America, Inc.

Paul G. Van WagenenChairman, President and Chief Executive OfficerPogo Producing Company

Randy E. VelardePresidentThe Plaza Group

Philip K. Verleger, Jr.PresidentPKVerleger, L.L.C.

Fred P. VigilAdministratorOil and Gas AdministrationJicarilla Apache Nation

John B. WalkerPresident and Chief Executive OfficerEnerVest Management Partners, Ltd.

Joseph C. Walter, IIIPresidentWalter Oil & Gas Corporation

L. O. WardChairman and Chief Executive OfficerWard Petroleum Corporation

Wm. Michael Warren, Jr.Chairman, President and Chief Executive OfficerEnergen Corporation

J. Robinson WestChairman of the BoardThe Petroleum Finance Company

Michael E. WileyHouston, Texas

Bruce W. WilkinsonChairman of the Board

and Chief Executive OfficerMcDermott International, Inc.

Barry A. WilliamsonAttorney At LawBarry A. Williamson, P.C.

Charles R. WilliamsonChairman of the Board

and Chief Executive OfficerUnocal Corporation

Mary Jane WilsonPresident and Chief Executive OfficerWZI Inc.

George M. YatesPresident and Chief Executive OfficerHEYCO Energy Group

John A. YatesPresidentYates Petroleum Corporation

Daniel H. YerginChairmanCambridge Energy Research Associates

Henry ZarrowVice ChairmanSooner Pipe Inc.



APPENDIX B B-1

APPENDIX B

Gregory L. ArmstrongChairman and Chief Executive OfficerPlains All American Pipeline, L.P.

Greg A. ArnoldPresident and Chief Operating OfficerTruman Arnold Companies

Alan L. BoeckmannChairman and Chief Executive OfficerFluor Corporation

Victor A. BurkChairmanOil & Gas GroupDeloitte & Touche LLP

Clarence P. Cazalot, Jr.President and Chief Executive OfficerMarathon Oil Company

Claiborne P. DemingPresident and Chief Executive OfficerMurphy Oil Corporation

Lynn Laverty ElsenhansPresidentShell Oil Company

Ronald A. EricksonChief Executive OfficerHoliday Companies

John D. GiglioExecutive DirectorNational Association of State Energy Officials

Lawrence J. GoldsteinPresidentPetroleum Industry Research Foundation, Inc.

S. Diane GrahamChair and Chief Executive OfficerSTRATCO, Inc. and Ecopath

William E. GreeheyChairman of the Board and

Chief Executive OfficerValero Energy Corporation

John B. HessChairman, President and

Chief Executive OfficerAmerada Hess Corporation

Ray L. HuntChief Executive OfficerHunt Oil Company


COMMITTEE ONREFINING AND INVENTORY ISSUES

COCHAIR, REFINING

Lee R. RaymondChairman and Chief Executive OfficerExxon Mobil Corporation

COCHAIR, INVENTORY

James J. MulvaChairman of the Board and Chief Executive OfficerConocoPhillips

GOVERNMENT COCHAIR

David K. GarmanActing Under SecretaryU.S. Department of Energy

SECRETARY

Marshall W. NicholsExecutive DirectorNational Petroleum Council

* * *

John A. KanebChairman and Chief Executive OfficerGulf Oil Limited Partnership

James W. KeyesPresident and Chief Executive Officer7-Eleven, Inc.

Richard D. KinderChairman and Chief Executive OfficerKinder Morgan Energy Partners, L.P.

Thomas E. LoveChairman and Chief Executive OfficerLove's Country Stores, Inc.

John Thomas MunroPresidentMunro Petroleum & Terminal Corporation

David J. O'ReillyChairman of the Board and Chief Executive OfficerChevronTexaco Corporation

Ross J. PillariPresidentBP America Inc.

John G. RicePresident and Chief Executive OfficerGE Energy

Henry A. Rosenberg, Jr.Chairman of the BoardRosemore, Inc.

Sam R. SimonPresident and Chief Executive OfficerAtlas Oil Company

Mitchell SteinhauseChairman of the BoardNew York Mercantile Exchange

Philip K. Verleger, Jr.PresidentPKVerleger, L.L.C.

B-2 OBSERVATIONS ON PETROLEUM PRODUCT SUPPLY

COMMITTEE ON REFINING AND INVENTORY ISSUES

Daniel H. YerginChairman

Cambridge Energy Research Associates

Armand S. AbayDirectorManufacturing SupportShell Oil Products

Kevin W. BrownExecutive Vice President, OperationsSinclair Oil Corporation

Lawrence J. GoldsteinPresidentPetroleum Industry Research Foundation, Inc.

Gary R. HemingerPresidentMarathon Ashland Petroleum LLC

William R. KlesseExecutive Vice President and

Chief Operating OfficerValero Energy Corporation

Robert A. LevinSenior Vice President, ResearchNew York Mercantile Exchange

Jerry E. MilhornVice President, Operations/EngineeringKinder Morgan Energy Partners, L.P.

Thomas F. MuellerGeneral ManagerRefining ServicesConocoPhillips

Joanne M. ShoreAnalysis Team LeaderOffice of Oil and GasEnergy Information AdministrationU.S. Department of Energy

Thomas J. SimonsManager, Strategy and PlanningGlobal RefiningChevronTexaco Global Downstream LLC

APPENDIX B B-3


REFINING SUBCOMMITTEEOF THE

NPC COMMITTEE ON REFINING AND INVENTORY ISSUES

SECRETARY

Benjamin A. Oliver, Jr.Senior Committee Coordinator

National Petroleum Council

* * *

CHAIR

Donald H. DaigleVice President, RefiningExxonMobil Refining and Supply Company

ASSISTANT TO THE CHAIR

Thomas R. EizemberSenior Planning AdvisorCorporate PlanningExxon Mobil Corporation

GOVERNMENT COCHAIR

Mark R. MaddoxActing Assistant SecretaryOffice of Fossil EnergyU.S. Department of Energy

ALTERNATE GOVERNMENT COCHAIR

James A. SlutzDeputy Assistant Secretary for Oil and Natural GasOffice of Fossil EnergyU.S. Department of Energy

Daniel H. BrusstarDirector, Energy ResearchNew York Mercantile Exchange

Ron ChittimSenior Refining AssociateAmerican Petroleum Institute

Terry W. DayProject Development AdvisorAmerica’s Area Engineering OfficeExxonMobil Research and

Engineering Company

William R. FingerSenior Associate Cambridge Energy Research Associates

Richard GrissomExecutive DirectorMarket AnalysisValero Energy Corporation

Michael E. LeisterManagerFuels TechnologyMarathon Ashland Petroleum LLC

Robert G. SlaughterPresidentNational Petrochemical and Refiners Association

James A. SmithDirector, Downstream OilCambridge Energy Research Associates

Richard H. WynnManagerFuels Planning, Quality and

Regulatory ComplianceCITGO Petroleum Corporation

David A. ZinamonSenior DirectorRefining and Environmental AffairsPIRA Energy Group


Frank A. VerrastroDirector

Energy ProgramCenter for Strategic & International Studies

SPECIAL ASSISTANTS

Jerry E. ThompsonChief Operations OfficerCITGO Petroleum Corporation

William R. VenoDirector, Global DownstreamCambridge Energy Research Associates

REFINING SUBCOMMITTEE OF THE NPC COMMITTEE ON REFINING AND INVENTORY ISSUES

Michael R. ConnerSurvey Statistician, Petroleum DivisionEnergy Information AdministrationU.S. Department of Energy

Corey T. Correnti Chief Operating OfficerProducts Supply and Trading BP Products North America Inc.

Terry W. DayProject Development AdvisorAmerica's Area Engineering OfficeExxonMobil Research and

Engineering Company

Gene EdwardsSenior Vice PresidentProduct Supply, Trading and

Wholesale MarketingValero Energy Corporation

Lawrence J. GoldsteinPresidentPetroleum Industry Research

Foundation, Inc.

James E. HollandVice President of OperationsKinder Morgan Inc.

Dennis M. HoustonExecutive Vice PresidentExxonMobil Refining and Supply Company

Michael E. LeisterManagerFuels TechnologyMarathon Ashland Petroleum LLC

Robert A. LevinSenior Vice President, ResearchNew York Mercantile Exchange

APPENDIX B B-5


INVENTORY SUBCOMMITTEEOF THE

NPC COMMITTEE ON REFINING AND INVENTORY ISSUES

SECRETARY

John H. Guy, IVDeputy Executive Director

National Petroleum Council

* * *

CHAIR

Philip L. FredericksonExecutive Vice PresidentCommercialConocoPhillips

ASSISTANT TO THE CHAIR

Thomas F. MuellerGeneral ManagerRefining ServicesConocoPhillips

GOVERNMENT COCHAIR

Mark R. MaddoxActing Assistant SecretaryOffice of Fossil EnergyU.S. Department of Energy

ALTERNATE GOVERNMENT COCHAIR

James A. SlutzDeputy Assistant Secretary for Oil and Natural GasOffice of Fossil EnergyU.S. Department of Energy

Aaron F. Brady Associate Director, Downstream Oil Cambridge Energy Research Associates

Daniel H. BrusstarDirector, Energy ResearchNew York Mercantile Exchange

Rosendo Cruz, Jr.Issues AdvisorU.S. Government RelationsExxon Mobil Corporation

Nick EdwardsDirector, Business AnalysisConocoPhillips

William R. EngibousManagerSupply Optimization Group West CoastChevronTexaco Global Supply and Trading

William R. FingerSenior Associate Cambridge Energy Research Associates

R. Dean ForemanSenior Economist – Corporate PlanningEconomics and Energy DivisionExxon Mobil Corporation

Richard GrissomExecutive DirectorMarket AnalysisValero Energy Corporation

James E. HegartyDirector, Clean Fuels Strategy ConocoPhillips

Nancy L. Johnson DirectorPetroleum Planning and Environmental AnalysisOffice of Fossil EnergyU.S. Department of Energy

Marianne S. KahChief EconomistConocoPhillips

L. Drew KronenbergerInventory Management AdvisorGlobal Logistics OptimizationExxonMobil Refining and Supply Company

Michael C. McParlandIndustry AnalystRefining and SupplyExxon Mobil Corporation

Ralph MullingerEconomistMarathon Oil Company

Olayinka I. OgunsolaTechnology AdvisorOffice of Oil and Natural GasOffice of Fossil EnergyU.S. Department of Energy

Edward D. PorterSenior EconomistPolicy Analysis DepartmentAmerican Petroleum Institute

Thondiyil PremkumarManagerIndustry Analysis and Operations SupportExxonMobil Refining and Supply Company


INVENTORY SUBCOMMITTEE OF THE NPC COMMITTEE ON REFINING AND INVENTORY ISSUES

SPECIAL ASSISTANTS

Gary C. LockhartVice President, Gasoline7-Eleven, Inc.

William R. VenoDirector, Global DownstreamCambridge Energy Research Associates

Philip K. Verleger, Jr. PresidentPKVerleger, L.L.C.

Michael K. WirthPresidentChevronTexaco Global Supply and Trading

Leonard RucinskiMarket Analysis ManagerValero Energy Corporation

Randall S. SharumDirector, Business AnalysisConocoPhillips

F. Dexter SutterfieldTechnology ManagerOffice of National Petroleum TechnologyU.S. Department of Energy

David A. ZinamonSenior DirectorRefining and Environmental AffairsPIRA Energy Group

APPENDIX B B-7

INVENTORY SUBCOMMITTEE OF THE NPC COMMITTEE ON REFINING AND INVENTORY ISSUES

Report Recommendation

• Regulations should be based on sound science,thorough analysis of cost effectiveness, and com-mercially proven technology.

• The NPC proposed that fuel quality changes haveminimum overlap to prevent resource constraints(financial and human); ULSD was recommendedno sooner than 2007. EPA should finalize timingand specifications for off-highway diesel sulfurreduction and national MTBE use in a timelymanner.

• Regulations should be defined with certainty inscope, timing, and requirements to allow therefining industry to make effective investmentsdecisions. Allow at least four years time fromadoption to implementation.

• Policy makers should recognize that policies orregulations favoring or promoting renewable oralternative fueling tends to discourage refineryinvestment.

Subsequent Action

• The ULSD regulation of a 15 ppm cap at retail wasimplemented as a technology forcing regulation.While some technology development hasoccurred, enforcement tolerances and down-grades in the distribution system raise seriousquestions about the ability to comply with therule. Commercially proven light duty engine andafter treatment technology did not and does notexist to use 15 ppm diesel to comply with NOxemissions specifications.

The NPC questions the adequacy of the non-roaddiesel cost-effectiveness analysis.

• The timing for ULSD was set in mid-2006, earlierthan the 2007 recommendation. The non-roaddiesel rules were finalized in 2004 for a mid-2007implementation. Regulations for MTBE continueto be a patchwork by individual states.

• EPA has not provided clarity to state fuel require-ments including oxygen waivers and NAAQS. Thelead time of four years has only been partially rec-ognized. NSR reforms necessary to improve reg-ulatory certainty have been promulgated but areunder litigation.

• Considerable uncertainty exists surrounding apossible Renewable Fuels Standard. Nationalenergy legislation continues to be debated butlegislation has yet to be passed.

APPENDIX C C-1

APPENDIX C

2000 REFINING STUDYRECOMMENDATIONS AND STATUS

Report Recommendation

• Reducing gasoline or on-highway diesel below 30ppm average should not be imposed until a basisfor sound conclusions about the cost, benefit,producibility, and deliverability of products isestablished.

• Changing the DI specification should not occuruntil further analysis of the cost effectiveness andpotential supply impacts is identified.

• Provide a streamlined permitting process andallocate necessary resources to support state andlocal agencies.

• Allow a portion of the emissions reduction result-ing from the use of lower sulfur fuels to offset thestationary source emissions resulting from newfacilities required in their manufacture.

• Requirements for the new source review shouldnot be retroactively reinterpreted.

• States and localities considering localized re-strictive fuel requirements should recognizeassociated increased costs and reduced reliabilityof supply.

• Enforcement requirements for complianceshould maintain the flexibility and capability ofdistribution system, i.e. primary enforcement atthe refinery gate, point of production, or import.

Subsequent Action

• EPA established a 30 ppm standard for gasoline atthe refinery gate. EPA did not follow this recom-mendation for diesel, imposing a 15 ppm standardat retail. Light duty engine technology remainsunproven and deliverability remains in question.

• As recommended, no DI specification has beenimposed by EPA. However, states (California andMichigan) are beginning to discuss including a DIspecification as part of state fuel requirements.

• A limited amount of streamlining has occurred.

• No low sulfur fuel offsets have been allowed in thepermitting of new facilities to produce low sulfurfuels.

• NSR reform is under litigation. However, noenforcement relief of reinterpretation of past NSRactions has been implemented.

• States and localities continue to impose restric-tive fuel requirements resulting in boutique fuels.Connecticut did revise the timing of their MTBEban to align with New York when advised ofpotential supply impacts.

• Enforcement of Tier 2 gasoline sulfur does main-tain distribution flexibility with primary refinerygate enforcement. Enforcement of ULSD at retailintroduces substantial issues for the supply anddistribution of ULSD starting in mid-2006.

C-2 OBSERVATIONS ON PETROLEUM PRODUCT SUPPLY

CUMULATIVE REGULATORY IMPACTS ON REFINERIES, 2000-2008 D-1

APPENDIX D

CUMULATIVE REGULATORY IMPACTSON REFINERIES, 2000-2008

2000 2001 2002 2003 2004 2005 2006 2007 2008

CALIFORNIA MTBE PHASE OUTREGIONAL HAZE3

NEW YORK MTBE PHASE OUT

GASOLINE TOXICS CONTROL

SECTION 126 PETITIONS

NEW SOURCE REVIEW ENFORCEMENT

RESIDUAL RISK

EXT.

TIER II GASOLINE SULFUR1

RENEWABLE FUELS MANDATE2

YEAR

Source: National Petrochemical & Refiners Association, March 2002.

ON-ROAD DIESEL4

ACTUAL TIME FRAME KNOWN OR BASED ON 36-48 MONTH COMPLIANCE SCHEDULE AFTER FINAL RULE ISSUED

COMPLIANCE REQUIREMENTS UNKNOWN AND TIME FRAME ESTIMATED

OFF-ROAD DIESEL5

REFINERY MACT II6

URBAN AIR TOXICS (AREA SOURCES)7

1. Longer compliance time for refineries in Alaska and Rocky Mountain states and small refineries covered by Small Business Regulatory Enforcement and Flexibility Act (SBREFA). Additional compliance time is available for these refineries if they produce ultra low sulfur highway diesel beginning in 2006.

2. Senate Energy bill (S. 517) proposes an ethanol mandate of 2.3 billion gallons in 2004 whichincreases to 5 billion gallons in 2012.

3. Regional Haze State Implementation Plans (SIPs) due 2005-2007.

Earliest compliance date. Schedule may be impacted by National Ambient Air Quality Standard (NAAQS) litigation.

4. Longer compliance time for small refiners covered by SBREFA.

5. Estimated effective date based on proposed heavy duty vehicle standards.

6. Compliance date may be harmonized with Tier II schedule.

7. Urban Air Toxics Strategy includes potential controls of gasoline loading facilities at refineries. Estimated compliance schedule.

The NPC recommends: “Immediate imple-mentation of comprehensive NSR reform is avery important policy step needed to improve

the climate for investment in domestic refineryexpansion. The NSR reforms promulgated by theAdministration, including the EquipmentReplacement Rule currently under judicial review,should be implemented as soon as possible.Attempts to delay or overturn the reforms should bevigorously opposed. Additional NSR reform propos-als regarding de-bottlenecking and project aggre-gation should be issued and finalized.”

This recommendation flows from several keyobservations and conclusions of the study group, asfollows:

Lack of grassroots domestic refinery construc-tion since the mid 70s reflects the historicalprofitability of the industry, more attractiveopportunities for expansion of existing refiner-ies and the economic availability of alternativesupplies.

Expansion and increased utilization of existingrefineries is typically more feasible, quicker,and less costly than grassroots refinery con-struction.

One of the significant factors affecting invest-ment decisions is the amount of futureuncertainty. While uncertainty cannot beeliminated, it can be exacerbated by certaingovernment actions. Where decisions re-garding large amounts of capital investmentare involved, increasing uncertainty tends toresult in minimization or delay of investmentto reduce risk. This results from the long-

lived nature of refining assets, where taking ashort term economic loss to await resolutionof uncertainty may be more attractive thaninvesting earlier in equipment that is notoptimum for the long term.

The NPC’s strong support for NSR reform reflectsthe fact that the national interest requires thatunnecessary, policy-based impediments to addition-al investment in domestic refinery capacity expan-sions be eliminated. No other federal policy has fos-tered as much uncertainty and discouragement ofdomestic refining investment as has EPA’s pastapplication and enforcement of the New SourceReview program.

By the end of the last decade, NSR requirementswere subject to a welter of conflicting interpreta-tions and court decisions. A sweeping reinterpreta-tion of the program at that time exacerbated thissituation, and finally brought NSR-related problemsto a head.

The Administration’s 2001 National Energy Policyrecommended that EPA review the potential impactof the NSR program on investment in new utilityand refinery capacity, energy efficiency and environ-mental protection. As part of the public reviewprocess, the refining industry submitted evidence toEPA that the existing NSR program resulted in fore-gone opportunities for additional capacity incre-ments, prevented increased production of cleanerfuels, and discouraged energy efficiency projects.

On June 13, 2002, EPA presented a report on NSRand recommendations for reform to the President.EPA found that the NSR program had impeded orresulted in the cancellation of projects that would

APPENDIX E E-1

APPENDIX E

NEW SOURCE REVIEW (NSR)REFORM BACKGROUND

maintain or improve reliability, efficiency or safetyof existing power plants and refineries. EPA con-cluded that reforms to the NSR program wouldremove barriers to pollution prevention projects,energy efficiency improvements, and investments innew technologies and modernization of facilities.

EPA has taken three significant actions on NSRreform in the past two years, as follows:

• December 31, 2002: EPA issued the final rule onNSR commonly referred to as the NSRImprovement Rule

• December 31, 2002: EPA issued a proposed ruleon NSR Routine Maintenance, Repair andReplacement (RMRR)

• October 27, 2003: EPA issued a final rule, knownas the Equipment Replacement Provision, whichaddresses a part of the Routine Maintenance,Repair and Replacement exclusion.

New Source Review Improvement Rule

The December 31, 2002 NSR Improvement Rulewas issued after more than a decade of rulemaking,including two proposed rules, one supplementalproposal, numerous public hearings, and considera-tion of over 130,000 comments. The rule made fourimportant changes to the NSR program:

• Calculating Emissions Increases and Establish-ing Emissions Baseline. The new NSR improve-ment rule provides for an actual-to-projectedactual test to calculate any projected actualincreases in emissions due to a physical or opera-tional change. Under the actual-to-potential test,EPA estimated post-change emissions based uponwhat a plant would theoretically emit if it operat-ed at maximum rates 24 hours a day, year-round.This made it impossible to make certain modestchanges in a facility without triggering NSR, evenif those changes did not actually increase emis-sions. In the NSR improvement rule, a facility isto use, in its permitting analysis, the amount ofemissions actually projected to occur after theproposed change instead.

Also, to more accurately measure actual emis-sions, account for variations in business cycles,and clarify what may be a “more representative”period, the improvement rule allows facilities touse any consecutive 24-month period in the pre-

vious decade as a baseline for pre-change actualemissions, as long as all current control require-ments are taken into account. The same 24-month period must be used for all pollutants.

• Plantwide Applicability Limits. A plantwideapplicability limit (PAL) is an option that facilitiesmay choose which allows greater operational flex-ibility to manage emissions on a plantwide basis.Under a PAL, a plant owner may decide to estab-lish a plantwide emissions cap based on actualemissions (any 24-month period in the last 10years). The plant owner may then make changesto the facility or individual units without trigger-ing the NSR permitting process, as long as thefacility’s overall emissions remain below theplantwide cap.

The PAL creates an emissions “bubble” in whichemissions from the facility are treated as a whole,rather than on a unit-by-unit basis. A facilityselecting this option is required to monitor emis-sions from all of its units.

• Clean Unit Provision. Under the improvementrule, plants that install “clean units” are allowedoperational flexibility if they continue to operatewithin permitted limits. This is meant to encour-age installation of state-of-the-art air emissionscontrols. Clean units must have an NSR permitor other regulatory limit that requires use of thebest air pollution control technologies.

• Pollution Control and Prevention Projects. Thisnew applicability test governs facilities that havegone through NSR review and have installed therequired best available control technology (BACT)(in attainment areas) or met the lowest achiev-able emissions rate (LAER) (in non-attainmentareas). Under the improvement rule, these facili-ties may make changes to the Clean Unit withouttriggering further NSR requirements, if thechange does not require alterations to the emis-sions limitations or work practices requirementsof their permit and the changes would not alterany physical or operational characteristics. CleanUnit status will last for up to 10 years. EPAintends this new applicability test to protect airquality, create incentives for sources to installstate-of-the-art controls, provide flexibility, andpromote administrative efficiency.

EPA has listed environmentally beneficial tech-nologies that qualify as pollution control projects.

E-2 OBSERVATIONS ON PETROLEUM PRODUCT SUPPLY

An owner or operator installing a PollutionControl and Prevention Project (PCPP) listed byEPA automatically qualifies for the NSR exclusionif there are no adverse air quality impacts. APCPP not listed may also qualify for the exclusionif the reviewing authority determines, on a case-specific basis, that the PCPP is environmentallybeneficial when used in a particular application.In the future, EPA may add specific PCPPs to thelist through rulemaking. EPA intends this exclu-sion to encourage the use of environmentallybeneficial emissions controls and to offer opera-tional flexibility.

The NSR reform rule has been challenged incourt by environmental groups and others. Thecase is pending in the U.S. Court of Appeals for theD.C. Circuit, and is expected to be resolved some-time in 2005. In the meantime, the court has twicerefused to stay implementation of the rule, and itsprovisions are currently in force in EPA directimplementation states and in states with EPA-approved programs that have incorporated the ruleinto their state programs. States have until January2006 to incorporate the rule into their state pro-grams. The current legal challenge to this rule hasat least temporarily limited its anticipated positiveimpacts on domestic refining investment and oper-ations.

Routine Maintenance, Repair andReplacement

When EPA published the NSR Improvement Ruleon December 31, 2002, it simultaneously proposedreforms to the routine maintenance, repair andreplacement (RMRR) exclusion. Conflicting inter-pretations of the RMRR exclusion from NSR haveled to confusion as to whether basic repairs andother traditional operations at a facility might trig-ger NSR and lead to significantly higher expendi-tures than contemplated for the repairs or otherroutine actions. The refining industry has urgedEPA to consider clarifications to RMRR as well asthe broader reforms included in the NSRImprovement Rule. To date, however, EPA has final-ized only the equipment replacement portion of itsproposal to reform RMRR.

EPA clarified the RMRR exclusion rule “toincrease environmental protection and promote the

implementation of necessary repair and replace-ment projects.” The agency explained that althoughNSR excludes repairs and maintenance activitiesthat are “routine,” a complex analysis had to be usedto determine what repairs meet that standard.

The RMRR rule is intended to remove disincen-tives for owners and operators of refineries andother manufacturing facilities from undertakingRMRR activities that would improve the safety, reli-ability, and efficiency of their plants. It is especiallyimportant that refinery owners and operators haveclarity on RMRR because these facilities run at veryhigh rates of utilization (compared to other manu-facturing plants) in order to provide sufficient sup-plies of petroleum products to meet consumerdemand. Therefore, uncertainty in the applicationof the RMRR exclusion has the potential to interferewith the efficient operation of refineries and theirability to provide reliable supplies of petroleumproducts.

Equipment Replacement Provision

EPA published its Equipment ReplacementProvision (ERP) rule on October 27, 2003. The ruleis intended to provide greater regulatory certaintywithout sacrificing environmental protection andthe benefits derived from the NSR program. TheERP provides an automatic exclusion from NSRrequirements for equipment replacement meetingthe following conditions:

• they involve replacement of any existing compo-nent(s) of a process unit with component(s) thatare identical to or serve the same purpose as thereplaced component(s);

• the fixed capital cost of the replaced compo-nent(s) plus the costs of any activities that arepart of the replacement activity (e.g., labor, con-tract services, major equipment rental, and asso-ciated repair and maintenance activities) do notexceed 20% of the current replacement value ofthe process unit; and

• the replacements do not alter the basic designparameters of the process unit or cause theprocess unit to exceed any emission limitation oroperational limitation that applies to any compo-nent of the process unit and is legally enforce-able.

APPENDIX E E-3

Like the NSR Improvement Rule, the equipmentreplacement provision was challenged by someenvironmental groups and others in the U.S. Courtof Appeals for the District of Columbia. The rule hasbeen stayed pending litigation and the lawsuit isexpected to be decided in 2005. Pending resolutionof the litigation, the positive impacts of this regula-tion on domestic refining investment and opera-tions have been severely limited. In the meantime,EPA has allowed additional opportunity for publiccomment on certain issues raised in administrativepetitions for reconsideration.

Additional NSR Reforms

EPA also made informal proposals addressingdebottlenecking activities and clarifying projectaggregation requirements. Both proposals wouldfurther the efficient operation of refineries andimprove the prospect for increased production andcapacity expansion through debottlenecking andsimilar projects.

Debottlenecking occurs when one part of a facili-ty is modified in such a way that throughput inother parts of the facility increases. Under currentrules, determining whether NSR applies to suchcomplex projects is difficult and can be time con-suming. EPA has proposed that sources should gen-erally look only at the unit undergoing the changewhen calculating actual emissions associated with aproject. Unless their permitted emissions limit isexceeded or increased, emissions from unitsupstream or downstream of the unit undergoing thechange should not be considered.

Aggregation refers to the situation in the facilityin which multiple projects are implemented in ashort period of time. In such a case, a difficult andcomplex analysis must be performed to determine ifthe projects should be treated separately or togeth-er (i.e., aggregated) under NSR. EPA’s proposal

establishes two criteria that would guide this deter-mination.

EPA has yet to finalize its debottlenecking andaggregation proposals, leaving significant uncer-tainty involving NSR application to multiple projectundertakings and debottlenecking. The studygroup strongly recommends that EPA finalize theseimportant proposals.

Response to Criticisms of the NSR Reform Proposals

There is much confusion about the purpose of theNSR program and the impact of the reform regula-tions. The program, initiated in 1974, requires pre-construction review of new and modified major sta-tionary sources of air pollution that cause emissionsincreases. EPA’s NSR webpage notes that “WhenCongress established the New Source ReviewProgram, it did so with a goal of providing for eco-nomic growth while maintaining or improving airquality,” and “Over time, the NSR program hasbecome more complex and complicated, due to theevolving nature of industrial practices and changesin the regulations and EPA’s interpretation of them.In response to concerns about this, EPA has workedfor nearly 10 years to simplify the NSR program.”[http://www.epa.gov/nsr/]

To quote EPA’s Fact Sheet [http://www.epa.gov/nsr/facts.html], “NSR is one of the many tools toensure the air quality goals are met. The key provi-sions of the Clean Air Act (CAA) include programsdesigned to protect human health and the environ-ment from the harmful effects of air pollution. Thefinal rule does not alter these protections. CAA pro-grams such as the National Ambient Air QualityStandards (NAAQS), the Acid Rain program, and theNOx SIP Call will ensure that the nation’s air qualitywill continue to improve.” In short, EPA believes thatNSR reform will not have a negative impact on airquality or lead to a significant increase in emissions.

E-4 OBSERVATIONS ON PETROLEUM PRODUCT SUPPLY




AND


AND AFFORDABILITY


NATIONAL PETROLEUM COUNCIL • DECEMBER 2004

recycledpaper

NATIONAL PETROLEUM COUNCIL1625 K Street, NW

Suite 600Washington, DC 20006

OBSERVATIO

NS O

N PETRO

LEUM PRO

DUCT SUPPLY

NATIO

NA

L PETROLEUM

COUN

CIL • DECEM

BER 2004

OBSERVATIONS ON PETROLEUM PRODUCT SUPPLY · 2006-02-28 · observations on petroleum product supply...

Documents

Transcript of OBSERVATIONS ON PETROLEUM PRODUCT SUPPLY · 2006-02-28 · observations on petroleum product supply...