Download:
pdf |
pdfU.S. Department of Transportation
Pipeline and Hazardous Materials
Safety Administration
Washington, D.C. 20590
Final Regulatory Impact Analysis
Hazardous Materials: Oil Spill Response Plans and Information Sharing for
High-Hazard Flammable Trains
Docket No.: PHMSA-2014-0105 (HM-251B)
Office of Hazardous Material Safety
February 2019
Table of Contents
REVISIONS AND UPDATES TO THE PRELIMINARY REGULATORY IMPACT ANALYSIS (RIA) ........3
EXECUTIVE SUMMARY ....................................................................................................................4
OMB CIRCULAR A-4 ACCOUNTING STATEMENT ..........................................................................7
1. INTRODUCTION ............................................................................................................................9
1.1. SUMMARY OF THE FINAL RULE .............................................................................................9
1.2. DETERMINATION OF NEED ...................................................................................................15
1.3. BASELINE ANALYSIS ...........................................................................................................20
1.4. UNIVERSE OF AFFECTED ENTITIES ......................................................................................29
1.5. ALTERNATIVES ANALYSIS ...................................................................................................32
1.6. TIMEFRAME FOR THE ANALYSIS ..........................................................................................34
2. COSTS .........................................................................................................................................35
2.1. COMPREHENSIVE OSRPS ....................................................................................................35
2.2. INFORMATION SHARING ......................................................................................................47
2.3. TOTAL COSTS ......................................................................................................................52
2.4. COST UNCERTAINTY AND SENSITIVITY ANALYSIS ..............................................................55
3. BENEFITS....................................................................................................................................63
3.1. BASELINE AND ASSUMPTIONS FOR BENEFITS ASSESSMENT.................................................63
3.2. EFFECTIVENESS OF OSRPS ..................................................................................................70
3.3. BENEFIT UNCERTAINTIES AND UNQUANTIFIABLE BENEFITS ...............................................73
4. APPENDIX A: RESEARCH AND DATA AVAILABILITY ON OSRP COSTS ...................................76
5. APPENDIX B: INCIDENT RESPONSE NARRATIVES.....................................................................77
5.1. PLAINFIELD, IL ....................................................................................................................77
5.2. MONEY, MS ........................................................................................................................80
5.3. MOSIER, OR ........................................................................................................................81
5.4. WATERTOWN, WI ................................................................................................................88
5.5. CULBERTSON, MT ...............................................................................................................90
6. APPENDIX C: EXECUTIVE ORDER 13771 ..................................................................................92
REVISIONS AND UPDATES TO THE PRELIMINARY REGULATORY IMPACT ANALYSIS (RIA) ........2
Page 1
EXECUTIVE SUMMARY ....................................................................................................................3
OMB CIRCULAR A-4 ACCOUNTING STATEMENT ..........................................................................6
1. INTRODUCTION ............................................................................................................................8
1.1. SUMMARY OF THE FINAL RULE .............................................................................................8
1.2. DETERMINATION OF NEED ...................................................................................................14
1.3. BASELINE ANALYSIS ...........................................................................................................18
1.4. UNIVERSE OF AFFECTED ENTITIES ......................................................................................28
1.5. ALTERNATIVES ANALYSIS ...................................................................................................31
1.6. TIMEFRAME FOR THE ANALYSIS ..........................................................................................33
2. COSTS .........................................................................................................................................34
2.1. COMPREHENSIVE OSRPS ....................................................................................................34
2.2. INFORMATION SHARING ......................................................................................................46
2.3. TOTAL COSTS ......................................................................................................................51
2.4. COST UNCERTAINTY AND SENSITIVITY ANALYSIS ..............................................................54
3. BENEFITS....................................................................................................................................62
3.1. BASELINE AND ASSUMPTIONS FOR BENEFITS ASSESSMENT.................................................62
3.2. EFFECTIVENESS OF OSRPS ..................................................................................................69
3.3. BENEFIT UNCERTAINTIES AND UNQUANTIFIABLE BENEFITS ...............................................72
4. APPENDIX A: RESEARCH AND DATA AVAILABILITY ON OSRP COSTS ...................................75
5. APPENDIX B: INCIDENT RESPONSE NARRATIVES.....................................................................76
6.1. PLAINFIELD, IL ....................................................................................................................76
6.2. MONEY, MS ........................................................................................................................79
6.3. MOSIER, OR ........................................................................................................................80
6.4. WATERTOWN, WI ................................................................................................................87
6.5. CULBERTSON, MT ...............................................................................................................89
6. APPENDIX C: EXECUTIVE ORDER 13771 ..................................................................................91
Page 2
Revisions and Updates to the Preliminary Regulatory Impact Analysis (RIA)
Since developing the preliminary Regulatory Impact Analysis (RIA), the Pipeline and Hazardous
Materials Safety Administration (PHMSA) has reviewed the public comments submitted to the
Notice of Proposed Rulemaking (NPRM) and continued our research of issues related to the
rulemaking and analysis of regulatory impacts. Table R1 provides an overview of the significant
revisions and updates made to the preliminary RIA.
Table R1. Updates to the Preliminary RIA
Analytical Area
Effect of the
Basis and Source(s)
Revision or Update
Further
Explanation in the
RIA
Oil spill size updated to Average spill size is
incorporate subsequent smaller than that
incidents
estimated at the
NPRM stage.
Industry-reported data
as recorded in OHMS
incident database
Section 3,
“Benefits”
Number of derailments
updated to include
subsequent years of
data
Fewer derailments
per year predicted
PHMSA and FRA
incident databases.
Section 3,
“Benefits”
Plan development and
review costs
These costs are
higher than the
preliminary RIA
Public comment and
additional research,
including input from
OSRP plan writers
Section 2, “Costs”
and Appendix A
Spill response
narratives
Additional
qualitative input on
relevant oil train
derailments and
responses
EPA, FRA, and
PHMSA data; State
input; external news
sources; public
comment
Section 3,
“Benefits” and
Appendix B
Page 3
Executive Summary
PHMSA, in consultation with the Federal Railroad Administration (FRA), is issuing a final rule
that expands the applicability of comprehensive oil spill response plans (OSRPs) based on
thresholds of crude oil that apply to an entire train consist. 1 Specifically, the final rule expands
the applicability for OSRPs so that no person shall transport 20 or more loaded tank cars of liquid
petroleum oil in a continuous block or a single train carrying 35 or more loaded tank cars of liquid
petroleum oil throughout the train consist, unless that person has implemented a comprehensive
OSRP. Furthermore, this action requires railroads to share additional information about highhazard flammable trains (HHFT; defined as a train transporting 20 or more loaded tank cars of
Class 3 flammable liquid in a continuous block or a single train carrying 35 or more loaded tank
cars of Class 3 flammable liquid throughout the train consist) with State and Tribal Emergency
Response Commissions (i.e., SERCs and TERCs) to increase community preparedness and
incorporate the voluntary use of the initial boiling point test (ASTM D7900) to determine
classification and packing group for Class 3 flammable liquids. 2
Each railroad subject to the final rule is required to prepare and submit a comprehensive OSRP
that includes a plan for responding, to the maximum extent practicable, to a worst-case discharge
and to a substantial threat of such a discharge of oil. The OSRP must also be submitted to PHMSA,
where it will be reviewed and approved by PHMSA personnel. PHMSA evaluated several
alternatives for establishing the threshold values for the volume of petroleum being transported
that would require a comprehensive response plan. These alternatives are discussed in greater
detail in Section 1.5, “Alternatives Analysis.”
Table ES 1 presents the annualized costs associated with the final rule by railroad class. 3
Table ES 1. Overview of Estimated Costs (millions) (Undiscounted, 3%, 7%)
Class of
Railroad
Undiscounted+
10-Year
Annualized
3% Discount Rate
10-Year
Annualized
7% Discount Rate
10-Year
Annualized
Oil Spill Response Plans
Class I
Class II
Class III
$6.30
$4.0
$15.2
$0.6
$0.4
$1.5
$5.6
$3.6
$13.5
$0.7
$0.4
$1.6
$4.9
$3.1
$11.8
$0.7
$0.4
$1.7
$0.5
$3.7
$0.5
Information Sharing
All Railroads
$4.7
$0.5
$4.2
A train consist is considered the rolling stock, exclusive of the locomotive, making up a train.
We note that the incorporation of ASTM D7900 test, which aligns with the API RP 3000, will not replace the
currently authorized initial boiling point testing methods, but rather serve as a testing alternative if one chooses to
use that method. PHMSA believes this provides flexibility and promotes enhanced safety in transport through
accurate packing group assignment. This requirement will impose no new costs.
3
“Class I railroad”, “Class II railroad”, and “Class III railroad” mean railroad carriers that have annual carrier
operating revenues that meet the threshold amount for Class I carriers, Class II carriers, and Class III carriers,
respectively, as determined by the Surface Transportation Board under section 1201.1–1 of title 49, Code of Federal
Regulations.” (49 U.S.C., subtitle V, part A, chapter 201, subchapter I, §20101(1).)
1
2
Page 4
Cost to Government
Government
Costs
Total
$2.1
$0.2
$1.9
$0.2
$1.7
$0.2
$32.3
$3.2
$28.9
$3.4
$25.2
$3.6
+ Figures in this table may not match sums from table ES 3 exactly due to rounding error.
Table ES 2 provides a summary of the estimated per carrier cost associated with the final rule
requirements, differentiated by cost category and class of railroad. For purposes of this analysis,
PHMSA has identified several categories of costs related to the development and implementation
of a comprehensive response plan. Those costs include: plan development, submission, and
maintenance; contract fees for designating an oil spill response organization (OSRO); and training
and exercises. We also identified costs to the Federal government for plan review and approval,
which are discussed in Section 2.2.3.
Table ES 2. Undiscounted Unit Cost per Railroad by Railroad Class
Category
Plan
Development
Outlay
Period
Class of Railroad
Class I
Year 1
Class II
Class III
Class I
Plan
Maintenance
Annual
Plan
Submission
Once every 5
years
Class II
Class III
Class I
Class II
Class III
Class I
OSRO Fee
Annual
Class II
Class III
Training and
Exercises
Information
Sharing
Class I
Year 1
Class II
Class III
Unit Cost Per
Railroad
$84,666*
$28,222*
$18,815*
$8,745
$2,915
$1,943
$21
$21
$21
$40,000
$6,000
$2,500
$66,475**
$42,305**
Year 1
All Railroads
$27,803**
$7,758
Annual
All Railroads
$2,365
*This cost represents the plan development cost per railroad in the implementation year. The final
rule requires each railroad to review its plan at least every 5 years from the date of the last approval.
PHMSA estimates the recurring burden with reviewing the initial plan is half of the burden needed
to develop the initial plan.
**This cost represents training and exercise costs in the implementation year (year 1). Subsequent
years have different costs due to different frequencies applicable to the training/exercise
requirements.
Page 5
Table ES 3 below provides a summary of the undiscounted costs by year for this 10-year period
by railroad class. Table ES 4 presents costs, benefits, and breakeven analysis by provision.
Table ES 3. Summary of Undiscounted 10-Year Costs by Railroad Class or Entity
(millions)
Year
Oil Spill Response Plans
Information Sharing
Class I Class II Class III
1
2
3
4
5
6
7
8
9
10
$1.4
$0.5
$0.5
$0.5
$0.5
$1.1
$0.5
$0.5
$0.5
$0.5
$0.9
$0.3
$0.3
$0.3
$0.3
$0.7
$0.3
$0.3
$0.3
$0.3
$2.8
$1.2
$1.2
$1.3
$1.3
$2.3
$1.3
$1.3
$1.3
$1.3
All Railroads
$1.1
$0.4
$0.4
$0.4
$0.4
$0.4
$0.4
$0.4
$0.4
$0.4
Costs to Federal
Government
Total
$0.6
$0.2
$0.2
$0.2
$0.2
$0.2
$0.2
$0.2
$0.2
$0.2
$6.8
$2.6
$2.6
$2.6
$2.6
$4.8
$2.6
$2.6
$2.6
$2.6
Table ES 4. 10-Year and Annualized Costs, Benefits, and Breakeven % by Provision
Provision
Benefits (7%)
Qualitative
Breakeven
Costs (7%)
10-Year: $21.4
million
Oil Spill
Response
Planning
• Improved Communication/Defined
Command Structure may improve response
• Pre-identified Access to Equipment and
Staging of Appropriate Equipment for
Response Zones
• Railroad employees and contractors trained to
the OSRP
Cost-effective if this
requirement reduces
the consequences of
spills by 6.7%.
Annualized: $3.1
million
Information
Sharing
• Improved Communication
• Enhanced Preparedness
Cost-effective if this
requirement reduces
the consequences of
spills by 1.2%.
10-Year:
$3.7 million
Annualized: $0.53
million
IBR of
ASTM
D7900
• Regulatory Flexibility
• Enhanced Accuracy in Packing Group
Assignments
--
No cost estimated
Cost-effective if this
requirement reduces
the consequences of
spills by 7.8%.
Total
Page 6
10-Year: $25.2
million
Annualized: $3.6
million
OMB Circular A-4 Accounting Statement
Executive Orders 12866 and 13563 require agencies to regulate in the “most cost-effective
manner,” to make a “reasoned determination that the benefits of the intended regulation justify its
costs,” and to develop regulations that “impose the least burden on society.” 4 PHMSA has
determined that this rulemaking is a significant regulatory action under Executive Order 12866,
Regulatory Planning and Review, and significant under U.S. Department of Transportation (DOT)
regulatory policies and procedures because substantial public interest in this rulemaking exists.
Regulatory analyses are required to: 5
1.
2.
3.
4.
5.
6.
7.
8.
9.
Describe the need for the regulatory action.
Define the baseline.
Set the time horizon of analysis.
Identify a range of regulatory alternatives.
Identify the consequences of regulatory alternatives.
Quantify and monetize the benefits and costs.
Discount future benefits and costs.
Evaluate non-quantified and non-monetized benefits and costs.
Characterize uncertainty in benefits, costs, and net benefits.
This final regulatory impact analysis was prepared in accordance with the guidance provided by
the Office of Management and Budget’s (OMB) Circular A-4 on the development of regulatory
analysis, as required under Section 6(a)(3)(c) of Executive Order 12866 and a variety of related
authorities.
(1993, October 4). Executive Order 12866. Federal Register, 58(190), 5173b–5174.
https://www.reginfo.gov/public/jsp/Utilities/EO_Redirect.jsp. Executive Order 13563. FR, 76(14), 3821–3823.
http://www.gpo.gov/fdsys/pkg/FR-2011-01-21/pdf/2011-1385.pdf.
5
U.S. Office of Management and Budget. Circular A-4,
https://obamawhitehouse.archives.gov/omb/circulars_a004_a-4/
4
Page 7
Category
Annualized monetized benefits ($ Million)
Annualized quantified, but unmonetized, benefits
Unquantifiable benefits
Impact
Source
---
RIA
RIA
•Improved communication/defined
command structure may improve response
•Trained railroad employees and/or contract
responders
•Improved communication
•Enhanced preparedness
•Regulatory flexibility
•Enhanced accuracy in packing group
assignments
RIA
Benefits
Costs
Annualized monetized
costs ($ Million)
Annualized quantified, but unmonetized, costs
Qualitative (un-quantified) costs
Annualized monetized transfers: “on budget”
From whom to whom?
Annualized monetized transfers: “off-budget”
From whom to whom?
Transfers
Other Analyses
Effects on State, local, and/or tribal governments
Effects on wages
Effects on growth
Page 8
$3.6 (7%)
RIA
$3.4 (3%)
RIA
----------
1. Introduction
1.1. Summary of the Final Rule
Summaries of and references to the final rule’s requirements in this RIA are included for analytical
purposes only. To understand the final rule’s requirements, as codified by PHMSA in Title 49,
Code of Federal Regulations (CFR), please review the regulatory text of the final rule.
The final rule expands the applicability of comprehensive OSRPs based on thresholds of crude oil
that apply to an entire train consist. Specifically, the final rule expands the applicability for OSRPs
so that no person shall transport a single train transporting 20 or more loaded tank cars of liquid
petroleum oil in a continuous block or a single train carrying 35 or more loaded tank cars of liquid
petroleum oil throughout the train consist unless that person has implemented a comprehensive
OSRP.
This action also requires railroads to share additional information with SERCs and TERCs to
increase community preparedness, and incorporate the voluntary use of the initial boiling point
test (ASTM D7900) to determine classification and packing group for Class 3 flammable liquids.
1.1.1. Comprehensive OSRPs
Each railroad subject to the rule must prepare and submit a comprehensive OSRP that includes a
plan for responding, to the maximum extent practicable, to a worst-case discharge and to a
substantial threat of such a discharge of oil. The OSRP must be submitted to PHMSA, where it
will be reviewed and approved by PHMSA personnel.
Each comprehensive OSRP must include: 6
•
Core Plan: A core plan includes an information summary and any components that do not
change between response zones. 7 Each plan must:
o Describe the railroad’s response management system, including the functional
areas of finance, logistics, operations, planning, and command.
o Demonstrate that the railroad’s response management system uses common
terminology (e.g., the National Incident Management System) and has a
manageable span of control, a clearly defined chain of command, and trained
personnel to fill each position.
o Include an information summary as required by § 130.120.
The following text is provided as an overview of the rule and does not replace regulatory text included in the final
rule.
7
A response zone means a geographic area along applicable rail route(s), containing one or more adjacent route
segments for which the railroad is required to plan for the deployment of, and provide spill response capabilities
meeting the planning requirements of § 130.130.
6
Page 9
o Certify that the railroad reviewed the National Contingency Plan (NCP) 8 and each
applicable Area Contingency Plan (ACP) 9 and that its response plan is consistent
with the NCP and each applicable ACP, as required by §§ 130.110 and 130.115.
o Include notification procedures and a list of contacts as required in § 130.125.
o Include spill detection and mitigation procedures as required in § 130.130.
o Include response activities and resources as required in § 130.130.
o Certify that applicable employees were trained per § 130.135.
o Describe procedures to ensure equipment testing and a description of the exercise
program per § 130.140.
o Describe plan review and update procedures per § 130.145.
o Submit the plan as required by § 130.150.
•
Response Zone Appendix: For each response zone, a railroad must include a response zone
appendix to provide the information summary and any additional components of the plan
specific to the response zones. For example, each response zone appendix must provide:
o A description of the response zone, including county(s) and State(s).
o A list of route sections contained in the response zone, identified by railroad
milepost or other identifier.
o Identification of environmentally sensitive areas.
o Identification of the location where the response organization will deploy from and
the location and description of equipment required.
Regarding NCP and ACP compliance, PHMSA believes that this requirement will provide a
formal communication framework that currently may not be in place or may be informal. At a
minimum, for consistency with the NCP, a comprehensive response plan must:
•
•
•
Demonstrate a railroad’s clear understanding of the Incident Command System and Unified
Command;
Include procedures to immediately notify the National Response Center; and
Establish provisions to ensure the protection of safety at the response site.
At a minimum, for consistency with the applicable ACP (or Regional Contingency Plan (RCP) for
areas lacking an ACP), the comprehensive response plan must:
The NCP is the federal government's blueprint for responding to hazardous substance releases, as well as oil spills.
See also https://www.epa.gov/emergency-response/national-oil-and-hazardous-substances-pollution-contingencyplan-ncp-overview
9
ACPs are developed to address the specific geographic scope of the incident. Such plans enable responders to
address incidents by helping to identify and coordinate the activities of the different government agencies and
private organizations involved in the response in that geographic area. See also https://www.epa.gov/oil-spillsprevention-and-preparedness-regulations/area-contingency-planning
8
Page 10
•
•
•
•
Address the removal of a worst-case discharge, and the mitigation or prevention of the
substantial threat of a worst-case discharge, of oil;
Identify environmentally sensitive or significant areas, along the route, which could be
adversely affected by a worst-case discharge;
Describe the responsibilities of the persons involved and of Federal, state, and local
agencies in removing a discharge and in mitigating or preventing a substantial threat of a
discharge; and
Identify the procedures to obtain any required federal and state authorization for using
alternative response strategies, such as in-situ burning and/or chemical agents.
PHMSA believes NCP and ACP compliance will provide the added benefit of a formal response
framework, and communication of command structures as well as the location of environmentally
sensitive or significant areas.
In addition, the final rule would require plan holders to certify that they have identified and ensured
by contract or other means the response resources which are available to arrive onsite within 12
hours after the discovery of a worst-case discharge or the substantial threat of such a discharge.
1.1.1.1. Pre-existing Oil Spill Response Requirements
The Clean Water Act (CWA), as amended by the Oil Pollution Act of 1990 (OPA 90), directs the
President, at section 1321(j)(1)(C), to issue regulations “establishing procedures, methods, and
equipment and other requirements for equipment to prevent discharges of oil and hazardous
substances from vessels and from onshore facilities and offshore facilities, and to contain such
discharges.” 10 The CWA directs the President to issue regulations requiring owners and operators
of certain vessels and onshore and offshore oil facilities to develop, submit, update, and in some
cases obtain approval of OSRPs. Executive Order 12777 delegated this responsibility to the
Secretary of Transportation for certain transportation-related facilities, and the Secretary delegated
this responsibility to DOT’s Research and Special Programs Administration (RSPA), PHMSA’s
predecessor agency. 11
On June 17, 1996, RSPA published a final rule at 49 CFR part 130 to carry out PHMSA’s delegated
authority under the CWA for motor carriers and railroads. 12 This rulemaking adopted general spill
response planning and response plan implementation requirements intended to prevent and contain
spills of oil during transportation.
Title 49 CFR part 130 requires a basic OSRP for oil shipments in a packaging having a capacity
of 3,500 gallons or more, which requires the preparation of a written plan that (1) “sets forth the
manner of response to discharges . . .,” (2) “takes into account the maximum potential discharge
of the contents from the packaging,” (3) “identifies private personnel and equipment available to
CWA § 311(j)(1)(C). See also 33 U.S.C. § 1321(j)(5); and CWA § (j)(5).
Executive Order 12777. Implementation of Section 311 of the Federal Water Pollution Control Act of October 18,
1972, as amended, and the Oil Pollution Act of 1990. Federal Register, 56(204), 54757—54770. Retrieved from
https://www.gpo.gov/fdsys/pkg/FR-1991-10-22/pdf/FR-1991-10-22.pdf
12
(1996, June 17). Oil Spill Prevention and Response Plans. Federal Register, 61(117), 30533–30543. Retrieved
from http://www.gpo.gov/fdsys/pkg/FR-1996-06-17/pdf/96-14611.pdf
10
11
Page 11
respond to a discharge,” and (4) “identifies the appropriate persons and agencies (including their
telephone numbers) to be contacted in regard to such a discharge and its handling, including the
National Response Center.” 13 The requirements for a basic response were issued as a prevention
and containment rule pursuant to § 1321(j)(1)(C) of the CWA.
Beyond a basic plan, the 1996 RSPA rulemaking specified that a comprehensive OSRP is required
for oil shipments in a package containing more than 42,000 gallons (1,000 barrels). Other
requirements from that rule specified that a comprehensive plan must: (1) include everything
required in the basic OSRP, (2) be consistent with the National Contingency Plan and Area
Contingency Plans, (3) identify a qualified individual with authority to implement removal and
facilitate communication between federal officials and spill response personnel, (4) identify and
ensure by contract response equipment and personnel to remove a worst case discharge, (5)
describe training equipment testing, and drills, and (6) be submitted to FRA. 14 The comprehensive
OSRP addresses minimum requirements for a plan specified by 33 U.S.C. 1321(j)(5)(D). The
1996 final rule accepted nationwide, regional, or other generic plans meeting the requirements.
The plan holder was not required to account for different response locations. Please note, this final
rule requires submission to PHMSA—not FRA—for review and approval of comprehensive plans,
in addition to changing the applicability and other comprehensive plan requirements.
1.1.1.2. Changes under the Final Rule (Basic vs. Comprehensive Plans)
Currently, most—if not all—of the rail community transporting oil, including crude oil transported
as a hazardous material, is subject to the basic OSRP requirement of 49 CFR 130.31(a), since
most—if not all—rail tank cars being used to transport crude oil have a capacity greater than 3,500
gallons. However, a comprehensive OSRP for shipment of oil is required only when the quantity
of oil is greater than 42,000 gallons per tank car. Accordingly, the number of railroads required
to have a comprehensive OSRP is much lower, or possibly nonexistent, because a very limited
number of rail tank cars in use would be able to transport a volume of 42,000 gallons in a car. 15
This final rule expands the applicability of comprehensive OSRPs to railroads transporting a single
train of 20 or more loaded tank cars of liquid petroleum oil in a continuous block or a single train
carrying 35 or more loaded tank cars of liquid petroleum oil throughout the train consist. Railroads
meeting this new applicability would need to develop, maintain, and implement a comprehensive
plan as described in this final rule.
1.1.2. Information Sharing
On May 7, 2014, DOT issued an Emergency Restriction/Prohibition Order in Docket No. DOTOST-2014-0067 (Order), 16 which required each railroad transporting 1 million gallons or more of
Bakken crude oil in a single train in commerce within the U.S. to provide certain information in
writing to the SERC for each state in which it operates such a train. Later that year, PHMSA
49 CFR 130.31(a)
49 CFR 130.31(b)
15
The 2014 Association of American Railroads’ (AAR) Universal Machine Language Equipment Register numbers
showed five tank cars listed with a capacity equal to or greater than 42,000 gallons, and none of these cars were
being used to transport oil or petroleum products.
16
Available at: https://www.regulations.gov/document?D=DOT-OST-2014-0067-0001
13
14
Page 12
proposed in HM-251 to codify and clarify the requirements of the Order in the Hazardous Materials
Regulations (HMR), and requested public comment on the various facets of that proposal (79 FR
45015, Aug. 1, 2014). Unlike many other proposals in the August 1, 2014 NPRM, the proposed
notification requirements were specific to a single train that contains 1 million gallons or more of
UN1267, petroleum crude oil, Class 3, sourced from the Bakken shale. In the final rule, HM-251
(80 FR 26643, May 8, 2015), PHMSA did not adopt the separate notification requirements
proposed in the HM-251 NPRM and instead relied on the expansion of the existing route analysis
and consultation requirements of § 172.820 to include HHFTs.
On December 4, 2015, President Obama signed into law the Fixing America’s Surface
Transportation Act of 2015 (“FAST Act”). The FAST Act includes the “Hazardous Materials
Transportation Safety Improvement Act of 2015” at Sections 7001 through 7311, which provides
direction for PHMSA’s hazardous materials safety program. Section 7302 directs the Secretary to
issue regulations that require Class I railroads to provide State Emergency Response Commissions
(SERCs) advanced notification of HHFTs traveling through their respective jurisdictions. Section
7302 requires Class I railroads to provide advanced notification and information on HHFTs to
SERCs consistent with the notification requirements in the Secretary’s May 2014 Emergency
Order in docket number DOT-OST-2014-0067. Section 7302 further requires SERCs receiving
this advanced notification to provide the information to law enforcement and emergency response
agencies upon request and directs the Secretary to establish security and confidentiality protections
for the electronic train consist information and advanced notification information required by
Section 7302.
In response to the FAST Act and to the public interest and feedback the Department previously
received related to its May 7, 2014 Emergency Order, this final rule adds a new section, 49 CFR
174.312, with information sharing requirements. As directed by the FAST Act, the information
requirements are generally consistent with the Order, but broaden the scope of trains covered by
the requirement. Consistent with the FAST Act, the regulation expands the notification
requirement to apply to all HHFTs, as defined in the HHFT final rule, not just trains transporting
1 million or more gallons of Bakken crude oil, and requires railroads to provide updates to the
notification for changes in volume greater than 25 percent. In addition, it would require railroads
to provide the required information to both SERCs and Tribal Emergency Response Commissions
(TERCs), or other appropriate state-delegated agencies. Finally, a railroad operating a train subject
to the Comprehensive Oil Spill Response Plan requirements would also need to provide the
relevant SERCs, TERCs, or other appropriate state or tribal agencies with the contact information
for qualified individuals specified in the plan.
In addition, § 174.312 requires a rail carrier operating an HHFT to provide on-going notifications
to each SERC, TERC, or other appropriate state-delegated entities meeting the following
requirements:
• A reasonable estimate of the number of HHFTs that the railroad expects to operate each
week, through each county within the State or through each tribal jurisdiction;
• The routes over which the HHFTs will operate;
•
A description of the hazardous material being transported and all applicable emergency
response information required by subparts C and G of part 172 of the HMR;
Page 13
•
HHFT point of contact: at least one point of contact at the railroad (including name, title,
phone number, and address) related to the railroad’s transportation of affected trains;
•
If a route is additionally subject to the comprehensive spill plan requirements, the
notification must include a description of the response zones (including counties and states)
and contact information for the qualified individual and alternate;
•
Railroads must update the notifications for changes in volume greater than 25 percent.
•
Notifications and updates may be transmitted electronically or by hard copy;
•
If the disclosure includes information that railroads believe is security sensitive or
proprietary and exempt from public disclosure, the railroads should indicate that in the
notification.
•
Each point of contact must be clearly identified by name or title and role (e.g., qualified
individual, HHFT point of contact) in association with the telephone number. One point
of contact may fulfill multiple roles; and
•
Copies of HHFT notifications made must be made available to the Department of
Transportation upon request.
1.1.3. Boiling Point
The initial boiling point (IBP) test (ASTM D7900) is not currently aligned with the testing
requirements authorized in the HMR, forcing shippers to continue to use the testing methods
authorized in § 173.121(a)(2). This misalignment results in a situation in which an industry best
practice for testing of crude oil (ASTM D7900 for initial boiling point) that was developed in
concert with PHMSA is not authorized by the HMR. Therefore, for initial boiling point
determination, PHMSA is proposing to incorporate by reference the ASTM D7900 test method
identified within American Petroleum Institute Recommended Practice (API RP) 3000, thus
permitting an industry best practice for testing Class 3 Packing Group (PG) assignments.
The API RP 3000 provides guidance on the material characterization, transport classification, and
quantity measurement for overfill prevention of petroleum crude oil for the loading of rail tank
cars. With regard to classification, this recommended practice concluded that for crude oil
containing any significant amount of light end components (e.g., methanes, ethane, propane,
butane and iso-butane), the recommended best practice is to test using American Society for
Testing and Materials (ASTM) D7900. The ASTM D7900 differs from the boiling point tests
currently in the HMR, because it is the only test which ensures a minimal loss of light ends.
PHMSA notes that the incorporation of ASTM D7900, which aligns with the API RP 3000, will
not replace the currently authorized testing methods, but rather will serve as a testing alternative
if one chooses to use that method. PHMSA believes that this provides flexibility and promotes
enhanced safety in transport through accurate PG assignment. This provision is voluntary and
would not impose any costs to industry.
Page 14
1.2. Determination of Need
1.2.1 Background
PHMSA’s mission is to protect people and the environment by advancing the safe transportation
of energy and other hazardous materials that are essential to our daily lives. To do this, PHMSA
establishes national policy, sets and enforces standards, educates stakeholders and the public,
conducts research to prevent accidents, and prepares the public and first responders to reduce the
consequences if accidents do occur.
The United States is now a global leader in crude oil production. U.S. Energy Information
Administration (EIA) projects that U.S. crude oil production will surpass the 9.6 million barrels
per day (b/d) record set in 1970 and plateau between 11.5 million b/d and 11.9 million b/d 17,
representing a total growth between 20 percent and 24 percent increase until 2040, or 0.7 percent
to 0.8 average annual growth. Thus, the final rule is necessary due to the expansion in U.S. energy
production, which has led to significant challenges in the transportation system. Expansion in oil
production has led to increasing volumes of product transported to refineries and other transportrelated facilities, such as transloading facilities. With a growing domestic supply, rail
transportation remains a flexible alternative to transportation by pipeline or vessel. While annual
crude-by-rail volumes have fallen from their peak in 2014 (382,034 thousand barrels), they were
175,701 thousand barrels in 2016 and 139,805 thousand barrels in 2017 as opposed to 23,788
thousand barrels in 2010 and 42,370 thousand barrels in 2011, or a nearly fivefold (487 percent)
increase from 2010 to 2017. 18 In 2018 through July, EIA reported a 10 percent increase in volumes
of crude by rail estimates over 2017.
U.S. EIA. Energy production (Reference case), pp. 19-20. Available at:
https://www.eia.gov/outlooks/aeo/pdf/AEO2018.pdf
18
U.S. EIA. Total Crude by Rail. Available at:
https://www.eia.gov/dnav/pet/hist/LeafHandler.ashx?n=pet&s=esm_epc0_rail_zamn-zamn_mbbl&f=a
17
Page 15
Figure 1 provides annual U.S. rail movements of crude oil from 2010 to 2017.
Figure 1. Annual Rail Movements of Crude Oil by Rail 19
This data is for the lower 48 states only. See
https://www.eia.gov/dnav/pet/hist/LeafHandler.ashx?n=PET&s=ESM_EPC0_RAIL_ZAMN-ZAMN_MBBL&f=M
19
Page 16
Rail accidents have tracked changes in rail shipments of crude oil—rising when rail shipments
increase in volume and falling when crude oil volumes fall. Figure 2 shows this rise in carloads
and derailments in recent years. 20
700,000
7
600,000
6
500,000
5
400,000
4
300,000
3
200,000
2
100,000
1
HHFT Derailments
Originating Carloads Per Year
Figure 2. Carloads of Crude Oil Shipped and Derailments, 2008–2016
0
0
2008
2009
2010
2011
2012
2013
2014
2015
2016
Year
Crude Carloads
HHFT Crude Derailments
Sources and notes: Vertical Axes present data on different scales. Originating Class I Carloads for 2008–2016 obtained
from the Surface Transportation Board Waybill sample. Derailments are from the PHMSA and FRA Incident Report
databases.
Based on these train accidents, the expectation of continued domestic crude oil production and
transportation, and the number of train accidents involving crude oil, PHMSA maintains that
improved oil spill response planning is essential to protecting people and the environment from
the risks of derailments involving large quantities of petroleum oil.
PHMSA has identified several recent derailments to illustrate the circumstances and consequences
of derailments involving petroleum oil transported in higher-risk train configurations: Plainfield,
IL (June 2017); Money, MS (April 2017); Mosier, OR (June 2016); Watertown, WI (November
2015); Culbertson, MT (July 2015); Heimdal, ND (May 2015); Galena, IL (March 2015); Mt.
Carbon, WV (February 2015); La Salle, CO (May 2014); Lynchburg, VA (April 2014);
Vandergrift, PA (February 2014); New Augusta, MS (January 2014); Casselton, ND (December
2013); Aliceville, AL (November 2013); and Parkers Prairie, MN (March 2013). The Heimdal,
20
Surface Transportation Board Waybill Sample and PHMSA Incident Report Database
Page 17
ND derailment and prior derailments were discussed in the NPRM and preliminary RIA. For
information on these derailments, please see the NPRM, preliminary RIA, and associated
regulatory docket (PHMSA-2014-0105). We discuss more recent derailments in this final RIA,
see Appendix B.
Separate from derailments occurring in the United States, PHMSA also considered the July 2013
derailment in Lac-Mégantic, Quebec. In response to this derailment, the NTSB issued Safety
Recommendation R-14-5, which recommended that PHMSA revise the spill response planning
thresholds prescribed in 49 CFR part 130 to require comprehensive OSRPs to effectively provide
for the carriers’ ability to respond to worst-case discharges resulting from accidents involving unit
trains or blocks of tank cars transporting petroleum products. 21 Thus, the revisions included in
this final rule were developed in response to this NTSB recommendation, as well as recent
derailments.
1.2.2
Market Failure
With respect to the role of insurance and liability in internalizing costs of oil spills, a market failure
at issue is that the shippers and rail companies are not insured against the full liability of the
consequences of incidents involving hazardous materials, including oil. Even with adequate
insurance, it is unclear whether full compensation for the consequences of events that may result
in severe injury or death is possible, resulting in external costs that go unrecompensed regardless
of the insurance carried by the railroad. Incidents involving severe negligence on the part of the
carrier may therefore result in harm that goes uncompensated. In addition, in the case of a
catastrophic event, a railroad company may become insolvent due to oil spill liabilities exceeding
its available capital and ability to pay. Despite potentially being a responsible party, the railroad
could “escape liability” in this way and costs would become social externalities. Additionally, to
process a claim to insurance compensation, those harmed must demonstrate real harm and value
lost. In the case of damage to the environment, the actual monetary value of lost or damaged assets
is difficult to determine. Relatedly, high information and coordination costs may pose a barrier to
aggregating claims of harm because the harm done to any affected entities may be lower than the
cost of obtaining evidence of harm, even though they may be substantial in the aggregate. As a
result, derailments involving petroleum oil may impose externalities or negative consequences.
Although the Agency does not believe that OSRPs would prevent or reduce the probability of a
derailment that may result in release of crude oil into the environment, such planning and
preparation can result in a more effective response after an oil spill occurs, and hence mitigate or
reduce the negative consequences.
As further evidence of potential market failure, PHMSA notes the typical limits of insurance
coverage for rail liabilities. In the RIA for the HM-251 rulemaking and PHMSA’s Rail Liability
Study, we indicated that, among Class I railroads, a self-insured retention of $25 million is
common, though it can be as much as $50 million, especially when toxic if inhaled (TIH) material
21
National Transportation Safety Board. (2014, January 21). Safety Recommendation R-14-4 through -6. Retrieved
from http://www.ntsb.gov/safety/safety-recs/recletters/R-14-004-006.pdf.
Page 18
is involved. 22 On the other hand, smaller regional and short line carriers, such as Class II and
Class III railroads, typically maintain retention levels well below $25 million, as they usually do
not have the cash flow to support substantial self-insurance levels. 23 Further, the maximum
coverage available in the commercial rail insurance market appears to be in the range of $1–1.5
billion per carrier, per incident. 24
While this level of insurance may be sufficient for the vast majority of accidents, it is inadequate
to cover some higher-consequence events. For example, the rail carrier responsible for the incident
at Lac-Mégantic, Quebec in July 2013 was covered for a maximum of $25 million in insurance
liability and had to declare bankruptcy because that coverage and the company’s remaining capital
combined were insufficient to pay for more than a fraction of the harm that was caused. Therefore,
rail carriers and shippers had insufficient coverage to bear the entire cost of “making whole” those
affected when an incident involving crude and ethanol shipment by rail occurred. Further, some
damages are unlikely to lead to liability, including any damages to the American public’s non-use
values of an area where a release occurs, as well as small amounts of per capita damages that can
be large overall if they affect a large number of people. For instance, if a release causes an
evacuation, the affected groups may not suffer enough harm to overcome the fixed costs of
litigating that harm, and coordination among the people affected may be difficult.
Another issue is that shippers, though responsible for buying or leasing the tank cars in which these
products are shipped and loading the material into the tank cars, do not generally bear any liability
for an incident once a rail carrier has accepted shipment, and rail carriers cannot refuse shipments
because of their common carrier obligation. 25 In addition, the rates that rail carriers can charge to
move these commodities are generally negotiated between the shipper and the carrier on a contract
basis and regulated by the Surface Transportation Board. Shippers do not generally bear liability
for their shipments while in transport and thus may lack an appropriate full incentive to ensure that
the tanks cars are adequate to appropriately address the level of risk.
The FAST Act required DOT to initiate a study on the levels and structure of insurance for railroad
carriers transporting hazardous materials, and submit a report to Congress on the results of the
study and recommendations for addressing liability issues with rail transportation of hazardous
materials. This report was completed and transmitted to Congress in November 2017. It is also
available at PHMSA’s website. 26
Pipeline and Hazardous Materials Safety Administration (PHMSA), “Hazardous Materials by Rail Liability
Study: Report to Congress,” November 29, 2017, available at:
https://www.phmsa.dot.gov/sites/phmsa.dot.gov/files/docs/news/57011/report-congress-hazardous-materials-railliability-study-nov-2017_1.pdf [hereinafter Rail Liability Study]
23
http://www.dot.gov/office-policy/transportation-hazardous-materials-insurance-security-and-safety-costs (section
3.4 Rail Liability and Insurance Controversy).
24
Rail Liability Study, pg. 31
25
See 49 U.S.C. § 11101. Available at: https://www.law.cornell.edu/uscode/text/49/11101
26
Rail Liability Study, available at: https://www.phmsa.dot.gov/news/hazardous-materials-rail-liability-study
22
Page 19
1.3. Baseline Analysis
This section details the regulatory baseline as well as the planning and response baseline related
to oil spill response planning. The table below provides a high-level summary of the anticipated
changes to baseline conditions if this rule were to become effective.
Table 1. Baseline Analysis of Regulatory Provisions
Provision
Oil Spill
Response Plans
(OSRPs)
No Action Baseline
Final Rule
Potential Change between the Baseline
and Final Rule
• Railroads transporting 3,500
gallons of oil in a tank car must
meet the requirements for “basic”
plans as set forth by 49 CFR part
130
• Based on current operating
procedures, railroads are not
required to meet the requirements
for “comprehensive” plans set
forth by 49 CFR part 130 and
thus plans are not aligned with
OPA 90 and 33 U.S.C. 1321
(j)(5)(D)
• Current plans lack alignment with
the larger Federal oil spill
regulatory regime (e.g., USCG,
BSEE, and EPA)
• Railroads currently plan for
derailments involving petroleum
oil through voluntary internal
planning, use of basic plans under
part 130, and non-regulatory
cooperation with OSROs and
non-OSRO contractors
• The current response inventory
available to railroads suggests
that OSRO resources are
available to arrive on site within
12 hours and to respond to a
WCD 27
• No change to basic plan
requirements under part 130
• Railroads transporting HHFT of
liquid petroleum oil need to
develop a comprehensive plan in
accordance with part 130
• Railroads’ comprehensive plans
will align with OPA 90 and 33
U.S.C. 1321 (j)(5)(D)
• Railroad’s comprehensive plans
will be more aligned with plans
used in other industries and
overseen by other Federal
agencies
• Railroads will determine and
include in their plan a WCD
planning volume for each
response zone based on their
operations and the final rule
definition
• Railroads will be required to
certify response resources are
available to respond to a WCD by
contract or other means
• Railroads will implement a
training program consistent with
or equivalent to PREP guidelines
• Railroads will submit
comprehensive plans and updates
to FRA for approval
• The quantity and type of response
resources available for response to
a WCD will be cited in the
comprehensive plan or will meet
OSRO classification standards
• 73 railroads will develop, maintain, and
submit to PHMSA a comprehensive plan
meeting the requirements. This includes a
requirement that railroads ensure by
contract or other means necessary response
resources to respond to a WCD. 28 Further,
this requires railroads to conduct training,
exercises, and drills in accordance with their
comprehensive plan and PREP guidelines
• The chain of events in a response are
anticipated to be better coordinated and
communicated, potentially quicker, and
better integrated into the Federal oil spill
response regulatory regime
• Responses to derailments involving
petroleum oil will implement
comprehensive plans, reducing damages
resulting from oil spills
27
Based on PHMSA’s analysis of available OSROs found in the USCG OSRO Response Resources Inventory
System and publicly-available Office of Pipeline Safety pipeline operator OSRPs. See the discussion in this
“Baseline Analysis” section.
28
Due to data uncertainty, PHMSA’s analysis assumes railroads do not have contracts in place. Costs are estimated
for all affected entities and do not account for the current level of compliance.
Page 20
Provision
Information
Sharing
IBR of ASTM
D7900
No Action Baseline
Final Rule
Potential Change between the Baseline
and Final Rule
• Railroads must notify SERCs and
TERCs of expected movements
of 1 million or more gallons of
Bakken crude through the
SERCs’ respective jurisdictions
• May 2014 DOT Emergency
Order (E.O.) remains in full force
and effect 29
• FAST Act instructs the Secretary
to issue regulations that require
each Class I railroad to provide
SERCs with information
consistent with the May 2014
DOT E.O.
• All railroads transporting HHFT
must notify SERCs and TERCs or
another appropriate statedelegated agency
• Responds to FAST Act mandate
by codifying requirements
consistent with the DOT E.O.
• Notifications are required for flammable
liquids in addition to Bakken crude oil (e.g.,
ethanol, crude oil sourced from other oilproducing regions)
• Notifications are required for railroads
transporting a HHFT, irrespective of the
1 million-gallon criteria from the DOT E.O.
• PHMSA estimates that 7 Class I and 40
Class II and III railroads were covered by
the DOT E.O., while an additional 131
railroads were not covered and will develop
SERC notifications under the rule.
• Shippers use the testing methods
authorized in § 173.121(a)(2) of
the HMR
• The initial boiling point (IBP) test
and practice recommended by
industry is ASTM D7900, within
API RP 3000, and cannot be used
for compliance with the HMR
• ASTM D7900 is incorporated into
the HMR and serves as a testing
alternative if a shipper chooses to
use that method
• Shippers have options in conducting an
initial boiling point test (regulatory
flexibility)
• Aligns with industry best practices
• Use of ASTM D7900 is voluntary and
would not impose any costs to industry
1.3.1. Current Regulatory Baseline
PHMSA views the regulatory baseline as the existing regulatory framework found in 49 CFR part
130, which provides the oil spill response requirements that PHMSA has already promulgated and
currently oversees. These regulations would continue to be in effect in the absence of this
rulemaking. Railroads transporting petroleum oil in tank cars that exceed 3,500-gallon capacity
are currently required to have a “written basic plan” (“basic plan”) meeting the response planning
requirements set forth for basic plans in 49 CFR 130.31(a). Based on this tank car threshold,
PHMSA assumes all railroads transporting petroleum oil have a basic plan.
Currently, 49 CFR part 130 includes requirements for comprehensive plans. A comprehensive
plan is required for railroads transporting petroleum oil in tank cars exceeding 42,000-gallon
capacity. In practice, railroads transporting petroleum oil do not currently need to fulfill these
comprehensive plan requirements because the tank cars used to transport petroleum oil are less
than 42,000-gallon capacity, and thus are not applicable according to the scope of the part found
in 49 CFR 130.2. The current 49 CFR part 130 requirements establish different regulatory
standards for response planning based on the existing applicability of the part and the extent of the
differences between the basic and comprehensive plan requirements. It is important to note that
the current thresholds are based on size of a containment vessel and not the amount of material
transported in a single train set or consist.
https://www.phmsa.dot.gov/news/phmsa-notice-regarding-emergency-response-notifications-shipmentspetroleum-crude-oil-rail-0
29
Page 21
Basic plans do not meet all the statutory requirements for response plans under the Oil Pollution
Act of 1990 found in 33 U.S.C. 1321(j)(5)(D)(i) through (vi), whereas the comprehensive plan
requirements are written in parallel to these statutory requirements. In the final rule, PHMSA
maintains that railroads transporting large quantities of petroleum oil in a single train set must meet
the statutory requirements for a response plan as prescribed in 33 U.S.C. 1321(j)(5)(D)(i) through
(vi) and the comprehensive plan requirements in 49 CFR part 130, which, according to commenter
input, must also be amended for clarification and further alignment with statute. PHMSA also
seeks to align 49 CFR part 130 with the interagency Federal regulatory framework and industry
best practices pertinent to oil spill response planning. For additional explanation of the regulatory
baseline, please refer to the “Background and Purpose” section of the final rule.
1.3.2. Current Planning and Response Baseline
The current regulations described in Section 1.3.1 of this RIA provide a minimum standard for
regulatory compliance. Based on comments to the ANPRM, NPRM, anecdotal evidence, and
additional research, PHMSA anticipates that many railroads are likely to meet the current basic
plan requirements required under 49 CFR part 130 and many may exceed these minimum
standards. Given current voluntary industry action with regard to oil spill response planning, this
section anticipates the planning and response changes that might be imposed by this rule. As we
anticipate that many railroads may voluntarily exceed the minimum standard for compliance, the
change to the current planning and response baseline is likely to be less than the change in the
regulatory baseline. To explore this possibility further, PHMSA analyzed the extent of the existing
Oil Spill Removal Organization (OSRO) network to develop a better understanding of the existing
coverage relative to potential derailment areas. This gives a better picture of the rule’s potential
impact on oil spill planning and response than the regulatory baseline.
We understand the OSRO network to be comprised of a variety of organizations that can provide
response resources and respond to an oil spill in the event of a derailment involving petroleum oil.
In this final rule, we define an OSRO as an entity that provides response resources. This definition
would include U.S. Coast Guard (USCG)-certified Oil Spill Removal Organizations. PHMSA
thus uses the acronym, OSRO, in a general sense as defined in the final rule, but also uses “USCG
OSRO” or “USCG-certified OSRO” to refer to an organization that has voluntarily chosen to
participate in the USCG’s classification program and has met USCG standards to be classified as
an OSRO. 30 USCG OSRO classification standards are found in Appendix B of 33 CFR part 154,
for marine transportation-related facilities, and Appendix B of 33 CFR part 155, for vessels. In
addition, USCG OSRO classification standards are discussed in the April 24, 2013 USCG Policy
Letter. 31
PHMSA notes that USCG has more than one classification standard for USCG OSROs. As such,
the USCG classification is not a “one-size-fits-all” approach, but allows for flexibility depending
on the applicable operating area as well as different tiers based on the quantity, type, and
availability of response resources. The USCG-designated operating areas are: Rivers/Canals;
Great Lakes; Inland: Nearshore; Offshore; and Open Ocean. These operating areas are defined in
33 CFR 154.1020. The different tiers that USCG designates are: Maximum Most Probable
30
31
http://www.americansalvage.org/email-files/MER-Policy-Letter-03-13-OSRO-Classification_signed.pdf
Ibid.
Page 22
Discharge (MMPD); Worst Case Discharge (WCD) 1; WCD 2; and WCD 3. The different tiers
correspond to different criteria for response resources, availability (response timeframes), and oil
recovery capability.
PHMSA’s Office of Pipeline Safety already uses the USCG OSRO classification standards and
criteria as part of their evaluation and approval of pipeline operator OSRPs. As such, PHMSA’s
Office of Hazardous Materials Safety (OHMS) worked with OPS to learn about their practices and
approval of pipeline operator plans. OPS referred OHMS to the USCG OSRO classification
standard of “Rivers/Canals” at the tier of WCD 1 as the most applicable standard for pipelines.32
PHMSA OHMS seeks to align our practices and regulation, whenever reasonable, with OPS. For
evaluation of response resource availability and capability, the standard comprising Rivers/Canals
and WCD 1 appears to be appropriate for railroads’ comprehensive plans. OPS’ use of USCG’s
Rivers/Canals OSRO standard lends support to the suitability of the USCG OSRO classification
system for the rail context and demonstrates that the classification system is applicable to both
water- and land-based spills. We note that there may be differences between pipeline and rail spill
contexts, but the Rivers/Canals OSRO standard was recommended over the other standards for our
analytical purposes (e.g., Inland: Nearshore).
USCG OSRO classification is important because, in USCG’s program, the holder of an oil spill
response plan can list an USCG OSRO in their plan rather than provide an extensive, detailed list
of response resources. For the comprehensive plan requirements, PHMSA has allowed regulatory
flexibility akin to the USCG program, such that a plan holder (railroad) citing a USCG-certified
OSRO is not required to provide a detailed listing of equipment, supplies, and personnel within
the comprehensive plan. PHMSA expects railroads affected by the final rule to primarily contract
with USCG-certified OSROs and to utilize this regulatory flexibility. PHMSA anticipates that
railroads may also contract with other (non-USCG-certified) OSROs to fulfill other, specific needs
specified within their core plans or within a given response zone. We emphasize, however, that
the final rule does not require a railroad to use a USCG-certified OSRO; the rule requires a standard
for response planning and allows the plan holder/operator discretion and flexibility in determining
the best way to meet that standard.
PHMSA also notes we have not required any specific regulation that would prevent the “sharing”
of OSROs between railroads. Whether an OSRO services more than one railroad in an area, or
within a response zone, is a private decision for the interested railroads to consider. However, it
is our understanding that USCG classification guidelines distinguish between “dedicated” and
“non-dedicated” resources. If resources are “non-dedicated,” this may affect the availability of
response resources and response times, and this could in turn affect the OSROs classification. 33
In light of this discussion, PHMSA assumes for the purposes of this analysis that a railroad meeting
the rule’s applicability will contract with a USCG-certified OSRO to comply with the response
PHMSA notes that the operating areas definitions can be specific and technical and the reader is advised to refer
to them in 33 CFR 154.1020. The “Rivers/Canals” operating area is a subset of the “Inland” operating area.
According to 33 CFR part 154, “Rivers and canals” means a body of water confined within the inland area,
including the Intracoastal Waterways and other waterways artificially created for navigation, that has a project depth
of 12 feet or less.
33
http://www.americansalvage.org/email-files/MER-Policy-Letter-03-13-OSRO-Classification_signed.pdf
32
Page 23
requirements of the final rule. USCG maintains a publicly-available registry of USCG-certified
OSROs, known as the Response Resource Inventory System (RRIS). 34 Using the RRIS and
publicly-available OSRPs submitted to the Office of Pipeline Safety, PHMSA analyzed the
network of USCG-certified OSROs alongside the rail network and thus the coverage of potential
derailment areas that currently exists (and would likely be sustained in the absence of
rulemaking). 35 These two publicly-available sources provided sufficient information to identify
the name and location of existing, USCG-certified OSROs and to map the network of response
coverage across the United States. PHMSA is aware that OSRO resources are mobilized from
master service agreements and from locations across USCG zones. However, PHMSA has chosen
to approximate the focal point of their coverage at the location given in the USCG OSRO RRIS
because the locations of specific response equipment are not widely available.
Given the USCG-certified OSRO’s location, we estimated the extent of the OSRO’s response
coverage by applying the assumption that the OSRO could travel to the site of a derailment at 35
miles per hour (mph). This assumption is consistent with the standards of the USCG OSRO
classification guidelines and as such, it is also consistent with the standards applied in PHMSA’s
approval of pipeline operator OSRPs. 36 Therefore, for response resources traveling by land, the
comprehensive OSRP will only be approved if all the necessary response resources are staged
within 420 miles of any point in the response zone. This requirement is similar to existing Federal
OSRP requirements under the USCG. To ensure response resources are adequately placed, USCG
gauges whether response resources can make it to a given location by assuming response resources
can travel 35 mph. We feel this 35 mph appropriately accounts for slower speeds due to
impediments or traffic, since the actual mechanical ability of land vehicles would often be greater
than 35 mph. We have incorporated this standard into the response planning requirements of this
rule.
As seen in the map below, the existing OSRO network affords a level of response coverage that
extends to the entirety of the continental U.S. PHMSA believes this lends support to our
assumption that all potential rail routes transporting petroleum oil in the continental U.S. could be
serviced by a USCG-certified OSRO in the event of a derailment within 12 hours or less, as
required in this final rule. Of course, it is likely that response resources arrive onsite well before
12 hours after the derailment, but this is the cutoff for inclusion of response resources in the
railroad’s response zone.
https://cgrri.uscg.mil/UserReports/WebClassificationReport.aspx
For publicly-available OPS plans, please see: http://www.phmsa.dot.gov/pipeline/oil-spill-response-plan/. For the
USCG OSRO Response Resource Inventory System, please see:
https://cgrri.uscg.mil/UserReports/WebClassificationReport.aspx
36
http://www.americansalvage.org/email-files/MER-Policy-Letter-03-13-OSRO-Classification_signed.pdf
34
35
Page 24
Data source and notes: OSRO data compiled using publicly-available OPS plans and USCG OSRO RRIS. * A
response timeframe of 12 hours or less assumes travel speed of 35 miles per hour “as the crow flies.” This is
equivalent to a distance of 420 miles. ** Rail network from US Census Bureau's 2015 national rail network,
available publicly at: https://catalog.data.gov/dataset/tiger-line-shapefile-2015-nation-u-s-rails-national-shapefile.
PHMSA also mapped the OSRO network with 6-hour halos, and while there was considerable
coverage, there were large notable areas of land that were not covered. These uncovered areas
coincided with portions of the rail network. See the following map with 6-hour halos for
comparison. The light grey portions of the map do not have USCG-certified OSRO coverage
according to our assumptions and available data.
Page 25
Data source and notes: OSRO data compiled using publicly-available OPS plans and USCG OSRO RRIS. *
Mapping the response timeframe of 6 hours or less assumes travel speed of 35 miles per hour “as the crow flies.”
This is equivalent to a distance of 210 miles. ** Rail network from US Census Bureau's 2015 national rail network,
available publicly at: https://catalog.data.gov/dataset/tiger-line-shapefile-2015-nation-u-s-rails-national-shapefile.
PHMSA expects that the existing OSRO network is sufficient to meet the final rule’s standards for
response planning; as such, PHMSA does not expect that the rule would result in incremental
equipment costs to railroads. PHMSA notes that, even if a railroad were considering expanding
the amount or types of response resources available, this action would not necessarily result in
incremental equipment costs. On this point, PHMSA notes the possibility of using “master service
agreements.” In such an agreement, a railroad could contract for a specific type of response
resource, and thereby ensure its availability in the event of a spill and meet the requirements of the
rule; however, the agreement itself is not expected to impose any retainer fee or other significant
costs to the railroad. This type of contract would only result in costs to the railroads in the event
of a spill which requires the contracted response resource.
PHMSA emphasizes that regulatory flexibility has been incorporated into the OSRP requirements.
Specifically, we enable railroads to retain a significant degree of discretion in the development of
their plans, so that railroads will be able to plan according to the needs of the areas in which they
operate. Put another way, the standard for response planning avoids what PHMSA considers to
be prescriptive terms (e.g., requiring a USCG-certified OSRO, requiring a set quantity of
equipment or any one type of equipment) for how railroads must meet the response planning
standard. This works well with the concept of “response zones” and we feel that regulatory
flexibility is justified given the wide variability of railroad operations and environments across the
U.S. We believe this flexibility minimizes many of the cost impacts to railroads which would be
incurred in response planning stages.
Page 26
In addition to minimizing costs, PHMSA also believes that regulatory flexibility allows the
regulated community to innovate to improve spill response. For example, railroads may be capable
of transporting response resources to the site of a derailment using their own railway. In addition,
it could be possible that response resources travel as part of the actual train configuration.
To supplement the discussion of existing response resources, we can direct the public to an
example of an OSRO equipment list from the National Response Corporation (NRC). 37 This
equipment list is publicly available on NRC’s website and provides an illustration of the equipment
that OSROs might be able to provide in response to an oil spill. PHMSA emphasizes that this is
an example only.
PHMSA also notes that servicing the rail network may require less response resources than other
transportation contexts. For example, worst-case discharges in the maritime context can involve
significantly greater volumes of petroleum oil than would be encountered in the rail context.38
Moreover, EIA data regarding domestic oil refinery receipts illustrates that rail transport was the
lowest volume mode among the different modes that delivered oil to refineries from 2010–2017.
In 2017, 80 percent of total receipts were delivered via pipeline with only 3 percent delivered via
rail. This may mean that the response resources required to address the risk of spills involving
crude-by-rail could be less than the level of resources needed to address the risk of spills involving,
for example, pipeline, tanker, barge or truck transport. See Table 2 for this EIA data.
See https://nrcc.com/dev/wp-content/myimages/2016/09/nrc-major-equipment-list.pdf or
http://nrcc.com/pdf/Website.pdf. Note: PHMSA is not endorsing this organization nor suggesting that this list is
what is required to meet the response planning requirements of this rule.
38
As examples, the Torrey Canyon tanker spill off the coast of England that led to the establishment of the National
Contingency Plan (NCP) in 1968 involved approximately 37 million gallons and the Exxon Valdez tanker spill
involved approximately involved nearly 11 million gallons. These figures were found in the following USCG and
NOAA links, respectively: (1)
https://www.dco.uscg.mil/Portals/9/DCO%20Documents/National%20Strike%20Force/foscr/ASTFOSCRSeminar/
References/FOSCGuideFinal.pdf?ver=2017-09-15-105040-910; and (2) http://response.restoration.noaa.gov/oiland-chemical-spills/significant-incidents/exxon-valdez-oil-spill.
37
Page 27
Table 2. Modal Share of U.S. Domestic Crude Oil Refinery Receipts, 2012-2016
100%
90%
80%
70%
60%
50%
40%
30%
20%
10%
0%
2012
2013
2014
2015
2016
U.S. Domestic Crude Oil Refinery Receipts by Truck
U.S. Domestic Crude Oil Refinery Receipts by Tank Cars (Rail)
U.S. Domestic Crude Oil Refinery Receipts by Barge
U.S. Domestic Crude Oil Refinery Receipts by Tanker
U.S. Domestic Crude Oil Refinery Receipts by Pipeline
U.S. Domestic Crude Oil Refinery Receipts (Thousand Barrels)
Transportation Mode
Pipeline
Tanker (ship)
Barge
Rail
Truck
Data
source:
2012
79.73%
7.74%
6.06%
1.21%
5.26%
Percent of Total Receipts by Year
2013
2014
2015
76.31%
74.18%
76.33%
8.57%
9.48%
9.58%
7.42%
7.53%
5.70%
2.64%
4.11%
3.23%
5.05%
4.70%
5.15%
Energy
Information
https://www.eia.gov/dnav/pet/pet_pnp_caprec_dcu_nus_a.htm
2016
79.63%
6.78%
5.10%
3.42%
5.08%
Administration,
1.3.3. Baseline Summary
The current regulations in 49 CFR part 130 provide a minimum standard for regulatory
compliance. PHMSA estimates that railroads are currently complying with this minimum standard
in the form of the “basic plan” required. Furthermore, PHMSA notes that many railroads may
exceed the minimum “basic plan” standards and may be partially meeting the comprehensive plan
requirements.
Given that similar oil spill response planning requirements are already in place for facilities,
pipelines, and vessels, PHMSA believes that sufficient response resources are currently available.
Page 28
In addition, this estimation is supported by our analysis of current response resource availability
through USCG-certified OSROs. Based on our analysis of publicly available OSRO data, PHMSA
finds evidence that the response standards required in this rulemaking are reasonable.
It is important to note that while this baseline analysis includes an examination of response
resources, the costs of the final rule are for planning and information sharing—not costs of
response. The costs of actually responding to a spill and related removal will be incurred by
responsible parties (e.g., railroad) irrespective of this rulemaking.
In light of these considerations, PHMSA does not expect that the rule will result in significant
additional resources being deployed to spills, or in the use of significant additional or more
expensive resources at spill locations. The same types and quantities of resources will be deployed
and used under this rule as compared to the status quo (absence of rulemaking)—the principal
change PHMSA expects is that resources may be deployed and used sooner or more effectively
through enhanced planning.
1.4. Universe of Affected Entities
1.4.1. Comprehensive OSRPs
The following entities would be subject to the final rule:
1. Any person transporting any liquid petroleum or non-petroleum oil in a quantity greater
than 42,000 gallons per packaging must submit a comprehensive plan meeting the
requirements of this subpart.
2. Any railroad that transports any single train carrying 20 or more tank cars of liquid
petroleum oil in a continuous block or 35 such cars in a single train must submit a
comprehensive plan.
a. In determining number of tank cars, that person or railroad is not required to include
tank cars carrying mixtures of petroleum oil not meeting the criteria for Class 3
hazardous material in 49 CFR 173.120 or containing residue.
3. A person or railroad meeting the requirements for a comprehensive plan need not submit a
plan if otherwise excepted in 49 CFR 130.2(c).
The requirement for any person transporting any liquid petroleum or non-petroleum oil in a
quantity greater than 42,000 gallons per packaging is consistent with the current requirement in
49 CFR 130.2. In the current Supporting Statement for Information Collection Request (ICR)
2137-0591, 39 PHMSA notes that no railroad has met the required threshold based on a carriage of
oil greater than 42,000 gallons per tank car, and thus the estimated number of entities currently
affected by the comprehensive plan requirements is zero.
Information Collection Request. Response Plans for Shipments of Oil - Supporting Statement. OMB Control No.
2137–0591, available at: http://www.reginfo.gov/public/do/PRAOMBHistory?ombControlNumber=2137-0591
39
Page 29
For determining the entities that will be affected by the threshold in this rulemaking, PHMSA uses
the definition of HHFT established in the “Enhanced Tank Car Standards and Operational Controls
for High-Hazard Flammable Trains – Final Rule” 40 published on May 8, 2015. PHMSA narrowed
the affected entities to only include railroads that transport crude oil and, in consultation with FRA,
revised the estimated number of Class III carriers that are subject to the rulemaking. FRA provided
PHMSA with this estimate based on 2013 STB confidential waybill sample and FRA inspector
observations of HHFT crude oil train movements. PHMSA believes the estimated number of Class
III impacted railroads is a conservative estimate and subject to uncertainty as not all Class III
railroads submit their carload waybills for inclusion in the STB waybill sample report. Based on
this assessment, PHMSA estimates there are 73 railroads (7 Class I, 11 Class II, and 55 Class III)
that would be subject to this rulemaking. While this rulemaking applies to all petroleum products
transported by rail, PHMSA believes that crude oil makes up a substantial percentage of the
petroleum being transported in the quantities specified in the rulemaking. Therefore, PHMSA
believes this estimate captures the universe of affected entities.
Table 3 presents the estimated number of affected entities for the oil spill response planning
provisions in the final rule. This includes estimates for Class I, Class II, and Class III railroads
that transport petroleum crude oil in the threshold established in this rulemaking. These estimates
were derived for the purpose of estimating the costs and benefits associated with the final rule.
Table 3. Universe of Affected Entities – Oil Spill Response Planning
Estimated Number of
Total Number of
Railroad Class
Affected Entities
Railroads
Class I
7
7
Class II
11
11
Class III
55
730
73
748
Total Universe
1.4.2. Information Sharing
The universe of affected entities for the information-sharing requirements is different from the
number of entities affected under the comprehensive response plan requirement. The applicability
of this requirement is derived in part from the HM-251 final rule; 41 specifically, the definition of
an HHFT and the information-sharing portion of the routing requirements relate to this final rule.
The applicability also relates to the FAST Act, which outlines the information-sharing requirement
in section 7302, paragraph (3).
The FAST Act specifies Class I railroads must provide advanced notification and information on
high-hazard flammable trains to each State Emergency Response Commission (SERC), consistent
with the notification requirements in the May 2014 Emergency Order. The FAST Act requires
that SERCs receiving this advanced notification must provide the information to law enforcement
80 FR 26643, pp 26643–26750. May 8, 2015. https://www.federalregister.gov/articles/2015/05/08/201510670/hazardous-materials-enhanced-tank-car-standards-and-operational-controls-for-high-hazard-flammable
41
“Hazardous materials: Enhanced tank car standards and operational controls for high-hazard flammable trains,”
May 8, 2015. 80 FR 26643.
40
Page 30
and emergency response agencies upon request. The FAST Act directs the Secretary to establish
security and confidentiality protections for electronic train consist information or advanced
notification.
The FAST Act limits the applicability of the advanced notification requirements for HHFT to the
Class I railroads (as described in section 20102 of title 49, United States Code). PHMSA is
requiring that the information-sharing requirements apply to all railroads with HHFT operations.
This requirement fulfills the Congressional mandate and is within PHMSA’s regulatory authority.
Through the authority of Federal hazmat transportation law and the delegation of this authority to
PHMSA by the Secretary, PHMSA is responsible for overseeing a hazmat safety program that
minimizes risks to life, property, and the environment inherent in the transportation of hazmat in
commerce. Thus, in requiring that the information-sharing requirements apply to all railroads with
HHFT operations, PHMSA believes we are faithfully addressing the provisions of the FAST Act,
as well as acting in accordance with our delineated authority by additionally addressing the
potential safety risks posed by the HHFT operations of Class II and Class III railroads. While we
acknowledge that the HHFT operations of Class II and Class III railroads are relatively limited in
comparison to those of Class I railroads, and thus are likely to pose fewer safety risks in the rail
transportation system, we maintain that the HHFT operations of Class II and Class III railroads
nonetheless pose safety risks that justify adherence to the information-sharing requirements of this
final rule. In other words, it is PHMSA’s belief at this time that the HHFT operations of Class II
and Class III railroads, however limited, do not warrant an exception to the information-sharing
requirements.
For these reasons, the potential universe of affected entities for the information-sharing provision
includes all Classes of railroads that transport HHFTs transporting crude petroleum oil and ethanol,
or 178 railroads (7 Class I, 11 Class II, and 160 Class III). Please note, the inclusion of rail carriers
that transport ethanol explains the difference between this potential universe for information
sharing and the number of affected entities for the comprehensive plan provisions.
For purposes of assessing costs for this provision, however, PHMSA assumes that there should be
no additional costs for Class I railroads to comply with this revision per AAR’s Circular OT-55,
requiring AAR members to provide bona fide emergency response agencies or planning groups
with specific commodity flow information covering all hazardous commodities transported
through the community for a 12-month period in rank order. We assume this includes the
information to be shared with SERCs and TERCs as required in this rule.
In addition, on May 7, 2014, DOT had issued an Emergency Restriction/Prohibition Order in
Docket No. DOT-OST-2014-0067 (Order). 42 That Order required each railroad transporting 1
million gallons or more of Bakken crude oil in a single train in commerce within the United States
to provide certain information in writing to the SERC for each State in which it operates such a
train. PHMSA determined that 40 Class II and Class III railroads were part of this order and have
already developed the required notification. Therefore, those Class II and Class III entities are
only subject to the update and submission requirements included in this rulemaking.
42
http://www.dot.gov/briefing-room/emergency-order
Page 31
Table 4 presents the estimated universe of affected entities for the information-sharing provisions
in the rule. For the number of affected entities for notification development, the calculation is as
follows: 178 total entities – 7 Class I entities – 40 Class II and Class III entities = 131 Class II and
Class III railroads. For the number of affected entities for updates, subtract the 7 Class I entities
from the overall number of entities due to the precedence of FAST Act requirements.
Table 4. Universe of Affected Entities – Information Sharing
Provision
Estimated Number of Affected Entities
Notification Development
Updates and Recordkeeping
131 Railroads
171 Railroads
1.5. Alternatives Analysis
For purposes of determining the applicability threshold, PHMSA evaluated four options that would
affect the number of entities subject to the comprehensive OSRP requirement. The details of these
four options are described below, and Table 5 presents the total and annualized costs for these
alternatives. PHMSA also considered other alternatives but did not analyze them because we
lacked sufficient information. These alternatives include more and less stringent response time
requirements for arriving at the scene of oil spill incidents, and more and less stringent definitions
of “worst case discharge.”
1.5.1. Alternative A: Tank cars carrying any quantity of petroleum crude oil
Under this alternative, any railroad carrying liquid petroleum of any volume would be required to
submit a comprehensive OSRP. PHMSA consulted with subject matter experts at FRA to estimate
the number of railroads that currently ship petroleum in any volume. Based on this evaluation,
PHMSA estimates that there are approximately 756 railroads (7 Class I, 11 Class II, and 738 Class
III) that transport liquid petroleum and thus would be subject to the requirement under this
alternative. This represents the most stringent application of the requirement, to all railroads, and
is the extreme upper bound.
Most of the approximately 738 small railroads that operate in the United States do not typically
transport hazardous materials, including petroleum oil. Some of these railroads are switching
yards, tourist operations, or in the “other” category and not “freight” category. Railroads carrying
a smaller volume of petroleum (below the threshold established in this rule) may be sufficiently
covered under the basic plan and do not meet the standards necessary to establish a comprehensive
plan. PHMSA holds that imposing an applicability of any other number of tank cars that is less
than 20 in a continuous block or 35 when dispersed throughout a train, would most likely be costly
or burdensome and yield limited safety benefits due to the impacts on small entities as well as
“manifest” train configurations involving petroleum oil. Therefore, PHMSA did not include a
quantified estimate of the costs associated with this alternative.
1.5.2. Alternative B: 20 cars in a unit or 35 cars in a consist carrying petroleum crude oil
(preferred option)
Under this alternative, any railroad carrying 20 or more tank cars of liquid petroleum oil in a
continuous block or 35 such cars on a single train would be required to submit a comprehensive
Page 32
OSRP. Under this alternative, the requirement would affect trains operating on all classes of track.
Based on an evaluation of the 2013 Waybill Sample data and consultation with FRA, PHMSA has
determined that approximately 73 railroads would be subject to the requirement under this
alternative. As described in Section 1.4, this estimate includes 7 Class I, 11 Class II, and 55 Class
III railroads. This is the requirement enacted by the final rule.
1.5.3. Alternative C: 20 cars in a unit or 35 cars in a consist carrying petroleum crude oil
operating on Class 3 track or higher
The quantity thresholds are the same as those of Alternative B. However, under this alternative,
only trains operating on Class 3 track or higher would be subject to the requirements. According
to 49 CFR 213.9, the maximum allowable speed for freight trains is 25 mph on a Class 2 track and
10 mph on a Class 1 track. FRA estimates that approximately 20 of the Class III railroads
identified in Alternative B would operate trains at speeds higher than 25 mph and thus operate on
Class 3 track or higher. The remaining 35 Class III railroads that transport petroleum in the
quantities required in this rule are assumed to operate on Class 1 or Class 2 track and are not
included in this alternative. Based on this estimate, PHMSA has determined that approximately
38 railroads (7 Class I, 11 Class II, and 20 Class III) would be subject to the requirement under
this alternative.
1.5.4. Alternative D: More than 70 cars in a consist carrying petroleum crude oil
Under this alternative, only railroads carrying more than 70 carloads of liquid petroleum oil in a
consist would be required to submit a comprehensive OSRP. Based on an evaluation of the 2013
Waybill data and consultation with FRA, PHMSA has determined that approximately 53 railroads
(7 Class I, 11 Class II, and 35 Class III) currently transport more than 70 carloads in a consist of
liquid petroleum and thus would be subject to the requirement under this alternative.
1.5.5. Alternative E: No Action
Under this alternative, the current thresholds in 49 CFR part 130 would apply for the development
of a comprehensive response plan. This requirement applies to any person transporting any liquid
petroleum or non-petroleum oil in a quantity greater than 42,000 gallons per packaging. In the
current Supporting Statement for Information Collection Request (ICR) 2137-0591, 43 PHMSA
determined that there may be up to 2 railroads that engage in the transportation of specifically
identified tank cars capable carrying a quantity of oil greater than 42,000 gallons. However, since
this regulatory requirement was promulgated, no railroad has met the required threshold based on
a carriage of oil greater than 42,000 gallons per tank car, thus the estimated number of entities
affected by this alternative is zero. Since there would be no additional costs associated with this
alternative, PHMSA did not include this alternative in the summary table below.
Information Collection Request. Response Plans for Shipments of Oil - Supporting Statement. OMB Control No.
2137–0591. http://www.reginfo.gov/public/do/PRAOMBHistory?ombControlNumber=2137-0591
43
Page 33
1.5.6. Summary
Under alternatives B through D, the number of Class I and Class II railroads affected by the
thresholds does not change. However, the number of Class III railroads that would be subject to
the final rule ranges from 55 to 20 railroads. The total costs over the 10-year period considered in
this analysis, discounted at 7 percent, range from $11 million to $4 million for Class III railroads.
Table 5 provides an overview of the 10-year total and annualized costs by railroad class for each
alternative evaluated. 44
PHMSA also looked at other plan requirements to evaluate potential options to reduce burden or
minimize costs for plan holders. For example, rather than requiring that plans be resubmitted every
5 years as is required by the U.S. Coast Guard (33 CFR 151.27), railroads would only be required
to re-submit their plans to PHMSA if updates are made to address new or different operating
conditions or information that would substantially affect the implementation of the response plan.
In addition, PHMSA identified several cost uncertainties and conducted sensitivity analyses that
are discussed in section 2.4 of this document.
Table 5. Overview of Total and Annualized Costs per Railroad Class by Alternatives
(7% Discount)
Alternative
B
C
D
Class of
Railroad
Number of
Affected
Railroads
Class I
Class II
Class III
Total
Class I
Class II
Class III
Total
Class I
Class II
Class III
Total
Total Cost (7%)
7
11
55
73
7
11
20
38
7
11
35
53
$4,929,142
$3,142,468
$11,770,121
$19,841,731
$4,929,142
$3,142,468
$4,280,044
$12,351,654
$4,929,142
$3,142,468
$7,490,077
$15,561,687
Annualized Cost (7%)
$701,799
$447,417
$1,675,800
$2,825,016
$701,799
$447,417
$609,382
$1,758,598
$701,799
$447,417
$1,066,418
$2,215,634
1.6. Timeframe for the Analysis
PHMSA estimates that the economic effects of this rulemaking, once finalized and adopted, will
be sustained indefinitely. Notwithstanding this, because of the difficulty of and uncertainty
associated with forecasting effects into the far future, PHMSA assumes a 10-year timeframe to
outline, quantify, and monetize the costs and benefits of the proposal and to demonstrate the net
effects of the proposal.
Table 5 only includes the costs for oil spill response planning, and does not include alternatives to the costs for
information sharing. Those costs would be the same under each of these scenarios for the railroads subject to the
information sharing provision.
44
Page 34
2. Costs
The following sections provide an overview of the estimated costs for railroads subject to the final
rule. For the development of comprehensive OSRPs, those costs include: plan development,
submission, and maintenance; contract services for OSROs; and training and exercises. For
information sharing, those costs include notification development, and periodic updates and
submission. Total costs are presented in Section 2.3 of this document. The costs to the Federal
Government for the review and approval of the comprehensive OSRPs and review of the
information sharing notifications were not included in the overall cost estimates for this rule.
However, an estimate of those costs is included in Section 2.2.3 of this document, and we have
provided a table in the “Total Costs” section to illustrate the increase in the total costs of the rule
if costs to the Federal Government were included.
2.1. Comprehensive OSRPs
2.1.1. Costs of Plan Development, Submission, and Maintenance
The final rule requires several specific elements to be included in the comprehensive OSRP. These
items are summarized below. The complete list of requirements can be found in the final rule.
This section will discuss the costs associated with the specific requirements for plan development,
submission, and maintenance. Sections 2.1.2 and 2.1.3, respectively, discuss the costs associated
with the requirement to ensure by contract or other means response resources and the costs
associated with conducting drills and training exercises.
2.1.1.1. Costs for Plan Development and Updates
Each railroad subject to the final rule must prepare and submit a comprehensive OSRP that
includes a plan for responding, to the maximum extent practicable, to a worst-case discharge and
to a substantial threat of such a discharge of oil. In addition, each railroad shall review its plan in
full at least once every 5 years from the date of the last approval.
A comprehensive plan must include:
•
Core Plan: A core plan includes an information summary and any components that do not
change between response zones.
•
Response Zone Appendix: For reach response zone, a railroad must include a response
zone appendix to provide the information summary and any additional components of the
plan specific to the response zones.
Currently, railroads are required to complete a basic OSRP for oil shipments in a package with a
capacity of 3,500 gallons or more. Railroads that are subject to comprehensive plans are required
to make several additions to what is already established in their basic plans. For example, these
additions include:
•
Ensure the plan is consistent with the NCP and applicable ACPs.
•
Identify a QI with full authority to implement response actions.
Page 35
•
Identify and ensure the availability of personnel and equipment to respond to worst-case
discharges.
•
Describe the training, equipment testing, exercises, and response activities that will be
undertaken to ensure safety and mitigate spills.
•
Describe the procedure for updating and resubmitting response plans when significant
changes occur.
In addition to the core comprehensive plan, railroads are also required to identify response zones
that are included as appendices to the core plan. Railroads will determine how many response
zones they need to develop. However, for purposes of estimating costs associated with this rule,
PHMSA, in consultation with FRA, evaluated the current network of crude oil transportation by
rail to determine an estimated number of response zones differentiated by railroad class. Based on
this evaluation, PHMSA has estimated that the average response zone will likely cover
approximately two States. PHMSA estimates that each Class I railroad will likely develop eight
response zones in addition to the core plan, Class II railroads will likely develop two response
zones, and Class III railroads will develop one response zone. Many Class III railroads’ operations
are limited to one to three states, so using two states made sense in developing a kind of lowest
common denominator for the size of a response zone. This would limit the impact on smaller
railroads.
For purposes of determining the amount of time that would be necessary to prepare the
comprehensive plan and response zone appendixes, PHMSA consulted with OSROs in
development of the NPRM and consulted with third party OSRP plan writers in response to
comments to the NPRM. We also evaluated estimates used in previous rulemakings requiring
non-railroad entities to develop and submit a comprehensive OSRP, as well as estimates used in
PHMSA’s information collection request and comments to the NPRM. 45,46 On the basis of
comments to the NPRM and input from a third party OSRP plan writer, PHMSA estimates the
level of effort will be similar for developing a Class I railroad or shortline core plan or a response
zone plan.
Table 6 provides an estimate of the amount of time necessary for each class of railroad to develop
the comprehensive plan and response zone appendices. PHMSA estimates it will take 40 hours of
senior time, 40 hours of administrative time, and 100 hours of mid-level staff for the core plan and
the same number of hours for developing a plan per response zone (RZ). The cost then is estimated
at $9,407 per core plan or RZ, and thus a total of $18,815 for developing a plan for a Class III
railroad as shown in the calculations below.
45
Nontank Vessel Response Plans and Other Vessel Response Plan Requirements – Final Regulatory Analysis and
Final Regulatory Flexibility Analysis for the Final Rule. USCG-2008-1070. May 2013.
http://www.regulations.gov/#!documentDetail;D=USCG-2008-1070-0046.
46
Information Collection Request. Response Plans for Onshore Oil Pipelines – Supporting Statement. OMB Control
No. 2137-0589.
Page 36
Table 6. Overview of Estimated Plan Development Hours per Railroad by Railroad Class
Class of
Railroad
Class I
Class II
Class III
Number of
Response Zones
Comprehensive Plan
Development Time
Response Zone Development
Time (per zone)
Total Hours
per Railroad
8 zones
2 zones
1 zone
180 hours
180 hours
180 hours
180 hours
180 hours
180 hours
1,620 hours
540 hours
360 hours
The hourly labor rate for a manager used to estimate the cost of initial plan development in the
implementation year is $75.53. This labor rate is based on the median wage estimate ($47.74)
from the Bureau of Labor Statistics (BLS) Occupational Employment and Wages, May 2016 for
the wage series “11-1021 General and Operational Managers.” 47 In addition, to calculate the
hourly wage rates for each year of the analysis, PHMSA inflated this wage by 56 percent to account
for fringe benefits. 48 Finally, the wage was adjusted for 2018 and subsequent years in this analysis
based on an estimated 1.1-percent annual growth rate in median real wages. 49 This results in an
estimated hourly wage of $75.53 for the manager involved in the initial development of the
comprehensive plan and response zones. Similarly, PHMSA estimated the hourly labor rates for
an administrative and a rail transportation worker at $42.50 50 and $46.86 51, respectively.
As discussed above, the initial plan development would require 180 hours of effort, which would
result in a cost in 2018 of approximately $9,407.20 per core plan (40 hours of manager time *
$75.53 per hour + 40 hours of administrative time * $42.50 + 100 hours of mid-level staff *
$46.86). As noted above, a similar level of effort will be required to develop a plan for a response
zone. The response zones would require an additional $9,407 per plan.
In addition to initial plan development, railroads subject to this rule are required to update their
plan to address new or different conditions or information, and review their plan in full at least
every 5 years from the date of the last approval. While it is likely that not all railroads will need
to completely revise their plans after each review period, PHMSA does not have information
available to estimate the extent of these updates. Therefore, for purposes of this analysis, PHMSA
assumes that all railroads subject to this rule would conduct a plan review and update every 5
years. This is consistent with assumptions made in OPS’ 2004 analysis for onshore pipelines.
https://www.bls.gov/oes/2016/may/oes111021.htm
BLS does not publish data on fringe benefits for specific occupations, but it does for broad industry groups in its
Employer Costs for Employee Compensation release. This regulatory evaluation uses the average hourly wage of
$25.75 and average hourly benefits of $14.49 for private industry workers in “transportation and warehousing” to
estimate that fringe benefits are equal to 56 percent ($14.49 / $25.75) of wages. Source: Bureau of Labor Statistics.
Table 10: Employer costs per hour worked for employee compensation and costs as a percent of total compensation:
Private industry workers, by industry group, December 2017. http://www.bls.gov/news.release/pdf/ecec.pdf
49
Based on the Median Usual Weekly Earnings (MUWE), in constant dollars, derived by BLS from the Current
Population Survey (Series LEU0252881600 – not seasonally adjusted), PHMSA estimated a 1.1% annual growth
rate.
50
This labor rate is based on the median wage estimate ($26.86) from the Bureau of Labor Statistics (BLS)
Occupational Employment and Wages, May 2016 for the wage series “43-6011 Executive Secretaries and
Administrative Assistants.”
51
This labor rate is based on the median wage estimate ($29.62) from the Bureau of Labor Statistics (BLS)
Occupational Employment and Wages, May 2016 for the wage series “53-4099 Rail Transportation Worker, all
other.”
47
48
Page 37
Further, PHMSA estimates the recurring burden with reviewing the initial plan as half of the
burden needed to develop the initial plan development.
Table 7 provides an overview of the estimated cost per railroad for the initial development of the
comprehensive plan in year 1. This cost estimate also includes review of the plan every 5 years.
Table 7. Estimated Plan Development Costs per Railroad by Railroad Class
Class of
Railroad
Class I
Class II
Class III
Level of
Effort
(Hours) for
Core Plan
180 hours
Level of
Effort
(Hours) for
Response
Zone(s)
Total Hours
per
Railroad
(core plan +
response
zones)
1,440 hours
360 hours
180 hours
1,620 hours
540 hours
360 hours
Weighted Average
Hourly
Compensation of
Railroad Employees
Plan
Development
Cost per
Railroad
$52.26͞
$84,665
$28,222
$18,814
In developing these estimates, PHMSA evaluated public comments in response to the ANPRM
and NPRM. In response to the ANPRM, PHMSA received public comments on the cost of plan
development that ranged from $12,000 to $500,000. AAR and ASLRRA estimated costs would
be $100K to $500K per railroad. It could be argued that PHMSA’s estimate for Class I plan
development costs are comparable to the lower-bound estimate given by AAR/ASLRRA in the
ANPRM ($84,666 vs. $100,000). Given the ANPRM contained no specific proposals, AAR and
ASLRRA noted they could not provide a specific or accurate cost estimate. Other comments to
the ANPRM regarding plan development costs were approximate and lacked detail.
In response to the NPRM, PHMSA received a few comments regarding plan development costs.
Unfortunately, they were not more detailed than the comments provided to the ANPRM. This
made it difficult to revise the cost estimates provided in the preliminary RIA. For example, API
asked that PHMSA “look more closely at the time and effort required to develop and implement
plans initially.” 52 In addition, AAR commented, “PHMSA’s hourly estimates for completing the
‘Core Plan and Response Zone’ documents are underestimated.” Please note, AAR and API did
not supply to the NPRM data or comparative estimates with which to revise our estimates.
Nevertheless, PHMSA engaged with third party OSRP plan writers in response to the comments
and was successful in receiving one set of updated estimates for the level of effort required to
develop a plan. These updated estimates for plan development are reflected in this RIA.
Specifically, the preliminary RIA estimated 80 hours to develop the core plan and 15 hours per
response zone. In this RIA, we estimate 180 hours of effort, split across 3 occupational groups, as
well as 180 hours for each response zone. This amounts to doubling the effort for core plan
development and increasing by 12-fold the effort estimated to create a single response zone
appendix. PHMSA feels this revision is responsive to the commenters’ concerns. If additional
data were supplied, PHMSA may have had the opportunity to evaluate other cost estimates, but
this was not the case.
Comment to HM-251B NPRM, American Petroleum Institute (API), pg. 2, available at:
https://www.regulations.gov/document?D=PHMSA-2014-0105-0322 [hereinafter “API comment”]
52
Page 38
PHMSA also considered estimates for plan development for non-railroad entities, such as pipeline
operators. For example, in the regulatory impact analysis for OPS’s 2005 final rule titled
“Response Plans for Onshore Oil Pipeline,” OPS estimated that per facility plan development costs
would range from $16,000 to $1.9M. The $1.9M estimate was for mega-facilities covering over
100 counties. We also mentioned this OPS-derived estimate in the preliminary RIA and did not
receive public comment suggesting that it, or an adaptation thereof, would be better suited for our
final analysis. In the supporting statement developed for the ICR to authorize the collection of
information required in the response plans for onshore pipelines, OPS estimated initial plan
development would require 40–80 hours 53 of engineering staff time per standalone plan. 54
Moreover, PHMSA evaluated estimates used for commercial, nontank vessels that are required by
USCG to develop and submit OSRPs. 55 In the final regulatory analysis developed for the Nontank
Vessel Response Plans and Other Vessel Response Plan Requirements (NTVRP), 56 the USCG
estimated that the initial OSRP developed by the plan holder would require 40 hours of labor time
from a senior staff member. 57
In comparison, PHMSA’s plan development cost estimates for railroads’ plans are lower than
AAR/ASLRRA ANPRM comments and OPS’ 2005 final rule, but higher than OPS’ ICR estimate.
Our estimates remain uncertain, and it must be considered that our analysis is a representation of
expected effects; not an infallible prediction of impacts. In further support of our final RIA
estimates, we believe it is likely that many of the railroads that would be subject to the final rule,
particularly Class I and Class II railroads, have already developed response plans that include
similar provisions to what is included in comprehensive plans. These railroads would only need
to update their plans to be consistent with the requirements and may not require the entire time
estimated for plan development. As such, it could be argued that our plan development cost
estimates are conservative, or over-estimates, in the sense that we present the costs as if the affected
railroad needed to develop a plan “from scratch.” We see this as an additional justification for our
overall plan development cost estimate, but acknowledge the uncertainty affecting these
estimations.
Further, while there are technical differences in the information that would be needed for a railroad
to develop a comprehensive OSRP (as compared to a pipeline- or vessel-related plan), the type of
information being requested is substantively similar to the requirements for these other entities.
Thus, PHMSA believes our revised estimates are reasonably responsive to public comment and
reflect a workable estimation of the cost to develop a comprehensive plan. Data uncertainties and
limitations limited the extent to which PHMSA could consider additional cost estimations.
Final Regulatory Evaluation of the Response Plan Requirement for Transportation-Related Onshore Oil Pipelines,
Office of Pipeline Safety (OPS), Research and Special Programs Administration (RSPA), August 2004, see pg. 7-7.
54
Information Collection Request. Response Plans for Onshore Oil Pipelines – Supporting Statement. OMB Control
No. 2137-0589.
55
33 CFR part 155.
56
Nontank Vessel Response Plans and Other Response Plan Requirements. Federal Register, 78(189), 60100–60135.
September 30, 2013.
57
Nontank Vessel Response Plans and Other Vessel Response Plan Requirements – Final Regulatory Analysis and
Final Regulatory Flexibility Analysis for the Final Rule. USCG-2008-1070. May 2013, available at:
http://www.regulations.gov/#!documentDetail;D=USCG-2008-1070-0046.
53
Page 39
2.1.1.2. Costs for Plan Submission
The final rule requires plans to be submitted to PHMSA.
In addition, any significant changes to the plan also must be submitted to PHMSA. Specifically,
under the final rule, railroads are required to update their plan to address new or different
conditions or information, and each railroad must review its plan in full at least every 5 years from
the date of the last approval. If a new or different operating conditions or information would
substantially affect the implementation of the response plan, the railroad must immediately modify
its plan to address such a change and must submit the change to PHMSA within 90 days. While
it is likely that not all railroads will need to submit an updated plan after each review period,
PHMSA does not have information available to estimate the likelihood of updates requiring a
submission. Therefore, for purposes of this analysis, PHMSA assumes that all railroads subject to
the rule will submit their plans every 5 years.
PHMSA estimates that it would take an administrative employee approximately 0.5 hours to
assemble and submit the final plan to FRA. The hourly labor rate used to estimate the cost of
initial plan development is $42.50. This labor rate is based on the median wage estimate ($26.86)
from the BLS Occupational Employment and Wages, May 2016 for the wage series “43-6011
Executive Secretaries and Executive Administrative Assistants.” 58 PHMSA inflated this wage by
56 percent to account for fringe benefits and adjusted the wage for 2016 and subsequent years in
this analysis based on an estimated 1.1-percent annual growth rate in median real wages.
Each plan submission would result in a cost of approximately $21 per plan (0.5 hours × $42.50 per
hour, rounded). PHMSA assumes the plans will be submitted electronically to PHMSA and would
not require any additional mailing costs. This cost would be the same for Class I, Class II, and
Class III railroads.
2.1.1.3. Costs for Plan Maintenance and Recordkeeping
The final rule requires each railroad to:
•
Maintain a copy of its plan at the railroad’s principal place of business.
•
Provide a copy of its core plan and the appropriate response zone appendix to each QI and
alternate.
•
Provide a copy of the information summary to each dispatcher in response zones identified
in the plan.
In addition, the final rule includes the following review or maintenance requirements for the
response plans:
•
Each railroad must include procedures to review the plan after a discharge requiring
activation of the plan to evaluate and record the plan’s effectiveness.
•
Each railroad shall update its plan to address new or different conditions or information.
Bureau of Labor Statistics. Occupational Employment and Wages, May 2016; Wage Series 43-6011 Executive
Secretaries and Executive Administrative Assistants. Retrieved from http://www.bls.gov/oes/current/oes436011.htm.
58
Page 40
•
If a new or different operating condition or information would substantially affect the
implementation of the plan, the railroad must modify its plan to address such a change and,
within 90 days of making such a change, submit the change to PHMSA.
While maintenance activities will vary from year to year, for purposes of this analysis PHMSA
assumes that annual maintenance and recordkeeping will require approximately 10 percent of the
total time of plan development. This estimate is consistent with maintenance estimates used in the
information collection for both onshore pipeline 59 and commercial vessel 60 OSRPs. This would
result in approximately 18 hours per year for the core plan and 18 hours per year for each response
zone for railroads subject to the final rule (180 hours / 10 = 18 hours).
In addition to questioning the initial plan development costs, API’s comment to the NPRM and
preliminary RIA indicated that PHMSA underestimated the time and effort required to update and
review plans annually. 61 While PHMSA is using the same percentage of plan development costs
to represent annual plan maintenance costs (10 percent), we have nevertheless revised upward
these costs since they are derived from the plan development costs. For example, a Class I railroad
was estimated in the preliminary RIA to spend 20 hours maintaining their comprehensive plan
annually; in the final RIA, it is estimated that this review would take 162 hours, an approximately
7-fold increase. Thus, by increasing plan development costs, we also increased plan maintenance
costs, and we believe we have reasonably responded to this public comment. Further, PHMSA
believes our annual plan maintenance cost may be a conservative estimate, since there may be
years in which no changes are made to the plan and no additional distribution is needed. Table 8
provides an overview of the costs of maintenance and recordkeeping requirements by railroad
class. The labor rate used is the same rate as the plan development, or $52.26 discussed in Section
2.1.1.
Table 8. Estimated Plan Maintenance and Recordkeeping Costs per Railroad by Railroad
Class
Class of
Railroad
Class I
Class II
Class III
Level of
Effort
(Hours) for
Core Plan
Level of Effort
(Hours) for
Response
Zone(s)
Total
Hours per
Railroad
18 hours
144 hours
36 hours
18 hours
162 hours
54 hours
36 hours
Average Hourly
Compensation of
Railroad Employee
$52.26
Cost per
Plan per
Railroad
$8,466
$2,822
$1,881
2.1.2. Costs for Response and Mitigation Activities for Comprehensive Plans
The final rule requires plan holders to identify and ensure by contract or other means the resources
necessary to remove, to the maximum extent practicable, a worst-case discharge and to mitigate
or prevent a substantial threat of a worst-case discharge. It requires railroads to describe in their
plans the response resources available to arrive onsite within 12 hours after the discovery of a
59
Information Collection Request. Response Plans for Onshore Oil Pipelines- Supporting Statement. OMB Control
No. 2137-0589.
60
Information Collection Request. Vessel and Facility Response Plans (Domestic and Int’l), and Additional
Response Requirements for Prince William Sound, Alaska. OMB Control No. 1625-0066.
61
See API comment, pg. 2
Page 41
worst-case discharge or the substantial threat of one. It requires railroads to implement their plans
for spills affecting navigable waters, the adjoining shorelines of navigable waters, or the natural
resources of the United States. However, PHMSA believes the actual costs of response cannot be
attributed to this regulatory action because the costs of response and other associated costs, such
as clean up and remediation of the derailment site, would be the responsibility of the responsible
party, irrespective of this rulemaking.
For purposes of this analysis, PHMSA assumes that railroads subject to this rule will contract and
use the services of a spill response provider, such as a USCG-certified OSRO, to comply with this
requirement in their response plans. Sometimes, this kind of contract involves a retainer fee that is
paid annually to the response provider in order to ensure appropriate resources are available in the
event of a worst-case discharge or the threat of one. In other cases, an annual retainer fee is not
required. We further assume that USCG certification as an OSRO is sufficient to ensure that the
OSRO has the necessary response resources to respond to a worst-case discharge as defined in this
final rule and that the retainer fee is sufficient to ensure the availability of all such resources.
The availability of response resources is separate from the deployment and use of these resources
to respond to spills. As noted above, PHMSA does not assign any cost to this final rule from
deploying or using response resources to respond to spills because these costs are expected to be
incurred by the responsible party irrespective of this rulemaking. Thus, PHMSA does not expect
that this rule will result in significantly different or additional resources to be deployed to spills.
In addition, we acknowledge that many railroads are likely to already have a contract in place with
an OSRO, with or without an annual retainer fee, signaling an existing level of compliance with
the requirements. However, PHMSA notes the importance of the regulatory baseline, as opposed
to the response planning baseline. Under 49 CFR part 130, railroads transporting HHFTs of
petroleum oil are not currently required to have a comprehensive plan and thus, are not required
to ensure by contract or other means the availability of resources to address a worst-case discharge.
This regulatory change is significant to the extent that PHMSA feels it is reasonable to present a
conservative estimate of response costs and specifically, the costs of retainer fees. We do this by
estimating that every affected entity will face an annual OSRO retainer fee, which represents the
cost of contracting response resources to comply with the rule. PHMSA does not have information
regarding the number of affected entities that already have contractual agreements in place, so we
are unable to make different assumptions. Again, some OSROs may not require an annual retainer
fee, but for the purposes of this analysis and the need to address uncertainty in this area, we present
a conservative approach to response cost estimation.
The primary drivers for the retainer fee costs are likely to be the geographic scope of coverage the
OSRO would provide and the scale of the worst-case discharge that would need to be covered.
While these costs will likely vary among response providers, PHMSA during the development of
the NPRM interviewed individuals from two OSRO companies to determine an estimate for these
services. Please see Appendix E in the preliminary RIA for the results of these interviews. 62
PHMSA estimated an OSRO retainer fee for each response zone based on these interviews and
additionally scaled the fee based on the size of the railroad. The number of response zones per
The preliminary RIA can be found in the Supporting Documents folder under Docket ID: PHMSA-2014-01050241. The questions and responses can be found in Appendix E of that document.
62
Page 42
railroad is the same as those estimated for plan development costs. PHMSA did not receive public
comment regarding retainer cost estimates, so is maintaining the original estimates in the final
RIA.
For additional background on potential OSRO retainer fees, PHMSA evaluated the estimates used
in previous rulemaking efforts for evaluating the potential OSRO fee. Based on a survey with
OSROs conducted as part of their rule development process, the USCG estimated that the average
annual OSRO fee was $224 per vessel. 63 However, the USCG also cited several factors that could
influence that cost, such as whether the OSRO was being contracted through the vessel’s
Protection and Indemnity club. In the 2005 final Onshore Pipeline Rule, OPS estimated an annual
response resource cost range of $10,000–$100,000 (in 2003 dollars) per facility. 64 Specifically,
OPS estimated OSRO fees would be $10,000 per year for pipeline facilities with a small worstcase discharge (defined as a less than 7,500 barrel discharge), $30,000 per year for facilities with
a medium worst-case discharge (7,501 – 35,000 barrel discharge), and $100,000 per year for
facilities with a large worst-case discharge (more than 35,000 barrels). OPS derived these
estimates from a consultation with an emergency response provider and also cited several factors
that could influence the cost for a facility, such as the volume of the worst-case discharge for which
facilities must provide response capabilities and whether facilities must also comply with U.S.
Environmental Protection Agency (EPA) regulations under OPA 90. In addition, PHMSA found
publicly available information from response providers that marketed “no retainers” or “no cost”
master service agreements. Additional discussion of our research is included in Appendix E in the
preliminary RIA.
Given the variability in the estimates used in the previous rulemakings, and the different units of
analyses (USCG used per vessel, OPS used per facility, and this RIA uses per railroad per response
zone), PHMSA focused on the estimate derived through the discussions with the two OSRO
companies mentioned above. One provider explained that they use a risk model to determine their
retainer fees, and typically derived facility retainer fees of $0.05/barrel; however, they have not
developed fees for railroads using this methodology. Another provider estimated that their typical
retainer fees for land-based facilities, including railroads, would range from approximately $2,500
to $10,000 per year and placed Class I railroads in the middle to lower end of that range. In
addition, they estimated that they would likely charge Class II railroads a retainer fee on the lower
end of the range, and Class III railroads may not even require a retainer fee depending on the
location of the railroad and their potential equipment needs. For purposes of this analysis, PHMSA
used this information as the primary source for estimating OSRO fees; however, a retainer fee was
included for Class III railroads. In Section 2.4.2, PHMSA conducted a sensitivity analysis
evaluating a lower or no-cost approach for a retainer fee for Class III railroads, as an alternative
viewpoint. PHMSA estimates that the annual retainer fee for an OSRO will range from $2,500 to
$5,000 for each response zone, differentiated by railroad class. Table 9 provides an overview of
the costs by railroad class.
63
Nontank Vessel Response Plans and Other Vessel Response Plan Requirements – Final Regulatory Analysis and
Final Regulatory Flexibility Analysis for the Final Rule. USCG-2008-1070, May 2013. Retrieved from
http://www.regulations.gov/#!documentDetail;D=USCG-2008-1070-0046
64
Pipeline Safety: Response Plans for Onshore Transportation-Related Oil Pipelines. Federal Register, 70(35),
8734–8748. February 23, 2005. http://www.gpo.gov/fdsys/pkg/FR-2005-02-23/pdf/05-3257.pdf
Page 43
Table 9. Estimated Annual OSRO Retainer Fee per Railroad by Railroad Class
Class of
Railroad
Class I
Class II
Class III
Retainer Fee per
Response Zone
Number of
Response Zones
Annual Cost
per Railroad
$5,000
$3,000
$2,500
8
2
1
$40,000
$6,000
$2,500
2.1.3. Cost for Training and Exercises
The final rule includes the following training and exercise requirements:
•
Training: The railroad must certify in the response plan that it conducted training to ensure
that:
o All railroad employees subject to the plan know their responsibilities under the
comprehensive spill response plan and the name of, and procedures for contacting,
the QI or alternate on a 24-hour basis.
o Reporting personnel must additionally know the content of the information
summary of the response plan, the toll-free telephone number of the National
Response Center, and the notification process required by the final rule.
•
Designated employees must receive the training required by this subpart at least once every
5 years.
Exercises: A railroad must implement and describe a drill program following the National
Preparedness for Response Exercise Program (PREP) guidelines. As an alternative, a
railroad choosing not to follow PREP guidelines must have a drill program that is
equivalent to PREP. The plan must include a description of the drill procedures and
programs the railroad uses to assess whether its response plan will function as planned,
including the types of drills and their frequencies.
2.1.3.1. Costs for Training
The final rule requires railroads subject to the rule to certify in their response plan that training has
been conducted for all employees subject to the plan and all reporting personnel. The purpose of
this training is to ensure that the applicable employees know their responsibilities under the plan,
the plan content, and the notification procedures. This training must be conducted at least once
every 5 years.
Based on discussions with subject matter experts at FRA, PHMSA determined that approximately
80 employees per Class I railroad, 40 employees per Class II railroad, and 16 employees per Class
III railroad would be subject to this training requirement. 65 For purposes of this analysis, PHMSA
assumes that 8 hours of training would be required per employee every 5 years and that training
would involve senior managers, using the senior management wage ($75.53). PHMSA anticipates
that railroads have likely conducted some training in accordance with basic plans and through
65
These estimates assume that railroads would need to train two staff members per shift, covering three shifts per
day, and that the total number of employees would vary by railroad size (class).
Page 44
voluntary actions, and notes that this estimate of 8 hours per training cycle is also consistent with
the training estimates derived in the onshore pipeline final rule and OPS’ accompanying regulatory
evaluation. 66 However, subject matter experts at FRA suggested that the training may take up to
40 hours per cycle. For this reason, PHMSA conducted a sensitivity analysis in Section 2.4.2 using
this scenario.
In addition to suggesting that plan development and annual maintenance cost estimates were low,
API also suggested in comment to NPRM that training costs were underestimated. Unfortunately,
API did not supply alternative training cost data with which to revise our preliminary estimates.
Moreover, API did not indicate quantitatively the extent to which PHMSA underestimated training
costs (e.g., 10 percent too low, 25 percent too low). Given the absence of data given in the public
comment period, PHMSA attempted to solicit additional information on training costs from third
party OSRP plan writers with training services experience, but did not obtain such information.
Ultimately, our training cost estimates resemble those given in the preliminary RIA, but with minor
revisions due to changes in salary input values.
Please note, this training ensures knowledge of the plan on the part of subject employees. Exercise
costs according to PREP guidelines are accounted for separately from training costs, although the
two areas are interrelated. The full extent of railroad training programs for hazmat and oil spill
response cannot be attributed to this rule; the costs presented here are a representation of the
incremental cost to railroads’ training and exercise programs to ensure knowledge of the plan and
alignment with PREP or PREP-equivalent guidelines. Further, there may be differences among
the affected railroads’ training and exercise programs; the costs estimated here are a generalization.
Table 10 provides an overview of the estimated training costs per training event for each class of
railroad.
Table 10. Estimated Training Costs per Railroad per Event by Railroad Class
Class of
Railroad
Class I
Class II
Class III
Number of
Employees
Hours per
Employee
Wage Rate
Total Cost per
Railroad per Event
80
40
16
8
8
8
$75.53/hour
$75.53/hour
$75.53/hour
$48,339
$24,170
$9,668
2.1.3.2. Costs for Exercises
The final rule requires railroads subject to the rule to implement an exercise program in accordance
with the USCG’s PREP guidelines, or an equivalent program. There are three categories of annual
requirements that apply to railroads subject to the rule:
1. QI Notification Exercises
2. Incident Management Team (IMT) Exercises
3. Equipment Deployment Exercises
66
Pipeline Safety: Response Plans for Onshore Transportation-Related Oil Pipelines. Federal Register, 70(35),
8734–8748. February 23, 2005. http://www.gpo.gov/fdsys/pkg/FR-2005-02-23/pdf/05-3257.pdf
Page 45
For purposes of this analysis, PHMSA referred to the USCG’s Nontank Vessel Response Plan and
Other Vessel Response Plan Final Regulatory Analysis, which includes several identical or similar
exercise requirements. 67 While the content of the exercises will vary depending on the
transportation context, PHMSA believes that the estimated number of staff members and time
burden would likely be similar for railroads.
First, the QI notification exercise is an annual exercise intended to ensure that the QI and alternate
QI identified in the response plan will respond as expected and carry out their duties in the event
of an incident. This drill includes contacting the QI by telephone or radio with a confirmation of
receipt. For purposes of this analysis, PHMSA assumes that the QI and employee conducting the
exercise will likely be senior members of the railroad staff or management team, and uses the
senior manager wage ($75.53) for this exercise. Each event will require 2 hours of the manager’s
time and 2 hours of either the QI or alternate QI’s time.
Second, the IMT exercise is an annual exercise in which the response plan is utilized to ensure that
the IMT is familiar with the plan and can use it effectively to conduct a response. At least one
IMT exercise in a triennial cycle must involve a worst-case discharge scenario. According to
interagency input, IMT exercises include practicing in a Unified Command structure,
implementing a Planning Cycle, generating an Incident Action Plan, and establishing an
Environmental Unit.
For purposes of this analysis, PHMSA assumes that the IMT exercise will take approximately
8 hours to complete and that that two senior managers from the railroad and two senior employees,
likely the QI and alternate QI, will participate, or 4 employees in total. The senior manager hourly
labor rate ($75.53) was used to estimate the labor rate for all employees involved in the IMT
exercise. In addition, travel may be required to support this exercise, and PHMSA estimates $500
per participant from the railroad to cover travel and lodging costs for the exercise. 68 A
representative from the OSRO, or a contractor that may have been used to develop the plan, may
also participate or help facilitate the exercise. PHMSA estimates that a contractor or an OSRO
may include an additional fee to participate in the exercise. For purposes of this analysis, PHMSA
assumes that such a fee will be $1,500 per year for each railroad subject to the rulemaking. This
fee estimate is similar to what was included in the USCG final regulatory evaluation referenced
above, and is also consistent with information received from PHMSA’s discussion with emergency
response providers. See Appendix E in the preliminary RIA for more information.
Third, the equipment deployment exercise is conducted annually and ensures that the response
equipment is appropriate for the operating environment in which it is intended to be used and that
operating personnel are trained in its operation. This exercise involves personnel who would
normally operate or supervise the operation of the equipment. In addition, plan holders are
responsible for ensuring that all equipment types cited in their respective plan are exercised,
whether the equipment is owned and operated by the plan holder or supplied through an OSRO.
According to the PREP guidelines, it is not necessary to deploy every piece of each type of
Nontank Vessel Response Plans and Other Vessel Response Plan Requirements – Final Regulatory Analysis and
Final Regulatory Flexibility Analysis for the Final Rule. USCG-2008-1070, May 2013. Retrieved from
http://www.regulations.gov/#!documentDetail;D=USCG-2008-1070-0046
68
This assumes $300 for travel costs and $200 for 1 night of lodging and incidentals per participant.
67
Page 46
equipment as long as all equipment is included in a periodic inspection and maintenance program
intended to ensure that the equipment remains in good working order. 69
For purposes of this analysis, PHMSA assumes that the equipment deployment exercise will
involve two senior managers from each covered railroad that will spend 8 hours preparing for the
exercise and 8 hours participating in the exercise. The hourly labor rate used for the senior
manager is $75.53. In addition, PHMSA assumes that approximately eight engineering staff
members from the covered railroad will participate in the 8-hour equipment deployment exercise.
The hourly labor rate used to estimate the cost for the railroad staff is $46.86. This labor rate is
based on the median wage estimate ($29.62) from the BLS Occupational Employment and Wages,
May 2016 for the wage series “53-4099 Rail Transportation Workers, all other.” 70 PHMSA
inflated this wage by 56 percent to account for fringe benefits, and adjusted the wage for 2018 and
subsequent years in this analysis based on a 1.10-percent annual growth rate in median real wages.
Travel also may be required to support this exercise, and PHMSA assumes $500 per participant
from the railroad will be incurred to cover travel and lodging costs for the exercise. Finally,
PHMSA assumes that there will be an additional $1,500 fee charged by the OSRO to provide
materials for the equipment deployment exercise.
Table 11 provides an overview of the costs of the three exercise requirements described in this
section by exercise, differentiated by type of exercise.
Table 11. Estimated Exercise Costs per Railroad per Exercise
Type of
Exercise
QI Notification
IMT Exercise
Equipment
Deployment
Managers/QI
Number
Hours
Labor
of
per
Rate
Managers Manager
Engineering Staff
Hours
Number
Labor
per
of Staff
Rate
Staff
Other Costs
Travel
Cost
OSRO
Fee
Total Cost
per
Railroad
per Exercise
2
4
2
8
$75.53
$75.53
0
0
0
0
$46.86
$46.86
$0
$2,000
$0
$1,500
$302
$5,917
2
16
$75.53
8
8
$46.86
$5,000
$1,500
$11,916
2.2. Information Sharing
Under the final rule, a rail carrier of an HHFT as defined in § 171.8 must provide the following
notification to the SERC, TERC, or other appropriate entities delegated by the State in which it
operates. Please see the final rule published in the Federal Register for the requirements codified.
We summarize these requirements here for convenience and analytical purposes only.
For purposes of this analysis, PHMSA divided the costs for this requirement into two sections: (1)
notification and development, and (2) periodic updates and submission. We did not receive public
comment concerning the information sharing estimates. Aside from accommodating more recent
https://www.bsee.gov/sites/bsee_prod.opengov.ibmcloud.com/files/final_2016_prep_guidelines.pdf (pg. 2-5)
Bureau of Labor Statistics. Occupational Employment and Wages, May 2016; Wage Series 53-4099 Rail
Transportation Workers, all other. Retrieved from: http://www.bls.gov/oes/current/oes534099.htm
69
70
Page 47
data on salaries, our methods for generating the information sharing estimates remain the same as
in the NPRM.
2.2.1. Notification Development
The information required to be shared must consist of the following:
•
A reasonable estimate of the number of HHFTs that the railroad expects to operate each
week, through each county within the State or through each tribal jurisdiction;
•
The routes over which the HHFTs will operate;
•
A description of the hazardous material being transported and all applicable emergency
response information required by subparts C and G of part 172 of this subchapter;
•
HHFT point of contact: at least one point of contact at the railroad (including name, title,
phone number and address) related to the railroad’s transportation of affected trains;
•
If a route is additionally subject to the comprehensive spill plan requirements, the
notification must include a description of the response zones (including counties and states)
and contact information for the qualified individual and alternate, as specified under section
130.104(a); and
•
Each point of contact must be clearly identified by name or title and role (e.g. qualified
individual, HHFT point of contact) in association with the telephone number. One point
of contact may fulfill multiple roles.
As discussed in Section 1.4.2 of this document, PHMSA estimates that 131 Class II and Class III
railroads would be subject to this requirement. Recall that 40 additional railroads, including all of
the Class I railroads, have already developed similar notifications, so the development costs for
these railroads were attributed to the Emergency Order, and not this final rule.
The provisions included in this final rule are fundamentally similar to those required under the
Emergency Restriction/Prohibition Order in Docket No. DOT-OST-2014-0067 (Order). 71
Therefore, for purposes of developing an estimate for the cost associated with this provision,
PHMSA used the same burden estimate developed for the ICR supporting statement for this
Emergency Order. 72 This supporting statement estimated that it will take approximately 30 hours
to complete each notification and send it to the appropriate SERC. In addition, the supporting
statement includes an assumption that each railroad subject to the requirement will develop
approximately 2.5 notifications.
Applying these estimates to the universe of affected entities for this final rule, PHMSA assumes
that approximately 328 notifications will be developed (2.5 x 131). PHMSA assumes that the
notification will be completed by a senior employee or manager, and uses a salary estimate of
$75.53.
http://www.dot.gov/briefing-room/emergency-order.
“Information Collection Supporting Statement Secretary’s Emergency Order Docket No. DOT-OST-2014-0067”
OMB Control No. 2130-0604. http://www.reginfo.gov/public/do/PRAViewICR?ref_nbr=201405-2130-003
71
72
Page 48
Based on 30 hours per notification, the total cost associated with this provision is estimated to be
$742,087 ($75.53/hour × 30 hours/notification × 327.5 notifications). In addition to sending the
notification to the appropriate SERC, railroads will also have to send the notification to the
appropriate TERC, as applicable. PHMSA assumes, however, that this addition will require
minimal additional burden and maintains the 30-hour estimate developed for the Emergency
Order.
2.2.2. Periodic Updates and Submission
In addition to the development of the notifications, railroads subject to the requirement must also:
• Update the notifications for changes in volume greater than 25 percent.
• Notifications and updates may be transmitted electronically or by hard copy.
• Copies of HHFT notifications made must be made available to the Department of
Transportation upon request.
PHMSA estimates that 171 Class II and Class III railroads would be subject to these requirements.
As with notification development, PHMSA evaluated burden estimates used in the Emergency
Order (EO) supporting statement to estimate the costs associated with these information-sharing
provisions. 73
For submission to DOT, the supporting statement estimates that approximately 8.33 percent of the
notifications would be subject to a request from DOT annually and that each request would require
1 hour of the railroad’s time per submission. PHMSA assumes that this action would be carried
out by a senior staff member or manager, and uses a $75.53 salary estimate. PHMSA estimates
that there will be approximately 428 notifications (171 railroads × 2.5 notifications) and that 36 of
those notifications will be requested by DOT each year (8.33 percent × 428 notifications). Based
on these assumptions, the annual cost for notification submissions to DOT would be $2,691
($75.53/hour × 1 hour × 36 notifications). The per-railroad cost calculation for submission to DOT
is somewhat abstract, but it would be approximately 0.21 notifications per railroad (8.33 percent
x 2.5 notifications = 0.21), meaning the per railroad annual cost would be approximately $15.73,
or $16.
The requirement in the Emergency Order for the railroad to update the notification, if the railroad
materially changed—defined as any increase or decrease of 25 percent or more—the estimated
volumes per week or frequencies of trains per week traveling through local communities, was
adopted in the rulemaking. The estimate for this provision was 4 hours per update in the E.O.
However, PHMSA assumes that the periodic updates will not require as much time as the updates
required by the EO. For purposes of this analysis, PHMSA assumes that the periodic update will
require 1 hour of a senior employee’s or manager’s time per month. To further evaluate this
assumption, PHMSA included an alternative estimate in Section 2.4.2, “Sensitivity Analysis,”
using the 4-hour estimate from the EO. It is uncertain how often an affected railroad would need
to update their notification, but we assume that each affected railroad would need to do so on a
monthly basis. This is an upper-bound estimation of the cost impact for this aspect of the
“Information Collection Supporting Statement Secretary’s Emergency Order Docket No. DOT-OST-2014-0067”
OMB Control No. 2130-0604. http://www.reginfo.gov/public/do/PRAViewICR?ref_nbr=201405-2130-003
73
Page 49
provision; many affected railroads may need to update less frequently, if their crude-by-rail
volumes do not change often.
Based on an estimate of 1 hour per update, and an assumption of one update per month, the monthly
cost for updates will be $32,289 (427.5 notifications × 1 hour/notification × $75.53/hour).
PHMSA assumes that the first year of the analysis would require 11 updates, and future years
would require 12 updates per year.
Table 12 provides an overview of the notification development, submission, and update costs per
railroad.
Table 12. Year 1 and Years 2–10 Average Costs per Railroad for Information Sharing
Year
Year 1
Years 2–10
Average
Requirement
Railroads
Affected
Unit Cost Per Railroad
Development
Updates
Submission to DOT
Updates
Submission to DOT
131
171
171
171
171
$5,665
$2,077
$16
$2,266
$16
Total Cost
per Railroad
Total Cost,
All Railroads
$7,758*
$1,099,959
$2,365
$404,440
*For example, the initial annual development cost for information sharing per railroad is estimated at $5,665 = 30
hours/notification × 2.5 notifications × $75.53/hour. The initial annual cost with monthly updates is estimated at
$2,077 = 2.5 notifications × 1 hour/notification/month × $75.53/hour × 11 months. Thus, the total initial annual cost
per railroad is estimated at $7,758 = $5,665 + $2,077 + $16.
2.2.3. Cost to the Federal Government
The final rule would require railroads to submit their OSRPs to PHMSA for review and approval.
In addition, any significant changes to the plan also would have to be submitted to PHMSA.
Specifically, under the final rule, railroads would be required to update their plan to address new
or different conditions or information, and each railroad must review its plan in full at least every
5 years from the date of the last approval. If new or different operating conditions or information
would substantially affect the implementation of the response plan, the railroad must immediately
modify its plan to address such a change and must submit the change to PHMSA within 90 days.
While it is likely that not all railroads will need to submit an updated plan to PHMSA after each
review period, PHMSA does not have information available to estimate the likelihood and timing
of updates requiring a submission. PHMSA assumes the review and approval of these plans would
require four full-time employees (FTE) after the initial plan submission (year 1) for railroads
subject to the OSRP requirement. In addition, 1 FTE would be required in the remaining years
(years 2-10, of this analysis) for the review of the updated plans submitted by the railroads.
For estimating the incremental costs of the review and approval of the ORSPs, PHMSA used
annual wage data from the Office of Personnel Management (OPM) to estimate wages for its staff
at the 2017 General Schedule (GS) level 14, step 1, wage class for the Washington-BaltimoreNorthern Virginia metropolitan area. 74 In accordance with the OMB Circular No. A-76 (M-07-02;
74
U.S. Office of Personnel Management. (2015). “2015 General Schedule (GS) Locality Pay Tables.”
Page 50
2006), PHMSA included a load factor of 36.45 percent for the Federal wage to account for fringe
benefits. 75 Finally, the Agency projected the wage forward using a 1-percent wage growth rate.
Table 13 presents an overview of the total cost to the Federal Government for the review and
approval of the OSRPs. The undiscounted total cost to the Federal Government is estimated to be
$2,053,943 for the 10-year period evaluated in this analysis. Using a 3-percent discount rate, the
total cost to the Federal Government is estimated to be $1,858,170, and using a 7-percent discount
rate, the total cost is estimated to be $1,653,464.
Table 13. Total Cost to Federal Government for Plan Review and Approval (undiscounted)
Year
Number
of FTE
1
4
2
1
3
1
4
1
5
1
6
1
7
1
8
1
9
1
10
1
Salary + Fringe
and Overhead
$153,973
$155,104
$156,247
$157,401
$158,567
$159,744
$160,933
$162,134
$163,347
$164,573
Total
Total Cost
$615,891
$155,104
$156,247
$157,401
$158,567
$159,744
$160,933
$162,134
$163,347
$164,573
$2,053,943
2.2.4. Information Sharing
The rule would require railroads to submit their notifications to DOT upon request. FRA estimated
in the ICR supporting for the Emergency Restriction/Prohibition Order in Docket No. DOT-OST2014-0067 76 that the review of each notification would require 1 hour of a GS-12, step 5
employee’s time. 77
PHMSA used hourly wage data from the Office of Personnel Management to estimate wages for
its staff at the 2017 GS level 12, step 5, using the wage class for the Washington-BaltimoreNorthern Virginia metropolitan area. 78 We converted this salary to an hourly wage, and estimated
an hourly wage. In accordance with the revised OMB A-76 (M-07-02; 2006), PHMSA included
Washington, DC. Retrieved from https://www.opm.gov/policy-data-oversight/pay-leave/salaries-wages/salarytables/pdf/2017/DCB_h.pdf
75
https://www.whitehouse.gov/sites/whitehouse.gov/files/omb/memoranda/2007/m07-02.pdf
76
http://www.dot.gov/briefing-room/emergency-order.
77
“Information Collection Supporting Statement, Secretary’s Emergency Order Docket No. DOT-OST-2014-0067”
OMB Control No. 2130-0604. http://www.reginfo.gov/public/do/PRAViewICR?ref_nbr=201405-2130-003
78
U.S. Office of Personnel Management. (2015). “2015 General Schedule (GS) Locality Pay Tables.”
Washington, DC. Retrieved from: https://www.opm.gov/policy-data-oversight/pay-leave/salaries-wages/salarytables/pdf/2017/DCB_h.pdf
Page 51
a load factor of 36.45 percent for the Federal wage to account for fringe benefits. 79 Finally, the
Agency projected the wage forward using a 1-percent wage growth rate.
Table 14 presents an overview of the total cost to the Federal Government for the review of
notifications. The undiscounted total cost to the Federal Government is estimated to be $22,148
for the 10-year period evaluated in this analysis. Using a 3-percent discount rate, the total cost to
the Federal Government is estimated to be $19,425, and using a 7-percent discount rate, the total
cost is estimated to be $16,577.
Table 14. Total Cost to Federal Government for Review of Notifications (undiscounted)
Year
1
2
3
4
5
6
7
8
9
10
Total
Hourly Salary (GS-12, step 5)
$59.50
$59.94
$60.38
$60.83
$61.28
$61.73
$62.19
$62.66
$63.12
$63.60
Number of Notifications
36
36
36
36
36
36
36
36
36
36
360
Annual Total
$2,142
$2,158
$2,174
$2,190
$2,206
$2,222
$2,239
$2,256
$2,272
$2,290
$22,148
Over the 10-year period, costs to the Federal government for both OSRPs and information
sharing are estimated to total approximately $2,076,091. The costs to the Federal government
for the approval of OSRPs and review of information sharing notifications are also reflected in
the “Total Costs” section of this document.
2.3. Total Costs
Based on the cost components described in sections 2.1 and 2.2, PHMSA estimates that the
undiscounted total cost over the 10-year period for this rule would be $32.3 million. Using a
discount rate of 3 percent, the total cost would be $28.9 million, and using a discount rate of 7
percent, the total cost would be $25.2 million.
Table 15 provides a summary of the total estimated costs associated with the final rule by railroad
class, including the undiscounted and discounted (3 percent and 7 percent) 10-year and annualized
costs.
79
https://www.whitehouse.gov/sites/whitehouse.gov/files/omb/memoranda/2007/m07-02.pdf
Page 52
Table 15. Summary of Undiscounted and Discounted Total and Annualized Costs (millions)
Class of
Railroad
Undiscounted
10-Year
Annualized
3% Discount Rate
10-Year
Annualized
7% Discount Rate
10-Year
Annualized
Oil Spill Response Plans
$6.30
$4.0
$15.2
Class I
Class II
Class III
$0.6
$0.4
$1.5
$5.6
$3.6
$13.5
$0.7
$0.4
$1.6
$4.9
$3.1
$11.8
$0.7
$0.4
$1.7
$0.5
$3.7
$0.5
Information Sharing
All Railroads
$4.7
$0.5
$4.2
Cost to Government
Government
Costs
Total
$2.1
$0.2
$1.9
$0.2
$1.7
$0.2
$32.3
$3.2
$28.9
$3.4
$25.2
$3.6
Table 16 provides a summary of the annualized costs, using a 7-percent discount rate only,
differentiated by each of the provisions evaluated and by railroad class.
Table 16. Summary of Annualized Costs by Provision (7% Discount Rate)
Category
Class of
Railroad
Class I
Plan Development
Class II
Class III
Class I
Plan Maintenance
Class II
Class III
Class I
Plan Submission
Class II
Class III
Class I
OSRO Fee
Class II
Class III
Class I
Training and Exercises
Class II
Class III
Information Sharing
All Railroads
Annualized Cost (7%)
$115,555
$60,529
$201,763
$65,243
$34,175
$113,916
$37
$58
$289
$299,600
$70,620
$147,125
$221,364
$282,035
$1,212,707
$530,567
As discussed in Section 1.6, PHMSA evaluated the impacts of this rulemaking over a 10-year
period. Table 17 provides an overview of the undiscounted costs for the private entities that would
be affected by the final rule over the 10-year period for each provision included in this analysis.
Page 53
Table 17. Summary of 10-Year Costs by Provision (undiscounted) (millions)
Year
1
2
3
4
5
6
7
8
9
10
Plan
Development
Plan
Maintenance
Plan
Submission
$1.9
$0.0
$0.0
$0.0
$0.0
$1.0
$0.0
$0.0
$0.0
$0.0
$0.2
$0.2
$0.2
$0.2
$0.2
$0.2
$0.2
$0.2
$0.2
$0.2
$0.0
$0.0
$0.0
$0.0
$0.0
$0.0
$0.0
$0.0
$0.0
$0.0
OSRO
Fees
Training and
Exercises
$0.5
$0.5
$0.5
$0.5
$0.5
$0.5
$0.5
$0.5
$0.5
$0.5
$2.5
$1.3
$1.3
$1.3
$1.3
$2.5
$1.3
$1.4
$1.4
$1.4
Information
Sharing
Cost to
Government
$1.1
$0.4
$0.4
$0.4
$0.4
$0.4
$0.4
$0.4
$0.4
$0.4
$0.6
$0.2
$0.2
$0.2
$0.2
$0.2
$0.2
$0.2
$0.2
$0.2
Table 18 provides an overview of the undiscounted cost for each cost category evaluated in this
analysis, differentiated by class of railroad.
Table 18. Undiscounted Unit Cost per Railroad by Railroad Class
Category
Plan
Development
Outlay
Period
Class of Railroad
Class I
Year 1
Class II
Class III
Class I
Plan
Maintenance
Annual
Plan
Submission
Once every 5
years
Class II
Class III
Class I
Class II
Class III
Class I
OSRO Fee
Annual
Class II
Class III
Training and
Exercises
Information
Sharing
Class I
Year 1
Class II
Class III
Year 1
All Railroads
Annual
All Railroads
Unit Cost Per
Railroad
$84,666*
$28,222*
$18,815*
$8,745
$2,915
$1,943
$21
$21
$21
$40,000
$6,000
$2,500
$66,475**
$42,305**
$27,803**
$7,758
$2,365
*This cost represents the plan development cost per railroad in the implementation year (year 1).
The final rule requires each railroad to review its plan at least every 5 years from the date of the
last approval. PHMSA estimates the recurring burden with reviewing the initial plan is half of the
burden needed to develop the initial plan development cost.
Page 54
Total
$6.8
$2.6
$2.6
$2.6
$2.6
$4.8
$2.6
$2.6
$2.6
$2.6
**This cost represents training and exercise costs in the implementation year (year 1). Subsequent
years have different costs due to different frequencies applicable to the training/exercise
requirements.
2.4. Cost Uncertainty and Sensitivity Analysis
2.4.1. Cost Uncertainty
In previous sections, we presented the estimated costs of the various provisions required in this
rulemaking. It is important to note areas where there are uncertainties regarding the cost
estimates, and where costs may be over- or underestimated. Table 19 below presents some areas
where cost uncertainties exist and whether the estimates used by PHMSA are likely to result in
overestimation or underestimation of costs.
Page 55
Table 19. Cost Uncertainties – Plan Development, Maintenance, OSRO Fees, Training, and Exercises
Cost Area
Direction of Uncertainty
Underestimation Overestimation Unknown
Plan
Development
X
Plan
Maintenance
X
OSRO Fees
X
Brief Explanation of Cause of Uncertainty
The Agency believes it is likely that many of the railroads that would be subject to
the final rule, particularly Class I and Class II railroads, have already developed
response plans that include similar provisions to what is included in the
comprehensive plans. These railroads would only need to update their plans to be
consistent with the requirements and may not require the entire time estimated for
plan development. However, AAR/ASLRRA submitted a comment to the ANPRM
suggesting that the plan development costs would be higher than PHMSA estimated
in the preliminary RIA. Also, AAR and API indicated in comments to the NPRM
that plan development costs were underestimated in the preliminary RIA. PHMSA
revised our estimates upwards in the final RIA, despite the lack of data submitted in
support of these commenters’ claims. Since there are reasons indicating both underand overestimation of plan development costs, we ultimately note “Unknown” for
this cost area.
The Agency assumes that railroads will have to make revisions to their plans
annually. However, revisions are only necessary if there are different conditions or
new information applicable to the plan, so there may be several railroads that do not
need to update their plans on an annual basis. API indicated that the costs to review
plans annually may be higher than estimated in the preliminary RIA. PHMSA
revised our estimate upwards in response. Nevertheless, based on the factors
identified, it is not known whether annual COSRP maintenance will be higher or
lower than estimated in the final RIA.
The Agency has limited information on the retainer fees railroads may incur to
contract with an OSRO. To be conservative, the Agency estimated that all railroads
subject to the plan would need an annual retainer fee even though we believe many
railroads, particularly Class I railroads, already have agreements in place with spill
response providers; however, there is uncertainty over the potential need for and cost
of new agreements. Retainer fees may not be needed at all in some cases, as some
OSROs may be willing to contract with a plan holder (railroad) simply on the basis
that their services will be solicited for future spills. Additionally, the costs of
existing contracts / agreements cannot be attributed to this final rule. We did not
receive public comment to the NPRM concerning our preliminary estimates. For
these reasons, we believe the OSRO fee costs are overestimated.
Page 56
Cost Area
Training and
Exercises
Total Cost
Direction of Uncertainty
Underestimation Overestimation Unknown
X
X
Brief Explanation of Cause of Uncertainty
The Agency has limited information on the number and types of employees that will
require training as a result of the final rule and the amount of time that training will
require. The railroad has discretion on how they want to administer the training
requirements, and the Agency believes it is likely that the response plan training will
be incorporated into other existing training requirements (e.g., Subpart H of Part
172—Training). Further, other Federal agencies may offer training funding for
which oil spill response training may qualify. Nevertheless, if a new training regime
needs to be developed, it may take longer than the estimated 8 hours per employee
considered in this analysis. API, in particular, noted in public comment that training
costs may be underestimated. However, we did not receive data with which to revise
our training costs estimates. We acknowledge training costs might possibly be
underestimated.
Several States have current or pending legislation or regulatory requirements related
to oil spill response planning for railroads. The Agency has limited information on
current compliance with these State requirements, but believes there will likely be
overlap between what is required at the State and Federal levels. Therefore, the
Agency believes several railroads may already be in compliance with some or part
of the comprehensive requirements based on State regulation, and will not be subject
to all of the costs included in this analysis. Further, many railroads have voluntarily
implemented oil spill response plans, and the costs of implementing voluntary
programs cannot be attributed to this rule. In the absence of rulemaking, many
railroads are proactively preparing for oil spills and seeking ways to improve their
spill preparedness and response.
Page 57
2.4.2. Sensitivity Analysis
To evaluate some of the cost uncertainties mentioned above, PHMSA conducted sensitivity
analyses on key input variables used in this analysis, and provided updated estimates for the total
cost of the rule. In Table 26 at the end of this section, PHMSA provides an overview of the low,
medium/selected, and high estimates.
2.4.2.1. OSRO Fees – Class III Railroads
As part of the evaluation of OSRO retainer fees, PHMSA has some examples of a “zero retainer
fee” policy for OSRO services. 80,81 For example, in the survey conducted with OSROs in
development of the NPRM and preliminary RIA, one company said that they would likely not
charge a retainer fee for Class III railroads. 82 For this sensitivity analysis, PHMSA adjusted the
retainer fees for Class III railroads from $2,500 to $0. Table 20 presents the revised undiscounted
costs by provision, and Table 21 presents the revised undiscounted costs by railroad class. This
adjustment would decrease the total costs to Class III railroads by $1,375,000 over the 10-year
period of this analysis.
Table 20. Total Undiscounted Costs by Provision with Adjusted Class III OSRO Fee
(millions)
Year
Plan
Development
Plan
Submission
OSRO
Fees
$0.2
$0.2
$0.2
$0.2
$0.2
$0.2
$0.2
$0.2
$0.2
$0.2
$0.0
$0.0
$0.0
$0.0
$0.0
$0.0
$0.0
$0.0
$0.0
$0.0
$0.3
$0.3
$0.3
$0.3
$0.3
$0.3
$0.3
$0.3
$0.3
$0.3
$1.9
$0.0
$0.0
$0.0
$0.0
$1.0
$0.0
$0.0
$0.0
$0.0
1
2
3
4
5
6
7
8
9
10
Training,
Exercises, and
Drills
Plan
Maintenance
Information
Sharing
$2.5
$1.3
$1.3
$1.3
$1.3
$2.5
$1.3
$1.4
$1.4
$1.4
Cost to
Goverment
$1.1
$0.4
$0.4
$0.4
$0.4
$0.4
$0.4
$0.4
$0.4
$0.4
Total
$0.6
$0.2
$0.2
$0.2
$0.2
$0.2
$0.2
$0.2
$0.2
$0.2
Table 21. Total Undiscounted Costs by Railroad Class with Adjusted Class III OSRO Fee
(millions)
Year
1
2
OSRPs
Class I
$1.4
$0.5
Class II
$0.9
$0.3
Information Sharing
Class III
$2.7
$1.1
All Railroads
$1.1
$0.4
Costs to
Government
$0.6
$0.2
For an example, see: http://petrochemrecovery.com/Emergency-Spill-Response-Management.html
Also see the preliminary RIA, Appendix E, question 5a.
82
Ibid.
80
81
Page 58
Total
$6.7
$2.4
$6.7
$2.4
$2.4
$2.4
$2.4
$4.6
$2.5
$2.5
$2.5
$2.5
Year
OSRPs
Class I
$0.5
$0.5
$0.5
$1.1
$0.5
$0.5
$0.5
$0.5
3
4
5
6
7
8
9
10
Class II
$0.3
$0.3
$0.3
$0.7
$0.3
$0.3
$0.3
$0.3
Information Sharing
Class III
All Railroads
$1.1
$1.1
$1.1
$2.2
$1.1
$1.1
$1.1
$1.1
$0.4
$0.4
$0.4
$0.4
$0.4
$0.4
$0.4
$0.4
Costs to
Government
$0.2
$0.2
$0.2
$0.2
$0.2
$0.2
$0.2
$0.2
Total
$2.4
$2.4
$2.4
$4.6
$2.5
$2.5
$2.5
$2.5
2.4.2.2. Training Requirements
The final rule would require all employees subject to the plan to receive initial training to ensure
that the applicable employees know their responsibilities under the plan, the plan content, and the
notification procedures. This training must be conducted at least once every 5 years. PHMSA
estimated this training would take 8 hours per employee in years 1, 5 and 10. However, subject
matter experts at FRA suggested that this requirement may take up to 40 hours to complete. To
evaluate that possibility, PHMSA adjusted the hours of training from 8 hours to 40 hours per
applicable employee. Table 22 provides an overview of the revised undiscounted cost estimate by
provision, and Table 23 provides an overview of the revised undiscounted costs by railroad class.
Under this scenario, the total costs would increase by $9,252,873 over the 10-year period of this
analysis.
Table 22. Total Undiscounted Costs by Provision with Adjusted Training Hours (millions)
Year
1
2
3
4
5
6
7
8
9
10
Plan
Development
Plan
Maintenance
Plan
Submission
OSRO
Fees
Training and
Exercises
Information
Sharing
Cost to
Government
Total
$1.9
$0.2
$0.0
$0.5
$7.0
$1.1
$0.6
$11.3
$0.0
$0.2
$0.0
$0.5
$1.3
$0.4
$0.2
$2.6
$0.0
$0.2
$0.0
$0.5
$1.3
$0.4
$0.2
$2.6
$0.0
$0.2
$0.0
$0.5
$1.3
$0.4
$0.2
$2.6
$0.0
$0.2
$0.0
$0.5
$1.3
$0.4
$0.2
$2.6
$1.0
$0.2
$0.0
$0.5
$7.2
$0.4
$0.2
$9.5
$0.0
$0.2
$0.0
$0.5
$1.3
$0.4
$0.2
$2.6
$0.0
$0.2
$0.0
$0.5
$1.4
$0.4
$0.2
$2.7
$0.0
$0.2
$0.0
$0.5
$1.4
$0.4
$0.2
$2.7
$0.0
$0.2
$0.0
$0.5
$1.4
$0.4
$0.2
$2.7
In addition, total costs for Class I railroads would increase by $2,756,175, Class II would increase
by $2,165,566, and Class III would increase by $4,331,132 over the 10-year period of this analysis.
Page 59
Table 23. Total Undiscounted Costs by Railroad Class with Adjusted Training Hours
(millions)
Year
1
2
3
4
5
6
7
8
9
10
Oil Spill Response Plans
Class I
$2.8
$0.5
$0.5
$0.5
$0.5
$2.5
$0.5
$0.5
$0.5
$0.5
Class II
$1.9
$0.3
$0.3
$0.3
$0.3
$1.8
$0.3
$0.3
$0.3
$0.3
Information Sharing
Class III
Cost to
Government
All Railroads
$4.9
$1.2
$1.2
$1.3
$1.3
$4.6
$1.3
$1.3
$1.3
$1.3
$1.1
$0.4
$0.4
$0.4
$0.4
$0.4
$0.4
$0.4
$0.4
$0.4
$0.6
$0.2
$0.2
$0.2
$0.2
$0.2
$0.2
$0.2
$0.2
$0.2
Total
$11.3
$2.6
$2.6
$2.6
$2.6
$9.5
$2.6
$2.6
$2.6
$2.6
2.4.2.3. Periodic Notifications
The final rule would require railroads to update their notifications when changes in volume are
greater than 25 percent. This aligns with the Emergency Order supporting statement. 83 The
Emergency Order included a requirement to update the notification if the railroad made any
material changes—defined as any increase or decrease of 25 percent or more—in the estimated
volumes per week or frequencies of trains per week traveling through local communities. While
the final rule only requires an update when changes in volume are greater than 25 percent, for
purposes of the analysis, we estimate that affected railroads would develop updates on a monthly
basis. As such, it is an upper-bound approximation of the potential costs that affected railroads
would face to update their information sharing notifications to SERCs, etc. The estimate for this
part of the information sharing provision was 4 hours per update in the E.O. supporting statement.
In our primary estimate for this analysis, we used 1 hour per notification. To address this
difference, we offer a sensitivity analysis showing the expected increase in costs if each update
required 4 hours, rather than 1 hour.
For this sensitivity analysis, PHMSA adjusted the estimate from 1 hour to 4 hours per notification
per railroad. Table 24 provides an overview of the revised undiscounted costs by provision, and
Table 25 provides an overview of the revised undiscounted costs by railroad class. Under this
scenario, the total cost would increase by $11,910,117 over the 10-year period of this analysis.
“Information Collection Supporting Statement Secretary’s Emergency Order Docket No. DOT-OST-2014-0067”
OMB Control No. 2130-0604. http://www.reginfo.gov/public/do/PRAViewICR?ref_nbr=201405-2130-003
83
Page 60
Table 24. Total Undiscounted Costs by Provision with Adjusted Periodic Notifications
(millions)
Yea
r
1
2
3
4
5
6
7
8
9
10
Plan
Developme
nt
Plan
Maintenanc
e
Plan
Submissio
n
OSR
O
Fees
$1.9
$0.0
$0.0
$0.0
$0.0
$1.0
$0.0
$0.0
$0.0
$0.0
$0.2
$0.2
$0.2
$0.2
$0.2
$0.2
$0.2
$0.2
$0.2
$0.2
$0.0
$0.0
$0.0
$0.0
$0.0
$0.0
$0.0
$0.0
$0.0
$0.0
$0.5
$0.5
$0.5
$0.5
$0.5
$0.5
$0.5
$0.5
$0.5
$0.5
Trainin
g/
Exercise
s
Informatio
n Sharing
Cost to
Governme
nt
$2.2
$1.6
$1.6
$1.6
$1.6
$1.6
$1.6
$1.6
$1.6
$1.7
$0.6
$0.2
$0.2
$0.2
$0.2
$0.2
$0.2
$0.2
$0.2
$0.2
$2.5
$1.3
$1.3
$1.3
$1.3
$2.5
$1.3
$1.4
$1.4
$1.4
Total
$7.2
$3.6
$3.6
$3.6
$3.6
$5.8
$3.7
$3.7
$3.7
$3.7
Table 25. Total Undiscounted Costs by Railroad Class with Adjusted Periodic Notifications
(millions)
Year
1
2
3
4
5
6
7
8
9
10
Oil Spill Response Plans
Class I
$1.4
$0.5
$0.5
$0.5
$0.5
$1.1
$0.5
$0.5
$0.5
$0.5
Class II
$0.9
$0.3
$0.3
$0.3
$0.3
$0.7
$0.3
$0.3
$0.3
$0.3
Information Sharing
Class III
$2.8
$1.2
$1.2
$1.3
$1.3
$2.3
$1.3
$1.3
$1.3
$1.3
All Railroads
$2.2
$1.6
$1.6
$1.6
$1.6
$1.6
$1.6
$1.6
$1.6
$1.7
Costs to
Government
$0.6
$0.2
$0.2
$0.2
$0.2
$0.2
$0.2
$0.2
$0.2
$0.2
Total
$7.2
$3.6
$3.6
$3.6
$3.6
$5.8
$3.7
$3.7
$3.7
$3.7
2.4.2.4. Combined Results
Table 26 provides a summary of the sensitivity analysis, including the undiscounted 10-year total
and annualized cost by railroad class. The “low scenario” is the total cost including the downward
adjustment of the OSRO fees for Class III railroads. The “medium scenario” is the total cost using
the primary estimate described in the RIA, and finally the “high scenario” is the total cost including
the upward adjustments for the hours of training and information sharing updates.
Page 61
Table 26. Combined Results: Undiscounted Total Costs under High, Medium, and Low
Scenarios (millions)
Class of
Railroad
Class I
Class II
Class III
All
Railroads
Federal
Governm
ent Costs
Total
Low Scenario
Medium Scenario
10 Year
10 Year
10 Year
10 Year
Total (3%)
Total (7%)
Total (3%)
Total (7%)
Oil Spill Response Plans
$5.6
$4.9
$5.6
$4.9
$3.6
$3.1
$3.6
$3.1
$12.3
$10.7
$13.5
$11.8
Information Sharing
$4.2
$1.9
$27.7
$3.7
$4.2
Costs To Government
$1.7
$24.2
$1.9
$28.9
Page 62
High Scenario
10 Year
10 Year
Total (3%)
Total (7%)
$8.2
$5.6
$17.6
$7.3
$5.0
$15.5
$3.7
$14.7
$12.6
$1.7
$25.2
$1.9
$47.9
$1.7
$42.0
3. Benefits
The OSRP requirements are designed to reduce the magnitude and severity of spills and improve
spill response, thereby reducing the environmental and other damages that spills may cause.
PHMSA faced data uncertainties that limited our ability to estimate, quantitatively, the benefits of
the final rule. This has been a continual challenge for PHMSA’s analysis of this rule, such that in
the preliminary RIA, PHMSA asked, “Are there quantifiable benefits to the proposed rule that
PHMSA has not identified? Can these benefits be monetized? How so?”84 After reviewing public
comment, we did not receive any new input on this issue, specifically new quantitative methods
or data to explore. As such, PHMSA has updated our breakeven analysis to identify the number
of gallons of oil that the final rule would need to prevent from being spilled in order for the final
rule’s benefits to at least equal its estimated costs. The breakeven analysis assumes the average
social cost of each gallon spilled is $218. 85 We explain the choice of this value for the cost per
gallon spilled in greater detail in the HM-251 RIA and preliminary RIA for the NPRM preceding
this final rule. 86 Additional benefits may also be incurred due to ecological and human health
improvements that may not be captured in the value of the avoided cost of spilled oil. These
benefits, which are difficult to quantify, are discussed further in Section 3.3 on a qualitative basis.
3.1. Baseline and Assumptions for Benefits Assessment
3.1.1. Baseline Incidents
To assess the baseline conditions that would be affected by the final rule, PHMSA evaluated data
provided in the Hazardous Material Incident Report Database. 87 Specifically, PHMSA evaluated
reported incidents that occurred in the United States from 2009–2017 involving liquid petroleum
transported by rail, and filtered them for only those incidents that likely involved HHFTs and
occurred on mainline track.
Figure 3 provides a count of the number of rail incidents involving liquid petroleum reported each
year from 2009 to 2016, along with crude oil carload volume for those years (2017 Waybill data
are not yet available).
See preliminary HM-251B RIA, pg. 90
Adjusted from $211 in June of 2015 to December 2017 dollars (the latest available) using the BLS CPI inflation
calculator, which can be found at https://data.bls.gov/cgibin/cpicalc.pl?cost1=25000&year1=198601&year2=201703
86
Final Regulatory Impact Analysis (RIA) – Hazardous Materials: Enhanced Tank Car Standards and Operational
Controls for High-Hazard Flammable Trains, Pipeline and Hazardous Materials Safety Administration (PHMSA),
May 2015, available at: https://www.regulations.gov/document?D=PHMSA-2012-0082-3442
87
PHMSA Office of Hazardous Materials (OHMS) Incident Reports Database Search, available at:
https://hazmatonline.phmsa.dot.gov/IncidentReportsSearch/Welcome.aspx
84
85
Page 63
700,000
7
600,000
6
500,000
5
400,000
4
300,000
3
200,000
2
100,000
1
HHFT Derailments
Originating Carloads Per Year
Figure 3. Carloads of Crude Oil Shipped and Rail Accidents (Derailments), 2000–2016
0
0
2008
2009
2010
2011
2012
2013
2014
2015
2016
Year
Crude Carloads
HHFT Crude Derailments
Sources and notes: Originating Class I Carloads for 2000–2016 obtained from the Surface Transportation Board Waybill sample.
Derailments are from the PHMSA and FRA Incident Report Databases and include derailments that are not summarized in the
RIA’s Appendix B.
A comprehensive OSRP would be required to cover those routes/railroads that haul HHFTs
carrying petroleum oil. We restrict our consideration of baseline societal damages that might be
mitigated by the final rule to those incidents involving petroleum oil HHFTs. The Agency has
identified 15 such incidents from 2012 through 2017. These events are presented in the Table 27
below.
Table 27. Petroleum Oil HHFT Derailments, 2012–2017
Year
Nearest Town
State
Quantity Released
2013
Parkers Prairie
MN
15,000
2013
Aliceville
AL
455,520
2013
Casselton
ND
474,936
2014
New Augusta
MS
50,350
2014
Vandergrift
PA
9,800
2014
Lynchburg
VA
30,000
2014
Evans
CO
7,932
Page 64
2015
Galena
IL
110,543
2015
Mount Carbon
WV
362,349
2015
Heimdal
ND
98,090
2015
Culbertson
MT
27,201
2015
Watertown
WI
1,000
2016
Mosier
OR
42,448
2017
Money
MS
24,653
2017
Plainfield
IL
28,245
2016 volumes are still slightly higher than the volumes seen in 2012, and EIA predicts U.S. crude
oil production volumes to remain high for the next decade and beyond. As a result, we expect
volumes going forward to remain relatively high by historic (pre-2012) standards.
3.1.2. Event Forecasting
One simple way to predict the number of future events based on the HHFT period is as follows:
The period of high-volume crude shipments starts in 2012 through 2017, providing a 6-year period.
We consider a 10-year analysis period going forward, so the analysis period is 1.67 times longer
than the observed period. There were 15 incidents in the observed period, so the predicted number
of events over the analysis period would be 15 x 1.67 = 25 incidents over 10 years, or 2.5 incidents
per year.
The Agency acknowledges that this methodology is somewhat simple, but no adverse comments
regarding this methodology, or suggestions for improvement, were provided by comments to the
docket. As noted at the NPRM stage, this approach enables consideration of the most recent
incidents. Given the wide swings in annual crude oil volumes shipped by rail over the past few
years, predicting future volumes is difficult. As a result, an estimate based on a prediction of future
volumes, and rates of derailment per unit of volume shipped, would be highly uncertain and prone
to significant inaccuracy. This simplified method may therefore produce a forecast as accurate, or
more accurate, than a more sophisticated method.
For example, we could base our predicted number of events on known incidents per unit of known
volume and apply that to predicted volume. For 2012–2016, there were 13 HHFT derailments
observed in 2,087,384 carloads shipped. PHMSA recently produced multiple new carload
forecasts due to a FAST Act mandate to revise the HM-251 RIA to incorporate new testing results
for the ECP brake provision of that rule. Using the “low” (i.e. lower volume) forecast from that
recent assessment, for 2019–2028 (an approximation of the 10-year analysis period), the Agency
estimated approximately 4.17 million carloads forecast to be shipped. This represents a significant
reduction in forecast volume from the industry-provided forecast used in the 2015 HM-251 final
rule, which forecast roughly 9.8 million carloads shipped by rail over the same period. Table 28
provides an overview of the carload estimate per year and the number of HHFT derailments.
Carload data are not currently available for 2017, so we calculate the rate of derailment per
thousand carloads using 2012-2016. Applying the rate per thousand carloads from the table below
(.006227891 derailments per thousand carloads) to the forecast number of carloads to be shipped
(4,174.77 thousand) yields an estimated 26 derailments over 10 years. This figure is very close,
but slightly above, the figure generated using a per-year average figure, as described above.
Page 65
Table 28. Number of Derailments and Carloads Estimates
Total Volume
Volume
Carload
HHFT
Spilled
Spilled/Carload
Year
Estimate
Derailments
(gallons)
Shipped
0
0
2012
237,932
0
945,456
2.08
2013
454,873
3
576,581
4
98,082
0.17
2014
525,231
5
599,183
1.14
2015
292,767
1
42,448
.145
2016
Total
2,087,384
13
1,685,169
0.81
Derailments per thousand
0.006227891
carloads
3.1.3. Expected Size of Events
There are fifteen incidents for which data reporting are complete, which we use to generate an
average for volume spilled per derailment. 88 The incidents and quantity released are presented in
Table 27 above. The 15 incidents produced a total estimated 1,738,067 gallons of product spilled,
resulting in an average spill size of 115,871 gallons. This figure is multiplied by the forecasted
number of incidents to obtain an estimated total volume spilled for the 10-year analysis period:
115,871 x 25 = 2,896,778 gallons.
Release volumes are strongly correlated with the number of cars hauling hazardous flammable
liquids in a train consist – the functional relationship can be easily understood by considering a
stylized example. Take a 100-car train that experiences a derailment involving 10 cars, with the
probability that any derailed crude oil car will puncture of 50 percent, and a punctured car assumed
to release its entire contents. If this train were hauling 20 carloads of crude oil the expected number
of crude oil cars that would derail is 2 and one car would puncture and release its entire contents,
for an expected release volume of 30,000 gallons. If we consider an identical derailment of a train
hauling 40 carloads of crude oil, we would expect 4 of the derailed cars to be hauling crude oil,
with a potential 2 cars puncturing and releasing their contents, for a total spill volume of 60,000
gallons. A 10-car derailment involving a 100-car unit train of crude oil would experience 10
derailed cars containing crude oil and 5 cars releasing product, for a total release quantity of
150,000 gallons. As the number of cars hauling crude oil increases, we expect the quantity released
to also increase.
Since this final rule addresses worst-case discharges or threats of worst-case discharges, we focus
on HHFT incidents, which are expected to be more severe than incidents involving trains hauling
smaller quantities of crude oil. The comprehensive OSRP requirements in the final rule apply only
to carriers and routes on which trains carrying 20 or more carloads of crude oil in a block or 35 or
more total carloads in a train consist. Because carriers and routes on which smaller volumes of
Technically, entities reporting incidents have a year to update spill volumes and other data related to incidents, so
2017 event reporting is not “complete,” although we do not expect the data to change significantly with respect to
the quantity released.
88
Page 66
crude oil are shipped are not necessarily covered under this requirement, we restrict our analysis
to the events involving trains that meet this threshold, to ensure that we do not include events that
may not be mitigated by the requirements.
Since this rule covers oil spill response requirements, any event that results in no release of product
should be ignored when considering societal damages and benefits. In addition, the events that
occurred prior to 2012 occurred in a much different environment than those which occurred with
the advent of high volume shipments of crude oil by rail. Prior to 2012, the annual carloads of
crude oil moved by rail numbered in the tens of thousands. Starting in 2012, crude oil volumes
increased markedly varying from more than 100,000 carloads per year to more than half a million.
The Agency believes that since the final rule requirements only apply to carriers and routes on
which high volumes of crude are moved, considering only those events involving consists hauling
relatively high volumes is appropriate. As explained above, these events will likely result in higher
volumes spilled.
3.1.4. Cost per Gallon Spilled of Oil
For the purposes of monetizing crude oil HHFT derailments, we continue to use the value used in
the HM-251 rule, which was $218 per gallon. 89 The Agency received no adverse comments
regarding this monetization figure, which was based on costs reported to the Agency for pipeline
spills. For a full explanation of the derivation of this figure, the reader is referred to the HM-251B
preliminary RIA and the 2015 HM-251 final rule RIA.
3.1.5. Adjusting for HHFT Rule
The incident data used in this analysis are historical and do not take into account recent revisions
to PHMSA’s regulations related to enhanced tank car standards and operational controls for
HHFTs. This rulemaking was finalized in May 2015 and is under implementation. In December
of 2017, PHMSA and FRA released a FAST Act mandated re-assessment of the ECP brake
requirement of the HM-251 final rule. 90 As part of this re-assessment, the Agencies re-assessed
the fleet composition of both crude and ethanol fleets, produced a new forecast for the quantity of
crude and ethanol shipped by rail, and recalculated the effectiveness rates of upgrading flammable
liquid tank cars in crude and ethanol service given the current fleet composition. While the 2015
rule is not yet fully in effect, for purposes of establishing a baseline universe of incidents, PHMSA
has reduced the estimated average release of liquid petroleum based on the current composition of
the crude oil fleet, and yearly effectiveness rates developed using the crude oil tank car figures
presented in the recent update to the HM-251 RIA that was conducted due to the aforementioned
FAST Act mandate.
PHMSA developed this figure for the 2015 HM 251 final rule and the NPRM attached to the OSRP rulemaking.
At that time the Agency produced an estimate of $211 per gallon. PHMSA updated this figure for inflation by using
the BLS CPI inflation calculator set to June 2015 for and adjusted to December of 2017 (the most recent available).
This produced a value of $217.97, which was rounded to $218. The BLS inflation calculator can be found online at
https://data.bls.gov/cgi-bin/cpicalc.pl?cost1=211.00&year1=201506&year2=201712
90
See 82 FR 48006 for the Federal Register notice requesting comments on the updated HM-251 RIA, or PHMSA2017-0102-0014 at www.regulations.gov for a PDF of the updated HM-251 RIA.
89
Page 67
Adjustment for current fleet composition is necessary because the crude oil fleet is currently
composed of CPC-1232 and DOT-117 tank cars, not unimproved DOT-111 tank cars. 91 All the
events presented above from years 2014 and prior involved unimproved DOT-111 tank cars. All
the incidents in 2015 and forward involved CPC-1232 tank cars. In order to obtain a baseline
expected release quantity given the current fleet composition, we adjust all events that occurred in
2014 and prior by the weighted effectiveness of the crude oil tank car fleet as currently composed
compared with a fleet composed of unimproved legacy DOT 111s. This adjustment reduces
expected releases based on 2014 and prior year incidents to approximately 66.7 percent of the raw
recorded figure. For example, the Casselton, ND derailment is reduced from 474,936 to 316,676
gallons – the expected release rate of the incident given the fleet as currently composed as
compared to a fleet composed of legacy DOT 111 cars. That adjustment reduces the adjusted
quantity spilled in incidents through 2017 to 1,390,336 gallons (down from 1,738,067) and the
average expected release to 92,689 gallons (down from 115,871 gallons). This change effectively
reduces the spill size of past events involving DOT 111 legacy cars to the expected spill size for
those events assuming they had involved a CPC-1232 or DOT-117 tank car.
Having reduced the expected spill size to account for already realized improvements in the crude
oil tank car fleet, we apply the expected yearly improvement by multiplying spill damages by one
minus the effectiveness rates presented below. These rates represent the remaining expected tank
car fleet upgrades over the course of the HM-251 implementation schedule (given current fleet
composition, this is the improvement from converting remaining jacketed and non-jacketed CPC
1232 tank cars to DOT 117R tank cars). Table 29 presents these effectiveness rates by year
reflecting the ECP update for the rule as a whole and for crude oil only. 92
Table 29. Estimated Effectiveness for the HM-251 Final Rule, 2019–2028
Year
Crude Only HHFT Effectiveness Post ECP Repeal
1
18.19%
17.35%
16.50%
15.93%
14.46%
13.82%
13.66%
13.37%
13.37%
13.37%
2
3
4
5
6
7
8
9
10
91
“Fleet Composition of Rail Tank Cars That Transport Flammable Liquids: 2013-2016,” Bureau of Transportation
Statistics (BTS), September 2017, available at: https://cms.bts.dot.gov/sites/bts.dot.gov/files/docs/browse-statisticalproducts-and-data/surveys/annual-tank-car-facility-survey/208061/fleet-composition-rail-tank-cars-flammableliquids-sept-5-2017.pdf [hereinafter “2017 BTS Fleet Report”]
92
Derivation of these figures is presented on pages 287–291 of the HM-251 RIA.
Page 68
The percentages in Table 29 represent the percent reduction in severity attributed to the HM-251
rule given the degree of implementation for crude oil transport only. These rates also reflect the
retirement of virtually all legacy DOT-111 tank cars from the crude oil fleet over the past 2 years.
The crude oil-only figures differ from those for the rule as a whole because cars in crude oil service
are on a faster retrofit schedule than those used in ethanol service, and because the crude oil fleet
has a significantly different composition compared to the ethanol fleet. Therefore, the specific
effectiveness rates improve more quickly for crude by rail than for ethanol. At present, legacy
DOT-111s still make up a significant portion of the ethanol fleet, whereas they have virtually
disappeared from crude oil service. 93 Because the crude oil fleet has a mix of cars that have a
lower probability of release in the event of derailment, the maximum effectiveness rate is lower
for crude than for ethanol, because upgrading these better cars to the new standard results in
smaller decreases in the probability of release. We subtract the crude oil-specific figures from 1
to calculate percent damages remaining given implementation of the HM-251 rule, and multiply
that figure by total societal damages as described above to obtain final baseline damages from
which to draw benefits for this final rule. As the Department has announced its intent to repeal
the ECP brake provision of the 2015 HM-251 final rule, the effectiveness attributed to ECP brakes
has not been considered in these calculations. The total expected damages are presented in Table
30.
Table 30. Estimated Societal Damages from Crude Oil HHFT Derailments (millions)
Total
Events
Adjusted
Monetized
HHFT
Damages
Estimated
Year
per
Monetized
Value1
Effectiveness
Per Event Cost of Final
year
Value
Rule
18.19%
1
2.5
$50.5
$41.3
$16.5
$6.8
17.35%
2
2.5
$50.5
$41.7
$16.7
$2.6
16.50%
3
2.5
$50.5
$42.2
$16.9
$2.6
15.93%
4
2.5
$50.5
$42.5
$17.0
$2.6
14.46%
5
2.5
$50.5
$43.2
$17.3
$2.6
13.82%
6
2.5
$50.5
$43.5
$17.4
$4.8
13.66%
7
2.5
$50.5
$43.6
$17.4
$2.6
13.37%
8
2.5
$50.5
$43.8
$17.5
$2.6
13.37%
9
2.5
$50.5
$43.8
$17.5
$2.6
13.37%
10
2.5
$50.5
$43.8
$17.5
$2.6
7% discount
3% discount
$321.4
$376.6
$25.1
$28.8
1 Calculated by multiplying 92,689 (estimate of average gallons released per event, adjusted to account for
improvements in the tank car fleet) times $218 (estimate of societal cost per gallon released) times 2.5 (estimate of
events per year).
93
See 2017 BTS Fleet Report, pg. 6
Page 69
As can be seen, the estimated costs of the final rule are approximately 7.6 - 7.8 percent of expected
damages resulting from the events that the final rule could affect, depending on whether benefits
and costs are discounted at 7 or 3 percent. This means that if this final rule achieves a reduction
of 8 percent in the total consequences of these events, benefits will be approximate to costs. This
is the basis of our breakeven analysis.
The Agency uses the breakeven analysis approach because we have faced difficulties in estimating
an accurate effectiveness rate for the comprehensive rail OSRP program codified in this final rule.
This difficulty stems in part from the fact that observational data on a Federal comprehensive rail
OSRP program does not exist; the applicability of 49 CFR part 130 prior to this rule meant that no
railroads were required to have a comprehensive OSRP on a regulatory basis. As such, developing
a rail OSRP effectiveness rate that is data-driven would most likely require retrospective regulatory
review, wherein PHMSA-OHMS compares and analyzes the severity of incidents prior to and after
the implementation of comprehensive plans in accordance with our revised regulations and tries
to control for confounding variables. Both USCG and PHMSA-OPS used this strategy of
retrospective review to generate their effectiveness rates; USCG conducted an “OPA 90
Programmatic Regulatory Assessment” in 2001 and OPS developed the “Final Regulatory
Evaluation of the Response Plan Requirements for Transportation-Related Onshore Oil Pipelines”
in 2004 after issuing the OPS interim final rule that established pipeline OSRPs. Given the
research and efforts of PHMSA-OPS and USCG, we feel it is appropriate to use their effectiveness
rates in order to proactively generate a benefits outlook for this rule. Furthermore, we believe that
a rate substantially higher than 7-8 percent could reasonably be expected. Other Agencies
(PHMSA–OPS, USCG, and EPA) have used much higher estimated effectiveness rates for
promulgating comprehensive OSRP requirements for other industries. These estimates are further
discussed in the following section of this final RIA (Section 3.2). PHMSA did not receive public
comment on rail OSRP effectiveness rates despite attempts to solicit this input in Section 4.3 of
the preliminary RIA.
3.2. Effectiveness of OSRPs
Several elements contribute to the overall cost of an oil spill. For example, incidents may involve
costs associated with property damage, emergency response, evacuation of residents or workers in
the surrounding areas, environmental damage, and transportation delays while the spill is being
cleaned up. Factors such as the population density where the spill occurred, proximity to a
sensitive resource, or circumstances of the accident will also affect the potential costs associated
with a spill.
While there is no overt mechanism by which OSRPs would reduce the frequency of incidents
occurring, based on an evaluation of the implementation of OSRPs for other industries (e.g.,
commercial vessels, pipelines), PHMSA has determined that having a comprehensive plan will
likely reduce the severity of incidents that occur, thereby also reducing the associated damages. In
addition, the comprehensive OSRP requirement might potentially reduce the frequency of spills
indirectly by increasing situational awareness.
In 2001, after promulgating several regulations directed at oil spill prevention, mitigation, cleanup,
and liability in response to broad mandates contained in OPA 90, the USCG and the Volpe
National Transportation Systems Center prepared a Programmatic Regulatory Assessment to
Page 70
evaluate the combined benefit, cost, and cost effectiveness of OPA 90 regulations. 94 As a part of
that analysis, a panel of seven private-sector and USCG experts was asked to assess the effects of
each rule on four major oil spill events. Based on the results of that assessment, effectiveness
factors for reducing the severity of incidents were estimated for several key requirements (e.g.,
double hulls, spill source control and containment, vessel response plans). The estimate from the
expert panel for effectiveness of vessel response plans was 14.8 percent. 95 In 2013, the USCG
used this same effectiveness factor (rounded to 15 percent) in the evaluation of benefits for the
Nontank Vessel Response Plans final rule. 96
As another example, OPS finalized in 2005 regulations establishing oil spill response planning
requirements for onshore oil pipelines. 97 In the regulatory evaluation for the rule, OPS used
historical data on spills to estimate that the response plan requirements reduced the quantity of oil
spilled by an average of approximately 806,000 gallons per year, or a 31.3-percent reduction in
damages associated with pipeline spills. If the requirements of this rule for rail OSRPs have
equivalent effectiveness, the rule would produce substantial net benefits, at either the USCG rate
or OPS rate.
Table 31 presents estimated benefits at these two effectiveness rates. If the requirements of this
rule for rail OSRPs have equivalent effectiveness, the rule would produce substantial net benefits,
at either the USCG rate or OPS rate.
Table 31. Benefits, Costs, and Net Benefits Using USCG and PHMSA Pipeline Effectiveness
Rates (millions)
Year
1
2
3
4
5
6
7
10 Year
Benefits, 14.8
Percent
Effectiveness
$6.1
$6.2
$6.2
$6.3
$6.4
$6.4
$6.5
10 Year
Benefits, 31.3
Percent
Effectiveness
$12.9
$13.1
$13.2
$13.3
$13.5
$13.6
$13.7
10
Year
Costs
Net Benefits,
14.8 Percent
Effectiveness
Net Benefits,
31.3 Percent
Effectiveness
$6.8
$2.6
$2.6
$2.6
$2.6
$4.8
$2.6
-$0.7
$3.6
$3.7
$3.7
$3.8
$1.7
$3.8
$6.1
$10.5
$10.6
$10.7
$10.9
$8.8
$11.0
Economic Analysis Division - John A. Volpe National Transportation Systems Center. (2001, May). 2001 OPA 90
Programmatic Regulatory Assessment (PRA) – Benefits, Costs, and Cost Effectiveness of Eleven Major Rulemakings
of the Oil Pollution Act of 1990. Retrieved from:
http://www.regulations.gov/#!documentDetail;D=USCG-1998-3417-0006
95
Ibid. pp 7–6.
96
USCG-2008-1070. (2013, May). Nontank Vessel Response Plans and Other Vessel Response Plan RequirementsFinal Regulatory Analysis and Final Regulatory Flexibility Analysis for the Final Rule, pp 30–31. Retrieved from
http://www.regulations.gov/#!documentDetail;D=USCG-2008-1070-0046
97
(2005, February 23). Pipeline Safety: Response Plans for Onshore Transportation-Related Oil Pipelines. Federal
Register, 70(35), 8734–8748. http://www.gpo.gov/fdsys/pkg/FR-2005-02-23/pdf/05-3257.pdf
94
Page 71
8
9
10
Discounted 7%
Discounted 3%
$6.5
$6.5
$6.5
$47.6
$55.7
$13.7
$13.7
$13.7
$100.6
$117.9
$2.6
$2.6
$2.6
$25.1
$28.8
$3.9
$3.9
$3.8
$22.5
$26.9
$11.1
$11.1
$11.1
$75.5
$89.1
Comparing the benefits under these alternative effectiveness rates using the high-cost scenario
from the cost sensitivity analysis section, the rule would still have positive net benefits.
The most likely drivers for the quantity spilled are the number of breached tank cars and the
quantity of oil carried by those cars. The rule is anticipated to reduce the damages that result
from the quantity spilled rather than reduce the quantity spilled in the event of a derailment.
Two spills of similar magnitude in the same environment would entail different levels of
damages depending on how well the response is coordinated, how long it takes for the elements
of response to arrive, how capable the response is upon arrival, the training of the railroad and
response personnel, the mitigation strategies applied and equipment used, etc. We also note that,
if this rule increases situational awareness of risk on the part of the rail industry, it may indirectly
reduce the number of derailments of trains hauling crude oil, and thereby reduce the total
quantity of oil spilled. However, the Agency is not able to estimate such an effect at this time.
Public comments to the NPRM on this issue were limited and did not serve in the effort to
quantify and monetize the benefits of the rule.
Nevertheless, in addition to the break-even analysis and comparison with other modes’
effectiveness rates, another way to look at the impact of this rule is to look at some measure of
industry revenue and compare that to the total cost of the rule. We can estimate that the typical
tank car hauls 667 barrels (approximately 28,000 gallons) of product. The EIA Short Term Energy
Outlook estimates that crude oil will average about $59 per barrel in 2019. So, each tank car hauls
product worth approximately 667 x $59 = $39,353. Dividing this figure into the undiscounted cost
of the rule, we find that approximately 769 carloads would yield revenue sufficient to cover the
cost of the rule. 769 carloads represents about 3/1000 of 2016 shipments (0.26 percent). Put
another way, the revenue from approximately 8 unit trains of crude oil would be enough to cover
the estimated costs of this rule.
3.2.1. Evidence of Potential Oil Spill Response Improvement
The Agency examined post-accident reports, news reports, and evidence provided by EPA, FRA,
and PHMSA personnel to develop a picture of the responses to recent rail crude oil spills. We
refer readers to the preliminary RIA for information on the May 6, 2015, Heimdal, ND derailment
and derailments that occurred prior to the Heimdal derailment. We summarize more recent
derailments in Appendix B of this document. The preliminary RIA also includes a more extensive
qualitative discussion of the potential benefits of a comprehensive plan, specifically how a
comprehensive plan may improve spill response.
Page 72
3.3. Benefit Uncertainties and Unquantifiable Benefits
In this section, the Agency describes uncertainties associated with the estimates it developed to
monetize damages associated with crude train derailments. These uncertainties stem from
variability in the estimation of damages and incident severity, and difficulties in forecasting future
events. For a more detailed discussion of these uncertainties, we refer the reader to the preliminary
RIA in the rulemaking docket.
3.3.1. Damage estimates for crude oil and ethanol spills
Comprehensive societal costs are especially uncertain or difficult to obtain for rail incidents for
the following reasons:
•
Immediate response and cleanup are often the only cost elements reported when an incident
occurs, rather than long-term effects for which the costs are difficult to measure or to prove
resulted from the spill.
•
Most research efforts have focused on crude oil maritime spills, which may have limited
applicability to rail incidents.
•
Relatively fewer studies examine spills in rail transport.
As a result of questionable cost reporting following incidents, and the wide range of estimates
available from the literature, in addition to the variance due to specifics of terrain, natural features,
and the type of crude oil spilled, there is a high degree of uncertainty regarding the costs imposed
on society by these incidents. Thus, the value used in this analysis to monetize oil spill costs may
overestimate or underestimate the true costs of rail oil spill incidents. Please see the preliminary
HM-251B RIA and HM-251 HHFT final rule for PHMSA’s literature review relevant to spill costs.
We did not receive adverse public comment regarding our estimate to monetize spill costs.
3.3.2. Incident Severity
Crude oil incidents vary in size and the amount of damages they inflict on society. An event in a
particularly environmentally sensitive area, or one that substantially degrades a high-value public
asset such as a source of drinking water, will impose higher costs than a comparable event in a less
sensitive area. In addition, the quantity released during an event varies due to several factors. One
such factor in this case is the quantity of crude oil present in a train consist when it derails. Another
factor is the type of tank car carrying a product (e.g., DOT-111, DOT-117).
The Agency attempted to accurately estimate the overall effectiveness of the tank car upgrades
mandated by the HM-251 final rule. A validated computer simulation model was used to estimate
the impacts of the various enhancements mandated in that final rule on the likelihood of puncture
in different derailment scenarios. Any computer simulation has limitations, so although the
Agency believes it used the best tool available to estimate the benefits of the tank car and other
enhancements mandated by HM-251, the simulation model may not perfectly replicate real-world
derailment scenarios. As a result, although we have applied the HM-251 effectiveness rate (see
section 3.1.5. above) to reduce the societal damages associated with HHFT derailments in
anticipation of that rule’s impacts on future events, those effectiveness rates may not perfectly
simulate the real-world impacts of the HM-251 rule. The true effectiveness of the rule may differ
Page 73
from the Agency’s estimates. Furthermore, it is unclear whether the simulations used by the
Agency would tend to over or underestimate the effects of the rule, so the direction of the
inaccuracy is also unknown.
3.3.3. Number of Crude Oil HHFT Incidents
The Agency faces several challenges in forecasting the future number of incidents involving
HHFTs carrying crude oil. The primary challenge is that there are only 6 years of data during the
era of high-volume crude oil shipments, and annual volumes have fluctuated widely during this
time period. Ideally, the forecast number of events would be based on the volume of product
shipped in HHFTs because it would be specific to the applicability of the rule. Failing that,
forecasting forward based on volume of product shipped would be a next-best solution. The short
timeframe of HHFT shipments of crude oil, however, presents challenges for forecasting into the
future, because one straightforward method for making such a forecast, if a sufficient timeline
existed, would use the past relationship between crude oil production and high-volume crude oil
shipments and extrapolate that trend forward using EIA’s Annual Energy Outlook production
forecast. The Agency could then estimate an incident rate over time and identify whether the
relationship between events and volume is stationary, increasing, or decreasing. Unfortunately, a
sufficiently long timeline to make such an extrapolation is lacking.
The Agency described two alternatives above in section 3.1.2 to forecast the number of events
going forward. One uses time as an independent variable and the number of events as the
dependent variable. Thus, given x number of years of HHFT crude oil shipments and y events, we
calculate y / x events per year and multiply that ratio by the analysis period. The second approach
uses volume rather than time as the independent variable, and is the approach used to forecast
future events in the recent RIA used to re-evaluate the efficacy of ECP brake systems. Neither
approach adjusts for the possibility that the rate of derailment may be changing over time due to
changes in industry practices, economic pressures, or other factors. The Agency has chosen to use
what it views as the more conservative (lower carload volume) of these two approaches described
in section 3.1.2. In any case, the Agency’s method, or any alternative method, is unlikely to
produce entirely accurate results, so the number of future events that may occur is another area of
uncertainty.
3.3.4. Non-Quantified Benefits
As noted above and in the preliminary RIA, it is unlikely that estimates of the cost of spilled oil
capture the entirety of the costs imposed on society and the environment from these incidents.
Some portion of the benefits associated with mitigating these events goes unquantified due to this
limitation. Secondly, the Agency has not applied any quantification of deaths and injuries
associated with these events. Although a response plan is unlikely to reduce immediate deaths and
injuries at the time of the incident (e.g., deaths or injuries associated with a building, vehicle, or
train operator being impacted by a collision or derailment or its immediate aftermath), planning
and training requirements may prevent injuries to railroad employees and response providers. The
reduction in risk of deaths and injuries during incident response has not been quantified. In
addition to uncertainty regarding the effect of the rule on preventing deaths and injuries, we can
only identify 1 injury and 0 deaths in crude oil HHFT incidents in the U.S through PHMSA 5800.1
incident reporting. The 1 injury is from the Mt. Carbon, WV derailment.
Page 74
Finally, contamination of soil or water, or pollution of air in the event of a fire, can impose longterm health consequences if people are exposed to the toxic chemicals introduced to the soil, water,
or air. Due to data uncertainties and limitations, the lack of quantitative data submitted as public
comment, and the analytical difficulty, PHMSA does not quantify long-term health consequences
from crude oil HHFT derailments. In addition, there are considerations that place uncertainty upon
the extent of long-term health consequences from crude oil HHFT derailments. For instance, the
air pollution from such events may be avoided by evacuation, and may be of relatively short
duration. While exposure to contaminated water or soil may pose some long-term risk, it may be
avoided through adequate cleanup and water quality monitoring.
Table 32 below presents the areas of spill cost uncertainty and the direction, if known, in which
they are likely to vary compared to the primary estimates used in this analysis.
Table 32. Benefit Uncertainties
Benefit
Area
Cost
imposed on
society
from an oil
spill
incident
Direction of Uncertainty
Underestimation Overestimation Unknown
Brief Explanation of Cause of
Uncertainty
Many of the oil spill studies examined that
estimate a comprehensive range of costs
find higher costs per gallon associated
with these incidents.
X
Incident
severity
X
Number of
HHFT
crude oil
incidents
X
Deaths and
injuries
prevented
X
Page 75
The severity of future incidents is
unknown and difficult to predict with
absolute certainty. In addition, at this
point the accuracy of the effectiveness
rates used to adjust our baseline for
implementation of the HM-251 rule is
unclear. That rule may be more or less
effective than estimated.
PHMSA’s methodology for predicting the
number of future events is conservative
when compared to an alternate approach
based on events per unit of volume of
crude shipped by rail.
Crude oil spill events pose the potential
for deaths and injuries to response
personnel, railroad employees, and the
public. PHMSA has not estimated or
monetize the number or severity of deaths
and injuries during spill response due to
uncertainties. Uncertainty surrounds
whether this rule’s requirements would
prevent fatalities or injuries and the extent
to which it can mitigate their severity.
4. Appendix A: Research and Data Availability on OSRP Costs
As noted elsewhere in this analysis, PHMSA received public comment indicating that some of our
preliminary cost estimations were low. While these comments did not include data or information
with which to revise our preliminary estimates, PHMSA attempted to obtain better data and
information. In developing our analyses, we aim to be data-driven and view input from
stakeholders, including subject matter experts, to be preferable to simply making new assumptions.
When public comments indicated our plan development and maintenance costs were low, we
thought about revising them upwards, but did not know to what extent we should do so.
In order to obtain better information to estimate plan-related costs, we reached out a small number
(~5) of third party (non-railroad) OSRP plan writers (“plan writers”) by phone and e-mail. We
asked for voluntary information about the level of effort required to develop a comprehensive plan
for the rail context, and how this level of effort would be allocated between the core plan and
different response zones. We also asked if the plan writers had experience responding to oil spills,
since some organizations provide spill response services similar to OSROs as well as develop
plans. In addition, we asked about experience giving or participating in PREP exercises, and for
the plan writers’ estimation of the effectiveness of comprehensive rail OSRPs.
While the plan writers were able to give background information, only 1 provided estimates for
the level of effort. PHMSA used this input to revise our estimation of plan development costs. It
is reflected in Section 2 of this final RIA, where it relays that a comprehensive plan is expected to
require 180 hours of effort to develop the core plan, and another 180 hours for each response zone.
Each block of 180 hours was further broken down into 40 hours of senior time, 40 hours of
administrative time, and 100 hours of mid-level staff time. This revision upward on plan
development costs was also reflected in plan maintenance/review costs, since plan review costs
are derivative of the amount of time to initially develop a plan (10 percent of initial development
costs).
Public comment also mentioned that training programs may cost more than estimated, but we were
unable to obtain additional input on this issue. For example, we did not succeed at obtaining
additional input on the cost of a GIUE exercise or more specific estimates on training event costs
that align with PREP. We further contend that the full extent of railroads’ training programs on
hazmat preparedness and oil spills cannot be attributed to this final rule; it is certainly possible that
railroads have made and will make voluntary decisions regarding training and exercise
preparedness that exceed the minimum requirements of this rule.
PHMSA would have benefited from more comprehensive information on these areas, especially
from the regulated community during the public comment process. What is reflected in this final
RIA is the result of our judgment, assumptions, reasoned approach to public comment, and the
available data and information given the agency’s resources and capabilities.
For additional comment discussion and response, please see the final rule preamble. This
discussion is included here for analytical purposes, namely to describe the agency’s representation
of the expected costs and benefits of this rule and how it evolved between the NPRM and final
rule.
Page 76
5. Appendix B: Incident Response Narratives
In developing this final rule, PHMSA collected and reviewed information from various sources
pertaining to derailments involving releases of crude oil. All accidents involving a release of
hazardous materials in transportation in the U.S., including derailments, are reported to PHMSA
and are recorded in the hazardous materials incident report database. 98 This searchable incident
report database is accessible at: https://hazmatonline.phmsa.dot.gov/IncidentReportsSearch. The
database includes the incident date, location (city and state), mode of transportation, carrier
information, shipper information, number of fatalities or injuries, monetary damages, type of
hazmat involved, quantity of hazmat released, and other information. Information from PHMSA
5800.1 incident reports helped in developing our analyses in support of this rulemaking.
Nonetheless, the information contained in PHMSA incident reports is not exhaustive, so we also
collected information from other resources, such as FRA Accident/Incident Reports, FRA
investigation reports, and EPA On-Scene Coordinator (OSC) reports, which are publicly available
at: http://epaosc.org.
In addition, PHMSA used some information gathered from common, publicly available news
sources (e.g., Internet-based sources).
In this final RIA, we provide summaries of the derailments that occurred in 2016 and 2017, as well
as summaries of the 2015 Watertown, WI and Culbertson, MT derailments, which were initially
referenced, but not discussed extensively, in the preliminary RIA. PHMSA provides narratives
and discussion of the circumstances and consequences of these derailments to the extent possible
amid limited information. PHMSA has identified these derailments as likely to have involved
trains transporting 20 or more tank cars of petroleum oil in a continuous block or 35 or more tank
cars dispersed throughout the train in conformance with the applicability of this rule. Furthermore,
these derailments resulted in releases of petroleum oil that harmed or posed a threat of harm to the
environment.
5.1. Plainfield, IL
Overview
According to FRA data, on June 30, 2017, a Canadian National Railway (CN) train operated by a
CN subsidiary consisting of 113 tank cars loaded with hazmat, including crude oil, derailed 19
cars near Plainfield, IL. 99 The PHMSA incident report suggests that 21 cars were derailed, not 19.
According to both FRA and PHMSA data, 5 cars released crude oil. 100 PHMSA’s 5800.1 report
indicates that 28,245 gallons were spilled, whereas the FRA Accident Detail report indicates
approximately 30,045 gallons. There were no evacuations, fires, or injuries reported.
See 49 C.F.R 171.15, 171.16.
FRA Accident Detail Report, querying for “Plainfield” in June 2017, available at:
https://safetydata.fra.dot.gov/OfficeofSafety/publicsite/Query/incrpt.aspx
100
PHMSA 5800.1 report, # I-2017070267, available at:
https://hazmatonline.phmsa.dot.gov/IncidentReportsSearch/Welcome.aspx
98
99
Page 77
Harm or Threat of Harm
This incident precipitated a response from the U.S. EPA, specifically Region V. 101 There is very
limited information on the EPA’s site for this oil spill. For other crude oil derailments, the EPA
OSC site has provided reports detailing response activities and how the oil affected the
environment, such as the water or land resources affected or threatened. For the Plainfield, IL
derailment, the EPA OSC states that the derailment “cause[d] oil to be released into the
environment” and corroborates that the spillage occurred.
Given the limited information available on the EPA OSC site for this derailment, we look at other
sources. One source indicates that the DuPage River was nearby, within 1,200 feet away from the
derailed train. 102 The same source relays that “EPA had not found any signs of oil in nearby
waterways” and “air monitoring did not suggest any problems.” 103 According to this source, the
Plainfield police chief said, at a news conference, “We believe by all standards that there is no
measurable risk to the community.” 104
Another news source indicated that the oil spill resulted in contaminated soil, and that the cleanup
of the spill in the days after the derailment would include the removal of this contaminated soil.105
This news report was released on July 4, 2017, indicating that cleanup was on-going for at least a
few days after the spill. It also relayed that the cleanup was “proceeding well and [would] continue
through the week.” On July 7, 2017, the same local news source relayed that “[o]n site monitoring
of air quality and ground contamination continues around the clock with no health issues identified
at this time.” 106 An article released in September 2017 claimed that the cleanup ultimately lasted
“weeks.” 107
External news sources indicate that the Federal Railroad Administration (FRA) started an
investigation into the Plainfield, IL derailment and that the investigation’s report would be
available on the FRA website when released. As of April 24, 2018, this report does not appear to
be released on the FRA website. 108 Nevertheless, the fact that FRA started an investigation
provides indication that the Plainfield, IL derailment met FRA’s general criteria for accident
101
U.S. EPA On-Scene Coordinator site for 6/30/17 Plainfield, IL derailment, available at:
https://response.epa.gov/site/site_profile.aspx?site_id=12263
102
“Train derailment dumps thousands of gallons of crude oil in Plainfield,” Alicia Fabbre, Chicago Tribune, July 1,
2017, available at: http://www.chicagotribune.com/news/local/breaking/ct-train-derailment-dumps-thousands-ofgallons-of-crude-oil-in-plainfield-20170701-story.html
103
Ibid.
104
Ibid.
105
“Cleanup Continues After Plainfield Train Derailment, Oil Spill,” Associated Press News Partner & Plainfield
Patch, July 4, 2017, available at: https://patch.com/illinois/plainfield/cleanup-continues-after-plainfield-trainderailment-oil-spill
106
“Train Speeds Increase, Cleanup Progressing After Plainfield Train Derailment,” Shannon Antinori, Plainfield
Patch, July 7, 2017, available at: https://patch.com/illinois/plainfield/train-speeds-increase-cleanup-progressingafter-plainfield-train-derailment
107
“Plainfield Derailment Could Leave Thousands Without Heat This Winter: Nicor,” Shannon Anitnori, Plainfield
Path, September 29, 2017, available at: https://patch.com/illinois/plainfield/250k-could-be-without-heat-winterthanks-plainfield-derailment
108
“eLibrary Search.” FRA Investigations of Railroad Accidents. U.S. Federal Railroad Administration, available
at: https://www.fra.dot.gov/eLib/Find#p1_z10_gD_lAC
Page 78
investigation, although this has not been confirmed. Among these criteria is the criterion that FRA
generally investigates accidents and incidents that involve “fire, explosion, evacuation, or release
of regulated hazardous materials, especially if it exposed a community to these hazards or the
threat of such exposure.” 109
In sum, the information on the effect or possible effect of the Plainfield, IL derailment on the
environment is limited. From what is available, it is certain that multiple tank cars released crude
oil, resulting in a spill near the DuPage River, but it does not appear that the spill entered the river.
Further, the spillage contaminated soil and required a cleanup effort. The extent length of the
cleanup is not clear, nor is it clear the quantity that was recovered versus the quantity that remained
in the environment. It does not appear to have affected air quality. Some sources indicate that
local residents could smell oil in the air. 110 While not related to environmental quality, there is
also indication from local news sources that local businesses were affected and closed in reaction
to the derailment and spill response. 111
Response and Discussion
Information regarding the response to the Plainfield, IL derailment is limited. In particular, the
EPA OSC site does not contain any reports detailing the spill response. In addition, it appears the
FRA investigation report is not yet available publicly, and the PHMSA incident report does not
detail the spill response.
Using external news sources, we find some commentary on the response effort. One source
relayed that, “First responders in Plainfield said prior training with the railroad helped them to
better prepare for a crude oil train derailment. [The first responders] said they had the situation
under control within two hours.” 112
This limited information does not necessarily point to possible improvements for the response to
the Plainfield, IL derailment.” For example, sources do not cite a lack of resources or an inability
to access the derailment site. The information summarized here points to limited spill impacts and
an effective response. However, the media reports refer to the incident as a “close call,” and the
Plainfield Fire Chief related the need to learn from the incident. 113
“Accident Data, Reporting, and Investigations,” FRA, available at: https://www.fra.dot.gov/Page/P0037 [see
“FRA Accident Investigations (general criteria)”]
110
“ ‘We were very lucky’: Plainfield’s escape from train disaster gives town pause,” Marni Pyke, Daily Herald,
July 10, 2017, available at: http://www.dailyherald.com/news/20170710/we-were-very-lucky-plainfields-escapefrom-train-disaster-gives-town-pause
111
“Plainfield Police Provide Update After Train Derailment,” Emily Florez and James Neveau, NBC 5 Chicago,
July 1, 2017, available at: https://www.nbcchicago.com/news/local/plainfield-police-update-cleanup-after-trainderailment-432049103.html
112
“Emergency Response Tested by Crude Oil Derailment,” Chris Coffey, NBC 5 Chicago, July 3, 2017, available
at: https://www.nbcchicago.com/investigations/emergency-responders-tested-crude-oil-train-derailment432349873.html
113
“ ‘We were very lucky’: Plainfield’s escape from train disaster gives town pause,” Marni Pyke, Daily Herald, July
10, 2017, available at: http://www.dailyherald.com/news/20170710/we-were-very-lucky-plainfields-escape-fromtrain-disaster-gives-town-pause
109
Page 79
In this light, if the incident had been more severe, a comprehensive plan would ensure that
resources sufficient to respond to a worst-case discharge would be available. For example, the
worst-case discharge planning volume for a response zone containing Plainfield, IL would be
approximately 500,000 gallons of oil ((113 tank cars * 30,000 gallons each) * 0.15), provided that
the train involved in the Plainfield, IL derailment were the largest train configuration expected to
operate in the area. Therefore, the actual spillage in the event (28,245 gallons), according to the
PHMSA incident form filed, was approximately 6 percent of this hypothetical WCD planning
volume. As such, it might be offered that the final rule would have afforded a high level of
response capability above and beyond the quantity spilled at the derailment.
In addition, one news source indicated that 30 different state, federal, local, and private agencies
were on-site, working to clean up the spill. 114 A comprehensive plan could improve the ability of
plan holders to cooperate across such a diverse array of public and private organizations and
improve coordination.
5.2. Money, MS
Overview
According to FRA data, on April 30, 2017, a Canadian National Railway (CN) train operated by a
CN subsidiary consisting of 114 tank cars loaded with hazmat, including crude oil, derailed 2 tank
cars near Money, MS. 115 One tank car released crude oil, spilling approximately 24,653 gallons
according to the PHMSA 5800.1 incident filing. 116 In the FRA Accident Detail report, 1 injury is
identified; however, in the PHMSA incident filing, no injuries are identified. In the FRA and
PHMSA reports, no evacuation is indicated; however, external new reports mention that “some
residents were evacuated.” 117 Both the PHMSA and FRA data indicate a fire occurred.Harm or
Threat of Harm
The information available on this incident is limited. EPA does not appear to have an OSC website
for this derailment and spill. However, from the PHMSA incident report, we have evidence that
the spill occurred and affected the environment. On the PHMSA 5800.1 form, we have indication
that “environmental damage” occurred. 118 Specifically, there is a binary indicator on the form for
whether environmental damage occurred and the 5800.1 filing for this incident has positive
indication (question #30). We have further indication that the spill cleanup involved “in-house”
as well as “other” resources. The response cost totaled $65,000 according to the same filing.
“Train derailment dumps thousands of gallons of crude oil in Plainfield,” Alicia Fabbre, Chicago Tribune, July 1,
2017, available at: http://www.chicagotribune.com/news/local/breaking/ct-train-derailment-dumps-thousands-ofgallons-of-crude-oil-in-plainfield-20170701-story.html
115
FRA Accident Detail Report, querying for “Money” in April 2017, available at:
https://safetydata.fra.dot.gov/OfficeofSafety/publicsite/Query/incrpt.aspx
116
PHMSA 5800.1 report, # X-2017060427, available at:
https://hazmatonline.phmsa.dot.gov/IncidentReportsSearch/Welcome.aspx [hereinafter # X-2017060427]
117
“Railroad Reopens After Fiery Crash In Mississippi Delta,” Associated Press, May 1, 2017, available at:
https://www.usnews.com/news/best-states/tennessee/articles/2017-05-01/railroad-reopens-after-fiery-crash-inmississippi-delta
118
See X-2017060427
114
Page 80
There is no apparent determination that the spill entered a waterway.
In addition to the PHMSA and FRA data, video footage from ABC News shows a smoke plume
emanating from the derailment, and other news sources refer to the derailment as a “fiery crash.” 119
Within the available sources, there does not appear to be any comment on air quality effects from
the fire and smoke. An additional news source relayed a comment from the Mississippi
Department of Environmental Quality that workers were removing the remaining oil on the day
following the derailment. 120
In sum, the available information is limited but we have indication from PHMSA and FRA data,
as well as external news sources, that the Money, MS derailment resulted in spillage and likely
harmed or posed a threat of harm to the environment.
Response and Discussion
Again, the information on this incident, including the spill response, is limited. We did not identify
commentary on how the response unfolded or the challenges faced. This makes it difficult to
assess the ways in which the final rule could have positively impacted the response to Money, MS.
However, on the PHMSA 5800.1 form, we have indication that weather conditions included
“heavy rain.” 121 A comprehensive plan requires preparation for a worst-case discharge, which
includes the potential for adverse weather. We do not know from the available public resources
how heavy rains may have affected the response, but a comprehensive plan could assist in planning
for these weather conditions during a response.
5.3. Mosier, OR
Overview
On June 3, 2016, a Union Pacific (UP) freight train derailed in Mosier, OR. 122 It had 96 cars, 94
of which carried hazmat, including crude oil. Sixteen tank cars were damaged or derailed, and 5
tank cars released crude oil. According to PHMSA incident data, approximately 42,448 gallons
of crude were released from these tank cars. 123 The derailment was adjacent to Highway 30,
“Freight train carrying crude oil derails in Money, Mississippi,” ABC News, April 29, 2017, available at:
https://www.google.com/search?q=money+mississippi+derailment&rlz=1C1GCEA_enUS754US754&oq=money&
aqs=chrome.0.69i59l2j69i57j69i60l3.1231j0j7&sourceid=chrome&ie=UTF-8
120
“Railroad Reopens After Fiery Crash in Mississippi Delta,” Associated Press, May 1, 2017, available at:
https://www.usnews.com/news/best-states/tennessee/articles/2017-05-01/railroad-reopens-after-fiery-crash-inmississippi-delta
121
See X-2017060427
122
https://safetydata.fra.dot.gov/OfficeofSafety/publicsite/Query/incrpt.aspx
123
PHMSA 5800.1 report, # X-2016060795, available at:
https://hazmatonline.phmsa.dot.gov/IncidentReportsSearch/Welcome.aspx [hereinafter # X-2016060795]
119
Page 81
Interstate 84, and the Columbia River. 124 There were no deaths or injuries, but an evacuation of
147 residents occurred and several cars caught fire. 125
Harm or Threat of Harm
The EPA On-Scene Coordinator Pollution/Situation Report (POLREP) #1 indicates the Mosier,
OR derailment as involving “[e]mergency response to a threat of discharge of oil into navigable
waters of the U.S.” 126 The derailment took place along a rail line at or very near the crossing of
Rock Creek, a tributary to the Columbia River. 127 EPA identified UP as the responsible party and
issued a Notice of Federal Interest (NOFI) on June 7, 2016. 128
It also resulted in damage to the local Waste Water Treatment Plan, which had to temporarily cease
operations. 129 Approximately 13,000 gallons of oil were sucked into the treatment plant. 130
According to EPA, a minor amount of oil discharged from the treatment plant outfall pipe into the
river, but this pipe was plugged and the oil removed. 131 This area was boomed and there were no
observable impacts to fish or wildlife. 132 POLREP #3 notes that the main treatment plant was
cleaned and repaired by UP and contractors to UP, and returned to service on June 20th, or more
than 2 weeks after the derailment and cessation of normal operations. 133
In POLREP #1, EPA noted that the tank car catching on fire resulted in an additional wildland fire
impacting 5-10 acres. 134 In POLREP #2, EPA revised the number of wildland acres affected
downward to “less than 1 acre,” and in POLREP #3 to 1.3 acres. 135,136 EPA also noted that if the
fire could not be controlled, the derailed oil train posed a “significant threat” to the river.137
Further, firefighting cooling tactics presented the risk that oil and oily water would be flushed into
Rock Creek, which was less than 200 yards away from the fire. 138 EPA noted the area is home to
federally listed endangered fish. 139 The day following the derailment, EPA noted a “rainbow
“Moiser Oil Train Derailment,” U.S. EPA, available at:
https://response.epa.gov/site/site_profile.aspx?site_id=11637 [hereinafter: “EPA OSC Mosier Main”]
125
Ibid.
126
“Mosier Oil Train Derailment POLREP #1,” U.S Environmental Protection Agency (EPA), June 3, 2016,
available at: https://response.epa.gov/site/sitrep_profile.aspx?site_id=11637&counter=27106 [hereinafter “Mosier
POLREP #1”]
127
Ibid.
128
“Mosier Oil Train Derailment POLREP #3,” U.S Environmental Protection Agency (EPA), June 3, 2016,
available at: https://response.epa.gov/site/sitrep_profile.aspx?site_id=11637&counter=27111 [hereinafter “Mosier
POLREP #3”]
129
Ibid.
130
EPA OSC Mosier Main
131
Ibid.
132
Ibid.
133
See Mosier POLREP #3
134
See Mosier POLREP #1
135
“Mosier Oil Train Derailment POLREP #2,” U.S Environmental Protection Agency (EPA), June 3, 2016,
available at: https://response.epa.gov/site/sitrep_profile.aspx?site_id=11637&counter=27109 [hereinafter “Mosier
POLREP #2”]
136
See Mosier POLREP #3
137
See Mosier POLREP #1
138
Ibid.
139
Ibid.
124
Page 82
sheen” on the Columbia River, but could not determine the pathway at that time. 140 The last
POLREP (#3) notes that no sheen was seen after that day, June 4, 2016. 141
In POLREP #3, EPA summarizes its estimation of the gallons lost to the spill and how it distributed
into the environment, specifically “13,000 gallons in the adjacent WWTP [wastewater treatment
plant] and piping, 16,000 gallons burned up and/or vaporized in air; and 18,000 gallons discharged
onto the group and in soil.” 142 Approximately 2 million gallons were used to put out the fire,
which drew the city’s backup drinking water well to very low levels. 143
In addition to impacts to water, EPA conducted air monitoring due to the smoke and fire in the
derailment area.
In addition to EPA reporting, ODEQ describes impacts to the environment, including disposal of
2,960 tons of petroleum-contaminated soil. 144 As relayed by ODEQ, the contaminated soil
contained approximately 18,000 gallons of oil. 145 With respect to water impacts, ODEQ states,
“[a] small amount of oil entered the Columbia River but the amount if unknown.” 146 The ODEQ
also mentions the possibility that “autumn rains” could lead to precipitation flushing additional oil
into the river, and that “[i]nitial testing of the wells indicate[s] that there is only nominal
contamination in the groundwater” at the tested wells south of the tracks, but “significant
groundwater contamination” was found at one of the wells installed on the north side of the
tracks. 147
On the “Mosier derailment Fact Sheet,” ODEQ relays that the community’s water supply had been
impacted by the fire response, and reinforces the EPA reporting claim that the community’s
wastewater treatment plant was damaged by the derailment and fire. 148 The fact sheet also
mentions a “light sheen of oil” in the Columbia River at the mouth of Rock Creek. 149 This aligns
with EPA reporting.
On the PHMSA 5800.1 form, there is indication that “environmental damage” resulted from the
derailment, as well as spillage and entry of hazmat into a waterway/storm sewer. 150 Response
costs were reported as $16,800 and remediation and cleanup costs as $7 million. 151 “In-house”
See Mosier POLREP #2
See Mosier POLREP #3
142
Ibid.
143
Ibid.
144
“Mosier UPRR Derailment,” Oregon Department of Environmental Quality, available at:
http://www.deq.state.or.us/Webdocs/Forms/Output/FPController.ashx?SourceId=6115&SourceIdType=11
[hereinafter “ODEQ Main”]
145
Ibid.
146
Ibid.
147
Ibid.
148
“Mosier derailment Fact Sheet 6-7-2016 – final with Spanish,” ODEQ, June 7, 2016, available at:
http://www.deq.state.or.us/Webdocs/Controls/Output/PdfHandler.ashx?p=4719c5c5-8592-4ee2-9a4b3476f448d2bbpdf&s=Mosier%20derailment%20Fact%20Sheet%206-7-2016%20%20final%20with%20Spanish.pdf
149
Ibid.
150
See X-2016060795
151
Ibid.
140
141
Page 83
cleanup is also indicated. 152 These data lend support other findings of harm or potential harm to
the environment.
In public comment to the NPRM, we have comments from the State of Washington Department
of Ecology, which offered this about the Mosier, OR derailment: “While this derailment was not
the worst-case scenario for which we have been preparing for, the impact on the local community
was significant.” 153 Washington State shares a waterfront with Oregon along the Columbia River
Gorge. 154
Several other commenters noted the importance of the Columbia River and its surrounding
environment, including Riverkeeper and its partner commenters and Lake Pend Oreille
Waterkeeper. 155 Riverkeeper references designated critical habitats for listed species, such as
salmon and steelhead habitat along the Columbia River. 156 As such, these commenters provide
further support that the Mosier, OR derailment posed harm or the threat of harm to the
environment.
Response and Discussion
There is a variety of information available regarding the response to and circumstances
surrounding the Mosier, OR derailment. This information includes multiple Pollution/Situation
Reports (POLREPs) on the U.S. EPA On-Scene Coordinator website for the Mosier derailment,
as well as the investigation report from FRA (HQ-2016-1136) and FRA Accident Detail
report. 157,158,159 In addition, a few public comments to the NPRM mentioned the Mosier, OR
derailment specifically, such as the comment from NTSB, the Washington Department of Ecology,
and Riverkeeper and partner comments. 160 We also have available the PHMSA 5800.1 report and
external news sources. 161 The Oregon Department of Environmental Quality has a repository of
information on the Mosier, OR derailment. 162
Ibid.
Comment from the Washington State Dept. of Ecology, posted Sept. 26, 2016, available at:
https://www.regulations.gov/document?D=PHMSA-2014-0105-0268 [hereinafter “Washington Dept. of Ecology
comment”]
154
Ibid.
155
Comments to the NPRM are available in the docket (PHMSA-2014-0105) at:
https://www.regulations.gov/docketBrowser?rpp=25&so=DESC&sb=commentDueDate&po=0&s=columbia&dct=P
S&D=PHMSA-2014-0105
156
“OSRP_PHMSAComments_CoalitionFinal,” Riverkeeper, posted Sept. 28, 2016, pg. 19, available at:
https://www.regulations.gov/document?D=PHMSA-2014-0105-0337 [hereinafter “Riverkeeper comment”]
157
See EPA OSC Mosier Main
158
“Accident Investigation Report: HQ-2016-1136,” Office of Railroad Safety, FRA, available at:
https://www.fra.dot.gov/eLib/details/L19462#p1_z5_gD_kmosier
159
Query for “Mosier” “OR” at https://safetydata.fra.dot.gov/OfficeofSafety/publicsite/Query/incrpt.aspx
160
“Hazardous Materials: Oil Spill Response Plans and Information Sharing for High-Hazard Flammable Trains,”
PHMSA, July 29, 2016, available at: https://www.regulations.gov/document?D=PHMSA-2014-0105-0240
161
See X-2016060795
162
See ODEQ Main
152
153
Page 84
According to EPA, 28 local and regional fire departments and hazmat teams responded to the
derailment from both OR and WA. 163 In addition, representatives from the Oregon Department of
Environmental Quality (ODEQ), the Washington Department of Ecology, and the Yakama Nation
were part of the Unified Command. 164 Further, there were several cooperating agencies beyond
those formed into the Unified Command, including the National Marine Fisheries Service,
National Oceanographic and Atmospheric Administration (NOAA), U.S. Coast Guard, U.S. Army
Corps of Engineers, the Federal Emergency Management Agency (FEMA), FRA, PHMSA and
others. 165 This is an example of the diverse array of organizations that can be involved in a
derailment, and a comprehensive plan is expected to enhance coordination among the various
parties involved.
POLREP #3 states that by mid-morning of June 4th, a contract Incident Management Team (IMT)
arrived at the Mosier, OR derailment site, having been mobilized by UP. 166 EPA claimed that this
IMT “brought needed structure” to the incident. 167 This statement, while broad, suggests that the
initial spill response could have been improved in some ways, by having additional “structure.”
PHMSA expects that a comprehensive plan would improve coordination during the initial spill
response and enhance the organization and structure of the spill response, specifically through
training requirements and ensuring that railroad employees know their responsibilities under the
plan and can make the proper notifications to other organizations as needed.
In addition, while we may not know exactly when the UP-mobilized IMT arrived, they arrived
mid-morning the day after the derailment, which had occurred around 12:15pm. 168 This suggests
the UP-mobilized IMT arrived on site more than 12 hours after the derailment. In the final rule,
plan holders are expected to identify and describe in the plan the resources which are available to
arrive onsite within 12 hours after the discovery of a worst-case discharge, or substantial threat of
a discharge. If the railroad’s plan had identified the IMT and its response resources as part of the
response zone containing Mosier, OR, there would be an expectation that these resources could
reasonably arrive on site in less time than was apparently needed to arrive at the Mosier, OR
derailment. If an incident like Mosier, OR occurred again, a comprehensive plan might encourage
a speedier arrival by response providers, such as a contract IMT. Please note, this is discussion
offered for the purposes of analyzing the benefits of this rule; please refer to the preamble and
regulatory text of the final rule, not the final RIA, for the language codified into the CFR and
explanation of agency intent.
POLREP #3 also describes challenges with respect to handling temporary storage and disposal of
crude oil contained within the derailed tank cars. According to this POLREP, a plan was needed
to build a temporary storage and transloading facility at a railyard outside of Mosier, and this plan
needed approval by the Unified Command and the city of The Dalles. 169 A comprehensive plan
could assist in planning for these types of spill response activities and encourage the dissemination
See EPA OSC Mosier Main
See Mosier POLREP #3
165
Ibid.
166
Ibid.
167
Ibid.
168
See EPA OSC Mosier Main and Mosier POLREP #3
169
See Mosier POLREP #3
163
164
Page 85
of lessons learned and best practices both within the plan holder’s organization as well as across
stakeholder groups.
POLREP #3 mentions discussion among the UC about resuming movement on the rail line. It
appears UP requested permission to resume operations on June 5th, roughly 2 days after the
derailment, even though several of the non-breached railcars still contained crude oil. 170 The UC
assessed and inspected the cars, and based on air monitoring around them, agreed that train traffic
could resume at a low speed given there was “no danger for fire or discharge of oil.” 171 PHMSA
believes a comprehensive plan would be applicable to ensuring workplace safety and making
decision such as the one exemplified here.
Additionally, POLREP #3 describes the type of equipment needed for workers if “elevated” levels
of air pollutants are discovered during air monitoring. 172 The “[w]orkers at the WWTP [Waste
Water Treatment Plant] were required to wear respirators until levels abated.” 173 PHMSA believes
a comprehensive plan would have a positive influence on ensuring workplace safety efforts, such
as this one.
According to the same source, the “first train” after operations resumed rolled through town during
a public meeting, upsetting some residents and “creating a security issue.” 174 The EPA OSC
website has a file repository, which includes a flyer for a public meeting. 175 It includes event and
contact information as well as information critical to safety. POLREP #3 cites, “[h]uge effects of
derailment and fire on community. Emotions raw, and concern high as oil trains continue to move
through their community and both sides of the Columbia River,” as well as “[i]ntense media and
political interests.” 176 A comprehensive plan may be applicable to improving the public’s
perception of risk if and when rail operations resume.
Moreover, POLREP #3 mentions that an “additional collection” of oil in the water treatment pipe
was discovered on June 6th. 177 It remains somewhat unclear, but it would appear oil did not reenter the river. 178 The main EPA OSC web page for the derailment suggests that it did not, as it
states, “no oil [was] observed in the River since June 4.” 179 A comprehensive plan may improve
the ability of the plan holder to identify sources of potential oil contamination and ensure that
monitoring for oil release is continual and sufficiently frequent.
POLREP #3 also mentions the transfer of lead oversight of soil and groundwater remediation to
the Oregon Department of Environmental Quality (ODEQ). 180 PHMSA believes a comprehensive
Ibid.
Ibid.
172
Ibid.
173
Ibid.
174
Ibid.
175
“Mosier Oil Train Derailment: Documents,” EPA, available at:
https://response.epa.gov/site/doc_list.aspx?site_id=11637
176
See Mosier POLREP #3
177
Ibid.
178
Ibid.
179
See EPA OSC Mosier Main
180
See Mosier POLREP #3
170
171
Page 86
plan would help in this type of coordination, specifically helping to understand roles and
responsibilities among the plan holder and Federal and other stakeholders. This seems especially
important as responsibilities are transferred and/or delegated. This POLREP notes that the
monitoring of groundwater would last “for at least one year.” 181 Again, the importance of this
monitoring and the length of time underscores this need to define roles and responsibilities and set
expectations for the spill’s response and long-term remediation.
In the PHMSA 5800.1 form, the description of events notes that, “[d]uring the re-positioning of
the car for transfer the bottom outlet valve handle was actuated on debris causing a release of 50
gallons of lading.” 182 PHMSA believes comprehensive plan may encourage the sharing of best
practices and lessons learned in re-positioning of derailed or potentially damaged tank cars.
Possibly there were techniques to prevent this unintended release during re-positioning and if so,
including these in training or exercise programs could improve spill response outcomes.
In the public comment from the State of Washington Dept. of Ecology, it is noted that the Mosier,
OR derailment demonstrated the need for a coordinated effort on the part of the rail carrier (plan
holder), the Federal government, and states whose waters are affected. 183 As provided in this
public comment, PHMSA believes that comprehensive plans will further this coordination by
aligning spill response efforts with the Clean Water Act’s national framework. 184
In the public comment from Riverkeeper, the commenter relays that the derailment in Mosier, OR
provides “some insight into the need for more rapid response.” 185 The comment notes that the
Governor invoked an emergency since the resources needed exceeded local resources. 186 The
comment suggests that a more rapid response would have contained the oil spill from reaching
waters and critical habitat. 187 PHMSA believes that comprehensive plans may contribute to faster
responses by improving response coordination and these plans include certification that the plan
holder can respond to and remove a worst-case discharge. For additional comment response,
please see the final rule preamble; this discussion is provided in the final RIA for analytical
purposes, in order to present potential and/or expected benefits of the final rule.
One external news source cites the potential for “high winds,” which could be considered an
adverse weather condition. 188 Another claims that the winds were “oddly quiet” for the region
during the time of the derailment and the “fire didn’t spread like it could have.” 189 PHMSA
believes comprehensive plans will encourage planning for weather conditions that are common to
the environments of a given response zone.
Ibid.
See X-2016060795
183
See Washington Dept. of Ecology comment
184
Ibid.
185
See Riverkeeper comment, pg. 20
186
Ibid.
187
Ibid.
188
“Mosier Oil Train Derailment,” Friends of the Columbia Gorge, available at: https://gorgefriends.org/protect-thegorge/mosier-oil-train-derailment.html
189
“How Bad Could the Mosier Oil Spill Have Been?” Emily Schwing, OPB, Aug. 10, 2016, available at:
https://www.opb.org/news/series/oil-trains/mosier-oil-spill-how-bad/
181
182
Page 87
Lastly, the comment from the NTSB include the following commentary on the response to the
Mosier, OR derailment: “We have found that despite voluntary outreach and community
awareness programs […], many communities and emergency responders are unprepared to cope
with derailments that involve fires fueled by crude oil. The Mosier, Oregon HHFT derailment on
June 3, 2016, is yet another example where the local emergency response community was illprepared to effectively respond to the derailment.” 190 PHMSA believes that comprehensive plans
will provide public safety and environmental protection that improves upon voluntary programs
and enhances the availability of resources to respond to worst-case discharges, or the threat of one.
5.4. Watertown, WI
Overview
On November 8, 2015, a Canadian Pacific (CP) freight train consisting of 110 tank cars loaded
with hazmat, including crude oil, derailed 15 tank cars near Watertown, WI. 191 Three of the
derailed tank cars were breached, spilling approximately 500 gallons of crude oil, according to the
report filed to FRA. According to other news reports, this amount was closer to 1,000 gallons. 192
In addition, the PHMSA 5800.1 report cites 1,000 gallons released. 193 The derailment forced the
evacuation of approximately 35 homes (41 people) as a precaution. 194 No fires or injuries were
reported.
Harm or Threat of Harm
In particular, the EPA OSC distinguishes a threat to waterways in the POLREP report by stating,
“Continued oversight until threat of release to waters of the US is abated.” 195 However, the
Watertown, WI derailment does not appear to have affected water resources, but a dry drainage
ditch that is a tributary to the Rock River was located approximately 50 feet from the release site. 196
EPA’s Pollution/Situation (“POLREP”) report indicates that the Watertown Fire Department was
able to contain the release on site. 197 The report indicates that the “standing liquid” was vacuumed
and CP would send plans to the local health department and State regarding remediation plans.
According to the same report, soil remediation was needed due to the spill.
Public comment from NTSB, posted Sept. 27, 2016, available at:
https://www.regulations.gov/document?D=PHMSA-2014-0105-0326
191
http://safetydata.fra.dot.gov/OfficeofSafety/publicsite/Query/incrpt.aspx
192
http://fox6now.com/2015/11/08/breaking-crews-on-scene-of-train-derailment-in-watertown/
193
PHMSA 5800.1 report, # I-2016010013, available at:
https://hazmatonline.phmsa.dot.gov/IncidentReportsSearch/Welcome.aspx [hereinafter # I-2016010013]
194
http://safetydata.fra.dot.gov/OfficeofSafety/publicsite/Query/incrpt.aspx
195
“cp watertown derailment POLREP #1,” U.S Environmental Protection Agency (EPA), November 8-9, 2015,
available at: https://response.epa.gov/site/sitrep_profile.aspx?site_id=11275&counter=25246
196
Ibid.
197
Ibid.
190
Page 88
An external news source relays the same information, including the lack of entry into waterways
and soil remediation activities, and adds that air monitoring readings did not exceed safe levels. 198
This news source indicates that the evacuation of 35 homes was conducted as a precaution.
The FRA issued an investigation report on this derailment, also stating that the crude oil did not
reach waterways and that the evacuation was precautionary. 199
The PHMSA 5800.1 report does not indicate environmental damages in the form’s question #30.
However, it confirms the spillage occurred, the quantity released, and that an in-house clean up
took place. It also provides an estimate of $500,000 for the response costs; however, for PHMSA
incident reporting, “Response Costs,” are identified separately from “Remediation/Cleanup
Cost.” 200 Thus, it is not clear if the response costs include spill response, or if the costs summing
up to $500,000 relate solely to other expenditures, such as costs incurred for local police and fire
department services. The PHMSA incident form for this derailment provides that the cost of
“Remediation/Cleanup” was $0. 201
Response and Discussion
Information on the spill response to the Watertown, WI derailment is limited. For example, the
FRA report focuses on the cause of the derailment, not the developments of spill response. In
addition, the available information from U.S. EPA is limited to POLREP #1, which suggests the
spill response was able to contain the spill on site, and does note any particular response challenges.
Similarly, the PHMSA 5800.1 incident form does not indicate challenges in the response. The
information provided by external news sources is also limited. One source includes comments
from the Watertown Fire Chief and video content from a news conference following the
derailment. 202 The fire chief relayed that there could have been a fire. 203 He also relayed training
activities that were undertaken by the fire department (e.g., tabletop functional exercises), and their
department’s hazmat planning was cited as ongoing for over a year. 204 He also mentioned the
State’s Mutual Aid Box Alarm System (MABUS).
While the available information does not point to specific response challenges, a comprehensive
plan would ensure that the plan holder has sufficient resources to respond to a worst-case discharge
(WCD). A worst-case discharge includes discharges resulting from fire and explosion, so had
there been a fire and additional spillage, a comprehensive plan would ensure this response
circumstance was accounted for. The Watertown, WI derailment resulting in approximately 1,000
gallons spilled, which is a mere fraction of the quantity that would constitute a worst-case
“Search for answers: Cleanup enters day 2 after train carrying crude oil derails,” Deandra Corinthios, Krystle
Kacner, and Katie Delong, Fox 6 News, Nov. 9, 2015, available at: http://fox6now.com/2015/11/09/cleanupcontinues-in-watertown-after-train-carrying-crude-oil-derails/ [hereinafter “Fox 6 News, Search for answers”]
199
“Accident Investigation Report HQ-2015-1095,” Office of Railroad Safety, U.S. Federal Railroad Administration
(FRA), available at: https://www.fra.dot.gov/eLib/details/L18543#p1_z10_gD_lAC_kwatertown
200
“Guide for Preparing Hazardous Materials Incidents Reports,” Pipeline and Hazardous Materials Safety
Administration (PHMSA), Jan. 2004, available at:
https://www.phmsa.dot.gov/sites/phmsa.dot.gov/files/docs/reporting_instructions_rev.pdf
201
See I-2016010013
202
See Fox 6 News, Search for answers
203
Ibid.
204
Ibid.
198
Page 89
discharge; as such, a comprehensive plan would assist in maintaining a large margin of
preparedness for smaller release incidents such as the Watertown derailment.
A comprehensive plan may also assist in planning for and disseminating tactics to deal with
different tank car conditions resulting from derailment or puncture. The video footage from the
local news conference captures some of the considerations facing railroads, response providers,
and officials in offloading product from a damaged or derailed car and moving these cars away
from the site as appropriate. A comprehensive plan can be used to consolidate a plan holder’s best
practices, as well as the industry’s.
5.5. Culbertson, MT
Overview
On July 16, 2015, a BNSF freight train consisting of 106 tank cars loaded with crude oil derailed
22 tank cars near Culbertson, MT. 205 Three of the derailed tank cars were breached, spilling
approximately 35,000 gallons of crude oil, according to one news source. 206 According to
reporting to FRA, 5 cars released and the quantity released was 27,201 gallons. 207 The derailment
forced the evacuation of approximately 50 people. 208 No fires or injuries were reported.
Harm or Threat of Harm
EPA does not appear to have an OSC website for the Culbertson, MT derailment. EPA OSC
websites often provide information on the extent of oil spills and whether they pose harm or the
threat of harm to water resources.
The FRA investigation report states, “No water ways were affected by the spilled crude oil.” 209 It
provides that a BNSF hazmat team contained the spilled product using earthen dams. 210 In an
FRA statement, the evacuation of nearby residents was described as a precaution. 211
The PHMSA 5800.1 incident form indicates spillage, but no fire, explosion, or environmental
damage. 212 In the form’s “Description of Events,” BNSF notes that “[t]he spill was contained by
natural barriers on-site and free liquids were vacuumed and recovered for later disposal.” 213 This
narrative also relays, “Cleanup for the crude oil in soil remains underway including excavation
205
http://safetydata.fra.dot.gov/OfficeofSafety/publicsite/Query/incrpt.aspx
Ibid.
207
http://safetydata.fra.dot.gov/OfficeofSafety/publicsite/Query/incrpt.aspx
208
Ibid.
209
“Accident Investigation Report HQ-2015-1075,” Office of Railroad Safety, U.S. Federal Railroad
Administration, available at: https://www.fra.dot.gov/eLib/Find#p1_z10_gD_lAC_kculbertson [hereinafter “FRA
Culbertson report”]
210
Ibid.
211
“FRA Statement on Culbertson Montana Derailment,” Press Release Number: 99-9999, FRA Public Affairs, July
17, 2015, available at: https://www.fra.dot.gov/eLib/details/L16736
212
PHMSA 5800.1 report, # X-2015080186, available at:
https://hazmatonline.phmsa.dot.gov/IncidentReportsSearch/Welcome.aspx [hereinafter X-2015080186]
213
Ibid.
206
Page 90
product recovery and remediation.” Further, the incident form specifies the response costs as
$145,000 and remediation/cleanup costs as $481,965. 214
An external news source provides that soil was contaminated and underwent
cleanup/remediation. 215 Another source relays that crude oil did not reach any waterways. 216
In sum, the available information describes that the soil may have been affected, but that
waterways were not directly affected. It is unclear whether there was a threat of harm to
waterways.
Response and Discussion
In general, the information available regarding the spill response is limited. FRA and PHMSA
data do not outline spill response challenges, nor does there appear to be an EPA OSC report to
refer to.
The FRA investigation report provides some information, however, on the emergency and spill
response, noting the presence of Roosevelt County Sheriff’s Department, Montana Highway Patrol
and the Culbertson, MT Volunteer Fire Department. 217 It also notes that a BNSF hazmat team
responded, and an FRA conclusion: “…the emergency response was both immediate and thorough
with all precautions taken to ensure the safety of the general and traveling public within the area
of the incident.” 218 The same report indicates that local fire and police resources were at the site
by 6:03pm, or 9 minutes after the derailment occurred. 219 It is not clear when the BNSF hazmat
team arrived, but they arrived capable of applying earthen dams and transloading the crude oil
from the damaged cars. 220
While the available information limits what can be said about spill response improvements, a
comprehensive plan ensures that a railroad transporting large quantities of oil has available
sufficient resources to respond to a worst-case discharge. For smaller-sized spills, the WCD
planning volumes provide a margin of safety.
The Culbertson, MT derailment is also an example of the diverse array of organizations that can
be involved at a crude oil train derailment. Comprehensive plans can enhance the ability of plan
holders to effectively coordinate with Federal, State, and local partners by identifying personnel
and contacts and ensuring notification procedures are delineated within the comprehensive plan.
Ibid.
“Culberton, Mont., Derailment Update.” KFYR-TV, July 17, 2015, available at:
http://www.kfyrtv.com/home/headlines/Culbertson-Mont-Derailment-Updates-316185901.html
216
“Montana train derailment spilled 35,000 gallons of crude oil,” Reuters, July 17, 2015, available at:
https://www.reuters.com/article/us-usa-derailment-montana/montana-train-derailment-spilled-35000-gallons-ofcrude-oil-idUSKCN0PR24R20150717
217
See FRA Culbertson report
218
Ibid.
219
Ibid.
220
Ibid.
214
215
Page 91
6. Appendix C: Executive Order 13771
On January 30, 2017, the President issued Executive Order 13771, titled “Reducing Regulation
and Controlling Regulatory Costs.” Since publication of the Executive Order, OMB has issued
guidance to all Executive agencies as to how to account for the cost and cost savings on all
regulatory, deregulatory, and other policy documents. This final rule results in costs; therefore,
this rulemaking is subject to the requirements of Executive Order 13771.
The annualized costs of this rulemaking, using a 7 percent discount rate, are estimated at
$3.37 million over 10 years. PHMSA projects that the costs described in years 2–6 of the current
10-year analysis would form a repeating pattern into perpetuity. That is, the costs that occur in
years 2–5 recur in years 7–10 as shown in the analysis, and year 6 costs would occur in year 11 if
the analysis were extended that far, and that pattern of year 6 costs recurring after 4 years of year
2–5 costs would continue to recur in future years. That same pattern of recurring costs would
continue indefinitely with no obvious non-arbitrary stopping point. These recurring costs are used
in developing perpetual cost estimates and perpetual annualized costs for the rulemaking as
required by Executive Order 13771 accounting conventions. This analysis produces perpetual
annualized cost expressed in 2016 dollars, and discounted to 2019 (the year the rule goes into effect
and cost realized) using a 7 percent discount rate, of $3.23 million, with corresponding perpetual net
present value costs of $46.18 million.
The next step, which is conducted solely for OMB reporting, is to convert the annualized numbers
into values as of 2016. The RIA was developed using 2016 figures, so there is no need for inflation
adjustment. However, Executive Order 13771 accounting conventions also call for discounting
future year dollar figures back to 2016. In this case, the costs would begin to be incurred in 2019,
so discounting back to 2016 values results in estimated annualized costs of $2.64 million and total
costs of $37.7 million.
Page 92
| File Type | application/pdf |
| File Modified | 2019-03-04 |
| File Created | 2019-02-27 |